1. This paper reports the effects of excess vitamin A on the development of hair follicles in organotypic cultures of embryonic mouse skin. Pieces of trunk skin, upper lip and lower jaw were cultivated on the surface of a clot of adult cock plasma and chicken embryo extract in slide preparations for 11 – 14 days. Changes in individual living follicles were observed daily and the findings confirmed from complete serial sections.

  2. In skin from the trunk of -day embryos the pelage follicles were initiated and underwent some normal differentiation, both in the presence and absence of added vitamin A (6·2 – 37·5 i.u./ml.).

  3. In skin from the trunk of 15-day embryos the pelage follicles differentiated normally for 4 days in both treated and untreated groups but then regressed in the vitamin-treated groups (12·5, 25·0 i.u./ml.), while normal differentiation continued and some keratinized hairs were formed in the control groups.

  4. Vibrissal follicles in the upper lip from a -day embryo differentiated normally for 4 days in excess vitamin A (12·5 i.u./ml.), and then some follicles became misshapen and developed gland-like lateral buds, while follicles in the control explant continued normal development and produced many keratinized hairs.

  5. All the vibrissal follicles originally present in four explants of upper Up from 14-day embryos differentiated normally for about 3 days in excess vitamin A (12·5 i.u./ml.) and then showed metaplastic changes. Many of them underwent a complete glandular metaplasia, being transformed into compound tubular glands with branching duct systems, terminal ‘alveolar’ swellings and occasional signs of mucous secretion. Some follicles retained traces of the original structure, and in a few the follicle eventually recovered its capacity for normal differentiation, producing simultaneously a small keratinized hair and a vigorously growing gland. The metaplastic glands were unlike any normal mammalian skin glands but resembled immature salivary glands. At the time of initiation of the metaplasia there was also an increase in the secretion of sebum or sebum-like material in the same hair follicles. In the control group without excess vitamin A the follicles continued normal differentiation and hair production until the experiment was terminated, and there was no metaplasia.

  6. Most of the vibrissal follicles in the explants of upper lip from 15-day embryos continued normal differentiation in the presence of excess vitamin A (12·5 i.u./ml.), although a higher proportion of the hairs were abnormal in structure than in the control group. In excess vitamin A a few of the initially least advanced follicles underwent glandular metaplasia, but no metaplasia was observed in the control group.

  7. Some of the differences between the behaviour of explants from embryos of different ages could be explained by the hypothesis that there is only a relatively short critical stage in follicle development during which vitamin A is capable of inducing metaplasia.

The demonstration of mucous metaplasia in chicken embryonic epidermis exposed to an excess of vitamin A (Fell & Mellanby, 1953) stimulated many further investigations (reviewed by Fell, 1964; Fell & Rinaldini, 1965; Dingle & Lucy, 1965), leading to new insights into control of differentiation and function at the cellular level.

The object of the present experiments was to study the effects of excess vitamin A on another keratin-producing system, the developing hair follicle, in organotypic cultures. Hairs and their follicles have received little attention from other investigators of vitamin A effects in vitro, although Fell & Mellanby (1953) referred to ‘evidence of an inhibitory effect on hair formation’ in some preliminary experiments with mouse embryonic skin, and New (1963) reported that ‘usually the development of hair follicles was suppressed’ in cultures of embryonic skin from the rat and mouse, both in the presence and absence of excess vitamin A.

Some unexpected and dramatic metaplastic changes in hair follicles exposed to an excess of vitamin A are described in this paper. New information obtained in the course of the same experiments about the effects of vitamin A on mam-malian epidermis (Hardy, 1965, 1967 a) and tissues of the oral cavity (Hardy, 1967 c) are reported elsewhere.

The mouse embryos used and the four experiments performed have already been described in the paper dealing with the epidermis (Hardy, 1967 a). The experiments are summarized in Table 1. Briefly, organotypic cultures of abdominal skin, upper lip and lower jaw were grown in a solid medium consisting of 3 drops of fowl plasma and 1 drop of chicken embryo extract on coverslips inverted over Maximow slides and incubated at 35·5°C. In each experiment the effect of adding certain concentrations of vitamin A was tested, and in Exp. 4, hydrocortisone was added to two of the four treatment groups to see whether vitamin A effects were suppressed. The explants were transferred to fresh clots every 3 or 4 days and subdivided when necessary to meet nutritional requirements.

Table 1

Summary of experiments and numbers of explants

Summary of experiments and numbers of explants
Summary of experiments and numbers of explants

All explants were fixed after 4 – 14 days, a few in buffered 2% osmic acid for 1 μ sections stained with toluidine blue, and the remainder in Zenker’s fluid for complete serial paraffin sections at 8 μ. The latter were stained (see Hardy, 1967 a) with Mayer’s haemalum, eosin and picric acid (HEP), Mayer’s haemalum and Alcian blue (HAB), or by the periodic acid Schiff procedure (PAS) in the presence or absence of diastase digestion and of counterstaining by Mayer’s haemalum.

The explants of skin from the trunk were either devoid of hair follicles (13|-day embryos) or contained pelage hair follicles in early stages of development (14-, 15-day embryos). Each explant of upper lip in Exps. 2 and 3 contained all the five rows of mystacial vibrissal follicles from one side of the face (Text-fig. 1) and each ‘half upper lip’ explant in Exp. 4 contained two or three rows. The lower jaws were bisected medially to provide explants of ‘half lower jaw’, of which those in Exp. 2 contained interramal vibrissal follicles while those in Exp. 4 contained submental follicles of a type intermediate between vibrissal and pelage.

Text-fig. 1

Head of mouse showing the main groups of vibrissae.

Text-fig. 1

Head of mouse showing the main groups of vibrissae.

The use of Maximow slides permitted daily microscopic examination of living explants and recording of the development stages of many follicles from day to day. Other follicles, however, were obscured. The stage of the most advanced follicle which was clearly visible in the living explant each day was arbitrarily chosen to record the stage of the explant, since there are wide variations in follicle development times and rates within one skin region both in vivo and in vitro. In a fixed explant the stage recorded was that of the most advanced follicle found in the complete serial sections, and there was good agreement between observations on living and fixed tissue. Stages 1-8 in hair follicle development in the mouse as originally defined for the pelage hairs by Hardy (1949), and a later subdivision of stage 3 (Hardy, 1967b), were used. Table 2 summarizes the criteria for recognizing these stages and the embryonic age at which each was first observed in sections of skin from the dorsolateral trunk (pelage) and the upper lip (vibrissae) respectively. The distinguishing morphological features of vibrissal follicles at all stages of development and their faster rate of development have been described previously (Davidson & Hardy, 1952).

Table 2

Summary of stages of development of pelage and vibrissal hair follicles in the mouse

Summary of stages of development of pelage and vibrissal hair follicles in the mouse
Summary of stages of development of pelage and vibrissal hair follicles in the mouse

To understand the peculiar orientation of mouse embryonic skin with developing hair follicles in coverslip cultures, the reader should consult earlier papers (Hardy, 1949, 1951). The overall behaviour of explants and the histology of the dermis and epidermis in the present series of experiments has already been described (Hardy, 1967a).

