Optimisation of murine organotypic slice culture preparation for a novel sagittal-frontal co-culture system

Journal of Neuroscience Methods 285 (2017) 49–57

Contents lists available at ScienceDirect

Journal of Neuroscience Methods journal homepage: www.elsevier.com/locate/jneumeth

Research paper

Optimisation of murine organotypic slice culture preparation for a novel sagittal-frontal co-culture system Sarah Joost a,∗ , Kazuto Kobayashi b , Andreas Wree a , Stefan Jean-Pierre Haas a a b

Department of Anatomy, Rostock University Medical Center, Gertrudenstraße 9, 18057 Rostock, Germany Department of Molecular Genetics, Institute of Biomedical Sciences, Fukushima Medical University School of Medicine, Fukushima 960-1295, Japan

h i g h l i g h t s • Time-saving optimisation of the preparation process for sagittal slice cultures. • First murine nigrostriatal organotypic slice co-culture system. • First nigrostriatal co-culture system using sagittal organotypic slices.

a r t i c l e

i n f o

Article history: Received 3 March 2017 Received in revised form 21 April 2017 Accepted 3 May 2017 Available online 4 May 2017 Keywords: Slice culture organotypic co-culture Nigrostriatal Embedding Axonal outgrowth Astroglia

a b s t r a c t Background: The nigrostriatal pathway is of great importance for the execution of movements, especially in the context of Parkinson’s disease. In research, analysis of this pathway often requires the application of severe animal experiments. Organotypic nigrostriatal slice cultures offer a resource-saving alternative to animal experiments for research on the nigrostriatal system. New method: We have established a time-saving protocol for the preparation of murine sagittal nigrostriatal slice cultures by using a tissue chopper and agarose embedding instead of a vibratome. Furthermore, we developed the first murine co-culture model and the first co-culture utilising sagittal slices for modelling the nigrostriatal pathway. Results: Sagittal nigrostriatal slice cultures show good overall tissue preservation and a high number of morphologically unimpaired dopaminergic neurons in the substantia nigra. Sagittal-frontal co-culture demonstrates massive outgrowth of dopaminergic fibres from the substantia nigra into co-cultured tissue. Comparison with existing methods: The use of a tissue chopper instead of a vibratome allows notable time-saving during culture preparation, therefore allowing optimisation of the preparation time. Sagittal co-cultures offer the opportunity to study dopaminergic fibres in their physiological environment and in co-cultured tissue from a different animal in the same culture system. Conclusion: We here present a possibility to optimise the slice culture preparation process with the simple means of using a tissue chopper and fast agarose embedding. Furthermore, our sagittal-frontal co-culture system is suitable for the observation of dopaminergic outgrowth in both co-cultured tissues. © 2017 Elsevier B.V. All rights reserved.

1. Introduction The nigrostriatal pathway is a part of the basal ganglia motor loop and therefore of vital importance for the execution of move-

Abbreviations: SN, substantia nigra pars compacta; CPu, caudate-putamen complex; MFB, medial forebrain; eGFP, enhanced green fluorescent proteinbundle; TH, tyrosine hydroxylase; DIV, days in vitro; PBS, phosphate-buffered saline; PFA, paraformaldehyde. ∗ Corresponding author. E-mail addresses: [email protected] (S. Joost), [email protected] (K. Kobayashi), [email protected] (A. Wree), [email protected] (S.J.-P. Haas). http://dx.doi.org/10.1016/j.jneumeth.2017.05.003 0165-0270/© 2017 Elsevier B.V. All rights reserved.

ments. It consists of the dopaminergic neurons of the substantia nigra pars compacta (SN) and their processes running through the medial forebrain bundle (MFB) into the caudate-putamen complex (CPu) (Fuxe et al., 2006). Degeneration of these neurons leads to Parkinson’s disease, the second most common neurodegenerative disorder and therefore a challenge for the public health system for today and in the future (Kalia and Lang, 2015). Analysing and manipulating the nigrostriatal pathway in vivo requires large experimental efforts involving severe animal experiments. Regarding the principles of ethical use of animals in testing, the three R’s (replacement, reduction, refinement; Russell and Burch, 1959), organotypic slice culture preparations could be a valuable alternative for studying and modelling the nigrostriatal

50

S. Joost et al. / Journal of Neuroscience Methods 285 (2017) 49–57

Fig. 1. Preparation of sagittal organotypic slice cultures. (A) The brain of a five day old mouse pup is rapidly dissected and (B) embedded in 4% agarose on the plastic cutting dish of a McIlwain tissue chopper. (C) The embedded brain is sagittally cut at 350 ␮m with the tissue chopper. (D) Excess agarose is cut off with a razor blade and (E) the sliced brain is carefully flushed from the plastic disc. (F) Under the dissection microscope, the sliced brain can be separated into sagittal slices with the aid of microspatulas. (G) Slices containing all components of the nigrostriatal pathway are chosen for culture (the slice marked with an asterisk is suitable for culture). (H) Slices can be transferred on a spatula in a drop of medium. (I) The liquid allows the slice to slip off the spatula onto the semiporous membrane. (J) Excess medium on the membrane will be soaked in after a few minutes and the slice is ready for culture.

