Genome-wide Gene Expression Profiling in Fission Yeast

Transcription

1 Genome-wide Gene Expression Profiling in Fission Yeast Jürg Bähler The Wellcome Trust Sanger Institute / Cancer Research UK

2 Post-genomic vs traditional experiments: Genes or gene products: Gene cloning Gene expression Gene deletion Protein localization Protein interactions Enzymatic activities 2 3 Vertical approach Horizontal approach n

3 Schizosaccharomyces pombe (Fission Yeast) unicellular eukaryote (fungus) genome recently sequenced: ~4 Mb, <5,000 genes easy to handle / genetics evolutionary distant to S. cerevisiae no beauty but what a beast!

4 Primer Design Annealing temperature: C; GC content: 40 60% Product length: bp; <2500 bp from gene end Products are 00% exon sequence Products have Blast score <400 with other fission yeast sequences All ORFs (nuclear and mitochondrial), pseudogenes, introns, bacterial control genes

5 Effects of probe size: Effects of probe position:

6 Unspecific hybridisation to similar probes:

7 5 -aminolink surface chemistry covalent attachment of DNA via 5 -aminolink modification single-stranded probes can be made after attachment probes can distinguish transcriptional direction entire length of probe accessible for hybridization high sensitivity

8 The fission yeast genome on a microarray 6050 spots printed in duplicate: 2,00 spots

9 Data Processing Pipeline Image Analysis Software: determine signal ratios (e.g., GenePix) InHouse program for initial data processing: filter weak and irreproducible signals, local normalization, quality control Data mining using various software (GeneSpring, Cluster, SAM, ArrayMiner, ) Oracle Database MIDAS: Plate tracking and storage of raw data

10 A Before normalization B Signal ratio After normalization C Signal ratio Running window 000 spots Signal ratio Local Normalization: Y position (µm) X position (µm)

11 log2(signal ratio) A log2(signal ratio) B log2(signal ratio) After normalization 0 C 3 Before normalization log0(signalcy3 signalcy5) log0(signalcy3 signalcy5) 6

12 Accuracy of signal ratios determined by spiking of S. cerevisiae RNA Spiked ratios Measured median ratios (range) Range of signals :2.9 (.8-2.) 490/270 2,080/32,060 :5 5.5 ( ) 420/90 58,320/0,990 :0 9.7 (5.8 -.) 380/40 35,890/5620 : ( ) 340/20 56,900/30

13 Reproducibility of signal ratios and intensities:

14 Reproducibility of array data: Measurement Mean SD (Range) CV (Range) Within array replicates 0.04 ( ) 4.4% (3.-6.2%) Technical repeats 0.04 ( ) 4.5% ( %) Biological repeats 0.07 ( ) 6.4% (4.9-8.%)

15 Three main projects: Normalized Intensity (log scale) MergeDC-GR-JM coloured by Venn Diagram 00 Drugs Sex 0 Rock n Roll 0. Stress Cell cycle Meiosis elu pat_processed

16 Expression Profiling During Sexual Differentiation: Juan Mata synchronized meiotic cell cultures transcription factor mutants genome-wide transcriptional program stage-specific gene expression regulatory sequence motifs and circuits

17 Mig u e l Fission yeast life cycle: Zygotic ascus Sexual differentiation Meiosis and sporulation Zygote formation Dormant ascospores Re-supply nutrients Conjugation G S Nutrient deprivation M G2 Vegetative cell cycle

18 Meiotic timecourses: major clusters of gene expression 000 N-starvation/pheromone induced 000 Early: S-phase/prophase Middle: meiotic divisions 000 Late: sporulation pat wt pat wt

19 0 Average Expression Profiles: 4 main clusters of successive gene expression waves -N / pheromone Early Middle Late Mean expression ratio pat wild type pat wt

20 Functions of early genes: 00 DNA replication 00 Chromosome cohesion Recombination 00 Nuclear movement Unknown 0

21 Known meiotic transcription factors are themselves regulated: 00 mei4 Expression ratio 0 rep res pat_ratios4 wt pat wt

22 Other transcription factors up-regulated during meiosis: 00 atf2 Expression ratio 0 atf pat_ratios4 wt pat wt

23 Characterization of atf function during meiosis: wt atf2 atf3 atf-dependent genes atf2 atf ~55% of late genes are atf2/3 dependent atf2 targets atf2 atf3 Expression ratio Time after meiotic induction

24 Transcriptional Regulation During Meiosis Pheromone response Early Middle Late?? ste rep mei4 atf2 atf3? rsv msn2 Glucose metabolism

25 Global regulation of genes up-regulated in N-starvation: Chromosome I Chromosome II 8 Gene density 4 0 Expression ratio Time (h) Time (h)