(1) Pelage hair follicles from the trunk of 5-day embryos (Exp. 1)

At the time of explantation, the skin contained the first-formed pelage follicles at the beginning of stage 2, evenly spaced over the skin surface. A few less developed follicles (stage 1) had appeared between them.

Observations on living explants and sections

During the first 4 days in vitro some normal follicle development took place in all treatment groups, and at least six explants reached stage 4, the normal stage for 19 days in vivo (Text-fig. 2). Some new follicles were initiated in most explants. Thereafter differences were noted in the living explants between control and vitamin-treated groups. Because of the curling of the epidermis around the dermis which frequently occurs in 15-day explants with the culture technique used, many follicles were prevented by poor nutrition from continuing to develop at the normal rate. However, in three of the eight control explants which were cultivated for 8 or more days, normal keratinized hairs were formed (Text-fig. 2), and six out of eight explants reached stage 4 at least. There was no development beyond stage 4 in either of the vitamin A-treated groups, and most of the follicles which could be distinguished in the living explants were seen to regress to earlier stages—3 a, 2 or 1. Many follicles disappeared completely. After 8 or more days of cultivation, none of the vitamin-treated explants were as advanced as stage 4. In one the follicles were observed regressing from stage 4 at 4 days to stage 2 at 8 days and then partially recovering, to stage 3 c after 11 days (Text-fig. 2).

Text-fig. 2

Graph showing rates of development of pelage follicles from 15-day embryos, × … ×, in trunk skin in situ; ● — ●, in explant showing maximum progress obtained in control medium (Exp. 1); ○ ‐ ‐ ○, in explant showing the maximum progress obtained in vitamin A medium (25·0 i.u./ml.). The meaning of the shaded bands in this and subsequent text-figs, is explained in the Discussion, p. 176.

Text-fig. 2

Graph showing rates of development of pelage follicles from 15-day embryos, × … ×, in trunk skin in situ; ● — ●, in explant showing maximum progress obtained in control medium (Exp. 1); ○ ‐ ‐ ○, in explant showing the maximum progress obtained in vitamin A medium (25·0 i.u./ml.). The meaning of the shaded bands in this and subsequent text-figs, is explained in the Discussion, p. 176.

(2) Pelage hair follicles from the trunk of -day embryos (Exp. 2)

Two control groups (combined in Table 1, but actually in media containing plasma from different batches) were compared with four groups receiving different levels of vitamin A. At the beginning of the experiment there were no signs of follicle development.

Observations on living explants and sections

The first follicles did not appear for 3 or 4 days, more slowly than in vivo. Once they began, some follicles proceeded to develop at about the normal rate, but when the experiment was terminated at 11 days the most advanced follicles were still only at stage 5 (Text-fig. 3). The numbers of successful explants in this experiment were too small for an evaluation of different treatments, but the following tentative conclusions were drawn: (a) the 37·5 i.u./ml. level of vitamin A, which was toxic to the skin (Hardy, 1967 a), was probably also inhibitory to the hair follicles (e.g. only one explant in four reached stage 2); (b) at all other levels of vitamin A (25·0, 12·5, 6·2 i.u./ml.) the development of hair follicles, though limited, was no more limited than in the control explants; (c) the most advanced hair follicles in the vitamin A-containing media appeared to be progressing normally (Plate 1, fig. A), except that in the living state they frequently had a swollen appearance or an irregular outline.

Text-fig. 3

Graph showing rate of development of pelage follicles from 1312-day embryos, ×···×, in trunk skin in situ; ○ ‐ ‐ ○, in explant (Exp. 2) showing the maximum progress obtained in vitamin A medium (6·2 i.u./ml.).

Text-fig. 3

Graph showing rate of development of pelage follicles from 1312-day embryos, ×···×, in trunk skin in situ; ○ ‐ ‐ ○, in explant (Exp. 2) showing the maximum progress obtained in vitamin A medium (6·2 i.u./ml.).

(3) Vibrissal follicles in the upper lip of a -day embryo (Exp. 2)

At the beginning of the experiment most of the largest mystacial vibrissal follicles in the two explants were at stage 3 a (Plate 1, fig. B), but an occasional one was at stage 3b. Many smaller follicles were at stages 2 and 1. The dermal papillae of most follicles were still shallow, and the hour-glass shape which is characteristic of the largest 14-day vibrissal follicles (Plate 1, figs. C, D) had not yet appeared.

Observations on living explants

After 2 days the control explant of upper lip reached stage 5 (Text-fig. 4) and the refractile cones of Henle’s layer of the inner root sheath were clearly visible. After 3 days it was at stage 6 and after 6 days at stage 8. The shape of the follicles was preserved, and well-developed hairs were seen. The vitamin-treated explant reached stage 4 after 2 days, when the follicles appeared swollen and flaccid. After 3 days, swellings were noted on the outer root sheaths of a few of the follicles (stage 3 c) and by 4 days these were well-defined buds. Elongation of the buds was observed in the remaining 3 days of culture, and the refractile cones of Henle’s layer were not seen in these follicles. Follicle differentiation remained 2 or 3 days behind that of the control explant, but continued until fixation at 7 days. Keratinization of the epidermis was delayed by 2 days in the vitamin-treated explant, but in other respects the structure of the skin appeared similar in the two cultures.

Text-fig. 4

Graph showing rates of development of vibrissal follicles from 1312-day embryo, ×···×, in upper lip in situ; ● — ●, in explant of upper lip (Exp. 2) in control medium; ○ ‐ ‐ ○, in explant of upper lip in vitamin A medium (12·5 i.u./ ml.) (the upper line represents maximum progress and the lower line minimum progress of those mystacial follicles which were most advanced at the beginning of the experiment); ▪ — ▪, in explant of lower jaw (interramal follicles) in control medium; □ ‐ ‐ □ in explant of lower jaw (interramal follicle) in vitamin A medium (12·5 i.u./ml.).

Text-fig. 4

Graph showing rates of development of vibrissal follicles from 1312-day embryo, ×···×, in upper lip in situ; ● — ●, in explant of upper lip (Exp. 2) in control medium; ○ ‐ ‐ ○, in explant of upper lip in vitamin A medium (12·5 i.u./ ml.) (the upper line represents maximum progress and the lower line minimum progress of those mystacial follicles which were most advanced at the beginning of the experiment); ▪ — ▪, in explant of lower jaw (interramal follicles) in control medium; □ ‐ ‐ □ in explant of lower jaw (interramal follicle) in vitamin A medium (12·5 i.u./ml.).

