Tbx20-related genes, mid and H15, are required for tinman expression, proper patterning, and normal differentiation of cardioblasts in Drosophila

Mechanisms of Development 122 (2005) 1056–1069 www.elsevier.com/locate/modo

Tbx20-related genes, mid and H15, are required for tinman expression, proper patterning, and normal differentiation of cardioblasts in Drosophila Ingolf Reima, James P. Mohlerb, Manfred Frascha,* a

Brookdale Department of Molecular, Cell and Developmental Biology, Mount Sinai School of Medicine, Box 1020, One Gustave L. Levy Pl., New York, NY 10029, USA b Department of Biological Sciences, Barnard College, New York, NY 10027, USA Received 23 February 2005; received in revised form 13 April 2005; accepted 19 April 2005 Available online 4 May 2005

Abstract Tbx20-related T-box genes have been implicated in the regulation of heart development in several vertebrate species. In the present report, we demonstrate that a pair of genes representing Drosophila orthologs of Tbx20, midline (mid) and H15, have important functions during the development of the Drosophila equivalent of the heart, i.e. the dorsal vessel. We show that mid is among the earliest known genes that are specifically expressed in all cardioblasts during early embryogenesis, and H15 expression is subsequently activated in the same cells. Mutant embryos lacking the activity of mid, or both mid and H15, are able to form dorsal vessels with largely normal numbers of cardioblasts and pericardial cells. Furthermore, the mutant cardioblasts express several general cardioblast markers such as Mef2 and Toll at normal levels. However, the expression of tinman (tin), which normally occurs in four out of six cardioblasts in each hemisegment of the dorsal vessel, is almost abolished. Conversely, the expression of the Dorsocross (Doc) T-box genes, which is normally restricted to the two Tin-negative cardioblasts in each hemisegment, is strongly expanded into the majority of cardioblasts in mid mutant and midCH15-deficient embryos. Altogether, the data from the loss-of-function phenotypes demonstrate that mid, and to a lesser degree H15, have important roles in establishing the metameric patterning of cardioblast identities, but not in specifying cardioblasts as such. Ectopic expression of mid causes ectopic tin expression and, less efficiently, produces extra cardioblasts. We propose that one of the major functions of mid and H15 during cardioblast development is the re-activation of tin expression at a stage when the induction of tin by Dpp in the dorsal mesoderm has ceased. Through this activity, mid and H15 are required for the normal functional diversification of cardioblasts and the expression of tin-dependent terminal differentiation genes within the dorsal vessel. q 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: T-box genes; Tbx20; Dorsal vessel; Heart development; Cardiomyocyte differentiation

1. Introduction Transcription factors containing NK homeodomains, GATA domains, and T-domains are known to play key roles in promoting heart development and differentiation in a variety of vertebrate and invertebrate animals (reviewed in Cripps and Olson, 2002; Zaffran and Frasch, 2002). Furthermore, mutations in several genes encoding these types of cardiogenic factors can cause congenital heart diseases in humans (reviewed in Gelb, 2004). However, * Corresponding author. Tel.: C1 212 241 0988; fax: C1 212 860 9279. E-mail address: [email protected] (M. Frasch).

0925-4773/$ - see front matter q 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.mod.2005.04.006

a complete understanding of the regulatory interactions among the genes encoding these cardiogenic factors and their exact developmental functions is still lacking. The equivalent of the heart in Drosophila, called dorsal vessel, consists of a relatively simple tube-like structure, which is composed of two rows of cardiomyocytes (called cardioblasts) that form the lumen and surrounding noncontractile pericardial cells (Rizki, 1978). Studies of Drosophila cardiogenesis have led to the identification of several regulatory genes and signaling processes that were subsequently found to have close molecular and functional counterparts in the control of vertebrate heart development. Most notably, the NK homeobox gene tinman (tin) is

I. Reim et al. / Mechanisms of Development 122 (2005) 1056–1069

expressed in the cardiogenic mesoderm and developing dorsal vessel of Drosophila and its activity is essential for generating the dorsal vessel (Azpiazu and Frasch, 1993; Bodmer, 1993). Likewise, the related NK homeobox gene Nkx2-5 is expressed in the cardiogenic mesoderm and developing heart of vertebrate embryos and its function is critical for the promotion, albeit not initiation, of normal heart development (reviewed in Harvey, 1996). GATA factor encoding genes, particularly GATA-4, -5 and -6, are expressed prominently during early cardiogenesis in vertebrate embryos and play critical roles during heart development as well (reviewed in Pikkarainen et al., 2004). In Drosophila, the GATA factor encoding gene pannier (pnr) shows transient expression in the cardiogenic mesoderm and is essential for dorsal vessel development (Gajewski et al., 1999; Alvarez et al., 2003; Klinedinst and Bodmer, 2003). Several members of the T-box family, most notably Tbx1, Tbx2, Tbx3, Tbx5, Tbx18, and Tbx20, are also involved in vertebrate cardiogenesis (reviewed in Plageman and Yutzey, 2005). There are indications that T-box genes also play roles during Drosophila cardiogenesis, but to date their functions have not yet been defined (Griffin et al., 2000; Lo and Frasch, 2001; Reim et al., 2003). Both in vertebrates and in Drosophila, the individual members of these three distinct classes of transcription factors are expressed in temporally and spatially overlapping patterns during cardiogenesis. These overlapping patterns are generated by complex regulatory networks among their respective genes, which have only been clarified partially, as well as by shared inductive signaling inputs (reviewed in Zaffran and Frasch, 2002). In addition, there is increasing evidence that protein–protein interactions between factors of the same or different classes are important for the execution of their specific transcriptional activities (e.g. Hiroi et al., 2001; Stennard et al., 2003). In the present manuscript, we identify two T-box genes, mid and H15, as important activators of tin expression in cardioblasts. mid (CG6634, formerly known as H15r) and its paralog H15 (Brook and Cohen, 1996) are orthologs of the vertebrate Tbx20 genes, which have also been implicated in the control of heart development in several species (reviewed in Plageman and Yutzey, 2005). We show that mid is expressed earlier and probably plays the more crucial role among the two genes in cardiogenesis. mid expression initiates during early stage 12 within the cardiogenic mesoderm, whereas H15 expression is activated only at late stage 12. The activation of these genes in these cells, which are presumptive cardioblasts, requires tin and pnr activity, and H15 expression is independent of mid. Both genes, mid and H15, continue to be co-expressed in all cardioblasts until the end of embryogenesis. In the absence of mid activity, tin expression in cardioblasts is severely reduced, a phenotype that is only slightly enhanced upon additional removal of H15. The reduction of tin expression in cardioblasts is accompanied by an equally dramatic expansion of Doc expression into tinnegative cardioblasts. The potency of mid to activate tin is further illustrated by ectopic expression experiments in the

1057

mesoderm, which result in an expansion of tin expression into additional dorsal mesodermal cells. However, mid is less potent in causing full fate transformations of these cells, which are mostly of pericardial origin, into cardioblasts. The total number of cardioblasts is not changed in mid mutants and is only slightly and variably reduced upon deletion of both mid and H15. Altogether, these observations argue against a role of mid and H15 in cardioblast specification. Rather, they show that mid via activating tin and repressing Doc in a subset of cardioblasts is required for the normal diversification of cardioblast identities. Hence, while the role of mid and H15 in generating a cardiac tube is minor, their roles in cardiac patterning and cardioblast differentiation are likely to be important for the normal functioning of the dorsal vessel in hemolymph circulation.

