![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 17, No. 3, pp. 348-353, February 1, 2003
1 Molecular Biology Institute, 2 Department of Molecular, Cell, and Developmental Biology, Biological Chemistry, and Human Genetics, University of California Los Angeles, Los Angeles, California 90095, USA
 |
ABSTRACT |
|---|
 Top
 Abstract
 Introduction
Results and Discussion
Materials and methods
References
|
|---|
The differentiation of Drosophila blood cells relies on a functional hierarchy between the GATA protein, Serpent (Srp), and multiple lineage-specific transcription factors, such as the AML1-like protein, Lozenge (Lz). Two major branches of Drosophila hematopoiesis give rise to plasmatocytes/macrophages and crystal cells. Serrate signaling through the Notch pathway is critical in the regulation of Lz expression and the specification of crystal cell precursors, thus providing a key distinction between the two lineages. The expression of Serrate marks a discrete cluster of cells in the lymph gland, a signaling center, with functional similarities to stromal signaling in mammalian hematopoiesis.
| |
Introduction |
|---|
|
Top
Abstract
Introduction
Results and Discussion
Materials and methods
References
|
|---|
Runx1, also known as Acute Myeloid Leukemia 1 (AML1), is
essential for the development of blood cells arising
from all lineages of definitive hematopoiesis in mice (Okuda et al.
1996
; Wang et al. 1996). In humans, AML1 is the most frequent target
for translocations resulting in acute myeloid leukemias (Werner et al.
1999). The family of GATA transcription factors is also used
reiteratively in multiple stages of blood development (Orkin 1998), and
a cofactor for the GATA proteins, FOG1, is required for erythropoiesis
and differentiation of megakaryocytes (Tsang et al. 1998). In studying the regulation of such transcription factors, in vitro differentiation assays have suggested that multiple signaling pathways are coordinately involved in hematopoiesis (Van Den Berg et al. 1998; Yoshida et al.
1998; Milner and Bigas 1999). Unfortunately, there is a paucity of in
vivo loss-of-function data involving signaling pathways that control
hematopoiesis in mammals due to difficulties such as pleiotropy,
redundancy, and early lethality.
There are significant differences with regard to the variety and
function of Drosophila hemocytes relative to mammalian blood cells, however, several molecular aspects of early hematopoiesis and
immunity have been evolutionarily conserved (Rehorn et al. 1996;
Dearolf 1998; Hoffmann et al. 1999; Lebestky et al. 2000). Two major
classes of Drosophila hemocytes are plasmatocytes, which can
function as macrophages that phagocytose invading pathogens and debris
from apoptotic cells (Tepass et al. 1994), and crystal cells that are
involved in the melanization of pathogens (Rizki et al. 1980).
serpent (srp), a Drosophila GATA homolog, is
expressed in all hemocyte precursors and is required for the
development of both classes of hemocytes (Rehorn et al. 1996; Lebestky
et al. 2000). lozenge (lz) encodes an AML1/Runt
domain family transcription factor (Daga et al. 1996) that is expressed
in the crystal cell precursors and is required for the specification of
this lineage (Rizki and Rizki 1981; Lebestky et al. 2000). glial
cells missing (gcm), a novel transcription factor
expressed exclusively in plasmatocytes in the embryo (Bernardoni et al.
1997), is required for their development. Its possible role in larval
hematopoiesis is less clear. Srp is essential for the expression of
both lz and gcm (Bernardoni et al. 1997; Lebestky et
al. 2000), creating a hierarchy of transcription factors controlling
the two major branches of hematopoiesis (Lebestky et al. 2000).
u-shaped (ush), a FOG homolog, also functions in
Drosophila hematopoiesis (Fossett et al. 2001). The molecular
similarities between Srp/Ush/Lz and GATA/FOG/AML1 suggest that aspects
of molecular mechanisms of blood development are shared between mammals
and Drosophila. Previous studies in Drosophila have
shown a role for JAK/STAT and Toll pathways in the proliferation of
hemocytes and immunity (Dearolf 1998; Mathey-Prevot and Perrimon 1998;
Qiu et al. 1998). However, no signaling pathway was known to specify a
commitment that distinguishes between lineages.
