- Email: [email protected]
Effects of Inducers on Inner and Outer Gastrula Ectoderm Layers of Xenopus laevis
Differentiat ion
Differentiation (1983) 23: 206-212
Springer-Verlag 1983
Effects of Inducers on Inner and Outer Gastrula Ectoderm Layers of Xenopus laevis Makoto Asashima' and Horst Grunz' Department of Biology, Yokohama City University, Yokohama 236, Japan Fachbereich 9 (Biologie), Zoophysiologie, Universitit Essen - GHS, Universitatsstr. 5, 4300 Essen 1, Federal Republic of Germany
Abstract. Gastrula ectoderm, isolated from Xenopus laeuis, was cultured in Holtfreter solution or modified Leibovitz medium (L-15) by the sandwich-method with or without inducer. The ectoderm (SD cell layers) consists of two cell sheets, representing a superficial (S) and a deep (D) layer. In the L-15 medium rather than in Holtfreter solution, the two cell layers separate out into distinct cell masses. This difference in cell affinity under certain experimental conditions could indicate that the deep layer contains endodermal cells. However, an endodermal character of the deep layer can be ruled out by induction experiments with vegetalizing factor or dorsal blastopore lip as inducers. Under the influence of vegetalizing factor the outer as well as the inner ectoderm layer differentiated into mesodermal derivatives such as notochord and somites. The results of the experiments with dorsal blastopore lip as inducer indicate that both inner and outer ectoderm layers are responsive to the neural stimulus. The lower neural competence of the outer ectoderm layer observed by several authors in normogenesis is discussed with regard to the hypothesis about short distance diffusion of the neuralizing factor and/or close cellto-cell contact between inducing tissue and ectodermal target cells.
Introduction It is well known that the gastrula ectoderm of amphibian embryos displays high responsiveness to inducing stimuli. For this reason competent ectoderm has been used as a test system in the implantation or sandwich technique by many investigators [9, 141. However, ectoderm of different amphibian species shows different reaction to inducing stimuli. Using Xenopus and Ambystoma embryos, Faulhaber [4] reported about differences of the reacting ectoderm to the induction of the dorsal blastopore lip when tested by xenoplastic transplantations. The blastopore lip of Xenopus induced neural tissues in Ambystoma host ectoderm more frequently than did Ambystoma dorsal lip in Xenopus host ectoderm. Moreover, there also are clear topographical differences in gastrula ectoderm between urodeles and anurans. While the ectoderm of the gastrula in anurans such as Xenopus and Rana consists of two distinct cell layers, in urodeles such as Triturus and Ambystoma ectoderm consists of a homogeneous cell layer (12, 13, 22, 2 To whom correspondence should be addressed
251. The double-layered arrangement in gastrula and neurula ectoderm of Xenopus has been described by several authors [1&12, 16, 19, 211. In the experiments reported here we show that the two ectodermal cell sheets separated into distinct cell masses when cultured in Leibovitz medium. These differences in cell affinity could indicate that the deep layer has the quality of endoderm. If so it should not be able to react on inducing stimuli. To answer this question inner or outer ectoderm layers were treated with either vegetalizing factor or dorsal blastopore lip as inducers in the sandwich-method.
Methods Eggs of Xenopus laevis were obtained by injecting males and females with primogonyl (Schering, Berlin). The males received two doses of 3001.U. each and the females one dose of 600 I.U.. The jelly coat of early gastrulae (stage 10, Nieuwkoop and Faber [17]) was removed at room temperature by treatment for 10-1 5 min in the following medium: 540 mg cysteine hydrochloride in 12 ml Holtfreter solution, corrected to pH 7.4 with NaOH. The jelly-free eggs were rinsed 10 times employing up to 500 ml sterile Holtfreter solution. Ectoderm (both layers, SD) of early gastrulae was isolated at an angle of about 30" around the animal pole. In further series the two ectodermal layers were separated in situ by fine glass needles into the superficial (S) layer and the deep @) layer as follows. First, only the superficial cell layer was carefully stripped off like the shell of an egg. This outer layer consisted of pigment-rich cells. Then the underlying (deep) layer of nearly pigment-free cells was removed. Two pieces each of the S-, D-, and SDlayers were combined with inducer in the form of a sandwich (Fig. 1). Explants of the control series were prepared in the same way omitting the inducer. Vegetalizing factor, which has mesodermal and endoderma1 inducing activity [3, 6, 1, was extracted from chicken embryos, purified, and prepared for the test as described elsewhere [3]. Dorsal blastopore lip as a neural inducer [20] was isolated from early gastrulae, followed by a sagittal median subdivision into two smaller pieces to simplify the combination with ectoderm. The vegetalizing factor in pellet form or dorsal blastopore lip was wrapped with inner or superficial ectoderm layer in the sandwich-technique (Fig. 1). Explants with or without inducer were cultured for up to five days at 20" C.Some control explants of series SD were cultured in modified Leibovitz medium [2, 5, 15,
207
Fig. 1. Diagrammatic representation of the sandwichmethod, showing the location of the cmployed ectoderm of early gastrula stage. Superficial +deep layer (SD-series), superficial layer (S-series), deep layer (D-series), i inducer
Tabk 1. Differentiations of the superficial and/or deep ectodermal layers of Xenopus laeuis cultured with or without vegetalizing factor or dorsal blastopore lip Control series
Experimental series
Superficial Superficial Deep +deep layer layer layers
Vegetalizing factor as inducer
Upper blastopore lip as inducer
Superficial Superficial Deep +deep layer layer layers
Superficial Superficial Dcep +deep layer layer layers
Total number of experiments Number of cases available for histogical examination Atypical epidermis
23
21
21
22
17
17
22
22
25
22
17
15
20
15
12
20
17
14
22 (100)
17 (100)
15 (100)
Epidermis Mesenchyme Melanophores
-
-
Brain Neural tube Neural structures not specified Neuroid
-
-
-
-
-
-
Notochord Myotomes Blood cells
-
-
-
-
1 1
-
(6) (6)
3 (20) 2 (13) 1 (7)
6 (30)
5 (33)
20(100)
15(100) 9 (60) 5 (33)
18 (90) 10 (50)
(5)
-
4 (20) 2 (10)
1 1
1
-
-
-
-
12(100)
20(100) 20(100)
17(100) 17(100)
14(100) 13 (93)
16 (80)
16 (94)
4 (29)
2 (17)
17 (85)
16 (94)
9 (64)
7 (58)
2 (10) 3 (15)
5 (29) 2 (12)
10 (83) 8 (67)
(7) (7)
-
-
2 (13)
-
5 (25) 12 (60) 3 (15)
(7) 9 (60) 1
-
7 (58) 7 (58)
-
5 (36)
-
-
-
-
18 (90) 19 (95)
16 (94) 15 (88)
14 (100) 14 (100)
-
1
(6)
2 (14)
Percentage of cases available for the histological examination in brackets
181. After four or five days culture only explants showing no obvious disruptions were fixed in Bouin’s fluid and block-stained with borax-carmine. After dehydration in ethanol, parafin sections of 10 pm were counterstained with aniline-blue-orange G.
