Herein, we present a protocol for the 3D culture of rat brain-derived glia cells, including astrocytes, microglia, and oligodendrocytes. We demonstrate primary cell culture, methacrylated hyaluronic acid (HAMA) hydrogel synthesis, HAMAphoto-polymerization and cell encapsulation, and sample processing for confocal and scanning electron microscopic imaging.
In the central nervous system, numerous acute injuries and neurodegenerative disorders, as well as implanted devices or biomaterials engineered to enhance function result in the same outcome: excess inflammation leads to gliosis, cytotoxicity, and/or formation of a glial scar that collectively exacerbate injury or prevent healthy recovery. With the intent of creating a system to model glial scar formation and study inflammatory processes, we have generated a 3D cell scaffold capable of housing primary cultured glial cells: microglia that regulate the foreign body response and initiate the inflammatory event, astrocytes that respond to form a fibrous scar, and oligodendrocytes that are typically vulnerable to inflammatory injury. The present work provides a detailed step-by-step method for the fabrication, culture, and microscopic characterization of a hyaluronic acid-based 3D hydrogel scaffold with encapsulated rat brain-derived glial cells. Further, protocols for characterization of cell encapsulation and the hydrogel scaffold by confocal immunofluorescence and scanning electron microscopy are demonstrated, as well as the capacity to modify the scaffold with bioactive substrates, with incorporation of a commercial basal lamina mixture to improved cell integration.
Inflammation of the central nervous system (CNS) has long been considered a hallmark of acute (e.g., ischemic stroke, traumatic brain and spinal cord injury) and chronic (e.g. Alzheimer's, Parkinson's, and Huntington's diseases) CNS injury, but is increasingly recognized as a causal contributor to neurodegenerative and neuropsychiatric disorders. Sustained or inappropriate inflammation can cause neural injury and demyelination (e.g. multiple sclerosis), and negatively influence brain development (e.g., schizophrenia, autism) and mood states (e.g., depression, anxiety, bipolar disorder). Further, novel therapeutic strategies using implantable devices (e.g.,brain-computer-interfaces1,2,3, deep brain stimulation4,5, intraspinal microstimulation6,7,8,9,10) generate a predictable inflammatory response at the interface between the device and the CNS resulting in a protective tissue response that can cause loss of efficacy or device failure over the lifetime of the implant11. Inflammation in the CNS is typically initiated by microglia, which function as the resident immune cells of the CNS responsible for tissue surveillance and mounting the foreign body response (reviewed12). Depending on the severity of an insult, the microglia signal and recruit additional cell types to an injury site. Specifically, the microglia activate the astrocytes, which in turn act as secondary inflammatory cells and form a dense protective barrier to contain an injury site13,14. Microglia can also initiate an activity cascade in the cells of the peripheral immune system, which can result in the breakdown of the BBB to permit immune infiltration (reviewed in reference15).
In the case of devices implanted into the CNS, tissue damage resulting from device insertion as well as the continued presence of the foreign device may initiate a process termed glial scarring. In this process, the microglia migrate to and proliferate at the site of injury. They also initiate the release of inflammatory factors to neutralize potential threats and recruit additional glial cells. Subsequently, activated astrocytes become hypertrophic and begin encapsulating the implanted device to form a continuous fibrous barrier16. Inflammatory signalling also serves to promote withdrawal of neuronal processes from the vicinity of the implant and eventually recruits fibroblasts to reinforce the developing glial scar17. The oligodendrocytes, responsible for sheathing neurons in myelin to enhance conductance, do not survive this process and distant cells are partitioned from the implant by the scar18. Glial scarring greatly reduces the function and lifetime of implanted devices, particularly for recording electrodes, and ultimately serves to limit the functionality of neural interfaces19.
Several approaches have been exploited to increase the biocompatibility and interface activity of implanted devices in the CNS20,21,22,23. An extensive review is available on the biocompatible design of these neural interfaces24.The most prominent strategies include surrounding the electrode with compatible coatings such as polyelthyleneglycol (PEG), polylactic-co-glycolic acids (PLGA)25, or enhancing the electrode with conductive polymers such as poly(ethylene dioxythiophene) (PEDOT), and polypyrrole (PPy)26,27,28,29,30,31. Bioactive coatings have also been employed to provide cues for neural tissue growth using ligands derived from extracellular matrices including collagens, fibronectins, and hyaluronic acids32,33,34,35,36,37. The bioactivity of these coatings has been further explored with growth factor release systems to emulate natural cell secretions30,38,39,40,41,42,43,44,45,46,47,48,49,50. Simultaneously, some research groups have opted to remodel the electrode geometry, flexibility, and composition to decrease the mechanical mismatch between device and tissue51,52,53,54,55,56,57. Altogether, these strategies have lead to many promising improvements in next-generation neural interfacial devices, however the long-term compatibility is an on-going issue and progress may be hampered by complex and time-consuming in vivo models.
