Friday,

Giaovoni Bosco

• in wild 80% of fly larvae of been injected by wasps
• Q: is there wasp/wasp competition inside the same chamber
• Flies produce 80% fewer eggs (apoptosis in ovary) when exposed to wasp.
• in separate chambers in fly condos, observe reduced egg laying in unexposed neighboring combos
• add GFP flies (naive) to exposed wt flies, see reduced egg laying in FGP
• how long does the repressed egg laying state last?
• physiological change in ovary also observed in GFP flies
• memory returns to normal over 240 hours (somewhat gradually)
• Orb2 mutants go back to normal egg laying within 24 hours (gene involved in memory)
• rut1 and Orb2 mutants cannot teach His-GFP flies to repress egg laying
• ninaB (vision) is required for response to wasp. See no effect in the dark or with blind flies.
• Learning also requires vision (Orb2, Adf1, dnc, rut required). Memory required for sustained oviposition.
• Requires 2 wings to teach. Only females teach. Males don’t help. Not auditory (can block auditory and olfactory)
• Compensentory increase in life span? (Otherwise what’s the adventage of decreasing your fitness, even in presence of the predator)
• In wild they seek out high ethanol food when egg depsition is low, and then it ramps up.
• Is it parasytic wasp specific? what about other wasps? Yes. Only respond to parasytic female wasps.

Charaterization of a long-distance neurotransmitter recycling pathway essential for Drosophila visual transmission. Ratna Chaturvedi, Hong-Sheng Li. Department of Neurobiology , University of Massachusetts Medical School, Worcester, MA

• local synaptic vesicle recycling can create interference
• long distance recycling (distal from the synapse) can avoid this.
• Does it occur in the fly?
• Histamine (analog of glutamate signaling in fly vision) metabolites are present in the glia cells (not just the neurons) suggests there is a longer range recycling.
• block gap junctions, see reduction in histamine in glial cells.
• these flies are also deffective in synaptic transmission.
• don’t know transporter (what was that about the gap junctions?)
• RNAi screen knockdown of BTA see reduced photo-signal transition. (Blind not lethal).
• only neurons express BTA, not Glia.
• only photoreceptor neurons damaged.
• Deleted BTA with CRISPR, acAGATACA motif validated RNAi
• BTA deletion flies accumulate carcinine in the lamina glia, and is not recycled back to the photoreceptors.
• use moving block ‘threat’ to assay visual responsiveness. BTA nulls blind.

Contribution of sex, genotype, and environment to individual gene expression profiles. Kseniya Golovnina1, Yanzhu Lin2, Zhenxia Chen1, Haiwang Yang1, Hina Sultana1, Brian Oliver1, Susan Harbison2. 1) NIH/NIDDK, Bethesda, MD; 2) NIH/NHLBI, Bethesda, MD.

• DGRP lines
• high variability even when same environment, sex, and genotype.
• How variable is gene expression among individual flies?
• 16 genotypes, 3 different environments
• 8 females, 8 males, do RNA-seq (768 libraries)
• Of 15695 genes, 95% differentially expressed due to sex
• 31% of genes with unknown function show sex specific expression differences. More often male specific up-regulation.
• Male expression values more variable than females
• Q: do you see a correlation in variabilty to being on the X chromosome?
• ID genes more sensitive to environment.
• find examples of variance explained by sex-environment / sex-genome co-variation.
• How was development stage controlled? : just age
• What was the environment variation: not stated.
• Variation in expression correlated to anything? When we look across all genes

The evolution of maternal mRNA deposition and zygotic genome activation across 14 Drosophila species. Joel Atallah, Susan E. Lott. Evolution & Ecology, University of California – Davis, Davis, CA.

• generated 144 libraries (14 species, 2 stages) look at zygotic genome activation
• look at divergence in zygotic / maternal FPK ratio for genes. Take correlated genes from mel and compare out to other species — correlation drops (not too surprising. but I suppose which gene change will be interesting)
• Look at maternal only, zygotic only, and both
• most transitions are the both to maternal or the both to zygotic
• very few examples in evolution of zygotic to maternal or maternal to zygotic: changes pass through an intermediate. (again, what you would likely guess)
• CAGTAG / TAGTeam sites are enriched in the promoters of genes that become zygotic / are zygotic. (Motif is well conserved. Maybe an 8mer instead of 7mer in pseudo).
• find motifs that are lineage specific

Three-step mechanism for the spatial and cell-cycle dynamics of pericentric chromatin. Eric Joyce, Tharanga Senaratne, Ting Wu. Department of Genetics, Harvard Medical School, Boston, MA.

