In the developing eye of Drosophila cell fate is controlled by a cascade of inductive interactions. Little is known about how the specificity of positional signalling is achieved such that directly adjacent progenitor cells reproducibly choose distinct developmental pathways. The determination of the R7 photoreceptor in each ommatidium depends on the presence of the sevenless protein which acts as a receptor for positional information on the R7 precursor. The rough gene encodes a homeodomain protein that plays an instructive role in the determination of the R3 and R4 photoreceptor cells. The use of ectopic expression of sevenless and rough has provided insight into the mechanisms of positional signalling and the normal function of rough. Ubiquitous expression of sevenless does not alter cell fate suggesting that the inducing signal is both spatially and temporally controlled. Conversely, ectopic expression of rough in the R7 precursor causes a transformation of R7 cells into Rl-6 type cells. This indicates that rough acts, similar to other homeobox genes, as a selector gene that determines the fate of single cells.

In the developing eye of Drosophila, specification of cell fate can be studied at the single cell level. This is facilitated by the remarkably precise arrangement of cells within both the developing and the mature retinal epithelium (Fig. 1A and IE). The compound eye consists of a repetitive array of about 800 identical unit eyes, or ommatidia, each of which is a precise assembly of a few distinct cell types (Fig. 1C; for review see Ready, 1989). The eight photoreceptor cells (R1 to R8) of the ommatidial unit can be grouped into three functional classes based upon position, spectral sensitivities, and projection pattern of their axons: R1 to R6, R7, and R8. During development they assemble and differentiate in a fixed temporal sequence: R8 is the founder cell in each ommatidium, subsequently R2/R5, R3/R4, and R1/R6 are added pairwise, and finally R7. Later, the non-neuronal elements, the cone cells and the pigment cells, become incorporated (Fig. IF).

Fig. 1.

Structure and development of the compound eye of Drosophila. Anterior is to the left. (A) Scanning electron micrograph of a wild-type eye. The enlargement shows the hexagonal arrangement of the individual ommatidia. (B) Schematic representation of cross-sections through the distal part of a wild-type and a sevenless mutant ommatidium. The positions of the photoreceptors R1-R7 are shown. In the sevenless mutants the R7 cell is missing. (C) Histological cross-sections through wild-type and (D) sevenless mutant eyes. (E) Eye imaginai disc stained with an antiserum against the sevenless protein visualizing the assembly of the ommatidia. Each of the stained clusters corresponds to a subset of cells in an ommatidium. Ommatidial assembly begins in a wave (arrow) that moves over the disc epithelium towards the anterior; cells anterior to this wave are unlabeled. The different developmental stages of ommatidial assembly are spatially displayed along the anterior-posterior axis. (F) Schematic representation of the inner 12-cell unit of an ommatidium at the larval stage. The different gray shades indicate the temporal sequence of assembly. R8 is the first cell in the cluster followed by the pairwise addition of photoreceptors R2/R5, R3/R4, R1/R6 and finally by R7.

Fig. 1.

Structure and development of the compound eye of Drosophila. Anterior is to the left. (A) Scanning electron micrograph of a wild-type eye. The enlargement shows the hexagonal arrangement of the individual ommatidia. (B) Schematic representation of cross-sections through the distal part of a wild-type and a sevenless mutant ommatidium. The positions of the photoreceptors R1-R7 are shown. In the sevenless mutants the R7 cell is missing. (C) Histological cross-sections through wild-type and (D) sevenless mutant eyes. (E) Eye imaginai disc stained with an antiserum against the sevenless protein visualizing the assembly of the ommatidia. Each of the stained clusters corresponds to a subset of cells in an ommatidium. Ommatidial assembly begins in a wave (arrow) that moves over the disc epithelium towards the anterior; cells anterior to this wave are unlabeled. The different developmental stages of ommatidial assembly are spatially displayed along the anterior-posterior axis. (F) Schematic representation of the inner 12-cell unit of an ommatidium at the larval stage. The different gray shades indicate the temporal sequence of assembly. R8 is the first cell in the cluster followed by the pairwise addition of photoreceptors R2/R5, R3/R4, R1/R6 and finally by R7.

