For a motor neuron to maintain the unwieldy meter-long axons that innervate the lower limbs requires all the cell’s parts to work together perfectly; the breakdown of one system can be the cell’s undoing. Motor neuron diseases such as amyotrophic lateral sclerosis (ALS) have been linked to mitochondria (see ARF related news story), axonal transport (see ARF related news story on Perlson et al., 2009), and RNA regulation (see ARF related news story). Now, two papers suggest that the root cause of hereditary spastic paraplegia (HSP), another motor neuron disease that occasionally overlaps with ALS, may come down to the turns and intersections of the endoplasmic reticulum. Part of this insight grew from research on Arabidopsis thaliana, a staple of plant geneticists.

A team of researchers from the National Institutes of Health and Harvard Medical School report in the August 7 issue of Cell that without atlastin-1, which is frequently mutated in the disease, the ER network cannot form properly. And in the August 20 Nature, researchers from Rice University in Houston, Texas, and the Eugenio Medea Scientific Institute in Conegliano, Italy, cement a role for atlastins in ER networks by showing that the protein drives membrane fusion in vitro. Unlike SNAREs or virus-mediated fusion, atlastins require energy from GTP to seal the membranes. “It’s the newest paradigm of membrane fusion,” said principal investigator James McNew of Rice University.

Other neurodegenerative diseases have been linked to the ER stress response (see ARF related news story on Saxena et al., 2009) and to its role in calcium homeostasis (see ARF related news story on Green et al., 2008), and reticulons—separate proteins that form ER tubules—may play both protective and toxic roles in Alzheimer disease (see ARF related news story on Shi et al., 2009). However, hereditary spastic paraplegia seems to be caused by malformations in the organelle itself. The disease, for which 45 genetic loci have been identified, causes a person’s legs to become weak and stiff. Unlike ALS, a relentlessly degenerative disease, HSP can occur during development and may not shorten a person’s lifespan.

The authors of the Cell paper found functional homologs for human atlastin in yeast and in plants. In the latter, in an uncanny similarity to neurons, the protein is essential for the proper formation of long, skinny root hairs—“the axons of the plant,” quipped Tom Rapoport of Harvard Medical School, one of the study’s principal investigators. This connection has inspired co-principal investigator Craig Blackstone, of the National Institute of Neurological Disorders and Stroke (NINDS) in Bethesda, Maryland, to consider plants as a model system for HSP.

A New Way to Fuse

Blackstone and colleagues, neuroscientists interested in spastic paraplegia, had been contemplating the atlastin-ER connection for several years. They previously showed that atlastin localizes to the ER, and mutations in it inhibit the formation of three-way ER intersections (Rismanchi et al., 2008). Rapoport, a cell biologist who studies the shaping of organelles, noticed that the atlastin-mutant ER morphology in that paper—long, unbranched tubules—resembled what his lab members had observed when they disrupted DP1/Yop1p and the reticulons (Voeltz et al., 2006). REEP1, a member of the DP1 family, is another gene commonly mutated in HSP. Blackstone and Rapoport teamed up, and recruited William Prinz, a former Rapoport lab member who now works at the National Institute of Diabetes and Digestive and Kidney Disorders in Bethesda, Maryland, to join them and contribute his yeast expertise. The joint first authors on the Cell paper are Junjie Hu, formerly of Harvard Medical School and now at Nankai University in Tianjin, China; Yoko Shibata of Harvard Medical School; and Peng-Peng Zhu of the NINDS.

The collaborators were able to link their favorite proteins together in co-immunoprecipitation experiments in rat brain extracts and COS-7 cells, suggesting that atlastins, reticulons, and DP1 interact. “They are almost certainly working in a complex,” Blackstone said. The reticulons and DP1 form the tubules, and atlastin joins them together, he suggested. Overexpression of atlastin mutants resulted in long, unconnected ER tubules, while overexpression of wild-type atlastin caused large sheets, as if all the tubules had fused together. To assay atlastin function, the researchers studied ER reconstitution in frog oocyte extracts. Anti-atlastin antibodies prevented the formation of an ER network, confirming the protein’s role in reticulation.

