Research Article: New Research | Neuronal Excitability Rapid ATF4 depletion resets synaptic responsiveness after cLTP https://doi.org/10.1523/ENEURO.0239-20.2021 Cite as: eNeuro 2021; 10.1523/ENEURO.0239-20.2021 Received: 8 June 2020 Revised: 14 April 2021 Accepted: 24 April 2021 This Early Release article has been peer-reviewed and accepted, but has not been through the composition and copyediting processes. The final version may differ slightly in style or formatting and will contain links to any extended data. Alerts: Sign up at www.eneuro.org/alerts to receive customized email alerts when the fully formatted version of this article is published. Copyright © 2021 Amar et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International license, which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed. 1 Title: Rapid ATF4 depletion resets synaptic responsiveness after cLTP 2 3 Abbreviated title: ATF4 resets synaptic responsiveness after cLTP 4 5 Fatou Amar1,2, Carlo Corona1,2,3, Johanna Husson2, Jin Liu1,2, Michael Shelanski1,2 and Lloyd 6 Greene1,2 7 1-Department of Pathology and Cell Biology, Columbia University Medical Center, Vagelos 8 College of Physicians and Surgeons, Columbia University, New York, NY 10032 9 2-The Taub Institute for Research on Alzheimer’s Disease and the Aging Brain, Columbia 10 University, New York, NY 10032 11 3-Current address, Burke Neurological Institute, White Plains, NY 10605 12 13 Corresponding authors: Lloyd A. Greene ([email protected]), Michael L. Shelanski 14 ([email protected]) 15 16 Number of pages: 37 17 18 Number of figures: 9 19 20 Number of words (Abstract): 228 21 22 Number of words (Significance) 95 23 24 Number of words (Introduction): 608 25 26 Number of words (Discussion): 1,369 27 28 Conflict of interest statement: The authors declare no competing financial interests. 29 30 Acknowledgments: This work was funded in part by a grant from the Zuckerman Foundation 31 and by the Henry and Marilyn Taub Foundation (MLS, CA, FA) and by NIH grant 32 5R01NS072050 (LAG). We thank Mr. Kaitao Zhao for excellent technical assistance. 33 34 35 36 37 38 39 40 41 42 43 44 45 46 1 47 48 49 50 51 Abstract: Activating transcription factor 4 (ATF4/CREB2), in addition to its well-studied role in 52 stress responses, is proposed to play important physiologic functions in regulating learning and 53 memory. However, the nature of these functions has not been well defined and is subject to 54 apparently disparate views. Here, we provide evidence that ATF4 is a regulator of excitability 55 during synaptic plasticity. We evaluated ATF4's role in mature hippocampal cultures subjected 56 to a brief chemical-LTP (cLTP) induction protocol that results in changes in mEPSC properties 57 and synaptic AMPA receptor density one hour later, with return to baseline by 24 hours. We find 58 that ATF4 protein, but not its mRNA, is rapidly depleted by about 50% in response to cLTP 59 induction via NMDA receptor activation. Depletion is detectable in dendrites within 15 min and 60 in cell bodies by 1 hour and returns to baseline by 8 hours. Such changes correlate with a 61 parallel depletion of phospho-eIF2a, suggesting that ATF4 loss is driven by decreased 62 translation. To probe the physiologic role of cLTP-induced ATF4 depletion, we constitutively 63 overexpressed the protein. Reversing ATF4 depletion by over-expression blocked the recovery 64 of synaptic activity and AMPA receptor density to baseline values that would otherwise occur 24 65 hours after cLTP induction. This reversal was not reproduced by a transcriptionally inactive 66 ATF4 mutant. These findings support ATF4's role as a required element in resetting baseline 67 synaptic responsiveness after cLTP. 68 69 Significance: The mechanism(s) by which synaptic responsiveness is reset after LTP are not 70 well understood. Resetting avoids LTP "saturation" and uncontrolled feed-forward potentiation 71 and may play a part in synaptic "scaling”. In the work reported here we have found that the 2 72 transcription factor ATF4 is translationally down-regulated following cLTP induction and acts as 73 a regulator of long-term synaptic plasticity, resetting the synapse to the de-potentiated state. Our 74 findings may serve to begin to reconcile the conflicting views regarding the role of ATF4 in 75 synaptic plasticity and further illuminate the role of ATF4 in learning and memory. 