The dependence of duodenal differentiation in chick embryos on pars distalis hormones

Several investigators, especially Moog and her co-workers in a series of elegant studies, have described the major biochemical and morphological changes which occur in the development of the duodenum of the chick embryo. Apparently these changes are necessary for the functional differentiation of the duodenum and include: (1) increases in duodenal length and diameter, (2) delimitation of the villi from the pre-villous ridges and subsequent elongation, and (3) a transient accumulation of erythrocytes in the villous stroma. The epithelial cells clothing the villi change in size, shape, cytoplasmic density and basophilia. Their nuclei undergo changes in location and morphology. The distribution of mitotic figures and goblet cells as well as the appearance, distribution and mobilization of cytoplasmic glycogen also change in characteristic ways. In addition, the appearance and maturation of the microvilli of the brush border of the mucosa cells parallel the appearance and increase in the amount of mucopolysaccharide and activity of alkaline phosphatase. These changes occur between the 15th and 20th (last) day of incubation, presumably in preparation for the ingestion of food after hatching (Hilton, 1902; Argeseanu & May, 1938; Moog, 1950, 1961; Moog & Wenger, 1952; Moog & Richardson, 1955; Kato, 1959; Romanoff, 1960; Hinni & Watterson, 1963; Stocum, 1966; Penttilâ & Gripen-berg, 1969).

Corticoids cause normal differentiation of the duodenum to occur precociously in vivo and in vitro (Moog & Richardson, 1955; Moog & Thomas, 1957; Moog, 1958, 1959a, b, 1962; Hayes, 1965a, b; Overton, 1965) and ACTH has the same effect in vivo (Moog & Ford, 1957). Thyroxine is apparently permissive in conditioning the sensitivity of the duodenum to adrenocorticoids (Moog, 1961) and encourages duodenal growth and villous morphogenesis (Stocum, 1966). Accordingly, ‘hypophysectomy’ (by partial decapitation) arrests duodenal epithelial cell differentiation at a 16- to 17-day-old normal level (Hinni & Watterson, 1963) but only retards the rate of duodenal growth and villous morphogenesis. Also the activity of the major electrophoretic component of duodenal alkaline phosphatase in partially decapitated chick embryos is reduced by 90% (Manwell & Betz, 1966). The subnormal duodenal development in partially decapitated embryos may be due to the absence of factors other than adeno-hypophyseal hormones (Hinni & Watterson, 1963). However, in other respects, the effects of partial decapitation have been shown to be the same as if the embryos had been hypophysectomized in the classical manner (Betz, 1967, 1968, 1970). Hinni & Watterson (1963) reported that cortisone and thyroxine alone or together apparently do not alleviate the arrested duodenal differentiation and suggested since STH (somatotrophic hormone) is essential to maintain the normal turnover rate of the duodenal mucosa of rats (Leblond & Carriere, 1955), it may also be necessary for normal duodenal differentiation in the chick embryo. Hinni & Watterson (1963) review the important literature concerning normal duodenal development, its hormonal modification and discussions of the theoretical functional significance of these changes.

The gross duodenal morphology appeared normal in ‘hypophysectomized’ chick embryos with a chorioallantoic graft of embryonic pars distalis gland (Betz, 1967) but the details of the correction were not analysed. The results of this study demonstrate that the pars distalis is essential for normal duodenal growth and differentiation in chick embryos.

Fertile eggs (Hyline strain 934 F) were selected at random and incubated without turning at 38 ± 0·5 °C and 60 ± 5 % relative humidity. There were four groups in the experimental design: (1) normal embryos, (2) ‘hypophysectomized’ embryos, (3) ‘hypophysectomized’ embryos which received a pars distalis homograft, and (4) ‘hypophysectomized’ embryos with grafts which apparently had not been vascularized and had atrophied. The embryos in group 1 received no further treatment. Those for groups 2 and 3 were handled and ‘hypophysectomized’ by surgical partial decapitation at stage 11 as described previously (Betz, 1967) except that in group 3 the hosts were 9·5 days old when they received chorioallantoic grafts of partes distales from donor embryos which ranged in age from 14 to 21 days. The experimental design, number of specimens in each group and age of the embryos at sacrifice are summarized in Table 1.