pathway. They allow to observe neurons of the nigrostriatal pathway in their physiological three-dimensional environment but do not require the execution of surgical procedures or other interventions on the living animal. Furthermore, they are easily accessible for analysis as well as pharmacological or mechanical lesioning of the dopaminergic system (Daviaud et al., 2013). For organotypic culture, brain tissue is sliced into 300–500 ␮m thick sections which are cultured in rollertubes (Gähwiler, 1981) or on semipermeable membranes following the Stoppini method (Stoppini et al., 1991). The latter technique is less effortful and therefore usually the method of choice for present studies. A broad range of brain regions can be cultured by this technique, e.g. hippocampus, cortex, striatum, SN, locus coeruleus, the basal forebrain, hypothalamus or the olfactory system (Humpel, 2015). Organotypic slices can be kept in culture for up to several weeks and therefore allow short- and long-term observations of the cultured tissue (Gogolla et al., 2006). Concerning slice cultures of the nigrostriatal system, there are basically two approaches: single slice cultures with sagittal slices or co-culture systems composed of frontal slices. In co-culture systems of the nigrostriatal pathway, frontal slices containing SN, CPu and cortex are apposed and dopaminergic fibre outgrowth from the SN towards the CPu can be analysed (Franke et al., 2003). Sagittal slices, on the other hand, contain all components of the nigrostriatal system in one slice and offer a model of the almost complete physiological architecture of the whole pathway. They are especially suited to study effects of lesioning experiments on the dopaminergic system (Kearns et al., 2006; McCaughey-Chapman and Connor, 2016; Ullrich et al., 2011). However, preparing, handling and cul-

ture of these large slices is considerably more challenging than the use of smaller frontal slices. The comparably large sagittal slices are usually prepared with the aid of a vibratome, which is a rather time-consuming procedure. In order to minimise the duration of the preparation procedure, we established a cutting technique for murine sagittal organotypic slices by using a McIllwain Tissue Chopper in combination with agarose embedding. Furthermore, for a detailed study of dopaminergic outgrowth we established a co-culture system of murine sagittal and striatal frontal slices. To our knowledge, this is the first description of a murine co-culture model of the nigrostriatal pathway and the first co-culture utilising sagittal slices. As a further refinement of the method, we used tissue from transgenic mice expressing enhanced green fluorescent protein (eGFP) under the control of the tyrosine hydroxylase (TH) promotor (Sawamoto et al., 2001). Therefore, in our model system, dopaminergic neurons and fibres are labelled by endogenous eGFP expression and can be observed throughout the whole culture period. 2. Materials and methods 2.1. Animals For this study, wildtype C57/BL6 mice or transgenic mice with a C57/BL6 background expressing eGFP under control of the rat TH promotor (Sawamoto et al., 2001) were used. The animals were

S. Joost et al. / Journal of Neuroscience Methods 285 (2017) 49–57

51

Fig. 2. Culture of nigrostriatal slices according to the air-liquid-interface method. (A) Organotypic brain slices are cultured on semiporous membranes. Culture medium below the membrane ensures moistening and nutrition of the tissue. Since the slice is only covered by a thin fluid film, gas exchange is possible. (B) Different brain regions can be identified in slices during preparation and culture and permit easy orientation even at the low magnification of a dissection microscope (Cb: cerebellum, CPu: caudate-putamen complex, Ctx: cortex, Hf: hippocampal formation, MFB: medial forebrain bundle, Ob: olfactory bulb, SN: substantia nigra, Th: thalamus).

