Synaptic memory and CaMKII

In the paper

“the potentiation typically is transient, lasting no longer than 20 min (Refs. 189–193, but see Refs. 194, 195).”

What the sources say

• ?Not checked — no full text of the sourcecites [189] source
• ?Not checked — no full text of the sourcecites [190] source
• ?Not checked — no full text of the sourcecites [191] source
• ?Not checked — no full text of the sourcecites [192] source
• Partly supported

  The Results (Fig. 4A,B; page not given in our copy) report that potentiation induced by depolarizing pulses often decayed back to control, supporting its transient nature. The reported decay time is 10–30 minutes, so the source does not support the claim that it lasts no longer than 20 minutes. (LLM, conf=0.88)

  cites [193] sourceExperimental — accuracy not yet measured
• Disputed

  In 'Induction of non-Hebbian LTP by postsynaptic depolarization' (page unavailable), the paper contrasts an earlier transient result with its own stable potentiation, observed for nearly an hour without a tendency to decay. The cited paper therefore provides an exception to the stated 20-minute limit, consistent with the citing sentence’s 'but see' qualification, rather than evidence for that limit. (LLM, conf=0.88)

  Checked by a reasoning model — may be wrong.

  cites [194] sourceExperimental — accuracy not yet measured
• ?Could not verify

  The Methods section describes how voltage-pulse LTP was induced, and “Postsynaptic Syt1,7 loss blocks LTP” discusses its mechanism. Neither passage states that the potentiation is typically transient or lasts no longer than 20 minutes; searches for voltage-pulse LTP duration and decay found no such statement in our copy. (LLM, conf=0.73)

  cites [195] sourceExperimental — accuracy not yet measured

In the paper

“The anatomical substrate for silent synapses is fingerlike dendritic protrusions referred to as filipodia (245), which account for roughly 30% of synapses in the mature brain (240).”

What the sources say

• Disputed

  The Results (page unavailable) report that filopodial synapses were 21% of identified contacts in developing CA1 during the first postnatal week and fell to 7.2% by postnatal day 12—not roughly 30% in the mature brain. The paper also states that dendritic filopodia disappear as the cortex matures; it does not report a 30% estimate for mature-brain synapses. (LLM, conf=0.82)

  Checked by a reasoning model — may be wrong.

  cites [245] sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [240] source

In the paper

“The existence of LTD avoids the problem of saturation, which could be essential for a mnemonic device (366–371).”

What the sources say

• ?Not checked — no full text of the sourcecites [366] source
• Supported

  In “II. Modification of Cortical Synapses” (page unavailable), the paper says that if synapses only increase, they eventually saturate and no information is stored; it therefore calls for conditions under which synapses decrease. Its proposed modification rule includes such decreases, supporting the claim’s point about avoiding saturation in a mnemonic device. (LLM, conf=0.91)

  cites [367] source
• Disputed

  In “The Stability Problem” (page unavailable), the paper groups LTP and LTD as learning-related plasticity mechanisms, then says such mechanisms can create destabilizing feedback that drives neuronal activity to saturation. It says nervous systems need other plasticity mechanisms to counteract those forces; it does not say that the existence of LTD avoids saturation. (LLM, conf=0.86)

  Checked by a reasoning model — may be wrong.

  cites [368] sourceExperimental — accuracy not yet measured
• Partly supported

  The paper’s discussion of memory lifetimes says that balancing synaptic potentiation and depression can improve memory performance. But its discussion of saturation describes effects distinct from the mechanism it proposes; our copy does not establish that LTD itself avoids saturation or that this is essential for a mnemonic device. (LLM, conf=0.75)

  cites [369] sourceExperimental — accuracy not yet measured
• ?Could not verify

  The Results section on single-synapse models describes LTD and says that both LTP and LTD should saturate, but does not say that the existence of LTD prevents synaptic saturation. The Discussion’s “Functional significance” section links maintained dynamic range to learning and memory, attributing it to the interaction of Hebbian and homeostatic plasticity rather than to LTD alone. I searched for the proposed LTD–saturation–memory link; no passage in our copy establishes it. (LLM, conf=0.72)

  cites [370] sourceExperimental — accuracy not yet measured
• ?Could not verify

  Section 4.4.3 discusses saturation of LTP/LTD after repeated spike pairings, and Section 4.4.1 discusses the need for synaptic stability to retain memories. Neither passage says that LTD prevents saturation in a memory device. Searches for LTD preventing saturation and for depression preventing synaptic-weight saturation found no passage making that connection. (LLM, conf=0.70)

  cites [371] sourceExperimental — accuracy not yet measured

In the paper

“These plateau potentials on their own do not change synaptic strength.”

What the sources say

• Disputed

  The Results section, “Plateau increase the amplitude of incoming input” (page unavailable), says plateau-induced place fields are consistent with increased EPSP amplitude—evidence of strengthened synaptic input. The Discussion (page unavailable) also describes plasticity produced by plateau potentials. The paper therefore makes the contrary point, although it does not isolate a plateau from all concurrent synaptic input. (LLM, conf=0.88)

  Checked by a reasoning model — may be wrong.

  cites [424] sourceExperimental — accuracy not yet measured
• Disputed

  The Role of BTSP section (page not provided) says Ca2+ plateau potentials drive BTSP and can induce plasticity in a single trial, contrary to the claim that they do not change synaptic strength. The paper does not separately test a plateau in the absence of synaptic input, so its statement does not establish that plateaus are sufficient without input. (LLM, conf=0.79)

  Checked by a reasoning model — may be wrong.

  cites [426] sourceExperimental — accuracy not yet measured

In the paper

“It is also agreed that the synaptic capture of AMPARs requires the PDZ domain interaction between TARPs and PSD-95 (275, 279, 294, 440).”

What the sources say

• Disputed

  The Introduction describes TARP–PSD-95 PDZ binding as a major mechanism of AMPAR localization, but the Discussion explicitly rejects it as a universal requirement. It reports that the GluA2 N-terminal domain can support synaptic anchoring even without γ8 PDZ ligands. (LLM, conf=0.89)

  Checked by a reasoning model — may be wrong.

  cites [275] sourceExperimental — accuracy not yet measured
• Supported

  In “The role of TARPs in AMPAR surface trafficking and synaptic targeting” (page not provided), the review says TARP PDZ-binding motifs bind PSD-95 and that TARPs are essential intermediaries in anchoring AMPARs at synapses. It also reports that disrupting stargazin–PSD-95 interactions prevents AMPAR complexes from clustering at postsynaptic densities. (LLM, conf=0.93)

  cites [279] source
• Partly supported

  The Discussion reports that a functional TARP PBM–PSD-95 PDZ interaction is necessary for TARP-assisted AMPAR accumulation at synapses. Its Limitations section narrows that conclusion to the TARP γ-8-based trafficking studied and notes that some synaptic transmission may be TARP independent. The unqualified claim that synaptic capture of AMPARs requires this interaction therefore goes beyond the paper's findings. (LLM, conf=0.85)

  cites [294] sourceExperimental — accuracy not yet measured
• Partly supported

  The section on TARP/MAGUK interactions says TARP–PSD-95 interactions stabilize AMPARs at the postsynaptic density and involve PSD-95 PDZ domains. It calls these interactions the most plausible AMPAR capture sites, but does not establish that this specific interaction is universally required; the reported requirement for the TARP binding motif comes from particular experimental conditions. (LLM, conf=0.84)

  cites [440] sourceExperimental — accuracy not yet measured

In the paper

“The more enduring storage of information, however, remains intact after brain silencing.”

• Also stated in: § 1. INTRODUCTION · plain

  “Thus, in cases where neuronal activity has been transiently silenced, as in barbiturate overdose (5) or brain cooling (6), memories remain intact.”

What the sources say

• ?Not checked — no full text of the sourcecites [5] source
• ?Could not verify

  The Results describe loss and restoration of EEG activity during circulatory arrest, but do not assess whether stored information survives that inactivity. The Comment raises a possible risk of neurocognitive deficits and calls for further study; it does not establish what happens to enduring memory storage. Searches for long-term memory storage and cognitive function after cooling found no passage resolving the claim. (LLM, conf=0.74)

  cites [6] sourceExperimental — accuracy not yet measured

In the paper

“Additionally, NMDAR-dependent LTP has been most tightly associated with intact animal behavioral memory (19, 43–46).”

What the sources say

• ?Not checked — no full text of the sourcecites [19] source
• ?Not checked — no full text of the sourcecites [43] source
• Partly supported

  The section on behavioral tests of memory reports that mice with impaired CaMKII/NMDAR complex formation had both reduced LTP and reduced spatial memory, supporting a link in intact animals. The paper does not compare NMDAR-dependent LTP with other forms of plasticity or establish that its association with behavioral memory is the tightest; the Conclusions say further behavioral work is needed. (LLM, conf=0.82)

  cites [44] sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [45] source
• Partly supported

  The Introduction (page unavailable) describes NMDA-initiated LTP in hippocampal CA1 as a candidate memory mechanism. The Discussion (page unavailable) says that interference with CaMKII can erase both LTP and memory, supporting a link between them. But the paper does not establish the comparative claim that NMDAR-dependent LTP is *most tightly* associated with behavioral memory in intact animals; it cautions that LTP is difficult to observe during learning. (LLM, conf=0.86)

  cites [46] sourceExperimental — accuracy not yet measured

In the paper

“However, it is now known that LTP can be induced by just depolarizing the postsynaptic cell with current through the recording electrode combined with low-frequency synaptic stimulation (40, 41).”

What the sources say

• Supported

  The paper’s opening summary describes low-frequency single-volley stimulation paired with depolarizing current pulses injected through the recording microelectrode. The Results report that pairing at 0.1 Hz increased the EPSP, and the Discussion identifies this lasting increase as LTP induced without high-frequency presynaptic activity. (LLM, conf=0.94)

  cites [40] source
• ?Not checked — no full text of the sourcecites [41] source

In the paper

“There is compelling evidence that for short forms of memory, often referred to as working memory, information is stored by ongoing activity in neuronal networks (1–4).”

What the sources say

• ?Not checked — no full text of the sourcecites [1] source
• Partly supported

  In “Working memory at the systems level,” the paper describes persistent neural activity during working-memory delays as evidence for maintaining representations. But the same section says persistent activity may not be necessary and discusses storage through short-term changes in synaptic weights. The claim overstates the paper’s position by presenting ongoing activity as the storage mechanism without that qualification. (LLM, conf=0.87)

  cites [2] sourceExperimental — accuracy not yet measured
• Supported

  The Introduction (p. 3) describes persistent neuronal activity during short-term memory tasks as a neural correlate of memory maintenance. It also discusses network mechanisms that sustain this activity, supporting the citing sentence’s broad claim. (LLM, conf=0.90)

  cites [3] source
• ?Not checked — no full text of the sourcecites [4] source

In the paper

“Importantly, this same depolarization in the absence of synaptic stimulation fails to induce LTP.”

What the sources say

• Supported

  The section on the duration of the pairing-induced EPSP increase reports that a second, unpaired input was unaffected by the current pulse. It also reports no observable EPSP increase when the pulse was delayed 400 ms after synaptic activation, supporting the claim that depolarization without coincident synaptic stimulation did not induce LTP in these experiments. (LLM, conf=0.91)

  cites [40] source
• ?Not checked — no full text of the sourcecites [41] source

In the paper

“it was not until the discovery of a Ca2+/calmodulin-dependent protein kinase, now referred to as CaMKII (9, 10), which became independent of Ca2+ (autonomous) after autophosphorylation (11–15)”

What the sources say

• Partly supported

  In “Possible Activation of the Kinase” (p. 840), the authors report that the treated Aplysia kinase becomes independent of added Ca2+/calmodulin. They describe autophosphorylation as “one attractive idea” for how this happens, not an established cause, so the claim overstates that link. (LLM, conf=0.88)

  cites [11] sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [12] source
• ?Not checked — no full text of the sourcecites [13] source
• ?Not checked — no full text of the sourcecites [14] source
• ?Not checked — no full text of the sourcecites [15] source

In the paper

“This representation is strikingly similar to actual CaMKII responses in synaptic spines stimulated by repetitive glutamate uncaging and, as in FIGURE 3, shows little integrated activation with T286A CaMKII, unless substantially higher frequencies are used (94).”

What the sources say

• Supported

  In the Results (Figures 2B–2C), T286A CaMKII activity fails to accumulate during repetitive glutamate uncaging at 0.49 Hz because it decays rapidly between pulses. The Results (Figures 3A–3F) then report that higher-frequency uncaging at 1.9 or 7.8 Hz raises its activity to a significantly higher plateau in stimulated spines. (LLM, conf=0.94)

  cites [94] source

In the paper

“The phosphorylation of T286 requires that the subunit being phosphorylated have Ca2+/CaM bound (82, 83), resulting in autophosphorylation.”