At sacrifice the grafts were fixed in 3% glutaraldehyde in 0·1 M phosphate buffer, pH 7 0 for 12 h at 0 °C. To ensure that the condition of the embryos in group 3 was due to hormones from the grafts and not from remnants of the host’s pars distalis, the heads were examined as before (Betz, 1965). The extra-embryonic membranes and yolk sacs were removed and the embryos rinsed in cold water and blotted. All heads were removed (Betz, 1967, 1968) to make the bodies comparable (parts of the head are missing in operated embryos). The headless bodies were weighed to the nearest 0·1 g and the length of the third right toe (the two distal segments of the phalanyx including the claw tip) was estimated to the nearest 0-5 mm. The duodenum was excised; the pancreas and adherent fat removed and the apex of the loop was cut into several pieces approximately 5 mm in length. Some pieces were fixed in glutaraldehyde as above. Representative gut samples from 18·75-day;-old normal embryos were fixed in Bouin’s fluid for 24 h at 22 °C. Samples from all embryos were fixed in 80% ethanol for 12 h at 4 °C for the localization of alkaline phosphatase activity (Gomori, 1941). The tissues fixed in glutaraldehyde were dehydrated and embedded in glycol methacrylate according to McCully (1966) and Feder & O’Brien (1968). Thin sections, 0·5–2μ thick, were cut with glass knives on a Porter-Blum, MT-1 Ultramicrotome and affixed to glass slides as described by Feder & O’Brien (1968). The tissues fixed in Bouin’s fluid and in 80 % ethanol were processed for paraffin techniques and sectioned on a rotary microtome set at 5 μ. The unfixed proximal and distal limbs of the duodena were immediately frozen at − 15 °C, dehydrated in vacuo at room temperature for 12 h and stored at − 15 °C no longer than 3 weeks. The storage time was kept constant within and between experimental groups. The dried duodena were weighed to the nearest 0·1 mg and homogenized in a glass system in glass-distilled water at 4 °C. Alkaline phosphatase activity of the homogenates adjusted to a concentration of 0·1 % was estimated by the method of Manning, Steinetz, Babson & Butler (1966). Histological and histochemical observations were made after treating the sections with one of the following procedures or stains. (1) Acid fuchsin and counterstained with toluidine blue as a general stain as’described by McCully (1966). (2) Periodic acid-Schiff’s reagent (PAS) for localization of glycogen and mucopolysaccharides. Sections were placed in 1 % chlorous acid 20 min (Rappay & Van Duijn, 1965); washed in tap water, 5 min; oxidized with 1 % periodic acid, 10 min; placed in Schiff’s reagent, 30 min; bleached in three sulfite washes, 10 sec each; counterstained with 1 % fast green, pH 2·0, 2 min; washed in tap water, dehydrated, cleared and mounted. Control sections were pretreated for 30 min to 2 h with salivary amylase to remove glycogen. (3) Gomori (1941) technique for localization of alkaline phosphatase activity. Pieces of duodena fixed in 80 % ethanol at 4 °C were dehydrated in two changes of absolute ethanol at 4 °C for 12 h each; cleared at room temperature in benzene three changes, 1 h each; infiltrated and embedded with histowax (m.p. 52 °C). Sections (5 μ) were mounted on slides, dried at 40 °C for 10 min, deparaffinized, cleared, hydrated and incubated 20 min in magnesium-free medium at 37 °C. (Hinni & Watterson, 1963). The sections were washed in cold tap water, treated with 2 % cobalt nitrate, 5 min, washed in tap water; treated with dilute ammonium sulfate (Gomori, 1952); washed in tap water; mounted in permount and photographed 2 days later.

Representative sections were photographed. The lengths of the villi and the heights of fourteen epithelial cells only at the tips of non-dilated villi were estimated with an ocular micrometer from thin sections stained with acid fuchsin-toluidine blue. The means, standard errors of the means of body weights, third toe lengths, dried duodenal weights, units of phosphatase activity and heights of the epithelial cells were calculated. References to similarities and differences within and between experimental groups in the text are based on the results of Student’s t test, which was used to estimate significant differences between the means, t values were chosen at P = 0 05.

