The neurovascular focus of the first Zilkha Symposium on Alzheimer’s Disease and Related Disorders, held April 4 in Los Angeles, started off with talks on the blood-brain barrier and midlife hypertension (parts one and two of this series).  Proper blood pressure is crucial not only to keep liquid moving inside the brain’s vessels, however. According to Maiken Nedergaard, University of Rochester, New York, arterial pressure also drags cerebrospinal fluid along the vessel’s outside, and this type of flow is equally important. Normally, the pulsatility of the brain’s arteries creates a convective force that draws CSF into the brain from the space around its artery walls and back out again along its veins. Gently propelled by this peristalsis, CSF penetrates deep inside the brain’s parenchyma. There, in the neighborhood of the neurovascular units, CSF exchanges with interstitial fluid, percolating through and around astroglial end-feet toward nearby veins, along which it then exits the brain. In so doing, it takes with it neuronal waste products, essentially rinsing neurons like so many pots and pans in a dishwasher (see March 2013 news story,  Jan 2014 Alzforum Webinar, Wikipedia video below). “This is how the brain exports its waste,” Nedergaard said. 

  

In Los Angeles, Nedergaard reported that this flow wanes as people get older. “We see a striking reduction of CSF influx into the depth of the brain with age,” Nedergaard said. For one, she said, new experiments show that the pulsatility of the small penetrating arteries weakens alarmingly. For another, the distribution of the aquaporin receptor changes. Normally, these receptors are located on the astrocytic end-feet facing the arteries, where they facilitate uptake of fluid from the peri-arterial space and drive glymphatic flow toward nearby veins. In old mice, the receptor occurs all over the astrocytes in a way that facilitates fluid retention inside this cell more than it aids directed flow from arteries to veins. This may be one way in which astrogliosis damages the aging brain, Nedergaard said.

It is well known that waste products such as Aβ and other aggregation-prone proteins are cleared not only alongside veins but also across the blood-brain barrier into the venous bloodstream (part two of this series). Even so, Nedergaard argued, Aβ and α-synuclein, being synaptic proteins, need an interstitial fluid flux to reach the BBB, whether they then get transported across or washed out by way of paravenous efflux.

Do We Sleep To Clear Our Minds?

Glymphatic clearance peaks in mice when they snooze. This is so not only because aquaporin receptors are more highly expressed in sleep, but also because the interstitial spaces themselves expand. “The interstitial volume decreases in wakefulness. It is a huge difference that would make a big impact on convection flux,” Nedergaard told the audience. Most of the molecular underpinnings of the glymphatic system remain to be understood; in the meantime, a good therapy for an aging brain may be to sleep more, Nedergaard said.

This, too, is a recommendation from a growing literature on sleep and neurodegenerative disease more broadly. It appears to be a bi-directional relationship. Sleep apnea or shift work in mid-life increases people’s risk for dementia, and once dementia has taken hold, it further alters sleep patterns, with frequent and active waking periods after midnight. Another example of this relationship may be REM sleep behavior disorder, which predisposes to Parkinson’s and dementia with Lewy bodies, both of which in turn count excessive daytime drowsiness among their symptoms. 

In healthy adults, the Aβ concentration in the cerebrospinal fluid is subject to a circadian fluctuation that is somehow lost as people reach their 60s and 70s. David Holtzman of Washington University, St. Louis, tries to define how neural activity, the wakefulness hormone orexin, and sleep deprivation influence the concentration of Aβ and tau in those very interstitial spaces that get flushed by glymphatic flow. 

NMR solution structure of orexin. Knocking out this hormone induces sleep and furthers glymphatic clearance of waste proteins, including those involved in Alzheimer’s.

Some years ago, Holtzman’s group showed that synaptic activity regulates Aβ levels in extracellular spaces, and that orexin increases Aβ levels in the interstitial fluid (Sep 2009 news story). At the Zilkha conference, Holtzman showed new data suggesting that orexin does not do this directly, by affecting neuronal signaling locally; it does so indirectly, via its influence on how much the mice sleep. Local orexin expression experiments by way of injecting lentiviral constructs into the brain indicated as much. So did orexin knockout mice crossed with APP/PS1 transgenic mice, which sleep more and have less soluble and deposited Aβ. Accordingly, other ways of changing the amount of sleep also influence the amount of Aβ in fluids; for example, certain sleep-inducing drugs may lower CSF Aβ in human volunteers, Holtzman said. 

Tau, the other hallmark protein of Alzheimer’s disease, is becoming a target of such studies, as well. Tau is not generally thought to be related to sleep; however, evidence is growing that it is being released into the interstitial space—more so during periods of high neuronal activity, i.e., wakefulness. Recently, Holtzman’s lab showed that driving up excitatory activity increases ISF tau (Feb 2014 news story). In Los Angeles, Holtzman noted that the half-life of tau in the ISF is too long to expect the fast spike in response to sleep deprivation that has been seen with Aβ; however, chronic sleep deprivation might drive up tau. Other labs have begun linking sleep to tau in transgenic AD mice (De Meco et al. 2014Rothman et al., 2013Cantero et al., 2012). 

What Have microRNAs Got to Do With it?

While some groups known for their work on the signature proteins of Alzheimer’s disease have started exploring physiological processes such as glymphatic clearance and sleep, other groups are venturing into the uncharted swaths of the human genome to better understand gene regulation in the disease. Bart de Strooper, KU Leuven, Belgium, is searching the non-coding tracts that make up the vast majority of human DNA for genetic elements that may be important in age-related neurodegeneration. “Do not call anything in the DNA ‘junk.’ We have to scrutinize all of it,” de Strooper said. In Los Angeles, he presented new data on the microRNA 132 in Alzheimer’s disease. 

