- Email: [email protected]
Maturation of thalamocortical synapses in the somatosensory cortex depends on neocortical AKAP5 expression
Neuroscience Letters 709 (2019) 134374
Contents lists available at ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
Research article
Maturation of thalamocortical synapses in the somatosensory cortex depends on neocortical AKAP5 expression Min Zhanga, Meifang Lub, Hao Huangb, Xiaoyan Liub, Haoran Suc, Hong Lib,
T
⁎
a
Department of Physiology, Anhui Medical College, Anhui 230601, China Department of Histology and Embryology, School of Basic Medical Sciences, Anhui Medical University, Anhui 230032, China c Department of Electronic Engineering, City University of Hong Kong, Hong Kong, China b
A R T I C LE I N FO
A B S T R A C T
Keywords: AKAP5 Thalamocortical Neocortex Topographic map Somatosensory cortex Development Barrel
Sensory cortex topographic maps consist of organized arrays of thalamocortical afferents (TCAs) that project into distinct areas of the cortex. Formation of topographic maps in sensory cortices is a prerequisite for functional maturation of the neocortex. Studies have shown that the formation of topographic maps and the maturation of thalamocortical synapses in the somatosensory cortex depend on the cyclic adenosine 5′-monophosphate(cAMP)-protein kinase A (PKA) signaling pathway. AKAP5 is a scaffold protein (also called AKAP79 in humans or AKAP150 in rodents; AKAP79/150) that serves as a signaling hub that links cAMP and PKA signaling. Whether AKAP5 plays a role in topographic map formation and the maturation of thalamocortical synapses during development of the somatosensory cortex is still unknown. Here, we generated cortex-specific AKAP5knockout mice (CxAKAP5KO) to examine its roles in somatosensory cortex development. We found that CxAKAP5KO mice displayed impaired cortical barrel maps. Electrophysiological recordings showed that the AMPA/NMDA ratio was reduced, and silent synapses were increased in thalamocortical synapses of CxAKAP5KO mice during postnatal development. Morphological analysis of layer IV cortical neurons demonstrated that dendritic refinement of these neurons was abnormal. These results indicate that AKAP5 is necessary for both topographic map formation and maturation of thalamocortical synapses as well as morphological development of cortical neurons in the somatosensory cortex.
1. Introduction Cyclic AMP (cAMP) activated protein kinase is important for synapse maturation and topographic map formation in somatosensory cortex during early postnatal development [1–4]. The topographic maps in the somatosensory cortex organize thalamic inputs and cortical neurons into barrels. Barrels consist of rings of dense granular cellbodies from cortical layer IV that surround a cell-sparse hollow, which contains cortical neuronal asymmetric dendrites and thalamus afferent axon innervations. The arrangement of barrels represents the arrangement of the facial whiskers [5,6]. The formation of barrels during development is largely believed to result from a combination of intrinsic genetic programs and activity-dependent mechanisms [7]. Specifically, during embryonic and fetal stages, cortical neuroepithelial cells express diffusible morphogens and transcription factors in specific areas to attract thalamocortical axon afferents during axon pathfinding; the maturation and refinement of these maps rely on neural activity-dependent mechanisms that occurs at postnatal stages [7,8]. Topographic
⁎
maps represent facial whiskers in the brain stem and thalamus called barrelettes and barreloids, respectively. Trigeminal nerves carrying whisker-specific information form synapses with principal sensory nucleus (PrV) neurons in the brain stem ipsilaterally, forming barrelettes. The PrV neurons then project to the contralateral ventral posteromedial nucleus (VPM) of the thalamus, forming barreloids. Barrelless mice (brl) lack the one-to-one barrel map in the somatosensory cortex due to loss of functional Adenylyl cyclase 1 (AC1) caused by transposon integration into the gene [9]. AC1 plays a critical role in the synthesis of cAMP mediated by the Ca2+/Calmodulin mechanism. Accumulating studies demonstrated that AC1-cAMP-PKA is involved in synaptic development and plasticity in the somatosensory cortex. For example, thalamocortical synapses of brl maintain immature states with reduced AMPA/NMDA response ratios and disruption of synapse plasticity, including long-term potentiation (LTP) and long-term depression (LTD) [4]. There are two components in the thalamocortical synaptic pathway: pre- and post-synaptic components, which express AC1 as well as PKA complex proteins. Previous studies showed that when
Corresponding author. E-mail address: [email protected] (H. Li).
https://doi.org/10.1016/j.neulet.2019.134374 Received 13 March 2019; Received in revised form 1 July 2019; Accepted 9 July 2019 Available online 13 July 2019 0304-3940/ © 2019 Elsevier B.V. All rights reserved.
Neuroscience Letters 709 (2019) 134374
M. Zhang, et al.
AKAP5flox/+) were perfused transcardially with ice-cold PBS and 4% PFA subsequently. The somatosensory cortex was separated and flatten as cytochrome oxidase staining described above. Tissues were cut tangentially into 50 μm-thick sections on a vibratome (Leica VT1000S). Next, sections (6 from each animal) were mounted and dried in a slide warmer at 37 °C for one day. Slides were dehydrated in gradient alcohol, and fixed in 10% formalin (Sigma-Aldrich), then stained 15 min in 2% cresyl violet solution. After dehydration in graded alcohol and xylene, slides were mounted with Aquamount. For More Material and Methods, please see Electronic Supplementary Material.
functional NMDA receptors are conditionally knocked out only in cortical excitatory neurons (Cx-NR1 KO), layer IV neurons fail to aggregate and barrels do not develop in the somatosensory cortex; however, thalamocortical afferent (TCA) clustering is largely normal, as are barrelettes and barrelloids [10,11]. Knocking out the type IIβ regulatory subunit of PKA (PKARIIβKO) has similar defects in postsynaptic barrel maps as Cx-NR1 KO mice [3], but critical-period plasticity is not impaired in these mice. PKARIIβ is expressed postsynaptically at thalamocortical synapses. These results demonstrated the importance of the postsynaptic PKA signaling pathway in regulating somatosensory cortex development. The PKA-anchoring protein 5 (AKAP5) has been shown to interact with the type II regulatory subunit of PKA and mediates cAMP-activated PKA signaling [12–14]. AKAP5 is a scaffold protein that is highly expressed in the cerebral cortex, and it is involved in regulating postsynaptic activity [13,15,16]. Several studies demonstrated that AKAP5 plays roles in regulating dendritic signaling as well as learning and memory [15–18]. To date, it is not clear if AKAP5 plays a direct role in barrel map formation and development of the somatosensory cortex or whether it is involved in the AC1- cAMP-PKA signaling pathway in activity-dependent topographic map refinement. Here, we produced a conditional knockout of AKAP5 in the neocortex (CxAKAP5KO) and examined topographic map formation and the electrophysiological properties of thalamocortical synapses in the somatosensory cortex. This study reveals that neocortical AKAP5 is important for development of the somatosensory cortex in terms of barrel map formation, electrophysiological maturation of thalamocortical synapses, and morphological maturation of layer IV granular neurons.
3. Results 3.1. Generation of cortex-specific AKAP5-knockout mice (CxAKAP5KO) To specifically delete AKAP5 in the neocortex, we crossed Emx1-Cre +/- males with AKAP5flox/flox females to generate cortex-specific AKAP5-knockout mice (CxAKAP5KO, Emx1- Cre+/-; AKAP5flox/flox) and control littermates (Emx1-Cre; AKAP5flox/+ and AKAP5flox/+ and AKAP5flox/flox). Since there were no differences observed between control genotypes, all control samples were pooled. At birth, CxAKAP5KO pups were indistinguishable from their control littermates. Moreover, there were no differences in body weight and length between CxAKAP5KO pups and their littermate controls during postnatal development. Western blots and quantitative Droplet Digital PCR (ddPCR) were used to determine the expression of AKAP5 in the neocortex of CxAKAP5KO, and we found that AKAP5 expression in CxAKAP5KO mice decreased to about 80% of that in controls (Fig. 1A, B).
2. Experimental procedures 3.2. Cortical barrels are absent in the somatosensory cortex of CxAKAP5KO mice
2.1. Animals
At first, we examined barrel map formation in CxAKAP5KO mice using cytochrome oxidase (CO) staining. Although CO staining showed that cortical barrels are grossly normal in terms of their presynaptic patterns in the somatosensory cortex of CxAKAP5KO mice at postnatal day 12 (Fig. 1C, D), the postsynaptic cortical ring-like barrel patterns in Nissl-stained sections of the somatosensory cortex of CxAKAP5KO mice are missing (Fig. 1E, F). These results demonstrated that lacking AKAP5 in the neocortex does not affect thalamocortical axon clustering into barrels, but that AKAP5 is required for neocortical neurons to form ringlike barrel cell clusters. We then examined thalamus barreloids and barrelettes in the brain stem via CO and Nissl staining. CO staining showed that both barreloids (Fig. 1G, H) and barrelettes (Fig. 1K, L) are normal in CxAKAP5KO pups. Nissl staining showed that there is no observed difference in the arrangement of neuronal cell bodies in the thalamus VPM and the PrV nucleus in the brain stem between control and mutant mice (Fig. 1I, J, M, N).
