Brief Communication | Published:

Protein kinase Mζ is necessary and sufficient for LTP maintenance

Nature Neuroscience volume 5, pages 295296 (2002) | Download Citation



Long-term potentiation (LTP), a persistent synaptic enhancement thought to be a substrate for memory1, can be divided into two phases: induction, triggering potentiation, and maintenance, sustaining it over time1,2. Many postsynaptic events are implicated in induction, including N-methyl-D-aspartate receptor (NMDAR) activation, calcium increases and stimulation of several protein kinases1; in contrast, the mechanism maintaining LTP is not yet characterized1. Here we show the constitutively active form of an atypical protein kinase C (PKC) isozyme, protein kinase M zeta (PKMζ), is necessary and sufficient for LTP maintenance.

Access optionsAccess options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.


  1. 1.

    & Science 285, 1870–1874 (1999).

  2. 2.

    , & Nature 335, 820–824 (1988).

  3. 3.

    & Annu. Rev. Neurosci. 10, 459–476 (1987).

  4. 4.

    FASEB J. 9, 484–496 (1995).

  5. 5.

    , & J. Neurosci. 17, 5357–5365 (1997).

  6. 6.

    , , , & Brain Res. 534, 201–208 (1990).

  7. 7.

    et al. Proc. Natl. Acad. Sci. USA 90, 8342–8346 (1993).

  8. 8.

    et al. Proc. Natl. Acad. Sci. USA 92, 11175–11179 (1995).

  9. 9.

    , , , & Mol. Cell. Biol. 19, 2180–2188 (1999).

  10. 10.

    et al. J. Cell. Biochem. 49, 239–250 (1992).

  11. 11.

    & J. Biol. Chem. 271, 15045–15053 (1996).

  12. 12.

    & J. Neurophysiol. 78, 2682–2692 (1997).

  13. 13.

    & Eur. J. Neurosci. 12, 4055–4062 (2000).

  14. 14.

    et al. J. Neurochem. 74, 807–817 (2000).

  15. 15.

    et al. Nat. Neurosci. 5, 316–324 (2002).

Download references


Supported by New York City Council Speaker's Fund for Biomedical Research (D.S.F.L.) and NIMH (L.S.B. and T.C.S.).

Author information


  1. Departments of Physiology, Pharmacology, and Neurology, SUNY Downstate Medical Center, 450 Clarkson Ave., Brooklyn, New York, 11203, USA

    • Douglas S.F. Ling
    • , Larry S. Benardo
    • , Peter A. Serrano
    • , Nancy Blace
    • , Matthew T. Kelly
    • , John F. Crary
    •  & Todd C. Sacktor


  1. Search for Douglas S.F. Ling in:

  2. Search for Larry S. Benardo in:

  3. Search for Peter A. Serrano in:

  4. Search for Nancy Blace in:

  5. Search for Matthew T. Kelly in:

  6. Search for John F. Crary in:

  7. Search for Todd C. Sacktor in:

Competing interests

The authors declare no competing financial interests.

Corresponding author

Correspondence to Todd C. Sacktor.

Supplementary information

PDF files

  1. 1.

    Supplementary Figure 1.

    Silver stain of baculovirus/Sf9-expressed PKMζ used in whole-cell recording pipette. Positions of molecular weight standards in kDa are shown on the left.

  2. 2.

    Supplementary Figure 2.

    No change in input resistance during potentiation by postsynaptic introduction of PKMζ. Upper graph, time course of the individual experiment shown in Fig. 1a and b (each point represents mean ± SEM of 4 consecutive EPSCs). Lower graph, no change in input resistance.

  3. 3.

    Supplementary Figure 3.

    Postsynaptic introduction of heat-inactivated PKMζ (0 pmol/min/ml phosphotransferase activity) had no effect on evoked EPSCs.

  4. 4.

    Supplementary Figure 4.

    Evoked EPSCs potentiated by PKMζ were suppressed by 93.0 ± 1.5% with exposure to 10 µM CNQX, a non-NMDA glutamate receptor antagonist (n = 3), indicating that the enhancement is predominantly of AMPAR responses.

  5. 5.

    Supplementary Figure 5.

    Potentiation of AMPAR responses by PKMζ was not secondary to activation of NMDARs. (a) Enhancement of EPSCs with intracellular introduction of PKMζ in cells voltage-clamped at -75 mV in the presence of 10 µM CPP. (b) Pooled data of the increase in evoked EPSCs by PKMζ in 10 µM CPP (mean ± SEM; n = 6 experiments). Unpaired t-test of PKMζ-mediated EPSC enhancement 12 min after the onset of whole-cell access showed no significant difference between experiments in the presence or absence of CPP.

  6. 6.

    Supplementary Figure 6.

