om2-202XjR4jR:102-01I R$02.00 jO 'I'm: ·JOll HNAI. OF INV EST1CATl VE DEHMATOI.O( ;Y. R:UI R- 12:\' 19R4 Copy ri ~hl ,e 19R4 by The Willi a ms & Wilkins Co.

Vol. 8:3 . No. 2 Printed in US. A .

Distribution of Epidermal Growth Factor Receptors in Rat Tissues During Embryonic Skin Development, Hair Formation, and the Adult

Hair Growth Cycle*

MARTIN R GREEN, D. PHIL_ AND JOHN R COUCH MAN , PH_D .

Biosciences Division, Unilever Research, Colwo rth IJaboratory, S hambrook, Bedfordshire, u.K.

In a previous study on neonatal rat skin (Green MR, Basketter DA, Couchman JR, Rees DA: Dev Bioi 100:506-512, 1983) a close positive correlation was found between epidermal growth factor (EGF) receptor tissue distribution and areas of potential epithelial cell prolifer ation. We now report on t he binding distribution of [1 ~r,I ]EG F, representing the tissue localization of availab le EGF receptors, during embryonic rat skin de ­velopment including hair follicle formation and the adult hair growth cycle. At 16 days embryonic devel­opment a relatively low receptor density is seen over all the epidermal cell layers but by 17 days, with the onset of very rapid epidermal proliferation, labeling increases and becomes restricted to the basal epidermal cells. Be­tween 17 and 20 days embryonic development, available receptors for EGF are consistently absent from epider­mal basal cells overlying the dermal condensates mark­ing the first stage of hair follicle development. This restricted and temporary loss of EGF receptors above these specialized mesenchymal condensates implies a role for the EGF receptor and possibly EGF or an EGF­like ligand in stimulating the epithelial downgrowth required for hair follicle development. In the anagen hair bulb, receptors for EGF are detected over the outer root sheath and the epithelial cell layers at the base of the follicle and show a correlation with the areas of epithelial proliferation in the hair bulb. During the ca­tagen and telogen phases of the hair cycle, receptors are observed in high numbers on all the undifferentiated or dedifferentiating cells of the degenerating epithelial strand and secondary hair germ. Dermal cells are, in general, less heavily labeled than the basal epithelial cells of skin except for the developing striated muscle (panniculus carnosus) in embryonic skin which is more heavily labeled.

The data are discussed in terms of a possible role for the EGF receptor and associated EGF or EGF-like li­gands in specific areas of epithelial tissue morphogenesis during embryonic skin maturation, hair follicle devel­opment, and hair cycling.

Epidermal growth factor (EGF) [1] is a low-molecular-weight polypeptide known to promote the growth of epit helia l cells in vivo and in vitro (for reviews see [2- 5]) and as such is worthy of consideration as a possible regulator in vivo of epit helia l growth in sk in _ EGF binds to specific cell-surface receptors in itiating a wide series of b iochemical events, including down

Manuscript received January 4,1984; accepted for publication March 20, 1984.

. * A preliminary report on part of this work has been presented elsewhere (Green MR, Couchman JR, Biochem Soc Trans, in press).

Reprint requests to: Dr. Martin R. Green , Biosciences Division, Unilever Research, Col worth Laboratory, Sharnbrook, Bedfordshire, U. K. MK44 I LQ.

Abbreviations: EGF: epidermal growth factor

regu lation of recepto rs [2,6], leadin g u lt imately to mitosis and cell proliferation.

We have previously exam ined t he distribution a nd number of EGF receptors in neo nata l rat skin [7] a nd found a close positive co rrelation between EG F receptor localizatio n and areas support ing epit helia l cell growth . Further, t he number of EGF receptors expressed on freshly isolated basal ep ide rma l cells from neonatal animals aged 1- 10 days decreased ra pidly with increasing age o f t he a nima l, in line wi t h the fa ll in basal ce ll mi tosis rate over the same t ime period. These data suggest t hat EGF receptor distribution a nd numbe r play a n important role in regu la ting epi t helia l t issue development in sk in . To explore t his co ncept furth er t his report doc umen ts t he distri­bution of EGF receptors in rat emb ryo nic ski n including t he morphologic cha nges involved in ha ir follicle development, and in t he cyclical events of hai r growth . We show t hat EGF recepto rs a re ge nera lly spatia lly co rrelated with the rapidly proliferating or undifferentiated epit he lial components of sk in during embryon ic development and t he ha ir growth cycle and provide evidence that EGF or a related ligand may be involved in t he induct ion of epiderma l downgrowths during t he init iation of ha ir follicle development.

