Neutralization of BDNF attenuates the in vitro protective effects of olfactory ensheathing cell-conditioned medium on scratch-insulted retinal ganglion cells

Abstract

Transplantation of olfactory ensheathing cells (OECs) becomes one of the promising strategies in restoring lost functions of injured central nervous system. Elevated level of expressed brain-derived neurotrophic factor (BDNF) was revealed in the previous studies to be related to the protective effects of OECs on injured cortical and brain stem neurons as well as retinal ganglion cells (RGCs), but no evidence has been obtained to demonstrate whether transplanted OECs protect injured central neurons directly by their secreted BDNF. In the present study, the effects of BDNF neutralization on the neuroprotection of adult OEC-conditioned medium (OEC-CM) on scratch-insulted RGCs were examined. The results showed that OEC-CM protected cultured RGCs from scratch insult, and neutralization of BDNF by BDNF neutralizing antibody attenuated such neuroprotection of the medium. It is thus concluded that neurotrophic factors including BDNF secreted by OECs can protect injured OECs in vitro and BDNF plays a major role in such a protection of OECs.

Fig. 1

Schematic diagram of the scratches (lines) and squares chosen for cell counting. Square 1 represents the injured area adjacent to the scratch and square 2 represents normal areas approximately 2 mm away from the scratch

Fig. 2

Phase contrast micrograph of cultured RGCs. a Intact RGCs distributed scatteredly when cultured for 7 days. b RGCs near the scratch (asterisk) had retracted processes (arrows) which were magnified and shown in the b. Bar = 100 μm

Fig. 3

Phase contrast micrograph of injured RGCs adjacent to the scratch in OEC-CM and NM groups. At 3- a and 6-day c time points, the processes of RGCs treated with OEC-CM (arrow heads) distributed to the area adjacent to the scratch (asterisk), while NM-treated RGCs appear to swell and collapse (arrow) shown in the b and d in this area at 3- b and 6-day d time points. Bar = 100 μm

Fig. 4

Morphology and numbers of β-tubulin III immunostained RGCs. Three days after scratch, RGCs in OEC-CM group a had bright, thick, and dense β-tubulin III immunoreactive processes (arrow) while cells in NM group b possessed thin and ruptured processes (arrow head). RGCs in the normal areas remained intact and showed no difference in the morphology of immunostained cells between OEC-CM c and CM d groups. Quantitative data of cell counting are presented. * and ** represent significant differences of P < 0.05 and P < 0.01, respectively, in comparison with NM group e. Dash lines indicate the location of the scratch and all the nuclei of cultured cells were shown with Hoechst 33342 counterstaining (blue/scattered oval nuclei). Bar = 50 μm (Color figure online)

Fig. 5

OD values of RGCs. OD value of injured RGCs treated with OEC-CM was significantly higher than that treated with NM (**P < 0.01) at 3-day, but not 6-day, time point. There was no significant difference between groups of intact RGCs treated with either OEC-CM or CM

Fig. 6

BDNF levels in concentrated NM, OEC-CM, and OEC extracts. BDNF level was detected through the western blot assay in OEC extract and concentrated OEC-CM, but not in concentrated NM

Fig. 7

Percentages of β-tubulin III positive RGCs. There was significant difference in the percentages of β-tubulin III immunostained surviving RGCs that were exposed to OEC-CM with or without BDNF neutralization (**P < 0.01). No significant difference could be found in the percentages of surviving RGCs between neutralized OEC-CM and NM treatments or between OEC-CM alone and OEC-CM plus non-specific IgG treatment