experiments was highest in the SD series and lowest in the D series. In the D series several explants did not curl up well after isolation or disintegrated into fragments during culture. A . Control Series (Explants Without Inducer)
Results The differentiation of early gastrula ectoderm cultured with or without either vegetalizing factor or blastopore lip is summarized in Table 1. Generally, the number of successful
1. SD-series (Superficial and Deep Ectoderm Layer)
Explants containing both ectodermal layers showed separation of both layers within 30 h culture in Leibovitz medium.
208
Fig. 2a. b. External views of the explants prepared from supe.rficial+deep layers without inducer. a Culture in L-15 medium, b culture in Holtfreter solution. d deep ectoderm layer, s superficial ectoderm layer. Magnification: x 100
Fig. 3. Section of an explant consisting of superficial and deep layers cultured without inducer in Holtfreter solution. s superficial layer, d deep layer. Magnification: x 180
The separation of the superficial ectoderm layer, containing black embryonic pigment, from the nearly pigment-free inner ectodermal layer could easily be observed by macroscopic examination (Fig. 2a). In Holtfreter solution the explants did not show this separation phenomenon into distinct spheres (Fig. 2b). The central parts of histological sections of these explants cultured in Holtfreter solution contained yolk-rich cell material, which were surrounded by cells containing pigment granules and much less yolk plate-
lets than the central mass (Fig. 3). In both media the ectoderm did not form any mesodermal or neural differentiation. 2. S-Series (Superficial Ectodem Layer) All 17 explants available differentiated into atypical epidermis (Fig. 4). In one case a part of the explant had differentiated into epidermal epithelium and into mesenchyme.
209
Fig. 4. Section of isolated superficial ectoderm layer. The tissue has differentiated into atypical epidermis. Magnification : x 280
Fig. 5. Section of isolated deep ectoderm layer. Sucker cells can be identified at the periphery. No further specific differentiations can be distinguished. su sucker cells. Magnification: x 180
3. D-series (Deep Ectoderm Layer)
The 15 cases available differentiated into atypical epidermis. After four day’s culture five of the explants rotated in the culture vessel, a macroscopic criterion that the ectoderm has differentiated into ciliated epidermis. By the histological observation it could be shown that all explants contained sucker cells (Fig. 5). In a few cases, mesenchyme (13%) and melanophores (7%) could also be observed. B. Experimental Series 1. Explants Treated with Vegetalizing Factor
a ) SD-Series (Superficial and Deep Ectoderm Layer). When the SD ectoderm layers were treated with vegetalizing fac-
tor, four out of 20 cases differentiated into mesodermal and neural tissues showing typical trunk/tail organization (notochord, somites, and neural tube). Based on 20 observations, notochord, somites, mesenchyme, and blood cells were observed in 5 , 12, 18, and 3 cases, respectively. Neural structures, e.g. brain and neural tube were identified in one and four cases, respectively.
b) S-Series (Superficial Ectoderm Layer). Notochord and neural tube formation were found in one out of 15 cases (Fig. 6). Nine (60%) explants have formed somites. In comparison to the SD-series in which mesenchyme and melanophores occurred in 90 and 50%, respectively, the same treatment in the S-series resulted in lower percentages of mesenchyme and melanophore differentiation (60 and 33%, re-
210
Fig. 6. Section of isolated superficial ectoderm layers treated with vegetalizing factor. The explant has differentiated into notochord (n) and somites (s). Magnification: x 170
Fig. 7. Section of isolated deep ectoderm treated with vegetalizing factor showing notochord (n), neural tube (nt), and endodermal epithelium (en). Magnification: x 130
spectively). One reason could be that the initial cell mass of the explants in the SD-series was higher than in the S-series. c ) D-series (Deep Ectoderm Layer). Twelve out of 17 explants could be used for examination. The remaining five
were rejected because the vegetalizing factor became detached from the explants during culture. The 12 cases available for examination differentiated into notochord (58%), somites (58%), and neural tube (58%). Mesenchyme and brain were formed in 83% and 17%, respectively. An explant with notochord and neural tube is shown in Fig. 7.