The gross morphology of the embryos in group 2 was defective but appeared normal in group 3 except for the lack of the derivatives of the head parts resected in the operation as described before (Betz, 1967, 1968). The embryos in group 4 were indistinguishable from untreated, partially decapitated embryos (group 2). In all parameters (Tables 1–5) development is arrested at the 16th or 17th day of incubation by ‘hypophysectomy’ and is restored to normal by one pars distalis graft. In group 4 the developmental pattern is not different from that in untreated ‘hypophysectomized’ embryos, indicating the effect exerted by the grafts occurs only if the cells are viable. The histology of the duodena from group 1 incubated 16·75, 17·75 and 19·75 to 21·5 days; group 2 incubated 19·75 to 21·5 days; groups 3 and 4 incubated 19·75 to 20·5 days are illustrated by Figs. 1-27. The cytology of a pars distalis graft is shown in Fig. 8.

In addition to the tabulated observations at 19·75 days the villi were apparently more numerous than at 18·75 days and the stage of differentiation was more variable than before or later. At 17·75 days of incubation amylase treatment of thin sections was required for 2 h to remove the PAS-positive granules from the normal mucosa cells but they were removed from thick, ethanol-fixed sections after only 30 min. At 18·75 days of incubation perinuclear vacuolations were common in the normal mucosa cells at the villous tips. There were low-intensity Gomori reactions which did not change from 16·75 to 19·75 days of development in parts of normal duodena other than the free border. At 20-5 days there were PAS-positive, amylase-fast vesicles in the brush border of many normal mucosa cells.

In ‘hypophysectomized’ embryos at 19·75 and 20·5 days the stage of development did not differ from 16·75- and 17·75-day normal embryos except at 19·75 days there were perinuclear vacuolations and the frequency of villi distended with red blood cells seemed greater. At 20·5 days the brush border was faintly PAS-positive in ethanol-fixed sections (5 μ) but not in thin sections. By 21·5 days the level of development was intermediate between the 15·75-and 16·75-day normal levels. The cells in the established pars distalis grafts were indistinguishable from normal (Fig. 8). Atrophic grafts were not sectioned.

This investigation determined that normal duodenal development occurred only in the presence of pars distalis tissue, whether in situ or ectopically as a graft. Most embryos were examined at 19·75 and 20·5 days of incubation (normally most landmarks of duodenal differentiation have occurred by this time) because the differences between the normal and ‘hypophysectomized’ embryos are so great correction would be obvious in group 3. In groups 2, 3 and 4 only a few embryos were examined after 19-75 days due to the characteristic high mortality in groups 2 (Betz, 1967, 1968) and 4. Also in group 3 the chorioallantoic circulation shuts off (day 20; before hatching), killing the grafts. Therefore the experiment is terminated because the source of hormones is lost. The normal embryos hatched at 20·5 days of incubation. Thus 21·5 and 22·5 days of ‘incubation’ are the first and second post-hatching days in group 1. However, in group 2 the embryos remain in the eggs (Betz, 1967). We used 14- to 21-day- old donors for pars distalis grafts as a source of hormones and did not detect differences in duodenal development within groups in group 3. This may be commensurate with the fact that the electrophoretic species of soluble proteins in the pars distalis at hatching are almost the same as in the adult gland (Manwell & Betz, 1966). In addition partes distales, from even younger, 10-day-old embryos correct or alleviate the defective development of partially decapitated embryos (Betz, 1965, 1967, 1968), indicating secretory differentiation of the pars distalis has occurred by that time or is synchronous with host development. The pars distalis grafts effected normal duodenal differentiation in partially decapitated embryos in the absence of the hypothalamus, epiphysis, pars nervosa and the brain parts of prosencephalic and anterior mesencephalic derivation. This demonstrates that these organs do not have important influences on duodenal development.

