Regulation of imaginal disc cell size, cell number ... - Semantic Scholar

Research Paper

1019

Regulation of imaginal disc cell size, cell number and organ size by Drosophila class IA phosphoinositide 3-kinase and its adaptor David Weinkove*†‡, Thomas P. Neufeld§, Thomas Twardzik¶, Michael D. Waterfield* and Sally J. Leevers* Background: Class IA phosphoinositide 3-kinases (PI 3-kinases) have been implicated in the regulation of several cellular processes including cell division, cell survival and protein synthesis. The size of Drosophila imaginal discs (epithelial structures that give rise to adult organs) is maintained by factors that can compensate for experimentally induced changes in these PI 3-kinaseregulated processes. Overexpression of the gene encoding the Drosophila class IA PI 3-kinase, Dp110, in imaginal discs, however, results in enlarged adult organs. These observations have led us to investigate the role of Dp110 and its adaptor, p60, in the control of imaginal disc cell size, cell number and organ size. Results: Null mutations in Dp110 and p60 were generated and used to demonstrate that they are essential genes that are autonomously required for imaginal disc cells to achieve their normal adult size. In addition, modulating Dp110 activity increases or reduces cell size in the developing imaginal disc, and does so throughout the cell cycle. The inhibition of Dp110 activity reduces the rate of increase in cell number in the imaginal discs, suggesting that Dp110 normally promotes cell division and/or cell survival. Unlike direct manipulation of cell-cycle progression, manipulation of Dp110 activity in one compartment of the disc influences the size of that compartment and the size of the disc as a whole. Conclusions: We conclude that during imaginal disc development, Dp110 and p60 regulate cell size, cell number and organ size. Our results indicate that Dp110 and p60 signalling can affect growth in multiple ways, which has important implications for the function of signalling through class IA PI 3-kinases.

Background

Addresses: *Ludwig Institute for Cancer Research, 91 Riding House Street, London W1P 8BT, UK and the Department of Biochemistry and Molecular Biology, University College London, Gower Street, London WC1E 6BT, UK. †The MRC Graduate Programme, University College London, UK. §Institute of Human Genetics, Department of Genetics, Cell Biology and Development, Box 206 Mayo, 420 Delaware Street SE, University of Minnesota, Minneapolis, MN 55455, USA. ¶Lehrstuhl für Genetik, Biozentrum, Am Hubland, D-97074 Würzburg, Germany. Present address: ‡The Netherlands Cancer Institute, Department of Molecular Biology, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands. Correspondence: Sally J. Leevers E-mail: [email protected] Received: 14 July 1999 Revised: 18 August 1999 Accepted: 19 August 1999 Published: 9 September 1999 Current Biology 1999, 9:1019–1029 http://biomednet.com/elecref/0960982200901019 0960-9822/99/$ – see front matter © 1999 Elsevier Science Ltd. All rights reserved.

During animal development, extracellular ligands utilise intracellular signalling pathways to coordinate patterning and growth (a term used here to describe an increase in mass), generating organs of a reproducible size and shape [1–3]. The size of an organ depends both upon the number of cells in that organ and the size of those cells. Thus, the growth of an organ can result from changes in cell size or cell number or both. Studies of growth in cell culture systems have focused on the regulation of changes in cell number rather than on the regulation of changes in cell size. In this paper, we investigate how changes in both cell size and cell number are regulated by the Drosophila phosphoinositide 3-kinase (PI 3-kinase), Dp110, and its adaptor, p60, to control the growth of an epithelial organ — the Drosophila imaginal disc.

molecules containing Src homology 2 (SH2) domains that recognise phosphorylated tyrosine residues located three amino acids amino-terminal to a methionine residue (pYXXM; in single-letter amino acid code where X denotes any amino acid; [5,6]). Upon ligand activation, a subset of receptor tyrosine kinases (RTKs) generate class IA PI 3-kinase adaptor binding sites by phosphorylating YXXM motifs either on themselves (for example the platelet-derived growth factor receptor) or on substrate proteins (for example insulin receptor substrate-1, IRS-1). Ligation of the adaptor SH2 domains to pYXXM peptides increases class IA PI 3-kinase activity in vitro and is thought to increase the access of the enzyme to substrates in the plasma membrane in vivo [7]. In addition, class IA PI 3-kinases associate with active Ras, which also increases kinase activity in vitro and might promote substrate access in vivo [8,9].

PI 3-kinases phosphorylate phosphoinositides on the 3′ hydroxyl group of the inositol ring and can be grouped into three classes (I–III) according to their primary structure, mode of regulation and substrate specificity [4]. Class IA PI 3-kinases are found in complex with adaptor

Class IA PI 3-kinase activation has several consequences that might affect organ growth. For example, class IA PI 3-kinases can promote cell proliferation, protect cells from apoptosis, increase protein synthesis and upregulate

1020

Current Biology Vol 9 No 18

glucose metabolism [7,10–12]. In addition, class IA PI 3-kinase activation affects membrane trafficking events, cytoskeletal rearrangements and cell differentiation [7,10]. It is unclear whether all of these effects are the direct consequence of class IA PI 3-kinase activation or secondary effects of more immediate responses. Most of these processes have been shown to be regulated by the serine/threonine kinases phosphoinositide-dependent kinase 1 (PDK1) and protein kinase B (PKB), both of which contain 3-phosphoinositide-binding pleckstrin homology (PH) domains [4,13–17]. In Caenorhabditis elegans the class IA PI 3-kinase, AGE-1, controls the decision to form dauer larvae, an alternative third larval stage adopted under conditions of overcrowding and starvation [18,19]. The cellular function of AGE-1 and the mechanism by which it alters the life cycle of the nematode is unclear. Previously, we have shown that Drosophila possess one class IA PI 3-kinase, Dp110, and one SH2-domain-containing adaptor, p60 [20–23]. Molecular and biochemical studies have failed to identify further isoforms of these proteins in Drosophila. We have investigated Dp110 function in imaginal discs, which are epithelial organs that grow extensively during larval development and are reorganised during metamorphosis to generate adult epidermal organs such as the eyes, wings and legs [24]. Over-expression of Dp110 in developing wing imaginal discs generates wings that are enlarged but patterned normally, whereas overexpression of a dominant-negative version of Dp110 results in smaller wings that are again patterned normally. Similarly, overexpression of wild-type or dominant-negative Dp110 in eye imaginal discs results in larger or smaller eyes, respectively, though in this case the ommatidial pattern is slightly perturbed [21]. This effect of a class IA PI 3-kinase on the development of Drosophila imaginal discs is an important observation, as numerous experiments have shown that imaginal disc size is remarkably resistant to perturbation. For example, when imaginal discs are transplanted from larvae into adult hosts, they grow to the same size regardless of the developmental stage of the transplanted disc [25,26]. This observation implies that, as with the size of many mammalian organs, the size of imaginal discs is controlled by intrinsic factors [2,3,27]. It has been shown that a number of cellular processes can be disrupted within the imaginal disc during its development without having a major impact on disc size. For example, when small regions of a disc are ablated or apoptosis is induced early in development, enhanced compensatory proliferation of the surviving cells results in the production of a disc that is of wild-type size and has normal patterning [28–30]. Other experiments have shown that increasing or reducing cell division in one compartment size of the wing imaginal disc increases or reduces cell number but does not affect final compartment or disc size [31,32]. Rather, the compartment continues to grow at

