[Morphogens: experimental illusion or reality?]

The main attention is paid to the critical analysis of experimental data on morphogenetically active substances, so called "morphogens". It is proposed to consider the morphogens as specific transmitters providing for definite phases of morphogenetic tissues interactions, rather than as vectors of "morphogenetic information". In the normal development, the most studied morphogenetic tissue interactions can be referred to as so called permissive inductions, since the cells of the vertebrate embryos (amphibians, avians) are early determined for development in the ectomeso--and endodermal directions. A slow progress in studying the morphogens can be due to the following causes. 1. Theoretical "inadequacy" of the former concepts on the essence and mechanisms of embryonic induction. The necessity to develop a new system of concepts in this area of developmental biology is stressed. 2. Incompleteness of knowledge about the properties of reacting tissues and the mechanisms of action of morphogens. The early gastrula ectoderm of amphibians, most frequently used for testing the morphogens, appears to be a heterogenous population of the cells with different properties and potencies. It is, therefore, impossible to standardize strictly the biotesting of morphogens. It is suggested that the use to this end of aggregates of cell "strains" from the gastrula ectoderm, rather than of the gastrula ectoderm itself, may be more adequate 3. Insufficiency of embryonic material for biochemical identification and isolation of natural morphogens. A study of so called heterogenous inductors might be of help; these latter can be considered as analogs of natural morphogens. But the similarity of natural and heterogenous inductors can be limited only by their final effect on target tissue. The data are provided on the chemical nature, properties and mechanisms of action for a number of natural and heterogenous inductors (vegetalizing, neuralizing, mesodermalizing, lens-inducing factors). A conclusion is drawn that specific antigens do exist normally but they should not be established as a special class of "informationally important" molecules. The information necessary for development is contained in target cells and the function of a morphogen consists in providing for a definite link in the chain of processes leading to the switching on or expression of one or another programme. Only syntheses of specific proteins can, apparently, be programmed, thus reflecting the "onset" of differentiation path for a cell.