McIlwain Tissue Chopper (Campden Instruments, Model TC752)

Standard Surgical Scissors (Fine Science Tools; item no. 14002-13)

Sterile Millicell 6-well culture plate inserts (Millipore, cat. no. PICM0RG50)

Opti-MEM™ (1×) [+Glutamax, +HEPES, +2.4 g/L Sodium Bicarbonate] (Gibco-Invitrogen cat. no. 51985-034)

Leibovitz’s L-15 medium is used as dissection and tissue chopping medium. Store in 50 mL aliquots at +4 °C.

Autoclave and sterilise two 100 mL Duran™ glass bottles and allow to cool down. Spray the outer surface of the bottles with 70% ethanol and transfer to the laminar flow safety cabinet. Prepare the slice culture media under aseptic conditions. Referring to Table 1, add the components listed for Media #1 and Media #2 to separate 100 mL bottles and mix gently. The media can then be poured into 2 × 50 mL Greiner™ centrifuge tubes (sterile), sealed, wrapped in parafilm and transferred to the fridge and stored at 4 °C for one to two weeks. Once re-opened, use within 1 week. It is advisable to prepare small volume batches of media rather than larger volumes that are opened more often.

On Culture Day 0, add 1.1 mL of Media #1 to each well of the 6-well plate. Media can be cold when positioning freshly chopped slices onto Millicell inserts. Transfer the 6-well plates to a warm humidified CO₂ incubator.

Timing: 20 min

Timing: 30 min

CRITICAL STEP: Ensure the tips of the scissors and forceps do not touch the brain to prevent any damage to the brain tissue.

12. Using a curved pointed forceps, peel apart the skull to expose the brain.

CRITICAL STEP: The brain tissue should not be allowed to dry out; hence steps 19 and 20 should be performed quickly.

21. Using a sterile Pasteur pipette, transfer the 400 µm hippocampal slices into a 60 mm Petri dish filled with ice-cold L-15 medium.

CRITICAL STEP: Make sure the brain tissue is fully intact with no damage during separation process.

Timing: 5 min

Timing: 15 min

CRITICAL STEP: Ensure that none of the hippocampal slices are in contact with each other and avoid placing slices at the edge of the insert to maximise culture growth and slice health.

CRITICAL STEP: Make sure that the insert membrane is fully wet and free of air bubbles underneath. Record the date and culture information on the 6-well plate lid using a black marker.

31. Place a culture insert, with 3 hippocampal slices, in each well of the 6-well plate (Figure 3). Repeat for the other 5 inserts.

Timing: 15 min

CRITICAL STEP: Media changes should be completed under sterile conditions in a laminar flow cabinet.

CRITICAL STEP: Antibiotics are not used in the culture media to minimize cytotoxicity and to improve tissue viability. Therefore, a high standard of aseptic technique and sterility is critical.

34. Based on the number of wells, calculate the volume of culture medium required (i.e. 1.1 mL per well plus 1 mL extra) and pipette into a Falcon tube.

NOTE: Warming the medium to 35.5 °C in the incubator should be performed 1–2 h ahead of time.

37. Transfer the 6-well plate with slices, and the new plate with fresh medium, from the incubator into the laminar flow hood.

CRITICAL STEP: Ensure no air bubbles are trapped under the insert.

41 Slices can be maintained and cultured for several weeks after plating. Four to 6 days after plating, the slices will be firmly attached to the insert. There is no need to coat the membrane with poly-D-lysine, poly-L-ornithine, or extracellular matrix.

ThermoFisher Scientific catalogue number C10638

Timing: 30 min

42. Allow the vials to warm to room temperature before opening.

Timing: 1 day

Timing: 1 week

CRITICAL STEP: Perform fixation under a fume hood and wear appropriate personal protective equipment. Formalin solution is carcinogenic.

57. Prepare the fixing solution (i.e. 10% formalin in PBS) and cool on ice 15 min before fixation.

NOTE: To enhance visualization of the slices, perform these fixation steps on a dark background.

63. Aspirate and remove the PBS.

NOTE: Examine the hippocampal slices under the microscope to assess their structural integrity. Immunofluorescence should only be performed on slices that display a well-preserved cytoarchitecture.

CRITICAL STEP: Exercise caution throughout all pipetting steps as the hippocampal slices can detach from the insert. To avoid this, gently apply each solution drop by drop at the edge of the insert away from the slices.

