Volume 135, Issue 5, Pages 960-973 (November 2008) A New Component in Synaptic Plasticity: Upregulation of Kinesin in the Neurons of the Gill-Withdrawal Reflex  Sathyanarayanan V. Puthanveettil, Francisco J. Monje, Maria Concetta Miniaci, Yun-Beom Choi, Kevin A. Karl, Eugene Khandros, Mary Ann Gawinowicz, Michael P. Sheetz, Eric R. Kandel  Cell  Volume 135, Issue 5, Pages 960-973 (November 2008) DOI: 10.1016/j.cell.2008.11.003 Copyright © 2008 Elsevier Inc. Terms and Conditions

Figure 1 ApKLC2 and ApKHC1 Are Induced by 5HT (A) Semiquantitative RTPCR analysis of ApKHC1 and ApKLC2 gene expression. Total RNA was isolated from pleural ganglia or pleural sensory clusters at different time points (0 min and 30 min) after five pulses of 10 μM 5HT treatments and from untreated controls. ApC/EBP was used as a positive control. ApActin and sensorin mRNAs were used to normalize kinesin mRNA levels. (B) Fold increase in mRNA levels of ApKHC1 and ApKLC2. The intensity of ApKLC2 and ApKHC1 bands (shown in [A]) were quantified and normalized to sensorin levels using IMAGEQUANT. (C) Real-time PCR analysis of changes in transcript levels of ApKLC2 and ApKHC1 in response to 5× 5HT. ApGAPDH mRNA was used for normalization of data. Fold changes were calculated according to Pfaffl, 2001 and Pffafl et al., 2002. (D) mRNA in situ analysis of ApKHC1 mRNA expression in sensory neurons (SN) and motor neurons (MN). Confocal projection images are shown. (E) Quantitation of ApKHC1 mRNA staining in sensory and motor neurons (shown in [D]) by mRNA in situ hybridization. Mean fluorescence intensities of labeled probe hybridizing to targets were quantified by analyzing the confocal images using METAMORPH (n = 6, ∗p < 0.01 for both SN and MN, Student's t test). (F) Western blot analysis of changes in protein levels of kinesin in response to 5× 5HT. Synaptophysin and β tubulin protein levels were used as loading controls and for normalization of data. (G) Fold increase in the protein levels of ApKHC and ApKLC (shown in [F]) in 30 min of 5HT treatment. (H) Immunocytochemical analysis of induction of ApKLC and ApKHC protein in 30 min of bath application of 5× 5HT in sensory neurons. Confocal projection images of two representative examples of immunostaining of ApKHC and ApKLC in sensory neurons are shown. (I) Quantitation of immunocytochemistry data presented in (H). Scale bar = 50 μm. Student's t test was performed to test levels of significance among different populations of specimens. The error bars represent SEM. Cell 2008 135, 960-973DOI: (10.1016/j.cell.2008.11.003) Copyright © 2008 Elsevier Inc. Terms and Conditions

Figure 2 5HT Upregulates Vesicle Trafficking in Sensory Neurons Aplysia sensory neuron-motor neuron cocultures were used to observe vesicle trafficking in sensory neurons, before and after the five pulses of 10 μM 5HT, using video enhanced contrast differential interference contrast (VEC-DIC) microscopy. (A) Three video micrographs captured at 2 s intervals are shown. Tracking of ∼300 nm size vesicle (accompanied by an asterisk in the figure) in the sensory neuron axon. These vesicles move with a velocity of about 200–325 nm per second. The bigger vesicles present in the video micrograph did not show any mobility in 10 min. The scale bar represents 2.25 μm. (B) Vesicles that are moving toward the distal processes were manually counted (∼300 nm size vesicles shown in Figure 2A) for 10 min before and after 5HT treatment. The means of percentage change in the number of these vesicles transported (n = 7, ∗p < 0.05., Student's t test) in 10 min following 5HT treated or mock treated controls are shown as bar graphs. The error bars represent SEM. Cell 2008 135, 960-973DOI: (10.1016/j.cell.2008.11.003) Copyright © 2008 Elsevier Inc. Terms and Conditions

Figure 3 Kinesins Are Necessary for the Induction of Long-Term Memory Storage in Aplysia (A and B) ApKHC1 antisense (AS) oligonucleotides microinjected into presynaptic sensory neurons does not affect STF (A) but block induction of LTF (B). (C and D) ApKLC2 AS oligonucleotides microinjected into presynaptic sensory neurons does not affect STF (C) but block induction of LTF (D). (E) Microinjection of ApKHC1 AS into postsynaptic motor neuron (MN) block induction of LTF. Sense ApKHC1 and ApKLC2 oligos were used as control for microinjection and electrophysiology. Changes in EPSP amplitudes (mean ± SEM) are shown in bar graphs. Student's t test was performed to test the level of significance among different populations. ∗p < 0.05. (F) Antisense oligonucleotides against ApKHC1 were microinjected into sensory neurons 24 hr after the application of five pulses of 5HT does not affect persistence of LTF. There was no significant difference (paired Student's t test, p > 0.5) between the neurons that received anisense oligonucleotide injection and neurons treated with 5HT alone. Cell 2008 135, 960-973DOI: (10.1016/j.cell.2008.11.003) Copyright © 2008 Elsevier Inc. Terms and Conditions