Cells are directed to their individual fates by inductive signals communicated by neighboring cells (Tomlinson and Ready, 1987). Differentiating cells are thought to express cell type specific signals which undetermined cells receive and interpret in order to choose their differentiation pathways. Mutations in genes whose products are involved in these inductive pathways can be grouped into two classes. Genes encoding products involved in the signalling pathway will show a non-autonomous mutant phenotype. Conversely, genes encoding products involved in the reception of the inductive signals and the subsequent execution of the developmental program will exhibit a cell autonomous mutant phenotype and are required intrinsically for a cell to differentiate correctly.

Three genes have been identified as important components in the communication mechanisms (Fig. 2). The sevenless mutation prevents the presumptive R7 cell from differentiating correctly and causes the cell to become a lens-secreting cone cell (Harris et al. 1976; Tomlinson and Ready, 1986). The sevenless gene product is a receptor tyrosine kinase required autonomously in the presumptive R7 cell and is thought to receive signals from the adjacent R8 cell (Basler and Hafen, 1988; Bowtell et al. 1988). The boss gene is required only in R8 for the R7 cell to develop appropriately (Reinke and Zipursky, 1988). The non-autonomy of boss suggests that it is part of the signalling machinery in R8. Similar to boss, rough is non-autonomously required in cells R2 and R5 for the correct specification of the neighboring R3 and R4 cells (Tomlinson et al. 1988; Fig. 2). The rough protein appears to act on the signalling side of the inductive pathway. Surprisingly, molecular analysis of rough revealed that it encodes a homeodomain protein and not a membrane protein or a secreted factor (Saint et al. 1988; Tomlinson et al. 1988). It has been postulated that rough functions in R2 and R5 as a transcriptional regulator for an R3/4 inducing signal (Tomlinson et al. 1988).

Fig. 2.

Mutations interfere with different steps of the assembly sequence of the ommatidial unit. Only the assembly of the 8-photoreceptor cell unit is shown schematically. Mutations in the rough gene interfere with the correct development of R3/4. Subsequent steps are disturbed which leads to the roughening of the eye. Mutations in the genes sevenless and boss interfere with a later step, namely the specification of the R7 photoreceptor cell. In the absence of either boss or sevenless the R7 precursor does not develop into a photoreceptor cell but into a non-neuronal cone cell.

Fig. 2.

Mutations interfere with different steps of the assembly sequence of the ommatidial unit. Only the assembly of the 8-photoreceptor cell unit is shown schematically. Mutations in the rough gene interfere with the correct development of R3/4. Subsequent steps are disturbed which leads to the roughening of the eye. Mutations in the genes sevenless and boss interfere with a later step, namely the specification of the R7 photoreceptor cell. In the absence of either boss or sevenless the R7 precursor does not develop into a photoreceptor cell but into a non-neuronal cone cell.

Although the characterization of mutations that disrupt ommatidial assembly permits the identification of genes involved in this process, it does not provide information on how these products function and interact such that directly adjacent cells select distinct developmental pathways. To address the question of the specificity of positional signalling we and others have started to examine the effects of ectopic expression of genes coding for sensory and instructive proteins. These analyses have led to important conclusions about the function of these genes and about the multipotency of retinal precursor cells. In this article we will review the results obtained from studying ectopic expression of sevenless and rough.

Ubiquitous expression of sevenless does not alter cell fate

Specification of cell fate by ligand - receptor interactions can be viewed in three different ways. (1) The ligand is ubiquitous but the receptor is expressed only locally. Specificity is controlled by the restricted expression of the receptor (Fig. 3A). (2) The ligand is localized and the receptor is ubiquitously expressed. In this case the specificity is controlled by the local expression of the ligand (Fig. 3C). (3) Both ligand and receptor are partially restricted. Specificity is controlled by the combination and overlap between the expression domains of both ligand and receptor (Fig. 3B).

Fig. 3.

Models showing how regional specificity can be achieved. Rounded squares represent undetermined cells; already differentiated cells are indicated by rhomboids. (A) Ubiquitous ligand and restricted expression of the receptor. Cell type restricted response, is controlled by the expression of the receptor. Prominent examples of this type of regulation are growth factor-mediated responses. (B) Partially restricted expression of signal and receptor. (C) Ubiquitous receptor and localized ligand. All undetermined cells contain receptors but only one contacts the ligand-presenting cell. Position-dependent specification of cell fate is likely to be governed by this mechanism.Υ receptor; •, diffusible ligand; ⟟, membrane bound ligand; stippling, cell that is determined in response to ligand-induced receptor activation.

