2.1 Experimental animals

Neonatal Sprague–Dawley (SD) rats were provided by the Department of Laboratory Zoology, Kunming Medical University, and all experiments were approved by the Ethics Committee of Kunming Medical University (Approval No. kmmu20220795).

2.2 Reagents and instruments

Neurobasal-A medium, B27, GlutaMAX-L, trypsin (TP), and fetal bovine serum (FBS) were purchased from Gibco Company; Dulbecco's modified Eagle's medium (DMEM) (high glucose, containing sodium pyruvate) was purchased from Biosharp Company; Penicillin–Streptomycin (PS) was purchased from Hyclone Company; Poly-l-lysine (PLL) was purchased from Solarbio Company; Neuronal β-III Tubulin (Tuj1) antibody was purchased from abcam Company; and antifluorescence Quenching mounting medium and 4’,6-diamidino-2-phenylindole (DAPI) were purchased from Beyotime Company. B2 The Biosafety cabinet (Thermo), dissecting microscope, and inverted phase-contrast microscope were manufactured by Nikon Corporation.

2.3 Configuration of culture medium

Basal medium: 89% DMEM + 10% FBS + 1% PS; Neuron-specific medium: 97% Neurobasal-A + 2% B27 + 1% GlutaMAX-L + 0.5% PS.

2.4 Pretreatment of culture plates

The plates were wrapped with 1× PLL at 37°C for 2 h before sampling and then recovered, and they were air-dried in a 37°C, 5% CO₂ incubator for later use.

2.5 Isolation of SPNs

The neonatal 24 h SD mammary rats were disinfected with a 75% alcohol spray and soaked for 3~5 min and then decapitated and killed. The epidermal tissue was peeled off along the dorsal midline, the upper end of the spinal cord was held with forceps, the upper dorsal root of the spinal column was separated with scissors, care was taken not to cut the viscera, the whole spinal column was separated and placed into a Petri dish containing phosphate-buffered saline (PBS), the spinal cord was blown out of the spinal cord cavity with a 10 ml syringe from the lumbosacral section of the spinal column, the spinal cord was immediately placed in a pre-chilled DMEM high-sugar medium, and the vascular membrane of the spinal cord was peeled off under a dissecting microscope. The stripped spinal cord tissue was placed in a 5 ml EP tube, cut into small pieces of 1 mm³ size with microscopic scissors, and digested by adding 1.5 ml of 0.25% TP at 37°C in a 5% CO₂ incubator for 10–15 min, shaking once every 5 min. After digestion, 2 ml of basal medium was added to terminate the digestion. The cells were gently blown 30 times with a barrel and held for 2 min. The supernatant was obtained and filtered into a 50 ml centrifuge tube with a 40 μm cell strainer, and then 3 ml of DMEM was added to the precipitated tissue mass. The above blowing steps were repeated and the cell suspension was aspirated and filtered; this was again repeated three times, centrifugation was performed at 1000 r.p.m. for 5 min, and the supernatant was removed. The cells were resuspended by adding the basal medium and then counted; the cell density was adjusted to 1.3 × 10⁶/ml. The cell suspension was inoculated into a culture plate (100 μl/well for 96-well plate, 500 μl/well for 24-well plate, 2 ml/well for 6-well plate) and incubated at 37°C and 5% CO₂ in an incubator; the basal medium was changed to neuronal medium after 4 h of incubation. After that, the half medium was changed every 3 days (d).

2.6 Purification of SPNs

2.7 Immunofluorescence identification

The cultured SPNs were washed once with PBS and fixed with 4% paraformaldehyde for 15 min, washed three times with PBS for 5 min/time, then permeabilized and blocked with a 0.3% Triton solution and 5% goat serum at 37°C for 30 min, followed by aspiration of the liquid and incubation with anti-Tuj1 mouse monoclonal antibodies (ab78078, 1:250) at 4°C overnight. Subsequently, after washing three times with PBS for 5 min/time, the samples were incubated with secondary antibodies. Dylight 488 goat antimouse IgG (A23210, 1:250, Abbkine) was incubated at 37°C for 1 h, then washed with PBS three times (5 min/time), and DAPI was added to counterstain the nuclei. The fluorescent images were observed using a Nikon three-dimensional inverted phase-contrast microscope.