211
2. Explants with Dorsal Blastopore Lip as an Inducer
The differentiation of mesodermal tissues such as notochord and somites was observed at a very high frequency (over 88%) in the following three series. The decision whether these differentiations originated from self-differentiation or were caused by induction from the blastopore lip, however, is not possible. Nevertheless, at least in part, the neural differentiations can be considered as inductions evoked in the reacting ectoderm rather than self-differentiation of the dorsal blastopore lip. a ) SD-Series (Superficial and Deep Ectoakrm Layer). Out of 20 examined cases, neural structures such as brain, eye, and lens were observed in 17 (85%), seven (35%), and five (25%), respectively. Notochord differentiated in 90% and somites in 95% of the cases available. b ) S-Series (Superficial Ectodermal Layer). Out of 17 cases brain and neural tube differentiated in 16 and 5 cases, respectively. Eye differentiations were found in three cases. Notochord and somites could be identified in 94% and 88%, respectively. c ) D-Series (Deep Ectoakrm Layer). Brain and neural tube were observed in nine and five, respectively, of 14 cases examined. In this series, ear vesicle formation was found in four cases (29%), but there was no differention of eyes at all.
Discussion
There exist clear differences in the topographic organization of urodelan and anuran gastrula ectoderm. The ectoderm of urodeles such as Triturus or Ambystoma consists of a single layer of homogeneous cells [13, 251. The ectoderm of anurans such as Xenopus or Rana, however, is composed of two distinct cell sheets, which can be distinguished by morphological criteria [12, 161. Moreover, differences in cell affinity between cells of the inner and outer ectoderm layer can be observed. Under certain experimental conditions (culture in Leibovitz-medium, L-15) we could show that in explants consisting of both ectodermal sheets the inner layer separates from the outer one. From our earlier experiments [5, 151 we know that L-15adapted to the requirements of amphibian cell culture is closer to physiological conditions than the hypotonic Holtfreter solution. Nevertheless Holtfreter solution is still the most suitable in the sandwich-technique [9] or implantation-method [14], because the hypotonic medium supports curling up and the formation of closed spheres. This is apparently one reason why a separation of inner and outer ectoderm layer does not take place in Holtfreter solution. On the other hand for the culture of single cells (for example study of cell affinity) the modified L-15 must be recommended. The separation of the inner and outer layers into distinct cell masses in L-15 indicates that the inner and superficial ectoderm layers show differences in cell affinity. That the inner layer does not represent endodermal cells, which could explain the separation from the superficial ectodermal layer, could be ruled out by our induction experiments with vegetalizing factor. It could be shown that the inner ectoderm layer will differentiate into somites and notochord after the treatment with vegetalizing factor. If the
inner ectoderm layer consisted of endodermal cells it could never react on the inducing stimulus of the vegetalizing factor. Columnar shaped cells, which can be observed in isolated inner ectoderm after five days culture must be considered as sucker cells rather than endodermal structures [23, 26, 271. As concerns the superficial ectoderm layer we could not ObServe a markedly lower mesodermal competence compared with the inner ectodermal layer. The differentiation of neural structures in the series with vegetalizing factor as inducer can be explained as the result of secondary cell interactions [l, 6, 151. Experiments with upper blastopore lip as an inducer indicate that both the superficial as well as the inner ectoderm layer is able to react on neural stimuli. Our results from homoplastic transplantations are of course of limited significance, because we cannot distinguish neural inductions as a result of self-differentiation of the living inducer (blastopore lip) on the one hand and neural tissue, which has been formed from the target cells (inner or outer ectoderm layer) on the other hand. Nevertheless, it can be assumed in agreement with xenoplastic transplantations [4] that preferentially the ectodermal cells differentiate into neural tissue, while the blastopore lip mainly forms notochord and somites. In contrast to the results of Sudarwati and Nieuwkoop [21] we could not observe a markedly lower competence of the outer ectoderm layer in comparison with the inner ectoderm layer. Sudarwati and Nieuwkoop combined endoderm of the vegetal pole with inner or outer ectoderm layer of Xenopus. These authors mentioned that the inner layer, in contrast to the series with outer ectoderm layer, quickly adhered to the isolated vegetal pole material. The close contact of vegetal pole material to inner rather than to outer ectoderm layer could explain why in the experiments of Sudarwati and Nieuwkoop the differentiation of neural structures was markedly higher in recombinates with inner ectoderm layer than in the series with outer ectoderm layer. In our experiments both inner and outer ectoderm layers quickly adhered to the blastopore lip, which served as inducing tissue. Thus in our experiments the neural competence of the outer and inner ectoderm layer seems to be equal. Our results, which show a strong neural competence of the superficial ectoderm layer, are not necessarily contradictory to the observations of other authors [19, 211 that in normogenesis the neural anlage arises mainly from the inner ectoderm layer, the superficial layer forming only the ependymal layer of the neural tube. In normogenesis of anurans the outer ectodermal layer is separated from the inducing underlying chordamesoderm by the inner ectoderm layer which acts as a barrier. Thus the low neural competence of the outer ectoderm layer in vivo could be explained by assuming that neural induction is triggered by diffusible morphogenetic factors over very short distances [24] or/and by close cell-tocell contacts between the inducing tissue and target cells [8]. The lower neural competence of the outer ectoderm layer can be explained as follows. Assuming that the neural competence of both cell layers, which show no close adhesion to each other, is essentially equal and that a diffusible morphogenetic factor starting from the chordamesoderm is responsible for the neural induction of the inner ectoderm layer, it must be concluded that the inducer is active over a short distance only and cannot reach the outer ectoderm layer without crossing the barrier of the inner layer and the gap between the inner and outer layer. Furthermore it cannot be excluded that the morphogenetic factor moves
21 2 very slowly and that it reaches the superficial ectodermal cells when they have already lost their neural competence. On the other hand we must take into account that the transfer of a neuralizing factor from the inducing tissue to the ectodermal target cells could take place in areas of close cell-to-cell contact only. This would explain why in normogenesis the neural anlage arises for the f a r greater part from the inner ectoderm layer, which has direct cell-tocell contact to the inducing underlying chordamesoderm. AcknowIedgements. The present work was supported by the Deutsche Forschungsgemeinschaft, Schwerpunktprogramm: ,,Steuerung der Differenzierung von ein- und wenigzelligen eukaryotischen Systemen", by the Ministerium fiir Wissenschaft und Forschung des Landes Nordrhein-Westfalen, and by a Grant-inAid for Scientific Research from the Ministery of Education, Science, and Culture of Japan. The authors thank especially Prof. Dr. Dr. H. Tiedemann, Berlin, who supplied us with vegetalizing factor.