Animal model-based approaches can limit the experimental throughput and increase costs of testing electrode biocompatibility. In vitro approaches using conventional cell culture techniques offer a more cost-effective alternative but fail to recapitulate much of the complexity of the interaction between device and tissue58. In particular, testing of surface coatings using 2D cell culture limits the modeling of electrode geometry and the influence of mechanical mismatch and micromotion thought to contribute to generating a host response contributing to device failure59,60.
To overcome problems associated with 2D cell culture, hydrogel cultures of neural cells have been developed for a wide variety of applications, pharmacological studies61, to direct neural cell differentiation62, to understand disease pathways63,64, or layered in co-culture with other cell types to model cell migration, neuroprotection, or to model tissue microenvironments61. Hydrogels are readily formed at different sizes and geometries can incorporate numerous types of primary or immortalized cell cultures, and are highly amenable to analysis by commonly used techniques such as confocal fluorescence microscopy. To create a model of the glial scarring process, we have recently developed and characterised a hyaluronic-acid based 3D hydrogel system for high-throughput testing of the glial response to implanted electrodes (Figure 1)65. This system has several distinct advantages: 1) primary glial cells (microglia, astrocytes, and oligodendrocytes) are encapsulated in a 3D matrix composed of polymers of hyaluronic acid, which is an endogenous extracellular matrix component; 2) the matrix stiffness can be 'tuned' to recreate the mechanical properties of brain or spinal cord tissue; and 3) cells can be encapsulated in the matrix in a rapid bench-top approach using photopolymerization with green light, limiting toxicity during encapsulation. This system enables key features of in vivo biocompatibility: devices are inserted into the hydrogel in a comparable manner to tissue, and the cellular response to implanted devices are monitored for a wide range of parameters65. These include mechanical mismatch between devices and the hydrogel coatings of various structures and electrical stimulation pulses. This system also includes oligodendrocyte and related precursors, which are often present and recruited in glial scars. Their damage, death, and phagocytosis by microglia are highly indicative of inflammatory injury and as a model reduced scarring or recovery, they have the capacity to demonstrate re-myelination of neurons66.
Herein we describe a method for synthesis and formation of hybrid hyaluronic acid hydrogels combined with commercially available basement membrane formulations to improve cell incorporation. Further, we will demonstrate the incorporation of primary cultured glial cells (microglia, astrocytes, and oligodendrocytes) and analysis of culture growth using immunocytochemistry and confocal microscopy.
The protocol for the brain tissue extraction from day 1 Sprague Dawley rat pups, euthanized by decapitation, was approved by the Animal Care and Use Committee at the University of Alberta.
1. Microglia and Astrocyte Isolations67,68
NOTE: All media for isolation and cell culture is preheated to 37 °C in a water bath. Hank's balanced salt solution (HBSS) has 1% penicillin-streptomycin (PS). All Dulbecco's modified Eagle's media with Ham's F12 nutrient mixture (DMEM/F12) is supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin (PS). Only mixtures with trypsin lack FBS. All materials in this protocol are sterile (filter, alcohol, purchased, and autoclaved) and every step subsequent to step number 1.8 is performed in a biosafety cabinet with aseptic technique.
2. Macromer Synthesis69
3. Gel Formation and 3D Encapsulation65
4. Microscopy
To model the neural tissue host response and the glial scar in a high throughput, in vitro system requires a 3D cell scaffold with a matrix material that is biocompatible, does not incur a cytotoxic event during in situ formation, and is modifiable with bioactive components to guide benevolent response. To this end, we have created a 3D cell scaffold system based on hyaluronic acid, and encapsulated a primary mixed glial cell population to study cell-cell interactions and glial bioreactivity. A brief visual study of cells and scaffold was performed. Morphology of cell subpopulations, including oligodendrocytes (CNPase in green), microglia (Iba1 in red), and astrocytes (GFAP in yellow), are shown by confocal immunofluorescent microscopy (Figure 3Ai-ii -Ci-ii). Combined scaffold and cell morphology are also shown by SEM (Figure 3Aiii-iv – Ciii-iv). In these images, both low (i and iii), and high (ii and iv) resolution representative images are shown. A 2D PLL coverslip coating (Figure 3A) was compared to a 3D 0.5% w/v HAMA scaffold (Figure 3B), and a mixture of 20% v/v basal lamina mixture in 0.5% w/v HAMA (Figure 3C). This was done to demonstrate the morphological differences in 2D and 3D cell culture, as well as the effect of introducing a biological lamina to the scaffold. A 3D reconstruction is also available in the Supplemental Information.