All animals were raised and treated following the Regulations and Rules for Experimental Animals at Anhui Medical University. Emx1-Cre +/- (Stock #: 005628, Jackson Laboratory) and AKAP5flox/+ (Stock #: 026694, Jackson Laboratory) mice were maintained and bred in a 12hour day and 12- hour night cycle. The genotype of pups was examined immediately after birth by PCR and the PCR products were confirmed by gene sequence. In every experiment, experimental mice (Emx1- Cre +/-; AKAP5flox/flox) and controls (Emx1-Cre: AKAP5flox/+ and AKAP5flox/+ and AKAP5flox/flox) were littermates. 2.2. Cytochrome oxidase staining Somatosensory cortex from controls (n = 9, AKAP5flox/+) and CxAKAP5KO (n = 8) were dissected at postnatal day 15 (P15), then flattened and fixed in 4% PFA for 2 h at room temperature. The fixed tissues were packed in 2% agarose and tangentially cut (parallel to layer 4) into 100 μm sections, then cytochrome oxidase (CO) staining was performed following a protocol described previously [7]. In brief, sections (3 sections from each animal, control animal: n = 6; CxAKAP5KO animal: n = 10) were firstly incubated with CO reaction solution at 4 °C for 10–14 h. When the staining was strong enough by visual inspection, sections were washed three times in PBS to stop the reaction and fixed with Aqua Polymount (Polysciences). For Barreloids CO staining (control animals: n = 4, CxAKAP5KO: n = 5, 3 sections per animal), thalamocortical sections were cut at 45 ° of sagittal plane across thalamus and somatosensory cortex. For Barrelettes CO staining (control: n = 3, CxAKAP5KO: n = 4, 3 section per animal), transverse sections were cut through brain stem.
3.3. Maturation of thalamocortical synapses is disrupted in CxAKAP5KO mice Under physiological conditions, the AMPA response increases during the second postnatal week at thalamocortical synapses. Previous studies demonstrated that the AMPA to NMDA ratio at thalamocortical synapses is abnormal in both brl mice [4] and cortex-specific AC1knockout mice [2]. We wanted to know whether the AMPA/NMDA ratio could be changed in the absence of cortical AKAP5. As such, we examined the development of thalamocortical synapses in AKAP5KO mice at P6-P9 and P10-P13. The AMPA response did not increase during the second postnatal week (after P10) with a significantly lower AMPA/ NMDA ratio at P10-P13 (control: 2.34 ± 0.52, n = 15, CxAKAP5KO: 0.78 ± 0.28, n = 18, P < 0.01, student’s t-test, Fig. 2A & B), however, no difference in the AMPA/NMDA ratio was seen before P10 (Fig. 2B). These results suggested that AMPA and NMDA receptors may
2.3. Nissl staining Nissl staining for somatosensory cortex were done at postnatal day 7 and 15. Experimental mice (n = 7) and control littermates (n = 6, 2
Neuroscience Letters 709 (2019) 134374
M. Zhang, et al.
Fig. 1. Cortical barrel pattern is disrupted in CxAKAPKO mice. (A) AKAP5 protein is significantly reduced in CxAKAP5KO somatosensory cortices compared to controls shown by western blot (left panel), and relative expression of AKAP5 protein is quantified and normalized with actin (right panel). (B) ddPCR showed similar reduction of AKAP5 mRNA in the somatosensory cortex of CxAKAP5KO mice. (C, D) CO staining shows grossly normal presynaptic thalamocortical afferents clustering into barrels in CxAKAP5KO mice at P12. (E, F) Nissl staining demonstrated that postsynaptic cortical L4 neurons fail to form barrels at P14. Barrelloids (G, H) and barrelettes (K, L) are normal in the absence of AKAP5 expression in the neocortex (P7). Scale bar is 400 μm in (C, D); 150 μm in (E, F); 500 μm in (G, H); 100 μm in (I, J, M, N) and 1 mm in (K, L).
Fig. 2. Evoked AMPA/NMDA ratios fail to increase during postnatal development of CxAKAP5KO mice. A sample of evoked AMPA and NMDA EPSC responses in thalamocortical synapses of control and CxAKAP5KO mice (A). AMPA/NMDA ratios fail to increase during the second postnatal week in CxAKAP5KO mice (B). Sample traces show evoked miniEPSCs after applying strontium in ACSF (C). The amplitude of miniEPSCs is reduced in CxAKAP5KO mice (D). Cumulative plot of evoked minEPSCs amplitude (E). **, P < 0.05; ***, P < 0.01 (unpaired student’s t-test). 3
Neuroscience Letters 709 (2019) 134374
M. Zhang, et al.
only silent synapses are present in CxAKAP5KO mice at early postnatal ages.
initially develop normally, but that the maturation of AMPA receptors is impaired in CxAKAP5KO mice during the second postnatal week. To estimate the efficiency of the AMPA response at a single thalamocortical synapse, we evoked miniature AMPA responses (miniEPSC) at thalamocortical synapses and recorded the effects. An evoked AMPA response was obtained by replacing Ca2+ with Sr2+ in the extracellular solution [4,6,19,21]. In the presence of Sr2+ in the extracellular solution, the evoked AMPA response was desynchronized into single miniEPSCs. We found that the amplitudes of evoked AMPA miniEPSCs were smaller in CxAKAP5KO mice compared to their littermate controls at P9 –P11 (Fig. 2C–E, control: 10.45 ± 0.24, n = 22; CxAKAP5KO: 7.61 ± 0.28, n = 25) indicating that there are fewer AMPA receptors at a single thalamocortical synapse and/or the quantity of the glutamate neurotransmitter released is reduced in CxAKAP5KO barrel cortices.
3.5. Abnormal morphology development of Layer IV neurons in the barrel cortex of CxAKAP5KO mice The dendrites of mature Layer IV (L4) neurons in the barrel cortex display asymmetric morphology; the majority of the dendrites of L4 spiny stellate cells orient toward the barrel’s hollow [6]. To examine if the loss of AKAP5 in the neocortex would affect the maturation of dendritic morphology in L4 neurons, we performed intracellular cell filling with biocytin on acute slices to display single cell morphology and reconstruct the dendritic morphology of L4 spiny stellate neurons (Fig. 4). We found that the dendritic asymmetric index of L4 neurons in CxAKAP5KO mice was significantly lower than control littermates (Fig. 4A-E, control: 0.91 ± 0.019, n = 22; CxAKAP5KO: 0.79 ± 0.044, n = 25, P < 0.01). This result indicates that the asymmetry of L4 neuron dendritic morphology is disrupted in CxAKAP5KO barrel cortices. Additionally, the total dendritic length (Fig. 4F, control: 1451 ± 47 μm; CxAKAP5KO: 2021 ± 169, P < 0.001) and maximum dendritic span in CxAKAP5KO L4 neurons were significantly larger (Fig. 4G, control: 172.45 ± 8.21; CxAKAP5KO 253.48 ± 13.54, P < 0.001) than those of control littermates. However, the dendritic branch points are significantly reduced compared to controls (Fig. 4H, control 10.72 ± 0.92; CxAKAP5KO: 6.78 ± 0.97, P < 0.01). These results demonstrated that the dendritic refinement of L4 neurons in barrel cortices was impaired in the absence of neocortical AKAP5.
