Identification of fat-cell enhancer regions in Drosophila melanogaster

Insect Molecular Biology (2002)

11

(1), 67–77

© 2002 Royal Entomological Society

67

Blackwell Science Ltd

Identification of fat-cell enhancer regions in

Drosophila melanogaster

J. M. Miller

1

, T. Oligino

2

, M. Pazdera

2

, A. J. López

2

and D. K. Hoshizaki

1

1

Department of Biological Sciences, University of Nevada, Las Vegas, 4505 Maryland Parkway, Box 454004, Las Vegas, Nevada 89154-4004, USA;

2

Department of Biological Sciences, Carnegie Mellon University, 4400 Fifth Ave, Pittsburgh, PA 15213, USA

Abstract

The insect fat body is a dynamic tissue involved inmaintaining homeostasis. It functions not only inenergy storage and intermediary metabolism but alsoin detoxification, communication and the immuneresponse. Some of these functions are confined todistinct groups of fat body cells. In

Drosophila mela-nogaster

, discrete precursor-cell clusters populate thefat body [Hoshizaki, D.K., Blackburn, T., Price, C.,Ghosh, M., Miles, K., Ragucci, M. and Sweis, R. (1994)Embryonic fat-cell lineage in


.

Development

120: 2489–2499; Hoshizaki, D.K., Lunz,R., Ghosh, M. and Johnson, W. (1995) Identification offat-cell enhancer activity in


using P-element enhancer traps.

Genome

38: 497–506;Riechmann, V., Rehorn, K.P., Reuter, R. and Leptin, M.(1998) The genetic control of the distinction between fatbody and gonadal mesoderm in

Drosophila

.

Develop-ment

125: 713–723]. Whether these clusters populatedefined morphological regions or whether they repres-ent the precursors to functionally similar groups of fat-body cells has not been formally demonstrated. Wehave identified a 2.1 kb enhancer region from

serpent

(

srp

), a GATA transcription factor gene that is sufficientto induce fat-cell formation. This enhancer regiondrives expression in specific groups of precursor-cellclusters, which we show give rise to defined regions ofthe mature embryonic fat body. We present evidencethat

srp

expression in different precursor fat cellsis controlled by independent

cis

-acting regulatory

regions, and we have tested the role of

trans

-actingfactors in the specification of some of these cells. Wesuggest that the different positional cues regulating

srp

expression, and therefore general fat-cell specification,might also be involved in the functional specializationof fat cells. This may be a common mechanism ininsects to explain the origin of biochemically distinctregions of the larval /adult fat body.

Keywords: enhancers,

Drosophila

, fat body, fat-celllineage, mesoderm.

Introduction

The insect fat body is a dynamic tissue that participatesin multiple biochemical functions, including energy storage,intermediary metabolism, detoxification, communicationand the immune response. The insect fat cell is the mainsite for the storage and production of proteins, lipids andcarbohydrates. It produces a variety of stage-specific,amino acid-storage proteins, including calliphorin in blow-flies (

Calliphora

) and drosophilin and hexamerins (e.g. thelarval serum proteins) in

Drosophila

(Keeley, 1985 andreferences therein). The fat body is also the primary bio-synthetic site of diacylglycerol and the major insect sugar,trehalose. Both diacylglycerol and trehalose are energy-storage molecules that are key to the survival of the animaland central for energy-intensive behaviour such as long-distance migration (e.g.

Locusta migratoria

; Becker

et al.

,1996; Vroemen

et al.

, 1998). Fat-body cells produce anumber of other significant proteins including vitellogeninsfor oocyte maturation in

Drosophila

and mosquitoes(

Aedes aegypti

) and diapause proteins and haemoglobin inmidge fly larvae (

Chironomus thummi

; reviewed in Keeley,1985). The fat body has been compared to the vertebrateliver in its role in trehalose biosynthesis and release andthrough its response to adipokinetic signalling, which isanalogous to the hormone (insulin)-mediated synthesisand release of glucose by the liver (Becker

et al.

, 1996).Interestingly, the fat body of the desert ant (

Cataglyphisniger

) is the major synthetic site of hydrocarbons involvedin communication among colony members (Soroker &Hefetz, 2000). The fat body also plays a central role in theinsect innate immune response (reviewed in Hoffmann

et al.

, 1996; Engstrom, 1999).

Received 15 August 2001; accepted after revision 12 October 2001. Cor-respondence: D. K. Hoshizaki, Department of Biological Sciences, Univer-sity of Nevada, Las Vegas, 4505 Maryland Parkway, Box 454004, Las Vegas,NV 89154-4004, USA. Tel.: +1 702 895 3266; fax: +1 702 895 3956; e-mail:[email protected]

IMB_310.fm Page 67 Wednesday, January 16, 2002 1:39 PM

68

J. M. Miller

et al

.

© 2002 Royal Entomological Society,

Insect Molecular Biology

,

11

, 67–77

The fat body is one of several tissues produced by theembryonic mesoderm (Hartenstein & Jan, 1992; Hoshizaki

et al.

, 1994; Technau, 1987). In general, the primordia foreach tissue type lie in defined positions determined bythe anterior–posterior and dorsal–ventral patterning of themesoderm. In

Drosophila

, pair-rule genes such as

even-skipped

help to establish intrinsic differences betweenmesodermal cells in a metameric pattern along the anterior–posterior axis (Azpiazu

et al.

, 1996), while along thedorsal–ventral axis,

decapentaplegic

(

dpp

) expression inthe dorsal embryonic ectoderm specifies the dorsal meso-derm through the maintenance of

tinman

(

tin

) expression(Frasch, 1995; Staehling-Hampton

et al.

, 1994). The estab-lishment and maintenance of morphologically distinct celltypes are controlled by homeotic genes that determine seg-ment identity through the regional activation of target genesnecessary for cell specification (Weatherbee & Carroll,1999; Weatherbee

et al.

, 1998). These patterning systemsprovide mesodermal cells with unique addresses thatserve as positional cues to establish distinct populationsof cells.

The fat-body precursors are organized in a metamericpattern. Within a given range of parasegments, they arelocated in defined dorsal-ventral positions (reviewed inRiechmann

et al.

, 1997). It is likely that the specificationof precursor fat cells requires the integration of differentanterior–posterior and dorsal–ventral positional cues andhomeotic gene information. The maturation of the fat bodyrequires the coordinated migration of cells from the fat-cellclusters to their final positions, where they are organizedinto a functional organ composed of three morphologicaldomains: the lateral fat body, the dorsal fat-cell projections,and the ventral fat-cell commissure (Campos-Ortega &Hartenstein, 1997; Hoshizaki

et al.

, 1994; Riechmann

et al.

,1998). A variety of studies have suggested that the fat bodyis also divided into different regions based on the diversebiochemical functions of this tissue (reviewed in Haunerland& Shirk, 1995).

