• Open Access

Serpent and a hibris reporter are co-expressed in migrating cells during Drosophila hematopoiesis and Malpighian tubule formation


  • Ruben D. Artero,

    1. Laboratory of Developmental Genetics, Dept of Genetics, University of Valencia, Dr. Moliner 50, ES-46100 Burjasot, Spain. E-mail: ruben.artero@uv.es
    Search for more papers by this author
  • Lidon Monferrer,

    1. Laboratory of Developmental Genetics, Dept of Genetics, University of Valencia, Dr. Moliner 50, ES-46100 Burjasot, Spain. E-mail: ruben.artero@uv.es
    Search for more papers by this author
  • Amparo Garcia-Lopez,

    1. Laboratory of Developmental Genetics, Dept of Genetics, University of Valencia, Dr. Moliner 50, ES-46100 Burjasot, Spain. E-mail: ruben.artero@uv.es
    Search for more papers by this author
  • Mary K. Baylies

    1. Developmental Biology Program, Sloan-Kettering Institute, Memorial Sloan-Kettering Cancer Center, New York, NY, USA
    Search for more papers by this author

Ruben D. Artero, Laboratory of Developmental Genetics, Dept of Genetics, University of Valencia, Dr. Moliner 50, ES-46100 Burjasot, Spain. E-mail: ruben.artero@uv.es


Motile mesodermal cells contribute several cell types to developing embryos. In Drosophila, blood cell precursors or prohemocytes, are first detected in the procephalic mesoderm by the expression of the GATA transcription factor Serpent. Once specified, a subset of prohemocytes migrate posteriorly to populate most of the embryo and further differentiate as plasmatocytes. Similarly, Drosophila nephrogenesis involves integration of posterior mesodermal cells into the Malpighian tubule primordia where these cells differentiate as stellate cells. Here we investigated the possibility that the immunoglobulin-domain protein Hibris and the GATA factor Serpent were co-expressed in motile mesodermal cells by using the hibris expression reporter P[w+]36.1 and antibody staining. We show that P[w+]36.1 reproduces the endogenous expression of hibris in several embryonic tissue types and organs, including mesectoderm, early mesoderm, pharyngeal musculature, hindgut, anal plates, posterior spiracles, and antennomaxillary complex. We find that both migrating prohemocytes and posterior mesodermal cells, before their integration into the Malpighian tubule primordia, simultaneously express the hibris reporter and Serpent. We also show that hibris function is not essential for prohemocyte migration out of the procephalic mesoderm NOR maintenance of Serpent expression in prohemocytes.


Mesodermal cell migration underlies several morphogenetic and cell specification processes in the Drosophila embryo. Two examples are hematopoiesis and renal tubule formation. Prohemocytes are derived from the head (procephalic) mesoderm and are first identifiable during embryonic stage 5 by the expression of Serpent (Srp), a GATA transcription factor required for hematopoietic development (Rehorn et al. 1996; reviewed by Evans et al. 2003). Prohemocytes will eventually differentiate into either plasmatocytes, which migrate out of the head region to populate the embryo, or crystal cells that generally remain localized near their point of origin in the embryo. For many cell types, including blood cells, receptor tyrosine kinases are essential for proper navigation (reviewed by Starz-Gaiano and Montell 2004). The Drosophila homolog of the human platelet-derived growth factor and vascular endothelial growth factor receptors (PDGFR and VEGFR, respectively), PVR, is highly expressed in the hemocytes and required for their directed migration. Embryos carrying a mutation in pvr still specify hemocytes, but these cells are unable to migrate properly. The same phenotype results from inactivation of the PVR ligands, which are normally localized along the hemocyte-migration route (Cho et al. 2002). Hemocytes have several significant roles in development, seeking out and removing dead cells, secreting and remodelling extracellular matrix components, monitoring the environment for pathogens and finally signaling to the larval fat body (Evans et al. 2003).