2. Results 2.1. mid and H15 are expressed in cardioblasts downstream of tin and pnr Based upon the known role of T-box genes in vertebrate cardiogenesis, we probed the embryonic mRNA expression of all previously uncharacterized Drosophila T-box genes with regard to their potential expression in developing cardiac tissues. As demonstrated below, a linked pair of Tbx20-related genes, CG6604/H15 and CG6634 (aka nmr2 and H15r), shows a highly specific pattern of expression in this tissue, in addition to other areas of expression at earlier stages. The earlier expression of CG6634 in ventral stripes, together with the map position of the gene and the ventral segmentation phenotype of midline (mid; Nu¨sslein-Volhard et al., 1984) mutant embryos led us to suspect that CG6634 corresponds to mid. During the course of our work, this tentative identification was confirmed experimentally by Buescher et al. (2004) who also determined that mid1 is a null allele. Henceforth, we name the two genes mid and H15. During embryonic stage 11, mid mRNA expression is observed in three clusters of Tin-positive cells within the posterior mesoderm, which may correspond to gonadal mesoderm precursors (Fig. 1A, arrow; van Doren, 2004). mid expression is not yet observed in the Tin-positive cells of the cardiogenic mesoderm at this stage, which occupy the dorsal-most areas of the mesoderm (Fig. 1A, arrow head). However, during early stage 12, mid expression initiates in a posterior-to-anterior fashion in small clusters of Tin-positive cells within the cardiogenic mesoderm (Fig. 1B). Based upon double-stainings for seven-up (svp)-LacZ (Fig. 1C; Gajewski et al., 2000; Lo and Frasch, 2001) and Ladybird (Lb; Jagla et al., 1997a; data not shown), these cell clusters correspond to the TinC and TinCLbC subsets of cardioblasts, whereas the SvpC cardioblasts are still negative for mid mRNA at this time (Fig. 1C). Accordingly, the two mid-positive cell clusters in each hemisegment are also positive for the Doc Tbox genes, which additionally mark mid-negative

1058

I. Reim et al. / Mechanisms of Development 122 (2005) 1056–1069

Fig. 1. Expression and regulation of mid and H15 mRNA during cardiogenesis. Fluorescent signals derived from mid and H15 mRNAs are shown in green and signals derived from various cardiogenic markers are shown in red. Panel I shows a dorsal view and all others show lateral views. (A) Stage 11 embryo showing mid mRNA in the gonadal mesoderm (gm) but not in the Tin-stained cardiogenic mesoderm (cm). (B) Early stage 12 embryo showing mid mRNA expression in two clusters of cardioblasts per hemisegment, which were identified as future TinC cells (Tin-CB) that are Ladybird (Lb)-positive and Lb-negative, respectively. (C) At early stage 12 the Svp-positive cardioblasts (Svp-CB) are still negative for mid mRNA (PC: Svp-positive pericardial cells). (D) At mid stage 12, mid mRNA is still restricted to two clusters per hemisegment, whereas Doc gene products remain expressed more broadly in the cardiac precursors.

I. Reim et al. / Mechanisms of Development 122 (2005) 1056–1069

1059

Fig. 2. Dorsal vessel development and cardioblast differentiation in mid mutant and mid, H15-deficient embryos. Shown are dorsal views of stage 15–16 embryos. Genotypes are shown on top of each column and color-coded probes to the left of each row. (A) Stage 16 control embryo showing expression of Mef2 and H15 mRNA in cardioblasts. (B) Homozygous mid1 mutant embryo at stage 16 displaying a full complement of cardioblasts, which express both Mef2 and H15. (C) Homozygous Df(2L)cl-h2 embryo (which lacks both mid and H15) at stage 16 with Mef2-expressing cardioblasts that are arranged in largely parallel rows and are only slightly reduced in number. (D–F) Stage 15, just prior to dorsal closure. Like the control embryos, homozygous mid1 mutant embryos and homozygous Df(2L)cl-h2 embryos show the presence of cardioblasts that express Mef2 and hand mRNA. In mid1 embryos, the number of cardioblasts is normal and in Df(2L)cl-h2 embryos it is slightly reduced. (G) Stage 16 control embryo showing Toll protein at the surface (mostly laterally and apically, i.e. towards luminal surface) of Mef2-labeled cardioblasts. (H) Stage 16 mid1 mutant embryo showing Toll and Mef2 staining in cardioblasts similar to control embryos. (I) Stage 16 Df(2L)cl-h2 embryo showing normal Toll and Mef2 expression levels in cardioblasts. Toll is seen more often also on basal surfaces of cardioblasts as compared to the control. [In (G–I), as well as in Fig. 5G,H, areas with ectodermal Toll staining on top of the dorsal vessel were eliminated from the respective scans prior to merging individual layers.] (J–L) Higher magnification views of TollCMef2 doubly-stained embryos as in (G) and (H), showing the heart and posterior aorta portions. Open arrowheads highlight examples for the normal absence of Toll from the basal side of cardioblasts while solid arrowheads point to cases of mis-localized Toll at the basal surface of cells.

(presumably SvpC) cardioblasts at this stage (Fig. 1D; Reim et al., 2003; I.R. and M.F., manuscript in preparation). At late stage 12, all cardioblasts, including the SvpC cells that are now losing Tin, have initiated mid-expression and align as a single row of cells along the dorsal margin (Fig. 1E). During this stage, H15 mRNA expression initiates as well (Fig. 1F), and subsequently, both mid and H15 are expressed in all cardioblasts of the developing dorsal vessel (Figs. 1G–I,2A). As expected, four of the six mid-stained cardioblasts in each hemisegment are positive and two are negative for Tin (Fig. 1G,I). None of the pericardial cells express mid or H15 (Fig. 1G–I, and data not shown).

As the above data indicate that mid is one of the earliest known markers specific for all early cardioblasts, we examined potential upstream regulators of mid expression. As shown in Fig. 1J,K, the two cardiogenic regulators tin and pnr are critical for the activation of mid expression in the cardioblast precursors. Because pnr is also expressed in the dorsal ectoderm, it may influence mid in the mesoderm indirectly. However, the observed rescue of mid expression in pnr mutant embryos with twidriven pnr expression suggests that the mesodermal component of pnr is crucial for mid activation in early cardioblasts (Fig. 1L).

3 (E) During late stage 12, mid mRNA expression is activated in all cardioblasts, whereas the Tin-positive and other types of pericardial cells are negative for mid (gm: expression of Tin and mid in the gonadal mesoderm). (F) At late stage 12, nuclear transcripts of H15 start appearing in Mef2-labeled cardioblasts. (G,H) At stage 14, mid and H15 mRNAs are expressed solidly in all cardioblasts. (I) Stage 16 embryo showing specific mid mRNA expression in all cardioblasts of the dorsal vessel. (J) Early stage 12 tin mutant embryo showing lack of mid mRNA expression in the usual area of the cardiac mesoderm (arrowhead). (K) Stage 12 pnr mutant embryo showing lack of mid mRNA expression and strongly reduced Tin expression in the cardiogenic region (arrowhead). Gonadal mesoderm (gm) expression of mid is unaffected. (L) pnr mutant embryo with twist-driven rescue of pnr expression in the mesoderm (2xPE-twi-GAL4/UAS-pnr; pnrVX6). Tin and mid mRNA expression in the cardiogenic mesoderm are rescued (arrowhead).

