Archs oral Biol. Vol. 37, No. 5, pp. 395410, Printed in Great Britain. All rights reserved

1992

0003-9969/92 $5.00 + 0.00 Copyright 0 1992 Pergamon Press Ltd

THE EFFECTS OF EPIDERMAL GROWTH FACTOR, TRANSFORMING GROWTH FACTORS ALPHA AND BETA AND PLATELET-DERIVED GROWTH FACTOR ON MURINE PALATAL SHELVES IN ORGAN CULTURE M. J. DIXON* 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 Ml3 9PT, U.K. (Received 29 January

1991; accepted 12 November 1991)

Summary-Palatal shelves isolated from day-13 embryonic mice were explanted on to the surfaces of collagen gels either singly or in pairs with their medial edges in contact, and cultured submerged in a 1 : I mixture of Dulbecco’s modified Eagle’s medium/Ham’s F12 medium. The medium was supplemented with either long/ml epidermal growth factor (EGF), long/ml transforming growth factor alpha (TGFa), 1ng/ml transforming growth factor beta (TGFP,) or 2 ng/ml platelet-derived growth factor (PDGF) all in the presence or absence of 2.5% donor calf serum (DCS). Cultures were terminated after 0, 24, 48 or 72 h and processed for histological and immunocytochemical examination. In serum-free medium and medium supplemented with 2.5% DCS the palatal epithelia differentiated in a manner similar to that seen in vir)o (oral, keratinization; nasal, pseudostratified, ciliated columnar cells and medial edge, epithelial degeneration). A similar pattern was obtained in serum-free medium supplemented with either EGF or TGFa. However in cultures with either EGF or TGFa plus 2.5% DCS present in the medium, medial-edge epithelial degeneration was inhibited and the oral epithelia were more heavily keratinized. The mesenchyme of such cultures stained more intensely for various extracellular matrix molecules. In TGFfi,-supplemented cultures (with, but especially without, serum supplementation) the epithelia were thin, medial-edge epithehal degeneration was marked, and the fibronectin content of the mesenchyme was increased. PDGF prevented medial-edge epithelial degeneration in the presence, but not in the absence, of serum; mesenchymal extracellular molecules were not as prevalent as with the EGF treatment. These results indicate that exogenous growth factors (including those present in serum) exert effects on organ-cultured mouse palatal shelves in a fashion similar to their effects in cell culture and that controlled physiological levels of such factors may be important in mouse palatal development. Key words: epidermal growth factor, transforming growth factor alpha, transforming growth factor beta, platelet-derived growth factor, extracellular matrix, palate development.

INTRODUCTION

During mammalian development the secondary palate arises as bilateral outgrowths from the maxillary processes (embryonic day 12 in mice). The shelves at first grow vertically lateral to the tongue (embryonic day 13 in mice) but at a precise stage rapidly rise to a horizontal position above it (embryonic day 14.5 in mice). The medial-edge epithelial cells of apposing shelves contact and fuse to form a midline epithelial seam (Greene and Pratt, 1976). The seam subsequently degenerates via the combined processes of cell death and epithelial-mesenchymal transformation/migration (Ferguson, 1988; Fitchett and Hay, 1989; Carette, Lane and Ferguson, 1991) allowing mesenchymal continuity across the palate. Simultaneously the epithelia along the nasal and the oral aspects differentiate into pseudostratified, cili-

*To whom all correspondence should be addressed. Abbreviarions:

DCS, donor calf serum; EGF, epidermal growth factor; PBS, phosphate-buffered saline; PDGF, platelet-derived growth factors; TGFa-transforming growth factor alpha; TGFfl, transforming growth factor beta.

ated columnar cells and stratified, squamous keratinizing cells, respectively (Greene and Pratt, 1976; Ferguson, 1988). Regional palatal epithelial differentiation is specified by the mesenchyme (Ferguson and Honig, 1984) probably via a complex interaction of extracellular matrix molecules and growth factors (Ferguson, 1987, 1988; Sharpe and Ferguson, 1988). In vitro, EGF inhibits medial-edge epithelial degeneration and induces keratinization in the presence of palatal mesenchyme or a suitable extracellular matrix (Hassell, 1975; Hassell and Pratt, 1977; Tyler and Pratt, 1980; Grove and Pratt, 1984). The EGF receptor and TGFcr, the presumptive embryonic homologue of EGF (Derynck, 1986), are intensely localized in the developing palate in vivo, particularly in the medial-edge epithelial seam (Ferguson, 1988; Dixon, Garner and Ferguson, 1991). EGF and TGFcr stimulate extracellular matrix synthesis by palatal mesenchyme in vitro (Pratt et al., 1980; Pratt, Kim and Grove 1984; Silver, Murray and Pratt, 1984; Turley, Hollenberg and Pratt, 1985; Ferguson, 1987; 1988). The apparent teratogenic effect of EGF in inhibiting medial-edge epithelial degeneration, in vitro, may be an artefact of the 395

M.

396

J. DIXONand M. W. J.

various culture conditions used: gradients from the medium/air interface, doses of EGF, presence or absence of serum, etc. (Hassell, 1975; Hassell and Pratt, 1977; Tyler and Pratt, 1980; Pratt et al., 1980, 1984; Grove and Pratt, 1984, Abbott and Pratt, 1987a, b). Additionally, such effects may result from inappropriate levels of EGF, presence of EGF at a stage of palatogenesis when it is not normally expressed in viva or disruption of the patterns of the spatiotemporal expression of growth factors and their receptors (Abbott and Birnbaum, 1990). Significantly, when EGF is injected into pregnant mice, it does not cause cleft palate, rather it actually reduces the incidence of cleft palate in the fetuses of cortisone-treated mothers (Kano, 1986). Isoforms of the TGFB family have recently been implicated as important regulators of palate development, by virtue of the temporal and spatial patterns of localization of their messenger RNAs (Fitzpatrick et al., 1990; Pelton ef al., 1990) and proteins (Heine et al., 1987; Sharpe and Ferguson, 1988; Gehris, D’Angelo and Green, 1991) during mouse palatogenesis. In vitro exogenous addition of TGFP, , increases the synthesis of mouse palatal glycosaminoglycans, proteoglycans and collagens (Sharpe and Ferguson, 1988; Foreman, Sharpe and Ferguson, 1991): important molecules in normal palate development (Ferguson, 1988). There appear to be few differences between the effects of the various TGF/l isoforms in stimulating extracellular matrix biosynthesis in vitro (for review see Roberts and Sporn, 1990) but this may reflect the role of culture conditions in minimizing the complex interactions (e.g. between different growth factors) that occur in vivo. The effects of exogenous TGF/l, on palatal epithelial and mesenchymal differentiation in organ culture are unknown, whilst there have been no studies of the role of PDGF in palate development. We have therefore now investigated the effects of TGFB, and PDGF, and re-evaluated the effects of EGF and TGFa, on palatal epithelial differentiation and extracellular matrix immunolocalization in vitro under defined conditions. MATERIALS AND METHODS