What the sources say

• ?Not checked — no full text of the sourcecites [82] source
• Partly supported

  The Results (Fig. 4) show that substrate subunits unable to bind Ca2+/calmodulin were not phosphorylated, supporting the requirement for the subunits tested there. But the Results (Fig. 5) also show that a C289P substrate subunit was autophosphorylated without Ca2+/calmodulin, so the claim is too absolute. Page numbers are unavailable in our copy. (LLM, conf=0.88)

  cites [83] sourceExperimental — accuracy not yet measured

In the paper

“Fourth, CaMKII has the intriguing property that once activated by Ca2+ the enzymatic activity toward substrates remains after Ca2+ removal (11–15).”

What the sources say

• Partly supported

  The Fig. 4 and Table III passages report that an Aplysia kinase treated with Ca2+/calmodulin and ATP could subsequently phosphorylate synapsin I under Ca2+-free conditions. The claim omits the ATP-dependent treatment and presents this as a general property of CaMKII activity toward substrates, whereas the paper demonstrates it for synapsin I under those conditions. (LLM, conf=0.85)

  cites [11] sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [12] source
• ?Not checked — no full text of the sourcecites [13] source
• ?Not checked — no full text of the sourcecites [14] source
• ?Not checked — no full text of the sourcecites [15] source

In the paper

“The structure of a CaMKIIN peptide to CaMKII shows it to bind across the surface groove, with docking sites designated as A, B, and C (3 oval indentations in FIGURE 2) (62).”

What the sources say

• Supported

  The Figure 5 caption describes CaMKIINtide bound to the CaMKII kinase-domain surface and identifies its interactions at docking sites A, B, and C. The section on the feedback inhibitor also reports the crystal structure of the bound peptide and its interactions at sites B and C. (LLM, conf=0.94)

  cites [62] source

In the paper

“Second, it is present in the brain at extraordinarily high levels, rivaling the levels of cytoskeletal proteins (63–67), suggesting that CaMKII may play a structural role (27, 68).”

What the sources say

• Partly supported

  The Discussion reports that CaMKII reaches 1–2% of total protein in some neurons and calls that level unusual. It compares this abundance with structural proteins, but gives substantially higher levels for tubulin (about 10%) and actin (about 7%); “rivaling” their levels overstates the comparison. (LLM, conf=0.86)

  cites [63] sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [64] source
• ?Not checked — no full text of the sourcecites [65] source
• ?Not checked — no full text of the sourcecites [66] source
• ?Not checked — no full text of the sourcecites [67] source

In the paper

“it is unique in having a COOH-terminal hub or association domain that assembles 12 subunits as a double-ringed, yoyolike holoenzyme (FIGURE 2, A AND B) (28).”

What the sources say

• Partly supported

  The sections on hub geometry and flexible linkers describe a distinctive ring-shaped hub built from vertical dimers, and report 12-subunit CaMKIIα holoenzymes. But the paper also reports 14-subunit forms and says the kinase domains are flexibly arranged around the hub, so it does not support an unqualified, fixed 12-subunit, double-ringed yo-yo shape for the whole holoenzyme. Page numbers were not available in our copy. (LLM, conf=0.86)

  cites [28] sourceExperimental — accuracy not yet measured

In the paper

“behavioral timescale synaptic plasticity (BTSP) (103–106), which is responsible for the rapid formation of hippocampal place fields.”

What the sources say

• Supported

  The Introduction says BTSP enables a new hippocampal CA1 place field to form in a single trial. The Discussion reports that BTSP produces synaptic changes large enough to rapidly reshape place-field activity, supporting the attributed role in rapid place-field formation. (LLM, conf=0.92)

  cites [103] source
• Supported

  The Abstract states that the authors discovered a synaptic potentiation that produces CA1 place fields in vivo in a single trial and named the rule behavioral time scale synaptic plasticity. The discussion likewise says BTSP can abruptly form new place fields in a single trial. (LLM, conf=0.94)

  cites [104] source
• Partly supported

  The Discussion says the results are consistent with a sizable fraction of de novo place fields forming via BTSP, particularly during novel experiences. The claim states this as an established causal role in rapid place-field formation, whereas the paper describes indirect correlates and acknowledges that fields can form through other mechanisms. (LLM, conf=0.88)

  cites [105] sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [106] source

In the paper

“Furthermore, although rarely cited, there are two rigorously controlled studies that failed to find any role for protein synthesis in late LTP (59, 60).”

What the sources say

• ?Not checked — no full text of the sourcecites [59] source
• Supported

  The Results section reports that late LTP induced by a single train persisted despite either of two protein-synthesis inhibitors. The Discussion says the authors verified severe inhibition of protein synthesis and found potentiation maintained for at least eight hours. This paper supports the citing sentence as one of the two studies, within its experimental conditions. (LLM, conf=0.88)

  cites [60] source

In the paper

“In essence, T286 autophosphorylation and CaM trapping would produce a frequency dependence for CaMKII activation.”

• Also stated in: § 3.1. CaMKII and Integration of Ca2+ Stimuli · boosted

  “This has, in fact, been demonstrated with purified CaMKII subjected to different frequencies and durations of Ca2+ spikes (69, 94).”

What the sources say

• ?Not checked — no full text of the sourcecites [69] source
• Partly supported

  The Results and Discussion (page unavailable) show that Thr286 autophosphorylation slows CaMKII activity decay and lowers the stimulation frequency needed to integrate Ca²⁺ signals. The paper discusses Ca²⁺/CaM binding and dissociation, but the passages do not establish CaM trapping as a cause of the frequency dependence. (LLM, conf=0.85)

  cites [94] sourceExperimental — accuracy not yet measured

In the paper

“As is the case with many kinases, CaMKII contains an autoinhibitory regulatory segment (FIGURE 2, A, B1, AND B2) that binds across the substrate binding groove, whose structure and function have been delineated by peptides, mutations, and crystallography (71–77).”

What the sources say

• ?Not checked — no full text of the sourcecites [71] source
• ?Not checked — no full text of the sourcecites [72] source
• Supported

  The Results identify a core pseudosubstrate segment in CaMKII’s autoinhibitory domain that anchors in the kinase’s peptide-binding site. The Discussion describes contacts positioning that segment at the peptide-binding site and a structural model in which the inhibitory domain fits into a channel on the protein surface. This supports the claim’s description of the segment binding across the substrate-binding region; the CaMKII structure presented here is a model. (LLM, conf=0.88)

  cites [73] source
• Supported

  The Introduction (page not available in our copy) describes CaMKII’s regulatory domain as autoinhibitory and says it partially blocks the substrate-binding site. The Discussion places a segment of that domain at or near the active site, consistent with the claimed binding across the substrate-binding groove; it also notes that docking is dynamic. (LLM, conf=0.87)

  cites [74] source
• Supported

  The Results section, “Structures of Autoinhibited Human CaMKII Isozymes,” states that the regulatory domain blocks the substrate-binding site in all human CaMKIIs. The Figure 2 caption likewise depicts the autoinhibited CaMKIIδ regulatory domain blocking that site, supporting the claim about the autoinhibitory segment. (LLM, conf=0.94)

  cites [75] source
• ?Not checked — no full text of the sourcecites [76] source
• Partly supported

  The Abstract identifies a CaMKII regulatory domain and reports that a peptide modeled on it inhibits kinase activity; the authors suggest it acts as an active site-directed inhibitory element. The available passage does not establish that the regulatory segment binds across the substrate-binding groove, so that structural detail is not supported by this source. (LLM, conf=0.82)

  cites [77] sourceExperimental — accuracy not yet measured

In the paper

“and reduces desensitization of GluN2B-containing receptors (129).”

What the sources say

• Partly supported

  The Discussion says Ser1303 phosphorylation is sufficient to reduce desensitization when intracellular chloride is low. The claim omits that condition: the Results report that a phospho-mimetic S1303D mutation instead increased apparent desensitization when chloride was the major intracellular anion. (LLM, conf=0.90)

  cites [129] sourceExperimental — accuracy not yet measured

In the paper

“This is despite the fact that CaM can activate CaMKII when only two of its four Ca2+ binding sites are occupied (112, 113).”

What the sources say

• Supported

  The Results section (“Activation of CaMKII by Mutant CaMs”) reports that calmodulin variants able to bind Ca²⁺ at only two sites activated CaMKII. The activation was partial and required higher calmodulin concentrations than activation by four-Ca²⁺-bound wild-type calmodulin, but the claim says only that activation can occur. (LLM, conf=0.94)

  cites [112] source
• Supported

  The Introduction states that CaM with two Ca2+ ions bound at its C-terminal sites binds CaMKII and activates autophosphorylation, though at a ten-fold lower catalytic rate than fully loaded CaM. The Discussion also describes CaM species with fewer than four bound Ca2+ ions as major CaMKII activators. (LLM, conf=0.94)

  cites [113] source

In the paper

“This binding of CaMKII to GluN2B is critical for the synaptic enhancement caused by CaMKII (see sect. 6) (119–122).”

What the sources say

• Supported

  In “Binding of CaMKII to NR2B Is Necessary for Synaptic Plasticity” (page not provided), the paper reports that reducing CaMKII association with NR2B blocked LTP and prevented spontaneous activity-driven potentiation. The Discussion concludes that binding to NR2B is necessary for LTP, supporting the claim that the binding is critical for synaptic enhancement. (LLM, conf=0.93)

  cites [119] source
• Supported

  The Results section on CA1 LTP reports that disrupting CaMKII binding to GluN2B reduced several forms of LTP by about half, and concludes that the binding is necessary for LTP to fully develop. This supports the claim that the interaction is critical for synaptic enhancement, though it is not required for all LTP. (LLM, conf=0.88)

  cites [120] source
• Supported

  In Results, under “The synaptic actions of CaMKII require binding to NMDARs” (page not provided), replacing GluN2B with a mutant that cannot bind CaMKIIα entirely prevented the synaptic enhancement produced by an active CaMKIIα mutant. This directly supports the claimed importance of CaMKII–GluN2B binding for that enhancement. (LLM, conf=0.93)

  cites [121] source
• ?Not checked — no full text of the sourcecites [122] source

In the paper

“Disrupting CaMKII binding prevents phosphorylation (124).”

What the sources say

• Partly supported

  The Results section reports that the GluN2BΔCaMKII mutation abolishes S1303 phosphorylation by CaMKII, and the Discussion says the mutation abolishes both CaMKII binding and S1303 phosphorylation. The paper does not establish that loss of binding itself prevents phosphorylation; it treats these as two effects of the mutation. (LLM, conf=0.84)

  cites [124] sourceExperimental — accuracy not yet measured

In the paper

“Phosphorylation of S1303 in heterologous cells enhances NMDAR current (127, 128)”

What the sources say

• ?Not checked — no full text of the sourcecites [127] source
• Supported

  In Results, “DAPK1 Activation Phosphorylates NR2B Subunit at Ser-1303” (page not provided), the authors report that activated DAPK1 increased NR1/NR2B receptor currents in HEK293 cells but not currents from receptors with the S1303A mutation. They conclude that phosphorylation at Ser-1303 enhances channel conductance. (LLM, conf=0.94)

  cites [128] source

In the paper

“overexpression of constitutively active CaMKII has little effect on synaptic NMDAR responses (130–132).”

What the sources say

• ?Not checked — no full text of the sourcecites [130] source
• Supported

  The Results report that CA1 neurons expressing constitutively active αCaMKII had no significant change in synaptic NMDAR EPSC amplitude compared with neighboring controls (0.83 ± 0.27 of controls; p = 0.54; n = 12). This supports the claim that overexpression has little effect on synaptic NMDAR responses. (LLM, conf=0.94)

  cites [131] source
• ?Not checked — no full text of the sourcecites [132] source

In the paper

“and is reversible (134).”

What the sources say

• Supported

  The Discussion explicitly states that CaMKII self-association is reversible, while noting that the time course depends on local Ca²⁺ levels or Thr286 phosphorylation. Figure 4 also reports that clusters disappeared within 2–5 minutes after Ca²⁺ was removed. (LLM, conf=0.95)

  cites [134] source

In the paper

“As proposed, the CTD peptide does bind in place of the regulatory domain, across the entire groove on the kinase, including the active site (61).”

What the sources say

• Supported

  In Results, the authors report that the GluN2B peptide docks at the active site and that peptides bind a shared surface of the CaMKII kinase domain. Their structural comparison also shows GluN2B engaging a pocket used by the regulatory segment, while the Discussion describes a continuous single binding site across the kinase surface. Together, these support the paraphrase; the copy provides no page numbers. (LLM, conf=0.86)

  cites [61] source

In the paper

“CaMKII in cellular clusters does contain T286P (134)”

What the sources say

• Supported

  In “Thr286 autophosphorylation of self-associated CaMKII,” clusters of wild-type GFP-CaMKII in stimulated HEK293 cells stained with a phospho-Thr286-specific antibody. This directly supports the claim that clustered cellular CaMKII can contain phosphorylated Thr286. (LLM, conf=0.93)

  cites [134] source

In the paper

“Shank3 resides in the pallium of the PSD, just below the core (139, 162).”

