The Interactive Fly
Genes involved in tissue and organ development
Drosophila serotonin receptors stimulate fluid secretion in the Malpighian tubules
Genes expressed in Malpighian tubule development
The tubules arise during embryogenesis as four protuberances extending from the proctodeum. As this tissue is considered ectodermal, Malpighian tubules are classified as an ectodermally derived secretory tissue. The protuberances grow, first by cell proliferation and then by extensive rearrangement of the cells, to produce elongated blind end tubes composed of a single-cell-layered epithelium (Hoch, 1994).
The development of Malpighian tubules reveals an essential role for the gene Krüppel. In each Malpighian tubule, one cell is singled out, the tip cell, whose function during embryogenesis is to promote cell division in its neighbours. The tip cell arises by division of a tip mother cell, which is selected from a cluster of equivalent cells, each expressing Krüppel in each tubule primordium. Each cluster is delineated by the expression of proneural genes; the selection of a single cell from each group involves lateral inhibition, mediated by the neurogenic genes. achaete is responsible for tip cell allocation, but Kr acts as the selector gene, responsible for tip cell fate. The tip cell directs the growth of the Malpighian tubules and organizes the mitotic response and migration of the other cells forming each tubule (Hoch, 1994). Therefore Krüppel is responsible for cell fate in the Malpighian tubules, a function quite distinct from Krüppel's role as a gap gene.
Drosophila possesses two pair of Malpighian tubules. The right pair of tubules project forward from their point of insertion within the hindgut and lie at the anterior end of the abdomen, and the left pair extend backwards so that their tips become attached to the posterior part of the hindgut. Each tubule pair unites to form a common ureter, which enters the intestine between the midgut and hindgut. The two anterior Malpighian tubules are classically described as comprising a distal initial segment and a proximal main segment, joined by a narrow transitional segment; the two posterior tubules, in contrast, were thought to consist solely of a main segment. Contemporary studies, using enhancer trap lines, which place reporter genes under the control of tissue specific enhancers, confirm this viewpoint and thus the nomenclature "initial," "transitional," and "main" segments has been adopted to described these genetically deduced domains (Sözen, 1997).
Enhancer trap studies reveal an unexpected complexity in Malpighian tubules in terms of both regions and cell types. Enhancer trap lines that delineate the initial and transitional segments of anterior tubules, reveal previously undescribed analogous domains in posterior tubules. It is also possible to subdivide the main segments. While the transitional-main segment boundary has been established in accordance with classical studies, an additional domain is found marking the lower third of the tubule and the ureter. This latter region, in turn, can be resolved into three subregions: a lower tubule and an upper and lower ureter (Sözen, 1997).
Previous studies have described just two tubule cell types: principal (type I) and secondary or stellate (type II). Both can be further subdivided. Principal cells, for example, comprise at least two distinct subpopulations. Thus there appear to be differences in otherwise indistinguishable cells with respect to enhancer trap expression patterns and presumably with respect to function as well. Type II cells are distributed evenly throughout the initial, transitional and main segments of posterior tubules and within the main segment of anterior tubules. None of the enhancer trap markers mark cells in the lower tubule or ureter, suggesting that the mechanism by which type II cells are specified respects the newly defined lower tubule boundary. Several lines mark a "tiny" cell type found in lower tubules and posterior midgut but do not mark the same genetic domain as stellate cells. Possibly these previously undescribed cells are counterparts of the myoendocrine cells recently described in Formica. These cells may monitor the fluid collected in the ureter and secrete neurohormones basally into the hemolymph to regulate muscle contractility or ion transport (Sözen, 1997).