3.4. Increased silent synapses in CxAKAP5KO mice Previous work [4] showed that brl mice have reduced silent synapses at thalamocortical synapses, so we examined whether there was a difference in the abundance of silent synapses between CxAKAP5KO mice and their littermate controls. NMDAR-only synapses with no AMPAR-mediated response when activated by glutamate are considered silent synapses [20]. NMDAR-only synapses can only be activated when there is AMPAR-mediated depolarization because of the voltage-dependent property of NMDAR. It is known that silent thalamocortical synapses are widely spread in the cortex during the first and the second postnatal weeks, but these numbers gradually reduce during postnatal development until the end of the critical period (P30) [20,22]. We applied a standard ‘minimal stimulation’ protocol described previously [4] to examine the abundance of silent synapses in P4-6 CxAKAP5KO mice and their littermates (Fig. 3). We first voltage-clamped the cell at −70 mV and started stimulation very low. After finding the minimal stimulation strength, we observed activation of a small number of synapses mixed randomly with response ‘failures,’ presumably caused by failure of vesicular release (Fig. 3A, lower panel). We then switched to +40 mV. The relative abundance of silent (NMDAR-only) synapses was estimated from the difference between the failure rates at −70 mV and +40 mV (Fig. 3B). A clear reduction in the failure rate from four out of five P4–6 control mice was observed when the holding potential was switched from −70 mV to +40 mV (Fig. 3B, left panel). The difference in failure rates, which is proportional to the percentage of thalamocortical synapses that are silent, was 22.5 ± 4% (n = 5; Fig. 3C) in control neurons. Interestingly, we observed increased failure rates at +40 mV in CxAKAP5KO pups (Fig. 3B, right panel). The difference in failure rates between −70 mV and +40 mV in CxAKAP5KO mice was 15.4% ± 7% (n = 6; Fig. 3C). These results revealed that more NMDA-
4. Discussion In this study, we produced a neocortical-specific knockout of AKAP5 and examined its role in barrel map formation and the electrophysiological maturation of thalamocortical synapses in the somatosensory cortex. We found impaired AMPA/NMDA ratios in evoked thalamocortical synapse responses in the somatosensory cortex of CxAKAP5KO mice. AMPA receptor-mediated miniEPSCs fail to increase during development and the number of NMDA-only silent synapses increased in these mice. The dendritic morphology of layer IV neurons in the somatosensory cortex of CxAKAP5KO mice is less asymmetric in shape with longer total length and fewer branch points at the second postnatal week. These results showed that AKAP5 is important for promoting development of electrophysiological maturation of thalamocortical synapses and morphological maturation of layer IV neurons in the somatosensory cortex. Previous studies on genetically modified mice have uncovered molecular mechanisms for barrel map formation that involve two major
Fig. 3. Silent synapses are increased at thalamocortical synapses of CxAKAP5KO mice. (A) Sample traces of AMPA receptor miniEPSC (upper panel) and record of success and failure under minimal stimulation (B). Summary of failure rates at -70mv and +40mv holding potentials (B). Difference in failure rates between controls and CxAKAP5KO mice (C), error bar: standard deviation. 4
Neuroscience Letters 709 (2019) 134374
M. Zhang, et al.
Fig. 4. Abnormal morphology development of cortical L4 neurons at P14. (A, B) Fluorescent images of biocytin-filled L4 neurons in the somatosensory cortex. (C, D) 3D morphology reconstruction of the dendritic trees of L4 neurons in (A) and (B). Dendritic asymmetric index is reduced in CxAKAP5KO L4 neurons (E). Total dendritic length is increased in CxAKAP5KO L4 neurons (F). Maximum dendritic span is increased in CxAKAP5KO L4 neurons (G), however, total branch points are reduced in CxAKAP5KO L4 neurons (H). Scale bar (A, B) = 50 μm. **, P < 0.05; ***, P < 0.01 (unpaired student’s t-test).
5
Neuroscience Letters 709 (2019) 134374
M. Zhang, et al.
Morphologically, layer IV neurons in the barrel cortex show similarly reduced dendritic asymmetry in CxAKAP5KO mice. Similar morphological abnormality of layer IV neurons has been found in PKARIIβKO and CxAC1 KO mice. These results further support that AKAP5 mediates PKA activation in regulating somatosensory barrel map formation. In summary, our results demonstrated that cortical AKAP5 expression is essential in the physiological and morphological maturation of the somatosensory cortex during postnatal development.
pathways: cAMP-PKA [4,6,8] and serotonin (5-HT) pathways [6,23–25]. Mice lacking AC1 (brl) do not form barrels pre- and postsynaptically [8,26]. The type IIβ regulatory subunit of PKA knockout mice (PKARIIβKO mice) show defects in postsynaptic barrel maps but presynaptical barrel maps are largely normal [6]. Layer IV neurons of cortex-specific knockout of an NMDA subunit (NR1) (Cx-NR1 KO) failed to aggregate into barrels, but TCA afferent clustering was normal [25]. These studies demonstrated that proper layer IV barrel organization can be disrupted independent of a gross defect in TCA patterning. NMDA–AC1–PKA signaling may be a central molecular pathway underlying activity-dependent regulation of barrel formation in the neocortex. Serotonin (5-HT) signaling is another pathway that is strongly implicated in barrel map development. During early postnatal development, 5-HT transporters (5-HTTs) are highly expressed at axon terminals of thalamocortical synapses. 5-HTTs take up excessive 5-HT in synapse clefts. Barrels do not form in single or double KOs of 5-HTT and monoamino oxidase A (MAOA), but can be rescued by knocking out the 5-HT1B receptor in 5-HTT and MAOA single or double mutants [27–29]. However, 5HT1B mutant mice themselves have normal barrel maps. These data indicate that 5-HT1B is essential for the disruption of barrels in the 5-HT signaling pathway, but is not necessary for barrel formation [29]. It is not clear whether these two pathways cross talk. Expression of AKAP5 during the developing cerebral cortex is increased during postnatal development (mouse ENCODE transcriptome data). We observed that the cell bodies of layer IV neurons failed to aggregate in CxAKAP5KO somatosensory cortices, indicating the prominent effect of AKAP5 in topographic map formation. The mechanism underlying this effect is an interesting question for further study. AKAP5 is believed to be a postsynaptic scaffold protein that serves as a signal hub that links, not only the PKA kinase pathway, but also PKC and other second messenger systems. The 5-HT1B receptor is a G-protein coupled receptor. Whether AKAP5 regulation of topographic map formation of the somatosensory cortex occurs merely through the cAMP-PKA pathway or the 5-HT signaling pathway or both is not yet clear. To answer this question, for example, examining barrel map formation in mice with a mutated PKA binding site in the AKAP5 protein will be an interesting future experiment. Lacking AKAP5 in the neocortex must not affect afferent clustering in the brain stem and spinal cord because barrelloids and barrelettes in CxAKAP5KO mice were largely normal, suggesting that the function of AKAP5 in topographic map formation is specific to the neocortex. Electrophysiological analysis of AMPA receptor responses of thalamocortical synapses in CxAKAP5KO mice revealed that AMPA receptors failed to increase during development. Not only does the AMPA/NMDA ratio in evoked thalamocortical synapses of CxAKAP5KO mice maintain immature stages, but silent synapses also fail to decrease in these mice. The maturation of thalamocortical synapses involves trafficking and insertion of AMPA receptors into the postsynaptic membrane. Our results are in line with a recent publication in which AKAP5 was shown to control AMPA receptor phosphorylation and cell surface targeting through cAMP-PKA signaling [16]. Synaptic maturation in the somatosensory cortex is thought to occur through LTP-like mechanisms that regulate the trafficking of AMPA receptors into the synapse [4,7,30]. Mice with a global knockout of PKARII have deficits in synaptic plasticity and TC synapse maturation. CxAKAP5KO mice show immature electrophysiological phenotypes similar to brl, CxAC1 KO, and PKARII β KO mice [3,4,25]. It is intriguing that we observed increased numbers of silent synapses in CxAKAP5KO mice compared to littermate controls. In contrast, brl mice do not form barrel patterns either pre- or postsynaptically. One possible reason for this difference could be that brlmice might have severe disruption of NMDA receptor expression to begin with since AC1 expression is globally absent in brl mice. Our results extend that molecular mechanisms known to regulate TC synapse maturation, which may partly associate with AC1/cAMP/PKA signaling.
