fat: Biological Overview | Evolutionary Homologs | Regulation | Developmental Biology | Effects of Mutation | References
Gene name - fat

Synonyms -

Cytological map position - 24D8

Function - transmembrane receptor

Keywords - cell cycle, planar polarity, tumor suppressor, Fat signaling pathway

Symbol - ft

FlyBase ID: FBgn0001075

Genetic map position - 2-12.0

Classification - cadherin domain, thrombospondin N-terminal-like domains

Cellular location - surface

NCBI links: Precomputed BLAST | Entrez Gene | UniGene |
Recent literature
Bosveld, F., Guirao, B., Wang, Z., Riviere, M., Bonnet, I., Graner, F. and Bellaiche, Y. (2016). Modulation of junction tension by tumor-suppressors and proto-oncogenes regulates cell-cell contacts. Development [Epub ahead of print]. PubMed ID: 26811379
Tumor-suppressor and proto-oncogenes play critical roles in tissue proliferation. Furthermore, deregulation of their functions is deleterious to tissue architecture and can result in the sorting of somatic rounded clones minimizing their contact with surrounding wild-type (wt) cells. Defects in somatic clones shape correlate with defects in proliferation, cell affinity, cell-cell adhesion, oriented cell division and cortical elasticity. Combining genetics, live-imaging, laser ablation and computer simulations, whether distinct or similar mechanisms can account for the common role of tumor-suppressor and proto-oncogenes in cell-cell contact regulation was analyzed. In Drosophila epithelia, Fat (Ft) and Dachsous (Ds) tumor-suppressors regulate cell proliferation, tissue morphogenesis, planar cell polarity and junction tension. By analyzing the time evolution of ft mutant cells and clones, this study shows that ft clones reduce their cell-cell contact with surrounding wt tissue in the absence of concomitant cell divisions and over-proliferation. This contact reduction depends on opposite changes of junction tensions in the clone bulk and its boundary with neighboring wt tissue. More generally, either clone bulk or boundary junction tensions is modulated by the activation of Yorkie, Myc and Ras yielding similar contact reductions with wt cells. Together these data highlight mechanical roles for proto-oncogene and tumor-suppressor pathways in cell-cell interactions.
Zhang, Y., Wang, X., Matakatsu, H., Fehon, R. and Blair, S.S. (2016). The novel SH3 domain protein Dlish/CG10933 mediates fat signaling in Drosophila by binding and regulating Dachs. Elife 5. PubMed ID: 27692068
Much of the Hippo and planar cell polarity (PCP) signaling mediated by the Drosophila protocadherin Fat depends on its ability to change the subcellular localization, levels and activity of the unconventional myosin Dachs. To better understand this process, this study performed a structure-function analysis of Dachs, and used this to identify a novel and important mediator of Fat and Dachs activities, a Dachs-binding SH3 protein that was named Dlish. It was found that Dlish is regulated by Fat and Dachs, that Dlish also binds Fat and the Dachs regulator Approximated, and that Dlish is required for Dachs localization, levels and activity in both wild type and fat mutant tissue. Evidence supports dual roles for Dlish. Dlish tethers Dachs to the subapical cell cortex, an effect partly mediated by the palmitoyltransferase Approximated under the control of Fat. Conversely, Dlish promotes the Fat-mediated degradation of Dachs.

Nakamura, A., Tanaka, R., Morishita, K., Yoshida, H., Higuchi, Y., Takashima, H. and Yamaguchi, M. (2017). Neuron-specific knockdown of the Drosophila fat induces reduction of life span, deficient locomotive ability, shortening of motoneuron terminal branches and defects in axonal targeting. Genes Cells [Epub ahead of print]. PubMed ID: 28488382
Mutations in FAT4 gene, one of the human FAT family genes, have been identified in Van Maldergem syndrome (VMS) and Hennekam lymphangiectasia-lymphedema syndrome (HS). The FAT4 gene encodes a large protein with extracellular cadherin repeats, EGF-like domains and Laminin G-like domains. FAT4 plays a role in tumor suppression and planar cell polarity. This study knocked down Drosophila fat in the nervous system, resulting in shortened life span and a defect in locomotion. Defects in synapse structure at neuromuscular junction and aberrations in a axonal targeting of photoreceptor neurons were also observed. The results indicate that Drosophila fat plays an essential role in formation and/or maintenance of neurons. Both VMS and HS show mental retardation and neuronal defects. It is therefore considered that these two rare human diseases could possibly be caused by the defect in FAT4 function in neuronal cells.
Wortman, J. C., Nahmad, M., Zhang, P. C., Lander, A. D. and Yu, C. C. (2017). Expanding signaling-molecule wavefront model of cell polarization in the Drosophila wing primordium. PLoS Comput Biol 13(7): e1005610. PubMed ID: 28671940
Cells throughout the wing primordium typically show subcellular localization of the unconventional myosin Dachs on the distal side of cells (nearest the center of the disc). Dachs localization depends on the spatial distribution of bonds between the protocadherins Fat (Ft) and Dachsous (Ds), which form heterodimers between adjacent cells; and the Golgi kinase Four-jointed (Fj), which affects the binding affinities of Ft and Ds. The Fj concentration forms a linear gradient while the Ds concentration is roughly uniform throughout most of the wing pouch with a steep transition region that propagates from the center to the edge of the pouch during the third larval instar. It is unclear how the polarization is affected by cell division and the expanding Ds transition region, both of which can alter the distribution of Ft-Ds heterodimers around the cell periphery. A computational model was developed to address these questions. In this model, the binding affinity of Ft and Ds depends on phosphorylation by Fj. It is assumed that the asymmetry of the Ft-Ds bond distribution around the cell periphery defines the polarization, with greater asymmetry promoting cell proliferation. The model predicts that this asymmetry is greatest in the radially-expanding transition region that leaves polarized cells in its wake. These cells naturally retain their bond distribution asymmetry after division by rapidly replenishing Ft-Ds bonds at new cell-cell interfaces. Thus it is predicted that the distal localization of Dachs in cells throughout the pouch requires the movement of the Ds transition region and the simple presence, rather than any specific spatial pattern, of Fj.

