We present an in vitro model to assess olfactory ensheathing glia (OEG) neuroregenerative capacity, after neural injury. It is based on a coculture of axotomized adult retinal ganglion neurons (RGN) on OEG monolayers and subsequent study of axonal regeneration, by analyzing RGN axonal and somatodendritic markers.
Olfactory ensheathing glia (OEG) cells are localized all the way from the olfactory mucosa to and into the olfactory nerve layer (ONL) of the olfactory bulb. Throughout adult life, they are key for axonal growing of newly generated olfactory neurons, from the lamina propria to the ONL. Due to their pro-regenerative properties, these cells have been used to foster axonal regeneration in spinal cord or optic nerve injury models.
We present an in vitro model to assay and measure OEG neuroregenerative capacity after neural injury. In this model, reversibly immortalized human OEG (ihOEG) is cultured as a monolayer, retinas are extracted from adult rats and retinal ganglion neurons (RGN) are cocultured onto the OEG monolayer. After 96 h, axonal and somatodendritic markers in RGNs are analyzed by immunofluorescence and the number of RGNs with axon and the mean axonal length/neuron are quantified.
This protocol has the advantage over other in vitro assays that rely on embryonic or postnatal neurons, that it evaluates OEG neuroregenerative properties in adult tissue. Also, it is not only useful for assessing the neuroregenerative potential of ihOEG but can be extended to different sources of OEG or other glial cells.
Adult central nervous system (CNS) neurons have limited regenerative capacity after injury or disease. A common strategy to promote CNS regeneration is transplantation, at the injury site, of cell types that induce axonal growth such as stem cells, Schwann cells, astrocytes or olfactory ensheathing glia (OEG) cells1,2,3,4,5.
OEG derives from the neural crest6 and locates in the olfactory mucosa and in the olfactory bulb. In the adult, olfactory sensory neurons die regularly as the result of environmental exposure and they are replaced by newly differentiated neurons. OEG surrounds and guides these new olfactory axons to enter the olfactory bulb and to establish new synapses with their targets in the CNS7. Due to these physiological attributes, OEG has been used in models of CNS injury such as spinal cord or optic nerve injury and its neuroregenerative and neuroprotective properties become proven8,9,10,11. Several factors have been identified as responsible of the pro-regenerative characteristics of these cells, including extracellular matrix proteases production or secretion of neurotrophic and axonal growth factors12,13,14.
Given the technical limitations to expand primary OEG cells, we previously established and characterized reversible immortalized human OEG (ihOEG) clonal lines, which provide an unlimited supply of homogeneous OEG. These ihOEG cells derive from primary cultures, prepared from olfactory bulbs obtained in autopsies. They were immortalized by transduction of the telomerase catalytic subunit (TERT) and the oncogene Bmi-1 and modified with the SV40 virus large T antigen15,16,17,18. Two of these ihOEG cell lines are Ts14, which maintains the regenerative capacity of the original cultures and Ts12, a low regenerative line that is used as a low regeneration control in these experiments18.
To assess OEG capacity to foster axonal regeneration after neural injury, several in vitro models have been implemented. In these models, OEG is applied to cultures of different neuronal origin and neurite formation and elongation—in response to glial coculture—are assayed. Examples of such neuronal sources are neonatal rat cortical neurons19, scratch wounds performed on rat embryonic neurons from cortical tissue20, rat retinal explants21, rat hypothalamic or hippocampal postnatal neurons22,23, postnatal rat dorsal root ganglion neurons24, postnatal mouse corticospinal tract neurons25, human NT2 neurons26, or postnatal cerebral cortical neurons on reactive astrocyte scar-like cultures27.