Embryos

Mice (strain MF) were mated overnight and screened for vaginal plugs the next morning. The day of finding such plugs was called day zero. Pregnant mice were killed by ether overdose and the gravid uteri aseptically removed. Individual embryos were dissected into sterile Hank’s balanced salt solution (Flow Laboratories, Rickmansworth, U.K.) and staged by external morphological characteristics (Theiler, 1972). Palatal shelves were dissected from the heads of decapitated Theiler stage-2 1 (embryonic day 13) embryos. Organ culture procedures

Type I collagen was extracted from rat tail tendons [for details, see Schor (1980)]. Three-dimensional hydrated collagen gels were cast by rapidly mixing 8.5 ml of the 2.0 mg/ml collagen solution with 1 ml of 10x concentrated culture medium (see below) and

FERGUSON

0.5 ml of 7.5% sodium bicarbonate (Flow Laboratories, Rickmansworth, U.K.). Two-ml portions were pipetted into 35-mm plastic tissue-culture dishes and allowed to gel at 37°C. Palatal shelves were explanted on to the surface of preformed gels either singly or in pairs with their tips accurately aligned such that their maximum convexities (the region that fuses first in uivo) were in contact. The shelves were cultured in I ml of a 1: 1 mixture of Dulbecco’s modification of Eagle’s Medium and Ham’s F-12 medium (DMEM/Fl2) supplemented with non-essential amino acids, 1 mM sodium pyruvate, 2 mM glutamine and 100 units/ml penicillin/streptomycin either in the presence or absence of 2.5% DCS (all Grand Island Biological Co., Paisley, U.K.). In some cases the medium was supplemented with either 10 ng/mI EGF (ICN Biomedicals, High Wycombe, U.K.) or 10 ng/ml TGFa (Peninsula Laboratories, St Helens, U.K.), or 1 ng/ml TGFfi, (R&D systems, Minneapolis, MN, U.S.A.) or 2 ng/ml purified PDGF (R&D Systems, Minneapolis, MN, U.S.A.). Cultures were maintained at 37°C in an atmosphere of 100% humidity and 5% carbon dioxide in air. After 1, 2 or 3 days in culture, representative palates plus a small amount of the underlying collagen gel were dissected out and processed for histological or immunocytochemical staining. Assay procedures

For histological examination, explantswerefixed for 24 h at 4°C in 4% phosphate-buffered paraformaldehyde, dehydrated through a graded series of alcohols, cleared in xylene and embedded in Fibrowax (BDH Chemicals, Poole, U.K.). Serial sections of 6 pm were cut in a transverse plane (equivalent to the frontal plane in viuo), stained in Gill’s haematoxylin and eosin and mounted in Eukitt synthetic mounting medium (BDH Chemicals, Poole, U.K.). For immunocytochemical staining, explants were snapfrozen in isopentane cooled in liquid nitrogen. The tissue was embedded in OCT embedding compound (BDH Chemicals, Poole, U.K.) and serial cryosections of 6 pm cut in a transverse plane in a Leitz 1720 Cryostat (Leitz, Wetzlar, Germany) at -20°C. The sections were fixed in acetone for 5 min, air-dried and stored at -70” until used for staining. The sections were stained with affinity-purified polyclonal antibodies against collagen types I, II, III, IV, V and VI, heparan sulphate proteoglycan, fibronectin and tenascin, or monoclonal antibodies against laminin and chondroitin-sulphate proteoglycan (Table I). In some cases, different antibodies against the same molecule were used as an internal control for epitope variation/masking. All antibodies listed in Table 1 were used to stain at least 10 sections from each palate analysed at each time point. The primary antibody was applied for I h followed by three 5 min washes in PBS. The appropriate fluorescein isothiocyanate-conjugated secondary antibody (Table 2) was applied for 1 h, the sections rewashed in PBS and mounted in non-fading aqueous mountant (Johnson et al., 1982). In each case, sections from anterior, middle and posterior regions of six different explants were stained. Controls were established by incubation with the secondary anti-

Growth factor influences on the palate

391

Table I. Primary antibodies used Primary antibody raised against Type I collagen (human) Type I collagen (native from rat) Type I collagen Type II collagen (bovine nasal cartilage) Type II collagen Type III collagen (rat skin) Type III collagen (human) Type III collagen Type IV collagen (human) Type IV collagen Type V collagen (bovine amnion) Type V collagen Type VI collagen (bovine placenta) Fibronectin (chick plasma) Laminin (EHS sarcoma) Laminin (mouse laminin) Tenascin (chick tenascin) Heparan sulphate proteoglycan (bovine kidney) Chondroitin sulphate proteoglycan

Primary antibody raised in Rabbit Sheep Goat Guinea pig Goat Sheep Rabbit Goat Rabbit Goat Guinea pig Goat Rabbit Goat Rabbit Rat (monoclonal) Rabbit

Rabbit

Mouse (monoclonal)

Source Institut Pasteur de Lyon, France Dr G. Rucklidge, Aberdeen, U.K. Seralab Dr S. Ayad, Manchester

1

1:51:lO 1:80 l:lOI:100

2

4 2

I:80 I:80

4 1

I:80 l:lOI:100 1:80 l:lOI:100 1: looI:200 l:lOOI:200 1:4001:lOOO l:lOOI:200

4 3

l:lOOI:200

1

ICN Biomedicals High Wycombe, U.K.

1:200I:800

6

RESULTS

Epithelial differentiation Differentiation of the palatal epithelium in vitro was similar to that seen in uiuo (oral, keratinization;

secondary antibodies used

Dilution Rabbit IgG Sheep IgG Guinea pig IgG Goat IgG Rat IgGAM Mouse IgG

1

4 1 4 1 5 1

nasal, pseudostratified, ciliated columnar cells; and medial-edge, epithelial degeneration) regardless of whether the palates were cultured in serum-free DMEM/FlZ (Fig. 1) or in medium supplemented with 2.5% DCS. Differentiation commenced after 48 h in culture (equivalent of 15 days in uiuo). The medial-edge epithelia of palates cultured under serum-free conditions usually degenerated earlier than in those cultured in serum-supplemented media but both were delayed by approx. 24 h compared with the in tliuo time-schedule (Fig. 1). Paired shelves showed medial-edge epithelial differentiation similar to those cultured in isolation. If their medial edges were accurately aligned and in contact they fused together to form an epithelial seam that subsequently degenerated, allowing mesenchymal continuity.

Table 2. Fluorescein isothiocyanate-conjugated Raised against

4 3

I:80 l:lOI:20 1:20

monoclonal antibodies) for staining mouse embryonic tissues using appropriate absorption controls (Whitby and Ferguson, 1991). Sections were viewed under incident light fluorescence in a Leitz Dialux 22EB microscope (Leitz, Wetzlar, Germany); photographs were taken on Kodak Ektachrome 160 ASA colour reversal film from which black and white prints were made.

1 2 3 4 5 6

I:20

Seralab Dr G. Rucklidge, Aberdeen, U.K. Institut Pasteur de Lyon, France Seralab Institut Pasteur de Lyon, France Seralab Dr S. Ayad, Manchester, U.K. Seralab Dr S. Ayad, Manchester, U.K. Dr D. Garrod, Southampton, U.K. E-Y Laboratories California, U.S.A. ICN Biomedicals High Wycombe, U.K Dr R. ChiquetEhrismann, Base], Switzerland Dr J. Anderson, Manchester, U.K.

bodies alone. We had previously established the specificity of these antibodies (including the mouse

Table 1 reference

Dilution

Secondary antibody (see Table 2l

Raised in

used

Source

Sheep Swine Goat Rabbit Sheep Goat

I:100 I:100 1: 100 1:40 I:20 1:40

Serotec, U.K. Serotec, U.K. Northeast Biomedicals Northeast Biomedicals ICN Biomedicals Ltd Zymed Laboratories, San Francisco, U.S.A.

