Modelling stem cell ageing: a multi-compartment continuum approach.

Stem cells are important to generate all specialized tissues at an early life stage, and in some systems, they also have repair functions to replenish the adult tissues. Repeated cell divisions lead to the accumulation of molecular damage in stem cells, which are commonly recognized as drivers of ageing. In this paper, a novel model is proposed to integrate stem cell proliferation and differentiation with damage accumulation in the stem cell ageing process. A system of two structured PDEs is used to model the population densities of stem cells (including all multiple progenitors) and terminally differentiated (TD) cells. In this system, cell cycle progression and damage accumulation are modelled by continuous dynamics, and damage segregation between daughter cells is considered at each division. Analysis and numerical simulations are conducted to study the steady-state populations and stem cell damage distributions under different damage segregation strategies. Our simulations suggest that equal distribution of the damaging substance between stem cells in a symmetric renewal and less damage retention in stem cells in the asymmetric division are favourable strategies, which reduce the death rate of the stem cells and increase the TD cell populations. Moreover, asymmetric damage segregation in stem cells leads to less concentrated damage distribution in the stem cell population, which may be more robust to the stochastic changes in the damage. The feedback regulation from stem cells can reduce oscillations and population overshoot in the process, and improve the fitness of stem cells by increasing the percentage of cells with less damage in the stem cell population.

Introduction

Stem cells are characterized by their ability to give rise to a variety of cell types through self-renewal and differentiation [1]. Although the process is very dynamic, the stem cell population is stable and remains almost steady. When a stem cell divides, each progeny has the potential to remain as a stem cell (self-renewal) or becomes a cell with a more specialized function (differentiation) (figure 1a). If the division produces the same type of cells, such as two stem cells and two differentiated cells, it is called a symmetric division. On the other hand, if the division produces two different types of cells, i.e. a stem cell and a differentiated cell, then it is called an asymmetric division. Available data suggest that most stem cells are able to switch between symmetric and asymmetric divisions, and the balance between these modes is controlled by various internal and external signals to produce appropriate numbers of stem and differentiated cells [1-3].

Figure 1.

Cell divisions and damage segregation in stem cells. (a) Three types of division in stem cell population: SR stands for symmetric renewal, ASR & D stands for asymmetric renewal and differentiation and SD stands for symmetric differentiation. (b) Commonly recognized ageing factors, such as protein aggregates, dysfunction organelles and DNA damage are segregated asymmetrically between two cells during division.

Stem cells can be found in most mammalian tissues, and they participate in tissue repair in response to damage and maintaining tissue homeostasis [4,5]. Research suggests that the decline in adult tissue maintenance and the increase in cancer formation might be a consequence of stem cell ageing [6]. Although age-related manifestation in stem cell population and function differ across tissues and organisms, decline in regenerative capacity due to depletion or dysfunction of stem cells is a hallmark [7]. Protein aggregates, dysfunction organelles and DNA damage are commonly identified as factors of ageing [4,5]. To slow down the accumulation of ageing factors, a hypothesis suggests that mitotic cells (actively dividing cells) might asymmetrically segregate damage away from the cell whose fate is to become a new stem cell [8]. The asymmetric inheritance of cellular components in dividing cells was first observed in yeast, and has been extensively studied over the past decades. In yeast, carbonylated proteins, extrachromosomal ribosomal DNA circles and dysfunction mitochondria are retained by a mother cell during asymmetric division, while its daughter is rejuvenated with little damage [9-11]. Recent evidence suggests that stem cells may employ a similar mechanism to protect one progeny from ageing [12-18] (figure 1b). An asymmetric partition of damaged proteins in stem cell division was observed in adult flies’ intestine and germline [12], mammalian stem cells [16], and murine neural stem cells [18]. Despite these emerging findings, it still remains elusive how stem cells cope with damage accumulation and how this is related to stem cell proliferation and differentiation.

Driven by the lack of knowledge of mechanisms regulating stem cell maintenance and differentiation in the ageing process, mathematical models have been employed to address key questions and provide quantitative insights into stem cell renewal and differentiation, as well as the decline of cellular functions in the ageing process. Several mathematical models were proposed to study stem cell population and ageing, which fall into two categories: individual-based modelling and continuous population modelling [19,20].

