protocol
Single-cell axotomy of cultured hippocampal neurons integrated in neuronal circuits Susana Gomis-Rüth1,5, Michael Stiess1,2,4,5, Corette J Wierenga3, Liane Meyn1,2 & Frank Bradke1,2 1Axonal Growth and Regeneration, Max Planck Institute of
Neurobiology, Martinsried, Germany. 2Axon Growth and Regeneration, Deutsches Zentrum für Neurodegenerative Erkrankungen, Bonn, Germany. 3Division of Cell Biology, Faculty of Science, Utrecht University, Utrecht, The Netherlands. 4Present address: Biozentrum, University of Basel, Basel, Switzerland. 5These authors contributed equally to this work. Correspondence should be addressed to F.B. ([email protected]).
© 2014 Nature America, Inc. All rights reserved.
Published online 4 April 2014; doi:10.1038/nprot.2014.069
An understanding of the molecular mechanisms of axon regeneration after injury is key for the development of potential therapies. Single-cell axotomy of dissociated neurons enables the study of the intrinsic regenerative capacities of injured axons. This protocol describes how to perform single-cell axotomy on dissociated hippocampal neurons containing synapses. Furthermore, to axotomize hippocampal neurons integrated in neuronal circuits, we describe how to set up coculture with a few fluorescently labeled neurons. This approach allows axotomy of single cells in a complex neuronal network and the observation of morphological and molecular changes during axon regeneration. Thus, single-cell axotomy of mature neurons is a valuable tool for gaining insights into cell intrinsic axon regeneration and the plasticity of neuronal polarity of mature neurons. Dissociation of the hippocampus and plating of hippocampal neurons takes ~2 h. Neurons are then left to grow for 2 weeks, during which time they integrate into neuronal circuits. Subsequent axotomy takes 10 min per neuron and further imaging takes 10 min per neuron.
INTRODUCTION Lesioning of axons upon central nervous system (CNS) injury can lead to the loss of function through the interruption of axonal connectivity. Subsequent to the injury, axotomized axons retract and degenerate. As CNS axons lack the ability to regenerate, therapy of such injuries and induction of functional recovery have proven to be difficult1,2. A key obstacle for developing effective therapeutic approaches is that we still know relatively little about the axonal behavior after axotomy and the underlying molecular mechanisms preventing regeneration. Experimental approaches to study axonal behavior post axotomy To better understand these mechanisms, different experimental approaches have been undertaken. A relatively complex approach is to perform a spinal cord injury in situ in the living animal. In these in vivo models, axonal tracts are injured by varying techniques, ranging from sharp transections to blunt contusions3–5. The regenerative behavior of axons is then studied by using conventional tracing methods and postmortem analysis6,7. Postmortem analysis, however, limits the analysis to fixed tissue and gives no insight into the dynamic behavior of axons after injury. To overcome these limits, live imaging of injured axons was established8–10. Thereby, single GFP-labeled axons in the spinal cord of transgenic mice are cut and the dynamic behavior of the transected axons is followed in vivo. In this paradigm, the lesion of surrounding tissue is minimized, but secondary effects, including inflammation or scar formation, cannot be avoided8,11. Although the damage to the surrounding tissue can be reduced by the use of two-photon laser axotomy12, a more reduced and defined environment is needed to study intrinsic regrowth properties of axons and the underlying cell biology. By injuring many axons at once, either by cutting or stretching them13–15, it is possible to study the response of injured axons. However, to assess how a single neuron responds to an injury, one needs to reveal the morphology of the whole injured neuron. Here we describe a protocol to study the intrinsic regenerative capacities of axons by performing single-cell axotomy in cultures of dissociated neurons16–24. This protocol was used 1028 | VOL.9 NO.5 | 2014 | nature protocols
in refs. 16,25. The most prominent advantage of single-cell axotomy is the ability to examine the effect of extrinsic and intrinsic factors on the growth capacities of injured axons in a controlled, reproducible and growth-permissive environment. As the neurons can be repeatedly observed in the cell culture dish, this paradigm allows analyzing the dynamic regenerative behavior of single cells. Furthermore, the approach permits the study of the effects of single extrinsic factors on the intrinsic regrowth capacities through specific manipulation of the neuronal environment by introducing different growth substrates, or by adding other specific cell types, including microglia or astrocytes26. Drugs or growth factors can be easily added, and their direct effects can be analyzed without being masked by the secondary effects of the injured tissue. The role of specific proteins in the regeneration process can be elucidated by manipulating their expression levels either by overexpression or RNAi-based downregulation. Therefore, single-cell axotomy enables deciphering the role of specific molecules or signaling pathways in the intrinsic regenerative capacities of axons27,28. Thus, it allows to dissect the molecular mechanisms of axon regeneration. Besides axon regeneration, axotomy serves as a valuable tool for studying the plasticity of neuronal polarization. By removing the axon, neuronal polarity is disrupted and repolarization can be analyzed in detail16–18,21. This includes changes in the neuronal cytoskeleton, axonal transport and the axonal initial segment29. Experimental design In this protocol, we focus on the use of rodent hippocampal neurons, the neuronal culture system commonly used to perform axotomy. These neurons represent a relatively homo geneous neuronal culture, they resemble the in situ morphology very well and they form extensive, synaptically coupled networks30. However, this approach can easily be adjusted to use other types of neurons. We first describe how a coculture of wild-type (WT) hippocampal neurons with sparsely fluorescently labeled neurons is set up
protocol Box 1 | Assessing maturity in regenerated axons Synaptically active, mature neurons that are integrated in neuronal networks can regenerate their axons or form new axons after axotomy16. To examine the success of axon regeneration after axotomy, it is key to identify and analyze the molecular composition of the regenerated or newly formed axons. In our laboratory, we perform several distinct experiments to assess the degree of maturation of the regenerated axons 1 and 5 d after the axotomy. Immunohistochemistry after 24 h. The presence of the axonal marker Tau-1 in regenerating axons after axotomy is assessed. In addition, the absence of the dendritic marker MAP-2 in dendrites transformed to axons after proximal axotomies is analyzed. Immunohistochemistry after 5 d. The axonal markers Tau-1 or SMI-312 and the dendritic marker MAP-2 can be used to specifically label axons and dendrites, respectively. To confirm the presence of synaptic molecules along the regenerating, GFP-positive axon, the distribution of presynaptic (e.g., synapsin-1 and synaptophysin) and postsynaptic markers (e.g., PSD-95) is observed (Fig. 6e).
