Fertile white hybrid hens eggs were incubated at 38˚C in a humidified atmosphere to obtain chick embryos at different developmental stages, ranging from 13 to 25 H.H. [16] , in order to study the expression pattern of HSPG-Perlecan. A total of 20 embryos were incubated up to stage 17 H.H. (corresponding to that of the lens vesicle, prior to differentiation of lens fibres), at which a single dose of 24 μl b-D-xyloside (a sulphated proteoglycans synthesis disruptor) 4 mM (Sigma) was injected subgerminally with a Hamilton microsyringe [12] , and the eggs were reincubated for 24 hours until stage 23 H.H. (corresponding to the development of lens fibres). Controls (15 embryos) used in the experiment were treated with a sterile saline solution with the same procedure. All the animal handling and procedures were supervised by the research and welfare service from the Valladolid University.

A total of 30 embryos were treated with heparinase as follows: once embryos reached stage 17 H.H., a small opening was made in the shell to expose the right eye of the embryo, after which the vitelline membrane was cut with tungsten needle and 12 nl of a heparinase II solution from Flavobacterium heparinum (Sigma) (10 U of enzyme dissolved in 25 μl of PBS), was microinjected into the subectodermal space in the area of the lens equator. As controls, we microinjected 15 embryos with the same amount of heat-inactivated enzyme as for the enzyme treated. Microinjection was performed with a microinjector (Medical Systems Corp, Greenvale, NY 11548, U.S.A, PLI-100). Following this, the eggs were sealed and re-incubated for a period of 18 hours until stage 21 H.H.

In both cases (β-D-xyloside and heparinase II treatment) a single dose of 200 nl of BrdU (0.3% solution) was microinjected in the outflow of the heart just 1 hour before the end of the reincubation period. The embryos were then removed from the extraembryonic membranes and fixed for 1 hr in Carnoy’s.

Histological sections (8 μ) from five different lens anlagen at each stage were washed in phosphate-buffered saline (PBS), pre-incubated with normal horse serum (1/20 in PBS) and incubated overnight at room temperature with a 1:1000 dilution of Anti Heparan Sulphate-Perlecan monoclonal antibody (Upstate Biotechnology). After two washings in PBS, the sections were reincubated for 30 minutes in fluorescein-conjugated goat anti-rat Ig-G FITC conjugate (Sigma) as a secondary antibody, mounted in Aquamount (Gurr) and examined under a Leica SPE laser confocal microscope. Control sections were prepared as described above but pre-immune serum was used as the primary antibody and no labelling was observed.

In others cases, lens anlagen histological sections (8 μm) were deparaffinised and BrdU was detected following standard procedures [17] . The sections were incubated in a solution containing a 1/100 dilution of monoclonal antibody to BrdU (Dako) for 30 minutes at room temperature. An avidin-extravidin system conjugated to peroxidase (mouse anti-rabbit 1/20 for 30 minutes and extravidin 1/20 for 10 minutes; Sigma) was used to detect the primary antibody and was developed with DAB. A Nikon microphot-FXA photomicroscope was employed to visualize and photograph the preparations.

In other cases, lens anlagen frozen sections (5 μm) were used to detect FGF2 following standard procedures. The sections were incubated in a solution containing a 1/1000 dilution of monoclonal antibody to FGF2 (anti-bovine FGF2 antibody from Sigma) for 30 minutes at room temperature. An ExtrAvidin stainig Kit antimouse system conjugated to peroxidase (from Sigma) was used to detect the primary antibody and was developed with DAB. A Nikon microphot-FXA photomicroscope was employed to visualize and photograph the preparations.

Finally, lens anlagen histological sections (8 µm) were deparaffinised, dehydrated and stained with Haematoxilin-Eosin technique following standard procedures. We used a Nikon microphot-FXA photomicroscope to visualize and photograph the preparations.

In this study we have used a monoclonal antibody which recognizes Perlecan, a type of Heparan Sulphate Proteoglycan present in basement membranes [18] [19] [20] . As we show in Figure 1, the antibody specifically labels basement membranes in the eye anlagen, including the surface ectoderm, lens and optic cup, at different stages of development.

At stage14 H.H., we show (Figure 1(a)) intense immunolabelling in the basement membrane of both the lens placode (similar to surface ectoderm basement membrane labelling) and the optic cup, although in some case both basement membranes seem to be in direct contact. During lens placode invagination (stages 15 - 16 H.H.) (Figure 1(b) and Figure 1(c)) there is still intense immunolabelling on the basement surface of the lens epithelium, however during this period intensity on the basement surface of the optic cup decreases. At these stages it is possible to observe several areas in which both basement membranes (lens and optic cup) are in close contact. A similar disposition in terms of Perlecan immunomarking can be seen at stages 16 - 17 H.H. (Figure 1(d)); furthermore, at these stages it can be observed how basement membrane immunolabelling in the lens anterior epithelium is restored as the final step in lens isolation from the ectodermal surface.

From stages 18 to 25 H.H. (Figure 1(e)-(h)), the lens capsule is the structure with the most intense immunolabelling in the eye anlage, appearing as a strong, continuous and apparently uniform layer on the basement side of the lens. At the beginning of lens fibre differentiation we cannot appreciate any labelling in the anterior and posterior epithelium cells.

