Objective: To explore the mechanism of optic nerve damage in glaucoma by study on structure of glial lamina cribrosa(LC) in rats.
Methods: Experimental study. Albino Swiss(AS) rats were divided into 3 groups. Bilateral eyes of 10 normal rats were employed to be group I (right eye ) and group II (left eye) . Group III was from the left eyes of 13 rats underwent artificially intraocular hypertension in the right eyes. All rats were perfused and fixed with electronic microscopy fixative (2% paraformaldehyde +2% glutaraldehyde). Trimmed optic nerves were embedded with resin. Serial 1.5 µm thick 'semithin' sections were cut, either (2 eyes from group III) longitudinally, through the optic nerve head (ONH) from the retinal end to the commencement of the optic nerve, or (31 eyes) transversely (cross-sections). Ultrathin sections were cut in the middle of glial LC. The morphological observation of glial LC was obtained by light microscopy and transmission electron microscopy. Bonferroni correction was used to counteract the multiple comparison of each group.
Results: Fortified astrocytes formed the main supportive structure of glial LC in all rats, including group I, group II and group III. Astrocytes were ranked as a fan-like radial array, firmly attached ventrally to the sheath of the LC by thick basal processes, but dividing dorsally into progressively more slender processes with only delicate attachments to the sheath. These fortified astrocytes form ventral stout basal end feet, radial array, axon free-'preterminal' layer before terminating in a complex layer of fine interdigitating delicate branches at the dorsal. LC astrocytes were highly and uniformly electron dense throughout all the cell processes. An equally striking feature of the astrocytic processes was their massive cytoskeletal 'strengthening' of longitudinal massed filaments and tubules. Especially, massive filaments accumulated as cytoskeletal cores to form 'scaffold' of fortified astrocytes. There was vulnerable area in the dorsal of glial LC. This vulnerable area was isomerisation in bilateral eyes and different rats. There was different space in the vulnerable area. These space could be divided into 3 grades, (-), (+) and (++) . The number of (-), (+) and (++)were 1, 6, 3 eyes in group I, 1, 5, 4 eyes in group II, 1, 7, 3 eyes in group III. The Kruskal-Wallis test was used for statistical evaluations. There was no statistical differences of the ratio of (-), (+) and (++) in group I, group II and group III(χ(2) = 3.35, P = 0.187>0.05;group I vs group II, Z = -1.048, P = 0.294;group I vs group III Z = -1.691, P = 0.091;group II vs group III,Z = -1.343, P = 0.179). The ratio of space (-)was significantly less than space (+) and space (++) in group I, group II and group III(χ(2) = 23.88, P < 0.05; (-) vs (+) , Z = -2.821, P = 0.005; (-) vs (++) , Z = -2.726, P = 0.006). The ratio of space (+)was much more than space (++) in group I, group II and group III(Z = -4.410, P < 0.05).
Conclusion: Glial isomerisation in LC may play a key role in glaucomatous optic nerve damage.