© 2014 Nature America, Inc. All rights reserved.
Synaptic exocytosis and endocytosis after 5 d. To confirm that the synaptic machinery is active, the regenerated, GFP-labeled axons are loaded with the red styryl dye FM 4-64. Synaptic loading of the dye is stimulated with high-potassium buffers, and the sites with functional synaptic machinery are detected by using fluorescence microscopy (Fig. 2d). Electron microscopy after 5 d. To visualize synaptic structures at the ultrastructural level in regenerated axons, we perform GRAB photoconversion (GFP recognition after bleaching) by inducing strong precipitation of 3,3′-diaminobenzidine (DAB) in GFP-expressing cells, adapting the protocol from previously published methods35–37. This electron-dense marker is used to relocalize the neurons of interest in the electron micrographs, and it allows the direct observation of synaptic vesicles in labeled regenerated or newly grown axons.
(Steps 1–12). Second, we explain in detail how single-cell axotomy on these labeled neurons is performed (Steps 13–26). Details of how to assess the physiological function of the regenerated axon are described in Box 1. In addition, different approaches to the molecular analysis of axon regeneration are described in Box 2. Other potential applications of the hippocampal culture with single-labeled neurons can be found below. We do not provide full details on how to culture hippocampal neurons, as alternative protocols are available, for example, in ref. 30. A key step for successful axotomy is the identification of the axonal process. This is very easy with developmentally young neurons31, but it can cause problems with mature neurons fully integrated into neuronal networks. Thus, to perform single-cell axotomy of mature neurons and to follow the regeneration of newly growing axons, it is necessary to label single neurons to distinguish them from other neurons and their processes. Labeling with a genetically encoded fluorescent protein expressed in neurons has proven to be an elegant and noninvasive labeling strategy. We mix dissociated hippocampal neurons from WT mice with a very low proportion (1–5%) of GFP-expressing hippocampal
neurons from genetically labeled mice (Fig. 1)32. This enables the simple and faithful recognition of the processes belonging to individual neurons and allows fast identification of the morphological changes taking place after axotomy. Generally, the surrounding neurons serve as growth controls for the axotomy experiment. If axotomy experiments are performed together with drug treatments, we recommend performing axotomies with neurons from the same culture in a different dish without drug treatment. This allows the exclusion of cultureand environment-specific differences. Neuronal culture with sparsely labeled neurons is also a great tool for analyzing the effects of different pharmacological treatments of differentiated neurons integrated in neuronal networks on a single-cell level. Furthermore, it also enables the analysis of the developmental defects of hippocampal neurons from genetically modified mice. The coculturing of GFP-expressing WT neurons can be used as an internal control when cocultured with neurons from genetically modified mice. In our laboratory, when we studied cell division cycle 42 (Cdc42)- or actin depolymerizing factor (ADF)/cofilin-knockout (KO) neurons33,34,
Box 2 | Molecular analysis of axonal regeneration Axotomy is a powerful approach for investigating how molecules and organelles change their behavior or distribution during axon regeneration and for identifying molecules necessary for the regeneration process. Furthermore, axotomy allows to study the plasticity and maintenance of neuronal polarity involving the cytoskeleton, axonal transport or the axonal initial segment16,18,19,29,31. To visualize proteins or organelles, it might be necessary to introduce DNA plasmids encoding fluorescently labeled proteins into hippocampal neurons. Furthermore, the knockdown of proteins by transfection of siRNA or shRNA constructs can help elucidate their role in axon regeneration. In our laboratory, the best transfection efficiency was obtained with Amaxa nucleofection before plating. However, a comprehensive discussion about different transfection methods of hippocampal neurons can be found elsewhere30. For long-term live observation, neurons should be transferred from the axotomy buffer back to the original medium after axotomy. If you are imaging red fluorophores, N-2 medium based on 1× MEM without phenol red should be used to avoid interference from the phenol red. Small incubator chambers placed on the fluorescence microscope with controlled temperature, CO2 and humidity regulation allow imaging for at least 24 h. We have had good experiences by using the heating stage and incubation unit from Ibidi when observing neurons for several days. nature protocols | VOL.9 NO.5 | 2014 | 1029
protocol Figure 1 | Main steps of the coculture of dissociated WT and GFP-expressing neurons. WT and GFP mice should be mated synchronously and their embryos should be isolated separately, but at the same time. Once the hippocampi are dissociated and the number of isolated neurons is counted, GFP-expressing and WT neurons can be plated at the desired ratio in the same dish. The neurons are plated on the coverslip with the grid and the wax dots upward. After 4–24 h in MEM-HS, the coverslips are flipped into dishes containing the glia monolayer and N-2 medium. After flipping, the wax dots should be facing the bottom of the dish.