1060

I. Reim et al. / Mechanisms of Development 122 (2005) 1056–1069

2.2. mid and H15 are not required for the formation of cardioblasts and the dorsal vessel The specific expression of mid and H15 in cardioblasts suggests a role of these genes in regulating the development of these cells. Stainings of late stage embryos for Myocyte enhancer factor-2 (Mef2; Bour et al., 1995; Lilly et al., 1995) demonstrated that this factor is expressed in cardioblasts of homozygous mid1 mutant embryos like in the wild type and the normal number of cardioblasts is formed (Fig. 2A,B). In addition, H15 is expressed normally in the cardioblasts of mid mutant embryos (Fig. 2A,B). Hence, mid is not required for the specification of cardioblasts, and despite the delayed expression of H15 as compared to mid, H15 expression in cardioblasts does not depend on mid. To test whether the absence of an effect on the expression of Mef2 and cardioblast specification upon mid mutation is due to functional redundancy with H15, we examined embryos homozygous for a deficiency, Df(2L)clh2, which uncovers both mid and H15. As shown in Fig. 2C, cardioblasts are formed and still express Mef2 upon loss of both mid and H15. Many embryos homozygous for this deficiency have an almost normal number of cardioblasts, although the average number is reduced to 92G5.6 cardioblasts (nZ10) as compared to 104 in wild type embryos. Hence, it is possible that mid and H15 have a minor and functionally-redundant role during the specification or for the survival of cardioblasts. However, we cannot rule out the possibility that this effect is caused by the loss of unrelated gene(s) that are uncovered by this deficiency. Because of the lack of major effects of mid and H15 on the formation of cardioblasts, we examined their potential role in the activation of additional cardioblast-specific markers. The expression of the bHLH encoding gene hand, which initiates in cardioblasts as well as in pericardial cells during mid stage 12 (i.e. shortly after mid) (Kolsch and Paululat, 2002), is also not affected in homozygous mid and mid, H15-deficient embryos (Fig. 2D–F). Likewise, the mRNA expression of how, which encodes a KH-domain/ STAR family protein is active in all cardioblasts in these mutant embryos as it is in the wild type (data not shown; Baehrecke, 1997; Lo and Frasch, 1997; Zaffran et al., 1997). The transmembrane receptor Toll is also expressed in cardioblasts, which is due to the activity of a cardioblastspecific enhancer element of Toll that receives inputs from both tin and Doc (Wharton and Crews, 1993; Wang et al., 2005). In control embryos, Toll protein localizes almost exclusively to the lateral and apical sides of cardioblasts (Fig. 2G,J; Wang et al., 2005). As shown in Fig. 2G–L, mid and H15 are not required for generating normal expression levels of Toll in cardioblasts. However, in addition to its normal sites of localization, the Toll protein is frequently also observed at the basal surface of cardioblasts in midC H15-deficient embryos (Fig. 2I,L), although its localization is rarely altered in mid mutants (Fig. 2H,K). Hence, mid and H15 may have redundant requirements for the expression of

yet unknown components controlling the cellular polarity of cardioblasts. However, because of the absence of a clear cell polarity phenotype in mid mutant embryos we cannot rule out the possibility that the observed phenotype in Df(2L)clh2 homozygotes is due to the absence of additional genes. Taken together, these observations show that mid is not required for the formation of cardioblasts and the expression of several general cardioblast-specific differentiation markers including Mef2, hand, Toll, and how. In addition, the dorsal vessel acquires its normal overall morphology in mid mutant embryos and in the fraction of mid, H15 deficient embryos that form close to normal numbers of cardioblasts (e.g. Fig. 2B,I). Hence, mid and H15 are not essential for major morphogenetic processes during dorsal vessel development, although they may be needed for the expression of certain differentiation genes that regulate more subtle aspects of dorsal vessel morphogenesis, such as the polarity of cells. 2.3. mid is required for proper activation of tin and spatial restriction of Doc expression in cardioblasts After their expression in all cardioblast progenitors at early and mid stage 12, tin and Doc become expressed in mutually exclusive subsets of cardioblasts from late stage 12 until the end of embryogenesis. Specifically, tin is then expressed in four cardioblasts in each hemisegment and Doc in the remaining two, which also express seven-up (svp) and, in the heart portion, form the ostia (Fig. 3A,D; Lo and Frasch, 2001; Reim et al., 2003). In addition, tin is expressed in pericardial cells. While the early expression of tin until mid stage 12 is normal in mid mutant and mid, H15 deficient embryos (data not shown), tin expression is strongly disrupted during its late phase. In homozygous mid1, mid2, mid1/Df(2L)cl-h2, and mid2/Df(2L)cl-h2 embryos, the number of cardioblasts expressing tin is severely reduced (Fig. 3B,E,F, and data not shown). Moreover, some of the residual Tin-positive cardioblasts express tin at very low levels. Unlike in the cardioblasts, the expression of tin in pericardial cells is not affected in these embryos (Fig. 3B,E,F). To test whether the residual expression of tin in some cardioblasts in these embryos is due to the function of H15, we stained homozygous Df(2L)cl-h2 embryos for Tin. However, these embryos still showed a small and slightly variable number of Tin-positive cardioblasts (Fig. 3G). These data demonstrate that mid has an important role as an activator of tin expression in the developing cardioblasts. H15 cannot substitute for this activity, although it may have a minor contribution to it. The phenotype of mid with regard to Doc expression in cardioblasts is the opposite as compared to tin. In embryos, homozygous for mid1 or Df(2L)cl-h2, and in mid/Df(2L)cl-h2 embryos, Doc expression is expanded into the great majority of cardioblasts (Fig. 3B,C,E–G). The complementarity of tin and Doc expression is still maintained in these situations, with the exception of cardioblasts that express tin at very low levels, which also show low levels of Doc expression

I. Reim et al. / Mechanisms of Development 122 (2005) 1056–1069

1061

Fig. 3. Roles of mid and H15 in regulating the expression of tin and Doc genes in cardioblasts. (A–C) show lateral views of stage 14 embryos while the remaining panels show dorsal views of stage 16 embryos. The top half of each panel shows double stainings for Tin and Doc, whereas the bottom half shows only the red channel to visualize Tin. (A) Control embryo showing four cardioblasts in each hemisegment with strong Tin signals (arrows) and two cardioblasts with low levels of residual Tin protein and strong Doc expression (arrowhead: Tin-positive pericardial cells). (B) Homozygous mid1 mutant, and (C), homozygous Df(2L)cl-h2 mutant embryo showing a majority of cardioblasts with low residual or absent Tin expression (arrows) and high levels of Doc expression. Most of the remaining Tin-positive cells correspond to pericardial cells (arrowhead). (D) Control dorsal vessel showing four TinC/DocK and two TinK/DocC cardioblasts in each hemisegment, and in addition TinC pericardial cells. (E) In a transheterozygous mid1/Df(2L)cl-h2 embryo, and (F), in a mid2/Df(2L)cl-h2 embryo, most of the cardioblasts, except for the anterior aorta, are negative for Tin and positive for Doc. The majority of the remaining Tinpositive cells correspond to pericardial cells. (G) Homozygous Df(2L)cl-h2 mutant embryo with relatively normal number of cardioblasts, most of which lack Tin expression and display Doc expression. Arrows indicate cardioblasts with low levels of both Tin and Doc.

(e.g. Fig. 3G, arrows). In extreme cases, essentially all cardioblasts in the main portion of the aorta and in the heart express Doc and lack Tin (Fig. 3F). In most cases, the expression patterns of tin and Doc in the cardioblasts of the anterior aorta are less affected (Fig. 3B,E,F; see also Fig. 4D). This is consistent with previous findings that this portion of the dorsal vessel is under different regulation and normally does not contain Doc/svp expressing cardioblasts (Alvarez et al., 2003; Lo and Frasch, 2003; Fig. 3D). 2.4. mid is required for proper differentiation and functional diversification of cardioblasts As mid is required for the proper establishment of the segmental tin and Doc expression patterns within the dorsal vessel, we asked whether the absence of mid and H15 activity

would affect the differential development of the Tin versus Doc-marked populations of cardioblasts. Unlike the general cardioblast markers described above, Drosophila Sulfonylurea receptor (Sur), predicted to encode a regulatory potassium channel subunit, displays high levels of expression within the four cardioblasts in each hemisegment that express tin but not in the two that express Doc in normal stage 16 embryos (Fig. 4A; Nasonkin et al., 1999; Lo and Frasch, 2001). In contrast to the controls, in homozygous mid1 mutant embryos Sur is not expressed at significant levels in most of the cardioblasts, with the exception of the anterior aorta that still expresses Sur to a variable extent (Fig. 4B). In embryos lacking both mid and H15, the reduction of Sur expression in the dorsal vessel appears even stronger and includes its anterior-most portion (Fig. 4C). As shown in Fig. 4D in a double staining for Tin protein and Sur mRNA, the few

