Here we describe a technique for studying hippocampal postnatal neurogenesis using the organotypic slice culture technique. This method allows for in vitro manipulation of adult neurogenesis and allows for the direct application of pharmacological agents to the cultured hippocampus.
Here we describe a technique for studying hippocampal postnatal neurogenesis in the rodent brain using the organotypic slice culture technique. This method maintains the characteristic topographical morphology of the hippocampus while allowing direct application of pharmacological agents to the developing hippocampal dentate gyrus. Additionally, slice cultures can be maintained for up to 4 weeks and thus, allow one to study the maturation process of newborn granule neurons. Slice cultures allow for efficient pharmacological manipulation of hippocampal slices while excluding complex variables such as uncertainties related to the deep anatomic location of the hippocampus as well as the blood brain barrier. For these reasons, we sought to optimize organotypic slice cultures specifically for postnatal neurogenesis research.
Adult neurogenesis in the mammalian hippocampus represents a remarkable example of the brain’s innate capacity for adaptability and plasticity. Dentate granule cells (DGCs) are generated from a renewable pool of neural progenitor cells in the hippocampal dentate gyrus, which is one of the two presently well-characterized neurogenic regions in the mammalian brain, and is thought to be particularly important for learning and memory. The hippocampus is part of the limbic system and has a deep location within the mammalian brain, which makes it a difficult target for precise pharmacological manipulation. Additionally, aberrant neurogenesis has been implicated in conditions, such as epilepsy, schizophrenia, and Alzheimer’s disease, which has prompted interest in understanding the influence of various pharmacological agents during the maturation and survival of newborn neurons. The distinction between postnatal and adult neurogenesis is blurred and previous studies have shown that many features of in vivo neuronal development in the early postnatal period and adulthood are similar25. Here we emphasize postnatal neurogenesis and suggest possible applications to adult neurogenesis.
Organotypic slice cultures provide an efficient in vitro method for studying various physiological properties of the mammalian hippocampus. The value of slice cultures prepared from rodent brains can be summarized in three main qualities: 1) the protocol is straightforward and requires readily available materials; 2) slice cultures allow for pharmacological studies that eliminate complex variables such as the deep anatomic location of the hippocampus and circumvents the blood brain barrier1; and 3) the well characterized structure of the hippocampus and tri-synaptic circuit is preserved2. Previous investigators have used the organotypic hippocampal culture to study synaptic development and physiology3,4, gliogenesis5-7, ischemic brain damage8,9, neuroprotection and neurorepair10-12 as well as epilepsy13-15.The slice cultures could also provide a useful model system allowing for the monitoring of cell development in conjunction with labeling of cells with green fluorescent protein (GFP) or other vital markers.
Slice cultures have also been previously employed to study postnatal hippocampal neurogenesis16-19, but one important factor in the majority of these studies is the well-characterized degeneration that results from explanting tissue from adult animals after approximately 2 weeks in vitro20,21. For this reason, slice cultures are typically prepared from early post-natal (P5-P10) mice or rat pups, which utilizes the improved viability of early postnatal brain tissue for culturing22. While previous studies have shown that the early postnatal and adult hippocampus differ with regards to synaptic physiology and the expression of specific neuronal subtypes23,24, there is substantial conservation of the choreographed developmental program that newborn dentate granule cells proceed through during maturation25. Additionally, recent studies have suggested that the physiological characteristics of newborn DGCs in culture are very similar to immature neurons in the acute hippocampal slice preparation16.
NOTE: All animal procedures conformed to the animal health and welfare guidelines of the Department of Comparative Medicine at the University of Toronto.
1. Preparation of Hippocampal Slices
2. Arrange Dissection Tools in Sterile Laminar Flow hood
3. Hippocampal Dissection
4. Feeding and Maintaining Organotypic Slices
5. Incubating Tissue Slices with Thymidine Analogues to Label Newborn Neurons
6. Tissue Fixation and Storage
7. Sectioning Tissue for Immunohistochemistry
Determining if organotypic cultures would be suitable for adult neurogenesis research required that they satisfy two main criteria: 1) that slices maintain characteristic morphological features of hippocampal slices after 10-21 days in vitro (DIV), and 2) that newborn DGCs can be quantified using standard immunohistochemical techniques commonly employed in adult neurogenesis research. Regarding the first criterion, Figure 1A and 1B highlight the preserved hippocampal morphology. Characteristic features such as the dentate gyrus (DG), CA1, and CA3 regions are easily identifiable.