Pregnant E17 WT and GFP mice
Isolate embryos
© 2014 Nature America, Inc. All rights reserved.
Isolate hippocampi and dissociate cells
the GFP labeling allowed an immediate recognition of the genotype. Furthermore, the observed phenotypes can be related to a cell-autonomous KO effect and not to effects of an unhealthy culture, as the KO neurons grow within a healthy WT culture. In the case of embryos from heterozygous mice, hippocampi of each KO embryos have to be dissociated and plated independently (together with the GFP-positive cells). Therefore, the help of a second person is necessary to prepare the WT and KO culture in parallel to finish the culture within a desired time of 2 h. At the end of the culture, a biopsy of the embryos is genotyped to identify the homozygous KO animals.
MATERIALS REAGENTS • Embryonic day (E)16.5 to 17.5 pregnant mice from the WT and GFP lines (expressing GFP under the ubiquitous ‘CAG’ promoter, composed of the cytomegalovirus (CMV) enhancer, a fragment of the chicken-actin promoter and rabbit-globin exons)13,15 ! CAUTION All animal experiments must be performed according to relevant institutional and governmental ethics and animal handling regulations. • Antibodies: mouse anti Tau-1, 1:5,000 (Chemicon International); goat anti MAP-2, 1:200 (Santa Cruz); rabbit anti-synapsin 1, 1:500 (Chemicon International); mouse anti-PSD-95, 1:1,000 (clone PSD-956G61C9, ABR Affinity BioReagents) and the matching secondary antibodies Alexa Fluor 555, 568 and 350 (Molecular Probes) CRITICAL These antibodies are only required if performing the immunohistochemistry described in Box 1. • FM4-64 dye (Molecular Probes, cat. no. T13320) • Cytosine-d-arabinofuranoside (ARA-C; Sigma) • Poly-l-lysine hydrobromide (Sigma-Aldrich, cat. no. P-2636) • Boric acid (H3BO3) • Borax (sodium tetraborate, Na2B4O7.10 H2O) • HBSS (Gibco (Life Technologies), cat. no. 14025-050) • HEPES (BIOMOL, cat. no. 05288) • Trypsin-EDTA (Gibco (Life Technologies), cat. no. 25300-054) • MEM (Gibco (Life Technologies), cat. no. 21430-020) • NaHCO3 • Glucose • Glutamine (Gibco (Life Technologies), cat. no. 25030-024) • MEM essential amino acids (Gibco (Life Technologies), cat. no. 11130-036) • MEM non-essential amino acids (Gibco (Life Technologies), cat. no. 1140-035) • Horse serum (HS) • NaOH • Insulin (Sigma-Aldrich, cat. no. I-5500) • HCl • Progesterone (Sigma-Aldrich, cat. no. P-8783) • Ethanol • Putrescine (Sigma-Aldrich, cat. no. P-5780) • Selenium (Sigma-Aldrich, cat. no. 325473) • Human apo-transferrin (Sigma-Aldrich, cat. no. T-2252) • d(+)glucose anhydrate • Sodium pyruvate (Sigma-Aldrich, cat. no. P-2256) • Albumin (Sigma-Aldrich, cat. no. A 5503) 1030 | VOL.9 NO.5 | 2014 | nature protocols
Mix at desired ratio and plate
Flip into dishes with cultured glia
• NaCl • KCl • MgSO4 • CaCl2 • DNQX (Tocris) • APV (Tocris) EQUIPMENT • Glass coverslips with engraved or photoetched grid (Cellocate; Electron Microscopy Sciences, cat. no. 72264-23) • Porcelain staining racks (Thomas Scientific, cat. no. 8542-E40) • Laboratory oven • Laminar flow hood • Stemi SV 6 Zeiss binocular incorporated into the laminar flow hood • Instrument set 1: one ~15-cm scissors (Fine Science Tools (F.S.T.), cat. no. 14002-14) and two forceps of 15–17 cm (F.S.T., cat. no. 11002-14) • Instrument set 2: one ~9-cm scissors (F.S.T., cat. no. 14084-08) and two forceps Dumont no. 5 (F.S.T., cat. no. 11251-30) • Instrument set 3: one curved spring scissors (F.S.T., cat. no. 15006-09), one forceps Dumont no. 5 (F.S.T., cat. no. 11251-30) and one curved forceps Dumont no. 7 (F.S.T., cat. no. 11271-30) • Instrument set 4: one small-blade scissors (F.S.T., cat. no. 15001-08) and two forceps Dumont no. 5 (F.S.T., cat. no. 11251-30) • Portable fluorescent lamp with 500/550-nm light range (Roth, cat. no. H466.1) • Plastic tissue culture dishes, 3 cm, 6 cm and 10 cm (Nunc) • Tissue culture flasks (Nunc) • Wilovert binoculars • Hemocytometer • Water bath, 37 °C • Cell culture incubators, 36.5 °C and 5% CO2 • Bunsen burner • Sterilization filters, 0.22 µm (MILLEX GP, Millipore) • pH meter (inoLab, level 1) • Inverted fluorescence microscope, comparable, for example, to a Zeiss Axio Observer including a 5% transmission filter (Neutralfilter (ND) 1.3, 5% Chroma) and Plan-Neofluar 1.5× and 25× objectives • Glass-bottom dishes, 30 mm (MatTek Corporation) • Glass capillaries (Science Products, cat. no. GB100TF-10) • Capillary puller (Glass puller P-97, Sutter Instrument) • Needle holder with a manual or electric micromanipulator adaptable to the microscope being used
© 2014 Nature America, Inc. All rights reserved.