In this study, we show that localized Notch signaling causes an
important early distinction between the crystal cell and plasmatocyte lineages. The Notch pathway controls binary cell fate decisions among
undetermined precursor cells in a multitude of developmental systems
(Artavanis-Tsakonas et al. 1999). We also find that a signaling center
expressing the ligand Serrate is important for differentiation and
proliferation of larval hemocytes in Drosophila.
| |
Results and Discussion |
|---|
|
Top
Abstract
Introduction
Results and Discussion
Materials and methods
References
|
|---|
Notch and Su(H) are required for hemocyte proliferation and development
Morphological and molecular differentiation of hemocytes can be
monitored in the larval lymph gland (Shrestha and Gateff 1982). Srp is
expressed in all hemocyte progenitors, whereas the expression of Lz is
largely restricted to a small subset of hemocytes in the first pair
(anterior-most) lobes of the lymph gland (Fig. 1A). In the Notch
temperature-sensitive allele, Nts1, Lz expression is
eliminated from the lymph gland at the nonpermissive temperature (Fig.
1B). Lz+ cells are also missing in the lymph glands of
Su(H)SF8/Su(H)AR9 larvae (Fig. 1C). As
Su(H) is the transcription factor that activates Notch target genes
(for review, see Weinmaster 2000), the canonical Notch/Su(H) pathway is
important for Lz expression in crystal cell precursors.
|
Using the FLP/FRT system (Golic 1991), we generated mutant clones of
N55e11 (Fig. 1D), Nts1 (Fig. 1E),
and Su(H)
47 (Fig. 1F) in the lymph
gland. Clones were marked by the absence of 
Gal expression and were
always small, reflecting an early requirement for the Notch pathway in
cell proliferation. Importantly, Lz+ cells were always
excluded from the mutant clones. Approximately 200 Lz+ cells
were counted for each genotype. We also generated positively marked
cells within the lymph glands that express an extracellular, dominant
negative form of Notch (NDN; Rebay et al. 1993). As in the
loss-of-function clones, marked Gal+ cells expressing
NDN do not coexpress Lz (Fig. 1G,H). Therefore, Lz+
cells are derived preferentially from wild-type precursors, once again
suggesting a requirement for Notch in the development of crystal cell
precursors. Rare exceptions to this rule have been seen (Fig. 1I),
presumably reflecting minor variability in the level of expression of
UAS-NDN in individual Gal+ cells.
Additionally, lymph glands that misexpress NDN are smaller
than wild-type controls and fewer Gal+ cells are seen.
This is not a result of increased apoptosis, as the fraction of cells
that stain with TUNEL is not increased in NDN lymph
glands compared with controls (Fig. 1J,K). Taken together, the above
genetic analyses establish that Notch signaling is critical for the
specification of the crystal cell lineage, but also has an early role
in the proliferation of hematopoietic cells in the lymph gland, similar
to the proliferative role of Notch in imaginal discs (Go et al. 1998;
Artavanis-Tsakonas et al. 1999).
Notch function is also required for crystal cell development in the
head mesoderm region during embryogenesis. The average number of
Lz+ crystal cell precursors in each bilateral cluster is
reduced significantly in N55e11 embryos [from 18 in
wild-type to 9 in mutant stage 12 embryos (n = 30); Fig. 1L,M]. The
residual crystal cell development in N55e11 embryos
likely reflects the maternal contribution of Notch (Morel and
Schweisguth 2000).
Serrate is the ligand for Notch in hematopoiesis
To identify the Notch ligand responsible for the development of
crystal cells, we investigated the possible role of the two Notch
ligands, Delta (Dl) and Serrate
(Ser) in larval hematopoiesis. Expression of Dl protein was
not detected in the lymph gland, and expression of Lz was not altered
in the DlM2/DlR9 genetic background (data
not shown). In contrast, Lz expression was absent in lymph glands of
SerBd3/+ larvae (Fig.
2A,B), and as a consequence, significantly
fewer circulating crystal cells were observed (Fig. 2C,D). As
SerBd3 is a dominant-negative allele, we used a null
allele, SerRx82, to generate loss-of-function
clones. Mutant clones usually did not contain Lz-expressing cells (Fig.
2E). However, consistent with Ser being required in signaling cells,
rare examples of Ser
/Ser mutant cells
at the edge of a clone were found to express Lz (Fig. 2F, arrowhead).
These results establish Ser as a ligand for Notch in the specification
of crystal cells.
|
To investigate the sufficiency of the Notch pathway in crystal cell
development, we used the Gal4/UAS system (Brand and Perrimon 1993) to ectopically express an activated form of Notch, Nact
(Fortini et al. 1993), or Ser in the lymph glands (Fig. 2G-M). The
number of Lz+ cells was dramatically increased in
hsp70-Gal4/UAS-Nact lymph glands (Fig. 2G,H). A more
modest increase in Lz+ cells was seen when Ser was
overexpressed in hsp70-Gal4/UAS-Ser larvae (Fig. 2I).
