Jul 10, 2024

Public workspaceCryo-electron tomography of thinned synapses

  • Anna Siegert1,2,
  • Arsen Petrovic1,3,
  • Thanh Thao Do1,2,
  • Florelle Domart1,
  • Rubén Fernández-Busnadiego1,3,2,4,5
  • 1University Medical Center Göttingen, Institute for Neuropathology, Göttingen, 37077 Germany;
  • 2Collaborative Research Center 1286 "Quantitative Synaptology", University of Göttingen, Göttingen, Germany;
  • 3Aligning Science Across Parkinson’s (ASAP) Collaborative Research Network, Chevy Chase, MD, USA;
  • 4Cluster of Excellence “Multiscale Bioimaging: from Molecular Machines to Networks of Excitable Cells” (MBExC), University of Göttingen, Göttingen, 37077, Germany;
  • 5Faculty of Physics, University of Göttingen, Göttingen, 37077, Germany
Open access
Protocol CitationAnna Siegert, Arsen Petrovic, Thanh Thao Do, Florelle Domart, Rubén Fernández-Busnadiego 2024. Cryo-electron tomography of thinned synapses. protocols.io https://dx.doi.org/10.17504/protocols.io.n92ldm188l5b/v1
License: This is an open access protocol distributed under the terms of the Creative Commons Attribution License,  which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Protocol status: In development
We are still developing and optimizing this protocol
Created: December 27, 2023
Last Modified: July 10, 2024
Protocol Integer ID: 92911
Keywords: ASAPCRN
Funders Acknowledgement:
Deutsche Forschungsgemeinschaft
Grant ID: EXC 2067/1-390729940
Deutsche Forschungsgemeinschaft
Grant ID: 448415290
Aligning Science Across Parkinson’s (ASAP) initiative
Grant ID: ASAP-000282
Deutsche Forschungsgemeinschaft
Grant ID: SFB1286/A12
Deutsche Forschungsgemeinschaft/NeuroNex
Grant ID: FE 1940
Abstract
Here we provide a protocol for cryo-electron tomography (cryo-ET) of thinned synapses within intact rat primary neuron cultures. This workflow relies on cryo-focused-ion-beam (FIB) milling to enable unrestricted access to synapses within neuronal cultures and to achieve samples sufficiently thin (~150 nm) for high resolution cryo-ET imaging. This protocol allows targeting of synapses with and without cryo-fluorescence light microscopy (cryo-FLM) correlation for FIB milling and cryo-ET.
Attachments
Guidelines
Primary hippocampal neuron co-cultures with glia
Mixed primary hippocampal cultures are prepared according to the Goslin, Asmussen and Banker protocol (Goslin et al., 1998) and comprise both neuronal and glial cells (mainly astrocytes and microglia). The preparation of these cultures on EM grids is rather simple and fast since the protocol does not deviate much from the conventional preparation on coverslips. However, many glial cells may be present which can be a disadvantage for certain applications (i.e. when live staining for markers that are present in both neurons and glia). Therefore, we also provide a protocol for the preparation of neuronal cultures with almost no glial cells (see accompanying protocol by Domart et al.).


Live staining with Synaptotagmin1 antibodies
Live labelling of neuronal cultures with antibodies directed against the luminal domains of Synaptotagmin1 (α-Syt1-ATTO647N) stains actively recycling pre-synapses and facilitates targeting synapse-rich regions for FIB milling. α-Syt1-ATTO647N is taken up into the presynaptic terminal during synaptic transmission upon synaptic vesicle (SV) recycling (Fig. 1A). Once an SV fuses with the pre-synaptic membrane, the luminal domain of Synaptotagmin1 is exposed to the extracellular space until the SV is recycled (Rizzoli, 2014). During this time, antibodies can bind. Therefore, live Synaptotagmin1-staining can be used to label the pool of actively recycling synapses (Hua et al., 2011; Kraszewski et al., 1995; Truckenbrodt et al., 2018; Wilhelm et al., 2010).

FIB milling
For the general FIB milling procedure and image correlation using the MAPS software please refer to the following protocols (https://www.protocols.io/view/cryo-fib-milling-protocol-formammalian-cells-kqdg3xpyqg25/v1 and https://dx.doi.org/10.17504/protocols.io.btyinpue).


