When induced by non-specific evocators, anterior neural tissue usually forms (Nieuwkoop, 1). Thus, some other factors are needed (Waddington's "individuators") to specify the regionality of forebrain, midbrain, hindbrain, and spinal cord. Otto Mangold (2) showed that the regional position of the blastopore lip-derived dorsal mesoderm determined the anterior-posterior specification of the neural tube. When transplanted into the blastocoel of early gastrulae, the anteriormost mesoderm induced cement glands and eyes, posterior-most mesoderm induced tails, and the intermediate mesoderm induced the appropriate intermediate structures. This could also be seen by transplanting the tissue at the time when it comprised the dorsal blastopore lip (3). Early dorsal blastopore lips gave rise to head structures; late dorsal blastopore lips induced new tails. To further study the organizer phenomenon, Holtfreter (3a) made a blastopore "sandwich," enclosing the dorsal blastopore lip between slices of undifferentiated ectoderm. The blastopore lips from the early gastrula stages induced the most archencephalic (anterior) structures, whereas the blastopore lips of later embryos caused the differentiation of the more posterior neural elements.
Two major sets of models were generated to explain these results. Both involved the notion of gradients. Here we have to digress briefly to discuss the concept of gradients in embryology. Until they were resurrected in the laboratory of Christiane Nüsslein-Volhard, "gradients" had been a "dirty word" for decades (4). There are several reasons for this, one of which was that they were ridiculed by a person who had been one of the leading investigators of such gradients, Thomas Hunt Morgan (5). The introduction of gradients into embryology is often credited to Boveri (6), who did introduce the term "Gefälle" (gradient) into his work. However, as Sander (7) has made explicit, Boveri's concept of a gradient is not the one that we have today. His is more like a step-gradient, wherein the cell membrane separating two cells also separates two concentrations of the same substance. His illustrations, however, show a graded character from the vegetal to the animal pole, and this makes it appear that there are threshold-related gradients in these embryos.
The threshold gradient idea was brought about primarily by Rünnstrom (8) and Hörstadius (9) in their studies of sea urchins. Their experiments indicated that a gradient of vegetal substance was found throughout the embryo, concentrated in the vegetal-most micromeres, and there was conversely a gradient of animalizing factor in the otherdirection. Their countryman T. Gustafson (10) represented this graphically as two intersecting gradients. Gavin de Beer and Julian Huxley (11), along with C.M. Child (12, 13), created the idea of a "gradient field" Here, a morphogenetic field existed that depended upon a system of induction (see 14). However, the cells responded to the inducer in a graded manner depending upon the inducer concentration. In that way, polarity could be generated. This was a very controversial idea. Morgan detested it because it postulated an alternative mode of inheritance (as in planaria dividing). Spemann (15) also did not agree with it, saying that while a streamof water could drive a water mill, the actual mountain cannot. In 1953, Leopold von Ubisch (16) would counter this argument with an example of the mountain cutting through a temperature gradient so that different types of vegetation would be found at different altitudes. Different genes could be activated by different amounts of substance. However, this remained strictly a hypothesis (for more detailed information, see 17).
Numerous investigators attempted to use gradients to explain the results of Mangold's experiments. The quantitative models postulated a gradient of a single molecule that would be formed within the dorsal mesoderm. The amount of this molecule would specify the regional differences in the overlying ectoderm. For instance, Dalcq and Pasteels (18) and Yamada (19) proposed that two intial substances at either end of the mesoderm would interact to produce a gradient of the actual morphogenetic substance (called "organicine" by the Belgian group). The fate of the overlying ectoderm was determined by the amount of organicine in the mesoderm. Quantitative models, such as that hypothesized by Lehmann (20), postulated that the dorsal mesoderm elaborated different substances at its anterior and posterior ends. The differences in the ratios of these substances produced by these interpenetrating gradients determined the type of neural tissue.
Evidence for the dual gradient model was obtained by Sulo Toivonen. Toivonen's mentor, Gunnar Ekman had worked in the laboratory of H. Braus, the close friend of Spemann who had given Spemann the double assurance idea. After Braus' death in 1924, Ekman spent some time in Spemann's laboratory in Freiburg. Ekmann was intrigued with Holtfreter's experiments on lens induction, and suggested that Toivonen find a lens-specific inducer (21). Toivonen's (22) research did not yield a lens inducer. However, he did find specific neural inducers. Adult guinea pig liver, for instance, only induced spinal cord and did not induce head structures. Conversely, the guinea pig bone marrow was an excellent inducer of heads, but did not induce the trunk neural tube. Similar results were also forthcoming from one of Holtfreter's graduate students, Siao-Hui Chuang (23), but whereas Chuang focused on whether the host environment might alter the induction (it did not), Toivonen was interested in the type of induction, "Leistungspezifitt", produced by these tissues. Moreover, instead of merely recognizing two specificities (head vs trunk) , he divided the neural axis into four regions: anterior (archencephalic) head structures, posterior (deuterencephalic) head structures, trunk structures, and tail structures. Needham (24) called this work of Toivonen's and Chuang's the third fundamental discovery about the organizer. (The first two were the discoveries that killed organizers still had inducing capacity and that killing non-inducing tissues could convert them into inducing ones).
