Saturday

Early zygotic dosage compensation in Drosophila melanogaster is Sxl dependent.

• Susan E. Lott1,2, Jacqueline E. Villalta3, Michael B. Eisen2,3. 1) Evolution and Ecology, University of California, Davis, Davis, CA; 2) Department of Molecular and Cell Biology, University of California, Berkeley, CA; 3) Howard Hughes Medical Institute, University of California, Berkeley, CA.
• Hox gene intro
• DNA binding domains highly conserved. Yet still high specificity
• Ubx activates svb responsible for producing trichomes
• Ubx everywhere, get A1 everywhere (has tricomes)
• ID enhancers (E3 and 7), both respond to Ubx the same way
• No canonical binding sites for homeobox / Ubx.
• chopped up enhancer and looked for which reions can bind ubx in vitro
• find cluster of low affinity Ubx binding sites.
• delete individual low affinity sits still see expression
• delete 2 (or 3) binding sites, abrogate expression.
• SELEX-seq — deep sequence screening for affinity range of TFs for all sequences.
• (AbdA binds same sequence as Ubx)
• Lowest affinity sites of the range that Ubx binds are the most Ubx specific.
• increase binding strength, start getting more expression in AbdA region.
• add high affinity binding sites more ectopic expression in anterior range, (less expression in native sites
• Ubx hets in minus one binding site, no change in WT, but change in Ubx hets
• low or high temp treatments also have stronger effects on site loss than wt at 25C

Low affinity binding site clusters confer Hox specificity and regulatory robustness.

• Justin Crocker1, Namiko Abe2, Lucrezia Rinaldi2, Alistair P. McGregor3, Nicolás Frankel4, Shu Wang5, Ahmad Alsawadi6,7, Philippe Valenti6,7, Serge Plaza6,7, François Payre6,7, Richard S. Mann2, David L. Stern1. 1) HHMI Janelia, Ashburn, VA; 2) Columbia University Medical Center; 3) Oxford Brookes University; 4) Universidad de Buenos Aires; 5) New Jersey Neuroscience Institute; 6) Centre de Biologie du Développement; 7) CNRS.

A cell type specific transcriptional repressor directs selective upregulation of terminal differentiation program.

• Jongmin Kim, Margaret Fuller. Stanford University, Stanford, CA.
• testis zinc finger was necessary for proper differentiation in the testes
• Made antibody, find protein is nuclear and only in differentiating cells not stem cells.
• phenotype similar to tMAC, another TF which activates testes specific transcripts.
• most genes strongly affected in tMAC not affected in tZNF
• group of genes strongly upregulated in tZNF KO: gut and neuronal RNAs
• tZnF is a repressor, tMAC activates a broad swath of genes, and tZnF keeps off a substantial fraction of these that are not wanted in testes.
• maybe tMAC has promiscuous binding? (what conserves these sites?)

Genome-wide futile cycling by Hairy transcriptional repressor reveals mechanism for development of nascent gene regulatory networks.

• Kurtulus Kok2, Ahmet Ay3, David Arnosti1,2. 1) Dept Biochem & Molec Biol, Michigan State Univ, East Lansing, MI; 2) Program in Genetics, Michigan State University, East Lansing, MI; 3) Departments of Biology and Mathematics, Colgate University, Hamilton NY.
• “shotgun model of transcriptional regulation”
• “Tijan era of precise TF binding is falling apart” – TFs bind where they are supposed to and where they are not supposed to.
• Fisher et al 2012 — lower occupancy sites tend not to drive expression — truly background binding
• early embryo: broad activators and specificity in short range repression.
• Short range repressors and long range repressors
• both short and long range interact with both Groucho and CtBP. Still a mystery.
• Chromatin?
• Hairy represses H4Ac across a long domain. Knirps has a very local effect on H4Ac
• over-express Hairy throughout the genome.
• more sites change than transcription change
• Hairy modifies chromatin at some sites without changing transcription.
• futile cycling? genome wide effects that may have no direct consequence?
• Hairy may target silent genes it doesn’t care about.
• How many active genes get Hairy binding when your over-express it?
• Hairy does not have epigentic

Single base differences in a shared cis-regulatory element are critical for rhodopsin expression in distinct photoreceptor subtypes.

