DEVELOPMENTAL DYNAMICS 206:41-26 (1996) FGF Receptor-1 (fZg)Expression Is Correlated With Fibre Differentiation During Rat Lens Morphogenesis and Growth R.U. DE IONGH, F.J. LOVICU, A. HANNEKEN, A. BAIRD, AND J.W. MCAVOY Department of Anatomy and Histology and The Sydney Institute for Biomedical Research, The University of Sydney, NSW 2006, Australia (R.U.d.I., F.J.L., J.W.M.); and Department of Cell Biology, The Scripps Institute, La Jolla, California 92037 (A.H., A B . ) Our previous studies indicate an ABSTRACT important role for fibroblast growth factor (FGF) in lens development. Here we study the expression of the flg variant of FGF receptor 1(FGFR1) during lens development by immunohistochemistry and in situ hybridisation. FGFRl was expressed throughout lens development. Prominent FGFRl immunoreactivity was associated with cell nuclei, particularly in differentiating lens fibres, suggesting internalisation and nuclear translocation of the receptor. FGFRl immunoreactivity was also associated with basolateral membranes of cells in the equatorial region and at lens sutures. FGFRl mRNA was only weakly expressed during early lens morphogenesis but expression increased with the onset of lens fibre differentiation. Once the lens acquired its distinct polarity, an anteroposterior gradient in both protein reactivity and mRNA s i g nal was evident. Anteriorly, central epithelial cells showed weak expression for FGFR1, whereas more posteriorly, in the germinative and transitional zones of the lens where cells maximally proliferate and undergo early stages of fibre differentiation, respectively, expression was significantly stronger. The anteroposterior gradient of increased expression of FGFRl in the lens coincides with the previously documented anteroposterior gradient of FGF stimulation. In lens epithelial explants, FGF stimulation was found to upregulate FGFRl expression. Such upregulation may be an important mechanism for generating a high level of FGF stimulation and ensuring a fibre differentiation response. In postnatal rat lenses, there was a significant age-relateddecline in FGFRl expression; this correlates with the reduced rate of lens fibre differentiation with age. Overall, these studies support the hypothesis that FGF and FGFRl are important for regulation of lens fibre differentiation throughout lens development. ferentiates into a spheroidal, transparent lens which has a highly ordered and polarised cytoarchitecture. The differentiated lens is composed of two distinct cell types, contained within a thick basement membrane, the lens capsule; elongated fibres, arranged in an anteroposterior axis, make up the bulk of the lens while a monolayer of cuboidal epithelial cells covers the anterior surface of the fibres (Fig. 1).Within the lens there are distinct spatial patterns of proliferation and differentiation which are established during morphogenesis and are maintained throughout life (Hanna and O’Brien, 1961; McAvoy, 1978a,b). Epithelial cells in the germinative region of the lens, just anterior to the lens equator, undergo proliferation and the progeny of these divisions migrate, or are displaced, posteriorly into the transitional zone of the lens where they elongate into fibres (see Fig. 1). In vitro studies have established that two members of the fibroblast growth factor (FGF) family (FGF-1 and FGF-2) stimulate rat lens epithelial cells to proliferate, migrate and undergo the structural (Lovicu and McAvoy, 1989,1992) and molecular (Chamberlain and McAvoy, 1989; Peek et al., 1992) changes characteristic of fibre differentiation. Moreover, these responses occurred in a progressive dose-dependent manner; cell proliferation, migration and fibre differentiation were induced at low, medium and high concentrations of FGF, respectively (McAvoy and Chamberlain, 1989). As the progression of these FGF-induced responses was similar to the anteroposterior pattern of cell behaviour in the intact lens, it was hypothesised that an anteroposterior gradient of FGF stimulation may influence lens polarity and growth patterns (McAvoy and Chamberlain, 1989; Schulz et al., 1993). Recent studies of FGF distribution in the eye have provided evidence for anteroposterior variations in the distribution and activity of FGF in the lens and ocular media. Cytoplasmic immunoreactivity for FGF-1 was found predominantly in equatorial regions of the de- 0 1996 Wiley-Liss, Inc. Key words: Lens development, FGF, FGF receptor, Fibre differentiation INTRODUCTION During vertebrate lens development, a sheet of competent ectodermal cells thickens, invaginates and dif0 1996 WILEY-LISS. INC. Received October 30,1995; accepted February 9, 1996. Address reprint reauestslcorresDondence to Dr. J. W. McAvov. Department of Anatomy and Histology, University of Sydney, Sydney NSW Australia 2006. F.J. Lovicu’s present address is Department of Cell Biology, Baylor College of Medicine, Houston, TX 77030-3498. FGFRl EXPRESSION DURING LENS DEVELOPMENT 413 FGFRl cDNA M6 R803 Fig. 2. Schematic diagram of the structure of FGFRl (adapted from Johnson and Williams, 1993). The structural features of the extracellular portion of FGFRl include an amino-terminal signal sequence (ss), three immunoglobulin-like domains (I, II, Ill), and an acidic box domain (a). Following a single transmembrane domain (tm), the cytoplasmic portion of the receptor contains a split tyrosine kinase domain (tk) and a carboxyterminal tail. The regions of the receptor recognised by the anti-FGFRI antibodies (designated R803, M15 and M6), used for immunohistochemistry, and the region encoded by the cDNA, used to prepare riboprobes, are indicated. Fig. 1. Midsagittal section of a 3 day-postnatal (P3)lens. Guboidal lens epithelial cells cover the anterior surface of the highly elongated fibre cells (fib). The six regions (I-VI) used for image analysis in this study are indicated: region I,incorporating the anterior epithelium; II, incorporating the germinative zone; 111, incorporating the transitional zone; IV, incorporating inner cortical fibres; V, incorporating mature fibres in the deeper cortex, and VI, incorporating the most mature fibres. aq, aqueous; c, cornea; cb, ciliary body; i, iris; vit, vitreous. Scale bar: 100 pm. veloping lens where cells actively proliferate and differentiate (de Iongh and McAvoy, 1992, 1993; Lovicu and McAvoy, 1993). Differential immunoreactivity for FGF-2 was found in the lens capsule with the posterior region staining more intensely than the anterior (de Iongh and McAvoy, 1993; Lovicu and McAvoy, 1993). Consistent with this, posterior but not anterior capsule was found to have FGF-associated biological activity (Schulz et al., 1993). Furthermore, analysis of the ocular media showed that more FGF was recovered from vitreous than aqueous (the former bathes the posterior part of the lens and the latter the anterior part of the lens; see Fig. 1)and only vitreous had fibre-differentiating activity. Fractionation of vitreous showed that virtually all of this activity was associated with FGF. These results are consistent with the hypothesis that anteroposterior differences in levels of FGF stimulation influence lens polarity and growth patterns. Recent support for this hypothesis comes from studies of transgenic mice which overexpressed a secreted form of FGF-1 in the lens (Robinson et al., 1995).In these mice, anterior epithelial cells differentiated into fibres resulting in a loss of lens polarity. A requirement for cellular responsivness t o FGF is the expression of appropriate receptors. FGFs mediate cellular responses by binding and activating specific cell surface receptor tyrosine kinases (Johnson and Williams, 1993). Four distinct but highly homologous FGF receptor genes (FGFR1-4) have been identified and within these four genes alternative splicing gives rise to numerous