The villi usually bend during growth; therefore the villous lengths were approximated from the unbent villi. The extent to which fixation, embedding and other procedural factors may affect the measurement of villus lengths was not determined. Body weights and toe lengths were recorded as indices of body growth which partially depends on pars distalis hormones (Betz, 1968, 1970). In our results the absolute reduction and increase in body weight in groups 2 and 3 differ from the results of Betz (1967, 1968) for unknown reasons but the difference between groups is almost equal. Our results indicate pars distalis grafts in ‘hypophysectomized’ embryos can, but do not always, restore body growth to normal. However, the between-group changes in toe length are in accord with the results of Betz (1968). The slight subnormality of the dry duodenal weights in group 3 may be due to reduced albumen ingestion by the operated embryos (Betz, 1968). The duodenal contents were not removed prior to determining dry weights. However, the data of Coulombre & Coulombre (1958), Hinni & Watterson (1963) and Stocum (1966) on the rate of duodenal growth (increases in length and diameter) agree with our data. We did not measure duodenal lengths and diameters because the peristaltic contractions which occur before and during fixation change the length and diameter of the intestine.

Von Pap (1933) proposed the erythrocyte aggregations in the lamina propria of the villi at 16-75 and 17-75 days of incubation arise de novo by erythropoiesis. He and Moog (1961) feel that they dilate the pre-villous ridges. Initially the cells at the tips of the pre-villous ridges are low columnar but become cuboidal or squamous after the red blood cell (RBC) aggregations develop. It is tempting to ascribe the changes in shape of these cells to an increase in the mechanical force supplied by the erythrocytes in accord with Moog (1961). The RBC aggregations normally disappear and the villi become vascularized on the 17th day of incubation (Von Pap, 1933). Thus, the increased frequency of villi with RBC aggregations in the duodena of the ‘hypophysectomized’ embryos may indicate the vascular development of the villi is subnormal as occurs in other organs of ‘hypophysectomized’ embryos (Betz, 1967) or perhaps the period of erythropoiesis is prolonged in these embryos.

The significance of the perinuclear vacuolations in the mucosal cells at 18·75 days is not apparent. However, they seem to be part of normal development (Hinni & Watterson, 1963). In ‘hypophysectomized’ embryos at 19·75 and 20·5 days the epithelial cells have perinuclear vacuolations (a characteristic of 18·75-day normal cells) but in other respects they are like 16·75-day normal epithelial cells, implying the cells are not as severely retarded. However, the phosphatase activity does not exceed the 15·75-day level. The development of the cytological and biochemical changes are not synchronous. This probably indicates that they are not developmental prerequisites for each other nor functionally interdependent but that individually they are affected differently by the endocrine environment; this is in agreement with the results of others (Moog & Nehari, 1954; Kato, 1959; Hayes, 1965 b). The PAS-positive amylasefast vesicles at the brush border on 20·5 days of incubation may be the light-microscopic equivalent of the vésiculations of microvilli which appear in vitro if cortisone is added to the medium (Hayes, 1965 b). If so they may be adrenocorticoid dependent in vivo. The heights of the mucosal cells reported by Hinni & Watterson (1963) are lower than ours, perhaps due to differences in fixation. In other respects, our results are in accord with those of others (Moog & Richardson, 1955; Moog, 1961; Hinni & Watterson, 1963) except for the location of glycogen in the mucosal cells. It is impossible to correlate our results with their reports since they did not observe the precautions of Pearse (1968) about glycogen fixation and we do not know how glutaraldehyde fixation affects glycogen. Crude salivary amylase removed the PAS-positive granules from thick but not thin sections unless incubation was extended to 2 h, perhaps contrary to expectation since there is less material in thin sections. Glutaraldehyde fixation may retard enzymic hydrolysis or hold glycogen more strongly than other fixatives especially since it causes proteins to cross-link (Swigart, Wagner & Atkinson, 1960). Also glycol methacrylate perhaps retards glycogen digestion. Further studies may clarify this problem. Presumably glycogen is an energy source for the synthesis of structural and enzymic proteins and is apparently mobilized by 20 days of development, perhaps because demand exceeds deposition (Moog, 1961; Hinni & Watterson, 1963).