a normal rate and therefore contains fewer bigger cells or more smaller cells. Finally, studies of the effects of mutations in the Minute class of genes, several of which encode components of the ribosomal machinery, suggest that inhibition of general protein synthesis also does not affect final disc size. Although organisms that are heterozygous for mutations in the different Minute genes grow at a reduced rate (presumably because the rate of protein synthesis is reduced), their development is delayed and many of them achieve wild-type adult body sizes [33]. In addition, clones of wild-type imaginal disc cells induced in a Minute–/+ background proliferate at an enhanced rate and consequently outgrow their Minute–/+ neighbours, but the final size and shape of the adult organ is unaffected [34]. Together, these data demonstrate that individual perturbation of cell survival, cell division or protein synthesis, all of which are processes thought to be regulated by class IA PI 3-kinases, does not affect imaginal disc size. Disc and adult organ size can be increased by overexpression of the secreted molecules Wingless, Decapentaplegic (Dpp) and Delta [35–37], by induction of clones of cells in which tumour suppressor genes are mutated [38], or by over-expression of activated Ras [39]. All these interventions simultaneously disrupt patterning and/or the structural integrity of the disc. Thus, the ability of Dp110 to modify organ size without affecting patterning is unusual and might reflect the fact that Dp110 and its adaptor, p60, are involved in the regulation of organ size. We have, therefore, further investigated the function of Dp110 and p60 in imaginal disc development and in the regulation of organ size.

Results Dp110 and p60 are essential genes required for larval growth

To investigate the requirement for the Drosophila class IA PI 3-kinase and its adaptor during development, mutations in the Dp110 and p60 genes were generated. First, the Dp110 and p60 genomic regions were characterised, then null alleles were generated by combining small Dp110 and p60 deletions with genomic constructs that rescued the activity of neighbouring genes also disrupted by the deletions. Dp110 is immediately distal of the Hairless gene (H), so small H and Dp110 deletions were generated by imprecise excision of the HD179 P element. The Dp110A deletion removes the 3′ end of Dp110 and all of the H gene (Figure 1a; see Materials and methods section). To generate null alleles of Dp110, an H genomic construct, P[gH], was recombined onto the Dp110A mutant chromosome. Dp110A P[gH] homozygotes display a lethality that is completely rescued by an additional genomic construct containing Dp110, termed P[gDp110]. As well as removing H and Dp110, a second deletion, Dp110B, is thought to remove additional genes because it displays a recessive lethality that is not rescued by P[gH] and P[gDp110]. Dp110B has therefore been used only in combination with Dp110A. The p60 gene is immediately proximal of

Research Paper Drosophila PI 3-kinase regulates cell and organ growth Leevers et al.

1021

Figure 1 (a)

(b) Distal

Proximal Df(2L)al

Proximal

p60A

Distal

p60B k14504 k00616 l(2)06751

Dp110A HD179 H

BamHI EcoRI Dp110

smo

gH

BamHI EcoRI

PLC21C

gR10 gDp110

1kb

dU2AF38 dSTI1 p60

319/11

1kb

(d)

(c)

gp60

Control

p60–/–

Control

p60–/–

Dp110–/– Current Biology

Mutation of Dp110 or p60 is lethal and inhibits larval growth. (a) A map of the Dp110 genomic region (92E12–13) showing the H and Dp110 genes, the Dp110A deletion and HD179 P element from which it was generated and the genomic fragments included in the constructs P[gH] and P[gDp110]. (b) A map of the p60 genomic region (21B8–C1) showing the p60, smo, dU2AF38, dSTI1, p60 and PLC21C genes, the p60A and p60B deletions, the proximal breakpoint of Df(2L)al, the k00616, k14504, l(2)06751 and 319/11 P element insertions and the genomic fragments included in the constructs P[gR10] and P[gp60]. For each gene, the intron–exon

structure, coding sequence (closed boxes) and non-coding sequence (open boxes) are indicated. (c) Whole mounts showing that Dp110– and p60– larvae are reduced in size. The control and p60– larvae shown are wandering larvae whereas the Dp110– larva is of the maximum size observed. (d) A comparison of the size of wing imaginal discs removed from wandering control and p60– larvae. No imaginal discs could be recovered from Dp110– mutant larvae. Larvae had the following genotypes. Control genotype: y w. Dp110–/– genotype: y w; Dp110A P[gH]/Dp110B P[gH]. And p60–/– genotype: y w; p60A/p60B; P[gR10].

dU2AF38 and dSTI1 at 21B8–C1 (Figure 1b; see Materials and methods section). The deletion mutants p60A and p60B remove p60, dU2AF38 and dSTI1. Null alleles of p60 can be generated by combining these deletions with P[gR10], a genomic construct containing dU2AF38 and dSTI1. The three genetic combinations p60A; P[gR10], p60B; P[gR10] and p60A/p60B; P[gR10] show a recessive lethality that is rescued completely by an additional genomic construct containing p60 (P[gp60]). The p60A deletion also removes the first exon of PLC21C. Although our analysis demonstrates that this exon is not essential for viability, p60A was used only in combination with p60B. In summary, we have generated null mutations in both Dp110 and p60 that can be used to study their function.

develop normally until the early third larval instar, either because maternal mRNA and/or protein provides Dp110 or p60 function (C. Coelho and S.J.L., unpublished data), or because Dp110 and p60 are not required for embryogenesis and early larval development. Dp110 mutant larvae (Dp110A P[gH]/Dp110B P[gH]) stop growing early in the third larval instar. Although these larvae continue to move and feed, surviving for approximately 20 days, they fail to increase in size (Figure 1c). Furthermore, they are transparent in appearance, possess no discernible imaginal discs and their fat bodies are reduced in size. Their nervous system, trachea and gut, however, appear normal. The p60 mutant larvae (p60A/p60B; P[gR10]) have a similar but less severe phenotype. They continue to grow in the third larval instar but at a reduced rate. When these larvae begin to wander (5 days after wild-type larvae) they are still reduced in size, with underdeveloped fat bodies and small imaginal discs (Figures 1c,d). Some p60 mutant

To investigate the impact of removing Dp110 or p60, the development of homozygous Dp110 or p60 mutant embryos was monitored. Dp110 and p60 mutant embryos

1022


Figure 2

(a)

larvae form small pupal cases but die without any indication of pupal development. These phenotypes demonstrate that Dp110 and p60 are essential genes and are consistent with Dp110 and p60 being required for normal larval imaginal disc growth. These experiments do not reveal, however, whether Dp110 and p60 are required in the discs themselves or in other larval organs that influence disc growth.

(b)

*

Dp110 and p60 are autonomously required for adult cells to achieve their normal size

To investigate whether Dp110 and p60 are required within the imaginal discs themselves, we generated mitotic clones of mutant cells in heterozygous Dp110 or p60 flies by using the Flipase (Flp) and Flp recognition target (FRT) site-specific recombination system [40]. Mutant phenotypes were examined in the adult eye, a highly ordered structure made up of repeated units or ommatidia, each of which contains the same pattern of differentiated photoreceptors and accessory cells. We found that both Dp110– and p60– eye clones form indented patches containing ommatidia that are reduced in size (Figures 2a,b). The internal eye structure was examined using tangential sections in which Dp110– clones (–/–) were distinguished from their corresponding sister clones or twin-spots (+/+) and the surrounding heterozygous cells (+/–) by an absence of red pigment granules (Figure 2c). The p60– clones of cells contain a small amount of pigment as a consequence of the presence of P[gR10], which contains the white+ marker, but they can still be distinguished from their twin-spots and the surrounding heterozygous tissue (Figure 2d).