subtypes. Common features of these receptors include an extracellular region composed of two or three immunoglobulin-like domains, a single transmembrane domain and a cytoplasmic region containing a split tyrosine kinase domain (Fig. 2). High affinity binding sites for FGF have been described in lens epithelial cells (Blanquet et al., 1989) and several studies have shown FGFR to be expressed in the lens of the mouse (Orr-Urtreger et al., 1991,1993; Marcelle et al., 1994; Peters et al., 1992, 1993), chicken (Heuer et al., 1990; Ohuchi et al., 1994) and rat (Wanaka et al., 1991) a t various stages of development. However, to date, no detailed study has been carried out on the temporal and spatial distribution of these receptors in the developing mammalian lens. In light of the distinct patterns of cell behaviour and FGF distribution in the lens, it was important to identify the expression patterns of FGF receptors. To elucidate further the role of FGFs during lens development, we examined the spatial and temporal expression of FGFRl (fig) during morphogenesis, differentiation and growth of the rat lens. At all ages studied, correlation of the patterns of lens cell behaviour and FGFRl expression suggest that FGFRl is linked with fibre differentiation, further implicating FGF as an important regulator of lens polarity and growth patterns throughout life. 414 de IONGH ET AL. CORNEAL EXPRESSION OF TYPES XI1 AND XIV COLLAGEN RESULTS synthesis of corneal-specific components of the secondImmunolocalisation of FGFRl ary stroma and stromal compaction. These data are During Development presented in Figure 2A. In thetypes epithelial cells, we found that the were level obof Two of FGFRl immunoreactivity is highest dayspepof mRNA for type XIV collagen tained with our panel of antibodies. Usingat the9 two development. It then a lower tide antibodies, M15decreases and R803and (seeremains Fig. 2), aat predomlevel at all the subsequent inantly punctate reactivitydevelopmental was found in stages all lenstested cells. (Fig. 2A). Counterstaining with Hoechst dye (data not shown) In contrast, stromal cells show a different al(X1V) of this reactivity was intranuclear. showed that most 9 to 13, thedomain stromal collagen mRNA profile. From In contrast, an antibody to thedays extracellular of concentration normalized to G3PDH remains relaFGFRl, M6, revealed predominantly membrane-assotively At day 9, this is approximately equal to ciatedconstant. reactivity. that of the epithelium. However, since the epithelial Nuclear immunoreactivity for FGFR1. Antibod11 days, the stromal values subsequently decrease ies R803 and M15 had similarat distribution patterns of value a t this timepoint is three times higher. After day reactivity in the developing lens at all stages studied. 13, stromal cells exhibit a sharp in their content of Initial experiments, carried outrise using antibody R803 type mRNA.and By paraformaldehyde-fixed day 17, a stromal cell and XIV M15 collagen on methanolcontains approximately more type XIV collagen frozen sections, showed 10-fold strong specific immunoreactivmRNA than an epithelial cell. ity associated with cell nuclei at all stages of development. Weak cytoplasmic or membrane-associated reactivity XI1 wasCollagen also observed at some stages. In general, Type paraformaldehyde fixation favoured the cytoplasmic Immunohistochemistry. antibody reactivity, but also resultedA inmonoclonal greater nonspecific against a region of theFor amino-terminal domain background staining. this reason NC3 all the data common both study the short andobtained long form splicemethanolvariants were using shown into this 5 dayshave of developof typefrozen XI1 collagen was used. Atstudies fixed sections. Previous shown ment, the antibody indicates that type XI1 that, in the case of FGF localisation, differentcollagen, antibodlike collagen XIV,regimes appears can to bemarkedly distributed throughout ies and fixation influence imthe primary stromapatterns (Fig. 3A).(Hanneken By 7 days, after primunolocalisation and the Baird, mary 1992).stroma has begun to swell, we observe immunoreactivity for type XI1 collagen largely concentrated in During early lens morphogenesis (embryonic day 11; the subepithelial and subendothelial regions of the corFig. 3A), the optic vesicle which is the precursor of the nea (Fig. 3B). The immunoreactivity for retina comes into subendothelial contact with the overlying presumpis At present throughout all stages obtype tive XI1 lenscollagen ectoderm. this stage, punctate reactivity served (Fig.was 3A-E). immunoreactivity is present for FGFRl foundThis ubiquitously in ectoderm (includseveral days before the onset of deposition of Desceming lens ectoderm), optic vesicle and the surrounding et’s membrane a t mesenchyme about 9 days of development, and is 3B). Additional undifferentiated (Fig. probably to the interface Descemet’s treatmentconfined of sections with Hoechstbetween dye confirmed that membrane and the stroma Discussion). the the punctate reactivity was(see associated with Under cell nuclei corneal epithelium, the matrix undergoes considerable (data not shown). The lens placode and optic vesicle modification its contentonofembryonic type XI1 collagen during day 12 (El21 to subsequentlyofinvaginate development. detect(Fig. form the opticInitially, cup andimmunohistochemically the lens pit, respectively able XI1 collagen is progressively from this 3C). type F'unctate nuclear FGFRl reactivitylost was seen in that by 10 days it is virtually being region, the lenssopit, optic cup, ectoderm and theabsent, surrounding comparable that 3D). givenBybyEthe negative control antimesenchymeto(Fig. l 4 the lens pit has closed body (Fig. 3C,H). Procedures to expose a vesicle (Fig. 3E). Elongation and potentially differentiato form masked epitopes (seelens Experimental Procedures) didleads not tion of the posterior cells into primary fibres reveal any masked immunoreactivity in this orstrucany to the lens vesicle assuming the characteristic other ocular region (dataisnot shown). By 12 days, howture of the lens which maintained throughout life; ever, subepithelial for type XI1with colof the lensimmunoreactivity is composed of primary fibres the bulk lagen returns to this zone (Fig. 3D). This correlates a monolayer of cuboidal epithelium covering their antemporally withAtthe onset of deposition of the mature terior surface. this stage, strong punctate nuclear Bowman’swas membrane (Fitch et al., 1994).cells, particureactivity found in all lens vesicle Finally, the region theprimary corneoscleral junction larly a t thein equator and inofthe lens fibres (Fig. (ocular limbus), the NC3 specific type XI1 antibody re3F). acts withfurther anterior and posterior the “scleral of the With development and(termed maturation spur”) in a temporospatial that reactivity is identiof FGFRl lens ansites anteroposterior patternpattern cal to that described (Sugrue, 1991) (see Disbecame evident withpreviously E20, postnatal and adult lenses cussion). 