Cell morphology of each cell type changed with the introduction of a 3D platform. In 2D adherent culture (PLL-coated glass coverslips), oligodendrocytes appeared typically round and formed small networks usually connecting with lightly labeled GFAP processes. Microglia were smaller than other cells and had small branching processes that did not extend far from their cell bodies (<20 µm). Astrocytes were more diversely shaped, some having a radial morphology with smaller cell bodies and extensive thin, branching processes, while others were fibrous with a flattened appearance and few broad processes that coat the surface. In the 3D HAMA culture, oligodendrocytes appeared in clusters having branched off into smaller spheres. Microglia were round with few processes, and astrocytes were either rounded or branched off radially into networks from a single cluster. Generally, these cells appeared to have outgrown from single points or remain in areas of the HAMA without making contact with other cells, which is unlike the 2D culture and is suggestive of limited mobility through the matrix. When 20% v/v of a basal lamina mixture was added to the photo-polymerization mixture, all glial subtypes integrated with the 3D platform, and formed more extensive, interactive branched structures. Oligodendrocytes extended bulbous processes radially similar to the non-basal lamina mixture morphologies, but additionally appeared to extend processes along the thin interconnected networks of astrocyte processes branching throughout the scaffold. Microglia were dispersed among these cell networks, having spread out with thin processes more consistent with their morphology in tissue. Overall, the morphologies suggest all three types of glia integrated into the HAMA-basal lamina mixture freely, potentially through either increased adherence to the mixed substrate or through increased ability to remodel the scaffold.
When observed by SEM, cell morphology was seen to corroborate the confocal micrographs. The HAMA matrix appeared as a smooth surface with 10-50 µm pores seen faintly beneath the surface. Cells were adherent to the surface of the HAMA and were also faintly visible in layers beneath. These glia did not form comparable morphologies to those noted in 2D culture, however with the addition of a basal lamina mixture, extensive networking was observed. Thick astrocyte fibers, with smaller clusters of microglia were observed both on the HAMA-basal lamina mixture surface and extending through the matrix in a seamless manner. This can be seen faintly as the matrix folded around the cells, suggesting that the cells may be altering the scaffold to formed into their desired conformations.
Figure 1: Schematic of methacrylated hyaluronic acid (HAMA)-based 3D cell culture and photopolymerization protocol. Glial cells (2 week primary rat brain culture) are mixed with HAMA and pipetted into a polydimethylsiloxane (PDMS) mold on a glass coverslip. This mixture is exposed to a high intensity green LED light to polymerize the hydrogel for 5 min. The mold is removed and the gel is incubated in media. Representative cells include microglia (red), astrocytes (yellow), and oligodendrocytes (green). Please click here to view a larger version of this figure.
Figure 2: Dissection of day 1 Sprague Dawley rat pup brain. Images of whole brain (A), dissection of cortices and cerebellum (B), a single cortex with meninges (C), peeling of the meninges with forceps (D), and the meninge-free cortex (E). Please click here to view a larger version of this figure.
Figure 3: Immunofluorescent confocal (i-ii) and scanning electron (iii-iv) micrographs of mixed glia cell populations on 2D PLL coatings (A), HAMA (B), and HAMA-basal lamina mixture (C). Lower (i, iii) and higher (ii, iv) magnification images are shown. Labels include Iba1 in red for microglia, GFAP in yellow for astrocytes, CNPase in green for oligodendrocytes, and a Hoechst nuclear stain. Scale bars = 25 µm for confocal images and 50 (iii) and 20 µm (iv) for scanning electron micrographs. Hydrogels were 0.5% w/v, and basal lamina mixture was 20% by volume. Please click here to view a larger version of this figure.
Supplemental Figure 1: 3D immunofluorescent micrographic reconstruction of HAMA-basal lamina mixture with mixed glia cell populations HAMA-basal lamina mixture. Labels include Iba1 in red for microglia, GFAP in yellow for astrocytes, CNPase in green for oligodendrocytes, and a Hoechst nuclear stain. Hydrogels were 0.5% w/v, and Getrex was 20% by volume. Please click here to download this file.
