Rosenblum Lab Molecular and Cellular Mechanisms Underlying Learning and Memory

    Memory and mRNA translation

    Many studies by our group and others have shown that mRNA translation into protein is required for memory consolidation, i.e., the conversion of labile, short term memory to relatively stable, long term memory (Gal-Ben-Ari and Rosenblum. 2012; Rosenberg et al. 2014). Since the foundation of the lab in 2001, we have been studying the regulation of mRNA translation into protein as a major molecular mechanism underlying learning and memory.

    Over the years, we have focused on specific molecules that serve as key regulators of protein synthesis, and shown them to be potential targets for memory enhancement in health and disease (e.g., mild dementia, Alzheimer's disease). These molecules include the eukaryotic initiation factor 2α (eIF2α) and its kinases (enzymes regulating its phosphorylation levels, and thereby, its activity). eIF2α has four known kinases, all known to be functional in the brain.

    We have shown that eIF2α and at least two of its kinases, protein kinase R (PKR) and PKR-like endoplasmic reticulum kinase (PERK), play an essential role in learning and memory. For example, eIF2α+/S51A mice (genetic reduction of eIF2α phosphorylation) have enhanced synaptic plasticity and enhanced hippocampal- and cortical-dependent learning (Costa-Mattioli et al. 2007). Furthermore, we have shown that treatment with a protein kinase R (PKR) inhibitor (C16) (rats, WT mice) or genetic deletion of eIF2α kinase PKR (PKR KO mice) results in enhanced memory in the conditioned taste aversion  paradigm (associative learning) and the novel taste learning paradigm (both paradigms are cortical-dependent) (Stern et al. 2013). Moreover, phosphorylation of eIF2α is increased with age in rats and mice, and in the ApoE4 mouse model of sporadic Alzheimer's disease, phosphorylation levels of PKR and eIF2α are increased (Segev et al. 2013), and PKR inhibition rescues memory impairment, as shown in the contextual fear conditioning paradigm (hippocampal-dependent) (Segev et al. 2015). Our findings regarding PKR, as a potential target for memory enhancement in health and disease, have been granted a patent (US 8,334,262 B2). Based on our findings regarding PKR and this patent, in 2015 we founded a start-up company (Protekt Therapeutics, part of the FuturX biotechnological incubator, Israel) aiming to develop PKR inhibitors, invested by Takeda Pharmaceuticals and Johnson & Johnson.

    In addition to PKR, we have shown that PERK plays an essential role in learning and memory and synaptic plasticity as well, and serves as the major regulator of eIF2α phosphorylation.  We have shown that reduction of either the expression levels (knockdown using a viral vector) or the activity of PERK (using a pharmacological inhibitor), in the insular cortex resulted in enhanced performance in the cortical-dependent novel taste learning and conditioned taste aversion paradigms (Ounallah-Saad et al. 2014). Similar inhibition of PERK activity or reduction of its expression levels in the CA1 region of the hippocampus in young adult male mice enhances neuronal excitability and improves cognitive function in the hippocampal depended trace fear conditioning (associative learning) paradigm. In addition, PERK knockdown or inhibition rescues the age-dependent cellular phenotype of reduced excitability and memory decline. Specifically, the reduction of PERK expression in the CA1 region of the hippocampus of middle-aged male mice using a viral vector rejuvenates hippocampal function and improves hippocampal-dependent learning (Sharma et al. 2018). This mechanism underlying aging both on the whole animal, behavioral level and on the neuronal function level points to PERK as a promising therapeutic target for age-dependent brain malfunction (US patent application 15335466).

    While eIF2α and its regulating kinases regulate the initiation (first) phase of protein synthesis, considered to be the rate limiting phase of protein synthesis, we have shown that another molecule, eukaryotic elongation factor 2 kinase (eEF2K), a major regulator of the elongation (second) phase of protein synthesis, is also essential for memory formation and synaptic plasticity consolidation  (Taha et al. 2013; Heise et al. 2017). eEF2K is the only known kinase of eukaryotic elongation factor 2 (eEF2). We have found that genetic deletion of eEF2K (knock-out, KO) in mice, which leads to complete loss of eEF2 phosphorylation, differentially affects hippocampal-dependent memory formation. From a clinical perspective, our results identify eEF2K as a potential novel target for antiepileptic drugs, since pharmacological and genetic inhibition of eEF2K can revert the epileptic phenotype in a mouse model of human epilepsy (Heise et al. 2017). Currently, we are examining the function of the eEF2 pathway in the dentate gyrus (DG) of the hippocampus. (Taha et al., manuscript in preparation).

    In our latest manuscript, we delved into the field of depression, and tested the hypothesis that both eEF2K and another kinase, Ca2+ /calmodulin-dependent protein kinase II (CaMKII), mediate the anti-depressant effect of ketamine. We identified CaMKII as new target for ketamine (and have a provisional patent on it, US application number 62/556,440). Interestingly, our results suggest that drugs selectively targeting calcium-calmodulin dependent kinases, specifically eEF2K, may offer a novel strategy for the treatment of major depressive disorder (Adaikkan et al. 2018).


