Anat Embryol (1991) 184:83-91
Anatomy and Embryology © Springer-Verlag 1991
Immunolocalisation of epidermal growth factor (EGF), EGF receptor and transforming growth factor alpha (TGF~) during murine palatogenesis in vivo and in vitro M.J. Dixon, J. Garner, and M.W.J. Ferguson Animal and Human Reproduction, Development and Growth Research Group, Department of Cell and Structural Biology, Stopford Building, University of Manchester, Manchester M13 9PT, UK Accepted April 23, 1991
Summary. The distribution of epidermal growth factor, the epidermal growth factor receptor and transforming growth factor alpha during murine palatogenesis was investigated immunocytochemically. On embryonic day 12 staining for transforming growth factor alpha was present throughout the palatäl mesenchyme, with little in the epithelia. On embryonic day 13 staining increased in the palatal epithelia and in the mesenchyme at the tip of the palate. As the palatal shelves fused together (embryonic day 14.5) intense staining for transforming growth factor alpha was seen in the midline epithelial seam and in the subjacent mesenchyme. On embryonic day 15 there was a generalised increase in palatal epithelial staining; this was most marked in the remnants of the degenerating epithelial seam. Mesenchymal staining was, however, uniform. Whilst palatal staining for epidermal growth factor was sparse, at all stages, staining for its receptor was present throughout the palatal epithelia and mesenchyme. This was most intense in the palatal medial edge epithelia at the time of midline epithelial seam degeneration. The regional and temporal differences in staining for the epidermal growth factor receptor and transforming growth factor alpha suggested that these molecules may play an important role in normal palate development in vivo, particularly in degeneration of the midline epithelial seam. Key words: Palate - Epidermal growth factor - Transforming growth factor alpha - Growth factors
Introduction Epidermal growth factor (EGF) consists of a single polypeptide chain of 53 amino acids and molecular weight 6,045 daltons. It has numerous biological effects, including the control of cellular proliferation, differentiation and extracellular matrix (ECM) synthesis (see reviews Offprint requests to: M.J. Dixon
by Hollenberg 1979; Cohen 1986). It exerts its actions by binding at an integral membrane glycoprotein receptor of 170 Kd (Cohen et al. 1982). The presence of functional EGF receptors on embryonic and extraembryonic tissue (Adamson et al. 1981; Adamson and Warshaw 1982; Adamson and Meek 1984) has suggested that EGF may play an important role in normal embryonic development, particularly during development of the secondary palate (Nexo et al. 1980). The palate arises as bilateral outgrowths from the maxillary proeesses (embryonic day 12 in mice). The outgrowths initially grow vertieally down the sides of the tongue (embryonic day 13 in reite). Each palatal shelf consists of a central core of neural crest-derived mesenchyme surrounded by a simple, histologically undifferentiated epithelium which is two- to three cells thick (Greene and Pratt 1976; Ferguson 1988). At a precise stage (embryonic day 14.5 in reite) the shelves rapidly elevate to a horizontal position above the tongue and contact each other. The medial edge epithelia (MEE) fuse together to form a midline epithelial seam which degenerates by cell death and epithelial-mesenchymal transformation (Ferguson 1988), allowing mesenehymal continuity across the palate. Simultaneously, the epithelia along the nasal and oral aspects of the palate differentiate into pseudostratified, ciliated columnar cells, and stratified, squamous keratinising cells respectively (Greene and Pratt 1976). Regional palatal epithelial differentiation is specified by the mesenchyme (Ferguson and Honig 1984). The mechanism of signalling is thought to involve the interaction of ECM molecules and soluble factors (Ferguson 1987, 1988). In vitro exogenous EGF affects palatal epithelial differentiation (Hassell 1975; Hassell and Pratt 1977; Abbott and Pratt 1987a, b; Sharpe and Ferguson 1988) mesenchymal cell division (Yoneda and Pratt 1981) and ECM synthesis (Silver et al. 1984; Turley et al. 1985). However, whilst EGF receptors have been detected from embryonic day 11 in mice, and increase in number up to embryonic day 18, particularly in the secondary palate, EGF itself is barely detectable by radioimmunoassay
84
before embryonic day 14.5. This suggests that the fetal form of mouse EGF differs from the adult (Nexo et al.
frontal cryosections were cut at 6 gm in a Leitz 1720 cryostat (Leitz-Wetzlar) at - 2 0 ° C. The sections were fixed in acetone for 5 min, air-dried and stored at - 7 0 ° C until used for staining. Representative sections from the anterior, middle and posterior regions of the developing palate (Diewert 1978) were selected. The primary antibodies (Table 1) were applied directly to the sections and incubated at 4° C in a humid atmosphere for 48 h, followed by three 5-minute washes in phosphate-buffered saline (PBS). The sections were blocked for 1 h with 0.5 M sodium chloride buffer containing 0.1% bovine serum albumin (Sigma Chemical Company, Poole), 0.05% Tween 20 (Sigma) and 5% fetal calf serum (Flow Laboratories, Rickmansworth). Excess blocking buffer was removed, and the appropriate gold-conjugated secondary antibody (Table 2) applied. After an incubation for 1 h at room temperature the sections were rewashed in PBS as above, and fixed overnight at 4° C in a solution of i % paraformaldehyde/1% glutaraldehyde (Agar Aids, London) in PBS. After fixation the slides were rinsed in PBS, then given ten 15-minute washes in deionised water. The sections were then silver enhanced using Intense M (ICN Biomedicals, High Wycombe). The enhancement procedure was monitored microscopically and stopped by washing in deionised water when a suitable signal/noise ratio had been achieved. The sections were then counterstained with erythrocin B, dehydrated, cleared and mounted in Eukitt synthetic mounting medium (BDH, Poole). Control sections were established by replacing the primary antibody with either serum from the animal in which it was raised (Table 1) or PBS, or by preabsorption of the primary antisera with either EGF or TGF«. The sections were viewed by bright field illumination using a Leitz Dialux 22EB microscope (Leitz-Wetzlar). Photographs were taken on Technical Pan film (Ilford; rated at I00 ASA) from which black and white prints were made.
1980). Transforming growth factor alpha (TGF«), which acts via the EGF receptor (Todaro et al. 1980) has been implicated as the embryonic homologue of EGF (Twardzik et al. 1982; Matrisian et al. 1982; Twardzik ]985; Freemark and Comer 1987). We have, therefore, immunolocalised EGF receptor, EGF and TGF~ in the developing mouse palate with the aim of investigating whether their distribution changes with developmental time, and whether these changes correlate with known morphogenetic/differentiativeevents.
Materials and methods MF strain mice were mated overnight. Fertilisation was assumed to have occurred at midnight prior to the day on which vaginal plugs were detected. Pregnant females were killed by ether overdose, and the gravid uteri were aseptically removed to sterile Hank's balanced salt solution (Flow Laboratories, Rickmansworth). Individual embryos were dissected from the uterine decidua, staged by the external features described by Theiler (1972) and decapitated. Palates were dissected from Theiler stage 21 (embryonic day 13) mouse heads, explanted onto the surface of preformed collagen gels either singly or in pairs, and grown in submerged culture for 24, 48 or 72 h. The culture media consisted of: 1. A 1 : 1 mixture of Dulbecco's modification of Eagle's Medium and Ham's F-12 medium (DMEM/F12) (Flow Laboratories, Rickmansworth) 2. DMEM/F12 supplemented with 2.5% donor calf serum (DCS) 3. DMEM/F12 plus 10 ng/ml TGFc~ (Peninsula Laboratories) 4. DMEM/F12 plus 10 ng/ml TGFΠplus 2.5% DCS Embryonic mouse heads (days 12 to 15; Theiler stages 20 to 23) or cultured palates were immediately snap-frozen in liquid nitrogen-cooled isopentane. The tissue was subsequently embedded in OCT embedding compound (BDH Chemicals, Poole) and serial
Tal)le 1. Primary antibodies used
Table 2. Gold-conjugated secondary anti-
bodies used
Results
In vivo staining patterns Embryonic day 12. S t a i n i n g for the E G F receptor is evenly d i s t r i b u t e d t h r o u g h o u t the p a l a t a l epithelia a n d mes-
Primary antibody raised against
Primary antibody raised in
Dilution used
Supplier
Secondary antibody (see Table 2)
TGFc~ (human) EGF (mouse) EGF-receptor (A431 cells)
Goat
1 :40
1
Rabbit
1 : 40
Biotope, Seattle ICN Biomedicals, High Wycombe
Mouse (monoclonal)
1:40
Table 1 reference no.