• Jens Rister, Claude Desplan. Department of Biology, New York University, New York City, NY.
• Broad ad restricted genes have a shared motif (called P3 or RCSI 11 bp)
• trpl GFP observe reporter in all cells, mutate motif, lose expression.
• also true for the motif in rhodopsin
• restricted rhodopsin actually have varients of these motifs.
• specific, subtle basepair changes completely define this specificity
• Question: are these motifs multermized in the enhancers? Comment on the organization of these sites relative to the promoter?
• change single basepair in the site and expand from subset of binding sites to all cells.
• created a Pph13 site? Yes. Can mutate
• see partial repeats upstream required for full expression.
• combine two restricted sites — get no expresion

Integration of repressive and patterning inputs at cardiac gene loci.

• Jemma Webber, (Rebay lab)
• Yan and Pointed regulate genes downstream of RTK signalling in bistable way.
• Yan (repressor) and pointed regulate muscle heart enhancer
• Yan binds upstream of eve promoter to D1 enhancer (Webber 2013)
• long range interactions occur betwen the HME and D1 (how do you distinguish these from enhancer promoter interactions)
• Pnt and Yan have similar binding profiles at eve locus.
• imaging shows Pnt and Yan colocalize in several of the mesoderm cells (though both have complex patterns)
• sequential ChIP fails to detect co-occupancy
• Pnt recruitment at D1 still occurs in Yan null
• Yan and Pnt co-occupy D1 region with Gro.
• Pnt is required for Gro recruitment (reduced in pnt null). Pnt may also be a repressor.

Differential binding and activation of enhancers by Bcd and Otd in the embryo.

• Rhea Datta, Danyang Yu, Stephen Small
• Bcd is missing from the genomes of most insects
• Otd plays a Bcd like role in other insects
• both K50 Homeodomain TFs TAATCc have same binding specificity
• does Otd bind Bcd targets? Bcd ~3000 peaks, Otd only 500, so no:
• Question: is the difference in Otd and Bcd binding concentration dependent?
• in late embryo Otd binds amny of the regions (870) of the regions bound by Bcd early
• Bcd and Otd bound to enhancers that are non-functional early.
• how are late stage enhancers kept off at the early stage?
• replace bcd with maternal gradient of Otd (otd/bcd swap):
• does not result in rescue: looks like bcd mutant by in large
• activates only a subset of Bcd target genes.
• swap bcd homeodomain in Otd: get rescue. Vice versa, no rescue.
• Use Bulyk micro-arrays to compare Bcd and Otd binding site strengths.
• But: Otd preferred sites are enriched in Bcd bound regions.
• Zelda and Hb sites are important
• Add Zelda sites to late acting enhancers — some of these get activation
• late responsive sites are enriched in tramtrack (maternally provided, rapidly degraded). Remove tramtrack repressor sites in late enhancer, get activation!
• what fraction have tramtrack sites?

Transcriptional activation by a low-complexity domain is a conserved feature of Zelda and orthologous proteins.

• Danielle Hamm, Eliana Bondra, Melissa Harrison. Dept. of Biomolecular Chemistry, University of Wisconsin, Madison, WI.
• MZT and Zelda
• how does Zelda mediate genome activation?
• 4 zinc finger region forms the DNA binding domain, accounts for 100% of the footprint.
• Mutation in any of zinc fingers abrogated transcription. Conclude all are necessary for binding.
• There is a splice form variant that lacks 3 of these 4 (but has most of the other sequence), ZLD-PD. No transcription activation from Zld-PD
• Zld-Pd reduces activitiy of Zld-PA (full-length)
• N terminal domain still essential for activity. Fused this to Gal4-DB. test with UAS-firefly.
• low complexity region 904-1300 has activator effect
• this plus the DNA binding Zn-fingers are sufficient to activate transcription.
• all zelda orthogos can activate (somewhat) transcription in D mel (out through Anaopholese)
• alternative splicing could be used to stop zelda activity.