References 1. Asahi K, Born J, Tiedemann H (1979) Formation of mesoder-
mal pattern by sccondary inducing interactions. Wilhelm Roux' Arch 187:231-244 2. Balls M, Ruben LN (1966) Cultivation in vitro of normal and neoplastic cells of Xenopus laeuis. Exptl. Cell Res 43:694-695 3. Born J, Geithe HP, Tiedemann H, Tiedemann H, KocherBecker U (1972) Isolation of a vegetalizing factor. HoppeSeyler's Z Physiol Chem 353: 1075-1084 4. Faulhaber I(1970) Das Induktionsvermogen von lebendem Urmundgewebe in ordnungsfremdem Wirtsektoderm. Wilhelm Roux's Arch 165:296302 5. Grunz H (1977) Differentiation of the four animal and the four vegetal blastomeres of the eightcell-stage of Triturus alpestris. Wilhelm Roux' Arch 181 :267-277 6. Grunz H (1979) Change of the differentiation pattern of amphibian ectoderm after the increase of the initial cell mass. Wilhelm Roux' Arch 187 :49-57 7. Grunz H, Multier-Lajous A-M, Herbst R (1975) The differentiation of isolated amphibian ectoderm with our without treatment with an inductor. A scanning electron microscope study. Wilhelm Roux's Arch 178:277-284 8. Grunz H, Staubach J (1979) Cell contacts between chordamcsoderm and the overlaying neuroectoderm (presumptive central nervous system) during the period of primary embryonic induction in amphibians. Differentiation 14: 5 9 4 5 9. Holtfreter J (1933) Nachweis der Induktionsfahigkeit abgetoteter Keimteile. Isolation- und Transplantationsversuche. Wilhelm Roux's Arch 128:584433 10. Keller RE (1975) Vital dye mapping of the gastrula and neurula of Xenopus lueuis. I. Prospective areas and morphogenetic movements in the superficial layer. Develop Biol42: 222-241
11. Keller RE (1976) Vital dye mapping of the gastrula and neurula of Xenopus laeuis. 11. Prospcctive areas and morphogenetic movements in the deep layer. Develop Biol51: 118-137 12. Keller RE (1980) The cellular basis of epiboly: An SEM study of deepcell rearrangement during gastrulation in Xenopus laeuis. J Embryo1 Exptl Morphol60:201-234 13. Lsvtrup S (1975) Fate maps and gastrulation in amphibia A critique of current views. Can J Zoo1 53:473479 14. Mangold 0 (1923) Transplantationsversuche zur Frage der Spezifizitiit und der Bildung der Keimblltter. Wilhelm Roux's Arch 100: 198-301 15. Minuth M, Grunz H (1980) The formation of mesodermal derivatives after induction with vegetalizing factor depends on secondary cell interactions. Cell Differentiation 9 :229-238 16. Nieuwkoop PD, Florschiitz PA (1950) Quelques caracteres speciaux de la gastrulation et de la neurulation del'oeuf de Xenopus Iaeuis, Daud. et de quelques autres Anoures. I'ere partie-Etude descriptive. Arch Biol (Liege) 61 :113-150 17. Nieuwkoop PD, Faber J (1967) Normal table of Xenopus laeuis (Daudin), second edition. North-Holland Publishing Company, Amsterdam 18. Rafferty KA Jr (1976) The physiology of amphibian cells in culture, In: Lofts B (ed) Physiology of the Amphibia. Vol. 111. Academic Press, New York, p 11 1-162 19. Rossi A (1965) Sul valore prospettico degli strati esterno ed interno della placca neurale degli anfibi anuri (Rana esculenta, Bufo hufo e Bufo uiridis). Boll Zoo1 32:991-1017 20. Spemann H, Mangold H (1924) o b e r Induktion von Embryonalanlagen durch Implantation artfremder Organisatoren. Wilhelm Roux's Arch 100:599-638 21. Sudarwati S, Nieuwkoop PD (1971) Mesoderm formation in anuran Xenopus laeuis. Wilhelm Row's Arch 166: 189-204 22. Szebenyi ES (1977) Frog development. In: Atlas of developmental embryology. Fairleigh Dickinson University Press, New Jersey, p 59-80 23. Takata C, Yamada T (1960) Endodermal tissues developed from the isolated newt ectoderm under the influence of guinea pig bone marrow. Embryologia 5:8-20 24. Toivonen S, Tarin D. SaxCn L, Tarin PJ, Wartiovaara J (1975) Transfilter studies on neural induction in the newt. Differentiation 4 : 1-7 25. Vogt W (1929) Gestaltungsanalyse am Amphibienkeim mit ortlicher Vitalfarbung. 11. Teil. Gastrulation und Mesodermbildung bei Urodelen und Anuren. Wilhelm Roux's Arch 120 :384706 26. Yamada T (1933) h e r die Determination der Haftdriisen bei Rana nigromaculata. J Faculty Sci Univ Tokyo 3 :239-254 27. Yamada T (1938) Weitere Analyse der Determination der Haftdriise bei Rana nigromaculata, mit einigen Bemerkungen iiber die Induktion anderer Kopforgane. J Faculty Sci Univ Tokyo 5:133-163
Received March 1982 / Accepted in revised form September 1982
Differentiation (1983) 23: 206-212
Springer-Verlag 1983
Effects of Inducers on Inner and Outer Gastrula Ectoderm Layers of Xenopus laevis Makoto Asashima' and Horst Grunz' Department of Biology, Yokohama City University, Yokohama 236, Japan Fachbereich 9 (Biologie), Zoophysiologie, Universitit Essen - GHS, Universitatsstr. 5, 4300 Essen 1, Federal Republic of Germany
Abstract. Gastrula ectoderm, isolated from Xenopus laeuis, was cultured in Holtfreter solution or modified Leibovitz medium (L-15) by the sandwich-method with or without inducer. The ectoderm (SD cell layers) consists of two cell sheets, representing a superficial (S) and a deep (D) layer. In the L-15 medium rather than in Holtfreter solution, the two cell layers separate out into distinct cell masses. This difference in cell affinity under certain experimental conditions could indicate that the deep layer contains endodermal cells. However, an endodermal character of the deep layer can be ruled out by induction experiments with vegetalizing factor or dorsal blastopore lip as inducers. Under the influence of vegetalizing factor the outer as well as the inner ectoderm layer differentiated into mesodermal derivatives such as notochord and somites. The results of the experiments with dorsal blastopore lip as inducer indicate that both inner and outer ectoderm layers are responsive to the neural stimulus. The lower neural competence of the outer ectoderm layer observed by several authors in normogenesis is discussed with regard to the hypothesis about short distance diffusion of the neuralizing factor and/or close cellto-cell contact between inducing tissue and ectodermal target cells.
