Multifunctional role of protein kinase C in regulating the formation and maturation of specific synapses

Abstract

Target-dependent increases in axon growth and varicosities accompany the formation of functional synapses between Aplysia sensory neurons and specific postsynaptic neurons (L7 and not L11). The enhanced growth is regulated in part by a target-dependent increase in the secretion of sensorin, the sensory neuron neuropeptide. We report here that protein kinase C (PKC) activity is required for synapse formation by sensory neurons with L7 and for the target-dependent increases in sensorin synthesis and secretion. Blocking PKC activity reversibly blocked synapse formation and axon growth of sensory neurons contacting L7, but did not affect axon growth of sensory neurons contacting L11 or axon growth of the postsynaptic targets. Blocking PKC activity also blocked the target-induced increase in sensorin synthesis and secretion. Sensorin then activates additional signaling pathways required for synapse maturation and synapse-associated growth. Exogenous anti-sensorin antibody blocked target-induced activation and translocation into sensory neuron nuclei of p42/44 mitogen-activated protein kinase (MAPK), attenuated synapse maturation, and curtailed growth of sensory neurons contacting L7, but not the growth of sensory neurons contacting L11. Inhibitors of MAPK or phosphoinositide 3-kinase also attenuated synapse maturation and curtailed growth and varicosity formation of sensory neurons contacting L7, but not growth of sensory neurons contacting L11. These results suggest that PKC activity regulated by specific cell-cell interactions initiates the formation of specific synapses and the subsequent synthesis and release of a neuropeptide to activate additional signaling pathways required for synapse maturation.

Figure 1.

Specific target induces an increase in sensorin expression. A, B, Phase-contrast micrographs of sensory neuron-target cultures (SN-L7 and SN-L11) 16 h after plating in the same culture dish. After fixation, the culture was processed for sensorin immunoreactivity. C, D, Sensorin immunoreactivity is greater in sensory neurons cultured with L7 compared with L11. Staining is confined to the sensory neurons. An ANOVA indicated a significant effect of target on sensorin immunostaining (df = 2, 16; F = 18.907; p < 0.001). A comparison for each compartment indicated that staining intensity was significantly greater when sensory neurons contact L7 (p < 0.01 for each compartment). The insets (a–c for SN-L7 and d–f for SN-L11) reveal the differences in staining in the sensory neuron axons and distal processes. Note that the sensory neuron axon that contacts L7 has intense staining within large granules. The axon stump (contacting the major processes of each target) accumulates sensorin, but the regenerated neurites contacting L7 are both more numerous and heavily stained. Scale bars, 25 μm.

Figure 2.

PKC activity mediates the target-induced increase in sensorin expression. Phase-contrast and corresponding epifluorescent images of sensorin immunoreactivity 16 h after plating cultures in Cont, the PKCI chelerythrine (2 μm), or the PI3K inhibitor LY294002 (2 μm; PI3KI). Staining in the sensory neuron cell body, axon, and axon stump are markedly reduced with PKCI. Note the absence of growth from the axon stump with treatment with PKCI. In contrast, staining for sensorin in the cell body and axon with control or PI3KI treatments are strong. Intense staining can be seen in numerous granules distributed throughout the axon in both the control and PI3KI-treated cells. Although overall growth in the presence of PI3KI was reduced (see below), staining in the varicosities remained strong. An ANOVA indicated a significant effect of treatment on sensorin staining (df = 4, 34; F = 82.767; p < 0.001). Individual comparisons indicated that staining in the cell bodies and axons of controls were significantly greater than staining in the cell bodies and axons of sensory neurons treated with PKCI (F = 45.522, p < 0.01 and F = 22.642, p < 0.01), but not significantly different from sensorin staining observed in cell bodies and axons of cultures treated with PI3KI. Staining for sensorin in the varicosities of sensory neurons treated with PI3KI was not significantly different from the staining in the varicosities of sensory neurons treated with control. Scale bars, 80 μm.

Figure 3.

