10 December 2010. In Alzheimer’s disease and other neurodegenerative disorders, misfolded proteins are seen banding together en masse, wreaking havoc in neurons. These findings have prompted research on potential treatments that rev up intracellular pathways to degrade and clear the rogue proteins. At the Society for Neuroscience annual meeting (SfN) held 13-17 November 2010 in San Diego, scientists talked about boosting autophagy in mouse models. One pharmacological approach improved pathology and behavior in tauopathy mice, while a genetic strategy showed similar benefits for a strongly amyloidogenic AD model.
Led by Wai Haung (Ho) Yu and Karen Duff, researchers at Columbia University Medical Center, New York, induced autophagy in the JNPL3 and rTg4510 tau transgenic models using trehalose, a disaccharide made by plants, fungi, and invertebrates. This compound has been shown to promote clearance of mutant huntingtin and α-synuclein in cell culture experiments (Sarkar et al., 2007), and to relieve motor dysfunction when administered orally in a Huntington’s mouse model (see Tanaka et al., 2004 and ARF related news story). The Columbia researchers wondered if the sugar could also relieve tauopathy in P301L transgenic mice with predominantly motor (JNPL3) or cognitive (rTg4510) phenotypes. Yu reported the results of these studies at SfN.
As a proof of concept, the scientists treated JNPL3 brain slice cultures with trehalose and other compounds that inhibit or induce autophagy. Trehalose did, in fact, promote autophagy (as judged by upregulation of LC3-II, a marker of autophagic vacuoles) and clear tau aggregates, as measured by a reduction in sarkosyl-insoluble tau detected using the human-tau specific antibody CP27 (Duff et al., 2000).
For the in vivo studies, the Columbia team then tested trehalose in preventive and therapeutic paradigms. They did the prevention study in JNPL3 mice, which develop neurofibrillary tangles, and behavioral and motor deficits that mimic human tauopathies (Lewis et al., 2000). Yu and colleagues gave JNPL3 mice 2 percent trehalose (or sucrose as a control) in their drinking water starting at 3.5 to four months of age, before tau pathology develops. After eight weeks, the treated group had lower levels of hyperphosphorylated tau (PHF1) and sarkosyl-insoluble tau (CP27) in the cortex than did the sucrose-fed controls. The trehalose group had better motor skills, as measured by rotarod and by how the animals fared on a hanging wire and when held by the tail. The researchers saw the same benefits in a therapeutic study in which they gave a four-week trehalose treatment to five-month-old mice with early-stage tau pathology.
Yu and colleagues also studied treatment in rTg4510 mice, which model AD and frontotemporal dementia and parkinsonism linked to chromosome 17 (FTDP-17) by showing tangle pathology, neuron loss, and cognitive impairment. They dripped trehalose in the drinking water starting at four to five months of age, when the mice already have aggregated tau. After eight weeks, treated rTg4510 mice had reduced levels of insoluble tau and aggregated p62, which is thought to bind tau and ferry it into the autophagic system for eventual degradation. Importantly, compared with sucrose-fed rTg4510 mice, the trehalose group also did better in the Morris water maze test of spatial learning and memory, though not quite as well as non-transgenic mice. Yu said his lab has a Tg4510 prevention study underway.
Company researchers seemed intrigued by the data. Yu told ARF he “fielded many questions from pharma,” including representatives from Pfizer, Bristol-Myers Squibb, Amgen, and Eli Lilly. Steve Jacobsen, who moved to Proteostasis in Cambridge, Massachusetts, after Pfizer bought his former company, Wyeth, said of the SfN data, “Overall I had a very favorable impression.” On a broader level, he noted that if a therapeutic strategy can be worked out for clearance of one misfolded, aggregated protein, there is a “good possibility it may clear other misfolded, aggregated proteins as well.”
In an SfN symposium entitled “AD-360°: Nonamyloid Mechanisms in Alzheimer’s Disease Pathogenesis” (see also Pimplikar et al., 2010), Ralph Nixon, New York University School of Medicine, spoke about a genetic approach for stimulating autophagy in TgCRND8 mice. This model has amyloid pathology starting at 8-10 weeks, with spatial memory loss developing soon after. Dun-Sheng Yang, of Nixon’s lab, established that TgCRND8 mice have impaired autophagy, as shown by grossly enlarged lysosomes, overabundance of ubiquitinated proteins, and defective proteolytic clearance of neuronal autophagic substrates including Aβ.
To reinvigorate the autophagy system in the TgCRND8 model, Yang and colleagues crossed them to mice genetically lacking cystatin B. Not to be confused with cystatin C, an AD genetic risk factor, cystatin B (CstB) inhibits the activity of cathepsins, enzymes that help break down proteins within lysosomes. Prior work in people (Rinne et al., 2002) and mice (Kopitar-Jerala and Turk, 2007) showed that reduced CstB activity correlates with increased cathepsin activity. Breeding TgCRND8 transgenic mice with CstB-knockout mice lifted the brakes on cathepsin activity, leading to enhanced autophagy as judged by high protein turnover rates in the crossed animals. The cross “represents what we consider a targeted intervention at the level of the lysosome,” Nixon said.
The NYU researchers analyzed the progeny at six months of age, when amyloidosis is well underway in the TgCRND8 strain. The autophagy defects were much reduced in the CstB-deleted transgenic mice, as shown by less intraneuronal Aβ, ubiquitinated proteins, and other autophagic substrates—including Aβ (see Yu et al., 2005)—clogging the system. CstB-deficient TgCRND8 mice also had less extracellular amyloid deposits and lower brain Aβ40 and Aβ42 levels. And on two cognitive/memory measures—fear conditioning and olfactory habituation (a newer test developed at NYU to measure memory for odors)—CstB-deleted TgCRND8 mice showed “significant improvement,” and were barely distinguishable from wild-type and CstB knockout controls, Nixon said. The findings will be published in the December 15 issue of Brain.
Nixon mentioned one caveat with the data. “Cystatin B deletion from birth creates its own pathology,” he told ARF. CstB-deleted mice model a form of childhood epilepsy, though they seem cognitively intact and their seizures are clinically invisible. Still, Nixon and colleagues were pleased to find that, despite the potential for “superimposed problems” from using CstB knockout mice, “our intervention was very effective in reversing the pathology [in TgCRND8 mice],” Nixon said. “One would not want to eliminate cystatin B from birth in a human to treat AD, but it's quite possible that once you get over the developmental period, a drug that mimicked the effect of eliminating cystatin B would be therapeutic.” To model a therapeutic paradigm, Yang and colleagues are now using viral vectors to induce RNA-mediated silencing of cystatin B in the brains of TgCRND8 mice.—Esther Landhuis.