Secondary antibody raised against
Secondary antibody raised in
Goat IgG
Rabbit
Rabbit IgG Mouse IgG
Goat Goat
ICN Biomedicals, High Wycombe
2 3
Dilution
Supplier
5 nm
1:40
10 nm 10 nm
1:25
ICN Biomedicals, High Wycombe Sigma, Poole Sigma, Poole
Size of particle (nm)
1:25
85
Fig. 1 A, B. Frontal sections through embryonic day 12 mouse heads. A Staining for the EGF receptor is evenly distributed throughout the palatal epithelia (e) and rnesenchyme (m). Bar 100 ~tm. B Staining for TGF~ is also evenly distributed throughout the palatal mesenchyme (m); little staining is present in the epithelium (e). Bar 150 Bm Fig. 2A, B. Frontal sections through embryonic day 13 mouse heads stained for the EGF receptor. A Staining is present throughout the palate (p) and is intense in the enamel organ (eo) of the incisor tooth germ. Bar 400 gin. B Palatal staining is present throughout the epithelia (e) and mesenchyme (m), and is most intense along the basal surfaces of the epithelium and along the epithelial-mesenchymal interface (arrowed). Bar 150 gm Fig. 3A, B. Frontal sections through embryonic day 13 mouse heads. A Staining for TGFc~ is present in the palatal epithelia (e) and mesenchyme (m), mesenchymal staining being most intense at the palatal tip (arrowed). Bar 200 gin. B Staining for EGF is evenly distributed throughout the palate (p) but is sparse. Bar 150 ~m
86 enchyme (Fig. 1 A). Whilst intense staining is present in the epithelia of the nasal passages, especially superficially, staining of the tongue, lateral maxilla and mandible is of similar intensity to that prescnt in the palate. Strong staining for TGFc~ is present throughout the palatal mesenchyme, little staining is, however, present in the palatal epithelia (Fig. 1 B). Similarly, little staining is present in the epithelia of the floor of the mouth, tongue or lateral maxilla. The epithelia of the nasal passages are, however, intensely stained, especially superficially. Conversely, staining for EGF is sparse, the mesenchyme being more intensely stained than the epithelia.
Embryonic day 13. Intense staining for the EGF receptor is present in the enamel organ of the incisor and molar tooth germs (Fig. 2A). This is most marked in the central region, which ultimately forms the stellate reticulum, and along the basal surfaces of the future internal and external enamel epithelia. Intense staining is also present in the epithelia of the nasal passages and vomeronasal organs, especially superficially. Staining for the EGF receptor is present throughout the palatal epithelia and mesenchyme, and is most intense along the basal surfaces of the epithelial cells and at the epithelial-mesenchymal interface (Fig. 2A, B). By embryonic day 13 epithelia, including that of the palate, stain more intensely for TGFc« Staining in the mesenchyme at the tip of the palatal shelf is more intense than that in the remainder of the palatal mesenchyme (Fig. 3 A). Intense staining is also present along the epithelial-mesenchymal interface of the palate and the floor of the mouth. Intense staining persists in the epithelia of the nasal passages and vomeronasal organs. It is also present in the central region of the enamel organ and the dental papilla mesenchyme of the tooth germs, although this is not as marked as for the EGF receptor. Staining for EGF remains sparsely and evenly distributed throughout the palate (Fig. 3 B). Unlike the EGF receptor and TGF~ there is no increased staining in the enamel organ. Embryonic day 14.5. Whilst staining for the EGF receptor is present throughout the palatal mesenchyme, that in the palatal epithelia is strongest basally (Fig. 4). Overall, there is little change in the staining pattern from embryonic day 13. However, there is a marked increase in staining in the epidermis and in the mesenchyme at the base of the nasal septum (Fig. 4). Palatal staining for TGFc~ is differentially expressed in an antero-posterior sequence. In the most anterior regions, where the palatal shelves have elevated and come into contact, but have not fused, staining is fairly evenly distributed throughout the epithelium (Fig. 5A). Mesenchymal staining is intense and most marked at the tips of the approximated shelves (Fig. 5A). Intense staining is also present in the blood vessels. Further posteriorly where the MEE of apposing shelves have fused together, extremely intense staining is seen along the midline epithelial seam (Fig. 5 B), particularly along the basal epithelial surfaces (Fig. 5 C), and in the subjacent
mesenchyme. A lower level of staining is present in the remainder of the palatal epithelia and mesenchyme. As with the EGF receptor, intense staining for TGF« is present in the epithelia of the nasal passages, the central region of the enamel organ (Fig. 5D), the dental papilla mesenchyme (Fig. 5D) and the mesenchyme at the base of the nasal septum. Staining for EGF is generally sparse, although it is most marked in the epithelia of the nasal passages and in the central region of the enamel organ. Staining is present throughout the palatal mesenchyme, with little staining in the epithelia (Fig. 6).
Embryonic da), 15. The staining patterns obtained with the anti-EGF receptor antibody were similar to those recorded for embryonic day 14.5, high levels of staining being present in the epithelia of the nasal passages and in the enamel organs. Staining is present throughout the palatal epithelia and mesenchyme, and is intense in the remnants of the degenerating epithelial seam (Fig. 7A,
B). Staining for TGFŒ is present in the epithelia of the nasal passages (Fig. 8 A), the central region of the enamel organ, the dental papilla mesenchyme and its interface with the internal enamel epithelium. Intense staining is also present in the epidermis. There is a generalised increase in epithelial staining for TGF«. This is marked in the dorsum of the tongue and in the palate, particularly high levels being detected in the remnants of the degenerating epithelial seam, at the nasal angle of the palate (Fig. 8 A, B) and along the epithelial-mesenchymal interface. By this stage of development, however, staining is uniform in the palatal mesenchyme; an area of increased staining adjacent to the degenerating seam is no longer apparent. Staining for EGF is increased from earlier stages. It is seen throughout the palatal mesenchyme and epithelia, that in the mesenchyme being more intense than that in the epithelium. Other staining patterns remain unchanged from earlier stages.
In vitro staining patterns The staining patterns obtained in vitro are the same as those obtained in vivo. Staining for the EGF receptor is evenly distributed throughout the palatal epithelia and mesenchyme on culture days 0 and 1 (equivalent of embryonic days 13 and 14 in vivo). Staining becomes more intense in the medial edge epithelium as it degenerates. On the day of explantation staining for TGFΠis uniformly distributed throughout the palatal epithelia; a small area of more intense staining is present in the mesenchyme at the palatal tip. As the shelves fuse (culture day 2) intense staining is present along the midline epithelial seam; this persists as the seam degenerates (culture day 3) (Fig. 9 A). Where TGFc~ is added to the culture medium in the presence of serum and MEE degeneration is inhibited (as described by Ferguson 1988; Sharpe and Ferguson
87
Fig. 4. Frontal seetion through an embryonic day 14.5 mouse head stained for EGF receptor. Intense staining is present in the palate (p) and in the nasal septum (ns) but is less intense in the tongue (t). Bar 150 gm
Fig. 5A-D. Frontal sections through embryonic day 14.5 mouse heads stained for TGFc« A Anteriofly, where the palatal shelves (p) are in contaet, but have not fused, epithelial staining is uniform. Intense staining is present in the mesenehyme beneath the medial
1988) the M E E does not stain intensely for T G F « , regardless of time in culture (Fig. 9 B). All control staining was negative (Fig. 10).
