Molecular identity of the synaptic tag
Which molecule marks a recently active synapse so that it can capture the proteins needed for lasting change?
Summary
Why it is open
Long-term potentiation (LTP), a lasting increase in synaptic strength, has an early phase of less than three hours and a late phase that needs new proteins, which are made mainly in the cell body even though LTP is specific to the stimulated inputs [1]. To explain this, Uwe Frey and Richard Morris proposed in 1997 that LTP creates a short-lasting "synaptic tag" at the potentiated synapse, set without protein synthesis, which sequesters the proteins needed for late LTP [1]. The hypothesis was later extended to learning: Roger Redondo and Morris write that new behavioural studies have translated it to the consolidation of newly formed memories [8]. How and where memories are stored more broadly is the problem of memory storage and retrieval.
Reviews list what a tag must do. It must be set by stimulation in an input-specific way, last less than about two hours unless extended, form without protein synthesis, stay relatively immobile and capture plasticity-related proteins (PRPs) [12]; Kelsey Martin and Kenneth Kosik add that it should be reversible [2]. In 2002 they wrote that kinases, adhesion molecules, the actin network, ion channels and local protein synthesis or degradation had all been proposed, and that "it is probable that neurons do not use a single synaptic tag to mark synapses, but a wide variety of them" [2].
No body sets a bar for a resolution. Yuan Lu and colleagues note that studies of candidate tags had relied on drug inhibitors in two-pathway experiments, and that a tag must also be shown to be "transiently and locally activated in a protein synthesis-independent manner by weak stimulation" [9]. Some molecules may be needed both for setting tags and for supplying PRPs [7]. Redondo and colleagues called for studies to "pin down the molecular identity of the tag set in motion by CaMKII activation" [7].
What is known
The original experiments (1997–2002). In hippocampal slices, Frey and Morris showed that weak stimulation, which ordinarily gives only early LTP, or strong stimulation in the presence of protein-synthesis inhibitors, produced protein-synthesis-dependent late LTP if strong stimulation had already been applied to another input to the same neurons; the tag decayed in less than three hours [1]. Martin and Kosik describe good evidence for a tag from Aplysia cell cultures and rat hippocampal slices [2].
Kinases (2005–2015). In rat hippocampal slices, Sreedharan Sajikumar, Julietta Frey and colleagues blocked PKMζ, a persistently active form of protein kinase C, with a zeta-pseudosubstrate inhibitory peptide (ZIP) and found that its activity maintained potentiation at both strongly and weakly stimulated (tagged) pathways; they called it the first LTP-specific PRP [3]. In 2007 they identified CaMKII (calcium/calmodulin-dependent protein kinase II) as an LTP-specific tag and ERK1/2 as an LTD-specific tag (for long-term depression) in apical dendrites of CA1 neurons, with PKA or PKMζ serving LTP tags in basal dendrites [4]. Redondo, Morris and colleagues, also in rat slices, found that a low dose of the CaMKII inhibitor KN-93 blocked tag setting while sparing PRP supply, whereas an inhibitor of a different kinase pathway (CaMKK) limited PRPs without affecting tags [7]. In rats trained weakly on an inhibitory avoidance task, a novel experience made the memory last; Diego Moncada, Fabricio Ballarini and Haydée Viola report that inhibitors of CaMKII, PKA and PKMζ around training impaired this, while an inhibitor of the ERK pathway and the protein-synthesis inhibitor anisomycin did not [12].
Actin and structure (2009–2011). Kenichi Okamoto, Miquel Bosch and Yasunori Hayashi note that CaMKII also binds actin filaments directly, and propose that the state of filamentous actin, regulated by CaMKII, sets a spine's capacity to bind postsynaptic proteins and so "may act as a synaptic tag" [6]. Binu Ramachandran and Frey found in rat slices that two structurally different inhibitors of actin polymerisation prevented late LTP and interfered with tagging, and proposed actin as a tag-specific molecule in apical CA1 dendrites [5]. Redondo and Morris revised the original hypothesis in 2011, writing that setting a tagged state and expressing LTP are dissociable, and that functional and structural plasticity differ in mechanism [8].
A receptor (2011). In knock-in mice whose TrkB, the receptor for brain-derived neurotrophic factor (BDNF), could be blocked by a specific drug, Lu and colleagues found that weak stimulation activated TrkB transiently, independently of protein synthesis and only at stimulated synapses; blocking it around the weak stimulation reduced late LTP on that pathway only [9]. Blocking TrkB during weak training also prevented a novel environment from turning short-term into long-term memory [9].