51 showing similar patterns of reactivity (Figs. 4A,C,D mRNA and 5). In anterior epithelial cells the nuclei showed 4Efor andthe 5). diffuse and fine punctate In isolated epithelia, thereactivities normalized(Figs. values At the oflens the germinative and mRNA bothequator, forms ofincluding al(XI1) collagen is extremely transitional there was diffuse cytoplasmic reac11 days of development (Fig. 2B and C, low a t 9 and zones, tivity as well as fine punctate nuclear reactivity. Comsolid line). This is consistent with the reduced level of mencing in immunof the transitional zone continuing subepithelial luorescence seenand a t day 10. The through theshort maturing cortical fibre cells, the nuclear rises sharply, to peak at value for to the form then reactivity became progressively more intense day 15 at a value about 20-fold higher than that at(Figs. day 4A,ItC,then D, G andto 5). Control sections with about half this levelincubated by day 17. In 11. falls either adsorbed immune serumof(Fig. 4H),form nonimmune contrast, the epithelial content the long mRNA IgG or a4%gradual bovineincrease serum albumin (BSA,todata not exhibits after 11 days, a level or no reactivity. shown) littleshort form a t 17 days. equal to showed that of the Membrane associated immunoreactivity for The isolated stromal tissue includes the adherent enFGFRl. Reactivity with M6from appeared to be predomidothelial cell layer, which, our immunofluoresnantlyresults, associated with source membrane and/or adjacent cence is the likely of mRNA for collagen cytoplasm lens cells. Atrecently embryonic stages, weak type XII. Inofsupport of this, in situ hybridizareactivity was observed in epithelial and early differtion has revealed that rabbit corneal endothelial cells, entiating fibre cells (data not shown). In postnatal rats, but not stromal fibroblasts, synthesize al(XI1) collagen reactivity wasetstronger and At more defined. For of example, 11days develmRNA (Zhan al., 1995). 9 and lens were reactive epithelialthecells of the P3 opment, normalized values for theweakly short form of along their basal surfaces, closely apposed to the lens type XI1 mRNA (Fig. 2B, dashed line) are vanishingly 6A). "his reactivity was is strongest a t the capsulewhile (Fig. that substantially small, for the long form equatorial Reactivity was also apparent t the greater at 9region. days, but becomes negligible at 11adays. apical11 surface of cells (epithelial-fibre howdays, both forms of the moleculejunction); are expressed After this waslevels shown to be nonspecific normal aever, t significant with mRNA for the when long form of mouse IgG was substituted for the primary antibody the molecule becoming relatively more abundant over 6B). 2B Strong specific reactivity wasthat ob(Fig.(Fig. time and C, dashedfibrillar lines). This suggests and posterior served a t the anterior the subendothelial type(Fig. XI16C) collagen after 11(Fig. days6D) is lens sutures where the endslong of opposing fibres meetwith and probably a mixture of both and short forms, overlap. mouse IgG wastime substituted for the ratio When of longnormal form increasing with of developM6, no specific labelling at lens sutures was observed ment. (Fig. 6E). DISCUSSION Expression of FGFRl mRNAis initiated in the emAvian corneal development During Development bryo a t about 3.5 days with the deposition of an epithe10 lially derived acellular primary stroma that is about During early lens development (E10-E12), FGFRl Fm thick (for reviews ubiquitously see Hay and in Revel, 1969; Hay, mRNA was expressed the optic primor1980). By 7). 5 days development, periocular that dia (Fig. Priorof to commencement of lenscells morphoof migrated centripedally along the posterior surface genesis (ElO), uniformly weak signals for FGFRl tranthe stroma a continuous layer. scripts were form detected in ectodermendothelial (arrowheads, Fig. Shortly thereafter, cells invade the 7A), optic vesicle andmesenchymal in the undifferentiated mesenstroma; this coincides with stromal swelling and furchyme between the optic vesicle and the ectoderm. In ther expansion continues to about day 12. earlymatrix E l l embryos, thethat signal in the neuroepithelium The invading fibroblasts secrete a set of fibrillar of the optic vesicle and diencephalon, relative to collaother V (Linsenmayer gens, composed of to collagen types I and tissues, appeared have increased, particularly in the et al., 1985) that differs thoseFig. deposited earlier by 7D). Transcripts basal ependymal layer from (arrows, the I and I1 (Linsenmayer et Fig. al., wereepithelium, still foundcollagens in the lens ectoderm (arrowhead, 1990). Theundifferentiated complement of mesenchyme. fibril-associated 7D) and At collagens E12, the also changes withfordevelopment, IX, for strongest signal FGFRl was type detected in example, the early being associated specifically with the primary lens vesicle. Elongating cells in the posterior stroma vesicle (Fitch al., 1988b), while and XIV, as andcollagens anteriorXI1 epithelial cells (open et arrow, Fig. 7G) shown appear in distinct in the secondshowedhere, strong signals. Signalslocations for FGFRl expression ary stroma. The fibrils that are deposited into were also evident in the anterior margins of the the swoloptic len immature secondary stroma, which has not yet cup (arrowheads, Fig. 7G) and the optic stalk (arrows, compacted, form orthogonally orientedundifferentiated bundles, sepaFig. 7G), whereas the surrounding rated by largeshowed spaces. weak signal for FGFRl tranmesenchyme By day the secondary stroma compacts to about No17, hybridisation signal above background was scripts. 50% of its thickness a t day 12, and has become com- Fig. 3. lmmunolocalisation of FGFRl during early lens morphogenesis. A, C, E Haematoxylinand phloxine stained sections of El 1, E l 2 and E l 4 embryos, respectively. 6: At E l l , strong punctate reactivity was detected in presumptive lens ectoderm, the optic vesicle and the surrounding mesenchyme.This reactivity appeared to be associated primarily with cell nuclei. D: At E12, similar punctate nuclear reactivity was detected in the lens pit, optic cup, ectoderm and to a lesser extent in the extraocular mesenchyme. F: At E14, punctate reactivity was detected in nuclei of all lens vesicle cells, and was particularly strong in cells at the equator and in the primary lens fibres. Cells of the optic cup, ectoderm and surrounding extraocular mesenchyme were also labelled. e, ectoderm; lv, lens vesicle; Ip, lens pit; m, mesenchyme; oc, optic cup; ov, optic vesicle; plf, primary lens fibres. Scale bars: (A,B), 50 pm; (C, D), 60 pm; (E, F), 100 pm. 416 de IONGH ET AL. detected when analogous sections at each stage were hybridised with the control sense probe (Fig. 7C, F, I). At later stages of ocular development, FGFRl expression appeared to be more intense and tissue-specific (Fig. 8). Consistent with the immunolocalisation of FGFRl in the lens, the hybridisation signal also increased anteroposteriorly. For example a t E16, weak hybridisation signals were detected in anterior lens epithelial cells (arrow, Fig. 8A) and stronger signals for FGFRl were detected in the equatorial regions of the developing lens where cells elongate into fibres (arrowheads, Fig. 8A). In more differentiated fibres, in the centre of the lens, the signal diminished markedly (asterisk, Fig. 8A). Relatively strong hybridisation signals were also detected in extraocular muscles (open arrow, Fig. 8A) and in mesenchyme surrounding the optic cup. The anteroposterior pattern of FGFRl expression in the lens was more evident a t E20 (Fig. 8C) and persisted in postnatal and adult lenses (data not shown). Transcripts for FGFRl were found in lens epithelial cells and early differentiating cortical fibres. A relatively weak signal was found in central epithelial cells, a moderate signal in the germinative zone, with the strongest signal in the transitional zone where epithelial cells differentiate into fibres (arrowheads, Fig. 8C). In the cortical fibres the signal was associated with the cell nuclei constituting the characteristic bow zone of the lens. As the cortical fibre cells become more differentiated and are displaced deeper into the lens cortex, the signal for FGFRl transcripts decreased until no appreciable signal was detected in the terminally differentiated fibres (Fig. 8C, asterisk). Quantitation of hybridisation signals using image analysis confirmed the anteroposterior changes in FGFRl expression within the E20 lens (see Fig. 9). Measurements of signal density in six defined lens regions (see Fig. 1) showed a progressive increase in FGFRl expression from region I to region