3 March 2006. A diagnosis, let alone a cure, for amyotrophic lateral sclerosis has proven hard to pin down, but recent papers offer some hope for both. In the February 26 Nature Genetics, researchers in Europe report that mutations in the protein angiogenin are strongly linked to the disease. Because angiogenin mediates the action of vascular endothelial growth factor (VEGF), which is of therapeutic benefit in mouse models of ALS, the finding bolsters the promise of future VEGF-related treatment for patients. On the diagnosis front, a proteomics study published in the February 15 Neurology online suggests that measuring levels of three cerebrospinal fluid (CSF) markers can identify ALS with high degrees of accuracy, specificity, and sensitivity in a small initial cohort of patients. Curiously, one of the markers is a fragment of the neuroendocrine protein VGF, which recently surfaced as a potential CSF biomarker for Alzheimer disease.
ALS, also known as Lou Gehrig’s disease, causes degeneration of motor neurons, leading to paralysis and eventually death. ALS is rare, and in most cases there is no known cause. However, about 10 percent of ALS cases are inherited, suggesting that in some cases, at least, genetic mutations are to blame. But to date, mutations in only one gene—that for the copper-zinc superoxide dismutase (SOD1)—are proven to directly cause the disease. Now, Orla Hardiman and colleagues at the Royal College of Surgeons, Dublin, Ireland, together with collaborators elsewhere in Europe and the U.S., report that mutations in angiogenin (ANG) are strongly linked to ALS in Scottish and Irish populations. The findings suggest that at the very least, these mutations dramatically increase susceptibility to the disease.
First author Matthew Greenway and colleagues previously reported that a single nucleotide polymorphism (SNP) in the ANG gene associates with ALS in Irish families (see Greenway et al., 2004). Greenway and colleagues now take that observation one step further. Writing in Nature Genetics, they report the sequencing of the coding and flanking regions of the gene in almost 3,000 individuals (about 1,600 ALS cases and almost 1,300 controls) from five countries: Ireland, Scotland, England, Sweden, and the U.S. The authors identified seven missense mutations among 15 individuals with ALS, 12 of them from Irish/Scottish ancestry.
Of those 15 carriers, only four had a family history of ALS, suggesting that rather than being directly causative, the mutations may act more as risk factors. In support of this, the screen also identified one of the mutations in an apparently healthy 65-year-old man. The data hark back to similar findings for Parkinson disease, in which mutations in the LRRK2 (leucine-rich repeat kinase 2) gene associate with both familial and sporadic forms of the disease (see ARF related news story).
Angiogenin is a ribonuclease with potent angiogenic activity. Of the seven mutations, five, including the one found in the healthy control, are in highly conserved positions that are deemed functionally critical. One of the other two mutations affects an amino acid that might be important for nuclear transfer, while the role of the seventh is uncertain. The authors suggest that the mutations “may result in haploinsufficiency for ANG protein function or a previously unknown gain of function that may cause autosomal-dominant ALS with low penetrance.”
How either gain or loss of ANG function might contribute to ALS pathology is not clear. The ribonuclease is expressed in response to hypoxia, or lack of oxygen, as is VEGF. Although no mutations in VEGF have been associated with ALS, Peter Carmeliet and colleagues at the University of Leuven, Belgium, showed that abolishing the hypoxia promoter on the VEGF gene induces ALS-like disease in mice (see ARF related news story). Subsequently, Carmeliet and virologist Nicholas Mazarakis at Oxford BioMedica, England, were able to delay onset of disease and increase lifespan in SOD1 mutant mice by using viral vectors to deliver VEGF to motor neurons (see ARF related news story). All these findings indicate that variations in VEGF and angiogenin, and perhaps other hypoxia-related genes, may have important roles in ALS pathology, and they suggest that VEGF-related therapies are worth a closer look. Carmeliet recently said in an interview with the Journal of Clinical Investigation that clinical trials for VEGF are underway (see JCI 2005).
If a treatment can be found for ALS, then diagnosis will take on even greater importance. Currently there is no simple test for the disease, and diagnosis often involves painstakingly eliminating other possibilities and watching how symptoms progress. Early diagnosis, which could someday lead to early intervention, is only possible for those rare cases attributable to SOD mutations.
Writing in Neurology, Giulio Pasinetti and colleagues at the Mount Sinai School of Medicine, New York, and elsewhere, present a proteomic approach to ALS diagnosis. Pasinetti and colleagues used mass spectrometry to analyze CSF from 36 ALS patients. They compared the profiles with those obtained from 21 controls and found that levels of three proteins were significantly lower in patient CSF. When these proteins were analyzed in combination, Pasinetti and colleagues found that they could identify ALS with 95 percent accuracy. The number of false negatives and false positives were promising (91 percent sensitivity, 97 percent specificity, respectively). Should these findings be validated in longitudinal studies with greater number of patients, this “three-protein” model could serve as the basis for an early diagnosis.
Of the three markers, the researchers were able to identify two. One was cystatin C, a cysteine protease inhibitor that is found in Bunina bodies, small intracellular inclusions that are found in motor neurons in ALS patients. This finding strengthens an initial report of cystatin C as a proteomic CSF marker for ALS by Robert Bowser and colleagues published last fall (Ranganathan et al., 2005). The other was a 4.8 kDa fragment of VGF, a neuroendocrine protein that is up-regulated by nerve growth factor. VGF has structural similarity with the chromogranins, which bind to mutant but not wild-type SOD (see ARF related news story). Studies also link CSF VGF with frontotemporal dementia (see Ruetschi et al., 2005) and AD (see Selle et al., 2005). Numerous studies have implicated cystatin C in AD genetically (see Alzgene page) and as a potential biomarker (see Carrette et al., 2003). This suggests these proteins may be fairly general markers for neurodegenerative diseases.—Tom Fagan.
Greenway MJ, Andersen PM, Russ C, Ennis S, Cashman S, Donaghy C, Patterson V, Swingler R, Kieran D, Prehn J, Morison KE, Green A, Acharya KR, Brown Jr. RH, Hardiman O. ANG mutations segregate with familial and "sporadic" amyotrophic lateral sclerosis. Nature Genetics advanced online publication. February 26, 2006. Abstract
Pasinetti GM, Ungar LH, Lange DJ, Yemul S, Deng H, Yuan X, Brown RH, Cudkowicz ME, Newhall K, Peskind E, Marcus S, Ho L. Identification of potential CSF biomarkers in ALS. Neurology. Expedited E-Pub, 15 February, 2006. Abstract