For the genomewide association study (GWAS), first author John Landers and senior author Robert Brown, both of the University of Massachusetts Medical School in Worcester, along with joint senior author Ammar Al-Chalabi of King’s College London, UK, led a group of scientists seeking genes involved in risk, age of onset, site of onset, and survival for sporadic ALS. They collected data on 1,821 people with ALS and 2,258 controls from centers in the U.S. and Europe. Among 288,357 SNPs catalogued, they found one hit, in intron 8 of the gene for kinesin-associated protein 3 (KIFAP3), which was associated with increased survival time. The researchers confirmed the statistical significance of the KIFAP3 SNP in an independent sample set of 538 people with ALS and 556 controls. The researchers found no new genes linked to risk or onset of sporadic ALS, nor did they confirm that genes fingered in other GWASs were truly relevant to disease (see ARF related news story).

“This is as mixed a population as you can get, and the numbers [of subjects] are pretty high,” said P. Hande Ozdinler of the Northwestern University Feinberg School of Medicine in Chicago, Illinois, who has worked with Brown in the past but was not involved in the current study. Those factors, she said, make the results convincing.

KIFAP3 is a cargo-binding subunit of a molecular motor that travels along microtubules. The SNP associated with longer survival, a cytosine instead of a thymine in intron 8, has an allele frequency of 28.9 percent, and 8.3 percent of the population is homozygous for the cytosine variant. Those homozygotes showed extended survival time, averaging 14 months, between the onset of ALS symptoms and death or need for an assistive breathing machine. (Heterozygotes showed a small benefit of two to three months.)

The intron 8 SNP led the scientists to a second SNP in the KIFAP3 promoter that is in linkage disequilibrium with their initial hit. The second SNP, 25 nucleotides up from the transcription start site, is a cytosine when linked to survival, and otherwise a guanine. The cytosine generates a putative binding site for the transcription factor family Sp1, members of which can both up- and downregulate transcription.

Landers and colleagues found evidence that KIFAP3 transcription is reduced in people homozygous for the cytosine alleles. They used quantitative RT-PCR of occipital lobe brain samples to show that KIFAP3 expression dropped by 41.1 percent in people homozygous for the cytosine variant in intron 8, compared to thymine homozygotes. Western blotting confirmed that KIFAP3 protein levels were also decreased, by 69.8 percent, in CC brain tissue compared to TT samples. In neuroblastoma cells transfected with the KIFAP3 promoter conjugated to luciferase, reporter activity dropped by 19.6 percent in the CC version, compared to the TT genotype.

The current study focused on sporadic ALS, which accounts for approximately 90 percent of cases, but could also be relevant to the familial form, the authors suggest. Among inherited cases, one-fifth are caused by mutations in the gene for superoxide dismutase 1 (SOD1). In a recent paper, researchers led by Toshiyuki Araki of the National Center of Neurology and Psychiatry in Tokyo, Japan, reported that mutant SOD1 binds, and may sequester, KIFAP3 in mice. KIFAP3 was associated with SOD1 aggregates in spinal cord sections from people who died from FALS as well (Tateno et al., 2009).

“Now we have something that is a direct drug target; if we could reduce [KIFAP3] in patients, there might be benefits,” Landers said. Yet his study is somewhat at odds with Araki’s, whose research indicated that KIFAP3 overexpression repaired transport deficits in cells expressing mutant SOD1—suggesting more KIFAP3, not less, is needed to combat ALS. “The transcriptional changes derived from the discovered SNP are extremely subtle,” Araki wrote in an e-mail to ARF. “In general, quantitative RT-PCR technology can be applied only to detect RNA expression differences greater than twofold; any expression changes smaller than twofold by qRT-PCR are always inconsistent and unreliable.”

Landers noted that the Western blots also showed reduction in KIFAP3, so he is confident that its transcription is genuinely reduced.

In addition, Araki pointed out that the authors did not analyze KIFAP3 expression in the corticospinal tract, the most relevant tissue for ALS. “Since there is no tissue specificity shown here regarding the reduction of KIFAP3 transcription, survival extension could well be due to non-neuronal roles of KIFAP3,” he suggested. Landers pointed out that it was challenging even to obtain brain tissue from people with the relevant genotypes, and it would be “extremely difficult” to collect spinal cord samples for similar analysis. Adding weight to Landers’s conclusions, Ozdinler noted that according to the Allen Brain Atlas, KIFAP3 is expressed in spinal motor neurons.

Axonal transport has been implicated in a number of neurodegenerative conditions (for review, see De Vos et al., 2008). Landers’s study “reinforces my belief that axonal transport is very important for motor neurons,” Ozdinler said. “Even a slight alteration may, in the long run, tip the balance” toward disease.

Modifying axonal transport, perhaps with a therapy targeting KIFAP3, might be one way to rebalance the scales. Another possibility to consider is a treatment aimed at muscle metabolism, according to the authors of the PLoS ONE paper. “Whether metabolic alterations and muscle contribute to the course of ALS is presently a hot debate in the field,” wrote Jean-Philippe Loeffler, the senior author, in an e-mail to ARF. Loeffler, first author Luc Dupuis, both of the University of Strasbourg, and colleagues studied transgenic mice that express uncoupling protein 1 under the muscle-specific promoter for creatine kinase (Couplan et al., 2002). Uncoupling protein 1, not normally expressed in muscle, causes mitochondria to produce less ATP, forcing muscle cells to ramp up their metabolism and consume lipids as fuel. This hypermetabolic state has been found in people with ALS (Desport et al., 2005).

Dupuis and colleagues found that the hypermetabolic animals suffered age-related deterioration of neuromuscular junctions leading to denervation. To further study the relationship between muscle metabolism and ALS, they crossed the hypermetabolic mice with a line expressing mutant human SOD1. Compared to single mutant mSOD1 mice, the double mutants developed disease at the same age, but progressed to end-stage approximately one week sooner than the single mutants.

“This paper is the first demonstration that altered muscle physiology is sufficient to produce a significant denervation process and motor neuron disease,” Loeffler wrote. The paper does not directly suggest a therapeutic pathway, he noted, but suggests “nutritional research may bring some help for better care of patients.”—Amber Dance.

References:
Landers JE, Melki J, Meininger V, Glass JD, van den Berg LH, van Es MA, Sapp PC, van Vught PWJ, McKenna-Yasek DM, Blauw HM, Cho TJ, Polak M, Shi L, Wills A-M, Broom WJ, Ticozzi N, Silani V, Ozoguz A, Rodriguez-Leyva I, Veldink JH, Ivinson AJ, Saris CGJ, Hosler BA, Barnes-Nessa A, Couture N, Wokke JHJ, Kwiatkowski TJ Jr, Ophoff RA, Cronin S, Hardiman O, Diekstra FP, Leigh PN, Shaw CE, Simpson CL, Hansen VK, Powell JF, Corcia P, Salachas F, Heath S, Galan P, Georges F, Horvitz HR, Lathrop M, Purcell S, Al-Chalabi A, Brown RH Jr. Reduced expression of the Kinesin-Associated Protein 3 (KIFAP3) gene increases survival in sporadic amyotrophic lateral sclerosis. PNAS Early Edition. 2009 May. Abstract

Dupuis L, Gonzalez de Aguilar JL, Echaniz-Laguna A, Eschbach J, Rene F, Oudart H, Halter B, Huze C, Schaeffer L, Boillaud F, Loeffler JP. Muscle mitochondrial uncoupling dismantles neuromuscular junction and triggers distal degeneration of motor neurons. PLoS ONE. 2009;4(4):e5390. Abstract