Dendritic, delayed, stochastic CaMKII activation in behavioural time scale plasticity

Animals

All experimental procedures were approved and carried out in accordance with the regulations of the Max Planck Florida Institute for Neuroscience Animal Care and Use Committee as per the guidelines by the US National Institutes of Health. C57/B6 mice were used. The mice were kept in 12-h light–dark cycle at 18–21 °C with 40–50% humidity. We also used Camk2a^T286A mice to test the requirement of CaMKII in BTSP experiments³⁰.

Plasmid constructs

We fused two monomeric dimVenus (Venus(A206K,Y145W)) and mouse eGFP (eGFP(A206K)) to rat CaMKIIα subunit (2dV-Camuiα) (Addgene, 220366)^8,31. T286A (Addgene, 220367) and T305D/T306D (Addgene, 220368) 2dV-Camuiα mutants were constructed by restriction digestion and ligation. To do the ER calcium imaging experiments, we used the ER-GCaMP6-210 plasmid (Addgene, 86919).

Organotypic hippocampal slice cultures and transfection

Organotypic hippocampal slices were prepared from wild-type or transgenic P4–P8 C57/B6 mouse pups of both sexes as previously described³². In brief, the animal was anaesthetized with isoflurane, after which it was quickly decapitated and the brain removed. The hippocampi were dissected and cut into 350-µm thick coronal hippocampal slices using a McIlwain tissue chopper (Ted Pella) and plated on hydrophilic PTFE membranes (Millicell, Millipore) fed by culture medium containing MEM medium (Life Technologies), 20% horse serum, 1 mM l-glutamine, 1 mM CaCl₂, 2 mM MgSO₄, 12.9 mM d-glucose, 5.2 mM NaHCO₃, 30 mM HEPES, 0.075% ascorbic acid and 1 µg ml^–1 insulin. The slices were incubated at 37 °C in 5% CO₂. After 7–12 days in culture, CA1 pyramidal neurons were transfected using biolistic gene transfer with 1.0 µm gold beads (8–12 mg) coated with 2dV-Camuiα (50 μg)³³. For CaMKII experiments, owing to the size of the CaMKII sensor, plasmid transfection using biolistic gene gun was the most effective. Transfection was done days in vitro day 7–10, and experiments were performed 2–7 days after transfection. The age of neurons would correspond to acute slices from juvenile animals³⁴.

Acute slice preparation

Male C57/B6 mice (P25–P35 or P45–P60) were sedated by isoflurane inhalation and perfused intracardially with a chilled choline chloride solution. The brain was removed and placed in the same choline chloride solution composed of 124 mM choline chloride, 2.5 mM KCl, 26 mM NaHCO₃, 4 mM MgCl₂, 1.2 mM NaH2PO₄, 10 mM glucose and 0.5 mM CaCl₂, pH 7.4 equilibrated with 95% O₂ and 5% CO₂. Coronal hippocampal slices (300 μm) from both hemispheres were cut using a vibratome (V1200, Leica) and maintained in a submerged chamber in artificial cerebrospinal fluid (ACSF; 127 mM NaCl, 2.5 mM KCl, 4 mM CaCl₂, 25 mM NaHCO₃, 1.25 mM NaH₂PO₄ and 25 mM glucose) at 32 °C for 1 h and then at room temperature in oxygenated ACSF.

Two-photon glutamate uncaging

Two-photon glutamate uncaging was performed during BTSP and structural LTP experiments in organotypic hippocampal cultures and in acute hippocampal slices as previously described^35,36. Experiments were performed in a small recirculating volume (about 8 ml) of continuously oxygenated ACSF containing 4 mM 4-methoxy-7-nitroindolinyl-caged-l-glutamate (MNI-caged glutamate). A Ti:Sapphire laser was tuned at a wavelength of 720 nm to uncage MNI-caged glutamate in a small region about 0.5 μm from the spine. For structural plasticity experiments, 30 uncaging pulses of 0.5 Hz train were given. The power of the laser was set to 2.7 mW measured at the objective. These structural plasticity experiments were performed in Mg²⁺-free ACSF (127 mM NaCl, 2.5 mM KCl, 4 mM CaCl₂, 25 mM NaHCO₃, 1.25 mM NaH₂PO₄ and 25 mM glucose) containing 1 μM TTX and 4 mM MNI-caged l-glutamate aerated with 95% O₂ and 5% CO₂. The BTSP experiments were performed in 2 mM Ca²⁺ and 1 mM Mg²⁺. Experiments were performed at room temperature (24–26 °C).