PKC activity is required to initiate synapse formation and synapse-associated growth. A, B, Nomarski-contrast (top) and corresponding epifluorescent (bottom) images of sensory neurites (within outlined area in each micrograph) after injection of dye into sensory neurons cultured with either L7 (SN-L7; A) or L11 (SN-L11; B), plated in the same dish, and treated with control or the PKC inhibitor chelerythrine (2 μm) for 16 h. In the Cont, contact with L7 induces significantly more growth and varicosities compared with contact with L11. Incubation with chelerythrine blocked all growth from the sensory neurons contacting L7 with the exception of a varicose-like protrusion from the axon stump. In contrast, growth from the sensory neuron contacting L11 in the same culture dish was unaffected by the chelerythrine. In the presence of chelerythrine, growth from the sensory neuron contacting L11 was greater than the growth of the sensory neuron contacting L7. Scale bars, 50 μm. C, The PKC inhibitor significantly reduced sensory neuron growth and varicosity formation only when sensory neurons contact L7 but not when they contact L11. The histogram summarizes the effect of the PKC inhibitor on sensory neuron growth and varicosity formation when contacting different targets. An ANOVA indicated a significant effect of target and PKC inhibitor on axon growth and number of varicosities formed (df = 3, 35; F = 93.722; p < 0.001). Individual comparisons indicated that chelerythrine compared with Cont significantly reduced growth and number of varicosities when sensory neurons contacted L7 (F = 94.355, p < 0.001 and F = 45.78, p < 0.001), whereas there was no significant difference between Cont and chelerythrine on sensory neuron growth and varicosities formed when sensory neurons contact L11. As expected for cultures treated with Cont, growth and varicosity formation were greater when sensory neurons contacted L7 compared with L11 (F = 24.044, p < 0.01 and F = 8.848, p < 0.01, respectively). In contrast, when incubated with chelerythrine, growth of sensory neurons contacting L11 was now significantly greater than the growth in contact with L7 (F = 36.29, p < 0.01 for axon growth and F = 6.098, p < 0.01 for varicosities).

Figure 4.

Synapse formation, growth, and sensorin expression resumes after washout of the PKC inhibitor. A, B, After washout of the PKCI (chelerythrine), sensory neurons form functional connections with L7. EPSPs were recorded once again in the same cultures after washout of the control medium or PKCI. Incubation with the PKCI resulted in no detectable EPSP. After washout, EPSP amplitudes increased in control cultures, and EPSPs were now recorded in cultures treated initially with the PKCI. Calibration: 10 mV, 25 ms. The histogram in B indicates the reversal of PKC inhibition on synapse formation. ANOVA indicated a significant effect of treatment on EPSP amplitude (df = 1, 17; F = 29.59; p < 0.001). Individual comparisons indicated that the presence of the PKCI (18 h) blocked synapse formation compared with control (F = 5.618; p < 0.01), but EPSP amplitude after washout of the inhibitor (PKCI-Washout; 42 h) was significantly increased (F = 13.923; p < 0.01) but was not significantly different from the EPSP amplitude for controls after 18 h. C, Growth and varicosity formation resume after washout of the PKCI. Nomarski-contrast and epifluorescent images (within outlined area in each micrograph) after dye injections reveal that growth and varicosities are considerable after PKCI washout and is consistent with a reversal of the blockade by the inhibitor. Because of the additional day of growth, control cultures have more growth and have formed more varicosities (F = 6.709, p < 0.01 and F = 7.459, p < 0.01). Scale bars, 50 μm. D, E, Sensorin expression returns to control levels after washout of the PKCI. Epifluorescent images of sensorin immunostaining in a 42 h control culture and in a 42 h culture 24 h after washout of the PKCI (D). Staining intensities in all compartments are comparable. Scale bar, 50 μm. The histogram in E indicates that staining in each compartment after PKCI washout was not significantly different from the staining in controls (normalized to 100%).

Figure 5.

