15 Mar 2019

New data expands on the benefits of gamma wave therapy previously reported in mouse models of Alzheimer’s disease. Researchers led by MIT scientist Li-Huei Tsai report that sound can entrain gamma oscillations in both the auditory cortex and hippocampus. This, they say, triggers microglia to scarf down plaques and improves spatial and recognition memory. Adding a flickering light stimulus at the same time extended the waves’ reach into brain areas beyond the sensory cortices, inducing oscillations in the medial prefrontal cortex and reducing plaques across the brain. The results appeared online March 14 in Cell.

“The translational relevance is overwhelming,” said Costantino Iadecola, Weill Cornell Medicine, New York, who was not involved in the work. “It raises the possibility of using a noninvasive, easily applicable way to reduce the pathology we think is causing the disease.”

Tsai and colleagues stunned the field two years ago when they reported that simply flashing a 40-Hz light at mice for an hour a day for a week reduced Aβ production and spurred microglia to chew up plaques in the visual cortex (Dec 2016 news). They called this technique GENUS, short for gamma entrainment using sensory stimulus. The noninvasive treatment got its start when graduate student Hannah Iaccarino wondered what would happen if she restored the brain’s gamma waves, which wane in both AD mouse models and in people with the disease. In the current paper, the authors tested whether sound could induce the same gamma waves. Would auditory GENUS affect amyloid and microglia in the same way, and would this benefit memory?

First authors Anthony Martorell and Abigail Paulson exposed 6-month-old 5XFAD mice to a series of tones either at 20-, 40-, 80-Hz, or at random intervals. Using silicone probes to measure electrical output in the auditory cortex, hippocampus, and medial prefrontal cortex (mPFC), they found that these tones entrained oscillations in both the auditory cortex and hippocampal area CA1. After a week, mice hearing the 40-Hz sounds recognized novel objects and remembered their locations better than mice that received 20- or 80-Hz or random stimulation. Mice on 40-Hz treatment also found a hidden platform in the Morris water maze faster than controls. Examining their brains, the researchers saw that soluble Aβ42 had dropped by half and Aβ40 by a third in the auditory cortex and hippocampus. Plaque number and size fell by about 60 percent relative to controls.

The sound treatment also affected glia and blood vessels in the auditory cortex and hippocampus. Microglia appeared enlarged, with shorter, more branched processes, and they ate up more plaques, while astrocytes became up to 20 percent more reactive. At the same time, blood vessels grew wider, some doubling their diameter (see image below). Almost 60 percent more Aβ co-localized with the LRP1 receptor, which helps shepherd Aβ out of the brain and into the blood to be cleared in the circulation. The microglial effect was also seen in 9-month-old APP/PS1 mice. However, if the researchers examined mouse brains a week after the treatment had stopped, microglia returned to their normal size.

Auditory tones seemed to affect tau, as well. In the P301S model of tauopathy, levels of hyperphosphorylated tau and tau seeding both dropped after a week of treatment relative to unstimulated controls.

Could the auditory stimulation combine with light therapy to spread the effect across the brain? Stimulating wild-type mice with coupled 40-Hz light and sound entrained gamma oscillations in the auditory cortex, the hippocampus, and also the mPFC, an area strongly affected in Alzheimer’s disease. The mPFC did not respond to either visual or auditory stimulation alone, suggesting light and sound together reached more areas of the brain. In 5XFAD animals, microglia again became hungrier for plaques in the auditory and visual cortices, as well as the hippocampus; this effect, too, extended to the mPFC. Plaque load fell by around two-thirds relative to controls in those brain regions, and by a third across the whole cortex. The microglia were more apt to cluster around plaques.

“Combined auditory and visual stimulation had wider effects than in just the auditory and visual cortex,” John Cirrito, Washington University School of Medicine in St. Louis, wrote to Alzforum (see full comment below). “Since these cortical areas are not the hardest hit in AD, it is promising that multiple modalities could be employed to have a broader impact across the brain to reduce pathology.”

“This is an important extension of their previous work,” said Marcus Raichle, also at WashU. He believes that the balance of excitatory and inhibitory excitation in cells is likely altered by inducing gamma waves, which could partially explain the effects. The results beg for trials in humans, Raichle said, though he noted that so far reducing plaques and tangles in people has failed to rescue memory.

