Immunolocalization of keratan sulphate in the human embryonic cornea and other human foetal organs

The human cornea consists of three different cellular layers: an anterior epithelium, an intermediate stroma and an endothelial monolayer that lines the posterior surface of the cornea (Hogan, Alvarado & Weddell, 1971). The stroma is the thickest section of the cornea and contains a fibroblastic cell type, the stromal cell, which is derived from the neural crest during embryogenesis (Hay, 1981). The matrix of the stroma consists of highly ordered orthogonal lamellae composed of type I collagen fibrils (Trelsted & Couloumbre, 1971) and of proteoglycans that contain keratan sulphate and chondroitin sulphate (Antonopoulos, Axelsson, Heinegård & Gardell, 1974; Axelsson & Heinegård, 1975; Muthiah, Stuhlsatz & Greiling, 1974).The organization of the collagen and proteoglycans in the stroma as well as the macromolecular characteristics of the proteoglycans are thought to be responsible for the structure and transparency of the cornea. Although the biochemical characteristics of corneal proteoglycans have been extensively studied (Anseth, 1961; Axelsson & Heinegård, 1978, 1980; Cöster, Cintron, Damle & Gregory, 1983; Dahl, Johnsen, Anseth & Prydz, 1974; Dahl & Laurent, 1982; Faltz et al. 1979; Gregory, Cöster & Damle, 1982; Handley & Phelps, 1972; Hart, 1976, 1978; Hart & Lennarz, 1978; Hassel, Newsome & Hascall, 1979; Klintworth & Smith, 1983; Matthews & Cifonelli, 1965; Meier & Hay, 1974; Meyer, Linker, Davidson & Weismann, 1953;

Toole & Trelstad, 1971; Trelstad, Hayashi & Toole, 1974), less is known about the sites of their production and how keratan sulphate is distributed in situ.

The present study was undertaken to examine the distribution of keratan sulphate in the human embryonic cornea as well as in other human foetal organs. For this purpose we took advantage of a recently developed monospecific antibody, which was raised against keratan sulphate (Zanetti, Ratcliffe & Watt, 1985). We report here that this antibody can be used to trace keratan sulphate in human foetal eyes as well as in other parts of the human embryo.

The primary material used in this study was received from 10-to 12-week-old human embryos obtained by a vacuum extraction abortion method (Brody, 1980). No apparently misdeveloped material was used, as judged by morphological examination. The foetal age in each case was estimated according to Shi et al. (1985). All material was processed within 6h after surgery. The foetal specimens delivered in collection vessels were initially diluted with an approximately equal volume of dextrose saline (4% dextrose (w/v) in 0·18% (w/v) aqueous NaCl, pH7·4, purchased from Steriflex plc, Nottingham, U.K.), sieved through a domestic plastic sieve with a hole size of 1 mm×l mm; whereafter excess blood and debris were washed through the sieve with two rinses each of approximately 200 ml of phosphate-buffered saline (PBS) lacking calcium and magnesium ions at pH 7·3 (solution A as described by Dulbecco & Vogt (1954) and obtained from Oxoid Ltd, U.K.). The contents of the sieve were then shaken and rinsed into a 37cm × 27cm white tray containing 700-1000 ml of PBS. Despite several transfers from one vessel to another, eye globes, umbilical cords, placental fragments, yolk sacks and other organs were frequently found.

For histology the organs were fixed between 4 and 6h after foetal aspiration. Routine histology was performed on formol-saline fixed material, which was subsequently embedded in paraffin and cut into 5 μm thick sections. The sections were stained in haematoxylin/eosin and examined and photographed in a Leitz inverted microscope with an attached camera system. For antigen localization foetal organs were snap-frozen in liquid nitrogen and cut with a cryostat into 10 μm thick sections. These sections were attached to Hendley Essex multispot glass slides precoated with 1% (w/v) gelatin and 0·01% (w/v) chrom alum in Analar water, air dried and stored at −70°C until use.

Thoroughly rinsed eye globes were microdissected and the entire cornea was cut out with a pair of corneal scissors to leave the limbus with the remaining eye bulb. After a thorough rinse in PBS the corneal button was placed with the endothelial side down (Hyldahl, Auer & Sundelin, 1982) in a 1 cm² well of a NUNC 24-well plate containing a 1:1 (v/v) mixture of Ham’s F12 and Dulbecco’s Modified Eagle’s Medium (DME) (Morton, 1970) supplemented with 10% foetal calf serum, 50 units of penicillin per ml, 50 μg of streptomycin sulphate per ml and 0·5 μg Fungizone per ml. All wells had been precoated with fibronectin (10μgml⁻¹ purified according to Engvall & Ruoslahti (1977) for 1 h at 4°C). The cultures were then stored in a humidified 5% CO₂/95% air mixture at 37°C for up to 10 days. The primary cultures were either fixed and prepared for antigen localization or removed for transfer by a trypsinization procedure. The cultures were briefly rinsed in prewarmed PBS and treated with cold (4°C) 0·25% trypsin (1:250, w/v; obtained from Difco Co., Detroit Mi, U.S.A.) in PBS until all cells had detached from the wells. The cells were then resuspended in 10% (v/v) foetal calf serum and spun at 200g for 2-3 min. The supernatant was discarded and the pellet resuspended in 10% (v/v) serum. The suspension was then seeded onto gelatinized 50mm tissue culture dishes and split ata routine ratio of 1:5 every 4-7 days. Primary cultures as well as passage 10 cultures were used for antigen localization in this study.