2.8 Cell Counting Kit-8 (CCK-8) assay

SPNs were inoculated in 96-well culture plates (1.3 × 10⁵/well) and incubated in a 5% CO₂ incubator at 37℃ for 18 and 20 d. Neurons were washed with PBS once, 100 μl of neuron medium and 10 μl of CCK-8 solution were added into each well of the 96-well plate to be tested, and incubated at 37℃ for 2 h. A microplate reader (BioTek Instruments, Inc.) was used to determine the OD values in each experimental well at 450 nm and to detect changes in cell proliferation capacity in each group.

2.9 Statistical analysis

All data in the experimental process are presented as mean ± SD and plotted using Graph Pad Prism 8 (Graph Pad Software). SPSS 21.0 was used for statistical analysis. Differences in cell viability between 18 and 20 d were analyzed using a t test of independent samples. The data were normally distributed. First, the Levene test was used to detect homogeneity of variance, and then the results of homogeneity of variance were used to determine whether the difference was statistically significant. p  <  0.05 indicated a significant difference.

3.1 In vitro culture of SPNs

The spinal cord was isolated by disinfection with 75% ethanol, decapitation, and killing of 1 d SD rats, digested with 0.25% TP for 10–15 min, centrifuged, and then inoculated using basal medium. Many stray cells and low purity of neurons were observed, as shown in the flow chart of SPN culture in Figure 1.

3.2 Purification of SPNs by Ara-C treatment

When the SPNs were first inoculated, the cells were round, translucent, and well refracted. Four hours after inoculation, most of the neuronal cells were already attached to the wall, the cells were round, small, and transparent, and synapses had not yet grown. After 1 d of inoculation, some synapses were visible, but the synapses were short and few. SPNs treated with not Ara-C at 3, 7, and 11 d in Figure 2A had significantly more mottled cells compared with those treated with Ara-C in Figure 2B–D, but the morphology and status of cells without Ara-C treatment were better. Compared with 3 μg/ml Ara-C treatment as shown in Figure 2B and 1 μg/ml Ara-C treatment as shown in Figure 2C, D, a large number of cells died, and the structure of SPNs was basically incomplete and in a poor state. Two days after 1 μg/ml Ara-C treatment (Figure 2C), it was observed that there were few miscellaneous cells but the number of neurons was less and the cell was in a poor state compared with Figure 2A, and the cell morphology continued to change and the number of cells continued to decrease after the fluid change. Before and after treatment with Ara-C, the reduction of heterogeneous cells was not obvious, but the status of neuron was affected (Figure 2D). At 4, 24, 48, and 72 h after seeding, SPNs were treated with 0.3, 0.6, 0.8, and 1 μg/ml of Ara-C, respectively. Ara-C did not have a good inhibitory effect on hybrid cells and damaged neurons, as shown in Figure 3A–D.

3.3 Differential adherent culture of SPNs

The differential adhesion method was used to culture neurons. It was found that almost all cells in PLL-coated culture plates adhered to the wall after 1 h of cell inoculation, and there were almost no cells in the supernatant after 1.5 h. There was no significant difference between the neuronal heterocells at 0.5 and 1 h, and the liquid exchange at 4 h. After 0.5 to 4 h of differential velocity, there were cells in the culture plates without PLL coating, but the number of cells was less than that of the original culture plate, and there were still heterogenous cells (Figure 4).

3.4 Naturally growing SPNs

On the 1st to 4th day after seeding of spinal neurons, there were few heterocytes; on the 5th to 10th day, a large number of heterocytes proliferated to cover neurons; on the 8th day, a small number of heterocyte cells were observed to aggregate; on the 11th day, the heterogeneous cells appeared aggregated and exposed and a single neuron was obviously visible; and on the 18th day, the axons of neurons aged and broke, and the cell state gradually deteriorated. After 22 d, the cells aged and axons disappeared (Figure 5). Immunofluorescence results showed a relative decrease in the number of parenchyma cells and an increase in the purity of naturally growing spinal neurons after 12 d (Figure 6). The cell viability of SPNs cultured for 18 d was better than that of SPNs cultured for 20 d (p < 0.001) (Figure 7).