(c)

+/+

–/–

+/–

(d) +/+

–/–

+/–

Current Biology

Dp110– and p60– ommatidia and cells in the adult eye are reduced in size. The FLP-FRT recombination system was used to generate mitotic clones 60 h ± 12 h AEL and adult eye phenotypes were examined. Scanning electron micrographs of (a) Dp110– and (b) p60– ommatidia that were indented (*) and reduced in size (arrows). Tangential sections demonstrate that (c) Dp110– and (d) p60– clones (–/–) contain smaller cells and photoreceptor rhabdomeres than their twin-spots (+/+) and the heterozygous tissue (+/–). The p60– and Dp110– adult eye clones were marked by a reduction in or absence of pigment (see Materials and methods section). The scale bar represents 10 µm. Flies had the following genotypes: (a) y w P[hs-FLP ry+]/+; P[gH ry+] FRT82B Dp110A/FRT82B M(3)95A P[w+], (b,d) y w P[hs-FLP ry+]/+; p60B FRT40A/P[w+ hs-π-myc] FRT40A; P[gR10, w+]/+ and (c) y w P[hsFLP ry+]/+; P[gH ry+] FRT82B Dp110A/FRT82B P[w+].

Dp110– cells, and to a lesser extent p60– cells, are significantly smaller in both cross-section (Figures 2c,d) and longitudinal-section (data not shown) than cells in the surrounding heterozygous and wild-type tissue. Thus, these experiments show that both Dp110 and p60 are autonomously required for eye cells to achieve their normal adult size. In addition, the mutant clones are consistently smaller than their twin-spots, suggesting that cell number as well as cell size might be reduced. Although reduced in size, the mutant photoreceptors can still differentiate and form rhabdomeres (which stain dark blue) by expanding and re-organising their internal membranes. Furthermore, almost every ommatidium contains the wildtype number of photoreceptors, suggesting that Dp110 and p60 are not essential for photoreceptor differentiation. The characteristic trapezoid arrangement of photoreceptor rhabdomeres within each ommatidium and the orientation of the ommatidia relative to one another are generally wild type in p60– clones, but are sometimes disrupted in Dp110– clones. Together, these data clearly demonstrate that in Drosophila the class IA PI 3-kinase, Dp110, and its adaptor, p60, are required in a cell-autonomous manner for the same process: the attainment of normal cell size. The


similarity of the mutant phenotypes for the two genes is consistent with the established role of the adaptor in the activation of class IA PI 3-kinases and represents the first direct comparison of the loss of function phenotypes of a class IA PI 3-kinase and its adaptor. Modulating the activity of Dp110 alters imaginal disc cell size

The data presented in the previous section demonstrate that Dp110 and p60 are necessary for adult eye cells to achieve their normal final size. A possible explanation for this requirement is that Dp110 drives growth solely in the final stages of eye development when the retinal cells increase in size after proliferation has ceased. Alternatively, Dp110 activity might be required for both proliferating and post-mitotic imaginal disc cells to achieve their normal size. To distinguish between these two possibilities, we investigated the effect of modulating Dp110 activity on the size of the proliferating cells of third instar wing imaginal discs. Thus, Dp110 and p60 transgenes under the control of GAL4 upstream activating sequences (UASs) were expressed in clones of cells induced at random locations in wing imaginal discs. This was achieved by removing a CD2 ‘stuffer cassette’ flanked by two FRT sites and inserted between the Actin 5C promotor and the GAL4 coding sequence using heat-shock inducible Flp [41]. Signalling through Dp110 was increased by overexpression of wild-type Dp110 and reduced by overexpression of a catalytically inactive and hence dominant-negative version of Dp110 (Dp110D954A) [21]. Greater inhibition of Dp110 signalling was achieved by overexpression of wild-type p60 or of p60 with part of the Dp110-binding site deleted (∆p60). Like the mammalian class IA PI 3-kinase adaptors, p60 and ∆p60 inhibit class IA PI 3-kinase signalling when overexpressed, presumably by competing with endogenous Dp110 and p60 complexes for binding sites on upstream activators (D.W. and S.J.L., unpublished observations; [42,43]). The effect of modulating Dp110 activity on cell size was quantified by flow cytometric analysis of dissociated imaginal disc cells co-expressing the gene encoding nuclear green fluorescent protein (GFP) together with the Dp110 or p60 transgenes. The forward scatter (FSC) of these cells (a measure of cell size) was compared with the FSC of control non-GFP-expressing cells from the same imaginal discs. The results presented in Figure 3a demonstrate that Dp110 expression dramatically increases imaginal disc cell size. In this experiment, the mean FSC was 150% of that of the control cell population. In contrast, the expression of Dp110D954A slightly reduces cell size (the mean FSC was 94% of control cells) whereas the expression of p60 or ∆p60 results in a more dramatic reduction in cell size (the mean FSC was 82% and 75% of control cells, respectively). Thus, modulating Dp110 activity alters the size of proliferating cells in the third instar disc epithelium. To investigate this effect more closely, we examined the size

1023

of Dp110-expressing and ∆p60-expressing cells at different stages of the cell cycle. This analysis demonstrated that Dp110-expressing cells are larger than control cells during both the G1 phase of the cell cycle and the S + G2 (Figure 3b). Conversely, ∆p60-expressing cells are reduced in size both in G1 and S + G2 (Figure 3c). Thus, Dp110 regulates cell size throughout the imaginal disc cell cycle. Inhibiting the activity of Dp110 reduces cell number

The increased size of Dp110-expressing cells might result either from an inhibition of cell division, or from accelerated biosynthesis. To distinguish between these two possibilities, the effect of Dp110 expression on cell number was examined. Clones of cells expressing GFP alone or GFP and Dp110 together were induced 72 hours after egg laying (AEL), and cell number was assessed 44 hours later (Figure 4a). As has been described [31], the number of cells in simultaneously induced clones can vary considerably. If large numbers of clones are analysed, however, the median value can be taken as a reliable indicator of cell number and can be used to establish cell doubling time (DT; Figure 4a) and the rate of increase in cell number (1/DT; Figure 4b, see Materials and methods section). Our analysis reveals that Dp110 expression does not alter the rate of increase in cell number. Thus, Dp110-expressing cells are larger because of increased growth and not because of the inhibition of cell division. Furthermore, this result demonstrates that, as far as can be detected with this technique, overexpression of Dp110 is not sufficient to increase cell number (Figures 4a,b). In contrast, ∆p60, p60 and to a lesser extent Dp110D954A expression does reduce the rate of increase in cell number (Figures 4a,b). Thus, attenuating Dp110 activity reduces cell size by reducing growth as opposed to increasing cell division. To further examine the effect that reducing Dp110 activity has on cell number, the number of cells in Dp110– clones was compared with the number of cells in their associated twin- spots. In agreement with the ability of ∆p60, p60 and Dp110D954A expression to reduce cell number, clones of Dp110– cells contain fewer cells than their twin-spots (Figure 4c). Taken together, these data demonstrate that the activity of Dp110 is necessary for cells to achieve their normal size and number. During our analysis by flow cytometry we noted that, compared with the control cell populations, a greater proportion of the Dp110-expressing cells were in the G2 phase of the cell cycle whereas a greater proportion of the ∆p60-expressing cells were in the G1 phase (Figures 3b,c). These observations are consistent with an increase in Dp110 activity increasing the rate of progression of cells through the G1/S transition but not through the G2/M transition of the cell cycle. Overexpression of Dp110 alters compartment size