Fig. 3.

Models showing how regional specificity can be achieved. Rounded squares represent undetermined cells; already differentiated cells are indicated by rhomboids. (A) Ubiquitous ligand and restricted expression of the receptor. Cell type restricted response, is controlled by the expression of the receptor. Prominent examples of this type of regulation are growth factor-mediated responses. (B) Partially restricted expression of signal and receptor. (C) Ubiquitous receptor and localized ligand. All undetermined cells contain receptors but only one contacts the ligand-presenting cell. Position-dependent specification of cell fate is likely to be governed by this mechanism.Υ receptor; •, diffusible ligand; ⟟, membrane bound ligand; stippling, cell that is determined in response to ligand-induced receptor activation.

The sevenless gene is expressed in a complex spatial and temporal pattern (Tomlinson et al. 1987). During the third instar larval period, it is almost exclusively expressed in the developing eye imaginai disc. Within the eye disc it is expressed transiently only in a subset of the ommatidial precursors (Fig. IE). Based on the expression pattern of sevenless it has been postulated that the local specification of R7 cells is controlled according to the model shown in Fig. 3B, by the combination of the partially restricted expression of both ligand and receptor (Tomlinson et al. 1987). To test this hypothesis, the sevenless gene was placed under the control of an inducible ubiquitous promoter and introduced into the germline of sevenless mutant flies (Basler and Hafen, 1989; Basler et al. 1989; Bowtell et al. 1989b). Ubiquitous expression of sevenless during development specifies R7 cells in correct positions but does not interfere with the development of other cells where sevenless is not normally expressed. This result suggests that the complex spatial and temporal regulation of sevenless gene expression does not contribute to the spatially restricted specification of R7 cells. Specificity of R7 selection may therefore be controlled by the local presentation of the sevenless ligand rather than by the restricted expression of the receptor (Fig. 3C).

Even if the expression of the ligand is restricted to a single cell, this cell may still contact multiple neighboring cells. For example, in the case of the R7 determination, the R8 cell which most likely produces the ligand for sevenless, is in contact with seven cells of which all may express the sevenless receptor, yet only the presumptive R7 cell initiates R7 development. It has been postulated that cell fate in the eye is controlled by the combination of different contacts between neighboring cells (Tomlinson and Ready, 1987). Alternatively, specificity may also be achieved if the ligand is not only locally but also temporally restricted, such that at the time of ligand presentation only the presumptive R7 cell is capable of responding since all other cells have selected a different pathway earlier. It is even conceivable that temporal control of ligand presentation alone is sufficient to control specificity. If for example the ligand for sevenless is expressed on all photoreceptor cells relatively late during the differentiation, R8, the oldest cells in the cluster, will be the first to express the ligand at a time when the R7 precursor is the only cell that can still respond to the inductive signal. Transient expression of surface proteins during the differentiation of neural cells is well documented (Patel et al. 1987). For an undetermined cell it might therefore not only be important what its neighbors are but also how old these neighbors are. The test of which of the two models, combinatorial cell contacts with multiple inducing ligands for a single fate, or the spatially and/or temporally restricted expression of a single inducing signal, is correct has to await the cloning of the gene for such a ligand. By ectopic expression of the ligand it should be possible to distinguish between the two models.

Indiscriminate expression of the homeobox gene rough is lethal

The rough gene is expressed in a small subpopulation of the ommatidial precursor cells. Persistent expression is observed only in the R2/5 and R3/4 cells (Kimmel et al. 1990). To examine the effect of ectopic expression of the rough gene we have designed a rough minigene under the control of the inducible hsp70 heat shock promoter. Repeated heat-induction at embryonic, larval and pupal stages resulted in lethality of the hsp-rough transformants. Under the same conditions, wild-type or hsp-sev transformants that contain a heat shock-inducible sevenless gene survived. Even single heat shocks at the white prepupal stage caused a substantially lower survival rate. Hatched flies that survived the heat shock have severely reduced eyes bordered anteriorly by a sharp vertical scar. The few ommatidia anterior to this scar were highly irregular whereas ommatidial columns posterior to the scar were normal. Similar results were obtained by Kimmel et al. (1990) except that they did not observe lethality associated with the ubiquitous expression of rough. Although we do not know what the reason for this discrepancy is, it is possible that the two hsp-rougA constructs used, differ in their level of expression upon heat induction.