398

M. J. DIXONand M. W. J. FERGLMN

Medial-edge epithelial degeneration occurred in both shelves in those cases where the medial-edge epi-

thelium of one shelf was aligned with either the nasal or the oral surface of another, but fusion did not

result, i.e. one shelf could not induce degeneration

in

the oral/nasal surface of another. When growth factors were added to the medium the effect on medial-edge epithelial development

Plate 1 Fig. I. Transverse sections through single palatal shelves cultured in serum-free medium for (A) 24, (B) 48 and (C) 72 h (bars = 200 pm). Progressive degeneration of the medial-edge epithelia (M) is apparent. Fig. 2. Transverse sections through single palatal shelves cultured for 72 h in (A) serum-free medium supplemented with long/ml TGFa or (B) medium supplemented with long/ml TGFa plus 2.5% DCS (bar = 200 pm). TGFa prevents medial-edge epithelial (M) degeneration in serum-containing (B), but not in serum-free (A), medium. In both cases TGFa stimulates migration of palatal mesenchymal cells out of the base of the explant into the collagen gel (arrows) [compare with Fig. l(C)]. Fig. 3. Transverse section through a single palatal shelf cultured for 24 h in serum-free medium supplemented with I ng/ml TGFP, (bar = 200 pm). Medial-edge epithelial (M) degeneration is apparent. The mesenchymal cells have a stellate appearance and are loosely arranged [compare with Fig. l(A)]. Plate 2 Fig. 4. Transverse sections through cultured palatal shelves stained for type III collagen. (A) On the day of explantation randomly arranged fibrils are present throughout the mesenchyme, staining is particularly intense along the epithelial-mesenchymal interface and in the subjacent mesenchyme (bar = 200 pm). (B) After 24 h in culture the fibrils beneath the medial-edge epithelium (M) are orientated perpendicular to the epithelial-mesenchymal interface (bar = 100 pm). (C) As the epithelial seam degenerates (48 h in culture) intense staining is present in the mesenchyme subjacent to those areas in which the epithelial seam (E) is intact. In those areas where the seam has degenerated a reduction in staining is evident but staining extends across the area previously occupied by the seam (arrowed) (bar = IOOpm). (D) In cultures supplemented with 10 ng/ml TGFa and 2.5% DCS (bar = 100 pm) intense staining for type III collagen persists beneath the medial-edge epithelium (M). Plate 3 Fig. 5. Transverse sections through cultured palatal shelves stained for type IV collagen (bars = 100 pm). On the day of explantation (A) and during epithelial seam formation staining is present in the basement membranes of the palate (P) and blood vessels (BV). As the seam degenerates (B) staining becomes discontinuous and punctate staining is seen in the mesenchyme immediately adjacent to the seam area (arrowed). This pattern is also’detected where the palates were cultured in isolation (C). In cultures supplemented with long/ml TGFa and 2.5% DCS (D), where medial-edge epithelial degeneration is prevented, the punctate dispersing pattern is not seen; rather, staining appears as a continuous line beneath the persistent medial-edge epithelium (M). Plate 4 Fig. 6. Transverse sections through cultured palatal shelves stained for fibronectin (bars = 100 pm). (A) After 72 h culture in serum-free medium, randomly arranged fibrils are present throughout the mesenchyme. The paired shelves have fused together but no intershelf discontinuity is apparent in the region formerly occupied by the epithelial seam (E). (B) After 72 h culture in medium supplemented with I ng/ml TGFB, staining. whilst being of a similar distribution, is more intense. Fig. 7. Transverse sections through cultured palatal shelves stained for tenascin. (A) On the day of explanatation, randomly arranged fibrils are concentrated in the mesenchyme along the future nasal (N) aspect and at the tip of the medial edge of the palate (M). Little staining is evident in the core mesenchyme (C) or beneath the future oral epithelium (0) (bar = 200 pm). After 48 h culture in serum-free medium (B), staining is most marked in the medial region of the palate in the mesenchyme beneath the epithelial seam (E). Staining is also increased along the oral aspect (0), particularly at the epithelial-mesenchymal interface (bar = 200 pm), In cultures supplemented with IO ng/ml TGFa and 2.5% DCS (C) there is a marked increase in tenascin staining, particularly in the mesenchyme beneath the medial-edge epithelium (M) (bar = 100 pm). Plate 5 Fig. 8. Transverse sections through single palatal shelves stained for chondroitin-sulphate proteoglycan (bars = 200 pm). (A) At the time of explantation diffuse, punctate staining is present in the mesenchyme and along the epithelial-mesenchymal interface beneath the future oral palatal epithelium (0). Staining is less marked at the medial edge (M) and absent from the nasal aspect (N). (B) After 72 h culture in serum-free medium staining is more intense and present in the mesenchyme and at the epithelial-mesenchymal interface of the oral (0), nasal (N) and medial (M) aspects of the palate. Staining also extends further into the mesenchymal core. (C) After 72 h culture in medium supplemented with IO ng/ml TGFa staining for chondroitin-sulphate proteoglycan is more intense and extends further into the mesenchymal core than in control cultures. (D) Conversely, in cultures supplemented with 1ng/ml TGF/?, a marked reduction in staining is apparent.

Growth

factor

influences

on the palate

399

:

.:

Plate

I

,

400

M. J. DIXON and M. W. J. FERGUS~N

Plate

Growth factor influences on the palate

Plate 3

401

402

M. J. DIXON and M. W. J. FERGUS~N

Plate 4

Growth factor influences on the palate

Plate 5

403

M. J. DIXONand M. W. J. FERGU.WN

404 Table 3. The effects of exogenous

growth

factors

on palatal

medial-edge

epithelial

(MEE) differentiation

in vitro

Medium DMEM/Fl2 DMEM/Fl2 + 2.5% DCS DMEM/Fl2 10 ng/ml

+ EGF

DMEM/FIZ + 10 ng/ml EGF +2.5% DCS DMEM/F12 + 10 ng/ml TGFG~ DMEM/F12 + 10 ng/ml TGFG( +2.5% DCS DMEM/Fl2 + 1 ng/ml TGFP, DMEM/Fl2 + 1ng/ml TGFB, +2.5% DCS DMEM/F12 + 2 ng/ml PDGF DMEM/F12 + 2 ng/ml PDGF +2.5% DCS

Hours in culture

Number of explants

Percentage with intact MEE

24 48 12 24 48 72 24 48 72 24 48 72 24 48 72 24 48 12 24 48 72 24 48 72 24 48 12 24 48 72