The test of this hypothesis was accomplished by Toivonen and his student, Lauri SaxEn. This partnership, as Hamburger (21) has noted, "left its mark on experimental embryology to this day." Toivonen and SaxEn (25, 26) placed an archencephalic inducer, guinea pig liver, together with a spinocaudal (mesodermal) agent, guinea pig bone marrow, into blastocoels or between sheets of competent ectoderm. In addition to inducing tail and archencephalic structures, the dual implants also formed the intermediate spinal cord structures that neither did alone. Moreover, in the Einsteck experiments, the induced structures were arranged along an axis from archencephalic to tail. Toivonen and SaxEn interpreted these results in terms of two gradients, each arising from its specific inducer. These gradients were projected upon the competent ectoderm. The neuralizing gradient was highest dorsally in the embryo, evenly distributed along the anterior-posterior axis. The mesodermalizing gradient (one must recall that the posterior end of the neural tube becomes mesoderm) is highest in the dorsal midline, extending only into the deuterencephalic area. Needham (24) reflected that this work "is assuredly the fourth fundamental experiment in this field."
Meanwhile Petr Nieuwkoop and his laboratory (1) in the Netherlands had revised the Waddington's (27) evocator/individuator and provided evidence for a two-step process of neural induction. The initial "activation" would induce competent ectoderm to become archencephalic structures. These neurally specified cells would secondarily be "transformed" into more caudal structures by some other factor. Nieuwkoop showed that this caudalizing factor existed in a gradient. Folds of competent ectoderm were implanted at various regions along the dorsal anterior-posterior axis. The distalmost region of each fold remained undifferentiated. The proximal region of each fold showed the specificities characteristic of the region wherein it was inserted into the host. The region between them developed more anterior specification than the region of insertion.
Confirmation of this two-step model was obtained by SaxEn, Toivonen, and T. Vainio (28). They induced competent ectoderm with either a forebrain inducer or a spinocaudal (mesodermalizing) inducer. They then separated the ectoderm from the inducers. And disaggregated the cells. The cells were then reaggregated as pellets. The pellet produced by the forebrain-induced cells gave rise primarily to forebrain structures, while the ectoderm reaggregated from spinocaudal-induced cells gave rise primarily to spinal cord. However, when the two groups of cells were aggregated together, the result was a preponderance of hindbrain structures, structures that neither group of cells would form well on their own.
However important this might be for a model, these neuralizing (activating) and mesodermalizing (transforming) gradients still had to be shown to exist within the actual neurulating embryo. This was done by Toivonen and Saxén (29) in 1968. The archencephalic regions of the neural plate of newt embryos were dissected out and placed in various proportions with posterior dorsal mesoderm from the same stage embryos. These combinations were then used as inducers. At 10:1 and 5:1 ratios of archencephalic cells and mesoderm, the neural cells remained forebrain structures (characterized by eyes and nose). However, at 5:2 ratios, hindbrain structures were seen in the explants for the first time, and at 1:1 ratios, spinal cord was observed. In other words, the posterior mesoderm could take cells that were already committed to become neural and then specify them into more caudal cells. The more the mesodermal portion of the ratio climbed, the more caudal the neural specification.
Recent advances using the techniques of molecular biology have refined the double gradient model. The neural inducers (such as Follistatin, Chordin, and Noggin) appear to induce an anterior (archencephalic) type of neural tissue. This appears to be posteriorized by a gradient of chemical coming from the posterior end. The posteriorizing chemical has been postulated to be a fibroblast growth factor, a Wnt protein, or a retinoid (all seem to posteriorize the neural tissue, and there is evidence for a posterior-to-anterior gradient of at least retinoic acid and eFGF (see 30).
1. Nieuwkoop, P. 1952. Activation and organization of the central nervous system. III. Synthesis of a new working hypothesis. J. Exp. Zool. 120: 83- 108.
2. Mangold, O. 1933. Über die Induktonsfähigkeit der verscheidenen Bezirke der Neurula von Urodelen. Naturwiss. 21: 761-766.
3. Mangold, O. and Spemann, H. 1927. Über Induktion von Medullarplatte durch Medullarplatte im jürgeren Keim. Roux' Arch. f. Entw. mech. 111: 341-422.
3a. Holtfreter, J. 1936. Regionale Induktionen in xenoplastische zusammengestzen Explantaten. Wilhelm Roux Entwick. Org. 134: 466-561.
