Presenilin-1 is perhaps best known as the business end of the γ-secretase complex that cleaves amyloid-β from its precursor. But over the years, non-catalytic roles for the protease have been proposed, including regulation of calcium release from intracellular stores. Yet another function for presenilin-1 (PS1) debuted in the August 14 Nature Neuroscience online. Researchers led by Jane Sullivan at the University of Washington, Seattle, reported that PS1 modulates homeostatic synaptic scaling, a type of synaptic plasticity that functions at the level of neuronal networks. Interestingly, PS1 regulates this scaling independently of both γ-secretase activity and calcium mobilization, suggesting a completely new modus operandi for this transmembrane protease. “This is very solid work and clearly an interesting observation,” said Ilya Bezprozvanny, University of Texas Southwestern Medical Center, Dallas. “Though how it relates to Alzheimer’s disease will depend on the role of this [synaptic scaling] mechanism in maintaining network activity in the brain,” he told ARF. The study of homeostatic scaling is still in its infancy and has mostly been conducted in vitro, but hints of its in-vivo relevance have emerged, Sullivan told ARF. “We’re opening up a new avenue for exploration with the possibility of new treatment strategies,” she said.

Homeostatic scaling is one of several mechanisms that keep neurons firing within optimal range. Neurons have the ability to respond to their environment by boosting or suppressing synaptic responses, but that can’t become runaway, said Sullivan. Homeostatic scaling evolved to ensure neurons don’t become over- or underactive in response to inputs. Research on the phenomenon has focused mostly on neuronal cultures from young animals, but more recent work has uncovered the same property in adult mice.

First author Kara Pratt made the connection between presenilin-1 and synaptic scaling when characterizing the properties of PS1-null mutants. Pratt was a graduate student at Gina Turrigiano's lab at Brandeis University, Waltham, Massachusetts, before joining Sullivan’s group as a postdoc. Turrigiano is one of the pioneers of homeostatic scaling research, and, being familiar with its properties, Pratt recognized that the PS1-negative neurons behaved as though deficient in the process. A typical test is to completely block neuronal activity with tetrodotoxin for 24 to 48 hours, then look for an uptick above baseline once that blockade is relieved. Unlike wild-type neurons, hippocampal neurons from PS1-negative mice failed to scale up electrical activity in this paradigm. Neurons expressing PS1 with the M146V familial AD mutation, which retains γ-secretase activity, also failed to scale up activity, while cells treated with a γ-secretase inhibitor behaved as did wild type. The evidence pointed to a γ-secretase-independent control by PS1 over synaptic scaling.

How does PS1 regulate this plasticity? Turrigiano’s group found that homeostatic scaling relies on a drop in calcium concentration in the soma and a quenching of calmodulin kinase IV (CaMKIV) activity (Ibata et al., 2008). Since Bezprozvanny and others linked PS1 to calcium mobilization (see ARF related news story and ARF news story), Sullivan said they expected that PS1-negative neurons jeopardized scaling because they maintained their calcium and CaMKIV activity. “But actually, that was not it. It was not operating in the way we expected,” said Sullivan. Scaling failed in PS1-null neurons even when the researchers blocked activation of the calmodulin kinase.

Instead, Sullivan pieced together a totally different scenario. The researchers had used a virus vector to express wild-type presenilin to rescue the synaptic defect. The vector was a gift from Rachael Neve. When working at McLean Hospital, Belmont, Massachusetts, Neve discovered, together with researchers at Nikolaos Robakis’s lab at the Mount Sinai School of Medicine, New York, that PS1 knockouts fail to activate Akt kinase (see ARF related news story on Baki et al., 2008). Sullivan realized that a class of compound Neve used to block Akt, phosphatidylinositol-3-kinase inhibitors, was recently reported to prevent homeostatic synaptic scaling. “We thought, 'Aha! Perhaps Akt deficiency is to blame,’” said Sullivan. Sure enough, when Pratt introduced a constituently active Akt into neurons expressing the M146V PS1, she rescued the scaling defect. The results suggest that Akt signaling may be deficient in familial, and even perhaps sporadic AD, according to the researchers.