Regarding the second criterion, Figure 1C (upper panel) provides a representative sample of newborn DGCs co-expressing the endogenous neuronal marker, Doublecortin (DCX) in green and the exogenous thymidine analogue, 5-Chloro-2′-deoxyuridine (CldU) in red. These neurons are located in the sub-granular zone of the hippocampal DG. In order to ensure correct birth dating of neurons, we identify CldU+ nuclei that co-express DCX. Confocal microscopy is needed at this stage to successfully identify double-labeled neurons because candidate cells must co-express the marker of interest throughout the Z-axis of the cell. Sample data obtained with such double-labeling yield approximately 17% of CldU+ cells that expressed DCX and 35% of CldU+ cells that expressed CaBP at 12 days after CldU application. The DCX value is very similar whereas the CaBP value is considerably lower than comparable figure obtained in vivo26. Standard tissue culture conditions may be responsible for a relatively low percentage of the CaBP+ cells.
Figure 2A presents the dissection steps that proceed from left to right (1-8): starting with decapitation of animal (1), removal of brain (2), transfer of brain to ice-cold dissection solution (3), dissection of hippocampus from left and right hemispheres (4), storage of dissected hippocampi in ice-cold dissection solution (5), transfer of both hippocampi to Stoelting tissue chopper and sectioning at 400 μm (6), separation of individual slices under dissecting microscope (7), and plating tissue on cell culture inserts (8). Proceeding in this manner helps maintain a sterile environment throughout the culture process.
Lastly, since the time-course of development is an important feature of hippocampal neurogenesis, we chose to incubate the cultured slices with CldU for exactly 2 hr after 3 DIV to label dividing neural stem cells. The narrow time window for CldU administration was chosen to improve the likelihood that labeled neurons constituted a homogeneous population of cells at approximately the same maturational stage (Figure 2). With regards to CldU labeling, one critical feature of neurogenesis for hippocampal function is that at a given time there is a heterogeneous population of dentate granule cells at various maturational stages27,28.
Figure 1. Sample fluorescence microscope photographs highlighting preserved hippocampal morphology. (A) Organotypic slice from 12 days post dissection immunolabeled for CldU (green) and CaBP (red), 20x air composite image (Scale bar= 500 µm). (B) Sample micrograph of slice from 21 days post dissection immunolabeled for CldU (red) and DCX (green) (opposite color-scheme of Figure 1A), 20x air (Scale bar= 100 µm). (C) Upper panel. Representative confocal microscope photograph of cells co-expressing ClXdU and endogenous immature neuronal marker, DCX. DGCs co-expressing DCX (green) and CldU (red) are counted as newborn neurons. Arrow indicates a double-labeled cell at early stage of development, 40X oil-immersion (Scale bar = 10 µm). Comparable cells have been observed 10 days post-labeling in vivo26. Lower panel. Representative fluorescent images of cells co-expressing CldU (green) and CaBP (red). Arrow indicates a double-labeled cell, 40X fluorescent micrograph (Scale bar = 10 μm). DG-dentate gyrus. GCL-granule cell layer. Please click here to view a larger version of this figure.
Figure 2. (A) Illustration of the sequential steps for hippocampal dissection in laminar flow clean bench. Dotted line indicates “unsterile” (left) and “sterile” (right) zones of the dissection area. (B) Timeline for organotypic slice cultures prepared from P7 rat pups (start). Notations indicate application of thymidine analogue, CldU*, and “treatment,” which can include various pharmacological agents suited to the experimental question. Cultures are fixed with paraformaldehyde at desired dwell times from CldU application (Fix Cultures). Please click here to view a larger version of this figure.