    Measuring mRNA translation in neuronal processes and somata by tRNA-FRET Koltun B, Ironi S, Gershoni-Emek N, Barrera I, Hleihil M, Nanguneri S, Sasmal R, Agasti SS, Nair D, Rosenblum K.Nucleic Acids Res. 2020 Jan 24. pii: gkaa042. doi: 10.1093/nar/gkaa042. [Epub ahead of print]

    Genetic or pharmacological reduction of PERK enhances cortical-dependent taste learning. Ounallah-Saad H, Sharma V, Edry E, Rosenblum K.J Neurosci. 2014 Oct 29;34(44):14624-32. doi: 10.1523/JNEUROSCI.2117-14.2014

    Local Inhibition of PERK Enhances Memory and Reverses Age-Related Deterioration of Cognitive and Neuronal Properties. Sharma V, Ounallah-Saad H, Chakraborty D, Hleihil M, Sood R, Barrera I, Edry E, Kolatt Chandran S, Ben Tabou de Leon S, Kaphzan H, Rosenblum K. J Neurosci. 2018 Jan 17;38(3):648-658. doi: 10.1523/JNEUROSCI.0628-17.2017. Epub 2017 Dec 1.

    Calcium/Calmodulin-Dependent Protein Kinase II and Eukaryotic Elongation Factor 2 Kinase Pathways Mediate the Antidepressant Action of Ketamine. Adaikkan C, Taha E, Barrera I, David O, Rosenblum K.Biol Psychiatry. 2018 Jul 1;84(1):65-75. doi: 10.1016/j.biopsych.2017.11.028. Epub 2017 Dec 5

    PKR Inhibition Rescues Memory Deficit and ATF4 Overexpression in ApoE ε4 Human Replacement Mice. Segev Y, Barrera I, Ounallah-Saad H, Wibrand K, Sporild I, Livne A, Rosenberg T, David O, Mints M, Bramham CR, Rosenblum K. J Neurosci. 2015 Sep 23;35(38):12986-93. doi: 10.1523/JNEUROSCI.5241-14.2015.








    Taste and the insular cortex

    The identification of a novel source of food as such and the recognition of a familiar source of food along with retrieval of its "safe" or "hazardous" labels is essential for the survival of all species, and is therefore evolutionarily conserved. As such, taste-related learning and memory paradigms are highly robust (Gal-Ben-Ari et al. 2012; Yiannakas and Rosenblum. 2017).

    Taste learning is unique compared to other learning forms dependent on other senses, especially in associative learning: the time interval during which association between the sensory stimulus (conditioned stimulus) and the unconditioned stimulus (a stimulus which triggers a physical or an  emotional response such as malaise, fear, satisfaction) lasts seconds to minutes in paradigms involving other senses, whereas paradigms involving the sense of taste have an association interval of up to 8 hours (Adaikkan and Rosenblum. 2015). The cortical gustatory region resides within the anterior part of the insular cortex and thus we aim to dissect out the role of the insular cortex in taste learning and interactions with other brain structures. However, the insular cortex is involved not only in taste information but in many other functions. We have identified different molecular mechanisms underlying taste learning which take place in the insular cortex (Belelovsky et al. 2005; Yefet et al. 2006; Costa-Mattioli et al. 2007; Elkobi et al. 2008; Belelovsky et al. 2009; Stern et al. 2013; Inberg et al. 2013; Rappaport et al. 2015; Levitan et al. 2016; Rosenberg et al. 2016a, 2016b; Inberg et al. 2016). We are continuing this line of research, but in addition are interested in identifying the specific cells and brain circuits which are part of the taste memory formation process. Using two-photon calcium imaging of defined gustatory cortex neurons in vivo, we have recently shown that conditioned taste aversion dynamically shifts neuronal population coding by recruiting neurons that project to the basolateral amygdala (Lavi et al., under revisions). In parallel, we measure neuronal activity in the insular cortex using tetrodes together with a unique system for delivering taste solutions (Salalha, Holzman, and Rosenblum, in progress). Another way to measure activity in the insular cortex is a mini-microscope, which we are setting up these days. Currently, we aim to put the identified molecular mechanisms within the context of specific cells and circuits in the insular cortex (Yiannakas, Kayal, and Rosenblum, in progress). 



    Activity of Insula to Basolateral Amygdala Projecting Neurons is Necessary and Sufficient for Taste Valence Representation.

    Kayyal H, Yiannakas A, Kolatt Chandran S, Khamaisy M, Sharma V, Rosenblum K.

    J Neurosci. 2019 Nov 20;39(47):9369-9382. doi: 10.1523/JNEUROSCI.0752-19.2019. Epub 2019 Oct 9

    A molecular mechanism underlying gustatory memory trace for an association in the insular cortex.

    Adaikkan C, Rosenblum K.

    Elife. 2015 Oct 9;4:e07582. doi: 10.7554/eLife.07582.

    ERK-dependent PSD-95 induction in the gustatory cortex is necessary for taste learning, but not retrieval.

    Elkobi A, Ehrlich I, Belelovsky K, Barki-Harrington L, Rosenblum K.

    Nat Neurosci. 2008 Oct;11(10):1149-51. doi: 10.1038/nn.2190. Epub 2008 Sep 7.

    Taste familiarity is inversely correlated with Arc/Arg3.1 hemispheric lateralization.

    Inberg S, Elkobi A, Edri E, Rosenblum K.

    J Neurosci. 2013 Jul 10;33(28):11734-43. doi: 10.1523/JNEUROSCI.0801-13.2013.

    The Insula and Taste Learning.

    Yiannakas A, Rosenblum K.

    Front Mol Neurosci. 2017 Nov 3;10:335. doi: 10.3389/fnmol.2017.00335. eCollection 2017. Review.






    © 2017 Sagol Department of Neurobiology, University of Haifa | Tel: +972-4-8280015 | Fax: +972-4-8240339 | Email:, Designed by: Shani Zylberman, Computing and Information Systems division