I11 and a progressive decline in deeper regions of the lens (regions IV to VI) where fibres are more mature and undergoing terminal differentiation. There was an approximate doubling of signal between regions I and I11 (P < 0.001) and a progressive significant decrease to region VI (P < 0.03). This pattern was similar at all ages analysed (data not shown for P3, P21, P100). In addition to spatial changes in FGFRl expression, temporal changes were also observed during lens development. Whilst the signal appeared to increase during early development (E10-E20), there was a marked age-related decrease in signal from P3 to PlOO (Fig. 10). Quantitation of hybridisation signals in region I11 of P3, P21 and P l O O rat lenses (arrowheads, Fig. 10) confirmed that there was a significant decrease (P<0.05) in FGFRl expression a t each age. Effect of FGF-2 on FGFRl Expression in Lens Explants Lens epithelial cell explants can be induced to differentiate into fibres by addition of FGF. To determine if fibre differentiation in vitro was associated with changes in FGFR1 expression, we quantified the hybridisation signal for FGFRl mRNA by grain counting in sections of 3-day-old lens explants cultured for four days with or without FGF-2 (Table 1). In control explants, which remained as a monolayer of epithelial cells, there were relatively low levels of FGFRl expression, with the peripheral region showing significantly greater FGFRl expression (P < 0.001) than the central region of the explant. Similar levels of expression were found in central (3.9 * 0.2 grains/100 pm2) and germinative (5.8 f 0.4 grainst100 p,m2) regions of 3-day-old rat lenses in situ, which had been sectioned and processed in parallel with the explants; these regions are equivalent to central and peripheral regions of lens explants, respectively. This indicates that the culture conditions did not affect receptor expression. Lens explants cultured with FGF-2 increased in thickness due to multilayering and cell elongation, which are morphological changes characteristic of early fibre differentiation (Lovicu and McAvoy, 1989). After 4 days exposure to FGF there was a three- to fourfold increase (P < 0.005) in density of hybridisation signal in the peripheral and central regions compared to control explants. The slight difference in levels of expression between the peripheral and central regions of neonatal rat lens explants cultured with FGF-2 was not significant. DISCUSSION In the present study we have investigated the spatiotemporal expression patterns of FGFRl during lens morphogenesis and differentiation using immunohistochemical and in situ hybridisation techniques. There was a strong concordance between the FGFRl protein and mRNA localisation patterns. The antibodies, R803 and M15, were raised against distinct peptide regions of the intracellular and extracellular domains of the receptor, respectively. The FGFRl reactivity of both antibodies was predominantly in the cell nuclei, although weak cytoplasmic reactivity was also detected. Nuclear reactivity for Fig. 4. lmmunolocalisationof FGFRl in eyes from E20 (A) P3 (E-G), P21 (C) and PlOO (D) rats. A: Adistinct anteroposterior pattern of FGFRl reactivity was evident at E20. Cells in the germinative zone (gz) showed fine punctate reactivity whereas cells in the transitional zone (tz) and cortical fibres showed a more intense and coarse punctate reactivity. In general, the punctate reactivity appeared to be primarily associated with cell nuclei and this was most obvious in the maturing fibres. Diffuse cytoplasmic and fine punctate reactivity was also found in the ciliary body and iris. Weak reactivity was also detected in the ganglion cell and neuroblast layers of the retina. B: Similar section to that in A, stained with haematoxylin and phloxine. A similar anteroposterior pattern of reactivity to that described for E20 was found in the P21 (C) and PlOO lens (D). E-G: Higher magnification of cells in the anterior epithelium (E), equatorial region (F) and early fibres (G) of the P3 lens showing FGFRl reactivity. H: No specific reactivity was observed with adsorbed control serum for R803 (Pi00 lens). cb, the ciliary body: g, ganglion cell layer; i, iris. Scale bar: (A-D,H), 100 km; (E-G) 40 km. 418 de IONGH ET AL. Fig. 5. lmmunolocalisation of FGFRI in P3 rat eye using antibody M15. The distribution pattern of FGFRl reactivity was similar to that obtained with R803. Reactivity was predominantly associated with lens cell nuclei; weakly in the lens epithelium and increasing anteroposteriorly with the strongest signal detected in differentiating fibres. The inset shows the fine punctate nuclear reactivity obtained with MI 5 in early lens fibres. Scale bar: 100 pm; inset, 18 ym. FGFRl has also been demonstrated recently in foetal cardiac myocytes using a peptide antibody similar to R803, raised against the carboxy terminus of f l g (Engelmann et al., 1993). Such nuclear localisation suggests that the whole receptor may be internalised and translocated to the nucleus. It has been demonstrated that, after FGF-1 stimulation, FGFR-1 trafficks to or near the nucleus as a cell cycle-regulated event (Prudovsky et al., 1994) similar to that observed for FGF-1 (Zhan et al., 1993). There is also some evidence that covalent complexes of FGF and receptor components are internalised and translocated to the nucleus (Shi et al., 1991a, b). In agreement with this we have shown that one of the FGFRl ligands, FGF-1, is also localised in lens cell nuclei in a similar distribution to that of FGFRl (Lovicu and McAvoy, 1993). It has been suggested that internalisation and nuclear translocation of peptide ligands andlor their receptors may partici- pate directly in regulation of gene expression (Jans, 1994). During early lens morphogenesis (ElO-E14), FGFRl was expressed in both lens and retinal precursors, suggesting that FGF may play a role in the inductive interactions and subsequent differentiation of both these tissues. Similarly, other studies have documented FGFRl expression in ectodermal and mesodermal lineages of early mouse (Yamaguchi et al., 1992) and rat (Wanaka et al., 1991) embryos and in presumptive lens and neural retina of chick (Heuer et al., 1990; Ohuchi et al., 19941, mouse (Orr-Urtreger et al., 1991) and rat (Wanaka et al., 1991) eyes. Together with previous studies which localised FGF-1 protein (de Iongh and McAvoy, 1993) and mRNA (unpublished observations) in the developing lens placode and retinal disc, the present findings are consistent with the suggestion that FGF-1 plays a role in the formation of these structures and this may involve an autocrine mechanism. In agreement with this, both FGF-1 and FGF-2 have been shown to induce proliferation and differentiation responses in cultured neuroepithelial cells (Murphy et al., 1990) and embryonic retinal cells (Guillemot and Cepko, 1992; Lillien and Cepko, 1992). However, to date, no studies on the influence of FGF on embryonic lens cells have been carried out. We have previously proposed that an anteroposterior gradient of increased FGF stimulation is involved in the maintenance of lens polarity and growth patterns (McAvoy and Chamberlain, 1989; Schulz et al., 1993). Clearly, this could be determined by the anteroposterior gradient of increased FGF distribution and activity that we have identified in the lens and the ocular media (see Introduction). However, the extent to which FGF stimulates a cell also depends on the level of FGFR expression. The present study has demonstrated an anteroposterior gradient of FGFRl expression in the intact lens with weak expression of both protein and mRNA in the central epithelium anteriorly, but increasingly strong expression in the germinative and transitional zones where cells maximally proliferate and undergo early stages of differentiation, respectively. The increased nuclear reactivity for FGFRl, shown by antibodies R803 and M15, in the germinative and transitional zones may reflect increased internalisation and nuclear translocation of the receptor during cell proliferation and early fibre differentiation. Increased expression of FGFRl in these regions, indicated by both the increased membrane-associated reactivity, shown by M6, and increased signal for mRNA, may indicate upregulation of FGFRl expression during cell proliferation and early fibre differentiation. Increased membrane-associated reactivity, shown by M6, was also detected in outer cortical fibres at the lens sutures. The significance of this reactivity at the lens sutures is not clear at present. Interestingly, although FGFRl protein persisted in nuclei of terminally differentiating fibres, the mRNA gradually decreased in mature fibres (regions V and FGFRl EXPRESSION DURING LENS DEVELOPMENT Fig. 6. lmmunolocalisationof FGFR in neonatal rat lens using monoclonal antibody M6. A: Lens epithelial cells showed specific reactivity along their basal surface which closely appose the lens capsule (arrowheads). 