(A) The Drosophila sex determination hierarchy consists of an alternative pre-mRNA splicing cascade, culminating in the production of sex-specific transcription factors encoded by dsx and fru P1. In Drosophila, the primary determinant of sex is the X chromosome to autosome ratio. In females, this ratio is one and leads to the production of SXL. sxl, tra, and tra-2 encode pre-mRNA splicing factors. SXL regulates sxl and tra pre-mRNA splicing, resulting in the production of TRA in females. TRA together with TRA-2 regulates the splicing of dsx and fru P1 pre-mRNAs. In males, dsx and fru P1 are spliced by the default pathway. Arrows indicate regulation of pre-mRNA splicing, and hourglass shapes indicate translation. The dotted line indicates inhibition of the dosage compensation (DC) pathway.
(B) Experimental overview for microarray comparisons. The genotypes and phenotypes of animals from which the microarray probe pairs are derived are indicated in columns with headings Target 1 and Target 2. For each genotype, the chromosomal sex is shown in parentheses. The rationale for the experimental comparison is indicated. The number of genes that met the statistical criteria (see text) is indicated in the column with heading “Genes.”
doi:0.1371/journal.pgen.0030216.g001

The 40 genes listed here showed significant differential expression in wild-type female versus male, female versus tra pseudomale, female versus dsx^D pseudomale, female versus dsx null XX, and male versus dsx null XY in 0–24-h adult heads. Columns contain FC ratio values from the microarray data: F/M, female over male; F/tra, female over tra; F/dsx^D, female over dsx^D; F/dsx, female over dsx null XX; M/dsx, male over dsx null XY. The white region contains genes (34) that showed activation or repression by both DSX^F and DSX^M in the same direction, in both sexes. The gray region contains genes (six) that showed DSX^F and DSX^M acting in opposite directions. (I.) Activation by both DSX^F and DSX^M with higher level of activation by DSX^F. (II.) Repression by both DSX^F and DSX^M with higher level of repression by DSX^F. (III.) Repression by both DSX^F and DSX^M with higher level of repression by DSX^M. (IV.) Activation by DSX^F and repression by DSX^M. (V.) Repression by DSX^F and activation by DSX^M. The symbols (+/−) in the first column are qualitative representations of activation (+) or repression (−), based on the microarray data.
doi:0.1371/journal.pgen.0030216.g003

In all panels, 0–24-h adult head frozen-sections are shown.
(A–C) In situ hybridization using cpn anti-sense probe.
(D–F) In situ hybridization using cpn sense control probe. The genotype of the tissue is shown at the left of the panels.
doi:0.1371/journal.pgen.0030216.g005

(A) Time (s) to first wing extension for CS male flies (n = 30, mean = 86.7, SE = 9.9) is significantly different (one-tailed t-test) than dpr-GAL4 (n = 26, mean = 42.3, SE = 9.6) (p < 2.95E-5), dpr-GAL4;UAS-gfp^RNAi (n = 50, mean = 39.4, SE = 9.3) (p < 1.73E-8), and dpr-GAL4;UAS-fru^RNAi (n = 66, mean = 26.2, SE = 3.1) (p < 3.00E-10) flies. Also, when compared to UAS-fru^RNAi (n = 38, mean = 85.3, SE = 11.7) male control flies, the time to first wing extension is significantly reduced in dpr-GAL4 (p < 2.21E-4), dpr-GAL4;UAS-gfp^RNAi (p < 2.28E-7), and dpr-GAL4;UAS-fru^RNAi (p < 2.71E-9) flies.
(B) The time to first attempted copulation (s) is significantly reduced in dpr-GAL4;UAS-fru^RNAi (n = 63, mean = 95.3, SE = 10.8) flies compared to CS (n = 27, mean = 178.3, SE = 18.3) (p < 3.20E-5), UAS-fru^RNAi (n = 33, mean = 218.5, SE=23.6) (p < 1.03E-6), dpr-GAL4;UAS-gfp^RNAi (n = 45, mean = 187.5, SE = 20.9) (p < 3.41E-5), and dpr-GAL4 (n = 24, mean = 202.2, SE = 31.1) (p < 1.45E-4) male flies. Error bars represent SE. Mann-Whitney rank-sum test: **, p < 1.0E-3; ***, p < 1.0E-4; ****, p < 1.0E-8.
doi:0.1371/journal.pgen.0030216.g007