21 20 22 18 21 21 17 23 22 17 22 21 10 12 12 10 13 13 8 14 13 8 9 8 8 II I1 8 14 11

100 25

depended upon whether or not serum was also present in the medium. The results are summarized in Table 3. The addition of 10 ng/ml EGF or TGFcl in the absence of serum caused a delay in medial-edge epithelial degeneration (evident after 48 h in culture) but did not prevent it. Ninety-five per cent of palates showed medial-edge epithelial degeneration after 72 h in cultures supplemented with EGF and 92% with TGFx supplementation [Fig. 2(A)]. By contrast, when EGF or TGFcr were added at the same concentration to serum-containing medium, a marked inhibition of medial-edge epithelial degeneration resulted. This effect was evident after 48 h, when inhibition of medial-edge epithelial degeneration was seen in 86% of cultures with EGF and 92% with TGFcr. This percentage inhibition changed little after 72 h [Fig. 2(B)] (81 and 92%, respectively). In cultures treated with EGF and DCS or TGFa and DCS the epithelia of the medial and oral aspects were more heavily keratinized than in control cultures. Both EGF and TGFc( promoted migration of mesenchyma1 cells into the collagen gel, either from the base of the explant or, where the medial-edge epithelium had degenerated, from the tip of the palatal shelf (Fig. 2). PDGF in the absence of serum did not prevent medial-edge epithehal degeneration: 8 1% of palates exhibited complete degeneration of the medial-edge epithelium after 72 h in culture, and the remaining 19% showed partial degeneration. In PDGF- and DCS-supplemented cultures the majority of palates

100 52 94 35 100 86 81 100 50 8 100 93 93 63

100 12 25 100 21 100 85 64

Percentage with partial degeneration of MEE

Percentage with complete degeneration of MEE

25 5

50 95

38 10 6 48 14

10 90 17 81

14 19 16 8

37 29

-

34 84

64 loo

33

55 75

46 19

27 81

15 27

9

displayed an intact medial-edge epithehum after both 48 and 72 h in culture (85 and 64%, respectively). Where present, medial-edge epithehal degeneration was very limited. The epithelia of the oral aspect of the palate were more heavily keratinized but mesenchymal migration into the collagen gel was no different to that observed in control cultures. By contrast, in TGFB,-supplemented cultures the nasal, oral but especially the medial-edge epithelia were thin and poorly differentiated. Medial-edge epithehal cell degeneration was precocious, 37% of palates exhibiting signs of degeneration after only 24 h in culture (Fig. 3) (compared with 0% in the controls). After 48 h only 7% of palates had an intact medial-edge epithelium; complete degeneration was evident in all palates after 72 h. The presence of DCS in TGF/?,-supplemented cultures reduced the effectiveness of TGFB, in promoting medial-edge epithelial degeneration. However, unlike similar cultures supplemented with EGF, TGFa or PDGF, the majority of palates (75%) showed complete medial-edge epithelial degeneration after 72 h. TGFB, supplementation also caused the mesenchyme to be more loosely arranged, mesenchymal cells having a stellate appearance separated by large intercellular spaces staining positively with alcian blue for glycosaminoglycans. Moreover, mesenchymal migration into the collagen gel was markedly inhibited compared with control cultures. TGF& does, however, promote the formation of large amounts of cartilage in the explants.

Growth factor influences on the palate Localization of extracellular matrix molecules Type I collagen. At all stages type I collagen was ubiquitously distributed throughout the mesenchyme where it appeared as a fibrous network around the cells. It was also concentrated at the epithelialmesenchymal interface but absent from the epithelia. In paired palates undergoing epithelial-seam degeneration, type I collagen appeared to be rapidly synthesized in the fusing areas such that no discontinuity was apparent between the mesenchyme of one shelf and that of the other. Type II collagen. No staining was seen at any stage, except in the cartilaginous blastemata of the TGF/l,-treated cultures. Type III collagen. Type III collagen was present throughout the mesenchyme at all the stages examined. Staining was intense at the epithelialmesenchymal interface. The pattern was of a randomly arranged fibrillar network. At explantation (embryonic day 13 in uioo) a higher concentration of staining was seen in the mesenchyme immediately beneath the medial-edge epithelium [Fig. 4(A)]. This gradient of staining became more marked with time in culture. After 24 h, type III collagen fibres at the epithelial-mesenchymal interface were orientated predominantly perpendicular to the basement membrane [Fig. 4(B)]. After 48 h in culture, intense staining was seen throughout the mesenchyme, particularly in the areas adjacent to the intact seam [Fig. 4(C)]. As the seam degenerated, staining (at a reduced intensity) rapidly became evident in the regions of mesenchymal continuity [Fig. 4(C)]. After 3 days in culture, when fusion was complete, a more even distribution of staining was present across the palate. Type IV collagen. Staining for type IV collagen was present in the basement membranes of the palate and blood vessels, but absent from the epithelia and mesenchyme. On the day of explantation [Fig. S(A)] and during seam formation (culture days 0 and 1) staining was seen as a continuous line. As the seam degenerated (day 2) the pattern became discontinuous and punctate staining was seen along the line previously occupied by the epithelial seam and in the mesenchyme immediately adjacent to it [Fig. 5(B)]. This pattern was detected regardless of whether the palates were cultured singly or in pairs [Figs 5(B) and (C)l. After degeneration of the medial-edge epithelium the only residual midline staining was present as a continuous line around occasional epithelial pearls: oral and nasal basement membrane staining was continuous, as before. Type V collagen. The patterns obtained with this antibody were similar to those described for type IV collagen. Type VI collagen. Type VI collagen was strongly localized in the palatal basement membranes at ail stages. At the time of explantation, staining of the mesenchymal interstitial matrix immediately subjacent to the epithelium was also present. This appeared as numerous, randomly arranged fibrils, many of which inserted into the basement membrane. There was no staining of the mesenchymal core or the epithelium. During the culture period, staining for type VI collagen increased in the mesenchyme, pre-

405

dominantly along the oral aspect. By the third day in culture, strong mesenchymal staining was seen beneath the oral and nasal palatal epithelia with relatively less on the medial aspect where the medialedge epithelium had degenerated. Laminin. The staining patterns obtained with Iaminin were identical to those obtained with type IV collagen at all stages. Fibronectin. Fibrillar staining for this molecule was evenly distributed throughout the mesenchymal matrix at all stages. Continuous staining was seen at the epithelial-mesenchymal interface. As the epithelial seam degenerated, staining was lost from the medial basement-membrane region. Simultaneously, fibronectin was rapidly synthesized in the mesenchyme so that no intershelf discontinuity was apparent [Fig. 6(A)]. Tenascin. On the day of explantation (embryonic day 13 in uiuo) staining for tenascin was present in the mesenchyme and at the epithelial-mesenchymal interface, but not in the epithelium. This appeared as numerous, brightly fluorescent fibrils concentrated in the mesenchyme beneath the future nasal aspect and the tip of the palate [Fig. 7(A)]. Little staining was evident in the central regions of the mesenchyme or beneath the future oral epithelium [Fig. 7(A)]. In the region immediately subjacent to the medial basement membrane the fibrils were orientated perpendicular to it, elsewhere they appeared more randomly arranged. After 24 h in culture, staining was concentrated at the medial edges of the palate, particularly at the epithelial-mesenchymal interface, from which numerous fibrils extended perpendicularly into the mesenchyme. Once the epithelial seam had formed, staining was most marked in the medial region of the palates [Fig. 7(B)]. Staining was also increased along the oral aspect both in the mesenchyme and at the epithelial-mesenchymal interface [Fig. 7(B)]. As the seam degenerated, staining remained brightest in the medial region of the palate. Heparan-sulphate proteoglycan. At the time of explantation, heparan-sulphate proteoglycan was localized most strongly in the basement membranes of the palate and blood vessels. Fainter staining was present in the mesenchyme but not the epithelium. This was more diffuse than that seen with collagen types I and III and fibronectin. Mesenchymal staining was most marked subjacent to the basement membrane. Staining of the basement membrane of the medialedge epithelia persisted during epithelial seam formation but decreased as it degenerated. The punctate dispersing pattern seen for laminin and type IV collagen was not seen for heparan-sulphate proteoglycan. Staining appeared to become more ubiquitous with increasing time in culture, the gradient between the core mesenchyme and that subjacent to the epithelium becoming less marked. Chondroitin-sulphate proteoglycan. Staining for chondroitin-sulphate proteoglycan was punctate and diffuse. It was present in the mesenchyme, especially along the epithelial-mesenchymal interface, but absent from the epithelium. At explantation, staining was confined to a narrow region of the mesenchyme along the future oral aspect and was especially intense at the epithelial-mesenchymal interface [Fig. 8(A)].

