Determination, induction and pattern formation in early amphibian embryos.

Determination (inducing) factors, the extracellular matrix, signaling pathways, transcription factors and genes interact in pattern formation and neural induction. Genes can either be activated or repressed. The animalvegetal and dorso-ventral polarities are determined in very early developmental stages. Factors of the TGF-ß superfamily in a graded distribution are involved in the determination of endoderm, mesoderm and ectoderm. The differentiation of mesoderm also depends on the animal ectoderm. Neural inducing factors have been partially purified.

Mesoderm differentiation in early amphibian embryos depends on the animal cap.

Embryos of Ambystoma mexicanum from the late morula to the late blastula stage were dissected and cultivated in varying combinations. The marginal zone (presumptive mesoderm) when isolated together with the vegetal region differentiated to notochord after dissection from early blastulae, but did not differentiate to other tissues. When isolated from middle to late blastulae, in addition myoblasts and mesenchyme were formed. The marginal zone isolated together with the animal region (presumptive ectoderm) differentiated to notochord, muscle, mesenchyme, renal tubules and mesothelium irrespective of the stage of dissection. Combination of isolated animal and vegetal regions did lead to the induction of mesodermal organs. The experiments suggest that further steps in the differentiation of mesodermal organs after the induction of mesoderm by the vegetalizing factor depend on factors from the animal region, which are involved in pattern formation.

The molecular mechanism of neural induction: neural differentiation of Triturus ectoderm exposed to Hepes buffer.

Ectoderm was isolated from early gastrula stages of Triturus alpestris and cultured in salt solution buffered with either bicarbonate or Hepes as the principal buffer substance. When bicarbonate was the principal buffer substance or when bicarbonate was omitted, the isolated ectoderm formed atypical epidermis. When Hepes was added as a buffer substance, neural tissue was formed in a high percentage of cases. The differentiation of neural tissue depends on the pH of the Hepes buffer. Hepes in the protonated form, which prevails at lower pH, seems to evoke neural differentiation at a much higher rate. Hepes could either enhance the NA+/H+ antiport system or it could directly bind to plasma membrane constituents. In both cases conformational changes in the plasma membrane could generate signals which finally lead to neural differentiation.

Neural-inducing activity of nuclei and nuclear fractions from Xenopus laevis embryos.

From embryos (Xenopus laevis) of different developmental stages nuclei were isolated which exert neural inducing activity in the biological test. The active material could partly be extracted from the nuclei. Experiments for the isolation of nuclear ribonucleoprotein (RNP) particles have shown that the activity is localized at least in part in these particles. On the other hand, some neural inducer is not detached from chromatin and the nuclear matrix even with ionic detergents. Inducing activity was found in germinal vesicles and to a higher degree in the cytoplasm of oocytes, but in a masked, biologically inactive state.

Biological activity of vegetalizing and neuralizing inducing factors after binding to BAC-cellulose and CNBr-sepharose.

Covalent binding to bromoacetyl-cellulose inactivates the vegetalizing factor. The bound factor is however still able to form a complex with an inhibitor for the factor. Covalent binding to CNBr-Sepharose likewise inactivates the vegetalizing factor. The neuralizing factor on the other hand is not inactivated when covalently bound to CNBr-Sepharose. When a crude fraction which contains the neuralizng factor as well as the vegetalizing factor is bound to CNBr-Sepharose the vegetalizing activity is greatly decreased whereas the neuralizing activity is not reduced. This suggests that the mechanisms of action of the two factors are quite different. Whereas the vegetalizing factor must be incorporated into the cells, the neuralizing factor interacts with the plasma membrane of competent ectoderm cells.

Influence of cyclic nucleotides on amphibian ectoderm.

Isolated amphibian (Triturus alpestris) gastrula ectoderm was treated with cyclic nucleotides for 24 h and cultured up to 12 days. Explants treated with$cyclic N6-Monobutyryl-adenosine-3'∶5'-monophosphate, cyclic Dibutyryladenosine-3'∶5'-monophosphate and cyclic Dibutyrylguanosine-3'∶5'-monophosphate in a concentration of 10-3 and 10-5 M did not differentiate into mesoderm- or endoderm-derived tissues. The number of explants with small neural and neuroid structures did not exceed the percentage found in the control series. Inductions could also not be obtained when ectoderm was dissociated prior to the treatment with cyclic nucleotides, or when theophylline (which inhibits phosphodiesterase) was added to the culture medium. The results are discussed with regard to the possible mode of action of the vegetalizing factor.

Proteoglycans with affinity for the neuralizing factor and the vegetalizing factor (activin A homologue).

Proteoglycans from chicken embryos bind neuralizing and vegetalizing inducing factors. The proteoglycan-factor complexes have no inducing activity. Enzymatic cleavage of the core proteins of the proteoglycans abolishes inhibition of the inducing activity by proteoglycans. The possible significance of the formation of complexes of inducing factors with proteoglycans is discussed.