Observations on histological sections

After 7 days in vitro the control explants contained many relatively thick (10 – 12 μ) keratinized hairs in follicles at stage 8 (Plate 2, figs. A – C). The vitamin A-treated explant had only a few very fine keratinized hairs (maximum diameter 4 – 5 μ) in follicles at stage 6, although all tissues were healthy and the dimensions of follicles and root sheaths were similar to those in the control explant. In the treated explant, three less advanced follicles (two at stage 5 and one at stage 4), which in the living state had shown lateral budding, were traced through serial sections. The main part of these follicles and the surrounding tissues were healthy and differentiating normally (Plate 2, figs. D – G). The buds were at first solid cylindrical outgrowths from the outer root sheath (Plate 2, fig. D) consisting of cuboidal epithelial cells. One had a terminal swelling (Plate 2, fig. G) and the beginning of a lumen (Plate 2, fig. F).

Several other glandular structures not seen in the living explants which were receiving excess vitamin A were discovered in the fixed tissue and two were traced through serial sections. They were both branching epithelial down-growths from the epidermis. One was mainly solid, with two unusually large sebaceous glands budding from it. The other had a duct with a long narrow lumen, a very small sebaceous gland and a solid terminal swelling with many mitoses. No trace of similar structures was found in the control explant. The significance of these epidermal appendages will be referred to in the discussion.

(4) Vibrissal follicles in the lower jaw of a -day embryo (Exp. 2)

At -days the three interramal follicles were at stage 1 in a papilla-like elevation of the skin in the mid-ventral line (Text-fig. 1). The development of two interramal follicles very close together was observed in the living control explant consisting of one half of the jaw, and sections after 11 days of cultivation showed one follicle at stage 5 (Text-fig. 4) and a second follicle base at stage 3 c joining the side of the first follicle. A third follicle close by, probably a pelage follicle, was observed first at 9 days and reached stage 3 a at 11 days. In the vitamin A-treated explant a single follicle, which was probably interramal, reached stage 2 at 2 days and stage 3 a at 4 days, but when fixed at 11 days was atrophic and had only reached stage 3 c. A few other atrophic follicles at stages 1 and 2 which were seen only in sections may have been submental vibrissal or pelage follicles. No budding from follicles was observed in the lower jaw explants.

Vibrissal follicles in the upper lip of 14-day embryos (Exp. 3)

The observations on budding from follicles made in Exp. 2 led to the design of Exp. 3, in which four pairs of carefully matched explants from the upper lip and lower jaw were compared. Some of the rapidly enlarging explants were divided after 4 days to ensure adequate nutrition, so that some groups eventually contained as many as seven explants.

Immediately after explantation the living mystacial follicles were seen to be at stages up to 3 b, and were a little more advanced than those in the -day embryos. The larger ones had acquired the characteristic hour-glass shape of vibrissal follicles (Plate 1, fig. D), and a well-defined connective tissue sheath (Plate 1, fig. C).

Observations on living explants

For the first 2 days in culture the vibrissal follicles developed normally although a little slowly in both treated and control groups (Text-fig. 5). After 3 days many follicles in both groups had reached stage 4, but in one explant in the vitamin A medium the epidermal parts of the follicles had rounded bases instead of being invaginated by dermal papillae.

Text-fig. 5

Graph showing rates of development of vibrissal follicles from 14-day embryos, ×···×, in upper lip in situ; ● — ●, in explants of upper lip (Exp. 3) in control medium; ⊙ ‐ ‐ ⊙, in explants of upper lip in vitamin A medium (12·5 i.u./ ml.) (the upper line represents maximum progress and the lower line minimum progress of the mystacial follicles which were most advanced at the beginning of the experiment); ▪ — ▪, in explants of lower jaw (submental follicles) in control medium; ⊡ – – ⊡ in explants of lower jaw (submental follicles) in vitamin A medium (12·5 i.u./ml.).

Text-fig. 5

Graph showing rates of development of vibrissal follicles from 14-day embryos, ×···×, in upper lip in situ; ● — ●, in explants of upper lip (Exp. 3) in control medium; ⊙ ‐ ‐ ⊙, in explants of upper lip in vitamin A medium (12·5 i.u./ ml.) (the upper line represents maximum progress and the lower line minimum progress of the mystacial follicles which were most advanced at the beginning of the experiment); ▪ — ▪, in explants of lower jaw (submental follicles) in control medium; ⊡ – – ⊡ in explants of lower jaw (submental follicles) in vitamin A medium (12·5 i.u./ml.).

At 5 days all of the five control explants containing epidermis had normally shaped follicles at stages 4, 5, 6 or 7, but no normal follicles were seen in any of the corresponding five vitamin-treated explants. Most of the follicles in the vitamin A group were long and thin, and had lost their dermal papillae, thick connective tissue sheaths and internal structure. Their epithelial outlines were irregular and buds were commencing to form from the outer root sheaths. At 7 days the follicles in control explants were still developing normally (stages 6, 7 and 8; Plate 3, figs. A, B), and small sebaceous glands with one to three differentiated sebaceous cells were observed in two out of five explants. In all of the four treated explants in which details could be observed clearly, most of the follicles were quite abnormal, and only one of normal appearance was observed, at stage 4. The others had no dermal papilla, and knobs or cylinders of cuboidal cells arose from one or more sites on the outer root sheath (Plate 3, figs. C, D). The thick connective tissue sheaths were absent from these irregular follicles. There was also an excessive development of sebaceous cells at this time. In every treated explant many of the follicles showed buds or parts of the original follicles which were filled with differentiated sebaceous cells and dark brown masses of sebum or sebum-like material.

At 9 days the control explants showed many normal follicles up to stage 8, but sebaceous glands were no longer seen. Most of the follicles in treated explants were abnormal, and the single normal follicle previously observed had now reached stage 7. The buds from other follicles had undergone extensive branching to form complex glandular but mainly non-sebaceous masses. Three out of four treated explants each contained in addition three or more sebaceous glands.

At 10 days all control explants showed a number of normally keratinized vibrissae, while only one of the vitamin-treated explants showed a follicle at stage 8 (Text-fig. 5). Sebaceous glands were not identified with certainty in any control explant, but were present in three out of four treated explants. The branching non-sebaceous glandular masses derived from follicles were seen in all treated but no control explants.

Some explants were fixed after 10 days and the remainder examined at 12 and 14 days. The two control tissues changed little during the remaining period of cultivation but the two vitamin-treated explants showed further proliferation of the non-sebaceous glands, and traces of patent ducts and terminal ‘alveoli’. The sebaceous secretion was also still very evident. The hair follicle in the treated explant which was previously observed at stage 8 was still visible, and two other follicles in the same explant reached stage 5. A second treated explant now had several normal-looking follicles at stage 4. Some of these ‘recovering’ follicles had acquired a fairly well developed connective tissue sheath, but still had an irregular epithelial outline and lacked the typical hour-glass shape.

