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.
Curr Biol
2004 Nov 09;1421:1916-23. doi: 10.1016/j.cub.2004.10.020.
Show Gene links
Show Anatomy links
Visualizing long-range movement of the morphogen Xnr2 in the Xenopus embryo.
Williams PH
,
Hagemann A
,
González-Gaitán M
,
Smith JC
.
???displayArticle.abstract???
One way in which cells acquire positional information during embryonic development is by measuring the local concentration of a signaling factor, or morphogen, that is secreted by an organizing center . The ways in which morphogen gradients are established, particularly in vertebrates, remain obscure, although various suggestions have been made for the mechanisms by which signaling molecules traverse fields of cells. These include simple diffusion, "cytonemes", filopodia, "argosomes", and "transcytosis". In this study, we use a functional EGFP-tagged ligand to visualize long-range signaling in the Xenopus embryo in real time. Our results show that the TGF-beta family member Xnr2 is secreted efficiently from embryonic cells, and a new method of tissue recombination allows us to investigate the way in which the morphogen traverses multiple cell diameters. This reveals that Xnr2 exerts long-range effects by diffusing rapidly through the extracellular milieu of nonexpressing cells. No evidence has been obtained for long-range signaling through cytonemes, filopodia, argosomes, or transcytosis. In demonstrating that long-range signaling in the early Xenopus embryo occurs by diffusion rather than by these alternative routes, our results suggest that different morphogens in different developmental contexts use different means of transport.
Figure 2. Tagged Xnr2 Constructs Retain Inducing Activity
(A) Schematic diagram showing tagged versions of Xnr2. EGFP was cloned between the endoprotease signal sequence (RPRR) and the
mature ligand, and a single HA tag was added immediately before the stop codon of Xnr2.
(B) The specific activity of EGFP-Xnr2 resembles that of native Xnr2. Xenopus embryos were injected at the 1 cell stage with the indicated
amounts of RNA and were then cultured to mid-blastula stage 8.5. Animal pole regions were isolated, cultured for a further 3 hr, and frozen.
Expression of Xbra, Derrie` re, and Goosecoid was analyzed by real-time RT-PCR with the LightCycler (Roche), and figures were normalized
to levels of ODC.
(C) Embryonic development is disrupted by similar concentrations of Xnr2 and EGFP-Xnr2. Embryos at the 1 cell stage were left uninjected
(C) or were injected with 2.5 pg RNA encoding Xnr2 (D) or 4 pg RNA encoding EGFP-Xnr2 ([E]; the molar equivalent of 2.5 pg Xnr2). In both
(D) and (E), ectodermal tissue is driven toward mesendodermal fates, leading to disruption of gastrulation.
Figure 3. Xnr2 Is Secreted from Producing Cells and Activates Target Genes over a Long Range
(A) Embryos injected with RNA (200 pg) encoding the indicated tagged versions of Xnr2 were allowed to develop to the early gastrula stage,
when they were treated as described below. Extracts were subjected to Western blotting with an anti-HA antibody. Slower-migrating bands
represent unprocessed forms of the Xnr2 proteins. Green arrows indicate mature processed EGFP-Xnr2-HA, and red arrows indicate mature
processed Xnr2-HA. (Lanes 1) Control embryos (lane 1) or injected embryos (lanes 2) were frozen immediately at early gastrula stage 10
(lanes 2 and 7), opened at the animal pole (without discarding any tissue) and cultured for 30 min before being frozen (lanes 3 and 6), or
opened and cultured in 250 g/ml porcine trypsin for 30 min before being frozen (lanes 4 and 7). Most secreted mature protein is lost after
opening the embryos (lanes 3 and 6), and what remains is further diminished by trypsin treatment (lanes 4 and 7). (Lanes 83) To ensure
that secreted mature ligand (lanes 3 and 6) was not degraded by proteases in the embryo culture medium, blastocoelic fluid was removed
from ten intact injected embryos and subjected to Western blotting. Extracts of whole embryos (lanes 9 and 10; 0.5 embryo equivalent) show
low levels of processed mature ligand. Secreted mature ligand is greatly enriched in blastocoelic fluid (lanes 12 and 13).