68. Once the permeabilization and blocking step is complete, use a sterile scalpel to cut the membranes from the Millicell inserts to obtain a smaller circular membrane containing all 3 hippocampal slices.

CRITICAL STEP: Ensure that the slices remain ‘face-up’ at all times. Use a pointed forceps to carefully transfer membranes and avoid folding of the membrane.

71. Add the appropriate concentration of primary antibodies (Table 5) in 400 µL blocking solution (i.e., PBS + 2% BSA + 0.1% Triton X) per insert.

To validate the structural integrity and evaluate the cellular composition and neurogenic potential of our organotypic hippocampal slice cultures, we performed immunofluorescence after 14 and 19 days in vitro using cellular markers for proliferation, neural progenitors, and differentiated neuronal populations. Neural stem/progenitor cell populations, within the dentate gyrus (DG), were examined using antibodies targeting Ki67 and SOX2 (Figure 5). Ki67+ proliferating cells (green) were detected throughout the granule cell layers, the subgranular zone, and the Hilus indicating in vitro cell division. SOX2-expressing neural precursor cells (magenta) were prominently localized along the SGZ, confirming preservation of the neurogenic stem cell niche in ex vivo slices. Higher magnification images demonstrated that Ki67+ proliferative cells were more numerous within the Hilus and SGZ than in the outer granule cell layers of the DG, consistent with in vivo spatial patterns of neurogenesis.

We next evaluated the levels of extracellular matrix and the density of perineuronal nets (PNNs) in the DG using WFA (Wisteria floribunda agglutinin) fluorescein (FITC). Slices were also co-labelled with the NSC and neural progenitor cell (NPC) marker Nestin (Figure 6). PNNs regulate synaptic plasticity and neuronal excitability which impact the rate of neurogenesis [20]. WFA staining (green) revealed preserved perineuronal nets within the Hilus and dentate granule cell layers, demonstrating maintenance of key ECM components in the hippocampal slices. Nestin+ progenitor cells were abundant and retained a well-organized spatial pattern of expression within the horseshoe shape of the DG, supporting the suitability of this slice culture system for studying neural precursor cell morphology and maturation processes.

To confirm the maintenance of functional neuronal circuits, particularly the presence of mossy fibre axonal tracts connecting the DG to CA3 pyramidal cells, slices were labelled for synaptophysin, a presynaptic vesicle protein, and the GluA1 subunit of the AMPA receptor, a marker of postsynaptic glutamate receptors (Figure 7). Synaptophysin expression was observed throughout the DG granule cell layer, with higher expression in mossy fibre axons projecting through the Hilus to CA3 region, indicating preserved presynaptic terminals. AMPA receptors (AMPAR) were abundantly expressed in the granule cell layer of the dentate gyrus. Merged images suggest the preservation of functional excitatory synaptic connections within cultured hippocampal slices as synaptophysin-positive axons and dendrites (green) formed dense networks throughout the slices and co-localized with AMPAR-expressing cells (magenta). These findings suggest that our organotypic slice culture system retains key cytoarchitectural features of hippocampal circuitry, facilitating the study of synaptic plasticity mechanisms alongside neurogenesis. Collectively, this immunofluorescence data demonstrate that our organotypic hippocampal slice cultures exhibit preserved in vivo-like cytoarchitecture with a defined SGZ niche containing a large pool of proliferative NPCs, mature extracellular matrix PNN structures in the DG, and mossy fibre-mediated connectivity between DG and CA3 regions. This characterization validates the slices as a reliable in vitro model for studying hippocampal neurogenesis and synaptic development.

To assess ongoing stem cell proliferation within the organotypic hippocampal slices, we performed an EdU incorporation assay coupled with NeuN and S100β immunofluorescence (Figure 8). At low magnification (×20), EdU+ nuclei (green) were distributed throughout the slice (Figures 8A–F) and represented the proportion of cells that remained mitotically active under our culture conditions. DAPI counterstaining revealed the overall cytoarchitecture of the DG, while NeuN (magenta) and S100β (red) immunolabelling distinguished mature neurons from astrocytes and oligodendrocytes, respectively.

We quantified the mean density of EdU+ cells in each dentate gyrus. Three ×20 magnification confocal images of the granule cell layers were captured and the total number of EdU+ cells per image were counted. We calculated an average of 24 (±9) EdU+ cells per 0.05 mm² in the DG region (Figure 8P). This result illustrates that the slice culture system maintains a consistent and reliable proliferative capacity, further validating its use as a platform for studying postnatal neurogenesis in an ex vivo hippocampal environment.

