Thursday Morning Platform Presentations

Yohnns Bellaiche: cell division in morphogenesis

• Intro: beautiful movie of zebrafish morphogenesis
• system: Pupal morphogenesis
• Goal: multiscale imaging –
  • 900 63x images (cellular resolution of whole tissue).
  • all 900 images every 5 min
• Can segment and color-code cells, for example by size.
• 50,000 cells, most junctions formed by cell division, formation of a single large interface.
• important for polarity maintenance, lies behind clonal patch.
• in cell culture, get actin ring that contracts to a point (not like long interphase in epithelial tissue)
• Neighboring cell forces are necessary to get a long interface (oblate cells, get constriction to a point contact).
• Two membranes invaginate to form long pairs of branches,
  • move over, myosin not must localized at interphase of cell, but also expressed inside cell edge of neighboring cell.
  • this is a response to the mechanical stress
  • this myosin helps hold the two membranes together so it invaginates as a narrow pennisula and not a big v (can laser ablate just the myosin in this cell and they spring apart into a v, but then recover).
• MyoII in neighboring cell is therefore required.
• Does membrane of neighboring cell get inserted between dividing cells?
  • Yes it does ingress!
  • but it retracts / degrades in the end.
  • F actin is recruited to the interface at the time of retraction of inserted membrane.
• Prevent F-actin function, it is the neighboring cells membrane that remains and two cells separate.
• Actin flow inside dividing cells drives withdrawl of the membrane
• Quantifying cell behaviors: size/shape change, cell rearrangememnt, cell division, and delamination. Plot spatial frequency of these events throughout the tissue.
• estimate junction stress from lattice image.

Genetic Conflicts During Meiosis Drive Innovation in Centromeric Proteins — Harmit Malik,

• Adapting to a changing / co-evolving environment
• genetic conflicts within the same genome
• conflict between ‘selfish centromeres’ and chromosome segregation?
• Male meiosis in plants and animals
  • one homolog can poison the other sibling chromosome sperm
  • highly adventageous to the one chromosome that will win
  • disadventageous as fertility drops 50%
• female meiosis (in plants and animals), only 1 cell wins the preferred position
  • if one chromosome can orient itself better (faster microtubule recruitment) can win more often. Selfish centromeres can increase transmission rate past 50%
  • examples: Robertsonian chromosomes (fusions of centromere-at-end chromosomes) These are normal, our chr2 is a Robersonian fusion still separate in great apes.
  • over-recruitment of centromeric proteins.
  • in male meiosis suffer fertility (male meiosis ‘likes symmetry’) (.1% of humans are Robertsonian)
• 60:40 segregation will sweep Drosophila in 10,000 years
• In Drosophila centromere boundaries / strengths are set up epigenetically
• heterochromatin protein over expression can encroach on centromere and vice versa.
• selfish centromeres try to get larger, penality paid in the males. intense evolution at centromere and centromeric proteins.
• Non-canonical nuclesome (H3 replaced by CENP-A) in centromeres
• swap Cid (fly chromosome centromeres) between simulans and yakuba, viability 95% (in both male and female F1) (viability not fertility?)
• predict reconstruction of ancesetoral Cid, no effect on F1 female viability but strong effect still in male.
• HP1b non-essential, but its daughter gene Umbrea does localize to centromere and is essential for normal cell division. Knockdown results in massive aneploidy.
• a constant turnover of essential centromeric proteins.
• C elegans ar holocentric (whole chromosome is a centromere, rather than concentrated at point on chromosome).
• Lepidoptera, Hemioptera, all lose CenH3 and have holocentric chromosome division.
• loss is specific and multiple times in the insect lineages.
• old genes can become dispensable and young genes essential.

Angela Stathopolous: mesoderm cell migration

Intro

• FGF ligands pyramus and Ths expressed in different tissues
• FGF signaling required for proper mesoderm spreading (should form a nice internal monolayer)
• cells don’t cross the midline
• in htl mutants, upper domain cells do not exhibit directed movement and cross back and forth across the midline.
• Role of leader cells — compotent to move in mutants or wt. FGF not so much chemo-attractant. More important in ensuring symmetric collapse of the mesoderm.
• Conducted genetic screen (311 UAS lines), found 10 genes that affect mesoderm spreading.