Human trials are underway. Tsai and co-author Edward Boyden co-founded Cambridge-based Cognito Therapeutics to translate their findings into potential therapies for humans. Cognito is running three clinical trials of combined auditory and visual stimulation in people at various stages of AD. Participants use wearable devices at home for about an hour per day for six to 12 months. The researchers are looking for changes in cerebrospinal fluid biomarkers, amyloid imaging, and cognitive tests. Company president Zach Malchano told Alzforum he expects to report results in the next year. “We want to be cautious about making sure we understand the therapies so they can be used appropriately. Although they are noninvasive, we want to be diligent and responsible.” Tsai said she is also exploring whether other sensory modalities—such as smell and touch—can be used to elicit gamma waves.

Meanwhile, other attempts at replicating the gamma effect and trying it in humans are popping up in the literature and on the internet. A recent manuscript on bioRχiv reported that two weeks of two-hour 40-Hz auditory stimulation activated microglia and reduced plaques in limbic and hippocampal regions of 6-month-old 5XFAD mice (Lee et al., 2018). In a pilot study, 10 days of light therapy had no effect on PiB-PET SUVR in six people with mild cognitive impairment or mild to moderate AD (Ismail et al., 2018). A slew of other iPad apps, desktop lamps, and light therapy kits are being tested and marketed.

“I anticipate that it will not be long before we know whether this intervention is effective in humans with MCI or AD,” Bruce Yankner, Harvard Medical School, wrote to Alzforum (see full comment below). He noted that the treatment effects appeared to be short-lived in the mice. “The reduction in amyloid load appears to dissipate seven days following the end of the stimulation period, suggesting that sustained treatment might be necessary,” he wrote. Investigating longer-term treatment regimens could test whether a more durable therapeutic response could be achieved, he said.

Other researchers urged that the vascular effect be explored further. “It would be of interest to examine additional properties of vascular changes, including blood-brain barrier permeability, fibrinogen deposition, brain endothelial and pericyte functions,” wrote Katerina Akassoglou, University of California, San Francisco, to Alzforum. Iadecola agreed, saying, “Before we can implicate vasculature factors in the clearance, we need to understand the effect of this stimulation on the resting cerebral blood flow.” He cautioned that the increased co-localization of Aβ with LPR1 could mean that the treatment causes cerebral amyloid angiopathy in these animals. Tsai said she is investigating that by tracking the fate of Aβ by two-photon microscopy.

“It is exciting to see recruitment of multiple brain areas by stimulation,” wrote Ksenia Kastanenka, Massachusetts General Hospital, Boston. “This is important to reversing AD pathology, which is widespread throughout the brain.” Gamma oscillations may not be the only brain waves beneficial for AD. Kastanenka found that optogenetically restoring slow oscillations during sleep stopped amyloid plaque deposition and prevented calcium overload in APP/PS1 mice (Kastanenka et al., 2017).—Gwyneth Dickey Zakaib

Comments

1. • David Attwell
     
   • Posted: 15 Mar 2019

   One of the earliest events in Alzheimer's disease is a decrease of blood flow (Iturria-Medina et al., 2016), which will reduce both the nutrient supply to neurons and amyloid removal. It is fascinating, therefore, that the auditory stimulus used in this study leads to a 50-100 percent increase in blood vessel diameter in auditory cortex and hippocampus (as assessed by the position of the basal lamina around vessels), which would be expected to greatly increase blood flow.

   It would be interesting to know whether the authors measured arteriole or capillary diameters, and whether live imaging could be used to see the time course of this change in vivo. I look forward to further work demonstrating that this result can be reproduced by other groups and establishing the mechanism of the effect.

   References:

   Iturria-Medina Y, Sotero RC, Toussaint PJ, Mateos-Pérez JM, Evans AC, Alzheimer’s Disease Neuroimaging Initiative. Early role of vascular dysregulation on late-onset Alzheimer's disease based on multifactorial data-driven analysis. Nat Commun. 2016 Jun 21;7:11934. PubMed.