Discovery of Novel Enhancers Using Natural Variation. Ashley Jermusyk, Sarah Gharavi, Gregory Reeves. Chemical and Biomedical Engineering, North Carolina State University, Raleigh, NC.

• using natural variation in DGRP lines to ID enhancers
• Quantifying variation in Kr posterior border between lines
• Question: have you tried hierarchical clustering on changes between lines? See if this clustering reflects correlations in mutations.
• grouped SNPs into 100 bp to 1 kb test enhancers.
• Question: did you swap back any of these SNPs?

Shadow enhancers enable Hunchback bifunctionality in the Drosophila embryo.

• M. Staller, B. Vincent, M. Bragdon, J. Estrada, Z. Wunderlich, A. DePace. Systems Biol, Harvard Med Sch, Boston, MA.
• eve 3/7 regulation
• general approach: map TF concentration to mRNA concentration with linear functions.
• Hb repressor model sufficient to explain 3/7 enhancer
• eve 3/7 expand in snaE>Hb. need to invoke Hb as an activator on 3/7
• How is Hb bifunctionality encoded in the genome?
• eve 3/7 enhancer doesn’t bend like the wildtype 3/7 stripes. It retracts like the repressor model.
• Kr proximal and distal respond differently.
• proximal expands, distal retract
• hunchback changes its promoter
• is this extension in stripe 7 temporal specific?
• Hb switches promoter (which may or may not change its sequence).

Changes in a P-MAD binding site underlie species diversity of wishful thinking patterning.

• Rob Marmion1, Milica Jevtic2, George Pyrowolakis2, Nir Yakoby1. 1) Biology Department and Center for Computational and Integrative Biology, Rutgers University, Camden, NJ; 2) Institute for Biology I, Albert-Ludwigs University of Freiburg, Freiburg, Germany.
• wishful thinking previously only implicated in axon guidance. Find a role in oogenesis.
• dpp is anteriorly localized.
• Wit is necessary for dpp signaling and its expression is regulated by the pathway.
• promoter bashing: characterize enhancer of Wit
• mutation of MAD sites disrupts reporter expression.
• Wit is wider in virilis (4 appendages)
• virilis enhancer drives wider expression in D mel (and patchy in posterior).

A genetic screen for new Polycomb group genes.

• James Kennison, Monica Cooper. Genomics of Differentiation, Eunice Kennedy Shriver National Institute of Child Health and Human Development, NIH, Bethesda, MD.
• Can we make a better screen than Sex combs?
• PREs have pairing sensitive silencing. PRE in transgene, 30-50% of insertion sites the homozygote eye color gets lighter.
• identified 5 PRE candidates that exhibit pairing sensitive silencing from scr. Focus on one from second intron of Scr. Biggest change from red to white
• Use FLP/FRT to generate clones in the eye (eye-FLP)
• mutagenize flies, screen for red conversion due to loss of pairing
• 370 – 400 lines
• 20% of mutants alter copy number of derepressed genes
  • derepression primary in posterior of eye, eye is rough / not dividing well. mutants survive as homozygotes.
  • Meiotic recombination strongly surpressed: can’t map and hint that something is wrong.
• 51 complementation groups. 30 have multiple alleles. 14 of these correspond to known PcG. Get everyone except mxc.
• do get extra sex combs from scr in new mutants. Need to map now.
• For 3 genes with a single allele, there are two ORFs.
• 22 of the genes map to a single transcription unit.
• 15 chromatin factors: Spps, grh, Dsp1, ocm, (previously implicated)
• also new ones: ftz-f1, Nf-YB, aminoacy-tRNA synthass and some protein kinases

Super-resolution imaging of chromatin nanostructure reveals tight coupling of epigenetic state and 3D genome organization.

• Alistair Boettiger1, Bogdan Bintu2, Jeff Moffitt1, Brian Beliveau4, Chaoting Wu4, Xiaowei Zhuang1,2,3. 1) Chemistry and Chemical Biology, Harvard University, Cambridge, MA; 2) Department of Physics, Harvard University, Cambridge, MA; 3) Howard Hughes Medical Institute, Harvard University, Cambridge, MA; 4) Department of Genetics, Harvard Medical School, Boston, MA.

