A genetic screen of transcription factors in the Drosophila melanogaster abdomen identifies novel pigmentation genes

Abstract

Gene regulatory networks specify the gene expression patterns needed for traits to develop. Differences in these networks can result in phenotypic differences between organisms. Although loss-of-function genetic screens can identify genes necessary for trait formation, gain-of-function screens can overcome genetic redundancy and identify loci whose expression is sufficient to alter trait formation. Here, we leveraged transgenic lines from the Transgenic RNAi Project at Harvard Medical School to perform both gain- and loss-of-function CRISPR/Cas9 screens for abdominal pigmentation phenotypes. We identified measurable effects on pigmentation patterns in the Drosophila melanogaster abdomen for 21 of 55 transcription factors in gain-of-function experiments and 7 of 16 tested by loss-of-function experiments. These included well-characterized pigmentation genes, such as bab1 and dsx, and transcription factors that had no known role in pigmentation, such as slp2. Finally, this screen was partially conducted by undergraduate students in a Genetics Laboratory course during the spring semesters of 2021 and 2022. We found this screen to be a successful model for student engagement in research in an undergraduate laboratory course that can be readily adapted to evaluate the effect of hundreds of genes on many different Drosophila traits, with minimal resources.

Keywords: Drosophila; CRISPR/Cas9; abdomen; development; gene regulation; pigmentation.

Fig. 1.

The TRiP transgenic gene editing system can be used for both overexpressing and knocking out genes of interest. a) Virgin females expressing either Cas9 or deactivated Cas9 fused to the VPR activation domain (dCas9–VPR) expressed in the abdominal midline driven by pnr were crossed to males with ubiquitous single guide RNAs. Progeny who received the Cas9 or dCas9–VPR–Gal4 driver and sgRNA were selected on the absence of dominant markers. b) Cartoon illustrates the 3 traits measured in this study: midline width, background color, and A6 stripe width. c) Genotypes of the parents and progeny in the knockout cross. In the knockout crosses, Cas9 can induce a frameshift mutation in the gene targeted by guide RNAs. These mutant gene alleles would produce a nonfunctional protein in the pnr expression domain. d) Genotypes of the parents and progeny in the overexpression cross. In the overexpression crosses, dCas9–VPR binds the promoter for a gene targeted by guide RNAs, recruiting transcription machinery to the gene of interest and ectopically expressing the gene in the pnr expression domain.

Fig. 2.

Changes among female flies to the anterior–posterior A6 stripe length, midline stripe width, and background pigmentation were observed in overexpression and knockout cross progeny. Two-tailed Student's t tests were used to compare targeted with control crosses, P < 0.001. a) Boxplot showing measurements of the A6 stripe in female flies compared with controls. Cartoon illustrates region of the fly measured (pink) and region affected by gene editing (green). b) Boxplot showing measurements of the midline stripe, assessed in the A4 segment of female flies, compared with controls. Cartoon illustrates region of the fly measured (pink) and region affected by gene editing (green). c) Boxplot showing calculated percent darkness of the A4 segment in female flies with a targeted transcription factor gene compared with controls. Cartoon illustrates region of the fly measured (pink) and region experiencing gene editing activity (green).

Fig. 3.

Noteworthy knockout tergite pigmentation phenotypes. Progeny of knockout crosses. Blue brackets highlight some notable phenotypes that were seen after imaging multiple samples, but are not representative of quantitative data. a) Knockout control abdomens. b–f) Gene knockouts featured here are b) CG10348, c) dsx, d) CG17806, e) sd, and f) spab. Knockouts for CG10348 and dsx demonstrate decreased pigmentation in the midline and increased pigmentation in the female A5/A6 regions, respectively. CG17806, sd, and spab knockouts resulted in shifts in background coloration. All other knockout crosses did not have significant phenotypes in the areas measured. KO, knockout.

Fig. 4.

Overexpression phenotypes with an increase of melanic pigmentation. Progeny of overexpression crosses. Blue brackets highlight some notable increases in dark pigmentation that were observed after imaging multiple samples, but are not representative of quantitative data. a) Overexpression control abdomens. b–l) Overexpressed genes featured here are b) abd-A, c) ato, d) unpg, e) C15, f) bigmax, g) Eip78C, h) Hr4, i) sbb, j) Su(var)3-7, k) tap, and (l) ush. OE, overexpression.

Fig. 5.

Overexpression phenotypes with a decrease in melanic pigmentation. Progeny of overexpression crosses. Blue brackets highlight some notable decreases in dark pigmentation that were observed across multiple samples, but are not representative of quantitative data. a) Overexpression control abdomens. b–g) Overexpressed genes featured here are b) bab1, c) Hey, d) Hr38, e) pdm3), f) slp2, and g) lab.

Fig. 6.

Defects in the development of the thorax and abdomen. a) Control thorax. b) The gene ato produces additional bristles on the thorax when overexpressed. c–e) When overexpressed, the genes c) abd-A, d) lab, and e) unpg produce a defect in the thorax. f–h) When knocked out, the genes f) Su(var)2-10, g) Su(z)12, and h) Ssrp produce a defect in the thorax. i) Control abdomens. j–l) When knocked out, the genes j) M1BP, k) Ssrp, and l) Su(z)12 produce a defect in the midline of the abdomen.