16 August 2012. Scientists know little about how amyloid-β deposits interact with the complex architecture of the brain. Are plaques seeded randomly, or do they develop preferentially in certain areas? In the August 15 Journal of Neuroscience, researchers led by Edward Stern at Bar-Ilan University, Ramat Gan, Israel, provide evidence for the latter. They measured Aβ distribution in the barrel cortex of AD mice, and found that plaques tended to form between the barrels, a region that is rich in inhibitory interneurons. “This is the first direct evidence that plaques disrupt columnar organization,” Stern told Alzforum. Although human brains do not contain barrels, our neocortex forms columns, suggesting these findings could apply to people, also, he added. Intriguingly, the data imply that plaques may disrupt inhibitory synaptic inputs more than excitatory, a hypothesis Stern is currently testing.
“[The paper] is a thoughtful piece of neuroanatomy that may tell us about underlying mechanisms of plaque formation. It will be interesting to see how these patterns compare to what is seen in humans,” Brad Hyman at Massachusetts General Hospital, Charlestown, wrote to Alzforum. He has collaborated with Stern in the past but was not involved in the current work.
Previous work by Hyman and colleagues showed that amyloid deposits distort neuronal connections, causing neurites to twist and synaptic responses to falter, but did not directly demonstrate effects on columns (see Knowles et al., 1999; Stern et al., 2004). Some studies of postmortem human brain have found that plaques cluster in the cortex (see Beach and McGeer, 1992), or stack in a columnar fashion (see Akiyama et al., 1993), but because cortical columns are not easily visualized in people, these studies could not directly place plaques in the neuronal architecture.
To get around this limitation, Stern and colleagues examined 20-month-old APP/PS1 transgenic mice. Mouse barrel cortex receives stimulation from the animal’s whiskers. This input arrives in layer IV of the cortex, where the barrels are easily seen as distinct physical structures. First author Shlomit Beker measured the plaque distribution in layer IV, and found significantly more deposits between barrels than within barrels. This resulted in a highly clustered pattern of plaques. Tracing the columns upward, the authors saw that this precise pattern did not hold in layer II/III, but that plaques in this layer clustered, indicating they still followed some columnar organization. By contrast, the authors found a more random distribution of plaques in the visual cortex, a region that does not organize into columns in mice.
Since most excitatory inputs to layer IV neurons are found within the barrels, and many inhibitory inputs originate from the septae between barrels, the data suggest that Aβ deposits may selectively weaken neuronal inhibition. Functionally, inhibitory interneurons in the septae mediate lateral inhibition of adjacent whiskers, sharpening the animal’s ability to discriminate edges and textures. Plaques might blunt this, in effect, widening the receptive fields of barrel neurons, Stern proposed. He is currently testing this idea using electrophysiological recordings. Intriguingly, other studies have shown that neurons near amyloid plaques become hyperactive (see ARF related news story). Some strains of AD mice are prone to epileptiform activity (see ARF related news story; ARF news story), and anti-epileptic drugs were recently reported to correct cognitive deficits in AD mice (see ARF related news story). Stern told Alzforum he is very interested in following up on the epilepsy connection, and is doing recordings from both awake and anesthetized mice to look at specific types of synaptic activity and how they relate to plaque distribution.
Thomas Beach at Banner Sun Health Research Institute, Sun City, Arizona, found the research intriguing.“ It is exciting that some of the implications and hypotheses arising from the finding can be tested in the transgenic mouse model used,” he wrote to Alzforum. “What remains tantalizing is the promise that such patterns of degeneration hold for understanding disease. (See full comment below.)
Tara Spires-Jones, who is also at Mass General and has worked with Stern in the past, agreed, suggesting that future work could look at the effects of inhibitory and excitatory circuitry on plaque deposition in the mouse. To extend the work to humans, researchers might examine postmortem brains to see if Aβ aggregates occur near inhibitory synapses, she said. She believes the paper also demonstrates that mouse barrel cortex can be a good model system for AD. Because barrel cortex is highly active, it may have some similarities to the default-mode network in human brains, which is constitutively active and one of the earliest areas to accumulate Aβ, she said. Barrel cortex is also very plastic even in adult rodents, and so could be useful for measuring whether plaque distribution predicts plasticity deficits.—Madolyn Bowman Rogers.
Beker S, Kellner V, Kerti L, Stern EA. Interaction between amyloid-β pathology and cortical functional columnar organization. J Neurosci. 2012 Aug 15;32(33):11241-9. Abstract