Cells from the group of posterior mesoderm that gives rise to the caudal visceral mesoderm migrate out of their original location to contribute to the developing embryo as a physiologically distinctive subset of Malpighian tubule cells: the stellate cells (Denholm et al. 2003). These cells not only migrate but also undergo a complex mesenchymal-to-epithelial transition as they integrate into the ectodermal epithelial buds. The normal incorporation of stellate cells and the later physiological activity of the mature tubules depend on the activity of hibris (hbs), an ortholog of the mammalian protein NEPHRIN. Hibris is a transmembrane immunoglobulin-like protein that shows extensive homology to Drosophila Sticks and stones (Sns) (Bour et al. 2000; Artero et al. 2001; Dworak et al. 2001). Hibris is expressed in embryonic visceral, somatic and pharyngeal mesoderm among other tissues such as the mesectoderm and the hindgut. In the somatic mesoderm, Hibris is restricted to a subset of fusion competent myoblasts and embryos that lack or overexpress hibris show a partial block of myoblast fusion, followed by abnormal muscle morphogenesis. Thus, together with other transmembrane proteins such as Kirre (Ruiz-Gomez et al. 2000), Roughest (Strunkelnberg et al. 2001), and Sns (Bour et al. 2000), it has been proposed that Hibris acts in cell-cell recognition and attraction during embryonic myogenesis (reviewed by Chen and Olson 2004). Indeed, Hibris and Roughest are detected in distinct interommatidial precursor cell populations during Drosophila eye morphogenesis and were found to mediate preferential cell adhesion between specific interommatidial cell types through heterophilic interactions (Bao and Cagan 2005). By employing the hibris reporter line P[w+]36.1, which we confirm that reproduces hibris expression in several embryonic tissue types, here we show co-expression with the GATA factor Serpent in migrating prohemocytes and in posterior mesodermal cells before their integration into the Malpighian tubule primordia. We also show that hibris function is not essential for prohemocyte migration out of the procephalic mesoderm and maintenance of their Serpent expression.

Material and Methods

Drosophila genetics

yw was used to reveal wild-type expression patterns. Fly stocks were hbs459, Df(2R)X28 and P[w+]36.1, all described in Artero et al. (2001). Briefly, hbs459 carries a non-sense codon mutation at position 553 in the hibris coding sequence and has been described as protein null. Df(2R)X28 deletes the hibris locus and was used in combination with the hbs459 chromosome to simultaneously remove hibris function but complement second site mutations present in the chromosomes. For expression studies a nuclear-LacZ reporter gene driven by 4 kb of genomic sequence upstream of the hibris transcription unit was constructed in pUAST and transgenic flies were raised. The line P[w+]36.1, in which the reporter construct inserted approximately 650 bp upstream of the hibris transcription start site, was generated using a P-homing approach (Taillebourg and Dura 1999). For phenotypic studies, relevant stocks were balanced with CyO P[w+wgen11lacZ] and homozygous mutant embryos were identified by the lack of β-galactosidase expression.

Histology techniques

Embryos were collected overnight at 25°C on grape juice plates daubed with yeast paste and were processed for whole-mount immunohistochemistry using standard procedures (essentially as described by Rushton et al. 1995) with modifications for fluorescent detection. Antibody dilutions were: anti-β-galactosidase (1:2000, mouse; Promega) and anti-Srp (1:1000, rabbit). Biotinylated antibodies were used in combination with Vector Elite ABC kit (Vector Laboratories, CA). Specimens were mounted in Vectastain (Vector) and microscope images were captured using an Axiocam camera mounted on a Zeiss Axiophot microscope (Zeiss). Immunofluorescent signals in co-localization studies were analyzed using a Leica TCS-NT confocal microscope. hibris RNA localization was detected using digoxigenin-labeled RNA probes using cDNA clone RE70806 as template (Berkeley Drosophila Genome Project), as described (O′Neil and Bier 1994). Figures were assembled in Adobe Photoshop.

