XB-ART-14768Development April 1, 1998; 125 (8): 1371-80.
A Xenopus homologue of aml-1 reveals unexpected patterning mechanisms leading to the formation of embryonic blood.
The Runt domain gene AML1 is essential for definitive hematopoiesis during murine embryogenesis. We have isolated Xaml, a Xenopus AML1 homologue in order to investigate the patterning mechanisms responsible for the generation of hematopoietic precursors. Xaml is expressed early in the developing ventral blood island in a pattern that anticipates that of later globin. Analysis of globin and Xaml expression in explants, in embryos with perturbed dorsal ventral patterning, and by lineage tracing indicates that the formation of the ventral blood island is more complex than previously thought and involves contributions from both dorsal and ventral tissues. A truncated Xaml protein interferes with primitive hematopoiesis. Based on these results, we propose that Runt domain proteins function in the specification of hematopoietic stem cells in vertebrate embryos.
PubMed ID: 9502719
Article link: Development
Genes referenced: agtr1 h4c4 runx1
Article Images: [+] show captions
|Fig. 1. XAML, CBFa2 and XRD sequence alignment. The predicted amino acid sequence for XAML (GenBank accesion number AF035446) is approximately 90% identical to CBFa2. Insertion/deletions and regions of the proteins which cannot be aligned are lower case. The darker shading indicates a high pair-mean score (obtained by the Macaw sequence alignment program).|
|Fig. 2. Expression of Xaml in Xenopus embryogenesis. Xaml (A-G,I) and a- globin (H) expression patterns, in all panels anterior is to the right. Bracket in D-F, 400 mm. Arrows indicate Xaml-expressing cells (C,I). (A) RTPCR of Xaml expression at different embryonic stages. (B) Dorsal view of Xaml expression in neuroblasts at stage 17. (C) Sagittal section through anterior endomesoderm of a stage 14 embryo, Xaml is in cells of the endomesoderm which make direct contact with the ectodermal layer, Bar, 20 mm. (D) Ventral view of a stage 14 embryo. (E) Ventral view of embryo shown in B; note that Xaml expression has expanded posteriorly. (F) Detail of EA Xaml expression at stage 22. At this stage Xaml expression refines into a V-shape, but some Xaml cells remain anterior to the VBI (right arrow). (G) Ventral Xaml expression in a stage 26 embryo extends to the proctodeum (arrow). (H) a-globin expression in a stage 26 embryo, has not yet reached the proctodeum (arrow) demonstrating that Xaml expression precedes a-globin in the VBI. (I) Transverse section of a stage 26 embryo at approximately 30% embryonic length. Xaml expression is seen in isolated cells of the lateral plate mesoderm, top panel is a high magnification view of the region boxed in the lower panel. Abbreviations: N, notochord, nt, neural tube, o, olfactory placode.|
|Fig. 3. Schematic representation of gastrulation in Xenopus. Presumptive EA Xamlexpressing cells are colored magenta. The embryo shown on the left is a stage 10 embryo; the brackets indicate the region of the embryo excised in explant experiments. The middle embryo shows a mid-gastrulation stage embryo. The embryo on the right is at a similar stage to the earliest that we detect Xaml by wholemount in situ hybridization. Abbreviations: A, archenteron, B, blastocoel cavity; BC, bottle cells; D, dorsal; V, ventral. (Adapted from Keller, 1991).|
|Fig. 4. Xaml in situ hybridization in dorsal and ventral explants. (A) Dorsal explant at control stage 15-17 showing positive Xaml staining. The position of Xaml expression is in a ventral location relative to the head folds, similar to the position of EA Xaml in a whole embryo. (B) Ventral explant at control stage 15-17 with lack of Xaml staining. (C) Dorsal explant at control stage 25-27 with lack of Xaml staining. (D) Ventral explant at control stage 25-27; note intense Xaml staining at the ‘anterior’ tip (pointing down) of the explant, this location of staining is similar to that seen with an a-globin probe (data not shown).|