Previous experiments, which utilised GAL4 under the control of the engrailed promotor (En–GAL4) to drive the

1024


Control

Dp110

(b)

Cell number

(a)

Cell number

Figure 3

Cell number

FSC (cell size)

X

Y

2C

4C

X

Cell number

Cell number

Cell number

FSC (cell size) Control Control

Dp110 and p60 modulate imaginal disc cell size throughout the cell cycle. (a) Clones of cells expressing GFP together with the indicated Dp110 or p60 transgenes were induced 72 h AEL. Wing imaginal discs were then dissected and their cells dissociated and analysed by flow cytometry 43 h later. The forward scatter (FSC) profiles of control cells (pink) versus GFP-expressing and transgeneexpressing cells (green) are shown. The effect of (b) Dp110 and (c) ∆p60 expression on cell size throughout the cell cycle was investigated by comparing FSC profiles of G1 cells with an approximate DNA content of 2C (X) or of S + G2 cells with an approximate DNA content of 3–4C (Y). Larvae were the following genotypes: y w hs-FLP/+; Act > CD2 > GAL4 UAS-GFPnls/UASDp110, Dp110D954A, p60 or ∆p60.

Control Dp110

Y

FSC (cell size)

FSC (cell size)

Dp110 Control

D954A

Control

(c)

∆p60

∆p60

Control

FSC (cell size)

X

Y

2C Cell number

Cell number

FSC (cell size)

Control

X

FSC (cell size)

Cell number

p60

Cell number

Cell number

FSC (cell size)

4C Y

FSC (cell size) Current Biology

expression of cell-cycle regulators in the posterior compartment of the wing imaginal disc, have shown that activation or inhibition of cell division in one compartment increases or reduces cell number without affecting the size of that compartment. Rather, that compartment contains more smaller cells or fewer bigger cells and growth is unaffected [31,32]. These results suggest that imaginal disc growth is not regulated through the cell cycle but by an additional mechanism that is dominant over the cell cycle. We have previously shown that overexpression of Dp110 increases adult wing and eye size, whereas overexpression of Dp110D954A reduces adult wing and eye size [21]. Thus, imaginal disc growth might be controlled by a mechanism that involves Dp110 and p60. To test this hypothesis using experiments directly comparable to the cell-cycle experiments described above, we examined the effect of En–GAL4-driven Dp110 or Dp110D954A expression on the growth of the posterior

compartment of the wing imaginal disc. In these experiments, posterior compartment size was visualised 120 hours AEL by co-expressing GFP with the Dp110 transgenes. The approximate area covered by the anterior and posterior compartments of several discs for each genotype was assessed, so that anterior compartment size, posterior compartment size, total disc size and the ratio of anterior compartment size to posterior compartment size could be determined. Although these experiments were performed using embryos collected from a 2 hour egg lay and reared at a constant density, imaginal disc size did vary considerably by the time of dissection. Despite this variation, we found that the size of the posterior compartment was significantly increased by Dp110 expression and reduced by Dp110D954A expression (Figures 5a,b). In contrast, the size of the anterior compartment for each genotype was not altered significantly. In addition, when the ratio of anterior to posterior compartment size was calculated for each disc, we found that Dp110 expression consistently reduces the ratio,


(a) 30

Control n = 153 DT = 9.4

(b)

20 10

15

0.1 0.075 0.05 0.025

50 40 30 20 10 0

Dp110 n = 81 DT = 9.7

Control Dp110 Dp110D954A p60 Dp60

0

Cell divisions per hour

Figure 4

Number of cells in twin-spot

whereas Dp110D954A expression consistently increases the ratio (Figure 5b; right hand graph). Thus, unlike direct manipulation of the cell cycle, manipulation of activity of Dp110 is sufficient to alter relative compartment size. Together, the experiments described demonstrate that Dp110 modulates both cell size and cell number, and that manipulation of Dp110 activity can override the intrinsic control mechanisms that normally regulate imaginal disc growth and maintain consistent compartment and organ size.

10 5

Class IA PI 3-kinases are consistently found in complex with SH2-domain-containing adaptors, and numerous studies have implied that these molecules depend upon each other for their function [7]. Similarly, Dp110 and p60 associate with each other and can be co-purified in approximately stoichiometric amounts [21,22]. This study provides the first genetic evidence, through mutational inactivation, that a class IA PI 3-kinase and its adaptor regulate the same biological process. Although the Dp110– and p60– phenotypes are similar, removing Dp110 has a consistently stronger effect on growth than removing p60 (Figures 1c,2c,d). It is possible that in p60– cells, Dp110 is still able to provide a limited growth-promoting signal, whereas in Dp110– cells, this would clearly not be the case. Consistent with Dp110 providing some function in the absence of p60, overexpression of Dp110 alone is sufficient to promote growth. Given that the small GTPase Ras also interacts with and activates class IA PI 3-kinases [8,9] and modulates imaginal disc cell size (D. Prober and B. Edgar, personal communication), it is possible that Ras activates Dp110 in response to RTK activation independently of p60. An alternative explanation for the relative severity of the Dp110– phenotype is suggested by the inhibition of growth by overexpressed p60. In Dp110– cells, free p60 might inhibit growth through pathways other than those involving Dp110. We consider this explanation unlikely, however, because Dp110– p60– larvae have similar phenotypes to Dp110– larvae, and because the effects of overexpressing p60 can be rescued by coexpression of Dp110 (D.W. and S.J.L., unpublished observations).

0 10

Dp110D954A n = 39 DT = 11.3

5 0 15

p60 n = 24 DT = 20.3

(c)

10 5 0 25 20 15 10 5 0

∆p60 n = 27 DT = 27.8

Number of clones

Number of cells in twin-spot

A class IA PI 3-kinase and its adaptor provide the same biological function

1 11 –5 – 21 15 – 31 25 – 36 35 – 51 40 –5 5

Discussion

1025

0 10 20 30 40 50 Number of cells in Dp110–/– clone

50 40 30 20 10 0

0 1 20 30 40 50 Number of cells in WT clone Current Biology

Inhibition of Dp110 reduces cell number. (a,b) Clones of cells expressing the gene encoding GFP together with the Dp110 or p60 transgenes indicated were induced 72 h AEL. Wing imaginal discs were then dissected 43 h later, fixed and the approximate number of cells per clone was determined. (a) Doubling time (DT) and (b) the rate of increase in cell number (1/DT) were calculated as described in the Materials and methods section. Larvae were of the same genotypes as those in Figure 3. (c) Dp110– clones contain fewer cells than their twin-spots. The FLP-FRT recombination system was used to generate mitotic clones of Dp110– cells 72 h AEL. Wing imaginal discs were dissected and analysed 43 h later. Clones marked by an absence of lacZ were identified and distinguished from their twin-spots (which contained 2 × lacZ), by staining with anti-lacZ, and nuclei were stained with Hoechst 33342. Larvae had the following genotypes: y w hs–FLP/+; P[gH ry+] FRT82B Dp110A/FRT82B P[arm-lacZ w+] or y w hs-FLP/+; FRT82B/FRT82B P[arm-lacZ w+].