Introduction It is well known that the gastrula ectoderm of amphibian embryos displays high responsiveness to inducing stimuli. For this reason competent ectoderm has been used as a test system in the implantation or sandwich technique by many investigators [9, 141. However, ectoderm of different amphibian species shows different reaction to inducing stimuli. Using Xenopus and Ambystoma embryos, Faulhaber [4] reported about differences of the reacting ectoderm to the induction of the dorsal blastopore lip when tested by xenoplastic transplantations. The blastopore lip of Xenopus induced neural tissues in Ambystoma host ectoderm more frequently than did Ambystoma dorsal lip in Xenopus host ectoderm. Moreover, there also are clear topographical differences in gastrula ectoderm between urodeles and anurans. While the ectoderm of the gastrula in anurans such as Xenopus and Rana consists of two distinct cell layers, in urodeles such as Triturus and Ambystoma ectoderm consists of a homogeneous cell layer (12, 13, 22, 2 To whom correspondence should be addressed
251. The double-layered arrangement in gastrula and neurula ectoderm of Xenopus has been described by several authors [1&12, 16, 19, 211. In the experiments reported here we show that the two ectodermal cell sheets separated into distinct cell masses when cultured in Leibovitz medium. These differences in cell affinity could indicate that the deep layer has the quality of endoderm. If so it should not be able to react on inducing stimuli. To answer this question inner or outer ectoderm layers were treated with either vegetalizing factor or dorsal blastopore lip as inducers in the sandwich-method.
Methods Eggs of Xenopus laevis were obtained by injecting males and females with primogonyl (Schering, Berlin). The males received two doses of 3001.U. each and the females one dose of 600 I.U.. The jelly coat of early gastrulae (stage 10, Nieuwkoop and Faber [17]) was removed at room temperature by treatment for 10-1 5 min in the following medium: 540 mg cysteine hydrochloride in 12 ml Holtfreter solution, corrected to pH 7.4 with NaOH. The jelly-free eggs were rinsed 10 times employing up to 500 ml sterile Holtfreter solution. Ectoderm (both layers, SD) of early gastrulae was isolated at an angle of about 30" around the animal pole. In further series the two ectodermal layers were separated in situ by fine glass needles into the superficial (S) layer and the deep @) layer as follows. First, only the superficial cell layer was carefully stripped off like the shell of an egg. This outer layer consisted of pigment-rich cells. Then the underlying (deep) layer of nearly pigment-free cells was removed. Two pieces each of the S-, D-, and SDlayers were combined with inducer in the form of a sandwich (Fig. 1). Explants of the control series were prepared in the same way omitting the inducer. Vegetalizing factor, which has mesodermal and endoderma1 inducing activity [3, 6, 1, was extracted from chicken embryos, purified, and prepared for the test as described elsewhere [3]. Dorsal blastopore lip as a neural inducer [20] was isolated from early gastrulae, followed by a sagittal median subdivision into two smaller pieces to simplify the combination with ectoderm. The vegetalizing factor in pellet form or dorsal blastopore lip was wrapped with inner or superficial ectoderm layer in the sandwich-technique (Fig. 1). Explants with or without inducer were cultured for up to five days at 20" C.Some control explants of series SD were cultured in modified Leibovitz medium [2, 5, 15,
207
Fig. 1. Diagrammatic representation of the sandwichmethod, showing the location of the cmployed ectoderm of early gastrula stage. Superficial +deep layer (SD-series), superficial layer (S-series), deep layer (D-series), i inducer
Tabk 1. Differentiations of the superficial and/or deep ectodermal layers of Xenopus laeuis cultured with or without vegetalizing factor or dorsal blastopore lip Control series
Experimental series
Superficial Superficial Deep +deep layer layer layers
Vegetalizing factor as inducer
Upper blastopore lip as inducer
Superficial Superficial Deep +deep layer layer layers
Superficial Superficial Dcep +deep layer layer layers
Total number of experiments Number of cases available for histogical examination Atypical epidermis
23
21
21
22
17
17
22
22
25
22
17
15
20
15
12
20
17
14
22 (100)
17 (100)
15 (100)
Epidermis Mesenchyme Melanophores
-
-
Brain Neural tube Neural structures not specified Neuroid
-
-
-
-
-
-
Notochord Myotomes Blood cells
-
-
-
-
1 1
-
(6) (6)
3 (20) 2 (13) 1 (7)
6 (30)
5 (33)
20(100)
15(100) 9 (60) 5 (33)
18 (90) 10 (50)
(5)
-
4 (20) 2 (10)
1 1
1
-
-
-
-
12(100)
20(100) 20(100)
17(100) 17(100)
14(100) 13 (93)
16 (80)
16 (94)
4 (29)
2 (17)
17 (85)
16 (94)
9 (64)
7 (58)
2 (10) 3 (15)
5 (29) 2 (12)
10 (83) 8 (67)
(7) (7)
-
-
2 (13)
-
5 (25) 12 (60) 3 (15)
(7) 9 (60) 1
-
7 (58) 7 (58)
-
5 (36)
-
-
-
-
18 (90) 19 (95)
16 (94) 15 (88)
14 (100) 14 (100)
-
1
(6)
2 (14)
Percentage of cases available for the histological examination in brackets
181. After four or five days culture only explants showing no obvious disruptions were fixed in Bouin’s fluid and block-stained with borax-carmine. After dehydration in ethanol, parafin sections of 10 pm were counterstained with aniline-blue-orange G.