|Fig. 5. Comparison of the effects of UV ventralization and LiCl on Xaml (A-D) expression and a-globin (D-G) expression. Arrowheads indicate untreated controls in F and G. The bracket indicates the region of a DAI=8 embryo where Xaml is increased but a-globin is absent (A,D,E). (A) LiCl-treated embryos at control stage 27 following in situ for Xaml. The top two embryos are DAI=8 the bottom two embryos are DAI=9. (B)Ventral view of an embryo probed for Xaml expression after UV treatment (left embryo); the arrow indicates the region of the UV embryo where Xaml expression is missing. A wild-type control embryo is shown on the right. (C) Dorsal view of embryos shown in B (see text). (D) LiCl embryos at control stage 27 (DAI=8) following whole-mount in situ hybridization for Xaml (right embryo) or a-globin (left embryo). (E) LiCl-treated embryos at control stage 27 following aT1 globin in situ. The top two embryos have DAI=8; the bottom two embryos have DAI 9. (F) a-globin expression in UV ventralized embryos at control stage 26 (G) a-globin expression in UV ventralized embryos at control stage 28-29. Abbreviations: o, olfactory placode; WT, untreated control.|
|Fig. 6. a-globin expression in embryos lineage traced with lacZ following in situ hybridization with an aT1 globin probe. b-gal staining gives a blue precipitate while a-globin staining is purple. (A) Embryos with a b-gal staining pattern typical of dorsally injected pigmented controls. Arrowhead indicates a-globin expression. (B) Embryos with b-gal staining typically seen in ventrally injected pigmented controls. Arrowhead indicates a-globin expression. (C) Ventral view of an embryo treated as in A. Note the significant overlap of the b-gal staining and the a-globin staining. (D) Ventral view of embryo shown in B.|
|Fig. 7. Benzidine staining of larvae (stage 45-50) injected in the VMZ, arrows point to the heart. (A) Control injected larva. (B) Larva injected with full-length Xaml mRNA. (C) Larva injected with Xrd mRNA. (D) Northern blot of RNA harvested from stage 38 embryos probed with a-globin and histone H4 (globin is reduced 26-fold in the bloodless embryos). Note that this is an earlier stage than is shown in A, B or C.|
|Fig. 8. Embryonic a-globin staining in embryos injected with XRD+lacZ mRNA (BH) or lacZ alone(A). Blue staining indicates cells that received the lacZ message, purple staining indicates cells expressing a-globin. (A) embryo injected ventrally with lacZ alone, note the significant overlap of blue and purple staining. (B) Embryo injected with XRD+lacZ and processed simultaneously with the embryos in C-H; note the normal a- globin expression in the VBI and the lack of blue staining in the VBI. (C,D) Embryos injected with lacZ staining overlapping the posterior VBI (i.e.. ventrally injected) showing significant inhibition of posterior a- globin. (E) Detail of embryo shown in D; note the small patch of a-globin-positive cells in the posterior of the VBI demonstrating that this embryo is at a similar developmental stage as that in A. (F,G) Embryos injected with XRD+lacZ with anterior targeting of the VBI typically seen with dorsally injected pigmented controls. Note the absence of a-globin in the anterior VBI (n=19). (H) Detail of embryo shown in G. Note the complimentary pattern of blue and purple staining. The blue cells are in close proximity to purple cells suggesting that XRD acts cell autonomously. In this experiment, we injected 60 embryos with either lacZ alone or with XRD+lacZ. Of these embryos, 54 injected with lacZ alone survived the entire procedure while 51 injected with XRD+lacZ survived. All 54 embryos injected with lacZ alone showed normal a- globin staining. Of the 51 embryos injected with Xrd+lacZ, 16 had normal a-globin expression and did not show XRD targetting to the VBI. 28 embryos (54%) showed abnormal a-globin staining and all of these showed b-gal staining in the VBI. In these experiments, there were an additional seven embryos injected with Xrd+lacZ that demonstrated normal a-globin staining and also contained b-gal in the VBI. This incomplete penetrance could be due to variations in the effective levels of XRD in these embryos. Consistent with this, we did not observe overlaps with b-gal and a-globin when higher levels of XRD mRNA (2 ng/blastomere) were injected.|