MicroRNA-132 precursor, secondary structure. [Wikimedia Commons image by permission of User:Ppgardne.]

MicroRNAs are small wisps of 18 to 23 nucleotides that bind to the 3’ UTR of target mRNAs and suppress their translation. They are encoded as independent genes, some in introns of other genes, others in what used to be dismissed as “junk” DNA. De Strooper’s lab started exploring microRNAs by deleting brain expression of dicer, an enzyme needed to process microRNAs. The resulting phenotype of tau phosphorylation, inflammation, and neuron loss intrigued the scientists sufficiently to look for microRNA binding sites in Alzheimer’s genes such as BACE and APP, where they indeed found them. Subsequent microRNA profiling of 49 AD brain samples from the Netherlands Brain Bank showed that expression of 41 microRNAs was different in the AD than control samples, and a follow-up study in AD brains from the MRC London Brain Bank confirmed the pattern. 

Of those 41, miRNA-132 proved to be markedly downregulated in the AD hippocampus and prefrontal and temporal cortex, and more so the farther advanced the patient’s disease had been by Braak staging. Deep sequencing and additional techniques confirmed that this finding is robust and reproducible, De Strooper told the audience.

It’s not just Alzheimer’s. This microRNA is down in Huntington’s, progressive supranuclear palsy, frontotemporal lobar degeneration, and amyotrophic lateral sclerosis. Berislav Zlokovic of the University of Southern California noted that stroke studies have shown the same thing. “We see this remarkable miRNA-132 downregulation long before neuron loss, so we think it is important,” de Strooper said. 

Sequence analysis predicts 407 miRNA-132 target genes in the human genome, including the genes for tau, GSK3β, and sirtuin1. In-vitro binding validated that, de Strooper said. To get a sense of what miRNA-132 might do, the scientists knocked it down it in zebrafish, which then had trouble swimming. In the developing zebrafish brain, the little RNA is strongly expressed in radial glia. These are progenitor cells that also guide migrating neurons. At the molecular level, miRNA-132 acts via notch, a γ-secretase substrate that is well known to be important in glial development. The microRNA affects notch by binding directly to the mRNA of the transcription factor CTBP2, which connects to notch via sirtuin-1.

Much remains to be found in this emerging pathway. “Even so, our data point in a speculative way toward a role of neurogenesis in brain aging, maybe some protective role in AD,” de Strooper said.

In the discussion, de Strooper said that the world of RNA should be explored with an emphasis on normal and diseased human brain for the time being, noting that the prospect of miRNA-based therapeutics appears premature.

The field’s growing interest in new cellular and molecular players in the pathogenesis of AD does not mean the original ones—APP and presenilin 1 and 2—have been set aside. At the Zilkha symposium, researchers presented new findings and model systems. Sam Sisodia of the University of Chicago discussed recently published research suggesting that Ab generation in cells other than excitatory neurons is important in Alzheimer’s pathogenesis (see Veeraraghavalu et al., 2014).

Rudy Tanzi of Massachusetts General Hospital, Charlestown, introduced a way of growing human neuronal precursor cells that express familial AD mutations of those genes in three dimensions by using a matrigel mixture; this is based on the work of Doo Yeon Kim of MGH.  As the cells secrete Aβ, the gelatinous consistency of the substrate restricts the peptide from diffusing away, and the cultures form both Aβ oligomers and amyloid plaques. Intriguingly, they also form hyperphosphorylated tau and even neurofibrillary tangles that stain with conventional silver stains, Tanzi told the audience. Beta- or γ-secretase inhibitors prevent the formation of both plaques and tangles in the cultures, whereas tau-based inhibitors prevent formation of tangles even in the continued presence of ample amyloid plaques, Tanzi said. This culture system might be useful for drug screening, and it could potentially recapitulate aspects of the pathology of other neurodegenerative diseases that feature secreted pathogenic proteins.—Gabrielle Strobel 

Comments

Make a Comment

To make a comment you must login or register.

Comments on this content

No Available Comments

References

News Citations

  1. It’s Not All About You, Neurons. Glia, Blood, Arteries Shine at Symposium
  2. Fluid Markers and Imaging Back Idea of Breached Blood-Brain Barrier
  3. Spinal Fluid Flush: Visualizing the Brain Drain With MRI
  4. Sleep Deprivation Taxes Neurons, Racks Up Brain Aβ?
  5. Neurons Release Tau in Response to Excitation

Webinar Citations

  1. Neurovascular Underpinnings of Alzheimer's Dementia

Paper Citations

  1. . Sleep deprivation impairs memory, tau metabolism, and synaptic integrity of a mouse model of Alzheimer's disease with plaques and tangles. Neurobiol Aging. 2014 Aug;35(8):1813-20. Epub 2014 Feb 15 PubMed.
  2. . Chronic mild sleep restriction accentuates contextual memory impairments, and accumulations of cortical Aβ and pTau in a mouse model of Alzheimer's disease. Brain Res. 2013 Sep 5;1529:200-8. PubMed.
  3. . Tau protein role in sleep-wake cycle. J Alzheimers Dis. 2010;21(2):411-21. PubMed.
  4. . Age-dependent, non-cell-autonomous deposition of amyloid from synthesis of β-amyloid by cells other than excitatory neurons. J Neurosci. 2014 Mar 5;34(10):3668-73. PubMed.

Further Reading

No Available Further Reading