Acknowledgements This work was supported by grants from the Natural Science Foundation of Anhui Province #KJ2018A0801 (MZ) and the National Natural Science Foundation of China: #31671064 (HL) and #81471159 (HL). Hong Li conceived the study and wrote the manuscript. Min Zhang, Meifang Lu, Hao Huang, and Xiaoyan Liu performed the experiments, Haoran Su carried out the statistical analysis. All authors have read and approved the final version of the manuscript. Hong Li’s present address: Department of Neuroscience, Yale University, U.S.A. Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.neulet.2019.134374. References [1] R.F. Watson, R.M. Abdel-Majid, M.W. Barnett, B.S. Willis, A. Katsnelson, T.H. Gillingwater, G.S. McKnight, P.C. Kind, P.E. Neumann, Involvement of protein kinase a in patterning of the mouse somatosensory cortex, J. Neurosci. 26 (2006) 5393–5401, https://doi.org/10.1523/JNEUROSCI.0750-06.2006. [2] T. Iwasato, M. Inan, H. Kanki, R.S. Erzurumlu, S. Itohara, M.C. Crair, Cortical adenylyl cyclase 1 is required for thalamocortical synapse maturation and aspects of layer IV barrel development, J. Neurosci. 28 (2008) 5931–5943, https://doi.org/ 10.1523/JNEUROSCI.0815-08.2008. [3] M. Inan, H.-C. Lu, M.J. Albright, W.-C. She, M.C. Crair, Barrel map development relies on protein kinase a regulatory subunit IIbeta-Mediated cAMP signaling, J. Neurosci. 26 (2006) 4338–4349, https://doi.org/10.1523/JNEUROSCI.3745-05. 2006. [4] H.-C. Lu, W.-C. She, D.T. Plas, P.E. Neumann, R. Janz, M.C. Crair, Adenylyl cyclase I regulates AMPA receptor trafficking during mouse cortical “barrel” map development, Nat. Neurosci. 6 (2003) 939–947, https://doi.org/10.1038/nn1106. [5] T.A. Woolsey, H. Van der Loos, The structural organization of layer IV in the somatosensory region (SI) of mouse cerebral cortex. The description of a cortical field composed of discrete cytoarchitectonic units, Brain Res. 17 (1970) 205–242 (Accessed September 29, 2018), http://www.ncbi.nlm.nih.gov/pubmed/4904874. [6] H. Li, M.C. Crair, How do barrels form in somatosensory cortex? Ann. N. Y. Acad. Sci. 1225 (2011) 119–129, https://doi.org/10.1111/j.1749-6632.2011.06024.x. [7] H. Li, S. Fertuzinhos, E. Mohns, T.S. Hnasko, M. Verhage, R. Edwards, N. Sestan, M.C. Crair, Laminar and columnar development of barrel cortex relies on thalamocortical neurotransmission, Neuron 79 (2013) 970–986, https://doi.org/10. 1016/j.neuron.2013.06.043. [8] T.K. Hensch, Critical period regulation, Annu. Rev. Neurosci. 27 (2004) 549–579, https://doi.org/10.1146/annurev.neuro.27.070203.144327. [9] R.M. Abdel-Majid, W.L. Leong, L.C. Schalkwyk, D.S. Smallman, S.T. Wong, D.R. Storm, A. Fine, M.J. Dobson, D.L. Guernsey, P.E. Neumann, Loss of adenylyl cyclase I activity disrupts patterning of mouse somatosensory cortex, Nat. Genet. 19 (1998) 289–291, https://doi.org/10.1038/980. [10] A. Datwani, T. Iwasato, S. Itohara, R.S. Erzurumlu, Lesion-induced thalamocortical axonal plasticity in the S1 cortex is independent of NMDA receptor function in excitatory cortical neurons, J. Neurosci. 22 (2002) 9171–9175 (Accessed September 29, 2018), http://www.ncbi.nlm.nih.gov/pubmed/12417641. [11] L.-J. Lee, T. Iwasato, S. Itohara, R.S. Erzurumlu, Exuberant thalamocortical axon arborization in cortex-specific NMDAR1 knockout mice, J. Comp. Neurol. 485 (2005) 280–292, https://doi.org/10.1002/cne.20481. [12] H.R. Robertson, E.S. Gibson, T.A. Benke, M.L. Dell’Acqua, Regulation of postsynaptic structure and function by an A-Kinase anchoring protein-membraneAssociated guanylate kinase scaffolding complex, J. Neurosci. 29 (2009) 7929–7943, https://doi.org/10.1523/JNEUROSCI.6093-08.2009. [13] Y. Lu, X. Zha, E.Y. Kim, S. Schachtele, M.E. Dailey, D.D. Hall, S. Strack, S.H. Green, D.A. Hoffman, J.W. Hell, A kinase anchor protein 150 (AKAP150)-associated protein kinase A limits dendritic spine density, J. Biol. Chem. 286 (2011) 26496–26506, https://doi.org/10.1074/jbc.M111.254912. [14] M.G. Gold, F. Stengel, P.J. Nygren, C.R. Weisbrod, J.E. Bruce, C.V. Robinson, D. Barford, J.D. Scott, Architecture and dynamics of an A-kinase anchoring protein 79 (AKAP79) signaling complex, Proc. Natl. Acad. Sci. 108 (2011) 6426–6431, https://doi.org/10.1073/pnas.1014400108.
6
Neuroscience Letters 709 (2019) 134374
M. Zhang, et al.
[15] M. Weisenhaus, M.L. Allen, L. Yang, Y. Lu, C.B. Nichols, T. Su, J.W. Hell, G.S. McKnight, Mutations in AKAP5 disrupt dendritic signaling complexes and lead to electrophysiological and behavioral phenotypes in mice, PLoS One 5 (2010) e10325, , https://doi.org/10.1371/journal.pone.0010325. [16] G.H. Diering, A.S. Gustina, R.L. Huganir, PKA-GluA1 coupling via AKAP5 controls AMPA receptor phosphorylation and cell-surface targeting during bidirectional homeostatic plasticity, Neuron 84 (2014) 790–805, https://doi.org/10.1016/j. neuron.2014.09.024. [17] A. Ostroveanu, E.A. Van der Zee, A.M. Dolga, P.G.M. Luiten, U.L.M. Eisel, I.M. Nijholt, A-kinase anchoring protein 150 in the mouse brain is concentrated in areas involved in learning and memory, Brain Res. 1145 (2007) 97–107, https:// doi.org/10.1016/j.brainres.2007.01.117. [18] B.J. Tunquist, N. Hoshi, E.S. Guire, F. Zhang, K. Mullendorff, L.K. Langeberg, J. Raber, J.D. Scott, Loss of AKAP150 perturbs distinct neuronal processes in mice, Proc. Natl. Acad. Sci. 105 (2008) 12557–12562, https://doi.org/10.1073/pnas. 0805922105. [19] M.A. Xu-Friedman, W.G. Regehr, Presynaptic strontium dynamics and synaptic transmission, Biophys. J. 76 (1999) 2029–2042, https://doi.org/10.1016/S00063495(99)77360-1. [20] J.T. Isaac, M.C. Crair, R.A. Nicoll, R.C. Malenka, Silent synapses during development of thalamocortical inputs, Neuron 18 (1997) 269–280 (Accessed September 23, 2018), http://www.ncbi.nlm.nih.gov/pubmed/9052797. [21] R.W. Rhoades, C.A. Bennett-Clarke, M.Y. Shi, R.D. Mooney, Effects of 5-HT on thalamocortical synaptic transmission in the developing rat, J. Neurophysiol. 72 (1994) 2438–2450, https://doi.org/10.1152/jn.1994.72.5.2438. [22] C.L. Young-Davies, C.A. Bennett-Clarke, R.D. Lane, R.W. Rhoades, Selective facilitation of the serotonin(1B) receptor causes disorganization of thalamic afferents and barrels in somatosensory cortex of rat, J. Comp. Neurol. 425 (2000) 130–138 (Accessed October 1, 2018), http://www.ncbi.nlm.nih.gov/pubmed/10940947. [23] C. Lebrand, O. Cases, C. Adelbrecht, A. Doye, C. Alvarez, S. El Mestikawy, I. Seif,
[24]
[25]
[26]
[27]
[28]
[29]
[30]
7
P. Gaspar, Transient uptake and storage of serotonin in developing thalamic neurons, Neuron 17 (1996) 823–835 (Accessed October 1, 2018), http://www.ncbi. nlm.nih.gov/pubmed/8938116. T. Iwasato, M. Inan, H. Kanki, R.S. Erzurumlu, S. Itohara, M.C. Crair, Cortical adenylyl cyclase 1 is required for thalamocortical synapse maturation and aspects of layer IV barrel development, J. Neurosci. 28 (2008) 5931–5943, https://doi.org/ 10.1523/JNEUROSCI.0815-08.2008. T. Iwasato, A. Datwani, A.M. Wolf, H. Nishiyama, Y. Taguchi, S. Tonegawa, T. Knöpfel, R.S. Erzurumlu, S. Itohara, Cortex-restricted disruption of NMDAR1 impairs neuronal patterns in the barrel cortex, Nature 406 (2000) 726–731, https:// doi.org/10.1038/35021059. C. Alvarez, T. Vitalis, E.A. Fon, N. Hanoun, M. Hamon, I. Seif, R. Edwards, P. Gaspar, O. Cases, Effects of genetic depletion of monoamines on somatosensory cortical development, Neuroscience 115 (2002) 753–764 (Accessed October 1, 2018), http://www.ncbi.nlm.nih.gov/pubmed/12435414. N. Salichon, P. Gaspar, A.L. Upton, S. Picaud, N. Hanoun, M. Hamon, E. De Maeyer, D.L. Murphy, R. Mossner, K.P. Lesch, R. Hen, I. Seif, Excessive activation of serotonin (5-HT) 1B receptors disrupts the formation of sensory maps in monoamine oxidase a and 5-ht transporter knock-out mice, J. Neurosci. 21 (2001) 884–896 (Accessed October 1, 2018), http://www.ncbi.nlm.nih.gov/pubmed/11157075. T. Vitalis, C. Fouquet, C. Alvarez, I. Seif, D. Price, P. Gaspar, O. Cases, Developmental expression of monoamine oxidases A and B in the central and peripheral nervous systems of the mouse, J. Comp. Neurol. 442 (2002) 331–347, https://doi.org/10.1002/cne.10093. C.B. Boylan, K.L. Kesterson, T.A. Crnko-Hoppenjans, M. Ke, T. Rizk, R.D. Mooney, R.W. Rhoades, The cortical vibrissae representation is normal in transgenic mice lacking the 5-HT(1B) receptor, Brain Res. Dev. Brain Res. 120 (2000) 91–93 (Accessed October 1, 2018), http://www.ncbi.nlm.nih.gov/pubmed/10727735. M.C. Crair, R.C. Malenka, A critical period for long-term potentiation at thalamocortical synapses, Nature 375 (1995) 325–328, https://doi.org/10.1038/375325a0.
