3D genome conformation and gene expression in fetal pig muscle at - PowerPoint PPT Presentation

3D genome conformation and gene expression in fetal pig muscle at late gestation Maria Marti Marimon 4 December 2019 1

Agronomic interest Factors responsible of piglets mortality: weight, genotype and maturity • Maturity of fetal muscle - Motor functions - Thermoregulation Voillet et al. 2016 2

Agronomic interest Factors responsible of piglets mortality: weight, genotype and maturity • Maturity of fetal muscle - Motor functions - Thermoregulation Voillet et al. 2016 Muscle transcriptome study (Voillet et al. BMC Genomics, 2014) • 90 d 110 d ↑ genes ↑ genes Transcriptional change muscle energy associated to 3D genome development metabolism organization? ↓ genes ↓ genes energy muscle metabolism development 3

3D genome architecture Doğan ES & Liu C, 2018 4

3D Genome dynamics during early development Dixon JR, et al., Nature 2015 Ke Y., et al., Cell. 2017 5

3D Genome dynamics during early development Dixon JR, et al., Nature 2015 Ke Y., et al., Cell. 2017 Cell differentiation Zygote genome activation expression programs 6

Experimental design 90 d 110 d Ø Gene expression (Voillet et al. 2014) ↑ genes ↑ genes 90 days gestation muscle energy development metabolism 110 days gestation ↓ genes ↓ genes energy muscle metabolism development ? 7

Experimental design 90 d 110 d Ø Gene expression (Voillet et al. 2014) ↑ genes ↑ genes 90 days gestation muscle energy development metabolism 110 days gestation ↓ genes ↓ genes energy muscle metabolism development ? ? Ø 3D Genome organization 3 fetuses (90 days gestation) In situ Hi-C on fetal muscle • Rep1-90 • Rep2-90 Rep3-90 • 3 fetuses (110 days gestation) Rep1-110 • Rep2-110 • Rep3-110 • Rao et al, 2014 8

Hi-C data analysis HiC-Pro (Servant et al. 2015) 476–685M read pairs/sample Paired-end (PE) reads 3.45 billion read pairs (total) Read 63-73% mapped pairs/sample alignment Detection of 122–283M valid pairs/sample valid pairs Raw Contact Maps Normalized Contact Maps TADs finding A/B compartments 9

Hi-C data analysis HiC-Pro (Servant et al. 2015) 476–685M read pairs/sample Paired-end (PE) reads 3.45 billion read pairs (total) Read 63-73% mapped pairs/sample alignment 100% 0,89% 2,11% 2,56% 2,30% 2,64% 3,78% 90% Detection of 80% 122–283M valid pairs/sample 43,09% valid pairs 45,92% 50,35% 51,07% 51,99% 70% 55,51% 60% 50% Raw Contact 40% Maps 30% 56,01% 51,79% 47,54% 45,45% 46,29% 20% 40,71% 10% Normalized 0% Contact Maps Rep1-90 Rep2-90 Rep3-90 Rep1-110 Rep2-110 Rep3-110 Trans valid pairs Cis long-range valid pairs Cis short-range valid pairs TADs finding High percentages of trans valid pairs Low percentages of cis short-range valid pairs A/B compartments 10

Hi-C data analysis HiC-Pro (Servant et al. 2015) Paired-end (PE) reads 500 kb 200 kb 50 kb Read alignment Detection of valid pairs Raw Contact 500 kb 200 kb 50 kb Maps Normalized Contact Maps TADs finding A/B compartments 11

Hi-C data analysis (TADs) Juicer: arrowhead (Neva et al., Cell Systems 2016; Rao et al., Cell 2014) Paired-end (PE) reads - 50 Kb resolution matrices - 1312 TADs per replicate on average Read - Average mean size: 1480 Kb alignment - Global conservation of TAD structure (74 – 79% of TAD boundaries from each TADs condition are identical to the other Detection of detection condition) valid pairs Raw Contact Maps Normalized Contact Maps TADs finding Kim et al., Cell 2016 A/B Li et al. 2016 compartments 12

Hi-C data analysis (TADs) CTCF CTCF TADs validation Consistent Hi-C data 13

Hi-C data analysis (A/B compartments) (Lieberman-Aiden et al., Science 2009) Paired-end (PE) reads Read alignment Detection of valid pairs Raw Contact Maps Normalized Contact Maps TADs finding 500 Kb resolution matrices A/B 682 compartments/replicate (average) compartments Median size 2.6 Mb – 3.5 Mb 14