    PKMζ potentiated spontaneous EPSCs. (a) Examples of spontaneous events recorded 1 and 10 min after initiation of whole-cell recording with 1.4 nM PKMζ in the pipette. The potentiation of sEPSCs by PKMζ was reversed by 1 µM chelerythrine. (b) Mean ± SEM of sEPSC amplitudes (1 min recording, 347 events; 10 min, 379 events; reversal by chelerythrine, 172 events; p < 0.0001, unpaired t-test between 1 and 10 min and no significant difference between 1 min and chelerythrine reversal). PKMζ did not change the frequency of sEPSCs significantly (98.5% ± 3.7 of initial values, n = 4).

  7. 7.

    Supplementary Figure 7.

    Whole-cell LTP by 4 tetanic trains paired with postsynaptic depolarization to -40 mV. (a) Representative EPSCs pretetanus (recorded at 10 min of establishing whole-cell access, immediately prior to tetanic stimulation) and posttetanus (30 min). (b) Time course of the individual experiment shown in (a). Each point represents mean ± SEM of 4 consecutive EPSCs. Tetanus is at arrow.

  8. 8.

    Supplementary Figure 8.

    The synaptic potentiation produced by PKMζ occludes LTP. (a) Upper traces, EPSCs at 1 and 10 min after the initiation of whole-cell recording with PKMζ, and posttetanus, at 30 min. No additional potentiation was observed. Lower traces, simultaneously recorded field potentials showed LTP. (b) Time course of individual experiment shown in (a). In whole-cell recordings, tetanization resulted only in PTP. Inset, time course of LTP in simultaneous field recordings. Each point represents mean ± SEM of 4 consecutive EPSCs. (c) Pooled data showing simultaneous LTP in field recordings from experiments presented in Fig. 2b (mean ± SEM; n = 5 experiments).

  9. 9.

    Supplementary Figure 9.

    Postsynaptic exposure to the dominant negative inhibitor PKMζ-K281W (20 nM) blocked LTP. (a) Silver stain of baculovirus/Sf9-expressed PKMζ-K281W used in the whole-cell recording pipette. (b) Evoked EPSCs recorded 1 and 10 min after introduction of PKMζ-K281W with cells held at -40 mV showed no obvious effect on synaptic responses mediated by AMPA and NMDARs. PKMζ-K281W also had no effect on passive membrane properties (not shown). (c) Upper traces, whole-cell EPSCs pre- and posttetanization, showing no persistent potentiation after exposure to PKMζ-K281W. Lower traces, simultaneously recorded field potentials showed LTP. (d) Time course of the individual experiment shown in (c). Each point represents mean ± SEM of 4 consecutive EPSCs. Tetanization resulted only in PTP. Inset, time course of simultaneous field recording showed LTP. (e) Pooled data showing simultaneous LTP in field recordings from experiments presented in Fig. 2c (mean ± SEM; n = 5 experiments). Whole-cell experiments immediately before and after those with PKMζ-K281W showed normal LTP.

  10. 10.

    Supplementary Methods.

Image files

  1. 1.

    Supplementary Figure 10.

    Persistent phosphorylation by PKMζ is necessary for LTP maintenance. (a) Staurosporine (100 nM in 0.001% DMSO) applied prior to the tetanic stimulation blocked the induction of LTP. Pooled field recording data are mean ± SEM of 4 experiments. (b) Representative field potential traces showing 1 min pretetanus, 1 hr posttetanus (immediately prior to drug application), and 5 hr posttetanus. Above, staurosporine (100 nM in 0.001% DMSO); center, chelerythrine (1 µM in 0.001% DMSO); below, myristoylated ζ-pseudosubstrate peptide (1 µM). For data presented in Fig. 3, one-way ANOVA showed significant differences among DMSO alone, staurosporine, chelerythrine, and ζ-pseudosubstrate peptide, 4 hr post-application [F(3, 14) = 10.11, p = 0.0008]. Post hoc unpaired t-test showed significant differences for both chelerythrine and ζ-pseudosubstrate peptide compared to both DMSO and staurosporine (p < 0.01). No significant differences were observed between DMSO and staurosporine, and between chelerythrine and ζ-pseudosubstrate peptide. Staurosporine effectively inhibited conventional PKCα, novel PKCε, and CaMKII at low nanomolar concentrations, but was ineffective on PKMζ up to 100 nM (n = 3; the SEM, not shown for clarity in Fig. 3, were less than 30% of the means for all data points; p < 0.02, Wilcoxon's independent samples test, PKMζ compared to PKCα, PKCε and CaMKII). Chelerythrine inhibited PKMζ at lower concentrations than that required to inhibit PKCα, PKCε, and CaMKII (n = 3; p < 0.02). The myristoylated ζ-pseudosubstrate peptide selectively inhibited PKMζ relative to CaMKII (n = 3; p < 0.02).

About this article

Publication history





Further reading