In our slides the first consistent appearance of a PAS-positive, amylase-fast material is at 19·75 days of incubation. Perhaps the difference between our results and those above is due to the small amount of polysaccharide which reacts with the PAS reagent in thin sections. The PAS-positive, amylase-fast material at the border could not be detected on 15·75 and 18·75 days, perhaps because the border contains another amylase-labile carbohydrate. In thick sections there was enough mucopolysaccharide after the labile component was removed to give a visible PAS reaction, but not in thin sections. Perhaps there is less of this material on days 15·75 and 18·75. Moog & Wenger (1952) suggest that the non-acidic mucopolysaccharide provides a cyto-skeletal framework which orients the phosphatase molecules, thus providing a suitable environment for activity and a finite number of enzyme accepting sites which limits enzyme accumulation. A mucopolysaccharide is an integral part of the surface of the microvilli (Ito, 1969; Millington, Critchley, Tovell & Pearson, 1969) and in addition phosphatase molecules are apparently glyco-proteins (Portmann, Rossier & Chardonnens, 1960).

We did not determine specific activities of alkaline phosphatase. Nevertheless our results are in accord with those of Moog & Richardson, (1955) and Moog (1961), who did determine specific activities. Activity expressed as units per mg total duodenal protein is apparently not a more valid assessment of true activity than units per mg dry duodenal weight, probably because the phosphatase is mostly confined to the brush border (Overton, Eichholz & Crane, 1965; Eichholz, 1969) which is a small fraction of the total dry weight or protein. The duodenum does not contain adipose tissue which would contribute to the weight, and the albumen ingested by chick embryos (Betz, 1968) would contribute to the protein content and dry weight. Our pilot experiments demonstrated that albumen has no significant alkaline phosphatase activity in accord with the results of Manning et al. (1966).

We deleted magnesium ions from the medium in the Gomori method (1941, 1952) and incubated for 20 min at 37 °C in accordance with Hinni & Watterson (1963), who reported the free border to be Gomori-negative prior to 17 days of incubation. Moog’s results (1961) concur even though she did not study embryos younger than 17 days. Also she added magnesium to the medium and varied the incubation time. With magnesium in the medium (Kabat & Furth, 1941) phosphatase activity is demonstrable at the free border as early as 14 days (Hancox & Hyslop, 1953). In our study duodena with less than 30 units of activity would not yield enough Gomori-positive precipitate to be seen, which indicates the resolving power of this modification of the Gomori technique with thin sections is not very good. In spite of the controversy (Pearse, 1968) concerning the validity of the Gomori technique for localizing alkaline phosphatase activity, apparently this enzyme is localized in the outer membrane of the microvilli (Clark, 1961; Holt & Miller, 1962; Goldfischer, Essner & Novikoff, 1964; Overton & Shoup, 1964; Eichholz & Crane, 1965; Overton et al. 1965; Hugon & Borgers, 1966; Mayahara, Hirano, Saito & Ogawa, 1967 ; Toner, 1968 ; Eichholz, 1969). The problem remains to show that the same localization occurs in vivo.

It has long been accepted that the microvilli are involved with absorption because they increase the surface area of the cells (Toner, 1968). Clark (1961), Overton & Shoup (1963, 1964) and Penttilä & Gripenberg (1969) showed that microvillous maturation parallels alkaline phosphatase accumulation in the brush border. The problem of correlating increases in enzymic activity and in absorptive capacity has not been resolved although Hudson & Levin (1968) have described the changes in electrical potential across the intestine of the chick embryo during development in correlation with the transport of hexoses and amino acids. Also Bogner (1961) demonstrated the absorption coefficient for several sugars increases significantly in the chick intestine within three days after hatching. The role played by alkaline phosphatase has not been elaborated, although Tosteson, Blaustein & Moulton (1961) have suggested that it may act as a ‘sodium pump’ coupled with the active uptake of sugars. However, the physiological significance of the structural and biochemical changes which occur during duodenal development remains obscure (Crane, 1965; Ugolev, 1965; Toner, 1968; Korn, 1969).