Ubiquitous expression of rough during development appears to interfere with the normal development of cells. Induction of rough in adult flies, however, has no effect. The greatly reduced eyes resulting from a single heat shock during the third instar stage indicates that ubiquitous expression of rough interferes with the formation of new ommatidial columns in the eye imaginai disc. Heat shock induction of sevenless and Ultrabithorax (Ubx) carried out in parallel experiments did not produce a similar phenotype. Hence we assume that this effect is specifically caused by rough.

Although ectopic expression of rough during embryonic development was lethal, we did not observe a detectable alteration in the cuticular pattern of the central or peripheral nervous system of heat-shocked hsp-rough transformants. Under the same conditions ubiquitous expression of Ubx caused an almost complete transformation of the head and thoracic segments into Al segments (Gonzàles-Reyes et al. 1990). We assume the ectopic expression of rough in unrelated cells causes a general disruption of a cellular function rather than inducing an alteration in cell fate.

Localized ectopic expression of rough using the sevenless enhancer

To limit ectopic expression of rough to a defined subset of cells in the developing eye imaginai disc, we used the sevenless enhancer sequences that control the sevenless expression pattern. In contrast to other tissue specific enhancers which are active only in differentiated cell types, such as the cis-acting sequences that control rhodopsin expression (Mismer et al. 1987), sevenless is expressed transiently in a subpopulation of ommatidial precursor cells prior to or at the time of their commitment (Tomlinson et al. 1987). sevenless is expressed strongly in cells R3, R4, R7 and the cone cells, and weakly in R1 and R6. It has been shown previously that a gene-internal fragment of the sevenless gene is responsible and sufficient for the temporally and spatially restricted expression of sevenless (Basler et al. 1989; Bowtell et al. 1989a). This regulatory element imposes the sevenless-specific expression pattern on heterologous promoters such as the hsp70 promoter. We inserted a restriction fragment containing the sevenless enhancer upstream of the hsp-rough construct, sev-hsp-rough, and generated germ-line transformants (Basler et al. 1990). A similar analysis was carried out independently by Kimmel et al. (1990).

Although the eyes of heterozygous sev-hsp-rough transformants appeared normal, histological sections revealed an irregular rhabdomere pattern in 50–80% of the ommatidia. Wild-type ommatidia contain 8 photoreceptor cells (R1 to R8). The rhabdomeres of R1 to R6 extend through the depth of the retina, and form an asymmetric trapezoid (Fig. 4A, C, and E). The rhabdomere of R7 is smaller in diameter than that of the outer photoreceptors and occupies the central position in the distal two thirds of the retina. This position is occupied proximally by the rhabdomere of R8 (Fig. 4E). Therefore in each plane of cross section, a highly ordered pattern of 7 rhabdomeres is visible (Fig. 4A,C, and E). Apical sections of sev-hsp-rough ommatidia also display seven rhabdomeres, but in an abnormal arrangement (Fig. 4B). The few undisturbed ommatidia can be used as an internal reference for the orientation of the rhabdomere pattern. Comparison with adjacent affected ommatidia indicates that photoreceptors R1 to R6 are normal, but that the cell between R1 and R6, corresponding to the R7 cell in wildtype, causes the observed irregularity: its rhabdomere is not in the central position, but is instead between that of R1 and R6 (Fig. 4B and 4E). Furthermore, its rhabdomere diameter is larger than that of a normal R7 cell, and similar to the size of the Rl-6 type rhabdomere. In contrast to wild-type R7 cells, these cells extend through the depth of the retina. Therefore in more basal sections we observe 8 rhabdomeres instead of 7 in the ommatidia containing transformed R7 cells: in addition to the small R8 rhabdomere surrounded by the rhabdomeres of R1 to R6, there is an additional rhabdomere present between R1 and R6, corresponding to the transformed R7 cell. Therefore in the sev-hsp-rough transformants, a large fraction of R7 cells assume all of the morphological characteristics of the Rl-6 type photoreceptor cells. We further demonstrated that in transformed flies homozygous for the ora mutation (Stark and Sapp, 1987), which causes specific degeneration of the rhabdomeres of the R1-R6 photoreceptor cells, the transformed R7 cells but not normal R7 cells also degenerate. We therefore conclude the expression of rough causes the transformation of the R7 precursor into an Rl-6 type photoreceptor.