1062

I. Reim et al. / Mechanisms of Development 122 (2005) 1056–1069

Fig. 4. Effects of mid and H15 mutation on the segmental diversification of cardioblasts. Shown are dorsal views of stage 16 embryos. (A) Control embryo stained for Mef2 and Sur mRNA. Sur mRNA levels are high in four cardioblasts (bracket) and barely detectable in the remaining two cardioblasts in each hemisegment. (B) mid1 mutant embryo stained as in (A). With some exceptions, particularly in the anterior aorta, Sur mRNA levels in cardioblasts are extremely low throughout the dorsal vessel. (C) Homozygous Df(2L)cl-h2 embryo stained as in (A). The number of cardioblasts with high levels of Sur mRNA is even further reduced as compared with mid mutant embryos. (D) mid1 mutant embryo stained for Sur mRNA and Tin. Only the residual Tin-positive cardioblasts express Sur (arrows). (E) Control embryo stained for Toll, Doc, and Tin (high magnification view of posterior portion of dorsal vessel with the heart marked by a dashed bracket). Cell surface staining with anti-Toll and nuclear staining with anti-Doc shows that the pairs of Doc-positive ostial cells (e.g. solid bracket) are elongated as compared to the Tin-positive cells in the heart and the cells in the aorta. (F) mid1 mutant embryo stained as in (E), showing that most of the extra Doc-positive cells in the heart are elongated like the ostial cells in normal embryos. (G) High magnification view of control embryo as in (E), stained for Wg, Doc and Tin. Wg signals mark the Doc-positive ostia, which consist of a pair of cells in each hemisegment of the heart portion. (H) Homozygous mid1 mutant embryo stained as in (G). While this example shows less than typical reductions of Tin expression in cardioblasts of the heart portion, the expression of ostia-specific Wg expression is expanded to a similar degree as that of Doc (solid brackets).

remaining cardioblasts with high levels of Tin are the ones that also retain Sur expression in mid mutant embryos. Hence, the observed effects of mid mutation on Sur expression are likely caused by the loss of tin expression. In the heart portion of the late dorsal vessel, the two Docexpressing cells in each hemisegment are known to give rise to cells forming the ostia, which are characterized by their elongated shape as compared to the Tin cardioblasts (Fig. 4E; Lo and Frasch, 2001; Molina and Cripps, 2001; Ponzielli et al., 2002). As shown in Fig. 4F, most of the extra Doc-positive cells in the heart of mid mutant embryos also display the elongated morphology, which suggests that the observed

changes in the Tin and Doc patterns are accompanied by transformations from ‘regular’ cardiomyocytes into ostia-like cells. In agreement with this notion, most of the extra Docpositive cells in the heart also express Wingless, a marker that is normally restricted to the two ostial cells in each hemisegment of the heart portion (Fig. 4G,H; Lo et al., 2002). 2.5. Ectopic mid causes expanded tin expression and can produce extra cardioblasts As mid is required for normal activation of tin in cardioblasts, we asked whether ectopic expression of mid

I. Reim et al. / Mechanisms of Development 122 (2005) 1056–1069

can cause an expansion of tin expression and perhaps the formation of ectopic cardioblasts in the mesoderm. As shown in Fig. 5B,D, ectopic expression of mid in the mesoderm with the SG30 driver (controlled by a combination of twi and how enhancers) does indeed lead to a broadening of tin expression in embryos of the stages 13–16. The expansion of tin is largely restricted to dorsal areas adjacent to the cardioblasts and pericardial cells that express tin in the normal situation (cf. Fig. 5A,C)

1063

and the effect is segmentally modulated. However, the cardioblast-specific marker H15 is mostly still restricted to a single row of tin-expressing cells on either side of the dorsal midline, although a few Tin-positive cells with H15 expression are seen adjacent to the normal cardioblast rows (Fig. 5B,D, arrows; cf. Fig. 5A,C). Similarly, cells expressing both Mef2 and H15 are found only occasionally outside the single rows of cardioblasts (Fig. 5F, arrows, cf. Fig. 5E). The total number of cells that co-express Mef2

Fig. 5. Ectopic expression of mid in the mesoderm causes expanded tin expression and produces ectopic cardioblasts. (A) Stage 14 wild type embryo (WT; lateral view) showing H15 expression in all cardioblasts and Tin in four cardioblasts in each hemisegment as well as in pericardial cells. (B) Stage 14 embryo with SG30-driven ectopic mesodermal expression of mid, showing expanded Tin expression (arrowhead) in cells adjacent to H15-marked cardioblasts and in some scattered cells further ventrally. H15 expression is occasionally expanded under these conditions as well (arrows). (C) Stage 15 wild type embryo (dorsal view) stained as in (A). (D) Stage 15 embryo with SG30-driven ectopic mesodermal mid expression showing ectopic Tin expression as in (B). Ectopic Tinpositive cells tend to cluster segmentally (arrowhead) and a few of them also express ectopic H15 (arrow). (E) Stage 15 wild type embryo stained for Eve protein (blue; Eve-PC: Eve-positive pericardial cells), H15 mRNA (green), and Mef2 protein (red). H15 and Mef2 are coexpressed in cardioblasts. (F) Stage 14 embryo with SG30-driven ectopic expression of mid in the mesoderm. The number of Mef2- and H15-labeled cardioblasts is largely normal although their arrangement is irregular and some ectopic Mef2C H15-positive cells are seen (arrows). Eve expression in pericardial cells is reduced. Mef2-labeled somatic muscle nuclei are present but arranged in a disorganized pattern. (G) Stage 16 embryo with 24B-driven ectopic expression of mid and stained for Tin and Toll. In the heart portion of the dorsal vessel, many cells express both Tin and Toll (white arrow), thus indicating ectopic cardioblast formation. White arrowheads indicate presumed pericardial and lymph gland cells and red arrowhead presumed somatic muscle cells that express Tin but not Toll. (H) Embryo with stage and genotype as in (G) but stained for Mef2 and Toll. The presence of additional Mef2C Toll-positive cardioblasts adjacent to the medial cardioblast rows of the heart portion (white arrow) indicates ectopic cardioblast formation.

1064

I. Reim et al. / Mechanisms of Development 122 (2005) 1056–1069

and H15 upon SG30-driven expression of mid is not strongly increased as compared to wild type embryos; however, these cardioblasts are not aligned as neatly and there are frequently small gaps in the cell rows (Fig. 5B,F). Mef2positive nuclei of the dorsal somatic muscles are still present although their arrangement is also disorganized (Fig. 5F). In many embryos, presumably those with highest levels of SG30-driven mid, there are larger gaps in the dorsal vessel and stronger reductions of Mef2-positive somatic muscle nuclei, which we attribute to cell death caused by high levels of ectopic mid. Therefore, the phenotypes in Fig. 5B,D,F likely represent the activities of intermediate levels of ectopic mid. However, with the 24B driver (controlled only by how enhancers) the reduction of mesodermal cell numbers is much less pronounced, which indicates that the detrimental effects of ectopic mid on cell viability occur mainly before mid stage 12. This makes it possible to study the consequences of higher levels of ectopic mid in the mesoderm during mid and late stages of embryogenesis on cardiac development. Although 24B-driven mid produces a similar extent of ectopic tin expression as SG30, a large number of the ectopic Tin-positive cells are also positive for other cardioblast markers such as Toll and Mef2 in many embryos (Fig. 5G,H). The effect is largely restricted to the heart portion. In agreement with the normal complementarity of Doc and Tin expression in cardioblasts, the ectopic

Tin/Toll/MEF2-positive cardioblasts are negative for Doc (data not shown). The 24B driver also causes disruptions of the somatic musculature, as judged by the Mef2 pattern (Fig. 5H). Together, these observations show that ectopic mid causes ectopic expression of tin in mesodermal cells near the heart and, less efficiently, their transformation into ectopic cardioblasts. 2.6. Ectopic mid causes reduced pericardial cell marker expression To obtain information on the identity of the cells that express ectopic tin we examined embryos with ectopic mid expression for the distribution of pericardial cell markers. In wild type embryos, eve and odd are expressed in complementary sets of pericardial cells, both of which also express zfh-1. In addition, eve is expressed in a single dorsal somatic muscle fiber and its founder in each segment (Fig. 6A,B). As shown in Fig. 6C,D, the expression of eve and odd in pericardial cells is severely reduced, and in some segments missing, upon SG30-driven ectopic expression of mid in the mesoderm. However, the expression of zfh-1 is much less affected. The somatic muscle expression of eve is reduced to a similar degree as in pericardial cells in these embryos (Fig. 6C,D). Thus it appears that the ectopic expression of tin upon ectopic mid expression with SG30 mainly occurs in

