doi: 10.1371/journal.pbio.2006838.
eCollection 2018 Dec.
An opposing function of paralogs in balancing developmental synapse maturation
Affiliations
- PMID: 30586380
- PMCID: PMC6324823
- DOI: 10.1371/journal.pbio.2006838
Item in Clipboard
An opposing function of paralogs in balancing developmental synapse maturation
PLoS Biol.
.
Abstract
The disc-large (DLG)-membrane-associated guanylate kinase (MAGUK) family of proteins forms a central signaling hub of the glutamate receptor complex. Among this family, some proteins regulate developmental maturation of glutamatergic synapses, a process vulnerable to aberrations, which may lead to neurodevelopmental disorders. As is typical for paralogs, the DLG-MAGUK proteins postsynaptic density (PSD)-95 and PSD-93 share similar functional domains and were previously thought to regulate glutamatergic synapses similarly. Here, we show that they play opposing roles in glutamatergic synapse maturation. Specifically, PSD-95 promoted, whereas PSD-93 inhibited maturation of immature α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid-type glutamate receptor (AMPAR)-silent synapses in mouse cortex during development. Furthermore, through experience-dependent regulation of its protein levels, PSD-93 directly inhibited PSD-95's promoting effect on silent synapse maturation in the visual cortex. The concerted function of these two paralogs governed the critical period of juvenile ocular dominance plasticity (jODP), and fine-tuned visual perception during development. In contrast to the silent synapse-based mechanism of adjusting visual perception, visual acuity improved by different mechanisms. Thus, by controlling the pace of silent synapse maturation, the opposing but properly balanced actions of PSD-93 and PSD-95 are essential for fine-tuning cortical networks for receptive field integration during developmental critical periods, and imply aberrations in either direction of this process as potential causes for neurodevelopmental disorders.
Conflict of interest statement
The authors have declared that no competing interests exist.
Figures
(A–F, I, J, L, M, O–Q) Sample traces of pyramidal neuron EPSCs from different brain areas using minimal stimulation of afferents of mice with different genotypes and ages. Sample traces (insets) and analysis of the peak values of AMPA receptor EPSCs, recorded at Vh = −60 mV (downward deflection), or composite glutamate receptor EPSCs, recorded at Vh = +40 mV (upward deflection), of successes (black) and failures (gray) of individual EPSCs are depicted. Scale bar: 20 ms and 25 pA. (A-F) Sample traces from V1 layer 2/3 pyramidal neurons with L4 afferents of P11/P28 WT (A/D), PSD-95 KO (B/E), and PSD-93 KO (C/F) mice. (G-H) Summary graphs of the fraction of silent synapses of WT (gray), PSD-95 KO (blue), and PSD-93 (red) mice at different developmental time points. Dots represent values of single neurons; horizontal bars are genotype averages. Recording scheme of layer 2/3 pyramidal neuron of V1 in coronal brain slices is depicted in inset, with patch pipette and bipolar stimulating electrode. *p < 0.05, **p < 0.01. See also S1 Fig. Underlying data for this figure can be found in S1 Data. (I, J, L, M) Sample traces of mPFC layer 2/3 pyramidal neuron with white matter afferents of P30/P15 WT (I/L), P30 PSD-95 KO (J), and P15 PSD-93 KO (M). (K, N) Summary graphs of the fraction of silent synapses of WT (gray), PSD-95 KO (blue), and PSD-93 KO (red) mice at different developmental time points. Recording scheme of layer 2/3 pyramidal neuron of the mPFC PrL in coronal brain slices is depicted in inset, with patch pipette and bipolar stimulating electrode. **p < 0.01, ***p < 0.001. Underlying data for this figure can be found in S1 Data. (O-Q) Sample traces of hippocampal CA1 pyramidal neuron with Schaffer collateral afferents of P20 WT (O), PSD-95 KD (P), and PSD-93 KO (Q). (R) Summary graphs of the fraction of silent synapses of WT (gray), PSD-95 KD (blue), and PSD-93 KO (red) mice at P20. Recording scheme of CA1 pyramidal neuron of V1 in coronal brain slices is depicted in inset, with patch pipette and bipolar stimulating electrode. **p < 0.01. Underlying data for this figure can be found in S1 Data. AMPA, α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid; CA1, Cornu Ammonis 1; EPSC, excitatory postsynaptic current; KD, knock-down; KO, knock-out; L4, layer 4; mPFC, medial prefrontal cortex; P, postnatal day; PrL, prelimbic cortex; PSD, postsynaptic density; Vh, holding potential; V1, primary visual cortex; WT, wild-type.