• screen for genes that affect pairing of peri-centric chromatin
• ID condensin II components, centromere components
• clustering is cell cycle dependent: stain CycB, more clustered in G1. (cycB is on in G1 only (?))
• centromeres get clustered in telophase
• knockdown centrosomes — still get division, centromeres don’t cluster in mitosis, remain unclustered in G1
• centromeres do not have an intrinsic ability to find eachother
• (Does this extend to pairing? of other loci?)
• Mcph1 RNAi, heterochormatin becomes disperse. No visible H3K9me2 when heterochromatin is disperse. Some repetive regions start getting expressed.
• centromeres redistribute in a condensin II dependent manner.

Mrg15-dependent binding of Cap-H2 to chromatin is required for chromosome organization and regulation of gene expression. Heather Wallace, Huy Nguyen … Giovanni Bosco.

• Condensin II drives axial compaction and un-pairing
• Mrg15 chromodomain protein interacts with condensin II (H3K36me3 interacting)
• required to mediate shortening and unpairing.
• Mrg15 binding sequence is required for its interaction with CapH2 (perhaps it targets CAP)
• FISH probes on chromosome measure compaction, (overexpress?) CapH2 GFP, chromosome shortens (distance > 2 um to closer than 2 um)
• ChIP-seq shows very high correspondence, both at active locations
• most of the same genes (95%) respond the same in RNAi knockdown of Mrg15 vs. Cap-H2 (same ~3400)
• most genes downregulated on loss of CapH2

Trimethylation of Histone H3 at lysine 27 by Polycomb Repressive Complex 2 and its role in epigenetic memory. Rory T. Coleman, Gary Struhl. Department of Genetics and Development, Columbia University College of Physicians and Surgeons, New York, NY

• Propose H3K27me3 carries the memory across cell-cycles
• DNA replication poses a challenge to histone-based memory (add unmodified histones)
• generate Ubx-LacZ PRE mini-gene with a FLP-outable Ubx-PRE (also containing GFP in FLP-out)
• look in wing disk where Ubx is silent. induce Pre excision,
• cells remember off state for 12 to 72 hours — location dependent (nodum remains repressed for 72 hrs) mostly de-repressed by 110.
• Only lose the memory of the off-state in actively dividing cells. G0 cells remain silenced.
• expect dilution of K27me3 through progressive rounds of cell division.
• after 12 hrs, drop to 32%, thereafter to 24% to 16% of H3K27me3. Rate of dilution is slower than predicted (7 cells divisions)
• Have you done ChIP for PRC2? ( propose ‘free ‘ PRC2 passes on the mark.) — PRC2 components only ChIP to the PRE. Don’t see near by.
• knockdown PRC1, lose repression while maintaining the H3K27me3, so K27me3 might be inheritable but is not intrinsically sufficiently repressive.
• Q: looks like a potentially great system to study the relative roles of PRC2 and PRC1.

Tip60 HAT Action in Environmental Enrichment induced Cognitive Restoration. Songjun Xu. Biology, Drexel University, Philadelphia, PA.

• Tip60 is an Alzheimer’s associated HAT
• Tip60 is required to induce neuroadaptive transcriptional response to environmental enrichment (EE). what is EE?
• 220 EE responsive genes, 43 sensitive to HAT,
• TIP60 rescues E induction of synaptic markers in the AD neurodegenerative fly brain (when was the AD fly model introduced?)