Screen results:

• Trol (HSPG) is required for furrow collapse
• Sdc is required for intercalation and differentiation.
• ectopic expression of pyr or ths in mesoderm leads to lumpy pile up of cells in mesoderm.
• E-cadherin (normally down regulated in mesoderm). knockdown or overexpress lose monolayer of mesoderm.
• E cad localization changed in htl mutants.
• propose snail is effected by FGF. sna up-regulated in absense of FGF, down-regulated by FGF.
• propose FGF regulates spreading by controlling adhesion state through E-cad regulation.

CVM cell migration

• long range 6 hour migration
• cells express heartless. Follow a track of cells expressing Pyr and Ths.
• In FGF mutants cells still migrate, but they intermix
• why do leader cells move faster? posterior cells hardly move…
• FACs to isolate CVM cells and RNAseq
• find enrichment in lipid synthesis in CVMs.
• sphingolipids regulate germ cell migration. watch comigration of germ cells and CVMs.
• lipid mutants show defects in migration.

Stas Shvartsman

Intro

• ERK signaling – used multiple times in development
• excessive ERK activation lead to many varieites of prevelant developmental abnormaliites.
• localized RTK responds to transient pulses of ERK throughout development in many tissues.
• ERK activates ind
• will describe mathematical model that captures behavior semi-quantitatively and makes several clear predictions.
• ### reconstructing dynamics
• don’t have a live sensor
• compare stills to database of live imaging movies to match morphology. record movies in the same vertically orienting PDMS device.
• linear regression model matches features of images (5 dimension 5 component PCA mapping). Accuracy is better than 3 min (presumably matching embryo to embryo)
• Question: does the 3 min accuracy come from intrinsic embryo to embryo variation, or uncertainty in the morphogenic mapping

Ind activation

• receptor uniformly expressed, activated by Vn and Spitz. Vein is dispensible, Spitz (regulated by rho) is essential – necessary and sufficient. Don’t see any consequence from removing vn.
• ERK signal grows as the t^3 over time.
• Pathway: Zld activated EGFR, Star and Spitz. Ligand grows linearly in time, receptor grows linear in time, integrated this rate over time, expect cubic growth.
• Ind is expressed when ERK crosses a basic threshold.
• delay induction in csw mutant, lower levels of ERK, takes longer to cross threshold, still turn on Ind.
• How is ind responding to ERK? – regulated by Cic. ERK relieves Cic repression.
• ERK first inactivates Cic, which is then exported and degraded.

implications

• upstream factors of ERK frequently mutated in developmental defects and cancer
• propose that cancer assoicated mutations will be lethal in development
• introduce mutations from human abnormalities into fly embryo.
• change ERK pulse. Both developmental abnormality and cancer mutations cause strong lethal phenotypes in fly.
• also developing computational tools.
• reduce Cic, also (reduce) ERK signalling

Maria Dominguez: Addressing complixity in Notch signaling and cancer

Intro

• finding out the unknown unknowns of cancer
• ‘peto’s paradox’: no relationship over species between cancer and organism. Mice have higher incidence of cancer than humans, whales have less.
• No / almost never cancer in naked mole rate. — lower metabolism, less Reactive Oxygen species
• Tumor cell communication is important.
• Nutrition is relevant.
• Size regulation / growth regulation / when to stop is relevant. — best understood in flies (?) Notch povides growth directions

Notch-Pten/Akt in cancer

• Delta upregulation minor overgrowth
• Delta + ATK upreguatlion dramatic cancer (clearly visible in eye tumor)
• ATK is downstream of INLS/IGF/RTK/PI3K and upstream of dTOR
• PTEN links NICD to Atk1
• Screened sigma-aldrich library of pharmological active compounds looking for effectors of (PTEN?)
• 150 flies screened per each 1000 compounds, ID 62 compounds that attenuate and 58 that enhance eye tumors
• 23 of the 62 reduce growth in human cultured tumor cells. ~6 of these don’t affect growth of non-tumor cells. (what happened to PTEN / NOTCH?)
• chose drug that has the strongest effect at lowest concentration.
• find atp syn Beta gene
• Warburg hypothesis: mitochondria dysfunction causes tumors.
• Test: overexpress Delta and knockdwon atp synthase beta -> tumors
• mitochondria dysfunction facilitates tumor growth through Delta/Notch signaling (these dysfunctional cells should die.
• Delta+ and PTEN-kd flies have elevated levels of ROS (using ROS Gstd1-GFP sensor).
• chemotherapy elevates ROS to kill tumor cells.