What the sources say

• Supported

  The Discussion states that Shank 3 is mostly located in the PSD pallium. The Results describe the pallium as the deeper layer of the PSD, consistent with its position below the core. The paper says “mostly,” rather than implying that every Shank3 molecule is confined there; the citing sentence is a reasonable paraphrase. (LLM, conf=0.92)

  cites [139] source
• Supported

  The ‘Shank Levels Increase in the Pallium’ section identifies Shank3 as a member of the Shank family. The ‘Pallial Layer of the PSD’ section places Shank labeling in the pallium, immediately below the PSD core. The paper describes the location for Shanks collectively rather than Shank3 separately. (LLM, conf=0.88)

  cites [162] source

In the paper

“Interestingly, CaMKIIα can also bind to a distinct domain of Shank3, which plays a role in gene transcription (163).”

What the sources say

• Supported

  The binding experiments identify a central Shank3 domain that directly binds activated CaMKIIα. The section on LTCC-dependent CREB phosphorylation and gene expression reports that disrupting this interaction impairs signaling to CREB and subsequent c-Fos expression in hippocampal neurons. The role in gene transcription is therefore through this signaling pathway, not direct transcription by Shank3. (LLM, conf=0.91)

  cites [163] source

In the paper

“Glutamate treatment of neuronal cultures causes CaMKII clustering at excitatory synapses (136–140).”

What the sources say

• Partly supported

  The Results and Discussion report that hippocampal cultures developed CaMKII clusters after treatment with CCCP in glucose-free medium, and that CaMKII labeling increased near postsynaptic densities. They do not report glutamate treatment, and the Discussion describes clustering around a synapse as a possibility rather than an observed result. (LLM, conf=0.86)

  cites [136] sourceExperimental — accuracy not yet measured
• Supported

  The Results report that glutamate/glycine treatment of hippocampal neuronal cultures increased CaMKII labeling at postsynaptic densities (passage 6). The Discussion describes this as reversible CaMKII association with PSDs and notes that earlier imaging found punctate CaMKII coincident with PSD-95 staining (passage 4). Here, “clustering at excitatory synapses” is supported as a paraphrase of that synaptic accumulation, rather than formation of the separate cytoplasmic clusters discussed in passage 5. (LLM, conf=0.88)

  cites [137] source
• ?Not checked — no full text of the sourcecites [138] source
• Partly supported

  The Results and Methods describe treating dissociated hippocampal neuronal cultures with NMDA or high K+, not glutamate. The Discussion reports that stimulation concentrated CaMKII at the postsynaptic density, while describing CaMKII self-clustering as a possibility rather than an established mechanism. Thus the synaptic accumulation is supported, but the specified glutamate treatment and definite clustering are not. (LLM, conf=0.86)

  cites [139] sourceExperimental — accuracy not yet measured
• Partly supported

  The Results section (‘Prolonged saline flush causes cytoplasmic redistribution of CaMKII’; page unavailable) reports CaMKII movement into spine heads and toward the postsynaptic density after a five-minute saline flush in rat hippocampal tissue. It does not report glutamate treatment of neuronal cultures. The Discussion (‘Technical issues’; page unavailable) says excessive glutamate-receptor activation is only one possible trigger of the reorganization, not an established cause in this experiment. (LLM, conf=0.88)

  cites [140] sourceExperimental — accuracy not yet measured

In the paper

“The second class of peptides is derived from an endogenous protein that inhibits CaMKII with nanomolar potencies (CaMKIIN) (149).”

What the sources say

• Supported

  The Results and Discussion report that the endogenous inhibitor CaM-KIIN inhibits CaM-KII with an apparent IC50 of 50 nM. They also report a 50 nM IC50 for the peptide derived from it, supporting the citing sentence’s description of nanomolar potency. (LLM, conf=0.94)

  cites [149] source

In the paper

“The monomeric units form a repeating chain of regulatory segment of one bound to the kinase groove of a second monomer and regulatory segment of that second monomer bound across the kinase groove of a third monomer, and so on.”

What the sources say

• Supported

  Split into 2 atoms — 2 confirmed, 0 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

  cites [62] source

In the paper

“since there are roughly 20–30 NMDARs in the PSD (66, 143) and ∼100 holoenzymes [∼1,200 CaMKII subunits (65)], after NMDAR activation only a small fraction of subunits per holoenzyme would be directly bound to GluN2B.”

What the sources say

• ?Not checked — no full text of the sourcecites [66] source
• Supported

  This source is cited for the estimate of roughly 20–30 NMDARs per PSD. The Discussion reports 20–30 NMDAR-type structures per PSD, and the EM tomography analysis reports 23 ± 5 in controls. The separate citation for the CaMKII holoenzyme count is not being judged here. (LLM, conf=0.93)

  cites [143] source
• ?Not checked — no full text of the sourcecites [65] source

In the paper

“Deletion of Shank3 has been reported to inhibit LTP (164–166).”

What the sources say

• Supported

  The Methods describe mice with a targeted deletion of Shank3 exons, and the Discussion reports reduced LTP in Shank3 heterozygous mice. Initial potentiation was normal, so the reported inhibition concerns sustained LTP rather than its onset. (LLM, conf=0.90)

  cites [164] source
• Supported

  The Results (Electrophysiology; page not specified) report that Shank3 null mice had significantly lower initial potentiation than controls and that it decayed to baseline by 40 minutes. The Discussion describes impaired LTP in these mice, supporting the claim that Shank3 deletion inhibits LTP. (LLM, conf=0.94)

  cites [165] source
• ?Not checked — no full text of the sourcecites [166] source

In the paper

“The requirements for clustering are consistent with the subunit capture conformation seen in a crystal structure of CaMKII truncated after a portion of the regulatory segment and lacking the CaM binding domain and the remaining COOH-terminal end of the protein (62).”

What the sources say

• Supported

  The crystal-structure passage describes a CaMKII construct containing the kinase domain and R1 portion of the regulatory segment, but lacking R2 and R3. The paper also describes the R1 segment of one subunit interacting with another kinase domain. The comparison with clustering requirements is the citing authors’ interpretation, not a finding attributed to this source. (LLM, conf=0.90)

  cites [62] source

In the paper

“A second difference is the specific high-affinity binding of the inactive β, but not α, to F-actin (167–170).”

What the sources say

• Partly supported

  The Results report that CaMKIIβ bound strongly to stabilized F-actin whereas CaMKIIα did not; the Discussion says inactive CaMKIIβ binds F-actin and activation reduces binding strength. The claim omits the stable-F-actin condition: the Results found weaker binding to dynamic F-actin. (LLM, conf=0.85)

  cites [167] sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [168] source
• ?Not checked — no full text of the sourcecites [169] source
• Partly supported

  The Discussion (page unavailable) says CaMKIIβ binds F-actin more strongly than CaMKIIα, but describes α as having a weaker interaction—not no interaction. The Results (page unavailable) also show that Ca2+/calmodulin promotes dissociation of the β–actin complex, consistent with binding in the inactive state. The categorical 'but not α' overstates the source. (LLM, conf=0.88)

  cites [170] sourceExperimental — accuracy not yet measured

In the paper

“expression of CaMKIIα in cells lacking both isoforms fully rescues LTP (121).”

What the sources say

• Supported

  The Results section (“The role of CaMKII in basal synaptic transmission and LTP”) explicitly reports that expressing CaMKIIα fully rescued LTP in the double knockout (DKO), meaning cells lacking both CaMKIIα and CaMKIIβ. (LLM, conf=0.95)

  cites [121] source

In the paper

“expression of CaMKIIβ in cells lacking both CaMKIIα and CaMKIIβ fails to rescue LTP (121)”

What the sources say

• Supported

  The Results section (‘The role of CaMKII in basal synaptic transmission and LTP’) states that expressing CaMKIIβ was unable to rescue LTP in the double-knockout cells. This directly supports the cited claim. (LLM, conf=0.95)

  cites [121] source

In the paper

“with intense induction protocols for LTP a component of LTP requires the activation of VDCCs (49–51, 193, 196).”

What the sources say

• ?Not checked — no full text of the sourcecites [49] source
• ?Not checked — no full text of the sourcecites [50] source
• Partly supported

  The Discussion (page unavailable) reports that high-frequency afferent activation produced prolonged potentiation that was much reduced by VDCC antagonists. But it says direct VDCC involvement in the potentiation seen after strong tetanization lacks evidence, and it does not establish that a component of tetanus-induced LTP requires VDCC activation. (LLM, conf=0.84)

  cites [51] sourceExperimental — accuracy not yet measured
• Partly supported

  The Discussion says L-type VDCCs generally do not contribute to LTP induction, while noting an exception for very high-frequency (200 Hz) tetani. That supports a role under an intense protocol, but this paper does not establish that a component of LTP requires VDCC activation under those conditions; its own nifedipine test found no effect on LTP induced by the tetanus it examined. (LLM, conf=0.84)

  cites [193] sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [196] source

In the paper

“Further studies showed that nitr-5 can evoke either an enhancement or a depression, depending on the amplitude and the duration of the Ca2+ signal (181).”

What the sources say

• Partly supported

  The Results section says earlier nitr-5 experiments produced either LTP (enhancement) or LTD (depression) after brief calcium elevations. But the paper attributes selective effects of calcium-signal amplitude and duration to its experiments with nitrophenyl-EGTA; it says nitr-5 could not produce a very high or prolonged calcium rise. Individual passage page numbers are unavailable in our copy. (LLM, conf=0.88)

  cites [181] sourceExperimental — accuracy not yet measured

In the paper

“The defect caused by CaMKIIβ deletion cannot be rescued by a mutant form of CaMKIIβ that cannot bind to F-actin (172).”

What the sources say

• Partly supported

  In Results, “Dissociation of CaMKIIβ from F-actin is necessary for functional and structural LTP” (page not available), the paper reports that the All A mutant failed to rescue the LTP reduction caused by CaMKIIβ shRNA. But All A retains F-actin binding; the binding-deficient All D mutant is described in Results, “Autophosphorylation sites on the F-actin binding region of CaMKIIβ regulate interaction with F-actin” (page not available), without a corresponding rescue result. The claim conflates the two mutants. (LLM, conf=0.86)

  cites [172] sourceExperimental — accuracy not yet measured

In the paper

“The discovery that NMDARs are highly permeable to Ca2+ (179, 180)”

What the sources say

• ?Not checked — no full text of the sourcecites [179] source
• Partly supported

  The paper’s numbered summary (points 5–6; page not available in our copy) reports a calcium-dependent reversal-potential shift and concludes that Ca2+ permeates NMDA channels. It does not characterize or quantify that permeability as high, so it supports calcium permeability but not the claim’s degree. (LLM, conf=0.78)

  cites [180] sourceExperimental — accuracy not yet measured

In the paper

“E: CA CaMKII containing the I205K mutation (n = 15 pairs) fails to potentiate synapses.”

What the sources say

• Supported

  The Results section (“CaMKII constitutive activity and the maintenance of LTP,” Fig. 10j,k) states that the I205K mutation completely prevented the enhancement produced by constitutively active CaMKIIα. The Discussion likewise says that this construct no longer affected synaptic responses when it carried I205K. (LLM, conf=0.94)

  cites [121] source

In the paper

“in one study (225) synapses lacking NMDARs had only a small reduction in the synaptic localization of CaMKII, suggesting contributions of other interacting proteins to its synaptic localization.”

What the sources say

• Supported

  The Results section (Figure 8a,b) reports that neurons lacking synaptic NMDARs showed only a ~15% reduction in CaMKII colocalization with PSD-95. It also proposes that densin and NMDAR subunits can each act as CaMKII docking sites, supporting the suggested contribution of another interacting protein. (LLM, conf=0.94)

  cites [225] source

In the paper

“The persistent accumulation of CaMKII at the PSD following an LTP-inducing stimulus is shown in FIGURE 6, A–C (217).”

What the sources say

• Supported

  The Results section, “Persistent increase in CaMKII density at the PSD,” reports that PSD-associated CaMKIIα increased after chemical LTP induction and remained elevated for at least one hour. The source illustrates this finding in its Figure 7A–C; the citing sentence’s Figure 6 reference does not change the finding attributed to the source. (LLM, conf=0.92)

  cites [217] source

In the paper

“D: constitutively active CaMKII (T286D-T305A/T306A) (CA CaMKII) in 2-amino-5-phosphonopentanoic acid (APV) (n = 8 pairs) potentiates synapses.”

What the sources say

• Supported

  The Results figure legend for Fig. 10h reports eight control–transfected cell pairs treated with APV. AMPAR EPSCs were higher in cells expressing constitutively active CaMKIIα T286D-T305A/T306A than in controls (296.9 ± 32 versus 124.2 ± 22.6; p = 0.007), supporting the stated potentiation. (LLM, conf=0.95)

  cites [121] source

In the paper

“G: cells in which CaMKII has been replaced with CaMKII containing the I205K mutation fail to express LTP.”