Fig. 4.

Transformation of R7 cells into Rl-6 type photoreceptor cells. Histological sections through wild-type (A, C) and sev-hsp-rough (B, D) eyes. As schematically illustrated in (E), both apical (A, B) and basal sections (C, D) are shown. Each photoreceptor cell has a microvillar stack of membranes containing rhodopsin, called the rhabdomere, that projects towards the center of the ommatidium. In wild-type, the R7 rhabdomere differs morphologically from the rhabdomeres of R1 to R6 with respect to position, diameter and length: it is located in the center of the trapezoid formed by the larger rhabdomeres of R1 to R6, it is smaller in diameter and it extends to only two thirds the depth of the retina. In many ommatidia of heterozygous sev-hsp-rough transformants, the rhabdomere of the R7 cell has all the characteristics of an Rl-6 type rhabdomere (arrow in B, D, hatched in E). In contrast to the seven rhabdomeres observed in basal sections of wild-type eyes (C), eight rhabdomeres are visible in altered ommatidia in corresponding sections of sev-hsp-rough transformants (D). Anterior is down. Magnification, × 1000.

Fig. 4.

Transformation of R7 cells into Rl-6 type photoreceptor cells. Histological sections through wild-type (A, C) and sev-hsp-rough (B, D) eyes. As schematically illustrated in (E), both apical (A, B) and basal sections (C, D) are shown. Each photoreceptor cell has a microvillar stack of membranes containing rhodopsin, called the rhabdomere, that projects towards the center of the ommatidium. In wild-type, the R7 rhabdomere differs morphologically from the rhabdomeres of R1 to R6 with respect to position, diameter and length: it is located in the center of the trapezoid formed by the larger rhabdomeres of R1 to R6, it is smaller in diameter and it extends to only two thirds the depth of the retina. In many ommatidia of heterozygous sev-hsp-rough transformants, the rhabdomere of the R7 cell has all the characteristics of an Rl-6 type rhabdomere (arrow in B, D, hatched in E). In contrast to the seven rhabdomeres observed in basal sections of wild-type eyes (C), eight rhabdomeres are visible in altered ommatidia in corresponding sections of sev-hsp-rough transformants (D). Anterior is down. Magnification, × 1000.

To find out whether rough had to be expressed in the R7 precursor or in a neighboring cell for the transformation of the R7 into an Rl-6 type cell we used mitotic recombination to generate eyes in which the sev-hsp-rough construct was only present in a subpopulation of cells. These results indicated that expression of rough in just the R7 precursor cell is sufficient for the transformation to occur (Basler et al. 1990; Kimmel et al. 1990). The autonomous requirement of sev-hsp-rough for the R7 transformation contrasts the apparent non-autonomy of the rough mutant phenotype: In the developing ommatidia, the rough protein is required in R2 and R5 for the ommatidia to develop correctly. In its absence from these two cells, R3 and R4 develop aberrantly, but R2 and R5, as judged from morphological criteria, develop normally. Given the role of the rough homeodomain protein as putative transcription factor, it has been proposed that rough controls in R2 and R5 the expression of a gene encoding the R3/4 inducing signal (Tomlinson et al. 1988). One way to reconcile this contradiction is to suggest that in the presumptive R7 cell, the rough protein transcriptionally activates the R3/4 inducing signals which, in an autocrine way, trigger the Rl-6 developmental pathway in R7. We think, however, that this is unlikely, and propose instead that rough serves an autonomous role in controlling the identity of the R2 and R5 cells. Since markers that distinguish between the identities of the outer photoreceptor cells are not available, the failure to obtain R2/5 identity in rough mutants could not be detected. According to this model the incorrect development of R3 and R4 in rough mutants is a mere consequence of the incomplete differentiation of R2 and R5. It appears therefore that rough, similar to other homeotic genes, is a selector gene that distinguishes between alternative fates of the cells where it is expressed. Although the progenitors of R2 and R5 enter the photoreceptor cell pathway, in rough mutants they fail to assume an R2/5 identity, which is manifested in their failure to induce R3 and R4. In wild-type, rough is also expressed in R3 and R4 (Kimmel et al. 1990). According to our model the presence of rough protein in R3 and R4 also causes these cells to assume an R2/5 identity and thereby distinguishes them from R1 and R6. The failure to assume this identity in rough mutant clones, however, cannot be detected because R3 and R4 do not recruit other photoreceptor cells. Similarly, a potential, incorrect signal sent by the transformed R7 cells does not induce a change in cell fate of neighboring cone and pigment cells, presumably because they are not competent to respond to such a signal.

