18 January 2013. You say “po-tay-toh,” and I say “po-tah-toh.” Or in the case of two sets of authors writing in the January 17 American Journal of Human Genetics online, one group says “cerebellar ataxia with spasticity,” while the other says “spastic paraplegia with ataxia.” Each found that mutations in β-glucocerebrosidase 2 (GBA2) causes the condition they studied. People with homozygous faulty GBA2 genes experienced both the disordered movements of ataxia and the muscle stiffness of spasticity. Clinically speaking, both spastic paraplegia and ataxia come in dozens of forms, but in these cases, the researchers were likely studying the same condition by two different names, said experts who spoke with Alzforum. “These diseases are essentially identical,” said Craig Blackstone of the National Institute of Neurological Disorders and Stroke in Bethesda, Maryland, who was not involved with either study.
GBA2 catalyzes the breakdown of the lipid glucosylceramide in the endoplasmic reticulum and plasma membrane. Its cousin, GBA1, made a name for itself as the enzyme deficient in Gaucher’s disease, Parkinson’s (see PDGene), and dementia with Lewy bodies. GBA1 performs the same function as GBA2, but in lysosomes. Without GBA1, lipids build up and can cause a variety of Gaucher’s symptoms, including skeletal abnormalities, liver enlargement, and mental disabilities. The finding that GBA2 mutations contribute to hereditary spastic paraplegia (HSP) and ataxia adds to other recent studies pushing lipid metabolism to the forefront of research on HSP (Tesson et al., 2012; Edvardson et al., 2008; reveiwed in Blackstone, 2012). “This disease shows how important lipid metabolism is for the brain,” said Ehud Goldin of the National Human Genome Research Institute in Bethesda, Maryland, who was not involved in the current studies.
Lipids also play a role in Alzheimer’s (e.g., see ARF related news story on Mielke et al., 2012; reviewed in Van Echten-Deckert and Walter, 2012) and Parkinson’s (see ARF related news story on Cullen et al., 2011, and Mazzulli et al., 2011).
First author Monia Hammer and senior author Andrew Singleton of the National Institute on Aging in Bethesda led the group that calls the GBA2 disease cerebellar ataxia with spasticity. Hammer, who is also a graduate student in the laboratory of Rim Amouri at the National Institute of Neurology in Tunis, Tunisia, was working with 10 Tunisian families who suffered recessively inherited cerebellar ataxia. All included consanguineous marriages. Having eliminated known ataxia loci as possible explanations, Hammer came to Singleton’s lab to perform single nucleotide polymorphism (SNP) genotyping and exome sequencing to identify mutations in the families.
The SNP genotyping told Hammer that three families shared a similar region on chromosome 9. All of the seven affected people in these families started out with ataxia but developed spasticity with time. The SNPs showed that two of those clans probably shared a common founder, along with a mutation identified by exome sequencing—a stop codon that prematurely cut off GBA2 at arginine 340. The third family possessed a missense mutation, arginine-873-histidine, predicted to interfere with the enzyme’s structure. “[GBA2] was the only gene shared by the three families,” Hammer said. Neither of these variants showed up in 40 neurologically normal control samples from Tunisia or 330 global controls. Further screening of Tunisians with ataxia, however, turned up yet another GBA2 variant, a stop codon at tyrosine 131.
Unbeknownst to Hammer and colleagues, the other group of researchers was zeroing in on the same gene. They were looking in cases of what they called complex HSP, because they included cerebellar ataxia. Co-first author Rebecca Schüle of the Hertie-Institute for Clinical Brain Research and Center for Neurology in Tübingen, Germany, was visiting the laboratory of Stephan Züchner at the University of Miami, Florida, to sequence the exomes of recessively inherited HSP samples. Schüle analyzed samples from her clinical practice as well as from that of Katrien Smets, also a first author, at the University of Antwerp in Belgium. In one family of Belgian origin, Schüle identified two nonsense mutations, truncating GBA2 at tryptophan 173 or threonine 492. And in a clan of Turkish descent, which also included married first cousins, she uncovered a stop codon at arginine 234.
Meanwhile, in yet another lab in Paris, France, Schüle’s co-first author Elodie Martin and senior author Giovanni Stevanin were also on GBA2’s trail, working with another consanguineous Tunisian kindred. Stevanin’s group, at the Pierre and Marie Curie University, had already narrowed the mutation site to a portion of chromosome 9 (Boukhris et al., 2010), and sequencing the exons in that region led them to a GBA2 arginine-630-tryptophan substitution. This variant, altering a conserved amino acid and predicted to alter protein structure, did not appear in 519 control samples from healthy subjects or 6,500 exomes in a public database. Finally, sequencing the coding portion of GBA2 in 95 more people with HSP identified a Portuguese person, again the product of a marriage of cousins, with the arginine-630-tryptophan substitution.