What the sources say

• Supported

  The Results section (‘The synaptic actions of CaMKII require binding to NMDARs’) reports that, on a CaMKII double-knockout background, expressing CaMKIIα I205K abolished LTP (Fig. 8g). The Discussion likewise states that LTP was absent with the CaMKIIα I205K mutant. (LLM, conf=0.94)

  cites [121] source

In the paper

“The second approach utilizes mutations in the CTD of GluN2B that cannot bind CaMKII as replacement for the wt (CaMKII L1298A/R1300Q) (120, 121, 230) or either competing with overexpressed full-length constructs (GluN2B R1300S/Q1303D) (231) or competing by expressing just the wt GluN2B CTD (839–1482) (232). The results are highly variable as reviewed in Ref. 233, ranging from complete block of LTP”

What the sources say

• Supported

  The Introduction identifies this paper’s L1298A/R1300Q GluN2B knock-in as a mutation approach that blocks CaMKII binding. The Results report approximately 50% less LTP in those mice. The citing sentence assigns the broader claim about effects ranging from complete block to no block to a separate review (Ref. 233), not to this paper. (LLM, conf=0.90)

  cites [120] source
• Partly supported

  In Results (‘The synaptic actions of CaMKII require binding to NMDARs’; page unavailable), the paper describes replacing GluN2B with a mutant that cannot bind CaMKII. It measures basal synaptic transmission in that replacement experiment, not LTP. The Discussion reports that earlier GluN2B-binding mutations impaired LTP, but does not establish the claimed range from complete block to no block. (LLM, conf=0.85)

  cites [121] sourceExperimental — accuracy not yet measured
• Partly supported

  The Results report that the paper’s GluN2B CaMKII-binding-site mutant showed no significant difference in theta-burst LTP from wild type. But its knockin is L1298A/R1300N/S1303D, not the L1298A/R1300Q variant named alongside this citation. The broader range from complete block to no block is attributed in the citing sentence to a separate review, not to this paper. (LLM, conf=0.87)

  cites [230] sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [231] source
• Supported

  The Materials and Methods describe a construct containing the NR2B C-terminal residues 839–1482, and the Results report that it reduced CaMKII binding to endogenous NR2B and impaired LTP. This supports the construct-specific point attached to this citation. The broader claim that results range from complete to no LTP block is assigned in the citing sentence to a separate review citation, not to this paper. (LLM, conf=0.91)

  cites [232] source

In the paper

“the results with the genetic deletion of CaMKIIα are more variable, some showing roughly a 50% block (87, 204–206) but some showing a complete block (121, 207, 208).”

What the sources say

• ?Not checked — no full text of the sourcecites [87] source
• ?Not checked — no full text of the sourcecites [204] source
• Partly supported

  The section “LTP is impaired in mice with adult deletion of αCaMKII” reports significantly impaired LTP after genetic deletion under both stimulation protocols (Fig. 4C,D). It does not describe a complete block of LTP or quantify a roughly 50% block; the paper’s statement that deletion was complete concerns αCaMKII protein, not LTP. (LLM, conf=0.84)

  cites [205] sourceExperimental — accuracy not yet measured
• Supported

  The Discussion reports that LTP in α-CaMKII mutant mice was reduced to about 50% of wild-type levels. The Results contrast this with an earlier study in which most mutant slices showed no LTP, supporting the sentence’s description of variable findings. This source supports the roughly 50% block for which its citation is offered. (LLM, conf=0.93)

  cites [206] source
• Supported

  The Discussion states that deleting CaMKIIα fully blocks LTP, supporting the complete-block result for which this paper is cited. The roughly 50% result is attached to other citations in the sentence; this paper’s approximately 50% reduction concerns AMPAR currents, not LTP blockade. (LLM, conf=0.90)

  cites [121] source

In the paper

“Numerous studies using peptide inhibitors have confirmed the block of LTP (200–203).”

What the sources say

• Supported

  The Results (passage 2) report that a peptide kinase inhibitor completely blocked LTP induction when a protease inhibitor cocktail was added. The paper also describes earlier studies in which postsynaptic peptide inhibitors blocked LTP development (Results, passage 7). Its finding concerns LTP induction, not maintenance. (LLM, conf=0.91)

  cites [200] source
• ?Not checked — no full text of the sourcecites [201] source
• Supported

  The Results section (‘tatCN21 inhibits induction but not maintenance of LTP’) reports that the peptide inhibitor tatCN21 blocked LTP generation. The Discussion also cites earlier studies in which other CaMKII inhibitors inhibited LTP induction. The supported block is of LTP induction, not maintenance. (LLM, conf=0.88)

  cites [202] source
• ?Not checked — no full text of the sourcecites [203] source

In the paper

“A photoactivatable form of CaMKII has recently been reported to induce LTP (215).”

What the sources say

• Supported

  The Introduction states that the authors developed photoactivatable CaMKII and demonstrated that its activation induces functional LTP at the single-spine level. The Discussion likewise says that paCaMKII allows LTP induction at single synapses. (LLM, conf=0.94)

  cites [215] source

In the paper

“activation of CaMKII results in the rapid translocation of CaMKII to the PSD (137, 140, 169, 216–220).”

What the sources say

• Partly supported

  The introduction says PSDs thicken during synaptic activity partly because of CaMKII translocation, and the discussion reports that glutamate induces reversible CaMKII association with PSDs. The paper supports rapid translocation after stimulation, but does not establish that activation of CaMKII itself causes it. (LLM, conf=0.83)

  cites [137] sourceExperimental — accuracy not yet measured
• Partly supported

  The Results (‘Prolonged saline flush causes cytoplasmic redistribution of CaMKII’; page unavailable) report movement toward the PSD within five minutes of a saline flush. The Discussion (‘Technical issues’; page unavailable) says the trigger remains unclear, so the paper does not establish that CaMKII activation causes that translocation. (LLM, conf=0.86)

  cites [140] sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [169] source

In the paper

“The primary interaction in the CaMKII-GluN2B complex is between the CTD, also referred to as C-tail, of the GluN2B subunit and the surface groove in the kinase domain of CaMKII exposed by its activation (61).”

What the sources say

• Supported

  The Results ('The CaMKII kinase domain has a single binding site'; page unavailable) show that a GluN2B peptide binds a single surface site on the kinase domain at the active site. The Introduction says CaMKII activation frees the substrate-binding site, and the Results ('The αD helix rotates outward to accommodate binding partners'; page unavailable) show that this movement presents a hydrophobic patch for GluN2B binding. The structure uses a GluN2B peptide rather than a full-length complex, but the described interaction matches the claim. (LLM, conf=0.88)

  cites [61] source

In the paper

“A–C: chemical long-term potentiation (chemLTP) induction results in persistent accumulation of CaMKIIα at the PSD.”

What the sources say

• Supported

  The “Persistent increase in CaMKII density at the PSD” section reports that increased PSD-associated CaMKIIα labeling persisted for at least one hour after cLTP induction. Figure 7’s caption explicitly describes persistent accumulation at the PSD in hippocampal synapses. (LLM, conf=0.95)

  cites [217] source

In the paper

“The first approach removed the GluN2B subunit, either by knockout or knockdown. The results vary from a complete block of LTP (226, 227) to a partial impairment (228, 229).”

What the sources say

• ?Not checked — no full text of the sourcecites [226] source
• Partly supported

  In Results, ‘NR2B Expression is Required for LTP’ (page not provided), NR2B RNAi reduced LTP to 1.0 ± 0.1-fold versus 1.7 ± 0.4-fold in controls, supporting the complete-block end of the stated range. This paper does not establish the claimed variation across knockout and knockdown studies or report partial impairment from its NR2B knockdown. (LLM, conf=0.84)

  cites [227] sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [228] source
• Supported

  The Discussion reports that the authors’ GluN2B mutants lost potentiation under a sub-saturating LTP protocol but showed normal LTP under a saturating protocol. This supports the partial-impairment end of the citing sentence’s range, for which this source is cited; the complete block is attributed to other references. (LLM, conf=0.91)

  cites [229] source

In the paper

“keeping in mind that this mutation is expected to also disrupt the binding to densin-180 and α-actinin-2 (234–236) and other PSD proteins.”

What the sources say

• Partly supported

  The Results and Discussion section shows that CaMKII binds densin-180 variants A, B, and D. It does not examine the CaMKII I205K mutation or show that this mutation disrupts binding to densin-180, α-actinin-2, or other PSD proteins. The residue 205 mentioned there belongs to a densin-180 fragment, not the CaMKII mutation. (LLM, conf=0.85)

  cites [234] sourceExperimental — accuracy not yet measured
• Partly supported

  The Discussion reports that CaMKIIα binds densin-180 and α-actinin, but at distinct sites: densin binds its association domain, whereas α-actinin binds its kinase domain. The paper identifies α-actinin-4 as a binding partner, not α-actinin-2, and does not test the I205K mutation or establish that it disrupts binding to these or other PSD proteins. The reported interactions support only the background premise of the prediction. (LLM, conf=0.86)

  cites [235] sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [236] source

In the paper

“Of particular interest is the finding that these clusters are juxtaposed across from presynaptic release site (251, 252, 260).”

What the sources say

• Supported

  The paper reports that GluA2-containing AMPARs are significantly enriched opposite presynaptic RIM1/2 nanoclusters. It also states that vesicle fusion occurs preferentially at sites directly opposing postsynaptic receptor ensembles (passages 1 and 7, labeled Abstract in our copy). (LLM, conf=0.94)

  cites [251] source
• Supported

  In “Are release sites and receptors aligned in nanocolumns?” (page unavailable), the review reports that presynaptic RIM nanoclusters align across the synaptic cleft with AMPAR nanoclusters, and that RIM localization predicts where fusion events occur. This supports the claim that AMPAR clusters lie opposite presynaptic release sites, though the paper notes that the precise architecture of those sites remains unresolved. (LLM, conf=0.90)

  cites [252] source
• Supported

  The Results section on presynaptic RIM and postsynaptic AMPAR nanodomains describes release sites positioned in front of AMPAR clusters. It also reports RIM–AMPAR apposition under control conditions and reduced apposition when NLG1-ΔCter is expressed. No page number is available in our copy. (LLM, conf=0.94)

  cites [260] source

In the paper

“the CA CaMKII containing the I205K mutation failed to enhance responses (FIGURE 6, E AND F).”

What the sources say

• Supported

  The Results section, “CaMKII constitutive activity and the maintenance of LTP,” states that the I205K mutation completely prevented the enhancement of synaptic transmission caused by constitutively active CaMKIIα (Fig. 10j, k). The Discussion likewise says the constitutively active kinase had no effect on synaptic responses when it harbored I205K. (LLM, conf=0.94)

  cites [121] source

In the paper

“Whereas CA CaMKII had its typical ∼2.5-fold enhancement (FIGURE 6, D AND F)”

What the sources say

• Supported

  The Results section’s Fig. 10 caption reports AMPAR EPSCs at 249.8 ± 26.3% of control for constitutively active CaMKIIα T286D-T305A/T306A—approximately a 2.5-fold enhancement. The experiment used overexpression in slice culture. (LLM, conf=0.95)

  cites [121] source

In the paper

“NMDARs tend to be concentrated in the middle of the PSD with little overlap with AMPARs (253–255).”

What the sources say

• Supported

  The Discussion reports that NMDAR clusters were mainly central in the PSD, surrounded by AMPAR clusters, and that the two types of cluster barely colocalized at the nanometric scale. This supports both parts of the citing sentence. (LLM, conf=0.94)

  cites [253] source
• Partly supported

  The 'Tangential distribution of NMDARs across the postsynaptic density' section (page unavailable) finds NMDAR labeling more frequent in the middle of the PSD. The 'Tangential distribution of AMPARs and NMDARs' section (page unavailable) describes AMPARs as more evenly distributed and proposes segregated microclusters, but does not establish that the receptors have little spatial overlap; it describes NMDAR clusters as interspersed among AMPARs in larger synapses. (LLM, conf=0.84)

  cites [254] sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [255] source

In the paper

“Not unexpectedly, LTP was entirely blocked, suggesting that the CaMKII-GluN2B complex is necessary for LTP (FIGURE 6G).”

What the sources say

• Supported

  In Results (“The synaptic actions of CaMKII require binding to NMDARs”), the authors report that the CaMKIIα I205K mutant abolished LTP when expressed on a CaMKII double-knockout background (Fig. 8g). The Discussion likewise states that LTP was absent in the I205K mutant. (LLM, conf=0.94)

  cites [121] source

In the paper

“expressing CA CaMKII failed to enhance AMPAR responses.”

What the sources say

• Partly supported

  In Results (‘The synaptic actions of CaMKII require binding to NMDARs’; page unavailable), a GluN2B mutant prevented enhancement by CaMKIIα T305A/T306A, but the paper does not identify that construct as constitutively active CaMKII. Results (‘CaMKII constitutive activity and the maintenance of LTP’) and the Discussion report that constitutively active CaMKII failed to enhance responses when CaMKII itself carried the I205K mutation. The claimed outcome is supported, but not for the stated combination of constructs. (LLM, conf=0.90)

  cites [121] sourceExperimental — accuracy not yet measured

In the paper

“Not unexpectedly, LTP was entirely blocked, suggesting that the CaMKII-GluN2B complex is necessary for LTP (FIGURE 6G).”