kept at 22 ± 2◦ C under a 12 h light/dark cycle with free access to water and standard diet. Each cage (720 cm2 ) contained up to 3 animals and was provided with litter, nesting material, a piece of wood for gnawing and a plastic tube with 4 cm in diameter and 10 cm in length for hiding. All animal-related procedures were conducted in accordance with the local ethical guidelines and the German federal animal welfare law. 2.2. Genotyping Tail samples were lysed in Direct PCR Lysis Reagent Tail (Peqlab, Germany) with 200 ␮g/ml Proteinase K (Peqlab, Germany) shaking for 16 h at room temperature. After centrifugation for 30 s at 6000 rpm, the supernatant was used for genotyping. For polymerase chain reaction, peqGOLD Taq-Polymerase “all inclusive” (Peqlab, Germany) was used following the manufacturer’s instructions. Primers for the TH-eGFP construct were AAGTTCATCTGCACCACCG (forward primer) and TGCTCAGGTAGTGGTTGTCG (reverse primer). Amplificates of 475 bp were detected by electrophoresis in Flash Gels (Lonza, Switzerland). 2.3. Brain dissection Postnatal mouse pups (P5) were separated from their mother and placed in a well ventilated box protected from light. They were kept warm on a warm-water-filled latex glove until preparation. For preparation, one pup at a time was rapidly decapitated with large scissors, the head was cleaned with cotton buds soaked in 70% ethanol and transferred under a horizontal laminar flow work bench (NUAIRE, USA). All following steps were conducted under sterile conditions. Sharp tweezers inserted into the orbital cavities were used to fixate the head. The scalp was cut with fine scissors and pulled sideward with forceps. Then the skull was cut using a clean fine scissor in the median plane and in the frontal plane anterior to the cerebellum (Fig. 1A). The top of the skull was gently removed with forceps and a fine spatula was used to sever the optical nerves and detach the brain from the cranial base. The brain was gently dropped into a petri dish filled with preparation medium at 8 ◦ C (DMEM/F12 supplemented with penicillin (100 U/ml) and streptomycin (100 ␮g/ml), all Sigma, Germany) and meninges and bigger blood vessels were removed with fine forceps under a dissection microscope. During the whole procedure, extreme caution was taken to avoid harming the very vulnerable brain tissue while keeping the preparation time as short as possible. All instruments

used for one animal were rinsed in sterile distilled water twice and then dipped in 70% ethanol and air-dried in the laminar flow hood before use for the next animal. 2.4. Embedding and sectioning Brains were sectioned with a McIllwain tissue chopper (Ted Pella, USA). For sectioning, the desired tissue is placed on a plastic disc which is clamped to the moving table of the tissue chopper. During slicing, the table with the specimen moves forward in defined steps while the cutting arm with an attached razor blade cuts the tissue by moving up and down. For slicing a whole brain, embedding of the tissue in agarose is essential to avoid attachment of the brain tissue to the blade and for fixation of the brain during slicing. Prior to the preparation procedure, 4% agarose (Biozym, Germany) in distilled water was autoclaved in bottles and sterilely distributed into aliquots of 7–8 ml which were sealed, allowed to harden and then stored at 4 ◦ C until use. Before the start of the preparation, one tube of agarose per animal was boiled up in a water bath and kept warm and liquid at 65 ◦ C until use. After the brain was dissected, the agarose of one tube was poured onto a plastic disc. For optimal attachment of the agarose to the disc, the whole disc should be covered with agarose. Caution has to be taken to avoid spilling over the edge of the disc, because a large amount of the agarose will flow down from the disc and the remaining agarose will not suffice to embed the whole brain. Directly after pouring the agarose, the brain was transferred to the disc with a spatula and allowed to sink into the warm agarose. At this time, the agarose has already cooled down to approximately 42–44 ◦ C. It is recommendable to check the agarose temperature with a thermometer during the first executions of the protocol. According to our experience, the experimenter quickly learns to assess the optimal agarose temperature by observing the viscosity of the agarose. With the spatula, some liquid agarose can be spread over the whole brain. Then the agarose was allowed to harden on a metal plate precooled at 4 ◦ C for approximately 3 min. Clamping the disc with the brain onto the moving table of the tissue chopper required trimming of the agarose with a razor blade (Fig. 1B). The disc was fixed on the rotatable moving table and the median plane of brain was adjusted parallel to the blade (Fig. 1C). Then the brain was sliced into 350 ␮m thick sagittal sections, excess agarose was removed with a razor blade (Fig. 1D) and the sliced brain including the surrounding agarose was carefully flushed into a petri dish with cold preparation medium (8◦ C) with the help of a pipette (Fig. 1E).