We describe here a 2.1 kb enhancer region from the

serpent

(

srp

) promoter that drives expression in a subset offat cells. The

srp

gene is one of three known

D. melanogaster

genes belonging to the GATA transcription-factor family(Brown & Castelli-Gair Hombria, 2000; Lin

et al.

, 1995;Ramain

et al.

, 1993; Winick

et al.

, 1993) and was initiallyidentified as a transcriptional activator of the

alcohol dehy-drogenase

(

Adh

) gene in fat cells (Abel

et al.

, 1993). SRPprotein is present at embryonic stage 10/11 (embryonicstages are those of Campos-Ortega & Hartenstein, 1997)in all fat-cell precursors (Sam

et al.

, 1996) and is the earli-est known gene to be expressed in the fat-cell lineage (Abel

et al.

, 1993; Rehorn

et al.

, 1996; Sam

et al.

, 1996). Geneticanalysis has revealed that

srp

is necessary for the mainten-ance of the fat-cell lineage (Sam

et al.

, 1996) and is sufficientto induce fat-cell formation within the mesoderm (Hayes

et al

., 2001). Thus, the activation of

srp

is a likely step in fat-cell specification. Because precursor fat cells arise fromstereotypic anterior–posterior and dorsal–ventral positionswithin the mesoderm, it is possible that a combination ofpositional information directs

srp

expression through differ-ent enhancers located within the

srp

regulatory region.The

srp

regulatory/promoter region spans at least 8 kb,based on the location of

srp

regulatory mutations (Rehorn

et al.

, 1996). Sequences within this region were identified asan

in vitro

target of the homeodomain transcription factor,Ultrabithorax (UBX) and contain putative UBX responseelements (Mastick

et al.

, 1995). We describe enhanceractivity that is associated with this region. By using thisenhancer region, we have traced the origin of specific mor-phological regions of the fat body and tested factors thatmight

trans

-activate

srp

. We discuss the possibility that thepositional information that directs the specification of pre-cursor fat cells is also responsible for the eventual biochem-ical differences found in the larval fat body.

Results

Developmental studies of the embryonic fat bodydemonstrate that fat cells originate from specific precursor-cell clusters that lie in the lateral, ventral, and dorsalmesoderm (Fig. 1; Hoshizaki

et al.

, 1994; Hoshizaki

Figure 1. Fat-cell development. (A, C, E) Lateral views of whole-mount embryos immunostained for SRP protein and (B, D, F) corresponding schematic drawings highlighting precursor fat-cell clusters and fat body domains. (A, B) Stage-10 embryos. The dorsal cell cluster is located in the dorsal mesoderm of PS13 and the primary cell clusters are located in the lateral mesoderm of PS 4–9. (C, D) Stage-11/12 embryos. The secondary ventral cell clusters are located in the ventral mesoderm of PS 3–11. A second group of subsidiary precursor fat cells are located as small cell clusters immediately posterior to the primary cell clusters in PS 4–9 and in the equivalent position in PS 10–12. (E, F) Stage-16 embryos. The fat body is made up of three morphological domains: the dorsal fat-cell projections, which extend in the anterior direction from the posterior–dorsal region of the lateral fat body; the lateral fat body which spans the lateral region of the embryo; and the ventral commissure, which extends from the anterior-lateral fat body and spans the ventral midline.


Fat-cell enhancer regions and fat body domains

69



,

11

, 67–77

et al

., 1995; Riechmann

et al

., 1998). At stage 10/11,precursor fat cells lie in the lateral mesoderm in a meta-merically repeating pattern (Fig. 1A). These cells makeup the primary cell clusters and are organized as seriallyduplicated clusters positioned in the

even-skipped

domain of parasegments (PS) 4–9 (Fig. 1A,B). Thesecell clusters are likely to populate most of the lateral fatbody (Hoshizaki

et al

., 1994; Riechmann

et al

., 1997).Within the dorsal mesoderm of PS 13 lies a large clusterof cells (Fig. 1A,B); this particular group of cells is likelyto make up the dorsal fat-cell projections (Fig. 1E,F;Riechmann

et al

., 1998). At stage 11, two secondarysets of cell clusters are identified that are serially dupli-cated along the anterior–posterior axis in the lateral andventral mesoderm (Fig. 1C,D; Riechmann

et al

., 1998).Each morphological region of the fat body is thought toarise from spatially distinct precursor-cell clusters, butthe lack of cell markers for specific precursor fat cells hasmade it difficult to confirm this idea.

Putative UBX response elements are associated with enhancer activity in a subset of

srp

-expressing cells

We have completed a genomic walk of the

srp

locus andhave positioned the

srp

gene within this region (Fig. 2Aand see Rehorn

et al.

, 1996). Within this walk is theA7.1ES fragment, which contains three putative UBXprotein-binding sites (Fig. 2B; Mastick

et al.

, 1995). We havetested this region for enhancer activity

in vivo

. A7.1ESwas subcloned into the P-element vector pCaSpeR-hs43-

lacZ

and transgenic animals were generated. Theactivity of the A7.1ES-

lacZ

reporter was studied inwhole-mount embryos. We found that the A7.1ES-

lacZ

reporter was active in a subset of

srp

-expressing cells,including cells that make up specific regions of the fatbody (Figs 3, 4).

The wild-type expression pattern of SRP protein hasbeen described in detail (Sam

et al.

, 1996). Briefly,

srp

isexpressed in fat-cell precursors and in the fat body. It is also

Figure 2. Map of genomic srp and sequence of A7.1ES. (A) A schematic representation of the srp gene and a portion of its regulatory region. λ genomic clones encompassing the srp walk are indicated as horizontal lines above the map and the location of the A7.1ES and A7.1EB fragments are shown below the map. Short arrows indicate the positions of the UBX response elements (UREs) previously identified in a yeast one-hybrid assay (Mastick et al., 1995). As mapped by Rehorn and co-workers (Rehorn et al., 1996), the two P-element inserts, srpas and srpPZ, are shown as triangles below the map and the location of the srp transcript is indicated by an arrow. Restriction sites are: B, BamHI; E, EcoRI; S, Sal I; Ss, SstI; X, XbaI; Xh, XhoI. (B) Genomic sequence of A7.1ES (GENBANK accession no. AF707404). Arrows indicate extended core sequences of the UREs and boxes indicate potential TIN-binding sites based on the published consensus TIN-binding site, 5′-TCCAAGTGG-3′ (Gajewski et al., 1997). Underlined are the EcoRI site (5′-GAATTC-3′) at position 1, the BamHI site (5′-TGCGCA-3′) at position 464 and the Sal1 site (5′-GTCGAC-3′) at position 1143. The A7.1ES sequence is defined as a EcoRI–Sal1 fragment (positions 1–1148) and the A7.1EB sequence is defined as an EcoRI–BamHI fragment (positions 1–469).


70

J. M. Miller

et al

.