Individual-based modelling simulates individuals or agents that have a unique set of state variables and usually interact with each other in the local environment [20,21]. The advantages of this approach include that it can take stochastic effects into consideration to describe phenomena on the level of individual cells, and that detailed molecular dynamics and cell–cell interaction can be incorporated. In [22], a stochastic model of stem cell organization was introduced to explain the observed heterogeneity of haematopoietic stem cells by the stochastic switching between the growth environments and the self-organizing process based on within-tissue plasticity properties. Assuming that cell proliferation is negatively affected by telomere shortening, an agent-based stochastic model [23] was proposed to study telomere-dependent stem cell replicative ageing. This model provides a good approximation of the qualitative growth of cultured human mesenchymal stem cells. In [24], mutation accumulation in large populations of stem cells was modelled by a discrete-time branching process where each division produces 0, 1 or 2 stem cell daughters, each of which randomly accumulates a mutation. This model demonstrated that symmetric division could reduce the risk of accumulating phenotypically silent heritable damage in individual stem cells. However, a major drawback of individual-based modelling is the computational inefficiency, especially when the population is large. When a population is large and homogeneous, continuous population model using ODEs [25-31] or PDEs [32-38] are more appropriate to describe the population dynamics. In [39], a maturity-structured two-compartment model was proposed to study the regulation of mammalian red blood cell production. The model consists of two transport equations describing population densities of mitotic cells and post-mitotic cells, and an ODE describing hormone dynamics. The model showed that a perturbation of blood-donation type leads to damped oscillatory return to normal status, and that an elevated random peripheral destruction of red blood cells leads to sustained oscillations. In [40], a three-compartment ODE model was applied to study the dynamics of stem cells, transit-amplifying cells and terminally differentiated (TD) cells in the olfactory epithelium of mice. The authors identified conditions on parameters for the stability of the system when negative feedback loops are present either as Hill functions or in more general forms. Their analysis suggested that two factors, autoregulation of the proliferation of transit-amplifying progenitor cells and low death rate of TD cells, enhance the stability of the system. In [41], the authors applied the mean-field approach to approximate an agent-based model for studying heterogeneity within the haematopoietic stem cell population. Their proposed PDE model can capture the key structure of the model including the ‘age’-structure of stem cells and improve the efficiency of the numerical algorithms. In [42], a system of PDEs was used to model mutation accumulation hierarchy and differentiation hierarchy of cells with stem cells on the top level and to examine cancer development and growth. In their model, maturity is treated as a continuous variable, while the number of mutation accumulation and telomere shortening are treated as discrete cell classes. The boundary conditions describe transition among different cell classes at division: cells lose telomeres and acquire mutations. The study showed that the more mutation classes and higher proliferation rate are sufficient to explain the faster growth of the cancer cell population.

To understand the effects of different damage segregation strategies on stem cell ageing, we propose a novel model to integrate stem cell proliferation and differentiation with damage accumulation in the stem cell ageing process, and feedback regulation from stem and TD cells. A system of hyperbolic PDEs is constructed to model two compartments in cell lineage: mitotic cells (stem cells) and post-mitotic cells (TD cells). It is assumed that the cell cycle progression of stem cells is a continuous process while stem cell division is discrete. The boundary conditions of the PDEs model the stem cell renewal and differentiation at division when damage segregation takes place. Cell death is modelled as an outcome of damage accumulation. Stem cell proliferation and differentiation are regulated by feedbacks from the population of TD cells and stem cells. Ageing effect is modelled through the inhibition from the damage accumulation on stem cell proliferation and self-renewal. Our analysis and numerical simulations are carried out to compare the effects of different regulation and damage segregation strategies on population dynamics and stem cell fitness, which have not been discussed in the previous studies of stem cell regulation.

Our simulations suggest that equal distribution of the damaging substance between stem cells in symmetric renewal and less damage retention in stem cell in asymmetric division are favourable strategies, which reduce the death rate of stem cells and increase TD cell populations. Also, asymmetric damage segregation in stem cells leads to less concentrated damage distribution in stem cells population, which may be more robust to the stochastic change in damage. Compared to the feedbacks solely from TD cells, the feedback regulation from stem cells (autoregulation) can reduce oscillations and population overshoot in the process, and improve the fitness of stem cells by increasing the percentage of stem cells with less damage in the stem cell population.