Name of Reagent/ Equipment | Company | Catalog Number | Comments/Description |
5-chloro-2'-deoxyuridine (CldU) | MP Biomedicals | 105478 | Hazardous, Carcinogenic |
Cell culture inserts, 30 mm diameter, 0.4 µm pore size | Thermo scientific | 140660 | Nuclon delta coating on these inserts provides better tissue adhesion and improves slice quality. |
Conical Centrifuge tubes (sterile) | Fisher Scientific | 14-432-22 | |
Dissector scissors (angled to side) | Fine Science Tools | 14082-09 | |
Minimum essential medium (MEM) | Gibco | 11095; liquid | Store at 4 °C |
Eclipse Ni-U fluorescent microscope | Nikon | ||
Glue for tissue | Krazy Glue | KG585 | Use minimum amount of glue to achieve adhesion as any tissue exposed to glue will be unusable for IHC. |
Hank’s Balanced Salt Solution (HBSS) (500 ml) | Gibco | 14025-092 | Store at 4 °C |
Horse Serum Heat Inactivated (500 ml) | Gibco | 16050-122 | Make 50 ml aliquots and store at -20 °C |
Kimwipes | Kimberly-Clarke | TW 31KYPBX | |
Modified glass pipettes (bottom of Pasteur pipette removed and edge smoothed with Bunsen flame) | |||
Petri Dish (100 mm x 15 mm) and (60 mm x 15 mm) | Fisher Brand | FB0875712 and FB0875713A | |
Scalpel blades #11 | Fine Science Tools | 10011-00 | |
Scalpel handle #3 | Fine Science Tools | 10003-12 | |
Serological Pipettes | Sorfa Medical Plastic Co. | P8050 | |
Standard Pattern forceps | Fine Science Tools | 11000-12 | |
Sterile vacuum filter | Thermo-Scientific | 565-0020 | |
Surgical Scissors | Fine Science Tools | 14054-13 | |
Syringe driven filter unit | Millipore-Millex | SLGP033RS | |
Tissue chopper with moveable stage | Stoelting | 51425 | |
Fine tip paintbrush |
Table 1. Supplies and Reagents
Solution | Ingredients and Instructions |
Dissection solution | a) 500 ml of Hank's Balanced Salt Solution (HBSS) (Gibco-14025-092). |
b) Add 2.2 g D-glucose. | |
c) Add 0.5 g Sucrose. | |
d) Add 1.787 g HEPES. | |
e) Mix for 30 min with magnetic stir plate. | |
f) Use pH meter to ensure solution has a final pH= 7.4. | |
g) Use osmometer to ensure final osmolality= 320-330 mOsm. | |
h) Sterilize solution in sterile laminar flow hood using vacuum filtration through 0.2 μm filter. | |
Serum-containing culture medium: 100 ml Minimum Essential Medium (MEM) (Gibco 11095), 50 ml Horse serum (Gibco 16050-122), 50 ml HBSS. | a) Add the following to 50 ml HBSS in beaker and dissolve in 37°C water bath. Mix with magnetic stirrer. |
b) 1.3 g D-glucose. | |
c) 36 mg MgSO4. | |
d) 17.6 mg Ascorbic acid. | |
e) 5 μl of 2M CaCl2 stock solution. | |
f) Add 50 μl Antibiotic-Antimycotic (100x stock, sterile; Gibco 15140-062). | |
g) 1 μg/ml Insulin. | |
h) Sterilize above solution by filtration through a 0.2 μm filter. | |
i) Mix filtered solution with 100 ml MEM and 50 ml Horse serum in laminar flow hood. | |
j) Make 50 ml aliquots in sterile conical centrifuge tubes (Fisher Scientific-14-432-22) and store at 4°C. | |
4% Paraformaldehyde fixative solution. | a) Prepare phosphate buffered saline (PBS) by adding the following to 300 ml of distilled H2O and mixing on magnetic stir plate. |
b) Add 2.7 g sodium phosphate monobasic (NaH2PO4). | |
c) Add 11.5 g sodium phosphate dibasic (NaHPO4). | |
d) Add 9.0 g sodium chloride (NaCl). | |
e) Heat approx. 700 ml of distilled H2O to 55 °C and turn off heat. | |
f) Add 40 g paraformaldehyde (PFA) and stir into 700 ml of water using magnetic stir plate. | |
g) Combine the PBS (a,b,c,d) and PFA (e,f) solutions, adjust the pH to 7.4 and top up to final volume of 1,000 ml. | |
0.1% Sodium Azide Solution | a) Add 1g of powdered sodium azide (NaN3) to 1 L of PBS solution. |
b) Mix using magnetic stir plate and store at 4°C. |
Table 2. Solutions and Recipes
Following CldU (or BrdU) administration, the timeline of application of pharmacological agents can be chosen to target newborn DGCs during particular developmental windows. For example, a hypothetical agent can be applied during the second week post-CldU injection, which is proposed to coincide with the age of immature neurons that are at a developmental stage where GABA is depolarizing. Future studies using this protocol could adapt the pharmacological agent and the window of exposure to “tailor” the approach to the specific experimental question of interest.