8 : If primary antibody was substituted with normal mouse IgG, there was no labelling of basal cell surfaces but there was nonspecific 419 labelling (curved arrow) at the apical surface of cells (epithelial-fibrejunction). C and D: At both anterior (C) and posterior (D) poles of the lens, strong fibrillar reactivity was observed at lens sutures. When normal mouse IgG was substituted for M6,no labelling was observed (E; posterior suture). Scale bars: (A, B), 100 pm; (C-E), 65 pm. 420 de IONGH ET AL. Fig. 7. FGFRI expression during early lens development (E10-E12). The darkfield micrographs show hybridisation signals obtained with antisense (A, D, G) and corresponding sense (C, F, I)FGFRI probes. B, E and H are brightfield micrographs of haematoxylin-stainedsections in A, D and G respectively. At E l 0 (A-C) uniformly weak signal was detected in presumptive lens ectoderm (arrowheads), optic vesicle and diencephalon, and in the surrounding mesenchyme. At E l 1 (D-F) there was no change in the ectodermal signal (arrowhead) but a slightly increased signal was detected in the basal ependymal layer of the neuroepithelium of the diencephalon (arrows) and optic vesicle which includes me reiinai disc. At E l 2 (G-I), strong signal was found in the early lens vesicle (open arrow), margins of the optic cup (arrowheads) and in the ependymal layer of the optic stalk (arrows). No specific hybridisation signal was evident with the sense probe at any age (C, F, I). d, diencephalon; e, ectoderm; Iv, lens vesicle; m, mesenchyme; oc, optic cup; os, optic stalk; ov, optic vesicle; rd, retinal disc. Scale bars: (A-C), 50 pm; (D-F), 100 pm; (G-I), 100 wm. FGFRl EXPRESSION DURING LENS DEVELOPMENT Fig. 8. FGFRl expression at El6 and E20. A At E16, strong signal for FGFRl was found in the equatorial lens where fibres differentiate (arrowheads) but only weak signals were detected in the anterior epithelium (arrow) and in more differentiated fibres in the centre of the lens (asterisk). Strong signals were also found in extraocular muscles (open arrow) and extraocular mesenchyme which may include the retinal pigmented epithelium. B: Brightfield micrograph of section shown in A. C: In the E20 lens there was a distinct anteroposterior pattern of expression with weak signals in the anterior epithelium which increased towards the 42 1 lens equator in the germinative zone. The strongest signals were in the transitional zone of the lens (arrowheads). Strong signals were also detected in ciliary body (arrow), dermis and hair follicles (curved arrow) of the eyelid and in mesenchyme of choroid and sclera. Weaker signals were detected in the cornea and stroma of the eyelids. D: Brightfield micrograph of same section shown in C, stained with haematoxylin. c, cornea; cs, choroid-sclera; e, epidermis; g, ganglion cell layer; I,lens; le, lens epithelium; nl, neuroblast layer. Scale bars (A, B), 250 pm; (C, D), 300 pm. 422 de IONGH ET AL. TABLE 1. Effect of FGF-2 on FGFR-1 Expression in Lens Epithelial Explants* Treatment Control FGFS Grain density (graindl00 pm2) Peripheral region Central region 3.8 ? 0.2 (36) 5.0 0.2 (50)** 17.0 2 0.6 (19)* 15.4 1.0 (22)* * * * S.E.M. The number of fields aData presented as mean countedkreatment is indicated in parentheses. *Significantly greater than corresponding control value (P<0.005). **Significantly greater than corresponding value for the central region (P<O.OOl). E 9a, .-> -Qa, .I- U Lens Regions Fig. 9. Quantitativeanalysis of FGFRl expression in different regions of the E20 lens. Using Tracor Northern Image Analysis System on darkfield images, the density of FGFRl hybridisation signals in six defined regions of the lens ( 1 4 1 , see Fig. 1) were measured. Values were normalised with respect to the mean value for lens region I. Each bar represents the mean ? S.E.M. of 4-7 determinations from separate sections. A similar pattern of FGFRl expression was found in lenses from P3, P21 and PI00 rats. VI) suggesting downregulation of FGFRl mRNA (possibly by decreased synthesis andfor increased degradation) during terminal differentiation. Another interpretation of the decreased signal density in the mature fibres is dispersement of the mRNA in these highly elongated cells. However, this would seem a less likely explanation, as the cells in region 111, which have undergone substantial cell elongation, have the highest signal density. Support for such a gradient of FGFRl expression in the lens contributing to the maintenance of the anteroposterior gradient of FGF stimulation is provided by in vitro studies (Lovicu and McAvoy, 1992; Richardson et al., 1992) of lens epithelial explant responses to FGF. These studies showed that cells from the periphery of lens epithelial explants, which incorporates the germinative zone (region I1 in Fig. l), were more responsive to FGF than cells from the centre of the explant (region I in Fig. 1).In the present study, in situ hybridisation with FGFRl probe on sections of lens epithelial explants showed that cells in the peripheral region of control explants had significantly greater FGFR mRNA expression than in the central region. Thus, a higher level of FGFRl expression in the cells of the peripheral region (germinative zone) may underlie their greater responsiveness to FGF. The fact that the anteroposterior increase in FGFRl expression coincides with the previously documented anteroposterior increase in FGF activity in the lens and ocular media led us to propose that FGF stimula- tion itself may upregulate expression of FGFRl. To investigate this we treated lens explants with a fibredifferentiating dose of FGF-2 and showed that, after 4 days treatment, FGFRl expression was indeed enhanced in both central and peripheral regions of the explants. These data suggest that when lens cells are exposed to high concentrations of FGF, such as would occur when they are exposed t o vitreous, they upregulate expression of FGFR1, thus potentially increasing their sensitivity to FGF stimulation. Such upregulation of FGFRl by FGF in the transitional zone of the lens would amplify the already existing gradient of FGF stimulation present in the eye and may be an important mechanism for generating a high level of FGF stimulation and ensuring a fibre differentiation response. As well as the regional differences in FGFRl expression, this study revealed a significant age-related decline in FGFRl expression in postnatal rat lenses. This may be related to the previously reported age-related decline in the ability of epithelial explants to respond to FGF (Richardson et al., 1990, 1992; Lovicu and McAvoy, 1992). Other studies have reported substantially greater levels of FGF receptors in embryonic