• Also stated in: § 7.2. CaMKII Targets · plain

  “As discussed in sect. 6, the formation of a CaMKII-GluN2B complex is essential for LTP”

What the sources say

• Supported

  The Results section (‘The synaptic actions of CaMKII require binding to NMDARs’; page unavailable) reports that preventing CaMKII binding to NMDARs abolished LTP. The Discussion identifies GluN2B as CaMKII’s primary binding partner at the postsynaptic density, supporting the claim about the CaMKII–GluN2B complex. (LLM, conf=0.90)

  cites [121] source

In the paper

“AMPARs form discrete clusters scattered around the PSD (256–259).”

What the sources say

• ?Could not verify

  The Results section on spine size reports discrete clusters of PSD-95 and presynaptic proteins, not AMPARs. The Discussion mentions AMPARs only as possible proteins involved in forming new nanomodules; it does not describe AMPAR clusters or their distribution around the PSD. Searches for AMPAR clustering and localization found no passage establishing the claim. (LLM, conf=0.73)

  cites [256] sourceExperimental — accuracy not yet measured
• Supported

  The “Nanoscale co-organization of PSD95 and AMPARs” section (page not supplied) reports that AMPARs localize to distinct nanodomains within the PSD. It also describes multiple nanodomains within individual PSDs, supporting the description of discrete clusters distributed across the PSD. (LLM, conf=0.91)

  cites [257] source
• Partly supported

  The Discussion states that AMPARs are present in the PSD and describes their capture through PSD-95. The Results describe discrete postsynaptic nanodomains of PSD-95, not AMPARs; our copy does not establish that AMPARs themselves form discrete clusters scattered around the PSD. (LLM, conf=0.82)

  cites [258] sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [259] source

In the paper

“Thus, it is now generally accepted that LTP expression is due to the rapid postsynaptic accumulation of AMPARs.”

What the sources say

• Supported

  The Long-Term Potentiation section says evidence turned opinion toward a postsynaptic mechanism for LTP expression. The Dynamic Regulation of AMPAR Membrane Trafficking section reports rapid receptor trafficking and AMPAR recruitment to synaptic spines after LTP induction, paralleling synaptic strengthening (page unavailable in our copy). (LLM, conf=0.86)

  cites [17] source
• ?Not checked — no full text of the sourcecites [18] source
• ?Not checked — no full text of the sourcecites [19] source
• Partly supported

  The sections on AMPAR trafficking and surface exocytosis describe rapid AMPAR accumulation at synapses and diffusion-trapping of AMPARs during initial potentiation. But the section on the cellular basis of plasticity says LTP expression is *largely* postsynaptic, with a possible presynaptic contribution at some synapses; the unqualified claim omits that limitation. Page numbers are unavailable in our copy. (LLM, conf=0.86)

  cites [20] sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [21] source
• Partly supported

  In “The AMPAR code and LTP,” the paper says the major form of LTP involves increased postsynaptic AMPAR density, supporting the accumulation mechanism. But it also identifies increased channel conductance as an alternative and describes cerebellar LTP that does not appear to involve AMPAR trafficking. The unqualified claim therefore overstates what the paper says. (LLM, conf=0.87)

  cites [22] sourceExperimental — accuracy not yet measured
• Supported

  The Introduction says there is general agreement that AMPA receptors are recruited to the postsynaptic membrane following LTP. The section on receptor delivery during LTP describes rapid incorporation of AMPA receptors as a mechanism for early synaptic potentiation, supporting the citing sentence’s broad paraphrase. (LLM, conf=0.86)

  cites [237] source

In the paper

“These synapses contain AMPARs, but LTP-inducing stimuli cause a rapid addition of AMPARs to the synapse (246–250).”

What the sources say

• Partly supported

  The passage labeled Abstract (page not provided) reports that LTP induction increased spine volume and synaptic current, and says spine volume is associated with the number of AMPA receptors in the postsynaptic density. It does not show that AMPARs were rapidly added to those synapses after stimulation; receptor addition was not directly measured. (LLM, conf=0.83)

  cites [246] sourceExperimental — accuracy not yet measured
• Partly supported

  The passage on AMPA-current mapping (labeled Abstract in our copy; no page given) reports that LTP-inducing paired stimulation rapidly increased AMPA-receptor-mediated currents at the stimulated spine, detectable within 3 minutes. A separate passage says spine enlargement might promote AMPA receptor accumulation, but the paper does not establish that receptors were rapidly added to the synapse. (LLM, conf=0.86)

  cites [247] sourceExperimental — accuracy not yet measured
• ?Could not verify

  The section “LTP leads to plasticity in spine head and neck geometry” (pp. 10–11) reports potentiated currents and changes in spine shape after LTP induction, but does not say that AMPARs were rapidly added to synapses. Searches for AMPAR insertion and rapid receptor addition found no passage addressing that mechanism in our copy. (LLM, conf=0.74)

  cites [248] sourceExperimental — accuracy not yet measured
• Supported

  The Results section ('Detecting exocytosis of endogenous AMPARs on dendrites following LTP'; page not provided) reports a rapid increase in the response of a potentiated spine and concludes that endogenous AMPARs move to synapses immediately after LTP induction. This supports the attributed rapid addition of AMPARs to spine synapses. (LLM, conf=0.92)

  cites [249] source
• Supported

  The Results report that LTP-inducing glutamate uncaging increased surface GluA1 in stimulated dendritic spines. The Discussion says AMPARs were recruited to those spines within approximately 10 seconds of LTP induction, supporting rapid addition; the Introduction describes their subsequent movement into and anchoring at the synapse. (LLM, conf=0.93)

  cites [250] source

In the paper

“Furthermore, expressing a constitutively active CaMKII on the background of GluN2B(L1298A-R1300Q) failed to enhance synaptic transmission (121).”

What the sources say

• Partly supported

  In Results (‘The synaptic actions of CaMKII require binding to NMDARs’; Fig. 9), a CaMKII-binding-deficient GluN2B mutant prevented the synaptic enhancement produced by CaMKIIα T305A/T306A. But the paper identifies the constitutively active construct as T286D-T305A/T306A and reports its failed enhancement with a CaMKIIα I205K mutation instead (Results, ‘CaMKII constitutive activity and the maintenance of LTP’; Fig. 10). Our copy therefore does not establish the claimed combination of constitutively active CaMKII and GluN2B(L1298A-R1300Q). (LLM, conf=0.88)

  cites [121] sourceExperimental — accuracy not yet measured

In the paper

“Phosphorylation of S831 of heterologously expressed GluA1 by CaMKII increases single-channel conductance (267, 268).”

What the sources say

• Supported

  The Results describe GluR1 (GluA1) expressed in HEK-293 cells with activated CaMKII and report increased phosphorylation at Ser-831. They also report a significant increase in single-channel conductance, with Ser-831 mutants mimicking the effect. (LLM, conf=0.94)

  cites [267] source
• Supported

  In Results, “Control of recombinant AMPA receptor function by CaMKII” (page unavailable), CaMKII increased mean single-channel conductance in HEK cells expressing GluA1; the effect was absent with the S831A mutation. The paper attributes the conductance increase to phosphorylation of GluA1-Ser831. It clarifies that larger conductance levels occur more often, rather than each level increasing in amplitude. (LLM, conf=0.93)

  cites [268] source

In the paper

“It is now well established that the delivery of AMPARs to synapses requires the binding of TARPs, via their PDZ binding motifs (PBMs), to PSD-95 (278, 279, 286, 289, 294).”

What the sources say

• ?Not checked — no full text of the sourcecites [278] source
• Supported

  In “The role of TARPs in AMPAR surface trafficking and synaptic targeting” (page not provided), the paper says TARPs target AMPARs to synapses through their PDZ-binding motifs, which bind PSD-95 and related scaffolding proteins. It also reports that disrupting the stargazin–PSD-95 interaction prevents AMPAR complexes from clustering at postsynaptic densities. (LLM, conf=0.90)

  cites [279] source
• ?Not checked — no full text of the sourcecites [286] source
• Supported

  The Introduction reports that a TARP lacking its PDZ-binding motif restored AMPAR surface expression but not synaptic expression, indicating that the motif is essential for synaptic targeting and/or retention. It also states that TARPs link AMPARs to PSD-95; the paper clarifies that an upstream Arg-rich motif contributes to this binding alongside the PDZ-binding motif. (LLM, conf=0.88)

  cites [289] source
• Partly supported

  The Discussion identifies TARP PBM binding to PSD-95 as a main requisite for synaptic AMPAR trafficking. But the Limitations of the study notes that roughly 35% of synaptic transmission may be TARP independent, so the unqualified claim that AMPAR delivery requires this interaction is broader than the paper establishes. Page numbers were not provided in our copy. (LLM, conf=0.87)

  cites [294] sourceExperimental — accuracy not yet measured

In the paper

“Previous studies have shown that the Rho GTPase Rac1 regulates synaptic AMPAR expression (311)”

What the sources say

• Supported

  The Figure 2 caption in the Discussion states that Rac1 activation recruits functional AMPARs to synapses in immature neurons. Figure 6 also reports that inhibiting Rac1 reduced synaptic responses. Together, these support the claim that Rac1 regulates synaptic AMPAR expression. (LLM, conf=0.93)

  cites [311] source

In the paper

“Surprisingly, expression of a phosphomimic mutation, S73D, prevents LTP expression (296).”

What the sources say

• Supported

  The Results section, “Expression of S73D PSD-95 Prevents LTP Expression,” reports that stimulated and control spine responses were indistinguishable in S73D PSD-95-expressing neurons and concludes that S73D prevents LTP. The Discussion likewise describes S73D as a phosphomimetic substitution that blocks LTP. (LLM, conf=0.95)

  cites [296] source

In the paper

“Furthermore, knocking out PSD-95 does not prevent LTP (299).”

What the sources say

• Supported

  The Results report that LTP was induced—and strongly enhanced—in hippocampal slices from PSD-95 mutant mice. The Discussion also describes enhanced LTP in the absence of PSD-95, supporting the claim that removing PSD-95 does not prevent LTP. (LLM, conf=0.94)

  cites [299] source

In the paper

“Experiments in which the CTD of GluA1 is deleted (Refs. 273–275, but see Refs. 276, 277) express LTP that is indistinguishable from that in neighboring control cells.”

What the sources say

• Partly supported

  In “No GluA1 C-tail domains are required for LTP” (page unavailable), two GluA1 C-tail truncations produced LTP comparable to control cells, and complete C-tail truncation rescued LTP when paired with GluA2. But “LTP requires a reserve pool of AMPARs” (page unavailable) reports substantially impaired LTP when completely truncated GluA1ΔC was used alone. The claim omits this condition-dependent result. (LLM, conf=0.88)

  cites [273] sourceExperimental — accuracy not yet measured
• Partly supported

  The Results (Figure 2F–G; page unavailable) report that mice with a truncated GluA1 CTD had LTP no different from wild-type controls. But those controls were cells from wild-type mice, not neighboring control cells. The paper’s recordings from adjacent control cells involved GluA1 whose CTD was replaced with the GluA2 CTD, rather than simply deleted. (LLM, conf=0.84)

  cites [274] sourceExperimental — accuracy not yet measured
• Partly supported

  The Discussion (page unavailable) says the GluA1 CTD PDZ ligand is not required for synaptic potentiation, which supports a narrower version of the claim. It does not report that deleting the entire GluA1 CTD produces LTP indistinguishable from neighboring control cells; it says the CTD’s requirement for synaptic plasticity needs further investigation. (LLM, conf=0.83)

  cites [275] sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [276] source
• Partly supported

  The Results section on adult mice (Figure 7B; page unavailable) reports paired-LTP of similar magnitude in GluA1C2KI and wild-type mice. But this paper replaces the GluA1 CTD rather than deleting it, compares mouse groups rather than neighboring cells, and reports impaired LTP with other protocols and in juvenile mice. Its Discussion says a separate CTD-deletion study found that paired-LTP could be induced, not that it was indistinguishable from neighboring controls. (LLM, conf=0.86)

  cites [277] sourceExperimental — accuracy not yet measured

In the paper

“Evidence has accumulated over the past decade indicating that AMPARs are not stand-alone receptors; rather, they are decorated by numerous types of auxiliary subunits that control the synthesis, trafficking, and biophysical properties of these receptors (278–281).”