Contents lists available at ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
Research article
Maturation of thalamocortical synapses in the somatosensory cortex depends on neocortical AKAP5 expression Min Zhanga, Meifang Lub, Hao Huangb, Xiaoyan Liub, Haoran Suc, Hong Lib,
T
⁎
a
Department of Physiology, Anhui Medical College, Anhui 230601, China Department of Histology and Embryology, School of Basic Medical Sciences, Anhui Medical University, Anhui 230032, China c Department of Electronic Engineering, City University of Hong Kong, Hong Kong, China b
A R T I C LE I N FO
A B S T R A C T
Keywords: AKAP5 Thalamocortical Neocortex Topographic map Somatosensory cortex Development Barrel
Sensory cortex topographic maps consist of organized arrays of thalamocortical afferents (TCAs) that project into distinct areas of the cortex. Formation of topographic maps in sensory cortices is a prerequisite for functional maturation of the neocortex. Studies have shown that the formation of topographic maps and the maturation of thalamocortical synapses in the somatosensory cortex depend on the cyclic adenosine 5′-monophosphate(cAMP)-protein kinase A (PKA) signaling pathway. AKAP5 is a scaffold protein (also called AKAP79 in humans or AKAP150 in rodents; AKAP79/150) that serves as a signaling hub that links cAMP and PKA signaling. Whether AKAP5 plays a role in topographic map formation and the maturation of thalamocortical synapses during development of the somatosensory cortex is still unknown. Here, we generated cortex-specific AKAP5knockout mice (CxAKAP5KO) to examine its roles in somatosensory cortex development. We found that CxAKAP5KO mice displayed impaired cortical barrel maps. Electrophysiological recordings showed that the AMPA/NMDA ratio was reduced, and silent synapses were increased in thalamocortical synapses of CxAKAP5KO mice during postnatal development. Morphological analysis of layer IV cortical neurons demonstrated that dendritic refinement of these neurons was abnormal. These results indicate that AKAP5 is necessary for both topographic map formation and maturation of thalamocortical synapses as well as morphological development of cortical neurons in the somatosensory cortex.
1. Introduction Cyclic AMP (cAMP) activated protein kinase is important for synapse maturation and topographic map formation in somatosensory cortex during early postnatal development [1–4]. The topographic maps in the somatosensory cortex organize thalamic inputs and cortical neurons into barrels. Barrels consist of rings of dense granular cellbodies from cortical layer IV that surround a cell-sparse hollow, which contains cortical neuronal asymmetric dendrites and thalamus afferent axon innervations. The arrangement of barrels represents the arrangement of the facial whiskers [5,6]. The formation of barrels during development is largely believed to result from a combination of intrinsic genetic programs and activity-dependent mechanisms [7]. Specifically, during embryonic and fetal stages, cortical neuroepithelial cells express diffusible morphogens and transcription factors in specific areas to attract thalamocortical axon afferents during axon pathfinding; the maturation and refinement of these maps rely on neural activity-dependent mechanisms that occurs at postnatal stages [7,8]. Topographic
⁎
maps represent facial whiskers in the brain stem and thalamus called barrelettes and barreloids, respectively. Trigeminal nerves carrying whisker-specific information form synapses with principal sensory nucleus (PrV) neurons in the brain stem ipsilaterally, forming barrelettes. The PrV neurons then project to the contralateral ventral posteromedial nucleus (VPM) of the thalamus, forming barreloids. Barrelless mice (brl) lack the one-to-one barrel map in the somatosensory cortex due to loss of functional Adenylyl cyclase 1 (AC1) caused by transposon integration into the gene [9]. AC1 plays a critical role in the synthesis of cAMP mediated by the Ca2+/Calmodulin mechanism. Accumulating studies demonstrated that AC1-cAMP-PKA is involved in synaptic development and plasticity in the somatosensory cortex. For example, thalamocortical synapses of brl maintain immature states with reduced AMPA/NMDA response ratios and disruption of synapse plasticity, including long-term potentiation (LTP) and long-term depression (LTD) [4]. There are two components in the thalamocortical synaptic pathway: pre- and post-synaptic components, which express AC1 as well as PKA complex proteins. Previous studies showed that when
Corresponding author. E-mail address: [email protected] (H. Li).
https://doi.org/10.1016/j.neulet.2019.134374 Received 13 March 2019; Received in revised form 1 July 2019; Accepted 9 July 2019 Available online 13 July 2019 0304-3940/ © 2019 Elsevier B.V. All rights reserved.
Neuroscience Letters 709 (2019) 134374
M. Zhang, et al.
AKAP5flox/+) were perfused transcardially with ice-cold PBS and 4% PFA subsequently. The somatosensory cortex was separated and flatten as cytochrome oxidase staining described above. Tissues were cut tangentially into 50 μm-thick sections on a vibratome (Leica VT1000S). Next, sections (6 from each animal) were mounted and dried in a slide warmer at 37 °C for one day. Slides were dehydrated in gradient alcohol, and fixed in 10% formalin (Sigma-Aldrich), then stained 15 min in 2% cresyl violet solution. After dehydration in graded alcohol and xylene, slides were mounted with Aquamount. For More Material and Methods, please see Electronic Supplementary Material.
functional NMDA receptors are conditionally knocked out only in cortical excitatory neurons (Cx-NR1 KO), layer IV neurons fail to aggregate and barrels do not develop in the somatosensory cortex; however, thalamocortical afferent (TCA) clustering is largely normal, as are barrelettes and barrelloids [10,11]. Knocking out the type IIβ regulatory subunit of PKA (PKARIIβKO) has similar defects in postsynaptic barrel maps as Cx-NR1 KO mice [3], but critical-period plasticity is not impaired in these mice. PKARIIβ is expressed postsynaptically at thalamocortical synapses. These results demonstrated the importance of the postsynaptic PKA signaling pathway in regulating somatosensory cortex development. The PKA-anchoring protein 5 (AKAP5) has been shown to interact with the type II regulatory subunit of PKA and mediates cAMP-activated PKA signaling [12–14]. AKAP5 is a scaffold protein that is highly expressed in the cerebral cortex, and it is involved in regulating postsynaptic activity [13,15,16]. Several studies demonstrated that AKAP5 plays roles in regulating dendritic signaling as well as learning and memory [15–18]. To date, it is not clear if AKAP5 plays a direct role in barrel map formation and development of the somatosensory cortex or whether it is involved in the AC1- cAMP-PKA signaling pathway in activity-dependent topographic map refinement. Here, we produced a conditional knockout of AKAP5 in the neocortex (CxAKAP5KO) and examined topographic map formation and the electrophysiological properties of thalamocortical synapses in the somatosensory cortex. This study reveals that neocortical AKAP5 is important for development of the somatosensory cortex in terms of barrel map formation, electrophysiological maturation of thalamocortical synapses, and morphological maturation of layer IV granular neurons.
3. Results 3.1. Generation of cortex-specific AKAP5-knockout mice (CxAKAP5KO) To specifically delete AKAP5 in the neocortex, we crossed Emx1-Cre +/- males with AKAP5flox/flox females to generate cortex-specific AKAP5-knockout mice (CxAKAP5KO, Emx1- Cre+/-; AKAP5flox/flox) and control littermates (Emx1-Cre; AKAP5flox/+ and AKAP5flox/+ and AKAP5flox/flox). Since there were no differences observed between control genotypes, all control samples were pooled. At birth, CxAKAP5KO pups were indistinguishable from their control littermates. Moreover, there were no differences in body weight and length between CxAKAP5KO pups and their littermate controls during postnatal development. Western blots and quantitative Droplet Digital PCR (ddPCR) were used to determine the expression of AKAP5 in the neocortex of CxAKAP5KO, and we found that AKAP5 expression in CxAKAP5KO mice decreased to about 80% of that in controls (Fig. 1A, B).
2. Experimental procedures 3.2. Cortical barrels are absent in the somatosensory cortex of CxAKAP5KO mice
2.1. Animals
At first, we examined barrel map formation in CxAKAP5KO mice using cytochrome oxidase (CO) staining. Although CO staining showed that cortical barrels are grossly normal in terms of their presynaptic patterns in the somatosensory cortex of CxAKAP5KO mice at postnatal day 12 (Fig. 1C, D), the postsynaptic cortical ring-like barrel patterns in Nissl-stained sections of the somatosensory cortex of CxAKAP5KO mice are missing (Fig. 1E, F). These results demonstrated that lacking AKAP5 in the neocortex does not affect thalamocortical axon clustering into barrels, but that AKAP5 is required for neocortical neurons to form ringlike barrel cell clusters. We then examined thalamus barreloids and barrelettes in the brain stem via CO and Nissl staining. CO staining showed that both barreloids (Fig. 1G, H) and barrelettes (Fig. 1K, L) are normal in CxAKAP5KO pups. Nissl staining showed that there is no observed difference in the arrangement of neuronal cell bodies in the thalamus VPM and the PrV nucleus in the brain stem between control and mutant mice (Fig. 1I, J, M, N).