Next, we captured ×63 magnification z-stack images of the DG (Figures 8G–L). As expected, NeuN-positive neuronal cell bodies were densely expressed in the dentate granule cell layer, whereas S100β-expressing glia were more dispersed throughout the subregions of the DG. Merged images demonstrated that EdU incorporation occurred in both neuronal and non-neuronal regions. Interestingly, orthogonal z-projections of the hippocampal slice illustrate that a layer of glial cells (red S100β+ cells) sits above the dentate granule neuronal layer (magenta NeuN+ cells) (Figure 8N). Moreover, EdU+ cells are present in both neuronal (Figure 8M) and glial cell layers (Figure 8O) as indicated by the white arrows. To quantify the cell types expressing EdU, we calculated the proportion of EdU+ cells expressing either NeuN, S100β or neither (−/−) markers (Figure 8Q).

S100β+ and EdU+ cells represented approximately 35–40% of the proliferative cell population. NeuN+ and EdU+ cells represented approximately 15% of the proliferative population. Finally, EdU+ cells that were double-negative (−/−) for S100β and NeuN+ made up approximately 45–50% of the proliferative cell population. These results illustrate the presence of a neural stem cell pool in the DG as well as NPCs that have actively differentiated into either astrocytes and oligodendrocytes (40%) or into mature post-mitotic neurons (15%).

Our protocol diverges from the traditional Stoppini formulation by implementing a phased transition from a 10% serum-based medium plus growth factors (NGF, EGF, FGF) to a 5% serum-containing medium with growth factors and defined supplements (B27 and N21). This acknowledges the shifting metabolic and trophic requirements of the hippocampal slice during the first few days of the recovery period to the maintenance phase for long-term cultures. Early studies utilized high concentrations of horse serum (up to 25%) to facilitate tissue flattening and initial survival [11, 12]. However, unrefined serum contains an unpredictable array of factors that varies lot-to-lot and by source and can enhance reactive gliosis, thus inhibiting the sensitive developmental mechanisms that drive NPC proliferation, migration and differentiation into functional neurons. To mitigate this reactive glial phenotype, more recent protocols have successfully pivoted toward chemically defined serum-free media. For instance, Mayerl and ffrench-Constant demonstrated that supplementing Neurobasal/B-27 media with the anti-inflammatory compound, indomethacin, directly dampens astrogliosis and actively preserves the ex vivo neurogenic potential of the hippocampal niche [21].

In our approach, the initial recovery phase utilizes a balanced combination of 6% horse serum (HS) and 4% foetal bovine serum (FBS). This pairing exploits the complementary properties of these additives, i.e., the mitogenic richness of FBS, laden with platelet-derived growth factor (PDGF) and fibroblast growth factors (FGF) which support the immediate proliferative needs of the subgranular zone (SGZ), while the milder profile of HS facilitates tissue stabilization without triggering excessive excitotoxicity. We have previously implemented a media switch on day 3 from 25% serum-containing salt solution to Neurobasal-A with B27-supplemented media and 0% serum for growing cerebellar slice cultures [22, 23]. However, we found that hippocampal slices were more sensitive to the abrupt change from high serum to a zero serum defined medium. As such, dropping the initial serum content to 10% followed by 7.5% on days 3–5 and finally to 5% from day 5 onwards, ensured the slices experienced a milder phased transition to a low serum environment.

The inclusion of MITO+ serum extender to both media #1 and #2 provides a concentrated suite of hormones and growth factors, including EGF, FGF, insulin, transferrin, and selenium (ITS), alongside progesterone and putrescine. This cocktail aims to maintain the necessary mitogenic stimuli for neural stem cells whilst facilitating a significant drop in serum concentration. This is particularly relevant given the work of Raineteau et al. [24] who showed that chronic EGF exposure can increase neurogenesis in organotypic hippocampal cultures but that horse serum (25%) can abolish this effect. Moreover, Namba et al. [25], demonstrated that while spontaneous neurogenesis occurs in organotypic slices, the neurogenic capacity of the dentate gyrus progressively declines in vitro, likely due to loss of niche signals and reactive changes, highlighting the need for defined culture conditions to sustain progenitor proliferation. By providing a defined mitogenic stimulus via MITO+ and nerve growth factor (NGF), we extend the window of active neurogenesis well into the third week in vitro. These findings mirror previous efforts to control neurogenesis in the subgranular zone niche using defined chemical additives. Previous studies have shown that the endogenous progenitor pool within postnatal slice cultures remains highly responsive to external chemical cues. For instance, Poulsen et al. [26] demonstrated that targeted media supplementation with growth factors (EGF and IGF-I) as well as small-molecule glutamate receptor antagonists such as NMDA blockers (MK-801 and APV) or the AMPA antagonist, CNQX, significantly increases the differentiation of pre-existing neural precursor cells into TUC-4+ dentate granule cells. Interestingly, these small-molecule inhibitors reduced baseline glutamatergic tone which synergises with biochemical mitogens to guide precursor cell fate without inducing cytotoxicity. By utilizing an optimized basal medium containing MITO+ serum extender and NGF, our protocol similarly harnesses this inherent plasticity and directs neuronal maturation, establishing a controlled baseline that can be easily combined with small-molecule modulators of neurogenesis.