This paper is structured as follows. The general description of our model is given in §2. In §3, a simple model without feedback regulations is presented to analyse the relation between population dynamics and various parameters. In §4, two more complex models with feedbacks from TD cells and stem cells are proposed to study different regulation mechanisms and the effect of segregation strategies.

Model description

In our mathematical model, a simplified conceptual model, we consider two types of cells: mitotic cells, which include stem cells and multiple progenitor cells, and post-mitotic cells, which include all TD cells. For simplicity, we call mitotic cells stem cells, and denote its population density by S; and we call post-mitotic cells TD cells, and denote its population density by T.

To consider the evolution of cell populations, we define both S(t, p, a) and T(t, p, a) as functions of time t and two other continuous biological state variables: cell cycle progression p and damage level a. To describe cell maturation and ageing processes, we make the following assumptions: By conservation law, a system of transport equations are derived to describe the evolution of S and T: and where V and U are rates of cell cycle progression for stem and TD cells, respectively; V and U are rates of damage accumulation for stem and TD cells, respectively. Note that various of feedbacks may regulate cell cycle progression and cell damage accumulations, and these functions will be specified in the following sections.

When the amount of damage a accumulates and reaches a certain threshold a*, stem cells die by an apoptosis-like process as a result of ageing via damage accumulation; similar to stem cells, when the amount of damaged substance reaches a certain threshold a, TD cells are removed by an apoptosis-like process.

Cell cycle progression p is an indicator variable, for which a stem cell divides when p increases to a threshold p*, unless damage has already reached a threshold a*, in which case the stem cell is removed before division. Although TD cells no longer divide, for simplicity, we keep cell cycle progression variable p in T, but assume the upper boundary for p is infinity.

The boundary conditions at p = 0 describe the reproduction/division process accompanied by the segregation of damage substances. First, we assume that there are three types of cell divisions: One of these three types of cell division occurs at the end of cell cycle p = p* with probability δ1, δ2 or δ3, where δ1 + δ2 + δ3 = 1 (figure 2a). These three probabilities may be regulated by various feedbacks and will be specified in later sections.

Figure 2.

Model descriptions of two-compartment stem cell system. Stem cells (mitotic cells) and TD cells (post-mitotic cells) are modelled as two compartments. (a) Stem cells renew themselves and replenish TD cells. The cell cycle progression (p) and damage accumulation (a) are modelled as continuous processes. Stem cell division and damage segregation take place at the end of its cell cycle (p = p*). Cells die when damage reaches a lethal threshold (a* or a). (b) In the simple model in §3, the proportions of three types of division are constants δ, and the damage segregation rules are fixed, i.e. α, β, γ are constants.

The daughter cells after division are two stem cells; this occurs with probability δ1.

The daughter cells after division are two TD cells; this occurs with probability δ2.

The daughter cells after division are one stem cell and one TD cell; this occurs with probability δ3.

Upon the completion of a cell cycle, the stem cell cycle progression p will be reset to zero and the damaged proteins are inherited from mother to daughters. The damage inheritance can be described by transition kernels r(a, a′) and r(a, a′). Based on the above assumptions, the boundary conditions for S and T cells at p = 0 are as follows: and The first term of the right-hand side of (2.3) represents the stem cell production process through type (i) cell division; the second term of the right-hand side of (2.3) represents the stem cell process through type (iii) cell division; the first term of the right-hand side of (2.4) represents the TD cell production process through type (ii) cell division; the second term of the right-hand side of (2.4) represents the TD cell production process through type (iii) cell division. The transition kernel r(a, a′) describes how daughter stem cells with damage a come from the mother cells with damage a′ after the i-th type of division; the transition kernel r(a, a′), describes how daughter TD cells with damage a come from the mother cells with damage a′ in the i-th type of division. The transition kernels satisfy the following conservation conditions: Since type (i) and type (ii) divisions induce two daughter stem cells and two daughter TD cells, respectively, the integrals of r1,(a, a′) and r2,(a, a′) equal 2; since type (iii) division induces one daughter stem cell and one daughter TD cell, the integrals of r3,(a, a′) and r3,(a, a′) both equal one. The last three integrals are based on the assumption that the amount of damage is conserved during cell division. No-flux conditions on the boundary a = 0 are imposed to ensure conservation of population in the direction of a and For TD cells, we assume that the population density of TD is zero when p is very large The populations of stem and TD cells at time t are given by the integrals