Diseussion Previous studies have implicated E G F in normal embryonic development (Adamson 1983). For example, E G F has been shown to prevent hyaline membrane disease in fetal lambs and rabbits by promoting epithelial growth and differentiation in the developing lung (Cat-
edge epithelium (arrowed), at the base of the nasal septum (ns) and in the blood vessels (bv). Bar 200 gin. B Further posteriorly, intense staining is present in the epithelium at the base of the nasal septum (ns) and along the midline palatal epithelial seam (s). Bar 100 ~tm. C With continued development, staining of the midline epithelial seam (s) becomes localised to the basal cells (arrowed). Bar 200 gin. D Intense staining is also present in the molar tooth germs, particularly in the central region of the enamel organ (eo) but also in the dental papilla mesenchyme (dp). Bar 150 Ixm
terton et al. 1979; Sundell et al. 1980). E G F receptors have been detected on murine embryonic tissue (Adamson 1983). The earliest time of detection is on trophoblast cells of a 5-day blastocyst cultured for three days (Adamson and Meek 1984). These receptors are functional in that they bind E G F and are down-regulated by excess E G F injected into the placenta or amniotic cavity (Adamson and Warshaw 1982). Nexo et al. (1980) used both a radioreceptor and radioimmunoassay to estimate the content of E G F in embryonic day 11.5 to 17.5 mouse embryos. E G F receptor binding was readily detected in the 11.5 day embryos,
88
Fig. 6. Frontal section through an embryonic day 14.5 mouse head. Staining for EGF is sparsely distributed throughout the palate (p). Bar 200 gm Fig. 7A, B. Frontal sections through an embryonicday 15 mouse head. Staining for the EGF receptor is present throughout the palatal epithelia (e) and mesenchyme (m), and is particularly intense in the remnants of the degeneratingepithelial seam (arrowed). Bar, A 150 gm, B 50 grn and rose steadily up to parturition, the rise being most marked in potential target tissues such as the secondary palate. However, in whole embryo extracts far greater amounts of EGF were detected by radioreceptor than by radioimmunoassay, suggesting that the fetal form of mouse EGF differs from the adult form. The nature of the fetal form of EGF suggested by Nexo et al. (1980) is uncertain. TGFc~, which shares considerable homology with EGF (Marquardt et al. 1984) and acts via the EGF receptor (Todaro et al. 1980), has recently been demonstrated in extracts of embryonic mice (Twardzik et al. 1982; Twardzik 1985), rats (Matrisian et al. 1982), sheep (Freemark and Comer 1987) and chicks (Mesiano et al. 1985). TGFc~ mRNA has also
been detected in pre-implantation mouse embryos (Rappolee et al. 1988). Twardzik (1985) has further demonstrated that there is an increase in TGF« expression between embryonic days 12 and 15 in the mouse. By contrast it has been reported that TGFe mRNA is not present at significant levels beyond day 10 ofmouse development (Lee et al. 1985; Wilcox and Derynck 1988). A1though this last study could not exclude the possibility that small populations of cells synthesised TGFe at levels below the detection threshold of the assay. TGFc~ may therefore be the fetal form of EGF. In the present study we have shown that on embryonic day 12 TGFc~ is localised throughout the palatal mesenchyme but is sparse in the epithelium. By embryonic
89
•
«
«
i,9' ~'«
,2
9B
m
10
P
P
Fig. 8A, B. Frontal sections through an embryonic day 15 mouse head stained for TGFc« Intense staining is present in the epithelia of the nasal passages (np) and tongue (t). This is most marked in the remnants of the degenerating epithelial seam (arrowed) and in the epithelia at the nasal angle of the palate (n). Bar, A 150 gm, B 25 gm Fig. 9A, B. Transverse sections through cultured palatal shelves stained for TGFcc A After 72 h of culture in serum/TGFc~-free
medium, staining is present throughout the palate, but is particularly intense in the midline epithetial seam (s). Bar 200 gm. B After 72 h of culture, in medium supplemented with 10 ng/ml TGFc~ and 2.5% donor calf serum, increased staining in the persistent medial edge epithelium (m) is not detected. Bar 100 lam Fig. 10. Frontal section through embryonic day 14.5 mouse head stained with secondary antibody alone (control section). The palates (p) are completely unstained. Bar 150 gm
90
day 13 staining is more intense in the palatal epithelium and in the mesenchyme at the tip of the shelf. After palatal shelf elevation there is a marked increase in TGF« staining particularly in the MEE during epithelial seam formation. As the epithelial seam degenerates there is a generalised increase in epithelial staining, particularly high levels being detected in the seam remnants. Conversely, over the corresponding period, staining for EGF remained sparse, whilst staining for the EGF receptor increased on the palatal MEE around the time of seam degeneration. TGF:~ and its interaction with the EGF receptor therefore appear important in palate development: both are temporally and spatially regulated, EGF is not. The finding that strong staining for the EGF receptor is detected in the MEE around the time of midline epithelial seam degeneration does not appear to agree with the results of Abbott et al. (1988) who reported that the MEE cells ceased to express EGF receptors late on embryonic day 13. This apparent anomaly may be due to differences in the properties of the antibodies used to detect the EGF receptor. Abbott et al. (1988) used a rabbit polyclonal antibody to EGF receptors purified from mouse liver. This does not react well with ligandoccupied receptors (Weller et al. 1987). Our antibody recognises the receptor regardless of whether it is occupied by ligand (EGF or TGFe) or not. Moreover, we have co-localised EGF and TGFc« Whilst EGF is sparsely distributed, TGF« is, for the most part, present in the same locations as the EGF receptor, which suggests that receptor-bound ligand was being recognised. Abbott et al. (1988) did not colocalise EGF or TGFe, rather they assessed EGF binding by [125I]-EGF autoradiography, which will localise unoccupied, accessible, cell-surface EGF receptors (see Green and Couchman 1985). Moreover, 12SI-EGF may not compete oft, bound, unlabelled EGF or TGFc« The absence of staining on the MEE and subjacent mesenchyme reported by Abbott et al. (1988) probably refiects a large number of EGF receptors with bound ligand rather than the absence of such receptors. Previous work has suggested that, in vitro, exogenous EGF causes cleft palate by preventing MEE cell death (Hassell 1975; Hassell and Pratt 1977; Abbott and Pratt 1987a, b). These studies used either pharmacological doses of EGF (up to 2 ~tg/ml) or serum-supplemented culture medium. EGF and TGFc~ have also been shown to stimulate cell proliferation in the developing palate (Yoneda and Pratt 1981). These results led Pratt (1987) to propose that on embryonic day 12 the palatal mesenchymal and oral and medial edge epithelial cells produce TGFc~, which then supports rapid growth of these cells during early palatal development. The MEE cells subsequently cease to produce TGF~ and, as a result, cease to synthesise DNA and die. By contrast, the oral epithelial cells continue to produce TGFŒ and differentiate into stratified, squamous keratinising cells. Our findings have, however, shown that intense staining for TGFc~ and the EGF receptor is present in the midline epithelial seam even after it has started to degenerate. This may
be in keeping with the observation that seam degeneration involves epithelial-mesenchymal transformation, and not just cell death as previously imagined (Ferguson 1988). Recent reports have suggested that TGFe stimulates ECM biosynthesis in either palatal organ cultures or in cultures of isolated palatal mesenchymal or epithelial cells (see reviews by Ferguson 1988; Sharpe and Ferguson 1988). It is particularly interesting that in epithelial cell culture TGFc~ causes an increase in staining for tenascin and type IX collagen in the MEE, as these molecules have been implicated in MEE degeneration (Ferguson 1988). The data from this study combined with those from cell/organ culture studies lead us to believe that the previously postulated developmental decline of EGF (or TGF«) and its receptor during medial edge epithelial seam formation is incorrect (Pratt 1987). Normally the levels of TGFc~ and the EGF receptor increase in the midline palatal epithelial seam. The TGF« is probably being synthesised by the mesenchyme, and binding to EGF receptors in the seam cells. Previous experiments, using pharmacological doses of EGF/TGF~ to induce cleft palate in vitro by inhibiting MEE degeneration (Pratt 1987), probably cause a perturbation of the above normal mechanisms, perhaps by down-regulation of the EGF receptor. This developmentally regulated, stage- and regionspecific event is important in signalling the mesenchymal-epithelial interactions which specify midline epithelial seam degeneration. The signalling may result from a combination of: 1. Direct effects on the epithelial cells e.g. on keratin profile 2. Stimulation of specific ECM molecule biosynthesis by the underlying mesenchyme cells, and the effects of such ECM molecules on the epithelial cells (Ferguson 1988) 3. Stimulation of ECM receptors on the epithelial cells (Sharpe and Ferguson 1988) 4. Stimulation of specific ECM molecule biosynthesis by the palatal epithelial cells, thus minimising the differences in the ECM interface between the palatal epithelial and mesenchymal cells. The result of this integrated growth factor/ECM signalling mechanism (Sharpe and Ferguson 1988) is that the basal epithelial cells migrate out of the seam and transform into mesenchyme cells (Ferguson 1988) whilst the post-mitotic suprabasal cells die. Physiological levels of TGF~ and its interaction with the EGF receptor appear to be important for normal palatal development. In this context it is perhaps significant that a genetic association has been observed between variation of the TGF« locus and the occurrence of clefting in human caucasian families (Ardinger et al. 1989). Acknowledgements. This work was supported by grants from the Wellcome Trust, Medical Research Council, Action Research for the Crippled Child and Birthright. M.J. Dixon was funded by a Wellcome Trust Research Training Fellowship grant.