protocol REAGENT SETUP Glial feeder cells These comprise cortical astroglia obtained from E16.5–E17.5 mouse embryos, and they are prepared and cultured as described previously30. Of these glia cells, ~105 cells should be plated ~1 week in advance in 6-cm Petri dishes in MEM-HS medium so that they reach a confluency of ~40% on the day of dissection. One day before the hippocampal neurons are dissected, the MEM-HS medium should be replaced by N2 medium. This conditioning of the medium is important for optimal neuronal growth. Full details of how to set up glial feeder cell plates can be found in ref. 30. ! CAUTION All animal experiments must be performed according to relevant institutional and governmental ethics and animal handling regulations. Coverslip coating solution Dissolve 1 mg/ml poly-l-lysine hydrobromide in borate buffer (1.24 g of boric acid (H3BO3), 1.90 g of Borax (sodium tetraborate, Na2B4O7·10 H2O) and H2O at pH 8.5, final volume of 400 ml) and sterilize it by filtration. Prepare the solution freshly before use. Culture dissection and axotomy buffer (HBSS + 7 mM HEPES) Combine 500 ml of HBSS with 3.5 ml of 1 M sterile HEPES solution at pH 7.25. This buffer can be stored at 4 °C for several months. CRITICAL Osmolarity should be kept constant, within 5–10%, between solutions in which the neurons are kept. Neuronal dissociation solution (Trypsin + 7 mM HEPES) Add 700 µl of 1 M HEPES to 100 ml of 0.05% Trypsin-EDTA at pH 7.25. Divide the solution into aliquots and store them at −20 °C for several months. Neuronal serum-rich medium (MEM-HS) Combine 300 ml of H2O, 50 ml of 10× MEM, 20 ml of 5.5% (wt/vol) NaHCO3, 15 ml of 20% (wt/vol) glucose, 5 ml of glutamine (200 mM, 100×), 10 ml of 50× MEM essential amino acids, 10 ml of 100× MEM non-essential amino acids and 50 ml of HS (heat-inactivated for 30 min at 56 °C, several batches). Adjust the pH to 7.3 with 1 M NaOH. Fill up with water up to 500 ml and sterilize the medium by filtration. This medium can be stored at 4 °C for up to 4 weeks. Stock solution for serum-free medium (N-2 stock) To prepare 50-ml aliquots, mix 5 ml of insulin (50 mg of insulin in 10 ml of 0.01 N HCl), 5 ml of progesterone (63 mg of progesterone in 100 ml of absolute ethanol, and from that solution 1 ml in 99 ml of H2O for storing solution), 5 ml of putrescine (161 mg of putrescine in 10 ml of H2O) and 5 ml of selenium (33 mg of selenium in 100 ml of H2O, and from that solution 1 ml diluted in 99 ml of H2O) with 500 mg of human apo-transferrin and bring the volume to 500 ml with N-MEM medium. Sterilize the medium by filtration, divide it into aliquots and store them at −20 °C for up to 6 months.
Supplement solution for serum-free medium (N-MEM) Combine 300 ml of H2O, 50 ml of 10× MEM, 20 ml of 5.5% (wt/vol) NaHCO3 filtered, 15 ml of 20% (wt/vol) d(+)glucose anhydrate, 5 ml of glutamine (100× solution) and 5 ml of pyruvate solution (1.1g of sodium pyruvate in 100 ml of H2O). Bring the solution up to 500 ml with H2O. The solution is usable for ~2 weeks after preparation when kept at 4 °C. Neuronal serum-free medium (N-2 medium) Combine 50 ml of N-2 stock, 450 ml of N-MEM supplement and 100 mg of 1% (wt/vol) ovalbumin (1 g of albumin added to 100 ml of N-MEM). Filter the medium with a cell culture vacuum filter bottle at pH 7.3. This medium can be stored at 4 °C for up to 4 weeks. Mitosis inhibitor ARA-C Combine 12 mg of ARA-C dissolved in 50 ml of H2O. Filter the solution, divide it into aliquots in 500-µl tubes and store them at −20 °C for several months. FM loading solution FM loading solution is 119 mM NaCl, 2.5 mM KCl, 2 mM MgSO4, 2 mM CaCl2, 25 mM HEPES, 30 mM glucose (pH 7.35), 10 µM DNQX, 50 µM APV and 20 µM FM4-64. We always prepare the solution freshly before use. EQUIPMENT SETUP Coverslips It is important that coverslips on which neurons can be unambiguously relocated after axotomy be used. We use coverslips with a laser-engraved grid. Originally, we used Cellocates (Eppendorf, not produced anymore). Currently, we use coverslips that have a photoetched gridded pattern on the surface. If you use these coverslips, clean them, sterilize them and add wax dots to the engraved side, as described in ref. 30. Afterward, place a big drop of the coverslip-coating solution on each coverslip and incubate it overnight at room temperature (21 °C). Do not allow the solution to dry out, as the poly-l-lysine crystals are toxic for neurons. We find that crystallization starts to occur 20 h after the cover solution is added. Wash the coverslips four times with sterile water and fill the dishes with 5 ml of MEM-HS medium. CRITICAL The poly-l-lysine must be added after cleaning, sterilizing and adding the wax dots. Fire-polished Pasteur pipettes Polish a Pasteur pipette at the flame of a Bunsen burner until the diameter of the opening is about one-fourth of the original. If the diameter is too wide, cells will not be completely dissociated; if it is too narrow, neurons might be mechanically injured. Precoating of the polished Pasteur pipettes with sterile 5% (wt/vol) BSA in PBS is recommended because it strongly decreases the sticking of tissue and cells to the pipette during dissociation.