(A-D, F, G, I, J, L) Sample traces of EPSCs using minimal stimulation of L4 afferents of DR WT mice at P11/P28 (A/B), DR PSD-93 KO (93KO) mice at P11/P28 (C/D), P28 WT mice with AAV-shLC/AAV-sh93 (F/G), PSD-93/95 dKO mice at P11/P28 (I/J), or DR at P28. Scale bar: 20 ms and 25 pA. (E, H, K, M) Summary graphs of the fraction of silent synapses of indicated genotypes/manipulation (inset) at indicated developmental time points. Dots represent values of single neurons. (K, M) Values of dKO (orange) and PSD-93 KO with sh95 are presented with white- or black-centered dots, respectively. Values of dKO and PSD-93 with sh95 are pooled for average value (horizontal bar). For comparison, average values from Fig 1 are plotted as thin horizontal bars without dots of individual values (H, K). Statistical difference between individual groups are presented in graphs and between genotypes in inset. Data are displayed as in Fig 1. **p < 0.01. Underlying data for this figure can be found in S1 Data. AAV, adeno-associated viral vector; dKO, double KO; DLG, disc-large; DR, dark-reared; EPSC, excitatory postsynaptic current; KO, knock-out; L4, layer 4; MAGUK, membrane-associated guanylate kinase; NR, normal-reared; P, postnatal day; PSD, postsynaptic density; shLC, short hairpin RNA against luciferase; sh93, short hairpin RNA against PSD-93; sh95, short hairpin RNA against PSD-95; WT, wild-type.
(A-H) Optically imaged activity and retinotopic maps in V1 of WT (A–D) and PSD-93 KO mice (E-H) during mid CP (−4), color-coded maps of retinotopy (bottom rows), histogram of OD scores (top right of panels), and color-coded OD maps (bottom right, including average ODI) are illustrated. In control mice of both genotypes (A and E), activity patches evoked by stimulation of the contralateral (contra) eye were always darker than those evoked by the ipsilateral eye (ipsi) stimulation, the average ODI was positive, and warm colors prevailed in the two-dimensional OD maps, indicating contralateral dominance. After 4 d of MD (MD eye illustrated as black spot), there was an OD shift towards the open eye in WT mice during mid (B) and late CP (C), and beyond the CP after dark rearing (D), whereas PSD-93 KO mice only showed an OD shift during mid CP (F), which was already absent in late CP (G) and not rescued after dark rearing (H). After MD in WT mice, the ODI histogram shifted leftwards (blue arrows), the ODI decreased, and colder colors prevailed in the OD maps (negative ODI values). In contrast, in PSD-93 KO mice, OD plasticity was absent beyond P28 (compare F and G), and the deprived, contra eye continued to dominate V1 (G). (I, J) Summary graph of ODI (I) and V1 activation (J). ○, open eyes; ●, deprived eyes. ODIs (I) and V1 activation (J) before (oo) and after (●o) 4 d MD in PSD-93 KO (red) and WT mice (light gray). Values of DR mice are illustrated in darker colors. Symbols in I represent ODI values of individuals; means are marked by horizontal lines. (J) V1 activation elicited by stimulation of the contra (co) or ipsi (ip) eye. Note that juvenile OD plasticity persisted in WT mice after the CP (beyond CP) after dark rearing (WT DR), but not in NR WT mice, whereas it was absent in both late CP (≥P28) and DR (≥P28) PSD-93 KO mice. Furthermore, all OD shifts after 4 d of MD were primarily mediated by reductions of deprived eye responses in V1 (J). *p < 0.05, **p < 0.01, ***p < 0.001. Underlying data for this figure can be found in S1 Data. ant, anterior; co, contralateral; contra, contralateral; CP, critical period; DR, dark-reared; ip, ipsilateral; ipsi, ipsilateral; KO, knock-out; MD, monocular deprivation; med, medial; NR, normal-reared; ODI, ocular dominance index; ODP, ocular dominance plasticity; P, postnatal day; PSD, postsynaptic density; V1, primary visual cortex; WT, wild-type.
