fat

DEVELOPMENTAL BIOLOGY

See the embryonic expression pattern of ft at the Berkeley Drosophila Genome Project Patterns of Gene Expression Site.

In discs of mature larvae the expression of the ft gene, detected by in situ hybridization with a ft specific probe, shows a spatially heterogeneous distribution. In wild-type mature wing discs, ft expression is ubiquitous but preferentially accumulates in the axillar region and in the basal posterior region, with graded differences away from these maxima. This pattern of expression is more diffuse in young imaginal discs (72-h AEL). Regions of high expression appear in the notum and in halteres in homologous regions to the wing disc. Leg and eye-antenna discs also show heterogeneous patterns of expression. Interestingly, these patterns are similar to those shown for dachsous (ds) which also encodes for a member of the cadherin superfamily of proteins (Garoia, 2000).

In ft¹⁸ (and ft⁴) overgrown discs the expression of ft mRNA is weak and appears restricted to the distal wing pouch. Since these alleles possibly do not correspond to transcriptional nulls, the reduction of transcription in ft mutants may reflect feed back control of ft gene expression (Garoia, 2000).

The integument of the Drosophila adult abdomen bears oriented hairs and bristles that indicate the planar polarity of the epidermal cells. Four polarity genes, frizzled (fz), prickle (pk), Van gogh/strabismus (Vang/stbm) and starry night/flamingo (stan/fmi) were examined in this study, and what happens when these genes are either removed or overexpressed in clones of cells was examined. The edges of the clones are interfaces between cells that carry different amounts of gene products, interfaces that can cause reversals of planar polarity in the clone and wild-type cells outside them. To explain, a model is presented that builds on an earlier picture of a gradient of X, the vector of which specifies planar polarity and depends on two cadherin proteins, Dachsous and Fat. It is conjectured that the X gradient is read out, cell by cell, as a scalar value of Fz activity, and that Pk acts in this process, possibly to determine the sign of the Fz activity gradient (Lawrence, 2004).

It has been proposed that, in the abdomen of Drosophila, morphogen gradients (Hh in the A compartment and Wg in the P compartment) organise a secondary gradient ('X'); the vector of X specifying the polarity of each cell. Although the composition of X is unknown, at least three proteins, Fj, Ds and Ft, are implicated. All three may be expressed, or be active, in bell-shaped distributions that peak near the A/P (Ds) or P/A (Fj, Ft) boundaries. Ds and Ft are transmembrane proteins in the cadherin superfamily; Fj probably acts in the Golgi. Ds and Ft are integrated into the membrane, suggesting that the X gradient itself may not be diffusible but instead might depend on information transfer from cell to cell (Lawrence, 2004).

How does Hh set up the X gradient? Although changing the real or perceived level of Hh does affect polarity, many clones (for example clones that lack Smo, an essential component of Hh reception) show there is no simple correlation between Hh concentration and polarity. For instance, large smo^- clones in the center of the A compartment are polarised normally, even though they are blind to Hh. Also, while smo^- clones in some regions of the A compartment do affect polarity, both mutant and wild-type cells are repolarised. Both these observations argue for some transfer of information about polarity between cells, a process that would be at least partly Hh independent. This process involves four genes (stan, fz, Vang and pk) that probably act downstream of ds, ft and fj (Lawrence, 2004).

Action of fat, four-jointed, dachsous and dachs in distal-to-proximal wing signaling

In the Drosophila wing, distal cells signal to proximal cells to induce the expression of Wingless, but the basis for this distal-to-proximal signaling is unknown. Three genes that act together during the establishment of tissue polarity, fat, four-jointed and dachsous, also influence the expression of Wingless in the proximal wing. fat is required cell autonomously by proximal wing cells to repress Wingless expression, and misexpression of Wingless contributes to proximal wing overgrowth in fat mutant discs. Four-jointed and Dachsous can influence Wingless expression and Fat localization non-autonomously, consistent with the suggestion that they influence signaling to Fat-expressing cells. dachs is identified as a gene that is genetically required downstream of fat, both for its effects on imaginal disc growth and for the expression of Wingless in the proximal wing. These observations provide important support for the emerging view that Four-jointed, Dachsous and Fat function in an intercellular signaling pathway, identify a normal role for these proteins in signaling interactions that regulate growth and patterning of the proximal wing, and identify Dachs as a candidate downstream effector of a Fat signaling pathway (Cho, 2004).

There is a progressive elaboration of patterning along the PD axis over the course of wing development. During the second larval instar, interactions among the Epidermal Growth Factor Receptor, Dpp and Wg signaling pathways divide the wing disc into a dorsal region, which will give rise to notum, and a ventral region, from which the wing will arise. An initial PD subdivision of the wing is then effected by signaling from the AP and DV compartment boundaries, which promotes the expression of two genes, scalloped and vestigial, that encode subunits of a heterodimeric transcription factor (Sd-Vg) in the center of the wing. This subdivides the wing into distal cells, which give rise to the wing blade, and surrounding cells, which give rise to proximal wing and wing hinge structures. The proximal wing is further subdivided into a series of molecularly distinct domains. Studies of Sd-Vg function in the wing led to the realization that the elaboration of this finer pattern depends in part upon signaling from the distal, Sd-Vg-expressing cells, to more proximal cells. Thus, mutation of vg leads to elimination, not only of the wing blade, where Vg is expressed, but also of more proximal tissue. Conversely, ectopic expression of Vg in the proximal wing reorganizes the patterning of surrounding cells (Cho, 2004 and references therein).

A key target of the distal signal is Wg, which during early third instar is expressed in a ring of cells that surround the SD-VG-expressing cells, and which later becomes expressed in a second, more proximal ring. Wg expression in the inner, distal ring within the proximal wing is regulated by an enhancer called spade-flag (spd-fg), after an allele of wg in which this enhancer is deleted (Neumann, 1996). Studies of this allele, together with ectopic expression experiments, have revealed that Wg is necessary and sufficient to promote growth of the proximal wing. Wg also plays a role in proximal wing patterning; it acts in a positive-feedback loop to maintain expression of Homothorax (Hth). The rotund (rn) gene has been identified as an additional target of distal signaling (Cho, 2004 and references therein).

This work identified Four-jointed (Fj), Dachsous (Ds), Fat and Dachs as proteins that influence signaling to proximal wing cells to regulate Wg and rn expression. Fj is a type II transmembrane protein, which is largely restricted to the Golgi. Null mutations in fj do not cause any obvious defects in the proximal wing. However, fj plays a role in the regulation of tissue polarity, yet acts redundantly with some other factor(s) in this process. Mutations in fat or ds can also influence tissue polarity. Although the molecular relationships among these proteins are not well understood, genetic studies suggest that fj and ds act via effects on fat, and both fj and ds can influence Fat localization in genetic mosaics (Cho, 2004 and references therein).

Interestingly, alleles of fj, ds and fat, as well as alleles of another gene, dachs, can result in similar defects in wing blade and leg growth. The similar requirements for these genes during both appendage growth and tissue polarity, together with the expression patterns of fj and ds in the developing wing, led to this investigation of their requirements for proximal wing development. All four genes influence the expression of Wg in the proximal wing, and genetic experiments suggest a pathway in which Fj and Ds act to modulate the activity of Fat, which then regulates transcription via a pathway that includes Dachs. These observations lend strong support to the hypothesis that Fj, Ds and Fat function as components of an intercellular signal transduction pathway, implicate Dachs as a key downstream component of this pathway, and identify a normal role for these genes in proximodistal patterning during Drosophila wing development (Cho, 2004).

The common feature of all of the manipulations of FJ and DS expression carried out in this study is that Wg expression, and by inference, Fat activity, can be altered when cells with different levels of Fj or Ds are juxtaposed. In the case of Fj, its normal expression pattern, and effects of mutant and ectopic expression clones are all consistent with the interpretation that juxtaposition of cells with different levels of Fj is associated with inhibition of Fat in the cells with less Fj and activation of Fat in the cells with more Fj. The influence of Ds, however, is more variable. Studies of tissue polarity in the eye suggest that Ds inhibits Fat activity in Ds-expressing cells, and/or promotes Fat activity in neighboring cells. The predominant effect of Ds during early wing development is consistent with this, but its effects in late discs are not. Studies of tissue polarity in the abdomen suggest that the Ds gradient might be interpreted differently by anterior versus posterior cells, and it is possible that a similar phenomena causes the effects of Ds to vary during wing development (Cho, 2004).

The influence of ds mutation on gene expression and growth in the wing is much weaker than that of fat. It has been suggested that Fj might influence Fat via effects on Ds, and fj mutant clones have been observed to influence Ds protein staining. The observations are consistent with the inference that both Ds and Fj can regulate Fat activity, but they do not directly address the question of whether Fj acts through Ds. They do, however, indicate that even the combined effects of Fj and Ds cannot account for FAT regulation, and, assuming that the strongest available alleles are null, other regulators of Fat activity must exist. It is presumably because of the counteracting influence of these other regulators that alterations in Fj and Ds expression have relatively weak effects. In addition, according to the hypothesis that Fat activity is influenced by relative rather than absolute levels of its regulators, the effects of Fj or Ds could be expected to vary depending upon their temporal and spatial profiles of expression, as well as on the precise shape and location of clones (Cho, 2004).

The observations imply the existence of at least two intracellular branches of the Fat signaling pathway. One branch involves the transcriptional repressor Grunge, influences tissue polarity, certain aspects of cell affinity, and fj expression, but does not influence growth or wg expression. An alternative branch does not require Grunge, but does require Dachs. Dachs is implicated as a downstream component of the Fat pathway, based on its cell autonomous influence on Fat-dependent processes, and by genetic epistasis. The determination that it encodes an unconventional myosin, and hence presumably a cytoplasmic protein, is consistent with this possibility. It also suggests that Dachs does not itself function as a transcription factor, and hence implies the existence of other components of this branch of the Fat pathway. This Grunge-independent branch influences Wg expression in the proximal wing and imaginal disc growth. However, further studies will be required to determine whether Dachs functions solely in Grunge-independent Fat signaling, or whether instead Dachs is required for all Fat signaling (Cho, 2004).

The observations that fj expression is regulated by Sd-Vg, and that fj is both necessary and sufficient to modulate the distal ring of Wg expression in the proximal wing, suggest that Fj influences the activity of a distal signal, which then acts to influence Fat activity. However, the relatively weak effects of fj indicate that other factors must also contribute to distal signaling, just as fj functions redundantly with other factors to influence tissue polarity. Since Ds expression is downregulated in a domain that is broader than the Vg expression domain, a direct influence of Vg on the Ds gradient is unlikely, and the essentially normal appearance of Wg expression in the proximal wing in fj ds double mutants implies that Ds is not a good candidate for the hypothetic factor Signal X. Rather, it is suggested that Ds acts in parallel to signaling from Vg-expressing cells to modulate Fat activity. This Vg-independent effect would account for the remnant of the distal ring that sometimes appears in vg null mutants. Importantly though, the observation that the phenotypes of hypomorphic dachs mutant clones on Wg expression are more severe than fj and ds suggests that the hypothesized additional factors also act via the Fat pathway. It is also noted that the limitation of Wg expression to the proximal wing even in fat mutant clones implies that Wg expression both requires Nubbin, and is actively repressed by distally-expressed genes (Cho, 2004).

The recovery of normal Wg expression by later stages in both fj and dachs mutant clones implies that the maintenance of Wg occurs by a distinct mechanism. Prior studies have identified a positive-feedback loop between Wg and Hth that is required to maintain their expression. It is suggested that once this feedback loop is initiated, Fat signaling is no longer required for Wg expression. Moreover, the recovery of normal levels of Wg at late stages suggests that this positive-feedback loop can amplify reduced levels of wg to near normal levels (Cho, 2004).

The distinct consequences of Vg expression and Fj expression in clones in the proximal wing suggest that another signal or signals, which are qualitatively distinct from the Fj-dependent signal, is also released from VG-expressing cells. When Vg is ectopically expressed, Wg is often induced in a ring of expression that completely encircles it. However, this is not the case for Fj-expressing clones. Both Vg- and Fj-expressing clones can activate rn and wg only within NUB-expressing cells, but Vg expression can result in non-autonomous expansion of the Nub domain, and this expansion presumably facilitates the expression of Wg by surrounding cells. Another striking difference between Vg- and Fj-expressing clones is that in the case of ectopic Fj, enhanced Wg expression is only in adjacent cells. By contrast, in the case of Vg, Wg expression initiates in neighboring cells, but often moves several cells away as the disc grows, resulting in a gap between Vg and Wg expression. This gap suggests that a repressor of Wg expression becomes expressed there, and recent studies have identified Defective proventriculus (Dve) as such a repressor (Cho, 2004).