The fat gene negatively controls cell proliferation in a cell autonomous manner. The Fat protein (with 5,147 amino acids) contains four major regions. Beginning at 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). Several cell behavior parameters of mutant alleles of fat ( ft) have been studied in Drosophila imaginal wing disc development. Mutant imaginal discs continue growing in larvae delayed in pupariation and can reach sizes several times those of wild-type. Their growth is, however, basically allometric. Homozygous ft cells grow faster than their twin cells in clones and generate larger territories, albeit delimited by normal clonal restrictions. Moreover, ft cells in clones tend to grow towards the wing proximal regions. These behaviors can be related to failures in cell adhesiveness and cell recognition. Double mutant combinations with alleles of other genes, e.g., of the Epidermal growth factor receptor pathway, modify ft clonal phenotypes, indicating that adhesiveness is modulated by intercellular signaling. In addition, mutant ft cells show smaller cell sizes during proliferation and abnormal cuticular differentiation; this is reflective of cell membrane and cytoskeleton anomalies, that are not modulated by the Egfr pathway (Garoia, 2000).

Fat also plays an important role in planar polarity. This phenomenon is evidenced by the coordinated orientation of ommatidia in the Drosophila eye. Planar polarity requires that the R3 photoreceptor precursor of each ommatidium has a higher level of Frizzled signaling than its neighboring R4 precursor. Two cadherin superfamily members, Fat and Dachsous, and the transmembrane/secreted protein Four-jointed play important roles in this process. The data support a model in which the bias of Frizzled signaling between the R3/R4 precursors results from higher Fat function in the precursor cell closer to the equator -- the cell that becomes R3. Evidence is also provided that positional information regulating Fat action is provided by graded expression of Dachsous across the eye and the action of Four-jointed, which is expressed in an opposing expression gradient and appears to modulate Dachsous function. It is suggested that the presence of relatively higher Ds function in the polar cell could result in a difference in Ft function between the R3/R4 precursors by either inhibiting Ft function in a cell-autonomous fashion or by stimulating Ft function in the equatorial cell. The difference in Ft function between the precursor cells biases Fz signaling so that the equatorial cell has higher Fz activity (Yang, 2002).

The first indication that Ft functions during PCP signaling in the eye came from examining flies homozygous for the weak, viable ft allele, ft1. A small fraction of ommatidia with reversed dorsal-ventral (d-v) polarity were consistently observed. To examine the effects of stronger ft alleles on ommatidial polarity, FRT-mediated mitotic recombination was used to generate clones of cells homozygous for the lethal ft alleles, ftl(2)fd or ftGr-V. Ommatidia located within the ft mutant tissue were constructed normally, but they frequently adopted the reversed d-v polarity form (~40% for ftl(2)fd, ~50% for ftGr-V). In addition to polarity reversals within the ft mutant tissues, occasional reversals of polarity in wild-type ommatidia bordering the polar side of the mutant tissue were also observed. This effect resembles the nonautonomous phenotypes previously reported for fz mutant clones in the eye (Yang, 2002).

The presence of a randomized pattern of d and v type ommatidia within ft mutant tissue suggests that Ft is required to correctly bias R3/R4 specification. To confirm that the initial pattern of R3 and R4 specification is randomized in ft mutant ommatidia, the expression pattern of an R4-specific marker, E(spl)mdelta0.5, was examined in ftl(2)fd and ftGr-V clones. This marker consists of a portion of the enhancer region of the E(spl) gene, a transcriptional target of N activation, fused to a ß-galactosidase (ß-gal) reporter. In wild-type ommatidia, this marker is initially expressed in both R3/R4 precursor cells, but then becomes stronger in the polar cell since this cell is specified as R4. Within the ft ommatidial clusters, the R4-specific marker is still strongly expressed in only one member of the R3/R4 pair. However, the cell expressing the R4 marker frequently occupies the position normally taken by the equatorial cell, indicating that the pattern of R3/R4 cell fate specification is reversed (Yang, 2002).

Because the presence of higher Fz signaling in the equatorial member of the R3/R4 precursor pair is a crucial determinant of R3/R4 specification, these results suggest that Ft may function in the placement or interpretation of positional cues controlling the equatorial/polar bias of Fz signaling. This model predicts that while Fz signaling should still occur in the absence of Ft, the pattern of Fz activation within an R3/R4 pair should be randomized with respect to the equator. Consistent with this prediction, two important differences between the ft and fz mutant phenotypes indicate that Fz signaling remains intact in ft ommatidia. (1) Previous studies have shown that many fz ommatidia are incorrectly formed and fail to have distinctly specified R3 and R4 cells. This phenotype indicates that the processes preventing both R3 and R4 precursor cells from adopting the same fate depend on Fz function and are inefficient when only Dl/N-mediated lateral inhibition is used to specify R3 and R4. (2) fz ommatidia frequently either fail to rotate or rotate incorrectly even when R3 and R4 cells are specified. In contrast, most ft ommatida contain uniquely specified R3 and R4 cells and rotate in the proper direction for their pattern of R3/R4 specification. These differences between the fz and ft phenotypes suggest that Fz signaling remains functional in the absence of Ft function (Yang, 2002).

To show that the absence of Ft function causes Fz signaling to occur in a randomly biased pattern within R3/R4 precursor pairs, eye discs containing ft mutant clones were immunostained for Flamingo (Fmi)/Starry Night (Stan) protein. Previous work has suggested that Fmi and Fz function together in a signaling complex at the proximal/distal (p/d) junctions between wing cells. Furthermore, the accumulation of Fmi at a p/d cell-cell boundary depends on the presence of differences in Fz signaling levels between the two cells. To first show that Fmi functions during PCP signaling in the eye, clones of cells homozygous for a strong loss-of-function fmi allele (fmiE59) were examined. Dramatic polarity defects were found that resembled those seen in fz ommatidia. The defects included aberrant rotation and a lack of distinct R3/R4 fates (Yang, 2002).