MATERIALS AND METHODS Au.toradiography

Autoradiography was performed as prev iously desc ribed 17J with minor modifications. The viable ski n explan t,s (about 0.2 x 0.5 cm) were incubated with gentle shaking at 22·C for 90 min in Hanks' balanced salt solut io.l containing 1 mg/ml bov ine serum albumin and 15 mM N-2- hydroxyethylpiperazine-N' -2-ethanesu lfonic ac id (Hepes) pH 7.3 (i ncubation medium ) with 6 nM '"" I-labeled EGF and where approp riate , 4 /lg/ml of unlabeled EGF. Tissues were ext~nsive ly washed in t he incubation medium for a further 60 min , fi xed overnight at 4· C in 3.5% paraformaldehyde in phosphate-buffered saline pH 7.4, processed for autoradiography using II ford K2 nuclear research emul­sion ]7], and developed, fi xed, and stained with hematoxyl in and eosin as previously described 17]. Autoradiographs of embryonic sk in were exposed for 6- 10 weeks and for 6- 12 weeks for the hair cycle studies. Epidermal growth factor was purified from mouse submaxi llary glands by the method of Savage and Cohen 18J and iodinated using the chloramine T method [9] to give an initial specific activity of about 120 /lCi / /lg. Control experiments, contai ning 4 /lg/ ml unlabeled EGF showed only background labe li ng on t he autoradiographs, confirming the specificity of t he binding of '""I-labeled mouse EGF to rat skin ]7J.

T£ssues

Dorsal skin was removed from embryos of 15- 20 days gestation wh ich had been removed by cesarean section from Wistar rats killed by CO2 asphyx iation. T he day after vaginal plug formation was desig­nated day 1 of gestation. For hair cycle studies, the dorsal skin of postnatal Wistar rats aged 6, 19- 23, and 50 days was shaved after noting the angle of hair growth and the skin samples excised. The shaved skin was cut in to strips about 0.2 x 0.5 cm with the direction of hair growth parallel to t he longer edge. Following tissue processing, sagitta l sections were cut in order to obtai n longitudinal cross-sections through hai r follicles.

Microscopy

Pa ired photographs were taken, using either interference-ref1ection microscopy ]7,10J or dark-lie ld illumination to highlight the s ilver

11 8

Aug. 1984 EGF RECEPTORS IN EMBRYONIC DEVELOPMENT AND HAIR CYCLING 119

gra i ns and bright.-fi eld illumination to display the histology, on I1ford XPI 400 or Kodak Ektachrome 400 film using a Leitz Ortholux II mic roscope equipped with epi -illumination, a super-pressure 50-W mercury lamp (Osram) , and phase contrast and interference-reflection opt ics (50 X NPL fluotar objective).

RESULTS

Embryonic Skin

Fetal rat skin develops rapidly from 16 days gestation to birth [11- 13] with the transformation of an undifferentiated epithelium of 3- 5 layers t hrough to a cornified, multilayered epithelium at parturItIOn. At 16 days gestatIOn, speCIfic label ­ing, reflecting the presence of unoccupied, accessible EGF receptors, is found over the stratum basale, stratum interme­dium, and periderm cell layers making up t he epidermis (Fig 1a,b). At 17 days gestation a higher level of receptors for EGF is detected , but the receptors, in nearly a ll the skin samples examined, are now largely restricted to the basal cells (Figs 1c, d 2a,b), demonstrating a loss of receptors from the upper e~idermal layers and an apparent increase in basal cell [' 2"1] EGF binding. Interestll1gly, between 17- 20 days, label IS con­sistently absent over epidermal cells immediately above dermal fibroblast condensates (arrows Figs 1e,d and 2a,b (17 days) and 19 h) (20 days); 18 and 19 days not shown) which mark the fir~t stage of hair follicle development (11- 14).