In strong fat mutants, the wing discs become enlarged and have extra folds and outgrowths in the proximal wing. The disproportionate overgrowth of the proximal wing is due to upregulation of Wg in this region, as demonstrated by its suppression by wg^spd-fg. At the same time, clones of cells mutant for fat overgrow in other imaginal cells, and fat wg^spd-fg discs are still enlarged compared with wild-type discs. Thus, Fat appears to act both by regulating the expression of other signaling pathways (e.g. Wg), and via its own, novel growth pathway. The identification of additional components of this pathway will offer new approaches for investigating its profound influence on disc growth (Cho, 2004).

Effects of Mutation or Deletion

Lethal mutations at the fat locus in Drosophila cause imaginal discs to continue to grow by cell proliferation far beyond their normal final size. During a greatly extended larval period, the overgrowing imaginal discs develop additional folds and lobes, but retain a single-layered epithelial structure. In the wing disc, the additional lobes originate in the proximal fold area, and in the extra tissue the cells are less columnar than normal. Mutant disc cells lack zonulae adherents as well as associated microtubules and microfilaments, and they show an abnormal distribution and reduced density of gap junctions. The effect on growth is disc-autonomous as shown by transplantation experiments. The overgrown imaginal discs retain the ability to differentiate adult cuticular structures, as shown by metamorphosis of discs after transplantation into wild-type larval hosts and by the ability of some mutant animals to develop to the pharate adult stage. The structures differentiated by mutant discs show many abnormalities including ingrowths, outgrowths, separated cuticular vesicles, and areas of reversed bristle polarity; some of these abnormalities suggest that the mutations interfere with cell adhesion as well as the control of cell proliferation. The fat locus is located in cytogenetic interval 24D5.6-7, and 18 alleles are known, including spontaneous, chemically induced, X-ray-induced, and dysgenic mutations (Bryant, 1988).

Recessive lethal mutations in the fat locus of Drosophila cause hyperplastic, tumor-like overgrowth of larval imaginal discs, defects in differentiation and morphogenesis, and death during the pupal stage. Clones of mutant cells induced by mitotic recombination demonstrate that the overgrowth phenotype is cell autonomous. Two recessive lethal alleles contain alterations in the fat coding sequence, and the dominant fat allele, Gull, contains an insertion of a transposable element in the 33rd cadherin domain. Thus, this novel member of the cadherin gene superfamily functions as a tumor suppressor gene and is required for correct morphogenesis (Mahoney, 1991).

The fat gene negatively controls cell proliferation in a cell autonomous manner. There are many ft alleles of different origin, most of which are pupal lethal, three are viable and one, ft^G (fat Gull) is a dominant antimorph. Bryant (1988) studied in some detail the phenotype of ft⁸ (previously known as fat floppy-disc, ft^fd) a pupal lethal with delayed entry in pupariation (8-9 days) and with abnormal-hyperplastic discs. Mutant ft⁸ pupae that survive to pharate adults show shorter and thicker leg joints, higher chaetae densities and deviant trichome and chaetae polarities. Wings are, however, of normal appearance. Delayed larvae show increasingly larger discs, with enlarged folds preferentially in the basal regions of the wing blade. However, transplants of these imaginal discs to metamorphosing wild-type hosts, did not show great departures from wild-type controls (Garoia, 2000).

Three different alleles: ft^G-rv, ft⁴ and a new one, isolated in a mutagenesis screen using P-element transposition and called l(2)79/18 have been examined in this study. l(2)79/18 was not associated with a P insertion and fails to complement other ft alleles; l(2)79/18 has been called ft¹⁸. ft^G-rv is a recessive lethal allele (Bryant, 1988) corresponding to the amorphic condition (Mahoney, 1991). Hemizygous ft^G-rv [ft^G-rv/Df(2L)sc19-1] larvae reach the pupal stage at 7-9 days and show hyperplastic discs (see Bryant, 1988). Homozygous ft⁴ larvae show a delay of 5-6 days until they reach pupariation and die as early pupae, prior to reaching the pharate stage. Their discs, in particular the wing disc, are many times larger than those of wild-type mature larvae, becoming larger the longer the larvae remain in culture. Homozygous ft¹⁸ larvae, in contrast, show less delay (up to 1-2 days) to pupariation, and 60% of the pupae reach the pharate stage. The pharate adults show abnormalities in cuticular patterns and structures in all appendages, similar to those of ft⁸ (Bryant, 1988). The adult wing of pharates is larger in size compared with the wild-type (1.5 times), with higher cell number (2.9 times) and higher trichome density (1.9 times). The shape of the wing is broad and short, enlarged in the A/P axis compared with controls. The basal wing region is more affected than distal regions, with vein abnormalities and abnormal trichome polarity. The chaetae pattern of the wing margin and the positioning of the vein sensillae are fairly normal, although chaetae are blunter and thicker (Garoia, 2000).

The two alleles ft⁴ and ft¹⁸ show different delays to pupariation. A comparison of sizes of the ft⁴ and ft¹⁸ wing discs from larvae of the same age reveals that ft¹⁸ grows faster than ft⁴. The sizes of ft⁴ and ft¹⁸ discs vary differently with the age of the larvae, suggesting a correlation between maximal disc wing size and delay to pupariation. Whether maximal wing disc size is allele specific was analyzed in flies carrying the ft⁴ and the ft¹⁸ mutations and a temperature-sensitive ecd¹ mutation. Homozygous ecd¹ larvae, when shifted from 20° to 29°C in the middle of third larval instar, fail to pupariate and survive as long as 3 weeks as larvae. Disc growth continues but levels off at maximal sizes after 20 days in ft⁴ ecd¹ and 15 days in ft¹⁸ ecd¹, whereas control ecd¹ wing discs stop growing at normal final size. These results indicate that extra growth is not indefinite and that different alleles reach different maximal sizes, being larger in ft⁴ than in ft¹⁸ discs. Similar results showing a maximal growth limit were reported by Bryant (1988) comparing 9-day-old discs of ft⁸ larvae with ft⁸ discs cultured in adult hosts for 21 days (Garoia, 2000).

The heteroallelic combinations ft⁴/ft¹⁸ and ft^G-rv/ft¹⁸ show a similar delay in pupariation (up to 1-2 days) and have a maximal disc size similar to that of ft¹⁸ homozygous. Both combinations reach the pharate stage with extreme ft phenotypes. Wing discs of ft^G-rv/ft⁴ combination are larger, larvae pupate at 7-9 days and die as pupae. In heteroallelic combinations with the deficiency for the locus, ft¹⁸ and ft^G-rv pupate at the same time and reach similar disc sizes, but ft⁴ pupates later and the discs reach larger sizes. Thus, delayed pupariation as in ft⁴ homozygous larvae or in discs maintained in non pupating hosts (ecd^-) allows for disc extra growth but only up to a maximal size. Combinations of the same ft alleles (and ft^G-rv; Bryant, 1988) with the weak viable allele ft⁺ are viable and of ft⁺ phenotype (Garoia, 2000).

The ft pharate adults (and mutant mosaic patches) show smaller cell surfaces or more density of trichomes. This effect on cell differentiation could correspond to smaller cell sizes during cell proliferation. Cell sizes of mature four days wing discs ft⁴ and ft¹⁸ were compared with those of wild-type. Cell densities in ft⁴ and ft¹⁸ discs are similar, and higher (1.2 times) than in wild-type discs. Faster growth rate and smaller cell size could be related to cluster size of cells in different cell-cycle stages. The wildtype stg expression that marks the G2/M transition, occurs in synchronic clusters of 5.1 +/- 2 cells throughout larval development. The sizes of stg clusters in ft¹⁸ and ft⁴ discs yield values in ft⁴ and in ft¹⁸ that are not different from those of wild-type discs (Garoia, 2000).

The study of cell behavior of lethal allelic conditions can be carried out in genetic mosaics. This allows for a study of phenotypes of mutant territories but also of their relationships with neighboring wild-type territories. Morphogenetic mosaics have been studied in large M⁺/ M⁺ (ft homozygous) clones, initiated in a M;ft heterozygous background as well as in clones of ft/ft and ft⁺/ft⁺ twin cells. The analysis of M⁺/M⁺ clones initiated at 60 +/- 12-h AEL associated with homozygosis of ft^G-rv, ft¹⁸ and ft⁴ reveals several features related with ft cell proliferation and differentiation. A fraction of M⁺ clones initiated at this age cross the D/V boundary and most clones tend to fill one or several intervein sectors. In wild-type wings the corresponding figures are 13.8% and clone sizes of 1.1 sectors on average. The corresponding figures for ft^G-rv are 25% and 1.3; 35.2% and 1.2 for ft¹⁸, and 19.2% and 0.8 for ft⁴, reflecting early growth advantages of mutant clones. In contrast with wild-type controls, ft clones tend to fill proximal regions of the wing blade without reaching distal regions of the wing. The distal borders of ft clones are rounded, i.e. cells grow there perpendicular to the A/P boundary. In addition, they are smooth in contrast with the indented borders found in M⁺ control clones. In all of these features, ft^G-rv and ft¹⁸ are more extreme than ft⁴ (Garoia, 2000).

ft clones cause outgrowths in proximal but not in distal wing regions. There are blisters of variable sizes that evaginate from the wing surface, and have irregular trichome polarities, usually perpendicular to the clone border. Trichome density (cell size) in clones (with or without outgrowths) is higher than in wild-type controls, 1.7 times in ft^G-rv, 2.2 times in ft¹⁸ and 1.4 in ft⁴ clones, indicating that mutant cell size is smaller than in wild-type. Mutant clones appearing between veins enlarge the corresponding sectors, with more cells than in control mosaics (ft^G-rv 1.4 times, ft¹⁸ 1.4 and ft⁴ 1.3). Mutant territories that overlap veins differentiate thicker than normal vein ribbons. In the wing margin ft M⁺ clones cause also higher density of chaetae. The ft chaetae (M⁺/M⁺) are smaller than the M/M⁺ heterozygous chaetae, despite not being M^- -- that is, they are not larger but rather are shorter and blunter. In these features ft^G-rv and ft¹⁸ are more extreme than ft⁴. In the notum, ft M⁺ clones are large, with more cells than control M⁺ clones, leading to an increase in the total notum surface. They contain many more chaetae (1.6 2 times in ft^G-rv, 1.7 in both ft¹⁸ and ft⁴), and higher cell (trichome) density (1.4 times in ft^G-rv, 1.4 in ft¹⁸ and 1.2 in ft⁴). Similar pattern deviations occur in the legs and head capsule (see Bryant, 1988). Tergite clones, however, show normal patterns of pigment, chaetae and trichomes (Garoia, 2000). Based on these features of clone size, trichome and chaetae density and chaetae size, the ft^G-rv and ft¹⁸ alleles appear again more extreme than the ft⁴ allele (Garoia, 2000).

The autonomous extra cell proliferation found in M⁺ mosaics can be quantitatively ascertained in a comparison of ft and ft⁺ homozygous cells in a twin test. In this test ft cells, either ft^G-rv, ft¹⁸ or ft⁴, are labelled with forked and the twin controls with crinkled (ck). Clones were initiated at 60 +/- 12 h AEL. The size (in cell number) of ft mutant clones was compared that of their ck twins. ft^G-rv and ft¹⁸ clones are much larger (average ratio of 6.5 and 14.9, respectively) than twin controls. The same holds for ft⁴ clones, although the ratio is only 3.0. Trichome density in these small clones is also like that in large M⁺ clones (1.7 in ft^G-rv, 1.7 in ft¹⁸ and 1.4 in ft⁴). This phenotype is cell autonomous because it appears even in small (late initiated) clones. As in Minute mosaics, clones that include two adjacent veins give rise to larger outgrowths, than those that include only one vein. These clones have increased cell density and deviant cell polarity, independently of clone size. ft⁴ clones are smaller than ft¹⁸ and ft^G-rv, and therefore fewer clones of this genotype overlap two adjacent veins (in these few cases they correspond to the largest clones). The cell viability of ft clones is like that of wildtype cells because the frequency (33% for ft^G-rv, 31% for ft¹⁸, 21% for ft⁴) off clones without ck twins corresponds to that of controls (27%) and to the distance of the Dp f⁺ to ck on the mitotic map (Garoia, 2000).