compared with adult rodent tissues (Olwin and Hauschka, 19891, as well as a decline in FGF receptor levels during chick ocular development (Olwin and Hauschka, 1990). The age-related decline in FGFR expression in postnatal rat lenses coincides with the reduced rate of fibre differentiation that occurs with age (Hanna and O'Brien, 1961; Cenedella, 1989). Overall, these results suggest that the responsiveness of lens cells to FGF, both within the lens and with aging, may be influenced, at least in part, by the level of FGFRl expression. FGF has been shown to induce different responses (proliferation, migration, differentiation) in lens cells depending upon its concentration (McAvoy and Chamberlain, 19891.However, it is still not known how these different responses are elicited by FGF. As different FGFRs display different ligand binding affinities (Johnson and Williams, 1993) as well as the potential for activating different signalling pathways (Vainikka et al., 1992; Wang et al., 1994), it is possible that the different responses of lens cells to FGF may be related to the differential expression of FGF receptors. This FGFRl EXPRESSION DURING LENS DEVELOPMENT 423 study clearly shows that expression of FGFRl changes in relation to the lens fibre differentiation response. Therefore, it will be important in future work to determine how expression of the other members of the FGF receptor family relate to patterns of lens cell behaviour. EXPERIMENTAL PROCEDURES Animals At various stages of gestation (E10, E l l , E12, E14, E l 6 and E20), pregnant Wistar rats were euthanased by C02 asphyxiation or an overdose of barbiturate, and the uterine horns were removed. Embryos were removed from the uterine horns and dissected free of extra-embryonic membranes in cold (4°C) phosphatebuffered saline (PBS), pH 7.4. Neonatal (3-day postnatal; P3), weanling (P21) and adult (P100) rats were euthanased by C02 asphyxiation and their eyes removed. Dissected embryos and eyes were immersed in Tissue Tek OCT compound (Miles Inc., Elkhart, IN) and frozen in isopentane cooled by liquid nitrogen. Specimens were stored in liquid nitrogen until sectioned. All procedures involving animals were in accordance with the National Health and Medical Research Council (Australia) guidelines and The Association for Research in Vision and Ophthalmology Handbook for the Use of Animals in Biomedical Research (USA).All protocols were approved by the Animal Ethical Review Committee of the University of Sydney, Australia. Lens Explant Culture Untrimmed lens explants from neonatal rats (P3) were prepared as described previously (McAvoy and Fernon, 1984) and cultured in Medium 199 with Earle’s salts (Cytosystems, Sydney, Australia), supplemented with 0.1%bovine serum albumin (Sigma, Sydney, Australia), 0.1 Fgiml L-glutamine, 50 IUlml penicillin, 50 p.g/ml Streptomycin and 2.5 Fg/ml Fungizone (all from Cytosystems, Sydney, Australia), with or without FGF-2 (40 ng/ml) for four days. Previous in vitro studies (Lovicu and McAvoy, 1989) have shown that after 4 days culture with FGF-2, explanted neonatal lens epithelial cells elongate and undergo early stages of fibre differentiation. At the end of the culture period, explants were quickly rinsed in ice-cold sterile PBS, im- Fig. 10. Fig. 10. Age-related decrease in FGFRl expression in postnatal rat eyes. Sections of whole eyes from P3 (A), P21 (6)and PlOO (C) rats, hybridised with anti-sense FGFRI probes under identical conditions, were exposed to the same autoradiographic film. Signal for FGFR1 mRNA was demonstrated in the lens (I) and surrounding ocular tissues, particularly in the ciliar body and iris, and the intensity clearly diminished with age. At all ages, strongest expression for FGFRl mRNA in the lens was in the cells of the transitional zone at the equatorial region (arrowheads). A strong signal was detected in the P3 lens (A); however, only a faint trace of signal was detected in the PlOO lens (C). In the older animals, signal present in the centre of the lens (the lens nucleus) was nonspecific. Scale bar: 1.5 mm. 424 de IONGH ET AL. mersed in OCT compound and frozen as described above. FGFRl Probes FGFRl antibodies. A panel of three anti-FGFRl antibodies (R803, M15, M6) were used to localise FGFRl immunoreactivity (see Fig. 2). R803 was an affinity-purified antiserum raised in rabbits against a synthetic amino acid fragment corresponding to 17 amino acids (806-822; CLPRHPAQLANGGLKRR)a t the carboxy terminal of human FGFR1. This antibody is specific for FGFRl (Gonzalez et al., 1996). M15 was a monoclonal antibody raised against a synthetic amino acid fragment corresponding to 20 amino acids (240-259; NHTYQLDVVERSPHRPILQA) between immunoglobulin-like (Ig) domains I1 and I11 of chicken FGFRl. By immunoblotting, this antibody has been shown to recognise the chicken form of FGFRl (cek-1) and the recombinant protein corresponding to the extracellular domain of human FGFRl (Hanneken et al., 1994, 1995). M6 was a monoclonal antibody raised against a baculovirus-expressed recombinant protein corresponding to the extracellular domain of human FGFRl. It recognises high molecular weight FGFRl from 3T3 and BHK cells and low molecular weight truncated FGFRl receptors (FGF binding proteins) from bovine calf serum but also has low levels of crossreactivity with FGFR2 and FGFR3 (Allen and Maher, 1993; Hanneken et al., 1994, 1995). FGFRl cDNA. A 300-bp rat FGFRl (flg)cDNA, isolated from a rat embryonic cDNA library (Wanaka et al., 1990), was used for in situ hybridisation experiments (see Fig. 2). This cDNA has been characterised (Wanaka et al., 1990) and used previously to identify FGFRl transcripts in rat tissues (Wanaka et al., 1990, 1991).The cDNA was linearised using Eco RI and Kpn I to obtain DNA templates. Using a RNA transcription kit (Stratagene, La Jolla, CA) and 35S-UTP (Amersham, Sydney, Australia), radiolabelled antisense and sense RNA probes were transcribed from these templates with T3 and T7 RNA polymerases, respectively. nus, Hawthorn, Vic. Australia): sheep anti-mouse IgG (for monoclonal antibody) or sheep anti-rabbit IgG (for polyclonal antibody). Occasional sections were also treated with 1 kg/ml bisbenzimide (Hoechst dye, Calbiochem, La Jolla, CA) to label nuclei. Sections were then rinsed with PBS-BSA and mounted. Representative sections were subsequently stained with haematoxylin and phloxine. In Situ Hybridisation Frozen sections of ocular primordia, eyes and explants were mounted on poly-L-lysine-coated gelatinised slides and fixed in 2.5% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4 for 20 minutes. Sections were rinsed in distilled water and then processed for in situ hybridisation based on a protocol described by Simmons et al. (1989). Sections were hybridised with the radiolabelled probe (specific activity lo7 cpdml) at approximately 55°C for 16 hours. After hybridisation, sections were digested with RNase A (20 kg/ml; Sigma, St. Louis, MO) a t 37°C for 30 minutes and rinsed in decreasing salt solutions for increasing stringency (from 2 x SSPE to 0.1 x SSPE, where 1 x SSPE = 15 mM NaCl, 1 mM NaH,PO,, 1 mM EDTA) supplemented with 1 mM dithiothreitol. Following a final high stringency wash (0.1 x SSPE, 45 minutes a t 65"C), slides were dehydrated and dried for autoradiography. To initially assess the strength of hybridisation signal, sections were exposed for 5 days to autoradiographic film @-Max Hyperfilm, Amersham, Sydney, Australia) which was developed according to manufacturer's instructions. Slides were then coated with a fine film of NTB-2 emulsion (Kodak, Sydney, Australia) and stored in light-tight boxes with dessicant at 4°C. Based on the signal obtained on the initial autoradiograph, slides were exposed for 2-3 weeks, after which they were developed (D-19, Kodak), rinsed and stained with