The PKMζ debate (2013–2026). In 2013 two groups, Anna Lee and colleagues and Lenora Volk, Richard Huganir and colleagues, reported that mice lacking PKMζ learn and remember normally, one also finding normal LTP, and that ZIP, the inhibitor behind most earlier evidence, still reverses LTP or erases a reward memory without PKMζ [10, 11]. Panayiotis Tsokas, Todd Sacktor and colleagues reported in 2016 that a related kinase, PKCι/λ, rises persistently and compensates in mice lacking PKMζ [13]. In July 2026 the same group reported that mice lacking both kinases lose late LTP and spatial long-term memory, but not short-term memory [15]; eLife's assessment called the study "important" and the results "convincing" [15].
Recent claims. In June 2024 Tsokas and colleagues reported in Science Advances that KIBRA, a postsynaptic scaffolding protein, binds PKMζ and anchors it at activated synapses, and that two drugs blocking this binding disrupt established late LTP and spatial memory in mice [14]. They propose that KIBRA acts as a synaptic tag for PKMζ, and speculate that its continual accumulation serves as "a persistent synaptic tag" during maintenance, in a process "distinct from the initial setting of the tag" [14]. No published response to the tag proposal was found as of 16 October 2026.
Main approaches
Kinases (CaMKII, PKA). Activation of a kinase at the synapse sets the tag, as CaMKII does in apical CA1 dendrites [4, 7]. Strongest evidence: CaMKII inhibition blocks tag setting but spares PRPs [7], and kinase inhibitors block behavioural tagging in rats [12]. Main difficulty: Redondo and colleagues cite evidence that phosphorylated CaMKII may spread beyond the stimulated synapses, an "apparent contradiction" they leave unresolved [7]; the kinase evidence came from drug inhibitors [9].
The actin cytoskeleton. The tag is a structural state of the spine set by actin remodelling [6]. Strongest evidence: actin polymerisation inhibitors prevent capture at weak pathways [5]. Main difficulty: the same inhibitors also impair early LTP, which the authors attribute to actin's role in early "housekeeping" processes [5].
Membrane receptors (TrkB). A relatively immobile membrane protein is more likely to be activated locally [9]. Strongest evidence: TrkB fits four of five proposed criteria for a tag, the authors write, and is needed for behavioural tagging [9]. Main difficulty: BDNF, its ligand, is itself implicated as a PRP, and the authors note that it "may play a more complex role in the tagging process" [9].
A persistent tag for PKMζ. KIBRA anchors PKMζ at potentiated synapses, sustaining late LTP and memory [14]. Strongest evidence: KIBRA–PKMζ complexes maintained month-old memory despite PKMζ turnover [14], and knocking out both PKMζ and its compensating relative abolishes late LTP and long-term memory [13, 15]. Main difficulty: this tag concerns maintenance, as distinct from the initial setting of a tag [14]; and from their knockout Volk and colleagues concluded that the effects of ZIP are independent of PKMζ [11].
Open sub-questions
- CaMKII at single spines. In a two-pathway experiment at single dendritic spines, does phosphorylated CaMKII stay confined to the weakly stimulated spine throughout the window in which it can capture proteins, or spread to neighbours [1, 7]?
- Actin after early LTP. Can actin dynamics be blocked after early LTP is established at a weak pathway so that capture fails while early LTP is unchanged [5, 6]?
- TrkB in the living animal. In the knock-in mice, does TrkB activation at a weakly stimulated input persist for as long as that input can still capture proteins, and stay confined to it in vivo [9, 12]?
- KIBRA at tag setting. Does blocking KIBRA–PKMζ binding only around weak stimulation prevent capture at that pathway in a two-pathway experiment, linking the persistent tag to the initial one [14]?
- Capture without atypical PKC. In mice lacking both PKMζ and PKCι/λ, can early LTP at a weak pathway still be converted to late LTP by strong stimulation elsewhere [1, 15]?
Work on this in Papex
If you have a testable answer to one of the sub-questions above, you can draft it as a private hypothesis in Papex, or prepare the draft from Claude or ChatGPT with the Papex connector. You review the exact version, sign it and decide whether to publish it; a public record that names this problem is listed on this page. A Papex record shows who signed which version and when. It is not peer review and does not mean the hypothesis is correct.