52

S. Joost et al. / Journal of Neuroscience Methods 285 (2017) 49–57

2.5. Slice culture conditions The slices were carefully separated with microspatula under a dissection microscope (Fig. 1F). Extreme care has to be taken to avoid punching of the slices with the spatula. Even if no injury of the tissue is visible, damaged slices are likely to develop necrosis in culture. Slices from the indicated cutting plane (Fig. 1G) were transferred to 30 mm diameter semiporous membrane inserts (Merck-Millipore, Germany) within a 6-well plate containing 1.2 ml of culture medium (50% MEM, 25% horse serum, 25% HBSS, 6.5 mg/ml glucose, 2 mM l-glutamine, 100 U/ml penicillin, 100 ␮g/ml streptomycin), prewarmed to 37 ◦ C. For transfer, the selected slice was carefully manoeuvred onto a large spatula and drawn out of the preparation medium. A fine spatula prevented the slice from slipping off the large spatula when passing the liquid surface. Then a small amount of preparation medium was carefully dropped onto the spatula with the slice (Fig. 1H). If the slice is surrounded by enough liquid, it will easily slip from the spatula when it touches the semiporous membrane of the membrane insert. Directly after transfer, the position of the slice can be slightly corrected with a fine spatula (Fig. 1I). The excess preparation medium on top of the membrane does not need removal since it diffuses through the membrane and distributes in the medium in approximately half an hour (Fig. 1J). The cultivation of slices at the air-liquid interface guarantees gas exchange as well as moistening and nutrition by the culture medium (Fig. 2A). Slices were incubated in a humidified incubator (Binder, Germany) at 37 ◦ C and 5% CO2 and medium was changed every second day. For the medium change, the membrane insert was lifted from the wellplate with forceps, the entire medium was aspirated and replaced by 1 ml of fresh medium. The exchanged volume of medium was chosen smaller than the desired medium volume of 1.2 ml because of the medium remaining adherent to the membrane insert. If the medium exceeds a volume of 1.2 ml, liquid may accumulate on top of the membrane and interfere with gas exchange of the slices. For daily macroscopic control and documentation of slice cultures, wellplates with membrane inserts were placed on the light source of the dissection microscope and photographed with a Canon D600 single-lens reflex camera with a 100 mm macro lens (Canon, Japan, Fig. 2B). 2.6. Sagittal-frontal nigrostriatal co-culture For sagittal-frontal co-cultures, transgenic mouse brains were cut into sagittal slices as described above. Before transferring the slices onto membrane inserts, the brainstem and cerebellum were dissected off directly caudal of the SN using a scalpel. Afterwards, wildtype mouse brains were cut into 350 ␮m thick frontal slices using the same technique as for sagittal slices. Slices containing the CPu were selected and the CPu was dissected out of the slice with a scalpel. The tissue was then transferred to the same membrane insert as the sagittal slice and positioned directly next to the cut caudal of the SN. 2.7. Immunofluorescence staining of slice cultures After 7–11 days of culture, membrane inserts with attached slices were washed thrice by carefully submerging the whole insert in phosphate-buffered saline (PBS, 0.1 M, pH 7.4). For fixation, membrane inserts were placed in 6-well plates with 2 ml 3.7% paraformaldehyde (PFA, Merck, Germany) in PBS and membrane inserts were additionally filled with 3–4 ml 3.7% PFA in PBS. Slices were fixed for 24 h at 4 ◦ C. Prior to immunostaining, PFA residues were removed by completely immersing the membrane inserts into PBS three times for 1 min. Afterwards, the membrane was cut out of the inserts and slices were separated from each other by cutting the mem-

Fig. 3. Development of macroscopic morphology of organotypic sagittal slices during culture. Directly after preparation (DIV 0), different brain structures are clearly recognisable. During the first days of culture, the tissue becomes opaque and seems to swell up. Starting at DIV 3, slices flatten down and attain more and more transparency.

brane with a scalpel. During the whole staining procedure, slices remained adherent to the membranes. The robust nature of the membrane supports the adherent slices and prevents them from stretching or creasing. All incubation steps were conducted freefloating in 24-well plates on a shaker. Slices were transferred to new wells by gripping the edge of the membrane with fine forceps.