For the Nature study, Italian scientist Andrea Daga, who was studying atlastin, recruited McNew for his expertise in membrane fusion. The joint first authors are Genny Orso and Diana Pendin in the Italian lab, and Song Liu in the Texas group. These researchers focused on Drosophila atlastin. Flies without atlastin exhibit motor problems, just as humans with atlastin mutations do (Lee et al., 2008). The scientists found effects similar to the Cell study: RNAi knockdown of atlastin led to ER fragmentation, while atlastin overexpression generated large, non-tubular structures.

McNew and colleagues offer further evidence that atlastins link membranes and splice them together. By co-immunoprecipitating myc- and HA-tagged atlastin from transfected HeLa cells, the scientists confirmed previous findings that the protein self-associates (Zhu et al., 2003). The distinctly labeled atlastins also were able to link membrane vesicles together in experiments with purified HeLa cell extracts. The researchers then reconstituted artificial vesicles with atlastins, and gave them the opportunity to fuse. They used two vesicle populations: one was unlabelled, the other contained phospholipids tagged with rhodamine or a compound that activates its fluorescence in a FRET pairing. When the two populations were mixed, the fluorescence decreased, indicating that the vesicles had fused and diluted the FRET partners.

Both sets of researchers found that GTP was necessary for atlastin activity. This makes atlastin the first membrane fusogen shown to require an energy source, McNew said. The researchers suspect that atlastins may reach across the membrane divide to bind one another—similar to the way SNAREs join membranes. Then, Rapoport said, “some miracle happens, and the membranes fuse.” Atlastins may be akin to mitofusins, large GTPases that are hypothesized, but not yet proven, to fuse mitochondrial membranes, McNew said (for review, see Benard and Karbowski, 2009).

The Other Atlastins

Having linked atlastins to ER fusion, Rapoport, Blackstone, and colleagues went looking for corresponding proteins in other organisms. “We know ER is present in every eukaryotic cell, but we could never find this atlastin protein in anything more primitive than a fruit fly,” Blackstone said. To track down the missing atlastins, the DP1 interaction proved the crucial clue. DP1 is an ortholog of yeast Yop1p, and for that reason Rapoport had been eyeing a protein called synthetic enhancer of Yop1p, or Sey1p. GFP-tagged Sey1p localized to the ER, and sometimes appeared concentrated at three-way junctions. The protein also co-immunoprecipitated with Yop1p and a yeast reticulon, Rtn1p.

Sey1p then led the researchers to another atlastin ortholog: RHD3 in plants. Short for root hair defective, this gene has been studied in Arabidopsis thaliana, the mouse ear cress that has long been a darling of plant geneticists. Along with abnormally short and wavy root hairs (Scheifelbein and Somerville, 1990), RHD3 mutants exhibit unbranched ER (Zheng et al., 2004). All plant cells express RHD3, but root hairs seem to be particularly vulnerable to RHD3 mutations—as are axons to atlastin mutations. The structures are so alike, Blackstone said, that textbook diagrams are eerily similar: “If you didn’t write ‘root hair,’ people would think that was an axon.”

So will neuroscientists someday be planting seeds instead of seeding neural cultures? “Sure, why not?” said John Fink of the University of Michigan in Ann Arbor, who was not involved with the current study. “We are talking about a very highly conserved cellular process that is conserved across kingdoms.” Root hair abnormalities are readily apparent in plants, and Arabidopsis boasts powerful genetics tools that would allow scientists to screen for suppressor or enhancer mutations. Similarly, drugs that attenuate the effects of an RHD3 mutation might, someday, do the same for people with HSP. Blackstone is not yet ready to invest in his own greenhouse, but he is open to working with green-thumbed collaborators.