All animals were raised and treated following the Regulations and Rules for Experimental Animals at Anhui Medical University. Emx1-Cre +/- (Stock #: 005628, Jackson Laboratory) and AKAP5flox/+ (Stock #: 026694, Jackson Laboratory) mice were maintained and bred in a 12hour day and 12- hour night cycle. The genotype of pups was examined immediately after birth by PCR and the PCR products were confirmed by gene sequence. In every experiment, experimental mice (Emx1- Cre +/-; AKAP5flox/flox) and controls (Emx1-Cre: AKAP5flox/+ and AKAP5flox/+ and AKAP5flox/flox) were littermates. 2.2. Cytochrome oxidase staining Somatosensory cortex from controls (n = 9, AKAP5flox/+) and CxAKAP5KO (n = 8) were dissected at postnatal day 15 (P15), then flattened and fixed in 4% PFA for 2 h at room temperature. The fixed tissues were packed in 2% agarose and tangentially cut (parallel to layer 4) into 100 μm sections, then cytochrome oxidase (CO) staining was performed following a protocol described previously [7]. In brief, sections (3 sections from each animal, control animal: n = 6; CxAKAP5KO animal: n = 10) were firstly incubated with CO reaction solution at 4 °C for 10–14 h. When the staining was strong enough by visual inspection, sections were washed three times in PBS to stop the reaction and fixed with Aqua Polymount (Polysciences). For Barreloids CO staining (control animals: n = 4, CxAKAP5KO: n = 5, 3 sections per animal), thalamocortical sections were cut at 45 ° of sagittal plane across thalamus and somatosensory cortex. For Barrelettes CO staining (control: n = 3, CxAKAP5KO: n = 4, 3 section per animal), transverse sections were cut through brain stem.
3.3. Maturation of thalamocortical synapses is disrupted in CxAKAP5KO mice Under physiological conditions, the AMPA response increases during the second postnatal week at thalamocortical synapses. Previous studies demonstrated that the AMPA to NMDA ratio at thalamocortical synapses is abnormal in both brl mice [4] and cortex-specific AC1knockout mice [2]. We wanted to know whether the AMPA/NMDA ratio could be changed in the absence of cortical AKAP5. As such, we examined the development of thalamocortical synapses in AKAP5KO mice at P6-P9 and P10-P13. The AMPA response did not increase during the second postnatal week (after P10) with a significantly lower AMPA/ NMDA ratio at P10-P13 (control: 2.34 ± 0.52, n = 15, CxAKAP5KO: 0.78 ± 0.28, n = 18, P < 0.01, student’s t-test, Fig. 2A & B), however, no difference in the AMPA/NMDA ratio was seen before P10 (Fig. 2B). These results suggested that AMPA and NMDA receptors may
2.3. Nissl staining Nissl staining for somatosensory cortex were done at postnatal day 7 and 15. Experimental mice (n = 7) and control littermates (n = 6, 2
Neuroscience Letters 709 (2019) 134374
M. Zhang, et al.
Fig. 1. Cortical barrel pattern is disrupted in CxAKAPKO mice. (A) AKAP5 protein is significantly reduced in CxAKAP5KO somatosensory cortices compared to controls shown by western blot (left panel), and relative expression of AKAP5 protein is quantified and normalized with actin (right panel). (B) ddPCR showed similar reduction of AKAP5 mRNA in the somatosensory cortex of CxAKAP5KO mice. (C, D) CO staining shows grossly normal presynaptic thalamocortical afferents clustering into barrels in CxAKAP5KO mice at P12. (E, F) Nissl staining demonstrated that postsynaptic cortical L4 neurons fail to form barrels at P14. Barrelloids (G, H) and barrelettes (K, L) are normal in the absence of AKAP5 expression in the neocortex (P7). Scale bar is 400 μm in (C, D); 150 μm in (E, F); 500 μm in (G, H); 100 μm in (I, J, M, N) and 1 mm in (K, L).
Fig. 2. Evoked AMPA/NMDA ratios fail to increase during postnatal development of CxAKAP5KO mice. A sample of evoked AMPA and NMDA EPSC responses in thalamocortical synapses of control and CxAKAP5KO mice (A). AMPA/NMDA ratios fail to increase during the second postnatal week in CxAKAP5KO mice (B). Sample traces show evoked miniEPSCs after applying strontium in ACSF (C). The amplitude of miniEPSCs is reduced in CxAKAP5KO mice (D). Cumulative plot of evoked minEPSCs amplitude (E). **, P < 0.05; ***, P < 0.01 (unpaired student’s t-test). 3
Neuroscience Letters 709 (2019) 134374
M. Zhang, et al.
only silent synapses are present in CxAKAP5KO mice at early postnatal ages.
initially develop normally, but that the maturation of AMPA receptors is impaired in CxAKAP5KO mice during the second postnatal week. To estimate the efficiency of the AMPA response at a single thalamocortical synapse, we evoked miniature AMPA responses (miniEPSC) at thalamocortical synapses and recorded the effects. An evoked AMPA response was obtained by replacing Ca2+ with Sr2+ in the extracellular solution [4,6,19,21]. In the presence of Sr2+ in the extracellular solution, the evoked AMPA response was desynchronized into single miniEPSCs. We found that the amplitudes of evoked AMPA miniEPSCs were smaller in CxAKAP5KO mice compared to their littermate controls at P9 –P11 (Fig. 2C–E, control: 10.45 ± 0.24, n = 22; CxAKAP5KO: 7.61 ± 0.28, n = 25) indicating that there are fewer AMPA receptors at a single thalamocortical synapse and/or the quantity of the glutamate neurotransmitter released is reduced in CxAKAP5KO barrel cortices.
3.5. Abnormal morphology development of Layer IV neurons in the barrel cortex of CxAKAP5KO mice The dendrites of mature Layer IV (L4) neurons in the barrel cortex display asymmetric morphology; the majority of the dendrites of L4 spiny stellate cells orient toward the barrel’s hollow [6]. To examine if the loss of AKAP5 in the neocortex would affect the maturation of dendritic morphology in L4 neurons, we performed intracellular cell filling with biocytin on acute slices to display single cell morphology and reconstruct the dendritic morphology of L4 spiny stellate neurons (Fig. 4). We found that the dendritic asymmetric index of L4 neurons in CxAKAP5KO mice was significantly lower than control littermates (Fig. 4A-E, control: 0.91 ± 0.019, n = 22; CxAKAP5KO: 0.79 ± 0.044, n = 25, P < 0.01). This result indicates that the asymmetry of L4 neuron dendritic morphology is disrupted in CxAKAP5KO barrel cortices. Additionally, the total dendritic length (Fig. 4F, control: 1451 ± 47 μm; CxAKAP5KO: 2021 ± 169, P < 0.001) and maximum dendritic span in CxAKAP5KO L4 neurons were significantly larger (Fig. 4G, control: 172.45 ± 8.21; CxAKAP5KO 253.48 ± 13.54, P < 0.001) than those of control littermates. However, the dendritic branch points are significantly reduced compared to controls (Fig. 4H, control 10.72 ± 0.92; CxAKAP5KO: 6.78 ± 0.97, P < 0.01). These results demonstrated that the dendritic refinement of L4 neurons in barrel cortices was impaired in the absence of neocortical AKAP5.
