When neurons stress out, they put the kibosh on protein production until conditions improve. If the stress persists, the shutdown can be disastrous for synaptic function. Now, a study led by Sergio Ferreira at the Federal University of Rio de Janeiro and Eric Klann at New York University reports that a small molecule inhibitor of the integrated stress response, called ISRIB, counteracted neuronal stress in the face of Aβ oligomers or rampant amyloidosis in mice. The treatment prevented synaptic deficits and memory loss, and even restored memory in plaque-ridden mice.

  • The small molecule inhibitor ISRIB counteracted stress response in mice.
  • ISRIB prevented synaptic and memory loss in Aβ toxicity model.
  • In amyloid-ridden mice, ISRIB restored synaptic function and memory.

The study, published February 2 in Science Signaling, adds to a growing body of evidence implicating a stress-induced slump in protein production in neurodegeneration, and further increases confidence in targeting this pathway for treatment of Alzheimer’s and related disorders, noted Giovanna Mallucci of Cambridge University in Cambridge, England, U.K.

The integrated stress response (ISR) exists for a good reason—to allocate precious cell resources to the most crucial tasks. One way the cell does this is by shutting down the bulk of new protein synthesis, allowing continued production of only essential proteins. Processes that rely on constant turnover of new proteins—including synaptic signaling and plasticity—take a hit (Trinh and Klann, 2013). 

The shutdown begins when stress-induced kinases, including PERK, PKR, and GCN2, phosphorylate the alpha subunit of elongation initiation factor-2 (eIF2α). In its unphosphorylated form, eIF2α functions as part of a heterotrimeric complex along with eIEF2β —a guanine nucleotide exchange factor—and eIF2γ, to initiate translation. However, once phosphorylated, eIF2α blocks GTP from binding to eIF2β, grinding protein synthesis to a halt. Phosphorylated eIF2α has been detected in the brains of people with neurodegenerative diseases, including AD, PD, and ALS. Previously, Ferreira, Klann and colleagues reported that suppression of eIF2α kinases alleviated deficits in synaptic plasticity and memory in mouse models of Aβ toxicity (Aug 2013 newsLourenco et al., 2013). 

Because kinases are notoriously difficult to target, and there are four of them that conspire to phosphorylate eIEF2α, the authors decided to disable the ISR downstream. They used ISRIB, a small molecule that binds eIF2β, preventing its deactivation by phosphorylated eIF2α (Sidrauski et al., 2013Zyryanova et al., 2018). Previous studies reported that the inhibitor stemmed neuronal damage in mouse models of cellular stress, prion disease, Down’s syndrome, and traumatic brain injury (Sidrauski et al., 2015; Halliday et al., 2015Zhu et al., 2019; Chou et al., 2017). 

Before testing the inhibitor in mouse models, first author Mauricio Oliveira and colleagues confirmed previous reports that the ISR was overactivated in the AD brain. Using western blot to probe cortical extracts from brain samples of eight people with AD, the researchers detected nearly double the amount of phosphorylated eIF2α as in samples from eight controls. In addition, AD brain samples had only half as much eIF2β as controls. The findings indicated that indeed, the ISR was elevated in the AD brain.

Could disabling the ISR prevent damage caused by Aβ oligomers? Previous studies had demonstrated that Aβ oligomers not only induce the stress response, but also hinder synaptic function and cause memory loss in mice. Here, the researchers reported that after injecting a single dose of 10 pmol Aβ oligomers into the cerebral ventricles of wild-type mice, the density of dendritic spines in the hippocampus crept downward over 12 days, and the mice began to forget the environment in which they had previously received a foot shock. Daily intraperitoneal injections of ISRIB—started on the same day as the oligomer injection—prevented both the spine loss and memory deficits. The researchers also found that the injected Aβ oligomers induced the ISR—as gauged by eIF2α phosphorylation and reduced protein synthesis, among other indicators—and that treatment with ISRIB prevented ISR activation downstream of eIF2α phosphorylation, and kept protein production up and running.

Might ISRIB treatment pass the higher bar of reversing damage in older mice burdened with Aβ plaques? To find out, the researchers treated 10- to 13-month-old APP/PS1 and wild-type mice for two weeks with daily doses. Compared to wild-type animals, APP/PS1 mice had about 20 percent fewer dendritic spines in the hippocampus. ISRIB restored spine density to nearly wild-type levels, but had no effect on spine density in wild-type mice. The researchers also found that treating hippocampal slices from APP/PS1 mice with ISRIB corrected defects in long-term potentiation, a measure of synaptic plasticity. These synaptic benefits translated into a cognitive boost. ISRIB-treated APP/PS1 mice performed at wild-type levels on spatial learning and contextual memory tasks.