PROCEDURE Mixed WT- and GFP-dissociated neuronal culture setup ● TIMING
Single-cell axotomy of cultured hippocampal neurons integrated in neuronal circuits Susana Gomis-Rüth1,5, Michael Stiess1,2,4,5, Corette J Wierenga3, Liane Meyn1,2 & Frank Bradke1,2 1Axonal Growth and Regeneration, Max Planck Institute of
Neurobiology, Martinsried, Germany. 2Axon Growth and Regeneration, Deutsches Zentrum für Neurodegenerative Erkrankungen, Bonn, Germany. 3Division of Cell Biology, Faculty of Science, Utrecht University, Utrecht, The Netherlands. 4Present address: Biozentrum, University of Basel, Basel, Switzerland. 5These authors contributed equally to this work. Correspondence should be addressed to F.B. ([email protected]).
© 2014 Nature America, Inc. All rights reserved.
Published online 4 April 2014; doi:10.1038/nprot.2014.069
An understanding of the molecular mechanisms of axon regeneration after injury is key for the development of potential therapies. Single-cell axotomy of dissociated neurons enables the study of the intrinsic regenerative capacities of injured axons. This protocol describes how to perform single-cell axotomy on dissociated hippocampal neurons containing synapses. Furthermore, to axotomize hippocampal neurons integrated in neuronal circuits, we describe how to set up coculture with a few fluorescently labeled neurons. This approach allows axotomy of single cells in a complex neuronal network and the observation of morphological and molecular changes during axon regeneration. Thus, single-cell axotomy of mature neurons is a valuable tool for gaining insights into cell intrinsic axon regeneration and the plasticity of neuronal polarity of mature neurons. Dissociation of the hippocampus and plating of hippocampal neurons takes ~2 h. Neurons are then left to grow for 2 weeks, during which time they integrate into neuronal circuits. Subsequent axotomy takes 10 min per neuron and further imaging takes 10 min per neuron.
INTRODUCTION Lesioning of axons upon central nervous system (CNS) injury can lead to the loss of function through the interruption of axonal connectivity. Subsequent to the injury, axotomized axons retract and degenerate. As CNS axons lack the ability to regenerate, therapy of such injuries and induction of functional recovery have proven to be difficult1,2. A key obstacle for developing effective therapeutic approaches is that we still know relatively little about the axonal behavior after axotomy and the underlying molecular mechanisms preventing regeneration. Experimental approaches to study axonal behavior post axotomy To better understand these mechanisms, different experimental approaches have been undertaken. A relatively complex approach is to perform a spinal cord injury in situ in the living animal. In these in vivo models, axonal tracts are injured by varying techniques, ranging from sharp transections to blunt contusions3–5. The regenerative behavior of axons is then studied by using conventional tracing methods and postmortem analysis6,7. Postmortem analysis, however, limits the analysis to fixed tissue and gives no insight into the dynamic behavior of axons after injury. To overcome these limits, live imaging of injured axons was established8–10. Thereby, single GFP-labeled axons in the spinal cord of transgenic mice are cut and the dynamic behavior of the transected axons is followed in vivo. In this paradigm, the lesion of surrounding tissue is minimized, but secondary effects, including inflammation or scar formation, cannot be avoided8,11. Although the damage to the surrounding tissue can be reduced by the use of two-photon laser axotomy12, a more reduced and defined environment is needed to study intrinsic regrowth properties of axons and the underlying cell biology. By injuring many axons at once, either by cutting or stretching them13–15, it is possible to study the response of injured axons. However, to assess how a single neuron responds to an injury, one needs to reveal the morphology of the whole injured neuron. Here we describe a protocol to study the intrinsic regenerative capacities of axons by performing single-cell axotomy in cultures of dissociated neurons16–24. This protocol was used 1028 | VOL.9 NO.5 | 2014 | nature protocols
in refs. 16,25. The most prominent advantage of single-cell axotomy is the ability to examine the effect of extrinsic and intrinsic factors on the growth capacities of injured axons in a controlled, reproducible and growth-permissive environment. As the neurons can be repeatedly observed in the cell culture dish, this paradigm allows analyzing the dynamic regenerative behavior of single cells. Furthermore, the approach permits the study of the effects of single extrinsic factors on the intrinsic regrowth capacities through specific manipulation of the neuronal environment by introducing different growth substrates, or by adding other specific cell types, including microglia or astrocytes26. Drugs or growth factors can be easily added, and their direct effects can be analyzed without being masked by the secondary effects of the injured tissue. The role of specific proteins in the regeneration process can be elucidated by manipulating their expression levels either by overexpression or RNAi-based downregulation. Therefore, single-cell axotomy enables deciphering the role of specific molecules or signaling pathways in the intrinsic regenerative capacities of axons27,28. Thus, it allows to dissect the molecular mechanisms of axon regeneration. Besides axon regeneration, axotomy serves as a valuable tool for studying the plasticity of neuronal polarization. By removing the axon, neuronal polarity is disrupted and repolarization can be analyzed in detail16–18,21. This includes changes in the neuronal cytoskeleton, axonal transport and the axonal initial segment29. Experimental design In this protocol, we focus on the use of rodent hippocampal neurons, the neuronal culture system commonly used to perform axotomy. These neurons represent a relatively homo geneous neuronal culture, they resemble the in situ morphology very well and they form extensive, synaptically coupled networks30. However, this approach can easily be adjusted to use other types of neurons. We first describe how a coculture of wild-type (WT) hippocampal neurons with sparsely fluorescently labeled neurons is set up
protocol Box 1 | Assessing maturity in regenerated axons Synaptically active, mature neurons that are integrated in neuronal networks can regenerate their axons or form new axons after axotomy16. To examine the success of axon regeneration after axotomy, it is key to identify and analyze the molecular composition of the regenerated or newly formed axons. In our laboratory, we perform several distinct experiments to assess the degree of maturation of the regenerated axons 1 and 5 d after the axotomy. Immunohistochemistry after 24 h. The presence of the axonal marker Tau-1 in regenerating axons after axotomy is assessed. In addition, the absence of the dendritic marker MAP-2 in dendrites transformed to axons after proximal axotomies is analyzed. Immunohistochemistry after 5 d. The axonal markers Tau-1 or SMI-312 and the dendritic marker MAP-2 can be used to specifically label axons and dendrites, respectively. To confirm the presence of synaptic molecules along the regenerating, GFP-positive axon, the distribution of presynaptic (e.g., synapsin-1 and synaptophysin) and postsynaptic markers (e.g., PSD-95) is observed (Fig. 6e).