(A-D) Optically imaged activity maps in V1 of WT mice with shLC (A, B) and with sh93 (C, D) during late CP (≥P28) before (A, C) and after 4 d of MD (B, D). Data displayed as in Fig 3. (E, F) Summary graph of ODI (E) and V1 activation (F). ○, open eyes; ●, deprived eyes. ODIs (E) and V1 activation (F) before (oo) and after (●o) 4 d MD in WT mice with shLC (green) and with sh93 (red). Data displayed as in Fig 3. *p < 0.05, **p < 0.01, ***p < 0.001. Underlying data for this figure can be found in S1 Data. ant, anterior; co, contralateral; contra, contralateral; CP, critical period; KO, knock-out; ip, ipsilateral; ipsi, ipsilateral; MD, monocular deprivation; ODI, ocular dominance index; ODP, ocular dominance plasticity; P, postnatal day; PSD, postsynaptic density; shLC, short hairpin RNA against luciferase; sh93, short hairpin RNA against PSD-93; V1, primary visual cortex; WT, wild-type.
(A, B) uEPSC (A) and success rate (B) of EPSCs evoked with minimal stimulation (recordings from Fig 1) in V1 slices of WT (black) and PSD-93 KO mice (red). Vertical dashed line illustrates time point of eye opening. Two-factor ANOVA for difference between genotypes and one-factor ANOVA with Tukey to test for differences between individual groups; *p < 0.05, **p < 0.01. Underlying data for this figure can be found in S1 Data. (C, D) Potency (C) and success rate (D) of EPSCs evoked with minimal stimulation from P28 WT mice, expressing shLC (green), sh93 (red), or PSD-93 KO with sh95 (orange) (recordings from Fig 2). Underlying data for this figure can be found in S1 Data. (E-G) mEPSC recordings with sample traces (E) of WT (top) or PSD-93 KO (bottom) and cumulative probability graph of mEPSC amplitude (F) and IEI (G) for WT (black) and PSD-93 KO (red) mice at P23 (P20–P26). Average mEPSC amplitude (F) and IEI (G) is illustrated in the inset. Number of layer 2/3 pyramidal neurons indicated in the foot of bar. KS test for equal distribution or t test for difference of means, *p < 0.05, **p < 0.01. Scale bar: 10 pA, 50 ms. Underlying data for this figure can be found in S1 Data. (H, I) PPR with different interstimulation intervals (Δt, 50 ms, 100 ms, 200 ms), with sample traces (H) for 50 ms and 100 ms for WT (H, top) and PSD-93 KO (H, bottom) at P24. Summary graph (I) with values for single neurons (dots) and average (horizontal line) for WT (black) and PSD-93 KO (red) mice at P25. Scale bar: 50 pA, 50 ms. Underlying data for this figure can be found in S1 Data. (J-L) Potency (K) and success rate (L) of NMDA receptor EPSCs (Vh = +40 mV) evoked with minimal stimulation from P24 WT or PSD-93 KO mice. To measure the potency, traces for successes for each cell were pooled and averaged, and the peak amplitude of the averaged trace measured and subtracted from the amplitude at the same time point of the averaged failure traces (J). Scale bar: 50 pA, 200 ms. Underlying data for this figure can be found in S1 Data. (M, N) mEPSC recordings plotted as cumulative probability graphs of mEPSC amplitude (M) and IEI (N) for WT (gray) or PSD-93/95 dKO (orange) mice at P23 (P20–P26). Average mEPSC amplitude (M) and IEI (N) are illustrated in the inset. Number of layer 2/3 pyramidal neurons indicated in the foot of bar. Data for WT are the same as in panels F and G. KS test for equal distribution or t test for difference of means, **p < 0.01. Underlying data for this figure can be found in S1 Data. dKO, double KO; EPSC, excitatory postsynaptic current; IEI, inter-event interval; KO, knock-out; KS, Kolmogorov-Smirnov; m, miniature; NMDA, N-methyl-D-aspartate; P, postnatal day; PPR, paired pulse ratio; PSD, postsynaptic density; shLC, short hairpin RNA against luciferase; sh93, short hairpin RNA against PSD-93; sh95, short hairpin RNA against PSD-95; uEPSC, unitary EPSC; Vh, holding potential; V1, primary visual cortex; WT, wild-type.