In these models, however, the regeneration assay relies on embryonic or postnatal neurons, which have an intrinsic plasticity that is absent in injured adult neurons. To overcome this drawback, we present a model of adult axonal regeneration in cocultures of OEG lines with adult retinal ganglion neurons (RGNs), based on the one originally developed by Wigley et al.28,29,30,31 and modified and used by our group12,13,14,15,16,17,18,32,33. Briefly, retinal tissue is extracted from adult rats and digested with papain. Retinal cell suspension is then plated on either polylysine-treated coverslips or onto Ts14 and Ts12 monolayers. Cultures are maintained for 96 h before they are fixed and then immunofluorescence for axonal (MAP1B and NF-H proteins)34 and somatodendritic (MAP2A and B)35 markers is performed. Axonal regeneration is quantified as a percentage of neurons with axon, with respect to the total population of RGNs and axonal regeneration index is calculated as the mean axonal length per neuron. This protocol is not only useful for assessing the neuroregenerative potential of ihOEG but can be extended to different sources of OEG or other glial cells.
NOTE: Animal experimentation was approved by national and institutional bioethics committees.
1. ihOEG (Ts12 and Ts14) culture
NOTE: This procedure is done under sterile conditions in a tissue culture biosafety cabinet.
2. Preparation of ihOEG (Ts12 and Ts14) for the assay
NOTE: This step must be done 24 h before RGN dissection and coculture.
3. Retinal tissue dissection
NOTE: 2-month old male Wistar rats are used as RGN source. Two retinas (one rat) for 20 wells of a 24-well cell dish. Autoclave surgical material before use. Papain dissociation kit is commercially purchased (Table of Materials). Follow the provider´s instructions for reconstitution. Reconstitute D,L-2-amino-5-phosphonovaleric acid (APV) in 5 mM stock and prepare the aliquots.
4. Immunostaining
5. Axonal regeneration quantification
NOTE: Samples are quantified under the 40x objective of an epifluorescence microscope. A minimum of 30 pictures should be taken on random fields, with at least 200 neurons, to be quantified for each treatment. Each experiment should be repeated a minimum of three times.
In this protocol, we present an in vitro model to assay OEG neuroregenerative capacity after neuronal injury. As shown in Figure 1, the OEG source is a reversible immortalized human OEG clonal cell line -Ts14 and Ts12-, which derives from primary cultures, prepared from olfactory bulbs obtained in autopsies15,17,18. Retinal tissue is extracted from adult rats, digested, and retinal ganglion neurons (RGN) suspension is plated on either PLL-treated coverslips or onto ihOEG monolayers, Ts14 or Ts12. Cultures are maintained for 96 h before they are fixed. Axonal and somatodendritic markers are analyzed by immunofluorescence and axonal regeneration is quantified.
Ts14 OEG identity is assessed by immunostaining with markers described to be expressed in ensheathing glia (Figure 2), such as S100 β (Figure 2A) and vimentin (Figure 2B); GFAP expression was also analyzed to discard astrocyte contamination (Figure 2C). As shown, Ts14 expressed S100 β and vimentin but not GFAP.
In the axonal regeneration assay, Ts14 regenerative capacity is compared to Ts12 in RGN-OEG cocultures, using PLL substrate as a negative control (Figure 3). Both the percentage of cells with axons as well as the average length of the regenerated axons were significantly higher in neurons cocultured on Ts14 monolayers, compared to neurons plated on either Ts12 cells or PLL (Figure 3D,E). Representative images show a lack of capacity of RGN to regenerate their axons over PLL or Ts12 cells (Figure 3A,B), while Ts14 stimulates the outgrowth of axons in RGN (3C).
Figure 1: Diagram of rat retinal ganglion neurons with olfactory ensheathing glia cells coculture, as a model of adult axonal regeneration. Immortalized human OEG (ihOEG) clonal cell lines -Ts12 and Ts14- derived from primary cultures from olfactory bulbs. Retinal ganglion neurons from adult rats are plated on either PLL-treated coverslips (negative control) or onto Ts14 or Ts12 monolayers. Cultures are maintained for 96 h before they are fixed and axonal and somatodendritic markers are analyzed by immunofluorescence. Percentage of neurons with axon and mean axonal length/neuron are quantified to assay RGN axonal regeneration. Please click here to view a larger version of this figure.