91
References Abbott BD, Pratt RM (1987a) Retinoids and epidermal growth factor alter embryonic mouse palatal epithelial and mesenchymal cell differentiation in organ culture. J Craniofac Genet Der Biol 7:219-240 Abbott BD, Pratt RM (1987b) Human embryonic palatal epithelial differentiation is altered by retinoic acid and epidermal growth factor in organ culture. J Craniofac Genet Dev Biol 7:241-265 Abbott BD, Adamson ED, Pratt RM (1988) Retinoic acid alters EGF receptor expression during palatogenesis. Development 102:853-867 Adamson ED (1983) Growth factors in development. In: Warshaw JB (ed) The biological basis of reproductive and developmental medicine. Elsevier, New York, pp 307-336 Adamson ED, Meek J (1984) The ontogeny of epidermal growth factor receptors during mouse development. Dev Biol 103:6270 Adamson ED, Warshaw JB (1982) Down-regulation of epidermal growth factor receptors in mouse embryos. Dev Biol 90:430434 Adamson ED, Deller MJ, Warshaw JB (1981) Functional EGF receptors are present on mouse embryo tissues. Nature 291 : 656-659 Ardinger HH, Buetow KH, Bell GI, Bardach J, Van DeMark DR, Murray JC (1989) Association of genetic variation of the transforming growth factor-alpha gene with cleft lip and palate. Am J Hum Geriet 45 : 348-353 Catterton WZ, Escobedo MB, Sexson WR, Gray ME, Sundell HW, Stahlman MT (1979) Effect of epidermal growth factor on lung maturation in fetal rabbits. Pediatr Res 13:10B108 Cohen S (1986) Epidermal growth factor. Bioscience Reps 6:10171028 Cohen S, Ushiro H, Stoscheck C, Chinkers M (1982) A native 170,000 epidermal growth factor receptor-kinase complex from shed plasma membrane vesicles. J Biol Chem 257:1523 1531 Diewert VM (1978) A quantitative coronal plane evaluation of craniofacial growth and spatial relations during secondary palate development in the rat. Arch Oral Biol 23 : 607-629 Ferguson MWJ (1987) Palate development: mechanisms and malformations. Ir J Med Sci 156:309-315 Ferguson MWJ (1988) Palate development. Development [Suppl 103] :41-60 Ferguson MWJ, Honig LS (1984) Epithelial-mesenchymal interactions during vertebrate palatogenesis. In: Zimmerman EF (ed) Current topics in developmental biology, vol 19. Palate development: normal and abnormal, cellular and molecular aspects. Aeademie Press, New York, pp 137 164 Freemark M, Comer M (1987) Epidermal growth factor (EGF)-like transforming growth factor (TGF) activity and EGF receptors in ovine fetal tissues: possible role for TGF in ovine fetal development. Pediatr Res 22: 609 615 Green MR, Couchman JR (1985) Differences in human skin between the epidermal growth factor receptor distribution detected by EGF binding and monoelonal antibody recognition. J Invest Derm 85:239-245 Greene RM, Pratt RM (1976) Developmental aspects of secondary palate formation. J Embryol Exp Morphol 36:225 245 Hassell JR (1975) The development of rat palatal shelves in vitro. An ultrastructural analysis of the inhibition of epithelial cell death and palate fusion by the epidermal growth factor. Dev Biol 45 : 90-102 Hassell JR, Pratt RM (1977) Elevated levels of cAMP alter the effect of epidermal growth factor in vitro on programmed cell
death in the secondary palatal epithelium. Exp Cell Res 106:5562 Hollenberg MD (1979) Epidermal growth factor-urogastrone, a polypeptide acquiring hormonal status. Vitam Horm 37:69110 Lee DC, Rochford R, Todaro GJ, Villarreal LP (1985) Developmental expression of rat transforming growth factor-e mRNA. Mol Cell Biol 5 : 3644-3646 Marquardt H, Hunkapiller MW, Hood LE, Todaro GJ (1984) Rat transforming growth factor type I: structure and relation to epidermal growth factor. Science 223 : 1079-1082 Matrisian LM, Pathak M, Magun BE (1982) Identification of an epidermal growth factor-related transforming growth factor Dom rat fetuses. Biochem Biophys Res Commun 107:761-769 Mesiano S, Browne CA, Thorburn GD (1985) Detection of endogenous epidermal growth factor-like activity in the developing chick embryo. Dev Biol 110: 23-28 Nexo E, Hollenberg MD, Figueroa A, Pratt RM (1980) Detection of epidermal growth factor-urogastrone and its receptor during fetal mouse development, Proc Natl Acad Sci USA 77:27822785 Pratt RM (1987) Role of epidermal growth factor in embryonic development. In: Sawyer RH (ed) Current topics in developmental biology, vo122. The molecular and developmental biology of keratins. Academic Press, New York, pp 175-193 Rappolee DA, Brenner CA, Schultz R, Mark D, Werb Z (1988) Developmental expression of PDGF, TGF-c~ and TGF-/? genes in pre-implantation mouse embryos. Science 241 : 1823-1825 Sharpe PM, Ferguson MWJ (1988) Mesenchymal influences on epithelial differentiation in developing systems. J Cell Sci [Suppl 10] : 195-230 Silver MH, Murray JC, Pratt RM (1984) Epidermal growth factor stimulates type-V collagen synthesis in cultured murine palatal shelves. Differentiation 27: 205-208 Sundell HW, Gray ME, Serenius FG, Escobedo MB, Stahlman MT (1980) Effects of epidermal growth factor on lung maturation in fetal lambs. Am J Pathol 100:707 726 Theiler K (1972) The house mouse. Development and abnormal stages from fertilization to four weeks of age. Springer, Berlin New York Todaro GJ, Fryling C, De Larco JE (1980) Transforming growth factors produced by eertain human tumor cells: polypeptides that interact with epidermal growth factor receptors. Proc Natl Acad Sci USA 77:5258-5262 Turley EA, Hollenberg MD, Pratt RM (1985) Effect of epidermal growth factor/urogastrone on glycosaminoglycan synthesis and accumulation in vitro in the developing mouse palate. Differentiation 28:279-285 Twardzik DR (1985) Differential expression of transforming growth faetor ~ during prenatal development of the mouse. Cancer Res 45:5413-5416 Twardzik DR, Ranchalis JE, Todaro GJ (1982) Mouse embryonic transforming growth factors related to those isolated from tumor cells. Cancer Res 42: 590-593 Weller A, Meek J, Adamson ED (1987) Preparation and properties of monoclonal and polyclonal antibodies to mouse epidermal growth factor (EGF) reeeptors: evidence for cryptic receptors in embryonal carcinoma cells. Development 100:351-363 Wilcox JN, Derynck R (1988) Developmental expression of transforming growth factors alpha and beta in mouse fetus. Mol Cell Biol 8:3415-3422 Yoneda T, Pratt RM (1981) Mesenchymal cells from the human embryonic palate are highly responsive to epidermal growth factor. Science 213 : 563-565
Anatomy and Embryology © Springer-Verlag 1991
Immunolocalisation of epidermal growth factor (EGF), EGF receptor and transforming growth factor alpha (TGF~) during murine palatogenesis in vivo and in vitro M.J. Dixon, J. Garner, and M.W.J. Ferguson Animal and Human Reproduction, Development and Growth Research Group, Department of Cell and Structural Biology, Stopford Building, University of Manchester, Manchester M13 9PT, UK Accepted April 23, 1991
Summary. The distribution of epidermal growth factor, the epidermal growth factor receptor and transforming growth factor alpha during murine palatogenesis was investigated immunocytochemically. On embryonic day 12 staining for transforming growth factor alpha was present throughout the palatäl mesenchyme, with little in the epithelia. On embryonic day 13 staining increased in the palatal epithelia and in the mesenchyme at the tip of the palate. As the palatal shelves fused together (embryonic day 14.5) intense staining for transforming growth factor alpha was seen in the midline epithelial seam and in the subjacent mesenchyme. On embryonic day 15 there was a generalised increase in palatal epithelial staining; this was most marked in the remnants of the degenerating epithelial seam. Mesenchymal staining was, however, uniform. Whilst palatal staining for epidermal growth factor was sparse, at all stages, staining for its receptor was present throughout the palatal epithelia and mesenchyme. This was most intense in the palatal medial edge epithelia at the time of midline epithelial seam degeneration. The regional and temporal differences in staining for the epidermal growth factor receptor and transforming growth factor alpha suggested that these molecules may play an important role in normal palate development in vivo, particularly in degeneration of the midline epithelial seam. Key words: Palate - Epidermal growth factor - Transforming growth factor alpha - Growth factors