As can be observed in Figure 1(e) and Figure 1(f), at these stages of eye development (E18 - 20 H.H.) immunolabelling on the surface ectoderm and neural retina decrease considerably; however, in some sections a point of contact can be perceived between the neural retina at the margin of the optic cup and the basement membrane of the lens equator (Figure 1(h)), both labelled by the Perlecan antibody. Finally, we detect the “pecten” (Figure 1(g)) as another structure showing strong immunolabelling during chick eye development.

With the aim to test the role of Heparan Sulphate-Perlecan proteoglycan on lens development, especially in the differentiation of lens fibres from epithelial cells, we carried out two experimental approaches: first, we used β-Dxyloside, a xylosiltransferase substrate which disrupts the assembly of sulphated proteoglycans including different forms of HSPG [21] ; secondly, we used Heparinase

type II, which specifically digest Heparan Sulphate including Perlecan [22] . Both treatments were administered at stage 17 H.H., which corresponds with the lens vesicle stage, before initial lens fibre differentiation, and were maintained until the 23 H.H., when the histological structure of the lens anlagen (Figure 2(a)) shows a polarized lens with a thin anterior epithelium, a transitional equatorial (marginal) area and a thick posterior epithelium displaying the elongated primary fibres with a central nucleus, closing the inner cavity.

As seen in Figure 2(d), BrdU immunolabelling showed the nucleus with DNA synthesis activity (which at these stages of development could be assimilated to mitotic activity) in the anterior epithelium, in which the homogeneous presence of a layer of BrdU-positive nucleus reaching the equatorial area could be observed; however, in the posterior epithelium, there was a total absence of BrdU-positive nucleus. This mitotic pattern was consistent with the dynamic cellular behaviour described during lens development [23] with cellular replication in the anterior epithelium, migration through the equatorial zone and differentiation in the posterior epithelium. As we described before, the eyes of stage 23 H.H. controls showed intense immunolabelling with the anti-perlecan antibody in the lens capsule (Figure 1(f)). Finally, lens capsule immunodetection of FGF2 show the absence of labelling in the anterior epithelium and a ubiquitous label in the lens fibre which was more intense in the posterior part of the lens capsule (Figure 2(j)).

All the β-Dxyloside treated embryos showed a severe, specific disruption in lens anlagen development. The histological study shows the lens anlagen with an undifferentiated posterior epithelium with no lens fibre growth, resembling the lens vesicle stage (Figure 2(b)). In other cases (Figure 2(c)), some areas of the posterior epithelium show a partial lens fibre growth while others remain undifferentiated and the lens vesicle cavity remains uncollapsed. In the β-Dxyloside treated embryos, BrdU immunodetection (Figure 1(e)) shows a clear disruption in the mitotic behaviour of the lens anlagen cells, with an increased number of positive nuclei in the anterior epithelium which appear disposed in several layers and a significant number of positive nuclei in the posterior epithelium which seems to remain in an undifferentiated state. In order to test the efficiency of β-Dxyloside treatment in HSPG-Perlecan lens capsule disruption, we studied the expression of this proteoglycan by immunohistohemistry. Figure 2(g) shows how, in the β-Dxyloside treated lens, there was a large decrease in Perlecan immunolabelling compared with the controls (Figure 2(f)), displaying a non-homogeneous distribution and extensive zones with no Perlecan expression, mainly located in the posterior epithelium. Finally the FGF2 lens anlagen immunolabelling was dramatically reduced in the fibre cells and posterior part of the lens capsule in β-Dxyloside treated lens (Figure 2(k)) with respect to that of the control embryos (Figure 2(h)).

In order to confirm the specificity of our results, we made a selective enzymatic digestion of HSPG (Perlecan) in the developing lens with Heparinase II microinjection in the lens anlagen equator. Heparinase II treated embryos (Figure 3(b)) show a hystolological appearance of the lens anlagen with the same alterations as in the β-D-Xyloside treated embryos, namely, a reduced thickness in the posterior epithelium, with a dramatic decrease in the length of the lens fibre cytoplasm (with the lens vesicle cavity remaining completely uncollapsed) compared with the control (Figure 3(a)); occasionally the Heparinase II treated embryos presented a completely undifferentiated posterior epithelium with no signs of lens fibre cytoplasmic growth (Figure 3(c)). Compared with the control

embryos (Figure 3(d)), BrdU immunodetection in the Heparinase II treated lens showed (Figure 3(e)) a greater number of BrdU positive nucleus in the anterior epithelium arranged in several layers and the abnormal presence of BrdU positive nucleus in the undifferentiated posterior epithelium. In order to test the effectiveness of Heparinase II enzymatic activity on HSPG-Perlecan in the chick embryo lens anlagen, we made an immunohistochemical study, which shows that the enzyme treated lens (Figure 3(g)) had greatly reduced HSPG-Perlecan in the lens capsule compared with the control embryos (Figure 3(f)), showing a non homogeneous labelling pattern with several areas of no expression of Perlecan, almost in the posterior capsule. Finally, as is showed in Figure 3(h), a dramatic decrease in FGF2 expression was detected, both in the lens fibres and in the posterior lens capsule, in the Heparinase II treated lens in comparison with the control embryos (Figure 3(i)).