“These finding are inherently interesting in that they link mutations known to cause AD with a mechanism of impaired synaptic scaling, thereby supporting an increasingly investigated hypothesis that there are subtle, underlying deficits in synaptic function occurring long before overt symptoms of AD pathology,” wrote Beth Stutzmann, Rosalind Franklin University, Chicago, Illinois, in an e-mail to ARF (see full text below).

Sullivan is interested in testing whether homeostatic synaptic scaling is also deficient in other models of AD. She noted recent work linking Aβ to caspase-3 activation, which has been implicated in regulation of Akt activation (see Jo et al., 2011). “There may be ways to generate the same [homeostatic synaptic scaling] deficit with Aβ, so we could possibly extend these observations to include non-presenilin-based causes,” she said.—Tom Fagan


  1. This is an exciting, concise study demonstrating that wild-type presenilin (PS) is required for a critical aspect of synaptic homeostasis, namely, synaptic scaling, which reflects a neuron’s ability to dynamically alter its responses in an activity-dependent manner. Pratt et al. describe how the absence of PS, or the expression of FAD-linked mutant PS, impairs the ability of neurons to "recalibrate" to changes in network activity. Notably, the spontaneous synaptic potentials in the PS-manipulated neurons appear inherently normal, but they lack the ability to respond to changing stimuli, likely due to impaired downstream PI3K/Akt signaling. These findings are inherently interesting in that they link mutations known to cause AD with a mechanism of impaired synaptic scaling, thereby supporting an increasingly investigated hypothesis that there are subtle, underlying deficits in synaptic function occurring long before overt symptoms of AD pathology.

    On a broader perspective, these findings also integrate into a larger arena regarding early, "below the radar" deficits in neuronal functioning linked to AD pathology. Much research is focused on the salient, late-stage features (e.g., amyloid aggregates, tangle formation, and cell death), but these likely represent the "end of the line" of the pathogenic cascade. Studies such as these demonstrate that there are subtly destructive mechanisms operating much earlier in the disease process that alter the synaptic circuitry and metaplasticity that support learning, memory, and higher cognitive functions. A critical mass of synapses that lack the ability to adapt to changing activity levels can threaten the whole circuit, not unlike a few rusty wires short-circuiting the whole machine. Although targeting the late-stage markers of AD may be a desirable therapeutic goal, it seems, in light of the present study by Pratt et al., as well as complementary studies (Müller et al., 2011; Goussakov et al., 2010), that identifying earlier, substantive changes in neuronal function may provide mechanisms to actually alter the course of cognitive loss in AD.


    . Constitutive cAMP response element binding protein (CREB) activation by Alzheimer's disease presenilin-driven inositol trisphosphate receptor (InsP3R) Ca2+ signaling. Proc Natl Acad Sci U S A. 2011 Aug 9;108(32):13293-8. PubMed.

    . NMDA-mediated Ca(2+) influx drives aberrant ryanodine receptor activation in dendrites of young Alzheimer's disease mice. J Neurosci. 2010 Sep 8;30(36):12128-37. PubMed.

    View all comments by Grace Stutzmann

Make a Comment

To make a comment you must login or register.


News Citations

  1. Presenilins Open Escape Hatch for ER Calcium
  2. Pump It Up—Presenilins Linked to ER SERCA Activity
  3. Lithium Takes Indirect Route to Inhibit GSK-3β

Paper Citations

  1. . Rapid synaptic scaling induced by changes in postsynaptic firing. Neuron. 2008 Mar 27;57(6):819-26. PubMed.
  2. . Wild-type but not FAD mutant presenilin-1 prevents neuronal degeneration by promoting phosphatidylinositol 3-kinase neuroprotective signaling. J Neurosci. 2008 Jan 9;28(2):483-90. PubMed.
  3. . Aβ(1-42) inhibition of LTP is mediated by a signaling pathway involving caspase-3, Akt1 and GSK-3β. Nat Neurosci. 2011 May;14(5):545-7. PubMed.

Further Reading

Primary Papers

  1. . Presenilin 1 regulates homeostatic synaptic scaling through Akt signaling. Nat Neurosci. 2011 Sep;14(9):1112-4. PubMed.