Cell size regulation by Dp110 and p60

Previously, we have shown that Dp110 expression increases and Dp110D954A expression reduces adult cell size [21]. In this report, we demonstrate that endogenous Dp110 and p60 are required for adult eye cells to achieve their normal size, and that this requirement is cell autonomous (Figure 2). In addition, we show that Dp110 does not solely affect post-mitotic growth during the later stages of disc development. Modulating Dp110 significantly affects the size of cycling cells from wing imaginal discs, throughout the cell cycle (Figure 3). As well as

increasing cell size, Dp110 expression increases the proportion of cells in the G2 phase of the cell cycle. Together, these data are consistent with Dp110 expression increasing biosynthesis and progression through the G1/S transition, and the imaginal disc cells being unable to respond to this increase in biosynthesis by passing through the G2/M transition. Consistent with this observation, the activation of the gene p110α class IA PI 3-kinase in rat embryo fibroblasts also promotes S phase

1026


Figure 5

0

D

p1

C

95

4A

0

95

0.5

tro l p1 10 10 D

5

on

1

Modulating Dp110 activity alters compartment size. (a) En–GAL4 was used to express UAS–GFP alone (control) or to co-express UAS–GFP with UAS–Dp110 (Dp110) or UAS–Dp110D954A (Dp110D954A) in the posterior compartment of the wing imaginal disc. Discs from female larvae were dissected 120 h AEL and analysed by fluorescence and light microscopy. The scale bar represents 100 µm. (b) The approximate areas covered by the anterior compartment, the posterior compartment and the anterior plus posterior compartments were analysed using Adobe Photoshop, and the ratio of anterior to posterior compartment area for each disc was determined. Control, n = 17; Dp110, n = 20; Dp110D954A; n = 13. Larvae had the following genotypes: En–GAL4 UAS–GFP/+, En–GAL4 UAS–GFP/UAS–Dp110, En–GAL4 UAS–GFP/UAS–Dp110D954A.

D

10

4A

1.5

tro p1 l 10 10 D

15

p1

4A

95

on

D

D

4A

95

tro l p1 1 10 0 D D

p1

D

on

tro l p p1 11 10 0 D

0

0

2

D

5

Anterior / posterior

20

on

5

Anterior + posterior

D

Area (µm2 x 104)

10

C

Area (µm2 x 104)

10

C

Posterior Area (µm2 x 104)

Anterior

(b)

Dp110D954A

Dp110

C

Control

(a)

Current Biology

entry but not progression through the entire cell cycle [44]. Thus far, mammalian class IA PI 3-kinases have not been implicated in the regulation of cell size, possibly because cell size is rarely monitored in tissue culture experiments. Alternatively, cell size and cell division might be more tightly linked in cultured mammalian cells, so that increased biosynthesis results in increased cell division rather than increased cell size. Cell number regulation by Dp110 and p60

The analyses of cell number in Dp110– clones and in clones of cells overexpressing Dp110D954A, p60 or ∆p60 demonstrate that Dp110 and p60 are required for cells to achieve their normal number. The effect on cell number in Dp110– clones is less severe than the effect of p60 or ∆p60 expression. A probable explanation for this difference is that overexpressed p60 or ∆p60 immediately inhibits signalling through Dp110. In Dp110– clones, significant amounts of Dp110 mRNA and protein, and hence signalling through Dp110, might persist for several cell divisions. Although inhibiting Dp110 reduces the rate of increase in cell number, ectopic expression of Dp110 does not detectably enhance the rate of increase in cell number within the time frame of the experiment shown (43 hours, Figure 4). When Dp110 is expressed throughout wing development, an approximately 7% increase in cell number is observed in the adult wings [21]. Considering the natural variation in cell number observed, this increase

would not have been detectable within the time frame of the experiment described here. There are a number of explanations for the failure of the Dp110 expression to significantly increase imaginal disc cell number. Firstly, the G2/M transition might be unable to respond to Dp110induced growth, as discussed above. In addition, several experiments have shown that the relationship between cell division and cell death in the developing wing imaginal disc is complex and involves compensatory feedback mechanisms. The induction of cell death can result in compensatory proliferation, and the induction of overproliferation is often accompanied by increased cell death (see the Introduction and [31,39,45,46]). Clearly, further experimentation will be required to establish the relationships between Dp110 activity, growth, cell proliferation and cell death in the developing imaginal disc. We have observed that inhibiting apoptosis by co-expression of p35 partially inhibits the ability of p60 or ∆p60 to reduce cell number (data not shown). Thus, inhibiting Dp110 might reduce cell number both through effects on cell division and cell survival. Dp110 and p60 regulate compartment and disc size

Previously, we have shown that Dp110 expression increases and Dp110D954A expression reduces adult eye and wing size [21]. In contrast, increasing or reducing cell division by manipulating the activity of direct regulators of the cell cycle does not affect disc or compartment size [31,32]. In a directly comparable experiment, we demon-


strate that modulating Dp110 activity does influence posterior compartment size and the overall size of the disc (Figure 5). Thus, unlike direct manipulation of the cell cycle, manipulation of the activity of Dp110 overcomes the intrinsic mechanism that normally maintains consistent compartment and organ size. This mechanism must have a number of properties. Firstly, it must be able to stimulate organ growth, through changes in cell number, cell size or both. Secondly, it must be able to respond to changes in organ size. Thirdly, it must be able to maintain the pattern of the disc. We suggest that Dp110 and p60 form part of this intrinsic mechanism for size control for the following reasons. Firstly, Dp110 activity can modulate organ growth through changes in cell size and cell number. Secondly, it is conceivable that production of a Dp110-activating ligand is regulated by organ size. Thirdly, Dp110 and p60 might be involved in integrating signals that coordinate growth and patterning. One way in which imaginal disc growth and patterning are thought to be coordinated is by the transforming growth factor β (TGFβ) homologue, Dpp [47], and synergy between TGFβ and class IA PI 3-kinase signalling has been observed [48,49]. Similarly, a family of chitinases called imaginal disc growth factors cooperate with insulin to stimulate imaginal disc cell proliferation [50]. Thus, Dp110 and p60 might be involved in integrating signals to control imaginal disc growth. Conservation of class IA PI 3-kinase signalling and function through evolution

Biochemical studies in mammalian cells and genetic analyses in C. elegans have shown that class IA PI 3-kinases are activated by insulin and insulin-like growth factor (IGF) receptors and their substrates and signal through a serine/threonine kinase cascade that includes PKB and p70S6kinase [4,11,13–17]. Recent data from several laboratories suggest that this signalling pathway is conserved in Drosophila. Treatment of Drosophila S2 cells with insulin or the expression of Dp110 activates Drosophila PKB and p70S6kinase (J. Verdu and M. Birnbaum, personal communication). Furthermore, mutation of the genes encoding Inr, an insulin receptor homologue, Chico, an insulin receptor substrate homologue, Drosophila p70S6kinase, and Drosophila PKB results in small flies containing cells that are reduced in size and/or number (H. Stocker and E. Hafen, personal communication; [51–53]). Thus, it is probable that Inr and Chico activate Dp110, which in turn transduces a growth-promoting signal through p70S6kinase and PKB. Starvation during larval development also inhibits imaginal disc growth and results in small flies with fewer, smaller cells [27,54]. This response enables Drosophila to survive through periods of low nutrition, because these flies are still fertile. Starvation or mutation of genes in the insulin/class IA PI 3-kinase signalling pathway also has an

