pubmed.ncbi.nlm.nih.gov

Genome-wide features of neuroendocrine regulation in Drosophila by the basic helix-loop-helix transcription factor DIMMED - PubMed

  • ️Thu Jan 01 2015

Genome-wide features of neuroendocrine regulation in Drosophila by the basic helix-loop-helix transcription factor DIMMED

Tarik Hadžić et al. Nucleic Acids Res. 2015.

Abstract

Neuroendocrine (NE) cells use large dense core vesicles (LDCVs) to traffic, process, store and secrete neuropeptide hormones through the regulated secretory pathway. The dimmed (DIMM) basic helix-loop-helix transcription factor of Drosophila controls the level of regulated secretory activity in NE cells. To pursue its mechanisms, we have performed two independent genome-wide analyses of DIMM's activities: (i) in vivo chromatin immunoprecipitation (ChIP) to define genomic sites of DIMM occupancy and (ii) deep sequencing of purified DIMM neurons to characterize their transcriptional profile. By this combined approach, we showed that DIMM binds to conserved E-boxes in enhancers of 212 genes whose expression is enriched in DIMM-expressing NE cells. DIMM binds preferentially to certain E-boxes within first introns of specific gene isoforms. Statistical machine learning revealed that flanking regions of putative DIMM binding sites contribute to its DNA binding specificity. DIMM's transcriptional repertoire features at least 20 LDCV constituents. In addition, DIMM notably targets the pro-secretory transcription factor, creb-A, but significantly, DIMM does not target any neuropeptide genes. DIMM therefore prescribes the scale of secretory activity in NE neurons, by a systematic control of both proximal and distal points in the regulated secretory pathway.

© The Author(s) 2015. Published by Oxford University Press on behalf of Nucleic Acids Research.

PubMed Disclaimer

Figures

Figure 1.
Figure 1.

Isolation of c929 (DIMM+) neurons from the adult brain. (A) Sample dissected adult brain used for sorting. (B) FACS-sorting of DIMM-positive neurons (bottom right) and DIMM-negative randomly sorted neurons (bottom left) from dissected adult fly brains. (C) Light microscopy of control gate cells lacking GFP expression (bottom) compared to high fluorescence gate cells (top). (D, D’) Electron micrographs depicting cells representative of ∼10 and ∼90%, respectively, of c929>GFP+ neurons after FACS and EM processing. (E) Electron micrograph showing a representative c929>GFP− neuron.

Figure 2.
Figure 2.

Deep sequencing of DIMM+ cell and control cell transcriptomes reveals enrichment of known peptidergic markers and little overlap with markers of major non-peptidergic neuronal subtypes. (A) The c929>GFP+ sample is highly enriched for dimm RNA (top). DIMM's direct target Phm is ∼3.5-fold enriched in the c929-GFP+ sample compared to the c929-GFP− sample. Likewise the known neurally-expressed bioactive peptides ilp2, ilp3 and Pdf transcripts are all enriched in c929-GFP+ neurons, while the ilp1 and ilp4 transcripts (poorly expressed in the brain) are not detected in either transcriptome. (B) Most, but not all, neuropeptide RNAs are strongly enriched in the c929-GFP+ cells compared to controls. This distribution compares favorably to known neuropeptide distributions among DIMM+ neurons (52).(C) c929-GFP+ cells are not enriched for gene markers of GABAergic, dopaminergic, cholinergic and glutamatergic neurons, compared to the c929-GFP− cell population.

Figure 3.
Figure 3.

DIMM binds to more than 300 loci. In addition to binding known target genes such as Phm and CG1275, DIMM binds hundreds of novel targets. (A) The schematic depicts genotypes and the experimental setup used for ChIP-chip. (B) DIMM binding is enriched near transcription start sites and (C) depleted near transcription termination sites. (D) A ‘metagene’ profile derived from an average of all genes in the genome shows striking enrichment of DIMM binding in the promoter/5′UTR sites, often peaking in the first 300 bp of a metagene. (E) ChIP-chip recapitulates known in vivo DIMM binding to the first intron of Phm. Novel targets include Creb/ATF family pro-secretory transcription factors CrebA and cryptocephal, as well as previously uncharacterized genes, such as CG4577. (F) DIMM binds preferentially to the first introns and transcriptional start sites, whereas DIMM binding over remaining gene introns is depleted.

Figure 4.
Figure 4.