,

11

, 67–77

expressed in several other tissues: the primordia of theanterior and posterior midgut; the cephalic mesoderm andhaemocytes; the amnioserosa primordium and amnio-serosa; and the lymph glands (see Fig. 2 in Sam

et al.

, 1996).Within the fat body, we detected A7.1ES-

lacZ

reporteractivity in only a subset of fat-body cells: the ventral- andposterior-most edge of the mature embryonic lateral fatbody, the dorsal fat-cell projection, and a portion of theventral commissure (Fig. 3). Using A7.1ES-

lacZ

as a cell-lineage marker, we have traced the origins of the aforemen-

tioned groups of fat body cells (Figs 4, 5). A7.1ES-

lacZ

reporter activity was detected at stage 10 in a single clusterof cells located within the dorsal mesoderm of paraseg-ment (PS) 13 (Fig. 4A). The cells of this cluster populatethe fat-cell projections and the posterior-most cells of thelateral fat body and are described in more detail in thefollowing section. By stage 11, the number of A7.1ES-

lacZ-

expressing cells in the aforementioned dorsal cellcluster has increased and we began to detect expression incells located in the ventral region of the mesoderm (Fig. 4B).By stage 12, strong expression was detected in the ventralmesoderm in serially duplicated cell clusters composed of4–6 cells located in PS 3–11 and in a smaller cluster in PS 12(Fig. 4C). The ultimate fate of the small cluster in PS 12 isnot known, because

β

-galactosidase activity ceased to bedetected in its cells after stage 13. By stage 14, the cell clustersin PS 3–11 fused to form a 1–2 cell-wide row that defines theventral edge of the lateral fat body (Fig. 4G,H). These cell clus-ters correspond to the secondary ventral fat-cell clusters.

The ventral secondary cell clusters in PS 3–5 alsocontribute to the ventral commissure, as further describedin detail below (Riechmann

et al.

, 1998). At stage 16, A7.1ES

Figure 3. A7.1ES is active in a subset of fat body cells.(A) Wild-type, stage-15 embryo stained for A7.1ES-lacZ activity (brown) and Adh transcripts (blue). (B) Enlargement of (A). A7.1ES is active in the dorsal fat-cell projections (overline) and in the posterior-most and ventral-most cells (bracket) of the lateral fat body. A7.1ES is also active in portion of the ventral commissure (see Fig. 5 and text for more detail).

Figure 4. A7.1 ES-lacZ reporter recapitulates a portion of the srp expression pattern. (A) Lateral view of a stage-10 embryo. Reporter activity is first detected in the cephalic mesoderm and in PS13 in a dorsal cluster of cells, which eventually gives rise to the dorsal fat-cell projection and the posterior-most lateral fat body. (B) Lateral view of a late stage-11 embryo. Prohaemocytes are detected as they migrate from the cephalic mesoderm (underline). Transient expression is detected in single cells in the ventral mesoderm (arrow) and later persists as nine clusters of cells. (C) Lateral and (D) dorsal view of a stage-12 embryo. (C) Reporter gene activity persists in bilateral clusters of cells located in PS 3–11, which are the ventral secondary cell clusters, and is detected in a small bilateral cluster of cells in PS 12. (D) Another small bilateral cluster of cells is detected in the ventral mesoderm of PS 13, adjacent to the dorsal cell cluster. Reporter gene activity is detected in the amnioserosa cells abutted to the posterior edge of the retracting germ band (asterisk in (D)) and the in large dorsal cell cluster in PS 13. Reporter gene activity is also detected in cells likely to be the precursors to the longitudinal visceral muscle fibres (bracket in (C)). (E) Dorsal and (F) lateral view of a late stage-12 embryo. Reporter gene activity is detected in the posterior amnioserosa cells (asterisk in (E)) and in the putative precursor longitudinal visceral muscle cells (bracket in (F)). The ventral cell cluster in PS13 has fused across the midline (arrow in (E)) and the bilateral ventral cluster in PS 12 is absent. (G) Lateral view of a stage 14/15 embryo. Secondary ventral cell clusters have begun to fuse to form the ventral edge of the lateral fat body. (H) Lateral and (I) ventral and (J) dorsal view of early stage-16 embryos. Reporter gene activity is present (H) in the ventral (brackets) and posterior cells of the lateral fat body (I) in one of the two bridges that form the ventral commissure (bracket) (see text and Fig. 5 for details) and (J) in the dorsal fat-cell projections (arrowheads) and posterior amnioserosa cells (star).


Fat-cell enhancer regions and fat body domains 71

© 2002 Royal Entomological Society, Insect Molecular Biology, 11, 67–77

was active in a portion of the ventral commissure thatmakes up the posterior bridge, including the bilateral horns,but was not active in the anterior bridge (Fig. 4I). Theventral commissure lies in the anterior region of the embryoand eventually spans the ventral midline (Fig. 5). It consistsof at least two fat-cell bridges that arise from cells of thesecondary ventral cell clusters in PS 3–5. The anteriorbridge is derived from cells in PS 3/4 and is located moreinternally than the posterior bridge (Fig. 5C). The posteriorbridge is derived from cells in PS 5 and is associated withthe ventral most cells of the lateral fat body (Fig. 5C). Bilat-eral horns of fat cells extend from the posterior bridge andcross the anterior bridge (Fig. 5).

The A7.1ES-lacZ reporter also was active in the cephalicmesoderm and in the prohaemocytes as they migrate fromthe cephalic mesoderm (Fig. 4B). Reporter gene activity,however, was absent in the majority of haemocytes bystage 14. A7.1ES-lacZ reporter expression was also detectedin the amnioserosa, but only in the posterior-most cells(Fig. 4D,E). At stage-12, we detected the novel expressionof A7.1ES-lacZ in cells that arose from the caudal pole(Fig. 4C). These cells subsequently appeared as rows ofcells ensheathing the midgut (Fig. 4F). Based on the mor-phology and position of these cells (Campos-Ortega &Hartenstein, 1997), it is likely that A7.1ES is active in thelongitudinal visceral muscle fibres and their precursors.

Expression of srp in the dorsal fat-cell cluster is controlled separately from the remaining fat body cells

A7.1ES has strong activity in posterior srp-expressingcells, i.e. the posterior amnioserosa and the dorsal cellcluster. We have restricted this posterior enhancer activityto the distal portion of A7.1ES, designated A7.1EB (Fig. 1A).Transgenic lines carrying the A7.1EB-lacZ reporter exhib-ited activity only in the posterior fat cells and their precur-sors and in the posterior amnioserosa cells (Fig. 6). Lyingpartially beneath the posterior precursor fat cells are srp-expressing cells that make up the posterior midgut primor-dium. The A7.1EB enhancer is not active in these cells. Thesimple expression pattern driven by A7.1EB has allowed usto carry out a detailed examination of the origin of theposterior fat cells, i.e. the dorsal fat-cell projections and the

Figure 5. The ventral commissure is composed of two fat-cell bridges.(A) Ventral and (C, D) ventral-lateral views of stage-16 wild-type embryo immunostained for SRP protein. (B) Schematic of the ventral commissure. The ventral commissure is composed of fat cells that span the ventral midline. The anterior bridge of cells extends from the secondary ventral cell clusters in PS 3 and 4 and fuse across the ventral midline. The posterior bridge extends from cells in the bilateral ventral cell cluster in PS 5 and also fuse across the ventral midline. (C) The posterior bridge lies in a more peripheral position within the embryo than the anterior bridge and is a continuation of the ventral-most cells of the lateral fat body. (D) Extending from the posterior bridge are two horns that extend in the anterior direction and crossover the anterior bridge (Second horn is out of the plane of focus).