1027

analogous effect in C. elegans, inducing the formation of dauer larvae. Dauer larvae re-enter the normal life cycle under favourable conditions, which again enables the organism to survive when food is scarce [19]. Furthermore, in mammals, insulin signalling is used to store energy when food is available and to utilise energy during periods of starvation. Together, these observations suggest that insulin/class IA PI 3-kinase signalling might have evolved as a pathway that enables organisms to respond to changes in nutrition and survive starvation. The cellular functions of Dp110 and p60

An important question arising from this work is how does signalling through Dp110 elicit changes at the cellular level that promote growth? Insulin-stimulated mammalian class IA PI 3-kinase activation regulates various processes that influence growth, including glycogen synthesis, glucose uptake and the translation of mRNAs with 5′ oligopyrimidine tracts. Furthermore, numerous reports have shown that class IA PI 3-kinases can regulate cell number through effects on cell proliferation and cell survival [7,10–12]. Various experiments suggest that manipulating cell division, cell survival or protein synthesis in isolation does not modulate disc size (see above and the Introduction). In contrast, when cell division is induced (by coexpression of cell cycle regulators, and apoptosis is simultaneously inhibited (by co-expression of p35), imaginal disc compartment size is increased [31]. This is accompanied by an extended larval period, which would allow increased biosynthesis, including protein synthesis, to occur. Thus, it is possible that class IA PI 3-kinase activity can modulate organ size because it simultaneously influences cell division, cell survival and biosynthesis.

Materials and methods Molecular genetics Dp110 genomic clones isolated from a KrSB10/SM1 genomic library in EMBL4 [55] were sequenced to reveal that Dp110 is immediately distal H at 92E12–13 and represents the transcript previously termed rea (Figure 1a) [56]. To isolate Dp110 deletions, independent excision lines were generated from the HD179 P element (additional lines were kindly provided by D. Maier and A. Preiss) and analysed by Southern blotting and PCR. Dp110A removes 9.3 kb from nucleotide 26 in the 5′ untranslated region (UTR) of H [57] to nucleotide 2296 in the Dp110 cDNA [21]. Conceptual translation reveals that, if expressed, Dp110A would encode amino acids 1–668 of Dp110 plus an additional four amino acids (RCAV, in single-letter amino acid code) before a stop codon is encountered [21]. We were unable to detect expression of this truncated protein (which would not contain the kinase domain), so it was assumed that the protein is either unstable or not expressed (data not shown). Dp110B removes all of H, Dp110 and the distal regions shown in Figure 1a so was used only in heterozygotes with Dp110A. Dp110A/Dp110A or Dp110A/Dp110B in combination with P[gH] were lethal but a genomic construct containing Dp110 (P[gDp110]) rescued that lethality completely (that is, with the expected proportion of rescued progeny and without developmental delay). Thus, P[gH], Dp110A and P[gH], Dp110A/P[gH], Dp110B are effective null mutants of Dp110. The p60 gene is approximately 14 kb proximal of smo, close to the distal breakpoint of Df(2L)al at 21B8–C1 (Figure 1b, [22,58]). Sequence analysis of the region revealed three genes close to p60

1028


[22,23] (Berkeley Drosophila Genome Project, unpublished observations): PLC21C, dU2AF38 [59] and dSTI1, a gene with strong homology to a stress inducible protein found in mouse, mSTI1, yeast, ySTI1, and human, hSTI1 (data not shown; [60]). The P element P[l(2)06751] disrupts dU2AF38, P[l(2)k00616] disrupts dST11, and P[l(2)k14504] disrupts both dU2AF38 and dST11. P[319/11] is inserted in an intron of PLC21C (data not shown; [59,61]). The p60A deletion is 18 kb in length and was generated by mobilisation and duplication of P[319/11], followed by deletion of the intervening region by cis-mediated excision. The p60B deletion is 11–19 kb in length and was created by imprecise excision of P[l(2)06751] [59]. The p60A and p60B breakpoints were mapped by Southern blotting and PCR. Both p60A and p60B complemented Df(2L)al but not l(2)P07651, l(2)k14504 or l(2)k00616. The lethality of l(2)P07651, l(2)k14504 or l(2)k00616 in trans with p60A or p60B was rescued by the genomic construct, P[gR10] (kindly provided by J. Hooper), which contains dU2AF38 and dSTI1, but not p60. The genotypes p60A/p60A, p60B/p60B and p60A/p60B in combination with P[gR10] were lethal, but a genomic construct containing only p60 (P[gp60]) rescued that lethality completely (that is, with the expected proportion of rescued progeny and without developmental delay). Thus, p60A/p60B, P[gR10] and p60B; P[gR10] are effective null mutants of p60. Although we cannot rule out the disruption of PLC21C by p60A or p60B, we think that is unlikely to have a significant effect on the phenotypes reported. The complementation of p60A and p60B with Df(2L)al and the viability of p60A and p60B in combination with P[gR10] plus P[gp60] suggests that PLC21C is not an essential gene or that alternative transcripts of this gene can be produced without the 5′ region.

Plasmid construction P[gDp110] was constructed by subcloning the distal 1.7 kb BamHI/EcoRI and adjacent 13 kb BamHI fragments indicated into pCaSpeR4 (Figure 1a). P[gp60] was constructed by ligating the BamHI/SacI fragment into pCaSpeR4. P[gH] has been previously described and termed HBS [57]. P[gR10] is the 10 kb EcoRI fragment indicated in Figure 1b cloned into pCaSpeR4 and was generated during the positional cloning of the gene smo [58]. A cassette encoding p60 tagged at the amino terminus with the haemagglutinin (HA) epitope was generated in the following way. Firstly, a Vent DNA polymerase PCR product was amplified from pDW1 (pBluescript SK containing a 900 bp p60 cDNA fragment that includes the 5′ UTR and the first 439 bp of p60 [22]) using a T7 primer and a primer containing a NotI site, initiation codon, BamHI site and codons 2–6 of p60: TCCCCGCGGTGGCGGCCGCACCACCATGGGATCCCAACCCTCCCCGCTCC. The product was digested with NotI and KspI and subcloned into the NotI/KspI-digested pDW1 to generate pDW2. Next, pDW2 was linearised with BamHI and annealed oligonucleotides encoding YPYDDVPDYA (an unintentionally modified version of the HA epitope tag) flanked by cohesive BamHI ends were inserted to generate pDW3 (sense oligonucleotide, GATCCTATCCATATGATGACGTACCAGATTACGCTG, antisense oligonucleotide, GATCCAGCGTAATCTGGTACGTCATCATATGGATAG). An XbaI site was then inserted into the XhoI site of the pDW3 polylinker with two annealed oligonucleotides to generate pDW4 (sense oligonucleotide, TCGACACCGCTCTAGAACG; antisense oligonucleotide, TCGACGTTCTAGAGCGGTG). A full-length p60 cDNA was then generated by linearising pDW4 with EcoRI and XbaI and inserting an EcoRI/XbaI fragment of the p60 cDNA containing the remainder of the coding sequence and 102 base pairs of 3′ UTR, thereby creating pBluescript–HAp60, pDW5. A cassette encoding ∆p60 with amino acids 466–558 deleted was generated by removing an internal 276 bp PstI fragment from pDW5 to generate pDW6. Then pUAST–HAp60 and pUAST–HA∆p60 were generated by subcloning blunt-ended NotI/XbaI fragments containing HAp60 and HA∆p60 into a modified version of pUAST which had been cut with XhoI/XbaI and blunt ended.