© 2014 Nature America, Inc. All rights reserved.
Synaptic exocytosis and endocytosis after 5 d. To confirm that the synaptic machinery is active, the regenerated, GFP-labeled axons are loaded with the red styryl dye FM 4-64. Synaptic loading of the dye is stimulated with high-potassium buffers, and the sites with functional synaptic machinery are detected by using fluorescence microscopy (Fig. 2d). Electron microscopy after 5 d. To visualize synaptic structures at the ultrastructural level in regenerated axons, we perform GRAB photoconversion (GFP recognition after bleaching) by inducing strong precipitation of 3,3′-diaminobenzidine (DAB) in GFP-expressing cells, adapting the protocol from previously published methods35–37. This electron-dense marker is used to relocalize the neurons of interest in the electron micrographs, and it allows the direct observation of synaptic vesicles in labeled regenerated or newly grown axons.
(Steps 1–12). Second, we explain in detail how single-cell axotomy on these labeled neurons is performed (Steps 13–26). Details of how to assess the physiological function of the regenerated axon are described in Box 1. In addition, different approaches to the molecular analysis of axon regeneration are described in Box 2. Other potential applications of the hippocampal culture with single-labeled neurons can be found below. We do not provide full details on how to culture hippocampal neurons, as alternative protocols are available, for example, in ref. 30. A key step for successful axotomy is the identification of the axonal process. This is very easy with developmentally young neurons31, but it can cause problems with mature neurons fully integrated into neuronal networks. Thus, to perform single-cell axotomy of mature neurons and to follow the regeneration of newly growing axons, it is necessary to label single neurons to distinguish them from other neurons and their processes. Labeling with a genetically encoded fluorescent protein expressed in neurons has proven to be an elegant and noninvasive labeling strategy. We mix dissociated hippocampal neurons from WT mice with a very low proportion (1–5%) of GFP-expressing hippocampal
neurons from genetically labeled mice (Fig. 1)32. This enables the simple and faithful recognition of the processes belonging to individual neurons and allows fast identification of the morphological changes taking place after axotomy. Generally, the surrounding neurons serve as growth controls for the axotomy experiment. If axotomy experiments are performed together with drug treatments, we recommend performing axotomies with neurons from the same culture in a different dish without drug treatment. This allows the exclusion of cultureand environment-specific differences. Neuronal culture with sparsely labeled neurons is also a great tool for analyzing the effects of different pharmacological treatments of differentiated neurons integrated in neuronal networks on a single-cell level. Furthermore, it also enables the analysis of the developmental defects of hippocampal neurons from genetically modified mice. The coculturing of GFP-expressing WT neurons can be used as an internal control when cocultured with neurons from genetically modified mice. In our laboratory, when we studied cell division cycle 42 (Cdc42)- or actin depolymerizing factor (ADF)/cofilin-knockout (KO) neurons33,34,
Box 2 | Molecular analysis of axonal regeneration Axotomy is a powerful approach for investigating how molecules and organelles change their behavior or distribution during axon regeneration and for identifying molecules necessary for the regeneration process. Furthermore, axotomy allows to study the plasticity and maintenance of neuronal polarity involving the cytoskeleton, axonal transport or the axonal initial segment16,18,19,29,31. To visualize proteins or organelles, it might be necessary to introduce DNA plasmids encoding fluorescently labeled proteins into hippocampal neurons. Furthermore, the knockdown of proteins by transfection of siRNA or shRNA constructs can help elucidate their role in axon regeneration. In our laboratory, the best transfection efficiency was obtained with Amaxa nucleofection before plating. However, a comprehensive discussion about different transfection methods of hippocampal neurons can be found elsewhere30. For long-term live observation, neurons should be transferred from the axotomy buffer back to the original medium after axotomy. If you are imaging red fluorophores, N-2 medium based on 1× MEM without phenol red should be used to avoid interference from the phenol red. Small incubator chambers placed on the fluorescence microscope with controlled temperature, CO2 and humidity regulation allow imaging for at least 24 h. We have had good experiences by using the heating stage and incubation unit from Ibidi when observing neurons for several days. nature protocols | VOL.9 NO.5 | 2014 | 1029
protocol Figure 1 | Main steps of the coculture of dissociated WT and GFP-expressing neurons. WT and GFP mice should be mated synchronously and their embryos should be isolated separately, but at the same time. Once the hippocampi are dissociated and the number of isolated neurons is counted, GFP-expressing and WT neurons can be plated at the desired ratio in the same dish. The neurons are plated on the coverslip with the grid and the wax dots upward. After 4–24 h in MEM-HS, the coverslips are flipped into dishes containing the glia monolayer and N-2 medium. After flipping, the wax dots should be facing the bottom of the dish.
Pregnant E17 WT and GFP mice
Isolate embryos
© 2014 Nature America, Inc. All rights reserved.