experiments was highest in the SD series and lowest in the D series. In the D series several explants did not curl up well after isolation or disintegrated into fragments during culture. A . Control Series (Explants Without Inducer)
Results The differentiation of early gastrula ectoderm cultured with or without either vegetalizing factor or blastopore lip is summarized in Table 1. Generally, the number of successful
1. SD-series (Superficial and Deep Ectoderm Layer)
Explants containing both ectodermal layers showed separation of both layers within 30 h culture in Leibovitz medium.
208
Fig. 2a. b. External views of the explants prepared from supe.rficial+deep layers without inducer. a Culture in L-15 medium, b culture in Holtfreter solution. d deep ectoderm layer, s superficial ectoderm layer. Magnification: x 100
Fig. 3. Section of an explant consisting of superficial and deep layers cultured without inducer in Holtfreter solution. s superficial layer, d deep layer. Magnification: x 180
The separation of the superficial ectoderm layer, containing black embryonic pigment, from the nearly pigment-free inner ectodermal layer could easily be observed by macroscopic examination (Fig. 2a). In Holtfreter solution the explants did not show this separation phenomenon into distinct spheres (Fig. 2b). The central parts of histological sections of these explants cultured in Holtfreter solution contained yolk-rich cell material, which were surrounded by cells containing pigment granules and much less yolk plate-
lets than the central mass (Fig. 3). In both media the ectoderm did not form any mesodermal or neural differentiation. 2. S-Series (Superficial Ectodem Layer) All 17 explants available differentiated into atypical epidermis (Fig. 4). In one case a part of the explant had differentiated into epidermal epithelium and into mesenchyme.
209
Fig. 4. Section of isolated superficial ectoderm layer. The tissue has differentiated into atypical epidermis. Magnification : x 280
Fig. 5. Section of isolated deep ectoderm layer. Sucker cells can be identified at the periphery. No further specific differentiations can be distinguished. su sucker cells. Magnification: x 180
3. D-series (Deep Ectoderm Layer)
The 15 cases available differentiated into atypical epidermis. After four day’s culture five of the explants rotated in the culture vessel, a macroscopic criterion that the ectoderm has differentiated into ciliated epidermis. By the histological observation it could be shown that all explants contained sucker cells (Fig. 5). In a few cases, mesenchyme (13%) and melanophores (7%) could also be observed. B. Experimental Series 1. Explants Treated with Vegetalizing Factor
a ) SD-Series (Superficial and Deep Ectoderm Layer). When the SD ectoderm layers were treated with vegetalizing fac-
tor, four out of 20 cases differentiated into mesodermal and neural tissues showing typical trunk/tail organization (notochord, somites, and neural tube). Based on 20 observations, notochord, somites, mesenchyme, and blood cells were observed in 5 , 12, 18, and 3 cases, respectively. Neural structures, e.g. brain and neural tube were identified in one and four cases, respectively.
b) S-Series (Superficial Ectoderm Layer). Notochord and neural tube formation were found in one out of 15 cases (Fig. 6). Nine (60%) explants have formed somites. In comparison to the SD-series in which mesenchyme and melanophores occurred in 90 and 50%, respectively, the same treatment in the S-series resulted in lower percentages of mesenchyme and melanophore differentiation (60 and 33%, re-
210
Fig. 6. Section of isolated superficial ectoderm layers treated with vegetalizing factor. The explant has differentiated into notochord (n) and somites (s). Magnification: x 170
Fig. 7. Section of isolated deep ectoderm treated with vegetalizing factor showing notochord (n), neural tube (nt), and endodermal epithelium (en). Magnification: x 130
spectively). One reason could be that the initial cell mass of the explants in the SD-series was higher than in the S-series. c ) D-series (Deep Ectoderm Layer). Twelve out of 17 explants could be used for examination. The remaining five
were rejected because the vegetalizing factor became detached from the explants during culture. The 12 cases available for examination differentiated into notochord (58%), somites (58%), and neural tube (58%). Mesenchyme and brain were formed in 83% and 17%, respectively. An explant with notochord and neural tube is shown in Fig. 7.