Fig. 6. Effects of ectopic mid expression and mid mutation on pericardial cell development. (A) Stage 14 wild type embryo (lateral view), and (B), stage 16 wild type embryo (dorsal view) stained for the pericardial cell markers Odd (green), Zfh-1 (red) and Eve (blue). Eve and Odd mark different subsets of Zfh-1positive pericardial cells (PC). Eve is also expressed in the dorsal somatic muscle #1 (DA1) and Odd in the lymph gland (LG). (C,D) Stage 14 and stage 16 embryo, respectively, with SG30-driven ectopic expression of mid in the mesoderm and stained as in (A,B). Odd and Eve expression in pericardial cells, DA1 muscles, and the lymph gland is reduced whereas Zfh-1 expression is largely normal (e.g. arrow). (E,F) Stage 14 and stage 16 embryo, respectively, deficient for mid and H15 and with normal patterns of Odd, Zfh-1, and Eve expression.

I. Reim et al. / Mechanisms of Development 122 (2005) 1056–1069

the cells that normally would have developed into pericardial cells and in some dorsal somatic muscle founders. The more lateral positions of several isolated Tin-expressing clusters in embryos with SG30 or 24B-driven mid expression also indicates that tin expression can occasionally be activated in dorsal muscle founders and precursors (see Fig. 5B,G). mid expression in these pericardial and somatic muscle cells interferes with their normal development as they lose proper marker expression and display altered arrangements. The absence of Mef2 and H15 and the presence of zfh-1 expression in pericardial cells with SG30-driven mid suggests that, even though they express tin, they are not fully transformed into functional cardiomyocytes. However, based upon the data described above, full transformations do occur, particularly with 24B-driven mid. Because of the observed reduction of pericardial cell markers with ectopic mid, we asked whether the expression of these markers is also affected (perhaps in the opposite fashion) in embryos that lack both mid and H15. However, we found that all three markers, eve, odd, and zfh-1, are expressed normally in pericardial cells of these embryos, as is eve in the dorsal somatic muscle (Fig. 6A,B,E,F). Altogether, our observations show that in the normal situation mid, and to a lesser degree H15, are required for normal patterning and differentiation of cardioblasts, but not for the development of pericardial cells. Normally, mid and H15 are not required for the specification of cardioblasts as such, although ectopic expression of mid can recruit some dorsal mesodermal cells into forming extra cardioblasts.

3. Discussion Our present study has identified the Tbx20-related T-box genes mid and H15 as important regulators of Drosophila cardiogenesis. Mid and H15 are the first known transcriptional regulators that become specifically expressed in all cardioblasts of the developing dorsal vessel in the mesoderm. By contrast, previously described transcription factors, including Tin, Doc proteins, Hand, and Mef2, are expressed in specific subsets of cardioblasts and/or in additional mesodermal cell types such as pericardial cells and somatic muscles (Lilly et al., 1994; Nguyen et al., 1994; Bodmer and Frasch, 1999; Lo and Frasch, 2001; Kolsch and Paululat, 2002). Accordingly, we find that the Tbx20orthologs from Drosophila have specific roles in regulating the diversification and differentiation of cardioblasts during dorsal vessel development. One of their key functions during this process is the normal activation of tin expression in cardioblasts. 3.1. mid activation in early cardioblasts downstream of tin and pnr The cardiac mesoderm during the transition from stage 11 to 12 includes the cardioblast progenitors, in which all

1065

three cardiogenic regulators, Tin, Pnr, and Doc, are co-expressed (I.R. and M.F., manuscript in preparation). Clonal analysis has shown that there are four of these progenitors within each hemisegment, two of which divide symmetrically into four Tin-expressing cardioblasts while the other two marked by Svp divide asymmetrically into one cardioblast and one pericardial cell each (Ward and Skeath, 2000). Our expression analysis indicates that mid expression is initially activated in the four cardioblasts arising from symmetric divisions and subsequently also in the two cardioblasts arising from asymmetric divisions. The delayed onset of mid expression in thoracic as compared to abdominal segments is consistent with this interpretation, as these segments form only asymmetrically-dividing cardioblasts (Alvarez et al., 2003). In all cases, it seems that mid becomes expressed in the cardioblasts shortly after their final divisions and not in their progenitor cells. Hence, mid represents the earliest specific marker for all cardioblasts. H15 is activated in cardioblasts with a significant delay as compared to mid, but as we have shown herein this delay is not due to a requirement for mid in H15 activation. Tin, Pnr, and Doc proteins are expressed in the entire cardiac mesoderm prior to mid and, during the period of mid and H15 activation, are still co-expressed in all cardioblasts. Correspondingly, we find that all three regulators are essential for activating mid and H15 expression in cardioblasts (Fig. 1; I.R. and M.F., manuscript in preparation). Because Tin, Pnr, and Doc are essential for cardioblast specification, whereas mid and H15 are not, we assume that these three cardiogenic regulators activate additional downstream genes in the developing cardioblasts that are required for their formation. Furthermore, as Tin, Pnr, and Doc are expressed more broadly in the early cardiogenic mesoderm, it will be interesting to determine the regulatory processes that prevent these factors from activating mid and H15 in pericardial cells. 3.2. The roles of mid and H15 in cardioblast development The presence of cardioblasts in late stage embryos lacking the function of mid or of both mid and H15 argues strongly against a normal function of these two genes in cardioblast specification. Although most homozygous Df(2L)cl-h2 embryos, unlike mid mutant embryos, have somewhat reduced numbers of cardioblasts, this phenotype could well be due to the lack of other genes near the mid, H15 locus. Rather than for their specification, mid and H15 are required for the establishment of normal gene expression patterns in the developing cardioblasts. Consistent with its earlier expression in developing cardioblasts, mid seems to be functionally more important than H15. A key gene regulated by mid in cardioblasts is tin. Although previous studies had shown that the early, broad expression of tin in the mesoderm and its subsequent expression in the dorsal mesoderm is regulated by twist and dpp, respectively (Bodmer et al., 1990; Frasch, 1995; Yin

1066

I. Reim et al. / Mechanisms of Development 122 (2005) 1056–1069

et al., 1997), the regulatory inputs with regard to tin activation in the cardioblasts proper were less well defined. However, it was known that tin activation in cardioblasts is controlled by a separate enhancer element, tinC, and requires signals from the FGF receptor Heartless (Yin et al., 1997; Michelson et al., 1998). Both our loss- and gainof-function experiments now identify mid as a critical activator of tin expression in the cardioblasts. Furthermore, in vitro DNaseI protection experiments identified several binding sites for Mid within the tinC element, which raises the possibility that mid is a direct regulator of tin (Patrick Lo and M.F., unpublished data). As we have shown genetically, H15 does not provide a redundant function with mid for the activation of tin. Presumably, this is due to the delayed onset of H15 expression, which occurs at a stage after the tinC enhancer has already been activated. The residual expression of tin in some cardioblasts at low or even normal levels in midK and mid, H15 deficient embryos indicates the presence of regulators acting in partially redundant fashion with mid. In addition to yet unidentified regulators, these could include Pnr, Doc, and autoregulatory Tin, all of which are expressed transiently during cardioblast-specific tin activation. In addition to their action through mid, some or all of these factors may have a direct contribution towards the activation of tin expression in cardioblasts. The observations with ectopic expression of mid are consistent with the notion that mid is normally required for activating tin expression in cardioblasts. Forced expression of mid in the mesoderm under the control of twi or of how enhancers, or both, causes ectopic expression of tin, mostly in cells that would normally have developed into pericardial cells and also in some somatic muscle cells. Although this is accompanied by the downregulation of the pericardial markers eve and odd, based upon their residual zfh-1 expression many of the cells with ectopic tin still appear to retain some properties of pericardial cells. Hence, the expression of mid is often not sufficient for a full transformation of pericardial and somatic mesodermal cells into cardioblasts. However, ectopic mid in these cells interferes with their normal development, including the expression of specific cell fate markers and their normal spatial arrangement. With a lower efficiency, we do observe some expansion of other cardioblast markers such as Mef2 and cardioblast-specific Toll in these embryos, particularly in dorsal/posterior areas of the mesoderm. Therefore, ectopic mid can transform certain cells into cardioblasts even though during normal development neither mid nor H15 are required for cardioblast specification. One possible explanation for this discrepancy could be that during early stages, ectopic mid can take on early functions that are normally exerted by the Doc T-box genes. Indeed, the absence of Doc genes leads to a complete loss of cardioblasts, whereas ectopic DocCtin expression causes expanded cardioblast formation similar to ectopic mid (I.R. and M.F., manuscript in preparation).