Exogenous anti-sensorin Ab blocked synapse maturation and synapse-associated growth. A, Synaptic efficacy is attenuated significantly with incubation with SEN Ab. Representative EPSPs were recorded in L7 after 20 h in the presence of Cont Ab or SEN Ab. Calibration: 5 mV, 25 ms. Incubation with SEN Ab significantly reduced EPSP amplitude (F = 14.509; p < 0.01). B, SEN Ab reduces significantly growth and varicosity formation when sensory neurons contact L7, but not L11. The histograms summarize axon growth and the number of varicosities formed by sensory neurons in contact with the major processes of the targets. An ANOVA indicated a significant effect of target and SEN Ab on synaptic efficacy, neuritic growth and number of varicosities formed (df = 6, 60; F = 58.125; p < 0.001). Individual comparisons indicated that SEN Ab compared with Cont Ab significantly reduced growth and number of varicosities when sensory neurons contacted L7 (F = 65.559, p < 0.001 and F = 64.318, p < 0.001, respectively), whereas there was no significant difference between Cont Ab and SEN Ab on growth and varicosities formed when sensory neurons contact L11. As expected for cultures treated with Cont Ab, growth and varicosity formation was greater when sensory neurons contacted L7 compared with L11 (F = 9.369, p < 0.01 and F = 20.9, p < 0.01, respectively). In contrast, when incubated with SEN Ab, growth of sensory neurons contacting L11 was now significantly greater than the growth in contact with L7 (F = 12.853; p < 0.01). C, Target and exogenous SEN Ab affect sensory neuron growth. Nomarski-contrast and corresponding epifluorescent image of sensory neurites (within outlined area in each micrograph) after dye injection are displayed for each condition. Each column contains the pair of cocultures (SN-L7 and SN-L11) in the same dish treated with either Cont Ab or SEN Ab. Note that contact with L7 induced significantly more growth and more varicosities compared with contact with L11 when cultures are incubated with Cont Ab. Incubation with SEN Ab significantly reduced growth of the sensory neuron contacting L7 to a level that was now less than the growth detected for the other sensory neuron in the same dish contacting L11. Scale bars, 100 μm.

Figure 6.

PKC activity is required both for initiating synapse formation and later for the secretion of sensorin. A, Addition of sensorin at plating does not rescue the blockade of synapse formation by the PKCI. Representative EPSPs recorded in L7 after stimulation of each sensory neuron (stimulation artifact) 16 h after plating cocultures in control, PKCI chelerythrine, or chelerythrine plus sensorin neuropeptide (80 ng/ml; PKCI + SEN) added to the cultures at time of plating. Very weak EPSPs were evoked in cultures treated with PKCI with or without sensorin. Calibration: 10 mV, 25 ms. The histogram summarizes the average EPSP amplitudes recorded after treatments. Incubation with the PKC inhibitor (PKCI or PKCI + SEN) blocked synapse formation. An ANOVA indicated a significant effect of treatment (df = 2, 60; F = 79.727; p < 0.001). Individual comparisons indicated that EPSP amplitude with control treatment was significantly greater than treatment with PKCI (F = 67.136; p < 0.01) or PKCI + SEN (47.034; p < 0.01). There was no significant difference in the EPSP amplitude in cultures treated with PKCI compared with cultures treated with PKCI + SEN. B, Addition of sensorin at 6 h after plating rescues the block of synapse formation by the PKC inhibitor. Representative EPSPs recorded in L7 after stimulation of each sensory neuron (simulation artifact) 22 h after plating cocultures in control, PKCI chelerythrine added 6 h after plating or PKCI + SEN (80 ng/ml) added to the cultures 6 h after time of plating. Note that PKCI added at 6 h reduced EPSP amplitude compared with control, whereas the addition of sensorin reversed the block produced by PKCI. Calibration: 10 mV, 25 ms. The histogram summarizes the average EPSP amplitudes recorded after treatments. Incubation with the PKC inhibitor 6 h after plating allowed the initiation of synapse formation but attenuated subsequent synapse maturation. The addition of sensorin reversed most of the attenuation produced by PKCI. An ANOVA indicated a significant effect of treatment (df = 2, 29; F = 12.195; p < 0.001). Individual comparisons indicated that EPSP amplitude with control treatment was significantly greater than treatment with PKCI at 6 h (F = 11.738; p < 0.01), but not significantly different from treatment with and PKCI + SEN at 6 h (p > 0.3). The addition of sensorin together with the PKCI at 6 h significantly increased EPSP amplitude (F = 4.64; p < 0.03) compared with incubation with PKCI alone at 6 h.

Figure 7.

