spib is required for primitive myeloid development in Xenopus

Abstract

Vertebrate blood formation occurs in 2 spatially and temporally distinct waves, so-called primitive and definitive hematopoiesis. Although definitive hematopoiesis has been extensively studied, the development of primitive myeloid blood has received far less attention. In Xenopus, primitive myeloid cells originate in the anterior ventral blood islands, the equivalent of the mammalian yolk sac, and migrate out to colonize the embryo. Using fluorescence time-lapse video microscopy, we recorded the migratory behavior of primitive myeloid cells from their birth. We show that these cells are the first blood cells to differentiate in the embryo and that they are efficiently recruited to embryonic wounds, well before the establishment of a functional vasculature. Furthermore, we isolated spib, an ETS transcription factor, specifically expressed in primitive myeloid precursors. Using spib antisense morpholino knockdown experiments, we show that spib is required for myeloid specification, and, in its absence, primitive myeloid cells retain hemangioblast-like characteristics and fail to migrate. Thus, we conclude that spib sits at the top of the known genetic hierarchy that leads to the specification of primitive myeloid cells in amphibians.

Figure 1

Xenopus spib identification. (A) Conserved protein motifs in spib and spi1; the PEST degradation domain and ETS DNA-binding domain. (B) ClustalW tree alignment on vertebrate spib and spi1 protein sequences and percentage identity over their entirety and DNA-binding domains. (C) Genome organization around the known vertebrate spib and spi1 genes, showing synteny between various vertebrate genomes. Genes shown in the same color represent putative orthologous genes. Expression pattern of Xenopus tropicalis spib and (D-G) Xenopus laevis spib paralogs (SpiBa and SpiBb, H-J) is indistinguishable and marks a population of primitive myeloid progenitors in the anterior ventral blood islands (aVBIs). spib expression is transient and first detected at stage 17. (F) Cleared embryo shows only mesodermal expression, and (J) spib expression is eventually down-regulated after spib-expressing cells associate with vitelline veins (arrowheads). Anterior is to the left, and dorsal to the top in all lateral views; panels D, G, and I,J represent ventral views. The following database sequences were used: Hs spib Q01892, Mm spib O35906, Xl spiba BC046671, Xl spibb BC130210, Xt spib IMAGE 7023083, Hs spi1 P17947, Mm spi1 P17433, Gg spi1 NP_990354, Xt spi1 BC098077, and Dr spi1 CAD58735 (outgroup NP_001008139 and P14921).

Figure 2

Primitive myeloid migration time-lapse video microscopy. (A) Experimental setup, stage-14 to -16 anterior ventral blood islands were transplanted from microruby-injected to noninjected embryos. Transplanted cells become migratory and leave the transplant to colonize the embryo in 4 to 8 hours. Throughout this period, primitive myeloid cells show different behaviors (Videos S1–S3). (B,C) Stage-18 brightfield/fluorescent composite of transplanted embryos. (D,E) Stills from supplementary movies. (Panel D and Video S2) Ventral view of primitive myeloid cells leaving the transplanted aVBI with “blebbing” behavior and low migratory speeds. (Panel E and Video S3) Primitive myeloid cells leaving the transplanted aVBI (stage 26, lateral view). At this stage, cells acquire elongated cell morphology and higher motility. Large dashed line shows the embryo contour, and the light dashed square shows the enlarged region shown in still frames. Colored arrowheads point and track the same cell. (D,E) Anterior view is shown to the left; dorsal, to the top. Time is shown in minutes. Images in panels B and C were obtained on a fluorescence stereoscope Leica MZ FLIII (Wetzlar, Germany) attached to a Sony CCD camera DXC-950 image capture system controlled by Northern Eclipse software 7.0 (Empix Imaging, Mississauga, ON). For panels D and E, the same image capture system was attached to an Olympus IX70 inverted fluorescent microscope; 0.1× MMR was used as imaging medium. (D) Total magnification 300× objective (Olympus LCPlan 20×/0.4 NA). (E) Total magnification 40× objective (Olympus UPlanFL 4×/0.13 NA). Photoshop CS2 (Adobe Systems, San Jose, CA) or ImageJ 1.38 (National Institutes of Health, Bethesda, MD) were used for image or time-lapse video processing.