Electrophysiology

Whole-cell patch-clamp electrophysiology experiments were combined with glutamate uncaging to induce BTSP at individual dendritic spines³⁶. The cells were first visualized in a bright field, or for the labelled cells, epifluorescence microscopy. The patch pipette (with a tip resistance of 2–5 MΩ) included the K⁺-based internal solution containing 145 mM K-gluconate, 14 mM phosphocreatine, 4 mM NaCl, 0.3 mM NaGTP, 4 mM MgATP, 3 mM l-ascorbic acid, 50–100 µM Alexa-594 and 10 mM HEPES (pH 7.4, 294 mOsm). In BTSP experiments, the EPSPs were measured under the current-clamp mode by a patch-clamp amplifier (MC-700B, Molecular Devices) and digitizer (National Instruments). After 2–5 min of dye loading, fluorescence from Alexa-594 was used to find dendritic spines in 2pFLIM. Uncaging-evoked EPSPs were induced on 1–2 spines on a dendrite by MNI-glutamate uncaging, about 0.5 µm away from the tip of the spine. The uncaging-evoked EPSP amplitude was 0.4–2 mV. Some BTSP experiments were performed in voltage-clamp configuration, for which the cells were held at −70 mV. The baseline glutamate uncaging-evoked EPSC amplitude was between 5 and 20 pA. These voltage-clamp experiments were performed with Cs⁺-based internal solution containing 130 mM Cs-methanosulfonate, 6 mM KCl, 10 mM HEPES, 4 mM NaCl, 0.3 mM NaGTP, 4 mM MgATP and 14 mM Tris-phosphocreatine (BTSP voltage-clamp protocol). Experiments were performed at room temperature (24–26 °C). In the CaMKII imaging experiments, similar to the above experiments, Alexa-594 dye (Thermo Scientific, 100 µM) was loaded as a structural marker. For experiments using APV, TTX, nifedipine, xestospongin C and MK-801, slices were incubated with the drugs for more than 30 min before BTSP experiments and were applied throughout the recording. Control experiments were done each day before the addition of the drug. For DMSO control experiments, a different ACSF solution was prepared each day. In experiments with thapsigargin and U73122, slices were incubated with the drugs for at least 60 min before the experiments. For QX314 experiments, we started the recordings 3–5 min after establishing the whole-cell patch clamp, and only the cells in which the BTSP protocol did not elicit any spiking during the current injection were considered for further analyses. All drugs were purchased from Tocris Biosciences unless specified otherwise. EPSPs were measured before and after the induction of BTSP. In all whole-cell recordings, the series resistance was monitored to be between 10 and 40 MΩ throughout the recording.

HeLa and HEK293FT cell maintenance, transfection and imaging

HeLa cells (American Type Culture Collection, CCL-2) and HEK293FT cells (Thermo Fisher) were grown in Dulbecco’s modified Eagle medium supplemented with 10% FBS at 37 °C in 5% CO₂. Plasmids were transfected into HeLa cells using Lipofectamine 3000 (Invitrogen). Imaging was performed 24–48 h following transfection in a HEPES-buffered ACSF solution (20 mM HEPES pH 7.3, 130 mM NaCl, 2 mM NaHCO₃, 25 mM d-glucose, 2.5 mM KCl and 1.25 mM NaH₂PO₄) with 2 mM CaCl₂ and 2 mM MgCl₂ by 2pFLIM as described below. When indicated, cells were stimulated with bath application of ionomycin (Tocris Biosciences) and then EGTA.