haematoxylin. Photography Sections were photographed using a Leitz Dialux 20 microscope equipped with normal, darkfield and epiillumination. Immunofluorescence micrographs were photographed with 400 ASA film (T-Max, Kodak) pushed to 1,600 ASA during processing and brightfield micrographs were photographed on 50 ASA film (Pan-F, Kodak), processed according to manufacturer's instructions. In situ hybridisation slides were viewed by darkfield illumination and photographed using 400 ASA film (T-Max, Kodak) processed according to manufacturer's instructions. Immunofluorescence Frozen sections through the ocular primordia (E10E20) and eyes (P3-PlOO) were mounted on gelatinised slides, fixed briefly in cold (- 20°C) methanol and allowed to dry at -20°C. Occasional sections were also fixed using 2.5% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) for 20 minutes a t room temperature followed by several PBS rinses. Sections were incubated for 25 minutes in PBS containing 0.1% BSA (PBS-BSA),supplemented with 3% normal goat serum, to block nonspecific binding. All antisera were diluted with this goat serum-supplemented PBS-BSA. Sections Image Analysis of FGFRl Hybridisation Signal Image analysis was used to compare the expression were incubated with the primary antibodies overnight at 4°C in a humidified chamber. After incubation, sec- of FGFRl in different regions of the intact lens from tions were rinsed in PBS-BSA and incubated for one rats of different ages and in lens explants cultured with hour a t room temperature with f luoroscein isothiocy- or without FGF. Intact lens. To quantify FGFRl expression in differanate (FITCI-conjugated secondary antibodies (Sile- FGFRl EXPRESSION DURING LENS DEVELOPMENT ent regions of the lens, the intensity of hybridisation signal was measured in six distinct regions (I-VI, see Fig. 1)of lens sections. Darkfield images were captured directly from slides by a sensitive videocamera (slowscan, Dage/MTI Inc., Michigan City, IN) and the hybridisation signals were measured as an optical density using a Tracor Northern Image Analysis system (Trator Inc., Middleton, WI). For each region, the area that contained specific signal was delineated using the cursor and the optical density of that area was measured. This technique offered a high level of resolution. As hybridisation and autoradiography conditions may have varied between experiments, values were first corrected for nonspecific background measurements and the data from each experiment were then normalised with respect to the mean count for region I from that experiment. Data from a t least three sections from each of three different experiments were then pooled and analysed. Values obtained for each of the different regions were compared using one-way analysis of variance and Student’s t-test. To compare the intensity of FGFRl hybridisation signal between lenses from rats of different ages, it was important that all samples be processed under identical conditions. To do this, sections of whole eyes from P3, P21 and PlOO rats, all hybridised with anti-sense FGFRl probes under identical conditions, were exposed to the same autoradiographic film for a period of 5 days. The density of hybridisation signal in region I11 (see Fig. 1)of the P3, P21 and PlOO lenses was determined by scanning the autoradiographic films with a personal HeNe laser densitometer (Molecular Dynamics, Sunnyvale, CA; wavelength 632.8 nm, resolution 50 pm) and quantifying the signal by image analysis software (ImageQuant, Molecular Dynamics). Data were compared using one-way analysis of variance and Student’s t-test as above. Lens explants. To examine the effects of FGF on FGFRl expression in different regions of lens explants, the intensity of hybridisation signal was measured by manual counting of exposed silver grains directly from darkfield images on the microscope. Representative transverse sections through the centre of lens explants from each treatment group were processed in parallel to ensure that hybridisation and autoradiography conditions were similar for each section. For each section, the lens explant was divided into central and peripheral regions (see Richardson et al., 1993): (i)the central region comprised the central third of the section and included lens epithelial cells from the anterior pole of the lens (equivalent to region I; see Fig. 1)and, (ii)the peripheral region comprised the remaining two-thirds of the section and included cells from the germinative zone (equivalent to region 11). An eyepiece graticule was used to divide each region into equal-sized fields which corresponded to a 21 x 21 Fm field when used in conjunction with a 40 x objective and the numbers of silver grains per field were counted. The numbers of silver grains were counted in every field which in- 425 cluded cells, with correction for partial filling of the field where necessary. For each treatment group, grain counts were made from a t least four different sections from two different explants. Counts were corrected for background which was quantified from regions of the slide adjacent to the sections; analysis of variance showed that there was no significant difference between different slides. One-way analysis of variance and Student’s t-test were used to compare values obtained from different regions of explants cultured with or without FGF. ACKNOWLEDGMENTS This work was supported by grant R01 EY03177 to J. McA from the National Eye Institute, Department of Health, Education and Welfare, USA, and by a grant from the National Health and Medical Research Council (NH & MRC), Australia. R.de I. acknowledges the support of a NH & MRC Biomedical Research Scholarship and a travelling scholarship from the Faculty of Medicine, University of Sydney, Australia. F.J.L. acknowledges the support of a Postdoctoral Research Fellowship from the Medical Foundation, University of Sydney, Australia. We gratefully acknowledge Dr. P.A. Maher and Dr. D. Lappi for providing FGFRl antibodies and Dr. J. Milbrandt for the rat FGFRl cDNA. Roland Smith is thanked for technical assistance with photography. We also thank The University of Sydney Electron Microscope Unit for the use of their image analysis facilities and Dr. C. Chamberlain for critical reading of the manuscript. REFERENCES Allen, L.E., and Maher, P.A. (1993) Expression of basic fibroblast growth factor and its receptor in an invasive bladder cell carcinoma cell line. J. Cell Physiol. 155:368-375. Blanquet, P.R., Patte, C., Fayein, N., and Courtois, Y. (1989) Identification and isolation from bovine epithelial lens cells of two basic fibroblast growth factor receptors that possess bFGF-enhanced phosphorylation activities. Biochem. Biophys. Res. Commun. 160: 1124-1 131. Cenedella, R.J. (1989) Aging and rates of lens cell differentiation in vivo, measured by a chemical approach. Invest. Ophthalmol. Vis. Sci. 30575-579. Chamberlain, C.G., and McAvoy, J.W. (1989) Induction of lens fibre differentiation by acidic and basic fibroblast growth factors. Growth Factors 1:125-134. de Iongh, R., and McAvoy, J.W. (1992) Distribution of acidic and basic fibroblast growth factors (FGF) in the foetal rat eye: Implications for lens development. Growth Factors 6:159-177. de Iongh, R., and McAvoy, J.W. (1993) Spatio-temporal distribution of acidic and basic FGF indicates a role for FGF in rat lens morphogenesis. Dev. Dynam. 198:190-202. Engelmann, G.L., Dionne, C.A., and Jaye, M.C. (1993) Acidic fibroblast growth factor and heart development: Role in myocyte proliferation and capillary angiogenesis. Circ. Res. 72:7-19. Gonzalez, A.M., Berry, M., Maher, P.A., Logan, A,, and Baird, A. (1996) A comprehensive analysis of the distribution of FGF-2 and FGFR-1 in the rat brain. Brain Res. (In press). Guillemot, F., and Cepko, C.L. (1992) Retinal fate and ganglion cell differentiation are potentiated by acidic FGF in an in uitro assay of early retinal development. Development 114:743-54. Hanna, C., and O’Brien, J.E. (1961) Cell production and migration in the epithelial layer of the lens. Arch. Ophthalmol. 