HAMA | Basal Lamina | NVP | TEA | EY | |
Working concentration | 0.5% w/v | 20% v/v | 0.10% | 0.10% | .01 mM |
12-well Plate | 2% w/v | 100% w/v | 10% v/v | 10% v/v | 1 mM |
1.4 mL Total | 350 µL | 280 µL | 14 µL | 14 µL | 14 µL |
Table 1: Example of Cell Encapsulation.
Component | Content | Incubation Time |
EtOH | 30% | 30 min |
EtOH | 50% | 30 min |
EtOH | 70% | 30 min |
EtOH | 90% | 20 min |
EtOH | 100% | 20 min |
EtOH | 100% | 20 min |
EtOH:HMDS | 75:25 | 20 min |
EtOH:HMDS | 50:50:00 | 20 min |
EtOH:HMDS | 25:75 | 20 min |
HMDS | 100% | 20 min |
HMDS | 100% | Overnight |
Table 2: Gradual Increments of EtOH and HMDS in Dehydration.
Towards the goal of generating a 3D culture system to model glial bioreactivity and the glial scarring process, we have developed a system that can support primary cultured microglia, astrocytes and oligodendrocytes and enables robust characterization of cell morphology and cell-cell interactions. From the micrographs shown, the morphology of each cell type was distinctly different with 2D, 3D-HAMA, and 3D HAMA-Basal lamina mixture platforms. In the 2D system, morphology was distinctly biased along the plane of the surface, but when compared to the 3D HAMA, microglia and astrocytes were generally smaller inside the matrix, with the exception of oligodendrocyte clusters. In the scanning electron micrographs, cells were generally more round shaped on the HAMA. This may be largely due to the cells being encapsulated, with limited capacity to modify the matrix for locomotion and free movement and growth of processes. Cell outgrowth in the HAMA was typically radial for both astrocytes and oligodendrocytes. While these cells extended processes, their organization suggest that their movement throughout the matrix and ability to interact with other cells was limited. In addition, previous work has shown freeze-fractured HAMA to be porous without interconnected networking, which may explain this lack of outgrowth65,70.It may be possible that some of these cells did not survive the photopolymerization, remaining preserved in the matrix, however viability was systematically assessed in our previous work, and remained 80% (0.5% w/v) over the course of a week65.
To improve the cell interaction with the matrix in 3D culture, we introduced a basal lamina mixture as a biocompatible and modifiable matrix component. Cell integration with the matrix was found to be markedly improved, with all cell types showing extensive processes compared to both the 2D culture and 3D HAMA. The basal lamina mixture is comprised of a variety basal lamina components and is primarily formulated to support glial differentiation of neural stem cell precursors71. It was not unexpected that the glial cells responded to the basal lamina mixture, and subsequently the surrounding 3D matrix, but the extent of integration suggests a healthy tissue-like environment. Comparing HAMA to HAMA-basal lamina mixture, the hydrogel morphology does not appear distinctly different, therefore cells may be more actively remodelling it. Considering the simplicity of adding this bioactive component, it is not unimaginable to use different laminas and/or protein cues to favour culture conditions for neurons. In a potential future direction, this system could be used to differentiate neurons from neural progenitors in conjunction with mixed glia for a variety of impactful myelination or neuro-inflammatory injury models.
In the event of serious and damaging CNS injury or biomaterial rejection, the resident tissue initiates inflammatory reactions capable of exacerbating neurodegeneration and demyelination. This typically results in the formation of a glial scar to partition the site of injury, which prevents regeneration. In this study, the methods were detailed for a methacrylated photo-polymerized hyaluronic acid-based 3D scaffold with the intention of modelling the initial cellular response behind glial scar formation; microglial reactivity, astrocyte recruitment and hypertrophy, and oligodendrocyte injury and withdrawal. A 3D cell scaffold was designed to incorporate all three glial cell types, and was further modified with a multi-component basal lamina mixture. It was determined, by confocal and scanning electron microscopy imaging, that that cells were encapsulated in the HAMA and were better able to integrate when the basal lamina mixture was introduced. These results could lead to several novel applications. HAMA alone may serve to create co-culture systems where the intention is to limit cell to cell interaction or study distil drug delivery using the gel as a drug loaded diffusion limited matrix. Hydrogels incorporating the basal lamina mixture into the HAMA matrix allow for a more consistent tissue-like environment capable of modeling acute inflammation, gliosis, or glial scarring, and enables further high-throughput characterization of glial bioreactivity to implants, drug and device development, and other strategies to reduce and recover inflammatory injury.