Dietary restriction reduces transposable element expression in aging Drosophila heads. Jason G. Wood … Stephen L. Helfand, Brown University

• How does chromatin change in age?
• in yeast, normally repressed loci (mating loci, telomere) become expressed with age.
• focus on repressive heterchromatin H3K9me3.
• a lot of the heterochromatin we don’t have assembled very well.
• with age, enrichment of H3K9me3 at centromeres decreases substantially (and chr4)
• Hp1 follows
• use LacZ moved into heterochromatin region, see increased expression in age.
• low calorie fed flies lose silencing later.
• how do native heterochromatin genes change expression with age? Some increase substantially with age (none decrease?)
• in calorie restriction less increase in expression (most decrease)
• Transposons: 2 types: cut and paste. Also retrotransposons, transcribe and jup. Lots of transposons. Pericentric region is gene poor but retro-transposon rich.
• transcriptional and post-transcriptional silencing mechanisms combat retro-transposons.
• rRNA depletion generally not designed for fly. Nugen custom depletion kit works better. TEs are repetitive, hard to align and quantify expression.
• used RepEntrich: assemble all elements into a single ‘gene’. (not at all clear how this works).
• Do any genes go down?

A novel chromatin factor Enhancer of Polycomb acts in somatic cells to maintain germ cell identity and activity in Drosophila adult testis. Lijuan Feng, Zhen Shi, Xin Chen. Biology, Johns Hopkins University, Baltimore, MD.

• Approach: knockdown E(Pc) and exmaine effect in germ-cell lineage
• compromised E(Pc) see extra mitotic division (overgrowth?)
• may divide unsynchronously (not all remain connected by fusome, EdU variation).
• E(Pc) enriched in promoter region of target genes that change in E(Pc) mutant, including Zfh-1.
• overexpression of Yan componsates in part for E(Pc) knockdown.
• E(Pc) is a component of Tip60.

Impacts of centromere misregulation on genome stability and cancer progression in a Drosophila model of glioblastoma. Nicole Beier, Renee Read, Gary Karpen

• many centromere kinetochore genes are overexpressed in cancers. e.g. CENPA and HJURP
• high levels of these proteins correlate to poor prognosis and susceptibility to radiation treatment.
• CID overexpression causes genome instability
• 70 fold overexpression find it throughout the genome. Leads to anneploidly
• Find increased breaks (with chromosome arm BAC paints)
• overexpress CID see fewer glia (kill cells)
• overexpress CAL1 (chaperone assembly factor for CID), no real change
• overexpress PI3K and CID get overgrowth (alone neither gives overgrowth)
• CID + CAL1 see decreased proliferation (opposite of expectation)
• endogenous CID is only enriched in centromere
• 2x expression see increase at promoters of transcribed genes
• expands into the gene-body on 70x expression.

A genome-wide resource for the analysis of gene function and protein localization in Drosophila.

• Mihail Sarov1, Christiane Barz2, Katja Finkl2, Marco Hein2, Stephan Janosch1, Nicole Plewka2, Bettina Stender2, Dana Suchold1, Vinay Vikas3, Matthias Mann2, Mani Ramaswami4, K. VijayRaghavan3, Pavel Tomancak1, Frank Schnorrer.
• 10,000 clones within their natural genomic environment
• goal: generate multifunction tags (super-folded GFP) good for spatial organization, protein purfication etc.
• ~12,000 genes cover by fosmids. ~11,000
• inserted onto third chromsome. 800 genes tagged.
• Do the proteins rescue? 15 don’t (including snail esg, and eya) TFs less likely to rescue.

Computational tissue labeling: Tissue and Cellular Recognition in Developing Drosophila Embryos. Soile V. E. Keranen

• older embryos have variable size nuclei
• two step segmentation: first isolate tissues and then cells
• one approach: hand annotation in every third slice.
• alternatively label tissue by gene expression.
• however 70+ tissue and sub-cell types.
• problem: only 3-5 colors. (sounds like an excellent system for MMECFISH
• get tissue by nuclei/DNA patterns
• then get nuclei. Use tissue specific nuclear parameters
• ‘Fly Annotator’ program (looks like Matlab)
• variable size and nuclear number between embryos (is this a staging issue?)
• compare left-and-right paired organisms (like salvary glands). Left salvary gland is longer, left anal pad is longer
• Training computer to automatically age rank embryos
  • overlay gut point clouds from embryos of different ages
• pyramid contex features allow identification of organs
• All 2-photon imaging.