In addition to the loss of tin expression, another major consequence of the loss of mid function is the expanded expression of Dorsocross in cardioblasts. Hence, mid normally acts to restrict Doc gene expression to two tinnegative cardioblasts in each hemisegment within the heart and the main portion of the aorta of the dorsal vessel. As mid is expressed in all six cardioblasts in each hemisegment, this repression could operate through either of two possible mechanisms. The repression could be direct and involve co-repressors that are restricted to the four Doc-negative cardioblasts in each hemisegment. Alternatively, it could be indirect and involve the transcriptional activation of a repressing gene in these cardioblasts. Recent observations from our laboratory together with those presented in the present report suggest that the latter mechanism is operating during this process and that tin could be the hypothetical repressor of Doc downstream of mid. For example, we have shown that forced expression of tin in all cardioblasts represses Doc (I.R. and M.F., manuscript in preparation). Altogether, these and previously published data suggest a model in which mid functions to activate tin in cardioblasts, while seven-up (svp) prevents mid from doing so in the two Svp-positive cardioblasts in each hemisegment (Gajewski et al., 2000; Lo and Frasch, 2001). Subsequently, tin represses Doc in the four Tin-positive cardioblasts in each hemisegment. By contrast, in mid mutant embryos tin is not activated in most cardioblasts and is unable to repress Doc in these cells. The disruption of the metameric 4 TinCC2 DocC pattern in the dorsal vessel of mid (and mid, H15) mutant embryos is expected to have several consequences that interfere with normal cardiogenesis. First, loss of tin expression would prevent activation of all downstream genes in the dorsal vessel that specifically require the activity of tin and do not respond to Doc. One likely example of this effect that we identified is the reduced expression of Sulfonylurea receptor (Sur), which is expected to affect the normal contractile activity of cardioblasts. Likewise, due to the expansion of Doc expression, terminal differentiation genes that are normally repressed by Doc in two cells per hemisegment would also not be activated upon loss of mid activity. Thus many, if not all, of the differentiation genes that are normally expressed only in the four TinC cardioblasts, as exemplified herein by Drosophila Sur, would not be expressed and not be able to endow this subpopulation of cardioblasts with their specific contractile functions and morphological properties. Conversely, differentiation genes specifically responding to Doc but not to tin would now be expressed in most cardioblasts, rather than in only two cells in each hemisegment. In the heart portion, the two DocC cells in each hemisegment normally form the ostia, which serve as inflow valves for the hemolymph (Lo and Frasch, 2001; Molina and Cripps, 2001; Ponzielli et al., 2002), whereas the TinC cells are thought to serve a predominant role in the contractility of the heart. The typical elongated shape of ostia cells that is

I. Reim et al. / Mechanisms of Development 122 (2005) 1056–1069

displayed by the majority of cardioblasts in the heart of mid mutant embryos does indeed argue for a transformation of ‘regular’ cardioblasts into ostia-like cells. Although it is unlikely that in mid mutant embryos all ectopic DocC cells in the heart would be able to form functional ostia, it is clear that the functional and morphological subdivision into inflow valve cells and regular contractile cardiomyocytes is severely disrupted. As a consequence of all these effects of the lack of mid activity, the dorsal vessel is expected to be functionally compromised and potentially unable to circulate the hemolymph. 3.3. Functional conservation and divergence of Tbx20 genes during cardiogenesis The specific expression and function of Tbx20-related genes during Drosophila and vertebrate cardiogenesis strongly suggests that certain aspects of the regulation and function of these genes have been conserved during evolution. Indeed, functional knock-down experiments in zebrafish and Xenopus have shown that Tbx20 (aka hrT) is required for the normal morphogenesis and functioning of the heart but not for the formation of cardiomyocytes per se, which is reminiscent of our findings for mid/H15 (Szeto et al., 2002; Brown et al., 2005). In zebrafish, loss of hrT function was reported to cause expanded Tbx5 expression in the heart, which is also reminiscent of our observations that loss of mid/H15 activity causes expanded expression of Doc genes in the dorsal vessel (Szeto et al., 2002). However, this phenotype was not observed in Xenopus (Brown et al., 2005). Furthermore, in neither species was there an effect on the expression of the tin-related gene Nkx2-5, at least until the stages when the linear heart tube is formed, which is different to the situation in Drosophila. Unlike in Drosophila, Tbx20 and Nkx2-5 remain largely co-expressed during vertebrate heart development, although it is not known whether this is true at the cellular level. Another difference is that in Drosophila the expression of mid and H15 depends on the early functions of tin and Doc, whereas at least in the mouse system, tbx20 is activated normally in Nkx2-5K/K and tbx5K/K animals (Stennard et al., 2003). Although some of these differences may be due to our incomplete knowledge of many of the details of the regulation and the functions of Tbx20-related genes in the different species and at different stages, in the aggregate they do suggest that there have been modifications of the regulatory circuitry of cardiogenic genes during the evolutionary separation of the vertebrate and arthropod lineages. 3.4. Postscript Just prior to submission of our study, two other manuscripts on mid (a.k.a. neuromancer2) and H15 (a.k.a. neuromancer1) were published (Miskolczi-McCallum et al., 2005; Qian et al., 2005), which differ in some of their key

1067

conclusions from one another and our study. For example, Miskolczi-McCallum et al. equate the two segmental cardioblasts that first express mid at stage 12 with cardioblast progenitors that would produce all six cardioblasts (and two pericardial cells) in each hemisegment via two rounds of cell divisions. Our observation is more compatible with the view that mid expression starts only after these divisions, but occurs sequentially in different cardioblast lineages. In addition, these authors conclude that mid and mid H15 mutant embryos have normal numbers of tin-expressing cardioblasts, whereas we show that the normal activation of tin expression in cardioblasts is one of the major functions of these genes. Most significantly, Qian et al. report that nmr genes (i.e. midCH15) are required for the specification of cardioblasts and that reduced nmr activity causes a switch from myocardial to pericardial cell fates, conclusions that are not supported by our data or those from Miskolczi-McCallum et al. These differences can partially be attributed to the different genetic tools used to knock-down or abrogate the activities of these genes. As a consequence, each of the three reports places the mid/H15 genes at a different position within the hierarchy of gene interactions during cardiac development in Drosophila. In particular, our data imply that mid first acts within the newly-specified cardioblasts and then (together with H15) also during subsequent stages to promote tin expression and proper differentiation of myocardial cells.