M. J. DIXON and M.

406

Staining was less intense at the medial edge of the palate and was absent from the nasal aspect [Fig. 8(A)]. After 24 h in culture, staining increased along the medial and oral aspects, extending further into the mesenchymal core. The epithelial-mesenchymal interface along the nasal aspect and the mesenchyme subjacent to it were also stained. Over the next 2 days in culture staining for chondroitin-sulphate proteoglycan extended further into the mesenchyme and became more intense [Fig. 8(B)]. Eflects of growth factors on localization of extracellular matrix molecules In EGF- or TGFa-supplemented, serum-free culture there was a generalized increase in staining for collagen types III, IV, V and VI, laminin, fibronectin, heparan-sulphate proteoglycan, chondroitin-sulphate proteoglycan and tenascin. There was, however, no alteration in their distribution. In control cultures, staining for collagen types IV, V and VI, laminin and heparan-sulphate proteoglycan was lost from the basement membrane beneath the medial-edge epithelium as it degenerated. Where EGF or TGFa were added in the presence of serum, staining for all these molecules persisted, in keeping with the persistence of the medial-edge epithelial seam [Fig. 5(D)]. In addition, the punctate dispersing pattern of staining associated with medial-edge epithelial degeneration was not seen [Fig. 5(D)]. In EGF/DCS- or TGFa DCS-treated cultures, a marked increase in type III collagen and tenascin staining was seen in the mesenchyme, particularly at the tip of the palate [Figs 4(D) and 7(C)]. Synthesis of chondroitinsulphate proteoglycan was also stimulated, staining extending further into the mesenchymal core than in control cultures [Fig. 8(C)]. In the absence of serum, PDGF produced no alterations to the control staining patterns. In the presence of serum, increased extracellular matrix staining was detected, the patterns being essentially the same as those in EGF- and DCS-treated cultures. TGFp,-treated cultures displayed increased extracellular matrix staining regardless of whether or not serum was present in the medium. The fibronectin content of the mesenchyme was markedly increased [Fig. 6(B)]. Conversely, staining for chondroitinsulphate proteoglycan was markedly decreased compared with the control cultures, particularly in the mesenchyme subjacent to the oral and medialedge epithelia [Fig. 8(D)]. Whilst no alteration was seen in collagen type I staining, there was increased staining for collagen types III and VI, tenascin and heparan-sulphate proteoglycan. DlSCUSSION

We have shown that embryonic mouse palatal shelves undergo their normal pattern of epithelial differentiation in oitro (i.e. oral, keratinization; nasal, pseudostratified, ciliated columnar cells; medial-edge, epithelial degeneration) when grown in submerged organ culture under defined, serum-free conditions. Whilst palates cultured in medium containing 2.5% DCS underwent a similar pattern of epithelial

W. J.

FERGIJSJN

differentiation, medial-edge epithelial degeneration was delayed. This is presumably because serum contains variable amounts of steroids and growth factors that may inhibit palatal development (Salomon and Pratt, 1979; Pratt ef al., 1980; Ferguson, Honig and Slavkin, 1984; Abbott and Pratt, 1987a, b). Submerged culture, as opposed to conventional culture on millipore filters at the air/gas interface (Ferguson et al., 1984), is important for experiments investigating the effects of growth factors. Preliminary experiments have shown that the penetration of growth factors into such explants at the air/gas interface is erratic, presumably because growth factors bind to the millipore filters and because of variable diffusion rates from the base of the tissue. It is important that previous experiments on palatal culture with EGF-supplemented media used culture at the air/gas interface (Hassell, 1975; Hassell and Pratt, 1977; Silver et al., 1984; Turley et al., 1985; Abbott and Pratt, 1987a, b; Abbott, Adamson and Pratt, 1988). Submerged culture, on the other hand, ensures adequate exposure of the explant to growth factors and precludes the establishment of artefactual gradients of growth factors induced by the culture conditions. Previous studies have shown that EGF inhibits medial-edge epithelial degeneration promoting keratinization (Hassell, 1975; Hassell and Pratt, 1977; Tyler and Pratt, 1980; Grove and Pratt, 1984; Abbott and Pratt, 1987a, b). However, as outlined in the Introduction, these studies used serum-supplemented media in air/gas interface cultures. The use of serum supplementation makes the interpretation of those experiments difficult as serum is undefined and contains many factors that may interact with EGF. We found that EGF and its presumptive embryonic homologue, TGFa (Derynck, 1986), did not inhibit medial-edge epithelial degeneration in palatal shelves cultured under serum-free conditions; inhibition only occurred in serum-containing medium. Similar results were obtained for PDGF, namely that it had no effect on medial-edge epithelial degeneration in serum-free conditions, but inhibited it in serum-containing cultures. As serum contains both EGF/TGFa and PDGF, it is tempting to speculate that it is the interaction between these two growth factors which is responsible for the differences observed when either factor is added alone to serum-free or serum-containing cultures. Indeed, our preliminary experiments involving addition of either TGFa or EGF, plus PDGF, to mouse palatal shelves organ-cultured under serum-free conditions indicate that this combination inhibits medial-edge epithelial degeneration. However, there are additional interactions, e.g. of TGFa/EGF/PDGF with basic fibroblast growth factor, another growth factor present in serum. These findings indicate that the physiological effects of growth factors during normal palatal development in vivo are likely to be the result of complex interactions between differing levels of differing growth factors at differing developmental times. One growth factor may modulate another’s (or its own) effects by, for example: up- or downregulation of the synthesis of the growth factor and/or its receptor (Fernandez-Pol, Klos and Hamilton, 1989; Gronwald, Seifert and Brown-Pope,