By 18 days gestation , hair germ downgrowths have developed with heavy labeli ng of the component epithelial cells, particu­larly toward the middle and lower end of the epithelial column. The labeling of the downgrowths is continuous with that of the epidermal basal ce lls. Further, a shar~ ly reduced n~mber of receptors is observed on the dermal papIlla (arrows FIg 1e,f) as is found in neonatal skin [7] and throughout the hair growth cycle (see below). Reduced labeling is a lso seen over the com­ponents making up t he connective tissue sheath enve loping t he downgrowth (Fig Ie.!).

At 20 days gestation (Fig 19,h) epidermal stratification is clearly apparent and EGF receptors, though fewer in number as judged by silver grain density, are sti ll restricted to the epiderma l basal cells, similar in pattern to that found for neonatal [7] and adult (not shown) epidermis.

Intense labeling of developing striated muscle in the lower dermis is seen at 17 and 18 days gestation (Figs le-f, 2a,b) which allows the presumptive muscle and dermal fibroblasts to be easily distinguished. At 20 days the muscle cell receptor number, as judged by silver grain density, appears reduced (not shown) and similar in level to neonatal skin [7].

The Hair Growth Cycle

The mammalian hair growth cycle [15,16] consists of a continuous sequence of morphologic events usually subdivided into 3 distinct phases, those of growth (anagen), breakdown (catagen) with regression and resorption of the hair follicle, and rest or dormancy (telogen). The distribution of receptors for EGF during the hair growth cycle is described here in sequential order commencing with mid-anagen.

Midanagen to Early Catagen

A typical example of the distribution of EGF recepto rs in the midanagen hair follicle similar to that already reported [7] is shown in Fig 3a,b. Heavy labe ling is found on the outer root sheath and the first few cell layers overlying the basement membrane separating t he hair matrix from t he connective t iss ue sheath particularly at the base of the hai r follicle. Label­ing is generally absent or very low over the basal epit helial cells encircl ing the upper end of t he dermal papilla though it is sometimes seen on a few basal and supra basal epithelia l cells over the lower extremity of the dermal papilla (arrows Fig 3a, b) . Prolonged incubation does not alter t his characteristic midanagen distribution, suggesting t hat the absence of label over the basal cells adjacent to the dermal papilla is not due to incomplete penetration of [' 2['I]EGF in to t he hair fo llicle. A

FIG 1. Distribution of label following incubation of fetal rat skin with 6 nM [' 25I1EGF and processing for autoradiography. Bright-field illumination is used on the left (a) to display the histology, and interference-reflection optics are used on the right (b) to highlight the silver grains. The silver grain distribution shows the location of unoc­cupied, accessible EGF receptors. a and b, Sixteen-day embryonic skin showing label over all the epidermal cell layers. c ana d, Seventeen-day embryonic skin with EGF receptors in the epidermis now largely restricted to the basal cells. The dermal condensate (arrows) underlies a region of greatly reduced epidermal cell labeling. e ana t, Eighteen­day embryon ic skin showing heavy labeling of the epithelial down ­growth and a striking reduction of label over the dermal papilla (ar ­rows ). g an.d h , Twenty-day embryonic epidermis. A dermal condensate (arrows) underlies a region of reduced epidermal labeling. S cale bars =

30 !lm.

120 GREEN AND COUCHMAN

FIG 2. Q and b, Seventeen-day embryonic skin, displayed as de­scribed in Fig 1, showing very heavy labeling of the developing striated muscle in the lower dermis. A dermal condensate (arrows) underlies a region of reduced epidermal cell labeling. Labeling of developing striated muscle can also be seen in Fig 1c-f. Scale bar = 30 lim.

study of early anagen (see below) also indicates that the lack of labeling found around the upper dermal papilla is not due to incomplete penetration of [12IiIJEGF but reflects an absence or prior occupation of EGF receptors. No labeling above back­ground is found over the dermal papilla cells. The EGF receptor distribution in the human scalp anagen hair follicle is spatially the same as that found for the rat (unpublished observation).

