|Select image to enlarge|
Sectorial domains of gene expression in the leg disc (which become stripes along the leg) plus one annular domain (hth). The flow of control is from en (upper right), which establishes the A-P axis, through dpp and wg, which polarize the D-V axis, to ac (lower left), which allocates cells to bristle vs. non-bristle fates.
Black areas chart mRNA (transcribed from endogenous or reporter gene) or protein levels during embryogenesis (E), late 3rd instar (3), early pupal period (P) or adult (A) stage. Protein levels are given for Hh (based on data from wing discs Δ), Dpp (range is not exactly known [1169, 4265]), and Wg Δ. Shading denotes degree of expression. Areas are mapped onto a disc, though some genes turn ON only after 3rd instar (e.g., ac) or stay ON into adulthood (e.g., en). The span of stages (e.g., '3-P') indicates when expression has been seen but does not connote confinement to that period. Wiring (> activation; | inhibition) illustrates a few genetic interactions (see text or sources cited below). AND means that both inputs are needed to activate the target gene. Thus, the dorsal hairy stripe is caused jointly by Dpp and Hh (neither alone is sufficient), while the ventral stripe is caused jointly by Wg and Hh . (The other two hairy stripes remain enigmatic.)
Genes (DO = details omitted): ac (achaete; DO: expression areas proximal to the tibia) , brk (brinker) , ci (cubitus interruptus) , dpp (decapentaplegic) [2069, 3122]Δ, en (engrailed) Δ, h (hairy) Δ, H15 (a T-box relative of omb) Δ, hh (hedgehog) Δ, hth (homothorax) Δ, omb (optomotor-blind) Δ, wg (wingless, whose stripe in pupal legs is ~5 cells wide ) Δ.
All of these genes are known to have LOF or GOF effects on pattern, except H15 (whose null phenotype is wild-type ) and omb. Two genes whose expression resembles that of H15 -- nkd  and Dfz3  (not shown) -- also have no LOF defects. Enhancer-trap line c409 (not shown) matches ci's pattern , as does aldehyde oxidase [2340, 4051]. See  and FlyView (flyview.uni-muenster.de)  for other enhancer-trap expression patterns.
Areas of expression that are not wedge-shaped (omitted) include arcs (e.g., GAL30A , grain , Sex combs reduced Δ, Ultrabithorax ), parabolas , a spot (at femoral chordotonal organ) , etc. [834, 1574, 2684, 4522]. Augmented from .
In some specimens the dpp-ON and wg-ON sectors appear as wide as shown [2954, 4277, 4849]Δ, but often they look narrower [231, 487, 3762]Δ. Likewise, the en domain sometimes displays a central acute angle as shown [2954, 3329]Δ, but it can also seem straight [1472, 4254] or even obtuse . These vagaries arise from variations in folding, viewing angle, and focal plane, as well as from inherent irregularities in the A/P (and other?) boundaries [1833, 4254]. In early pupae a gap arises between the en domain and the dorsal h sector , though it is not be as large as implied here.
N.B.: Hh, Dpp, and Wg only seem to define D-V sectors of h expression . Therefore, one remaining mystery is: what factors define h-ON sectors along the A-P axis? Another is: what factors define the edges of ac-ON stripes that are not created by h | ac inhibition?