The research also suggests to Blackstone that the ER is the key to HSP. “In general, this is going to be a disease of ER morphology,” he said. Atlastin and REEP1, both now linked to ER structure, are among the most common genes mutated in people with HSP. Rounding out the top three is spastin, which is involved in severing microtubules. Microtubules and the ER are closely associated, however, and Rapoport noted that one isoform of spastin localizes to the ER. Blackstone suspects that his discoveries may also be relevant to some forms of ALS, which can include spastic paraplegia.

“It is quite a major advance in the HSP field,” said Evan Reid of the University of Cambridge in the UK, who was not involved with the current work. “Now we will all be looking at tubulation processes.” Fink, for his part, is not ready to accept all HSPs under the ER umbrella. “This expands our understanding of what atlastin can do,” he said, but noted that the protein could have other functions through which mutations contribute to disease. For example, HSP mutant proteins have also been linked to endosomes (Tsang et al., 2009) and axonal transport (Kasher et al., 2009). The next challenge will be to directly link atlastins and ER morphology to axon degeneration and dysfunction.—Amber Dance.

Hu J, Shibata Y, Zhu PP, Voss C, Rismanchi N, Prinz WA, Rapoport TA, Blackstone C. A class of dynamin-like GTPases involved in the generation of the tubular ER network. Cell. 2009 Aug 7;138(3):549-61. Abstract

Orso G, Pendin D, Liu S, Tosetto J, Moss TJ, Faust JE, Micaroni M, Egorova A, Marinuzzi A, McNew JA, Daga A. Homotypic fusion of ER membranes requires the dynamin-like GTPase atlastin. Nature. 2009 Aug 20;460(7258):978-83. Abstract


Make a Comment

To make a comment you must login or register.

Comments on News and Primary Papers

  1. This work provides a fascinating suggestion of molecular conservation in the control of polarized cell growth in plant root hairs and mammalian neurons. Several plant biology labs, including my own, have studied Arabidopsis root hair development for many years with the hope that it would provide fundamental insights into mechanisms of regulated cell expansion in eukaryotes. It appears that this recent publication has validated this view, and it is reasonable to expect that the further combined analysis of the atlastin-1/RHD3 protein (and perhaps other related molecules) in both neurons and root hairs will be a fruitful approach.

    View all comments by John Schiefelbein
  2. In 2006, after I read the function of Yop1 in ER membrane curvature, I believed that RHD3 and its homologs could play a primary role in the tubulation of the ER network, probably with a physical interaction with Yop1. Hu et al. provide a key evidence for what I believed.

    Yes, I always tells my students who are very much interested in a medical related career that the outgrowth of root hairs shares many similarity with outgrowth of neurites. With the advantage of genetics and a wealth of cell biology tools available in Arabidopsis, it is reasonable to expect that further research in RHD3 should shed light on the onset of the human disease HSP.

    View all comments by Hugo Zheng


News Citations

  1. DC: Magnifying Glass on Poisoned Mitochondria in ALS
  2. Slowly But Surely, Dynein Delivers Stress Signals in ALS Model
  3. New Gene for ALS: RNA Regulation May Be Common Culprit
  4. ER Struggles in Motor Neurons That Fall to ALS
  5. Pump It Up—Presenilins Linked to ER SERCA Activity
  6. Too Much of a Good Thing—A Toxic Aβ Blocker?