Strikingly, misexpression of Nact caused Lz+ cells
to appear in the second lobe (Fig. 2K), where they are rarely seen in
wild type (Fig. 2J). To create small clones of cells with sustained
misexpression of Ser, we used the AyGal4 system (Ito et al.
1997). In contrast to the Gal control (Fig. 2L), the misexpression
of Ser resulted in a dramatic increase in the number of crystal cell
precursors differentiating in the second lobe (Fig. 2M). Thus, the
misexpression of Ser can create ectopic signaling centers, which then
induce Lz expression.
Serrate defines a signaling center
Expression of Ser was monitored using an antibody raised against
the Ser protein and using a Ser-Gal reporter (Bachmann and Knust
1998; Fig. 3). Many Ser-expressing cells
were found to be clustered at the posterior end. Additionally,
scattered Ser+ cells were also seen throughout the gland
(Fig. 3A). A high level of Ser-Gal expression is also observed in
the cells at the posterior tip, immediately adjacent to a pericardial
cell (Rugendorff et al. 1994; Fig. 3B-D). Given the unique pattern of
Ser expression in this discrete cluster of cells, we designate this
region as the Posterior Signaling Center (PSC). A number of
Ser+ cells that are adjacent to the PSC appear to have broken
off from the cluster (Fig. 3A,F, inset), suggesting that Ser+
cells may all be derived from the PSC and actively migrate into the
body of the lymph gland to seed differentiation of the crystal cells.
Future real-time cell imaging will be necessary to definitely establish
cell migration patterns.
|
Ser+ cells are seen at the PSC, but many are also found inside the gland. In either case, there is no overlap between cells that express Lz and Ser (Fig. 3H,I). As Ser function is critical for the development of Lz+ cells, these results suggest that cell-cell interaction between the two subpopulations is needed for crystal cell development. Consistent with this notion, many examples of Lz+ cells that are immediately adjacent to Ser+ cells were seen (Fig. 3H,I), suggestive of an inductive relationship between these two subpopulations. When Ser mutant clones were generated in the lymph gland (Fig. 2E,F), we frequently found lymph gland lobes that entirely lack Lz+ cells, although the entire gland was not mutant. This further supports the hypothesis that Ser+ cells are clustered initially to form a signaling center and later disperse and participate in the induction of Lz+ cells.
As with all the other cells in the lymph gland, the PSC cells express Serpent (GATA; Fig. 3E-G). The expression of Ser is the first observed indication that distinguishes the PSC cells from the remainder of the Srp+ progenitors within the lymph gland. Additionally, the PSC displays a number of interesting characteristics that distinguish this region from the rest of the lymph gland. First, although Ser is required for the expression of Lz, the PSC is spatially distant from the region in which the majority of crystal cells differentiate (Fig. 3J). Second, Ser expression in the PSC is unaffected in Nts1 mutants raised at the nonpermissive temperature (Fig. 3K). Also, in creating clones of NDN in the lymph gland (Fig. 1H), the majority of cells that express NDN are found in the PSC region, and these cells do not apoptose compared with cells in the rest of the lymph gland (Fig. 3L). These results suggest that although the Ser+ cells of the PSC signal through the Notch pathway, they themselves are refractory to the Notch signal and do not require Notch for their own development. Finally, the Ser+ cells of the PSC do not incorporate BrdU administered in vitro for 1 h (Fig. 3M) or through overnight (18 h) feeding of third instar larvae (Fig. 3N), unlike the majority of cells within the lymph gland that actively proliferate in the third instar (Fig. 3M,N). Thus, the Ser+ cells of the PSC represent a distinct signaling cell population that rarely proliferates, but is important for proliferation of hemocyte precursors and for their differentiation into crystal cells.