(A) Overview of results from statistical tests of microarray data leading to the identification of sex-differentially expressed genes regulated by tra, dsx, or neither in adult head tissues.
(B) Overview of results from statistical tests of microarray data leading to the identification of genes regulated downstream of DSX activity and the modes of DSX regulation for these genes.
doi:0.1371/journal.pgen.0030216.g002

Fold-change (FC) from averaged microarray data for each validated gene is shown at the top of each panel. Each genotype of the target pair compared in the microarray experiment is indicated at either side of the gradient bar. White in the gradient bar represents no difference in gene expression between the two genotypes (FC = 1), and the white to black gradient indicates increasingly higher expression differences in the genotypes indicated. The FC in gene expression between the two genotypes is shown below the gradient bar in the gray-scaled box. At the bottom of each panel are the results of RT-PCR analysis, for which the expression of a given gene relative to rp49 ribosomal expression is reported. Microarray and RT-PCR results for (A) cpn, (B) dpr-family members, and (C) capa. (A) RT-PCR showed that cpn expression relative to rp49 expression was significantly higher (p < 0.015, 3.5-fold) in dsx^D female (mean = 9.32, SE = 1.99) than in CS female (mean = 2.64, SE = 0.49) adult heads.
(B) dpr2, dpr7, dpr14, and dpr20 microarray expression, with higher expression in males than fru P1 males. dpr RT-PCR expression relative to rp49 expression was significantly higher (p < 0.05, 4.4-fold) in CS male (mean = 3.47, SE = 0.94) than fru P1 male (mean = 0.78, SE = 0.004) adult heads.
(C) RT-PCR showed that capa expression relative to rp49 expression was significantly higher (p < 9.2E-4, 1.4-fold) in CS male (mean = 0.063, SE = 0.003) than fru P1 male-dissected (mean = 0.047, SE = 0.003) CNS tissue. Error bars represent SE. One-tailed t-test: *, p < 0.05; **, p < 0.001.
doi:0.1371/journal.pgen.0030216.g004

In all images, dpr-GAL4; UAS-nlsGFP (hereafter called dpr expression) is shown in green.
(A–C) dpr expression in the 0–24-h adult male CNS. (A) Anterior brain z-section. (B) Posterior brain z-section. (C) Anterior ventral nerve cord z-section.
(D–F) dpr expression in the 0–24 h adult female CNS. (D) Anterior brain z-section. (E) Posterior brain z-section. (F) Anterior ventral nerve cord z-section.
(G–H) dpr and FRU^M (red, immunohistochemistry) expression in the 0–24-h adult male CNS. (G) dpr and FRU^M expression in the anterior male brain. Colocalized expression of dpr (green) and FRU^M (red) is observed in the median bundle neurons (white box). dpr is expressed in cells in the axonal tract extending from the median bundle (arrow), and in the pars intercerebralis (PI). (H) dpr and FRU^M expression in the 0–24-h adult male ventral nerve cord. Colocalized expression of dpr (green) and FRU^M (red) is observed in the first thoracic ganglion (white box). (J) dpr (green) and FRU^M (red) expression in 40× magnification of median bundle neuron region (white box in [G], independent brain). (K) dpr (green) and FRU^M (red) expression in 40× magnification of first thoracic ganglion region (white box in [H], independent VNC). dpr and FRU^M expression in the 48-h pupal male brain (I) and ventral nerve cord (L). Z-section thickness was ∼1 μm for (A–I), (L), and ∼0.5 μm for 40× images (J–K).
doi:0.1371/journal.pgen.0030216.g006