Introduction Epidermal growth factor (EGF) consists of a single polypeptide chain of 53 amino acids and molecular weight 6,045 daltons. It has numerous biological effects, including the control of cellular proliferation, differentiation and extracellular matrix (ECM) synthesis (see reviews Offprint requests to: M.J. Dixon
by Hollenberg 1979; Cohen 1986). It exerts its actions by binding at an integral membrane glycoprotein receptor of 170 Kd (Cohen et al. 1982). The presence of functional EGF receptors on embryonic and extraembryonic tissue (Adamson et al. 1981; Adamson and Warshaw 1982; Adamson and Meek 1984) has suggested that EGF may play an important role in normal embryonic development, particularly during development of the secondary palate (Nexo et al. 1980). The palate arises as bilateral outgrowths from the maxillary proeesses (embryonic day 12 in mice). The outgrowths initially grow vertieally down the sides of the tongue (embryonic day 13 in reite). Each palatal shelf consists of a central core of neural crest-derived mesenchyme surrounded by a simple, histologically undifferentiated epithelium which is two- to three cells thick (Greene and Pratt 1976; Ferguson 1988). At a precise stage (embryonic day 14.5 in reite) the shelves rapidly elevate to a horizontal position above the tongue and contact each other. The medial edge epithelia (MEE) fuse together to form a midline epithelial seam which degenerates by cell death and epithelial-mesenchymal transformation (Ferguson 1988), allowing mesenehymal continuity across the palate. Simultaneously, the epithelia along the nasal and oral aspects of the palate differentiate into pseudostratified, ciliated columnar cells, and stratified, squamous keratinising cells respectively (Greene and Pratt 1976). Regional palatal epithelial differentiation is specified by the mesenchyme (Ferguson and Honig 1984). The mechanism of signalling is thought to involve the interaction of ECM molecules and soluble factors (Ferguson 1987, 1988). In vitro exogenous EGF affects palatal epithelial differentiation (Hassell 1975; Hassell and Pratt 1977; Abbott and Pratt 1987a, b; Sharpe and Ferguson 1988) mesenchymal cell division (Yoneda and Pratt 1981) and ECM synthesis (Silver et al. 1984; Turley et al. 1985). However, whilst EGF receptors have been detected from embryonic day 11 in mice, and increase in number up to embryonic day 18, particularly in the secondary palate, EGF itself is barely detectable by radioimmunoassay
84
before embryonic day 14.5. This suggests that the fetal form of mouse EGF differs from the adult (Nexo et al.
frontal cryosections were cut at 6 gm in a Leitz 1720 cryostat (Leitz-Wetzlar) at - 2 0 ° C. The sections were fixed in acetone for 5 min, air-dried and stored at - 7 0 ° C until used for staining. Representative sections from the anterior, middle and posterior regions of the developing palate (Diewert 1978) were selected. The primary antibodies (Table 1) were applied directly to the sections and incubated at 4° C in a humid atmosphere for 48 h, followed by three 5-minute washes in phosphate-buffered saline (PBS). The sections were blocked for 1 h with 0.5 M sodium chloride buffer containing 0.1% bovine serum albumin (Sigma Chemical Company, Poole), 0.05% Tween 20 (Sigma) and 5% fetal calf serum (Flow Laboratories, Rickmansworth). Excess blocking buffer was removed, and the appropriate gold-conjugated secondary antibody (Table 2) applied. After an incubation for 1 h at room temperature the sections were rewashed in PBS as above, and fixed overnight at 4° C in a solution of i % paraformaldehyde/1% glutaraldehyde (Agar Aids, London) in PBS. After fixation the slides were rinsed in PBS, then given ten 15-minute washes in deionised water. The sections were then silver enhanced using Intense M (ICN Biomedicals, High Wycombe). The enhancement procedure was monitored microscopically and stopped by washing in deionised water when a suitable signal/noise ratio had been achieved. The sections were then counterstained with erythrocin B, dehydrated, cleared and mounted in Eukitt synthetic mounting medium (BDH, Poole). Control sections were established by replacing the primary antibody with either serum from the animal in which it was raised (Table 1) or PBS, or by preabsorption of the primary antisera with either EGF or TGF«. The sections were viewed by bright field illumination using a Leitz Dialux 22EB microscope (Leitz-Wetzlar). Photographs were taken on Technical Pan film (Ilford; rated at I00 ASA) from which black and white prints were made.
1980). Transforming growth factor alpha (TGF«), which acts via the EGF receptor (Todaro et al. 1980) has been implicated as the embryonic homologue of EGF (Twardzik et al. 1982; Matrisian et al. 1982; Twardzik ]985; Freemark and Comer 1987). We have, therefore, immunolocalised EGF receptor, EGF and TGF~ in the developing mouse palate with the aim of investigating whether their distribution changes with developmental time, and whether these changes correlate with known morphogenetic/differentiativeevents.
Materials and methods MF strain mice were mated overnight. Fertilisation was assumed to have occurred at midnight prior to the day on which vaginal plugs were detected. Pregnant females were killed by ether overdose, and the gravid uteri were aseptically removed to sterile Hank's balanced salt solution (Flow Laboratories, Rickmansworth). Individual embryos were dissected from the uterine decidua, staged by the external features described by Theiler (1972) and decapitated. Palates were dissected from Theiler stage 21 (embryonic day 13) mouse heads, explanted onto the surface of preformed collagen gels either singly or in pairs, and grown in submerged culture for 24, 48 or 72 h. The culture media consisted of: 1. A 1 : 1 mixture of Dulbecco's modification of Eagle's Medium and Ham's F-12 medium (DMEM/F12) (Flow Laboratories, Rickmansworth) 2. DMEM/F12 supplemented with 2.5% donor calf serum (DCS) 3. DMEM/F12 plus 10 ng/ml TGFc~ (Peninsula Laboratories) 4. DMEM/F12 plus 10 ng/ml TGFΠplus 2.5% DCS Embryonic mouse heads (days 12 to 15; Theiler stages 20 to 23) or cultured palates were immediately snap-frozen in liquid nitrogen-cooled isopentane. The tissue was subsequently embedded in OCT embedding compound (BDH Chemicals, Poole) and serial
Tal)le 1. Primary antibodies used
Table 2. Gold-conjugated secondary anti-
bodies used
Results
In vivo staining patterns Embryonic day 12. S t a i n i n g for the E G F receptor is evenly d i s t r i b u t e d t h r o u g h o u t the p a l a t a l epithelia a n d mes-
Primary antibody raised against
Primary antibody raised in
Dilution used
Supplier
Secondary antibody (see Table 2)
TGFc~ (human) EGF (mouse) EGF-receptor (A431 cells)
Goat
1 :40
1
Rabbit
1 : 40
Biotope, Seattle ICN Biomedicals, High Wycombe
Mouse (monoclonal)
1:40
Table 1 reference no.
