Exploring the relation between evolutionary gene age, gene expression and chromatin 3D structure in cancer

Flavien Raynal, Benoît Aliaga, Kaustav Sengupta, Dariusz Plewczynski, Vera Pancaldi

Monday, September 25^th

6th course on Computational Systems Biology of Cancer: models of data, data for models

Institute Curie, Paris

2

Pancaldi, V., Carrillo-de-Santa-Pau, E., Javierre, B.M. et al. Integrating epigenomic data and 3D genomic structure with a new measure of chromatin assortativity. Genome Biol 17, 152 (2016).

Botta et al. 2010, Intra‐ and inter‐chromosomal interactions correlate with CTCF binding genome wide

KS Sandhu et al. 2012, Large-scale functional organization of long-range chromatin interaction networks

Boulos et al. 2017 Multi-scale structural community organisation of the human genome

Mourad et al. 2017 Uncovering direct and indirect molecular determinants of chromatin loops using a computational integrative approach

Norton et al. 2018 Detecting hierarchical genome folding with network modularity

Huang et al. 2020 A subset of topologically associating domains fold into mesoscale core-periphery networks

We can use networks to study 3D chromatin structure

PCHiC or Hi-C-seq

What is chromatin assortativity?

3

Pancaldi, V., Carrillo-de-Santa-Pau, E., Javierre, B.M. et al. Integrating epigenomic data and 3D genomic structure with a new measure of chromatin assortativity. Genome Biol 17, 152 (2016).

Newman, M. E. J. (2002). Assortative Mixing in Networks. Physical Review Letters, 89(20), 208701.

Madrid-Mencia et al. 2020  Using GARDEN-NET and ChAseR to explore human haematopoietic 3D chromatin interaction networks

Assortativity, or homophily in social networks, is the property of networks in which nodes with similar characteristics tend to contact each other more than expected by chance (Newman, 2002).

ChAs > 0

ChAs = 0

ChAs < 0

Cancer and tumor microenvironment

4

• Tumors are a complex object.

​

• They can be divided into three “big” components:
• Cancer cells
• Immune cells
• Fibroblasts and other microenvironment components

​

• For each cell type, wide or narrow heterogeneity can explain drug resistance (immunotherapy) and tumor plasticity.

​

• Understanding the cell phenotype is an important key if we want to overcome this disease.

​

• Which mechanisms are involved in normal and cancer cell phenotype and its heterogeneity?

The atavistic theory of cancer

5

Reduced heterogeneity/plasticity

Well-defined phenotype

Large fluctuations, motion across landscape

Plasticity – in response to external signals

Atavism is the theory that some individual animals for some reason, revert back to an earlier evolutionary type.

Atavistic theory of cancer: cancer cells re-express early genes and forget the multi-cellular context

The phenotype and the epigenome landscape

6

Genes

7

Gene ages and their expression in cancer

Trigos et al. 2017 Altered interactions between unicellular and multicellular genes drive hallmarks of transformation in a diverse range of solid tumors

Trigos et al. 2019 Somatic mutations in early metazoan genes disrupt regulatory links between unicellular and multicellular genes in cancer

Gene ages and 3D structure

Trigos et al. 2019 Somatic mutations in early metazoan genes disrupt regulatory links�between unicellular and multicellular genes in cancer�Trigos et al 2017 Altered interactions between unicellular and multicellular genes drive hallmarks of transformation in a diverse range of solid tumors.

​

Genes classed by age

​

Old genes

-more housekeeping roles

-altered connectivity in PPIs in cancer

-more expressed through Oncogenesis

Low TAI = high exp of early genes

​

​

​

​

​

​

�Atavistic theory:

Cancer cells re-express early genes

Forget multi-cellular context

Gene ages on human monocyte PP network

A lower TAI corresponds to higher expression of genes from earlier phylostrata

Do we have a link between gene age and gene expression variability?

9

Trigos et al. 2017 Altered interactions between unicellular and multicellular genes drive hallmarks of transformation in a diverse range of solid tumors

Trigos et al. 2019 Somatic mutations in early metazoan genes disrupt regulatory links between unicellular and multicellular genes in cancer

Gene ages and gene expression variability is linked in immune cells

10

Which relation between epigenetic markers and gene ages in monocytes?

11

Active marks are highly enriched around old genes

Polycomb complexes target mainly EM genes

DNA methylation and gene ages?

12

MM gene promoters are highly enriched in CpG

MV of MM genes reflect newest genes expression variability

In conclusion, in monocytes, gene ages are related to …

13

Study the 3D genome structure during differentiation and cancer

14

hESC

B cell

Leukemia / Lymphoma

Undifferentiated

Healthy differentiated

Cancerous

Polycomb target genes

15

Whole genome

By gene ages

Polycomb target gene levels decrease through cancerogenesis

Polycomb target gene levels increase through differentiation

3D chromatin structure in network

16

hESC

3D chromatin structure alterations in differentiation and cancer

17

hESC

B cell

Leukemia

ChAS

ChAS

ChAS

Conclusion

18

• Variability of gene expression is associated to 3D chromatin structure

​

• In terms of chromatin 3D structure, we think evolutionary older genes are found at the core of the chromatin network, distributed in a set of cores, also broadly corresponding to the least variable genes.

​

​

• We identify consistent changes during oncogenesis in the spatial organization of genes (maybe specific to evolutionary classes), reinforcing the important role that old genes and their organization in the nucleus play in cancer phenotypes.

​

• These findings suggest a potential atavistic theory confirmation at the level of chromatin organization.

Thank you for your attention

19

Group Leader:   Vera Pancaldi

Postdocs:   Malvina Marku

  Leila Khajavi

  Benoît Aliaga

Yann Aubert

Engineers:   Flavien Raynal

  Abdel Mounim Essabbar

Julie Bordenave

PhD students: Alexis Hucteau (co-sup J-E. Sarry)

  Matthieu Genais (co-sup B. Segui)

Jacobo Solórzano (co-sup Y. Martineau)

Marcelo Hurtado

  Hugo Chenel

Hafida Hamdache

​

Visiting student:   Kaustav Sengupta (Plewczynski lab)

3D polymer representation

20

hESC

B cell

Leukemia

Work in progress

Kaustav Sengupta

Two aims of this work

21

We will focus on two components of the tumors:

Study the inter-individual gene expression variability in immune cells

Study the 3D genome structure during differentiation and cancer

• Do we have a relationship between gene age and gene expression variability in immune cells?

​

• If yes, which epigenetic mechanisms are involved in gene expression variability?

​

• Why is it important?

​

• If we understand the mechanisms of variability in immune cells, we can improve immunotherapy.

​

• Do we have a relationship between gene ages and 3D genome structure?

​

• Why is it important?

​

• The epigenomic modifications can explain cancer cell plasticity, and then we hope to develop a future strategy to cut off cancer cell development.

​