Mechanisms of cell differentiation: Affinity of a morphogenetic factor to DNA.

A morphogenetic factor which induces inTriturus gastrula ectoderm tissues which are derived from mesoderm and endoderm has been extracted from chicken and amphibian embryos. The factor which is protein in nature has been obtained from chicken embryos in a highly purified state.The biological activity of the chicken factor is partially inhibited when the factor is combined with chicken DNA or sonicated chicken DNA.When the 3H-labelled factor is combined with sonicated DNA and then centrifuged on a sucrose gradient the factor migrates in part with the DNA. This indicates that the factor is bound to DNA.The inferences from these results are discussed with regard to the possible mechanism of action of the factor and the molecular mechanism of differentiation.

Neural differentiation of amphibian gastrula ectoderm exposed to phorbol ester.

Ectoderm from early gastrula stages of amphibians was isolated and treated with phorbol 12-myristate 13-acetate. The ectoderm formed neural tissue and in a few cases also mesenchyme and melanophores. The control explants formed atypical epidermis. In explants treated with phorbol 12-myristate 13-acetate the mitotic rate was increased.

The activation of a neuralizing factor in the neural plate is correlated with its homoiogenetic-inducing activity.

Neural plates which are induced in the dorsal ectoderm of Triturus by the underlying mesoderm acquire, in turn, neural-inducing activity. This process is correlated with the appearance of neural-inducing activity in the microsomal fraction of the neural plate homogenate. The high-speed supernatant also acquires inducing activity after neural induction, but to a lesser extent. The experiments suggest that a masked neuralizing factor, which is already present in the ectoderm, is in part activated and exported from the inducing neural plate cells.

Basic fibroblast growth factor can induce exclusively neural tissue in Triturus ectoderm explants.

Ectoderm was isolated from early gastrulae of Triturus alpestris and induced with recombinant basic fibroblast growth factor (b-FGF). Neural tissue differentiated in about 38% of the explants which were induced by 2,5 µg/ml FGF. These explants do not contain other tissues, or contain only small amounts of mesenchyme and melanophores which are probably derived from induced neural crest. It is therefore unlikely that these neural tissues are secondarily induced. The other explants contain predominantly blastema tissue, endothelium/ mesothelium, small amounts of skeletal muscle and, rarely, notochord besides neural tissues. The mitotic rate was enhanced in about 20% of the induced explants. Possible mechanisms for the unexpected neural-inducing activity of b-FGF in Triturus ectoderm are discussed.

Partial characterization of neural-inducing factors from Xenopus gastrulae Evidence for a larger protein complex containing the factor.

High (Mr ≈ 90-110 kDa) and low (Mr ≈ 15-30 kDa) molecular weight forms of neural-inducing factors have been found in the supernatant of Xenopus gastrula homogenate. The factors, which are protein in nature, have been partially purified by size exclusion high-performance liquid chromatography (HPLC) and sodium dodecyl sulphate (SDS)-polyacrylamide gel electrophoresis. The factor of smaller size, which could be derived from a precursor, is associated with other proteins in a larger complex. The neural-inducing factors are not irreversibly inactivated after chemical deglycosylation with trifluoromethansulfonic acid. The neural-inducing protein which is found in ribonucleoprotein (RNP)-particles was partially purified by hydrophobic chromatography. Possible relationships of the factors in different subcellular fractions and their physiological significance are discussed.

Biological activity of the vegetalizing factor: Decrease after coupling to polysaccharide matrix and enzymatic recovery of active factor.

The inducing activity of the vegetalizing factor decreases after covalent coupling to CNBr-Sepharose with reduced binding capacity. The residual inducing activity is probably due to the release of a small amount of the factor from Sepharose beads. Covalent coupling to activated CH-Sepharose completely inactivated the vegetalizing factor, whereas the neuralizing factor retained its full activity. The biological activity was also very much reduced when the vegetalizing factor was bound to Sephadex beads, a derivative of dextran. Fully active factor was recovered after enzymatic degradation of the dextran matrix with dextranase. The experiments suggest that the neuralizing factor acts on the cell surface of ectoderm cells, whereas the vegetalizing factor must probably be internalized to become biologically active.

Formation of mesodermal pattern by secondary inducing interactions.

A highly purified vegetalizing factor induces endoderm preferentially in amphibian gastrula ectoderm. After combination of this factor with less pure fractions, a high percentage of trunks and tails with notochord and somites are induced. The induction of these mesodermal tissues depends on secondary factors which may act on plasma membrane receptors of the target cells. The secondary factors are probably proteins as they are inactivated by trypsin or cellulose-bound proteinase K. They are not inactivated by thioglycolic acid.The implication of these findings for tissue determination and differentiation in normal development in relation to the anlageplan for endoderm and mesodermal tissues is discussed.