Figure 2: Identity of ihOEG cell line Ts14. Immunofluorescence images of Ts14 in culture, labeled with anti-S100 β (panel A, green) and vimentin (panel B, red). GFAP expression (panel C, red) was also analyzed to discard astrocyte contamination. Nuclei are stained with DAPI (blue). Please click here to view a larger version of this figure.
Figure 3: Assay for axonal regeneration in cocultures of OEG lines with adult retinal ganglion neurons (RGNs). (A–C) Immunofluorescence images showing somatodendritic labelling with 514 antibody, which recognizes microtubule-associated protein MAP2A and B, in red, and with axon-specific SMI31 antibody in green, against MAP1B and NF-H proteins. Green arrows indicate RGN axons (SMI31-positive: green) and yellow arrows indicate neuronal bodies and dendrites (514 positive: red and yellow). (D,E) Graphs show mean and standard deviation of the percentage of neurons exhibiting axons and the axonal regeneration index, a parameter reflecting the mean axonal length (µm) of axons per neuron. A minimum of 30 pictures (40x) were taken on random fields and quantified for each cell sample. Experiments were performed in triplicate, from three different rats (N = 3), retinal tissue pooled from both eyes, with duplicates for each experimental condition (each glia population tested). Asterisks indicate the statistical significance: *p < 0.05, **p < 0.01, ***p < 0.001, NS: non significance (ANOVA and post hoc Tukey test comparisons between parameters quantified for Ts14 vs Ts12, Ts14 vs PLL, and Ts12 vs PLL). Please click here to view a larger version of this figure.
OEG transplantation at CNS injury sites is considered a promising therapy for CNS injury due to its constitutive pro-neuroregenerative properties7,8,9. However, depending on the tissue source—olfactory mucosa (OM-OEG) versus olfactory bulb (OB-OEG)—or the age of the donor, considerable variation exists in such capacity26,31,33,36. Therefore, it is of importance to have an easy and reproducible in vitro model to assay the neuroregenerative capacity of a given OEG sample, before initiating in vivo studies. In this protocol, adult rats’ axotomized RGN are cocultured onto a monolayer of the OEG to assay. Subsequent analysis of RGN axonal and somatodendritic markers by immunofluorescence is performed to assess RGN axonal regeneration.
An initial difficulty of the assay is the source of OEG. In this work, we use reversible immortalized human OEG (ihOEG) clonal lines, previously established and characterized by our group15,16,17,18, which provide an unlimited supply of homogeneous OEG. Two of these ihOEG cell lines are Ts14, which maintains the regenerative capacity of the original cultures and Ts12, a low regenerative line that is used as a low regeneration control in these experiments18 Nevertheless, although technical limitations exist to expand human primary OEG cells, they can also be obtained from nasal endoscopic biopsies—OM—or, in case of OB-OEG, from cadaver donors.
Preparation of monolayer OEG cultures is a crucial procedure, as too many cells could cause the coculture to detach from the plate. Therefore, prior to OEG preparation for the assay, it is recommended that the user determines the optimal number of cells to be plated, depending on their size and division rate.
Another critical issue is the retinal tissue dissociation, after retina dissection. It is necessary to break up the tissue fragments, following incubation in the dissociation mix. If done too vigorously, the cells will be destroyed, but tissue fragments will be left intact if done too weakly. In order to obtain a homogeneous cell suspension, we suggest filling and emptying a Pasteur pipette 10–15 times, with a tip of intermediate diameter, while avoiding bubbling. Pasteur pipettes with wide tips can be narrowed using a Bunsen burner.
To assess the capacity of different glial populations to foster adult neurons’ axonal regeneration, we have determined that 96 h is the time interval that best suits the aim because: 1) it is the longest time to maintain the culture alive without disturbing the OEG monolayer; and 2) it is the time needed for neurons to grow axons long enough to reveal differences between the regenerative capacities of different OEG populations or other non-regenerative cells (i.e., fibroblasts12,13,14,15,16,17,18,32,33). It would certainly be interesting to determine the time course of the regeneration process, as it could provide information about the differential regenerative properties of different glial populations, at shorter times of the co-culture. In our hands, for regenerative glia, the time course between 72–96 h is quite similar for all the cell lines, although axons are shorter at 72 h (unpublished data). Also, 96 h of co-culture, permits to study OEG-dependent mechanisms of adult axonal regeneration12,14.