Tomogram acquisition

For the general tomogram acquisition procedure please refer to the following protocol (dx.doi.org/10.17504/protocols.io.6qpvr3442vmk/v1).
Following tomogram reconstruction using the IMOD software (https://bio3d.colorado.edu/imod/doc/tomoguide.html), the mature synapses should contain the following characteristic features (see Fig. 2D)


Materials

Culturing primary rat hippocampal neurons on EM grids: Preparation of the EM grids

  • EM grids (e.g. SiO2 film on Au mesh, R1/2 or R2/2, Quantifoil)
  • 35 mm glass bottom dishes (e.g. Thermo Fisher Scientific)
  • 35 mm dishes with 4 inner rings (e.g. Greiner Bio-One)
  • 1 mg/mL poly-L-lysine in borate buffer
  • Sterile water
  • Plasma cleaner (e.g. PDC-32G-2, Harrick Plasma Cleaner)
  • UV light

Primary hippocampal neuron co-cultures with glia

AB
DMEM supplemented with
FCS10 %
L-glutamine 2 mM
penicillin/streptavidin (DMEM10%FCS) 1 %

AB
Neurobasal plus medium supplemented with
B272 %
Glutamax1 %


Live staining with Synaptotagmin1 antibodies

  • ReagentSynaptotagmin1 antibody luminal domain Synaptic SystemsCatalog #105 311AT1 (c=1mg/ml)

AB
Neurobasal plus medium supplemented with
B272 %
Glutamax1 %


Plunge freezing

  • 5 % glycerol in Tyrode solution
  • Automated or manual plunger

Cryo-FLM for identification of synapse-rich regions
Microscope: cryo-light microscope (here: THUNDER Imager EM Cryo CLEM, Leica)

FIB milling Microscope: cryo-FIB-SEM microscope (here: Aquilos 2, ThermoFisher Scientific)

Tomogram acquisition
Microscope: cryo-transmission electron microscope (here: 300 kV Krios G4 Cryo TEM, ThermoFisher Scientific)


Culturing primary rat hippocampal neurons on EM grids - Preparation of the EM grids
Culturing primary rat hippocampal neurons on EM grids - Preparation of the EM grids
Glow discharge the EM grids (SiO2 film on Au mesh, R 1/2 or R 2/2, Quantifoil) with a plasma cleaner (Duration00:00:30 , medium voltage) followed by Duration00:30:00 under UV light.

30m 30s
Place 4 grids in a 35 mm dish. Glass bottom dishes or dishes with 4 inner rings can be used.
Coat the EM grids with Concentration1 mg/mL poly-L-lysine Duration02:00:00 at Temperature37 °C and 5 % CO2.

2h
Wash 3 times with sterile water and keep the grids in sterile water at Temperature4 °C before use.

Wash
Culturing primary rat hippocampal neurons on EM grids - Primary hippocampal neuron co-cultures with glia
Culturing primary rat hippocampal neurons on EM grids - Primary hippocampal neuron co-cultures with glia
Prepare a hippocampal cell suspension from E19 rat embryos.

Dilute the hippocampal cell suspension to a concentration of 200-300,000 cells per mL in DMEM10%FCS.
For the glass bottom dish, replace the water from the dish with Amount500 µL of pre-warmed DMEM10%FCS. For the dishes with the 4 inner rings immediately plate the cells after removing the water.

Pipetting
Plate Amount100 µL of cell suspension dropwise on each EM grid. Incubate at Temperature37 °C and 5 % CO2.

Incubation
Pipetting
After Duration02:00:00 , add Amount500 µL of pre-warmed DMEM10%FCS per dish.

2h
Pipetting
Incubate DurationOvernight at Temperature37 °C and 5 % CO2.

2h
Incubation
Overnight
The following day, replace the medium with Amount2 mL of pre-warmed Neurobasal plus medium supplemented with 2 % B27 plus and 1 % Glutamax.

Pipetting
Live staining with Synaptotagmin1 antibodies
Live staining with Synaptotagmin1 antibodies

Note
Live fluorescent labelling is performed immediately prior to vitrification by plunge freezing. When adding or removing the medium from culture dishes containing EM grids with primary hippocampal cultures, be careful not to move or flip the grids and pipet slowly and gently.