Secondary antibody raised against
Secondary antibody raised in
Goat IgG
Rabbit
Rabbit IgG Mouse IgG
Goat Goat
ICN Biomedicals, High Wycombe
2 3
Dilution
Supplier
5 nm
1:40
10 nm 10 nm
1:25
ICN Biomedicals, High Wycombe Sigma, Poole Sigma, Poole
Size of particle (nm)
1:25
85
Fig. 1 A, B. Frontal sections through embryonic day 12 mouse heads. A Staining for the EGF receptor is evenly distributed throughout the palatal epithelia (e) and rnesenchyme (m). Bar 100 ~tm. B Staining for TGF~ is also evenly distributed throughout the palatal mesenchyme (m); little staining is present in the epithelium (e). Bar 150 Bm Fig. 2A, B. Frontal sections through embryonic day 13 mouse heads stained for the EGF receptor. A Staining is present throughout the palate (p) and is intense in the enamel organ (eo) of the incisor tooth germ. Bar 400 gin. B Palatal staining is present throughout the epithelia (e) and mesenchyme (m), and is most intense along the basal surfaces of the epithelium and along the epithelial-mesenchymal interface (arrowed). Bar 150 gm Fig. 3A, B. Frontal sections through embryonic day 13 mouse heads. A Staining for TGFc~ is present in the palatal epithelia (e) and mesenchyme (m), mesenchymal staining being most intense at the palatal tip (arrowed). Bar 200 gin. B Staining for EGF is evenly distributed throughout the palate (p) but is sparse. Bar 150 ~m
86 enchyme (Fig. 1 A). Whilst intense staining is present in the epithelia of the nasal passages, especially superficially, staining of the tongue, lateral maxilla and mandible is of similar intensity to that prescnt in the palate. Strong staining for TGFc~ is present throughout the palatal mesenchyme, little staining is, however, present in the palatal epithelia (Fig. 1 B). Similarly, little staining is present in the epithelia of the floor of the mouth, tongue or lateral maxilla. The epithelia of the nasal passages are, however, intensely stained, especially superficially. Conversely, staining for EGF is sparse, the mesenchyme being more intensely stained than the epithelia.
Embryonic day 13. Intense staining for the EGF receptor is present in the enamel organ of the incisor and molar tooth germs (Fig. 2A). This is most marked in the central region, which ultimately forms the stellate reticulum, and along the basal surfaces of the future internal and external enamel epithelia. Intense staining is also present in the epithelia of the nasal passages and vomeronasal organs, especially superficially. Staining for the EGF receptor is present throughout the palatal epithelia and mesenchyme, and is most intense along the basal surfaces of the epithelial cells and at the epithelial-mesenchymal interface (Fig. 2A, B). By embryonic day 13 epithelia, including that of the palate, stain more intensely for TGFc« Staining in the mesenchyme at the tip of the palatal shelf is more intense than that in the remainder of the palatal mesenchyme (Fig. 3 A). Intense staining is also present along the epithelial-mesenchymal interface of the palate and the floor of the mouth. Intense staining persists in the epithelia of the nasal passages and vomeronasal organs. It is also present in the central region of the enamel organ and the dental papilla mesenchyme of the tooth germs, although this is not as marked as for the EGF receptor. Staining for EGF remains sparsely and evenly distributed throughout the palate (Fig. 3 B). Unlike the EGF receptor and TGF~ there is no increased staining in the enamel organ. Embryonic day 14.5. Whilst staining for the EGF receptor is present throughout the palatal mesenchyme, that in the palatal epithelia is strongest basally (Fig. 4). Overall, there is little change in the staining pattern from embryonic day 13. However, there is a marked increase in staining in the epidermis and in the mesenchyme at the base of the nasal septum (Fig. 4). Palatal staining for TGFc~ is differentially expressed in an antero-posterior sequence. In the most anterior regions, where the palatal shelves have elevated and come into contact, but have not fused, staining is fairly evenly distributed throughout the epithelium (Fig. 5A). Mesenchymal staining is intense and most marked at the tips of the approximated shelves (Fig. 5A). Intense staining is also present in the blood vessels. Further posteriorly where the MEE of apposing shelves have fused together, extremely intense staining is seen along the midline epithelial seam (Fig. 5 B), particularly along the basal epithelial surfaces (Fig. 5 C), and in the subjacent
mesenchyme. A lower level of staining is present in the remainder of the palatal epithelia and mesenchyme. As with the EGF receptor, intense staining for TGF« is present in the epithelia of the nasal passages, the central region of the enamel organ (Fig. 5D), the dental papilla mesenchyme (Fig. 5D) and the mesenchyme at the base of the nasal septum. Staining for EGF is generally sparse, although it is most marked in the epithelia of the nasal passages and in the central region of the enamel organ. Staining is present throughout the palatal mesenchyme, with little staining in the epithelia (Fig. 6).
Embryonic da), 15. The staining patterns obtained with the anti-EGF receptor antibody were similar to those recorded for embryonic day 14.5, high levels of staining being present in the epithelia of the nasal passages and in the enamel organs. Staining is present throughout the palatal epithelia and mesenchyme, and is intense in the remnants of the degenerating epithelial seam (Fig. 7A,
B). Staining for TGFŒ is present in the epithelia of the nasal passages (Fig. 8 A), the central region of the enamel organ, the dental papilla mesenchyme and its interface with the internal enamel epithelium. Intense staining is also present in the epidermis. There is a generalised increase in epithelial staining for TGF«. This is marked in the dorsum of the tongue and in the palate, particularly high levels being detected in the remnants of the degenerating epithelial seam, at the nasal angle of the palate (Fig. 8 A, B) and along the epithelial-mesenchymal interface. By this stage of development, however, staining is uniform in the palatal mesenchyme; an area of increased staining adjacent to the degenerating seam is no longer apparent. Staining for EGF is increased from earlier stages. It is seen throughout the palatal mesenchyme and epithelia, that in the mesenchyme being more intense than that in the epithelium. Other staining patterns remain unchanged from earlier stages.
In vitro staining patterns The staining patterns obtained in vitro are the same as those obtained in vivo. Staining for the EGF receptor is evenly distributed throughout the palatal epithelia and mesenchyme on culture days 0 and 1 (equivalent of embryonic days 13 and 14 in vivo). Staining becomes more intense in the medial edge epithelium as it degenerates. On the day of explantation staining for TGFΠis uniformly distributed throughout the palatal epithelia; a small area of more intense staining is present in the mesenchyme at the palatal tip. As the shelves fuse (culture day 2) intense staining is present along the midline epithelial seam; this persists as the seam degenerates (culture day 3) (Fig. 9 A). Where TGFc~ is added to the culture medium in the presence of serum and MEE degeneration is inhibited (as described by Ferguson 1988; Sharpe and Ferguson
87
Fig. 4. Frontal seetion through an embryonic day 14.5 mouse head stained for EGF receptor. Intense staining is present in the palate (p) and in the nasal septum (ns) but is less intense in the tongue (t). Bar 150 gm
Fig. 5A-D. Frontal sections through embryonic day 14.5 mouse heads stained for TGFc« A Anteriofly, where the palatal shelves (p) are in contaet, but have not fused, epithelial staining is uniform. Intense staining is present in the mesenehyme beneath the medial
1988) the M E E does not stain intensely for T G F « , regardless of time in culture (Fig. 9 B). All control staining was negative (Fig. 10).