During axonal regeneration quantification, it is important to take a minimum of 30 pictures at 400 augments (40x objective), at different random areas of the coverslip, but following the complete axons of the photographed neurons. Therefore, the experimenter must take serial pictures in the chosen areas to measure the real axonal lengths.
Other in vitro approaches have also been developed to evaluate OEG regenerative functions. In these models, OEG is applied to cultures of different neuronal origin and, in response to glial coculture, neurite formation and elongation are assayed19,20,21,22,23,24,25,26,27. However, the regeneration assay relies on embryonic or postnatal neurons, which have an intrinsic plasticity absent from injured adult neurons. This model consisting of adult axonal regeneration in cocultures of OEG lines with adult retinal ganglion neurons (RGNs) overcomes this drawback. In addition, we are dissecting adult retinas, and because we cut optic nerve and axons retract in the process of dissection, we obtain neuronal bodies clean of myelin, to perform the coculture. This is the difference with other parts of the adult CNS, where myelin can hinder very much with the dissection to obtain clean neurons for the coculture.
Based on the one originally developed by Wigley et al.28,29,30,31, we highlight the following improvements in the protocol. First, the use of neurobasal medium supplemented with B27 as OEG-RGN coculture medium, which allows growth of neuronal cells and positively affects the reproducibility of the experiment. Second, we characterize and quantify axonal regeneration by using a specific marker of the axonal compartment; and third, we use an additional direct parameter, the mean axonal length/neuron, that assesses the axonal growth regenerative potential of OEG.
In summary, we consider that this is a simple, reproducible, time saving, and medium-cost assay, not only useful for assessing the neuroregenerative potential of ihOEG, but also because it can be extended to different sources of OEG or other glial cells. Moreover, it could be used as a valuable proof of concept of the neuroregenerative potential of an OEG or glial sample, before translation to in vivo or clinical studies.
The authors have nothing to disclose.
This work was financially supported by project SAF2017-82736-C2-1-R from Ministerio de Ciencia e Innovación to MTM-F and by Fundación Universidad Francisco de Vitoria to JS.
antibody 514 | Reference 34 | Rabbit polyclonal antiserum, which recognizes MAP2A and B. | |
antibody SMI-31 | BioLegend | 801601 | Monoclonal antibody against MAP1B and NF-H proteins |
anti-mouse Alexa Fluor 488 antibody | ThermoFisher | A-21202 | |
anti-rabbit Alexa Fluor 594 antibody | ThermoFisher | A-21207 | |
B-27 Supplement | Gibco | 17504044 | |
D,L-2-amino-5-phosphonovaleric acid | Sigma | 283967 | NMDA receptor inhibitor |
DAPI | Sigma | D9542 | Nuclei fluorescent stain |
DMEM-F12 | Gibco | 11320033 | Cell culture medium |
FBS | Gibco | 11573397 | Fetal bovine serum |
FBS-Hyclone | Fisher Scientific | 16291082 | Fetal bovine serum |
Fluoromount | Southern Biotech | 0100-01 | Mounting medium |
ImageJ | National Institutes of Health (NIH-USA) | Image software | |
L-Glutamine | Lonza | BE17-605F | |
Neurobasal Medium | Gibco | 21103049 | Neuronal cells culture medium |
Papain Dissociation System | Worthington Biochemical Corporation | LK003150 | For use in neural cell isolation |
PBS | Home made | ||
PBS-EDTA | Lonza | H3BE02-017F | |
Penicillin/Streptomycin/Amphotericin B | Lonza | 17-745E | Bacteriostatic and bactericidal |
Pituitary extract | Gibco | 13028014 | Bovine pituitary extract |
Poly -L- lysine (PLL) | Sigma | A-003-M |