Organotypic hippocampal slices are often grown in culture at sub-physiological temperatures between 35 and 36 °C [9, 12, 17]. Studies have shown that mild hypothermia is neuroprotective [27, 28] and thus we maintained slices at a constant temperature of 35.5 °C for the duration of the culture. The transition at day 3 to a Neurobasal-A backbone, supplemented with B27 and N21, represents a move toward a more physiological maintenance phase. Neurobasal-A, optimized for the survival of post-mitotic neurons, lacks the high glutamate levels found in standard basal media, thereby reducing chronic excitotoxicity [19]. The B27 supplement is essential for its antioxidant properties; it contains catalase, superoxide dismutase, and vitamin E, which mitigate the oxidative stress inherent in high-oxygen interface cultures. Furthermore, the addition of N21 (also known as NS21) provides a more defined and consistent source of essential fatty acids and vitamins compared to original B27 lots. Research by De Simoni et al. [29] underscored the importance of such defined media for the development of CA1 pyramidal neurons and the formation of mature dendritic spines. Our observation of high synaptophysin and AMPA receptor expression suggests that this supplemented environment supports not only the birth of new neurons but their successful integration into functional, mature synaptic networks.

Our immunofluorescent results quantitatively validate the fidelity of this slice culture system. The presence of dense Ki67 and SOX2 clusters within the SGZ, alongside high levels of Nestin, confirms that the primary neurogenic reservoir remains localized and active. This spatial restriction is critical; as Namba et al. [25] noted, the loss of neurogenic potential in many slice models is often preceded by the dispersal of these progenitor populations.

Our EdU lineage tracing data reveals the following differentiation profile:

15% NeuN+ (neurons): Representing the terminal differentiation of progenitors into a neuronal lineage, a process fundamentally guided by the local hippocampal environment.

The preservation of the extracellular matrix (ECM), evidenced by robust Wisteria floribunda agglutinin (WFA) staining for perineuronal nets (PNNs), is perhaps the most significant advantage of this 3D system. PNNs are critical for stabilizing the synaptic architecture and defining the boundaries of the neurogenic microenvironment. In the absence of these structures, as is often seen in 2D monolayer cultures or emergent organoid models, the mechanical and biochemical cues required for directed neurogenesis are lost.

Although organotypic slice cultures retain the principal cytoarchitecture of the hippocampal formation, tissue preparation necessarily involves axotomy of major afferent and efferent projections. Removal of the entorhinal cortex abolishes perforant pathway input to the dentate gyrus, while tissue sectioning transects commissural and associational fibres that normally provide recurrent and interhemispheric connectivity. As a result, the trisynaptic hippocampal circuit is effectively deafferented and functionally isolated from key extrinsic inputs. This loss of connectivity triggers reactive synaptogenesis and substantial network reorganization, including mossy fibre sprouting into the inner molecular layer of the dentate gyrus and remodelling of recurrent CA3 collateral circuitry [29–32]. These anatomical adaptations increase recurrent excitation and reshape network dynamics, progressively promoting hyperexcitability and lowering the threshold for epileptiform activity [33].