The RNA binding protein Arrest (Aret) regulates myofibril maturation in Drosophila flight muscle.

• M. Spletter1, C. Barz1, A. Yeroslaviz1, C. Schönbauer1, D. Gerlach3, I. Ferreira1, M. Sarov4, A. Stark2, B. Habermann1, F. Schnorrer1. 1) Max Planck Institute for Biochemistry, Martinsried, Germany; 2) Research Institute of Molecular Pathology, Vienna, Austria; 3) Boehringer Ingelheim RCV GmbH & Co KG, Vienna, Austria; 4) Max-Planck-Institute of Cell Biology and Genetics, Dresden, Germany.
• fast contracting flight muscles have fibular structure
• all other muscles have tubular organization. How do these differences arise?
• spalt major is a master regulator of fibular muscle (KO -> convert to tubular muscle).
• mRNA seq in spalt KO
• RNA binding protein Arrest is regulated by spalt. Arrest KO muscles pull themselves apart.
• Arret is expressed in cyto and nuclei early in development and in immature fibrils. later it moves into the nucleus. (72 hours)
• Sarcomeres widen (factor of 1.5) in WT. This widening growth fails in arret KD
• seq Arret KO, 90% of exons change — splicing effect more than tx effect
• Aret promotes the inclusion of fibrilliar exons, inhibits the inclusion of tubular exons.
• Arrest phenotype looks like classic myosin over-activity defect (muscles over contract and snap)
• myosin mutant blocks this pull the muscle apart phenotype.
• Strn is one of the differentiatially spliced genes.

Ultraconserved core elements are an essential feature of insect enhancers.

• Thomas Brody, Ward Odenwald. Neural Cell-Fate Determinants Section, NINDS/NIH, Bethesda, MD.
• late neuroblact ultraconserved enhancer
• enhancer consists of clusters of conserved octomer bindings sites (repeat sequences)
• conservation of mouse to lamprey (2.5 By) including insertion deletion lack.
• (260 My) Anapholeses to D mel.

Saturday evening talks

Preventing age-related metaplasia promotes homeostasis of the gastrointestinal tract and extends lifespan.

• Hongjie Li1,2, Yanyan Qi1, Heinrich Jasper1 1) Buck Institute for Research on Aging, Novato, CA; 2) University of Rochester, Biology Department, Rochester, NY.
• up-regulation of Jak-STAT pathway leads to gut degeneration
• inhibition of Jak-STAT matains gut homeostasis and extends lifetime.

Regulation of metabolism and insulin sensitivity by Sir2 in Drosophila.

• Rebecca A. S. Palu, Carl S. Thummel. Human Genetics, University of Utah School of Medicine, Salt Lake City, UT.
• Sir2 (ideintified in yeast associated with aging) is a metabolic and epigenetic regulator
• involved in a variety of metabolic processes. Details remain unclear (aging, diabetes, Obesity etc)
• sir2 mutants become increasingly insulin resistant with age of animal
• old sir2 mutants clear glucose much more slowly than wt or young sir2 mutants
• hyperglycemia sets in early. then insulin resistant obesity sets in. then glucose intolerance (about a week apart each).
• RNA seq mutants at 2 week time point.
• strong degree of overlap between genes regulated by sir2 and genes regulated by hnf4 and foxo (known targets of sir2 in mammals. invovled in glucose matabolism)
• Propose Sir2 de-acytelates these TFs to change gene expression to change metabolism in response to nutritional cues.’

Systemic organ wasting induced by localized expression of the secreted insulin/IGF antagonist Imp L2.

• Young Kwon1, Wei Song1, Ilia Droujinine1, Yanhui Hu1, John Asara3,4, Norbert Perrimon1,2.
• Wasting: process of losing mass (opposite of growth)
• systemic organ wasting associated with extreme starvation, infection, severe infection, cancer and other diseases.
• cancer cahexia — cancer associated wasting
• Hippo signaling is a master regulator of growth via Yki activation, leads to localized cell proliferation.
• Fly phenotype: ovary atrophy, fat body disappears, bloated, translucent abdomen.
• Impl2 RNAi rescues the systemic organ wasting phenotypes

Mechanism of Body Fat Regulation by Split ends.