26 Stress Response Mechanisms: Dongrong Chen Mark Toone, Nic Jones (Paterson, Manchester) various stresses: heat, osmotic, toxic metals, oxidative stress: time, dose, different oxidants general vs specific stress responses acute vs adaptive stress response long-term adaptation to stress regulatory sequence motifs and circuits

27 Eukaryotic stress-activated protein kinase pathways Vertebrates Drosophila S. cerevisiae S. pombe MEKK MEKK ASK? Ssk22 Ssk2 Win Wak MEK MKK4/7 MKK3/6 Hep Pbs2 Wis SAPK JNK p38 Bsk Hog Sty Trxn factor targets cjun ATF2 DJun Msn4 Msn2? Pap Atf Cellular responses Immune cell activation Development DNA repair Apoptosis Control cell proliferation Dorsal closure Immune response Osmotic stress response Oxidative stress response Osmotic stress response UV response Meiotic program Multidrug resistance Heavy metal resistance

28 Importance of stress and stress response pathways in human pathology: - Ethanol induced liver damage - Ischemic hepatitis - Ischemia/reperfusion injury catalyzed by Xanthine Oxidase- transplantation biology - cerebrovascular injury - Parkinson s disease - Amyotrophic lateral sclerosis - Atherosclerosis - Ageing (senescence) - Cancer Homeostasis: - Kidney cells exposed to dramatic fluctuations in osmolarity - Metabolism of toxic compounds in the liver - Apoptosis - Immune system - Resistance to DNA damaging agents

29 CESR: Core Environmental Stress Response H 2 O 2 Cd Heat Sorb. MMS Average Expression Profiles 8 4 CESR induced 682 Genes Expression ratio CESR repressed 0.25 H 2 O 2 Cd Heat Sorb. MMS >8x induced >8x repressed

30 The transcriptional response to stress consists of general and stress-specific gene induction: CESR H 2 O 2 CESR Sorbitol CESR Heat CESR MMS CESR Cd

31 Regulation of stress response genes: wild-type H 2 O 2 Cd Heat Sorb. MMS sty atf H 2 O 2 Cd Heat Sorb. MMS H 2 O 2 Cd Heat Sorb. MMS 40 CESR genes Stress-specific genes >6x induced >6x repressed H 2 O 2 Heat Cd Sb MMS

32 Regulation of stress response genes: sty atf % Dependent genes H 2 O 2 Cd Heat Sorb. MMS H 2 O 2 Cd Heat Sorb. MMS CESR Non-CESR

33 Different regulation of stress response genes in two yeasts: Budding Yeast Fission Yeast stress stress 2 stress 3 stress stress 2 stress 3 Stress activated MAPK pathway core stress response genes core stress response genes

34

35 Growth media influence stress gene expression: CESR induced CESR repressed Meiotic divisions

36 Increasing cell density and stress gene expression: CESR induced CESR repressed

37 Increasing cell density and stress gene expression: Growth curve: 0.8 Cell density (OD) Time (minutes)

38 Expression Profiling During Cell Cycle: Gabriella Rustici synchronized cell cultures cell cycle and transcription factor mutants periodic (stage-specific) gene expression gene function ( guilt by association ) regulatory sequence motifs and circuits

39 All genes: 2 experiments 0 Normalized Intensity (log scale) merged 972 elu and cdc25 b&r elu cdc25 block and release TC3 time

40 ~200 periodic genes: 7 different experiments

41 Average expression profiles of four main clusters (7 experiments/58 samples):

42 4 major waves of transcription: 3 Normalized Intensity (log scale) 2 Expression ratios time Minutes after synchronization

43 Average gene expression profiles: 2.00 Normalized Intensity (log scale) M G S G2 M G S G2 Expression ratios time Minutes after synchronization

44 Principal component analysis:

45 Major Conclusions: 4 distinct waves of transcription, with gap during much of G2 phase where no genes seem to be regulated ~200 periodic genes (4% of genome): 26 of those genes have been described as cell-cycle regulated, while majority of remaining genes have unknown function Study regulation of periodic cell cycle transcription using various mutants, overexpressors, and Chip-chip Compare and contrast with periodically expressed genes in other organisms

46 General and simple recommendations: Repeat biological experiments to get statistically sound data Plan and design experiments carefully / controlled and standardized conditions Compare data from different experiments The more data the better! Explore data with various tools

47

48 Juan Mata Gavin Burns Gabriella Rustici Dongrong Chen Rachel Lyne Chris Penkett Jürg Bähler Mark Toone (Paterson) Nic Jones (Paterson) Alvis Brazma (EBI) Katja Kivinen (EBI) Helen Parkinson (EBI) Paul Nurse (CRUK) John Sgouros (CRUK) Sanger Institute: Dave Vetrie Cordelia Langford Oliver Dovey Val Wood Roger Pettett Rob Andrews Adam Butler Kate Rice Bart Barrell

49 Genome-wide Gene Expression Profiling in Fission Yeast The Wellcome Trust Sanger Institute / Cancer Research UK