(A, B) Immunofluorescence labeling of semi-thin sections with overview (A) and enlargement of boxed area (B) of mouse (P40) visual cortex for PSD-95 (blue), PSD-93 (green), and Munc13-1 (red). Upper panels illustrate fluorescence for single channels and lower panels for two channels, with PSD-95/PSD-93 (left), PSD-93/Munc13-1 (middle), and PSD-95/Munc13-1 (right). White arrowheads depict puncta with only one paralog colocalized with Munc13-1, and white arrows depict puncta of paralogs not colocalized with Munc13-1. Scale bar: 5 μm. (C-E) Quantification of synaptic colocalization (C) of PSD-95 in PSD-93/Munc13-1 positive puncta (blue) and of PSD-93 in PSD-95/Munc13-1 positive puncta (green). Synapse density, defined as puncta with PSD-93/Munc13-1 colocalization (D) in WT and PSD-95 KO mice. Similarly defined PSD-95 synapse density in WT and PSD-93 KO mice. Number of animals in foot of bar. See also S5 Fig. Underlying data for this figure can be found in S1 Data. (F) Protein levels in crude synaptosomes of the visual cortex in P28 WT (gray) or PSD-93 KO mice (red). Values of samples from each mouse were normalized to the average value of WT mice for each indicated protein on a western blot and presented as the relative amount compared with the WT band intensity. Sample bands for each protein and genotype are illustrated on the right. t test, *p < 0.05. (G-H) Developmental profile of synaptic proteins PSD-95 (blue), PSD-93 (red), SAP97 (gray), and SAP102 (green) from crude synaptosomal fractions from V1 of WT mice. Sample blots (G) are illustrated for the indicated proteins, and quantified protein levels, normalized to the adult levels at P90, are plotted against the postnatal day (P, H). n = 4–5 (mice). Values for PSD-95 are from a previous report [8]. Underlying data for this figure can be found in S1 Data. (I) Comparison of synaptic protein levels from crude synaptosomal fractions from DR (black) and NR (light gray) mice at P28. Protein levels were assessed as described in panel F. Number of mice is indicated in the foot of the bar. t test, *p < 0.05. Underlying data for this figure can be found in S1 Data. CP, critical period; DR, dark-reared; EO, eye opening; Glu, glutamate receptor subunit; KO, knock-out; Munc13-1, Mammalian uncoordinated 13–1; NR, normal-reared; P, postnatal day; PSD, postsynaptic density; SAP, synapse-associated protein; V1, primary visual cortex; WT, wild-type.
(A, B, D, E, G, H, J) Sample traces of V1 layer 2/3 pyramidal neuron EPSCs with minimal stimulation of P28 WT mice with AAV-GFP (A), with AAV-PSD-93α2 (B), with AAV-sh95 + PSD-93α2 (J), P11 WT mice with AAV-GFP (D), with AAV-PSD-95α (E), P28 PSD-93 KO mice with AAV-GFP (G) and with AAV-PSD-93α2 (H), with sample traces (inset) and analysis of the peak values of AMPA receptor EPSCs (downward deflection) or composite glutamate receptor EPSCs (upward deflection) of successes (black) and failures (gray) of individual EPSCs. Scale bar: 20 ms and 25 pA. Schematic drawing of stereotactic injection of indicated AAV into the visual cortex of P0 mice. (C, F, I, K) Summary graphs of the fraction of silent synapses of indicated manipulation (color code). Dots represent value of single neuron. Values for PSD-95 KO and WT mice (K) were obtained from Fig 1. *p < 0.05; **p < 0.01. Underlying data for this figure can be found in S1 Data. AAV, adeno-associated viral vector; AMPA, α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid; EPSC, excitatory postsynaptic current; GFP, green fluorescent protein; KO, knock-out; P, postnatal day; PSD, postsynaptic density; sh95, short hairpin RNA against PSD-95; V1, primary visual cortex; WT, wild-type.