Observations on histological sections

Three of the four control explants fixed after 10 days and the two fixed at 14 days showed essentially normal vibrissal follicles at stage 8, with relatively thick and normally keratinized hairs (Plate 3, figs. E, F). Some assumed distorted shapes, probably due to abnormal tensions in the explants, but none showed budding from the outer root sheath or abnormal gland formation. The sebaceous glands, which in vivo remain very small in vibrissal follicles, had disappeared from these explants. In each follicle the dermal papilla remained fully invaginated and was usually healthy and the thick connective tissue sheath was still present (Plate 3, fig. F). The fourth control explant fixed at 10 days was derived by subdivision at 4 days when it consisted of epidermis and dermis with no hair follicles. Newly initiated follicles, which were observed in the living explant 3 days later (i.e. after 7 days in vitro), were at stage 3 b at 10 days, and also appeared normal. The latter were probably pelage follicles, although it was not possible to establish this from morphological criteria. The sections stained in HAB showed alcian-positive reactions only in the locations where they are found regularly in normal active hair follicles in vivo (Braun-Falco, 1958).

By contrast, the vitamin-treated explants sectioned after 10 and 14 days showed a wide spectrum of hair follicle and gland differentiation. Follicles were found in various stages of departure from normal (Plate 3, figs. G, H; Text-fig. 5), and only one explant contained hairs. The three minute hair tips were found in follicles which had normal dermal papillae and root sheaths, and also some large sebaceous glands and extensively branching tubular glands (Plate 4, fig. A). Several other hair follicles had normal inner root sheath cones, sebaceous glands and branching tubular glands, but no hairs. Some had both types of glands and shrunken dermal papillae, but no root sheaths. The majority of follicles originally present were represented only by a solid downgrowth from the epidermis, from which the sebaceous and non-sebaceous glands were derived (Plate 4, fig. B), but some were lacking even this remnant of the original structure (Plate 4, fig. C). In two explants all the follicles originally present were completely transformed into sebaceous glands and tubular branching glands (Plate 4, fig. D; Text-fig. 5). There were also a few small follicles which, like those in a control explant referred to above, were initiated in vivo, and were developing normally up to stage 4 without any buds, but all the vibrissal follicles originally present showed some deviations from the behaviour of those in control explants.

Nearly all the vitamin-treated explants had many differentiated sebaceous cells and some sebum-like secretory masses produced by the characteristic holocrine process (Plate 4, fig. E).

The non-sebaceous branching tubular glands were the most prominent feature of the vitamin-treated explants, and in many instances by repeated branching they grew as large as the follicles in control medium. In nearly all explants there were some patent ducts opening either directly on to the surface of the epidermis (Plate 4, fig. B), or leading to a cyst (Plate 4, fig. F) which usually communicated by another duct with the skin surface. In the glands which were best differentiated the collecting ducts had a wide lumen and a two-layered wall of compact cuboidal or columnar cells with a smooth inner border (Plate 4, fig. B). Terminal bars and terminal webs were recognized in some ducts. The cysts (Plate 4, fig. F) were lined for the most part by a single layer of flattened cells which sometimes had an irregular border. The smaller ducts derived by branching had a single layer of cuboidal cells with a smooth border (Plate 4, fig. F). Most of the glands had some terminal swellings or ‘alveoli’ containing cuboidal cells (Plate 4, fig. G) which could sometimes be distinguished by paler nuclei and a more eosinophilic cytoplasm. The nuclei were central or slightly basal. Many terminal swellings did not show any central lumen in the rather thick sections but the majority had a few very small and irregular intercellular canals (Plate 4, figs. G, H). Although most of the cells in terminal swellings did not stain with Alcian blue, both of the explants grown for 14 days had occasional cells of the same morphological type whose cytoplasm was filled with Alcian-positive and therefore presumably mucous granules. One of the explants fixed after 10 days had Alcian blue staining material in some lining cells and in the lumen of several ducts (Plate 4, fig. I).

Vibrissal follicles in the lower jaw of 14-day embryos (Exp. 3)

The submental group of follicles, of intermediate type between pelage and vibrissal follicles, were at stage 1 or stage 2 when the experiment began. Conditions were not very favourable for the growth of hair follicles in the lower jaw explants, but there were some indications of a difference between control and treated groups. In the control group, follicles were observed at various stages in living explants (Text-fig. 5), and after fixation at 10 or 14 days, eight normal follicles at stage 8 were found in the sections from two explants and some follicles at earlier stages in the sections from two more. In the vitamin-treated group, follicles were identified in only one of the living explants, and after 10 or 14 days only one normal follicle at stage 8 and a few other follicles at earlier stages were found in sections from three explants. There was no sign of abnormal budding or gland development from these follicles.

Vibrissal follicles in the upper lip of 15-day embryos (Exp. 4)

Each explant contained two or three rows of follicles of which the most advanced was at stage 3 c, 4 or 5 and the least advanced at stage 1 (Text-fig. 6). Some of the follicles at stage 5 had a small swelling at the side to mark the position of the future sebaceous gland, although differentiated sebaceous cells do not appear in the embryo until about a day later, when the hairs begin to form.

Text-fig. 6

Graph showing rates of development of vibrissal follicles from 15-day embryos, ×···×, in upper lip in situ; ● — ●, in explants of upper lip (Exp. 4) in control medium (initially most advanced follicles); ○ ‐ ‐ ○, in explants of upper lip in vitamin A medium (12·5 i.u./ml.) (initially most advanced follicles); ▴ — ▴, in explants of upper lip in control medium (initially least advanced follicles); △ ‐ ‐ △, in explants of upper lip in vitamin A medium (12·5 i.u./ml.) (initially least advanced follicles).

Text-fig. 6

Graph showing rates of development of vibrissal follicles from 15-day embryos, ×···×, in upper lip in situ; ● — ●, in explants of upper lip (Exp. 4) in control medium (initially most advanced follicles); ○ ‐ ‐ ○, in explants of upper lip in vitamin A medium (12·5 i.u./ml.) (initially most advanced follicles); ▴ — ▴, in explants of upper lip in control medium (initially least advanced follicles); △ ‐ ‐ △, in explants of upper lip in vitamin A medium (12·5 i.u./ml.) (initially least advanced follicles).

Observations on living explants

After 3 days many follicles identified in the original explants had progressed to stage 6, 7 or 8 (Text-fig. 6), and the extent of progress was similar in all treatment groups. Sebaceous cells were recognized in about half of the explants. After 4 days there was further development of hair follicles in all groups, and more differentiated sebaceous cells were observed. Although some effects of vitamin A on the epidermis were apparent at 3 and 4 days (Hardy, 1967 a), no significant difference was found between treatment groups with respect to hair follicles. At 6 days, however, some differences showed up in the follicles. In the control group all 6 explants had reached stage 8 in follicle development and in the hydrocortisone (HC) group five out of six explants were at this stage, but in the vitamin A group only one explant and in the A + HC group only three explants out of six were at stage 8. Some follicles in one explant in the A group had lost their well-defined refractile Henle’s layer. The number of keratinized hairs observed after 6 and 7 days in vitro in the two groups receiving vitamin A was only about half as great as in the control groups, while the number in the HC group was greater.