Dilute α-Syt1-ATTO647N at 1:500 in medium from the culture dish containing EM grids.
Pipetting
Remove all the remaining medium from the culture dish and replace by the diluted antibody in culture medium.
Incubate for 30-45 min at Temperature37 °C and 5 % CO2 to allow the uptake of α-Syt1-ATTO647N into pre-synapses.

Incubation
Wash twice with Neurobasal plus medium supplemented with 2 % (v/v) B27 plus and 1 % (v/v) Glutamax, pre-incubated at Temperature37 °C and 5 % CO2.

Wash
Plunge freezing
Plunge freezing
Remove all the medium and replace it by 5 % (v/v) glycerol in Tyrode solution prewarmed at Temperature37 °C and incubate for 2-5 min.

Incubation
Immediately after glycerol incubation, plunge freeze grids in a mixture of liquid ethanepropane (37 % ethane, 63 % propane) cooled at Temperature-195 °C .

Cryo-FLM for identification of synapse rich regions
Cryo-FLM for identification of synapse rich regions
Acquire an overview of the grid in both bright field and fluorescence channels.
Imaging
Identify squares with cell bodies using bright field and fluorescence data (Fig. 1B).

Note
α-Syt1-ATTO647N-positive puncta localize around the cell body, as well as along the neuronal processes, and together with the outline of the cell body in the bright field channel, serving as a guidance for the identification of neuronal cell bodies.

Imaging
Acquire z-stacks of the squares containing cell bodies deemed suitable for FIB milling (cell bodies should not be localized on the grid bar).
Overlay z-stacks (or maximum intensity projections (MIPs) of the z-stacks) with the scanning electron microscopy (SEM) images prior to FIB milling and position the lamella milling pattern according to α-Syt1-ATTO647N positive puncta around the cell body (Fig. 1C, D).

Note
The cryo-CLEM step prior to FIB milling may be omitted by experienced users, and lamellae milling patterns may be placed directly at regions near the cell body with a high synaptic density (neuropil, Fig. 2).


Fig. 1: (A) Schematic representation of α-Syt1-ATTO647N antibody uptake into the pre-synapse during live staining. (B) Representative cryo-FLM images of primary hippocampal cultures live stained with αSyt1-ATTO647N, acquired with a Leica THUNDER Imager EM Cryo CLEM with a 50x objective (0.9 N.A.) at -195 °C. (C) Scanning electron microscopy (SEM) image of the same neuronal cell body as seen in (B) without (left) and with overlayed α-Syt1-ATTO647N-fluorescence (right). (D) Ion beam (IB) image showing the lamella positioning (white rectangle) following α-Syt1-ATTO647N fluorescence at the junction between neuronal cell body and grid surface (left) and SEM image of a fine milled lamella (right).
Fig. 1: (A) Schematic representation of α-Syt1-ATTO647N antibody uptake into the pre-synapse during live staining. (B) Representative cryo-FLM images of primary hippocampal cultures live stained with αSyt1-ATTO647N, acquired with a Leica THUNDER Imager EM Cryo CLEM with a 50x objective (0.9 N.A.) at -195 °C. (C) Scanning electron microscopy (SEM) image of the same neuronal cell body as seen in (B) without (left) and with overlayed α-Syt1-ATTO647N-fluorescence (right). (D) Ion beam (IB) image showing the lamella positioning (white rectangle) following α-Syt1-ATTO647N fluorescence at the junction between neuronal cell body and grid surface (left) and SEM image of a fine milled lamella (right).

Imaging
FIB milling
FIB milling
Identify neuronal cell bodies in SEM and ion-beam (IB) induced images. Neurons at DIV15 appear as bulges of 15-30 µm diameter with processes surrounding the soma (Fig. 2A).

