Expression of rn and roe in third instar imaginal discs. (A-D) Late third instar rn⁸⁹ imaginal discs stained with X-gal. Expression is seen as a wide ring in the leg disc (A) and the antennal portion of the eye-antenna disc (B, arrow). Note the lack of detectable expression in the eye portion (B, arrowhead). Expression is also evident in the central region and the notum of the wing disc (C) and in the central region of the haltere disc (D). (E-G) In situ hybridization to wild type using a rn-specific probe in early (E), mid (F), and late (G) third instar discs. Expression of rn is seen in a pattern similar to that of X-gal in rn⁸⁹. In the leg disc (l), the expression of rn is transient and evident only during 80-96 hours after egg laying. In contrast, the expression of rn in the wing (w) and haltere (h) is found throughout the third instar larval stages. (H-K) In situ hybridization of wild-type eye-antenna discs using rn-specific (H), roe-specific (I,K) and rn/roe common 3′ (J) probes. (H) Expression of rn is found only in the antennal portion (arrow), and (I) roe only in the eye portion of the eye-antennal disc (arrowhead). (J) Using the rn/roe common 3′ probe we detect the combined pattern of rn and roe and both the antennal (arrow) and the eye portion (arrowhead) show expression. (K) Expression of roe is found in a band of 4-6 cells at the morphogenetic furrow.

Rescue of the rn leg phenotype. (A-H) Adult male forelegs and (I) quantification of the number of tarsal segments. (A) Wild-type leg with sex comb (s.c.), 5 distinct tarsal segments (T1-5), and a claw at the tip of the 5th tarsus. (B) rn¹⁶/rn²⁰, a genetic null. The sex comb is completely missing in all cases and the five tarsi appear fused into one segment. Note, however, that the claw is still present. (C) rn^Δ2–2/rn²⁰, which acts as a genetic null. (D) rn^GAL4#5/rn²⁰, a hypomorphic allelic combination. The sex comb is present and appears normal. The claw is normal. However, the tarsi are fused into two to three tarsal-like segments. (E) UAS-rn /+;rn^GAL4#5/+. UAS-rn causes no obvious disruption of the leg. (F) Rescue of rn mutants in UAS-rn /+;rn^GAL4#5/rn²⁰. The rn cDNA, expressed using the GAL4/UAS system, rescues the leg phenotype. (G) UAS-roe/+;rn^GAL4#5/+. UAS-roe has negative effects when expressed in the rn pattern. (H) UAS-roe/+;rn^GAL4#5/rn²⁰. UAS-roe is unable to rescue rn mutants. (I) Quantification of tarsi in wild type, rn mutants and rescue flies. The apparent number of tarsal segments was determined in rn mutants and rescue flies (>20 flies and >120 legs/genotype). The rescue is statistically significant to P<0.001 using a two-tailed t-test. Error bars represent the standard deviation. Temperature for rescue is 18°C though similar results were observed at 22°C (not shown).

(A,B) Expression of sev-GAL4, visualized by crossing to UAS-lacZ and staining for anti-β-gal (green), in relation to Elav (red) expression. Expression of sev-GAL4 commences posterior to the morphogenetic furrow in subsets of photoreceptors, as evident by the overlap with Elav. In addition, sev-GAL4 expression is observed in cells adjacent to the developing photoreceptors, most likely corresponding to mystery and cone cells. (C,D) Misexpression of rn (C) in UAS-rn/+;GMR-GAL4/+ and roe (D) UAS-roe/+;GMR-GAL4/+ both lead to disruptions in the morphology and size of the adult eye. These include an apparent loss of pigment cells and bristle cells, as well as the presence of patches of necrotic tissue (black).

Rescue of roe. (A-D) Adult eyes and (E) quantification of photoreceptors. (A) Wild type. (B) rn¹⁶/rn²⁰ a roe null combination displays a small and rough eye. (C) UAS-roe can rescue roe. sev-GAL4/UAS-roe;rn¹⁶/rn²⁰ have larger and apparently less rough eyes than roe. (D) UAS-rn fails to rescue roe. sev-GAL4/UAS-rn;rn¹⁶/rn²⁰ eye shows no sign of rescue, instead an apparent enhancement of the roe phenotype. (E) Quantification of the rescue of roe mutants. Adult eyes were sectioned and the number of Elav-positive cells in each ommatidia was counted. Wild-type ommatidia carry the typical seven (R1-7) photoreceptors (the R8 photoreceptor cell body is located slightly offset and was not included). In roe mutants we find an average of 5.7 photoreceptors, which is rescued to 6.3 by providing roe activity using UAS-roe (P<0.04). Using UAS-rn we find no evidence of rescue and roe ommatidia contain an average of 5.6 photoreceptors per ommatidia. In addition we find ommatidia with 4 or sometimes only 3 photoreceptors, something not observed in the other genotypes, indicating a negative action of UAS-rn.

(A-D) Mid third instar larval and (E,F) pupal leg imaginal discs. Expression of Bab in wild-type (A) and rn¹⁶/rn²⁰ (B) leg discs show that neither the pattern nor the intensity of Bab staining is affected in rn. Expression of Ser in wild type (C,E) and rn¹⁶/rn²⁰ (D,F). In wild-type leg discs (C) Ser expression is observed as a ring in the first tarsal fold (arrow) and in the proximal furrow (arrowhead). In rn¹⁶/rn²⁰ leg discs (D) Ser expression is down-regulated in the tarsal fold (arrow) but maintained in the proximal furrow (arrowhead). Similarly, in pupal leg discs Ser appears to be down-regulated in the presumptive tarsal area where rn is normally expressed. Compare bracketed areas in (E) wild type and in (F) rn¹⁶/rn²⁰.