As in large M⁺ clones, borders of small ft clones in twins are smooth and in some cases they also show outgrowths evaginating from the wing surface. It is interesting to notice here that large and small ft clones do not merely abut adjacent veins, as control clones do, but very frequently cross over the veins and stop a few cells beyond the territory differentiating vein histotype running parallel to them. The frequency of vein overlapping clones is 64.7% for ft^G-rv, 62.5% for ft¹⁸ and 67.5% for ft⁴ clones compared respectively to 25, 23 and 27.5% for twin ck clones. The observation that M⁺ ft cells grow preferentially in proximal regions, avoiding the distal wing, is also reflected in a twin test. The topological location of the mother cell of the twin in the anlage can be ascertained in the adult mosaics as the point where the two members of the twin come into contact. The mutant cells preferentially grow towards the base of the wing, while their twins grow towards more distal regions (that holds for ft¹⁸ and ft^G-rv and less so for ft⁴). This at least accounts for 80% in ft¹⁸, 71% in ft^G-rv and 61% in ft⁴; exceptions to the rule appear in the proximal and most anterior wing regions. This effect is evident in the frequent separation of the two members of the twin (18% in ft¹⁸). Abnormal cell differentiation in vein thickening, chaetae density and chaetae morphogenesis in these small clones is, as expected, like that in large M⁺ clones (Garoia, 2000).

The abnormal behavior of ft mutant cells in cell proliferation, as reflected in abnormal disc sizes and shapes prompted an analysis of possible genetic interactions with mutations in other genes that affect these parameters. Such interactions can be studied in doubly mutant combinations, either in the morphology of imaginal discs of doubly mutant lethal larvae, or in genetic mosaics. Since ft mutant discs become increasingly more abnormal with the age of the larva, comparisons of phenotypes of the mutant combinations are difficult. The selector gene of the dorsal compartment, ap, causes in the mutant condition (ap^rK568) a major reduction of the wing pouch. The double homozygote ft¹⁸ aP^rK568 is lethal (L1-2 phenoeffective phase), suggesting early synergetic effects, possibly in organs other than imaginal discs. The nub² allele of the nubbin gene causes smaller wing disc and a strong reduction of growth in the proximodistal axis of the wing. Surprisingly the doubly mutant combination ft¹⁸;nub² causes a strong delay in pupation (from 2 to 5 days), allowing the wing discs to be considerably enlarged. These effects are apparently synergetic because 5 days old discs are larger than ft¹⁸ controls. The extreme folding of 7-10 day old discs indicates overgrowth of the wing pouch despite being mutant for nub²; this could simply reflect epistasis of ft¹⁸ over nub (Garoia, 2000).

The Drosophila fat gene is exclusively expressed in ectodermal derivatives, both in the embryo and imaginal tissues (Mahoney, 1991) and this paper shows that ft is transcribed in all imaginal discs. The Fat protein contains four major regions. Beginning by the N-terminus there are 34 cadherin-like domains, five EGF-like repeats interspersed with two laminin A-G chain motifs, a transmembrane domain and a novel cytoplasmic domain (Mahoney, 1991; Ponassi, 1999). All these domains appear in the FAT cadherin homolog of vertebrates, except for its cytoplasmic domain (Ponassi, 1999). Classic vertebrate cadherins contain only the cadherin domains. A Drosophila homolog to these cadherins is the dachsous gene. It is not known how the Drosophila ft gene functions in cell adhesion, but the presence of EGF-like repeats in other C-terminal regions also suggests a role in cell-cell signaling, as a receptor or as a ligand. The molecular description of some mutations of ft ( ft¹⁰, ft¹⁵, ft^G and ft^G-rv) shows that they encode truncated proteins containing different fractions of the cadherin region but lacking the EGF domains and more proximal domains. A ft^G revertant ( ft^G-rv), resulting from a complex rearrangement within the ft gene, may correspond to a ft null allele (Mahoney, 1991). All the lethal alleles are phenotypically similar. They could correspond to functional nulls, if the cadherins cannot be anchored to the cell membrane and lack the putative signaling regions, but the remaining secreted cadherin domains could act as dominant negative competitors of normal cadherins. This is the explanation given for ft^G (Mahoney, 1991). Developmental genetic studies have been carried out in four alleles ( ft⁸, Bryant, 1988; ft^G-rv, ft⁴ and ft¹⁸; Garioa, 2000). Of these alleles only the molecular nature of the ft^G-rv allele is known (Mahoney, 1991). ft⁴ and ft¹⁸ are transcribed, but the signal is weak and appears only in the tip of the wing blade. This low expression could mean that ft transcription is dependent on normal ft activity (Garoia, 2000).

The phenotypic consequences of ft alleles are multiple: (1) most alleles are pupal lethals; (2) imaginal discs reach large sizes, with abnormally convoluted folds; (3) pupariation is delayed; (4) proliferating embryonic and adult cells are smaller than wild-type cells; and (5) differentiation of trichomes and chaetae is affected, not only in terms of their size and polarity, but also in their numbers and patterning (Bryant, 1988 and Garoia, 2000). Interestingly, hyperplasia is associated with an abnormal pattern of gene expression, as visualized in two-dimensional protein gels of ft⁴ discs. They show 6-8 fewer spots than wild-type; 11 abundant polypeptides are either new or absent and another 11 show significant quantitative differences with wild-type (Fernandez-Santaren, 1998). The three alleles studied in this work differ in several respects: ft^G-rv hemizygotes ft¹⁸ homozygotes and heteroallelic combinations show maximal delays to pupariation of 1-2 days, whereas ft⁴ larvae may grow for a further 4 days before pupariation. Experiments in ecd¹ permissive larval environment, in which pupariation is delayed, however, indicate that ft mutant discs may continue growing but only up to a maximal size. The ft¹⁸ and ft^G-rv alleles show higher growth rates and more extreme phenotypes than homozygous ft⁴ in mosaics. Thus the ft⁴ allele may be more hypomorphic than the others, but by itself causes delays in pupariation which allow discs to grow larger than in the other alleles that pupate earlier (Garoia, 2000).

How could these mutational alterations of ft account for the complex syndrome of mutant phenotypes at the cellular level? Clonal mutant phenotypes allow for study of developmental parameters of the extreme alleles. Mosaic patches (M⁺ and twins) of the three alleles show overgrown or enlarged territories with more, and smaller, cells than controls. The autonomous extra growth in mosaics occurs everywhere in all the wing disc derivatives examined, within an otherwise normal pattern. In proximal wing regions these extra growths may lead to large bulges. These clones, however, respect the compartmental and vein restrictions. Mosaic extra growth is locally restricted to the clones, the surrounding wild-type territories retaining their wild-type aspect and sizes without signs of positional accommodation. These findings indicate that the mutant territories are simply modifying some proliferation parameters of an otherwise normal, allometrically growing wing disc (Garoia, 2000).

The abnormal cell behavior in clones suggests failures in cell adhesiveness and cell recognition. Thus, ft mutants clones contain trichomes with abnormal polarities, mainly pointing to clone borders and have smooth borders indicating incompatibilities with wild-type cells. In distal wing regions, clone borders are perpendicular to the proximo-distal axis and not parallel, as in controls, indicating preferential allocation of post-mitotic cells to proximal wing positions. This is more manifest in twin clones where the mutant partner grows proximally. In contrast, mutant territories show abnormal cell differentiation patterns, including thicker or wider veins and patterns with more chaetae at higher density, suggesting a role of ft in patterned cell differentiation. The same holds for the territorial expression of wg and dpp expanded to wider territories in mutant discs as compared with wild-type. Thus, whereas the mutant features of trichome polarity and smooth clonal borders can be directly explained by adhesiveness failures, the wider differentiation patterns of chaetae and veins could reveal failures in cell communication in the subdivision of pre-pattern territories. The latter effects are possibly not mediated by failures in lateral inhibition because ft and N or Ax double combinations do not increase the phenotypes beyond mere additivity. ft mutant alleles also affect the differentiation of chaetae; they are shorter and have blunter shafts, suggesting an additional role of ft also in final cell differentiation, possibly related to cytoskeleton anomalies observed in EM pictures of imaginal disc cells (Bryant, 1988; Garoia, 2000).

Overgrowth and abnormal cell behavior of ft cells during proliferation seem to be related to failures in cell signaling and/or cell adhesion. The normal expression patterns of genes with regional specificity, such as ap, en, wg and dpp shown in large ft discs indicate that the overgrowth of the discs is allometric, i.e., not associated with duplications or abnormal local regeneration. This is also the case for other hyperplastic discs, like lethal giant discs (lgd) whose discs show ectopic expression of dpp and wg. Overgrowth of mutant discs in delayed pupariating hosts (ecd^- mutant background) reaching a maximal size, indicates again that overgrowth is allometric, i.e. normal cell proliferation parameters are maintained. Thus, enlarged territories in clones included between normal restriction boundaries (compartments and veins), could be due to failures in cell-cell signaling needed to intercalate positional values defined by borders of clonal restrictions (Garoia, 2000).

Genes that are required for cell proliferation control in Drosophila imaginal discs were tested for function in the female germ-line and follicle cells. Chimeras and mosaics were produced in which developing oocytes and nurse cells were mutant at one of five imaginal disc overgrowth loci (fat, lgd, lgl, c43 and dco) while the enveloping follicle cells were normal. The chimeras were produced by transplantation of pole cells and the mosaics were produced by X-ray-induced mitotic recombination using the dominant female-sterile technique. The results show that each of the genes tested plays an essential role in the development or function of the female germ line. The fat, lgl and c43 homozygous germ-line clones fail to produce eggs, indicating a germ-line requirement for the corresponding genes. Perdurance of the fat plus gene product in mitotic recombination clones allows the formation of a few infertile eggs from fat homozygous germ-line cells. The lgd homozygous germ-line clones give rise to a few eggs with abnormal chorionic appendages, a defect thought to result from defective cell communication between the mutant germ-line and the nonmutant follicle cells. One allele of dco prevents egg development when homozygous in the germ line, whereas another allele has no effect on germ-line development. Fs(2)Ugra, a recently described follicle cell-dependent dominant female-sterile mutation, allows the analysis of egg primordia in which fat, lgd or lgl homozygous mutant follicle cells surround normal oocytes. The results show that the fat and lgd genes are not required for follicle cell functions, while absence of lgl function in follicles prevents egg development (Szabad, 1991).

Recessive lethal mutations of the lethal(2)giant discs (l(2)gd) and lethal(2)fat (l(2)ft) loci of Drosophila melanogaster cause imaginal disc hyperplasia during a prolonged larval stage. Imaginal discs from l(2)ft l(2)gd or Gl(2)gd double homozygotes show more extensive overgrowth than in either single homozygote, and double homozygous l(2)ft l(2)gd mitotic clones in adult flies show much more overgrowth than is seen in clones homozygous for either l(2)gd or l(2)ft alone. dachsous (ds) also acts as an enhancer of l(2)gd, producing dramatically overgrown discs and causing failure to pupariate in double homozygotes. The comb gap (cg) mutation, which also interacts with ds, greatly enhances the tendency of imaginal discs from l(2)gd larvae to duplicate as they overgrow. If l(2)gd homozygotes are made heterozygous for l(2)ft, then several discs duplicate, indicating that l(2)ft acts an a dominant enhancer of l(2)gd. l(2)ft also acts as a dominant enhancer of l(2)gd, and conversely l(2)gd acts as a dominant modifier of l(2)ft. The enhancement of overgrowth caused by various mutant combinations is accompanied by changes in expression of Decapentaplegic and Wingless. These results show that tumor suppressor genes act in combination to control cell proliferation, and that tissue hyperplasia can be associated with ectopic expression of genes involved in pattern formation (Buratovich, 1997).

High-resolution two-dimensional (2D) gel electrophoresis coupled with computer analysis has been used to construct a quantitative protein database of Drosophila mature wing imaginal discs. The level of expression for all of the detected proteins has been quantitatively determined. This database has been used to evaluate changes in the patterns of protein synthesis in wing imaginal discs from two Drosophila melanogaster mutants with abnormal wing disc development: fat and two different alleles of lethal (2) giant disc. Patterns of pulse-labeled proteins of the different mutants show variations in both qualitative and quantitative parameters of synthesis. In this comparison, specific sets of protein changes characteristic of both alleles of the same locus and a set of protein changes common to both loci have been detected. How the abnormal expression of these proteins relates to the abnormal process of mutant hyperplasia is discussed (Santaren, 1998).