211
2. Explants with Dorsal Blastopore Lip as an Inducer
The differentiation of mesodermal tissues such as notochord and somites was observed at a very high frequency (over 88%) in the following three series. The decision whether these differentiations originated from self-differentiation or were caused by induction from the blastopore lip, however, is not possible. Nevertheless, at least in part, the neural differentiations can be considered as inductions evoked in the reacting ectoderm rather than self-differentiation of the dorsal blastopore lip. a ) SD-Series (Superficial and Deep Ectoakrm Layer). Out of 20 examined cases, neural structures such as brain, eye, and lens were observed in 17 (85%), seven (35%), and five (25%), respectively. Notochord differentiated in 90% and somites in 95% of the cases available. b ) S-Series (Superficial Ectodermal Layer). Out of 17 cases brain and neural tube differentiated in 16 and 5 cases, respectively. Eye differentiations were found in three cases. Notochord and somites could be identified in 94% and 88%, respectively. c ) D-Series (Deep Ectoakrm Layer). Brain and neural tube were observed in nine and five, respectively, of 14 cases examined. In this series, ear vesicle formation was found in four cases (29%), but there was no differention of eyes at all.
Discussion
There exist clear differences in the topographic organization of urodelan and anuran gastrula ectoderm. The ectoderm of urodeles such as Triturus or Ambystoma consists of a single layer of homogeneous cells [13, 251. The ectoderm of anurans such as Xenopus or Rana, however, is composed of two distinct cell sheets, which can be distinguished by morphological criteria [12, 161. Moreover, differences in cell affinity between cells of the inner and outer ectoderm layer can be observed. Under certain experimental conditions (culture in Leibovitz-medium, L-15) we could show that in explants consisting of both ectodermal sheets the inner layer separates from the outer one. From our earlier experiments [5, 151 we know that L-15adapted to the requirements of amphibian cell culture is closer to physiological conditions than the hypotonic Holtfreter solution. Nevertheless Holtfreter solution is still the most suitable in the sandwich-technique [9] or implantation-method [14], because the hypotonic medium supports curling up and the formation of closed spheres. This is apparently one reason why a separation of inner and outer ectoderm layer does not take place in Holtfreter solution. On the other hand for the culture of single cells (for example study of cell affinity) the modified L-15 must be recommended. The separation of the inner and outer layers into distinct cell masses in L-15 indicates that the inner and superficial ectoderm layers show differences in cell affinity. That the inner layer does not represent endodermal cells, which could explain the separation from the superficial ectodermal layer, could be ruled out by our induction experiments with vegetalizing factor. It could be shown that the inner ectoderm layer will differentiate into somites and notochord after the treatment with vegetalizing factor. If the
inner ectoderm layer consisted of endodermal cells it could never react on the inducing stimulus of the vegetalizing factor. Columnar shaped cells, which can be observed in isolated inner ectoderm after five days culture must be considered as sucker cells rather than endodermal structures [23, 26, 271. As concerns the superficial ectoderm layer we could not ObServe a markedly lower mesodermal competence compared with the inner ectodermal layer. The differentiation of neural structures in the series with vegetalizing factor as inducer can be explained as the result of secondary cell interactions [l, 6, 151. Experiments with upper blastopore lip as an inducer indicate that both the superficial as well as the inner ectoderm layer is able to react on neural stimuli. Our results from homoplastic transplantations are of course of limited significance, because we cannot distinguish neural inductions as a result of self-differentiation of the living inducer (blastopore lip) on the one hand and neural tissue, which has been formed from the target cells (inner or outer ectoderm layer) on the other hand. Nevertheless, it can be assumed in agreement with xenoplastic transplantations [4] that preferentially the ectodermal cells differentiate into neural tissue, while the blastopore lip mainly forms notochord and somites. In contrast to the results of Sudarwati and Nieuwkoop [21] we could not observe a markedly lower competence of the outer ectoderm layer in comparison with the inner ectoderm layer. Sudarwati and Nieuwkoop combined endoderm of the vegetal pole with inner or outer ectoderm layer of Xenopus. These authors mentioned that the inner layer, in contrast to the series with outer ectoderm layer, quickly adhered to the isolated vegetal pole material. The close contact of vegetal pole material to inner rather than to outer ectoderm layer could explain why in the experiments of Sudarwati and Nieuwkoop the differentiation of neural structures was markedly higher in recombinates with inner ectoderm layer than in the series with outer ectoderm layer. In our experiments both inner and outer ectoderm layers quickly adhered to the blastopore lip, which served as inducing tissue. Thus in our experiments the neural competence of the outer and inner ectoderm layer seems to be equal. Our results, which show a strong neural competence of the superficial ectoderm layer, are not necessarily contradictory to the observations of other authors [19, 211 that in normogenesis the neural anlage arises mainly from the inner ectoderm layer, the superficial layer forming only the ependymal layer of the neural tube. In normogenesis of anurans the outer ectodermal layer is separated from the inducing underlying chordamesoderm by the inner ectoderm layer which acts as a barrier. Thus the low neural competence of the outer ectoderm layer in vivo could be explained by assuming that neural induction is triggered by diffusible morphogenetic factors over very short distances [24] or/and by close cell-tocell contacts between the inducing tissue and target cells [8]. The lower neural competence of the outer ectoderm layer can be explained as follows. Assuming that the neural competence of both cell layers, which show no close adhesion to each other, is essentially equal and that a diffusible morphogenetic factor starting from the chordamesoderm is responsible for the neural induction of the inner ectoderm layer, it must be concluded that the inducer is active over a short distance only and cannot reach the outer ectoderm layer without crossing the barrier of the inner layer and the gap between the inner and outer layer. Furthermore it cannot be excluded that the morphogenetic factor moves
21 2 very slowly and that it reaches the superficial ectodermal cells when they have already lost their neural competence. On the other hand we must take into account that the transfer of a neuralizing factor from the inducing tissue to the ectodermal target cells could take place in areas of close cell-to-cell contact only. This would explain why in normogenesis the neural anlage arises for the f a r greater part from the inner ectoderm layer, which has direct cell-tocell contact to the inducing underlying chordamesoderm. AcknowIedgements. The present work was supported by the Deutsche Forschungsgemeinschaft, Schwerpunktprogramm: ,,Steuerung der Differenzierung von ein- und wenigzelligen eukaryotischen Systemen", by the Ministerium fiir Wissenschaft und Forschung des Landes Nordrhein-Westfalen, and by a Grant-inAid for Scientific Research from the Ministery of Education, Science, and Culture of Japan. The authors thank especially Prof. Dr. Dr. H. Tiedemann, Berlin, who supplied us with vegetalizing factor.