With the onset of early catagen there appears to be no significant change in tissue receptor distribution. Fig 3c,d (50-day rat) shows a hair follicle entering catagen at the end of the second hair growth cycle with some cell necrosis and follicle thinning and it has a similar receptor distribution to the midanagen hair follicle. This indicates that a gross change in EGF receptor population and localization is not involved in the initiation of catagen.

Midcatagen to Telogen

At midcatagen (Fig 3e,f; 20-day rat ) the degenerating epithe­lial strand is heavily labeled while, in sharp contrast, the partially compacted dermal papilla (arrows Fig 3e,f) shows very little or no labeling above background. The un degraded strand of connective tissue sheath fibroblasts remaining below the dermal papilla as the epithelium retracts toward the skin sur­face is poorly labeled. At telogen no label is found over the compacted dermal papilla (arrows Fig 3g,h) while the secondary hair germ is heavily labeled with less label present over the undifferentiated cells of the outer capsule in the club hair. Very little labeling is found on the keratinized club hair and the

F IG 3. Distribution of EGF receptors in anagen, catagen, and telo­gen hair folli cles from postnatal rats. Photographs are disp layed as described in Fig 1 except for Fig 3d where dark-field illumination is used to highlight the silver grains. a and b, Anagen hair follicle from 6-day rat skin showing heavy labeling of the outer root sheath and the epithelial cells (ar rows ) adjacent to the lower end of the dermal papilla. Characte ristically a low level of labeling is seen over the dermal papilla fibrob lasts. c and d, Early catagen hair fo llicle from 50-day rat skin showing evidence of necrotic cells in the upper bulbar region and no significant changes in EGF receptor distribution compared wi th the anagen hair fo llicle. e and r, Midcatagen hair follicle from 20-day rat skin. The dermal papilla remains unlabeled (arrows) while the degen-erating epitheli al cell strand containing the presumptive secondary hair

c

germ is heavily labe led. g and h, Telogen hair fo llicle from 20-day rat 9 skin. The dermal papilla is unlabeled (arrows) while the secondary germ has a high level of EGF receptors. Scale bars = 30 lim.

Vol. 83, No. 2

Aug. 1984 EGF RECEPTORS IN EMBRYONIC DEVELOPMENT AND HAIR CYCLING 121

FIG 4. Localization of EGF receptors during stages of anagen hair follicle development in the 21-day-old rat. a and b, Very early anagen hair follicle showing intense labeling of the proliferating epithelial downgrowth. The cells at the tips of the down growth appear to ha~e a reduced level of labeling (arrows) . c and d, Early anagen hair follIcle with heavy labeling of the outer root sheath and the epithelial layers su rrounding the dermal papilla. A small reduction in label density is apparent ove r the cells surrounding the upper part of the dermal papilla. e and [, A more mature anagen hair follicle showing clear depletIOn of the cell layers around the top and middle of the dermal papilla. The distribution of EGF receptors now approximates that shown in Fig 3a,b. Fig 4a- [ were all obtained from 5 serial sections of skin . Scale bars = 30 I'm.

partially keratinized elements of the capsule lying between t he undifferent iated outer cells and the club hair. The remaining components of the outer root sheath are labeled (not shown), being cont inuous with the undifferentiated marginal cells of the sebaceous duct and the basal cells of the infundibulum and the epidermis [7) .