Mutations in 17 genes (26 deficiencies) were characterized that interact with Arm^over and/or Arm^under. Interaction strength varies from deficiency to point mutation, suggesting that several genes in the original deficiencies could have contributed to, or modified, the interaction. Only for 7 of the 17 genes have interactions been identical between the point mutation and the corresponding starting deficiency. The 17 genes were sorted into four groups. Group 2 consists of genes required for cell adhesion: This group includes shotgun (which encodes DE-cadherin), as expected. Also uncovered were fat (ft) and dachsous (ds). These two genes encode nonclassical cadherin characterized by a huge extracellular domain containing up to 35 cadherin repeats and a bipartite Arm binding site. Interactions with these two mutants are similar to those observed with shotgun (DE-cadherin), the only difference being that ft interacts more weakly than shg with Arm^over. In addition to genes encoding cadherins (classical and nonclassical), interactions have been observed with some of the genes known to be essential (directly or indirectly) for the assembly or maintenance of adherens junctions -- stardust (sdt), discs-large (dlg), and crumbs (crb). These interact in the same direction as shg; however, the suppression of Arm^under is always weaker and only dlg^M52 enhances Arm^over to the same extent as zw3^M11 (Greaves, 1999).

Developmental compartments and planar polarity in Drosophila

Planar polarity refers to the asymmetry of a cell within the plane of the epithelium; for example, cells may form hairs that point in a posterior direction, or cilia may beat uniformly. This property implies that cells have information about their orientation; it is of interest to understand the nature of this information. Relevant also is the body plan of insects, which, in the ectoderm and somatic mesoderm, consists of a chain of alternating anterior and posterior compartments -- basic units of development with independent cell lineage and subject to independent genetic control. Using the abdomen of adult Drosophila, genes required for normal polarity were either removed or constitutively expressed in small clones of cells and the effects on polarity were observed. Hitherto, all such studies of polarity genes have not found any difference in behavior between the different compartments. This study shows that the three genes, four-jointed, dachsous, and fat, cause opposite effects in anterior and posterior compartments. For example, in anterior compartments, clones ectopically expressing four-jointed reverse the polarity of cells in front of the clone, while, in posterior compartments, they reverse behind the clone. These three genes have been reported by others to be functionally linked. This discovery impacts on models of how cells read polarity. At the heart of one class of models is the hypothesis that cell polarity is determined by the vector of a morphogen gradient. Evidence is presented that cell polarity in the abdomen depends on at least two protein gradients (Fj and Ds), each of which is reflected at compartment borders. Consequently, these gradients have opposing slopes in the two compartments. Because all polarized structures made by abdominal cells point posteriorly, it is surmised that cells in each compartment are programmed to interpret these protein gradients with opposite signs, pointing up the gradient in one compartment and down the gradient in the other (Casal, 2002).

Two other genes resemble fj with regard to compartment-specific effects: dachsous (ds) and fat (ft). In both cases, UAS transgenes cannot be easily made, so only the effects of removing the genes have been studied. Like ds, ft encodes a huge molecule with many cadherin repeats, and as with ds, null mutant flies do not develop. The mutant imaginal discs grow excessively, and there are some effects on the polarity of bristles. Clones of ft^- cells in otherwise wild-type discs are abnormally large; in the abdomen, these clones tend to be creased, as if they were trying to grow beyond their normal compass (Casal, 2002).

In the A compartments of the tergites, ft^- clones tend to disturb and reverse polarity behind the clone, while, in the P compartments, they tend to reverse in front. Thus, ft^- clones, like ds^- and fj^- clones, have opposite effects on polarity in A and P. When the ft^- clones are near the A/P boundary, they behave as would be expected from the provenance of the cells neighboring the clone: clones at the back of the A compartment fail to reverse the P cells behind (P cells normally reverse in front of a ft^- clone), and P clones fail to reverse A cells in front of them (A cells normally reverse behind a ft^- clone) (Casal, 2002).

Thus, ft^- clones, like ds^- and fj^- clones, have opposite effects on polarity in A and P. Further, the effects of ft^- clones are similar to those of fj^- clones but are opposite those of UAS-fj and ds^- clones. For example, in the A compartment, hairs point toward ft^- clones but away from UAS-fj clones, whereas, in P, they point away from ft^- clones but toward UAS-fj clones. Using the logic deployed with fj and ds, it is inferred that Ft activity is reflected like that of Fj, forming a peak at the segment boundary and declining to a trough at the A/P boundary. But note that ft^- clones can cause polarity reversals anywhere within A, as well as in anterior P -- but fj^- clones do so only in anterior A. This difference argues for a model in which Fj is produced only by cells flanking the segment boundary, acting non-autonomously on cells further away, whereas Ft activity might be required autonomously in all cells, with any differential in Ft activity between neighboring cells determining their polarity (Casal, 2002).

The three genes ds, ft, and fj are functionally linked: mutations in all three damage the tarsi in a similar way; ds and ft encode similar cadherin molecules, and they and fj interact genetically. For the Drosophila eye, it has been proposed that the products of ds, ft, and fj work together in a linear pathway in the developing ommatidia. This pathway begins with a gradient of Wg and leads to the differential activation of Fz in the presumptive R3 and R4 cells. According to this model, graded Wg spreads into the eye from sources at the dorsal and ventral poles, induces Ds expression, represses Fj expression, and thereby generates reciprocal Ds and Fj gradients. Fj activity then represses Ds activity and reinforces this reciprocity. In turn, the Ds gradient then patterns the activity of Ft, which is ubiquitously expressed. Finally, the gradient of Ft activity promotes the activation of Fz in the more equatorial cell and directs it to become the R3 cell, while the more polar cell becomes the R4 cell (Casal, 2002).

The present results point to parallels between the action of Fj, Ds, and Ft in the eye and abdomen. In both cases, a morphogen (Wg in the eye, and Hh in the abdomen) appears to govern polarity through the induction of reciprocal gradients of Fj and Ds expression. Further, in the abdomen, Hh organizes polarity at least in part through the induction of Wg. Hence, as in the eye, peak Wg activity occurs where fj is repressed and where ds is expressed. Finally, the results suggest that the gradient of Ds in the abdomen is reciprocal to that of Ft activity, consistent with the model proposed for the eye. These parallels suggest that the three genes are part of a mechanism common to the eye and abdomen and presumably elsewhere (Casal, 2002).

How might A and P cells be programmed so that bidirectional activity gradients of Fj, Ds, or Ft would lead to a unidirectional slope of Fz activity? It is suggested that a transcription factor, Engrailed, encoded by the selector gene that distinguishes P from A cells, also alters the response of P cells relative to A cells, so that in A cells, Fz might accumulate at the cell edge where Fj is lowest, while, in the P cells, it might accumulate where it is highest. The result would be a localized accumulation of Fz along the posterior edge in all cells, whether in A or P. A precedent comes from yeast, where haploid (a or a) cells bud axially near prior budding sites, while diploid (a/a) cells bud in a bipolar fashion at the site farthest from the previous bud. In yeast, this switch in polarity is also governed by transcription factors encoded by the mating-type locus (Casal, 2002).

In the abdomen, there are observations that do not fit with a simple linear pathway as proposed for the eye. For example, hair polarities are not randomized in fj^-, ds^-, or ft^- mutant tissues, and even entirely fz^- flies show relatively normal polarity in most regions. Nevertheless, consistent changes in polarity are generated by disparities in the activity of each of these polarity genes, usually across clone borders. Hence, cell polarity may depend on multiple signals of which the mutually reinforcing effects of Fj and Ds are but one example (Casal, 2002).

Mutations that cause a reduction in cell division are common, but those, such as ft, that cause increased growth are rare. The ft gene may be a link between planar polarity and growth — it has been suggested that a morphogen gradient may control both. If the slope or vector of a morphogen is used to specify planar polarity, the local steepness of that same gradient might provide a measure of dimension. This measure would then help determine the probability of cell division and apoptosis, regulate the rate of net growth, and limit the final size (Casal, 2002).

Nonautonomous planar polarity patterning in Drosophila: Dishevelled-independent functions of Frizzled

The frizzled (fz) gene of Drosophila is required for planar polarity establishment in the adult cuticle, acting both cell autonomously and nonautonomously. These two activities of fz in planar polarity are temporally separable in both the eye and wing. The nonautonomous function is dishevelled (dsh) independent, and its loss results in polarity phenotypes that resemble those seen for mutations in dachsous (ds). Genetic interactions and epistasis analysis suggest that fz, ds, and fat (ft) act together in the long-range propagation of polarity signals in the eye and wing. Evidence has been found that polarity information may be propagated by modulation of the binding affinities of the cadherins encoded by the ds and ft loci (Strutt, 2002).

There are a number of reasons for thinking that fz nonautonomous activity in the eye is closely related to ds and ft function. The phenotypes of clones lacking early fz function are similar to those of ds clones and ft clones. Furthermore, there are strong genetic interactions between these factors. Finally, an epistasis test between the clonal phenotypes of fz and ds gives an apparently additive (or possibly synergistic) phenotype. These results are consistent with fz acting jointly with ds and ft in the nonautonomous propagation of polarity information. A similar function for ds has been suggested on the basis of studies in the wing, it having been shown that ds nonautonomously affects trichome polarity and that it is likely to be involved in the maintenance or propagation of an fz-dependent nonautonomous polarity signal (Strutt, 2002).

Thus, overall data from both the eye and wing support fj acting upstream of ds and ft, which then act jointly with fz nonautonomous function in the long-range propagation of polarity information. Uncharacterized mechanisms of intercellular signaling then lead to autonomous activation of fz and assembly of asymmetric polarity protein complexes. Note is taken of the contrast with the recent suggestion that ds and ft act directly upstream of the autonomous function of fz (Strutt, 2002).

Other factors or mechanisms must also be involved in nonautonomous propagation of polarity information, in order to explain all of the observations. For instance, complete loss of fj function does not result in a loss of polarity patterning in the wing, indicating that there must be other upstream patterning factors. Furthermore, clones of fj and ft give stronger nonautononomous phenotypes in a central portion of the wing, whereas ds and fz seem to give rather similar phenotypes throughout. This suggests that there are other modulators of pathway activity that have region-specific effects (Strutt, 2002).

Groups of cells lacking fj function tend to round up into tight foci, appearing to have greater affinity for each other than for their fj-expressing neighbors. Furthermore, in mutant cells abutting fj-expressing neighbors, the cadherins Ds and Ft are preferentially found at the cell junctions touching fj⁺ cells. These observations support the notion that one role of fj in wing patterning is to alter the adhesive properties of cells and also of the cadherins Ft and Ds. It is also noteworthy that loss of ft activity results in Ds no longer being tightly localized in the apical junctional zone of cells and that, similarly, loss of ds seems to result in reduction of apical Ft localization (Strutt, 2002).

It is speculated that a gradient of fj activity in the wing might lead to graded Ds/Ft activity and, hence, cell adhesion. Such a gradient of cell adhesion constitutes a possible mechanism for the long-range transmission of polarity information, although direct evidence for this is lacking. It is noteworthy that fj, ft, and ds mutations also all result in truncations of the wing on the proximodistal axis, and it is possible that this phenotype is in some way due to effects on cell adhesion (Strutt, 2002).

Interestingly, the effect of fj clones on Ds/Ft is cell autonomous. It was suggested that, on the basis of its amino acid sequence and in vitro studies, fj encodes a secreted factor and that this property could explain its nonautonomous phenotypes. These results indicate that at least some functions of fj are cell autonomous (Strutt, 2002).

Drosophila genetic variants that change cell size and rate of proliferation affect cell communication and hence patterning

The role of genetic variants that affect cell size and proliferation in the determination of organ size has been investigated. Genetic mosaics of loss or gain of function were used in six different loci, which promoted smaller or larger than normal cells, associated with either smaller or larger than normal territories. These variants have autonomous effects on patterning and growth in mutant territories. However, there is no correlation between cell size or rate of proliferation on the size of the mutant territory. In addition, these mosaics show non-autonomous effects on surrounding wild-type cells, consisting always in a reduction in the number of non-mutant cells. In all mutant conditions the final size (and shape) of the wing is different from normal. The phenotypes of the same variants include higher density of chaetae in the notum. These autonomous and non-autonomous effects suggest that the control of size in the wing is the result of local cell communication defining canonic distances between cells in a positional-values landscape (Resino, 2004).

Size of insect organs is sex- and species-specific. In the Drosophila wing, where most of the studies on size control have been carried out, the determination of the size of imaginal disc is disc-autonomous. Young imaginal discs transplanted to the abdomens of adult flies grow after several days of culture, irrespective of hormonal and nutritional conditions, to a maximal size that corresponds to that of mature imaginal discs. Minute mosaics and regeneration experiments reveal that a final normal size is attained irrespective of the rate of cell proliferation. Clonal analysis of cell proliferation in wild-type wings show regional differences related to specification or differentiation, indicative of local as opposed to global control of organ size. Size of the growing imaginal disc depends on the allocation of postmitotic cells along the main axes of the wing in regimes that change with developmental time. There is no indication that cell proliferation or cell allocation relates to the position of cells with respect to distances to compartments boundaries, where postulated diffusible morphogens are at maximal concentration (Resino, 2004).