Figure 4 Overexpression of Full-Length ApKHC1 in Sensory Neurons Causes an Increase in EPSPs (A) Different optical sections of ApKHC1-EGFP expressing sensory neuron (SN) shows that it does not localize in the nucleus and that ApKHC1-EGFP is present in distal processes of the sensory neuron. (B) ApKHC1-EGFP or EGFP plasmid were microinjected into sensory neuron connected to motor neuron (MN) and EPSPs were measured at different time points. Changes in EPSP amplitude (mean ± SEM) are shown as bar graphs. Cell 2008 135, 960-973DOI: (10.1016/j.cell.2008.11.003) Copyright © 2008 Elsevier Inc. Terms and Conditions

Figure 5 Protein Cargos of ApKHC (A) Cartoon representation of kinesin complex showing microtubule tracks, kinesin heavy- and light-chain and cargos. (B) Silver-stained gel showing isolation of kinesin complex from Aplysia pleural ganglia by coimmunoprecipitation (CoIP) with the ApKHC antibody. ApTOR (Aplysia Target Of Rapamycin) antibody was used as a control. CoIP (+ ApKHC Ab, + ApTOR Ab), no antibody control (beads only) and antibody alone (ApKHC Ab and ApTOR Ab) are shown. Protein bands corresponding to ApKHC (1) and ApKLC (2) were excised and verified by mass spectrometric sequencing. (C) Identification of protein cargos by candidate approach. ApKHC complex was immunoprecipitated and analyzed by western blot hybridization using antibodies for bassoon, piccolo, neurexin, neuroligin, K Channel, kinesin light chain and heavy chain. (D) Protein cargos are upregulated following 5HT application. Western blot analyses of soluble proteins (5–10 μg) isolated from pleural ganglia after 5HT treatments are shown. Synaptophysin and β tubulin were used as loading controls. (E) Coomasie staining of a typical gel used for this experiment. (F) Fold increase of protein levels of cargo in response to 5HT. Cargo protein levels were normalized to synaptophysin levels. The error bars represent SEM. (G and H) Piccolo levels are critical for the induction of LTF in Aplysia. Piccolo antisense (Pic-AS) oligonucleotides microinjected into sensory neurons did not affect STF (G) but blocked induction of LTF (H). A sense (Pic-S) piccolo oligo was used as control for microinjection and electrophysiology. Pic-AS injection into motor neurons did not show any effect on facilitation suggesting that its effect is specific to sensory neurons. Student's t test was performed to test the level of significance among different populations. ∗p < 0.05. Changes in EPSP amplitudes (mean ± SEM) are shown in bar graphs. Cell 2008 135, 960-973DOI: (10.1016/j.cell.2008.11.003) Copyright © 2008 Elsevier Inc. Terms and Conditions

Figure 6 Necessity and Sufficiency of ApKHC1 on LTF (A) CREB is required for the LTF at 12 and 24 hr. Microinjection of CRE elements into sensory neurons block the LTF at 12 or 24 hr induced by five pulses of 5HT. Mutant CRE (mCRE) that do not bind to phosphoCREB was used as a control. Neither the wild-type nor mutant CRE had any effect on basal EPSPs. (B) Role of CREB in the ApKHC1 induced increase in EPSPs. CRE injection into sensory neurons do not block ApKHC1 induced increase in EPSPs at 12 hr. (C) Piccolo AS oligo injection into sensory neurons do not block ApKHC1 induced increase in EPSPs at 12 hr. Piccolo S oligo was used as a control for microinjection and electrophysiology. (D) ApKHC1 is required for the 5HT-induced increase in EPSPs at 12 hr. Sensory neurons were injected with ApKHC1 AS or S oligos (as control) and were exposed to 5HT. Data was analyzed by ANOVA followed by Tukey-Kramer multiple comparison test. ∗p < 0.05. Percent changes in mean EPSP amplitudes (mean ± SEM) are shown in the bar graphs. Cell 2008 135, 960-973DOI: (10.1016/j.cell.2008.11.003) Copyright © 2008 Elsevier Inc. Terms and Conditions

Figure 7 A Model for Programmed Learning Response in Aplysia Results discussed in this manuscript and those previously published suggest an upregulation of transcription, local protein synthesis and molecular transport induced by five pulses of 5HT are critical for long-term synaptic plasticity. We define the process that couples all these critical components in concert to produce long-lasting changes in synaptic plasticity as programmed learning response. We suggest that kinesin is a key coordinator of this process in presynaptic as well as in postsynaptic neurons. Synaptic stimulation of sensory neurons by repeated applications of 5HT activates transcription in the nucleus (1), local protein synthesis at the synaptic sites (2), and upregulate kinesin heavy and light chain, and some of its cargos (3). Protein cargos transported by kinesin are used for the initiation of LTF, while RNA cargos may be used for the maintenance of LTF. This upregulated kinesin mediated transport in pre- and postsynaptic neurons induced by repeated applications of 5HT (3) represents a new component of long-term memory storage. Cell 2008 135, 960-973DOI: (10.1016/j.cell.2008.11.003) Copyright © 2008 Elsevier Inc. Terms and Conditions