Microscopy and histology Adult flies were prepared for scanning electron microscopy (SEM) by fixing in 5% glutaraldehyde (4 h), performing a serial dehydration into

100% ethanol, and incubating in 100% amyl acetate (1 h). Critical point drying, gold coating and SEM at 10 kV were performed according to standard procedures using a JEOL 4500LV scanning electron microscope. Adult eyes were fixed, embedded and sectioned as described [62]. Transmitted light images were collected with a Leaf Microlumina colour charge coupled device camera (ISS, Greater Manchester, UK [63]). Dissected discs from wandering third instar larvae were fixed and stained for immunofluorescence as described [21] and examined by conventional fluorescent microscopy using a Photonic Science CV12 camera or by confocal microscopy using a Zeiss laser scanning microscope 510.

Flow cytometry and cell number analysis Flow cytometry was carried out as previously described [31]. Cell doubling times were calculated using the formula DT = logN/log2 (h) in which N, median number of cells per clone and h, age of clone; the rate of increase in cell number is 1/DT.

Acknowledgements We are very grateful to Martin Heisenberg, Dieter Maier, Anette Preiss, Jim Posakony, David Rudner, Donald Rio, Joan Hooper, Bruce Edgar, Ernst Hafen and Konrad Basler for fly stocks and Markus Noll for the genomic library. We thank Krishna Pitrola, Alan Entwistle, Mark Tumaine, Derek Davies and Heather Phillips for technical assistance and Bruce Edgar, Martin Heisenberg, Thomas Raabe, Juan Riesgo-Escovar, Ernst Hafen, Bart Vanhaesebroeck, Lindsay MacDougall and Carmen Coehlo for useful discussions. In addition, the advice of Jennifer Rohn and Helen McNeill in the preparation of this manuscript is greatly appreciated. S.J.L. is a Royal Society University Research Fellow. This work was supported by the Ludwig Institute for Cancer Research and the Royal Society.

References 1. Raff MC: Size control: the regulation of cell numbers in animal development. Cell 1996, 86:173-175. 2. Conlon I, Raff M: Size control in animal development. Cell 1999, 96:235-244. 3. Stern DL, Emlen DJ: The developmental basis for allometry in insects. Development 1999, 126:1091-1101. 4. Vanhaesebroeck B, Leevers SJ, Panayotou G, Waterfield MD: Phosphoinositide 3-kinases: a conserved family of signal transducers. Trends Biochem Sci 1997, 22:267-272. 5. Panayotou G, Gish G, End P, Truong O, Gout I, Dhand R, et al.: Interactions between SH2 domains and tyrosine-phosphorylated platelet-derived growth factor beta-receptor sequences: analysis of kinetic parameters by a novel biosensor-based approach. Mol Cell Biol 1993, 13:3567-3576. 6. Songyang Z, Shoelson SE, Chaudhuri M, Gish G, Pawson T, Haser WG, et al.: SH2 domains recognize specific phosphopeptide sequences. Cell 1993, 72:767-778. 7. Wymann MP, Pirola L: Structure and function of phosphoinositide 3-kinases. Biochim Biophys Acta 1998, 1436:127-150. 8. Rodriguez-Viciana P, Warne PH, Dhand R, Vanhaesebroeck B, Gout I, Fry MJ, et al.: Phosphatidylinositol-3-OH kinase as a direct target of Ras. Nature 1994, 370:527-532. 9. Rodriguez-Viciana P, Warne PH, Vanhaesebroeck B, Waterfield MD, Downward J: Activation of phosphoinositide 3-kinase by interaction with Ras and by point mutation. EMBO J 1996, 15:2442-2451. 10. Rameh LE, Cantley LC: The role of phosphoinositide 3-kinase lipid products in cell function. J Biol Chem 1999, 274:8347-8350. 11. Thomas G, Hall MN: TOR signalling and control of cell growth. Curr Opin Cell Biol 1997, 9:782-787. 12. Shepherd PR, Withers DJ, Siddle K: Phosphoinositide 3-kinase: the key switch mechanism in insulin signalling. Biochem J 1998, 333:471-490. 13. Hemmings BA: PtdIns(3–5)P3 gets its message across. Science 1997, 277:534. 14. Downward J: Lipid-regulated kinases: some common themes at last. Science 1998, 279:673-674. 15. Leevers SJ, Vanhaesebroeck B, Waterfield MD: Signalling through phosphoinositide 3-kinases: the lipids take centre stage. Curr Opin Cell Biol 1999, 11:219-225. 16. Balendran A, Casamayor A, Deak M, Paterson A, Gaffney P, Currie R, Downes CP, Alessi DR: PDK1 acquires PDK2 activity in the presence of a synthetic peptide derived from the carboxyl terminus of PRK2. Curr Biol 1999, 9:393-404.