Observations on histological sections

Explants sectioned after 7 and 11 days revealed additional differences between the treated and control groups. While the majority of explants in all groups had some follicles which had reached stage 7 or 8, there were differences in the numbers and the quality of hairs produced. Table 3 shows that each of the two groups receiving vitamin A had about three-quarters as many stage 8 follicles as the control group, but only one-third as many perfectly keratinized fibres (which stain bright yellow in HEP and remain unstained in HAB). The remaining hairs in all groups were pink, or pink and yellow, in HEP, and deep turquoise blue in HAB, and showed an unusual amount of bending, irregular outlines, imperfect cuticular scales, swollen hair cortex, loss of birefringence or some combination of these features denoting faulty keratinization, even though many of them had emerged beyond the skin surface and grown considerably in length.

Table 3

Numbers of keratinized hairs observed in explants of upper lip of 15-day embryos (Exp. 4)

Numbers of keratinized hairs observed in explants of upper lip of 15-day embryos (Exp. 4)
Numbers of keratinized hairs observed in explants of upper lip of 15-day embryos (Exp. 4)

The four treatment groups were similar with respect to the development of sebaceous glands. Differentiated sebaceous cells were identified in sections from one, two or three of the six explants in each group.

Although the formation of non-sebaceous glandular buds from hair follicles was not nearly as extensive in the 15-day upper lip skin as in the 14-day skin, there were indications that the same tendency to gland formation existed. In the vitamin A group, three of the four explants fixed after 11 days showed budding from a few of the smallest and least advanced vibrissal follicles (Text-fig. 6). One explant had a well-developed branching gland-type structure with a solid terminal bud, and a duct with a wall three cells thick and a lumen opening at the skin surface. A second explant had three budding structures derived from hair follicles which were identified in living explants—(a) a solid downgrowth from the epidermis with several buds but no dermal papilla, (b) a solid downgrowth with some sebaceous secretion, and (c) a bilobed gland with a duct lumen opening at the skin surface and the beginning of lumina in the two lobes (Text-fig. 6). The third explant had a follicle, whose dermal papilla had been lost, with a lateral bud containing some sebaceous secretion. In the A + HC group one of the four explants fixed after 11 days had a follicle with an elongated lateral bud, and a second explant had buds beginning to form from an outer root sheath. Neither the control group nor the one receiving hydrocortisone alone showed any of these budding or glandular structures, and all of their vibrissal follicles differentiated in a more normal fashion (Text-fig. 6).

Metaplasia of vibrissal follicles

The most radical changes in response to vitamin A occurred in the vibrissal follicles of the upper lip in Exp. 3. This appears to be a true metaplasia, in which cells of the outer root sheath or less differentiated follicle cells changed the direction of their development and formed branching tubular glands. Sometimes an entire hair follicle was transformed into an entire gland, and it seemed that most of the follicle cells of epidermal origin were involved, so that a potentially complex keratin-forming organ changed into an equally complex secretory type of organ. Other hair follicles were able to pursue the two developmental pathways simultaneously, and produce both a keratinized hair and a secretory-type gland. In Exps. 2 and 4 a few follicles from the upper lip underwent similar changes, and it is possible that the glandular structures which were attached directly to the epidermis in these experiments were also derived originally from hair follicle rudiments.

The exact nature of the glands formed remains an open question. They are morphologically distinct from sebaceous glands, which are usually simple saccular glands in the mouse, and even in their more elaborate forms such as the human Meibomian glands are branched saccular but not tubular glands. Furthermore, the metaplastic glands were not undergoing holocrine secretion or producing fatty substances. They could also be distinguished from the eccrine sweat glands of mammals, which are unbranched coiled tubular glands arising from the epidermis independently, and which are absent from the hair-bearing areas in the mouse. They differ also from the apocrine sweat glands, the unbranched coiled tubular glands with wide terminal lumina which are attached to the hair follicles in many species but not in the mouse. In species possessing apocrine sweat glands a single gland begins as a bud from the ental side of a hair follicle (i.e. in the obtuse angle between follicle and skin surface) above the neck of the sebaceous gland. The metaplastic glands, on the other hand, arose by budding at one or more points from any level of the follicle.

The origin of the metaplastic glands differed in several ways from that of mammary glands, the only remaining major group of glands derived from the mammalian epidermis. The mammary glands of the mouse develop from large rounded epidermal buds after a lag period of about 5 days (Balinsky, 1950 a, b) with a primary sprout from which secondary and later branches are derived. Mammary glands are characterized both in vivo and in vitro by early duct lumen formation and the absence of terminal swellings or alveoli during the early weeks of development (Hardy, 1950) unless suitable hormones are added to those already present in the usual biological media (Lasfargues & Murray, 1959).

Caution must be exercised in interpreting the nature of the secretory process in glands which are obviously immature. Certainly the metaplastic glands showed no signs of holocrine activity, and, in the absence of any evidence of apocrine secretion, they would be classified as merocrine or eccrine skin glands. The cells composing the terminal swellings corresponded morphologically to the ‘special serous’ (i.e. non-zymogenic) type discussed by Jacoby & Leeson (1959), but a few of these cells were filled with mucous granules and thus resembled the occasional ‘mucoserous’ cells of human mixed salivary glands (Bloom & Fawcett, 1962). One may therefore refer to the glandular metaplasia of hair follicles but there is as yet insufficient evidence for regarding this as a specifically mucous metaplasia such as is found in chick epidermis (Fell & Mellanby, 1953).

The vitamin A-induced glands bore a surprising resemblance to some structures of ectodermal but not epidermal origin—the developing salivary glands. The vitamin A-induced glands were compared with the developing mouse salivary glands as studied both in vivo and in vitro by Borghese (1950), and with the salivary glands which developed in the explants of lower jaw in the present experiments (Hardy, 1967 c). (The metaplastic glands were found only in the upper lip explants while the salivary glands were found only in the lower jaw explants, so there was no possibility of confusing them). The metaplastic glands resembled the submandibular rather than the sublingual glands in their general aborizing form, the histology of large and small ducts and the development of cysts from ducts in vitro, the histology of terminal swellings, and even the occasional appearance of morphologically ‘serous’ but histochemically ‘mucous’ cells. Since both the metaplastic glands and the salivary glands (Jacoby, 1959; Jacoby & Leeson, 1959) were still immature when the tissues were fixed, it is not known whether the subsequent differentiation of the two groups of glands would show a divergence.

Most of the ‘specialized’ glands in mammalian skin are merely elaborations of typical sebaceous and sweat glands. However, Lyne, Molyneux, Myktowycz & Parakkal (1964) found an exception in the chin-gland of the rabbit, which originated as an apocrine sweat gland from a hair follicle, but resembled the metaplastic glands of the mouse in two departures from the usual pattern. It underwent extensive branching to form a system of ducts, and it formed some eccrine secretory lobules as well as the usual apocrine ones.

Another transformation of hair follicles was described by Dawe (1963) in mice affected by polyoma virus. The sequence of events could only be inferred from a series of autopsy specimens, but there were some striking similarities .to the changes actually observed in the present study in response to vitamin A. Extensively branching tubular secretory glands and branching sebaceous glands developed from the side of hair follicles.