S. Joost et al. / Journal of Neuroscience Methods 285 (2017) 49–57

After washing the slice cultures with PBS 3 × 10 min, slices were blocked with 3% normal horse serum (Sigma, Germany) and 1% Triton X-100 in PBS for one hour at room temperature. The primary antibody (sheep-anti-TH, 1:500, Merck-Millipore Cat# AB1542, RRID:AB 11213126; goat-anti-ChAT, 1:400, MerckMillipore Cat# AB144P, RRID:AB 11214092; goat-anti-GFAP, 1:200, Sigma-Aldrich Cat# SAB2500462, RRID:AB 10603437) was diluted in PBS with 1% normal horse serum and 0.5% Triton X-100. Due to the tissue thickness, incubation times of four days at 4 ◦ C were required for complete penetration of the slice. After incubation, slices were washed in PBS 3 × 30 min. The secondary antibody (mouse-anti-goat, Cy3-conjugated, 1:400, Jackson ImmunoResearch Labs Cat# 205-165-108, RRID:AB 2339065; donkey-anti-sheep, Cy3-conjugated, 1:500, Jackson ImmunoResearch Labs Cat# 713-165-003, RRID:AB 2340727) was diluted in PBS with 1% normal horse serum and 0.5% Triton X-100 and incubated four days at 4 ◦ C. Afterwards, slices were washed 3 × 30 min in PBS, counterstained with DAPI (0.2 ng/ml, Roth, Germany) for 10 min and DAPI residues were removed by washing with PBS 3 × 10 min. For mounting, slices were transferred to SuperFrost Plus (ThermoFisher Scientific, USA) slides in a bowl with distilled water with the aid of forceps gripping the edges of the membrane. The membrane remained between the slide and the slice culture. Then, the slices were allowed to dry on a heating plate at 37 ◦ C until the edges of the membrane slightly start to curl after 5–10 min. If slices dried completely, the membrane would become too corrugated for proper mounting. On the other hand, if slices were not dried long enough, they would not stick to the slide sufficiently. Slices were covered with few drops of fluorescence mounting medium and glass coverslips. 2.8. Perfusion and cryosectioning For the comparison of cultivated slices and the in vivo situation, mice were perfused at P5, brains were cryosectioned and stained immunohistochemically. Prior to perfusion, animals received a lethal overdose of sodium phenobarbital i. p. (60 mg/kg) and were immediately transcardially perfused with 4 ml cold (4◦ C) 0.9% saline and 10 ml 3.7% PFA in PBS. Brains were dissected and postfixed in 3.7% PFA in PBS overnight at 4 ◦ C. For cryoprotection, fixated brains were incubated in 20% sucrose in PBS at 4 ◦ C overnight and then snap-frozen for 5 min in −50 ◦ C isopentane. Sagittal sections of 20 ␮m were cut on a cryostat (Jung, CM Leica, Germany), transferred to Superfrost Plus glass slides (ThermoFisher Scientific, Germany) and dried on a heating plate at 37 ◦ C for 15 min. 2.9. Immunofluorescence staining of frozen sections Immunofluorescence staining was conducted on mounted sections. The staining procedure was identical to the staining of slice cultures with the exceptions of shorter incubation times and lower Triton X-100 concentrations. In brief, sections were washed 3 × 10 min in PBS and blocked with 3% serum and 0.05% Triton X-100. The same primary and secondary antibodies as described above were diluted in PBS containing 1% serum and 0.025% Triton X-100 and incubated overnight at 4◦ . After each antibody incubation, sections were washed thrice in PBS. Counterstaining with DAPI (0.2 ng/ml, Roth, Germany) was conducted for 5 min, sections were washed again three times in PBS, embedded in fluorescence mounting medium and coverslipped. 2.10. Microscopic images Microscopic images were taken using a confocal microscope C1 with the image acquisition software EZ-C1 (Nikon, Japan) or an

53

Aristoplan fluorescence microscope (Leitz, Germany) with a digital camera (SPOT 7.3 Three Shot Color, Diagnostic Instruments Inc, USA). Brightness and contrast of the images were adjusted for optimal recognisability of stained structures. 3. Results and discussion 3.1. Preparation The preparation procedure for sagittal nigrostriatal slice cultures offers a fast and simple method for modelling the nigrostriatal pathway. A skilled experimenter needs approximately 15 min for the whole procedure per mouse pup. One mouse brain at P5 yields 2–4 sagittal slices with all components of the nigrostriatal pathway. 3.2. Macroscopic development of sagittal nigrostriatal slice cultures The macroscopic appearance of slice cultures changed during culture (Fig. 3). Directly after preparation, structures like the CPu, MFB or hippocampal formation were visible, but after the first day in vitro (DIV) the tissue became opaque and seemed to swell up slightly. From DIV 3 onwards, the slices began to flatten and simultaneously became more transparent. This development started at the frontal cortex and spread over the whole slice until DIV 7. During this process, slices tightly attached to the membrane and remained adherent through all subsequent staining procedures. Flattening and transparency of slices as well as attachment to the membrane is described as a sign of viability (Humpel, 2015) and demonstrates the preservation of intact tissue in slice culture in general. 3.3. Preservation of the nigrostriatal pathway in organotypic slice cultures in comparison to the in vivo situation The most important criterion for evaluating the use of organotypic slice cultures as a model for the nigrostriatal pathway is the preservation of the dopaminergic cells in the SN after culture. The transgenic mouse used in this study allows easy recognition of these cells by expressing eGFP under the control of the TH promotor. After 7 DIV, a large number of TH-eGFP-positive cells can be found in the SN of organotypic slice cultures (Fig. 4A). They exhibited an unimpaired neuronal morphology with triangular perikarya and axonal and dendritic processes (Fig. 4C). This state is very similar to the situation in vivo in mice at the preparation age of five days (Fig. 4B and D). However, due to the low thickness of cryosections (20 ␮m) compared to slice cultures (350 ␮m at the beginning of culturing) TH-eGFP-expressing cells in vivo give the impression of a lower cell density in the SN. The processes of dopaminergic neurons exit the SN and proceed through the MFB towards the CPu. The low eGFP expression in cell processes scarcely sufficed for satisfying illustration, so immunofluorescence stainings against TH were conducted for the detection of TH-positive fibres. In vivo, a continuous bundle of THpositive fibres ran from the SN along the MFB towards the forebrain where they fanned out to innervate the CPu (Fig. 4G). In slice cultures, the fibre density after 11 DIV was distinctly reduced. Only single fibres made their way from the SN in direction to the CPu, ending before reaching their final target (Fig. 4E, F). This suggests the assumption that dopaminergic processes degenerate during culture while cell bodies stay viable and morphologically unimpaired. Another explanation could be an impairment of anterograde axonal transport in dopaminergic processes resulting in failure of TH transport. However, the dopaminergic neurons begin to grow new processes towards the CPu which explains why the processes are not yet reaching the CPu after 11 DIV. The reason of the fibre