Moog (1962) and Toner (1968) consider the progressive accumulation of enzymes during development as evidence of adaptation to absorptive function. Moog (1962) and Penttilä & Gripenberg (1969) propose that the small intestine passes suddenly from a prolonged non-functional state into active function and point to the suggestive evidence of the upsurge in alkaline phosphatase activity which always occurs shortly before function begins. Presumably in the chick this means the duodenum begins to function at or shortly after the time of hatching. The age at which the duodenum of the chick becomes functional is not known but between 11 and 15 days of incubation albumen is ingested, digested and utilized (Witschi, 1949; Carinci & Manzoli-Guidotti, 1968; Hinsch, 1967; Betz, 1970) and if chick embryos do not ingest albumen they are dwarfed (Betz, 1968). The proventricular glands begin to secrete acid on day 13 and show histochemical signs of heightened secretory activity on days 14 and 15 (Toner, 1965; Hinsch, 1967). This in addition to the results of Hudson & Levin (1968) indicates the intestine functions, at least to some extent, before the changes which have been associated with ‘functional’ differentiation of the duodenum by Moog and her co-workers Hinni & Watterson (1963) and Baxter-Grillo (1969a, b) have progressed very far. Thus, the developmental changes in the duodenum are probably not prerequisites for absorption, digestion or transport but they may be necessary to support the increased rates of function after hatching. It seems unlikely but the ingested albumen may be stored in the gut until functional onset or the digestion products may be absorbed in a more posterior part of the small intestine.

This study demonstrates the indispensability of pars distalis hormones in normal duodenal differentiation in the chick embryo during the last 4 days of development. The developmental rate of the duodenum in ‘hypophysectomized’ embryos apparently does not simply slow down because differentiation is arrested at the 16·75- or 17·75-day level, even if incubation is prolonged for 2 or as many as 6 days beyond the time of hatching (Hinni & Watterson, 1963). One or more pars distalis hormones probably affect duodenal differentiation directly (perhaps STH, prolactin and/or ACTH) or ACTH and TSH may stimulate the synthesis of adrenocorticoids and thyroxine (Betz, 1970). Betz (1965, 1967, 1968) demonstrated that pars distalis grafts improve the subnormal growth of ‘hypophysectomized ‘embryos, perhaps indicating that the embryonic pars distalis secretes STH or prolactin or both. Also Bates, Miller & Garrison (1962) have demonstrated that bovine growth hormone and ovine prolactin have a splanchnotropic effect in hypophysectomized pigeons.

The possibility that STH is essential for duodenal differentiation in the chick embryo is consistent with the results of Leblond & Carriere (1955), and Turner (1966) points out that STH and thyroxine are complementary in many species and may be essential for the full expression of other hormones. Also a growthpromoting activity (bio-assayed in Rana temporaria larvae) which appears on the 15th day of incubation is confined to the caudal region of the pars distalis in the chick embryo (Enemar, 1967). Presumably, STH rather than prolactin is responsible for this growth-promoting activity. See review of Betz (1970). However, the time at which STH may be released in vivo in amounts which may stimulate duodenal development is not known. It is tempting to suggest the upsurge in the rate of duodenal development at 18 days of development coincides with the release of STH and perhaps other hormones. Other interpretations are possible.

Testosterone (Carriere, 1966) and estrogen (Bullough, 1946) can stimulate mucosal mitogenesis. The effects of sex steroids on duodenal differentiation have not been reported but may not be inconsiderable since the pars distalis— gonadal axis probably becomes established at least by 13·5 days of incubation (Woods & Weeks, 1969). However, there was no apparent sexual dimorphism in duodenal development in normal embryos. A more precise analysis of the hormones involved in duodenal differentiation in chick embryos is in progress.