[Action of sulfhydryl compounds on embryonic inducing factors]. / Wirkung von Sulfhydrylverbindungen auf embryonale Induktionsfaktoren.

Crude extracts from 9 days old chicken embryos containing neuralizing and mesodermalizing inducing factors as well as purified mesodermalizing factor were incubated with thioglycolic acid and with 2-Mercaptoethanol. The fractions were tested by implanting into early gastrulae ofTriturus orAmbystoma. The mesodermalizing factor is inactivated whereas the neuralizing factor does not lose its activity.

Activation of a neuralizing factor in amphibian ectoderm.

Isolated gastrula ectoderm has no neural-inducing activity and does not differentiate into neural tissues. It has, however, a high neural-inducing capacity, but the inducing factors are present in a masked, inactive form. The inducing factors are partially activated by homogenization and by freezing of the homogenate and are fully activated by treatment with ethanol. The relative distribution of inducing factors in different subcellular fractions changes after treatment with demecolcine and cytochalasin B or after autolytic incubation of the homogenate. The inducing activity of the high-speed supernatant is enhanced under these conditions. The experiments suggest that the activation of neuralizing factor(s) depends on the release from complex structures. Cytoskeletal elements seem to be involved. When early neural plate homogenate was fractionated, the high-speed supernatant showed neural-inducing activity. This is in contrast to the high-speed supernatant from the ectoderm homogenate, which shows no such activity.

Isolation of plasma membranes from Xenopus embryos.

Plasma membranes were isolated in high yield from Xenopus gastrulae by repeated sedimentation in discontinuous sucrose gradients. Most of the yolk was separated by lowspeed sedimentation before centrifugation on the discontinuous sucrose gradients. The isolation of plasma membranes was followed by covalent labelling of the surface of dissociated gastrula cells with diazoniobenzene sulphonate, by electron microscopy and the distribution of enzymatic markers. The isolated plasma membranes have a low neural inducing activity as compared to other cell constituents.

Neural induction in amphibians : Transmission of a neuralizing factor.

The neural plates of very early neurula stages of Triturus alpestris were removed, the material which is released from the extracellular space between mesoderm and neural plate to the medium in which the embryos were dissected was isolated and extracted with phenol. The protein isolated from the phenol layer showed neural inducting activity. Proteoglycans isolated from the aqueous layer did not show such inducing activity. These results together with previously published experiments (Wilhelm Roux's Arch 184: 285-299) suggest that a neuralizing factor which is released from the mesoderm acts on the inner surface of the overlying dorsal ectoderm.

Inducing activity of subcellular fractions from amphibian embryos.

The homogenate from unfertilized eggs, gastrulae, neurulae and hatched embryos ofXenopus laevis was fractionated by differential centrifugation and subsequent repeated centrifugation on discontinuous sucrose gradients. A high archencephalic-neural inducing activity was found in RNP particles, which were released from the high-speed ("microsomal") sediment by treatment with EDTA, and in a fraction of heterogeneous small vesicles. The highest archencephalic inducing activity was observed in RNP particles from unfertilized eggs and from gastrulae. RNP particles isolated from hatched embryos had a lower inducing activity. The neuralizing factor can be extracted from the small vesicles with pyrophosphate buffer at pH 8.6, but it is not solubilized with a non-ionic detergent (Triton X 100). The high-speed supernatant from the gastrula homogenate contains soluble neuralizing factor, whereas the supernatant from egg homogenate has a low inducing activity. The plasma membrane fraction (isolated from gastrulae) also has only a low inducing activity. The possible significance of the subcellular distribution of neuralizing factors for the transmission of neuralizing inducer from the mesoderm to competent gastrula ectoderm and the processing of signals which are generated on the plasma membrane of induced cells is discussed.

A mesoderm-inducing factor from a Xenopus laevis cell line : Chemical properties and relation to the vegetalizing factor from chicken embryos.

We have compared the chemical properties and biological activities of the mesoderm-inducing factor that is secreted by the Xenopus XTC cell line with the vegetalizing factor from chicken embryos. The inducing activity of the factors was tested in different concentrations on totipotent ectoderm either by implantation into early gastrulae of Triturm alpestris or by application of solutions to isolated ectoderm of early gastrulae of Xenopus laevis. Both factors have similar properties. They are not irreversibly inactivated after treatment with 6 M urea or with phenol at 60° C. Reduction with thioglycolic acid inactivates the factors completely. The inducing activity of XTC-conditioned medium decreases only slightly after treatment with 50% formic acid. The apparent molecular mass and the isoelectric point of the factors are similar. The XTC factor was partially purified by size-exclusion and reversed-phase high-pressure liquid chromatography and by isoelectric focusing. The possible relationship of these factors to transforming growth factor ß is discussed.