54

S. Joost et al. / Journal of Neuroscience Methods 285 (2017) 49–57

Fig. 4. Preservation of the nigrostriatal pathway in organotypic slice cultures compared to the situation in vivo. (A) The SN from a slice culture after seven days of culture shows a large number of TH-eGFP expressing neurons. (B) TH-eGFP-positive cells in the SN of a five day old mouse. (C) Detailed view of TH-eGFP-positive cells from the SN of an organotypic slice after 7 DIV. (D) Close-up on TH-eGFP-expressing cells from a five day old mouse. (E) Overview of a representable slice directly after preparation. The position of (F) is marked by a square. (F) TH-containing fibres were stained immunohistochemically in an organotypic slice culture after 11 DIV. Fibres originate from the SN and are directed towards the CPu. (G) Immunostaining of TH-expressing fibres reveals TH-positive fibres extending from the SN to the CPu in the MFB in a 5 day old mouse. (H) Cholinergic interneurons of the CPu are immunohistochemically stained against ChAT in an organotypic slice culture at 8 DIV. (I) The CPu of five day old mice contains ChAT-positive cholinergic interneurons.

S. Joost et al. / Journal of Neuroscience Methods 285 (2017) 49–57

55

Fig. 5. Glial morphology in organotypic slice cultures in comparison to the situation in vivo. (A) The morphology of astrocytes in organotypic slice cultures at DIV 7 is revealed by immunostaining against GFAP. The high number of processes is typical for reactive astrocytes. (B) Immunohistochemically marked astrocytes in a 5 day old mouse show the morphology of resting astrocytes. (C) GFAP-positive cells in the cortex of organotypic slice cultures at DIV 7 show the long parallel processes of radial glial cells. (D) In the cortex of five day old mice, GFAP-immunostaining demonstrates the presence of radial glia. (E) At the edges of organotypic slice cultures at DIV 7 (dashed line), GFAP-stained glial cells show outgrowth from the slice.

degeneration might be found in involuntary truncation of the MFB during slicing as suggested by Daviaud et al. (2014). To address this issue, Beurrier et al. (2006) introduced a technique with the cutting plane tilted by 10◦ . On these slices, they were able to show the functional integrity of the nigrostriatal pathway directly after preparation but did not pursue this technique further in slice cultures. On the other hand, Kearns et al. (2006) showed a completely preserved nigrostriatal pathway after three weeks of culturing murine organotypic slices. The target structure of the dopaminergic neurons of the SN is the CPu. It is mainly characterised by GABAergic projection neurons and cholinergic interneurons. The latter ones were stained exemplarily by immunofluorescence staining against the marker enzyme choline acetyl transferase (ChAT). Both organotypic slice cultures after 8 DIV (Fig. 4H) and in vivo sections from five day old mice (Fig. 4I) showed numerous ChAT-positive neurons in the CPu, thereby demonstrating the good preservation of this tissue during culture.

3.4. Glial morphology in organotypic slice cultures A central advantage of organotypic slice culture is the preserved composition of different tissue-specific cell types in their physiological three-dimensional arrangement. In case of brain slices, this especially includes glial cells in addition to neuronal cell populations. For this reason, we analysed the glial morphology by immunohistochemical staining against glial fibrillary acidic protein (GFAP). After 7 DIV, organotypic slice cultures show a high number of GFAP-positive astrocytic cells with numerous processes, the typical morphology of activated astrocytes (Fig. 5A). The in vivo situation in five day old mice on the contrary exhibits fewer and smaller astrocytes with considerably fewer processes, which are likely to be resting astrocytes (Fig. 5B). A massive astrogliosis is typical for severe injuries of brain tissue (Pekny et al., 2016), so its occurrence in organotypic brain slices as a reaction towards the slicing procedure is not unexpected. Some groups even utilised this reaction for the establishment of a slice culture model of reactive astrogliosis (Damiani and O’Callaghan, 2007; Dean et al., 2011).