The authors have nothing to disclose.
The authors are grateful for financial support from NSERC, CFI, AIHS, Alberta Health Services, and the Davey Endowment for Brain Research.
1. Materials for HAMA synthesis and photopolymerization | |||
Hyaluronic acid (HA) | Sigma-Aldrich | 53747-10G | Streptococcus equi, MW: 1.5 – 1.8 X 10^6 |
Methacrylic anhydride (MA) | Sigma-Aldrich | 275585-100ML | |
Sodium hydroxide (NaOH) | Sigma-Aldrich | 221465-25G | |
Ethanol (EtOH) | Commerical Alcohols Inc. | Anhydrous | |
Phosphate buffered saline (pH 7.4) tablets | Fisher Scientific | 18912014 | |
Triethanolamine (TEA) | Sigma-Aldrich | 90279-100ML | |
1-Vinyl-2pyrrolidinone (NVP) | Sigma-Aldrich | V3409-5G | |
EosinY (EY) | Sigma-Aldrich | E6003-25G | |
Polydimethylsiloxane (PDMS) Sylgard 184 Silicone Elastomer Kit | Dow Corning | ||
3-(Trimethoxysilyl)propyl methacrylate | Sigma-Aldrich | 440159-100ML | |
Beaker (100 mL) | Corning | 1000-100 | |
Beaker (500 mL) | Corning | 1000-600 | |
pH paper (Labstick) | Sigma-Aldrich | 9580 | |
Name | Company | Catalog Number | Comments |
2. Materials for glial cell isolation and cell culture | |||
P1-2 Sprague Dawley rat pups | Charles River | CD Sprague Dawley rat strain code 001 | |
Dissector scissors – slim blades (small) | Fine Science Tools | 14081-09 | |
Surgical scissors – Toughcut (large) | Fine Science Tools | 14130-17 | |
Fine forceps (Dumont #5) | Fine Science Tools | 11521-10 | |
Curved fine forceps (Dumont #7) | Fine Science Tools | 11271-30 | |
Hank's balanced salt solution (HBSS) | Gibco | 14170-112 | |
Dulbecco's modified Eagle's medium and Ham's nutrient mixture F-12 (DMEM/F12) | Gibco | 11320-033 | |
Penicillin-streptomycin (PS) | Gibco | 15140-122 | |
Fetal bovine serum (FBS) | Gibco | 12483-020 | |
0.25% Trypsin-ethylenediaminetetraacetic acid (EDTA) | Gibco | 25200-072 | |
Poly-L-lysine (PLL) | Sigma-Aldrich | P-6282 | |
50 mL conical centrifuge tube | Fisher Scientific | 05-539-13 | |
15 mL conical centrifuge tube | Fisher Scientific | 05-539-5 | |
12 well Tissue culture treated plates (Cellstar) | Greiner Bio-One | 665 108 | |
10 mL serological pipette | Fisher Scientific | 13-676-10F | |
25 mL serological pipette | Fisher Scientific | 12-676-10K | |
Petri dish (60 mm X 15 mm) | Fisher Scientific | FB0875713A | |
Petri dish (100 mm X 15 mm) | Fisher Scientific | FB0875712 | |
Microscope Coverslip (18 mm) | Fisher Scientific | 12-545-100 18CIR | |
Name | Company | Catalog Number | Comments |
3. Materials for microscopy (confocal and scanning electron microscopy) | |||
Mouse monoclonal anti-CNPase | abcam | ab6319 | |
Rabbit anti-Iba1 | Wako Laboratory Chemicals | 019-17741 | |
Chicken anti-GFAP | abcam | ab4674 | |
Hoechst 33342 | Fisher Scientific | 62249 | |
Fluoromount-G | Fisher Scientific | 00-4958-02 | |
Formalin | Sigma Aldrich | HT501128-4L | Buffered (10%) |
Triton X-100 | Fisher Scientific | BP151-500 | |
Horse Serum | Gibco | 16050-122 | |
Paraformaldehyde | Electon Microscopy Sciences | 157-8 | Buffered (8%) |
Guteraldehyde | Electon Microscopy Sciences | 16019 | Buffered (8%) |
Osmium tetraoxide | Electon Microscopy Sciences | 19152 | Buffered (2%) |
Hexamethyldilazane (HMDS) | Electon Microscopy Sciences | 16700 | |
Ethanol (EtOH) | Electon Microscopy Sciences | 15055 | Anhydrous |
Microscope Slide (25 X 75 X 1 mm) | VWR International | 48311-703 |