Figure 6. The A7.1EB-lacZ reporter is expressed in the posterior of the embryo. (A) Lateral and (B) dorsal view of a stage-11 embryo. Reporter gene activity is detected in the dorsal cell cluster of PS13 (square bracket) and in a small bilateral ventral cell cluster (arrow). (C) Lateral and (D) dorsal view of a stage-12 embryo. As the germband retracts, the bilateral ventral clusters of cells in PS 13 fuse across the ventral midline. Reporter gene expression is also detected in the posterior-most amnioserosa cells (asterisk in (C, D)) and in the dorsal cell cluster (bracket in (C, D)). (E) Lateral and (F) dorsal view of a late stage-12 embryo. As the germband fully retracts, the dorsal cell cluster begins to separate into two subgroups of cells (square brackets, 1 and 2) and the ventral cluster becomes located below the dorsal cell cluster (arrow in (E)) and is out of the plane of focus in (F). Reporter gene expression is still detected in the posterior amnioserosa cells (star). (I) Lateral and (J) dorsal view of a stage-16 embryo. A subgroup of cells from the dorsal cell cluster (bracket 1 from (G)) forms the dorsal fat-cell projection (bracket 1 in (I, J)) and the second subgroup (bracket 2 from (G)) along with the ventral cell cluster of PS 13 form the posterior-most portion of the lateral fat body.


72 J. M. Miller et al.


posterior–lateral fat body, without the difficulties presentedby the underlying SRP-expressing posterior midgut primor-dium. Using A7.1EB-lacZ, we mapped the dorsal cell clus-ter in the mesoderm relative to the engrailed (en) stripes inthe ectoderm (Fig. 7). The majority of the dorsal cell clusterlies between en stripes 13 and 14, which mark the anteriorportions of the parasegments in the ectoderm (Lawrence,1992). A few dorsal cells lie beneath the en stripes, but arenot positioned beyond the boundaries of these two stripes.

Using the A7.1EB-lacZ reporter, we detected strongexpression at stage 11 in the dorsal cell cluster of PS 13and in a single cell located immediately ventral to this largecluster (Fig. 6A,B). By early stage 12, the single cell formeda small cluster of approximately 4–6 cells. Although thiscell cluster is located in the ventral mesoderm, its behaviourat later stages revealed that it is distinct from the secondaryventral cell clusters. During germband retraction, the bilat-eral ventral PS 13 cell cluster fused across the ventralmidline (Fig. 6D) and later contributed to the posterior-mostcells of the lateral fat body (Fig. 6I,J). At late stage 12, thedorsal cell cluster began to separate, and by stage 13, twodistinct subgroups of cells could be identified (Fig. 6E,G).The dorsal-most subgroup forms the dorsal fat-cell projec-tion and the other subgroup coalesces with the fusedventral cluster to form the posterior region of the lateral fatbody (Fig. 6I,J).

UBX affects the morphology of the fat body

The precursor fat-cell clusters are located in defined posi-tions within the specific segments of the mesoderm.Because UBX-target sequences map to A7.1EB (Fig. 2),we tested whether the Ubx gene might play a role as a tran-scriptional regulator of srp. Because Ubx is expressed inthe mesoderm of PS 6–12 but not of PS 13 (Akam &Martinez-Arias, 1985; White & Wilcox, 1984), Ubx might func-tion to repress the formation of a dorsal cell cluster inanterior segments. We examined the fat-cell phenotypescaused by both the loss-of-function and mis-expression of

Ubx. The mis-expression of Ubx throughout the mesodermwas achieved by employing the GAL4/UAS targeted geneexpression system of Brand & Perrimon (1993).

We found that Ubx does not have a role in the specifica-tion of the dorsal cell cluster. Loss of Ubx function did notinduce ectopic dorsal cell clusters in the anterior segmentsnor did mis-expression of Ubx throughout the mesoderm(including PS13) repress dorsal fat-cell specification (datanot shown). We note, however, that mis-expression of Ubxin PS 13 leads to morphological alterations of the dorsal fat-cell projections. The fat-cell projection is normally 6–8 cellsin width at its broadest region. In the Ubx mis-expressionembryos, the dorsal fat-cell projections extended properlyin the anterior direction, but they were composed of only asingle row of cells (Fig. 8B). This change in cell numberwas also reflected in the number of the A7.1EB-lacZ posi-tive cells (data not shown). On the other hand, the loss ofUbx occasionally caused a slight reduction in the number ofcells present in the projections (Fig. 8C). Taken togetherthese data suggest that Ubx does not play a direct role in

Figure 7. The dorsal cell cluster is located between en stripes 13 and 14. (A) Lateral and (C) dorsal view of a stage 11 A7.1EB-lacZ embryo immunostained for ENGRAILED and β-galactosidase protein. (B) is a higher magnification of (A). The dorsal cell cluster lies in the mesoderm beneath the ectodermal en stripes 13 and 14. Cells of the cluster extend within the boundaries of the stripes but do not extend beyond them.

Figure 8. Mis-expression of Ubx leads to a morphological alteration of the dorsal fat-cell projection. (A) Dorsal view of wild-type (B) twi-GAL4; UAS-Ubx, and (C) Ubx mutant stage-16 embryos immunostained for SRP protein. (A) wild-type dorsal fat-cell projection (arrow) (B) Mis-expression of Ubx reduces the width of the projections (arrow) to 1–2 cells compared to the normal 6–8-cell width and causes the loss of the lymph glands. (C) Loss of Ubx has little or no effect on the number of cells contributing to the projection (arrow). The lymph glands have hypertrophied (arrowhead) as previously described by Mastick et al. (Mastick et al., 1995).




the trans-activation of srp, but that it is involved in determin-ing segment identity and can affect the differentiation ofcells that contribute to the dorsal fat-cell projections.