56

S. Joost et al. / Journal of Neuroscience Methods 285 (2017) 49–57

Fig. 6. Outgrowth of TH-positive processes in organotypic slice cultures. (A) In a frontal-sagittal co-culture system, a sagittal slice from a TH-eGFP-expressing mouse is cut directly posterior to the SN to remove the brain stem. CPu-containing tissue from the frontal slice of a wildtype mouse is positioned directly next to the SN of the sagittal slice. The border between the slices is marked by a dashed line and SN, CPu, hippocampal formation (Hf) and olfactory bulb (Ob) are labelled in the sagittal slice. (B) The frontal-sagittal co-culture from (A) after 11 days of culture. The dashed line indicates the border between the slices. The asterisk labels the position of the SN and the square marks the position of (D). (C) In co-cultures of a TH-eGFP-positive sagittal slice and wildtype CPu tissue, TH-eGFP-expressing processes from cells of the SN (position marked by asterisk in (B)) expand into the grafted wildtype tissue after 11 days in culture. The dashed line indicates the border between the sagittal and frontal slice. (D) TH-eGFP-expressing processes spread several hundred micrometers into wildtype CPu-tissue after 11 days of culture. The position is indicated by the square in (B). (E) Brain stem and cerebellum of a sagittal slice were cut off in direct proximity to the SN before cultivating the slice for 9 days. The cutting line is illustrated by the dashed line and the square shows the position of (F). (F) Processes of TH-eGFP-positive cells spread out of the slice culture after 9 DIV. The border of the slice is indicated by the dashed line and the position of this image is marked by the square in (E).

In the cortex of nigrostriatal slice cultures after 7 DIV, GFAP staining reveals numerous parallel GFAP-positive fibres perpendicular to the slice surface (Fig. 5C). This morphology is typical for radial glial cells which are characteristic for prenatal and early postnatal development of the cortex (Kriegstein and Alvarez-Buylla, 2009). Alternatively, the fibre orientation might have been induced by the artificial plain surface that originates from the cutting procedure. The same morphology can be found in vivo in five day old mice (Fig. 5D). In 14 day old mice, on the contrary, no radial glial cells can be detected (data not shown), so the preservation of radial

glial cells in organotypic slice cultures demonstrates a retention of the developmental state of the cortex at the moment of brain preparation. Concentrating on the edges of the cultured slices, outgrowth of glial cells is obvious (Fig. 5E). This correlates well with the observation of adhesion of the slice to the semiporous membrane after a few days of culture. Slices stick considerably tough to the membrane due to this glial outgrowth, so even after days of shaking for free-floating staining procedures, slices stay firmly attached to the membranes.

S. Joost et al. / Journal of Neuroscience Methods 285 (2017) 49–57

3.5. Outgrowth of TH-positive processes in sagittal-frontal co-cultures The growth of TH-positive fibres in organotypic slice cultures from the SN in direction towards the CPu hints at a large potential of fibre regeneration in this slice culture model. To further analyse this capacity, we established a co-culture model consisting of sagittal slices from transgenic TH-eGFP-positive animals and frontal slices from wildtype mice. In the sagittal slices, brain stem and cerebellum were cut off directly posterior to the SN before transferring the slice onto the membrane. Quadratic sections of CPu tissue were dissected out from the frontal slices and positioned directly next to the sagittal slices (Fig. 6A). After 11 DIV, the outgrowth of TH-eGFPpositive fibres into the wildtype CPu tissue was analysed (Fig. 6B). The border between transgenic and wildtype tissue was hardly detectable which hints at optimal survival and integration of the neighbouring tissues. The SN was located in direct proximity to the junction of the slices and a large amount of delicate THpositive fibres stretched out from the SN into the surrounding tissue (Fig. 6C). The border between the slices was often crossed and some fibres even grew several hundred micrometres into the wildtype CPu tissue (Fig. 6D). To further challenge the ability of fibre growth, we also cultivated sagittal slices without brain stem and cerebellum as described above, but without adding of wildtype CPu next to the SN (Fig. 6E). Even without tissue to grow into, TH-eGFP-expressing fibres extended from the SN, leaving the slice and growing on the bare membrane (Fig. 6F). The large potential of dopaminergic fibre regeneration in organotypic slice cultures has already been demonstrated in frontal co-culture systems consisting of SN, CPu and sometimes also cortex tissue (Heine and Franke, 2014). These model systems are especially used to assess the influence of growth factors or pharmacological compounds on fibre outgrowth (Heine et al., 2013; Jaumotte and Zigmond, 2005). A special benefit of co-culture systems is the possibility to combine tissues from different transgenic animals to study the influence of spatially limited overexpression or knock-down of e.g. growth factors on fibre outgrowth. For this purpose, the introduction of a murine co-culture system of the nigrostriatal pathway is particularly advantageous because of the high availability of murine transgenic model systems. 4. Conclusion We here present a time-saving alternative for the preparation of nigrostriatal sagittal slice cultures by using a tissue chopper combined with agarose embedding of the brain instead of a vibratome. The obtained slice cultures show well preserved TH-positive neurons in the SN and robust outgrowth of dopaminergic fibres. Furthermore, we introduced a sagittal-frontal co-culture system nicely suited to observe fibre outgrowth from the SN. Funding sources This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Acknowledgements We would like to thank Dr. Jakob Wolfart and M. Sc. Deborah Laker (Oscar Langendorff Institute of Physiology, Rostock University Medical Center) for the detailed introduction into slice culture