Saving Spines. Hippocampal neurons in APP/PS1 mice had fewer dendritic spines compared to wild-type mice. Treatment with ISRIB restored spine density to nearly wild-type levels. [Courtesy of Oliveira et al., Science Signaling, 2020.]

Although squelching the stress response restored synaptic function and memory, it did not rid the mice of Aβ plaques. ISRIB-treated APP/PS1 mice had just as many as did untreated mice, although they were smaller and denser. The researchers speculated that changes in microglial function could underlie this last finding, though there were no overt changes in microglial number or Iba1expression with treatment. In all, the findings suggest that ISRIB can restore synaptic function even in the continued presence of substantial amyloid deposition.

The findings also hint that low, repeated doses of ISRIB might boost protein synthesis without causing side effects. This would be welcome news, because in previous studies in mouse models of amyloidosis, one dose was ineffective and daily doses of 5 mg/kg—20 times higher than what Oliveira and colleagues used—were toxic (Briggs et al., 2017; Johnson and Kang, 2016). PERK inhibitors—which block the ISR upstream of eIF2α—are also toxic (Halliday et al., 2015). 

“The paper points to dysregulation of protein synthesis as a cause of synapse loss and memory impairment in AD,” commented Peter Giese of King’s College London. “This work also suggests that synapse numbers and learning abilities can be restored by compounds like ISRIB in the early stages of the disease.” He added that the mouse models used in the study do not develop tau pathology, so more work is needed to establish the validity of the approach in people with AD.

“Converging evidence suggests that the ISR may be a central molecular switch for memory consolidation, applicable to a wide range of neurodegenerative disorders and diseases, brain injury, and aging,” commented Susanna Rosi and Elma Frias of the University of California, San Francisco. “These findings strengthen the idea that targeting the ISR may represent an effective therapeutic strategy to ameliorate AD-associated memory deficits.”

They noted that Calico Life Sciences, which recently licensed ISRIB, will begin a Phase 1 trial, partnered with Abbvie, of its eIF2β activator, ABBV-CLS-8262, in people with ALS this year (press release).—Jessica Shugart


  1. This paper provides additional data for the integrated stress response (ISR) as playing an active role in models of neurodegenerative disease. This response pathway both impacts protein synthesis and leads to stress-granule formation. Prior studies suggest this stress response is chronically activated in animal models and patients with ALS and prion disorders. Moreover, ISR pathways are important in synaptic function and plasticity.

    Here, Oliviera and colleagues demonstrate that ISRIB, which interrupts this stress-response pathway and precludes its shutdown of protein synthesis, improves phenotypes in two models of Alzheimer’s disease. Perhaps most interesting is the corrections observed in synaptic function in these models. 

    While valuable, it is important to note that these studies were both performed in overexpression model systems, which may overrepresent the importance of these pathways in AD disease pathogenesis. As such, further studies validating these findings will be needed to fully interpret the role of this pathway in AD.


  2. Targeting Protein Synthesis Translational Control to Treat Dementia.

    Memory disorders pose a major threat to public health and are responsible for an enormous economic and social burden. Long-term memory formation diminishes with normal aging and among many clinical conditions, including brain injury, and neurodegenerative diseases. The integrated stress response (ISR)—a cell-intrinsic translation regulatory network—constitutes the underlying pathway of cognitive dysfunction in numerous disorders.

    Indeed, activation of the ISR impairs long-term memory formation. The ISR is a conserved intracellular signaling pathway that serves as a universal protein synthesis regulator for the eukaryotic initiation factor eIF2 (eIF2α), via the guanine nucleotide exchange factor (eIF2β). Increased phosphorylation of eIF2α (eIF2α-P) halts global protein synthesis and reprograms translational upregulation of a select subset of mRNAs. ISR activation is detected in postmortem brains from individuals with, and animal models of, Alzheimer’s disease (AD), Parkinson’s disease, Huntington’s disease, amyotrophic lateral sclerosis (ALS), traumatic brain injury, and Down’s syndrome. At preclinical level, cognitive deficits in mouse models of Alzheimer’s disease, traumatic brain injury, and aging result from ISR activation. Notably, a small molecule that inhibits the integrated stress response (ISRIB) stabilizes eIF2β even in the presence of eIF2α-P, thus restoring global protein synthesis. Targeting the ISR, and specifically eIF2β activity via ISRIB, represents a compelling approach to restore memory consolidation in numerous pathological conditions.