Interestingly, rough induced transformation of R7 into an Rl—6 type photoreceptor is dependent on the sevenless and boss gene. When sev-hsp-rough is crossed into a sevenless or boss mutant background we observed a complete absence of cells in the R7 position. Therefore sevenless and boss are epistatic over rough in the R7 precursor. In sevenless mutants the R7 precursor appears never to enter the photoreceptor cell pathway. Rough expression in the absence of either sevenless or boss is not sufficient to initiate the photoreceptor cell development. This suggests that specification of photoreceptor cell fate requires more than one function. Boss and sevenless appear to be involved in the initiation of this pathway in R7, whereas rough functions in the specification of photoreceptor cell identity. Rough expression in R7 can compete with or even override the step that leads to the specification of the R7 cell fate. The dependence of rough function on a prior commitment of the rough expressing cell to photoreceptor development is consistent with its normal function in R2 and R5, since these cells apparently initiate photoreceptor cell development in the absence of rough (Tomlinson et al. 1988). Although rough controls photoreceptor cell identity, it is not sufficient to confer neural fate to undetermined cells.

The developing eye as an experimental system to study position-dependent cell fate specification has its limitations in that experimental manipulations such as laser ablation of individual cells and single cell transplantations are not possible. These techniques have been extremely useful in the Drosophila embryo (Technau and Campos-Ortega, 1986) and in the grasshopper (Doe et al. 1985) for assaying the developmental potential of individual cells in an altered environment. This shortcoming can be overcome by molecular genetic manipulation of the system. Genes controlling developmental decisions or specifying positional information can be fused to heterologous promoters and reintroduced into the germline. In this way positional values in the developing field can be changed without the inherent side effects of experimental manipulation. The specificity of the molecular genetic manipulation depends largely on the genes used and, just as importantly, on the specificity of the control elements that confer ectopic expression. Ectopic expression may be achieved by fusing the structural gene to an inducible promoter such as the hsp70 promoter. Indiscriminate expression of genes, however, is often lethal, and this lethal phenotype is then difficult to relate to the normal function of the gene. Similarly, the results reported here on the effect of ubiquitous expression of rough are difficult to compare with the normal function of rough in specifying photoreceptor cell identity in postmitotic cells. In contrast, by using the sevenless enhancer to drive rough expression, we expressed rough ectopically in only a small number of related cells, and indeed obtained specific information on the function of rough. Although the promoters of other genes have been characterized that are specifically expressed in subsets of ommatidial cells (i.e. rhodopsin, Mismer et al. 1987), the sevenless enhancer is unique in that it is active early, during the determination process of retinal precursor cells. This approach will become more widely applicable as more stage- and tissue-specific enhancers are identified and characterized by the enhancer trap method (O’Kane and Gehring, 1987; Wilson et al. 1989; Bier et al. 1989).

The dominant phenotype observed by the ectopic expression of rough demonstrates that the restricted rough expression observed in wild-type is critical for proper differentiation of several cell types in the compound eye. In contrast, the restricted expression pattern of sevenless, a putative receptor for an inductive signal, is not important for the correct determination of retinal precursor cells (Basler and Hafen, 1989; Bowtell et al. 19896). This reflects the different roles of rough and sevenless in cell fate specification. While rough has an instructive role, sevenless acts as a sensory protein in cell fate specification.