If control of cell proliferation is local, the question arises as to how this is achieved. Can variations in cell size affect the final size of the organ or its proliferation parameters? These variations can be produced using mutations, usually lethal in organisms, and have to be studied in genetic mosaics. Mosaics of haploid territories (with half the cell size of diploid cells) led to bigger territories with more cells than diploid territories. Male wings have less and smaller cells than females, characteristics that are locally autonomous in gynandromorphs. For mutations that affect cell size, it has to be considered that they cause different perturbations that may affect other cellular parameters in addition, such as cell viability, proliferation rate or cell adhesion, which make difficult the interpretation of the phenotype. Thus, the insufficient function of genes involved in cell cycle progression, such as string (stg), cdc2 and cyclins or E2F (cycE positive regulator), may retard the cell cycle and cause cell mortality, an increase in cell size and smaller mosaic territories in otherwise apparently normal sized discs. Mutant cells in these mosaics do not differentiate properly. On the contrary, over-expression of the same cell cycle genes (i.e. stg, cycE, cycD-cdk4) or of their activators (i.e., E2F) in imaginal disc clones cause acceleration of their characteristic phases of the cell cycle, as well as a reduction of cell size (except cycD-cdk4 combination) and an increase in number of cells of the mutant territory compared with control cells in apparently normal sized mosaic wing discs. These effects are more extreme in some genetic combinations (e.g., cycE-stg) because they cause an acceleration of the whole cell cycle. These studies conclude that cell size reduction/increase is 'compensated' by increment/decrement in cell number in the mutant territory, as if the organ would compute a global normal size, because the mutant wing disc territories have an apparent wild-tupe size. This interpretation is biased by the fact that those mosaics show high cell mortality. When this is prevented with the coexpression of P35, the extra growth of the mutant territories in discs and clones is even higher, leading to abnormally shaped mutant territories. The over-expression of the cycD-cdk4 combination in the eye reaches the adult stage and causes larger and abnormally shaped mutant territories. These studies have not analyzed non-autonomous effects in non-mutant territories of the same discs (Resino, 2004 and references therein).

Less drastic mutant effects associated with cell viability are obtained with mutant perturbations in the signal transduction and reception of the insulin pathway. As a rule, loss of function of Drosophila Insulin Receptor (Inr), chico or Dp110 causes reduction in both cell size and cell number of mutant territories. This is similar to what happens in wild-tupe flies exposed to malnutrition or premature metamorphosis. This holds for each member of the insulin receptor pathway except for Drosophila S6 kinase (S6K), because S6K loss of function only reduces cell size but not cell number. On the contrary, the gain of function of genes of this pathway causes larger cells and an increase in the number of cells of the mutant territory in mosaics. The loss of function of myc in diminutive mutants leads to smaller flies, with smaller cells, in addition to poor cell viability. Its overexpression causes larger cells but not larger territories, suggesting that in this latter condition (but not the former) the wing size in globally controlled by a normalizing compensating mechanism (Resino, 2004 and references therein).

The results show a great heterogeneity in the response of regional size to genetic perturbations that cause variations in cell size during cell proliferation. In fact, both smaller or larger than normal cell size may accompany normal, larger or smaller mutant territories. In the present paper, the effects on cell proliferation of mutant conditions in six loci that cause smaller and larger cell sizes have been studied. Of these, one corresponds to a new gene and five to previously studied genes that affect cell size. They were chosen as examples of the cell behavior variants, as representatives of mutant effects on cell size (larger and smaller than normal) and rate of proliferation (slower and faster than normal). The choice was made without considering the genetic/molecular bases of the corresponding wild-tupe alleles, in any case mechanistically far separated from the analyzed phenotype. Their autonomous effects in mutant territories and in the mosaic wing as a whole were studied: nonautonomous effects were documented as well (Resino, 2004).

Adult cell size is measured by the exposed planar surface of the cuticle cells. In principle, this may not reflect the size of the proliferating cells, when organ size is determined. However, in some of the cases examined in this study, cell dissociation has revealed by direct estimation the larger or reduced cell size in the proliferating wing disc cells. In others, cell size during growth is inferred by the mutant effects on pattern formation, a process that precedes final cell differentiation, as in the notum pattern of microchaetae. This pattern results from the singularization of sensory organ mother cells (SOMC) in a field of epidermal cells through a process of lateral inhibition in a field of proneural clusters. Thus, the final pattern reveals cell-cell interactions or communication, as observed in the form of cell projections emanating from epidermal cells. It holds for all mutant and genetic combinations examined in this study that the pattern, number and density of chaetae are all altered in the notum (in the mutant Dmcdc2^E1-24 cells fail to differentiate chaetae). In all cases, chaetae appear more densely spaced (separated by less epidermal cells) associated with an increase in the total number of chaetae. These variations to the wild-tupe condition suggest that mutant cells have impaired the capacity to signal among themselves to define spaced SOMC singularization. Whether this is or is not associated with cell size in individual cases is not known. These pattern effects reveal abnormal cell communication between cells during cell proliferation (Resino, 2004).

Although less easy to measure in mosaic nota, there is a phenotypic association of variable cell size with a reduction (in l(3)Me10, gig^Me109, Dp110^D945A) or an increase (EP(3)3622, ft^a13, Dp110-CAAX) in notum sizes. But there is no apparent causal relation between both parameters of cell size and number of cells making the adult notum. Perhaps cell viability associated with the mutation, as in l(3)Me10 and gig^Me109, may account for the observed lack of correlation between both parameters. However, these effects on notum size in other cases may also reflect failures in cell-cell communication leading to more or less cell proliferation (Resino, 2004).

The relationship between cell size and growth can be more readily measured in the wing. The studied genetic variants can be grouped, based on variations in these parameters, as follows:

• (1) Reduction in cell size and in number of cells within a wing sector is associated with a reduction of the size of the mosaic sector (l(3)Me10 and Dp110D954A). This is a common phenotype in many mutant conditions such as in mosaics of vein and torpedo, both of the Epidermal growth factor receptor (EGFR) pathway. This situation is very common to many loss of function alleles of many genes (Resino, 2004).
• (2) Reduction in cell size and increase in number of cells is associated with increase in the sector area (EP(3)3622 and ft^a13) (Resino, 2004).
• (3) Increase in cell size and reduction in number of cells in mosaic sectors is associated with a reduction in the sector area in gig^Me109 and Dmcdc2^E1-24 (Resino, 2004).
• (4) Increase in cell size and increase in cell number is associated with an increase in the mosaic sector area in Dp110-CAAX (Resino, 2004).

The autonomous effects on reduced clone size can result from the poor viability of mutant cells (l(3)Me10 or Dmcdc2^E1-24), as shown in twin clonal analysis and cell death monitoring. The increased clone size of EP(3)3622, ft^a13 or Dp110-CAAX reflects higher than normal cell proliferation, however there are no correlations between cell size and clone size. Despite this lack of correlation it holds for all mutants examined in this study that, concerning the non-autonomous effects on growth in the mosaic wing sector: the non-mutant cells of the sector are always reduced in number. No cases were found in which the reduction or increase in sector size by the presence of mutant territories is compensated by wild-tupe cells to obtain a normal sized sector (Resino, 2004).

The mosaic wings show, in addition to autonomous effects within mutant sectors, non-autonomous effects in the rest of the wing. It holds for all cases studied that wings with entire or mosaic wing sectors show a reduction in the total area of the wing or more in particular in non-mosaic areas (sectors or compartments) of the wing. This phenomenon is designated as 'positive' or 'negative' accommodation, depending on its correlation with the size of the mutant region. This phenomenon could be easily trivialized for mutations that cause size reduction and 'positive accommodation'. It is arguable that there are not enough cells in the mutant territories to confront with normal growing cells abutting the clone, the sector or the mutant compartment. 'Positive accommodation' could result from adjustment between poorly growing cells and normal ones. However this large effect hardly explain 'negative accommodation' for the whole wing. 'Negative accommodation' occurs in mosaic wings with mutant territories with more cells than normal, such as EP(3)3622, ft^a13 or Dp110-CAAX (Resino, 2004).

Reduction in the size of non-mutant territories in mosaic wings cannot be explained either by delay in development (mosaic flies hatch at the same time as sib controls) or age of clone initiation. It cannot be explained either by cell death, because there is enough time for extraproliferation to reach normal sized wings, since it occurs in mosaics where cell death has been massively induced in Gal4 territories. 'Negative accommodation' is surprising because one would expect that larger than normal mutant territories should provide adjacent wild-tupe cells with more growth signals (Resino, 2004).

To account for this 'negative accommodation' it is postulated that mutant cells do not convey among themselves and to wild-tupe cells sufficient signals necessary for them to proliferate. These signals may depend on cell-cell communication. In the notum it has been seen that failures in cell-cell communication may account for abnormal chaetae patterning and notum size. The same may apply to the wing blade, although there are not enough pattern elements to support this inference (Resino, 2004).

A model has been proposed to explain controlled cell proliferation, based on local cell-cell signalling, as opposite to reception of graded amounts of morphogens emanating from compartment boundaries, such as Dpp and Hedgehog or Wingless. The Entelechia model (Interactive Fly editor's note: 'Entelechia' is a Greek term coined by Aristotle for the complete reality or perfection of a thing, and refers to the process of coming into being) states that cell proliferation results from local interactions between neighboring cells. In these interactions, cells compute positional values, presumably expressed in the cell membrane. Positional value discrepancies elicit cell division and readjustment of positional values of daughter cells to those of neighboring cells. These values differ along the two main axes of the wing, A/P and Pr/Ds. Cell proliferation occurs within clonal boundaries; those of compartments in the early disc and other boundaries, such as veins, later. In these boundaries the interchange of some type of signals help to increase positional values at the border, eliciting cell division, cascading down to intermediate regions with minimal values. Cell proliferation is intercalar and driven by differences in positional values between cells with lower and higher values. These minimal differences may reflect canonic efficiencies ('increments') in transduction of signals (ligands/receptors) between neighboring cells. Cell division ceases in the anlage when cells in the boundaries reach maximal values and their increments, between all the cells of a region become minimal. The anlage has then reached the Entelechia condition of growth, characteristic of the organ, the sex and species (Resino, 2004).

An organ such as the wing, grows co-ordinately through compartments and clonal boundaries because maximal positional values result from cell interactions at both sides of the boundaries. In this respect compartments or wing sectors are not independent units of cell proliferation. This was first seen in bithorax-Complex (bx-C) mutants, where either the A or P compartments of the haltere were transformed to A or P compartments of the wing. The untransformed A or P haltere compartments contain now more cells, and the transformed ones less than a wild-tupe A or P wing compartment. This accommodation is explained as due to the reduced extent of the compartment boundary between apposed mutant and nonmutant compartments. Similar accommodation effects have been already reported in other mutant conditions, such as mutants of the EGFR pathway in extramacrochaetae (emc) and in nubbin (nub). In the latter case, the presence of proximal wing mutant territories causes a distal reduction in growth in all the wing compartments (Resino, 2004).

The Entelechia model helps to understand the behavior of mosaic wings for the mutants examined in this study. In all cases, clones or regions with smaller or larger cells and with less or more cells than normal, cause autonomous effects on growth in mutant territories but also a non-autonomous 'accommodation' in the rest of the wing formed by wild-tupe cells. It should be emphasized that the effects on proliferation between mutant and non-mutant territories are reciprocal; the non-mutant territories rescuing proliferation in the mutant territories and vice versa. It is hypothesized that failures in cell communication of positional values to/from neighboring mutant or non-mutant cells affect the 'increment' values of the model. This leads to reduced proliferation in both genetic territories between cells because cells cannot generate higher positional values and thus promote intercalar proliferation. This finding indicates that the size of territories does not depend on distances from diffusible morphogen sources, measured either in physical terms or in number of cells, or on other postulated parameters such as measuring global cell mass or wing length. How would these global dimensions be defined, and how would they be computed by individual cells? How would one explain that mosaic territories separated from compartment boundaries (or morphogen sources) can affect the growth of wild-tupe territories far away all over the wing? It seems rather that cell proliferation control depends on local cell interactions (cell-cell communication) that define positional values throughout the whole growing organ (Resino, 2004).

fourjointed interacts genetically with ft and ds in planar polarity and proximodistal patterning

four-jointed (fj) is required for proximodistal growth and planar polarity in Drosophila tissues. It encodes a predicted type II transmembrane protein with putative signal peptidase sites in its transmembrane domain, and its C terminus is secreted. Fj has therefore been proposed to act as a secreted signalling molecule. Fj protein has a graded distribution in eye and wing imaginal discs, and is largely localized to the Golgi in vivo and in transfected cells. Forms of Fj that are constitutively secreted or anchored in the Golgi were assayed for function in vivo. Cleavage and secretion of Fj is shown to not be necessary for activity, and Golgi-anchored Fj has increased activity over wild type. fj has similar phenotypes to those caused by mutations in the cadherin-encoding genes fat (ft) and dachsous (ds). fj is shown to interact genetically with ft and ds in planar polarity and proximodistal patterning. It is proposed that Fj may act in the Golgi to regulate the activity of Ft and Ds (Stutt, 2004).

