Promoting optic nerve regeneration in adult mice with pharmaceutical approach

Abstract

Our previous research has suggested that lack of Bcl-2-supported axonal growth mechanisms and the presence of glial scarring following injury are major impediments of optic nerve regeneration in postnatal mice. Mice overexpressing Bcl-2 and simultaneously carrying impairment in glial scar formation supported robust optic nerve regeneration in the postnatal stage. To develop a therapeutic strategy for optic nerve damage, the combined effects of chemicals that induce Bcl-2 expression and selectively eliminate mature astrocytes--scar forming cells--were examined in mice. Mood-stabilizer, lithium, has been shown to induce Bcl-2 expression and stimulate axonal outgrowth in retinal ganglion cells in culture and in vivo. Moreover, astrotoxin (alpha-aminoadipate), a glutamate analogue, selectively kills astrocytes while has minimal effects on surrounding neurons. In the present study, we sought to determine whether concurrent applications of lithium and astrotoxin were sufficient to induce optic nerve regeneration in mice. Induction of Bcl-2 expression was detected in the ganglion cell layer (GCL) of mice that received a lithium diet in compared with control-treated group. Moreover, efficient elimination of astrocytes and glial scarring was observed in the optic nerve of mice treated with astrotoxin. Simultaneous application of lithium and astrotoxin, but not any of the drugs alone, induced robust optic nerve regeneration in adult mice. These findings further support that a combinatorial approach of concurrent activation of Bcl-2-supported growth mechanism and suppression of glial scarring is required for successful regeneration of the severed optic nerve in adult mice. They suggest a potential therapeutic strategy for treating optic nerve and CNS damage.

Fig. 1. Intake of lithium-containing diet increases lithium concentration in serum and upregulates Bcl-2 expression

(A) Graph showing a significant increased in lithium concentration in lithium-treated mice than the control wild-type (wt) mice consumed control diet (Student’s t-test, P < 0.00001). Values are mean ± s.d., (B) RT-PCR results show the induction of Bcl-2 mRNA in the retinas of lithium-treated mice. Note the Bcl-2 mRNA expresses in the retinas derived from lithium-treated wt mice and the positive control Bcl-2 transgenic (tg) mice but not in the wt mice that consumed a control diet. The expression of house-keeping gene, G3PDH, is served as an internal control

Fig. 2. Localization of lithium induced Bcl-2 expression in adult mouse retina

Epifluorescence photomicrographs of retinal sections derived from mice fed with control diet (A–C) or lithium-containing diet (D–F). DAPI staining localizes the nuclear layers in retina (B, E). Arrow marks the co-localization of Bcl-2 protein in ganglion cell layer (GCL). The insert is an enlarged image from the arrowhead pointed cells (F). Note there is undetectable signal of Bcl-2 protein in the control retina (A–C). Scale = 10 µm

Fig. 3. Simultaneously administration of lithium and astrotoxin promote robust regeneration of the severed optic nerves in adult wt mice

Montages of photomicrographs showing optic nerve sections in adult wt mice 8 days after optic nerve crush (A–L). First row: Cresyl violet staining identifies the crushed site of optic nerve and marked by asterisk. Second row: GAP-43 staining reveals the length regenerating axons along the crushed optic nerve. Extensive GAP-43 positive axons regenerate posterior to the crush site derived from the mice receiving lithium and astrotoxin simultaneously (J) but no regenerating axons were observed posterior to the crushed site in the other groups (B and F). Third row: Strong GFAP staining extends along the crushed optic nerve. In the groups of mice receiving astrotoxin, there is a track of GFAP-negative area along the optic nerve (G and K). In the group of mice receiving lithium and astrotoxin simultaneously, the overlay shows that the regenerating axons fulfill the GFAP-negative area, which is astrocyte-ablated, of the crushed optic nerve (I). Scale = 250 µm. Fourth row: Growth cone-like structure of regenerating axons in the crushed optic nerve. In the mice received lithium-containing diet and astrotoxin simultaneously, many GAP-43 labeled regenerating axons with growth cone-like structure (arrow marked) exhibits at their expanding tip (M and N). Scale = 5 µm

Fig. 4. A graph showing the length of regenerating axons in the crushed optic nerve

The length of regenerating axons in the mice simultaneously receiving lithium (Li) and astrotoxin (AA) is significantly longer than that of the other groups (P < 0.001). The numbers of regenerating axons measured in the groups of control, Li, AA and Li + AA are 8.17 ± 4.09, 84.4 ± 28.2, 52.2 ± 48.9 and 435 ± 68.2, respectively. Values are mean ± SEM

Fig. 5. Lithium-containing diet exerts no effect on prevention of RGCs death

(A) The graph shows the density of surviving RGCs, which is FluoroGold (FG) pre-labeled, in normal retinas (N) and retinas with optic nerve lesion (L) in the wt mice. There is no significant difference between the corresponding groups with or without lithium-treated (P > 0.05). Photomicrographs showing the surviving FG-labeled RGCs in normal retina (B) and lesioned retina (C) derived from lithium-treated mice. Arrow marks the microglia with a feature of elongated cell body and fine processes. Scale = 50 µm. Li −ve, mice received control food chow; Li +ve, mice received lithium-containing food chow