We next investigated the temporal relationship between Notch signaling
and Lz expression. Ser is expressed robustly in the second instar
larval lymph gland (Fig. 4A), however, very
few Lz+ cells, if any, are seen at this stage (Lebestky et
al. 2000). Whereas Lz is spatially restricted to the first lobe of the
lymph gland, Ser is also seen in further posterior lobes (Fig. 4B). Consistent with these temporal and spatial relationships, expression of
Ser is independent of lz (Fig. 4C). The expression of a
12XSu(H)-Gal reporter line (Go et al. 1998) can be used as a
read-out for Notch signaling. Upon activation of the Notch pathway, the
Su(H) protein causes expression of Gal. In the first lobe of the
lymph gland, a small number of cells that receive Notch signal are
marked by Gal expression (Fig. 4D). This expression of Gal is
dependent on Notch (Fig. 4E), but is unaffected in lz
null mutants (Fig. 4F). This further substantiates the fact that Notch
activation is upstream of Lz function. When lymph glands expressing
12XSu(H)-Gal were stained with the 
-Lz antibody (Fig. 4G-I),
many examples of Gal+ cells that also express Lz protein
are seen (Fig. 4I), further supporting that signaling by Notch is
important for cells to become Lz+. However, cells that
express Gal, but not Lz, and those that express Lz alone are also
seen. The simplest explanation for this pattern is that a cell
receiving a Notch signal is a prohemocyte. As a cell initiates Lz
expression, it seems refractory to the Notch signal but the perdurance
of the stable Gal protein is still evident. Finally, as this cell
initiates a program for crystal cell differentiation, it continues to
express Lz. Mature crystal cells in circulation have been shown to be
Lz+ (Lebestky et al. 2000). Whether Drosophila
hemocytes undergo maturation as in mammals, can be tested once suitable
markers are identified.
|
Previous studies have shown that a hierarchy of transcription factors
is responsible for the differentiation of two independent classes of
hemocytes (Lebestky et al. 2000). Results presented here demonstrate
that Notch signaling among Srp+ hemocyte precursors directly
or indirectly regulates the expression of Lz. This signaling creates a
molecular distinction between the two major branches of hematopoietic
lineage and results in the differentiation of crystal cells (Fig. 4J).
The expression and function of Ser in the lymph gland marks a
putative signaling center that highlights certain developmental and
molecular similarities between hematopoietic tissues of
Drosophila and mammals. The existence of discrete
hematopoietic microenvironments within the bone marrow, in which
stromal cells influence the proliferation and differentiation of the
HSCs has been proposed in many mammalian studies (Bianco et al. 2001).
In fact, stromal signaling via the mammalian Ser homolog, Jagged1, has
been implicated in the expansion and self-renewal of HSCs
(Varnum-Finney et al. 1998). However, in vivo identification and
characterization of such signaling microenvironments within the bone
marrow is difficult to achieve. Although Drosophila hemocytes
are quite distinct from mammalian blood cells, spatial relationships
with regards to signaling pathways in the Drosophila lymph
gland will lead to a better understanding of how similar patterning
events affect HSC populations in the marrow.
In mice, AML1/Runx1 is expressed in the embryonic dorsal aorta in
hematopoietic clusters associated with an endothelial layer (North et
al. 1999). The Notch ligand, Jagged1, is also expressed in the
embryonic aorta at this stage (Loomes et al. 1999), but any possible
role it may have in the generation of hematopoietic clusters has not
been investigated. Furthermore, independent studies have implicated
both Notch and AML1 in the progressive commitment and
maturation of T cells (Aster and Pear 2001; Reizis and Leder 2002). The
regulatory relationship between Notch and lz in
Drosophila suggests similar interactions may exist between the
Notch pathway and AML1 in mammalian hematopoiesis. Investigating such
regulatory relationships is especially valuable, as AML1 is critically
needed for all definitive hematopoiesis and is also the most frequent target in acute myeloid leukemias (Downing et al. 2000).
| |
Materials and methods |
|---|
|
Top
Abstract
Introduction
Results and Discussion
Materials and methods
References
|
|---|
Flp-out clones expressing UAS-NDN or
UAS-Ser were generated using the Ay-Gal4 system (Ito
et al. 1997) and marked with either UAS-nuclear
Gal or UAS-GFP. To activate hsp70-flp,
progeny were maintained at 18°C until the first larval instar,
heat-shocked at 37°C for 40 min, and then returned to 18°C until
the dissection. hsp70-Gal4/UAS-Nact or
hsp70-Gal4/+ UAS-Ser/+ were maintained at 18°C until
early third instar and heat-shocked twice at 37°C for 40 min, then
returned to 18°C until dissection. N55e11,
Nts1, Su(H)47, or
SerRx82 mutant clones were generated using the
hsp70-FLP/FRT system (Golic 1991). Crosses were maintained at
18°C, heat-shocked for 1 h in the second instar for
N55e11 and Su(H)47, and in the
first instar for Nts1 and
SerRx82. For Figure 1, Nts1 was
raised at 29°C either from first instar until third instar, or for 18 h in the early third instar before dissection in the third instar, in
both cases showing loss of Lz+ cells.
| |
Acknowledgments |
|---|
We thank S. Artavanis-Tsakonas, J. Posakony, F. Schweisguth, Y. Hiromi, K. Irvine, E. Knust, and K. Matthews of the Bloomington Stock Center for providing fly stocks; R. Reuter and K. Irvine for generously providing antibodies; and Volker Hartenstein and members of the Banerjee laboratory for suggestions and comments. This work was supported by an NIH grant (RO1 HL67395) to U.B. and the U.S. Public Health Service NRSA (GMO7185) to T.L.