Fluorescence-coupled size-exclusion chromatography

Expression vector DNA (2 µg) including Camuiα were transfected into HEK293S GnTI- cells (2 × 10⁶ cells per well in 6-well plates) cultured in FreeStyle 293 (Thermo Fisher) using TransIT2020 transfection reagent (Mirus Bio). Cells were collected 48 h after transfection, washed with ice-cold PBS and sonicated in 250 µl TBS (20 mM Tris-HCl (pH 8.0) and 200 mM NaCl) using a Misonix Sonicator 3000 (3 times, 30 s, power level of 9.0). The lysate was ultracentrifuged at 70,000 r.p.m. for 10 min (TLA110 rotor). The supernatant (20 µl) was loaded onto a Superose-6 size-exclusion chromatography column (10/300 GL; GE Healthcare), pre-equilibrated with TBS, and run at a flow rate of 0.4 ml min^–1. The eluent from the Superose-6 column was detected using a fluorometer (RF-10AXL, Shimadzu) with the following settings: excitation, 475 nm; emission, 507 nm; time increment, 0.5 s; integration time, 1 s; and recording time, 75 min. The fluorescence-coupled size-exclusion chromatography data points were plotted using OriginPro graphic software (OriginLab v.9.5).

Fluorescence correlation spectroscopy

HEK293FT cells (Thermo Fisher) were transfected with the plasmids using Lipofectamine 3000 (Thermo Fisher) and cultured for 2 days at 37 °C and 5% CO₂. After washing the plate wells once in PBS buffer, the cells were lysed for 5 min with M-PER mammalian protein extraction reagent (Thermo Scientific), including Halt protease inhibitor (Thermo Scientific) and 5 mM EDTA. The lysates were centrifuged at 20,000g for 10 min and the supernatants were used for fluorescence correlation spectroscopy (FCS) measurement by diluting 2–15-fold in PBS buffer including the protease inhibitor. The FCS measurements were performed at 23 °C under a two-photon microscope without laser scanning, equipped with a Ti:Sapphire laser (Chameleon Ultra II, Coherent) tuned to a wavelength of 920 nm. The time-correlated single-photon counting data were collected for 60–120 s using a water-immersion objective (LUMPlanFL N ×60 NA 1.0 W, Olympus) directly immersed in 300 µl of the lysate solution, a single-photon counting board (Time Harp 260, PicoQuant) and a software of TTTR mode real-time correlator in TimeHarp 260 (v.3.0). Data analysis was performed using FoCuS-point software³⁷.

Optical CaMKII inhibition experiments

The CaMKII inhibition experiments were performed in organotypic hippocampal slices using previously described paAIP2 (ref. ¹²). In these experiments, slices were virally infected with 0.5–1 µl AAV mixture per slice (containing AAV9-Camk2a-Cre at 2 × 10¹² vg per ml (1:1,000 dilution, Addgene (105558-AAV9) and rAAV8-DIO-CBA-paAIP2-mEGFP at 4.2 × 10¹² vg per ml, UNC GTC Vector) at days in vitro 4–6 and imaged or patched at days in vitro 10–13. Cells with strong eGFP expression were used for experiments. Labelled cells were patched with the K⁺-based internal solution (see above) plus Alexa-594 dye in the patch pipette as described above. LED light stimulation (470 nm, M470L5, Thorlabs) was used to activate paAIP2.

Two-photon microscopy and 2pFLIM

Custom-built two-photon fluorescence lifetime imaging microscopy was used to perform 2pFLIM as previously described³⁸. 2pFLIM imaging was performed using a Ti:Sapphire laser (Coherent, Chameleon or Spark Alcor 920 nm (Spark Lasers)) at a wavelength of 920 nm with a power of 1.0–1.4 mW. Fluorescence emission was collected using a water-immersion objective (×60, NA 0.9, Olympus), divided with a dichroic mirror (565 nm) and detected with two separated photoelectron multiplier tubes placed after the wavelength filters (Chroma, 510/70-2p for green and 620/90-2p for red). Both red and green fluorescence was detected with photoelectron multiplier tubes with a low transfer time spread (H7422P40; Hamamatsu). Photon counting for fluorescence lifetime imaging was performed using a time-correlated single-photon counting board (Time-harp 260, Pico-Quant) using custom software (https://github.com/ryoheiyasuda/FLIMage_public). 2pFLIM images were collected at 64 × 64 pixels at the frame rate of 7.8 Hz (128 ms per frame), and the time course was filtered with a moving average over 30 frames. A second Ti:Sapphire laser tuned at a wavelength of 720 nm was used to uncage MNI-caged glutamate.

