A gradient of matrix-bound FGF-2 and perlecan is available to lens epithelial cells

Abstract

Fibroblast growth factors play a key role in regulating lens epithelial cell proliferation and differentiation via an anteroposterior gradient that exists between the aqueous and vitreous humours. FGF-2 is the most important for lens epithelial cell proliferation and differentiation. It has been proposed that the presentation of FGF-2 to the lens epithelial cells involves the lens capsule as a source of matrix-bound FGF-2. Here we used immunogold labelling to measure the matrix-bound FGF-2 gradient on the inner surface of the lens capsule in flat-mounted preparations to visualize the FGF-2 available to lens epithelial cells. We also correlated FGF-2 levels with levels of its matrix-binding partner perlecan, a heparan sulphate proteoglycan (HSPG) and found the levels of both to be highest at the lens equator. These also coincided with increased levels of phosphorylated extracellular signal-regulated kinase 1 and 2 (pERK1/2) in lens epithelial cells that localised to condensed chromosomes of epithelial cells that were Ki-67 positive. The gradient of matrix-bound FGF-2 (anterior pole: 3.7 ± 1.3 particles/μm2; equator: 8.2 ± 1.9 particles/μm2; posterior pole: 4 ± 0.9 particles/μm2) and perlecan (anterior pole: 2.1 ± 0.4 particles/μm2; equator: 5 ± 2 particles/μm2; posterior pole: 1.9 ± 0.7 particles/μm2) available at the inner lens capsule surface was measured for the bovine lens. These data support the anteroposterior gradient hypothesis and provide the first measurement of the gradient for an important morphogen and its HSPG partner, perlecan, at the epithelial cell-lens capsule interface.

Keywords: ERK1/2; FGF-2; lens capsule; perlecan.

Fig. 1

Measurement of FGF-2, collagen IV and perlecan at the epithelial cell – lens capsule interface by immunoelectron microscopy. Bovine lens capsules were flat-mounted and washed in deionised water for one hour to remove epithelial cells. They were then incubated with either rabbit anti-collagen IV polyclonal antibody (Abcam, UK), or rabbit anti-FGF-2 polyclonal antibody (Calbiochem, USA) or mouse anti-perlecan monoclonal antibody (Chemicon International, USA) for 2 h at room temperature, followed by an hour of incubation with the appropriate secondary antibodies (either goat anti-rabbit or goat anti-mouse IgG both conjugated with 10 nm gold; BBInternational, Cardiff, UK). Negative controls were incubated with PBS instead of primary antibodies. Immunogold labelled capsules were viewed in a Hitachi SU-70 FEG scanning electron microscope (Hitachi Hight-Technologies Europe GmbH, Germany). Each labelling was independently repeated three times. A. An example of a lens capsule prepared for scanning electron microscopy. Three images were then randomly taken at the equator and the anterior and posterior poles (black squares). B. The secondary electron image shows the capsular surface and its compacted meshwork. C. A backscatter electron image of the same area as in panel B shows a large number of white gold particles that indicate collagen IV labelling. D is the combined secondary/backscatter image of panels B and C to show the location of gold particles on the lens capsule. E–M: Representative images show the immunogold labelling of collagen IV (E–G), FGF-2 (H–J) and perlecan (K–M) in different regions of the inner surface of the bovine lens capsule. More gold particles were present after collagen IV labelling than after FGF-2 and perlecan labelling (white arrows). N–P: The quantification of gold particles detected on lens capsules after Collagen IV, FGF-2 and perlecan labelling. Significant differences tested by Student's t-Test are shown (*). B–M. Scale bars = 100 nm.

Fig. 2

Regional and subcellular distribution of pERK1/2 in the bovine lens epithelium. A. A diagram to show the dissection of a lens epithelium for immunoblotting. B. Representative immunoblotting bands of lens epithelial cells in the CZ and GZ + TZ assayed for total ERK1/2, phosphorylated ERK1/2 and actin. Total cell proteins in the CZ and GZ + TZ were extracted and 10 μg was analysed by immunoblotting with mouse anti-ERK1/2 monoclonal antibody, rabbit anti-phospho-ERK1/2 monoclonal antibody (Thr202/Tyr204) (both from cell signalling, USA), and mouse anti-β-actin antibody (MP Biolmedicals, LLC., USA). C. Quantification results of the immunoblotting signals. The ERK1/2 and pERK1/2 signals were quantified and the relative band density was obtained using the formula based on previous reports (Wang et al., 2010; Wang et al., 2009): (D_ERK1/2/D_β-actin)_GZ+TZ/(D_ERK1/2/D_β-actin)_CZ and (D_pERK1/2/D_β-actin)_GZ+TZ/(D_pERK1/2/D_β-actin)_CZ, where D represents the band density. D_ERK1/2/D_β-actin adjusts the density of each band against standard band β-actin. This experiment was independently repeated three times. By the two tailed Student's t-test, the difference between pERK1/2 levels in the CZ and GZ+TZ was 0.051. D–G. A representative mitotic cell labelled with pERK1/2. The flat-mounted bovine lens epithelium was stained with pERK1/2, Ki-67 (Dako, Denmark) and DAPI. Images with pERK1/2-positive cells were taken using a Zeiss LSM 510 Meta scanning confocal microscope (Carl Zeiss Inc., Jena, Germany). The DAPI (D) and Ki-67 (E) labelled condensed nuclear chromosomes suggest that this cell is in M phase. Punctate phosphorylated ERK1/2 staining is distributed along the chromosomes (F). Scale bar = 100 μm.