In Drosophila, the atypical cadherins Ft and Ds are good candidates for being the ultimate targets of fj activity. They are required for both planar polarity and PD patterning, and have similar mutant phenotypes to fj. In addition, fj interacts genetically with ds and ft in both planar polarity and PD patterning. Interestingly, ds fj double mutants have surprisingly strong phenotypes, which were qualitatively different to those of the single mutants, including duplications or transformations of limb structures. However, no such phenotypes are seen in any of the double mutant combinations, suggesting that the duplications/transformations may be specific to the combination of chromosomes used in classical experiments. The current results instead show that mutations in fj enhance the phenotypes of both ft and ds hypomorphic mutations, suggesting that these genes act in a common pathway (Stutt, 2004).

Epistasis experiments further demonstrate that ds is required to mediate fj function, and therefore ds acts downstream of fj; this is in agreement with data based on clonal analysis of ds and fj. Interestingly, recent experiments have also revealed a role for fj in regulating the intracellular distribution of Ds and Ft. In wild-type tissue, Ds and Ft colocalize at apicolateral membranes, and their localization is mutually dependent. Inside fj mutant clones, Ds and Ft localization is largely unaltered. However, in the row of mutant cells immediately adjacent to wild-type tissue, Ft and Ds preferentially accumulate on the boundary between fj⁺/fj^- cells. In addition, cells inside the fj clones appear to be 'rounded-up', suggesting that they prefer to adhere to each other rather than to non-mutant cells. Thus, it is thought that fj modulates the activity and intermolecular binding properties of Ft and Ds (Stutt, 2004).

An interesting point to note is that both ds and ft show planar polarity phenotypes as homozygotes, whereas fj only shows polarity phenotypes on the boundaries of mutant clones. The fj phenotypes have been explained by models in which fj acts redundantly to regulate the production of a gradient, the direction of which determines polarity. Thus, in homozygotes the direction of the gradient is unchanged, and animals show no major defects; but at clone boundaries there is a discontinuity in the direction of the gradient, leading to inversions of polarity. This model can be extended to suppose that Fj may modulate Ds/Ft activity, but that it does not act as a simple on-off switch; rather Ds/Ft retain some activity even when Fj is not present (Stutt, 2004).

In the absence of a known enzymatic function for Fj, the mechanism by which it might modulate Ft and Ds activity remains uncertain. It is speculated that since Fj acts intracellularly, it is possible that it promotes or mediates the post-translational modification of Ds and/or Ft proteins, and that these molecules mediate the non-autonomous signalling functions of Fj. However, the large size of the Ft and Ds gene products (5147 and 3380 amino acids, respectively) renders the analysis of their post-translational modification highly challenging (Stutt, 2004).

Interactions between Fat and Dachsous and the regulation of planar cell polarity in the Drosophila wing

It has been suggested that a proximal to distal gradient of the protocadherin Dachsous (Ds) acts as a cue for planar cell polarity (PCP) in the Drosophila wing, orienting cell-cell interactions by inhibiting the activity of the protocadherin Fat (Ft). This Ft-Ds signaling model is based on mutant loss-of-function phenotypes, leaving open the question of whether Ds is instructive or permissive for PCP. Tools have been developed for misexpressing ds and ft in vitro and in vivo, and these have been used to test aspects of the model. (1) This model predicts that Ds and Ft can bind. Ft and Ds are shown to mediate preferentially heterophilic cell adhesion in vitro, and each stabilizes the other on the cell surface. (2) The model predicts that artificial gradients of Ds are sufficient to reorient PCP in the wing; the data confirms this prediction. (3) Loss-of-function phenotypes suggest that the gradient of ds expression is necessary for correct PCP throughout the wing. Surprisingly, this is not the case. Uniform levels of ds drive normally oriented PCP and, in all but the most proximal regions of the wing, uniform ds rescues the ds mutant PCP phenotype. Nor are distal PCP defects increased by the loss of spatial information from the distally expressed four-jointed (fj) gene, which encodes putative modulator of Ft-Ds signaling. Thus, while the results support the existence of Ft-Ds binding and show that it is sufficient to alter PCP, ds expression is permissive or redundant with other PCP cues in much of the wing (Matakatsu, 2004).

Several gain-of-function findings are consistent with previous loss-of-function findings, and support the model that Ft-Ds signaling is sufficient to influence wing PCP. Ft and Ds preferentially bind in vitro. Patterned misexpression of ds is sufficient to alter wing PCP, consistent with its proposed role as a ligand. The effects of ft or ds misexpression on the direction of hair polarization are usually the opposite of those previously reported from ft or ds loss of function. The direction of hair polarity induced by ectopic ft is usually the opposite of that induced by ectopic ds, consistent with the proposal that Ds binding inhibits Ft activity. Finally, the effects of Ft misexpression are reduced in a ds mutant background, consistent with the proposed role of Ft as a receptor (Matakatsu, 2004),

Nonetheless, the data also show that the proximal to distal gradient of ds expression is not necessary for PCP throughout the wing, despite the distal defects observed in loss-of-function ds mutants. Instead, the experiments show that uniform ds misexpression can rescue the PCP defects caused by a ds mutation in all but the most proximal portions of the wing. Thus, ds is permissive for PCP in most of the wing, and there must be another polarity cue in the distal wing that is sufficient to orient PCP in the presence of uniformly transcribed ds. The experiments indicate that this distal cue is not provided by the distally expressed Fj protein: distal PCP is not disrupted either by uniform misexpression of both ds and fj, or by uniform misexpression of ds in a fj null mutant (Matakatsu, 2004),

It remains possible that the distal cue functions by regulating Ft-Ds signaling. These studies tested the PCP inputs from the patterns of ds and fj transcription, but unknown factors might post-transcriptionally regulate the forms of Ds or Ft protein produced, or their availability at the cell surface. It also is possible that Ft activity is spatially regulated by binding partners other than Ds. ft mutants have stronger PCP and disc overgrowth defects than do ds mutants, and misexpression of ft still causes PCP defects in a ds mutant lacking detectable cell surface protein (Matakatsu, 2004),

Alternatively, the cue may be provided by a mechanism that is completely independent of Ft or Ds. One often-proposed candidate is signaling via the Drosophila Wnts, especially given their patterned (distal or marginal) expression. However, although the misexpression of Drosophila wnt4 can disrupt wing PCP, PCP defects have not been reported in Drosophila Wnt mutants (Matakatsu, 2004),

Although a ds gradient is not required for PCP in most of the wing, it is possible that such a gradient is required locally in the portion of the wing near and proximal to the anterior cross vein. Proximal ds mutant PCP defects could not be rescued with uniform Ds expression, and the data suggest that this is not simply a failure caused by insufficient Ds levels. Thus, the view is favored that this sharp Ds gradient acts as a PCP cue in the proximal wing. If so, this indicates that the cues that orient PCP in the wing are not generally distributed; rather, the wing may be a patchwork of different regions that rely on different cues. This would provide a mechanism for locally altering PCP during evolution without globally affecting polarity in the wing (Matakatsu, 2004),

The hypothesis that Ft acts as a receptor and Ds acts as a ligand for PCP is based, not only on the uniform expression pattern Ft, but also on epistasis experiments in the eye, where the PCP activity of ds clones appears to depend on the presence of ft. Wing PCP can also be disrupted by the expression of a truncated form of Ds lacking its intracellular domain, which is consistent with Ds acting as a ligand (Matakatsu, 2004),

However, this study has shown that misexpressed Ft retains PCP activity in a ds mutant that eliminates detectable cell surface Ds. Thus, Ft activity is apparently not strictly dependent on patterned Ds expression. Again, this is consistent with the greater severity of ft mutant phenotypes compared with ds, and with the finding that uniform misexpression of ft but not ds can cause PCP defects. Since there is no evidence for homophilic Ft binding, the unbound Ft molecule may have basal PCP activity. Alternatively, low-level homophilic binding or heterophilic binding to some unknown ligand may activate Ft in the absence of Ds (Matakatsu, 2004),

It is not yet known how Ft-Ds interactions regulate the polarized redistribution of the core polarity proteins in the older pupal wing. The cytoplasmic domains of Ft and Ds contain potential regions for ß-catenin binding, and ft and ds mutants can enhance the effects of ß-catenin (Armadillo) misexpression. However, although expression of DE-cadherin in vitro results in a detectable concentration of Armadillo at the cell membrane, no similar effects were detected after expression of ft or ds. Moreover, clones homozygous for a strong armadillo mutation do not affect PCP. It has also been suggested that the cytoplasmic domain of Ft binds to and changes the activity of Grunge, the Drosophila homolog of the Atrophin transcriptional co-repressor, but it is not known whether this interaction is altered by Ft-Ds binding (Matakatsu, 2004),

The studies examining the timing of Ds activity suggest that its effects on the polarization of the core polarity proteins are likely to be indirect, since Ds acts before the polarized redistribution of the core polarity proteins within cells can be detected. Patterned misexpression of Ds at later stages, during the time of core protein polarization, has no effect on PCP. The period sensitive to ds misexpression is roughly congruent with the period of early Fz activity; if loss of Fz is limited to a period from 6 to 24 hours AP it leads to distinct, ds-like PCP defects. Thus, early Fz and Ds activity may be linked, or they may share a common target (Matakatsu, 2004),

The only known sign of cell polarization during the stages sensitive to Ds and early Fz activity is the redistribution of the Widerborst PP2A regulatory subunit from the anterior-proximal side to the distal side of wing cells at some time between 8 and 18 hours AP. Reductions in Widerborst activity can disrupt the polarized redistribution of Fmi and Dsh, suggesting an instructive role. However, Widerborst polarization is not affected by ectopic Fz expression, making it less likely that Widerborst polarization mediates early Fz activity (Matakatsu, 2004 and references therein),

A final interesting feature of the results is the preferentially heterophilic binding observed between Ft and Ds in vitro. This result is consistent with analyses of protein distribution within and adjacent to ft and ds mutant and overexpression clones. With the exception of the desmosomal cadherins, this kind of binding is unusual for cadherin-like proteins (Matakatsu, 2004),

A number of mammalian Fat-like (Fat1, Fat2, Fat3, XP_227060) and Ds-like (Protocadherin 16, Cdh23) proteins have been identified. Mutations and knockouts have been examined for a few of these; however, conjectures about the bases of the mutant phenotypes have largely assumed that these proteins mediate homophilic cell adhesion. It will be interesting to see whether the preferentially heterophilic interactions observed in Drosophila are preserved in similar mammalian proteins (Matakatsu, 2004),

The tumor suppressor gene fat modulates the EGFR-mediated proliferation control in the imaginal tissues of Drosophila melanogaster

Molecules involved in cell adhesion can regulate both early signal transduction events, triggered by soluble factors, and downstream events involved in cell cycle progression. Correct integration of these signals allows appropriate cellular growth, differentiation and ultimately tissue morphogenesis, but incorrect interpretation contributes to pathologies such as tumor growth. The Fat cadherin is a tumor suppressor protein required in Drosophila for epithelial morphogenesis, proliferation control and epithelial planar polarization, and its loss results in a hyperplastic growth of imaginal tissues. While several molecular events have been characterized through which fat participates in the establishment of the epithelial planar polarity, little is known about mechanisms underlying fat-mediated control of cell proliferation. Evidence is provided that fat specifically cooperates with the epidermal growth factor receptor (EGFR) pathway in controlling cell proliferation in developing imaginal epithelia. Hyperplastic larval and adult fat structures indeed undergo an amazing, synergistic enlargement following to EGFR oversignalling. Such a strong functional interaction occurs downstream of MAPK activation through the transcriptional regulation of genes involved in the EGFR nuclear signalling. Considering that fat mutation shows di per se a hyperplastic phenotype, a model is suggested in which fat acts in parallel to EGFR pathway in transducing different cell communication signals; furthermore its function is requested downstream of MAPK for a correct rendering of the growth signals converging to the epidermal growth factor receptor (Garoia, 2005).