Surprisingly, Ubx had a dramatic effect on the formationof the ventral commissure, even though the precursors tothe commissure lie in PS 4–5 and not in the UBX domain.It has been previously shown that loss of Ubx activityresults in strong derepression of srp within the secondaryventral cell clusters in PS 6–12 (Fig. 9B) and see Fig. 4 in(Mastick et al., 1995). We have followed up on this obser-vation and have found that in Ubx mutants, a third bridgearises from PS 6 (Fig. 9E). Thus, it appears that UBX mightbe directly or indirectly involved in repressing the formationof fat-cell bridges. To further test this involvement, we askedwhether UBX is sufficient to suppress the formation of theendogenous ventral commissure by employing the GAL4/UAS system of Brand & Perrimon (1993). In twi-GAL4;UAS-Ubx embryos, we found that both bridges of the ven-tral commissure were absent (Fig. 9F). These data suggestthat Ubx is involved in suppressing the differentiation of thecommissure in posterior segments through its role as asegment identity gene. Ubx might repress a hierarchy ofgenes that control whether the fat-cell bridges will arisefrom the ventral-most cells of the lateral fat body.

Role of tin and dpp in the specification of the secondary ventral and dorsal cell clusters

The dorsal cell cluster lies in the dorsal mesoderm. Wehypothesize that only cells within the dorsal mesoderm arecompetent to acquire a dorsal fat-cell fate, and that srp,through the A7.1EB enhancer, might be responsive tofactors that establish the dorsal mesoderm. Key to the forma-tion of the dorsal mesoderm is the NK homeodomainprotein tin (Kim & Nirenberg, 1989). tin is expressed inthree phases during embryonic development. In the firstphase, it is expressed in a pan-mesodermal fashion. In the

second phase, tin expression is lost within the mesodermexcept in the dorsal region where it is maintained by theaction of dpp in the overlying ectoderm. In the third phase,tin expression is restricted to and marks the precursor tothe heart (Azpiazu & Frasch, 1993; Bodmer, 1993; Yin &Frasch, 1998).

Based on the different characterized roles for tin, wepredicted that the second phase of tin expression is importantfor the eventual establishment of the dorsal cell cluster.Moore et al. (1998), however, suggest that early tin expres-sion is important in the specification of fat cells based upona comparison of the phenotypes associated with the loss oftin and dpp activity. This comparison can distinguish betweenthe first and second phase of tin expression because thesecond phase, but not the first phase of tin expression, isdependent upon dpp. In tin mutants, in addition to the lossof visceral, heart and dorsal somatic muscle (Azpiazu &Frasch, 1993; Bodmer, 1993), there is a loss of some lateralfat cells based upon the location and number of srp-expressing cells (see Fig. 4 in Moore et al., 1998). In a dppmutant, more fat cells appear to be present than in a tin mutant,although because of the gross developmental defectsassociated with the loss of dpp, it is difficult to determinewhether any fat cells are absent (see Fig. 4 in Moore et al.,1998). Based upon this comparison, it was suggested thatearly tin is involved in the specification of fat cells (Mooreet al., 1998).

We have re-examined tin’s role in fat-cell developmentusing the A7.1EB-lacZ reporter gene as a cell marker forthe dorsal cell cluster and the dorsal fat-cell projections.The reporter gene was introduced into both tin- and dpp-mutant backgrounds using standard genetic crosses. Wefound tin-mutant embryos had few, if any, cells in the dorsalcluster, and concomitantly, there was a loss of dorsal fat-cell projections (and the posterior-most cells of the lateralfat body) (Fig. 10C,D). To determine whether this loss is due

Figure 9. Mis-expression of Ubx leads to morphological alterations of the ventral commissure. (A) Ventral-lateral view of stage-13 wild-type (B) Ubx-mutant and (C) twi-GAL4; UAS-Ubx embryo immunostained for SRP protein. (B) Loss of Ubx results in the de-repression of srp in the ventral cell clusters in PS 6–12 while (C) mis-expression of UBX throughout the mesoderm has little or no effect on srp expression in the ventral cell clusters. (D) Dorsal view of stage-16 wild-type (B) Ubx-mutant and (C) twi-GAL4; UAS-Ubx embryos immunostained for SRP protein. (B) Loss of Ubx results in the formation of an ectopic fat-cell bridge from PS 6 (bracket) while (C) mis-expression of Ubx throughout the mesoderm represses the formation of the ventral commissure.

IMB_310.fm 3 Wednesday, January 16, 2002 1:39 PM



to the first or second phase of tin expression, we examineddpp-mutant embryos. We found that reporter gene activitywas nearly absent in the location of the dorsal cell cluster,and concomitantly the dorsal fat-cell projections wereabsent (Fig. 10F,G). These data suggest that the specifica-tion of the dorsal cell cluster requires formation of the dorsalmesoderm and depends upon the second phase of tinexpression. To further examine the role of dpp-dependentexpression of tin, we examined twi-GAL4; A7.1EB-lacZ;UAS-dpp embryos where the dpp was expressed through-out the mesoderm. In these experimental embryos, the dor-sal cell cluster was expanded to fill most of PS 13, and thedorsal fat-cell projection was not formed (Fig. 10H–J).

Discussion

Over 20 years ago, Rizki & Rizki (1978) suggested that thelarval fat body has a segmental origin. Recent studies havefirmly established the segmental nature of the precursor fatcells and have identified different classes or groups ofmetamerically repeating cell clusters that contribute to theembryonic fat body (Abel et al., 1993; Hoshizaki et al.,1994; Riechmann et al., 1998). In our efforts to understandthe molecular mechanism underlying the commitment toand specification of a fat-cell fate, we have focused ourattention on srp, a transcription factor gene involved in thespecification (Hayes et al., 2001) and differentiation of fatcells (Sam et al., 1996). We present here the first analysisof srp regulatory regions active in the secondary ventral cellclusters and the dorsal cell cluster. Because the precursorfat cells arise from unique anterior–posterior and dorsal–ventral positions within the mesoderm, it is likely that acombination of positional information and input from differ-

ent homeotic genes are involved in the establishment andmaintenance of fat cells that populate the different regionsof the fat body. The capability of srp to induce fat-cell forma-tion (Hayes et al., 2001) suggests that this patterning informa-tion is integrated at the srp locus through the use of thedifferent srp fat-cell enhancers. We have examined UBX asa possible direct activator of srp, but find no evidence thatUbx plays a role in establishing the dorsal cell cluster or thesecondary ventral cell clusters. Although Ubx-mutantembryos displayed morphologically altered fat bodies, Ubxdoes not appear to directly control srp’s activation orrepression, rather that Ubx appears to function in maintain-ing the segment-specific characteristics of fat-cell clusters.

In the dorsal–ventral axis, we find that the dorsal cellcluster and secondary ventral cell clusters are sensitive todorsal–ventral patterning cues. The persistent expressionof tin in the dorsal region of the mesoderm subdivides themesoderm. The secondary ventral cell clusters are locatedin the ventral mesoderm in PS 3–11, and thus lie outside thistin domain. The loss of tin activity has no effect on the spec-ification of these cells (data not shown); however, the spec-ification of these cells can be repressed by the expansionof the dorsal mesoderm into the ventral region by the mis-expression of dpp through the mesoderm (data not shown).