(B) Xnr2 and EGFP-Xnr2 activate Xbra at long range in animal cap conjugates. 100 ng FLDx (B), 100 ng FLDx 10 pg Xnr2 (C), 100 ng
FLDx 16 pg EGFP-Xnr2 (D), or 100 ng FLDx 300 pg constitutively active ALK4 (ALK4*) were injected into Xenopus embryos at the 1 cell
stage. Animal caps were dissected and juxtaposed with uninjected animal caps. They were cultured for 3 hr, fixed, and processed by in situ
hybridization for Xbra expression (blue) and by immunocytochemistry for FLDx (red). Upregulation of Xbra was never observed in control
conjugates ([B]; n 29), whereas both Xnr2 ([C]; 20/26) and EGFP-Xnr2 ([D]; 18/26) can upregulate Xbra in cells distant from the source. In
a separate experiment, constitutively active ALK4 was found to act in a strict cell-autonomous fashion in all cases ([E]; n 8). Xnr2 exerted
long-range effects in five out of eight cases in this experiment, whereas EGFP-Xnr2 behaved cell-autonomously, perhaps because insufficient
RNA was injected.
(F) Xnr2 and EGFP-Xnr2 activate Xbra at long range after injection of RNA into single blastomeres at the 64 cell stage of intact Xenopus
embryos. 2 ng FLDx (F), 2 ng FLDx 30 pg Xnr2 (C), 2 ng FLDx 40 pg EGFP-Xnr2 (D), or 2 ng FLDx 30 pg constitutively active ALK4
(ALK4*) were injected into single blastomeres of Xenopus embryos at the 64 cell stage. Embryos were allowed to develop to early gastrula
stage 10.5 and were then processed by in situ hybridization for Xbra expression (blue) and by immunocytochemistry for FLDx (red). Upregulation
of Xbra was never observed in control embryos ([F]; n 14), whereas both Xnr2 ([G]; 14/22) and EGFP-Xnr2 ([H], arrow; 7/16) cause upregulation
of Xbra in cells distant from the source. Circles of Xbra expression, as shown in (G), were observed only once in response to EGFP-Xnr2 (H),
suggesting that this tagged molecule has a slightly more restricted range than the parent protein. Constitutively active ALK4 was unable to
activate nonautonomous expression of Xbra ([I]; n 16, of which 13 showed exogenous Xbra expression).
Figure 4. Neither Argosomal Transport Nor Transcytosis Is Responsible for the Long-Range Effects of EGFP-Xnr2
(A) Schematic diagram showing two animal caps juxtaposed to allow observation of tagged proteins in real time. Animal caps from embryos
injected with the desired RNAs are dissected at the late blastula stage and juxtaposed on a fibronectin-coated coverslip.
(A) Long-range signaling occurs in this experimental setup as it does in the experiments described in Figure 3. Animal caps derived from
embryos injected with 400 pg EGFP-Xnr2 and 2 ng FLDx were juxtaposed as described in (A) and cultured for 4 hr before being processed
by in situ hybridization for Xbra expression (blue) and by immunocytochemistry for FLDx (red). Long-range upregulation of Xbra was observed
in all 15 conjugates examined. No Xbra expression was observed in control conjugates lacking EGFP-Xnr2 (n 14; data not shown), and
long-range signaling was not observed in experiments in which constitutively active ALK4 replaced EGFP-Xnr2, although in one case (out of
14), Xbra was activated in cells immediately adjacent to the FLDx-labeled tissue (data not shown). In the images that follow, the passage of
tagged ligand from one animal cap to another is followed by confocal microscopy. With the exception of (B) below, all images represent
optical sections taken at the level of the nucleus of the cell adjacent to the coverslip. The field of view in (C)C″) and (D)D″) resembles that
outlined in (A).
(B) Z series projection of cells labeled with the membrane marker CFP-GPI. Blastomeres extend projections into their immediate environment,
but none extend further than a single cell diameter. The scale bar represents 20 m.
(C″) EGFP-Xnr2 does not exert long-range effects with argosomes. RNA (400 pg) encoding EGFP-Xnr2 (C″) was coinjected into Xenopus
embryos with RNA (500 pg) encoding the membrane marker CFP-GPI (C). Animal caps from such embryos were juxtaposed with uninjected
caps. EGFP-Xnr2 in receiving caps was not surrounded by CFP-GPI-labeled membrane but is present in abundance in the extracellular space.
(C) is a merged image of (C) and (C″). The scale bars represent 20 m.
(D″) EGFP-Xnr2 does not exert long-range effects by transcytosis. Animal caps expressing EGFP-Xnr2 (D″) were juxtaposed with caps
expressing CFP-GPI (D). EGFP-Xnr2 in the receiving cap is extracellular and is not detected in CFP-GPI-bounded cells or vesicles. (D) is a
merged image of (D) and (D″). The scale bars represent 20 m. Amounts of injected RNA are as described in (C)C″).