What the sources say

• ?Not checked — no full text of the sourcecites [278] source
• Partly supported

  The sections “Transmembrane AMPAR auxiliary subunits beyond TARPs” and “TARPs form an integral component of native AMPAR complexes” describe auxiliary subunits associated with AMPARs that affect trafficking and gating. “TARPs and AMPAR biogenesis” reports evidence for a role in biosynthesis and ER export, but says whether TARPs act in earlier steps such as folding and subunit assembly remains undetermined. The claim overstates the evidence by saying auxiliary subunits control AMPAR synthesis. (LLM, conf=0.84)

  cites [279] sourceExperimental — accuracy not yet measured
• Partly supported

  The sections “Glutamate receptors and auxiliary subunits” and “Conclusions” say that AMPARs contain auxiliary subunits that regulate trafficking and channel properties. Our copy does not say that those subunits control AMPAR synthesis. (LLM, conf=0.88)

  cites [280] sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [281] source

In the paper

“sLTP is induced by a two-photon glutamate uncaging LTP protocol onto single spines that leads to a rapid and long-lasting enlargement of spines that parallels the enhanced AMPAR responses.”

What the sources say

• Supported

  The Abstract reports that two-photon glutamate uncaging at single spines induced rapid enlargement, persistent in small spines, and increased AMPA-receptor-mediated currents at the stimulated synapse. It also reports a strong correlation between AMPA-current potentiation and enlargement of the same spine head. (LLM, conf=0.93)

  cites [247] source

In the paper

“Indeed, spines contain unusually high levels of actin (307–309)”

What the sources say

• ?Not checked — no full text of the sourcecites [307] source
• Supported

  The Introduction’s “Actin organization in dendritic spines” section calls actin the major cytoskeletal component of dendritic spines. The Figure 1 description also says the spine head typically contains a dense network of actin filaments, supporting the qualitative claim that spines are especially rich in actin; the copy gives no numerical comparison. (LLM, conf=0.88)

  cites [308] source
• Supported

  In “Cortactin is enriched in dendritic spines and translocates with glutamate stimulation” (page not available), the paper describes a concentration of F-actin in dendritic spines, revealed by phalloidin labeling. This supports the citing sentence’s broad claim that spines contain high levels of actin. (LLM, conf=0.87)

  cites [309] source

In the paper

“and this dispersal is required for chemLTP (304).”

What the sources say

• Supported

  In Results, under “SynGAP Synaptic Dispersion Is Required for LTP,” the authors report that a construct retaining SynGAP in spines during chemLTP blocked LTP-associated spine enlargement and receptor insertion. They conclude that SynGAP relocation is required for these downstream effects. (LLM, conf=0.89)

  cites [304] source

In the paper

“Imaging studies have reported that this phosphorylation triggers movement of SynGAP out of the PSD (304, 305)”

What the sources say

• Supported

  The Results section reports that phosphorylation of SynGAP at S1108 and S1138 promotes its dissociation from PSD-95 and dispersion from spines. It also reports that phosphorylated SynGAP was dispersed from the PSD upon LTP. Page numbers were not supplied. (LLM, conf=0.94)

  cites [304] source
• Partly supported

  The Discussion reports that NMDA stimulation moves both SynGAP isoforms out of the PSD core and that inhibiting CaMKII blocks this redistribution. The paper links the movement to CaMKII-mediated phosphorylation, but its imaging shows movement out of the PSD core, not out of the PSD complex as the claim states (Discussion; page not available). (LLM, conf=0.88)

  cites [305] sourceExperimental — accuracy not yet measured

In the paper

“These studies show that many synaptic proteins are remarkably dynamic, including AMPARs (237, 249, 327, 328), NMDARs (329), CaMKII (330), and scaffolding proteins, such as PSD-95 (331).”

What the sources say

• ?Could not verify

  Citation 331 is attached to PSD-95, so the question is whether this source describes that scaffolding protein as dynamic. The Introduction section on lateral diffusion identifies PSD-95 as an anchor for mobile AMPA receptors, but does not describe PSD-95 itself as mobile or dynamically exchanged. Searches for PSD-95 mobility, turnover, and trafficking found no passage establishing that point. (LLM, conf=0.72)

  cites [237] sourceExperimental — accuracy not yet measured
• Supported

  The citation is attached to the AMPAR portion of the sentence, not to NMDARs, CaMKII, or PSD-95. The Introduction describes AMPAR delivery by lateral diffusion and exocytosis, and the Discussion reports that GluR1-containing AMPARs move from the spine surface into synapses during LTP. (LLM, conf=0.92)

  cites [249] source
• ?Not checked — no full text of the sourcecites [327] source
• Supported

  This source is cited for AMPARs, not the other proteins listed in the sentence. In the introduction, the paper says AMPARs diffuse rapidly across synaptic and extrasynaptic membranes; the section on AMPAR mobility reports that 70% of GluR1-AMPARs were mobile in the postsynaptic membrane. Page numbers were not provided in our copy. (LLM, conf=0.94)

  cites [328] source
• ?Not checked — no full text of the sourcecites [329] source
• ?Not checked — no full text of the sourcecites [330] source

In the paper

“In the LLPS study, autophosphorylation is necessary for persistence of the condensate.”

What the sources say

• Supported

  The section “Autophosphorylation of CaMKII is required for persistence of protein condensate” states that T286 autophosphorylation is required to maintain condensates after Ca²⁺ dissipates, although it is not needed for their initial formation. The Discussion repeats this distinction. (LLM, conf=0.95)

  cites [323] source

In the paper

“Numerous studies have shown that blocking postsynaptic exocytosis by a variety of means blocks LTP/sLTP (195, 249, 250, 332–338).”

What the sources say

• Supported

  The Abstract reports that blocking postsynaptic Syt1 and Syt7 expression abolished LTP, and that a dominant-negative Syt1 mutant blocked postsynaptic AMPA-receptor exocytosis and LTP. The “Syt1-C2A*B* blocks LTP” section likewise reports blocked LTP and LTP-induced AMPA-receptor exocytosis. The cited paper therefore supports the LTP part of this multi-citation statement; it does not need to establish the collective claim about numerous studies on its own. (LLM, conf=0.90)

  cites [195] source
• Partly supported

  The Discussion (page not available in our copy) says several studies found that agents targeted to block exocytosis can inhibit LTP, supporting the LTP portion of the claim. It does not establish that blocking exocytosis blocks structural LTP (sLTP), and it cautions that some agents may inhibit LTP through effects other than exocytosis. (LLM, conf=0.84)

  cites [249] sourceExperimental — accuracy not yet measured
• Partly supported

  The results describing Fig. 2D report that tetanus toxin, which inhibits vesicle-dependent exocytosis, partially blocked structural plasticity. This supports a role for postsynaptic exocytosis in sLTP, but not the claim that blocking it fully blocks LTP/sLTP; the paper does not establish that result by a variety of means. (LLM, conf=0.88)

  cites [250] sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [332] source
• Supported

  The Results section reports that postsynaptic complexin knockdown blocks LTP, and the Discussion states that earlier studies found postsynaptic SNARE-mediated membrane fusion necessary for LTP. Together, these support the cited sentence’s point that interfering with postsynaptic exocytosis can block LTP. Page numbers are unavailable in our copy. (LLM, conf=0.90)

  cites [333] source
• Partly supported

  The Results section, “Postsynaptic Syntaxin-3 is Required for AMPAR Delivery During LTP” (page unavailable), reports that postsynaptic syntaxin-3 knockdown blocked the cultured LTP-like response but impaired, rather than abolished, LTP in acute slices. The Results section on SNAP-47 likewise reports impaired slice LTP. These findings support the role of postsynaptic exocytosis in LTP, but not an unqualified claim that the interventions blocked LTP or structural LTP (sLTP). (LLM, conf=0.83)

  cites [334] sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [335] source
• ?Not checked — no full text of the sourcecites [336] source
• Partly supported

  The Results section ('Resolving distinct exocytic insertion modes'; page not available) reports that potentiating stimuli increase AMPA-receptor exocytic events, and the Discussion says receptor delivery plays a critical role in initiating synaptic potentiation. The paper does not report that blocking postsynaptic exocytosis—by any means—blocks LTP or sLTP. (LLM, conf=0.82)

  cites [337] sourceExperimental — accuracy not yet measured
• Supported

  The Results section, “Stx4 is Required for Long-Term Potentiation in CA1 Hippocampus” (page unavailable), reports that an acute postsynaptic block of Stx4 function nearly abolished LTP. As one of several citations attached to the sentence, this paper supports an example of blocking postsynaptic exocytosis to block LTP. (LLM, conf=0.90)

  cites [338] source

In the paper

“In both cases addition of Ca2+/CaM in the absence of ATP or with the T286A mutant was sufficient for the LLPS or complex formation.”

What the sources say

• ?Not checked — no full text of the sourcecites [78] source
• Supported

  The section ‘Autophosphorylation of CaMKII is required for persistence of protein condensate’ (page unavailable) reports that condensates formed with Ca²⁺ in the absence of ATP and with the T286A mutant. The paper distinguishes initial formation from persistence: in both conditions, the condensates dissolved after Ca²⁺ was chelated. (LLM, conf=0.93)

  cites [323] source

In the paper

“and that the Rho GTPases Cdc42 and RhoA are required for LTP and sLTP (312, 313).”

What the sources say

• Supported

  The paper reports that inhibiting Rho or Cdc42 signaling impaired spine enlargement, with Rho affecting the transient phase and Cdc42 the sustained phase (Fig. 4; page unavailable). It also reports that both inhibitors impaired induced synaptic potentiation (Supplementary Fig. 10), supporting the claim about structural and functional LTP. (LLM, conf=0.88)

  cites [312] source
• Supported

  The ‘Plasticity of spine heads’ section says functional and structural plasticity both require activation of small GTPases. The ‘Regulation of actin cytoskeleton’ section identifies Cdc42 and RhoA among the GTPases involved in LTP and reports that inhibiting their respective effectors inhibits structural LTP. (LLM, conf=0.89)

  cites [313] source

In the paper

“Furthermore, when AMPAR/TARPs (represented by the CTD of the auxiliary TARP subunit Stargazin) and PSD-95 are added to the system, activation of CaMKII in the presence of ATP partitions AMPAR/TARPs and NMDARs into two different phases (FIGURE 8).”

What the sources say

• Supported

  The Segregation of AMPAR and NMDAR section reports that, after Ca²⁺ activation, the Stargazin-tail proxy for AMPAR and GluN2B partitioned predominantly into different regions of a phase-in-phase condensate with CaMKII and PSD-95. Figure 2 describes this protein mixture in the presence of Mg²⁺-ATP. The paper also notes that ATP is not required to initiate the segregation, but the claim says only that ATP was present. (LLM, conf=0.93)

  cites [323] source

In the paper

“However, actin is largely excluded from the core of the PSD, interacting with such proteins as Shank and SAPAP located in the deeper, cytoplasmic-facing layer of the PSD referred to as the pallium (66, 162, 322).”

What the sources say

• ?Not checked — no full text of the sourcecites [66] source
• Partly supported

  The Conclusion (page unavailable) places the pallium between the PSD core and the actin cytoskeleton, and the Pallial Layer section places Shank in that deeper layer. The Actin and the PSD section describes possible links between actin and Shank-binding proteins, but does not establish that actin is largely excluded from the core. It also does not describe actin interacting with SAPAP; the Core Layer section places SAPAP/GKAP at the interface between core and pallium. (LLM, conf=0.86)

  cites [162] sourceExperimental — accuracy not yet measured
• Partly supported

  The Discussion (page unavailable in our copy) says the PSD-95/SAPAP/Shank complex interfaces with the actin cytoskeleton at a deeper layer of the PSD. It does not say that actin is largely excluded from the PSD core or call the deeper layer the pallium. (LLM, conf=0.86)

  cites [322] sourceExperimental — accuracy not yet measured

In the paper

“However, in the LLPS study, even after long incubation the CaMKII-GluN2B CTD dissipated after removal of Ca2+ unless the kinase was allowed to autophosphorylate, albeit slowly.”

What the sources say

• Partly supported

  The 'Autophosphorylation of CaMKII is required for persistence of protein condensate' section reports that CaMKII–GluN2Bc condensates dissolved after Ca²⁺ chelation when ATP was absent or CaMKII could not autophosphorylate at T286. It does not establish the claim's 'even after long incubation' qualification. Its turbidity results instead report a return to baseline within 30 seconds after chelation for the T286A mutant, rather than slow dissipation. (LLM, conf=0.86)

  cites [323] sourceExperimental — accuracy not yet measured

In the paper

“The authors show that a mixture of the GluN2B CTD and CaMKII undergoes liquid-liquid phase separation (LLPS) when Ca2+/CaM binds to CaMKII in either the absence or presence of ATP.”