Early Anagen

The proliferating migrating germ is heavily labeled (Fig 4a, b), though the t ip of the migrating front of epithelial cells appears, in part, to be depleted of EGF receptors (arrows, Fig 4a,b), perhaps indicating that these cells are committed to cell migration and invasion as opposed to cell proliferation. These migrating cells would also lack epithelial cell-cell interactions at t heir leading edges and may be interacting with dermal mat rix components lying close to the newly synthesized base­ment membrane. t

At a later stage in anagen, receptors are still retained over cells, 3 or 4 layers deep, lining the dermal papilla (Fig 4c,d) though the number of available EGF receptors at the upper end of the dermal papilla is apparently reduced. As the follicle approaches midanagen (Fig 4e,() with clearly defined streaming of cells in the upper bulbar region , receptors are largely lost from the epithelial tissues adjacent to the upper half of the dermal papilla, with the receptor distribution now gradually approaching that found for midanagen (Fig 3a,b). Figs 4a-f are micrographs of hair follicles taken from 5 serial skin sections (21-day rat) showing that it is highly unlikely that experimental variations or incomplete ligand penetration are responsible for these observed changes in silver grain dist ribution.

Throughout the rat hair growth cycle the dermal papilla fibroblasts undergo considerable morphologic changes [15,16); however, at no stage is there evidence for anything other than a very low number of accessible EGF receptors on these cells.

DISCUSSION

There is now increasing evidence for important roles for EGF or EG F -like molecules in tissue embryonic maturation [3,17) including the development of the lung [18,19], secondary palate [20,21]' and skin [1,1 7,22,23). In addition, the binding of [1 2SI) EGF to a range of fetal tissues has been detected [6,24-26), EGF receptor kinase activity has been observed during rodent embryonic development in various tissues including mouse skin [27], and EGF or EGF -like molecules have been found in mouse fetal homogenates [24) . The distribution of EGF receptors in mouse palatal shelves from 13-day embryos has been described [24) with receptors being detected over all t he epithelial cell layers, similar in pattern to that found here in 16-day embry­onic rat epidermis (Fig 1a,b). Furthermore, skin development in fetal lambs in vivo is accelerated by infusion ofEGF, causing hypertrophy of the sebaceous and sweat glands and t hickening of the epidermis [17,22], and studies in vitro have shown that EGF can stimulate epidermal proliferation, keratinization, and thymidine uptake in chick embryonic epidermis [1 ,23]. Epider­mal growth in the rat is rapid from 16- 17 days gestation until birth [11- 13) and from 17-20 days (Figs 1c-h, 2a,b) . Consistent with this, we find t hat receptors for EGF are closely and preferent ially associated with basal cells which predominantly have proliferative capacity in t he epidermis. Growth, matura­t ion, and differentiat ion of the striated muscle in the lower dermis is rapid over the same t ime period and these events coincide with an initial very high level of detectable EGF receptors on the muscle cells at 17 days gestation which de­creases with age. This localization of EGF receptors on rapidly proliferating cells implies a role for the EGF receptor, in conjunction with EGF or an EGF-related ligand [4,24,28) in

t Couchman JR, Gibson WT: Submitted.

122 GREEN AND COUCHMAN

embryonic epidermal development as well as some types of mesenchymal tissue maturation.

The apparent increase in EGF receptor number and their "polarization" exclusively to the basal cells which occurs be­tween 16 and 17 days gestation (F igs la- d; 2a,b) coincides with, or lags briefly behind, the increase in commitment of epidermal ce lls to mitosis [13] as measured by the ["H]thymidine labeling index. Several ultrastructural changes relating to enhanced epidermal differentiation also occur over the same t ime period in rat skin [29,30]. Bauer [29] reports that keratohyaline gran­ules and "compound" granules appear in suprabasal cells at 17 days gestation (adjusted to compare with this study), the ob­servations being consistent with those of Bonneville [30]. Fur­ther, the relative amount of fibrous proteins present in rat epidermis increases from 16 to 20 days gestation [31] in line with the enhanced epidermal keratinization over this period. In 20-day embryonic rabbit skin when the epidermis is 2 cell layers thick and at an early stage of structural development, commitment to terminal epidermal differentiation is marked by t he appearance of a "high"-molecular-weight 65 K keratin protein [32]. It is not possible at this stage to link any of these observations, however the possibility arises that the polariza­tion and increase in EGF receptors described here between 16 and 17 days gestation repesent a biochemical process related to in the onset of enhanced embryonic epidermal proliferation and/or keratinization or a biochemical change involved in the commitmen t to terminal epidermal differentiation.