An important criterion for determining that slice cultures are a valid model for postnatal neurogenesis research is the ability to stain and quantify newborn neurons in the hippocampus. The two main findings in support of this hypothesis were that microscopic analysis revealed immunohistochemical reactivity for CldU and endogenous protein markers in the same neurons. When used in combination with endogenous neuronal markers, thymidine analogues such as BrdU and CldU are powerful tools for neurogenesis research.
The application of thymidine analogues, such as BrdU, via intraperitoneal injections is commonly utilized in neurogenesis research to label neurons undergoing S-phase of mitosis29. Similar approaches can be employed in organotypic cultures with certain modifications. For example, previous studies administered BrdU (0.5 μM for 3 days) to slice cultures after ~14 DIV18. Reviewing the data presented in that paper reveals that some of the metrics used for quantifying neurogenesis do not employ the standard techniques used in the neurogenesis field, i.e., stereological quantification30. For example, when reporting the co-expression of BrdU and Neuronal-nuclei (NeuN) positive cells, they indicate a total number of cells “per culture” instead of providing information regarding tissue area or volume.
Subsequent studies improved on this method by sectioning the cultured slices to individual 10 μm sections, which improved visualization and immunohistochemistry protocols by allowing antibodies to more readily permeate the tissue samples31. Bunk et al.28 reported double labeling with BrdU (10 μM for 3 days) as the number of co-labeled cells per 10 μm section, but did not provide information about the comparative area or specific hippocampal region studied i.e., CA1, CA3 or DG. Additionally, analysis of the confocal and fluorescence microscopy images does not convincingly show that hippocampal morphology was successfully maintained.
Importantly, both studies used a BrdU exposure period of 3 days, which has associated drawbacks. BrdU labeling has greatly aided neurogenesis studies by allowing investigators to track newly divided cells in various brain regions. However, BrdU toxicity has also been well characterized. Its use has been shown to cause morphological and behavioral abnormalities32,33 and negative effects on cell cycle, differentiation, migration and survival of neural stem cells34-36. The prolonged administration of BrdU in the previously mentioned studies may have introduced confounding variables that altered hippocampal physiology and while some side effects from BrdU administration may be unavoidable, our experimental protocol was designed to limit some of these complications by incubating the tissue with thymidine analogues for 2 hr. Additionally, we chose to use CldU instead of BrdU because it showed better solubility than BrdU when preparing the incubating solution. Although the 3 day protocol may be useful for certain experimental designs e.g., maximizing the labeling of proliferating cells, this 2 hr protocol has an advantage of pulse-labeling of a relatively small population of cells which can be studied at desired survival times (see Figure 2B).
By comparing the level of neuronal production following two different methods of BrdU application, Namba et al. made an important contribution to labeling techniques in organotypic slice cultures37. The authors compared intraperitoneal (I.P.) injection of BrdU (50 mg/kg) in postnatal day 5 (P5) rats with in vitro cultures that received culture medium containing 1 μM BrdU for 30 min immediately following explantation of tissue. They report no statistically significant difference between in-vivo and early in vitro BrdU injection in cultured tissue. The authors did not present clear images outlining the hippocampal structure but they report BrdU immunoreactivity as percentages of total cells in the granule cell layer. While they employ stereological counting, providing a measure by area or volume would be valuable. In general, the cited studies present organotypic cultures as a thorough detailing of postnatal hippocampal slice cultures, with applications for the study of neurogenesis and pharmacological perturbations. Using this technique, hippocampal slices can be maintained for up to 21 days in vitro (DIV) and drugs can be added to the medium at any point in the culture period to study the effect on neurogenesis.