Neither set of authors knew what the other was up to until the actual papers came out. “I think it is probably the same disease,” Schüle said, citing the combination of spasticity and ataxia, the early age of onset, and the slow progression in both sets of participants. Combining data from the two papers, she added, means the condition is more common than she thought. She expects to identify more people with GBA2 deficiency in her practice.
How do these eight different mutations cause disease? Since five are truncations and the three substitutions are predicted to alter enzyme activity, a loss-of-function mechanism is likely. Supporting this idea, Schüle and colleagues found no measurable GBA2 activity in a white blood cell line from one of their subjects.
The group led by Stevanin examined disease pathology in zebrafish embryos with the GBA2 orthologue knocked down by antisense technology. Some of the fish exhibited a curly tail, and a quarter could barely move. When poked, they swam slowly and covered a shorter distance than normal fish. Their spinal motor neurons developed abnormally, with short, overly branched axons. Adding back the GBA2 gene, in the form of human GBA2 mRNA, fixed the defects, while mutant GBA2 was unable to repair the problem. GBA2 knockdown mice also exist; curiously, they show no major neurological defects (Yildiz et al., 2006), but their remaining 50 percent normal GBA2 activity may explain this.
The mechanism of GBA2 deficiency is uncertain. Indeed, the molecular and cellular pathology resulting from GBA1 loss in Gaucher’s disease is also poorly understood, said Pablo Sardi of Genzyme, a Sanofi company based in Framingham, Massachusetts. Genzyme makes a GBA1 substitute, Cerezyme®, which is used to treat Gaucher's. Hammer and colleagues wrote that the buildup of glucosylceramides in the endoplasmic reticulum or plasma membrane of neurons might interfere with calcium homeostasis, creating neurological disease (Lloyd-Evans et al., 2003). Calcium toxicity has been linked to many neurodegenerative diseases, including AD and ALS.
GBA losses may have broader implications. The GBA1 mutations behind Gaucher’s disease have been linked to increased risk for Parkinson’s and Lewy body disease in carriers of mild or single mutations (see ARF related news story). Some forms of HSP also include juvenile-onset parkinsonism, Blackstone noted. Schüle said she and Smets looked for evidence of PD in the GBA2 mutation carriers and their families, but found none. However, they have not seen patients in their seventies or older. The phenotype ascribed to GBA2 mutations may broaden as researchers analyze more carriers, Blackstone suggested. Scientists might also find that GBA2 mutations that do not completely abolish enzyme activity are linked to milder neurological symptoms, Goldin posited.
Researchers hoping to treat GBA2 deficiency can look to Gaucher’s treatment for guidance. Enzyme replacement therapy often works for GBA1, but the replacement protein does not enter the central nervous system. Genzyme is now working on getting its GBA1 therapy across the blood-brain barrier, Sardi said.
If the body cannot break down glucosylceramides, can drugs persuade cells to produce less? That is the reasoning behind Gaucher's therapies that inhibit glucosylceramide synthase, such as miglustat, made by the Swiss company Actelion. Combination therapy might also work. A synthase inhibitor with enzyme replacement therapy appears to work better than one treatment or the other for mice modeling other lysosomal diseases, Sardi said. Synthase inhibitors cross the blood-brain barrier, but some also inhibit GBA2 (Nietupski et al., 2012). Sardi said Genzyme has a synthase inhibitor that ignores GBA2.
Researchers are also developing chaperone therapies to boost GBA1’s ability to find lysosomes (isofagomine; Sun et al., 2011). A similar strategy might help people who express full-length but malformed GBA2, suggested Martin and colleagues.—Amber Dance.
Hammer MB, Eleuch-Fayache G, Schottlaender LV, Nehdi H, Gibbs JR, Arepalli SK, Chong SB, Hernandez DG, Sailer A, Liu G, Mistry PK, Cai H, Shrader G, Sassi C, Bouhlal Y, Houlden H, Hentati F, Amouri R, Singleton AB. Mutations in GBA2 cause autosomal-recessive cerebellar ataxia with spasticity. Am J Hum Genet. 2013 Feb 7;92. Abstract
Martin E, Schüle R, Smets K, Rastetter A, Boukhris A, Loureiro JL, Gonzalez MA, Mundwiller E, Deconinck T, Wessner M, Jornea L, Oteyza AC, Durr A, Martin JJ, Schöls L, Mhiri C, Lamari F, Züchner S, De Jonghe P, Kabashi E, Brice A, Stevanin G. Loss of function of glucocerebrosidase GBA2 is responsible for motor neuron defects in hereditary spastic paraplegia. Am J Hum Genet. 2013 Feb 7;92. Abstract