The loss of detectable basal cell epidermal EGF receptors above dermal fibroblast condensates in 17 to 20 days gestation rat skin (Figs le,d; 19,h; 2a,b) suggests another important role for the EGF receptor in embryonic skin maturation. The most straightforward interpretation of this observation is that the EGF receptors on the basal cells have been occupied, down regulated, or otherwise masked by molecules synthesized by the proximal dermal condensates. Alternatively, EGF receptors may not be synthesized by the epidermal cells proximal to the dermal condensates or may have been internalized or degraded by an EGF-independent process. This loss of receptors from the basal cells may also be linked with the transient inability of the same epidermal cells to incorporate I"H]thymidine (see [13] for a discussion). These observations suggest that EGF or a related ligand [4,24 ,28] could be involved in epithelial induc­tion by a mesenchymal t issue, and play a role in dermal/ epidermal interactions (33) required for hair development. To our knowledge this is the first biochemical evidence showing a localized change in the epithelium connected with the initiation of hair fo llicle formation.

Mouse EGF is known to influence hair development and the hair growth cycle [2,17,34- 37]. EGF, administered by cutaneous injection, promotes epidermal growth and keratinization in neonatal mice [2,34,36) but at the same time inhibits hair follicle development and hair growth [34-36J. In addition, high doses of EGF cause a transitory cessation of wool growth, leading to a weakness or complete break in the wool fiber thus allowing the fleece to be removed by hand [37]. These data indicate a differing response of hair follicles and epidermis to administration of the growth factor and also suggest that the effects of high levels of exogenous submaxillary gland EGF may not accurately reflect the physiologic functions of naturally occurring EGF(s) in skin.

With one exception, throughout the hair growth cycle we always find large numbers of receptors able to bind EGF associated with the proliferating [38] and undifferentiated com­ponents of the hair follicle; receptors are absent only over the ce ll layers lying next to the upper end of t he dermal papilla during full anagen and early catagen (Fig 3a-d). Receptors are absent or reduced over the partly or fully differentiated ele­ments of the hair shaft and cells which have started to differ­entiate [7], underlining t he close association of EGF receptors with the areas of potential or actual epithelial proliferation in

Vol. 83, No.2

skin [7]. During midcatagen the ce lls comprising the resorbing epithelial strand are heavily labeled (Fig 3e,f) . These cells do not proliferate but have the potential to do so, in that a proportion give rise to the secondary hair germ. At telogen (Fig 3g,h), the secondary hair germ is characterized by a high level of receptors for EGF. These receptors are available for EGF binding and thereby provide a potential mechanism for the initiation of cell proliferation at the start of anagen . The presence of EGF receptors during t he remainder of the growth cycle raises the possibility of their involvement in regulating proliferation in the hair bulb and hair development.

This role for the EGF receptor requires the presence of a trigger ligand able to bind to the receptor [2]. A potential source of the necessary diffusable EGF or EGF-like [4,24,28] protein able to bind to the EGF receptor could be the dermal papilla, which is essential for hair development [39,40], and, interest­ingly, we find no labeling or a very low level of labeling associated with this structure throughout the hair cycle (Figs 3a- h; 4a-f). The cells of the dermal papilla may simply have a low number ofEGF receptors or may not express EGF receptors at all. A third possibility is that "down regulation" or occupation of dermal papilla receptors has occurred in an autocrine fashion through secretion, by the dermal papilla cells themselves, of ligands able to bind to EGF receptors. The ['H) thymidine labeling index [41] and proliferation [42] of the dermal papilla cells is low, except during anagen substage 4 [41] before the hair has emerged above skin surface [15], when a remarkable 10- to 20-fold increase in labeling index is detected [41). Pierard and de la Brassine [41) find a graded decrease in labeling in ["H]thymidine labeling index moving further out from the anagen hair follicle and suggest that this could rise from a gradient of stimulation issued from the hair bulb, providing further indirect support for the presence of a diffusable li­gand(s) synthesized by the dermal papilla during the hair cycle.