Juergen Knoblich, (IMBA) Stem Cells

Intro

• stem cell number rapidly decreases during development.
• Timely loss of stem cells is important to avoid child tumors
• Drosophila neuroblasts (neural stem cells). Responsible for generating all the neurons and glia in adult brain.
  • Type I: dived assymetrically, differentiatin cell divides into 2 neurons. (mice)
  • Type 2 proliferative cell is generated from first asymetric division and then divide and differentiate. (flies and humans)

Drosophila neuroblasts

• Drosophila pupal NBs shrink before disappearing (in diameter before differentiating),
  • consequence of metabolic limits of non-eating pupae?
  • Follow neuroblast division in culture to test.
  • due larval and pupal NBs culture differently? Pupal neuroblasts become smaller with cell divsion, larval ones do not. Not related to division rate.
• known growth pathways do not appear to be involved. — Genetic screen:
• express luciferase in neuroblasts.
• Find 300 RNAi lines that affect behavior, none appear to be off-targets
• 8 of these map to Mediator complex.
• knockdown mediator components (Med27, Med10), get cell-type specific effect: more neuroblasts, less shrinkage and death. Increased diameter of neuroblasts.
• Question: Does it matter which subunits of mediator you knockdown? (Is mediator functioning in a modular way?)’
• Mediator is required for Ecdysone response (major hormone that triggers instar changes, spikes in pupation.
• Mediator regulates metabolism.
• NB cell cycle exit requires alpha-KGDHC (knock this down and they don’t exit / differentiate). (Gate keeper of Krebs Cycle, catalyzes the first irreversible step of Krebs. Generates NADH for respitory chain.
• Mediator is required for upregulation of oxidative phosphorylation regulation.
• Proliferative tissues (staying same size) don’t do Krebs, rely on anaerobic metabolism.
• Metabolism change effects change in cell fate (rather than v.v.)

Thursday Evening Session: Organogenesis and gametogenesis

Weischaus lab Nucleolus formation

• rRNA synthesis starts by cc11. Nucleolus doesn’t form until cc13.
• phase separation model
• RNA PolI and fibriliarin form in the absence of rDNA (but are much weaker).
• rDNA is a nucleation site for fibrilian and PolI. Sites are less stable without it.
• What explains the difference in intensity? If it is just a seeding effect, one might expect it is just time.

Yiqin Ma (Buttitta Lab, U Michigan): Examining chromatin structure in G0

• Multiple types of G0: Reversible-quiescence, terminal differentiation, and senescence.
• will study terminal differention G0 in fly wing (naturally synchronized)
• exit cell cycle for last time to G0 24 hrs after pupal formation.
• Flexible G0 phase 24 hrs to 36 hr (can still exit). Thereafter robust G0.
• PcG /PRC2 and H3K9 methyltransferases up regulated on G0 exit
• heterochromatin histones and proteins are disrupted (based on immune staining)
• heterochromatin modifications are tightly associated with cell cycling status.
• K27me3 and HP1 not required for cell cycle exit.
• screen for other regulators,

the chromatin modifier trithorax regulates systemic signaling during imaginal disk regeneration

• use UAS GAL4 to drive reaper under TS-Ga80
• HS induce ablation (vero 90% of wing pouch). Use wing size to quanitfy amount of regeneration.
• Tune wildtype response to have 50% regeneration
• Find that Trithorax dominant increases regeneration and Trx mutant decreases.
• trx mutants regenerate half as much as wildtypes.
• trx hets still form regenerating blastemea — same amount of tissue is regenerating as in wildtype.
• tissue is not dying
• regenerating animals delay pupariation (depndent of insulin like peoptide 8 and retinoid signaling). Trx hets delay for less long.
• rescue pupation delay using UAS dilp8 — regenerate more wing.
• dlip8 is regulated by JNK signaling.

How is growth coordinated to maintain isometry (proper size and proportion)

• growth rate of eye reduced when wing disk is damaged and has to regrow.
• damaged imaginal tissues release dilp8 to coordinate growth.
• NOS overexpression also reduces growth rate of disks.
• experiment: irraiate larva, shield eyes with lead tape.
• NOS is necessary for growth delay in eye (remove NOS lose delay)
• NOS is not required for developmental delay but is a mediator of dilp8
• propose NOS is inhibitory on growth during feeding and activating afterwards.