4. Experimental procedures 4.1. Drosophila strains and crosses The mutant stocks Df(2L)cl-h2, mid1, mid2, and pnrVX6 were obtained from the Bloomington Stock Center. The mid1 and mid2 alleles were recently characterized molecularly by Buescher et al. (2004). The tin346 allele was described in Azpiazu and Frasch (1993). Mutant stocks were re-balanced with either CyO, wg-lacZ or TM3, eve-lacZ to allow discrimination of homozygous mutants from balancer-carrying embryos by absence of b-Galactosidase staining. Oregon R was used as wild type control. UAS-pnr (2nd chromosome insertion) was obtained through Robert A. Schulz (University of Texas M.D. Anderson Cancer Center, Houston). The svp enhancer trap line AE127 (svplacZ) is a gift from Yasushi Hiromi (National Institute of Genetics, Mishima, Japan). For ectopic expression homozygous GAL4 and UAS-mid flies were crossed and embryos were collected after development at 28 8C. The lines 24B-GAL4 (how-GAL4), SG30-GAL4 (a combination of twi-GAL4 on the X chromosome and 24B-GAL4 on the 3rd), and 2xPE-twiGAL4 were used as GAL4 drivers (Brand and Perrimon, 1993; Greig and Akam, 1993; Baker and Schubiger, 1996).

1068

I. Reim et al. / Mechanisms of Development 122 (2005) 1056–1069

4.2. Generation of UAS-mid transgenes

Acknowledgements

To generate UAS-mid lines, the cDNA insert of the BDGP full-length EST clone RE27439 (Stapleton et al., 2002) was removed from the pFLC-1 vector as a 3.0 kb BamHI–NotI fragment and re-inserted between the BglII and NotI sites of pUAST. Several independent lines were established using standard transformation methods (Rubin and Spradling, 1982).

We thank Hsiu-Hsiang Lee, Patrick Lo, and Stephane Zaffran for sharing materials. Confocal laser scanning microscopy was performed at the MSSM-Microscopy Shared Resource Facility, supported from NIH-NCI shared resources grant (R24 CA095823) and NSF Major Research Instrumentation grant (DBI-9724504). The research was supported by a grant from NIH-NICHD (HD30832) to M.F. and an American Heart Association/Harriet and Robert Heilbrunn Research Fellowship in Cardiovascular Genetic Research to I.R.

4.3. Staining of embryos Embryo fluorescent immunostainings and in situ hybridizations were done essentially as described in Knirr et al. (1999). For double stainings involving two primary antibodies from the same species, embryos were stripped of antibodies after the tyramide reaction by incubating them three times for 10 min in 250 mM glycine/HCl (pH 2.5), 0.1% Tween80 and washing four times with PBT prior to incubation with the second primary antibody. The following antibodies were used: rabbit anti-Mef2 (1:750; gift from Hanh Nguyen, Albert Einstein College, New York) (Bour et al., 1995), rabbit anti-Tin (1:750) (Yin et al., 1997), guinea pig anti-Doc2C3 (1:400 or 1:1200 if detection used Tyramide Reagent after in situ hybridization) (Reim et al., 2003), mouse monoclonal anti-Lbe (1:10) (Jagla et al., 1997b), guinea pig anti-Eve (1:400) and rat anti-Odd (1:500) (Kosman et al., 1998), rabbit anti-Zfh-1 (1:2000) (Broihier et al., 1998), rabbit anti-Toll (1:500; gift from S. Wasserman, UC San Diego), mouse anti-Wg (4D4, 1:30; Developmental Studies Hybridoma Bank, University of Iowa, under the auspices of NICHD), rabbit antib-Galactosidase (Promega; 1:1500), mouse polyclonal anti-Galactosidase (Sigma, 1:200). Primary antibodies were detected with FITC-, Cy3-, or Cy5-conjugated AffiniPure goat anti-rabbit IgG (HCL) (1:200; Jackson Immuno Research Laboratories). For anti-Lbe, anti-Odd, anti-Wg and anti-Toll, detection biotinylated secondary antibodies (1:500) were used in combination with the Vectastain ABC Kit (Vector Laboratories) and fluorescent Tyramide Reagents (PerkinElmer). Digoxigenin-labeled mid anti-sense RNA in situ probe was prepared from the BDGP cDNA RE27439, and the hand and Sur in situ probes from a genomic clone obtained by genomic PCR (Zaffran and Frasch, unpublished). The H15 in situ probe was obtained from a fragment containing the first two exons and the first intron. The fragment was amplified by PCR from genomic DNA using primers CTCGAGAATACGCAAATTACGATAAATAGC and GCTTTACCTGGGCTGGTGATTCC and cloned into pCRII-TOPO (Invitrogen). Another short mid probe without the T-box coding region to rule out any possible crosshybridization with H15 was used to confirm that the expression seen with the RE27439-derived probe is due to mid. Laser scanning microscopy was performed with Leica TCS-SP and Zeiss LSM 510 META microscope systems.

References Alvarez, A.D., Shi, W., Wilson, B.A., Skeath, J.B., 2003. pannier and pointedP2 act sequentially to regulate Drosophila heart development. Development 130, 3015–3026. Azpiazu, N., Frasch, M., 1993. tinman and bagpipe: two homeo box genes that determine cell fates in the dorsal mesoderm of Drosophila. Genes Dev. 7, 1325–1340. Baehrecke, E., 1997. who encodes a KH RNA binding protein that functions in muscle development. Development 124, 1323–1332. Baker, R., Schubiger, G., 1996. Autonomous and nonautonomous Notch functions for embryonic muscle and epidermis development in Drosophila. Development 122, 617–626. Bodmer, R., 1993. The gene tinman is required for specification of the heart and visceral muscles in Drosophila. Development 118, 719–729. Bodmer, R., Frasch, M., 1999. Genetic determination of Drosophila heart development. In: Harvey, R., Rosenthal, N. (Eds.), Heart Development. Academic Press, San Diego, CA, pp. 65–90. Bodmer, R., Jan, L.Y., Jan, Y.N., 1990. A new homeobox-containing gene, msh-2, is transiently expressed early during mesoderm formation of Drosophila. Development 110, 661–669. Bour, B., O’Brien, M., Lockwood, W., Goldstein, E., Bodmer, R., Taghert, P., et al., 1995. Drosophila MEF2, a transcription factor that is essential for myogenesis. Genes Dev. 9, 730–741. Brand, A.H., Perrimon, N., 1993. Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118, 401–415. Broihier, H., Moore, L., Van Doren, M., Newman, S., Lehmann, R., 1998. zfh-1 is required for germ cell migration and gonadal mesoderm development in Drosophila. Development 125, 655–666. Brook, W., Cohen, S., 1996. Antagonistic interactions between wingless and decapentaplegic responsible for dorsal–ventral pattern in the Drosophila leg. Science 273, 1373–1377. Brown, D., Martz, S., Binder, O., Goetz, S., Price, B., Smith, J., Conlon, F., 2005. Tbx5 and Tbx20 act synergistically to control vertebrate heart morphogenesis. Development 132, 553–563. Buescher, M., Svendsen, P., Tio, M., Miskolczi-McCallum, C., Tear, G., Brook, W., Chia, W., 2004. Drosophila T box proteins break the symmetry of hedgehog-dependent activation of wingless. Curr. Biol. 14, 1694–1702. Cripps, R., Olson, E., 2002. Control of cardiac development by an evolutionarily conserved transcriptional network. Dev. Biol. 246, 14–28. Frasch, M., 1995. Induction of visceral and cardiac mesoderm by ectodermal Dpp in the early Drosophila embryo. Nature 374, 464–467. Gajewski, K., Fossett, N., Molkentin, J., Ra, S., 1999. The zinc finger proteins Pannier and GATA4 function as cardiogenic factors in Drosophila. Development 126, 5679–5688.