Do discrete physiological properties map to the genetic domains that have been identified? With respect to a number of different transport processes, this is indeed the case. The obvious functional property of the tubule is to secrete urine. It has been reported that the initial segment of Drosophila anterior tubule does not secrete detectable fluid, that the lower third of the tubule is reabsorptive, and that only the main segment is responsible for fluid production. High levels of proton-pumping V-ATPases energize apical plasm membranes of several epithelia, including Malpighian tubules. The B-subunit of Drosophila V-ATPase is expressed in the initial and transitional segments and is much weaker in the reabsorptive main segment. The main segment consists entirely of the large, principal cell. This provides the first evidence that cation transport into the lumen of Malpighian tubules may be a unique property of principal, rather than type II, cells. A putative aquaporin has been cloned in Drosophila: this channel is found in stellate cell basolateral membranes. Given that stellate cells are found in secretory but not reabsorptive tubule regions, they may well play an essential role in fluid secretion. Another function ascribed to Malpighian tubules is the secretion of organic metabolites. This function is confined to the main segment. It is clear that the staining pattern for alkaline phosphatase precisely matches the lower tubule boundary, and is associated with the reabsorptive, rather than the secretory, domain of the tubule (Sözen, 1997).
Secretion by Malpighian tubules is under hormonal control. CAP2b, a cardioacceleratory peptide, is present in Drosophila and stimulates Malpighian tubule fluid secretion via cGMP, which in turn stimulates the nitric oxide signaling pathway. Liquid chromatography analysis of adult Drosophila reveals the presence of a CAP2b-like peptide, that coelutes with Manduca sexta CAP2b and synthetic CAP2b and that has CAP2b-like effects on the M. sexta heart. CAP2b stimulation elevates tubule cGMP levels but not those of cAMP. Both CAP2b and cGMP increase the transepithelial potential difference, suggesting that stimulation of vacuolar ATP action underlies the corresponding increases in fluid secretion (Davies, 1995).
Calcium mobilization in identified cell types within an intact renal epithelium (the Drosophila Malpighian tubule) was studied by GAL4-directed expression of an aequorin transgene. Aequorin is a Ca2+ sensitive liminescent protein isolated from the coelenterate Aequorea victoria. It is a complex of apoaequorin, a 21 kDA polypeptide, and coelenterazine, a hydrophobic luminophore. Aequorin is used for monitoring Ca2+ changes. CAP2b, causes a rapid, dose-dependent rise in cytosolic calcium in only a single, genetically-defined, set of 77 principal cells in the main (secretory) segment of the tubule. In the absence of external calcium, the CAP2b-induced calcium response is abolished. In Ca2+-free medium, the endoplasmic reticulum Ca2+-ATPase inhibitor, thapsigargin, elevates [Ca2+]i only in the smaller stellate cells, suggesting that principal cells do not contain a thapsigargin-sensitive intracellular pool. Assays for epithelial function confirm that calcium entry is essential for CAP2b to induce a physiological response in the whole organ. The data suggest a role for calcium signaling in the modulation of the nitric oxide signaling pathway in this epithelium. CAP2b must act to increase fluid secretion rates solely by an initial rise of [CA2+]i in principal cells. CAP2b stimulates tubule Nitric oxide synthase activity. It is probable that the CAP2b induced rise in [CA2+]i is sufficient to trigger the activation of Drosophila calcium sensitive Nitric oxide synthase. The maximal CAP2b concentrations employed elevate principal cell calcium levels from 87 to 255 nM, a value close to the EC50 of Drosophila NOS. This implies that Drosophila Nos is responsive over the range of the CAP2b concentrations employed. This may account for the observation that thapsigargin treatment results in increased basal cGMP levels that are not further increased on CAP2b stimulation. Thus the data provide strong evidence for a calcium-mediated link between CAP2b and NOS/cGMP activation of fluid secretion. The GAL4-targeting system allows general application to studies of cell-signaling and pharmacology that does not rely on invasive or cytotoxic techniques (Rosay, 1997).
Other hormones likely to be involved in Malpighian tubule function are the leucokinins. Leucokinins are a group of widespread insect hormones. In tubules, their major action is to raise chloride permeability through stellate cells by binding to receptors on the basolateral membrane, and so ultimately to enhance fluid
secretion (Julian Dow, personal communication).