Paper Citations

  1. . A switch in retrograde signaling from survival to stress in rapid-onset neurodegeneration. J Neurosci. 2009 Aug 5;29(31):9903-17. PubMed.
  2. . A role for motoneuron subtype-selective ER stress in disease manifestations of FALS mice. Nat Neurosci. 2009 May;12(5):627-36. PubMed.
  3. . SERCA pump activity is physiologically regulated by presenilin and regulates amyloid beta production. J Cell Biol. 2008 Jun 30;181(7):1107-16. PubMed.
  4. . Reduced amyloid deposition in mice overexpressing RTN3 is adversely affected by preformed dystrophic neurites. J Neurosci. 2009 Jul 22;29(29):9163-73. PubMed.
  5. . Atlastin GTPases are required for Golgi apparatus and ER morphogenesis. Hum Mol Genet. 2008 Jun 1;17(11):1591-604. PubMed.
  6. . A class of membrane proteins shaping the tubular endoplasmic reticulum. Cell. 2006 Feb 10;124(3):573-86. PubMed.
  7. . Loss of spastic paraplegia gene atlastin induces age-dependent death of dopaminergic neurons in Drosophila. Neurobiol Aging. 2008 Jan;29(1):84-94. PubMed.
  8. . Cellular localization, oligomerization, and membrane association of the hereditary spastic paraplegia 3A (SPG3A) protein atlastin. J Biol Chem. 2003 Dec 5;278(49):49063-71. PubMed.
  9. . Mitochondrial fusion and division: Regulation and role in cell viability. Semin Cell Dev Biol. 2009 May;20(3):365-74. PubMed.
  10. . Genetic Control of Root Hair Development in Arabidopsis thaliana. Plant Cell. 1990 Mar;2(3):235-243. PubMed.
  11. . A GFP-based assay reveals a role for RHD3 in transport between the endoplasmic reticulum and Golgi apparatus. Plant J. 2004 Feb;37(3):398-414. PubMed.
  12. . The hereditary spastic paraplegia proteins NIPA1, spastin and spartin are inhibitors of mammalian BMP signalling. Hum Mol Genet. 2009 Oct 15;18(20):3805-21. PubMed.
  13. . Direct evidence for axonal transport defects in a novel mouse model of mutant spastin-induced hereditary spastic paraplegia (HSP) and human HSP patients. J Neurochem. 2009 Jul;110(1):34-44. PubMed.
  14. . A class of dynamin-like GTPases involved in the generation of the tubular ER network. Cell. 2009 Aug 7;138(3):549-61. PubMed.
  15. . Homotypic fusion of ER membranes requires the dynamin-like GTPase atlastin. Nature. 2009 Aug 20;460(7258):978-83. PubMed.

Further Reading


  1. . A class of dynamin-like GTPases involved in the generation of the tubular ER network. Cell. 2009 Aug 7;138(3):549-61. PubMed.
  2. . Homotypic fusion of ER membranes requires the dynamin-like GTPase atlastin. Nature. 2009 Aug 20;460(7258):978-83. PubMed.
  3. . Mutations in a newly identified GTPase gene cause autosomal dominant hereditary spastic paraplegia. Nat Genet. 2001 Nov;29(3):326-31. PubMed.
  4. . Spastic paraplegia, optic atrophy, and neuropathy: new observations, locus refinement, and exclusion of candidate genes. Ann Hum Genet. 2009 May;73(Pt 3):382-7. PubMed.
  5. . Apolipoprotein E4 domain interaction induces endoplasmic reticulum stress and impairs astrocyte function. J Biol Chem. 2009 Oct 2;284(40):27273-80. PubMed.
  6. . Gp78, an ER associated E3, promotes SOD1 and ataxin-3 degradation. Hum Mol Genet. 2009 Nov 15;18(22):4268-81. PubMed.
  7. . Quantitative and functional analyses of spastin in the nervous system: implications for hereditary spastic paraplegia. J Neurosci. 2008 Feb 27;28(9):2147-57. PubMed.
  8. . Isoform-specific increase of spastin stability by N-terminal missense variants including intragenic modifiers of SPG4 hereditary spastic paraplegia. Eur J Neurol. 2007 Dec;14(12):1322-8. PubMed.

Primary Papers

  1. . A class of dynamin-like GTPases involved in the generation of the tubular ER network. Cell. 2009 Aug 7;138(3):549-61. PubMed.
  2. . Homotypic fusion of ER membranes requires the dynamin-like GTPase atlastin. Nature. 2009 Aug 20;460(7258):978-83. PubMed.