Immunostaining eye imaginal discs for Fmi has revealed a dynamic pattern of localization within developing ommatidia. In young clusters prior to ommatidial rotation, Fmi is localized to the cell membranes of both of the R3/R4 precursors. However, as the clusters initiate rotation, Fmi shows asymmetric localization and is eventually concentrated at the cell-cell junction between the R3 and the R4 precursors and at the polar border of the R4 precursor, where R4 abuts a cell of undetermined fate. To confirm that this asymmetric localization depends on Fz signaling, Fmi localization was examined in eye discs from fz mutant animals (fzH51/fzKD4a). In the absence of functional Fz, Fmi was uniformly localized to the cell membranes of R3/R4 precursors without obvious signs of asymmetry. Further support for the use of asymmetric Fmi localization as a marker for Fz signaling came from observations that GFP-tagged forms of Fz and Dsh, whose localization to the p/d boundaries in wing cells are also Fz dependent and coincide with that of Fmi, similarly colocalize with Fmi within ommatidial clusters. Having established that Fmi localization can be used as a marker for Fz signaling, Fmi localization was examined in ft clones. Asymmetric localization of Fmi within individual clusters is readily apparent in ft tissue. However, the pattern of asymmetric localization is often reversed and correlates with the pattern of R3/R4 specification as shown by the direction of rotation of those clusters (Yang, 2002).

Intense Fmi accumulation is often apparent anterior to the morphogenetic furrow along the ft clonal border. Since differences in Fz levels between adjacent wing cells can recruit Fmi to cell-cell boundaries, this result suggests that differences in the levels of Ft function between adjacent cells can result in differences in their levels of Fz signaling. Thus, these experiments raise the possibility that asymmetric Fz signaling between the R3/R4 precursor cells occurs because one precursor cell has more Ft function than does its neighbor. For example, Ft function might be consistently higher in the equatorial precursor cell and act to either enhance Fz signaling in that cell or to inhibit Fz signaling in the neighboring polar precursor cell. This model makes two predictions. (1) Removal of Ft function only from the precursor cell that normally has higher Ft function should lead to reversal, rather than randomization, of the pattern of R3/R4 specification. In contrast, removal of Ft from the precursor with lower Ft function should have little effect on R3/R4 specification. (2) The ability of differences in Ft signaling within a precursor pair to determine the pattern of R3/R4 specification should depend on the presence of Fz activity (Yang, 2002).

The effect of removing Ft from one member of an R3/R4 precursor pair was assayed using clones of homozygous ft cells generated in a heterozygous background. Because there are no strict lineage relationships among ommatidial cells, the ommatidia along the clonal border were composed of random combinations of ft-/+ and ft-/- cells. Among these combinations were mosaic ommatidia in which only one member of the R3/R4 precursor pair possessed functional Ft. Because ft ommatidia rotate in the direction corresponding to their pattern of R3/R4 specification, the fate of the equatorial and the polar precursors in such R3/R4 mosaic ommatidia could be determined. In ommatidia that adopted the correct polarity for their position in the eye, R3 and R4 were derived from the equatorial and polar precursor cells, respectively, while in reversed polarity ommatidia, the R3 was derived from the polar precursor cell and R4 from the equatorial cell. Analysis of 166 R3/R4 mosaic ommatidia revealed that in all 71 examples where Ft function was absent from the polar R3/4 precursor cell, the equatorial cell became R3 and the resulting ommatidia displayed normal polarity. In contrast, loss of Ft function from the equatorial cell resulted in the polar cell becoming R3 and a reversal of ommatidial polarity in 88 of 95 examples. These results demonstrate that Ft normally acts in the equatorial cell to promote the correct pattern of R3/R4 specification. These data also indicate that in ommatidia lacking Ft function only in the equatorial cell, the Ft present in the polar cell actively directs the reversal of the pattern of R3/R4 specification. This is apparent in the observation that removal of Ft function from only the equatorial cell usually reverses the R3/R4 specification pattern. In contrast, loss of Ft function from both precursor cells leads to a random pattern of R3/R4 specification. The ability of strong differences in Ft function to override the normal positional cues instructing R3/R4 specification is consistent with the idea that the normal equatorial/polar bias in Fz signaling results at least in part from the presence of higher Ft function in the equatorial precursor cell (Yang, 2002).

To demonstrate that Fz is required for differences in Ft levels between the R3/R4 precursor cells to dictate the pattern of R3/R4 specification, ft clones were generated in the eyes of fzKD4a/fzH51 animals. Analysis of 23 R3/R4 ft mosaic ommatidia has revealed that differences in Ft function failed to determine the pattern of R3/R4 development in the absence of Fz. ft-/- R3/R4 precursor cells in the fz background adopted either the R3 or R4 fate with roughly equal frequency (39% specified as R4, 61% as R3), in marked contrast to ft-/- cells in mosaic ommatidia generated in a fz+ background (95% specified as R4, 5% as R3) (Yang, 2002).

These data support a model in which a higher Fz signaling in the equatorial member of each R3/R4 precursor pair results from the presence of higher Ft function in that cell. The next experiments sought to discover the mechanisms that control the level of Ft function within each R3/R4 precursor pair. One possibility is that Ft could be expressed in a graded fashion from the equator toward the poles. To investigate this possibility, antibodies were raised against the intracellular domain of Ft and wild-type eye discs were stained. It was found that Ft is not expressed in a detectable graded fashion. Instead, Ft appears to be uniformly distributed along most cell membranes and overlaps with Dsh-GFP prior to the asymmetric localization of Dsh-GFP in the region near the morphogenetic furrow where R3/R4 cell fate decisions are being made (Yang, 2002).