Fig. 2: (A) Lamella positioning (white rectangle) without fluorescence correlation. (B) Fine milled lamella corresponding to the position chosen in (A), approximately 150 nm thin. (C) Low magnification cryotransmission electron microscopy (TEM) overview of the same lamella as seen in (B) with a zoomed inset indicating the position chosen for tomogram acquisition based on synapse characteristics (as described in 6. Tomogram acquisition) that can be identified on low magnification TEM overviews. (D) Tomographic slice corresponding to the reconstructed tilt series (tomogram) of the position indicated in (C) acquired at a pixel size of 2.94 Å/px with a 300 kV Krios G4 Cryo TEM from ThermoFisher Scientific.
Fig. 2: (A) Lamella positioning (white rectangle) without fluorescence correlation. (B) Fine milled lamella corresponding to the position chosen in (A), approximately 150 nm thin. (C) Low magnification cryotransmission electron microscopy (TEM) overview of the same lamella as seen in (B) with a zoomed inset indicating the position chosen for tomogram acquisition based on synapse characteristics (as described in 6. Tomogram acquisition) that can be identified on low magnification TEM overviews. (D) Tomographic slice corresponding to the reconstructed tilt series (tomogram) of the position indicated in (C) acquired at a pixel size of 2.94 Å/px with a 300 kV Krios G4 Cryo TEM from ThermoFisher Scientific.

Imaging
Place the lamella milling patterns in front or to the side of the cell body (Fig. 2A), in the vicinity of the synaptic rich region. This avoids having the lamella going directly through the cell body, thus increasing the chance of capturing synapses (Fig. 2B, C, D).
Imaging
Mill at an angle of 9-12 ° and a lamella width of 12-20 µm.
Tomogram acquisition
Tomogram acquisition

Note
In general, the cumulative electron dose is kept around 120 e-/Å2. We acquire tomograms with a magnification of 2.94 or 1.89 Å/px in most cases.

The presynaptic terminus should contain densely packed synaptic vesicles with an average radius between 30-40 nm. Often, one should be able to observe the "active zone", the region containing tethered synaptic vesicles in close proximity to the plasma membrane. Rarely, membrane fusion events can also be observed, whereby the synaptic vesicle membrane is connected to the plasma membrane. Very often, a synaptic vesicle cluster is associated with microtubules. Additionally, a clearly distinguishable mitochondrion is often seen. Depending on the lamella thickness, actin filaments are also evident.

Imaging
A synaptic cleft of approximately 20-30 nm containing dense material should also be apparent.
Contrary to the presynaptic terminus, the postsynaptic region is relatively featureless. The postsynaptic density (PSD), often a characteristic feature in conventional EM studies, is less apparent. The postsynaptic region often appears dense (the grey scale value is darker). Large macromolecular complexes like ribosomes or actin filaments are often present.
Protocol references
Goslin, K., Asmussen, H., & Banker, G. (1998). Rat Hippocampal Neurons in Low-DensityCulture. In G. Banker & K. Goslin (Eds.), Culturing nerve cells (pp. 339–370). MIT Press. https://doi.org/10.7551/mitpress/4913.003.0020
Hua, Y., Sinha, R., Thiel, C. S., Schmidt, R., Hüve, J., Martens, H., Hell, S. W., Egner, A., & Klingauf, J. (2011). A readily retrievable pool of synaptic vesicles. Nature Neuroscience, 14(7), 833–839. https://doi.org/10.1038/nn.2838
Kraszewski, K., Mundigl, O., Daniell, L., Verderio, C., Matteoli, M., & Camilli, P. de (1995). Synaptic vesicle dynamics in living cultured hippocampal neurons visualized with CY3-conjugated antibodies directed against the lumenal domain of synaptotagmin. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 15(6), 4328–4342. https://doi.org/10.1523/JNEUROSCI.15-06-04328.1995
Rizzoli, S. O. (2014). Synaptic vesicle recycling: Steps and principles. The EMBO Journal, 33(8), 788–822. https://doi.org/10.1002/embj.201386357
Truckenbrodt, S., Viplav, A., Jähne, S., Vogts, A., Denker, A., Wildhagen, H., Fornasiero, E. F., & Rizzoli, S. O. (2018). Newly produced synaptic vesicle proteins are preferentially used in synaptic transmission. The EMBO Journal, 37(15). https://doi.org/10.15252/embj.201798044
Wilhelm, B. G., Groemer, T. W., & Rizzoli, S. O. (2010). The same synaptic vesicles drive active and spontaneous release. Nature Neuroscience, 13(12), 1454–1456. https://doi.org/10.1038/nn.2690