Understanding epigenetic changes in aging stem cells--a computational model approach

Abstract

During aging, a decline in stem cell function is observed in many tissues. This decline is accompanied by complex changes of the chromatin structure among them changes in histone modifications and DNA methylation which both affect transcription of a tissue-specific subset of genes. A mechanistic understanding of these age-associated processes, their interrelations and environmental dependence is currently lacking. Here, we discuss related questions on the molecular, cellular, and population level. We combine an individual cell-based model of stem cell populations with a model of epigenetic regulation of transcription. The novel model enables to simulate age-related changes of trimethylation of lysine 4 at histone H3 and of DNA methylation. These changes entail expression changes of genes that induce age-related phenotypes (ARPs) of cells. We compare age-related changes of regulatory states in quiescent stem cells occupying a niche with those observed in proliferating cells. Moreover, we analyze the impact of the activity of the involved epigenetic modifiers on these changes. We find that epigenetic aging strongly affects stem cell heterogeneity and that homing at stem cell niches retards epigenetic aging. Our model provides a mechanistic explanation how increased stem cell proliferation can lead to progeroid phenotypes. Adapting our model to properties observed for aged hematopoietic stem cell (HSC) clones, we predict that the hematopoietic ARP activates young HSCs and thereby retards aging of the entire HSC population. In addition, our model suggests that the experimentally observed high interindividual variance in HSC numbers originates in a variance of histone methyltransferase activity.

Keywords: DNA methylation; aging of stem cells; clonal competition; histone modification; mathematical model; population dynamics.

Figure 1

Multi-scale model of epigenetic drifts in aging cells. (A) Time and length scales covered by our model. The basic unit of the model is the nucleosome, the histones of which undergo H3K4 (de-) modification. The average H3K4me3 level controls the access of CpG-methylating enzymes to nearby CpG sites. Both H3K4 and DNA methylation interfere with the transcription factor-network of the cell that is defined by its genome. All cells are equipped with the same genome but underlie individual aging and thus reside in individual regulatory states. Aging depends on signals provided by the environment. We consider two environments α and Ω. (B) In the α environment mimicking a stem cell niche (blue) cells are quiescent, while in the Ω environment (orange), they can proliferate and differentiate with rate R and q, respectively. Differentiated cells are removed from the system. Cells make transitions between α and Ω with probabilities P_α and P_Ω. (C) Regulatory profiles of individual cells: Transcription, H3K4me3 modification, and DNA-methylation profiles (red-yellow-white: low–high values) for a cell fixed in environment α (upper row) and Ω (lower row). H3K4me3 modification and DNA methylation are given as the fraction of modified nucleosomes and methylated CpG sites associated with the gene, respectively. Time is given in computational time steps Δt. Proliferation is required for changes in DNA methylation, while changes of H3K4me3 modification occur also in quiescent cells. (D) Age-dependent phenotypic changes. Initially the nucleosomes carry H3K4me3 modifications which protect the associated DNA from becoming methylated. Over time this modification is lost enabling methylation of the nearby CpG sites and stable silencing of the associated gene.

Figure 2

Simulation results for an age-independent phenotype: (A) Regulatory states for each gene averaged separately over all cells in the α- and Ω-environment (red-yellow-white: low-high, 200 Δt ~ 1 generation). Genes are sorted by the average H3K4me3 modification in Ω. Values for transcription and H3K4me3 modification are normalized to their maximum value. H3K4me3 and DNA-methylation are given as the fraction of modified nucleosomes and methylated CpG sites associated with the gene, respectively. Three different sets of genes C1–C3 can be defined (see text). H3K4me3 decreases for C2-genes in both environments and for C1a-genes in Ω. In Ω cell proliferation allows for increasing DNA methylation, while DNA of quiescent cells in α remains un-methylated (not shown). (B) Variance of the states for each gene averaged separately over all cells in the α- and Ω-environment. (green-brown-white: low–high, all values normalized to their maximum value). Genes sorted as in (A). Regulatory states of C2-genes, in particular H3K4me3 states, are more heterogeneous in α compared with Ω. (C) Number of cooperative nucleosomes associated with the genes. (D) Differential expression of the genes in the two environments (lnα–lnΩ). The genes of set C2 and of set C1a show higher expression (yellow, white) in the niche.

Figure 3

Age-related phenotype feedback on epigenetic states. (A) Simulated cell numbers for decreased differentiation rate q (D_NOVO = 0.3, TS = 2). Shown are cell numbers in α (black: young, gray: old) and in Ω (red: young, pink: old). (B) Simulated cell numbers for decreased proliferation rate R (D_MAIN = 0.8, TS = 2). Colors as in A. Inserts: Differences in histone and DNA methylation between systems without and with ARP. Changes in phenotype controlling genes (red) and other C1a-genes (black) are shown as averages over all cells of the system. (A) In case of a dominant ARP, aging of all C1a-genes becomes accelerated, that is, histone modification (DNA methylation) in the system without a phenotype is larger (smaller) compared to the system with an ARP. (B) In case of a recessive ARP, aging becomes selectively retarded in C1a-genes controlling the ARP but not in the other C1a-genes.

Figure 4

Simulated hematopoietic stem cell (HSC) aging. (A) The number of HSCs in the niche α increases with age, while the HSC number in Ω remains constant. Colors were chosen as in Fig. 3A. (B) Clonal composition of the system. Numbers at the top indicate the number of clones present in the system. Each clone is represented by an individual color. (C) The color saturation indicates the fraction of aged cells. Aged and young cells coexist for long times. (D) The average generation number is reduced in the HSC- (red) compared to the control system. (E,F) Differences in the epigenetic states in the HSC system compared to the system without an ARP. Phenotype controlling genes (red) and other C1a-genes (black) show a tendency for higher modification of the associated nucleosomes (E) and lower DNA methylation (F) in HSCs. (G) Emergence of the ARP sensitively depends on the histone modification rate k_M. Decreasing k_M (left, 0.6 k_M,0) strongly accelerates aging, while increasing it (right, 1.4 k_M,0) decelerates it. (H) As a consequence, a high interindividual variance of the cell number is observed assuming individuals with differences in k_M (left), in contrast to a group of individuals with identical k_M (right). Black open symbols are results for 10 different individuals. Red dots are average values.