PKA inhibitor failed to affect synapse formation, growth, and varicosity formation by sensory neurons and sensorin expression. EPSP amplitude was not affected by incubation with the PKA inhibitor (5 μm KT5720) for 18 h after plating. A, Growth and varicosity formation was not affected by incubation with the PKA inhibitor for 18 h. Nomarski-contrast and corresponding epifluorescent image of sensory neurites (within outlined area in each micrograph) after dye injection are displayed for each condition. Growth from the axon stump of both sensory neurons is extensive and has numerous varicosities. ANOVA indicated no significant effect of treatment, and individual comparisons indicated no significant difference in growth and varicosity formation by the sensory neurons. Scale bars, 100 μm. B, C, Sensorin expression in sensory neurons is not affected by incubation with the PKA inhibitor KT5720. Epifluorescent images of sensorin immunostaining in a control culture and in a culture incubated with the PKA inhibitor KT5720 (B). Staining intensities in all compartments are comparable. Scale bar, 100 μm. The histogram in C indicates that staining in each compartment of the sensory neurons exposed to KT5720 was not significantly different from the staining in controls (normalized to 100%).

Figure 8.

Contact with L7, which increased the secretion of sensorin, resulted in the activation and translocation of p42/44 MAPK in sensory neurons. A, Epifluorescent images of sensory neurons immunostained for phospho-MAPK. The sensory neurons contacting L7 (SN-L7) and the sensory neuron contacting L11 (SN-L11) were plated in the same dish and exposed to control Ab for 20 h before fixation. The other dishes contained SN-L7 cocultures incubated with SEN Ab. Note that staining for phospho-MAPK in the nuclei of the SNs is less than staining in the cytoplasm when SNs contact L11 or when SN-L7 cocultures are incubated with SEN Ab. Scale bar, 25 μm. B, Summary of the effect of target and incubation with SEN Ab on staining intensity (arbitrary units) for phospho-MAPK in the entire cell body, nucleus, and cytoplasm of sensory neurons. An ANOVA indicated a significant effect of treatment and target (df = 4, 30; F = 41.493; p < 0.001). Individual comparisons indicated that overall cell body and nuclear staining were significantly greater in sensory neurons contacting L7 incubated with Cont Ab compared with sensory neurons contacting L7 and incubated with SEN Ab (F = 9.605, p < 0.01 and F = 46.78, p < 0.01, respectively) or compared with sensory neurons contacting L11 and incubated with Cont Ab (F = 11.272, p < 0.01 and F = 40.259, p < 0.01, respectively). Staining and its distribution in sensory neurons contacting L7 and incubated with SEN Ab were not significantly different from the staining in sensory neurons contacting L11.

Figure 9.

MAPK inhibitor U0126 blocked synapse maturation and synapse-associated growth. A, Synaptic efficacy was attenuated significantly with incubation with inhibitor U0126 compared with the control compound U0124. EPSPs were recorded in L7 with a single action potential in each sensory neuron plated with L7 for 18 h in the presence of inhibitor U0126 (2 μm) or control U0124 (2 μm). Incubation U0126 significantly reduced but did not abolish synaptic efficacy (F = 20.495; p < 0.01). Calibration: 5 mV, 25 ms. B, Incubation with MAPK inhibitor U0126 reduced significantly both growth and varicosity formation when sensory neurons contacted L7, but not L11. The histograms summarize axon growth and the number of varicosities formed by sensory neurons in contact with the major processes of the targets. An ANOVA indicated a significant effect of the target and inhibitor U0126 on synaptic efficacy, axon growth, and number of varicosities formed (df = 6, 52; F = 58.591; p < 0.001). Individual comparisons indicated that MAPK inhibitor U0126 compared with control U0124 significantly reduced growth and number of varicosities when sensory neurons contacted L7 (F = 60.601, p < 0.001 and F = 38.351, p < 0.01, respectively), whereas there was no significant difference between incubation with U0126 and U0124 on growth and number of varicosities formed by sensory neurons contacting L11. As expected for cultures treated with the control U0124, growth and varicosity formation were greater when sensory neurons contacted L7 compared with L11 (F = 7.408, p < 0.01 and F = 16.932, p < 0.01, respectively). In contrast, when incubated with the inhibitor U0126, growth by sensory neurons contacting L11 was now significantly greater than sensory neuron growth in contact with L7 in the same culture dishes (F = 21.322; p < 0.01). C, Target and inhibitor U0126 affect sensory neuron growth. Nomarski-contrast and corresponding epifluorescent image of sensory neurites (within outlined area in each micrograph) after dye injection are displayed for each condition. Each column contains the pair of cocultures (SN-L7 and SN-L11) in the same dish treated with either control U0124 or MAPK inhibitor U0126. Contact with L7 induces significantly more growth and more varicosities compared with L11 when cultures are incubated with U0124. Incubation with U0126 significantly reduced growth of sensory neurons contacting L7 to a level that was now less than the growth detected for the sensory neuron in the same dish contacting L11. Scale bars, 100 μm.