Key references
- Frey, U. & Morris, R. G. M. (1997). Synaptic tagging and long-term potentiation. Nature 385(6616), 533–536. 10.1038/385533a0 · Checked 16 Oct 2026
- Martin, K. C. & Kosik, K. S. (2002). Synaptic tagging — who's it? Nature Reviews Neuroscience 3(10), 813–820. 10.1038/nrn942 · Checked 16 Oct 2026
- Sajikumar, S., Navakkode, S., Sacktor, T. C. & Frey, J. U. (2005). Synaptic tagging and cross-tagging: the role of protein kinase Mζ in maintaining long-term potentiation but not long-term depression. Journal of Neuroscience 25(24), 5750–5756. 10.1523/JNEUROSCI.1104-05.2005 · Checked 16 Oct 2026
- Sajikumar, S., Navakkode, S. & Frey, J. U. (2007). Identification of compartment- and process-specific molecules required for "synaptic tagging" during long-term potentiation and long-term depression in hippocampal CA1. Journal of Neuroscience 27(19), 5068–5080. 10.1523/JNEUROSCI.4940-06.2007 · Checked 16 Oct 2026
- Ramachandran, B. & Frey, J. U. (2009). Interfering with the actin network and its effect on long-term potentiation and synaptic tagging in hippocampal CA1 neurons in slices in vitro. Journal of Neuroscience 29(39), 12167–12173. 10.1523/JNEUROSCI.2045-09.2009 · Checked 16 Oct 2026
- Okamoto, K., Bosch, M. & Hayashi, Y. (2009). The roles of CaMKII and F-actin in the structural plasticity of dendritic spines: a potential molecular identity of a synaptic tag? Physiology 24(6), 357–366. 10.1152/physiol.00029.2009 · Checked 16 Oct 2026
- Redondo, R. L., Okuno, H., Spooner, P. A., Frenguelli, B. G., Bito, H. & Morris, R. G. M. (2010). Synaptic tagging and capture: differential role of distinct calcium/calmodulin kinases in protein synthesis-dependent long-term potentiation. Journal of Neuroscience 30(14), 4981–4989. 10.1523/JNEUROSCI.3140-09.2010 · Checked 16 Oct 2026
- Redondo, R. L. & Morris, R. G. M. (2011). Making memories last: the synaptic tagging and capture hypothesis. Nature Reviews Neuroscience 12(1), 17–30 (published online 2010). 10.1038/nrn2963 · Checked 16 Oct 2026
- Lu, Y., Ji, Y., Ganesan, S., Schloesser, R., Martinowich, K., Sun, M. et al. (2011). TrkB as a potential synaptic and behavioral tag. Journal of Neuroscience 31(33), 11762–11771. 10.1523/JNEUROSCI.2707-11.2011 · Checked 16 Oct 2026
- Lee, A. M., Kanter, B. R., Wang, D., Lim, J. P., Zou, M. E., Qiu, C. et al. (2013). Prkcz null mice show normal learning and memory. Nature 493(7432), 416–419. 10.1038/nature11803 · Checked 16 Oct 2026
- Volk, L. J., Bachman, J. L., Johnson, R., Yu, Y. & Huganir, R. L. (2013). PKM-ζ is not required for hippocampal synaptic plasticity, learning and memory. Nature 493(7432), 420–423. 10.1038/nature11802 · Checked 16 Oct 2026
- Moncada, D., Ballarini, F. & Viola, H. (2015). Behavioral tagging: a translation of the synaptic tagging and capture hypothesis. Neural Plasticity 2015, 650780. 10.1155/2015/650780 · Checked 16 Oct 2026
- Tsokas, P., Hsieh, C., Yao, Y., Lesburguères, E., Wallace, E. J. C., Tcherepanov, A. et al. (2016). Compensation for PKMζ in long-term potentiation and spatial long-term memory in mutant mice. eLife 5, e14846. 10.7554/eLife.14846 · Checked 16 Oct 2026
- Tsokas, P., Hsieh, C., Flores-Obando, R. E., Bernabo, M., Tcherepanov, A., Hernández, A. I. et al. (2024). KIBRA anchoring the action of PKMζ maintains the persistence of memory. Science Advances 10(26), eadl0030. 10.1126/sciadv.adl0030 · Checked 16 Oct 2026
- Tsokas, P., Hsieh, C., Grau-Perales, A., Tcherepanov, A., Kwok, L., Rodriguez-Valencia, L. et al. (2026). PKMζ-PKCι/λ double-knockout demonstrates atypical PKC is crucial for the persistence of hippocampal LTP and spatial memory. eLife 15, RP110499. 10.7554/eLife.110499 · Checked 16 Oct 2026
Source
Open, as listed by Wikipedia · Unsolved problems in neuroscience. Checked 08 Oct 2026.
The status is the source’s. Papex does not decide whether a problem is solved.
Records on this problem
No public record addresses this problem yet.