57

techniques. Furthermore, thanks are due to Susan Lehmann and Laura Hiepe (Department of Anatomy, Rostock University Medical Center) for excellent technical support. References Beurrier, C., Ben-Ari, Y., Hammond, C., 2006. Preservation of the direct and indirect pathways in an in vitro preparation of the mouse basal ganglia. Neuroscience 140 (1), 77–86. Damiani, C.L., O’Callaghan, J.P., 2007. Recapitulation of cell signaling events associated with astrogliosis using the brain slice preparation. J. Neurochem. 100 (3), 720–726. Daviaud, N., Garbayo, E., Schiller, P.C., Perez-Pinzon, M., Montero-Menei, C.N., 2013. Organotypic cultures as tools for optimizing central nervous system cell therapies. Exp. Neurol. 248, 429–440. Daviaud, N., Garbayo, E., Lautram, N., Franconi, F., Lemaire, L., Perez-Pinzon, M., et al., 2014. Modeling nigrostriatal degeneration in organotypic cultures, a new ex vivo model of Parkinson’s disease. Neuroscience 256, 10–22. Dean, J.M., Riddle, A., Maire, J., Hansen, K.D., Preston, M., Barnes, A.P., et al., 2011. An organotypic slice culture model of chronic white matter injury with maturation arrest of oligodendrocyte progenitors. Mol. Neurodegener. 6 (1), 46, http://dx.doi.org/10.1186/1750-1326-6-46. Franke, H., Schelhorn, N., Illes, P., 2003. Dopaminergic neurons develop axonal projections to their target areas in organotypic co-cultures of the ventral mesencephalon and the striatum/prefrontal cortex. Neurochem. Int. 42, 431–439. Fuxe, K., Manger, P., Genedani, S., Agnati, L., 2006. The nigrostriatal DA pathway and Parkinson’s disease. J. Neural Transm. 70 (Supplementum), 71–83. Gähwiler, B.H., 1981. Organotypic monolayer cultures of nervous tissue. J. Neurosci. Methods 4 (4), 329–342 http://www.sciencedirect.com/science/ article/pii/0165027081900030. Gogolla, N., Galimberti, I., DePaola, V., Caroni, P., 2006. Preparation of organotypic hippocampal slice cultures for long-term live imaging. Nat. Protoc. 1 (3), 1165–1171. Heine, C., Franke, H., 2014. Organotypic slice co-culture systems to study axon regeneration in the dopaminergic system ex vivo. Methods Mol. Biol. (Clifton, N.J.) 1162, 97–111. Heine, C., Sygnecka, K., Scherf, N., Berndt, A., Egerland, U., Hage, T., et al., 2013. Phosphodiesterase 2 inhibitors promote axonal outgrowth in organotypic slice co-cultures. Neurosignals 21 (3–4), 197–212. Humpel, C., 2015. Organotypic brain slice cultures: a review. Neuroscience 305, 86–98. Jaumotte, J.D., Zigmond, M.J., 2005. Dopaminergic innervation of forebrain by ventral mesencephalon in organotypic slice co-cultures: effects of GDNF. Brain Res. Mol. Brain Res. 134 (1), 139–146. Kalia, L.V., Lang, A.E., 2015. Parkinson’s disease. Lancet (Lond., Engl.) 386 (9996), 896–912. Kearns, S.M., Scheffler, B., Goetz, A.K., Lin, D.D., Baker, H.D., Roper, S.N., et al., 2006. A method for a more complete in vitro Parkinson’s model: slice culture bioassay for modeling maintenance and repair of the nigrostriatal circuit. J. Neurosci. Methods 157 (1), 1–9. Kriegstein, A., Alvarez-Buylla, A., 2009. The glial nature of embryonic and adult neural stem cells. Annu. Rev. Neurosci. 32, 149–184. McCaughey-Chapman, A., Connor, B., 2016. Rat brain sagittal organotypic slice cultures as an ex vivo dopamine cell loss system. J. Neurosci. Methods 277, 83–87. Pekny, M., Pekna, M., Messing, A., Steinhauser, C., Lee, J.-M., Parpura, V., et al., 2016. Astrocytes: a central element in neurological diseases. Acta Neuropathol. (Berl.) 131 (3), 323–345. Russell, W.M.S., Burch, R.L., 1959. The Principles of Humane Experimental Technique. Universities Federation for Animal Welfare, London. Sawamoto, K., Nakao, N., Kobayashi, K., Matsushita, N., Takahashi, H., Kakishita, K., et al., 2001. Visualization, direct isolation, and transplantation of midbrain dopaminergic neurons. Proc. Natl. Acad. Sci. U. S. A. 98 (11), 6423–6428. Stoppini, L., Buchs, P.A., Muller, D., 1991. A simple method for organotypic cultures of nervous tissue. J. Neurosci. Methods 37 (2), 173–182. Ullrich, C., Daschil, N., Humpel, C., 2011. Organotypic vibrosections: novel whole sagittal brain cultures. J. Neurosci. Methods 201 (1), 131–141.