The lack of a graded Ft expression suggested that Ft might be regulated by proteins that are themselves expressed in gradients in the eye disc. Two previous findings prompted a test of the role of dachsous (ds, which encodes a Fat-like transmembrane protein containing 27 cadherin repeats in its extracellular domain), as a Ft regulator: (1) loss of Ds function in the wing causes planar polarity defects; (2) removal of a single copy of ds suppresses defects caused by a dominant ft mutation, suggesting that Ds might act in conjunction with Ft. To determine whether Ds provides positional information regulating Ft function in the eye, ds expression was examined in the eye imaginal disc using a ds-lacZ enhancer trap (a ß-gal reporter gene inserted in the first intron of ds) that faithfully reproduces ds expression during third instar larval development. Interestingly, ds-lacZ is expressed in a graded pattern that is high at the two poles and low at the equator. In order to confirm that this gradient of ds transcription results in a gradient of Ds protein expression, antibodies were raised against the intracellular domain of Ds and used to stain eye discs. In the region near the morphogenetic furrow where the R3 and R4 fates are being specified, graded expression of Ds from the poles was readily apparent. In older ommatidia in the posterior region of the disc, Ds, like Ft, accumulates in a subset of cells surrounding each ommatidium (Yang, 2002).

If Ds is an important regulator of Ft, then the absence of Ds should lead to randomized d-v polarity. This prediction was examined using animals homozygous for the ds loss-of-function mutation dsUA071. While Ds function is important for normal viability, a few homozygous dsUA071 mutant animals survive for a few days after eclosion. Similar to ft ommatidia, the ds ommatidia randomly adopted either d or v polarity. Similar results were observed in animals transheterozygous for dsUA071 and another strong ds allele (ds38K) and in marked clones of homozygous dsUA071 cells (~40% polarity reversals). The examination of ds mutant clones also showed the presence of nonautonomous effects on the polarity of neighboring wild-type tissue along the equatorial border of the clone (Yang, 2002).

When the pattern of R3/R4 specification and Fz signaling in ds eye discs was examined by staining for either the E(spl)mdelta0.5 marker or Fmi, the results were very similar to those described above for ft mutant clones. In both cases, the mutant ommatidia exhibited polarized Fmi localization and R4 reporter expression, but the direction of polarization was randomized. Thus, Ds, like Ft, plays an essential role in the establishment of properly biased Fz signaling during R3/R4 specification (Yang, 2002).

The requirement for Ds during the establishment of ommatidial polarity and the gradient of ds expression suggested that higher ds expression in the polar member of the R3/R4 precursor pair might contribute to the normal pattern of R3/R4 specification by modulating Ft function. For example, the presence of higher Ds in the polar precursor cell might either inhibit Ft function within the polar cell or else promote Ft function in the neighboring equatorial precursor cell. This proposal makes several predictions: (1) loss of Ds function from the polar cell, which would reverse the relative levels of Ds within an R3/R4 precursor pair, should lead to reversals in R3/R4 specification pattern and reversals of ommatidial polarity; (2) in contrast, loss of Ds from the equatorial cell, which does not change the direction of the ds gradient within an ommatidium, should have little effect on polarity; (3) loss of Ds from the polar cell should have no effect if that cell also lacks functional Ft (Yang, 2002).

To analyze the effects of a lack of Ds function in one member of an R3/R4 precursor pair, homozygous dsUA071 clones were generated in heterozygous animals and R3/R4 mosaic ommatidia were examined. The loss of Ds function from the equatorial R3/R4 precursor cell had only a mild effect on ommatidial polarity (13% polarity reversals in 127 examples. In contrast, loss of Ds function from the polar cell led to polarity reversals in 43% of 98 examples. These data demonstrate that Ds acts primarily in the polar precursor cell and are consistent with the idea that the graded ds expression contributes to the normal pattern of R3/R4 specification by providing a higher Ds level to the polar cell of each R3/R4 precursor pair. However, the observation that differences in Ds function between the R3/R4 precursor cells are less effective than Ft differences at directing the pattern of R3/R4 specification suggests that, in addition to its primary role in the polar precursor cell, Ds may also play a role in regulating Fz signaling from the equatorial cell (Yang, 2002).

In order to test whether Ds regulates Fz signaling by modulating Ft function, the genetic interaction of ft and ds in specifying R3/R4 cell fates was examined. Since removal of Ds from the polar precursor cell frequently causes this cell to become R3 rather than R4 while the absence of Ft from an R3/R4 precursor cell leads to its specification as R4, the effects of removing both Ds and Ft from the polar cell were examined using marked clones of dsUA071;ftGr-V double mutant cells. Polarity reversals occurred in only 2.5% (2 out of 80 examples) of the R3/R4 mosaic ommatidia in which the polar precursor cell lacked Ds and Ft. This result contrasts with the 43% polarity reversals observed when only Ds function was lost from the polar cell. Indeed, the effect of removing both Ds and Ft functions was remarkably similar to that of removing Ft alone. In each case, the mutant precursor cell was preferentially specified as R4 (80% and 98%, respectively). Thus, the absence of Ft function from one R3/R4 precursor cell determines the pattern of R3/R4 specification in a manner that is largely independent of input from Ds. These results suggest that Ds acts upstream of Ft and are consistent with the idea that the presence of higher Ds levels in the polar R3/R4 precursor directs correct ommatidial polarity by ensuring that Ft activity is higher in the equatorial precursor cell (Yang, 2002).

To further explore how Ds might act in regulating Ft, it was asked whether Ft protein level is altered in ds mutant clones. Staining for Ft was carried out in ds clones and it was found that the level of Ft is increased slightly within the clones. This result suggests that one mode of Ds regulation might be to antagonize Ft protein expression or stability (Yang, 2002).

