Transforming growth factor-beta in the chicken fundal layers: an immunohistochemical study.

In the chicken model of myopia, it has first been shown that imposing defocus to the retina results in active remodelling of the sclera which, in turn, results in axial length changes of the eye. Transforming growth factor-beta (TGF-beta) is one of the scleral growth modulators but its cellular localization in the fundal layers, colocalization and function are not well known. The aim of the current study was to investigate the cellular distribution of the three isoforms TGF-beta1, 2 and 3 by immunohistochemical labelling. Furthermore, the effects of visual experience that induces refractive errors on TGF-beta2 labelling were examined. Transversal cryostat sections of the fundal layers were analyzed by indirect immunofluorescent labelling and cell counts. Visual experience was changed by having the chicks wear either diffusers, or positive or negative lenses of 7D power in front of the right eyes for various periods of time. Left eyes served as uncovered controls. All TGF-beta isoforms were localized in both scleral layers. In choroid, diffuse labelling of all isoforms was found. In retina, TGF-beta1 and 3 were detected in bipolar, amacrine and ganglion cells and TGF-beta2 in amacrine and ganglion cells. To further characterize these cells, double-labelling with known amacrine and bipolar cell markers was performed (calbindin, cellular retinoic acid binding protein (CRABP), Islet1, Lim3 and protein kinase C (PKC)). TGF-beta1, 2 and 3 could be colocalized with calbindin and CRABP in single amacrine cells. TGF-beta1-positive bipolar cells were immunoreactive to Lim3. TGF-beta1 and 3 were never colocalized with PKC in bipolar cells. Also, colocalization with peptides known to be involved in myopia development in chicks, such as glucagon, or vasointestinal polypeptide and the key enzyme for dopamine synthesis, tyrosine hydroxylase, was not observed. Lenses or diffusers, worn by the chicks for various periods of time, had no effect on TGF-beta2 immunoreactivity in choroid or sclera, or on the number of TGF-beta2 (active and latent form) expressing amacrine cells. This result did not change when the two identified populations of TGF-beta2 expressing amacrine cells (one calbindin-positive and the other CRABP-positive) were separately considered. Also no modulation was seen in choroid, although an earlier study had found changes in TGF-beta2 mRNA after lens treatment. The lack of any visually-induced changes in retina or choroid suggests that TGF-beta may not represent a key molecule in the retino-choroidal signalling cascade although it has previously been shown to have a primary role in scleral remodelling. Copyright 2010 Elsevier Ltd. All rights reserved.