The dorsal cell cluster lies within the dorsal mesoderm.Loss of the dorsal mesoderm (through the loss of tin or dppexpression) results in the loss of the dorsal cell cluster;conversely, extension of the dorsal mesoderm results inthe expansion of the dorsal cell cluster. tin clearly plays agenetic role in the specification of the dorsal cell cluster. Atthis time, we cannot distinguish between a direct or indirectrole of tin in this process. TIN might be a direct transcrip-tional activator of srp or TIN might indirectly activate srp

Figure 10. Specification of the dorsal cell cluster requires the formation of the dorsal mesoderm. Lateral views of (A, B) A7.1EB-lacZ wild-type and (C, D) A7.1EB-lacZ; tin-mutant embryos immunostained for β-galactosidase. (C) In a tin-mutant embryo at stage 12, the dorsal cell cluster consists of a few cells as revealed by the loss of lacZ-expressing cells and results in (D) the loss of the fat cell projections (and posterior-most cells of the lateral fat body). (A1) and (C1) are higher magnification of (A) and (C), respectively. Dorsal view of (E) A7.1EB-lacZ wild-type and (F) dpp; A7.1EB-lacZ embryos. (F) A dpp-mutant embryo (approximately stage-12) has reduced number of cells in the dorsal cell cluster. The PS 13 ventral cell cluster appears unaffected, and the wild-type fusion of these two cell clusters takes place across the ventral midline. (G) A dpp-embryo (approximate stage-16) exhibits little to no reporter gene activity. Dorsal view of (H) stage-12 and (I) stage-15 twi-GAL4; UAS-dpp; A7.1EB-lacZ embryos immunostained for β-galactosidase. Mis-expression of dpp causes the ventral expansion of the dorsal fat-cell cluster to occupy most of PS 13. (J) Lateral view of (I).




through downstream factors intrinsic to the dorsal meso-derm. It is likely, however, that the activation of srp in thedorsal cell cluster will require both TIN and other dorsalmesoderm-intrinsic factors. We find that within the A7.1EBsequence there are TIN-binding sites at positions 53–36, 200–181 and 257–274 that contain single base-pairdifferences from the consensus TIN binding site5′[TCAAGTGG]3′ (Gajewski et al., 1997). The presence ofthese sites raises the possibility that specification, at leaston the dorsal–ventral axis, requires transcriptional activa-tion of srp through the binding of TIN to A7.1EB. We hopethat further dissection of the A7.1EB sequence will help toidentify the relevant cis-acting sequences that play a role inintegrating the combination of positional information cuesthat activate srp and specify the dorsal cell cluster.

Our analysis of srp regulatory regions reveals that differ-ent enhancers control srp expression in specific precursor-cell clusters, and that these precursor cells give rise to fatcells that populate specific regions of the fat body. Addi-tional factors intrinsic to the region of the mesoderm fromwhich the precursors arise are likely to help define the over-all morphology of the fat body. It has not escaped our atten-tion that our observations also provide a model for theorigin of functionally distinct regions in the fat body. The fatbody is a diverse and extremely complex tissue which isinvolved in many metabolic processes; it is not a function-ally homogenous tissue and some of its functions arerestricted to cells lying within specific regions of this tissue(reviewed in Haunerland & Shirk, 1995). These so-calledfunctional domains have mainly been characterized withinthe orders diptera and lepidoptera. For several Drosophilaspecies, eye pigment biosynthesis is compartmentalizedwithin the fat body. The tryptophan pryrolase metabolitekynurenine is localized in the anterior region of the larval fatbody, whereas the pteridine-precursor isoxanthopterin isrestricted to the posterior regions (Rizki, 1961). Addition-ally, in Drosophila melanogaster, cells in the posterior larvalfat body store more protein granules than do the cells ofthe anterior fat body (Tysell & Butterworth, 1978). In theflesh fly (Sarcophaga peregrina), anterior fat body protein(AFP), which is similar to mammalian regucalcin (a calcium-binding liver protein), is almost exclusively expressed in theanterior regions of the larval fat body (Nakajima & Natori,2000). The fat bodies of the Indianmeal moth (Plodia inter-punctella) and the corn earworm (Helicoverpa zea) alsocompartmentalize the accumulation of storage proteinswithin distinct regions (reviewed in Haunerland & Shirk,1995). Because specific precursor-cell clusters populatedefined regions of the fat body, we suggest that the combi-natorial positional cues regulating srp expression mightalso participate in pre-patterning different biochemicalfunctions of the embryonic fat body. These positional cuesmight regulate as-yet unidentified genes in specific precur-sor fat cells to specify their biochemical capability and to

define different functional domains within the fat body.Further experiments to precisely map functional domainswithin the embryonic fat body along with the characteriza-tion of the remaining enhancers of srp will be necessaryto test this idea and to establish the genetic hierarchy lead-ing to fat-cell specification by srp and the specialization offat-body cells.

Experimental procedures

Genomic walk

A cDNA clone of srp (a generous gift from T. Abel) was used toscreen an EMBL-3 SP6/T7 l Drosophila melanogaster genomiclibrary (Clontech) using standard molecular procedures (Maniatiset al., 1982). The λ clones were restriction mapped and aligned.Our map is in agreement with the published map of Rehorn andco-workers (Rehorn et al., 1996).

Drosophila stocks

The transgenic lines carrying reporter gene constructs of A7.1EBor A7.1ES were generated by standard microinjection techniquesusing pCaSpeR-hs43-lacZ, a vector specifically designed to testfor enhancer and tissue-specific regulatory elements in transgenicflies (Thummel & Pirrotta, 1992). A7.1ES is a 1.2 kb EcoRI–Sal1fragment and A7.1EB is an EcoRI–BamHI 0.47 kb subclone ofA7.1ES. The Ubx9.22 mutant has a 1587 bp deletion that removesthe final 1.4 kb of the 50 kb intron, the splice acceptor site of the3′ exon, and 48 codons of the homeobox, and has a sequence ofsix nucleotides inserted between the deletion breakpoints (Wein-zierl et al., 1987). The tin346 mutant (generously provided by M.Frasch) has a 103 bp deletion of the transcription unit (Azpiazu &Frasch, 1993). dppH48 is a loss of function allele (Andrew et al.,1997; Held & Heup, 1996; Mason et al., 1997). Ubx and dpp wereectopically expressed using the GAL4-UAS system derived fromyeast (Brand & Perrimon, 1993). The stocks used in ectopicexpression experiments were: twi-GAL4 (Baylies & Bate, 1996;Greig & Akam, 1993), UAS-Ubx (Akam, 1996), and UAS-dpp(Staehling-Hampton & Hoffmann, 1994). These three lines wereprovided by the Bloomington Stock Center. The twi-GAL4 driver con-tains the entire regulatory region of the twi promoter and directsectopic expression in a pattern similar to that of the endogenoustwi gene. twi-driven expression is first detected in the mesodermand in the mesectoderm at gastrulation. Uniform expressionpersists in the mesoectoderm and in the mesoderm until early stage10 and late stage 11, respectively. twi-driven expression alsopersists in a subset of muscle-cell progenitors until at least stage12 (Baylies & Bate, 1996).

