Combinatorial control of temporal gene expression in the Drosophila wing by enhancers and core promoters

Abstract

Background: The transformation of a developing epithelium into an adult structure is a complex process, which often involves coordinated changes in cell proliferation, metabolism, adhesion, and shape. To identify genetic mechanisms that control epithelial differentiation, we analyzed the temporal patterns of gene expression during metamorphosis of the Drosophila wing.

Results: We found that a striking number of genes, approximately 50% of the Drosophila transcriptome, exhibited changes in expression during a time course of wing development. While cis-acting enhancer sequences clearly correlated with these changes, a stronger correlation was discovered between core-promoter types and the dynamic patterns of gene expression within this differentiating tissue. In support of the hypothesis that core-promoter type influences the dynamics of expression, expression levels of several TATA-box binding protein associated factors (TAFs) and other core promoter-associated components changed during this developmental time course, and a testes-specific TAF (tTAF) played a critical role in timing cellular differentiation within the wing.

Conclusions: Our results suggest that the combinatorial control of gene expression via cis-acting enhancer sequences and core-promoter types, determine the complex changes in gene expression that drive morphogenesis and terminal differentiation of the Drosophila wing epithelium.

Figure 1

Early stages of wing differentiation. (A) To illustrate the morphogenetic process of wing-disc elongation, wing tissue from late third larval instar (L3), 2 h after puparium formation (APF), and 36 h APF was dissected and stained for DNA. The wing margin (red), wing veins (orange), and notum (purple) are indicated in developing tissue and the adult fly. (B) Between L3 and 36 h APF, each wing epithelial cells adopts a cell-type-specific shape, and differentiates a wing hair. Time courses of wings stained for either DE-cadherin (to visualize apical cell shape) or F-actin (to visualize hair formation) are shown. Images are centered on a presumptive wing vein. Developmental stages are indicated. (C) Flow cytometric analysis demonstrates changes in DNA content associated with cell-cycle exit in the wing. At L3 and 2 h APF, presumptive wing cells asynchronously proliferate. By 6 h APF, most cells in the wing temporarily arrest in the G2 phase of the cell cycle, leading to a relatively synchronized final cell cycle between 14 and 24 h APF (represented here by 18 h APF). By 24 h APF, cell proliferation is no longer detected in the wing epithelium, and nearly all cells arrest with a G1 DNA content. Dissected wing tissues stained for DNA are shown for each developmental stage. (D) To determine the changes in gene expression associated with wing morphogenesis and cell cycle exit, RNA was collected from six developmental time points between L3 and 36 h APF (corresponding to images shown in (C)), and microarray analysis was performed. Using L3 as a reference sample, the number of transcripts that exhibit a significantly different level of expression is listed for each time point.

Figure 2

Dramatic changes in gene expression take place during wing differentiation. A compressed heat map shows significant global gene expression changes for 8338 genes in the wing between late L3 and 36 h APF (proliferating L3 wing tissue was used as the reference sample). Each row corresponds to a single gene and each column represents an individual time point. Expression values (log₂) are color coded according to the legend at the top. Using k-means clustering (see Materials and Methods), genes were grouped into 30 clusters based on expression profile similarities. Expression plots are shown for selected k-means clusters and approximately aligned to corresponding regions of the heat map. Plots of all 30 k-means clusters are shown in Additional file 3. For each cluster, the normalized log₂ expression level of each gene (grey lines) is plotted as a function of time (x-axis). The magenta line represents the average expression of all genes within a cluster. Examples of genes and enriched gene ontology terms for the selected clusters are listed at right.