Results and Discussion

The P element insertion line P[w+]36.1 reproduces hibris embryonic expression pattern

The hibris reporter P[w+]36.1 was isolated as a P homing event that recapitulated endogenous expression in the mesoderm during the genetic characterization of the hibris gene (Artero et al. 2001; see material and methods). Because Hibris is a membrane bound protein, thus making co-localization with transcription factors difficult, the nuclear localization of the hibris reporter P[w+]36.1 has been used in Drosophila eye morphogenesis and Malpighian tubule development to follow hibris expression (Denholm et al. 2003; Bao and Cagan 2005). In both cases, the hibris loss of function phenotype was consistent with the expression data, thus reinforcing that the reporter recapitulated endogenous hibris expression. To specifically address the degree of overlap between the endogenous and the reporter patterns of expression in the embryo, we detected hibris transcription and β-galactosidase production from the P[w+]36.1 reporter in parallel experiments. Worth noting is that the simultaneous immunodetection of β-galactosidase and Hibris protein is not feasible due to the different fixation conditions required to detect each antigen.

hibris transcripts are strongly expressed in the ventral midline (mesectoderm) and visceral mesoderm precursors in late stage 10 embryos, whereas expression in the somatic mesoderm begins slightly later. P[w+]36.1 embryos recapitulated the midline and mesodermal expression of hibris, with the expected delay for a reporter and an overall weaker signal (Fig. 1A-B). Stage 14 embryos revealed coincident hibris RNA and β-galactosidase expression in the hindgut, pharyngeal muscles, antennomaxillary complex, anal plates, and posterior spiracles (Fig. 1C-H and not shown). However, whereas hibris was clearly expressed in the heart precursors and epidermal muscle attachments, both by in situ hybridization and immunostaining (Fig. 1C; Artero et al. 2001), we only detected very weak expression in the heart precursors of stage 13 embryos and a rather general expression in the epidermis of late P[w+]36.1 embryos, which in addition to ectopic, may include the endogenous expression (not shown).

Figure 1.

A-H. The P[w+]36.1 reporter line recapitulates several aspects of the hibris expression pattern in the embryo. Transcription from the hibris gene (A, C, E, G) and β-galactosidase production from the P[w+]36.1 reporter line (B, D, F, H) were detected by in situ hybridization and immuhistochemistry, respectively. (A) Ventrolateral view of a late stage 10 embryo showing hibris RNA in ventral midline (arrowhead) and visceral mesoderm precursors (bent arrow). The hibris reporter reveals a similar expression pattern in ventrolateral views of stage 11 embryos (B), which includes midline (arrowhead) and mesodermal expression (bent arrow). Dorsal views of stage 14 embryos (C, D) showing hibris transcripts in the heart precursors (white arrowhead), hindgut (bent arrow) pharyngeal muscles (black arrowhead), epidermal muscle attachment sites (arrow) and posterior spiracles (asterisk; C). P[w+]36.1 reporter embryos express β-galactosidase in pharyngeal muscles, posterior spiracles and hindgut (symbols used as in C), but fail to show strong signal in epidermal muscle attachment sites and heart precursors. Stage 14 embryos revealed overlapping expression of hibris and the P[w+]36.1. reporter in the antennomaxillary complex (E, F; arrowheads) and anal plates (G, H; bent arrows). Anterior is to the left in all panels.

The hibris reporter line P[w+]36.1 is expressed in a multipotent hemapotoietic cell population

During the study of the involvement of hibris in myoblast fusion (Artero et al. 2001, Dworak et al. 2001) and mesodermal cell integration into Malpighian tubule primordia (Denholm et al. 2003), we noticed several examples of β-galactosidase-positive cells in the cephalic mesoderm of hibris reporter embryos. To establish their identity, we investigated the possibility that the hibris reporter detected prohemocyte cell migration out of the head mesoderm by simultaneously detecting the prohemocyte marker Srp. In these experiments β-galactosidase production in P[w+]36.1 embryos revealed a population of migrating mesodermal cells in late stage 11 embryos, which simultaneously express the prohemocyte marker Srp thus confirming their identity as prohemocytes (Fig. 2). These cells first migrate dorsally and then posteriorly towards the tip of the embryo's tail, which, at this developmental time, is adjacent to the cephalic region. Prohemocytes do not migrate at random towards a given destination but instead they follow stereotyped routes, probably interpreting adhesion clues on the surfaces that serve as substrate for their movement.