Immunohistochemical staining and in situ hybridization to whole-mount embryos

The generation of SRP antibody and immunohistochemistry towhole-mount embryos has been previously described (Hoshizakiet al., 1994; Sam et al., 1996). Goat anti-rabbit IgG alkaline phos-phatase conjugate, anti-β-galactosidase monoclonal antibody,and goat anti-mouse IgG alkaline phosphatase conjugate wereobtained from Promega. The anti-engrailed monoclonal antibodydeveloped by C. Goodman was obtained from the DevelopmentalStudies Hybridoma Bank (University of Iowa, Department of




Biological Sciences, Iowa City, IA 52242) (Patel et al., 1989).Double labelling with anti-β-galactosidase and Adh anti-sense RNAwas carried out as described by Lloyd and Sankonju (Lloyd &Sakonju, 1991). Adh anti-sense RNA was prepared from a 1.3 kbregion of Adh coding sequence subcloned from P13E-3 (Heberleinet al., 1985) and inserted into pGEM1 at the Sal1–BamHI sites(kindly provided by K. Hales). The RNA probe was synthesizedusing a digoxygenin-labelled uracil triphosphate as described bythe manufacturer (Boehringer Mannheim). Chromogenic sub-strates were X-phosphate and NBT (Boehringer Mannheim).

Microscopy

Embryos were equilibrated in mounting solution (50% glycerol,150 mM NaCl, 10 mM Tris-HCl pH 8.0). Microscopic analyses wereperformed on a Zeiss Axioplan2 microscope using Nomarskioptics. Whole-mount embryos were photographed onto Kodaktungsten slide film with a 35 mm camera attached to the micro-scope and digitized by scanning at 600 dpi using a U-max S-12scanner. Digital images of whole-mount embryos were taken witha Kodak MDS120 digital camera attached to the microscope. Bothtypes of images were assembled into figures using CorelDraw8software.

Acknowledgements

We thank T. Abel for the srp cDNA clone, M. Frasch forthe tin346 mutant stock and K. Hales for the Adh-containingplasmid. This work was supported in part by grants from theCouncil for Tobacco Research (no. 2270), the SouthernNevada EPSCoR Women in Science and Engineeringprogram awarded to D.K.H. and by a generous gift to D.K.H.from Dr Steven and Mrs Miriam Shearing (NorthstarFoundation). This research was also supported by a BasilO’Connor Award from the March of Dimes Foundation toA.J.L. and by grants awarded to J.M.M. by the GraduateStudent Association (University of Nevada, Las Vegas).J.M.M. has been supported by the Marjorie Barrick fellow-ship awarded by the Graduate College (University ofNevada, Las Vegas), a U.N.L.V. Regents Award Program(RAP) fellowship, and a Southern Nevada EPSCoRWomen in Science and Engineering (WISE) programfellowship, each of which are gratefully acknowledged.

References

Abel, T., Michelson, A.M. and Maniatis, T. (1993) A DrosophilaGATA family member that binds to Adh regulatory sequencesis expressed in the developing fat body. Development 119:623–633.

Akam, M. (1996) Personal communication to FLYBASE,(FBrf0131210).

Akam, M.E. and Martinez-Arias, A. (1985) The distribution of Ultra-bithorax transcripts in Drosophila embryos. EMBO J 4: 1689–1770.

Andrew, D.J., Baig, A., Bhanot, P., Smolik, S.M. and Henderson,K.D. (1997) The Drosophila dCREB-A gene is required fordorsal/ventral patterning of the larval cuticle. Development124: 181–193.

Azpiazu, N. and Frasch, M. (1993) tinman and bagpipe: twohomeo box genes that determine cell fates in the dorsalmesoderm of Drosophila. Genes Dev 7: 1325–1340.

Azpiazu, N., Lawrence, P.A., Vincent, J.P. and Frasch, M. (1996)Segmentation and specification of the Drosophila mesoderm.Genes Dev 10: 3183–3194.

Baylies, M.K. and Bate, M. (1996) twist : a myogenic switch inDrosophila. Science 272: 1481–1484.

Becker, A., Schloder, P., Steele, J.E. and Wegener, G. (1996) Theregulation of trehalose metabolism in insects. Experientia 52:433–439.

Bodmer, R. (1993) The gene tinman is required for specification ofthe heart and visceral muscles in Drosophila. Development118: 719–729.

Brand, A.H. and Perrimon, N. (1993) Targeted gene expression asa means of altering cell fates and generating dominant pheno-types. Development 118: 401–415.

Brown, S. and Castelli-Gair Hombria, J. (2000) Drosophila grainencodes a GATA transcription factor required for cell rearrange-ment during morphogenesis. Development 127: 4867–4876.

Campos-Ortega, J.A. and Hartenstein, V. (1997) The EmbryonicDevelopment of Drosophila Melanogaster. Berlin/Heidelberg:Springer-Verlag.

Engstrom, Y. (1999) Induction and regulation of antimicrobialpeptides in Drosophila. Dev Comp Immunol 23: 345–358.

Frasch, M. (1995) Induction of visceral and cardiac mesoderm byectodermal Dpp in the early Drosophila embryo. Nature 374:464–467.

Gajewski, K., Kim, Y., Lee, Y.M., Olson, E.N. and Schulz, R.A.(1997) D-mef2 is a target for Tinman activation duringDrosophila heart development. Embo J 16: 515–522.

Greig, S. and Akam, M. (1993) Homeotic genes autonomouslyspecify one aspect of pattern in the Drosophila mesoderm.Nature 362: 630–632.

Hartenstein, V. and Jan, Y.N. (1992) Studying Drosophila embryo-genesis with P-lacZ enhancer trap lines. Wilhelm Roux’s ArchDev Biol 201: 194–220.

Haunerland, N.H. and Shirk, P.D. (1995) Regional and functionaldifferentiation in the insect fat body. Annu Rev Entomol 40:121–145.

Hayes, S., Miller, J. and Hoshizaki, D. (2001) serpent, a GATA-liketranscription factor gene, induces fat-cell development in Dro-sophila melanogaster. Development 128: 1193–1200.

Heberlein, U., England, B. and Tjian, R. (1985) Characterization ofDrosophila transcription factors that activate the tandem pro-moters of the alcohol dehydrogenase gene. Cell 41: 965–977.

Held, L.I. and Heup, M.A. (1996) Genetic mosaic analysis ofdecapentaplegic and wingless gene function in the Drosophilaleg. Dev Genes Evol 206: 180–194.

Hoffmann, J.A., Reichhart, J.M. and Hetru, C. (1996) Innate immu-nity in higher insects. Curr Opin Immunol 8: 8–13.

Hoshizaki, D.K., Blackburn, T., Price, C., Ghosh, M., Miles, K.,Ragucci, M. and Sweis, R. (1994) Embryonic fat-cell lineage inDrosophila melanogaster. Development 120: 2489–2499.