Model-driven data visualization and quantitative animation of developmental signaling. Bomyi Lim (Stas Shvartsman lab)

• with fixed tissue temporal staging is difficult
• live imaging not available for many things (like dpERK)
• want to combine different stains
• use membrane length in blastoderm
• use PCA in gastrulation. 5 components do a pretty good job
• use a linear regression with 5 projection coefficients to
• Q: If you apply the method to snapshots from live movies of different embryos, can you measure temporal variability in coordination of growth.
• ERK is activated only in cells where rho was expressed (spitz must be short range).
• add ERK from model and Ind from in situ stain to histone-RFP movie to create 3 color movie of ERK.

High-throughput Investigation of Drosophila Brains via Structure Based Similarity. Florian Ganglberge

• 3d confocal stacks of 1000s of Gal4 driver lines expressed in the brain.
• want to annotate neurons.
• challenges: variable intensity, registration errors, background from other neurons labeled.
• approach: ‘slow bulky Matlab script’. few hours for 1 search
• optimized the method by downscaling image resolution, sparsification: remove noise and background.
• added tool to braingaizer.org. 7x faster
• can highlight the neuron in one image and find all the gal drivers that have that neuron ON in the database of images.
• structure based queries work better than direct overlap (direct overlap is low).

An automated image analysis tool to track cell divisions during Drosophila axis elongation.

• Michael F. Z. Wang1, Rodrigo Fernandez-Gonzalez1,2,3. 1) Institute of Biomaterials and Biomedical Engineering, University of Toronto, Toronto, Canada; 2) Cell and Systems Biology, University of Toronto, Toronto, Canada; 3) Developmental and Stem Cell Biology Program, The Hospital for Sick Children, Toronto, Canada.
• convergent extension intro
• cell division is resolved along AP axis, contributing to elongation.
• explain watershed algorithm (almost identical to matlab tutorial):
• Threshold, dilation, inversion, distance transform, now watershed.
• temporal alignment using cross correlation in paired binned images.
• use ratio of perimeter to area (relative to circle) – this dereases and then increases in preparation for division
• the ratio of the major and minor axes also helps ID pinching off cleavage furrow.
• can now measure cell division orientation quantitatively for lots of cells. (compared to manual annotation).
• look at the role of tension in the A-P resolution of division.

FlyVar: a database for genetic variation in Drosophila melanogaster.

• Rui Chen1,2, Lichu Jiang1, Yong Chen4, Nele Haelterman3, Hugo Bellen2,3,5, Fei Wang4. 1) HGSC,Baylor College of Medicine, Houston, TX; 2) Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, Texas; 3) Program of Developmental Biology, Baylor College of Medicine, Houston, Texas; 4) Information Processing, Department of Computer Science and Technology, Fudan University, Shanghai, China; 5) Howard Hughes Medical Institute.
• genome sequencing cost still dropping fast.
• bottleneck is now to ID bases that change
• high degree of variation in drosophila: 1 variation per 500 bp
• don’t have genome of our earlier lab strains
• lots of fly lines / mutants!
• Approach to find the good stuff: remove parental varients. remove benign variants, remove unrelated genes, add additional specific filters.
• FlyVar documents the benign polymorophisms.
• If you know locus and you sequence it, you may find mutliple SNPs. Tool can help ID which were meaningful?

Identification of novel drug targets for Tuberous Sclerosis Complex by cross-species synthetic screens combining CRISPR based knockouts with RNAi. Benjamin E. Housden,

• Tsc complex (mutated in rare benign cancer — simplest cancer genetics)
• theraputic strategy is weak: inhibits overactivated mTor.
• synthetic lethal screen: kill cells only which house the mutation.
• e.g. Ben Haley’s double hairpin approach
• why not use combinitorial RNAi?
  • incomplete delivery
  • incomplete knockdown
  • off target effects
  • limitations are compounded in combinitorial screens
• use single RNAi screens in CRISPR generated mutant cell lines
• goal: generate homogeneous cell cultures (challenging in tetraploid fly cells)
• CRISPR with GFP co-transfection.
• Drosophila cells don’t like to grow alone. They like colonies.
• change culture conditions to survive sort, and 16% grow from single cell into full culture. Nice, Ajaz should talk to these guys
• can create homozygous mutant lines. e.g. for TCS1 and TSC2
• mutants keep on growing without serum as expected, are larger as expected.
• RNAi screen all phosphotases and kinases in the genome.
• only explore RNAi that have no effect on WT
• synthetic lethal effects conserved through human cells.