Figure 2.

A-F. A hibris reporter is co-expressed with Serpent in the precursors of the Drosophila hematopoietic lineage. Single confocal sections of mid to late stage 11 Drosophila embryos showing ventral (A-C) and lateral (D-F) views of β-galactosidase expression from the reporter construct P[w+]36.1 (A, D), Srp expression (B, E) and the corresponding overlap between both images (C, F). Arrowheads point to prohemocyte cells, actively migrating at this stage in development, showing both reporter and Srp expression. The bent arrow highlights an example of hibris-specific expression that the P[w+]36.1 reporter reproduces, in this case expression in the ventral midline.

The broad differentiation potential that hematopoietic cells show in vivo has been attributed to transdifferentiation and/or fusion with resident cells. Using the Cre/lox recombination method, it has been demonstrated that bone-marrow-derived cells can fuse with cells in liver, heart and brain thus raising the possibility that cell fusion may contribute to the development or maintenance of these cell types (Alvarez-Dolado et al. 2003). Given the involvement of the Hibris protein in myoblast fusion during Drosophila muscle development (Artero et al. 2001; Dworak et al. 2001), and our results showing that a hibris reporter line is expressed in prohemocytes in Drosophila, it is tempting to propose the Hibris ortholog in mice as a potential protein involved in such cell fusion events.

Serpent is expressed in the putative precursors of Malpighian tubule stellate cells

Because Srp is expressed in actively migrating mesodermal cells committed to the hematopoietic fate, we tested whether Srp was similarly expressed in other motile cells: the precursors of stellate cells during Malpighian tubule formation. In these studies we used hibris expression, from reporter construct P[w+]36.1, as marker to reveal putative stellate cell precursors in the surroundings of Malpighian tubules (Denholm et al. 2003). We detected several examples of Srp and β-galactosidase-positive cells closely apposed to Malpighian tubule primordia, which we interpret as mesodermal cells in the process of integrating into the tubule (Fig. 3). These results offer another example of correlation between hibris reporter and Srp expression in actively migrating mesodermal cells.

Figure 3.

A-I. Putative stellate cell precursors express Serpent before integration into Malpighian tubules. Confocal images showing early stage 13 Drosophila embryos stained to detect Srp expression (A, D, G), β-galactosidase expression from the P[w+]36.1 reporter construct (B, E, H) and the corresponding overlap (C, F, I). All images are dorsal views. Arrowheads point to mesodermal cells simultaneously expressing the hibris reporter construct P[w+]36.1 and Srp in the surroundings of developing Malpighian tubules.

hibris function is not essential for prohemocyte migration

Because Hibris is a cell adhesion protein, we next tested whether hibris function was required for prohemocyte migration out of the cephalic mesoderm. In these experiments we detected Srp expression in a hibris mutant background and found several examples of Srp-positive migrating cells (Fig. 4). These results indicate that hibris function was dispensable for the initial migration of prohemocytes. However, we did not follow prohemocytes to their final destinations and thus cannot rule out later defects due to hibris loss of function. These results also indicate that hibris is not required for maintenance of Srp expression since we did not detect a reduction in Srp expression in a hibris mutant background.

Figure 4.

A–C.hibris is not essential for prohemocyte initial migration. Late stage 11 Drosophila embryos of the genotype hbs459/Df(2R)X28 stained with an anti-Srp antibody and shown in ventral (A, B) and lateral (C) views. Arrowheads point to several examples of mesodermal cells that stain positive for the prohemocyte marker Srp in a hibris mutant background.


This work was supported in part by research grants from the Spanish Ministerio de Educación y Ciencia (SAF2003-03536) to R.A., the New York Academy of Medicine-Speaker′s Fund for Biomedical Research Toward the Science of Patient care, and the National Institutes of Health (GM 56989) to M.B.