Diseussion Previous studies have implicated E G F in normal embryonic development (Adamson 1983). For example, E G F has been shown to prevent hyaline membrane disease in fetal lambs and rabbits by promoting epithelial growth and differentiation in the developing lung (Cat-
edge epithelium (arrowed), at the base of the nasal septum (ns) and in the blood vessels (bv). Bar 200 gin. B Further posteriorly, intense staining is present in the epithelium at the base of the nasal septum (ns) and along the midline palatal epithelial seam (s). Bar 100 ~tm. C With continued development, staining of the midline epithelial seam (s) becomes localised to the basal cells (arrowed). Bar 200 gin. D Intense staining is also present in the molar tooth germs, particularly in the central region of the enamel organ (eo) but also in the dental papilla mesenchyme (dp). Bar 150 Ixm
terton et al. 1979; Sundell et al. 1980). E G F receptors have been detected on murine embryonic tissue (Adamson 1983). The earliest time of detection is on trophoblast cells of a 5-day blastocyst cultured for three days (Adamson and Meek 1984). These receptors are functional in that they bind E G F and are down-regulated by excess E G F injected into the placenta or amniotic cavity (Adamson and Warshaw 1982). Nexo et al. (1980) used both a radioreceptor and radioimmunoassay to estimate the content of E G F in embryonic day 11.5 to 17.5 mouse embryos. E G F receptor binding was readily detected in the 11.5 day embryos,
88
Fig. 6. Frontal section through an embryonic day 14.5 mouse head. Staining for EGF is sparsely distributed throughout the palate (p). Bar 200 gm Fig. 7A, B. Frontal sections through an embryonicday 15 mouse head. Staining for the EGF receptor is present throughout the palatal epithelia (e) and mesenchyme (m), and is particularly intense in the remnants of the degeneratingepithelial seam (arrowed). Bar, A 150 gm, B 50 grn and rose steadily up to parturition, the rise being most marked in potential target tissues such as the secondary palate. However, in whole embryo extracts far greater amounts of EGF were detected by radioreceptor than by radioimmunoassay, suggesting that the fetal form of mouse EGF differs from the adult form. The nature of the fetal form of EGF suggested by Nexo et al. (1980) is uncertain. TGFc~, which shares considerable homology with EGF (Marquardt et al. 1984) and acts via the EGF receptor (Todaro et al. 1980), has recently been demonstrated in extracts of embryonic mice (Twardzik et al. 1982; Twardzik 1985), rats (Matrisian et al. 1982), sheep (Freemark and Comer 1987) and chicks (Mesiano et al. 1985). TGFc~ mRNA has also
been detected in pre-implantation mouse embryos (Rappolee et al. 1988). Twardzik (1985) has further demonstrated that there is an increase in TGF« expression between embryonic days 12 and 15 in the mouse. By contrast it has been reported that TGFe mRNA is not present at significant levels beyond day 10 ofmouse development (Lee et al. 1985; Wilcox and Derynck 1988). A1though this last study could not exclude the possibility that small populations of cells synthesised TGFe at levels below the detection threshold of the assay. TGFc~ may therefore be the fetal form of EGF. In the present study we have shown that on embryonic day 12 TGFc~ is localised throughout the palatal mesenchyme but is sparse in the epithelium. By embryonic
89
•
«
«
i,9' ~'«
,2
9B
m
10
P
P
Fig. 8A, B. Frontal sections through an embryonic day 15 mouse head stained for TGFc« Intense staining is present in the epithelia of the nasal passages (np) and tongue (t). This is most marked in the remnants of the degenerating epithelial seam (arrowed) and in the epithelia at the nasal angle of the palate (n). Bar, A 150 gm, B 25 gm Fig. 9A, B. Transverse sections through cultured palatal shelves stained for TGFcc A After 72 h of culture in serum/TGFc~-free
medium, staining is present throughout the palate, but is particularly intense in the midline epithetial seam (s). Bar 200 gm. B After 72 h of culture, in medium supplemented with 10 ng/ml TGFc~ and 2.5% donor calf serum, increased staining in the persistent medial edge epithelium (m) is not detected. Bar 100 lam Fig. 10. Frontal section through embryonic day 14.5 mouse head stained with secondary antibody alone (control section). The palates (p) are completely unstained. Bar 150 gm
90
day 13 staining is more intense in the palatal epithelium and in the mesenchyme at the tip of the shelf. After palatal shelf elevation there is a marked increase in TGF« staining particularly in the MEE during epithelial seam formation. As the epithelial seam degenerates there is a generalised increase in epithelial staining, particularly high levels being detected in the seam remnants. Conversely, over the corresponding period, staining for EGF remained sparse, whilst staining for the EGF receptor increased on the palatal MEE around the time of seam degeneration. TGF:~ and its interaction with the EGF receptor therefore appear important in palate development: both are temporally and spatially regulated, EGF is not. The finding that strong staining for the EGF receptor is detected in the MEE around the time of midline epithelial seam degeneration does not appear to agree with the results of Abbott et al. (1988) who reported that the MEE cells ceased to express EGF receptors late on embryonic day 13. This apparent anomaly may be due to differences in the properties of the antibodies used to detect the EGF receptor. Abbott et al. (1988) used a rabbit polyclonal antibody to EGF receptors purified from mouse liver. This does not react well with ligandoccupied receptors (Weller et al. 1987). Our antibody recognises the receptor regardless of whether it is occupied by ligand (EGF or TGFe) or not. Moreover, we have co-localised EGF and TGFc« Whilst EGF is sparsely distributed, TGF« is, for the most part, present in the same locations as the EGF receptor, which suggests that receptor-bound ligand was being recognised. Abbott et al. (1988) did not colocalise EGF or TGFe, rather they assessed EGF binding by [125I]-EGF autoradiography, which will localise unoccupied, accessible, cell-surface EGF receptors (see Green and Couchman 1985). Moreover, 12SI-EGF may not compete oft, bound, unlabelled EGF or TGFc« The absence of staining on the MEE and subjacent mesenchyme reported by Abbott et al. (1988) probably refiects a large number of EGF receptors with bound ligand rather than the absence of such receptors. Previous work has suggested that, in vitro, exogenous EGF causes cleft palate by preventing MEE cell death (Hassell 1975; Hassell and Pratt 1977; Abbott and Pratt 1987a, b). These studies used either pharmacological doses of EGF (up to 2 ~tg/ml) or serum-supplemented culture medium. EGF and TGFc~ have also been shown to stimulate cell proliferation in the developing palate (Yoneda and Pratt 1981). These results led Pratt (1987) to propose that on embryonic day 12 the palatal mesenchymal and oral and medial edge epithelial cells produce TGFc~, which then supports rapid growth of these cells during early palatal development. The MEE cells subsequently cease to produce TGF~ and, as a result, cease to synthesise DNA and die. By contrast, the oral epithelial cells continue to produce TGFŒ and differentiate into stratified, squamous keratinising cells. Our findings have, however, shown that intense staining for TGFc~ and the EGF receptor is present in the midline epithelial seam even after it has started to degenerate. This may
be in keeping with the observation that seam degeneration involves epithelial-mesenchymal transformation, and not just cell death as previously imagined (Ferguson 1988). Recent reports have suggested that TGFe stimulates ECM biosynthesis in either palatal organ cultures or in cultures of isolated palatal mesenchymal or epithelial cells (see reviews by Ferguson 1988; Sharpe and Ferguson 1988). It is particularly interesting that in epithelial cell culture TGFc~ causes an increase in staining for tenascin and type IX collagen in the MEE, as these molecules have been implicated in MEE degeneration (Ferguson 1988). The data from this study combined with those from cell/organ culture studies lead us to believe that the previously postulated developmental decline of EGF (or TGF«) and its receptor during medial edge epithelial seam formation is incorrect (Pratt 1987). Normally the levels of TGFc~ and the EGF receptor increase in the midline palatal epithelial seam. The TGF« is probably being synthesised by the mesenchyme, and binding to EGF receptors in the seam cells. Previous experiments, using pharmacological doses of EGF/TGF~ to induce cleft palate in vitro by inhibiting MEE degeneration (Pratt 1987), probably cause a perturbation of the above normal mechanisms, perhaps by down-regulation of the EGF receptor. This developmentally regulated, stage- and regionspecific event is important in signalling the mesenchymal-epithelial interactions which specify midline epithelial seam degeneration. The signalling may result from a combination of: 1. Direct effects on the epithelial cells e.g. on keratin profile 2. Stimulation of specific ECM molecule biosynthesis by the underlying mesenchyme cells, and the effects of such ECM molecules on the epithelial cells (Ferguson 1988) 3. Stimulation of ECM receptors on the epithelial cells (Sharpe and Ferguson 1988) 4. Stimulation of specific ECM molecule biosynthesis by the palatal epithelial cells, thus minimising the differences in the ECM interface between the palatal epithelial and mesenchymal cells. The result of this integrated growth factor/ECM signalling mechanism (Sharpe and Ferguson 1988) is that the basal epithelial cells migrate out of the seam and transform into mesenchyme cells (Ferguson 1988) whilst the post-mitotic suprabasal cells die. Physiological levels of TGF~ and its interaction with the EGF receptor appear to be important for normal palatal development. In this context it is perhaps significant that a genetic association has been observed between variation of the TGF« locus and the occurrence of clefting in human caucasian families (Ardinger et al. 1989). Acknowledgements. This work was supported by grants from the Wellcome Trust, Medical Research Council, Action Research for the Crippled Child and Birthright. M.J. Dixon was funded by a Wellcome Trust Research Training Fellowship grant.