3.4. Increased silent synapses in CxAKAP5KO mice Previous work [4] showed that brl mice have reduced silent synapses at thalamocortical synapses, so we examined whether there was a difference in the abundance of silent synapses between CxAKAP5KO mice and their littermate controls. NMDAR-only synapses with no AMPAR-mediated response when activated by glutamate are considered silent synapses [20]. NMDAR-only synapses can only be activated when there is AMPAR-mediated depolarization because of the voltage-dependent property of NMDAR. It is known that silent thalamocortical synapses are widely spread in the cortex during the first and the second postnatal weeks, but these numbers gradually reduce during postnatal development until the end of the critical period (P30) [20,22]. We applied a standard ‘minimal stimulation’ protocol described previously [4] to examine the abundance of silent synapses in P4-6 CxAKAP5KO mice and their littermates (Fig. 3). We first voltage-clamped the cell at −70 mV and started stimulation very low. After finding the minimal stimulation strength, we observed activation of a small number of synapses mixed randomly with response ‘failures,’ presumably caused by failure of vesicular release (Fig. 3A, lower panel). We then switched to +40 mV. The relative abundance of silent (NMDAR-only) synapses was estimated from the difference between the failure rates at −70 mV and +40 mV (Fig. 3B). A clear reduction in the failure rate from four out of five P4–6 control mice was observed when the holding potential was switched from −70 mV to +40 mV (Fig. 3B, left panel). The difference in failure rates, which is proportional to the percentage of thalamocortical synapses that are silent, was 22.5 ± 4% (n = 5; Fig. 3C) in control neurons. Interestingly, we observed increased failure rates at +40 mV in CxAKAP5KO pups (Fig. 3B, right panel). The difference in failure rates between −70 mV and +40 mV in CxAKAP5KO mice was 15.4% ± 7% (n = 6; Fig. 3C). These results revealed that more NMDA-
4. Discussion In this study, we produced a neocortical-specific knockout of AKAP5 and examined its role in barrel map formation and the electrophysiological maturation of thalamocortical synapses in the somatosensory cortex. We found impaired AMPA/NMDA ratios in evoked thalamocortical synapse responses in the somatosensory cortex of CxAKAP5KO mice. AMPA receptor-mediated miniEPSCs fail to increase during development and the number of NMDA-only silent synapses increased in these mice. The dendritic morphology of layer IV neurons in the somatosensory cortex of CxAKAP5KO mice is less asymmetric in shape with longer total length and fewer branch points at the second postnatal week. These results showed that AKAP5 is important for promoting development of electrophysiological maturation of thalamocortical synapses and morphological maturation of layer IV neurons in the somatosensory cortex. Previous studies on genetically modified mice have uncovered molecular mechanisms for barrel map formation that involve two major
Fig. 3. Silent synapses are increased at thalamocortical synapses of CxAKAP5KO mice. (A) Sample traces of AMPA receptor miniEPSC (upper panel) and record of success and failure under minimal stimulation (B). Summary of failure rates at -70mv and +40mv holding potentials (B). Difference in failure rates between controls and CxAKAP5KO mice (C), error bar: standard deviation. 4
Neuroscience Letters 709 (2019) 134374
M. Zhang, et al.
Fig. 4. Abnormal morphology development of cortical L4 neurons at P14. (A, B) Fluorescent images of biocytin-filled L4 neurons in the somatosensory cortex. (C, D) 3D morphology reconstruction of the dendritic trees of L4 neurons in (A) and (B). Dendritic asymmetric index is reduced in CxAKAP5KO L4 neurons (E). Total dendritic length is increased in CxAKAP5KO L4 neurons (F). Maximum dendritic span is increased in CxAKAP5KO L4 neurons (G), however, total branch points are reduced in CxAKAP5KO L4 neurons (H). Scale bar (A, B) = 50 μm. **, P < 0.05; ***, P < 0.01 (unpaired student’s t-test).
5
Neuroscience Letters 709 (2019) 134374
M. Zhang, et al.
Morphologically, layer IV neurons in the barrel cortex show similarly reduced dendritic asymmetry in CxAKAP5KO mice. Similar morphological abnormality of layer IV neurons has been found in PKARIIβKO and CxAC1 KO mice. These results further support that AKAP5 mediates PKA activation in regulating somatosensory barrel map formation. In summary, our results demonstrated that cortical AKAP5 expression is essential in the physiological and morphological maturation of the somatosensory cortex during postnatal development.
pathways: cAMP-PKA [4,6,8] and serotonin (5-HT) pathways [6,23–25]. Mice lacking AC1 (brl) do not form barrels pre- and postsynaptically [8,26]. The type IIβ regulatory subunit of PKA knockout mice (PKARIIβKO mice) show defects in postsynaptic barrel maps but presynaptical barrel maps are largely normal [6]. Layer IV neurons of cortex-specific knockout of an NMDA subunit (NR1) (Cx-NR1 KO) failed to aggregate into barrels, but TCA afferent clustering was normal [25]. These studies demonstrated that proper layer IV barrel organization can be disrupted independent of a gross defect in TCA patterning. NMDA–AC1–PKA signaling may be a central molecular pathway underlying activity-dependent regulation of barrel formation in the neocortex. Serotonin (5-HT) signaling is another pathway that is strongly implicated in barrel map development. During early postnatal development, 5-HT transporters (5-HTTs) are highly expressed at axon terminals of thalamocortical synapses. 5-HTTs take up excessive 5-HT in synapse clefts. Barrels do not form in single or double KOs of 5-HTT and monoamino oxidase A (MAOA), but can be rescued by knocking out the 5-HT1B receptor in 5-HTT and MAOA single or double mutants [27–29]. However, 5HT1B mutant mice themselves have normal barrel maps. These data indicate that 5-HT1B is essential for the disruption of barrels in the 5-HT signaling pathway, but is not necessary for barrel formation [29]. It is not clear whether these two pathways cross talk. Expression of AKAP5 during the developing cerebral cortex is increased during postnatal development (mouse ENCODE transcriptome data). We observed that the cell bodies of layer IV neurons failed to aggregate in CxAKAP5KO somatosensory cortices, indicating the prominent effect of AKAP5 in topographic map formation. The mechanism underlying this effect is an interesting question for further study. AKAP5 is believed to be a postsynaptic scaffold protein that serves as a signal hub that links, not only the PKA kinase pathway, but also PKC and other second messenger systems. The 5-HT1B receptor is a G-protein coupled receptor. Whether AKAP5 regulation of topographic map formation of the somatosensory cortex occurs merely through the cAMP-PKA pathway or the 5-HT signaling pathway or both is not yet clear. To answer this question, for example, examining barrel map formation in mice with a mutated PKA binding site in the AKAP5 protein will be an interesting future experiment. Lacking AKAP5 in the neocortex must not affect afferent clustering in the brain stem and spinal cord because barrelloids and barrelettes in CxAKAP5KO mice were largely normal, suggesting that the function of AKAP5 in topographic map formation is specific to the neocortex. Electrophysiological analysis of AMPA receptor responses of thalamocortical synapses in CxAKAP5KO mice revealed that AMPA receptors failed to increase during development. Not only does the AMPA/NMDA ratio in evoked thalamocortical synapses of CxAKAP5KO mice maintain immature stages, but silent synapses also fail to decrease in these mice. The maturation of thalamocortical synapses involves trafficking and insertion of AMPA receptors into the postsynaptic membrane. Our results are in line with a recent publication in which AKAP5 was shown to control AMPA receptor phosphorylation and cell surface targeting through cAMP-PKA signaling [16]. Synaptic maturation in the somatosensory cortex is thought to occur through LTP-like mechanisms that regulate the trafficking of AMPA receptors into the synapse [4,7,30]. Mice with a global knockout of PKARII have deficits in synaptic plasticity and TC synapse maturation. CxAKAP5KO mice show immature electrophysiological phenotypes similar to brl, CxAC1 KO, and PKARII β KO mice [3,4,25]. It is intriguing that we observed increased numbers of silent synapses in CxAKAP5KO mice compared to littermate controls. In contrast, brl mice do not form barrel patterns either pre- or postsynaptically. One possible reason for this difference could be that brlmice might have severe disruption of NMDA receptor expression to begin with since AC1 expression is globally absent in brl mice. Our results extend that molecular mechanisms known to regulate TC synapse maturation, which may partly associate with AC1/cAMP/PKA signaling.