Previous studies have demonstrated that four-jointed (fj), which encodes a type II transmembrane/secreted protein, is expressed in a graded fashion from the equator toward the poles of the eye imaginal disc. Fj has been implicated in the regulation of ommatidial polarity based on the observation that reversals of ommatidial polarity occur along the polar border of fj mutant clones. However, fj mutant flies and the interior of fj clones only rarely display polarity defects. This has suggested that much of Fj action during ommatidial polarity formation may be redundant with other signals. The role of Fj was investigated further by asking whether Fj might regulate Ds or Ft function. To determine whether Fj acts predominantly in one member of the R3/R4 precursor pair, strong loss-of-function fj mutant (fjN7) clones were generated, and R3/R4 mosaic ommatidia were examined. The loss of Fj function from the polar precursor cell, which expresses lower levels of Fj, had little effect on ommatidial polarity (6% reversals in 70 cases). In contrast, loss of Fj function from only the equatorial precursor cell led to polarity reversals in 71% of the 52 cases. These results indicate that Fj is mainly required in the equatorial R3/R4 precursor cell during the determination of ommatidial polarity. However, when only the equatorial cell lacks functional Fj, the Fj produced by the polar cell frequently is able to reverse the pattern of R3/R4 specification. These data suggest that the Fj gradient may play a role in directing R3/R4 specification by providing more Fj activity to the equatorial precursor cell (Yang, 2002).

The frequent polarity reversals caused by loss of Fj from the equatorial cell contrasts markedly with the paucity of polarity defects in fj mutant animals. A possible explanation for this difference is that Fj might act by modulating Ds function. Thus, loss of Fj from an entire eye would leave a Ds gradient in place to direct ommatidial polarity. In contrast, loss of Fj from only the equatorial precursor may reverse the gradient of Ds function within an R3/R4 pair. If this model were correct, differences in Fj function between the R3/R4 precursor cells should be ineffective at biasing cell fate specification in the absence of Ds. However, differences between the R3/R4 precursor cells in their levels of Ds function should be able to direct ommatidial polarity even in the absence of Fj (Yang, 2002).

The first prediction was tested by examining fj clones generated in dsUA071 flies. Since the lack of a recognizable equator in homozygous ds mutant eyes made it impossible to designate normal versus reversed polarity ommatidia in fjN7 clones, the fj R3/R4 mosaic ommatidia were analyzed by measuring the frequency with which the fj-/- precursor cell became R4. When fj mosaic ommatidia were assayed in ds mutant flies, the fj-/- precursor cell was equally likely to become R3 or R4 (54% as R4, 46% as R3). In contrast, the fj-/- cell became R4 in 84% of cases when fj mosaic clones were generated in a ds+ background. These data show that differences in Fj function can only specify the pattern of R3/R4 specification when Ds is present. The ability of differences in Ds levels to influence R3/R4 specification in the absence of Fj function was examined by generating dsUA071 clones in either wild-type or homozygous fjN7 flies. When ds mutant clones were induced in wild-type background, the ds-/- cell has 68% chance of becoming an R3. Interestingly, when ds clones were induced in fj animals, this frequency was increased to 85%, showing that the lack of Fj not only fails to block the action of Ds but instead enhances the ability of Ds differences to dictate the pattern of R3/R4 specification. These results support a model in which Fj acts upstream of Ds, perhaps as a regulator of Ds function (Yang, 2002).

Since fj and ds are expressed in opposing gradients in the eye, one possible mechanism by which Fj could regulate Ds function would be to inhibit ds transcription. This possibility was investigated by examining ds-lacZ expression in clones of cells that either lack Fj function or overexpress Fj. The lack of detectable effect on ds-lacZ expression in these experiments indicates that Fj does not act by antagonizing ds expression (Yang, 2002).

This analysis supports the idea that positional information controlling Fz signaling during ommatidial development is provided by the opposing gradients of fj and ds expression. The question arises as to how these gradients are established. Previous work has shown that a major determinant of the fj expression gradient is Wg, a secreted Wnt class ligand that negatively regulates fj expression and that is expressed at high levels at the two poles of the eye disc. To test whether the Wg gradient also contributes to the regulation of ds expression, clones of cells in which Wg signaling was either ectopically activated or reduced were examined in animals carrying the ds-lacZ reporter. Ectopic activation was achieved by overexpressing a constitutively activated form of Armadillo (Arm) and resulted in a dramatic increase in ds-lacZ expression. The effects of attenuating Wg signaling were assayed in clones of cells homozygous for the hypomorphic armH8.6 mutation. ds-lacZ expression was severely reduced in these clones. Combined with previous studies of fj-lacZ expression, these data suggest that the ds and fj expression gradients result in large part from the presence of a gradient of Wg signaling that increasingly activates ds and inhibits fj expression near the poles. It is worth emphasizing that the receptor mediating the effects of Wg on fj and ds expression is likely to be another member of the Fz family, perhaps dFrizzled2 (dFz2), rather than Fz itself. This is evident from the observation that fj-lacZ expression is not affected by the loss of Fz function (Yang, 2002).

Genetic epistasis experiments strongly suggest that Fj, Ds, and Ft act together to control PCP signaling. Further support for their placement in a common pathway came from an analysis of fj-lacZ expression. Fj can regulate its own expression. fj overexpression in eye disc clones represses fj-lacZ expression within the clones. When fj-lacZ expression is examined in ft and ds eye disc clones, the loss of Ft leads to an increase in fj-lacZ reporter expression, while the loss of Ds leads to a reduction in fj-lacZ expression. These results suggest that Fj regulates its own expression via a pathway involving Ft and Ds. While the importance to ommatidial polarity of the fj transcriptional regulation by Fj, Ft, and Ds is presently unclear, these data provide further evidence that these proteins can act in a common signaling pathway (Yang, 2002).

