Effects of glial cell line-derived neurotrophic factor deletion on ventral mesencephalic organotypic tissue cultures

Abstract

Glial cell line-derived neurotrophic factor (GDNF) is potent for survival and promotion of nerve fibers from midbrain dopamine neurons. It is also known to exert different effects on specific subpopulations of dopamine neurons. In organotypic tissue cultures, dopamine neurons form two diverse nerve fiber growth patterns, targeting the striatum differently. The aim of this study was to investigate the effect of GDNF on the formation of dopamine nerve fibers. Organotypic tissue cultures of ventral mesencephalon of gdnf gene-deleted mice were studied. The results revealed that dopamine neurons survive in the absence of GDNF. Tyrosine hydroxylase immunoreactivity demonstrated, in gdnf knockout and wildtype cultures, nerve fiber formation with two separate morphologies occurring either in the absence or the presence of astrocytes. The outgrowth that occurred in the absence of astrocytes was unaffected by gdnf deletion, whereas nerve fibers guided by the presence of astrocytes were affected in that they reached significantly shorter distances from the gdnf gene-deleted tissue slice, compared to those measured in wildtype cultures. Treatment with GDNF reversed this effect and increased nerve fiber density independent of genotype. Furthermore, migration of astrocytes reached significantly shorter distances from the tissue slice in GDNF knockout compared to wildtype cultures. Exogenous GDNF increased astrocytic migration in gdnf gene-deleted tissue cultures, comparable to lengths observed in wildtype tissue cultures. In conclusion, cultured midbrain dopamine neurons survive in the absence of GDNF, and the addition of GDNF improved dopamine nerve fiber formation - possibly as an indirect effect of astrocytic stimulation.

Fig. 1

Viable TH-positive neurons in gdnf gene-deleted cultures. Representative views of TH-positive neurons in WT (a, n = 26), gdnf^+/− (b, n = 22), and gdnf^−/− (c, n = 14) fetal ventral mesencephalic tissue, cultured for 12 DIV, showed no apparent morphological differences. Scale bar: a–c = 50 μm.

Fig. 2

Nerve fiber formation from midbrain WT organotypic cultures. Fetal ventral mesencephalic tissue cultures showing TH (red) and S100 (green) immunoreactivity at 12 DIV. DAPI staining (blue) revealed the presence or absence of cell nuclei. The TH-positive fibers occurred in two different morphological growth patterns, one in the absence of other cells as revealed by a lack of DAPI labeling (a) and the other in the presence of S100-positive astrocytes (b, c). TH-positive nerve fibers growing in the absence of astrocytes and other cell types were radiating from the tissue slice, without changing direction, as observed at the distal end of the fiber outgrowth (a). The glia-mediated TH-positive nerve fibers had branched and formed a network (c), and no TH-positive fibers extended outside the astrocytic layer in the most distal regions from the tissue slice (b). DAPI staining showed the presence of cells other than S100-positive astrocytes between the migrating S100-positive cells (c, see arrows). TH (red), S100 (green), and DAPI (blue) are represented in panels a–c and S100 (green), and DAPI (blue) are represented in panel d. Scale bar: a–c = 50 μm.

Fig. 3

GDNF effects on length and density of TH-positive outgrowth from ventral mesencephalic cultures. Graphs showing TH-positive glia-guided outgrowth at 12 DIV; (a) distance from tissue slice and (b) nerve fiber density. The glia-guided TH-positive outgrowth reached significantly shorter distances from the tissue slice in gdnf^−/− compared to WT (a), and measurements were performed from the periphery of the tissue slice to the distal end nerve fibers reached. Treatment with GDNF (10 ng/ml medium) reversed the effect of the deletion (a). No significant effect of GDNF treatment was seen on the length of glia-guided TH-positive fibers in gdnf^+/− and WT cultures (a). The addition of GDNF resulted in significantly increased nerve fiber density of glia-guided TH-positive outgrowth in cultures from all genotypes. In figure b, significance comparison of nerve fiber density was performed between GDNF treatment within each genotype. *p < 0.05, **p < 0.01, ***p < 0.001; a: WT n = 26, WT + GDNF n = 25; gdnf^+/− n = 17; gdnf^+/− + GDNF n = 15; gdnf^−/− n = 12; gdnf^−/− + GDNF n = 13; b: WT n = 8, WT + GDNF n = 9; gdnf^+/− n = 10; gdnf^+/− + GDNF n = 9; gdnf^−/− n = 7; gdnf^−/− + GDNF n = 7.

Fig. 4

GDNF effects on nerve fiber formation. Fetal ventral mesencephalic organotypic tissue cultures (ts) at 12 DIV showing TH-positive nerve fibers growing from WT (a, representative of n = 26), WT treated with GDNF (10 ng/ml medium) (b, representative of n = 25), gdnf^−/− (c, representative of n = 14), and gdnf^−/− treated with GDNF (10 ng/ml medium) (d, representative of n = 13). The TH-positive nerve fibers reached shorter distances in gdnf^−/− cultures (c) compared to WT cultures (a). The addition of GDNF (10 ng/ml medium) counteracted the effect of the knockout (d). Scale bar: a–d = 100 μm.

Fig. 5

Quantification of distances for the non-glia-guided and glia-guided TH-positive nerve fiber outgrowth. Measurements were performed from the periphery of the tissue slice to the distal end the nerve fibers had reached. Ventral mesencephalic tissue slices were cultured for 12 DIV. The non-glia-guided TH-positive outgrowth reached significantly longer distances from the tissue slice compared to the glia-guided TH-positive outgrowth in all genotypes. The glia-guided TH-positive outgrowth reached significantly shorter distances from the tissue slice in gdnf^−/− compared to WT (*p < 0.01). Significance comparison of non-glia-guided with the glia-guided TH-positive outgrowth of each genotype. ***p < 0.001, non-glia-guided: WT n = 24, gdnf^+/− n = 14, gdnf^−/− n = 14; glia-guided: WT n = 26, gdnf^+/− n = 22, gdnf^−/− n = 12.

Fig. 6

Astrocytic migration in gdnf^−/− tissue cultures. Fetal ventral mesencephalic tissue was cultured for 12 DIV. S100-immunoreactivity showed astrocytic migration from tissue slice (left in pictures) from WT (a), gdnf^−/− (b), WT treated with GDNF (10 ng/ml medium) (c), and gdnf^−/− treated with GDNF (10 ng/ml medium) (d) cultures. The S100-positive astrocytes migrated over shorter distances in gdnf^−/− cultures (b) compared to WT cultures (a). The addition of GDNF (10 ng/ml medium) counteracted the effect of the deletion (d). Additional GDNF did not exert any effect on the S100-positive migration in WT (c). Panels a–d are representing n's of 26 = WT, 25 = WT + GDNF, 14 = gdnf^−/−, 13 = gdnf^−/− + GDNF. Scale bar: a–d = 200 μm.

Fig. 7

S100-positive astrocytic migration from ventral mesencephalic tissue slice at 12 DIV. The distance was calculated from the periphery of the tissue slice to the distal end the astrocytes had migrated. The migrating S100-positive astrocytes reached significantly shorter distances from the tissue slice in gdnf^−/− compared to gdnf^+/− and WT. The shorter distance in gdnf^−/− was counteracted by the addition of GDNF (10 ng/ml medium). Additional GDNF did not exert any significant effect on the S100-positive migration in gdnf^+/− and WT cultures. **p < 0.01, WT n = 23, WT + GDNF n = 21, gdnf^+/− n = 22, gdnf^+/− + GDNF n = 16, gdnf^−/− n = 7, gdnf^−/− + GDNF n = 8.