7 December 2010. Though it could well be the subcortical brain region hit hardest in Alzheimer’s and Parkinson’s, the blue spot, or locus ceruleus, has only recently begun to draw wider attention among neurodegeneration scientists. New work presented at this year’s annual meeting of the Society for Neuroscience, held 13-17 November 2010 in San Diego, could help place this lesser-known cluster of noradrenergic neurons more squarely on the AD research map. In a nanosymposium, Doug Feinstein of the University of Illinois, Chicago, reported that restoring dwindling supplies of brain noradrenaline reduced cognitive deficits in an AD mouse model that has robust pathology. Recent publications have also solidified connections between the locus ceruleus and AD.
Locus ceruleus (LC) neurons die and brain noradrenaline levels drop with normal aging (Marien et al., 2004), even more so in AD (German et al., 1992). AD mouse models also have these features (see German et al., 2005; O’Neil et al., 2007). Several recent studies have gone a step further by demonstrating functional consequences of LC degeneration in neurological disease models. Ahmad Salehi, Veterans Affairs Palo Alto Health Care System, and colleagues at Stanford University, restored contextual memory in APP-overexpressing Down’s syndrome transgenic mice (Ts65Dn) by giving subcutaneous infusions of L-threo-3,4-dihydroxyphenylserine (aka L-DOPS), a brain-penetrant noradrenaline precursor (Salehi et al., 2009 and ARF related news story). Though the mice already had significant LC degeneration, hippocampal neurons that were downstream targets of LC cells still responded to norepinephrine by firing when treated with the β1- and β2-adrenergic receptor agonist isoproterenol, the authors noted. More recently, Feinstein’s lab used a similar strategy to slow disease progression in experimental autoimmune encephalomyelitis (EAE) mice modeling multiple sclerosis (Simonini et al., 2010). At SfN, Feinstein presented new data suggesting the approach can help in AD mouse models.
This study used Bob Vassar’s 5XFAD strain, which has three mutations in its human APP transgene as well as two presenilin-1 mutations. Starting at six weeks of age, these mice develop robust amyloid pathology. They lose neurons and show massive glial inflammation leading to cognitive impairment by five to six months. Sergey Kalinin of Feinstein’s group established that the LC is compromised in these animals: LC neurons had rampant inflammation, as judged by immunostaining, which revealed a 50 percent increase in GFAP-positive cells compared to wild-type LC cells. Furthermore, 5XFAD LC neurons had unusually large cell bodies, likely an effect of stress, Feinstein said.
The researchers administered L-DOPS to 4.5-month-old 5XFAD mice, which are free of cognitive impairment but on the verge of decline. The treatment involved subcutaneous L-DOPS injections along with two other drugs given intraperitoneally. The first was carbidopa, which blocks conversion of L-DOPS to noradrenaline in the periphery. The second was atomoxetine, a noradrenaline reuptake inhibitor. The combination specifically boosts noradrenaline in the brain. The mice were treated three times a week for four weeks, and one month later, faced tests of spatial learning and memory before being sacrificed for biochemical and immunohistochemical analyses.
The postmortem analysis showed that the L-DOPS treatment increased CNS noradrenaline levels and reduced plaque burden and astrocyte activation in the hippocampus and frontal cortex. Behaviorally, the treated mice fared better than controls. In one set of experiments, control- and L-DOPS-treated 5XFAD mice were assessed on the Morris water maze four times daily for three days. “The control group basically didn’t learn,” Feinstein reported. After 12 trials, these mice were taking almost as long to find a hidden escape platform as they did in the first trial. By the tenth trial, however, L-DOPS-treated mice did start finding the platform faster than the control animals. But they did not perform as well as wild-type mice, which require fewer trials to learn the platform location, and once they learn, find the platform more quickly, Feinstein noted.
On a related water maze probe test, L-DOPS-treated transgenics showed better long-term memory than did the untreated 5XFAD group. This version of the test measures time spent in the quadrant that had contained the hidden platform seven days prior. During that one-week lag, untreated transgenic mice forgot where the platform was, whereas the L-DOPS-treated group showed some recall. At the meeting, Feinstein proposed that these cognitive benefits may involve increases in neurotrophin levels, as mRNA analysis showed higher expression of nerve growth factor (NGF) and brain-derived neurotrophic factor (BDNF) in L-DOPS-treated mice relative to the control group. Expression of the Aβ-degrading factors neprilysin and insulin-degrading enzyme (IDE) also increased in treated mice, raising the possibility that upregulation of these enzymes may mediate L-DOPS’s amyloid-lowering effects, Feinstein said.
The use of L-DOPS in these studies is interesting, in part because it “could work even after many of the LC neurons degenerate,” commented David Weinshenker of Emory University in Atlanta, Georgia. This is because the enzyme that converts L-DOPS to noradrenaline (i.e., aromatic acid decarboxylase) is present in dopaminergic and serotonergic neurons, as well as noradrenergic neurons. Second, L-DOPS is already known to be safe. The compound (aka Droxidopa) has been used in Japan to treat “freezing gait” symptoms in PD, Weinshenker said. In the U.S., it is being tested in a Phase 3 trial of PD patients with neurogenic orthostatic hypotension, a condition where loss of peripheral adrenergic receptors in the sympathetic nervous system causes reduced blood pressure. “If L-DOPS has efficacy in the APP mice, it could quickly be translated to humans and open up a new avenue of treatment for AD,” Weinshenker noted.
A BioMed Central Medical Genetics paper published online on November 11 reinforces the link between noradrenergic neuron loss and AD. European researchers genotyped 1,757 AD patients and 6,294 elderly controls, and homed in on a single-nucleotide polymorphism that decreases activity of dopamine β-hydroxylase, which converts dopamine to noradrenaline. In their analysis, this SNP (-1021T) associated with increased AD risk (Combarros et al., 2010). The work was led by senior investigator Donald Lehmann of the Oxford Centre for Gene Function in the U.K. and first author Onofre Combarros of the University of Cantabria in Santander, Spain.
Furthermore, in a recent study, Steffen Rossner, University of Leipzig, Germany, and colleagues probed various regions of mouse and human AD brain for expression of glutaminyl cyclase (QC). This is the enzyme that catalyzes formation of N-terminally truncated Aβ peptides into pyroglutamate Aβ, a highly aggregation-prone species that has attracted increasing attention from AD researchers (see ARF Chicago story and ARF San Diego story). In the August issue of Acta Neuropathologica, first author Markus Morawski and colleagues report finding QC expression in locus ceruleus neurons, but not in adjacent brain structures, of mouse and human AD samples (Morawski et al., 2010). In line with this, AD brain samples had intraneuronal pyroglu-Aβ and extracellular pyroglu-Aβ deposits in the LC. The authors take these data as support for a “scenario in which human QC-expressing LC neurons are intoxicated by formation of pyroglutamate Aβ,” Rossner wrote in an e-mail to ARF. “Additionally, QC and/or pyroglu-Aβ may be released at hippocampal synapses of LC neurons, thus contributing to hippocampal pyroglu-Aβ deposition. This mechanism may contribute to the known degeneration of LC neurons in AD and in subsequent noradrenergic denervation of the hippocampus.” Alzforum will host a Webinar on the role of the LC in AD on 16 December 2010.—Esther Landhuis.