Planar polarization in the Drosophila eye is evident in the consistent orientation of the ommatidial units with respect to the equator and poles of the eye. The key step in this process is the establishment of a higher level of Fz PCP signaling activity in the equatorial member of the R3/R4 precursor pair. This data is relevent to understanding the mechanisms by which the R3/R4 precursor cells within an ommatidium sense which cell is closer to the equator. It has been shown that Ft and Ds, play essential roles in this process. Support for this conclusion comes from analyzing ommatidial development in the absence of Ft or Ds activity. In either case, the orientation of ommatidial polarity and Fz-dependent signaling events are randomized with respect to the equator. From this data, a model is proposed for the establishment of PCP in which the presence of higher Ft function in the equatorial cell allows it to attain a higher level of Fz signaling than its polar neighbor. This idea is supported by analysis of ommatidia in which one member of the R3/R4 precursor pair lacks Ft function. This mosaic analysis has shown that Ft normally functions from the equatorial precursor cell (the R3) to promote correct ommatidial polarity and that experimental alteration resulting in the absence of Ft from the equatorial precursor cell allows Ft present in the polar cell to direct the polar cell to become an ectopic R3 (Yang, 2002).

A second feature of the proposed model is that greater Ft function in the equatorial cell results from the presence of greater Ds function in the adjacent polar precursor cell. Important support for this idea comes from mosaic experiments showing that Ds acts primarily in the polar precursor cell and from genetic epistasis experiments between ft and ds that are consistent with Ds acting through the modulation of Ft. The final element of the model is that the presence of greater Ds function in the polar precursor cell results from two sources. ds transcription is graded (high at the poles, low at the equator) across the eye disc due to a stimulation of ds transcription by the secreted ligand Wg, which is expressed at the poles. A second source of Ds regulation appears to come from the action of Fj, which is expressed in a graded fashion from the equator and acts in the equatorial cell, in which it is more highly expressed, to promote R3 fate. Importantly, results from genetic epistasis experiments between fj and ds are consistent with Fj regulating ommatidial polarity by modulating Ds function (Yang, 2002).

While the proposed model provides a genetic framework for an understanding of how Ft, Ds, and Fj collaborate to regulate Fz signaling and the orientation of planar polarity, many issues remain that will only be answered by biochemical analysis of the signaling pathway. Among the most important of these is whether the regulation at each step of the pathway occurs in a cell-autonomous or in a nonautonomous 'crosstalk' manner. Because the specification of the R3 and R4 cells is determined by a competition between them for the higher Fz signaling state, two possible mechanisms exist for each step in the proposed pathway. Using the regulation of Ft by Ds as an example, the presence of higher Ds in the polar cell may act in a cell-autonomous manner to inhibit Ft function, causing the polar cell to have less Ft activity than its equatorial neighbor. Alternatively, Ds in the polar cell could promote Ft function in the neighboring equatorial cell, perhaps as a result of cadherin domain interactions between Ds and Ft across the junction of the two cells. It is worth noting that cadherin domain interactions of Ft or Ds with Fmi may also play an important role in Fz regulation. For example, the ability of Fmi to inhibit Fz signaling in the wing raises the possibility that Ft on the surface of the equatorial cell might bind to and stimulate the activity of Fmi in the neighboring polar cell. Addressing these different mechanisms of action will require detailed understanding of the architecture of the cell-cell junction formed between the R3/R4 precursor cells (Yang, 2002 and references therein).

One important implication of this study regards the role of Wnt proteins in regulating PCP. While the central role of Fz in PCP signaling has led to the proposal that a Wnt gradient might directly control the orientation of planar polarization by acting as a graded Fz ligand, appropriately expressed Wnts have not been reported. This analysis provides a possible explanation for the directional bias of Fz signaling that does not rely directly on a Wnt molecule to provide a graded Fz activating signal. It is important to note, however, that these experiments do not rule out the possibility that a Wnt molecule might participate in planar polarity signaling in the eye by providing a basal level of Fz activation that is then modified by the action of Fj, Ds, and Ft (Yang, 2002).

Thus a model has been provided for the directional biasing of Fz signaling in the Drosophila eye. These data illustrate how a diffusible morphogen, Wg in this case, can direct PCP by establishing opposing gradients of ds and fj expression that then regulate Ft and Fz. This analysis raises the possibility that the core of this proposed mechanism, the ability of opposing gradients of ds and fj expression to directionally bias Ft and Fz signaling, might comprise an evolutionarily conserved cassette that is deployed, under the control of different morphogens, to orient PCP signaling in other tissues. Several reports support this idea. (1) recent work has shown that Fj regulates PCP signaling in the Drosophila wing and is expressed in a gradient along the p/d axis; (2) it has been shown that Ds also regulates PCP in the wing and is expressed in a pattern complementary to that of Fj. Together, these results suggest that Ds and Fj collaborate to regulate planar polarity signaling in Drosophila tissues other than the eye. (3) It has been suggested that mice affected by the Ames waltzer hearing loss syndrome, which results from mutations in the cadherin superfamily member Pch15 and is characterized by extensive disorganization and degeneration of the cochlear and vestibular neuroepithelium, have early defects in the orientation of their outer hair cells (Alagramam, 2001; Raphael, 2001). This early phenotype may represent a defect in the establishment of PCP and thus suggest a conserved role in PCP signaling for other nonclassical cadherins besides Ft and Ds (Yang, 2002 and references therein).