Every living cell must detect, and respond appropriately to, external signals. Thus, the functions of intracellular second messengers, such as guanosine 3'5'-cyclic monophosphate (cGMP), adenosine 3'5'-cyclic monophosphate (cAMP), and intracellular calcium, are intensively studied. However, artifact-free manipulation of these messengers is problematic, and simple pharmacology may not allow selective intervention in distinct cell types in a real, complex tissue. A method has been devised by which second messenger levels can be manipulated in cells of choice using the GAL4/UAS system. By placing different receptors (rat atrial natriuretic peptide [ANP] receptor and Drosophila serotonin receptors [5HTDro7 and 5HTDro1A]) under UAS control, they can be targeted to arbitrary defined populations of cells in any tissue of the fly, and second messenger levels can be manipulated simply by adding the natural ligand. The potential of the system is illustrated in the Drosophila renal (Malpighian) tubule, where each receptor has been shown to stimulate fluid secretion, to act through its cognate second messenger, and to be blocked by appropriate pharmacological antagonists. The results have uncovered a new role for cGMP signaling in tubules and also demonstrate the utility of the tubule as a possible in vivo test bed for novel receptors, ligands, or agonists/antagonists (Kerr, 2004).
Ectopic expression of the rat ANP receptor (GC-A) in tubules was achieved under control of both principal cell and stellate cell GAL4 drivers and a heat-shock (hs) promoter. Expression of the GC-A transgene was confirmed by RT-PCR. Expression of GC-A in tubules confers sensitivity to ANP, with resultant production of cGMP. Measurement of cAMP levels in tubules that express GC-A in principal cells (using the c42 GAL4 driver), stellate cells (using the c742 driver), or ubiquitously, show that cGMP levels are stimulated neither by GC-A receptor, nor by ANP, alone. ANP raises cGMP in a dose-dependent manner in c42-GC-A and c724-GC-A tubules, with a nearly 4-fold maximal increase in cGMP levels in c42-GC-A tubules and a 2-fold cGMP increase in c724-GC-A tubules. Since stellate cells make up only a small fraction of the tubule volume, the apparently lower fold increase in stellate cell cGMP probably reflects a larger absolute rise in cGMP in these cells. The EC50 for ANP in both principal and stellate cells is similar, 10-8 M (Kerr, 2004).
Similar increases in fluid transport are observed upon stimulation with ANP, when GC-A is expressed either in only principal cells or in stellate cells. Although cGMP signaling has been shown in principal cells, a diuretic role for cGMP in stellate cells had not been demonstrated before. The effects of cGMP in principal and stellate cells are additive; when GC-A is expressed ubiquitously under hs control, maximal secretion rates are higher than when it is targeted to either cell alone (Kerr, 2004).
cGMP signals in principal cells may activate CNG-type calcium channels, resulting in calcium increase and fluid transport. Targeted expression of the calcium reporter, aequorin, was used to measure changes in intracellular calcium ([Ca2+]i) in GC-A tubules. In tubule principal cells, a biphasic elevation of [Ca2+]i is observed upon ANP stimulation, followed by a sustained secondary rise in [Ca2+]i. By contrast, no change in stellate cell [Ca2+]i was observed upon ANP challenge. Thus, major cellular targets for cGMP in principal cells are CNG channels. Stellate cells, however, must contain uncharacterized cGMP-activated targets that modulate fluid transport (Kerr, 2004).
ANP-mediated cGMP signaling and increased fluid transport are a result of specific ligand-receptor interactions, since the ANP antagonist, anantin, abolishes ANP stimulation of both cGMP and fluid transport in hs-GC-A tubules in a dose-dependent manner (Kerr, 2004).