La différenciation duodénale chez les embryons de poulet normaux entre 15,75 et 20,5 jours d’incubation est caractérisée par les changements suivants:

1. Le poids sec augmente de 4,1 à 12,4 mg.

2. L’activité de la phosphatase alcaline croît de moins de 12 à 426 unités.

3. La longueur des villosités augmente de six fois.

4. La hauteur de cellules épithéliales aux extrémités des villosités augmente de 12,9 à 25,9

5. Les cellules muqueuses sont d’abord cylindriques et courtes, puis cubiques, puis cylindriques et élevées.

6. Les noyaux, d’abord sphériques, deviennent ovoïdes.

7. Les figures mitotiques, initialement distribuées sur tout l’épithélium, s’observent par la suite uniquement sur les cryptes de Lieberkühn.

8. Le glycogène cytoplasmique, apparaît après 15,75 jours et est mobilisé après 20,5 jours.

9. Un mucopolysaccharide apparaît à la bordure en brosse des cellules muqueuses après 16,75 jours et augmente progressivement.

10. L’activité de la phosphatase alcaline, mise en évidence par la technique de Gomori, appraît faiblement au niveau de la bordure en brosse, après 16,75 jours et devient ensuite plus intense.

11. Le poids frais et la longueur du troisième orteil après 19,75 et 20,5 jours d’incubation ont été pris comme indices de la croissance.

    Dans les embryons ‘hypophysectomisés’, après 19,75 et 21,5 jours:

12. Le niveau de la différenciation duodénale est le même que celui d’embryons normaux âgés de 16,75 à 17,75 jours.

13. Les embryons ‘hypophysectomisés’, chez qui un greffon de la pars distalis chorioal-lantoïdienne a été implanté, présentent un duodénum normal.

14. Chez les embryons ‘hypophysectomisés’ où les greffons s’étaient atrophiés, le niveau de la différenciation duodénale est similaire à celui des embryons ‘hypophysectomisés,’ non traités.

15. Chez les embryons de poulet, la différenciation duodénale dépend des hormones de la ‘pars distalis’.

This work was supported by a grant to T. W. Betz from the National Research Council of Canada.

These photomicrographs are of thin cross-sections of the duodena of chick embryos at various stages of development, except that Fig. 8 is a photomicrograph of a section of a pars distalis homograft in the chorioallantois of a partially decapitated chick embryo. All were stained with acid fuschin-toluidine blue except Figs. 9–13 which were treated with PAS and counterstained with fast green. All are magnified×380 except Figs. 8, 9, 13–14 which are×1000,×240 and×100 respectively.

Fig. 1. 19·75-day-old normal embryo. The epithelial cells are columnar; the brush border is well defined and the stroma is dense. 220 units of phosphatase activity.

Fig. 2. 19·75-day-old normal embryo. This section represents the least well developed duodenum examined at this stage of incubation. Note the apical vacuolation of the cytoplasm (arrow). 150 units of phosphatase activity.

Fig. 3. 19·75-day-old ‘hypophysectomized’ embryo. The cells at the villous tips are low columnar and the stroma is loose. Perinuclear vacuolations (arrow) are common. 18 units of phosphatase activity. The level of differentiation does not exceed day 16·75 normal (Fig. 7).

Fig. 4. 19·75-day ‘hypophysectomized’ embryo with a pars distalis graft. 220 units of phosphatase activity.

Fig. 5. 19·75-day ‘hypophysectomized’ embryo with a pars distalis graft. The epithelial cells are columnar and the stroma is compact. 245 units of phosphatase activity.

Fig. 6. 19·75-day ‘hypophysectomized’ embryo with an atrophic pars distalis graft. Compare with the untreated’hypophysectomized’embryo (Fig. 3). 24 units of phosphatase activity.

Fig. 7.16·75-day-old normal embryo. The epithelial cells (arrow) are cuboidal and the stroma contains well developed erythrocyte aggregations. The cuboidal cells average 11·2μ in height. 21 units of phosphatase activity.

Fig. 8. Pars distalis homograft in the chorioallantois of a 20·5-day ‘hypophysectomized’ chick embryo. The cytology of the cells appears normal. Note the granular cytoplasm (arrow).

Fig. 9. 19·75-day normal embryo. Goblet cells are near the tips of the villi (arrow) and the brush border is intensely PAS-positive. 220 units of phosphatase activity. Fig. 10. 19·75-day ‘hypophysectomized’ embryo. Note the erythrocyte aggregations in the stroma and the PAS-positive granules in the cytoplasm of the epithelial cells (arrow). 20 units of phosphatase activity.