Another study suggests that fat is involved not in the establishment of R3 and R4 cell fate, but instead functions in establishing the dorsal-ventral midline (equator) during eye morphogenesis. The Drosophila eye is a polarized epithelium in which ommatidia of opposing chirality fall on opposite sides of the eye's midline, the equator. The equator is established in at least two steps: photoreceptors R3 and R4 adopt their fates, and then ommatidia rotate clockwise or counterclockwise in accordance with the identity of these photoreceptors. Two cadherins, Fat (Ft) and Dachsous (Ds), play a role in conveying the polarizing signal from the D/V midline in the Drosophila eye. In eyes lacking Ft, the midline is abolished. In ft and ds mutant clones, wild-type tissue rescues genetically mutant tissue at the clonal borders, giving rise to ectopic equators. These ectopic equators distort a mosaic analysis of these genes and have led to a possible misinterpretation that ft and ds are required to specify the R3 and R4 cell fates, respectively (Yang, 2002). Interpretation of these data supports a significantly different model in which ft and ds are not necessarily required for fate determination. Rather, they are involved in long-range signaling during the formation of the equator, as defined by the presence of an organized arrangement of dorsal and ventral chiral ommatidial forms (Rawls, 2002).

ft has long been known for its role in proliferation control. The identification of new ft alleles in a FLP/FRT screen has revealed a role for Ft in establishing epithelial polarity. In ft422 null clones, approximately 52.5% of ommatidia—including mosaic and genetically mutant ommatidia—exhibit defects in polarity. Of these, 50.5% are inverted on their D/V axis. The remaining 2.0% of ommatidia are inverted on their A/P axis or on both their A/P and D/V axes. Furthermore, 98% of mosaic ommatidia that are phenotypically mutant are inverted on their D/V axis (Rawls, 2002).

Dorsoventrally inverted ommatidia are not randomly distributed within ft clones. Rather, they are preferentially localized toward the polar border such that the phenotypically mutant ommatidia are found in the polar region of the clone and phenotypically wild-type ommatidia are found along the equatorial border. The consequence of this biased distribution of ommatidia is an 'inverted equator' (originally called a pseudoequator) within the mutant clone, in which the points of opposing trapezoids face each other. Inverted equators in ft clones consistently arise approximately two rows from the equatorial border of the clone. These inverted equators were seen in 35/41 (85%) ft422 clones. The 15% of ft clones with no apparent ectopic equator were either small or long and narrow and therefore not broad enough to detect this phenotype (Rawls, 2002).

While a small percentage of ft clones lie along the equator, of over 200 ft clones examined, none cross the equator. A closer examination of these clones has revealed that the position of the endogenous equator, which can be identified in neighboring wild-type tissue, gets shifted by one to two ommatidial rows along the mutant border to accommodate the ft mutant clone. These observations suggest that the juxtaposition of ommatidia with high versus low Ft activity influences the placement of the equator (Rawls, 2002).

ds, known for its role in morphogenesis, plays a role in setting up polarity in the eye. The bias of D/V:A/P errors is similar to that described for ft clones, although fewer ommatidia are disrupted. Approximately 29% of ommatidia (both mosaic and genetically mutant) in ds38K clones (strong hypomorphic allele) display D/V inversions, while fewer than 1% of genetically mutant ommatidia display A/P inversions (517 ommatidia scored from 12 ds38k clones). Inverted equators also occur in ds mutant clones at a similar frequency as is seen with ft: 15 out of 18 clones (83%) had inverted equators. However, it is interesting to note that ds inverted equators arise along the equatorial border of the clone rather than within the clone, as is seen with ft. Finally, although ectopic ds equators are rare, they do occur. This phenotype might be more penetrant in a null allele of Ds function; however, no such alleles have been reported (Rawls, 2002).

Inverted and/or ectopic equators have also been observed in mirr, fng, and four-jointed (fj). One significant difference between ft/ds and these genes is that, in these other examples, mutant tissue nonautonomously disrupts wild-type tissue, generating a contiguous patch of nonautonomous D/V inversions. In ft, nonautonomous effects are never observed in wild-type ommatidia. In ds, the nonautonomy can extend many rows beyond the clone and may be separated by up to eight rows of unaffected ommatidia. It is appealing to speculate that this 'extended' nonautonomy is an effect of the twin-spot clone, in that a difference in relative amounts of Ds activity reverses the polarity. However, the scattered occurrence of these inverted ommatidia and the fact that a range cannot be seen in the number of affected ommatidia make this hypothesis somewhat unsatisfying. In order to test this 'twin-spot' hypothesis unambiguously, again, a protein null allele of ds is necessary (Rawls, 2002).

The clonal phenotypes of ft and ds suggest that they are involved in establishing equators. Since the establishment of the equator is known to involve long-range signaling, the nonautonomous effects seen in the clonal phenotypes might be a consequence of a requirement for ft and ds in the transduction of a global patterning signal. To remove any effects of long-range signaling, mutant eyes were generated completely devoid of wild-type Ft using the EGUF system (Stowers, 1999). In contrast to ft clones, in EGUF-ft eyes, the D/V axis is so severely perturbed that the endogenous equator is abolished. The greater degree of disruption observed in EGUF-ft eyes compared to mosaic clones suggests that wild-type tissue communicates with mutant tissue, perhaps via cell-cell relay of the signal transduced by Ft. If this is the case, then the presence of the inverted equator within the mutant clones is established as a consequence of signaling from wild-type tissue (Rawls, 2002).

The EGUF-ds phenotype mimics the clonal ds phenotype: 31% of ommatidia display D/V inversions. In EGUF-ds eyes, the endogenous equator is evident. This may be due to the weaker nature of this allele. Alternatively, it may suggest that Ds plays a more modulating role in establishing the D/V midline than does Ft (Rawls, 2002).

Ft and Ds act nonautonomously in the eye. In ft clones, the majority of polarity defects occur in the polar region of the clone. In contrast, ommatidia in the equatorial region of the clone are phenotypically wild-type, suggesting that wild-type ommatidia outside the equatorial boundary of the clone rescue genetically mutant ommatidia within the equatorial region of the clone. Furthermore, wild-type tissue on the polar border of the clone does not rescue mutant ommatidia within the polar region of the clone, indicating rescue takes place only in an equatorial to polar direction and not from the poles to the equator. If this is the case, the Ft signal is propagated in a directional fashion from wild-type tissue at the equatorial border into the mutant clone. Finally, mutant tissue never nonautonomously affects wild-type tissue -- in over 200 clones analyzed, no inverted ommatidia were seen in which all eight photoreceptors were wild-type. In contrast to ft, rescue takes place in a polar to equatorial direction in ds clones (Rawls, 2002).