Growth factor influences on the palate

1989; Ranganathan and Getz, 1990; Roberts and Sporn, 1990); the synthesis of extracellular matrix molecules that bind the growth factor, either maintaining it in an active form or neutralizing its action (Fava and McClure, 1987; Yamaguchi, Mann and Ruoslahti, 1990); the synthesis and degradation of extracellular matrix molecules and their receptors (Roberts and Sporn, 1990; Merwin et al., 1990), which may in turn affect the cell’s shape, receptor display and response to other growth factors (Sharpe and Ferguson, 1988); and by acting at signalling pathways distal to the receptor for other growth factors (Massague, 1985; Like and Massague, 1986). Such findings caution against experiments involving the addition of growth factors to developing palates in vitro in the presence of undefined factors in serum. They also indicate that experiments involving the exogenous application of growth factors in physiologically meaningful combinations to palate organ cultures under serum-free conditions may provide more relevant insights into regulatory events in vivo than the application of one growth factor alone. This thesis is supported by the results from the TGF/?,-supplemented cultures, where addition of TGF/?, in serum-free conditions induced precocious medial-edge epithelial degeneration, whilst in the presence of serum, this effect was greatly attenuated. TGF/?, , flz and fi, are each known to be inhibitory to the growth of epithelial cells, antagonizing the mitogenie effects of TGFcr and EGF (Massague, 1985; Like and Massague, 1986; Coffey et al., 1988). Interestingly, TGFP, and /I3 mRNAs are localized predominantly in the palatal medial-edge epithelia, particularly on embryonic day 14, when the midline epithelial seam is forming and degenerating (Fitzpatrick et al., 1990; Pelton et al., 1990). This localization pattern, together with our finding that exogenous TGF/I, induces precocious medial-edge epithelial degeneration, is highly suggestive of the possibility that TGFfl isoforms play an important role in medial-edge epithelial differentiation and seam disruption in vivo. TGF,!I may down-regulate DNA synthesis and cell division in the medial-edge epithelium. Interestingly, this down-regulation occurs despite the presence of large numbers of EGF receptors and bound ligand on the medial-edge epithelium (Dixon et al., 1991). In other systems, TGFB inhibits epithelial cell growth by an action distal to the EGF receptor (Massague, 1985; Like and Massague, 1986, Coffey, 1988) and this may be its mechanism of action in the medial-edge epithelium. Addition of TGFj3, to palatal organ cultures also induced abnormal cartilage formation in the mesenchyme. TGFfl is known to stimulate chondrogenesis (Roberts and Sporn, 1990) and, interestingly, TGF/$ transcripts localize to the centre of the fused palate at embryonic day 15 in vivo (Fitzpatrick et al., 1990) the area that subsequently differentiates into the cartilage of the midpalatal suture. The palatal midline epithelial seam disappears by a combination of death of the superficial epithelial cells and mesenchymal transformation of the basal epithelia (Ferguson, 1988; Fitchett and Hay, 1989). Recently, we have acquired provisional evidence that these transformed medial-edge epithelial cells migrate

407

up and down to the oral and nasal epithelial triangles, where they subsequently revert to a classical epithelial phenotype and colonize the oral and nasal epithelia. Interestingly, submerged organ cultures of either single or paired palatal shelves here exhibited all the signs, e.g. transient punctate staining for basement membrane molecules, of this medial-edge epithelial migration, despite the assertion that organ culture is supposed to accentuate medial-edge epithelial death (Fitchett and Hay, 1989). Accumulation of growth factors, e.g. TGFa (Dixon et al., 1991) TGFfl,, TGF/.?, (Fitzpatrick et al., 1990; Pelton et af., 1990) in the medial-edge epithelium and their interaction may be important in stimulating this migration whilst inhibiting, at least transiently, medial-edge epithelial division. Moreover, these growth factors increase the synthesis of extracellular matrix molecules, e.g. tenascin, type III collagen, hyaluronic acid, particularly adjacent to the medial-edge epithelium; these molecules may be important in facilitating medialedge epithelial migration during seam disruption. Moreover, the accumulation of such matrix molecules and mesenchymal cells adjacent to the midline epithelial seam means that mesenchymal continuity is rapidly established after seam disruption, ensuring the strength of palatal fusion. In this context, it is interesting that EGF/TGFa stimulated the migration of mesenchymal cells out of either the base of the explant or the disrupting medial edge and into the collagen gel. Such stimulation may reflect the physiological role of these growth factors in vivo in promoting mesenchymal cell migration and extracellular matrix biosynthesis across the disrupting midline epithelial seam. Importantly, some molecules, e.g. type III collagen and tenascin, were orientated with their fibres perpendicular to the medial-edge basement membrane. This orientation, also seen in vivo (Ferguson, 1988) may reflect the ordered parallel alignment of collagen fibres Seen behind the trailing edge of a fibroblast/mesenchymal cell migrating within a collagen gel (Bilozur and Hay, 1989). Such a pattern suggests that the growth factors localized within the medial-edge epithelium in vivo (and in vitro) may be chemotactic to palate mesenchyme cells, causing them to migrate to the medial-edge epithelium ready to colonize rapidly the area vacated by the migrating medial-edge epithelium during seam dispersal. Further evidence for this view is given by the observation that the medial-edge epithelial cells themselves synthesize a number of extracellular matrix molecules, e.g. tenascin, type III collagen, hyaluronic acid, before and during seam disruption. These molecules may blur the boundary between seam epithelia and mesenchyme, facilitate migration of medial-edge epithelial cells during seam disruption, and ensure the strength of palatal fusion during seam disruption. Addition of growth factors to organ cultures of palatal shelves also caused changes in their extracellular matrix profile. Under growth factor and serum-free conditions collagen types I and III and fibronectin were ubiquitously distributed in the palate at all stages. As discussed, these molecules are rapidly synthesized and accumulate in the fusion zone at the time of medial-edge epithelial degeneration. Laminin and collagen type IV were localized in the basement

408

M. J. DIXONand M. W. J. FERGUKIN

membranes of the palate and blood vessels. At the time of medial-edge epithelial degeneration, but not before, staining becomes punctate and appears in the mesenchyme adjacent to the area formerly occupied by the seam or the medial edge in a single-palate organ culture. This pattern is probably due to the migrating epithelial cells carrying with them fragments of the basement membrane, which are rapidly degraded, resulting in loss of this staining pattern. Interestingly, this pattern was observed regardless of whether the palates were cultured singly or in pairs, suggesting that shelf contact is not necessary for transformation and migration of the medial-edge epithelial cells. EGF and TGFcr had a generalized stimulatory effect on extracellular matrix production, a marked increase in type III collagen and tenascin staining being found in the mesenchyme at the medial edge of the palate when serum was also present. These results are similar to those obtained in cell cultures of mouse embryonic palatal mesenchymal cells alone (Ferguson, 1988; Sharpe and Ferguson, 1988). EGF has been shown to stimulate synthesis of type V collagen and fibronectin in cultured murine palatal shelves (Silver et al., 1984). Turley et al. (1985) also noticed that EGF stimulated glycosaminoglycan production in organ cultures of murine palates; the outcome of cellulose acetate electrophoresis suggested that this was due in part to an increase in the amount of chondroitin sulphates. We found that EGF markedly increased staining for chondroitinsulphate proteoglycan. Previous studies on the effects of growth factors on mouse palatal extracellular matrix production used organ cultures at the air/gas interface and SainteMarie processing techniques for immunocytochemistry (Silver et al., 1984; Turley et al., 1985). Therefore, whilst we are in agreement about the stimulatory effect of EGF on extracellular matrix biosynthesis, we believe that the patterns previously described are artefactual and reflect diffusion of growth factors from the base of the explant and a loss of some extracellular matrix components during processing. Our present findings were derived from submerged organ cultures and used frozen immunocytochemistry and therefore more accurately reflect the true patterns. We have also shown that TGF/?, stimulates extracellular matrix accumulation in palatal-shelf organ culture, the increase being most marked for fibronectin and tenascin. TGF/?, has been shown to stimulate extracellular matrix synthesis by mouse embryonic palatal mesenchymal cells (Sharpe and Ferguson, 1988). TGF/.?, stimulates the accumulation of extracellular matrix molecules both by increasing their synthesis and decreasing their degeneration (Roberts and Sporn, 1990). It also stimulates synthesis of glycosaminoglycans, proteoglycans (Chen, Hoshi and McKeehan, 1987; Bassols and Massague, 1988; Roberts and Sporn, 1990) and tenascin (Pearson et al., 1988). We found that TGFB, appeared to inhibit production of chondroitin-sulphate proteoglycan, particularly in the core mesenchyme. Analysis of glycosaminoglycan synthesis by palatal mesenchymal cells has shown that TGF/I, markedly stimulates the production of hyaluronic acid and