• Kelsey Jensen, Tânia Reis. Department of Medicine, University of Colorado Medical School, Aurora, CO.
• use sucrose PBS solution to screen float vs sinking L3 larvae to ID fat regulation
• (screen performed by Reis in Hariharan lab in 2010)
• trying to find new genes that regulate fat
• use carnegie FlyTrap project to screen 33 lines which have GFP. 18 of which are expressed in the fat body
• next step: knock these genes own specifically in the larval fat body.
• gene split ends (conserved between flies and mammals) not previously linked to metabolism.
• over-expressed copy of gene leads to high density low fat larvae
• Is there a behavioral / eating / moving change?
• KD larvae over-eat (may contribute to being fat).
• overexpression larvare eat normally.
• Film movement: both over-expression and lean larvae move slower than their controls.
• fat larvae actually starve first (opposite to expectation). Can’t utilize their fat properly? perhaps this leads to the over-eating phenotype?
• increase in fat metabolism, decrease in sugar metabolism
• Split ends (Spen)
• now testing expression levels of homolog in mice (MINT). different levels of Spen in fat liver than in lean liver.

Saturday Evening Session

Angela DePace

• cooperativity mostly thought about in terms of pairwise interactions
• long distance cooperativity?
• ‘linear framework’: Markov chain transition between microstates
• ‘Ahsendorf et al’ BMC Biology 2014
• need more than pairwise cooperativities to get hill like functions
• asymptotically approach hill coefficient equal to N sites
• doesn’t really matter if assume all bound yields transcription or any bound yield transcription.
• away from equilibrium more easily get sharp switching.

Michael White

• Goal: quantitative relationship between sequence and gene expression
• Hh signaling makes Ci an activator. Otherwise it is a repressor
• Gene expression is proportional to occupancy
• in middle of gradient, activation wins at low affinity and repressor wins at high affinity, (because repressor is more cooperative and switches sharper).
• 3-strong sites, repressed. (cooperative binding dependent?). delete 2 of the activator sites, now it’s an activator.

Golding: Quantifying the stochastic kinetics of Gene regulation

• Heng Xu and Anna Sokac
• How does binding of TFs drive stochastic process of mRNA production?
• how do multiple TFs regulate a single gene?
• Progress:
  • Measure concentration of a given TF factor
  • promoter activity
  • TF binding
• in individual nuclei of WT embryos
• use stochastic theoretical analysis to connect this to binding
• have the endogenous species (check)
• have ‘absolute numbers’ (maybe….)
• can we interpret the full dataset rather than averaging over nuclei?
• 2 step promoter model
• Kon is Bcd dependent. Describe that as the Hill function of Bcd function.
• signal to noise is around 1%. But pool many nuclei and this may be enough
• linear proportionality to 3 loci measured by ChIP (but I don’t believe quantitative ChIP data either)
• cycle 12 embryos, see a double plateau of bcd concentration at the transcription site as a function of AP position.
• propose Hb binding represses hb.
• Two color control?

Reinitz

• triplet code
• Cis-regulatory code: on in correct cell type, off in correct cell type.
• need inputs, outputs, and a way to compare them.
• need to work at cellular resolution.
• will discuss control of the even-skipped
• need to rely on some phenomenology (careful guessing). Function tends to reside in multiple sites. Can get substantial amounts of synergy.
• Regulatory mechanisms considered: protein binding DNA steric competition, cooperative binding, short range repression by quenching (150 bp) phenomelogical
• coactivation (150 bp) phenomological
• activation by arrhenius rate law.
• PWMs are determined independently
• 3-7 + 2 fusion
• sepsid stripe 2 works in D mel (‘shit eating flies’) Homology in stripe 2 is extremely weak (2 binding sites or so). Predict that sepsid stripe 2 will be posteriorly shifted.
• most activation in stripe 2 was coming from co-activated Hb sites
• Hb-3 amplifies signal but does not give expression by itself.