In summary, we find high levels of EGF receptors on the proliferating and/or undifferentiated epithelial components of skin during embryonic development and the hair growth cycle and provide evidence that the EG F receptor and associated ligand(s) may playa role in embryonic t issue maturation and hair cycling.

We thank G. Boxall and D. Dix for skilled histological assistance, C. B. Evans and S. Hawes for processing photomicrographs, G. West­gate for assistance in identifying skin containing catagen hair fo llicles, and D. A. Basketter for helpful discussions.

REFERENCES 1. Cohen S: The stimulation of epidermal proliferation by a specific

protein (EGF). Dev BioI 12:394-407, 1965 2. Carpenter G, Cohen S: Epidermal growth factor. Annu Rev

Biochem 48:193-216, 1979 3. Gospodarowicz D: Epidermal growth factor and nerve growth fac­

tors in mammalian development. Annu Rev Physiol 43:251-263 1981 '

4. Das M: Epidermal growth factor: mechanisms of action. Int Rev Cytol 78:233- 256, 1982

5. Schlessinger J, Schreiber AB, Levi A, Lax T, Libermann T, Yarden Y: Regulation of cell proliferation by epidermal growth factor. eRC Crit Rev Biochem 14:94- 111, 1983

6. Adamson ED, Warshaw JB: Down-regulation of epidermal growth factor receptors in mouse embryoes. Dev BioI 90:430-434, 1982

7. Green MR, Basketter DA, Couchman JR, Rees DA: Distribution and number of epidermal growth factor receptors in skin is related to epithelial cell growth. Dev BioI 100:506- 512, 1983

8. Savage CR, Cohen S: Epidermal growth factor and a new derivative. J BIOI Chern 247:7609- 7611,1972

9. Carpenter G, Cohen S: 12S1-Labelled human epidermal growth factor. J Cell BioI 71:159- 171, 1976

10. Curtis ASG: The mechanism of adhesion of cells to glass. A study by mterference reflection microscopy. J Cell Bioi 20:199-215 1964 '

11. Fraser DA: The development of the skin of the back of the albino rat until the eruption of the first hairs. Anat Rec 38:203-223 1928 '

12. Hanson J: The histogenesis of the epidermis in rat and mouse. J Anat 81:174- 197,1947

13. Stern lB, Dayton L, Duecy J: The uptake of tritiated thymidine by

A ug. 1984 EGF RECEPTORS IN EMBRYONI C DEVELO PMENT AND HAIR CYCLI NG 123

14.

15. 16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

t he dorsal epide rmis of the fetal and newborn rat. Anat Rec 170:225- 234, 1971 .

Westgate GE, Shaw DA, Harrap GJ , Couchman J R: Im munohIs­toc hemica l loca lIzatIOn of basement membra ne components dur­ing ha ir follicle morphogenesis. J Invest Dermatol 82:259- 264, 1984 ..

C hase HB: Growt h of t he ha lT. PhyslOl Rev 34:113-126, 1954 Kli gma n AM : The huma n hair cycle. J Invest Dermatol 33:307-

316, 1959 R Thorburn GO, Waters MJ , Young I , Dolling M, Bunt ine 0 ,

H opkins PS: Epiderma l Growt h Factor: A Critical Factor in Feta l Maturation? The Fetus and Independent Life (Ciba Foun ­dation Symposium 86) . London, P it ma n, 1981, pp 172- 198

Ca tte rton WZ, Escobedo MB,Sexson WR, Gray ME, Sundell HW, Stahlma n MT: Effect of epIdermal growt h factor on lung matu­ration in fetal rabbI ts. Pedlatr Res 13:104- 108, 1979

S undell HW, Gray ME, Seven ius FS, Escobedo MB, Stahlman MT: E ffects of epIdermal growt h factor on lung maturatIOn 111

fetal la mbs. Am J Pathol 100:707-726, 1980 H asse ll JR: The development of rat pa lata l shelves in vitro. Dev