91
References Abbott BD, Pratt RM (1987a) Retinoids and epidermal growth factor alter embryonic mouse palatal epithelial and mesenchymal cell differentiation in organ culture. J Craniofac Genet Der Biol 7:219-240 Abbott BD, Pratt RM (1987b) Human embryonic palatal epithelial differentiation is altered by retinoic acid and epidermal growth factor in organ culture. J Craniofac Genet Dev Biol 7:241-265 Abbott BD, Adamson ED, Pratt RM (1988) Retinoic acid alters EGF receptor expression during palatogenesis. Development 102:853-867 Adamson ED (1983) Growth factors in development. In: Warshaw JB (ed) The biological basis of reproductive and developmental medicine. Elsevier, New York, pp 307-336 Adamson ED, Meek J (1984) The ontogeny of epidermal growth factor receptors during mouse development. Dev Biol 103:6270 Adamson ED, Warshaw JB (1982) Down-regulation of epidermal growth factor receptors in mouse embryos. Dev Biol 90:430434 Adamson ED, Deller MJ, Warshaw JB (1981) Functional EGF receptors are present on mouse embryo tissues. Nature 291 : 656-659 Ardinger HH, Buetow KH, Bell GI, Bardach J, Van DeMark DR, Murray JC (1989) Association of genetic variation of the transforming growth factor-alpha gene with cleft lip and palate. Am J Hum Geriet 45 : 348-353 Catterton WZ, Escobedo MB, Sexson WR, Gray ME, Sundell HW, Stahlman MT (1979) Effect of epidermal growth factor on lung maturation in fetal rabbits. Pediatr Res 13:10B108 Cohen S (1986) Epidermal growth factor. Bioscience Reps 6:10171028 Cohen S, Ushiro H, Stoscheck C, Chinkers M (1982) A native 170,000 epidermal growth factor receptor-kinase complex from shed plasma membrane vesicles. J Biol Chem 257:1523 1531 Diewert VM (1978) A quantitative coronal plane evaluation of craniofacial growth and spatial relations during secondary palate development in the rat. Arch Oral Biol 23 : 607-629 Ferguson MWJ (1987) Palate development: mechanisms and malformations. Ir J Med Sci 156:309-315 Ferguson MWJ (1988) Palate development. Development [Suppl 103] :41-60 Ferguson MWJ, Honig LS (1984) Epithelial-mesenchymal interactions during vertebrate palatogenesis. In: Zimmerman EF (ed) Current topics in developmental biology, vol 19. Palate development: normal and abnormal, cellular and molecular aspects. Aeademie Press, New York, pp 137 164 Freemark M, Comer M (1987) Epidermal growth factor (EGF)-like transforming growth factor (TGF) activity and EGF receptors in ovine fetal tissues: possible role for TGF in ovine fetal development. Pediatr Res 22: 609 615 Green MR, Couchman JR (1985) Differences in human skin between the epidermal growth factor receptor distribution detected by EGF binding and monoelonal antibody recognition. J Invest Derm 85:239-245 Greene RM, Pratt RM (1976) Developmental aspects of secondary palate formation. J Embryol Exp Morphol 36:225 245 Hassell JR (1975) The development of rat palatal shelves in vitro. An ultrastructural analysis of the inhibition of epithelial cell death and palate fusion by the epidermal growth factor. Dev Biol 45 : 90-102 Hassell JR, Pratt RM (1977) Elevated levels of cAMP alter the effect of epidermal growth factor in vitro on programmed cell
death in the secondary palatal epithelium. Exp Cell Res 106:5562 Hollenberg MD (1979) Epidermal growth factor-urogastrone, a polypeptide acquiring hormonal status. Vitam Horm 37:69110 Lee DC, Rochford R, Todaro GJ, Villarreal LP (1985) Developmental expression of rat transforming growth factor-e mRNA. Mol Cell Biol 5 : 3644-3646 Marquardt H, Hunkapiller MW, Hood LE, Todaro GJ (1984) Rat transforming growth factor type I: structure and relation to epidermal growth factor. Science 223 : 1079-1082 Matrisian LM, Pathak M, Magun BE (1982) Identification of an epidermal growth factor-related transforming growth factor Dom rat fetuses. Biochem Biophys Res Commun 107:761-769 Mesiano S, Browne CA, Thorburn GD (1985) Detection of endogenous epidermal growth factor-like activity in the developing chick embryo. Dev Biol 110: 23-28 Nexo E, Hollenberg MD, Figueroa A, Pratt RM (1980) Detection of epidermal growth factor-urogastrone and its receptor during fetal mouse development, Proc Natl Acad Sci USA 77:27822785 Pratt RM (1987) Role of epidermal growth factor in embryonic development. In: Sawyer RH (ed) Current topics in developmental biology, vo122. The molecular and developmental biology of keratins. Academic Press, New York, pp 175-193 Rappolee DA, Brenner CA, Schultz R, Mark D, Werb Z (1988) Developmental expression of PDGF, TGF-c~ and TGF-/? genes in pre-implantation mouse embryos. Science 241 : 1823-1825 Sharpe PM, Ferguson MWJ (1988) Mesenchymal influences on epithelial differentiation in developing systems. J Cell Sci [Suppl 10] : 195-230 Silver MH, Murray JC, Pratt RM (1984) Epidermal growth factor stimulates type-V collagen synthesis in cultured murine palatal shelves. Differentiation 27: 205-208 Sundell HW, Gray ME, Serenius FG, Escobedo MB, Stahlman MT (1980) Effects of epidermal growth factor on lung maturation in fetal lambs. Am J Pathol 100:707 726 Theiler K (1972) The house mouse. Development and abnormal stages from fertilization to four weeks of age. Springer, Berlin New York Todaro GJ, Fryling C, De Larco JE (1980) Transforming growth factors produced by eertain human tumor cells: polypeptides that interact with epidermal growth factor receptors. Proc Natl Acad Sci USA 77:5258-5262 Turley EA, Hollenberg MD, Pratt RM (1985) Effect of epidermal growth factor/urogastrone on glycosaminoglycan synthesis and accumulation in vitro in the developing mouse palate. Differentiation 28:279-285 Twardzik DR (1985) Differential expression of transforming growth faetor ~ during prenatal development of the mouse. Cancer Res 45:5413-5416 Twardzik DR, Ranchalis JE, Todaro GJ (1982) Mouse embryonic transforming growth factors related to those isolated from tumor cells. Cancer Res 42: 590-593 Weller A, Meek J, Adamson ED (1987) Preparation and properties of monoclonal and polyclonal antibodies to mouse epidermal growth factor (EGF) reeeptors: evidence for cryptic receptors in embryonal carcinoma cells. Development 100:351-363 Wilcox JN, Derynck R (1988) Developmental expression of transforming growth factors alpha and beta in mouse fetus. Mol Cell Biol 8:3415-3422 Yoneda T, Pratt RM (1981) Mesenchymal cells from the human embryonic palate are highly responsive to epidermal growth factor. Science 213 : 563-565