Fig. 11. 19·75-day ‘hypophysectomized’ embryo with a pars distalis graft. The stage of differentiation is not different from normal. 210 units of phosphatase activity.

Fig. 12. 16·75-day normal embryo. Note the localization of aggregates of PAS-positive granules especially in the apical cytoplasm of the cell at the tips of the villi (arrow). The brush border is faintly PAS-positive. 19 units of phosphatase activity.

Fig. 13. 19·75-day ‘hypophysectomized’ embryo. Note the distribution of PAS-positive granules in the cytoplasm of the epithelial cells. Note the frequency and size of erythrocyte aggregations (arrow). 20 units of phosphatase activity.

Fig. 14. 19·75-day ‘hypophysectomized’ embryo with a pars distalis graft. Note the lack of large erythrocyte aggregations (arrow) and length of villi. 175 units of phosphatase activity.

These photomicrographs are of thin cross-sections of the duodena of chick embryos in various stages of development. All were treated with PAS and counterstained with fast green except Figs. 18 and 26 which were stained with acid fuchsin-toluidine blue and Figs. 21 and 22 which were treated with the Gomori technique. The magnification is×380 except Figs. 23 and 24 which are×1000.

Fig. 15. 20·5-day normal embryo. Note the PAS-positive brush border, the absence of cytoplasmic granules and the apical goblet cell (arrow). 350 units of phosphatase activity.

Fig. 16. 20·5-day ‘hypophysectomized’ embryo. The pattern of distribution of PAS-positive granules in the cytoplasm of epithelial cells at the tips of the villi (arrow) is similar to 16·75-day normals, but the brush border is PAS-negative. 21 units of phosphatase activity.

Fig. 17. 20·5-day ‘hypophysectomized’ embryo with an atrophic graft. Note similarity to the 17·75-day normal (Fig. 19) and the 20·5-day hypophysectomized embryo (Fig. 16). 20 units of phosphatase activity.

Fig. 18. 20·5-day ‘hypophysectomized’ embryo with a pars distalis graft. The stage of cytological differentiation is more advanced than in 19·75-day normals. 400 units of phosphatase activity.

Fig. 19.17·75-daynormal embryo. Compare theamountand location of PAS-positive granules (arrow) with Fig. 16. 26 units of phosphatase activity.

Fig. 20. 17·75-day normal embryo. The section was treated with salivary amylase for 2 h prior to PAS treatment. Note reduction of PAS-positive cytoplasmic granules (arrow) compared to Fig. 19. Note the absence of a PAS-positive brush border. 28 units of phosphatase activity.

Fig. 21. 20·5-day normal embryo. Note the heavy deposition of Gomori precipitate showing alkaline phosphatase activity localization at the brush border of the epithelial cells (arrow) and its relative absence in other parts. Green filter in the light path. 395 units of phosphatase activity.

Fig. 22.20·5-day’hypophysectomized’embryo with a graft. Note similarity to Fig. 21. (Compare area at arrow.) 420 units of phosphatase activity.

Fig. 23. 20·5-day normal embryo. Note the PAS-positive brush border and the absence of cytoplasmic granules. Note also the goblet cell (arrow). 350 units of phosphatase activity.

Fig. 24. 20·5-day ‘hypophysectomized’ embryo with a graft. The stage of differentiation is normal (cf. Fig. 23). 400 units of phosphatase activity.

Fig. 25. 21·5-day (hatched) normal chick. Note the number of goblet cells and the intense PAS-positive brush border (arrow).

Fig. 26. 21·5-day ‘hypophysectomized’ embryo. Note the PAS-positive granules in the cytoplasm of the cuboidal epithelial cells on the tips of the villi (arrow). 25 units of phosphatase activity.

Fig. 27. 21·5-day ‘hypophysectomized’ embryo. The epithelial cells arecuboidal and the stroma is filled with erythrocytes. These villi are less well developed than those of the 17·75-day-old embryo (Fig. 19). 22 units of phosphatase activity.