Our aim was to label a discrete, relatively homogenous population of DGCs by providing a brief ‘pulse’ application of CldU for 2 hr. One commonly employed strategy for studying neurogenesis involves identification of a neuron’s maturational stage via immunohistochemical staining for various endogenous markers with a thymidine analogue. Confocal and fluorescence microscopy confirmed the presence of nuclei that incorporated the thymidine analogue CldU and were therefore actively undergoing mitosis during the culture period. Figure 2 provides evidence that immunohistochemical protocols commonly used for in-vivo tissue analysis can be adapted for slice cultures.
Specifically, a commonly used method in neurogenesis research is to perform immunohistochemical staining for the microtubule associated protein, DCX, which is predominantly expressed in immature neurons from day 3-21 and the mature neuronal marker, Calbindin (CaBP), which is fully expressed following 28 days post-mitosis. The phenotype of CldU+ cells was determined using these endogenous markers26.
Improved methods for maintaining slice cultures for longer periods may have the additional benefit of allowing more CldU+ neurons to reach the mature, CaBP+ stage. At present, one limitation of the organotypic culture approach is that the tissue is continuously changing during the culture period. For example, immediately following hippocampal dissection and plating of slices, the tissue has a width of approximately 400 μm. However, after 2-3 weeks in the incubation chamber, tissue slices will begin to thin, which results in a final width between 250-350 μm. This limits the amount of tissue that can be used for immunohistochemistry and should be considered when planning how many animals to use for a project. Additional experiments will help characterize the functional changes in hippocampal physiology that occur in vitro.
The protocol for sectioning and staining hippocampal slices was developed to analyze cellular and morphological changes taking place during the culturing period. Slice cultures provide an opportunity to test the effect of various pharmacological agents as hippocampal DGCs pass through distinct developmental stages during maturation and represent a valuable tool for future adult neurogenesis studies.
The authors have nothing to disclose.
This work was supported by a research grant MOP 119271 to JMW by the Canadian Institute of Health Research. The authors would like to thank Ms. Yao Fang Tan for her outstanding technical assistance.
Name of Reagent/ Equipment | Company | Catalog Number | Comments/Description |
5-chloro-2'-deoxyuridine (CldU) | MP Biomedicals | 105478 | Hazardous, Carcinogenic |
Cell culture inserts, 30mm diameter, 0.4µm pore size | Thermo scientific | 140660 | Nuclon delta coating on these inserts provides better tissue adhesion and improves slice quality. |
Conical Centrifuge tubes (sterile) | Fisher Scientific | 14-432-22 | |
Dissector scissors (angled to side) | Fine Science Tools | 14082-09 | |
Minimum essential medium (MEM) | Gibco | 11095; liquid | Store at 4°C |
Eclipse Ni-U fluorescent microscope | Nikon | ||
Glue for tissue | Krazy Glue | KG585 | Use minimum amount of glue to achieve adhesion as any tissue exposed to glue will be unusable for IHC. |
Hank’s Balanced Salt Solution (HBSS) (500 mL) | Gibco | 14025-092 | Store at 4°C |
Horse Serum Heat Inactivated (500 mL) | Gibco | 16050-122 | Make 50 mL aliquots and store at -20°C |
Kimwipes | Kimberly-Clarke | TW 31KYPBX | |
Modified glass pipettes (bottom of Pasteur pipette removed and edge smoothed with Bunsen flame) | |||
Petri Dish (100mm x 15mm) and (60mm x 15mm) | Fisher Brand | FB0875712 and FB0875713A | |
Scalpel blades #11 | Fine Science Tools | 10011-00 | |
Scalpel handle #3 | Fine Science Tools | 10003-12 | |
Serological Pipettes | Sorfa Medical Plastic Co. | P8050 | |
Standard Pattern forceps | Fine Science Tools | 11000-12 | |
Sterile vacuum filter | Thermo-Scientific | 565-0020 | |
Surgical Scissors | Fine Science Tools | 14054-13 | |
Syringe driven filter unit | Millipore-Millex | SLGP033RS | |
Tissue chopper with moveable stage | Stoelting | 51425 | |
Fine tip paintbrush |