Figure 10.

PI3K inhibitor LY294002 blocked synapse maturation and synapse-associated growth. A, Synaptic efficacy was attenuated significantly with incubation with the PI3K inhibitor compared with the control. EPSPs were recorded in L7 with a single action potential in each sensory neuron plated with L7 for 18 h in the presence of inhibitor LY294002 (2 μm) or control. Incubation with the inhibitor significantly reduced synaptic efficacy (F = 50.964; p < 0.01). Calibration: 5 mV, 25 ms. B, Incubation with PI3K inhibitor reduced significantly both growth and varicosity formation when sensory neurons contacted L7, but not L11. The histograms summarize axon growth and the number of varicosities formed by sensory neurons in contact with the major processes of the targets. An ANOVA indicated a significant effect of the target and PI3K inhibitor on synaptic efficacy, axon growth, and number of varicosities formed (df = 3, 19; F = 24.937; p < 0.001). Individual comparisons indicated that the PI3K inhibitor compared with control significantly reduced growth and number of varicosities when sensory neurons contacted L7 (F = 26.167, p < 0.01 and F = 23.262, p < 0.01, respectively), whereas there was no significant difference between incubation with LY294002 and control on growth and number of varicosities formed by sensory neurons contacting L11. As expected for cultures treated with the control, growth and varicosity formation were greater when sensory neurons contacted L7 compared with L11 (F = 5.509, p < 0.02 and F = 7.955, p < 0.01, respectively). In contrast, when incubated with the PI3K inhibitor, growth by sensory neurons contacting L11 was now significantly greater than sensory neuron growth in contact with L7 in the same culture dishes (F = 6.491, p < 0.01 for growth and F = 3.995, p < 0.05 for varicosities). C, Target and inhibitor LY294002 affect sensory neuron growth. Nomarski-contrast and corresponding epifluorescent image of sensory neurites (within outlined area in each micrograph) after dye injection are displayed for each condition. Each column contains the pair of cocultures (SN-L7 and SN-L11) in the same dish treated with either control or PI3K inhibitor LY294002. Contact with L7 induces significantly more growth and more varicosities compared with L11 when cultures are incubated with control. Incubation with LY294002 significantly reduced growth of sensory neurons contacting L7 to a level that was now less than the growth detected for the sensory neuron in the same dish contacting L11. Scale bars, 100 μm.

Figure 11.

Summary of the events and signaling pathways mediating the formation and maturation of sensory neuron synapses with a specific postsynaptic target. Interaction between appropriate cells leads to PKC-dependent initiation of synapse formation and growth followed by a PKC-dependent increase in the synthesis and secretion of the neuropeptide sensorin [for the role of PKC activity in regulating sensorin synthesis and secretion in activity-dependent LTF, see Hu et al., (2007)]. We do not know whether enhanced or constitutive PKC activity is required to initiate synapse formation (dashed arrow a). Because sensorin application in the presence of PKC inhibition failed to initiate synapse formation, sensorin's capacity to modulate synapse maturation may require a prerequisite step that is completed during the initial formation of the synapse (dashed arrow b). PKA activity is required neither for initial synapse formation nor for the increase in sensorin synthesis and secretion. Secreted sensorin is required for synapse maturation and growth with the appropriate target by activating and translocating MAPK into the nuclei of the sensory neurons. PI3K activity is also required for synapse maturation, but we do not have direct evidence that PI3K is activated by sensorin during this phase of synapse formation (dashed arrow c). Activation of PI3K by sensorin is required for sensorin-dependent LTF at mature sensory neuron synapses (J. Y. Hu and S. Schacher, unpublished observations).