(A, B) Scheme of the VWT apparatus (A) showing the trapezoid water-filled pool, the midline divider, choice line, and the monitors on which visual stimuli are projected. For measuring visual acuity, mice were trained to swim towards the sinusoidal vertical grating (B; rewarded stimulus). For measuring orientation discrimination, mice were trained to swim towards square-wave vertical grating (G; rewarded stimulus), and the initial horizontal grating was turned by 5° per trial to approach the vertical orientation. (C, H) Number of training blocks to learn the visual acuity procedure (C) for WT, PSD-93 KO, and PSD-93/95 dKO mice or orientation discrimination procedure (H) for WT, PSD-95 KO, or PSD-93 KO mice. Values of individual mice are presented as dots and mean as horizontal line. Underlying data for this figure can be found in S1 Data. (D, I) Visual acuity (D) or orientation discrimination (G) threshold for indicated mouse groups. **p < 0.01. Underlying data for this figure can be found in S1 Data. (E) Schematic representation of the looming procedure with a mouse in a dark arena with a shelter hut and a bright ceiling, on which the dark looming spot is presented. Time course of looming spot presentation is illustrated at the top. (F) The fraction of responses of three consecutive trials (24 h apart) is plotted against WT and PSD-93/95 dKO mice. Values of individual mice are presented as dots and mean as horizontal line. **p < 0.01. Underlying data for this figure can be found in S1 Data. dKO, double KO; DLG, disc-large; KO, knock-out; MAGUK, membrane-associated guanylate kinase; PSD, postsynaptic density; VWT, visual water task; WT, wild-type.
Similar articles
-
Spine dynamics of PSD-95-deficient neurons in the visual cortex link silent synapses to structural cortical plasticity.Proc Natl Acad Sci U S A. 2021 Mar 9;118(10):e2022701118. doi: 10.1073/pnas.2022701118. Proc Natl Acad Sci U S A. 2021. PMID: 33649238 Free PMC article.
-
Progressive maturation of silent synapses governs the duration of a critical period.Proc Natl Acad Sci U S A. 2015 Jun 16;112(24):E3131-40. doi: 10.1073/pnas.1506488112. Epub 2015 May 26. Proc Natl Acad Sci U S A. 2015. PMID: 26015564 Free PMC article.
-
Synaptic state-dependent functional interplay between postsynaptic density-95 and synapse-associated protein 102.J Neurosci. 2013 Aug 14;33(33):13398-409. doi: 10.1523/JNEUROSCI.6255-11.2013. J Neurosci. 2013. PMID: 23946397 Free PMC article.
-
Postsynaptic density-membrane associated guanylate kinase proteins (PSD-MAGUKs) and their role in CNS disorders.Neuroscience. 2009 Jan 12;158(1):324-33. doi: 10.1016/j.neuroscience.2008.07.068. Epub 2008 Aug 13. Neuroscience. 2009. PMID: 18773944 Review.
-
Synaptic trafficking of glutamate receptors by MAGUK scaffolding proteins.Trends Cell Biol. 2007 Jul;17(7):343-52. doi: 10.1016/j.tcb.2007.07.005. Epub 2007 Jul 20. Trends Cell Biol. 2007. PMID: 17644382 Review.
Cited by
-
A mild impairment in reversal learning in a bowl-digging substrate deterministic task but not other cognitive tests in the Dlg2+/- rat model of genetic risk for psychiatric disorder.Genes Brain Behav. 2023 Dec;22(6):e12865. doi: 10.1111/gbb.12865. Epub 2023 Sep 13. Genes Brain Behav. 2023. PMID: 37705179 Free PMC article.
-
Defects in early synaptic formation and neuronal function in Prader-Willi syndrome.Sci Rep. 2023 Jul 25;13(1):12053. doi: 10.1038/s41598-023-39065-x. Sci Rep. 2023. PMID: 37491450 Free PMC article.
-
Shank2 identifies a subset of glycinergic neurons involved in altered nociception in an autism model.Mol Autism. 2023 Jun 14;14(1):21. doi: 10.1186/s13229-023-00552-7. Mol Autism. 2023. PMID: 37316943 Free PMC article.
-
A Deficiency of the Psychiatric Risk Gene DLG2/PSD-93 Causes Excitatory Synaptic Deficits in the Dorsolateral Striatum.Front Mol Neurosci. 2022 Jul 28;15:938590. doi: 10.3389/fnmol.2022.938590. eCollection 2022. Front Mol Neurosci. 2022. PMID: 35966008 Free PMC article.
-
Differential Methylation Profile in Fragile X Syndrome-Prone Offspring Mice after in Utero Exposure to Lactobacillus Reuteri.Genes (Basel). 2022 Jul 22;13(8):1300. doi: 10.3390/genes13081300. Genes (Basel). 2022. PMID: 35893036 Free PMC article.
References
Publication types
MeSH terms
Substances
Grants and funding
LinkOut - more resources
Full Text Sources
Molecular Biology Databases
Research Materials
