Click here to close
Hello! We notice that you are using Internet Explorer, which is not supported by Xenbase and may cause the site to display incorrectly.
We suggest using a current version of Chrome,
FireFox, or Safari.
Activin has direct long-range signalling activity and can form a concentration gradient by diffusion.
McDowell N
,
Zorn AM
,
Crease DJ
,
Gurdon JB
.
???displayArticle.abstract???
BACKGROUND: Activin has strong mesoderm-inducing properties in the early Xenopus embryo, and has a long-range signalling activity that activates genes in cells distant from a source in a concentration-dependent way. It has not yet been established what mechanism of signal transmission accounts for this and other examples of long-range signalling in vertebrates. Nor is it known whether activin itself acts on distant cells or whether other kinds of molecules are used for long-range signalling. Here we have used a well characterised model system, involving animal caps of Xenopus blastulae treated with activin or transforming growth factor beta, to analyze some fundamental properties of long-range signalling and of the formation of a morphogen gradient.
RESULTS: We find that cells distant from the source of activin require functional activin receptors to activate Xbrachyury, a result suggesting that activin itself acts directly on distant cells and that other secondary signalling molecules are not required. We also find that the signals can be transmitted across a tissue that cannot respond to it; this argues against a relay process. We provide direct evidence that labelled activin forms a concentration gradient emanating from its source and extending to the distant cells that express Xbrachyury. Lastly, we show that there is no inherent polarity in the responding tissue that influences either the direction or rate of signalling.
CONCLUSIONS: The long-range signalling mechanism by which activin initiates the transcription of genes in a concentration-dependent manner depends on a process of rapid diffusion and the establishment of an activin gradient across the tissue. It cannot be explained by a relay or wave propagation mechanism. Activin itself is the signalling molecule to which distant cells respond.
Figure 1.
Activin does not induce Xbra via secondary signalling. Two types of directional conjugate were constructed (see text). (a) A decoated animal cap of RLDx-labelled, wild-type cells separates the ActRIIBdn cap (above) from the activin beads. The lower wild-type cap supporting the conjugate acts as a positive control. (b) No Xbra band is seen in the upper ActRIIBdn cells although scattered staining can be seen at the distal edge of the decoated cells and in the bottom control cap. (c) RLDx labelling distinguishes the wild-type decoated cells from the dominant-negative cells. In none of the samples was an Xbra band seen in the body of the ActRIIBdn animal cap (total 47). (d) Experimental design for ActRIIBexd: the activin signal passes from the beads into the RLDx-labelled animal caps. Following bead removal, one of the activin-influenced caps is then placed next to either an ActRIIBexd or a wild-type cap. (e,g) An Xbra band has been induced in the cells of the wild-type cap (12 out of 17 conjugates with 5 beads of either 40 or 16 nM). (f,h) No Xbra band, however, is present in the ActRIIBexd cells (0 out of 18 conjugates). The black lines demarcate the boundaries.
Figure 2.
TGFβ2 can act as a morphogen and pass across tissue composed of cells unable to respond to it. (a–c) TGFβ2 can act as a morphogen. (a) Two TGFβ2 beads were sandwiched between an animal cap supplied with TGFβIIR and a non-responsive (wild-type), rhodamine-labelled animal cap. (b) With weak beads (1 nM), the band in the receptor injected cells is in the vicinity of the beads (5 out of 5 conjugates), while stronger beads (20 nM) induce a band further away (5 out of 5 conjugates) (c). The black lines demarcate the boundary between the receptor injected and RLDx cells (not shown). (d) A modified version of the three-layer conjugate, described previously in Figure 1a. The middle decoated layer (RLDx) is wild-type and hence cannot respond to the TGFβ2 signal from the beads, while cells of the upper animal cap can respond since they contain TGFβIIR. (e) A strong band of Xbra expression is induced in the upper responding cap (26 conjugates out of 26). (f) Rhodamine labelling demonstrates that the wild-type decoated tissue forms a continuous layer separating the receptor injected cells from the beads.
Figure 3.
35S-labelled activin protein can spread from beads into tissue. (a) Synthetic mRNA encoding activin βB was micro-injected into Xenopus oocytes that were cultured for 3 days in the presence of 35S-methionine and 35S-cysteine. The labelled proteins in the conditioned oocyte medium were resolved by 15% SDS-PAGE in reducing conditions and visualised by fluorography. Oocytes injected with activin mRNA (+) contain both the mature processed activin (∼14 kDa) as well as a larger (∼40 kDa) secreted form with a size consistent with that of unprocessed activin. (b) Affigel beads were loaded with labelled activin oocyte medium and implanted into animal cap conjugates. After 4 h of culturing, the beads were recovered from the tissue. The labelled proteins present in the tissue or still bound to the beads were resolved by 15% SDS-PAGE in reducing conditions and visualised by fluorography. Lane 1 shows protein bound to beads before implantation. Lane 2 shows the proteins remaining on the beads after removal from tissue and lane 3 shows the activin protein present in the tissue
Figure 4.
Visualisation of an activin morphogen gradient in tissue. Between 10 and 40 pg (1000–4000 cpm per bead; see Materials and methods) of 35S-labelled activin βB from oocyte medium was loaded on to Affigel beads, the beads were implanted in animal cap conjugates and after 4 h of culturing conjugates were fixed, sectioned and subjected to autoradiography. (a,b) Low magnification of an activin bead in a conjugate; (a) dark field view; (b) bright field view; the white area in (a) is a dark field view of autoradiographic grains. (c) A dark field view of (a) at higher magnification. (d) Hoechst nuclear stain; the dotted circle indicates the position of the bead. (e) Bright field view of a section through a conjugate in which the bead was dislodged before autoradiography. (f) Bright field view of an activin-bead containing conjugate, in which the section shown has passed ∼25 μm above the level of the edge of the bead. (g) Dark field view of a conjugate with two beads loaded with 35S-methionine and 35S-cysteine only (no activin); the flat morphology shows no response to activin. (h) High magnification of a bright field autoradiograph shows a concentration gradient. (i) NIH image software was used to quantitate the average silver grain density over sections of tissue. The results were plotted relative to distance from the edge of the bead position (referred to as tissue edge). The graph depicts the concentration gradient of activin across the tissue.
Figure 5.
The polarity of responding tissue does not affect direction or rate of signalling. (a,b) The design of the experiment: the outer impermeable surface coat was removed from an animal cap and the tissue placed in either an inverted or normal orientation. Cells expressing Xbra RNA are the same distance from an activin bead, whether the upper decoated cap is in a normal orientation (c) or 180 degrees inverted (d). The lower animal cap has a normal surface coat (pigmented) and has been injected with rhodamine lysinated dextran (RLDx). The black line (c,d) demarcates the boundary between decoated and rhodamine labelled cells (e,f).