   View all comments by David Attwell
2. • John Cirrito
     Washington University
   • Posted: 15 Mar 2019

   This is very interesting. I was fascinated by the visual stimulus data a couple years ago, and now the same by the auditory stimulus. This paper goes into a lot of detail to counter the hypothesis that changes in neuronal activity driving a suppression in Aβ production.

   I agree with the authors’ argument that is likely a clearance effect and not activity. First, they detect changes in microglia and astrocytes that are consistent with a clearance mechanism. Second, GENUS reduces pathology in seven days. I find it difficult to believe activity would have a rapid effect in reducing plaques that fast, whereas enhanced clearance could be that rapid. Third, GENUS worked in their exogenous tau seeding experiment; altered activity could affect seeding, I guess, but the more likely explanation is enhanced clearance.

   That several modalities at 40Hz are capable of suppressing AD-related pathology is very exciting. It suggests that the underlying mechanism can be engaged via multiple routes that could reach beyound just visual and auditory cortex. Adding to that, combined auditory/visial stimulation had wider effects than in just the auditory and visual cortex; since these cortical areas are not the hardest hit in AD, it is promising that either multiple modalities could be employed or combinations could be used to have a broader impact across the brain to reduce pathology.

   View all comments by John Cirrito
3. • Bruce Yankner
     Harvard Medical School
   • Posted: 15 Mar 2019

   This new study is an elegant tour de force linking neural network function to the cognitive and pathological features of AD. Although application to humans remains to be established, the protean effects of gamma entrainment using auditory and visual stimulation are impressive, with improved memory, reduction of amyloid and tau pathology, and effects on microglia, astrocytes and vascular function in AD and FTD mouse models. The reduction in amyloid load from auditory stimulation appears to dissipate seven days following the end of the stimulation period, suggesting that sustained treatment might be necessary. It would be of interest to investigate this with longer-term treatment regimens to know if a more durable therapeutic response could be achieved.

   A surprising finding was that changes in pathology and behavior could not be explained by overall changes in neuronal firing rate, suggesting that some other higher-order function of gamma entrainment, perhaps number or type of neurons entrained, might be relevant. Another interesting observation is that the effects of gamma entrainment on Aβ pathology are mediated, at least in part, by microglial activation. How gamma stimulation, but not other neuronal firing patterns, regulate microglial biology and induce Aβ clearance is an important question.

   I anticipate that it will not be long before we know whether this intervention is effective in humans with MCI or AD.

   View all comments by Bruce Yankner

Make a Comment

To make a comment you must login or register.

References

News Citations

1. Flashy Treatment Synchronizes Neurons, Lowers Aβ in Mice 7 Dec 2016

Paper Citations

1. Lee J, Ryu S, Kim HJ, Jung J, Lee B, Kim T. 40 Hz acoustic stimulation decreases amyloid beta and modulates brain rhythms in a mouse model of Alzheimers disease. BioRxiv. August 20, 2018. BioRxiv.
2. Ismail R, Hansen AK, Parbo P, Brændgaard H, Gottrup H, Brooks DJ, Borghammer P. The Effect of 40-Hz Light Therapy on Amyloid Load in Patients with Prodromal and Clinical Alzheimer's Disease. Int J Alzheimers Dis. 2018;2018:6852303. Epub 2018 Jul 30 PubMed.
3. Kastanenka KV, Hou SS, Shakerdge N, Logan R, Feng D, Wegmann S, Chopra V, Hawkes JM, Chen X, Bacskai BJ. Optogenetic Restoration of Disrupted Slow Oscillations Halts Amyloid Deposition and Restores Calcium Homeostasis in an Animal Model of Alzheimer's Disease. PLoS One. 2017;12(1):e0170275. Epub 2017 Jan 23 PubMed.

Further Reading

Papers

1. Buss SS, Fried PJ, Pascual-Leone A. Therapeutic noninvasive brain stimulation in Alzheimer's disease and related dementias. Curr Opin Neurol. 2019 Apr;32(2):292-304. PubMed.

News

1. Excited Neurons Release More Aberrant Tau 29 Jun 2016
2. Rocking Improves Sleep and Memory in Adults 25 Jan 2019
3. Brain-Wave Patterns Distinguish Parkinson’s Subtypes 24 Aug 2018
4. Disturbed Sleep Exerts Toll on Memory and Neurodegeneration 21 Dec 2017