Branching and oscillations in the epigenetic landscape of cell-fate determinationby Jomar Fajardo Rabajante, Ariel Lagdameo Babierra

Progress in Biophysics and Molecular Biology

About

Year
2015
DOI
10.1016/j.pbiomolbio.2015.01.006
Subject
Molecular Biology / Biophysics

Text

Accepted Manuscript

Branching and oscillations in the epigenetic landscape of cell-fate determination

Jomar Fajardo Rabajante, Ariel Lagdameo Babierra

PII: S0079-6107(15)00007-3

DOI: 10.1016/j.pbiomolbio.2015.01.006

Reference: JPBM 980

To appear in: Progress in Biophysics and Molecular Biology

Received Date: 25 September 2014

Revised Date: 5 January 2015

Accepted Date: 18 January 2015

Please cite this article as: Fajardo Rabajante, J., Babierra, A.L., Branching and oscillations in the epigenetic landscape of cell-fate determination, Progress in Biophysics and Molecular Biology (2015), doi: 10.1016/j.pbiomolbio.2015.01.006.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

M

AN

US

CR

IP

T

AC

CE

PT

ED

ACCEPTED MANUSCRIPT 1

Branching and oscillations in the epigenetic landscape of cell-fate determination

Authors: Jomar Fajardo Rabajantea,*, Ariel Lagdameo Babierraa aInstitute of Mathematical Sciences and Physics, University of the Philippines Los Baños,

College, Laguna 4031 Philippines *Corresponding author. E-mail address: jfrabajante@up.edu.ph. Present address: Shizuoka

University, Hamamatsu, Japan.

Abstract. The well-known Waddington’s epigenetic landscape of cell-fate determination is not static but varies because of the dynamic gene regulation during development. However, existing mathematical models with few state variables and fixed parameters are inadequate in characterizing the temporal transformation of the landscape. Here we simulate a decisionswitch model of gene regulation with more than two state variables and with time-varying repression among regulatory factors. We are able to demonstrate multi-lineage differentiation at different timescales that portrays the branching canals in Waddington’s illustration. We also present a repressilator-type system that activates suppressed genes via sustained oscillations in a flattened landscape, hence providing an alternative scheme for cellular reprogramming. The time-dependent parameters governed by gradient-based dynamics regulate cell differentiation, dedifferentiation and transdifferentiation. Our prediction integrates the theories of branching and structural oscillations in cell-fate determination, which reveals key temporal patterns of cell differentiation and associated diseases, such as cancer.

Keywords. gene regulatory network, stem cells, pluripotency, synthetic biology, multistability, attractor

M

AN

US

CR

IP

T

AC

CE

PT

ED

ACCEPTED MANUSCRIPT 2

Main text:

Waddington’s epigenetic landscape illustrates the canalization in the cell differentiation and fate determination process [1-4]. The topography of Waddington’s illustration represents the developmental pathways of tissues formed from totipotent and pluripotent cells to terminallydifferentiated specialized cells (Fig. 1a). Various theoretical studies have quantified

Waddington’s epigenetic landscape and are able to predict bistability in gene regulatory networks (GRNs) [5-9]. However, many of the mathematical models only consider at most two regulatory factors, and focus on static epigenetic landscape represented by fixed parameter values. In reality, the topography of Waddington’s illustration is dynamic and involves many regulatory factors (high-dimensional [10,11]). The parameters that represent gene regulation are indeed changing during the development of an organism [10,12-17].

Mathematical models with two regulatory factors and fixed parameter values only describe a particular temporal scenario in cell differentiation.

The mechanisms that regulate gene expression, such as kinetics of gene regulatory factors (GRFs) and the structure of GRF-GRF interaction, influence the outcome of cell-fate determination [7,10,12,15-18]. Waddington observed that changes in these mechanisms could alter the epigenetic landscape leading to cell-lineage switching [1]. The changes in the GRFGRF interaction do not necessarily entail mutations but can be due to normal processes. In mathematical point-of-view, the variations in gene regulation can be represented by modifications in the parameter values of the quantitative models. Bifurcation analyses of existing models have been done [5,7,12,16,17], but most of them do not provide elaborate illustrations of cells trailing the dynamic pathways governed by more than two GRFs. Here we present numerical illustrations of cells trailing different epigenetic routes such that the pathways transform due to changes in the strength of repressive interaction among multiple

GRFs (see Fig. 1b and Box 1 for the mathematical model). The GRFs in the model (Fig. 1b)

M

AN

US

CR

IP

T

AC

CE

PT

ED

ACCEPTED MANUSCRIPT 3 have mutual repression because a mature cell expresses only one phenotype and constrains the expression of the other phenotypes.

An example of GRN with mutual repression is the coarse-grained mesenchymal transcription network shown in Figure 1c. The upper module consists of lineage-specifying master genes, namely, PPAR-γ, RUNX2 and SOX9 [18]. Auto-activation is a common property of master genes [10]. In mouse embryonic stem cells, the up-regulation of PPAR-γ by retinoic acid and insulin, while inhibiting the growth of RUNX2 and SOX9, steers adipogenesis (fat formation). The up-regulation of RUNX2 by retinoic acid and BMP4 represses PPAR-γ and SOX9 and drives osteogenesis (bone formation). On the other hand, retinoic acid and TGF-β up-regulate SOX9 leading to chondrogenesis (cartilage formation) and to the inhibition of PPAR-γ and RUNX2 [18]. The activation of the upper module (Fig. 1c) represents cell differentiation towards specific lineages and is hindered by the module of pluripotency factors (lower module in Fig. 1c).