These results support the hypothesis of a relevant functional interaction between ft and genes of the EGFR pathway. When EGFR signalling is raised in fact, an amazing, non-additive increase in ft-induced proliferation is observed. ft clones in UAS-rho or UAS-raf^GOF wings where an EGFR oversignalling is induced show the same distinctive features of those induced in a wild-type background (tissue hyperplasia, reduced cell size, loss of planar polarity), with a phenotype much more severe concerning outgrowth number and dimensions. On the contrary, in experiments where the EGFR signalling is reduced, ft-induced hyperproliferation results are partially suppressed. The same trend is observed in the adult eye; the ft head-capsule where rho or raf^GOF were ectopically expressed was particularly enlarged (Garoia, 2005).

The most dramatic effects were however observed in the eye and wing imaginal discs where the EGFR signalling was increased in the presumptive dpp expression domains. The controls did not show significant phenotypes; conversely, in ft discs severe effects were observed, including non-autonomous aberrations in the disc morphology. In the ft raf^DN eye disc a strong non-autonomous reduction of the eye presumptive territory was observed that may be caused by decrease of proliferation rate and/or by increase of cell death. Non-autonomous cell death may result from cell competition, ectopic cell proliferation or experimentally induced apoptosis. An increase in non-autonomous cell death was observed not only in ft raf^DN discs but in all the ft/UAS combinations analyzed. Thus, cell death does not seem to be the mechanism through which the growth deficit occurs in the ft raf^DN eye disc, suggesting a more complex interaction between ft and EGFR activity in the modulation of eye disc growth (Garoia, 2005).

Therefore ft interacts with the EGFR pathway modulating its proliferative signal; ft function is therefore involved ingrowth regulation, allowing the structures to correctly interpret signals incoming from EGFR that are essential for eye and wing to acquire the final shape. No interaction was found with mutants involved in pathways other than the EGFR cascade (N, dpp and wg), making the role of ft in the control of the imaginal disc growth specifically dependent on its interaction with the EGFR pathway (Garoia, 2005).

The proximal ft clone allocation along the wing blade arises from an alteration in the growth direction and not from a different viability of the cells relative to their layout in the proximo-distal axis; it is then obvious that ft cells show a greater 'affinity' for the proximal region of the wing. The ft mutant phenotype seems anyway to be influenced by EGFR signalling also with respect to the proximalization of the clones; the distribution is indeed more homogeneous along the wing blade if the EGFR signalling is altered. It is interesting to notice that an EGFR signalling reduction or increase produces, in this case, the same biological effect, while the proliferative phenotype is directly correlated to the activity of the EGFR effectors. The activity of the EGFR cascade is spatially and temporally modulated during development, and in the wing disc it is gathered in the hinge and vein presumptive regions. Even if there are no evidences that the EGFR signalling plays a role in the P/D patterning of the wing blade, recent studies have shown that a gradient of EGFR activity is required for the correct P/D development of the leg. In the experimental conditions used, the MS1096-GAL4 driver creates an almost homogeneous EGFR signal along the wing blade, determining a quasi wild-type distribution of ft clones. This role in the modulation of the differential distribution of ft mutant cells along the P/D wing axis suggests for the EGFR pathway an involvement in the morphogenetic events that control the final shape of the wing (Garoia, 2005).

The ft-induced hyperplasia is associated with an abnormal pattern of gene expression, as visualized in two-dimensional protein gels of ft mutant imaginal discs. Although no data is available to hypothesize a cytoplasmic interaction between ft and the EGFR signalling, an alteration in MAPK expression could not be excluded. The ft mutant phenotypes, however, do not include the differentiation defects typical of the EGFR pathway genes, whose activity is indeed not altered since no significant modifications of the activated dpERK levels or patterns were detected in the ft tissues with respect to the wild-type (Garoia, 2005).

The results shown in this paper suggest that the interaction between ft and EGFR takes place at the proliferation level, while differentiation signals controlled by the EGFR pathway appear unaffected. With the aim to find some mechanisms that could explain the synergic phenotype of ft and EGFR mutations, the transcriptional levels of yan, dmyc and pnt, genes involved in proliferation control whose function is regulated by the EGFR cascade, were studied in ft and wild-type imaginal tissues. The results of semi-quantitative RT-PCR trials showed in ft tissues an increase of the transcription levels of yan and dmyc, whereas pnt was unaffected. The Dmyc transcription factor, the unique Drosophila homologue of the Myc family of proto-oncogenes, plays a central role in the control of cell growth in Drosophila. Overexpression of ras is capable to increase post-transcriptionally the Dmyc protein levels, promoting the G1-S transition via the increase of CycE translation. The increase in the Dmyc levels, however, affects growth rate but not proliferation, since the shortening of the G1 phase is balanced by the compensatory lengthening of G2, resulting in an increase in cell size but not in cell number. ft mutation otherwise induces an increase of cell proliferation without altering the cell size. Taken together, these results indicate that ft mutation affects not only the G1-S transition via Dmyc but also the G2-M transition, since the coordinated stimulation of the two cell-cycle checkpoints is necessary to increase the proliferation rate in Drosophila imaginal discs. Interestingly, the transcription level of pnt was unaffected in ft mutant discs. pnt is an ETS transcriptional activator that plays a central role in the mitosis control mediated by the EGFR signalling cascade; several studies however suggest the presence of additional Pnt-independent effectors in EGFR-mediated mitosis control. The ft control of the G2–M transition may involve EGFR effectors other than pnt, or molecules functioning through different signalling pathways. The yan gene is another component of the ETS transcriptional regulator family involved in the EGFR signalling. Phosphorylation by MAPK affects stability and subcellular localization of Yan, resulting in a rapid down-regulation of its activity. Yan functions as a fairly general inhibitor of differentiation, allowing both neuronal and non-neuronal cell types to choose between cell division and differentiation in multiple developmental contexts and recent studies indicate that the mammalian homologue of the Drosophila yan, TEL, is overexpressed in tumors. In the Drosophila developing eye yan is expressed in all undifferentiated cells and is down regulated as cells differentiate, so a high yan activity in ft mutant discs is correlatable with the observed proliferative advantage of ft cells (Garoia, 2005).

There are several indications that EGFR signalling can trigger different responses by different activity levels: in the Drosophila eye disc, differentiation requires high signalling levels, whereas lesser EGFR activity promotes mitosis and protects against cell death. These findings indicate that EGFR signalling may coordinate partially independent processes, transferring graded activity to the nucleus, rather than triggering 'all or none' responses. The simultaneous increase of activity in both growth promoters (dmyc) and differentiation repressors (yan) in ft mutant imaginal discs suggests the presence of a mechanism that shifts the EGFR nuclear equilibrium towards a level insufficient to induce differentiation but adequate for promoting cell growth and proliferation (Garoia, 2005).

These results indicate that, in the Drosophila imaginal discs, ft function is necessary for the correct interpretation of the multiple EGFR signals that coordinate proliferation, and that its loss causes misinterpretation of proliferation stimuli leading to tissue overgrowth. This effect may be due, at least in part, to the transcriptional regulation of genes involved in EGFR signalling. Nevertheless, the hyperplastic phenotype of ft mutations cannot be completely ascribed to its role in modulating signals transduced by EGFR, according to the very partial rescue observed utilizing dominant-negative alleles of the pathway. These results suggest that the ft function is not restricted to the modulation of EGFR signals, but controls different developmental events involved in imaginal discs morphogenesis (Garoia, 2005).

Several findings indicate that cadherin-catenin complexes may interact with growth factor receptors. The association of cadherins with growth factor receptors allows the assembly of a locally active apparatus that is essential for the generation of correct cell-cell signalling, as suggested by the downregulation of E-cadherins observed in mammalian tumors. Furthermore, E-cadherins were found to be a direct biochemical target of the EGFR pathway, suggesting a close relation of these molecules with the modulation of cell-cell communication. The only partial homology between the Ft protocadherin and the classic E-cadherins, and the lack of data about interactors for the cytoplasmic domain of ft makes a direct comparison of their function very difficult. Taken together, the data suggest a novel mechanism through which ft tumor suppressor gene and EGFR pathway cooperate in the control of proliferation and morphogenesis in Drosophila imaginal tissues (Garoia, 2005).

The orientation of cell divisions determines the shape of Drosophila organs: Mutations in the Drosophila PCP genes dachsous (ds) and fat (ft) cause changes in the shape of different organs

Organ shape depends on the coordination between cell proliferation and the spatial arrangement of cells during development. Much is known about the mechanisms that regulate cell proliferation, but the processes by which the cells are distributed in an orderly manner remain unknown. This can be accomplished either by random division of cells that later migrate locally to new positions (cell allocation) or through polarized cell division (oriented cell division; OCD). Recent data suggest that the OCD is involved in some morphogenetic processes such as vertebrate gastrulation, neural tube closure, and growth of shoot apex in plants; however, little is known about the contribution of OCD during organogenesis. The orientation patterns of cell division was examined throughout the development of wild-type and mutant imaginal discs of Drosophila. The results show a causal relationship between the orientation of cell divisions in the imaginal disc and the adult morphology of the corresponding organs, indicating a key role for OCD in organ-shape definition. In addition, a subset of planar cell polarity genes was found to be required for the proper orientation of cell division during organ development (Baena-López, 2005).

Drosophila imaginal discs are a classical model system for studying general mechanisms involved in the control of organ growth and patterning. The imaginal discs are epithelial structures that originate from the embryonic ectoderm, and, after a period of cell proliferation during the larval stages, give rise to most adult organs. The wing disc is divided into lineage units known as compartments. The boundaries between compartments play a key role in the control of wing disc growth and patterning. Analysis of mitotic recombination clones in animals and plants allows for tracing the descendants of single marked cells during development. These experiments have shown a clear correlation between the shape of the clones and adult morphology of the organ where the clones are studied, i.e., clone growth defines organ shape. Most clones in the wing blade are very elongated and grow along the proximal-distal (P/D) axis of the wing, perpendicular to the D/V border. In contrast, clones within the wing margin grow along the D/V border. This study has considered the dorsal-ventral (D/V) boundary as a reference to measure the orientation of cell divisions during wing disc development. A striking relationship was observed between the shape of the clones and the orientation of cell divisions. Thus, the majority of cells divide along the proximal-distal (P/D) axis of the wing blade; 59.4% (n = 549) of mitoses form angles higher than 55° with respect to the D/V boundary, while only 13.8% are lower than 35°. In contrast, in the wing margin, most cells divide nearly parallel to the D/V boundary, forming angles lower than 35° (71.4%; n = 70). Furthermore, the characteristic shape of each intervein region is also reflected in the cell-orientation patterns. Thus, 65.8% (n = 116) of the mitotic figures studied in intervein regions C and D, where clones are very elongated and grow perpendicular to the wing margin, are nearly perpendicular to the D/V border, whereas in regions A and E, where clones are wider and grow parallel to the D/V border, only 50.3% of mitoses (n = 81) have this orientation (Baena-López, 2005).

Early-induced clones also show an elongated shape along the P/D axis of the wing. Accordingly, most of the cell divisions in the second instar wing discs appear with a P/D orientation. In everted wings in pupae, most cells also divide preferentially along the P/D axis, indicating that the correlation between the orientation of cell divisions and the shape of the clones is maintained throughout development (Baena-López, 2005).

Interestingly, the orientation of postmitotic daughter cells, analyzed in clones of two cells, conserves the positions determined by the angle of the OCD. The orientation of the first cell division tends to be maintained in subsequent divisions, since it is observed that 57% (n = 121) of clones of four cells form straight lines of one cell width. Although these results suggest that the cell relocation plays a minor role to define the clone shape, it cannot be ruled out that this process might refine clone shapes and therefore organ shape. Finally, it is suggested that the width of the clones mainly depends on the general probability of OCD (Baena-López, 2005).

To evaluate the general requirement of OCD in organogenesis, the patterns of mitotic orientation were examined in other wing disc regions. The thorax shows an isodiametric morphology, and mitotic recombination clones grow isodiametrically. Accordingly, no preferential orientation of the planar axes of cell divisions was found. In the peripodial epithelium, it was also observed that the orientation of cell divisions define the characteristic shape of the clones. As in the wing blade, postmitotic cells show the same orientation of previous cell divisions (Baena-López, 2005).