Hi-C data analysis (A/B compartments) (Lieberman-Aiden et al., Science 2009) Paired-end (PE) reads Gene density in A/B compartments Gene expression in A/B compartments Read alignment Detection of valid pairs Raw Contact Maps Normalized Contact Maps TADs finding 500 Kb resolution matrices A/B 682 compartments/replicate (average) compartments Median size 2.6 Mb – 3.5 Mb 15

A/B compartments Bins assigned to the same compartment type: - 83.3% in all 6 replicates Good consistency of results across replicates 16

Genomic regions switching compartments 17

Genomic regions switching compartments Variability between conditions: 3.3% switching bins (52 Mb) 90 d è 110 d 43.3% (AAA è BBB) 56.7% (BBB è AAA) 18

A/B compartments and gene expression 19

A/B compartments and gene expression Switching regions are associated to transcriptional changes 20

Genome-wide fragmentation during the muscle maturation process Number distribution of compartments 21

Genome-wide fragmentation during the muscle maturation process Number distribution of compartments Fragmentation of genome compartmentalization 22

Differentially distal genomic regions Ø Filtering, normalization and detection of bin pairs with significant number of contacts (method: Generalized Linear Model “GLM” functionality of edgeR) 500 Kb 200 Kb Total bin pairs with any count 9,262,199 3,844,272 Differential bin pairs 10,183 (0.11%) 3,417 (0.09%) 23

Differentially distal genomic regions Ø Filtering, normalization and detection of bin pairs with significant number of contacts (method: Generalized Linear Model “GLM” functionality of edgeR) 500 Kb 200 Kb Total bin pairs with any count 9,262,199 3,844,272 Differential bin pairs 10,183 (0.11%) 3,417 (0.09%) logFC (bin pair) = log2 [ (counts at 110 days) / (counts at 90 days) ] Positive logFC = more counts “contacts” at 110 days than at 90 days = genomic regions closer to each other Negative logFC = more counts “contacts” at 90 days than at 110 days = genomic regions closer to each other 24

Gene expression in differentially distal genomic regions Distributions of logFC expression values of probes mapped to different categories of genomic regions 25

Gene expression in differentially distal genomic regions Distributions of logFC expression values of probes mapped to different categories of genomic regions The expression values of probes in genomic regions closer at either 90 or 110 days are significantly lower 26

Differential interacting regions (90-110 days of gestation) 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 27

Differential interacting regions (90-110 days of gestation) cis bin pairs 81.8% 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 28

Differential interacting regions (90-110 days of gestation) 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 29

Differential interacting regions ( cis ) Positive logFC Negative logFC 30

Differential interacting regions ( cis ) Positive logFC Negative logFC 31

Differential interacting regions ( cis ) Positive logFC Negative logFC Large dynamic differential regions (90-110 days gestation) 32

Differential genomic regions ( trans ) Positive logFC Negative logFC 33

Differential genomic regions ( trans ) Positive logFC Negative logFC Telomeric regions Negative logFC ptel qtel 34

Differential genomic regions ( trans ) Positive logFC Negative logFC Telomeric regions Negative logFC ptel qtel Preferential clustering of telomeres at 90 days 35

Differential genomic regions ( trans ) Positive logFC Negative logFC Telomeric regions Negative logFC ptel qtel Preferential clustering of telomeres at 90 days 36

Preferential associations of telomeres (90 days gestation) A SSC2pter-SSC9qter SSC13qter-SSC9qter SSC15qter-SSC9qter 37

General output Ø Changes in genome structure at late gestation è switching A/B compartments è genome-wide fragmentation è differentially interacting regions (telomeres) Ø These changes are associated with variations in gene expression 90 d 110 d ↑ genes muscle ↑ genes energy development metabolism Gene expression (Voillet et al. 2014) ↓ genes energy ↓ genes muscle Expression changes associated metabolism development to the switching regions Expression changes associated 3D structure to differentially distal regions 3,1% regions switching compartment Up to 10,000 differential interacting pairs 38

Hi-C working team: Experiments: Hervé Acloque & Florence Mompart Sequencing: Diane Esquerré Data analysis: Sylvain Foissac , Sarah Djebali , Matthias Zytnicki & David Robelin Statistic analysis : Nathalie Vialaneix Cytogenetic team: Yvette Lahbib-Mansais Martine Bouissou-Matet Funding: SCALES projet (CNRS): Pierre Neuvial & Nathalie Vialaneix 39