The tissue polarity genes fz and stbm are required to specify R3 and R4. These cells then regulate the direction of ommatidial rotation. Given the importance of these two cells in the establishment of polarity, ommatidia were examined that were mosaic for ft within the R3 and R4 pair. In the majority of cases, the Ft+ cell becomes R3. A similar analysis of the other developmental pairs of photoreceptors, R1/R6 and R2/R5, revealed that there is a strong tendency for the Ft+ cell to adopt the fate of the anterior (R1 and R2) rather than the posterior (R5 and R6) photoreceptor cell (Rawls, 2002).

There are two possible interpretations of these data. First, ft may be involved in specifying the anterior photoreceptor fates, as the data imply. However, if Ft is required to specify the R3 fate, at least a small fraction of ommatidia should be seen in which no R3 fate is specified in ft mutant tissue (i.e., ommatidia that have two R4s), as is the case in fz mutants, this phenotype is not seen in ft clones. Alternatively, this finding could reflect the link between how cells are recruited into the growing ommatidium, how ommatidia rotate to establish polarity, and how this process is disrupted in ft mutant ommatidia (Rawls, 2002).

In doing this mosaic analysis, it was essential to recognize that a property inherent to eye development is that ommatidia that arise at the polar border of a clone predominantly recruit their polar cells from wild-type tissue and their equatorial cells from mutant tissue. Phenotypically mutant ommatidia occur only in the polar region of ft clones. This phenotype complicates the analysis and makes it difficult to draw conclusions regarding the specific cell(s) in which Ft is required for cell fate (Rawls, 2002).

In wild-type clones, the cells that are recruited from the polar side of the clone (R4, R5, and R6) will face the posterior side of the clone at the end of rotation. In ft clones, ommatidia that are both phenotypically mutant and mosaic occur only at the polar boundary of the clone. Since these ommatidia are almost always D/V inversions, then they will have recruited their ft+ cells from the polar side of the clone, but rather than these polar-derived cells becoming posterior photoreceptors (R4, R5, and R6) as they would have in wild-type, they become anterior photoreceptors (R1, R2, and R3). Together, these factors create an artifactual bias in which the anterior photoreceptor cells are ft+ and the posterior cells are ft-. Consequently, even though the data appear to indicate ft is required for the anterior cell fates (R1, R2, R3), a more detailed analysis reveals that the nature of the ft phenotype introduces a developmental bias that must be considered (Rawls, 2002).

The mosaic analysis of ds mutant clones revealed a trend the opposite of that described for ft. In ommatidia mosaic for ds in the pairs R1/R6, R2/R5, and R3/R4, the majority of photoreceptors are wild-type for the posterior fates (R4, R5, and R6). Since D/V inversions are found on the opposite side of the clone in ds compared to ft, this is the expected result if one applies the same logic as described above for ft. As with ft, no functional autonomy can be assigned to a single cell (Rawls, 2002).

A contrasting interpretation of a mosaic analysis of ft and ds, presented by Yang (2002), suggests ft and ds are required to specify the fates of photoreceptors R3 and R4, respectively. However, mosaic analyses of ft and ds are inherently biased due to the clonal phenotypes, as described above. This bias might mask a role for ft or ds in the R3/R4 fate decision, but currently there is no compelling evidence for such a functional requirement. The genetic data reported here and in Yang (2002) are insufficient to draw conclusions about the role of ft and ds in fate specification. Extensive experimentation and a better understanding of mechanism are necessary to discriminate between a role for ft in global D/V signaling versus a requirement for ft and ds in specification of the R3 and R4 cell fates (Rawls, 2002).

Early acting genes, for example, mirror (mirr) and fringe (fng), specify the dorsal and ventral halves of the eye, respectively, thereby setting up the D/V boundary. The ft clonal phenotype suggests ft might mediate D/V boundary formation. To address this possibility, the effect of ft on mirr expression was assessed. The expression pattern of the enhancer trap line 8A5, in which the mirr promoter drives expression of white, is unaffected in EGUF-ft eyes. In these eyes, mirr expression remains restricted to the dorsal half of the eye, indicating that ft is required either downstream of or in parallel to mirr (Rawls, 2002).

A proposed model for ft activity differs from that of Yang (2002) because this study (1) takes into account the observations that ectopic equators are generated in mutant clones and that the ft clonal phenotype differs significantly from the EGUF-ft phenotype and (2) considers fundamental properties of eye development in conjunction with the clonal phenotype. A model is proposed in which ft conveys D/V positional information to developing ommatidia to create the D/V midline (Rawls, 2002).

Ft functions to inhibit D/V signaling in the wing and haltere. The data are consistent with this proposal -- new equators are generated in ft mutant clones in the eye. It is proposed that a consistent level of Ft activity throughout the eye inhibits the D/V signaling required to form the equator. At the equator, Ft activity must be inhibited. The molecule that inhibits Ft could be expressed in several ommatidial rows encompassing the future midline. In a ft mutant clone, the phenotype is rescued for two rows in the equatorial region of the clone, suggesting the D/V signal propagated by Ft can be relayed for a distance of two ommatidial rows. Therefore, it is proposed that in the presence of the regulatory protein, the D/V signal can also be relayed an equivalent distance at the endogenous equator. When the D/V signal reaches its minimum, tissue with Ft activity apposes tissue without Ft activity, and it is at this point that the equator is established (Rawls, 2002).


cDNA clone length - 15,966

Bases in 5' UTR - 192

Bases in 3' UTR - 332


Amino Acids - 5,147

Structural Domains

The fat locus encodes a novel member of the cadherin gene superfamily: an enormous transmembrane protein of over 5000 amino acids with a putative signal sequence, 34 tandem cadherin domains, four EGF-like repeats, a transmembrane domain, and a novel cytoplasmic domain (Mahoney, 1991).

fat: Biological Overview | Evolutionary Homologs | Regulation | Developmental Biology | Effects of Mutation | References

date revised: 25 July 2002

Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.