Expression of rough in the R7 progenitor has demonstrated that the R7 precursor is multipotent. It can assume four distinct fates depending on the state of activity of the sevenless and rough gene products (Table 1): In wild-type, sevenless protein is activated, probably by boss, which results in an R7 fate. If rough is also present in the cell, it can become an Rl-6 type photoreceptor cell. If sevenless is not activated, the cell becomes a normal cone cell. We assume that, as in the case of normal cone cells, high levels of rough protein abort development of this cell. The observed multipotency indicates that the precision with which cell fate is determined in the eye depends in part on the spatially and temporally restricted expression of positional cues and additionally on the combination of different intracellular signal transducers that respond to the inductive stimulus.

Table 1.

Potential fates of the R7 precursor cell depending on different combinations of sevenless and rough activity

Potential fates of the R7 precursor cell depending on different combinations of sevenless and rough activity
Potential fates of the R7 precursor cell depending on different combinations of sevenless and rough activity
Basler
,
K.
and
Hafen
,
E.
(
1988
).
Control of photoreceptor cell fate by the sevenless protein requires a functional tyrosine kinase domain
.
Cell
54
,
299
311
.
Basler
,
K.
and
Hafen
,
E.
(
1989
).
Ubiquitous expression of sevenless’, position-dependent specification of cell fate
.
Science
243
,
931
934
.
Basler
,
K.
,
Siegrist
,
P.
and
Hafen
,
E.
(
1989
).
The spatial and temporal expression pattern of sevenless is exclusively controlled by gene-internal elements
.
EMBO
8
,
2381
2386
.
Basler
,
K.
,
Yen
,
D.
,
Tomlinson
,
A.
and
Hafen
,
E.
(
1990
).
Reprogramming cell fate in the developing Drosophila retina: transfromation of R7 cells by extopic expression of rough. Genes Dev.
4
,
728
739
.
Bier
,
E.
,
Vaessin
,
H.
,
Shepherd
,
S.
,
Lee
,
K.
,
McCall
,
K.
,
Barbel
,
S.
,
Ackerman
,
L.
,
Carretto
,
R.
,
Uemura
,
T.
,
Grell
,
E.
,
Jan
,
L. Y.
and
Jan
,
Y. N.
(
1989
).
Searching for pattern and mutation in the Drosophila genome with a P-lacZ vector
.
Genes Dev.
3
,
1273
1287
.
Bowtell
,
D. D. L.
,
Kimmel
,
B. E.
,
Simon
,
M. A.
and
Rubin
,
G. M.
(
1989a
).
Regulation of the complex pattern of sevenless expression in the developing Drosophila eye
.
Proc. natn. Acad. Sci. U.S.A.
86
,
6245
6249
.
Bowtell
,
D. D. L.
,
Simon
,
M. A.
and
Rubin
,
G. M.
(
1988
).
Nucleotide sequence and structure of the sevenless gene of Drosophila melanogaster. Genes Dev.
2
,
620
634
.
Bowtell
,
D. D. L.
,
Simon
,
M. A.
and
Rubin
,
G. M.
(
1989b
).
Ommatidia in the developing Drosophila eye require and can respond to sevenless for only a restricted period
.
Cell
56
,
931
936
.
Doe
,
C. Q.
and
Goodman
,
C. S.
(
1985
).
Early events in insect neurogenesis: II. The role of cell interactions and cell lineage in the determination of neuronal precursor cells
.
Devi Biol.
Ill
,
206
-
219
.
Gibson
,
G.
and
Gehring
,
W. J.
(
1988
).
Head and thoracic transformations caused by ectopic expression of Antennapedia during Drosophila development
.
Development
102
,
657
675
.
Gonzàlez-Reyes
,
A.
,
Urquia
,
N.
,
Gehring
,
W. J.
,
Struhl
,
G.
and
Morata
,
G.
(
1990
).