Acknowledgements This work was supported by grants from the Natural Science Foundation of Anhui Province #KJ2018A0801 (MZ) and the National Natural Science Foundation of China: #31671064 (HL) and #81471159 (HL). Hong Li conceived the study and wrote the manuscript. Min Zhang, Meifang Lu, Hao Huang, and Xiaoyan Liu performed the experiments, Haoran Su carried out the statistical analysis. All authors have read and approved the final version of the manuscript. Hong Li’s present address: Department of Neuroscience, Yale University, U.S.A. Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.neulet.2019.134374. References [1] R.F. Watson, R.M. Abdel-Majid, M.W. Barnett, B.S. Willis, A. Katsnelson, T.H. Gillingwater, G.S. McKnight, P.C. Kind, P.E. Neumann, Involvement of protein kinase a in patterning of the mouse somatosensory cortex, J. Neurosci. 26 (2006) 5393–5401, https://doi.org/10.1523/JNEUROSCI.0750-06.2006. [2] T. Iwasato, M. Inan, H. Kanki, R.S. Erzurumlu, S. Itohara, M.C. Crair, Cortical adenylyl cyclase 1 is required for thalamocortical synapse maturation and aspects of layer IV barrel development, J. Neurosci. 28 (2008) 5931–5943, https://doi.org/ 10.1523/JNEUROSCI.0815-08.2008. [3] M. Inan, H.-C. Lu, M.J. Albright, W.-C. She, M.C. Crair, Barrel map development relies on protein kinase a regulatory subunit IIbeta-Mediated cAMP signaling, J. Neurosci. 26 (2006) 4338–4349, https://doi.org/10.1523/JNEUROSCI.3745-05. 2006. [4] H.-C. Lu, W.-C. She, D.T. Plas, P.E. Neumann, R. Janz, M.C. Crair, Adenylyl cyclase I regulates AMPA receptor trafficking during mouse cortical “barrel” map development, Nat. Neurosci. 6 (2003) 939–947, https://doi.org/10.1038/nn1106. [5] T.A. Woolsey, H. Van der Loos, The structural organization of layer IV in the somatosensory region (SI) of mouse cerebral cortex. The description of a cortical field composed of discrete cytoarchitectonic units, Brain Res. 17 (1970) 205–242 (Accessed September 29, 2018), http://www.ncbi.nlm.nih.gov/pubmed/4904874. [6] H. Li, M.C. Crair, How do barrels form in somatosensory cortex? Ann. N. Y. Acad. Sci. 1225 (2011) 119–129, https://doi.org/10.1111/j.1749-6632.2011.06024.x. [7] H. Li, S. Fertuzinhos, E. Mohns, T.S. Hnasko, M. Verhage, R. Edwards, N. Sestan, M.C. Crair, Laminar and columnar development of barrel cortex relies on thalamocortical neurotransmission, Neuron 79 (2013) 970–986, https://doi.org/10. 1016/j.neuron.2013.06.043. [8] T.K. Hensch, Critical period regulation, Annu. Rev. Neurosci. 27 (2004) 549–579, https://doi.org/10.1146/annurev.neuro.27.070203.144327. [9] R.M. Abdel-Majid, W.L. Leong, L.C. Schalkwyk, D.S. Smallman, S.T. Wong, D.R. Storm, A. Fine, M.J. Dobson, D.L. Guernsey, P.E. Neumann, Loss of adenylyl cyclase I activity disrupts patterning of mouse somatosensory cortex, Nat. Genet. 19 (1998) 289–291, https://doi.org/10.1038/980. [10] A. Datwani, T. Iwasato, S. Itohara, R.S. Erzurumlu, Lesion-induced thalamocortical axonal plasticity in the S1 cortex is independent of NMDA receptor function in excitatory cortical neurons, J. Neurosci. 22 (2002) 9171–9175 (Accessed September 29, 2018), http://www.ncbi.nlm.nih.gov/pubmed/12417641. [11] L.-J. Lee, T. Iwasato, S. Itohara, R.S. Erzurumlu, Exuberant thalamocortical axon arborization in cortex-specific NMDAR1 knockout mice, J. Comp. Neurol. 485 (2005) 280–292, https://doi.org/10.1002/cne.20481. [12] H.R. Robertson, E.S. Gibson, T.A. Benke, M.L. Dell’Acqua, Regulation of postsynaptic structure and function by an A-Kinase anchoring protein-membraneAssociated guanylate kinase scaffolding complex, J. Neurosci. 29 (2009) 7929–7943, https://doi.org/10.1523/JNEUROSCI.6093-08.2009. [13] Y. Lu, X. Zha, E.Y. Kim, S. Schachtele, M.E. Dailey, D.D. Hall, S. Strack, S.H. Green, D.A. Hoffman, J.W. Hell, A kinase anchor protein 150 (AKAP150)-associated protein kinase A limits dendritic spine density, J. Biol. Chem. 286 (2011) 26496–26506, https://doi.org/10.1074/jbc.M111.254912. [14] M.G. Gold, F. Stengel, P.J. Nygren, C.R. Weisbrod, J.E. Bruce, C.V. Robinson, D. Barford, J.D. Scott, Architecture and dynamics of an A-kinase anchoring protein 79 (AKAP79) signaling complex, Proc. Natl. Acad. Sci. 108 (2011) 6426–6431, https://doi.org/10.1073/pnas.1014400108.
6
Neuroscience Letters 709 (2019) 134374
M. Zhang, et al.
[15] M. Weisenhaus, M.L. Allen, L. Yang, Y. Lu, C.B. Nichols, T. Su, J.W. Hell, G.S. McKnight, Mutations in AKAP5 disrupt dendritic signaling complexes and lead to electrophysiological and behavioral phenotypes in mice, PLoS One 5 (2010) e10325, , https://doi.org/10.1371/journal.pone.0010325. [16] G.H. Diering, A.S. Gustina, R.L. Huganir, PKA-GluA1 coupling via AKAP5 controls AMPA receptor phosphorylation and cell-surface targeting during bidirectional homeostatic plasticity, Neuron 84 (2014) 790–805, https://doi.org/10.1016/j. neuron.2014.09.024. [17] A. Ostroveanu, E.A. Van der Zee, A.M. Dolga, P.G.M. Luiten, U.L.M. Eisel, I.M. Nijholt, A-kinase anchoring protein 150 in the mouse brain is concentrated in areas involved in learning and memory, Brain Res. 1145 (2007) 97–107, https:// doi.org/10.1016/j.brainres.2007.01.117. [18] B.J. Tunquist, N. Hoshi, E.S. Guire, F. Zhang, K. Mullendorff, L.K. Langeberg, J. Raber, J.D. Scott, Loss of AKAP150 perturbs distinct neuronal processes in mice, Proc. Natl. Acad. Sci. 105 (2008) 12557–12562, https://doi.org/10.1073/pnas. 0805922105. [19] M.A. Xu-Friedman, W.G. Regehr, Presynaptic strontium dynamics and synaptic transmission, Biophys. J. 76 (1999) 2029–2042, https://doi.org/10.1016/S00063495(99)77360-1. [20] J.T. Isaac, M.C. Crair, R.A. Nicoll, R.C. Malenka, Silent synapses during development of thalamocortical inputs, Neuron 18 (1997) 269–280 (Accessed September 23, 2018), http://www.ncbi.nlm.nih.gov/pubmed/9052797. [21] R.W. Rhoades, C.A. Bennett-Clarke, M.Y. Shi, R.D. Mooney, Effects of 5-HT on thalamocortical synaptic transmission in the developing rat, J. Neurophysiol. 72 (1994) 2438–2450, https://doi.org/10.1152/jn.1994.72.5.2438. [22] C.L. Young-Davies, C.A. Bennett-Clarke, R.D. Lane, R.W. Rhoades, Selective facilitation of the serotonin(1B) receptor causes disorganization of thalamic afferents and barrels in somatosensory cortex of rat, J. Comp. Neurol. 425 (2000) 130–138 (Accessed October 1, 2018), http://www.ncbi.nlm.nih.gov/pubmed/10940947. [23] C. Lebrand, O. Cases, C. Adelbrecht, A. Doye, C. Alvarez, S. El Mestikawy, I. Seif,
[24]
[25]
[26]
[27]
[28]
[29]
[30]
7
P. Gaspar, Transient uptake and storage of serotonin in developing thalamic neurons, Neuron 17 (1996) 823–835 (Accessed October 1, 2018), http://www.ncbi. nlm.nih.gov/pubmed/8938116. T. Iwasato, M. Inan, H. Kanki, R.S. Erzurumlu, S. Itohara, M.C. Crair, Cortical adenylyl cyclase 1 is required for thalamocortical synapse maturation and aspects of layer IV barrel development, J. Neurosci. 28 (2008) 5931–5943, https://doi.org/ 10.1523/JNEUROSCI.0815-08.2008. T. Iwasato, A. Datwani, A.M. Wolf, H. Nishiyama, Y. Taguchi, S. Tonegawa, T. Knöpfel, R.S. Erzurumlu, S. Itohara, Cortex-restricted disruption of NMDAR1 impairs neuronal patterns in the barrel cortex, Nature 406 (2000) 726–731, https:// doi.org/10.1038/35021059. C. Alvarez, T. Vitalis, E.A. Fon, N. Hanoun, M. Hamon, I. Seif, R. Edwards, P. Gaspar, O. Cases, Effects of genetic depletion of monoamines on somatosensory cortical development, Neuroscience 115 (2002) 753–764 (Accessed October 1, 2018), http://www.ncbi.nlm.nih.gov/pubmed/12435414. N. Salichon, P. Gaspar, A.L. Upton, S. Picaud, N. Hanoun, M. Hamon, E. De Maeyer, D.L. Murphy, R. Mossner, K.P. Lesch, R. Hen, I. Seif, Excessive activation of serotonin (5-HT) 1B receptors disrupts the formation of sensory maps in monoamine oxidase a and 5-ht transporter knock-out mice, J. Neurosci. 21 (2001) 884–896 (Accessed October 1, 2018), http://www.ncbi.nlm.nih.gov/pubmed/11157075. T. Vitalis, C. Fouquet, C. Alvarez, I. Seif, D. Price, P. Gaspar, O. Cases, Developmental expression of monoamine oxidases A and B in the central and peripheral nervous systems of the mouse, J. Comp. Neurol. 442 (2002) 331–347, https://doi.org/10.1002/cne.10093. C.B. Boylan, K.L. Kesterson, T.A. Crnko-Hoppenjans, M. Ke, T. Rizk, R.D. Mooney, R.W. Rhoades, The cortical vibrissae representation is normal in transgenic mice lacking the 5-HT(1B) receptor, Brain Res. Dev. Brain Res. 120 (2000) 91–93 (Accessed October 1, 2018), http://www.ncbi.nlm.nih.gov/pubmed/10727735. M.C. Crair, R.C. Malenka, A critical period for long-term potentiation at thalamocortical synapses, Nature 375 (1995) 325–328, https://doi.org/10.1038/375325a0.