To mitigate these circuit-level artefacts and better preserve physiological patterns of connectivity, several complementary approaches can be adopted. Retention of entorhinal cortex input (Figure 1B) through organotypic entorhino-hippocampal co-cultures preserves perforant pathway connectivity and reduces the extent of deafferentation imposed by conventional hippocampal slice preparation [34, 35]. By maintaining this major afferent projection, these slice preparations provide a closer approximation of native hippocampal network architecture. However, we have found that hippocampal slices, particularly the dentate gyrus, appear to survive better without the neocortex and entorhinal cortex attached. Pharmacological strategies may also be used to limit excessive network activity; however, glutamatergic signalling is itself a key regulator of injury-induced circuit plasticity. Consequently, glutamate receptor antagonism modifies rather than prevents reactive remodelling. For example, blockade of NMDA receptors promotes axonal sprouting, whereas signalling through AMPA/kainate receptors is required for reactive sprouting and the recovery of functional connectivity following deafferentation [36, 37]. Any intervention targeting glutamatergic transmission may therefore alter the trajectory of circuit reorganization rather than restore baseline excitability of the native hippocampal network.

A further limitation inherent to organotypic slice cultures is the mechanical trauma of tissue sectioning, which initiates inflammatory and glial responses characteristic of long-term culture adaptation. The magnitude of this tissue injury is influenced by the slicing method. The McIlwain tissue chopper enables rapid and highly reproducible preparation of large numbers of slices and therefore remains widely used for organotypic culture protocols. In contrast, vibratome sectioning reduces tissue compression and shear forces during cutting, resulting in improved preservation of tissue architecture at the slice surface and enhanced morphological integrity of the preparation [18, 38]. However, these advantages must be balanced against longer preparation times. Vibratome sectioning is considerably slower and extends the interval between tissue harvest and culture, increasing exposure to ex vivo metabolic stress. In contrast, the speed of tissue chopping minimises this period and may facilitate more rapid recovery of cellular metabolism following culture establishment. Regardless of the preparation method used, tissue sectioning induces a robust reactive glial response. Consistent with this, our immunofluorescence analysis revealed pronounced GFAP-positive astrogliosis in cultured hippocampal slices (Figure 9). Importantly, reactive gliosis and neuroinflammatory signalling are persistent features of long-term organotypic cultures, reflecting the combined effects of axotomy, deafferentation, and adaptation to the in vitro environment [39, 40]. Consequently, selection of a slicing strategy involves a practical trade-off between minimising acute mechanical damage and minimising preparation-associated metabolic stress.

The organotypic hippocampal slice culture protocol presented here is also compatible with emerging three-dimensional imaging and tissue-processing approaches. Organotypic slice cultures can be combined with tissue-clearing and refractive index matching methods, such as ScaleS, as well as expansion microscopy (ExM) and light-sheet fluorescence microscopy (LSFM), to improve imaging depth and visualization in thick tissue preparations [41–43]. For neurogenesis studies, LSFM may facilitate long-term, low-phototoxicity imaging of progenitor cell behaviour, enabling tracking of cell division, migration and neuronal integration over time. In parallel, ExM enables nanoscale-resolution visualization of protein distributions and subcellular structures within intact tissue, including synaptic receptor localization and microcircuit organization [43, 44]. Together, these complementary approaches could support high-resolution volumetric analyses of lineage progression, neuronal maturation and circuit integration in hippocampal slice cultures ex vivo.

While human iPSC-derived organoids have become a prominent feature of modern neuroscience, they frequently lack the structural maturity and laminar precision found in our rodent organotypic slices. Organoids often suffer from stochastic cell placement and the absence of a defined vascular-like scaffold or a mature, organized ECM. Specifically, the mossy fibre pathway and the precise orientation of the radial glial scaffold, which are both pre-assembled in our slice model, are essential for the authentic integration of new neurons into the tri-synaptic circuit. As we move toward a more comprehensive understanding of neurodevelopmental disorders, the preservation of this functional anatomy becomes paramount. Our refined protocol, by honouring the structural template of the hippocampus while modernizing the biochemical support system, offers a high-fidelity platform for long-term neurogenesis assays. It bridges the gap between simple monolayer cell cultures and the complexity and inaccessibility of the hippocampus during in vivo studies.

Arguably, the most appropriate ex vivo system to investigate dentate granule cell physiology is human organotypic hippocampal slice cultures which have been generated by several research groups [45, 46]. These valuable research tools bridge the gap between iPSC-derived brain organoids and in vivo functional imaging techniques, but the excised samples of human brain tissue are extremely rare, lack consistency and are often biopsied from patients with brain tumours or epilepsy, for example, [47]. For most basic neuroscientists without neurosurgical collaborators; if the precise anatomical structure of the hippocampal circuitry and SGZ niche cytoarchitecture remains central to an investigation, then rodent organotypic brain slices are valuable models for interrogating the synaptic plasticity and functional maturation of hippocampal neural networks.