Effect of vitamin A excess on sebaceous glands

In Exp. 3 the glandular metaplasia in excess vitamin A was accompanied by a marked increase in the differentiation and secretion of sebaceous cells, but the sebaceous glands of control explants either did not form at all or began to develop and then disappeared. In Exp. 2 and 4 the majority of follicles in vitamin A-treated groups did not undergo metaplasia and their sebaceous glands were similar to those in the control groups. However, in the few metaplastic follicles produced by vitamin A excess, sebaceous differentiation and secretion were also greater than that in the non-metaplastic follicles of the control group. In all three experiments, differentiated sebaceous cells were located mostly in the usual position at the upper part of the follicle neck, but occasionally, like the non-sebaceous buds, they appeared at other levels along the follicle. Thus it seems that vitamin A may have a stimulating effect on sebaceous glands, at least under the conditions in which it also produces a glandular metaplasia of a non-sebaceous type. Sebaceous glands were apparently stimulated to differentiate and secrete rather than to grow faster or to undergo branching.

Interference with hair follicle development and hair differentiation caused by excess vitamin A

A number of effects of vitamin A excess on hair follicles were inhibitory to the follicles without adversely affecting the rest of the skin. The metaplastic changes in vibrissal follicles from the upper lip in Exps. 2 – 4 were all accompanied by some inhibition of normal follicle development. Even the follicles which achieved some normal development as well as glandular metaplasia showed a delay in follicle differentiation, and their hairs were either abnormally fine or absent. The majority of mystacial follicles from 15-day embryos (Exp. 4) differentiated normally and did not undergo glandular metaplasia, but the large number of imperfectly keratinized hairs can be attributed to the hypervitamino-sis. A small number of observations on vibrissal follicles from the lower jaw in Exp. 2 and 3 suggested a similar delay or inhibition of follicle development in the presence of excess vitamin A without any inhibition of epidermal growth or differentiation. Pelage follicle development also was inhibited in Exp. 1, and most of the follicles regressed. While the regression in 25·0 i.u./ml. vitamin A might perhaps be attributed to a toxic effect of the medium, since there were also a few degenerative changes in the epidermis and dermis (Hardy, 1967 a), this is unlikely to be the explanation for follicle regression in 12·5 i.u./ml., when the skin was just as healthy and vigorously growing as that in control medium.

Failure of hydrocortisone to affect hair follicles

Very little change in epidermal differentiation (Hardy, 1967 a) or in hair follicle development was brought about by adding hydrocortisone to either control medium or vitamin A medium in Exp. 4. This was unexpected, since hydrocortisone in similar concentrations antagonized the action of vitamin A in many organ culture systems (Lasnitzki, 1965), and also when administered alone affected rat skin in vitro (Weissmann & Fell, 1962). It is necessary to check the activity of the hydrocortisone preparation and to do further experiments before drawing any conclusions. The observations are included in this paper merely to provide complete data on vitamin A effects, since experimental work had to be discontinued for several years.

The ‘recovery’ of some follicles affected by vitamin A

Text-figures 2 and 5 and many reported observations on individual explants suggest that follicles which are severely affected by vitamin A after 4 to 6 days in culture may later recover some capacity for normal differentiation even though they are still in vitamin A medium and might have attached metaplastic glands. One would first suspect that this might be due to a falling off of vitamin A concentration in the medium, but from published analyses of similar media (Fell & Mellanby, 1953) it seems unlikely that the concentration would fall below an effective dose. In Exp. 3, the introduction of fresh vitamin A solutions on the 9th day failed to check this ‘recovery’ effect. Furthermore, other vitamin A responses continued in full force in the same explants (Hardy, 1967 a, c), and the metaplastic glands, once initiated, did not show any signs of regression towards the end of the experiments. Probably there is a real recovery of the ability of hair follicles to differentiate normally after some time in the high vitamin medium.

Is there a critical period in follicle development for the metaplastic action of vitamin A ?

It was surprising that the extensive metaplasia of mystacial follicles produced by vitamin A which was so widespread in all the 14-day explants was less common in a -day explant, infrequent in the 15-day explants and absent from the vibrissal follicles of the lower jaw. Perhaps there is not only a critical time interval between the administration of vitamin A and the appearance of a morphological change, but also a critical period in follicle development at which the metaplastic action of vitamin A can operate. In many investigations by Fell and her colleagues (see Fell & Rinaldini, 1965) and in the present series of experiments (Hardy, 1967 a, c) most of the histological changes in skin and other epithelial tissues began after 3 or 4 days of vitamin A treatment. Whatever the earlier critical time and critical stage may be for the significant changes at the biochemical level, what the histologist observes is a budding from vibrissal follicles if they are at the later dermal papilla or early hair cone stage (stages 3 c, 4) after about 3 days in excess vitamin A. To illustrate the hypothesis of a critical stage, in Text-figs. 2-6 the supposed ‘histological critical stage’ and ‘histological critical time’ are indicated by shaded bands. The majority of mystacial follicles from 14-day embryos reached stage 3 c or 4 at the critical time (Text-fig. 5), and so metaplasia was widespread. New follicles (not indicated in the figure) arising in vitro many days later were unaffected. The submental follicles from 14-day embryos treated with vitamin A reached stage 4 several days too late and did not undergo metaplasia. The largest mystacial follicles from a -day embryo for technical reasons developed more rapidly than those from 14-day embryos and reached stage 4 just before the critical time (Text-fig. 4). These were not greatly affected, but a few of the next largest follicles underwent a moderate degree of metaplasia. The majority of mystacial follicles developed a little later and were unaltered. The interramal follicles, developing even later, were also unaffected. The largest and medium-sized upper lip follicles from 15-day embryos were well past stage 4 at the critical time (Text-fig. 6) and did not show metaplasia, but some of the latest vibrissal follicles to develop in these explants met the conditions for the metaplastic change.

It is not known whether pelage follicles would undergo metaplasia given the right conditions, but it may be significant that those which reached stage 4 just after the postulated critical time (Text-fig. 2) underwent extensive regression, while those which were not initiated until after the critical time (Text-fig. 3) proceeded to develop normally. On the other hand, the failure of pelage follicles to undergo metaplasia may be another instance of the lower responsiveness to vitamin A of skin on the trunk as compared with skin on the extremities (Hardy, 1967a). This and many other questions, in particular several relating to the role of the mesenchyme, await further experimentation.

Métaplasie glandulaire des follicules pileux et autres réactions à un excès de vitamine A dans des cultures de peau de Rongeur

  1. Cet article rapporte les effets d’un excès de vitamine A sur le développement des follicules pileux dans des cultures organotypiques de peau embryonnaire de Souris. Des morceaux de peau du tronc, de la lèvre supérieure et de la mâchoire inférieure sont cultivés à la surface d’un coagulum de plasma de Coq adulte et d’extrait embryonnaire de Poulet, pendant 11 à 14 jours. Les transformations dans les follicules vivants sont observées chaque jour et les résultats confirmés par des coupes sériées complètes.

  2. Dans la peau du tronc d’embryons de 13,5 jours, les follicules du pelage apparaissent et commencent une différenciation normale; à la fois en présence et en absence de vitamine A (6,2 – 37,5 u.i./ml).

