Vertical and Horizontal Signals from the Organizer

Our discussion of the organizer so far has discussed those signals that go vertically from the organizer to the overlying ectoderm. We now know that there is a second set of signals that is produced by the dorsal blastopore lip and sent horizontally through the plane of the ectoderm (Figure 1). Recent evidence suggests that both vertical induction through the chordamesoderm and horizontal (planar) induction through the ectoderm are necessary for complete embryonic induction.

Figure 1 Two modes of inducing the dorsal axis. In the planar mechanism, molecules are transferred from the dorsal blastopore lip tissue through the plane of the ectoderm. In the vertical mechanism, soluble molecules from the dorsal-blastopore-lip-derived chordamesoderm induce the cells above them to become neural tissue. (After Doniach, 1993.)

What roles might these signals play? First, there is some evidence that planar signals may be involved in the neuralizing activities of the organizer. Perhaps they provide the missing signals that activate neurulation (as opposed to the signals that block ventralization). If planar signals traveling from the dorsal blastopore lip through the ectoderm are responsible for neural induction, then the original source of such signals should be the epithelium of the dorsal marginal zone rather than the deep mesenchyme cells of that organizer zone. Shih and Keller (1992) have found this to be the case. They repeated the Spemann and Mangold experiment, but instead of using the entire dorsal marginal zone (DMZ), they transplanted either the epithelial cells or the deep cells of the DMZ (which they labeled with fluorescent dextran particles). The epithelial cells had all the inductive properties of the Spemann organizer and differentiated into mesodermal tissue. This epithelium was also able to rescue embryos that had been "ventralized" by UV irradiation. Neither the deep cells of the DMZ nor the ventral marginal cells could accomplish such organizer activities. A recently discovered inducer protein, Xenopus nodal-related-3 (Xnr3) has been found in this superficial layer of the organizer, and it can convert animal pole caps into anterior neural ectoderm. Interestingly, unlike other inducers, it does not dorsalize the mesoderm. It is not yet known if this protein is part of the planar signaling system (Hansen et al., 1997).

However, the planar signals are not seen to be sufficient for neural induction. Nieuwkoop and Koster (1995) prevented vertical induction from occurring during Xenopus gastrulation and found that no neural differentiation took place. Furthermore, if the fibronectin binding fragment, RGD, is injected into the blastocoel of Rana pipiens gastrulae, the axial mesoderm fails to migrate toward the animal pole. Rather, it splits into two streams that involute horizontally along the equator of the embryo, forming two laterally located notochords. Each notochord induces a neural plate, but a neural plate does not form in the dorsal ectoderm, where the planar signals would have spread (Saint-Jeannet and Dawid, 1994). So in this model, the planar signals are redundant or may support the vertical signals from the notochord.

In the second model, the planar signals may be important in contributing to the regional specificity of induction. Doniach and her colleagues (1992) showed that instructive, positionally specific information is provided by planar signals passing through the ectoderm. When explants are taken from early Xenopus gastrulae such that the ectoderm retains contact with the dorsal blastopore lip but never contacts the mesoderm, not only are the pan-neural markers NCAM and NF-3 induced in the ectoderm, but four position-specific neural markers–engrailed-2, Krox-20, XlHbox1, and XlHbox6–are expressed in the explant ectoderm in the appropriate anterior- posterior sequence (Figure 2). It appears, then, that the horizontally inductive signals from the dorsal blastopore lip are sufficient for inducing the anterior-posterior neural pattern. Ruiz i Altaba (1992) has also confirmed extensive neural patterning in these exogastrulae, showing that the pattern of neural markers in the exogastrulae reflects the normal pattern except in the forebrain and ventral regions. He also provides evidence that the transmission of these horizontal signals is through the notoplate (the ectoderm above the notchord). (Ruiz i Altaba, 1990, 1992).

Figure 2 Expression pattern of neural markers induced by contact with the dorsal blastopore lip in plane of ectoderm. (A) Sagittal section of early Xenopus gastrula showing where cuts were made. (B) Explant depicting the anterior-posterior polarity expected from fate map: white region is epidermis; stippled region is presumptive mesenchyme; black region is dorsal mesoderm; striped region is archenteron roof. The explants were placed under coverslips to prevent migration of the mesoderm. (C) Expression of neural markers in the control embryo, stage 21. Homeobox genes engrailed-2 and XlHbox6 are expressed at the hindbrain-midbrain border and in the spinal cord, respectively; zinc finger protein gene Krox-20 is expressed in rhombomeres 3 and 5 of the hindbrain. (D) The same order of expression is seen in the ectoderm of those explants retaining a connection to the dorsal blastopore lip. (After Doniach et al., 1992.)

However, more recent studies on both urodele and anuran embryos (Chen et al., 2000) claim that planar signalling is not sufficient to generate the anterior-posterior pattern of neural markers. This group prepared their exogastrulae in a manner that mechanically excised the ectoderm from the mesoderm and endoderm at the beginning of gastrulation, leaving only a small connection between the dorsal mesoderm and the ectoderm. They then covered these exogastrulae with filter paper to prevent their folding back on themselves and, in so doing, providing vertical signals. In these instances, the only neural induction that was seen was in the bridge between the ectoderm and endomesoderm.

In the third model, the planar signals complement the vertical signals in the creation of the neural tube. The planar signals appear to be involved in inducing the convergent extension of the hindbrain and spinal cord ectoderm adjacent to it (while the folding of the neural plate into a neural tube appears to be induced by the notochord). (Keller et al., 1992; Nieuwkoop and Koster, 1995). We are still trying to fit all the pieces of the induction puzzle together, while new pieces are being discovered. Spemann predicted that scientists would find that the embryo used more than one mechanism ("double assurance") to accomplish its ends. The embryo may well be using both planar and vertical signals to induce its nervous system.

Literature Cited

Chen, Y., Hollemann, T., Pieler, T., and Grunz, H. 2000. Planar signalling is not sufficient to generate a specific anterior/posterior neural pattern in pseudoexogastrulae explants from Xenopus and Triturus. Mech. Devel. 90: 53-63.

Ruiz i Altaba, A. 1990. Neural expression of the Xenopus homeobox gene Xhox3: evidence for a patterning neural signal that spreads through the ectoderm. Development 108:595-604.

© All the material on this website is protected by copyright. It may not be reproduced in any form without permission from the copyright holder.

HOME :: CHAPTER 10

PREVIOUS :: NEXT