References 1. Asahi K, Born J, Tiedemann H (1979) Formation of mesoder-
mal pattern by sccondary inducing interactions. Wilhelm Roux' Arch 187:231-244 2. Balls M, Ruben LN (1966) Cultivation in vitro of normal and neoplastic cells of Xenopus laeuis. Exptl. Cell Res 43:694-695 3. Born J, Geithe HP, Tiedemann H, Tiedemann H, KocherBecker U (1972) Isolation of a vegetalizing factor. HoppeSeyler's Z Physiol Chem 353: 1075-1084 4. Faulhaber I(1970) Das Induktionsvermogen von lebendem Urmundgewebe in ordnungsfremdem Wirtsektoderm. Wilhelm Roux's Arch 165:296302 5. Grunz H (1977) Differentiation of the four animal and the four vegetal blastomeres of the eightcell-stage of Triturus alpestris. Wilhelm Roux' Arch 181 :267-277 6. Grunz H (1979) Change of the differentiation pattern of amphibian ectoderm after the increase of the initial cell mass. Wilhelm Roux' Arch 187 :49-57 7. Grunz H, Multier-Lajous A-M, Herbst R (1975) The differentiation of isolated amphibian ectoderm with our without treatment with an inductor. A scanning electron microscope study. Wilhelm Roux's Arch 178:277-284 8. Grunz H, Staubach J (1979) Cell contacts between chordamcsoderm and the overlaying neuroectoderm (presumptive central nervous system) during the period of primary embryonic induction in amphibians. Differentiation 14: 5 9 4 5 9. Holtfreter J (1933) Nachweis der Induktionsfahigkeit abgetoteter Keimteile. Isolation- und Transplantationsversuche. Wilhelm Roux's Arch 128:584433 10. Keller RE (1975) Vital dye mapping of the gastrula and neurula of Xenopus lueuis. I. Prospective areas and morphogenetic movements in the superficial layer. Develop Biol42: 222-241
11. Keller RE (1976) Vital dye mapping of the gastrula and neurula of Xenopus laeuis. 11. Prospcctive areas and morphogenetic movements in the deep layer. Develop Biol51: 118-137 12. Keller RE (1980) The cellular basis of epiboly: An SEM study of deepcell rearrangement during gastrulation in Xenopus laeuis. J Embryo1 Exptl Morphol60:201-234 13. Lsvtrup S (1975) Fate maps and gastrulation in amphibia A critique of current views. Can J Zoo1 53:473479 14. Mangold 0 (1923) Transplantationsversuche zur Frage der Spezifizitiit und der Bildung der Keimblltter. Wilhelm Roux's Arch 100: 198-301 15. Minuth M, Grunz H (1980) The formation of mesodermal derivatives after induction with vegetalizing factor depends on secondary cell interactions. Cell Differentiation 9 :229-238 16. Nieuwkoop PD, Florschiitz PA (1950) Quelques caracteres speciaux de la gastrulation et de la neurulation del'oeuf de Xenopus Iaeuis, Daud. et de quelques autres Anoures. I'ere partie-Etude descriptive. Arch Biol (Liege) 61 :113-150 17. Nieuwkoop PD, Faber J (1967) Normal table of Xenopus laeuis (Daudin), second edition. North-Holland Publishing Company, Amsterdam 18. Rafferty KA Jr (1976) The physiology of amphibian cells in culture, In: Lofts B (ed) Physiology of the Amphibia. Vol. 111. Academic Press, New York, p 11 1-162 19. Rossi A (1965) Sul valore prospettico degli strati esterno ed interno della placca neurale degli anfibi anuri (Rana esculenta, Bufo hufo e Bufo uiridis). Boll Zoo1 32:991-1017 20. Spemann H, Mangold H (1924) o b e r Induktion von Embryonalanlagen durch Implantation artfremder Organisatoren. Wilhelm Roux's Arch 100:599-638 21. Sudarwati S, Nieuwkoop PD (1971) Mesoderm formation in anuran Xenopus laeuis. Wilhelm Row's Arch 166: 189-204 22. Szebenyi ES (1977) Frog development. In: Atlas of developmental embryology. Fairleigh Dickinson University Press, New Jersey, p 59-80 23. Takata C, Yamada T (1960) Endodermal tissues developed from the isolated newt ectoderm under the influence of guinea pig bone marrow. Embryologia 5:8-20 24. Toivonen S, Tarin D. SaxCn L, Tarin PJ, Wartiovaara J (1975) Transfilter studies on neural induction in the newt. Differentiation 4 : 1-7 25. Vogt W (1929) Gestaltungsanalyse am Amphibienkeim mit ortlicher Vitalfarbung. 11. Teil. Gastrulation und Mesodermbildung bei Urodelen und Anuren. Wilhelm Roux's Arch 120 :384706 26. Yamada T (1933) h e r die Determination der Haftdriisen bei Rana nigromaculata. J Faculty Sci Univ Tokyo 3 :239-254 27. Yamada T (1938) Weitere Analyse der Determination der Haftdriise bei Rana nigromaculata, mit einigen Bemerkungen iiber die Induktion anderer Kopforgane. J Faculty Sci Univ Tokyo 5:133-163
Received March 1982 / Accepted in revised form September 1982