    In this study Oliveira and colleagues analyzed postmortem brain tissues from Alzheimer’s disease patients and found increases in eIF2α-P and decreases in eIF2β protein levels. These findings strengthen the idea that targeting the ISR may be an effective therapeutic strategy to ameliorate AD-associated memory deficits.

    Next, the authors used two different mouse models of AD (acute injections of Aβ oligomers and mice carrying the APP Swedish mutation characterized by age dependent accumulation of Aβ [APPswe/PS1∆E9 mice]). Systemic low-dose injections of ISRIB over several days rescued measures of synaptic function and memory deficits. Interestingly, while the treatment reduced Aβ plaque size, it also resulted in an increase in plaque density. These results suggest that while ISR interference directly affects neuronal function (measured by LTP, dendritic spine density, and memory measures), it also modulates Aβ pathology. While the results are promising, further investigations are needed to understand the downstream molecular mechanisms by which ISR activation modulates AD pathophysiology, specifically Aβ plaques and tau pathology, the other key component of AD.

    Converging evidence suggests that the ISR may be a central molecular switch for memory consolidation, applicable to a wide range of neurodegenerative disorders and diseases, brain injury, and aging (image below). Inhibition of the ISR has been shown to reverse cognitive and synaptic deficits induced by traumatic brain injury, aging, and Down’s syndrome (Chou et al., 2017; Krukowski et al., 2020; Krukowski et al., 2020; Zhu, 2019). ISRIB shows no overt toxicity in mice and promises to become a powerful treatment for cognitive dysfunction.

    Fig. 1 from Costa-Mattioli and Walter, Science 368:eaat5314, Review 2020.

    Since the discovery of ISRIB in 2013 at the University of California, San Francisco, by the Walter lab, the drug has been licensed to Calico, which recently announced with Abbvie its use in clinical trials. Specifically, its lead elF2β activator, ABBV-CLS-7262, is currently progressing through Phase 1 with the plan to begin a trial later this year in patients with ALS (press release). 

    Alzheimer’s disease was first described in the early 1900s and so far, current treatments have only been able to partially address AD-associated symptoms. While ISRIB may or not be a cure for AD, it is promising that targeting a pathway shared by many pathological conditions could greatly advance pharmacological treatments for dementia.


    . The integrated stress response: From mechanism to disease. Science. 2020 Apr 24;368(6489) PubMed.

    . Small molecule cognitive enhancer reverses age-related memory decline in mice. Elife. 2020 Dec 1;9 PubMed.

    . Integrated Stress Response Inhibitor Reverses Sex-Dependent Behavioral and Cell-Specific Deficits after Mild Repetitive Head Trauma. J Neurotrauma. 2020 Jun 1;37(11):1370-1380. Epub 2020 Feb 11 PubMed.

    . Inhibition of the integrated stress response reverses cognitive deficits after traumatic brain injury. Proc Natl Acad Sci U S A. 2017 Aug 1;114(31):E6420-E6426. Epub 2017 Jul 10 PubMed.

    . Small molecule ISRIB suppresses the integrated stress response within a defined window of activation. Proc Natl Acad Sci U S A. 2019 Feb 5;116(6):2097-2102. Epub 2019 Jan 23 PubMed.

    . Activation of the ISR mediates the behavioral and neurophysiological abnormalities in Down syndrome. Science. 2019 Nov 15;366(6467):843-849. PubMed.

  3. This paper from Oliveira et al. shows exciting data in AD mouse models that add to the growing body of evidence for repressed protein synthesis rates contributing to memory failure and neuronal loss in neurodegenerative diseases. Critically, it further builds confidence in dysregulated proteostasis as a drug target for the treatment of these disorders. Reduced translational rates are seen in the brain across the range of mouse models of neurodegenerative diseases, from Parkinson’s to prion to ALS and frontotemporal dementia, associated with overactivation of the Unfolded Protein Response (UPR), particularly the PERK branch. The resultant high levels of PERK-P and its downstream target eIF2α-P lead to repression of translation (protein synthesis rates) that starves synapses of essential proteins and leads to synapse loss and eventually neurodegeneration.