4.1 Spinal cord tissue acquisition

In the rat spinal cord, there are various types of cells: neurons, oligodendrocytes, astrocytes, microglia, fibroblasts, and endothelial cells.⁹ The primary culture of SPNs will inevitably be contaminated by other cell types, so in our method of isolating and purifying SPNs from the spinal cord, we use a dissecting microscope to carefully dissect the tiny capillaries and dura from the spinal cord to reduce contaminating cells, because fibroblasts and endothelial cells are the two major cellular components of the meningeal covering and accompanying tiny capillaries.¹⁰ When we stripped the tiny capillaries and dura as much as possible we can effectively reduce or even remove fibroblasts and endothelial cells from the culture. In addition, to sufficiently dissociate cells and harvest more viable SPNs, we chose to trypsinize tissue. While trypsinization is acute, it can cause direct damage and irritation to dissociated tissue, so we use a low concentration of TP, which leads to relatively mild digestion and fewer dead cells. By controlling the pH, temperature, concentration, and time in real time, overdigestion and cell damage can be avoided.^{11, 12} Papain has also been widely used to digest and culture primary neurons.^{13, 14} Studies have compared the digestion efficiency and culture effect of TP and papain, and found that it is comparable to that of papain. Compared with the papain group, the TP group had more cortical neurons, larger cell bodies, longer axonal lengths, and fewer impurities.¹⁵

4.2 Ara-C kills dividing and proliferating cells

Ara-C is an antimetabolite that affects pyrimidine metabolism, which is catalyzed by deoxycytidine kinase to arabinosine diphosphate or glycytidine triphosphate in vivo, which can inhibit the activity of DNA polymerase and interfere with nucleotide incorporation into DNA. Inhibition of nucleotide reductase and blocking of the reduction of cytosine nucleotide to deoxycytosine nucleotide, which specifically acts on the S phase, thus affect DNA synthesis in the S phase of dividing cells and kill dividing cells.^{16, 17} Neuronal cells are highly differentiated cells that do not proliferate, and Ara-C has almost no direct inhibitory effect on their growth, but can disrupt the neuronal cell growth environment by inhibiting the migration and proliferation of non-neuronal cells, leading to a decrease in the secretion of trophic factors such as the glial cell line-derived neurotrophic factor (GDNF) in non-neuronal cells, which disrupts the neuronal cell growth environment and thus indirectly affects the growth of SPN cells.¹⁸ Therefore, the rational use of Ara-C is the key to purification. If Ara-C is added too early or at too high a concentration, the coexistence environment between neurons and glial cells will be destroyed, glial cells will die in large numbers, and the GDNF secreted by glial cells to promote neuronal survival and morphological differentiation will be markedly reduced; if it is added too late or at too low a concentration, glial cells will divide and proliferate in large numbers and consume a large amount of nutrients in the culture medium, which will also inhibit the survival of spinal neurons. The continuous presence of Ara-C in the culture system also exerts toxic effects on neurons.^{19, 20} We tried for the first time to treat SPNs at 4, 24, 48, and 72 h of inoculation with concentrations of 0.3, 0.6, 0.8, and 1 μg/ml of Ara-C, respectively, but low concentrations of Ara-C did not inhibit promiscuous cells well and may also damage neurons by prematurely disrupting the normal physiological environment of neurons. Therefore, Ara-C is not a good choice in the purification process of SPNs.

4.3 Differential velocity adherent with the PLL coating

The differential velocity adherent method is a purification method that induces little damage. It achieves the purpose of separating and purifying neurons according to the different adhesion ability of various cells in nerve tissue.²¹ In the culture system, the most easily adherent cells are fibroblasts, followed by glial cells, and finally neurons.²² PLL encapsulation can improve the adhesion ability of cells.^{23, 24} Therefore, we assume that PLL has different adhesion ability to different cells and will separate fibroblasts and other heterocytic cells. However, it was observed during the experiment that PLL-coated pore cells were basically stuck to the wall after 0.5 h of inoculation, and only a small number of cells and few neurons were observed after the replacement supernatant was re-inoculated. Therefore, this method is not very useful for purifying neurons.

4.4 Natural growth

Neurobasal-A without serum medium is dedicated to neuronal culture with B27 nerve growth factor. It prevents the growth of glial cells and other non-neuronal cells, promotes neuronal growth, and prolong neuronal survival time.²⁵ In the process of culture, we changed the amount of fluid by half every 3 days to remove the suspended cells with inhibited growth, leaving purer neurons and improving the purity of neurons.

In conclusion, SPNs naturally grew in Neurobasal-A medium with B27 to obtain high purity neurons, without adding Ara-C and differential adherent operation, which is easy to operate. SPNs on Days 12–18 can be selected as experimental cells.