  3. Dans la peau du tronc d’embryons de 15 jours, les follicules du pelage se différencient normalement pendant 4 jours dans les groupes traités et non traités mais régressent ensuite dans les groupes qui reçoivent de la vitamine (12,5, 25,0 u.i./ml), tandis que la différenciation normale continue et que des poils kérati-nisés se forment dans les groupes témoins.

  4. Des follicules de vibrisses dans la lèvre supérieure d’un embryon de 13,5 jours se différencient normalement pendant 4 jours en présence d’un excès de vitamine A (12,5 u.i./ml.), puis certains follicules deviennent malformés et produisent des bourgeons latéraux d’apparence glandulaire; les follicules dans l’explant témoin poursuivent un développement normal et produisent de nombreux poils kératinisés.

  5. Tous les follicules de vibrisses présents au départ dans 4 explants de lèvre supérieure d’embryons de 14 jours se différencient normalement pendant environ 3 jours, en présence d’un excès de vitamine A (12,5 u.i./ml.), puis présentent des altérations métaplasiques. Beaucoup d’entre eux subissent une métaplasie glandulaire complète, et sont transformés en glandes tubulaires composées, avec un système de canaux ramifiés, des gonflements ‘alvéolaires’ terminaux et à l’occasion des signes de sécrétion muqueuse. Certains follicules gardent des traces de leur structure originelle, et quelques-uns récupèrent éventuellement leur capacité de differentiation normale; ceux-ci produisent alors simultanément un petit poil kératinisé et une glande à groissance vigoureuse. Les glandes métaplasiques ne ressemblent à aucune glande normale de la peau des Mammifères, mais ressemblent à des glandes salivaires immatures. Au moment où la métaplasie apparaît, il y a également une augmentation de la sécrétion de sebum, ou de matériel analogue à du sebum, dans le follicule pileux. Dans le groupe témoin non soumis à la vitamine A en excès, les follicules continuent la différenciation normale et la formation de poils jusqu’au terme de l’expérience et aucune métaplasie ne se manifeste.

  6. La plupart des follicules de vibrisses dans les explants de lèvre supérieure d’embryons de 15 jours continuent la différenciation normale en présence d’un excès de vitamine A (12,5 u.i./ml.), pourtant une proportion plus importante des poils présentent un structure anormale que dans le groupe témoin. Avec un excès de vitamine A quelques-uns des follicules les moins avancés au départ subissent la métaplasie glandulaire; aucune métaplasie n’apparaît dans le groupe témoin.

  7. Certaines des différences de comportement entre les explants provenant d’embryons d’âges différents pourraient être expliquées par l’hypothèse de la brièveté du stade critique du développement folliculaire pendant lequel la vitamine A peut induire la métaplasie.

The author is grateful to Mr Vincent Blancuzzi for careful and patient assistance in the preparation of thousands of serial sections and to Mrs Joline Spivack and others for technical help. Dr J. P. W. Gilman of the University of Guelph kindly read the manuscript. Professor Margaret R. Murray of Columbia University College of Physicians and Surgeons and Professor Dame Honor Fell, D.B.E., F.R.S., of the Strangeways Research Laboratory gave encouragement in many ways, and Dr Richard P. Bunge provided in his laboratory an excellent medium for the growth and differentiation of research scientists. This research was supported by U.S.P.H.S. N.I.H. Grant NB 04235 and National Multiple Sclerosis Society Grant 328 administered by Dr Richard P. Bunge.

Balinsky
,
B. I.
(
1950a
).
On the prenatal growth of the mammary gland rudiment in the mouse
.
J. Anat.
84
,
227
35
.
Balinsky
,
B. I.
(
1950b
).
On the developmental processes in mammary glands and other epidermal structures
.
Trans. R. Soc. Edinb.
62
,
1
31
.
Bloom
,
W.
&
Fawcett
,
D.
(
1962
).
A Textbook of Histology.
London and Philadelphia
:
W. B. Saunders
.
Borghese
,
E.
(
1950
).
The development in vitro of the submandibular and sublingual glands of Mus musculus.
J. Anat.
84
,
287
318
.
Braun-Falco
,
O.
(
1958
).
The histochemistry of the hair follicle
. In
The Biology of Hair Growth,
pp.
65
90
.
London and New York
:
Academic Press
.
Davidson
,
P.
&
Hardy
,
M. H.
(
1952
).
The development of mouse vibrissae in vivo and in vitro.
J. Anat.
86
,
342
56
.
Dawe
,
C. J.
(
1963
).
Skin-appendage tumors induced by polyoma virus in mice
.
Natn. Cancer Inst. Monogr.
10
,
459
88
.
Dingle
,
J. T.
&
Lucy
,
J. A.
(
1965
).
Vitamin A, carotenoids and cell function
.
Biol. Rev.
40
,
422
61
.
Fell
,
H. B.
(
1964
).
The experimental study of keratinization in organ culture
. In
The Epidermis,
pp.
61
81
.
New York
:
Academic Press
.
Fell
,
H. B.
&
Mellanby
,
E.
(
1953
).
Metaplasia produced in cultures of chick ectoderm by high vitamin A
.
J. Physiol., Lond.
119
,
470
88
.
Fell
,
H. B.
&
Rinaldini
,
L. M.
(
1965
).
The effects of vitamins A and C on cells and tissues in culture
. In
Cells and Tissues in Culture,
1
, pp.
659
99
.
London and New York
:
Academic Press
.
Hardy
,
M. H.
(
1949
).
The development of mouse hair in vitro with some observations on pigmentation
.
J. Anat.
83
,
364
84
.
Hardy
,
M. H.
(
1950
).
The development in vitro of the mammary glands of the mouse
.
J. Anat.84, 388–93
.
Hardy
,
M. H.
(
1951
).
The development of pelage hairs and vibrissae from skin in tissue culture
.
Ann. N.Y. Acad. Sci.
53
,
546
51
.
Hardy
,
M. H.
(
1965
).
Effect of excess vitamin A on rodent epidermis and developing hair folilcles in vitro.
Excerpta Medica, 19, Section 1, no. 10: XII
.
Hardy
,
M. H.
(
1967a
).
Responses in embryonic mouse skin to excess vitamin A in organotypic cultures from the trunk, upper lip and lower jaw
.
Exp. Cell. Res.
46
,
367
84
.
Hardy
,
M. H.
(
1967b
).
The differentiation of hair follicles and hairs in organ culture
.
Advances in Biology of Skin, 9 (in the Press)
.
Hardy
,
M. H.
(
1967C
).
Responses in oral epithelium and tissues of the tongue and lower jaw of the embryonic mouse to excess vitamin A in organotypic cultures (in preparation)
.
Jacoby
,
F.
(
1959
).
Observations on the post-natal development of the mouse submaxillary gland
.
J. Anat.
93
,
579
.
Jacoby
,
F.
&
Leeson
,
C. R.
(
1959
).
The post-natal development of the rat submaxillary gland
.
J. Anat.
93
,
201
16
.
Lasfargues
,
E. Y.
&
Murray
,
M. R.
(
1959
).
Hormonal influences on the differentiation and growth of embryonic mouse mammary glands in organ culture
.
Devi. Biol.
1
,
413
35
.
Lasnitzki
,
I.
(
1965
).
The action of hormones on cell and organ cultures
. In
Cells and Tissues in Culture,
1
, pp.
591
658
.
London and New York
:
Academic Press
.
Lyne
,
A. G.
,
Molyneux
,
G. S.
,
Myktowycz
,
R.
&
Parakkal
,
P. F.
(
1964
).
The development, structure and function of the submandibular cutaneous (chin) glands of the rabbit
.
Aust. J. Zool.
12
,
340
8
.
New
,
D. A. T.
(
1963
).
Effects of excess vitamin A on cultures of skin and buccal epithelium of the embryonic rat and mouse
.
Br. J. Derm.
75
,
320
5
.
Weissmann
,
G.
&
Fell
,
H. B.
(
1962
).
The effect of hydrocortisone on the response of fetal rat skin in culture to ultra-violet irradiation
.
J. exp. Med.
116
,
365
80
.