Pilot experiments have shown that Drosophila tubules are insensitive to 5HT; there is thus scope to modulate second messengers by ectopic expression of the cognate GPCRs. The Drosophila 5HT7Dro receptor raises cAMP levels in cultured cells by activating adenylate cyclase. When 5HT7Dro is expressed in either principal or stellate cells using the appropriate GAL4 drivers, 5HT induces dose-dependent production of cAMP. EC50 for 5HT in both lines was 10−7 M. Similarly, the heat shock construct elicits elevation of [cAMP] upon 5HT treatment. Controls (non-heat-shocked hs-5HT7Dro and heat-shocked wild-type or parental lines) show no response to 5HT. Furthermore, there is no detectable impact on cGMP levels when 5HT7Dro is driven either in principal cells, stellate cells, or ubiquitously (Kerr, 2004).
As would be expected from the literature, increased [cAMP] in principal cells stimulates fluid transport. However, as for cGMP, a previously undocumented diuretic role of cAMP in stellate cells was uncovered, and the maximal rates of 5HT-induced fluid secretion observed with c42-5HT7Dro and c724-5HT7Dro tubules appear to be additive: the sum of maximal rates is approximately equal to the maximal 5HT-induced rate observed in heat-shocked hs-5HT7Dro tubules. It was thus hypothesized that, if stellate cells have the machinery to respond functionally to cAMP and cGMP, then they are also likely to have the machinery to produce the signals (Kerr, 2004).
As with GC-A, the possibility that [Ca2+]i may be altered upon activation of 5HT7Dro was investigated. In principal cells of 5HT7Dro/c42-aeq tubules, 5HT stimulation results in a biphasic elevation of [Ca2+]i, similar to that seen in GC-A/c42-aeq tubules. However, no [Ca2+]i rise was observed in 5HT-stimulated 5HT7Dro/c710-aeq tubules. These results are consistent with both cAMP and cGMP acting on a CNG channel that is expressed only in principal, and not in stellate, cells (Kerr, 2004).
It proved possible to reproduce the known pharmacology of this 5HT receptor. The antagonist (+)-butaclamol almost completely attenuated 5HT-stimulated production of cAMP in a dose-dependent manner, with an IC50 of 2.5 × 10−8 M. This agrees precisely with values obtained for the receptor in cell lines. A maximal dose of (+)-butaclamol (10−5 M) also reduced 5HT-stimulated fluid transport in hs-5HT7Dro tubules (Kerr, 2004).
Another Drosophila 5HT receptor, 5HT1ADro, is known to mobilize intracellular calcium. Expression of 5HT1ADro in either principal or stellate cells results in 5HT-induced calcium responses. As would be expected from the actions of capa and leucokinin (ligands known to act through [Ca2+]i), 5HT also stimulated fluid transport, when 5HT1ADro was expressed in either principal cells or stellate cells, and no effect was seen without a GAL4 driver. Although a comprehensive dose-response curve was not performed, 5HT-stimulated fluid transport was inhibited by yohimbine. The effective concentration for yohimbine in tubules (10−5 M) is comparable with the Ki (18 μM) obtained for this receptor in cell lines (Kerr, 2004).
The 5HT1ADro receptor has been previously shown to inhibit adenylate cyclase in vitro. Accordingly, cAMP and cGMP were measured in 5HT-stimulated c42-5HT1ADro and c724-5HT1ADro tubules. Activation of the 5HT1ADro receptor in tubules does not affect cyclic nucleotide levels, and thus the stimulatory effects of 5HT on fluid transport are solely due to stimulation of increased [Ca2+]i (Kerr, 2004).
Although receptor guanylate cyclases (like GC-A) are self-sufficient, the strategy of modulating second messengers by ectopic expression of GPCRs depends on the expression of cognate G proteins in the target cell type. A priori, it was expected that most G proteins would be widely expressed, but this was confirmed in tubules by RT-PCR with primers against all known G protein α, β, and γ subunits. As predicted, all the Gα subunit genes found in Drosophila are expressed in tubules. Tubules also express the full complement of genes encoding the Gγ subunit; the only G protein subunit that does not appear to be expressed in tubules is Gβ76C (Kerr, 2004).