chondroitin sulphate in sparse but not confluent cell cultures (Sharpe and Ferguson, 1988). Stimulation of hyaluronic acid biosynthesis by TGF/3, has been demonstrated biochemically in organ cultures of mouse palatal shelves (Foreman et al., 1991) and can be inferred from the stellate appearance of the mesenchymal cells, separated by large alcian blue-positive intracellular spaces, seen in the present study. Interestingly, these appearances are very similar to those observed in uioo before palatal shelf elevation where the accumulation and hydration of palatal hyaluronic acid is thought to play a crucial role in that elevation (Ferguson, 1988). Palatal mesenchymal TGF/? may play a critical role in regulating the synthesis of such glycosaminoglycans and hence in regulating elevation. Importantly, TGFB, stimulates the synthesis of specific extracellular matrix molecules, e.g. fibronectin and tenascin by palate mesenchyme cells, particularly adjacent to the medial-edge epithelia. TGFB can also stimulate the synthesis of integrin receptors for such extracellular matrix molecules on epithelial cells (Ignotz and Massague, 1987). Thus the same growth factor, TGF/I,, could signal an epithelial-mesenchymal interaction by its divergent effects on mesenchymal and epithelial cells: promoting extracellular matrix biosynthesis by the mesenchyme and the development of receptors for such extracellular matrix molecules by the epithelium. Moreover, even in the absence of such direct effects, the accumulation of specific extracellular matrix molecules within differing regions of the palate may modulate the response of cells within and adjacent to such areas to various growth factors (Ingber, Madri and Folkman, 1987; Roberts and Sporn, 1990). The matrix molecules, e.g. fibronectin, may even bind and sequester growth factors, e.g. TGFB, (Fava and McClure, 1987), releasing them for activity on either a different cell population and/or at a later developmental time, e.g. during palatal bone formation. In summary, the addition of EGF or TGFa, both of which have been localized in the developing palate (Ferguson, 1988; Dixon et aI., 1991), does not inhibit medial-edge epithelial degeneration unless serum is present in the medium. Conversely, TGF/?, , which is also present in the developing palate (Heine et al., 1987; Sharpe and Ferguson, 1988; Fitzpatrick et al., 1990; Pelton et al., 1990; Gehris et al., 1991) induces precocious medial-edge epithelial degeneration under serum-free conditions, this effect being less marked in the presence of serum. Growth factors acting either alone or in concert with other growth factors and extracellular matrix molecules may therefore influence palatal epithelial differentiation. The growth factors also stimulated the synthesis of specific extracellular matrix molecules by the palate mesenchyme; such molecules being important in shelf elevation, epithelial and mesenchymal cell differentiation and consolidation of the recently fused palate. Acknowledgements-This work was supported by grants from the Wellcome Trust, MRC and Action Research for the Crippled Child. M. J. Dixon was a Wellcome Trust Research Training Fellow at the time of conducting this research.

Growth factor influences on the palate REFERENCES Abbott B. D. and Pratt R. M. (1987a) Retinoids and epidermal growth factor alter embryonic mouse palatal epithehal and mesenchymal cell differentiation in organ culture. J. craniofac. Genet. devl Biol. 7, 219-240. Abbott B. D. and Pratt R. M. (1987b) Human embryonic palatal epithelial differentiation is altered by retinoic acid and epidermal growth factor in organ culture. J. craniofac. Genei. devl Biol. 7, 241-265.

Abbott B. D. and Birnbaum L. S. (1990) Retinoic acid induces alterations in the expression of growth factors in embryonic mouse palatal shelves. Teratology 42,597-610. Abbott B. D., Adamson E. D. and Pratt R. M. (1988) Retinoic acid alters EGF receptor expression during palatogenesis. Development 102, 853-867. Bassols A. and Massague J. (1988) Transforming growth factor /3 regulates the expression and structure of extracellular matrix chondroitin/dermatan sulfate proteo&cans. J. biol. Chem. 263. 3039-3045. B&&r M. E. and Hay E. D. (1989) Cell migration into neural tube lumen provides evidence for the fixed cortex theory of cell motility. Cell Motility Cyloskeleton 14, 4699484. Carette M. J. M., Lane E. B., Ferguson M. W. J. (1991) Differentiation of murine embryonic palatal epithehum in culture: selective cytokeratin expression distinguishes between oral, medial edge and nasal epithelial cells. Differenriaiion. 47, 149-16 1. Chen J.-K., Hoshi H. and McKeehan W. L. (1987) Transforming growth factor type p specifically stimulates synthesis of proteoglycan in human adult arterial smooth muscle cells. Proc. natn. Acad. Sci., U.S.A. 84, 5287-5291. Coffey R. J., Bascom C. C., Sipes N. J., Graves Deal R., Weissman B. E. and Moses H. L. (1988) Selective inhibition of growth related gene expression in murine keratinocytes by transforming growth factor beta. Molec. Cell Biol. 8, 3088-3093.

Derynck R. (1986) Transforming growth factor-a: structure and biological activities. J. Cell Biochem. 32, 293-304. Dixon M. J., Garner J. and Ferguson M. W. J. (1991) Immunolocalisation of epidermal growth factor (EGF), EGF receptor and transforming growth factor alpha (TGFa) during murine palatogenesis in ciao and in vitro. Anal. Embryol. l&I, 83-91. Fava R. A. and McClure D. B. (1987) Fibronectin-associated transforming growth factors. J. Cell Physiol. 131, 184-189.