I. Reim et al. / Mechanisms of Development 122 (2005) 1056–1069 Gajewski, K., Choi, C., Kim, Y., Schulz, R., 2000. Genetically distinct cardial cells within the Drosophila heart. Genesis 28, 36–43. Gelb, B., 2004. Genetic basis of congenital heart disease. Curr. Opin. Cardiol. 19, 110–115. Greig, S., Akam, M., 1993. Homeotic genes autonomously specify one aspect of pattern in the Drosophila mesoderm. Nature 362, 630–632. Griffin, K.J., Stoller, J., Gibson, M., Chen, S., Yelon, D., Stainier, D.Y., Kimelman, D., 2000. A conserved role for H15-related T-box transcription factors in zebrafish and Drosophila heart formation. Dev. Biol. 218, 235–247. Harvey, R., 1996. NK-2 homeobox genes and heart development. Dev. Biol. 178, 203–216. Hiroi, Y., Kudoh, S., Monzen, K., Ikeda, Y., Yazaki, Y., Nagai, R., Komuro, I., 2001. Tbx5 associates with Nkx2-5 and synergistically promotes cardiomyocyte differentiation. Nat. Genet. 28, 276–280. Jagla, K., Frasch, M., Jagla, T., Dretzen, G., Bellard, F., Bellard, M., 1997a. ladybird, a new component of the cardiogenic pathway in Drosophila required for diversification of heart precursors. Development 124, 3471–3479. Jagla, K., Jagla, T., Heitzler, P., Dretzen, G., Bellard, F., Bellard, M., 1997b. ladybird, a tandem of homeobox genes that maintain late wingless expression in terminal and dorsal epidermis of the Drosophila embryo. Development 124, 91–100. Klinedinst, S.L., Bodmer, R., 2003. Gata factor Pannier is required to establish competence for heart progenitor formation. Development 130, 3027–3038. Knirr, S., Azpiazu, N., Frasch, M., 1999. The role of the NK-homeobox gene slouch (S59) in somatic muscle patterning. Development 126, 4525–4535. Kolsch, V., Paululat, A., 2002. The highly conserved cardiogenic bHLH factor Hand is specifically expressed in circular visceral muscle progenitor cells and in all cell types of the dorsal vessel during Drosophila embryogenesis. Dev. Genes Evol. 212, 473–485. Kosman, D., Small, S., Reinitz, J., 1998. Rapid preparation of a panel of polyclonal antibodies to Drosophila segmentation proteins. Dev. Genes Evol. 208, 290–294. Lilly, B., Galewsky, S., Firulli, A., Schulz, R., Olson, E., 1994. D-MEF2: a MADS box transcription factor expressed in differentiating mesoderm and muscle cell lineages during Drosophila embryogenesis. Proc. Natl Acad. Sci. USA 91, 5662–5666. Lilly, B., Zhao, B., Ranganayakulu, G., Paterson, B., Schulz, R., Olson, E., 1995. Requirement of MADS domain transcription factor D-MEF2 for muscle formation in Drosophila. Science 267, 688–693. Lo, P.C., Frasch, M., 1997. A novel KH domain protein mediates cell adhesion processes in Drosophila. Dev. Biol. 190, 241–256. Lo, P.C., Frasch, M., 2001. A role for the COUP-TF-related gene seven-up in the diversification of cardioblast identities in the dorsal vessel of Drosophila. Mech. Dev. 104, 49–60. Lo, P.C.H., Skeath, J.B., Gajewski, K., Schulz, R.A., Frasch, M., 2002. Homeotic genes autonomously specify the anteroposterior subdivision of the Drosophila dorsal vessel into aorta and heart. Dev. Biol. 251, 307–319. Lo, P.C., Frasch, M., 2003. Establishing A-P polarity in the embryonic heart tube: a conserved function of Hox genes in Drosophila and vertebrates?. Trends Cardiovasc. Med. 13, 182–187. Michelson, A.M., Gisselbrecht, S., Zhou, Y., Baek, K.H., Buff, E.M., 1998. Dual functions of the heartless fibroblast growth factor receptor in development of the Drosophila embryonic mesoderm. Dev. Genet. 22, 212–229. Miskolczi-McCallum, C.M., Scavetta, R.J., Svendsen, P.C., Soanes, K.H., Brook, W.J., 2005. The Drosophila melanogaster T-box genes midline and H15 are conserved regulators of heart development. Dev. Biol. 278, 459–472. Molina, M., Cripps, R., 2001. Ostia, the inflow tracts of the Drosophila heart, develop from a genetically distinct subset of cardial cells. Mech. Dev. 109, 51–59.

1069

Nasonkin, I., Alikasifoglu, A., Ambrose, C., Cahill, P., Cheng, M., Sarniak, A., et al., 1999. A novel sulfonylurea receptor family member expressed in the embryonic Drosophila dorsal vessel and tracheal system. J. Biol. Chem. 274, 29420–29425. Nguyen, H., Bodmer, R., Abmayr, S., McDermott, J., Spoerel, N., 1994. DMef2: a Drosophila mesoderm-specific MADS box-containing gene with a biphasic expression profile during embryogenesis. Proc. Natl Acad. Sci. USA 91, 7520–7524. Nu¨sslein-Volhard, C., Wieschaus, E., Kluding, H., 1984. Mutations affecting the pattern of the larval cuticle in Drosophila melanogaster I. Zygotic loci on the second chromosome. Roux’s Arch. Dev. Biol. 193, 267–282. Pikkarainen, S., Tokola, H., Kerkela, R., Ruskoaho, H., 2004. GATA transcription factors in the developing and adult heart. Cardiovasc. Res. 63, 196–207. Plageman, T., Yutzey, K., 2005. T-box genes and heart development: putting the ‘T’ in heart. Dev. Dyn. 232, 11–20. Ponzielli, R., Astier, M., Chartier, A., Gallet, A., Therond, P., Semeriva, M., 2002. Heart tube patterning in Drosophila requires integration of axial and segmental information provided by the Bithorax Complex genes and hedgehog signaling. Development 129, 4509–4521. Qian, L., Liu, J., Bodmer, R., 2005. neuromancer Tbx20-related genes (H15/midline) promote cell fate specification and morphogenesis of the Drosophila heart. Dev. Biol. 279, 509–524. Reim, I., Lee, H.H., Frasch, M., 2003. The T-box-encoding Dorsocross genes function in amnioserosa development and the patterning of the dorsolateral germ band downstream of Dpp. Development 130, 3187–3204. Rizki, T.M., 1978. The circulatory system and associated cells and tissues. In: Ashburner, M., Wright, T.R.F. (Eds.), The Genetics and Biology of Drosophila. Academic Press, New York/London, pp. 397–452. Rubin, G.M., Spradling, A.C., 1982. Genetic transformation of Drosophila with transposable element vectors. Science 218, 348–353. Stapleton, M., Carlson, J., Brokstein, P., Yu, C., Champe, M., George, R., et al., 2002. A Drosophila full-length cDNA resource. Genome Biol. 3 (RESEARCH0080). Stennard, F., Costa, M., Elliott, D., Rankin, S., Haast, S., Lai, D., et al., 2003. Cardiac T-box factor Tbx20 directly interacts with Nkx2-5, GATA4, and GATA5 in regulation of gene expression in the developing heart. Dev. Biol. 262, 206–224. Szeto, D., Griffin, K., Kimelman, D., 2002. hrT is required for cardiovascular development in zebrafish. Development 129, 5093–5101. van Doren, M., 2004. Development of the somatic gonad and fat bodies. In: Sink, H. (Ed.), Drosophila Muscle Development, Landes, Georgetown, TX, http://www.eurekah.com/abstract.php?chapidZ2028andbookidZ 162andcatidZ20. Wang, J., Tao, Y., Reim, I., Gajewski, K., Frasch, M., Schulz, R.A., 2005. Expression, regulation, and requirement of the Toll transmembrane protein during dorsal vessel formation in Drosophila. Mol. Cell Biol. 25, 4200–4210. Ward, E., Skeath, J., 2000. Characterization of a novel subset of cardiac cells and their progenitors in the Drosophila embryo. Development 127, 4959–4969. Wharton, K., Crews, S., 1993. CNS midline enhancers of the Drosophila slit and Toll genes. Mech. Dev. 40, 141–154. Yin, Z., Xu, X.-L., Frasch, M., 1997. Regulation of the Twist target gene tinman by modular cis-regulatory elements during early mesoderm development. Development 124, 4871–4982. Zaffran, S., Frasch, M., 2002. Early signals in cardiac development. Circ. Res. 91, 457–469. Zaffran, S., Astier, M., Gratecos, D., Semeriva, M., 1997. The held out wings (how) Drosophila gene encodes a putative RNA-binding protein involved in the control of muscular and cardiac activity. Development 124, 2087–2098.