A functional hierarchy of homeotic genes that is not based on cross-regulatory interactions
.
Nature
334
,
78
80
.
Hafen
,
E.
,
Basler
,
K.
,
Edstroem
,
J. E.
and
Rubin
,
G. M.
(
1987
).
Sevenless, a cell-specific homeotic gene of Drosophila, encodes a putative transmembrane receptor with a tyrosine kinase domain
.
Science
236
,
55
63
.
Harris
,
W. A.
,
Stark
,
W. S.
and
Walker
,
J. A.
(
1976
).
Genetic dissection of the photoreceptor system in the compound eye of Drosophila melanogaster. J. Physiol.
256
,
415
439
.
Kimmel
,
B.
,
Heberlein
,
U.
and
Rubin
,
G. M.
(
1990
).
The horneo domain protein rough is expressed in a subset of cells in the developing Drosophila eye where it can specify photoreceptor cell subtype
.
Genes Dev.
4
,
712
727
.
Klemenz
,
R.
,
Weber
,
U.
and
Gehring
,
W. J.
(
1987
).
The white gene as a marker in a new P- element vector for gene transfer in Drosophila. Nucl. Acids Res.
15
,
3947
3959
.
Mismer
,
D.
and
Rubin
,
G. M.
(
1987
).
Analysis of the promoter of the ninaE opsin gene in Drosophila melanogaster. Genetics
116
,
565
578
.
O’Kane
,
C. J.
and
Gehring
,
W. J.
(
1987
).
Detection in situ of genomic regulatory elements in Drosophila. Proc. natn. Acad. Sci. U.S.A.
84
,
9123
9127
.
Patel
,
N. H.
,
Snow
,
P. M.
and
Goodmann
,
C. S.
(
1987
).
Characterization and cloning of fasciclin III: a glycoprotein expressed on a subset of neurons and axon pathways in Drosophila. Cell
48
,
975
988
Ready
,
D. F.
(
1989
).
A multifaceted approach to neural development
.
Trends Neurosci.
12
,
102
110
.
Reinke
,
R.
and
Zipursky
,
S. L.
(
1988
).
Cell-cell interaction in the Drosophila retina: the bride of sevenless gene is required in photoreceptor cell R8 for R7 cell development
.
Cell
55
,
321
330
.
Saint
,
R.
,
Kalionis
,
B.
,
Lockett
,
T. J.
and
Elizur
,
A.
(
1988
).
Pattern formation in the developing eye of Drosophila melanogaster is regulated by the homeobox gene, rough. Nature
334
,
151
154
.
Stark
,
W. S.
and
Sapp
,
R.
(
1987
).
Ultrastructure of the retina of Drosophila melanogaster. the mutant ora (outer rhabdomeres absent) and its inhibition of degeneration in rdgB (retinal degeneration-B)
.
J. Neurogenetics
4
,
227
240
.
Technau
,
G. M.
and
Campos-Ortega
,
J. A.
(
1986
).
Lineage analysis of transplanted individual cells in embryos of Drosophila melanogaster. I. The method
.
Roux’s Arch. Dev. Biol.
195
,
389
398
.
Tomlinson
,
A.
,
Bowtell
,
D. D. L.
,
Hafen
,
E.
and
Rubin
,
G. M.
(
1987
).
Localization of the sevenless protein, a putative receptor for positional information in the eye imaginai disc of Drosophila. Cell
51
,
143
150
.
Tomlinson
,
A.
,
Kimmel
,
B. E.
and
Rubin
,
G. M.
(
1988
).
Rough, a Drosophila homeobox gene required in photoreceptors R2 and R5 for inductive interactions in the developing eye
.
Cell
55
,
771
784
.
Tomlinson
,
A.
and
Ready
,
D. F.
(
1986
).
Sevenless: a cell-specific homeotic mutation of the Drosophila eye
.
Science
231
,
400
402
.
Tomlinson
,
A.
and
Ready
,
D. F.
(
1987
).
Neuronal differentiation in the Drosophila ommatidium
.
Devi Biol.
120
,
366
376
.
Wilson
,
C.
,
Pearson
,
R. K.
,
Bellen
,
H. J.
,
O’Kane
,
C. J.
,
Grossniklaus
,
U.
and
Gehring
,
W. J.
(
1989
).
P-element-mediated enhancer detection: an efficient method for isolating and characterizing developmentally regulated genes in Drosophila. Genes Dev.
3
,
1301
1313
.