66:103-107. 426 de IONGH ET AL. Hanneken, A,, and Baird, A. (1992) Immunolocalization of basic fibroblast growth factor: Dependence on antibody type and tissue fixation [letter]. Exp. Eye Res. 54:lOll-1014. Hanneken, A,, Ying, W., Ling, N., and Baird, A. (1994) Identification of soluble forms of the fibroblast growth factor in blood. Proc. Natl. Acad. Sci. USA 91:9170-9174. Hanneken, A., Maher, P., and Baird, A. (1995) High affinity immunoreactive FGF receptors in the extracellular matrix of vascular endothelial cells-implications for the modulation of FGF2. J . Cell Biol. 128:1221-1228. Heuer, J.G., von Bartheld, C.S., Kinoshita, Y., Evers, P.C., and Bothwell, M. (1990) Alternating phases of FGF receptor and NGF receptor expression in the developing chicken nervous system. Neuron 5:283-296. Jans, D.A. (1994) Nuclear signaling pathways for polypeptide ligands and their membrane receptors. FASEB J . 8:841-847. Johnson, D.E., and Williams, L.T. (1993) Structural and functional diversity in the FGF receptor multigene family. Adv. Cancer Res. 6O:l-41. Lillien, L., and Cepko, C. (1992) Control of proliferation in the retina: Temporal changes in response to FGF and TGF-a. Development 115:253-266. Lovicu, F.J., and McAvoy, J.W. (1989) Structural analysis of lens epithelial explants induced to differentiate into fibres by fibroblast growth factor (FGF). Exp. Eye Res. 49:479-494. Lovicu, F.J., and McAvoy, J.W. (1992) The age of rats affects the response of lens epithelial explants to fibroblast growth factor: An ultrastructural analysis. Invest. Ophthalmol. Vis. Sci. 33:26292278. Lovicu, F.J., and McAvoy, J.W. (1993) Immunolocalisation of acidic FGF, basic FGF and heparan sulphate proteoglycan in the rat lens: Implications for lens polarity and growth patterns. Invest. Opbthalmol. Vis. Sci. 34:3355-3365. Lovicu, F.J., de Iongh, R., and McAvoy, J.W. (1996) Expression of FGF-1 and FGF-2 mRNA during lens morphogenesis, differentiation and growth. (unpublished). Marcelle, C., Eichmann, A., Halevy, O., BrBaut, C., and Le Douarin, N.M. (1994) Distinct developmental expression of a new avian fibroblast growth factor receptor. Development 120:683-694. McAvoy, J.W. (1978a) Cell division, cell elongation and distribution of a-,p- and y-crystallins in the rat lens. J . Embryol. Exp. Morphol. 44:149-165. McAvoy, J.W. (197813) Cell division, cell elongation and the co-ordination of crystallin gene expression during lens morphogenesis in the rat. J . Embryol. Exp. Morphol. 45:271-281. McAvoy, J.W., and Chamberlain, C.G. (1989)Fibroblast growth factor (FGF) induces different responses in lens epithelial cells depending on its concentration. Development 107:221-228. McAvoy, J.W., and Fernon, V.T.P. (1984) Neural retinas promote cell division and fibre differentiation in lens epithelial explants. Curr. Eye Res. 3:827-834. Murphy, M., Drago, J., and Bartlett, P.F. (1990) Fibroblast growth factor stimulates the proliferation and differentiation of neural precursor cells in uitro. J . Neurosci. Res. 25:463-475. Ohuchi, H., Koyama, E., Myokai, F., Nohno, T., Shiraga, F., Matsuo, T., Matsuo, N., Taniguchi, S., and Noji, S. (1994) Expression patterns of two fibroblast growth factor receptor genes during early chick eye development. Exp. Eye Res. 58:649-658. Olwin, B.B., and Hauschka, S.D. (1989) Cell type and tissue distribution of the fibroblast growth factor receptor. J . Cell Biochem. 39: 443-454. Olwin, B.B., and Hauschka, S.D. (1990) Fibroblast growth factor levels decrease during chick embryogenesis. J . Cell Biol. 110:503-509. Orr-Urtreger, A,, Givol, D., Yayon, A,, Yarden, Y., and Lonai, P. (1991) Developmental expression of two murine fibroblast growth factor receptors, flg and bek. Development 113:1419-1434. Orr-Urtreger, A,, Bedford, M.T., Burakova, T., Arman, E., Zimmer, Y., Yayon A,, Givol, D, and Lonai, P. (1993) Developmental localization of the splicing alternatives of the fibroblast growth factor receptor-2. Dev. Biol. 158:475-486. Peek, R., McAvoy, J.W., Lubsen, N.H., and Schoenmakers, J.G.G. (1992) Rise and fall of crystallin gene messenger levels during fibroblast growth factor induced terminal differentiation of lens cells. Dev. Biol. 152:152-160. Peters, K., Ornitz, D., Werner, S., and Williams, L.T. (1993) Unique expression pattern of the FGF receptor 3 gene during mouse organogenesis. Dev. Biol. 155:423-430. Peters, K.G., Werner, S., Chen, G., and Williams, L.T. (1992) TWO FGF receptors are differentially expressed in epithelial and mesenchymal tissues during limb formation and organogenesis in the mouse. Development 114:233-243. Prudovsky, I., Savion, N., Zhan, X., Friesel, R., Xu, J., Hou, J., McKeehan, W.L., and Maciag, T. (1994) Intact and functional fibroblast growth factor (FGF) receptor-1 trafficks near the nucleus in response to FGF-1. J . Biol. Chem. 269:31720-31724. Richardson, N.A., and McAvoy, J.W. (1990) Age-related changes in fibre differentiation of rat lens epithelial cells exposed to fibroblast growth factor. Exp. Eye Res. 50:203-211. Richardson, N.A., McAvoy, J.W., and Chamberlain, C.G. (1992) Age of rats affects response of lens epithelial explants to fibroblast growth factor. Exp. Eye Res. 55:649-656. Richardson, N.A., Chamberlain, C.G., and McAvoy, J.W. (1993) IGF-1 enhancement of FGF-induced lens fiber differentiation in rats of different ages. Invest. Ophthalmol. Vis. Sci. 343303-3312. Robinson, M.L., Overbeek, P.A., Verran, D.J., Grizzle, W.E., Stockard, C.R., Friesel, R., Maciag, T., and Thompson, J.A. (1995) Extracellular FGF-1 acts as a lens differentiation factor in transgenic mice. Development 121:505-514. Schulz, M., chamberlain, C.G., de Iongh, R.U., and McAvoy, J.W. (1993) Acidic and basic FGF in ocular media: Implications for lens polarity and growth patterns. Development 118:117-126. Shi, E., Kan, M., Xu, J., and McKeehan, W.L. (1991a) 16-kilodalton heparin binding (fibroblast) growth factor type one appears in a stable 40-kilodalton complex after receptor-dependent internalization. J . Biol. Chem. 2665774-5779. Shi, E., Kan, M., Xu. J., Morrison, R., and McKeehan, W.L. (1991b) Direct linkage of heparin binding (fibroblast) growth factor type one to a fragment of its extracellular domain after receptor-dependent internalization. J. Cell Biol. (Suppl.) 115:416a. Simmons, D.M., Arriza, J.L., and Swanson, L.W. (1989) A complete protocol of in situ hybridization of messenger RNA in brain and tissues with radiolabelled single-stranded RNA probes. J. Histotechnol. 12:169-181. Vainikka, S., Partanen, J., Bellosta, P., Coulier, F., Birnbaum, D., Basilico, C., Jaye, M., and Alitalo, K. (1992) Fibroblast growth fact o r 4 shows novel features in genomic structure, ligand binding and signal transduction. EMBO J . 11:4273-4280. Wanaka, A., Johnson, E.M., and Milbrandt, J . (1990) Localization of FGF receptor mRNA in adult rat central nervous system by in situ hybridization. Neuron 5:267-281. Wanaka, A., Milbrandt, J., and Johnson, E.M. (1991) Expression of FGF receptor gene in rat development. Development 111:455-468. Wang, J.K., Gao, G., and Goldfarb, M. (1994) Fibroblast growth factor receptors have different signalling and mitogenic potentials. Mol. Cell. Biol. 14181-188. Yamaguchi, T.P., Conlon, R.A., and Rossant, J . (1992) Expression of the fibroblast growth factor receptor FGFR-l/fZg during gastrulation and segmentation in the mouse embryo. Dev. Biol. 152:75-88. Zhan, X., Hu, X.,, Friesel, R., and Maciag, T. (1993) Long term growth factor exposure and differential tyrosine phosphorylation are required for DNA synthesis in BALB/c 3T3 cells. J . Biol. Chem. 268: 9611-9620.