Although a particular target cell type might not express cognate G proteins, an RT-PCR strategy compared with measurement of second messengers should be informative. As well as the apoaequorin calcium reporter, others are now available, and successful cAMP and cGMP measurements in the tiny (160-cell) tubule suggest that radioimmunoassay is sufficiently sensitive for most Drosophila tissues of interest (Kerr, 2004).
Obviously, the target cell must not normally express receptors for, or respond to, the ligand of choice. In the case of Drosophila, this condition is satisfied; there is no atrial natriuretic peptide-like sequence encoded within the Drosophila genome. For 5HT receptors or other GPCRs, more caution must be exercised; nonetheless, standard experimental controls would quickly identify any problems. In the case of 5HTDro1A, expression has only been documented in some cells of the embryonic nervous system; 5HTDro7 expression has been found in cells of the embryonic ventral midline, and in adult head, but not body. It is thus likely that these constructs will be useful in most Drosophila nonnervous tissues (Kerr, 2004).
This study demonstrates successful ectopic expression of vertebrate and Drosophila receptors in Malpighian tubules and quantifies the effects of such expression on signal transduction pathways and physiological output. There are clear results from this approach: (1) the interactions between 3 second messenger pathways were studied in unprecedented detail, in two cell types, in an organotypic context; (2) the results validated all that was known about signaling through known neuropeptides in the tubule; (3) a diuretic role for cAMP and cGMP in stellate cells was demonstrated, inviting the intriguing question as to which extracellular ligands normally activate these pathways; (4) this conserved pharmacology of a vertebrate receptor (rat GC-A) expressed in a model organism illustrates another possibility for functional genomics, i.e., that novel genes could be characterized relatively cheaply and easily by ectopic expression in a model organism where detailed organotypic-phenotypic analysis is possible. The Drosophila Malpighian tubule is an ideal such 'test bed' for genes where organotypic analysis may be important for normal function (Kerr, 2004).
The utility of this experimental system is more general, allowing sensitive and specific intervention in second messenger signaling in any Drosophila tissue for which there exists a GAL4 driver. Combined with the ready availability of real-time calcium reporters in Drosophila and the possibility of measuring cAMP and cGMP similarly, using transgenic FRET reporters, this simple model organism now has an impressive genetic toolbox for cell signaling studies (Kerr, 2004).
For more information about Malpighian tubule function, see The Drosophila melanogaster Malpighian tubule WWW page, maintained by Julian Dow at the University of Glasgow.
In addition, the Dow laboratory maintains a sensitive map illustrating the physiology of Malpighian tubule secretory cells. Functioning of the V-ATPase is described at Julian Dow's V-ATPase site.
Davies, S. A., et al. (1995). CAP2b, a cardioacceleratory peptide, is present in Drosophila and stimulates tubule fluid secretion via cGMP. Am. J. Physiol. 269: R1321-1326. 8594932
Hoch, M., Broadie, K., Jackle, H. and Skaer, H. (1994). Sequential fates in a single cell are established by the
neurogenic cascade in the Malpighian tubules of Drosophila. Development 120: 3439-3450. 7821213
Kerr, M., Davies, S. A. and Dow, J. A. T. (2004). Cell-specific manipulation of second messengers: A toolbox for integrative physiology in Drosophila. Curr. Biol. 14: 1468-1474. 15324663
Rosay, P., et al. (1997). Cell-type specific calcium signalling in a Drosophila
epithelium. J. Cell Sci. 110: 1683-1692
Sözen, M. A., et al. (1997). Functional domains are specified to single-cell resolution in a Drosophila epithelium. Proc. Natl. Acad. Sci. 94: 5207-5212
back to a list of genes expressed in Malpighian tubule development Genes involved in tissue and organ development
Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.
The Interactive Fly resides on the
Society for Developmental Biology's Web server.