17. Coffer PJ, Jin J, Woodgett JR: Protein kinase B (c-Akt): a multifunctional mediator of phosphatidylinositol 3-kinase activation. Biochem J 1998, 335:1-13. 18. Morris JZ, Tissenbaum HA, Ruvkun G: A phosphatidylinositol-3-OH kinase family member regulating longevity and diapause in Caenorhabditis elegans. Nature 1996, 382:536-539. 19. Thomas JH, Inoue T: Methuselah meets diabetes. Bioessays 1998, 20:113-115. 20. MacDougall LK, Domin J, Waterfield MD: A family of phosphoinositide 3-kinases in Drosophila identifies a new mediator of signal transduction. Curr Biol 1995, 5:1404-1415. 21. Leevers SJ, Weinkove D, MacDougall LK, Hafen E, Waterfield MD: The Drosophila phosphoinositide 3-kinase Dp110 promotes cell growth. EMBO J 1996, 15:6584-6594. 22. Weinkove D, Leevers SJ, MacDougall LK, Waterfield MD: p60 is an adaptor for the Drosophila phosphoinositide 3-kinase, Dp110. J Biol Chem 1997, 272:14606-14610. 23. Albert S, Twardzik T, Heisenberg M, Schneuwly S: Isolation and characterization of the droPIK57 gene encoding a new regulatory subunit of phosphatidylinositol 3-kinase from Drosophila melanogaster. Gene 1997, 198:181-189. 24. Cohen S: In Imaginal disc development, The development of Drosophila melanogaster. Edited by Bate M, Martinez-Arias A. Vol. II. pp 747-842, 1993. Cold Spring Harbor: Cold Spring Harbor Laboratory Press, USA;1993:747-842. 25. Garcia-Bellido A: Transplantation of developing larval organs into an adult environment. J Insect Physiol 1965, 11:1071-1078. 26. Bryant PJ, Levinson P: Intrinsic growth control in the imaginal primordia of Drosophila, and the autonomous action of a lethal mutation causing overgrowth. Dev Biol 1985, 107:355-363. 27. Bryant PJ, Simpson P: Intrinsic and extrinsic control of growth in developing organs. Q Rev Biol 1984, 59:387-415. 28. Simpson P, Berreur P, Berreur-Bonnenfant J: The initiation of pupariation in Drosophila: dependence on growth of the imaginal discs. J Embryol Exp Morphol 1980, 57:155-165. 29. Haynie JL, Bryant PJ: The effects of X-rays on the proliferation dynamics in the imaginal wing disc of Drosophila melanogaster. Wilhelm Roux Archives 1977, 183:85-100. 30. Milan M, Campuzano S, Garcia-Bellido A: Developmental parameters of cell death in the wing disc of Drosophila. Proc Natl Acad Sci USA 1997, 94:5691-5696. 31. Neufeld TP, de la Cruz AF, Johnston LA, Edgar BA: Coordination of growth and cell division in the Drosophila wing. Cell 1998, 93:1183-1193. 32. Weigmann K, Cohen SM, Lehner CF: Cell cycle progression, growth and patterning in imaginal discs despite inhibition of cell division after inactivation of Drosophila Cdc2 kinase. Development 1997, 124:3555-3563. 33. Lindsley DL, Zimm GG: The Genome of Drosophila melanogaster. 1992. Academic Press. 34. Morata G, Ripoll P: Minutes: mutants of Drosophila autonomously affecting cell division rate. Dev Biol 1975, 42:211-221. 35. Zecca M, Basler K, Struhl G: Direct and long-range action of a wingless morphogen gradient. Cell 1996, 87:833-844. 36. Burke R, Basler K: Dpp receptors are autonomously required for cell proliferation in the entire developing Drosophila wing. Development 1996, 122:2261-2269. 37. Dominguez M, de Celis JF: A dorsal/ventral boundary established by Notch controls growth and polarity in the Drosophila eye. Nature 1998, 396:276-278. 38. Bryant PJ, Watson KL, Justice RW, Woods DF: Tumor suppressor genes encoding proteins required for cell interactions and signal transduction in Drosophila. Dev Suppl 1993, 239-249. 39. Karim FD, Rubin GM: Ectopic expression of activated Ras1 induces hyperplastic growth and increased cell death in Drosophila imaginal tissues. Development 1998, 125:1-9. 40. Xu T, Rubin GM: Analysis of genetic mosaics in developing and adult Drosophila tissues. Development 1993, 117:1223-1237. 41. Pignoni F, Zipursky SL: Induction of Drosophila eye development by decapentaplegic. Development 1997, 124:271-278. 42. Hara K, Yonezawa K, Sakaue H, Ando A, Kotani K, Kitamura T, et al.: 1-Phosphatidylinositol 3-kinase activity is required for insulinstimulated glucose transport but not for RAS activation in CHO cells. Proc Natl Acad Sci USA 1994, 91:7415-7419. 43. Rodriguez-Viciana P, Warne PH, Khwaja A, Marte BM, Pappin D, Das P, et al.: Role of phosphoinositide 3-OH kinase in cell transformation and control of the actin cytoskeleton by Ras. Cell 1997, 89:457-467.

1029

44. Klippel A, Escobedo MA, Wachowicz MS, Apell G, Brown TW, Giedlin MA, et al.: Activation of phosphatidylinositol 3-kinase is sufficient for cell cycle entry and promotes cellular changes characteristic of oncogenic transformation. Mol Cell Biol 1998, 18:5699-5711. 45. Asano M, Nevins JR, Wharton RP: Ectopic E2F expression induces S phase and apoptosis in Drosophila imaginal discs. Genes Dev 1996, 10:1422-1432. 46. Du W, Vidal M, Xie JE, Dyson N: RBF, a novel RB-related gene that regulates E2F activity and interacts with cyclin E in Drosophila. Genes Dev 1996, 10:1206-1218. 47. Serrano N, O’Farrell PH: Limb morphogenesis: Connections between patterning and growth. Curr Biol 1997, 7:R186-R195. 48. Kimura KD, Tissenbaum HA, Liu Y, Ruvkun G: daf-2, an insulin receptor-like gene that regulates longevity and diapause in Caenorhabditis elegans. Science 1997, 277:942-946. 49. Krymskaya VP, Hoffman R, Eszterhas A, Ciocca V, Panettieri RA, Jr.: TGF-beta 1 modulates EGF-stimulated phosphatidylinositol 3kinase activity in human airway smooth muscle cells. Am J Physiol 1997, 273:L1220-1227. 50. Kawamura K, Shibata T, Saget O, Peel D, Bryant PJ: A new family of growth factors produced by the fat body and active on Drosophila imaginal disc cells. Development 1999, 126:211-219. 51. Chen C, Jack J, Garofalo RS: The Drosophila insulin receptor is required for normal growth. Endocrinology 1996, 137:846-856. 52. Böhni R, Riesgo-Escovar J, Oldham S, Brogiolo W, Stocker H, Andruss BF, et al.: Autonomous control of cell and organ size by CHICO, a Drosophila homolog of Vertebrate IRS1-4. Cell 1999, 97:865-875. 53. Montagne J, Stewart MJ, Stocker H, Hafen E, Kozma SC, Thomas G: Drosophila S6 kinase: a regulator of cell size. Science 1999, in press. 54. Held LI: Pattern as a function of cell number and cell size on the second-leg basitarsus of Drosophila. Wilhelm Roux Archives 1979, 187:105-127. 55. Baumgartner S, Bopp D, Burri M, Noll M: Structure of two genes at the gooseberry locus related to the paired gene and their spatial expression during Drosophila embryogenesis. Genes Dev 1987, 1:1247-1267. 56. Maier D, Stumm G, Kuhn K, Preiss A: Hairless, a Drosophila gene involved in neural development, encodes a novel, serine rich protein. Mech Dev 1992, 38:143-156. 57. Bang AG, Posakony JW: The Drosophila gene Hairless encodes a novel basic protein that controls alternative cell fates in adult sensory organ development. Genes Dev 1992, 6:1752-1769. 58. Alcedo J, Ayzenzon M, Von Ohlen T, Noll M, Hooper J: The Drosophila smoothened gene encodes a seven-pass membrane protein, a putative receptor for the hedgehog signal. Cell 1996, 86:221-232. 59. Rudner DZ, Kanaar R, Breger KS, Rio DC: Mutations in the small subunit of the Drosophila U2AF splicing factor cause lethality and developmental defects. Proc Natl Acad Sci USA 1996, 93:10333-10337. 60. Blatch GL, Lassle M, Zetter BR, Kundra V: Isolation of a mouse cDNA encoding mSTI1, a stress-inducible protein containing the TPR motif. Gene 1997, 194:277-282. 61. Torok T, Tick G, Alvarado M, Kiss I: P-lacW insertional mutagenesis on the second chromosome of Drosophila melanogaster: isolation of lethals with different overgrowth phenotypes. Genetics 1993, 135:71-80. 62. Tomlinson A, Ready DF: Neuronal differentiation in the Drosophila ommatidium. Dev Biol 1987, 120:366-376. 63. Entwistle A: A comparison between the use of a high-resolution CCD camera and 35 mm film for obtaining coloured micrographs. J Microscopy 1998, 192:81-89.

Because Current Biology operates a ‘Continuous Publication System’ for Research Papers, this paper has been published on the internet before being printed. The paper can be accessed from http://biomednet.com/cbiology/cub — for further information, see the explanation on the contents page.