What the sources say

• Supported

  The ‘CaMKII undergoes LLPS with GluN2B carboxyl tail’ section says Ca²⁺/calmodulin triggers LLPS-mediated condensates containing CaMKII and GluN2Bc. Figure 1 describes condensate formation with ATP in panel C and without Mg²⁺-ATP in panel F. Page numbers are not provided in our copy. (LLM, conf=0.94)

  cites [323] source

In the paper

“Although inhibiting Kalirin function alone has little effect on LTP (318, 319)”

What the sources say

• ?Not checked — no full text of the sourcecites [318] source
• Supported

  The Results section reports that kalirin-deficient mice showed LTP induction in both groups, with only a modest deficit at later time points. This supports 'little effect,' though it does not mean LTP was entirely unaffected. (LLM, conf=0.88)

  cites [319] source

In the paper

“The phase separation resulting from GluN2B CTD and CaMKII mixed in the presence of Ca2+/CaM remains intact after the removal of Ca2+, but only if T286 is phosphorylated.”

What the sources say

• Supported

  The Figure 1 caption reports that CaMKII–GluN2Bc condensates formed in the presence of Ca²⁺ and persisted after EGTA was added, whereas condensates with the T286A mutant did not persist. The Discussion states that CaMKII T286 autophosphorylation is required for persistence after Ca²⁺ decreases. GluN2Bc is identified in the results as the GluN2B carboxyl tail. (LLM, conf=0.94)

  cites [323] source

In the paper

“CaMKII, in turn, phosphorylates Tiam1.”

What the sources say

• Supported

  The Results section, under “Reciprocal activation of CaMKII and Tiam1 in the complex,” explicitly reports phosphorylation of Tiam1 by CaMKII and says it increases Tiam1’s RacGEF activity. (LLM, conf=0.95)

  cites [314] source

In the paper

“In summary, perhaps the strongest case for the requirement of downstream CaMKII targets in sLTP, and likely LTP, is the actin cytoskeleton (30, 313, 321)”

What the sources say

• ?Not checked — no full text of the sourcecites [30] source
• Supported

  The ‘Regulation of actin cytoskeleton’ section (page not supplied) says functional and structural LTP require actin reorganization and identifies actin-regulating molecules downstream of CaMKII. ‘Plasticity of spine heads’ likewise says both forms require CaMKII activation and actin polymerization. The source supports the substance of the claim, though it does not rank actin as the ‘strongest’ case. (LLM, conf=0.90)

  cites [313] source
• Supported

  The sections on transient Ca²⁺ elevation and rapid structural remodeling (pages not provided) describe CaMKII signaling to molecules that reorganize the actin cytoskeleton during sLTP. The latter section states that CaMKII-activated GTPases are required for sLTP and that loss of two of them impairs LTP. The citing authors’ description of this as the “strongest case” is their assessment, not a comparison attributed to the source. (LLM, conf=0.88)

  cites [321] source

In the paper

“Postsynaptic synaptotagin1 and 7 (Syt1/Syt7) have also been reported to be required for LTP (195).”

What the sources say

• Supported

  The section “Postsynaptic Syt1,7 loss blocks LTP” reports that combined postsynaptic loss of Syt1 and Syt7 blocks LTP. The Summary clarifies that they are functionally redundant: LTP requires Syt1 or Syt7, not each individually. (LLM, conf=0.94)

  cites [195] source

In the paper

“but, as discussed, CaMKII appears not to be required for AMPAR exocytotic events (250).”

What the sources say

• Supported

  The Discussion states that the authors’ data indicate CaMKII is not required for AMPAR exocytosis. This directly supports the citing sentence’s hedged claim. (LLM, conf=0.95)

  cites [250] source

In the paper

“Fifth, a comparison of the synaptic accumulation of SEP-GluA1 by glutamate uncaging, with and without prior bleaching of surface receptors, indicates that much of the accumulation is due to lateral diffusion and not exocytotic insertion of receptors (250).”

What the sources say

• Supported

  The Results (passage 3) compare SEP-GluA1 increases with and without surface-receptor bleaching and estimate that 70–90% of the increase comes from diffusion, versus 10–30% from newly exocytosed receptors. The Discussion (passage 6) describes diffusion of preexisting surface receptors as the main source of spine recruitment. No page numbers are supplied in our copy. (LLM, conf=0.94)

  cites [250] source

In the paper

“There are rather few examples of CaMKII directly initiating exocytosis (348, 349).”

What the sources say

• ?Not checked — no full text of the sourcecites [348] source
• Partly supported

  In “The role of CaMKII in NT secretion” (p. 11), the authors report that inhibiting CaMKII strongly decreased BDNF and NT-3 release and say their results are best reconciled with a direct action on secretion. This provides an example relevant to the claim, but the paper does not establish that such examples are generally rare, and it frames the direct action as an interpretation rather than a demonstrated mechanism. (LLM, conf=0.84)

  cites [349] sourceExperimental — accuracy not yet measured

In the paper

“Third, most of these studies have tagged GluA1 on the NH2 terminus. This tagging severely impacts the trafficking of the GluA1-containing receptor (343, 344).”

What the sources say

• Supported

  The Discussion says GFP-tagged GluA1 cannot traffic to synapses without a potentiating stimulus and attributes the discrepancy with constitutive GluA1 trafficking to its N-terminal GFP tag. This supports the claim about impaired trafficking, specifically for synaptic trafficking with an N-terminal GFP tag. (LLM, conf=0.85)

  cites [343] source
• ?Not checked — no full text of the sourcecites [344] source

In the paper

“There is support that CaMKII can mediate the activity-dependent exocytosis of BDNF (349), but the linkage between TrkB activation and the synaptic accumulation of AMPARs remains unclear.”

What the sources say

• Supported

  The Results (p. 6) report that two CaMKII inhibitors strongly reduced BDNF-GFP secretion in hippocampal neurons. The Discussion (p. 9) describes activity-dependent postsynaptic BDNF secretion as critically dependent on CaMKII activation, supporting the cited claim. (LLM, conf=0.94)

  cites [349] source

In the paper

“Consistent with this result is the finding that interfering with the surface diffusion of AMPARs by receptor cross-linking markedly impairs LTP (345–347).”

What the sources say

• Supported

  The paper reports that antibody cross-linking limited AMPAR surface diffusion and that early LTP was completely abolished when the cross-linking antibody was continuously infused into acute hippocampal slices (passage 3). It also reports marked attenuation of potentiation after antibody injection in vivo (passage 7). (LLM, conf=0.94)

  cites [345] source
• Supported

  The Results section, “Cross-linking of bAP-GluA2 by NA precludes LTP in the hippocampal CA1 region” (page not provided), reports that LTP was almost completely abolished when GluA2-containing AMPARs were cross-linked, while control slices showed normal LTP. The paper also reports that this cross-linking reduces AMPAR surface diffusion. (LLM, conf=0.94)

  cites [346] source
• ?Not checked — no full text of the sourcecites [347] source

In the paper

“The first evidence that CaMKII contributes to synaptic transmission came from expression of peptide inhibitors (150, 154).”

• Also stated in: § 8. SYNAPTIC TRANSMISSION IS MAINTAINED BY THE CONSTITUTIVE ACTION OF CaMKII · boosted

  “More recent experiments with either RNAi (211) or CRISPR (121) confirmed the role of CaMKII in maintaining synaptic responses.”

What the sources say

• Partly supported

  The Results section reports that expression of the CaMKII peptide inhibitors CaMKIIN or AIP reduced synaptic AMPAR responses, supporting a contribution to baseline synaptic transmission. The copy does not establish that this was the first evidence, and its reported shRNA experiment targets CaMK4, not CaMKII; it provides no CRISPR evidence for CaMKII. (LLM, conf=0.86)

  cites [150] sourceExperimental — accuracy not yet measured
• Supported

  The Results report that cells expressing the CaMKII-inhibiting peptide CN19 had reduced EPSCs compared with nearby untransfected cells. The Discussion states that postsynaptic expression of CN19 reduces basal transmission. The cited sentence attributes evidence from peptide-inhibitor expression to this paper; it does not attribute the separate RNAi or CRISPR methods to it. (LLM, conf=0.90)

  cites [154] source
• Partly supported

  The Results (page unavailable) show that knocking down CaMKIIα in cultured hippocampal neurons decreased mEPSC amplitude, supporting a contribution to synaptic strength. The Materials and Methods identify shRNA as the knockdown method. Our copy does not establish that the first evidence came from expressed peptide inhibitors, or show a CRISPR experiment. (LLM, conf=0.88)

  cites [211] sourceExperimental — accuracy not yet measured

In the paper

“By tagging AMPARs with a pH-sensitive tag (SEP-GluA1), the insertion of AMPARs into the surface membrane can be directly visualized during LTP-inducing protocols (249, 250, 335, 337, 338, 340).”

What the sources say

• Supported

  The Introduction explains that pH-sensitive SEP tagging distinguishes fluorescent surface receptors from nonfluorescent intracellular receptors and permits monitoring receptor exocytosis during plasticity. In Results (“Small contribution of AMPAR exocytosis to synaptic potentiation during LTP”), the authors report visualizing GluR1-SEP fluorescent spots after inducing cLTP, consistent with receptor insertion into the surface membrane. Page numbers are not supplied in our copy. (LLM, conf=0.93)

  cites [249] source
• Supported

  The Results (Fig. 1) describe imaging SEP-GluA1 during LTP induced by two-photon glutamate uncaging and explain that its pH-sensitive signal selectively reports surface receptors. The Results (Fig. 2) then use prebleaching to distinguish newly exocytosed receptors from receptors already on the surface. Page numbers are not provided in the supplied passages. (LLM, conf=0.94)

  cites [250] source
• ?Not checked — no full text of the sourcecites [335] source
• Supported

  In “Real-time visualization of surface AMPAR insertion in hippocampal slice culture” (page not provided), the paper describes tagging GluR1, also called GluA1, with pH-sensitive SEP to visualize surface receptors. It reports imaging discrete insertion events and increased event frequency during potentiating chemical stimulation; the “Live imaging” methods identify the glycine/zero-magnesium and forskolin/IBMX conditions. The support is for chemical potentiation protocols, not necessarily every LTP-inducing protocol. (LLM, conf=0.89)

  cites [337] source
• Partly supported

  The Results section, “AMPA Receptors are Stably Inserted Into the Spine Plasma Membrane Via Fusion of Spine Endosomes” (page unavailable), reports direct imaging of SEP-GluR1 insertion after Bic/Gly stimulation. The paper describes Bic/Gly as increasing network activity; its separately described LTP-induction protocol was used for electrophysiology, not SEP-GluR1 imaging. It therefore supports the visualization claim but not specifically visualization during a protocol it identifies as LTP-inducing. (LLM, conf=0.84)

  cites [338] sourceExperimental — accuracy not yet measured
• Partly supported

  The Results section, “Capturing insertion events of GluRs” (page unavailable), reports direct visualization of surface GluR1 insertion using a pH-sensitive pHluorin tag. However, the paper describes LTP induction in separate whole-cell recordings, not visualization of insertion during an LTP-inducing protocol; the Discussion attributes that latter observation to another study. (LLM, conf=0.87)

  cites [340] sourceExperimental — accuracy not yet measured

In the paper

“Fourth, in most cases the AMPARs are inserted via exocytosis into the dendritic shaft, some distance from the synapse (220, 249, 250, 337).”

What the sources say

• Partly supported

  The Results (page not provided) identify probable exocytosis-related patches on dendrites away from postsynaptic densities. But they also find patches on cell bodies; the paper does not establish that insertion occurs into the dendritic shaft in most cases. (LLM, conf=0.85)

  cites [220] sourceExperimental — accuracy not yet measured
• Supported

  The Introduction (page not provided) says that during LTP, AMPAR exocytosis occurs primarily onto the dendritic shaft. The Discussion (page not provided) distinguishes this dendritic insertion from incorporation into synapses, which the paper attributes mainly to lateral movement. (LLM, conf=0.90)

  cites [249] source
• Partly supported

  The Abstract and Results report AMPAR exocytosis in dendrites near stimulated spines, including sites about 2 μm away (Fig. 3B). But the paper also observes exocytosis directly in stimulated spines and does not establish that dendritic-shaft insertion occurs in most cases. (LLM, conf=0.84)

  cites [250] sourceExperimental — accuracy not yet measured
• Supported

  The section on exocytic insertion outside synapses reports that observed dendritic insertion events occurred outside recognizable spines, and post hoc staining placed them outside PSD-95-marked postsynaptic sites. The Discussion describes extrasynaptic exocytosis in the dendritic shaft as a major pathway for AMPAR surface insertion, while acknowledging that direct insertion into spines cannot be excluded. (LLM, conf=0.88)

  cites [337] source

In the paper

“In addition, pharmacological inhibition of CaMKII activity has generally been reported not to affect synaptic responses (198, 200–203, 357, 358).”