76 77 Introduction 78 79 Although often considered in pathological contexts in the brain as a responder to stress (Pitale et 80 al., 2014), much evidence has pointed to roles for the transcription factor ATF4 (also called 81 CREB2) in physiologic neuronal function (Bartsch et al., 1995; Costa-Mattioli et al., 2007; Hu et 82 al., 2015; Liu et al., 2014; Pasini et al., 2015). ATF4's features make it especially suitable for a 83 potential role in synaptic plasticity. It is detectably expressed in neurons and in particular in 84 neuronal processes, and undergoes retrograde transport to cell bodies and nuclei where it 85 regulates a variety of genes relevant to neuronal function (Lai et al, 2004; Sun et al., 2013; 86 Baleriola et al., 2014; Pasini et al., 2016; Liu et al., 2018). ATF4 is also subject to rapidly 87 regulated expression (Ameri and Harris, 2008). The latter property arises from ATF4's fast 88 turnover and regulated translation via eukaryotic translation initiation factor 2a (eIF2a). When 89 eIF2a is in a non-phosphorylated state, it promotes global translation, but suppresses translation 90 of a subset of mRNAs including the mRNA encoding ATF4 (Harding et al, 2000). Conversely, 91 when eIF2a is phosphorylated by a defined set of kinases, global translation is reduced and 92 translation of the subset of messages including that encoding ATF4 are selectively enhanced 93 (Harding et al, 2000; Ameri and Harris, 2008). 94 3 95 While there is general agreement that ATF4 appears to play an important role in synaptic 96 plasticity as well as in learning and memory, there are divergent views about whether such 97 activities are positive or negative (Bartsch et al., 1995; Chen et al., 2003; Costa-Mattioli et al., 98 2005, 2007; Hu et al., 2015; ILL-Raga et al., 2013; Trinh et al., 2012; Liu et al., 2014; Pasini et 99 al., 2015). Such disparate findings may, in part, reflect that much of the evidence is based on 100 indirect control of ATF4 expression/activity by manipulating eIF2a phosphorylation levels or by 101 blockade with dominant-negative inhibitors that might also inhibit other members of the b-ZIP 102 transcription factor family to which ATF4 belongs. In past work, to more directly gauge ATF4's 103 physiologic role in brain, we have directly manipulated its expression in hippocampal neurons in 104 culture and in vivo and reported that chronic ATF4 down regulation or depletion decreases 105 mushroom spine density, reduces excitatory synapses, produces deficits in long-term spatial 106 memory and behavioral flexibility, impairs both long-term potentiation (LTP) and long-term 107 depression (LTD) as well as glutamatergic function, and diminishes GABABRs trafficking (Liu 108 et al., 2014; Pasini et al., 2015; Corona et al., 2018). Strikingly, while ATF4 over-expression 109 reverses such parameters, it does not elevate them above baseline. 110 111 Here, to further understand ATF4's role in neuronal plasticity, we have examined both its 112 regulation and function in a form of long-term potentiation (LTP), a cellular mechanism 113 associated with learning and memory (Bliss and Collingridge, 1993; Chen and Tonegawa, 1997; 114 Dudai, 2004; Malenka and Nicoll, 1999; Pittenger and Kandel, 2003). Although LTP must last 115 for several hours to promote long-term changes in memory, it also must be capable of "resetting" 116 to unstimulated levels for a variety of reasons including avoidance of saturation and promotion 117 of additional rounds of plasticity (Moser et al., 1998), providing synaptic scaling (Turrigiano, 4 118 2008) and forestalling excessive excitatory activity that could lead to seizures (Bliss and Cooke, 119 2011) or even neuronal damage (McEachern and Shaw, 1999). In this context, we report that 120 chemically-induced LTP (cLTP) evoked by brief exposure to glutamate leads to a rapid decrease 121 of ATF4 protein levels in hippocampal neurons and that reversal of such depletion by ATF4 122 overexpression blocks long-term resetting of the LTP mechanism back to baseline. These 123 findings suggest that ATF4 acts as a feedback regulator of synaptic plasticity associated with 124 LTP. 125 126 Material and methods 127 DNA constructs 128 Lentiviral constructs were generated as previously described (Liu et al., 2014). 129 Lentivirus preparation 130 For in vitro experiments, the second-generation packaging system (which generates replication- 131 deficient lentivirus) was used for all experiments to prepare lentivirus (Zufferey et al., 1997). 132 Packaging vectors psPAX2 and pMD2.G were obtained from Addgene. In summary, lentiviral 133 constructs for overexpression were co-transfected with the packaging vectors into HEK293T 134 cells with calcium phosphate. Supernatants from HEK293T cells infected with virus were 135 collected 24 and 48 hours after transfection. After centrifugation at 1000 rpm for 10 min, the 136 supernatants were passed through a 0.45 μm PDVF filter unit (Nalgene) then concentrated 20– 137 30x by centrifugation in an Amicon Ultra centrifugal filter (100 K) (Millipore) following the 138 manufacturer's instructions. Viruses were aliquoted and stored at −80°C. Viral titers ranged from 139 1–5 × 106 infectious units/μl. 5 140 Cell culture and infection 141 Primary hippocampal cultures were prepared as previously described (Corona et al., 2018; Liu et 142 al., 2018).