Glass slides with 7-to 10-day-old primary cultures or passage 10 stromal cells (Hyldahl, 1984) were subjected to antigen analysis and prepared by four different means of fixation. First, cells were air dried at room temperature and postfixed in acetone for 15 s at 4°C. Secondly, the cells were fixed in 20% acetic acid/80% ethanol (v/v) for 10 min. Thirdly, the cells were fixed in 96% (v/v) aqueous ethanol/glacial acetic acid, 99:1 (v/v), Engelhardt’s modification of Saint-Marie’s fixative (Saint-Marie, 1962; Engelhardt, Goussev, Shipova & Abelev, 1971). Fourthly, the cells were fixed in 3·7% formaldehyde in PBS (v/v) for 8 min at room temperature followed by incubation in absolute methanol on ice for 5 min (Zanetti et al. 1985).

MZ15 is a monoclonal antibody to keratan sulphate, which recognizes an antigenic determinant that is sensitive to keratanase digestion (Zanetti et al. 1985). It was obtained from a fusion of NS1 myeloma cells with spleen cells of a Balb/c mouse that had been immunized with pig laryngeal chondrocytes. The antibody has wide species cross-reactivity and recognizes both corneal and skeletal keratan sulphate (Zanetti et al. 1985).

Slides with sections of human foetal organs, primary corneal explants or passage 10 stromal cells were rinsed briefly in PBS at room temperature. The slides were subsequently incubated in MZ15 (ascitic fluid diluted at 1/1000 in PBS supplemented with 2% foetal calf serum (v/v) and 0·l% (w/v) sodium azide) in a humid atmosphere at room temperature for 30 min. After a brief rinse in PBS the specimens were incubated in fluorescein isothiocyanate(FITC)-conjugated rabbit antimouse immunoglobulin G (IgG) (Miles, diluted to 1/30) for 30 min at room temperature as above. The reaction was terminated by a rinse in PBS, whereafter the slides were mounted in 30% glycerol/70% PBS (v/v). Slides exposed to second antibody only were used as negative controls. In one case (snap-frozen placenta) the background binding of the se∞nd antibody was too high to draw any conclusions about MZ15 reactivity. It was assumed that this was due to the presence of Fc receptors in placental tissue, and placental sections were therefore preincubated in human plasma diluted 1:50 in PBS (v/v) for 20 min to block the Fc receptors and then briefly rinsed in PBS before MZ15 staining. The slides were examined in a Leitz photomicroscope using epifluorescence illumination. To produce photographs of fluorescent images, Ilford HP5 400 ASA/27 DIN film was used.

Fig. 1 shows histological cross-sections from a 10-week post-fertilization human embryonic eye, from which it can be seen that the cornea is clearly separated from the anterior surface of the lens (Fig. 1A). Fig. 1B shows that the embryonic cornea, like its adult counterpart, consists of an outer epithelium, an intermediate stroma and an inner endothelial monolayer. Thus, it was possible to search for keratan sulphate in all three layers.

Fig. 2 shows the localization of keratan sulphate in the embryonic cornea as revealed by antibody MZ15 staining. The staining was most intense in the endothelium and around Descemet’s membrane, the basal membrane that separates the endothelium from the corneal stroma. The intermediate stroma was also stained with MZ15, albeit at lower intensity. This difference becomes more evident in Fig. 4, which shows a section of Fig. 2 at higher magnification. In contrast, the corneal epithelium was found to be MZ15-negative. Staining with the second FITC-conjugated antibody revealed a weak background staining evenly distributed over the entire cornea (Fig. 3). It has been previously suggested that corneal cells explanted in vitro alter their expression of proteoglycans (Conrad & Dorfman, 1974; Dahl & Cöster, 1978; Gnadinger and Schwager-Hubner, 1975a,b; Klintworth & Smith, 1976, 1981; Yue, Baum & Silbert, 1976). To test this hypothesis we have explanted cells from human embryonic corneas and subjected them to 7-10 days’ incubation in vitro. This interval was chosen because it takes at least 7 days for cells from different corneal layers to detach fully from the corneal bulk and attach firmly to the tissueculture surface (Hyldahl et al. 1982; Hyldahl, 1984, 1985, and unpublished observations). Different fixation procedures were thereafter used, as described in Table 1, but in no case did we detect any MZI5-positive cells. Nor did we reveal any binding of this antibody to human corneal stromal cells that had undergone 10 passages after explantation in vitro (Table 1). This suggests that corneal cells grown in vitro are phenotypically different from their in vivo counterparts — at least with respect to MZ15 staining.