Plate 1

Fig. A. Section through explant of trunk skin from -day embryo after 11 days in 6·2 i.u./ ml. vitamin A medium (Exp. 2), showing pelage follicle developed normally to stage 5. d.p. = dermal papilla; o. = outer root sheath; He. = Henle’s layer; Hu. = Huxley’s layer; cu.= cuticle of inner root sheath; cu. = cuticle of developing hair; co. = cortex of developing hair (not yet keratinized). HEP, × 250.

Fig. B. Section through upper lip of 13|-day embryo (control to Exp. 2) showing vibrissal follicle at advanced stage 3 a. Note shallow dome-shaped dermal papilla (d.p.), and beginning of condensation of mesenchyme cells to form connective tissue sheath (c.t). HEP, × 250.

Fig. C. Section through upper lip of 14-day embryo (control to Exp. 3) showing vibrissal follicle at stage 3 b with dermal papilla (d.p.), hour-glass shape and connective tissue sheath (c.t.). HEP, × 100.

Fig. D. Vibrissal follicle in fig. C at the same magnification as fig. B to show the greater follicle length and deeper dermal papilla invagination. HEP, × 250.

Figs. A – C. Sections through explant of upper lip of -day embryo cultivated for 7 days in control medium (Exp. 2), showing normal vibrissal follicles at stage 8. Figs. B and C illustrate the middle and lower portions of the same follicle and fig. A the upper portion of a similar follicle with emerging hair tip (h.). HEP, × 250.

Figs. D– G. Sections through a single vibrissal follicle in explant of upper lip from the same embryo as figs. A– C but cultivated for 7 days in 12·5 i.u./ml. vitamin A medium (Exp. 2), showing gland-like structure arising from follicle. Note enlarged outer root sheath but otherwise normal differentiation of follicle layers to stage 5. l .= lumen. HEP, × 250.

Plate 3

Fig. A. Entire living unstained explant from upper lip of 14-day embryo after 7 days in control medium (Exp. 3) showing numerous vibrissal follicles at stages 6 – 8 attached to central mass of keratinizing epidermis, d.p. = Dermal papilla; c.t. = connective tissue sheath. × 40.

Fig. B. A follicle at stage 8 in the upper part of fig. A. d.p. = margin of dermal papilla; o. = outer root sheath; He. = Henle’s layer; co. = developing hair cortex (not keratinized at this level in follicle), × 100.

Fig. C. Entire living unstained explant from upper lip of same embryo after 7 days in 12·5 i.u./ ml. vitamin A medium, showing loss of normal structure and budding from follicles, × 40.

Fig. D. A follicle at lower left of fig. C which consisted only of a budding mass of cuboidal epithelial cells, × 100.

Fig. E. Section through entire explant shown in fig. A after 10 days in control medium, showing central epidermis and well-developed vibrissal follicles cut at various angles. HAB, × 40.

Fig. F. Enlargement of part of field in fig. E to show follicle structure, d.p. = Dermal papilla; c.t. = connective tissue sheath; h. = keratinized hair. HAB, × 100.

Fig. G. Section through entire explant shown in fig. C after 10 days in 12·5 i.u./ml. vitamin A medium, showing central epidermis surrounded by numerous glands and one hair follicle. HAB, × 40.

Fig. H. Enlargement of part of field in fig. G, showing hair follicle at stage 5 (f.) and branching tubular glands (g.). HAB, × 100.

Plate 4

Fig. A. Section through explant of upper lip from 14-day embryo after 14 days in 12·5 i.u./ ml. vitamin A (Exp. 3), showing large healthy vibrissal follicle (stage 8) cut obliquely and containing a minute keratinized hair (A.). Attached to follicle is a branching tubular gland with cystic enlargement of some ducts. HAB, × 67.

Fig. B. Section through another part of explant shown in fig. A with parts of two branching tubular glands. Gland on left has a duct opening directly to epidermis. Gland on right has, attached to its duct, an epidermal mass (f.) representing the original hair follicle. HAB, × 67.

Fig. C. Section through same explant as in figs. A and B, showing two glandular trees. No remnants of the original follicle were found in association with the gland on left. HAB, × 67.

Fig. D. Section through another explant of upper lip from 14-day embryo after 10 days in 12·5 i.u./ml. vitamin A medium (Exp. 3) showing four former hair follicles whose complete transformation into glands was observed in the living state. At left, left centre and right centre are three branching tubular glands derived from three follicles, and at the upper right is the sebaceous gland portion of a glandular structure derived from the fourth follicle. HEP, × 67.

Fig. E. Section through explant of upper lip from 14-day embryo after 10 days in 12·5 i.u./ml. vitamin A medium (Exp. 3) showing differentiated sebaceous cells (s.c.) at left, and sebaceous cells breaking down to produce sebum (se.) at right. HEP, × 250.

Fig. F. Another part of explant shown in figs. A – C showing transverse sections of a large cyst and several small ducts (d.) and terminal portions of branching tubular glands. HEP, × 250.

Fig. G. Part of section seen in fig. D enlarged to show oblique section of small duct (on left) and two terminal swellings with small lumen (l.) surrounded by cuboidal (l’alveolar’) cells. HEP, × 250.

Fig. H. Another part of explant shown in figs. A-C and F with small gland ducts (d.) and ‘alveoli’, the latter with a very small lumen (l.) of an intercellular canal. HAB, × 250.

Fig. I. Duct of tubular gland in same explant showing mucopolysaccharide reaction. The dark shade of the cytoplasm of marked cell (c.) and of strands crossing the duct from this cell are due entirely to Alcian blue staining. HAB, × 1000.

Plate 1

M. H. HARDY

Plate 1

M. H. HARDY

Plate 2

M. H. HARDY

Plate 2

M. H. HARDY

Plate 3

M. H. HARDY

Plate 3

M. H. HARDY

Plate 4

M. H. HARDY

Plate 4

M. H. HARDY