Figure 3

Recruitment of primitive myeloid cells to embryonic wound sites in Xenopus laevis. spib (A-C), mpo (XPOX2) (D,E), and scl (G-I) whole mount in situ hybridization in wounded embryos. Three hours is sufficient for the recruitment of myeloid cells expressing spib or mpo to the scrape wound outline (dashed line contour in the inset). scl-expressing cells are not recruited to the wound site, and primitive myeloid cell recruitment occurs before the appearance of a fully functional vascular network.

Figure 4

Xenopus tropicalis primitive myeloid marker expression analysis by RT-PCR and WMISH. (A-F) spib, spi1, cebpa, mpo, mmp7, and globin single embryo qRT-PCR and ventral views of WMISH at stages 18 and 23. Among the transcription factors, (A) spib is the first to be specifically expressed in primitive myeloid cells, (B) followed by spi1. (C) Earlier expression of cebpa is due to its broad mesodermal expression before its restriction to the aVBI. (D) mpo is a neutrophil marker; (E) mmp7, a macrophage marker; and (F) globin, a erythroid marker. Error bars represent the SD of at least 3 independent experiments.

Figure 5

spib is necessary for primitive myeloid differentiation. spib-depleted embryos show absence or reduced expression of mpo, spi1, and mmp7 but not cebpa. (A-O) WMISH and (P,Q) RT-PCR analysis on spib ATG and e1i1 morphants. ATG*** is a 3-mismatch ATG morpholino. (P) spib ATG morpholino reduces but does not eliminate all spib transcripts, (Q) whereas spib e1i1 morpholino eliminates functional spib mRNA at 10 ng. (A-F) Severe reduction on the number of spib- and mpo-expressing cells and (J-O) marked absence of spi1 and mmp7 expression on spib morphants. (G-I) Under the same experimental conditions, cebpa-expressing cells are not affected. All embryos are shown in lateral views with anterior to the left. The number of representative embryos is shown at the bottom right corner of each panel.

Figure 6

Effect of spib knockdown in primitive myeloid progenitors. (A-I) Primitive macrophages defined by the expression of spi1 and mmp7 are absent from spib-depleted embryos. (J-L,O-Q,T-V) WMISH analysis of aVBI in spib morphants; at the earliest point we can identify a pool of primitive myeloid progenitors by the coexpression of spib, mpo, and cebpa. (M,R,W) Quantification of the number of embryos that express spib, mpo, and cebpa and (N,S,X) size of primitive myeloid progenitor pool, by the area of tissue-expressing progenitor markers. Error bars represent the SD of the number of embryos analyzed (n). (O) Amount of tissue was calculated using area ratios of a and a′. Similar results were obtained in either ATG or e1i1 morphants and are in contrast with the uninjected and ATG mismatch control group (CTL). Student t test P values lower than .001 are marked as * and considered statistically significant; P values higher than .05 are not statistically significant and marked as ***.

Figure 7

Analysis of migratory potential and hemangioblast-like characteristics of primitive myeloid progenitors in spib-depleted embryos. Migration away from the aVBI is the first morphologic sign of primitive myeloid differentiation and is absent in spib-depleted embryos. (A-I) In spib-depleted embryos, we never observed isolated cells far from the aVBI expressing spib, mpo, and cebpa at a time when migration should have already set in. Usual down-regulation of scl and fli (▿, J,M) is eliminated in spib morphants (▼, K,L and N,O). The scl⁺fli⁺ hemangioblast-like characteristics are maintained for longer in a small subset of spib-depleted aVBI tissue.