BioI 45:90- 102, 1975 H assell JR Pratt RM: Elevated levels of cAMP alte rs t he effect of

epiderm~ 1 growt h factor in vitro on programmed cell death in t he secondary palatal epIthelIum. Exp Cell Res 106:55- 62, 1977

D olling M, Thorburn GD, Young IR: Effects of epidermal growth factor on the sk1l1 of the fetal lamb (abstr). J Anat 136:656, 1983

Bertsc h S Ma rks F: Effect of foetal calf serum and epidermal growt h ~actor on DN~ synthesis in explants of chick embryo epidermIS. Nature 251.517-519, 1974

N exo E Hollenberg MD, Figueroa A, Pratt RM: Detection of epide;mal growt h fac tor-urogastrone and its receptor during fetal mouse development. Proc Natl Acad SCI USA 77:2782-2785, 1980 .

Ada mson ED, Deller MJ , Warsh~w JB: FunctIOnal EGF receptors are present on mouse embryo t Issues. Nature 291:656-659, 1981

U day P Devaskar MD: Epidermal growth factor recepto rs in fetal and ~aternal rabbit lung. Biochem Biophys Res Commun 107·714- 720, 1982

Horts~h M , Schlessenger J, Gootwine E: Webb CG: Appeara nce of functiona l EC F recepto r kll1ase dunng rodent embryoge l11sls. EMBO J 2:1937- 1941, 1983

D e La rco JE, T odaro GJ: Growth factors fro m murine sa rcoma virus t ra ns formed cells. Proc Natl Acad SCI USA 75:4001- 4005,

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

1978 Bauer FA: Differentiation and keratinisation of fetal rat skin.

Dermatologica 145:16- 36, 1972 Bonneville MA: Observations on epidermal diffe rent iation in the

feta l rat. Am J Anat 123:1 47- 164, 1968 Dale BA, Stern IB, Rabin M, Huang L-Y: T he identificat ion of

fibrous proteins in fetal rat epide rmis by electrophoretic and immunological techniques. J Invest Dermatol 66:230- 235, 1976

Ba nks-Schlegel SP: Keratin a lte rations during embryonic epide r­ma l diffe rentiation: a presage of adul t epide rmal maturation. J Cell BioI 93:551-559, 1982

Cohen J : Dermis, epidermis a nd dermal papillae in te racting, Ad­vances in Biology of Skin , vol IX, Hair Growth. Edited by W ~ontagna, RL Dobson. New York , Pergamon P ress, 1967, pp 1-

Steidler NE, Reade PC: Histomorphological effects of epidermal growt h factor on skin and ora l mucosa in neonata l mice. Arch Oral Bioi 25:37- 44, 1980

Moore GPM, Pana retto BA, Robertson D: Effects of epidermal growth facto r on hair growt h in the mouse. J Endocri nol 88:293-299, 1981

Moore GPM, Panaretto BA, Robertson D: Epide rmal growth facto r delays t he development of the epidermis and t he hair follicles of mice during growth of t he first coat. Anat Rec 205:47- 55, 1983

Moore GPM, Panaretto BA, Robertson D: Inhibi t ion of wool growth in merino sheep following administration of mouse epi ­dermal growth factor a nd a derivative. Aust J BioI Sci 35:163-172, 1982

Spearman RIC: The structure and function of t he fully developed follIcle, T he Physiology a nd Pathophysiology of the skin , 4th ed. Edited by A J a rrett. New York , Academic P ress, 1977, pp 1293-1349

Oliver RF: Whis ker growth a fte r removal of t he de rma l papilla and lengths of follicle in t he hooded rat. J Embryol Exp Morphol 15:331- 347, 1966

Oliver RF: The experimental induction of whisker growth in the hooded rat by Implanta tion of dermal papillae. J Embryol Exp Morphol 18:43- 51, 1967

P ierard CE, de la Brassine M: Modulation of derma l activity during hair growth in t he rat. J Cutan Pathol 2:35- 41, 1975

Wessels NK, Roessner KD: Non-pro liferation in dermal conden­sates of mouse vibrissae and pelage hairs. Dev BioI 12:419- 433 1965 '