Figure 3

Enrichment of core promoter types and regulatory motifs within gene expression clusters. (A) Clusters were grouped based on temporal patterns of expression, which are represented by the heat map (y-axis is time). Average expression values for genes within a cluster are shown. For each cluster, the most significantly enriched gene ontology (GO) term is listed (only when p < 10^-5). (B) All 30 clusters were examined for statistically significant (z-score > |3|) enrichment (red) or depletion (green) of ten core Drosophila promoter motifs (see Materials and Methods). The search space was defined as the 100 bp spanning the transcriptional start site (−60 to +40 bp). Certain core promoter motifs were frequently enriched in the same clusters. For example, DRE, CP-1, CP-7, and CP-6 motifs were typically enriched within a similar subset of clusters. In addition, these clusters tended to decrease expression over time. In contrast, INR, DPE, and MTE motifs were often found together in clusters that increased expression over time. TATA motifs were rarely enriched or depleted. (C) All 30 clusters were examined for significant enrichment or depletion of 87 known Drosophila transcription factor binding motifs. The search space was defined as the 1 kb immediately upstream of the transcriptional start site. A group of motifs known to regulate cell growth, protein synthesis, and the cell cycle (e.g., E2F, Myc, Dref, Mad/Med/Brinker) were enriched and frequently found together in clusters that decreased expression over time and upon cell-cycle exit. Clusters that decrease expression over time were also depleted for a group of motifs associated with tissue differentiation (e.g., Cad, Croc, En, Ubx). These differentiation-associated motifs were instead enriched in clusters that increased expression during pupal time points.

Figure 4

Motif-enrichment validations. Gene clusters with significant motif enrichment were compared to published datasets. For example, cluster 30 was enriched for EcR/USP binding sites, so genes within cluster 30 were compared to independently identified EcR/USP target genes (A). This analysis was repeated for clusters 29 and 13, which were both enriched for the starvation-associated motif (B), the Myc motif (C), and the Drosophila E2F motif (D). In every case but one, significant enrichment for independently validated target genes was observed. Confirmed Myc target genes within Cluster 13 are generally induced by Myc expression, whereas cluster 29 genes are generally repressed by Myc expression (see heat map (C)). Similarly, verified E2F targets within clusters 13 and 29 are generally upregulated and downregulated by E2F expression, respectively (see heat map and expression plot (D)).

Figure 5

Changes in core-promoter binding proteins may affect wing differentiation. (A) Expression data from the wing developmental time course for genes that encode core promoter binding proteins, general transcription components, and TBP-associated factors (TAFs). Each row corresponds to a single gene and each column represents an individual time point. Expression values (log₂) are color coded according to the legend at the bottom. Specific groups of genes are indicated. The complete clustering of all 72 genes is provided in Additional file 7. (B) RT-PCR and western blot analyses were used to examine RNA or protein levels for selected factors in pupal wing tissue. By microarray analysis sa, which is a tTAF, did not increase expression at the time points shown. This was verified by western blot. Levels of the tTAF Mia increased in the wing, both by RT-PCR (at 24 h APF) and western blot (at 36 h APF). Mia-specific bands at 35 and 70 kD (indicated by asterisks) are observed with anti-Mia antibody in select tissues (N. Haugen and D. Wassarman, personal communication). BEAF decreased in the wing over time. Acetylated histone H4 served as a loading control. (C-F) Using UAS-RNAi lines, engrailed-Gal4 was used to inhibit either white (w), or nht in the posterior wing. Tubulin-Gal80^TS was used to limit RNAi expression from the second larval instar until eclosion. Compared to controls (C), nht reduction affected posterior wing growth and cuticle integrity (D). Staining for F-actin reveals developing wing hairs at 34 h APF (E). Compared to the anterior control, expression of nht RNAi led to delay in wing hair formation in the posterior (F). nht^z5946 hemizygotes, exhibit ectopic vein and multiple wing hair phenotypes (G, inset). Two nht alleles in trans (nht^z5347/nht^z5946) at 25°C exhibit ectopic vein (indicated by arrowheads in H, with 20X magnification in I). Approximately 10% of nht^z5347/nht^z5946 females exhibit patches of thin, small wing hairs (outlined by dashed line in J).