Isolate hippocampi and dissociate cells
the GFP labeling allowed an immediate recognition of the genotype. Furthermore, the observed phenotypes can be related to a cell-autonomous KO effect and not to effects of an unhealthy culture, as the KO neurons grow within a healthy WT culture. In the case of embryos from heterozygous mice, hippocampi of each KO embryos have to be dissociated and plated independently (together with the GFP-positive cells). Therefore, the help of a second person is necessary to prepare the WT and KO culture in parallel to finish the culture within a desired time of 2 h. At the end of the culture, a biopsy of the embryos is genotyped to identify the homozygous KO animals.
MATERIALS REAGENTS • Embryonic day (E)16.5 to 17.5 pregnant mice from the WT and GFP lines (expressing GFP under the ubiquitous ‘CAG’ promoter, composed of the cytomegalovirus (CMV) enhancer, a fragment of the chicken-actin promoter and rabbit-globin exons)13,15 ! CAUTION All animal experiments must be performed according to relevant institutional and governmental ethics and animal handling regulations. • Antibodies: mouse anti Tau-1, 1:5,000 (Chemicon International); goat anti MAP-2, 1:200 (Santa Cruz); rabbit anti-synapsin 1, 1:500 (Chemicon International); mouse anti-PSD-95, 1:1,000 (clone PSD-956G61C9, ABR Affinity BioReagents) and the matching secondary antibodies Alexa Fluor 555, 568 and 350 (Molecular Probes) CRITICAL These antibodies are only required if performing the immunohistochemistry described in Box 1. • FM4-64 dye (Molecular Probes, cat. no. T13320) • Cytosine-d-arabinofuranoside (ARA-C; Sigma) • Poly-l-lysine hydrobromide (Sigma-Aldrich, cat. no. P-2636) • Boric acid (H3BO3) • Borax (sodium tetraborate, Na2B4O7.10 H2O) • HBSS (Gibco (Life Technologies), cat. no. 14025-050) • HEPES (BIOMOL, cat. no. 05288) • Trypsin-EDTA (Gibco (Life Technologies), cat. no. 25300-054) • MEM (Gibco (Life Technologies), cat. no. 21430-020) • NaHCO3 • Glucose • Glutamine (Gibco (Life Technologies), cat. no. 25030-024) • MEM essential amino acids (Gibco (Life Technologies), cat. no. 11130-036) • MEM non-essential amino acids (Gibco (Life Technologies), cat. no. 1140-035) • Horse serum (HS) • NaOH • Insulin (Sigma-Aldrich, cat. no. I-5500) • HCl • Progesterone (Sigma-Aldrich, cat. no. P-8783) • Ethanol • Putrescine (Sigma-Aldrich, cat. no. P-5780) • Selenium (Sigma-Aldrich, cat. no. 325473) • Human apo-transferrin (Sigma-Aldrich, cat. no. T-2252) • d(+)glucose anhydrate • Sodium pyruvate (Sigma-Aldrich, cat. no. P-2256) • Albumin (Sigma-Aldrich, cat. no. A 5503) 1030 | VOL.9 NO.5 | 2014 | nature protocols
Mix at desired ratio and plate
Flip into dishes with cultured glia
• NaCl • KCl • MgSO4 • CaCl2 • DNQX (Tocris) • APV (Tocris) EQUIPMENT • Glass coverslips with engraved or photoetched grid (Cellocate; Electron Microscopy Sciences, cat. no. 72264-23) • Porcelain staining racks (Thomas Scientific, cat. no. 8542-E40) • Laboratory oven • Laminar flow hood • Stemi SV 6 Zeiss binocular incorporated into the laminar flow hood • Instrument set 1: one ~15-cm scissors (Fine Science Tools (F.S.T.), cat. no. 14002-14) and two forceps of 15–17 cm (F.S.T., cat. no. 11002-14) • Instrument set 2: one ~9-cm scissors (F.S.T., cat. no. 14084-08) and two forceps Dumont no. 5 (F.S.T., cat. no. 11251-30) • Instrument set 3: one curved spring scissors (F.S.T., cat. no. 15006-09), one forceps Dumont no. 5 (F.S.T., cat. no. 11251-30) and one curved forceps Dumont no. 7 (F.S.T., cat. no. 11271-30) • Instrument set 4: one small-blade scissors (F.S.T., cat. no. 15001-08) and two forceps Dumont no. 5 (F.S.T., cat. no. 11251-30) • Portable fluorescent lamp with 500/550-nm light range (Roth, cat. no. H466.1) • Plastic tissue culture dishes, 3 cm, 6 cm and 10 cm (Nunc) • Tissue culture flasks (Nunc) • Wilovert binoculars • Hemocytometer • Water bath, 37 °C • Cell culture incubators, 36.5 °C and 5% CO2 • Bunsen burner • Sterilization filters, 0.22 µm (MILLEX GP, Millipore) • pH meter (inoLab, level 1) • Inverted fluorescence microscope, comparable, for example, to a Zeiss Axio Observer including a 5% transmission filter (Neutralfilter (ND) 1.3, 5% Chroma) and Plan-Neofluar 1.5× and 25× objectives • Glass-bottom dishes, 30 mm (MatTek Corporation) • Glass capillaries (Science Products, cat. no. GB100TF-10) • Capillary puller (Glass puller P-97, Sutter Instrument) • Needle holder with a manual or electric micromanipulator adaptable to the microscope being used
© 2014 Nature America, Inc. All rights reserved.