In order to study the expression of keratan sulphate in other parts of the human embryo, different organs from 10-to 12-week-old human embryos were collected, snap-frozen, cut with a cryostat and finally stained with MZ15. The results are summarized in Table 2. It was found that keratan sulphate could not be detected in the umbilical cord or in the foetal yolk sack. Nor did MZ15 bind to foetal intestine, liver, suprarenal glands or to the posterior parts of the human foetal eye (i.e. the inner or outer layers of the optic cup or the foetal lens). In contrast, we found that foetal cartilage bound this antibody in abundant quantities (Fig. 5). We also observed a traceable binding of MZ15 to sections of human placenta (Fig. 6).

This study has taken advantage of the recent development of a monoclonal antibody to keratan sulphate (Zanetti et al. 1985). The antibody MZ15 was used to study the distribution of keratan sulphate in the human embryonic cornea. It was found that MZ15 bound to human corneal stroma, suggesting the presence of keratan sulphate in this tissue. This result is in line with a number of reports showing that keratan sulphate is an important constituent of corneal stroma in a variety of species (Anseth, 1961; Axelsson & Heinegård, 1975; Dahl, Johnson, Anseth & Prydz, 1974; Faltz et al. 1979; Gregory et al. 1982; Hart, 1976, 1978; Klintworth & Smith, 1983; Meier & Hay, 1974; Toole & Trelsted, 1974). However, we found that the binding of MZ15 was significantly more intense to the corneal endothelium and Descemet’s membrane. Even though it has been reported that Descemet’ s membrane from bovine eyes contains keratan sulphate (reviewed by Gospodarowicz & Fuiji, 1981; Hyldahl, 1983) this is the first time that human embryonic corneal endothelial cells have been shown to contain large quantities of keratan sulphate. Sections from other human foetal organs were used as controls and it was found that foetal cartilage stained intensely with MZ15. Placental sections also stained with MZ15, albeit at lower intensity. In contrast, umbilicus and foetal yolk sack were found to be MZ 15negative. Other differentiated organs of the human embryo such as intestine, liver, the suprarenal glands, the inner and outer layers of the optic cup and the foetal lens were also negative with respect to the binding of this antibody.

When cells from the human embryonic cornea were explanted and allowed to proliferate into a cell monolayer, they lost their reactivity with the MZ15 antibody. No binding could be detected irrespective of fixation procedure either in primary or late-passage cultures. This may well represent an adaptation to in vitro conditions, which involves a decrease in the keratan sulphate expression. Even though it has been shown by Dahl et al. (1974) that rabbit stromal cells in culture synthesize small quantities of keratan sulphate, it is significant that we have been unable to trace any MZ 15-positive material in either early or late-passage cultures of cells from human embryonic corneas. A possible but less probable explanation for this apparent inconsistency may be that a putative keratan sulphate molecule produced by stromal cells grown in tissue culture is qualitatively different from the keratan sulphate detected in situ. It then follows that the MZ15 antibody would fail to detect such molecules. This hypothesis would be in line with a recent report by Kenney, Benya, Nimni & Smith (1981), who clearly demonstrated that rabbit corneal endothelial cells dramatically alter their pattern of collagen synthesis after explantation in vitro. It remains to be shown, however, that human embryonic corneal stromal cells grown in culture produce a keratan sulphate molecule that is different from that produced in the human cornea in vivo.

The development of a monoclonal antibody that recognizes keratan sulphate in human corneal tissue has important practical implications. The antibody may be useful in detecting aberrant keratan sulphate distribution patterns in human corneas. Among the disorders in which such a diagnostic tool would be helpful is the macular corneal dystrophy – a recessively inherited disease, which is characterized by a general diffuse haze throughout the corneal stroma and the accumulation of irregularly shaped deposits that tend to be anterior in the central region of the stroma (Maumenee, 1981). Recent studies (Hassel, Newsome, Krachmer & Rodriguez, 1980) have demonstrated that the cloudiness in corneas from patients with macular corneal dystrophy results from the failure to synthesize a keratan sulphate or a mature keratan sulphate proteoglycan. It can thus be envisaged that a monoclonal antibody that recognizes human keratan sulphate in tissue sections will eventually become useful also in medical practice..

The authors thank Professor Christopher F. Graham FRS for valuable support throughout this study. This work was supported by the Cancer Research Campaign. Louise Hyldahl was on sabbatical leave at the University of Oxford supported by Carmen and Bertil Regner’s fund for Ophthalmological Research, Axel Hirsch’s travel fund for surgeons, The International federation for University Women (IFUW), the Swedish Society for Medicine and the research funds of Karolinska Institutet.