4. Lawrence, P. A. 1992. The Making of a Fly: The Genetics of Animal Design. Blackwell Scientific, London.
5. Mitman, G, and Fausto-Sterling, A. 1992. Whatever happened to Planaria? C. M. Child and the physiology of inheritance. In The Right Tool for the Job: At Work in Twentieth-Century Life Sciences. (ed. A. E. Clarke and J. H. Fujimura). Princeton University Press, Princeton.
6. Boveri, T. 1910. Die Potenzen der Ascaris-Blastomeren bei abgeänderter Furchung, zugleich ein Beitrag zur Frage qualitativ-ungleicher Chromosomen-Teilung. Festschrift für Richard Hertwig, vol. 3. Gustav Fischer, Jena.
7. Sander, K. 1994. Of gradients and genes: Developmental concepts of Theodor Boveri and his students. Roux' Arch. Dev. Biol. 203: 295-297.
8. Runnström, J. 1929. Über selbstdifferenzierung und Induktion bei dem Seeigelkeim. Roux' Arch. f. Entw. mech. Org. 117: 123-145.
9. Hörstadius, S. 1935. Über die Determination in Verlauf der Eiachse bei Seeigeln. Pubbl, Stn. Zool. Napoli 14: 251.
10. Gustafson, T. 1965. Morphogenetic significance of biochemical patterns in sea urchin embryos. In The Biochemistry of Animal Development I (R. Weber, ed.). Academic Press, Pp. 139-202.
11. de Beer, G. R. (1951). Embryos and Ancestors. Revised edition. Oxford University Press, Oxford.
12. Child, C. M. 1915. Individuality in Organisms. University of Chicago Press, Chicago.
13. Child, C. M. 1941. Patterns and Problems of Development. University of Chicago Press, Chicago.
14. Gilbert, S. F., Opitz, J., and Raff, R. A. 1996. Resynthesizing evolutionary and developmental biology. Developmental Biology 173: 357-372.
15. Spemann, H. 1936. Experimentelle Beitruage zu einer Theorie der Entwicklung. Verlag Julius Springer, Berlin.
16. von Ubisch, L. Entwicklungsprobleme. Gustav Fischer, Jena.
17. Wolpert, L. 1985. Gradients, position, and pattern: a history. In A History of Embryology (T. J. Horder, J. A. Wikowski, and C. C. Wylie, eds.). Cambridge University Press, Cambridge. Pp. 347-362.
18. Dalcq, A. and Pasteels, J. 1938. Potentiel morphogénètique régulation et "axial gradients" de Child. Mireau point des bases physiologiques de la morphogénèse. Bull. Acad. Méd. Belg. 6 sér. 3: 261-308.
19. Yamada, T. 1940. Beeinflussung der differenzierungsleistung des isolerten Mesoderms von Molchkeimen durch zugefügtes Chorda- und Neural-material. Okajimas Fol. Anat. Jap. 19: 131-197.
20. Lehmann, F. E. 1950. Die Morphogenese in ihrer Abhängigkeit von elementaren biologischen Konstituenten des Plasmas. Rev. Suisse Zool. 57: Suppl. 1: 141-151.
21. Hamburger, V. 1988. The Heritage of Experimental Embryology: Hans Spemann and the Organizer. Oxford University Press, Oxford.
22. Toivonen, S. 1938. Spezifische Induktionsleistungen von abnormen Induktoren im Implantatversuch. Ann. Soc. Zool.-Bot. Fenn. Vanano 6 (5):1-12.
23. Chuang, H.-H. 1938. Spezifische Induktionsleistungen von Leber und Niere im Explantationsversuch. Biol. Zbl. 58: 472-480.
24. Needham, J. 1968. Organizer phenomena after four decades: A retrospective and prospect. In Haldane and Modern Biology. (K. R. Dronamraju, ed.) Johns Hopkins University Press, Baltimore. Pp. 277-298.
25. Toivonen, S. and Saxén, L. 1955. Über die Induktion des Neuralrohrs bei Trituruskeim als simultane Leistung des Leber- und Knochen-markgewebes vom Meerschweinchen. Ann. Acad. Sci. Fenn. Ser. A, IV, 30: 1 -29.
26. Toivonen, S. and Saxén, L. 1955. The simultaneous inducing action of liver and bone marrow of the guinea pig in implanation and explantation experiments with embryos of Triturus. Exp. Cell Res. Suppl. 3: 346-357.
27. Waddington, C. H. 1940. Organisers and Genes. Cambridge University Press, Cambridge.
28. Saxén, L., Toivonen, S., and Vainio, T. 1964. Initial stimulus and subsequent interactions in embryonic induction. J. Embryol. Exp. Morphol. 12: 333-338.
29. Toivonen, S. and Saxén, L. 1968. Morphogenetic interaction of presumptive neural and mesodermal cells mixed in different ratios. Science 159: 539-540.
30. Gilbert, S. F. 1997. Developmental Biology. Fifth edition. Sinauer Associates, Sunderland, MA.