Ca²⁺ imaging

Ca²⁺ imaging was performed by loading calcium dyes Cal-590 (50–100 μM, AAT Bioquest) together with a structural marker Alexa-488 (100 μM, Thermo Fisher Scientific). The Ca²⁺ sensor intensity measurements were collected at 64 × 64 pixels at the frame rate of 7.8 Hz with 2pFLIM (lifetime information was not used). The Ca²⁺ response was calculated by normalizing the intensity with the intensity of Alexa-488. The membrane voltage was also recorded during Ca²⁺ imaging under the current-clamp mode. In a subset of experiments, uncaging-evoked EPSPs were measured before and after BTSP induction. For simultaneous Ca²⁺ and CaMKII imaging experiments, Ca²⁺ was normalized to the average of the first 100 frames before the induction of BTSP. For Ca²⁺ imaging experiments in acute hippocampal slices, we performed a whole-cell patch clamp with electrodes (4–6 MΩ) loaded with Cs⁺-based internal solution (see above) plus Cal-590 (50 µM). We measured the baseline Ca²⁺ for 2–4 min and then applied a published BTSP protocol¹, whereby Schaffer collaterals were stimulated with bipolar electrodes 10 times at 20 Hz and paired with postsynaptic current injection (300 pA for 300 ms) with a delay of 750 ms for 5 times. Then, Ca²⁺ imaging was resumed for another 2–4 mins. The Ca²⁺ events were detected using a custom Python code, whereby 3 times the standard deviation of the baseline noise was used as a detection threshold after the subtraction of the basal trend line obtained by linear regression.

2pFLIM analysis

2pFLIM analysis was performed as previously described³⁹. To measure the fraction of the donor that was undergoing FRET with the acceptor (binding fraction), we fit a fluorescence lifetime curve summing all pixels over an entire image with a double exponential function convolved with the Gaussian pulse response function as follows:

where τ_AD is the fluorescence lifetime of the donor bound with the acceptor, P_D and P_AD are the fraction of free donor and donor undergoing FRET with the acceptor, respectively, and H(t) is a fluorescence lifetime curve with a single exponential function convolved with the Gaussian pulse response function:

in which τ_D is the fluorescence lifetime of the free donor, τ_G is the width of the Gaussian pulse response function, F₀ is the peak fluorescence before convolution and t₀ is the time offset, and erfc is the complementary error function.

To generate the fluorescence lifetime image, we calculated the mean photon arrival time, <t>, in each pixel as follows:

Then, the mean photon arrival time was related to the mean fluorescence lifetime, <τ>, by an offset arrival time, t₀, which was obtained by fitting the entire image as follows:

For analysing fluorescence lifetime in regions of interests (ROIs) (spines or dendrites), we calculated the fluorescence lifetime by fitting the decay curve with equation (1), assuming τ_D, τ_AD, τ_G and t₀ are constants within each image session. To measure the CaMKII time of occurrence and peak lifetime change in BTSP and control experiments, the raw traces were first normalized using the first 100 frames as baseline and then the normalized data were smoothened using a moving average of 60 data points. Following this processing, the time of CaMKII peak and amplitude was manually calculated on individual CaMKII traces.

Statistics and reproducibility

All values are presented as the mean ± s.e.m. unless otherwise noted. The number of independent measurements or cells (n) is indicated in figures or figure legends. For electrophysiology experiments, the recordings were performed on 9–26 neurons (1 neuron per slice) from at least 2 different litters. For imaging experiments, the experiments were independently performed on 12–82 dendrites from at least 5 neurons (1 neuron per slice) from at least 2 different litters. For pharmacology experiments, 1–2 control experiments were performed on the same slices before the specific drug was added to the ACSF. In experiments for which DMSO was used as a vehicle, we performed control experiments on different days but on the slices made from the same batch as used in the experiments. Unpaired two-tailed Student’s t-test was used to compare two independent samples. Paired two-tailed Student’s t-test was used to compare dependent variables (before–after, soma–dendrite). One-way ANOVA followed by Dunnett’s multiple comparison test was used to compare more than two independent samples. Two-way ANOVA followed by Dunnett’s or Tukey’s multiple comparison test was used to compare grouped datasets. Correlation analysis was done by computing Pearson correlation coefficients. Data were organized in Microsoft Excel (v.2016). Data smoothening, statistical tests and P values are noted in each figure legend and were computed using GraphPad Prism (v.7.03, 9.5). Schematics of Figs. 1a, 3a and 5a and Extended Data Fig. 13 were created using Microsoft PowerPoint (v.2016) and Adobe Illustrator (v.27.9.1).

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.