    Earlier data from the authors had shown the effects of genetically restoring protein synthesis by targeting PERK or eIF2α kinases in the related integrated stress response (ISR) in AD mice, backed up by genetic and pharmacological targeting of PERK in various mouse models. ISRIB and molecules targeting the pathway downstream of eIF2α-P, such as the repurposed drug trazodone, have been particularly exciting due to lack of toxic side effects (both on- and off-target) seen with PERK inhibitors that limited enthusiasm for UPR inhibition as a therapeutic strategy.

    Previous work has shown that the small molecule ISRIB restores translation and memory in Down’s syndrome, prion disease, traumatic brain injury and in aged mice, but data from AD models was controversial—likely due to dosing regimens. Here, Oliviera and colleagues show that regular dosing with ISRIB at lower levels restores protein synthesis, memory, and synaptic plasticity in different mouse models of AD—both transgenic APP/PS1 mice and Aβ oligomer treated wild-type mice. The brains of patients with these disorders, notably Alzheimer’s disease and tauopathies, show high levels of PERK-P and eIF2α-P and Oliviera further found reduced levels of eIF2β subunits in patients’ brains and a key component of the translational apparatus and “target” of eIF2α-P.

    It will be fascinating to see if patients with AD, like the mouse models, have reduced protein synthesis rates due to high eIF2α-P levels, as suggested both by histological findings and by the western blotting data from Oliviera. Then, the use of ISRIB-like molecules, or repurposed drugs such as trazodone, which act in a similar manner, would be predicted to restore crucial protein synthesis rates in patient brains, boosting memory and neuronal survival. Oliviera and colleagues’ elegant paper further increases confidence in targeting this pathway for treatment of Alzheimer’s and related disorders. 

Make a Comment

To make a comment you must login or register.


News Citations

  1. Boosting Protein Translation PERKs Up Synapses in Alzheimer's Mice

Paper Citations

  1. . Translational control by eIF2α kinases in long-lasting synaptic plasticity and long-term memory. Neurobiol Learn Mem. 2013 Oct;105:93-9. Epub 2013 May 22 PubMed.
  2. . TNF-α mediates PKR-dependent memory impairment and brain IRS-1 inhibition induced by Alzheimer's β-amyloid oligomers in mice and monkeys. Cell Metab. 2013 Dec 3;18(6):831-43. PubMed.
  3. . Pharmacological brake-release of mRNA translation enhances cognitive memory. Elife. 2013;2:e00498. PubMed.
  4. . Binding of ISRIB reveals a regulatory site in the nucleotide exchange factor eIF2B. Science. 2018 Mar 30;359(6383):1533-1536. PubMed.
  5. . The small molecule ISRIB reverses the effects of eIF2α phosphorylation on translation and stress granule assembly. Elife. 2015 Feb 26;4 PubMed.
  6. . Partial restoration of protein synthesis rates by the small molecule ISRIB prevents neurodegeneration without pancreatic toxicity. Cell Death Dis. 2015 Mar 5;6:e1672. PubMed.
  7. . Activation of the ISR mediates the behavioral and neurophysiological abnormalities in Down syndrome. Science. 2019 Nov 15;366(6467):843-849. PubMed.
  8. . Inhibition of the integrated stress response reverses cognitive deficits after traumatic brain injury. Proc Natl Acad Sci U S A. 2017 Aug 1;114(31):E6420-E6426. Epub 2017 Jul 10 PubMed.
  9. . Role of Endoplasmic Reticulum Stress in Learning and Memory Impairment and Alzheimer's Disease-Like Neuropathology in the PS19 and APPSwe Mouse Models of Tauopathy and Amyloidosis. eNeuro. 2017 Jul-Aug;4(4) Epub 2017 Jul 14 PubMed.
  10. . A small molecule targeting protein translation does not rescue spatial learning and memory deficits in the hAPP-J20 mouse model of Alzheimer's disease. PeerJ. 2016;4:e2565. Epub 2016 Oct 19 PubMed.

External Citations

  1. press release

Further Reading


  1. . Small molecule ISRIB suppresses the integrated stress response within a defined window of activation. Proc Natl Acad Sci U S A. 2019 Feb 5;116(6):2097-2102. Epub 2019 Jan 23 PubMed.

External Links

  1. DNL343

Primary Papers

  1. . Correction of eIF2-dependent defects in brain protein synthesis, synaptic plasticity, and memory in mouse models of Alzheimer's disease. Sci Signal. 2021 Feb 2;14(668) PubMed.