Ferguson M. W. J. (1987) Palate development: mechanisms and malformations. Ir. J. Med. 156, 309-315. Ferguson M. W. J. (1988) Palate development. Development 103 (SuppI.), 4160. Ferguson M. W. J. and Honig L. S. (1984) Epithelialmesenchymal interactions during vertebrate palatogenesis. In Current Topics in Developmenral Biology (Ed. Zimmerman E. F.), vol. 19., pp. 1377164. Academic Press, New York. Ferguson M. W. J., Honig L. S. and Slavkin H. C. (1984) Differentiation of cultured palatal shelves from alligator, chick and mouse embryos. Anaf. Rec. 209, 231-249. Femandez-Pol J. A., Klos D. J. and Hamilton P. D. (1989) Modulation of transforming growth factor alpha dependent expression of epidermal growth factor receptor gene by transforming growth factor beta, triiodothyronine and retinoic acid. J. Cell Biochem. 41, 159-170. Fitchett J. E. and Hay E. D. (1989) Medial edge epithelium transforms to mesenchyme after embryonic palate shelves fuse. Devl Biol. 131, 455474. Fitzpatrick D. R., Denhez F., Kondaiah P. and Akhurst R. J. (1990) Differential expression of TGF beta isoforms in murine palatogenesis. Developmenf 109, 585-595. Foreman D. M., Sharpe P. M. and Ferguson M. W. J. (1991) Comparative biochemistry of mouse and chick secondary-palate development in vivo and in vitro with

409

particular emphasis on extracellular matrix molecules and the effects of growth factors on their synthesis. Archs oral Biol. 36, 457-471.

Gehris A. L., D’Angelo M. and Greene R. M. (1991) Immunodetection of the transforming growth factors /I 1 and 82 in the developing murine palate. Devl Biol. 35, 17-24.

Greene R. M. and Pratt R. M. (1976) Developmental aspects of secondary palate formation. J. Embryol. exp. Morph. 36, 225-245.

Gronwald R. G. K., Seifert R. A. and Bowen-Pope D. P. (1989) Differential regulation of expression of two platelet derived growth factor receptor subunits by transforming growth factor-b. J. biol. Chem. 264, 8120-8125. Grove R. I. and Pratt R. M. (1984) Influence of epidermal growth factor and cyclic-AMP on growth and differentiation of palatal epithelial cells in culture. Devl. Biol. 106, 427-437.

Hassell J. R. (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. Devl Biol. 45, 90-102. Hassell J. R. and Pratt R. M. (1977) Elevated levels of CAMP alters the effect of epidermal growth factor in vitro on programmed cell death in the secondary palatal epithelium. Expl Cell Res. 106, 5562. Heine U. I., Munoz E. F., Flanders K. C., Elhngsworth L. R., Lam H.-Y. P., Thompson N. L., Roberts A. B. and Sporn M. B. (1987) The role of transforming growth factor beta in the development of the mouse embryo. J. Cell Biol. 105, 2861-2876.

Ignotz R. A. and Massague J. (1987) Cell adhesion protein receptors as targets for transforming growth factor /I action. Cell 51, 189-197. Ingber D. E., Madri J. A. and Folkman J. (1987) Endothelial growth factors and extracellular matrix regulate DNA synthesis through modulation of cell and nuclear expansion. In Vitro 23, 387-394. Johnson G. D., Davidson R. S., McNamee K. C., Russell G., Goodwin D. and Holborrow E. J. (1982) Fading of immunofluorescence during microscopy: a study of the phenomenon and its remedy. J. Immunol. Meth. 55, 231-242.

Kano Y. (1986) Effects of intraperitoneal administration of epidermal growth factor to maternal mice on the palatal morphogenesis of fetuses. J. Osaka Univ. dent. Sot. 31, 7496. Like B. and Massague J. (1986) The antiproliferative effect of type beta transforming growth factor occurs at a level distal from receptors for growth activating factors. J. biol. Chem. 261, 13,42613,429. Massague J. (1985) Transforming growth factor beta modulates the high affinity receptors for epidermal growth factor and transforming growth factor alpha. J. Cell Biol. 100, 1508-1514. Merwin J. R., Anderson J. M., Kocher O., Van ltallie C. M. and Madri J. (1990) Transforming growth factor beta I modulates extracellular matrix organisation and cellcell junctional complex formation during in vitro angiogenesis. J. Cell Physiol. 142, 117-128. Pearson C. A., Pearson D., Shibahara S., Hofsteenge J. and Chiquet-Ehrismann R. (1988) Tenascin: cDNA cloning and induction by TGF-/I. EMBO JI 7, 2677-2981. Pelton R. W., Hogan B. L. M., Miller D. A. and Moses H. L. (1990) Differential expression of genes encoding TGF’s p 1, 82 and /33 during murine palate formation. Devl Biol. 141, 456460. Pratt R. M., Yoneda T., Silver M. H. and Salomon D. S. (1980) Involvement of glucocorticoids and epidermal growth factor in secondary palatal development. In Cur rent Research Trends in Prenatal Craniofacial Development (Eds Pratt R. M. and Christiansen R. L.). pp. 235-252. Elsevier, North Holland.

410

M. J. DIXONand M. W. J. FERGUS~N

Pratt R. M., Kim C. S. and Grove R. I. (1984) Role of

glucocorticoids and epidermal growth factor in normal and abnormal palatal development. Current Topics in Developmental l%ology (Ed. Zimmerman E. F.), Vol. 19, pp. 81-101. Academic Press, New York. Ranganathan G. and Getz M. J. (1990) Co-operative stimulaGon of specific gene transcription by epidermal growth factor and transforming growth factor type fi 1. J. biol. Chem. MS, 3001-3004. Roberts A. B. and Sporn M. B. (1990) The transforming growth factor /3s. In Peptide Growth Facfors and their Receptors (Eds Sporn M. B. and Roberts A. B.), pp~ 419472. Springer, Berlin. Salomon D. S. and Pratt R. M. (1979) Involvement of glucocorticoids in the development of secondary palate. Differentiation 13, 141-154.

Schor S. L. (1980) Cell proliferation and migration on collagen substrata in vitro. J. Cell Sci. 41, 159-175. Sharpe P. M. and Ferguson M. W. J. (1988) Mesenchymal influences on epithelial differentiation in developing systems. J. Cell. Sci. 10 (Suppl.), 195230.

Silver M. H., Murray J. C. and Pratt R. M. (1984) Epiderma1 growth factor stimulates type V collagen synthesis in cultured murine palatal shelves. Differentiation 27, ._ 205-208. Theiler K. (1972) The House Mouse. Developmenr and Normal Stages from Fertilization to Four Weeks of Age.

Springer, Berlin. Turley E. A., Hollenberg M. D. and Pratt R. M. (1985) Effect of epidermal growth factor/urogastrone on glycosaminoglycan synthesis and accumulation in vitro in the developing mouse palate. Differentiation 28, 279-285. Tyler M. S. and Pratt R. M. (1980) Effect of epidermal growth factor on secondary palatal epithelium in vitro. Tissue isolation and recombination studies. J. Embryol. exp. Morph. 58, 93-106.

Whitby D. J. and Ferguson M. W. J. (1991) The extracellular matrix of lip wounds in fetal, neonatal and adult mice. Development 112, 651-668. Yamaguchi Y., Mann D. M. and Ruoslahti E. (1990) Negative regulation of transforming growth factor-/? by the proteoglycan decorin. Nature 346, 281-284.

The effects of epidermal growth factor, transforming growth factors alpha and beta and platelet-derived growth factor on murine palatal shelves in organ culture.

Palatal shelves isolated from day-13 embryonic mice were explanted on to the surfaces of collagen gels either singly or in pairs with their medial edg...
6MB Sizes 0 Downloads 0 Views