DIMM binds to conserved E-boxes located close to the centers of DIMM ChIP-chip binding peaks. (A, B) DIMM preferentially binds to two palindromic E-boxes, CATATG and CAGCTG. Both E-boxes match precisely known DIMM E-boxes in the Phm 1st intron enhancer (21) ; the CATATG E-box matches the SELEX-predicted DIMM-binding E-box (63). (C) Strand-specific conservation of the CATATG E-box in intronic DIMM binding peaks compared to ∼800 randomly selected intronic CATATG E-boxes. (D) Strand-specific conservation of the CAGCTG E-box in intronic DIMM binding peaks compared to ∼800 randomly selected intronic CAGCTG E-boxes. (E) The CATATG E-box shows an enrichment for the CCATATG variant, particularly around the centers of binding peaks. (F) The CATATG E-box is enriched in the center of DIMM-bound peaks (red) and lacks enrichment in CLOCK-bound peaks (blue). CLOCK data from (26). (G) Multiple CATATG and CAGCTG E-boxes are seen dispersed throughout peak length, but they appear concentrated near the center of DIMM-bound peaks. (H) DIMM-occupied genomic regions are GC-rich, particularly near the centers.

Figure 5.
Figure 5.

Receiver operating characteristic (ROC) curve of different models for classification of DIMM bound versus unbound putative sites based on DNA sequence and shape features derived from the binding site's flanking regions. Area under the curve (AUC) values were calculated from true positive rates plotted against false positive rates for classification of bound versus unbound sequences containing the CATATG E-box. Different models, including a nucleotide sequence model (red), a DNA shape model (combining minor groove width, roll, propeller twist and helix twist; blue), a shape-augmented sequence model (using both sequence and the four DNA shape features; green) and a sequence-based model augmented by only minor groove width (MGW; magenta), were used to distinguish between bound and unbound DIMM target sites. The AUC values for the CATATG E-box are given in the legend, and the respective values for the CAGCTG E-box were 0.5980, 0.5921, 0.5941 and 0.5999.

Figure 6.
Figure 6.

DIMM-binding sites correlate with expression of structural and functional elements of the regulated secretory pathway. (A) DIMM binds to a core set of genes encoding biosynthetic, neuropeptide processing enzymes that are highly expressed by c929-DIMM+ cells. Gene expression from this study compared to RNA-Seq data from (55) obtained from purified nuclei of octopaminergic and Kenyon neurons adult brain neurons. (B) Intersection of ChIP-chip identified DIMM target genes and the c929-DIMM+ cell transcriptome. (C) DIMM does not bind to the defined neuropeptide-encoding genes of the Drosophila genome.

Figure 7.
Figure 7.

Schematic overview of the DIMM transcriptional output responsible for orchestration of the RSP in NE cells. Illustration of a small conventional neuron (bottom) next to a larger DIMM+ NE cell. Various DIMM targets are selected to feature the proximate and distal levels of the secretory pathway that DIMM supports—including nuclear factors that govern specific gene expression, RNA-binding proteins that govern specific translation, Golgi-related factors that regulate the secretory process and LDCV components (in the expanded box).

Similar articles

Cited by

References

    1. Burbach J.P., Luckman S.M., Murphy D., Gainer H. Gene regulation in the magnocellular hypothalamo-neurohypophysial system. Physiol. Rev. 2001;81:1197–1267. - PubMed
    1. Helfrich-Forster C., Winter C., Hofbauer A., Hall J.C., Stanewsky R. The circadian clock of fruit flies is blind after elimination of all known photoreceptors. Neuron. 2001;30:249–261. - PubMed
    1. Rao S., Lang C., Levitan E.S., Deitcher D.L. Visualization of neuropeptide expression, transport, and exocytosis in Drosophila melanogaster. J. Neurobiol. 2001;49:159–172. - PubMed
    1. Beuret N., Stettler H., Renold A., Rutishauser J., Spiess M. Expression of regulated secretory proteins is sufficient to generate granule-like structures in constitutively secreting cells. J. Biol. Chem. 2004;279:20242–20249. - PubMed
    1. Lee A.H., Chu G.C., Iwakoshi N.N., Glimcher L.H. XBP-1 is required for biogenesis of cellular secretory machinery of exocrine glands. EMBO J. 2005;24:4368–4380. - PMC - PubMed

Publication types

MeSH terms

Substances