The publication costs of this article were defrayed in part by payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 USC section 1734 solely to indicate this fact.
| |
Note added in proof |
|---|
Duvic et al. (2002) have also recently studied the role of Ser
and Notch in hematopoiesis.
| |
Footnotes |
|---|
[Keywords: Serrate; Notch; Su(H); crystal cell; Drosophila; hematopoiesis]
Received November 8, 2002; revised version accepted December 9, 2002.
3 Present address: Department of Biology 216-76, California Institute of Technology, Pasadena, California 91125, USA.
4 These authors contributed equally to this work.
5 Corresponding author.
E-MAIL banerjee{at}mbi.ucla.edu; FAX (310) 206-9062.
Article and publication are at http://www.genesdev.org/cgi/doi/10.1101/gad.1052803.
| |
References |
|---|
|
Top
Abstract
Introduction
Results and Discussion
Materials and methods
References
|
|---|
CiteULike 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J. A. Martinez-Agosto, H. K.A. Mikkola, V. Hartenstein, and U. Banerjee The hematopoietic stem cell and its niche: a comparative view Genes & Dev., December 1, 2007; 21(23): 3044 - 3060. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Minakhina, M. Druzhinina, and R. Steward Zfrp8, the Drosophila ortholog of PDCD2, functions in lymph gland development and controls cell proliferation Development, July 1, 2007; 134(13): 2387 - 2396. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Williams Drosophila Hemopoiesis and Cellular Immunity J. Immunol., April 15, 2007; 178(8): 4711 - 4716. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Nakagawa, M. Ichikawa, K. Kumano, S. Goyama, M. Kawazu, T. Asai, S. Ogawa, M. Kurokawa, and S. Chiba AML1/Runx1 rescues Notch1-null mutation-induced deficiency of para-aortic splanchnopleural hematopoiesis Blood, November 15, 2006; 108(10): 3329 - 3334. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Bataille, B. Auge, G. Ferjoux, M. Haenlin, and L. Waltzer Resolving embryonic blood cell fate choice in Drosophila: interplay of GCM and RUNX factors Development, October 15, 2005; 132(20): 4635 - 4644. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S.-H. Jung, C. J. Evans, C. Uemura, and U. Banerjee The Drosophila lymph gland as a developmental model of hematopoiesis Development, June 1, 2005; 132(11): 2521 - 2533. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Soderhall, Y.-A Kim, P. Jiravanichpaisal, S.-Y. Lee, and K. Soderhall An Ancient Role for a Prokineticin Domain in Invertebrate Hematopoiesis J. Immunol., May 15, 2005; 174(10): 6153 - 6160. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Davidson, R. Tirouvanziam, L. A. Herzenberg, and J. S. Lipsick Functional Evolution of the Vertebrate Myb Gene Family: B-Myb, but Neither A-Myb nor c-Myb, Complements Drosophila Myb in Hemocytes Genetics, January 1, 2005; 169(1): 215 - 229. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.-J. Zettervall, I. Anderl, M. J. Williams, R. Palmer, E. Kurucz, I. Ando, and D. Hultmark A directed screen for genes involved in Drosophila blood cell activation PNAS, September 28, 2004; 101(39): 14192 - 14197. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. B. Milchanowski, A. L. Henkenius, M. Narayanan, V. Hartenstein, and U. Banerjee Identification and Characterization of Genes Involved in Embryonic Crystal Cell Formation During Drosophila Hematopoiesis Genetics, September 1, 2004; 168(1): 325 - 339. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Okajima, A. Xu, and K. D. Irvine Modulation of Notch-Ligand Binding by Protein O-Fucosyltransferase 1 and Fringe J. Biol. Chem., October 24, 2003; 278(43): 42340 - 42345. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Fossett, K. Hyman, K. Gajewski, S. H. Orkin, and R. A. Schulz Combinatorial interactions of Serpent, Lozenge, and U-shaped regulate crystal cell lineage commitment during Drosophila hematopoiesis PNAS, September 30, 2003; 100(20): 11451 - 11456. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