Hoshizaki, D.K., Lunz, R., Ghosh, M. and Johnson, W. (1995)Identification of fat-cell enhancer activity in Drosophila mela-nogaster using P-element enhancer traps. Genome 38: 497–506.

Keeley, L.L. (1985) Physiology and biochemistry of the fat body. In:Comprehensive Insect Physiology, Biochemistry and Pharma-cology (Kerkut, G.A. and Gilbert, L.I., eds), Vol. 3, pp. 211–248. Oxford: Pergamon Press.




Kim, Y. and Nirenberg, M. (1989) Drosophila NK-homeobox genes.Proc Natl Acad Sci USA 86: 7716–7720.

Lawrence, P.A. (1992) The Making of a Fly. Cambridge, MA:Blackwell Scientific Publications.

Lin, W.H., Huang, L.H., Yeh, J.Y., Hoheisel, J., Lehrach, H., Sun, Y.H.and Tsai, S.F. (1995) Expression of a Drosophila GATA tran-scription factor in multiple tissues in the developing embryos.Identification of homozygous lethal mutants with P-elementinsertion at the promoter region. J Biol Chem 270: 25150–25158.

Lloyd, A. and Sakonju, S. (1991) Characterization of two Dro-sophila POU domain genes, related to oct-1 and oct-2, andthe regulation of their expression patterns. Mech Dev 36:87–102.

Maniatis, T., Sambrook, J. and Fritsch, E. (1982) Molecular Clon-ing. A. Laboratory Manual. Cold Spring Harbor, NY: ColdSpring Harbor Laboratory Press.

Mason, E.D., Williams, S., Grotendorst, G.R. and Marsh, J.L.(1997) Combinatorial signaling by Twisted Gastrulation andDecapentaplegic. Mech Dev 64: 61–75.

Mastick, G.S., McKay, R., Oligino, T., Donovan, K. and Lopez, A.J.(1995) Identification of target genes regulated by homeoticproteins in Drosophila melanogaster through genetic selectionof Ultrabithorax protein-binding sites in yeast. Genetics 139:349–363.

Moore, L.A., Broihier, H.T., Van Doren, M. and Lehmann, R. (1998)Gonadal mesoderm and fat body initially follow a commondevelopmental path in Drosophila. Development 125: 837–844.

Nakajima, Y. and Natori, S. (2000) Identification and characteriza-tion of an anterior fat body protein in an insect. J Biochem(Tokyo) 127: 901–908.

Patel, N.H., Martin-Blanco, E., Coleman, K.G., Poole, S.J., Ellis,M.C., Kornberg, T.B. and Goodman, C.S. (1989) Expression ofEngrailed proteins in arthropods, annelids, and chordates. Cell58: 955–968.

Ramain, P., Heitzler, P., Haenlin, M. and Simpson, P. (1993)pannier, a negative regulator of achaete and scute in Dro-sophila, encodes a zinc finger protein with homology to thevertebrate transcription factor GATA-1. Development 119:1277–1291.

Rehorn, K.P., Thelen, H., Michelson, A.M. and Reuter, R. (1996)A molecular aspect of hematopoiesis and endoderm develop-ment common to vertebrates and Drosophila. Development122: 4023–4031.

Riechmann, V., Irion, U., Wilson, R., Grosskortenhaus, R. andLeptin, M. (1997) Control of cell fates and segmentation in theDrosophila mesoderm. Development 124: 2915–2922.

Riechmann, V., Rehorn, K.P., Reuter, R. and Leptin, M. (1998) The

genetic control of the distinction between fat body and gonadalmesoderm in Drosophila. Development 125: 713–723.

Rizki, M.T.M. (1961) Intracellular localization of kynurenine in thefatbody of Drosophila. J Biophys Biochem Cyt 9: 567–572.

Rizki, T.M. and Rizki, R.M. (1978) Larval adipose tissue of home-otic bithorax mutants of Drosophila. Dev Biol 65: 476–482.

Sam, S., Leise, W. and Hoshizaki, D.K. (1996) The serpent geneis necessary for progression through the early stages of fat-body development. Mech Dev 60: 197–205.

Soroker, V. and Hefetz, A. (2000) Hydrocarbon site of synthesisand circulation in the desert ant Cataglyphis niger. J InsectPhysiol 46: 1097–1102.

Staehling-Hampton, K. and Hoffmann, F.M. (1994) Ectopic decap-entaplegic in the Drosophila midgut alters the expression of fivehomeotic genes, dpp, and wingless, causing specific morpho-logical defects. Dev Biol 164: 502–512.

Staehling-Hampton, K., Hoffmann, F.M., Baylies, M.K., Rushton, E.and Bate, M. (1994) dpp induces mesodermal gene expressionin Drosophila. Nature 372: 783–786.

Technau, G.M. (1987) A single cell approach to problems of cell lin-eage and commitment during embryogenesis of Drosophilamelanogaster. Development 100: 1–12.

Thummel, C. and Pirrotta, V. (1992) New pCaSpeR P-elementvectors. Disease 71: 150.

Tysell, B. and Butterworth. F.M. (1978) Different rate of proteingranule formation in the larval fat body of Drosophila mela-nogaster. J Insect Physiol. 24, 201–206.

Vroemen, S.F., Van der Horst, D.J. and Van Marrewijk, W.J. (1998)New insights into adipokinetic hormone signaling. Mol CellEndocrinol 141: 7–12.

Weatherbee, S.D. and Carroll, S.B. (1999) Selector genes andlimb identity in arthropods and vertebrates. Cell 97: 283–286.

Weatherbee, S.D., Halder, G., Kim, J., Hudson, A. and Carroll, S.(1998) Ultrabithorax regulates genes at several levels of thewing-patterning hierarchy to shape the development of the Dro-sophila haltere. Genes Dev 12: 1474–1482.

Weinzierl, R., Axton, J.M., Ghysen, A. and Akam, M. (1987) Ultra-bithorax mutations in constant and variable regions of theprotein coding sequence. Genes Dev 1: 386–397.

White, R.A.H. and Wilcox, M. (1984) Distribution of Ultrabithoraxproteins in Drosophila. EMBO J 4: 2035–2043.

Winick, J., Abel, T., Leonard, M.W., Michelson, A.M., Chardon-Loriaux, I., Holmgren, R.A., Maniatis, T. and Engel, J.D. (1993)A GATA family transcription factor is expressed along theembryonic dorsoventral axis in Drosophila melanogaster.Development 119: 1055–1065.

Yin, Z. and Frasch, M. (1998) Regulation and function of tinmanduring dorsal mesoderm induction and heart specification inDrosophila. Dev Genet 22: 187–200.


Identification of fat-cell enhancer regions in Drosophila melanogaster

Documents

Transcript of Identification of fat-cell enhancer regions in Drosophila melanogaster