What the sources say

• ?Not checked — no full text of the sourcecites [198] source
• Supported

  The Figure 7 caption reports no decline in control or potentiated synaptic responses after the CaMKII inhibitor was applied. This paper therefore provides an example supporting the cited statement, although its finding concerns responses after LTP induction; the Discussion separately reports that the inhibitor blocks LTP induction. (LLM, conf=0.86)

  cites [200] source
• ?Not checked — no full text of the sourcecites [201] source
• Supported

  The Discussion (page unavailable) says that CaMKII inhibitors, including tatCN21, generally do not cause rundown of synaptic transmission, while identifying exceptions. The Results (page unavailable) report that tatCN21 did not cause rundown of basal excitatory postsynaptic responses. This supports the claim as a statement about basal synaptic responses, not LTP induction. (LLM, conf=0.88)

  cites [202] source
• ?Not checked — no full text of the sourcecites [203] source
• ?Not checked — no full text of the sourcecites [357] source

In the paper

“However, as mentioned above, most genetic deletion studies have failed to observe effects on baseline synaptic transmission (Refs. 43, 204, 205, but see Ref. 206).”

What the sources say

• ?Not checked — no full text of the sourcecites [43] source
• ?Not checked — no full text of the sourcecites [204] source
• Supported

  In the section “LTP is impaired in mice with adult deletion of αCaMKII,” synaptic transmission was unaffected after adult deletion: fiber-volley amplitude and postsynaptic fEPSP slope did not differ significantly from controls. The authors say the fEPSP finding is consistent with previous Camk2a knockout studies, supporting the cited statement in the context of αCaMKII deletion studies. (LLM, conf=0.88)

  cites [205] source

In the paper

“Second, immunogold labeling studies find that AMPARs within spines, e.g., in spine vesicles, are scarce, except for large spines containing a spine apparatus (290).”

What the sources say

• Partly supported

  The Figure 9 caption (page unavailable in our copy) reports AMPAR labeling in the spine apparatus and extrasynaptic membrane most often at large spines, while small immunonegative spines usually lack labeling at those locations. It does not establish that AMPARs in spine vesicles are scarce, and it also identifies AMPARs at synaptic junctions within spines; the claim is therefore too broad about receptors ‘within spines.’ (LLM, conf=0.86)

  cites [290] sourceExperimental — accuracy not yet measured

In the paper

“The slow time course is not due to slow access of the inhibitors to the site of action, since paAIP2, a photoactivatable peptide inhibitor (203) that acts within seconds of light exposure, has the same slow time course (361).”

What the sources say

• ?Not checked — no full text of the sourcecites [203] source
• Partly supported

  In Results (Blocking CaMKII inhibits synaptic transmission; page not specified), blue light rapidly exposes paAIP2, yet inhibition develops slowly, with kinetics similar to myr-CN27 and myr-AIP. The authors say this *suggests* slow access is not the cause; the citing sentence states that conclusion as fact. The copy describes rapid activation but does not specify that paAIP2 acts within seconds. (LLM, conf=0.85)

  cites [361] sourceExperimental — accuracy not yet measured

In the paper

“In addition, LTP-inducing stimuli result in the persistent accumulation of CaMKII in the PSD (FIGURE 6, A–C) (217).”

What the sources say

• Supported

  The Results section (‘Persistent increase in CaMKII density at the PSD’) reports that PSD-associated CaMKIIα increased after chemical LTP induction and remained elevated for at least one hour. The Discussion likewise describes a persistent increase in CaMKIIα levels in the PSD. (LLM, conf=0.94)

  cites [217] source

In the paper

“It is well established that the Ca2+/CaM-dependent binding and phase separation (LLPS) of CaMKII to GluN2B can occur in the absence of enzymatic activity (169, 323, 376).”

What the sources say

• ?Not checked — no full text of the sourcecites [169] source
• Supported

  The Discussion (page not specified in our copy) says that Ca²⁺ triggers the initial formation of CaMKII condensates and that this formation is independent of kinase activity. The Introduction describes Ca²⁺/calmodulin-dependent binding of CaMKII to the GluN2B tail. The paper distinguishes initial formation from persistence after Ca²⁺ falls, which does require T286 autophosphorylation; the claim concerns what can occur, not whether it persists. (LLM, conf=0.93)

  cites [323] source
• Partly supported

  The Introduction states that Ca2+/CaM stimulation induces CaMKII–GluN2B binding, and the Discussion states that kinase activity is not required for that binding (page numbers unavailable in our copy). The paper does not address phase separation or LLPS, so it does not support that part of the claim. (LLM, conf=0.90)

  cites [376] sourceExperimental — accuracy not yet measured

In the paper

“However, numerous previous studies have failed to reverse established LTP (Refs. 198, 200–203, but see Ref. 357).”

What the sources say

• ?Not checked — no full text of the sourcecites [198] source
• Supported

  In the Results, the authors report that applying a CaMKII inhibitor after LTP induction did not produce significantly more decay than a control peptide. The discussion (‘Is a kinase involved in LTP maintenance?’) likewise states that they observed no decay relative to controls. This paper supports the cited sentence as one example of a study that failed to reverse established LTP; the other cited studies account for ‘numerous.’ (LLM, conf=0.90)

  cites [200] source
• ?Not checked — no full text of the sourcecites [201] source
• Supported

  The Discussion reports that previous LTP-maintenance studies found no effect from CaMKII-derived inhibitors and cites three such studies. The Results also report that applying tatCN21 after LTP induction did not affect its maintenance. The paper acknowledges contrasting reports, but the claim says that numerous studies failed to reverse LTP, not that all did. (LLM, conf=0.89)

  cites [202] source
• ?Not checked — no full text of the sourcecites [203] source

In the paper

“Deleting NMDARs in individual neurons or replacing GluN1 with a pore-dead mutant in utero, thus preventing NMDAR-dependent LTP, should prevent generation of constitutive CaMKII activity. This indeed is the case (361).”

What the sources say

• Supported

  The Results section, “The origin of the constitutive action of CaMKII,” reports that CaMKII inhibitors had no depressive effect in neurons lacking GluN1 or expressing pore-dead GluN1. The Discussion interprets both experiments as showing that these cells lack constitutive CaMKII activity, consistent with the claim. (LLM, conf=0.91)

  cites [361] source

In the paper

“After the transient inhibition of CaMKII with the photoactivatable inhibitor (paAIP2) (203), there is no recovery (361).”

What the sources say

• ?Not checked — no full text of the sourcecites [203] source
• Supported

  The Results section, “The origin of the constitutive action of CaMKII” (Figure 4B), reports no recovery after blue-light activation of paAIP2 ended, despite the inhibitor’s action reversing in about 40 seconds. The Figure 4 caption likewise says AMPAR synaptic transmission remained depressed after light exposure. (LLM, conf=0.94)

  cites [361] source

In the paper

“Thus, if one were to transiently inhibit CaMKII and reduce baseline transmission, the magnitude of LTP should be larger. Again, experiments confirm this prediction (361).”

What the sources say

• Supported

  The Results section, Figure 5 caption, reports that transient paAIP2 activation reduced baseline AMPAR responses and enhanced subsequent LTP compared with a neighboring control cell. The Discussion likewise states that LTP was approximately twice as large after transient CaMKII blockade. (LLM, conf=0.95)

  cites [361] source

In the paper

“Chelating postsynaptic Ca2+ has no effect on this CaMKII constitutive activity (361).”

What the sources say

• Supported

  In Results, “The origin of the constitutive action of CaMKII” (page not provided), the authors report that loading cells with the Ca2+ chelator BAPTA did not prevent the depressive effect of a CaMKII inhibitor. This supports the claim about the constitutive activity’s resistance to postsynaptic Ca2+ chelation, as assessed by its effect on synaptic transmission. (LLM, conf=0.90)

  cites [361] source

In the paper

“Furthermore, two-photon glutamate uncaging experiments on single spines indicate that synapses that already express functional AMPARs can be further enhanced by inducing LTP (246–248).”

What the sources say

• Supported

  The Methods (page unavailable) report measuring uncaging-evoked currents at individual spines at −70 mV. The results passage (indexed as Abstract; page unavailable) reports that the single-spine LTP protocol produced a lasting 99 ± 17% increase in those currents, supporting enhancement of spines with a pre-existing functional response. The paper also relates these currents and spine changes to AMPA receptors. (LLM, conf=0.88)

  cites [246] source
• Supported

  The supplied passages (labelled Abstract; no page given) describe two-photon glutamate uncaging at single spines and AMPA-receptor-mediated currents measured before and after stimulation. They report that currents already detectable at the same locations were potentiated after stimulation, with potentiation persisting for at least 30 minutes in some small spines. (LLM, conf=0.93)

  cites [247] source
• Supported

  The Electrophysiology section (p. 28) reports baseline glutamate-uncaging currents recorded at −70 mV in the presence of MgCl2, consistent with functional AMPAR responses before LTP induction. The LTP results (p. 10) report that two-photon uncaging subsequently potentiated currents in the targeted spines by about 75%. (LLM, conf=0.89)

  cites [248] source

In the paper

“T286 phosphorylation plays a role in the maintenance of liquid-liquid phase separation (LLPS) (323).”

What the sources say

• Supported

  The Discussion states that CaMKII T286 autophosphorylation is required for protein condensates to persist after Ca²⁺ decreases, while their initial formation is independent of kinase activity. The segregation results likewise say that persistent segregation after Ca²⁺ recedes requires T286 phosphorylation. This supports a role in maintaining LLPS, not in initiating it. (LLM, conf=0.94)

  cites [323] source

In the paper

“Indeed, based on quantitative estimates of CaMKII distribution in spines (380), it was proposed that the active CaMKII bound to the NMDAR was small (∼1%).”

What the sources say

• Partly supported

  The Discussion ('Fraction of tightly bound CaMKII') estimates that about 1% of spine CaMKII is bound to NR2B under basal conditions, but distinguishes that from the fraction activated by binding, which it estimates at about 0.2%. The claim captures that the active pool is small but assigns it the percentage for the bound pool. (LLM, conf=0.90)

  cites [380] sourceExperimental — accuracy not yet measured

In the paper

“Recent biochemical evidence suggests that CaMKII may also satisfy this second requirement, i.e., for subunit exchange (28, 406–408).”

What the sources say

• Supported

  The section “Activation-Triggered Subunit Exchange in CaMKII” reports that activation leads to mixing of subunits from different CaMKII holoenzymes, observed by colocalization of differently labeled subunits. This supports the hedged claim that evidence suggests CaMKII can undergo subunit exchange. (LLM, conf=0.92)

  cites [28] source
• Supported

  The Results and discussion section reports significant subunit exchange between CaMKII holoenzymes after Ca2+/CaM activation. The Concluding remarks specify that this was demonstrated in vitro for human CaMKIIα, supporting the citing sentence’s hedged claim of biochemical evidence for subunit exchange. (LLM, conf=0.94)

  cites [406] source
• Supported

  The Results and discussion section reports a biochemical FRET experiment showing that CaMKII-β exchanges subunits after stimulation with Ca2+/CaM and ATP. The Introduction and concluding remarks also describe activation-triggered subunit exchange in CaMKII. This supports the citing sentence’s hedged claim that CaMKII may satisfy a requirement for subunit exchange. (LLM, conf=0.94)

  cites [407] source
• Supported

  The Introduction describes experiments in which activated CaMKII samples showed co-localization or increased FRET, evidence consistent with subunit exchange between holoenzymes. The Conclusions also discuss activation-dependent subunit exchange, while noting that its precise role awaits further study. This supports the citing sentence’s cautious wording. (LLM, conf=0.91)

  cites [408] source

In the paper

“In particular, CaMKII is known to turn over rapidly, with a half-life of 2–4 days (403–405).”

What the sources say

• Partly supported

  The Results table reports a 3.81-day half-life for CaMKIIβ2 in cultured cortical neurons, within the claimed range. It also reports 5.03 days for CaMKIIγ, so the 2–4-day range is not supported for CaMKII generally. (LLM, conf=0.86)

  cites [403] sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [404] source
• Partly supported

  The Results section on protein localization and function reports that CaMK2 subunits have different half-lives. It does not give a 2–4-day range for CaMKII or establish that all its subunits turn over rapidly; the numerical half-lives quoted elsewhere in the Results concern other proteins. No page number is available in our copy. (LLM, conf=0.78)

  cites [405] sourceExperimental — accuracy not yet measured

In the paper

“In a series of classic studies it was found that LTD is induced by the activation of a high-affinity Ca2+ cascade resulting in the activation of PP1 (395–397).”

What the sources say

• ?Not checked — no full text of the sourcecites [395] source
• ?Could not verify

  No passages from our copy were returned for searches on LTD, calcium signaling, and PP1 activation, including broader searches for “LTD” and “synaptic.” The available evidence therefore does not establish whether this paper supports the claim. (LLM, conf=0.70)

  cites [396] sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [397] source

In the paper

“Despite these findings suggesting a critical role of T286, it should be noted that Chang et al. (94) reported that LTP can be expressed in the absence of T286 phosphorylation.”

What the sources say

• Supported

  The Introduction states that increasing stimulation frequency allowed the T286A mutant to reach a level sufficient to induce structural and electrophysiological LTP. The Results also report rescue of structural plasticity in T286A mice with higher-frequency stimulation. The claim says LTP can occur without T286 phosphorylation, not that it occurs under standard induction conditions. Page numbers were not supplied. (LLM, conf=0.93)

  cites [94] source