The general requirement of the OCD in the definition of wing disc morphology led to a study of the contribution of this process to the development of other organs. The pattern of cell-division orientations was examined during eye disc development. The dorsoventral midline of the Drosophila eye is known as the equator and defines a line of mirror-image symmetry, with ommatidia on each side having opposite chirality. This clonal boundary also plays an important role in the patterning and growth of the eye. Clones in the eye grow symmetrically oblique with respect to the equator. During eye development, an indentation known as the morphogenetic furrow (MF) marks the front of a wave of differentiation that sweeps from posterior to anterior across the disc. The orientation of cell divisions was monitored with respect to the MF in the region anterior to the furrow. The orientation of mitotic spindles in most mitotic figures in the ventral half of the eye discs form angles between 40° and 60°, whereas in the dorsal half, they form angles between 120° and 140°. In the eye disc, the postmitotic cell allocation again reflects the orientation of mitosis. Attempts were made to measure the orientation of cell divisions in a tube-shaped organ like the leg, but the highly folded epithelial organization of these discs prevented the reaching of any clear conclusion. However, clones in the adult legs appear narrow, running proximodistally over several joints. The results indicate that the OCD may be a general mechanism to generate shape during organogenesis (Baena-López, 2005).

One interesting group of genes involved in the orientation of cell divisions during Zebrafish gastrulation and growth of shoot apex in Arabidopsis is composed of planar cell polarity (PCP) genes. The function of these genes is evolutionary conserved to define the polarity and positional information of the cells within an epithelium. Mutations in the Drosophila PCP genes dachsous (ds) and fat (ft) cause changes in the shape of different organs: wings and eyes are rounder than the wild-type, whereas legs are wider and shorter. The activity of these genes is required upstream of other PCP genes that do not affect adult organ shape such as the core PCP genes strabismus (stbm), prickled (pk), and flamingo (fmi) or the effector PCP gene multiple wing hair (mwh). Whether the abnormal organ shapes, observed in ds and ft mutant backgrounds, are associated with changes in the clone shape and the orientation of cell division was analyzed (Baena-López, 2005).

Mitotic recombination clones in ds mutant wing discs show rounded shapes, losing the elongated shape of wild-type clones. Interestingly, loss-of-function clones for ds or ft fail to adopt their typical enlarged shape in the wing blade or in the eye, showing a rounder shape than wild-type clones. In contrast to control wing discs and wild-type clones, it was observed that the orientation of cell divisions is randomized in ds mutant cells (wing discs or ds mutant clones). Accordingly, postmitotic cells do not show any preferential orientation. The rounded shape and disturbed OCD is also observed in clones of ds- or ft-expressing cells and in adult wings where ds or ft is ectopically expressed using the UAS/G4 system. Similar results are observed in the eye imaginal discs. These results suggest that the polarization cues mediated by PCP genes are required for the control of the orientation of cell division and, therefore, are involved in organ shape definition. These findings show that some of the genes required for PCP are acting in early stages of the development, suggesting that the base for PCP may be established during larval development. The conserved function of PCP genes controlling OCD in different organisms suggests that this process might be an evolutionary mechanism in organ shape definition. Although the mechanistic details underlying the new functions of PCP genes are unknown, it is likely that the asymmetric cell distribution of a classical core of PCP proteins is not required. Finally, similar phenotypes were observed in both the lack and excess of function for ds and ft, suggesting that the default organ shape is circular when the cell polarization is disrupted (Baena-López, 2005).

To conclude, it is proposed that the shape of organs can be accounted for by the oriented pattern of cell divisions rather than postmitotic relocation of proliferating cells. However, how the preferential orientation is topologically determined is poorly understood. It is speculated that the orientation of cell divisions in different territories during development may result from the integration of signals coming from restriction borders and territorial local cell interactions. The existence of clusters of cells, synchronized in the same stage of the cell cycle, support the idea of a local control of cell proliferation. It has been found that the orientations of cell divisions in these clusters tend to be highly aligned, suggesting also a local control of OCD superimposed to more general and long-range signals. Although the signals exchanged between neighboring cells to determine local OCD are still unknown, it is likely that a subset of PCP genes could be involved in this process (Baena-López, 2005).

Separating the adhesive and signaling functions of the Fat and Dachsous protocadherins

The protocadherins Fat (Ft) and Dachsous (Ds) are required for several processes in the development of Drosophila, including controlling growth of imaginal discs, planar cell polarity (PCP) and the proximodistal patterning of appendages. Ft and Ds bind in a preferentially heterophilic fashion, and Ds is expressed in distinct patterns along the axes of polarity. It has thus been suggested that Ft and Ds serve not as adhesion molecules, but as receptor and ligand in a poorly understood signaling pathway. To test this hypothesis, a structure-function analysis of Ft and Ds was performed, separating their adhesive and signaling functions. It was found that the extracellular domain of Ft is not required for its activity in growth, PCP and proximodistal patterning. Thus, ligand binding is not necessary for Ft activity. By contrast, the extracellular domain of Ds is necessary and sufficient to mediate its effects on PCP, consistent with the model that Ds acts as a ligand during PCP. However, evidence is also provided that Ds can regulate growth independently of Ft, and that the intracellular domain of Ds can affect proximodistal patterning, both suggestive of functions independent of binding Ft. Finally, it is shown that ft mutants or a dominant-negative Ft construct can affect disc growth without changes in the expression of wingless and Wingless target genes (Matakatsu, 2006).

Chief amongst the findings of this study is that Ft activity is not simply a byproduct of changes in cell-cell adhesion. The FtΔECD construct lacks almost the entire extracellular domain and cannot bind or stabilize Ds in vitro or in vivo. Nonetheless, it can rescue the lethality, overgrowth and PCP defects of ft alleles that should be null for any adhesive or receptor function, and in a wild-type background can disrupt proximodistal patterning. This suggests that the intracellular domain of Ft can act in the absence of binding between endogenous Ft and Ds, or indeed between Ft and any other extracellular ligand, as long as sufficient levels are expressed (Matakatsu, 2006).

Conversely, it was found that a form of Ft lacking the intracellular domain (FtΔICD) failed to rescue overgrowth in ft mutants. In fact, this form acts as a strong dominant negative, inducing overgrowth of wild-type and ft mutant imaginal discs. This occurs despite the ability of FtΔICD to stabilize endogenous cell surface Ds and Ft, raising the possibility that FtΔICD binds to Ds and Ft is a way that blocks their activities. The possibility that FtΔICD alters the activity of some additional, unknown player cannot be ruled out. Although lethality prevents determining whether FtΔICD can rescue ft mutant PCP defects, expression of FtΔICD in wild-type wings also disrupts PCP. These PCP defects are weaker than those observed in ft mutants, suggesting that FtΔICD might have stronger effects on growth control than PCP (Matakatsu, 2006).

In contrast to Ft, the extracellular domain of Ds is sufficient for its effects on PCP. The DsΔICD construct lacks almost the entire intracellular domain, but nonetheless can rescue the PCP defects of strong ds mutants and disrupt PCP in wild-type wings. The DsΔECD construct, however, cannot bind or stabilize Ft and cannot rescue ds mutant PCP defects or influence PCP in wild-type wings. The results thus support the hypothesis that in PCP Ds acts chiefly as a ligand for Ft, modulating its activity (Matakatsu, 2006).

Nonetheless, the possibility that the intracellular domain of Ds has some PCP activity within the context of the whole protein cannot be ruled out, and the conservation of large regions of the Ds intracellular domain in its vertebrate homologs dachsous 1 and dachsous 2 suggests that Ds may have activity beyond that of a ligand. Thus, it is intriguing that expression of DsΔECD can disrupt another ds-sensitive phenotype, crossvein spacing in wild-type wings. Since crossvein spacing defects can result from either gains or losses in Ds or Ft function, it is possible that this defect is caused by disrupting the function of endogenous Ds, and thus the ability of that Ds to signal via Ft. However, DsΔECD did not cause any obvious change in the levels of endogenous Ds. Moreover, loss of Ds normally causes visible destabilization of cell surface Ft, and no changes were seen in Ft levels in cells misexpressing DsΔECD (Matakatsu, 2006). ds mutations can also enhance the overgrowth observed in mutants that lack the intracellular domain of Ft, indicating that in overgrowth, Ds activity is not completely dependent on regulating the activity of the intracellular domain of Ft. In this respect, overgrowth differs from PCP; ft mutants and ds ft double mutants produce identical PCP phenotypes. The result could be explained if Ds regulates growth via its intracellular domain. Alternatively, Ds may be acting as an extracellular ligand for a binding partner other than Ft (Matakatsu, 2006).

The results support the hypothesis the Ft signals via its intracellular domain in growth control, PCP and proximodistal patterning. Similarly, it is likely that the intracellular domain of Ds contributes to proximodistal patterning and perhaps growth control. The conservation of long stretches of the intracellular domain of Ft and Ds in the vertebrate homologs Fat4, dachsous 1 and dachsous 2 also suggests that there is conserved binding to intracellular factors (Matakatsu, 2006).

There are no known binding partners for the intracellular domain of Ds or dachsous-like proteins. The intracellular domain of Drosophila Ft also lacks the ENA-VASP binding sites that mediate at least some of the function of vertebrate Fat1 in vitro. The intracellular domain of Drosophila Ft can bind the atrophin Grunge, and genetic evidence suggests a link between Grunge and PCP. However, it is not yet clear if Grunge acts downstream of Ft, nor is it clear how atrophins, which act as transcriptional co-repressors, could polarize cells. grunge mutants also do not apparently reproduce the effects of ft mutants on disc growth or on wg expression in the prospective wing hinge (Matakatsu, 2006).

Some evidence suggests that Ds and Ft regulate growth and patterning by altering either the expression of wg in the prospective wing hinge or the response to Wg signaling. However, the current results make it unlikely that this can explain all but a small part of the overgrowth phenotype. The overgrowth induced by ft mutations or FtΔICD occurs without any consistent change in the expression of Wg target genes Dll or Vg, or in the expression of wg. Moreover, FtΔICD induced overgrowth in the entire wing disc, but whereas increased Wg signaling can induce overgrowth in the hinge, in the prospective wing blade Wg signaling reduces growth. The results are consistent with the failure of mutants in the Wg signaling pathway to modify the ft overgrowth phenotype (Matakatsu, 2006).

A recent study has suggested a possible link between overgrowth and Ras signaling; mild reductions in Ras function that have little effect on the growth of wild-type cells can block the overgrowth observed in ft mutant clones. It remains to be seen whether Ft can actually affect Ras signaling, or whether this represents the convergence of the two pathways on a shared target (Matakatsu, 2006).

Because Ds is expressed in an apparently graded fashion along the axes of polarity, it was suggested that Ds provides a global cue that orients PCP in the eye, wing and abdomen. But whereas patterned Ds misexpression is sufficient to reorient PCP, and patterned Ds expression does appear to be necessary for normal PCP in the eye, in the wing uniform Ds expression is able to rescue most of the ds mutant PCP defects. This suggests that most of the PCP defects in ds mutant wings are caused, not by a change in the spatial regulation of Ds-Ft signaling, but rather by the loss of a basal level of signaling required for the proper activity of some other polarizing cue. These results left open the possibility that Ft activity is being spatially regulated by an extracellular ligand other than Ds. However, this study shows that ft mutant PCP defects can be substantially rescued by uniform expression of FtΔECD, a form of Ft that cannot bind Ds, or probably any other ligand (Matakatsu, 2006).

There is, however, a region in the proximal wing where PCP defects cannot be rescued with uniform expression of either Ds, Ft, or FtΔECD. This is also the region of the wing where there is a boundary or sharp gradient between proximal regions with high and distal regions with low ds expression. Thus, it remains possible that Ds and Ft activities are permissive in much of the wing but, in the proximal wing, spatially instructive. The different sensitivities of different regions to changes in Ds and Ft may reflect localized differences in the strength of other partially redundant polarizing cues (Matakatsu, 2006).

REFERENCES

Alagramam, K. N., et al. (2001). The mouse Ames waltzer hearing-loss mutant is caused by mutation of Pcdh15, a novel protocadherin gene. Nat. Genet. 27: 99-102. 11138007

Baena-López, L. A., Baonza, A. and GarcÌa-Bellido, A. (2005). The orientation of cell divisions determines the shape of Drosophila organs. Curr. Bio. 15: 1640-1644. 16169485

Bryant, P. J., Huettner, B. H., Held, L. I., Ryerse, J. and Szidonya, J. (