protocol REAGENT SETUP Glial feeder cells These comprise cortical astroglia obtained from E16.5–E17.5 mouse embryos, and they are prepared and cultured as described previously30. Of these glia cells, ~105 cells should be plated ~1 week in advance in 6-cm Petri dishes in MEM-HS medium so that they reach a confluency of ~40% on the day of dissection. One day before the hippocampal neurons are dissected, the MEM-HS medium should be replaced by N2 medium. This conditioning of the medium is important for optimal neuronal growth. Full details of how to set up glial feeder cell plates can be found in ref. 30. ! CAUTION All animal experiments must be performed according to relevant institutional and governmental ethics and animal handling regulations. Coverslip coating solution Dissolve 1 mg/ml poly-l-lysine hydrobromide in borate buffer (1.24 g of boric acid (H3BO3), 1.90 g of Borax (sodium tetraborate, Na2B4O7·10 H2O) and H2O at pH 8.5, final volume of 400 ml) and sterilize it by filtration. Prepare the solution freshly before use. Culture dissection and axotomy buffer (HBSS + 7 mM HEPES) Combine 500 ml of HBSS with 3.5 ml of 1 M sterile HEPES solution at pH 7.25. This buffer can be stored at 4 °C for several months. CRITICAL Osmolarity should be kept constant, within 5–10%, between solutions in which the neurons are kept. Neuronal dissociation solution (Trypsin + 7 mM HEPES) Add 700 µl of 1 M HEPES to 100 ml of 0.05% Trypsin-EDTA at pH 7.25. Divide the solution into aliquots and store them at −20 °C for several months. Neuronal serum-rich medium (MEM-HS) Combine 300 ml of H2O, 50 ml of 10× MEM, 20 ml of 5.5% (wt/vol) NaHCO3, 15 ml of 20% (wt/vol) glucose, 5 ml of glutamine (200 mM, 100×), 10 ml of 50× MEM essential amino acids, 10 ml of 100× MEM non-essential amino acids and 50 ml of HS (heat-inactivated for 30 min at 56 °C, several batches). Adjust the pH to 7.3 with 1 M NaOH. Fill up with water up to 500 ml and sterilize the medium by filtration. This medium can be stored at 4 °C for up to 4 weeks. Stock solution for serum-free medium (N-2 stock) To prepare 50-ml aliquots, mix 5 ml of insulin (50 mg of insulin in 10 ml of 0.01 N HCl), 5 ml of progesterone (63 mg of progesterone in 100 ml of absolute ethanol, and from that solution 1 ml in 99 ml of H2O for storing solution), 5 ml of putrescine (161 mg of putrescine in 10 ml of H2O) and 5 ml of selenium (33 mg of selenium in 100 ml of H2O, and from that solution 1 ml diluted in 99 ml of H2O) with 500 mg of human apo-transferrin and bring the volume to 500 ml with N-MEM medium. Sterilize the medium by filtration, divide it into aliquots and store them at −20 °C for up to 6 months.
Supplement solution for serum-free medium (N-MEM) Combine 300 ml of H2O, 50 ml of 10× MEM, 20 ml of 5.5% (wt/vol) NaHCO3 filtered, 15 ml of 20% (wt/vol) d(+)glucose anhydrate, 5 ml of glutamine (100× solution) and 5 ml of pyruvate solution (1.1g of sodium pyruvate in 100 ml of H2O). Bring the solution up to 500 ml with H2O. The solution is usable for ~2 weeks after preparation when kept at 4 °C. Neuronal serum-free medium (N-2 medium) Combine 50 ml of N-2 stock, 450 ml of N-MEM supplement and 100 mg of 1% (wt/vol) ovalbumin (1 g of albumin added to 100 ml of N-MEM). Filter the medium with a cell culture vacuum filter bottle at pH 7.3. This medium can be stored at 4 °C for up to 4 weeks. Mitosis inhibitor ARA-C Combine 12 mg of ARA-C dissolved in 50 ml of H2O. Filter the solution, divide it into aliquots in 500-µl tubes and store them at −20 °C for several months. FM loading solution FM loading solution is 119 mM NaCl, 2.5 mM KCl, 2 mM MgSO4, 2 mM CaCl2, 25 mM HEPES, 30 mM glucose (pH 7.35), 10 µM DNQX, 50 µM APV and 20 µM FM4-64. We always prepare the solution freshly before use. EQUIPMENT SETUP Coverslips It is important that coverslips on which neurons can be unambiguously relocated after axotomy be used. We use coverslips with a laser-engraved grid. Originally, we used Cellocates (Eppendorf, not produced anymore). Currently, we use coverslips that have a photoetched gridded pattern on the surface. If you use these coverslips, clean them, sterilize them and add wax dots to the engraved side, as described in ref. 30. Afterward, place a big drop of the coverslip-coating solution on each coverslip and incubate it overnight at room temperature (21 °C). Do not allow the solution to dry out, as the poly-l-lysine crystals are toxic for neurons. We find that crystallization starts to occur 20 h after the cover solution is added. Wash the coverslips four times with sterile water and fill the dishes with 5 ml of MEM-HS medium. CRITICAL The poly-l-lysine must be added after cleaning, sterilizing and adding the wax dots. Fire-polished Pasteur pipettes Polish a Pasteur pipette at the flame of a Bunsen burner until the diameter of the opening is about one-fourth of the original. If the diameter is too wide, cells will not be completely dissociated; if it is too narrow, neurons might be mechanically injured. Precoating of the polished Pasteur pipettes with sterile 5% (wt/vol) BSA in PBS is recommended because it strongly decreases the sticking of tissue and cells to the pipette during dissociation.
PROCEDURE Mixed WT- and GFP-dissociated neuronal culture setup ● TIMING
