The Sandwich Explant of the Dorsal Marginal Zone and Prospective Neural Tissue

This explant is made by sandwiching the inner deep surfaces of two dorsal sectors, taken from two individual gastrulae, together (Fig. 6). Both the invo luting (PM/endodermal) and noninvoluting (prospective neural) regions of this explant converge and extend, whereas the animal-most ectodermal region, corresponding to the prospective mid- and forebrain, remains bulbous (52,66,67) (Fig. 6). Since Organizer tissues of this explant are in edgewise or planar apposition to the ectoderm, it is assumed the planar signals must induce the ectoderm to acquire a neural fate. However, some thought on the requirements for demonstrating planar induction suggests that the interpretation of the behavior of this sandwich explant is more complex than might first appear.

4.2.1. Is the Assayed Behavior Already Autonomous When the Test for Induction Is Done?

The neural convergent extension or marker expression in sandwich explants of early gastrulae is not necessarily induced by planar signals from the Organizer. They may have been induced previously or patterned by cytoplasmic localizations, and may be autonomous behaviors by gastrulation. Testing for autonomous expression at the early gastrula is not a trivial problem, particularly in the case of the SC and hindbrain, since these regions are short and difficult to manipulate (52) .

4.2.2. Do Contaminating Cells Provide Vertical Signaling?

The second problem is that mesodermal cells may contaminate the interior of the noninvoluting, prospective neural tissue, providing vertical signals in a preparation supposedly devoid of vertical signals. In sandwiches made at stage 10 to 10+, the leading edge of the mesodermal mantle likely has begun involution, as described in Subheading 1.3.2. and Fig. 2. If these mesoderm cells are not removed from the inner surface of the explant, they have access to the fibronectin-rich inner surface of the blastocoel roof. Under these conditions, these cells are highly invasive (41) and will invade the core of sandwich and provide vertical signals (Fig. 7A). If sandwich explants are made early, at stage 10-, PM will be found at the vegetal end of the explant. These PM cells can also migrate animally beneath the neural tissue, both as individuals and in streams, and even lead the more posterior cells in an invasion of the core of the explant (Fig. 7B) (see also Fig. 10 in ref. 9). As few as four or five notochordal cells can change the expression of neural genes in the overlying ectoderm (see ref. 80).

There are several methods one can use to solve the problem of invasion. The best method, of course, is to remove all the potential invading cells from the inner surface of the vegetal end of the explant, as well as adherent cells higher in the marginal zone. They are easy to identify (Fig. 7A; see also Fig. 10 in ref. 9). One could also monitor such invasions with markers expressed by the potentially invading tissue. However, one must know what markers this tissue should express, and the test for marker expression must be sufficiently sensitive at the appropriate stage.

Fig. 7. Time-lapse video microscopy of the inner surface of an open-faced explant shows the tendency of mesodermal cells to migrate animally on the inner surface of the prospective neural tissue. As the bulk of the mesoderm converges and extends vegetally (arrows, A), a number of individual mesodermal cells break off from the posterior edge of the mesoderm and migrate on the inner surface of the prospective neural ectoderm (pointers, A). In some cases, large tongues of mesodermal cells may migrate animally (white arrows, B) while the rest of the mesoderm converges (C) and extends (E) vegetally (black arrows, B). Photos courtesy of John Shih.

Fig. 7. Time-lapse video microscopy of the inner surface of an open-faced explant shows the tendency of mesodermal cells to migrate animally on the inner surface of the prospective neural tissue. As the bulk of the mesoderm converges and extends vegetally (arrows, A), a number of individual mesodermal cells break off from the posterior edge of the mesoderm and migrate on the inner surface of the prospective neural ectoderm (pointers, A). In some cases, large tongues of mesodermal cells may migrate animally (white arrows, B) while the rest of the mesoderm converges (C) and extends (E) vegetally (black arrows, B). Photos courtesy of John Shih.

A better way to document the absence of such invasion and to demonstrate convincingly that only planar induction has occurred is to appose an Organizer from a fluorescently labeled embryo next to an unlabeled sandwich of ectoderm that, in isolation, does not show neural development. Any labeled cells from the inducing tissue that invade, regardless of phenotype, will be noticed (see Fig. 7C in ref. 52).

Finally, a positive control for the effect of contamination can be done. Contamination is allowed, both a little of it and a lot of it, to determine if it makes a difference in neural convergent extension or marker expression, relative to cases thought to have no contamination. For example, neither a little nor a lot of contamination increased the amount of convergence and extension of the neural region, and thus we concluded that vertical signals had no detectable effect on convergent extension induced by planar signals alone (52).

Was this article helpful?

0 0

Post a comment