Based on this model, we will study the population dynamics under two scenarios:

For the simplest case where no feedbacks are involved in regulating the stem cell division and differentiation, the dynamics of stem cell population mainly depend on stem cell damage segregation rules. Different segregation rules may result in exploded, stabilized or extinct stem cell pool.

Feedbacks from TD cells and stem cells are introduced and population evolutions are simulated to study the effect of different damage segregation rules.

Model without feedback regulations

In this section, we start with the simple model where no feedbacks are included, and we additionally make the following assumptions to simplify the model (figure 2b): Under these assumptions, the transition kernels in the boundary conditions are Dirac delta functions and the boundary conditions in (2.3) and (2.4) become and Now, we study the role of the damage segregation rules in the long-term behaviour of the stem cell population. Since the segregation rule is fixed, after sufficiently long time, the cellular damage at the end of cell cycle, temporarily assuming no death, converges to a limit damage band where ω1 and ω2 are the minimum and the maximum of α1, α2, γ1, respectively. The derivation of damage limit band can be found in appendix A 1.

Cell cycle progression velocities V = v and U = u are constants, so are the damage accumulation rates V = v and U = u.

The portions of three types of divisions δ are constants.

When stem cells divide, damage is partitioned into portions α1 and α2 between stem cells, β1 and β2 between TD cells, or γ1 and γ2 between stem and TD cells, where α1 + α2 = 1, β1 + β2 = 1 and γ1 + γ2 = 1. Without loss of generality, we assume α1 ≤ α2 and β1 ≤ β2.

The population dynamics turns out to depend on the proportions of three division types of stem cells and the position of the lethal threshold a*/v with respect to the limit damage band [p*/v(1 − ω1), p*/v(1 − ω2)]. Before proceeding, we define a key lumped parameter for studying the long-term population behaviour Here, the parameter f is called the self-renewal fraction. Using the limit damage band and the self-renewal fraction, we can find some conditions for different long-term behaviours.

Proposition 3.1.

Stem cells become extinct, if the renewal fraction f < 1/2 or the limit damage band is completely above the lethal threshold, i.e. a*/v < p*/v(1 − ω1).

Proposition 3.2.

Assume that the limit damage band lies completely below the lethal threshold, i.e. a*/v > p*/v(1 − ω2) and Then stem cell population blows up if the renewal fraction f > 1/2, or is eventually conserved if f = 1/2.

The proofs of propositions 3.1 and 3.2 can be found in appendix A 2. In the situations shown in propositions 3.1 and 3.2, either damage accumulation does not affect cells or no cell can survive after sufficiently long time. Next, we consider the more intriguing intermediate situations as follows, based on the following assumptions on the renewal fraction and the lethal threshold a*/v lies within the limit damage band

Under the above assumptions in (3.3), there are many situations of damage segregation rules, and more importantly not all of the situations are biologically meaningful. Here, we will focus on one situation: the damage retention in stem cells through asymmetric division is smaller than the damage segregation portions in symmetric stem cell renewal, i.e., This setting is based on the biological observation that asymmetric division is a favourable mechanism for stem cell lineage to remove the damage. For avoiding the case that the lethal threshold a*/v lies below p*/v(1 − α1), i.e. all asymmetrically renewing stem cells are destined to die, we further assume that By the assumptions (3.3)–(3.5), we obtain that Under the assumption (3.6), the stem cell population may approach zero or blow up in the long-term behaviour. The following proposition provides a condition to guarantee extinction of stem cells, as well as a condition for population blow-up, with the proof given in appendix A 2.

Proposition 3.3.

Assume that (3.6) holds.

Define If there exists a combination of δ such that , then the stem cell population is eventually zero.

If δ2 = 0 and then the stem cell population goes to infinity, or is bounded below to avoid stem cell extinction.

From proposition 3.3, we can obtain several necessary conditions to maintain the stem cell population. Since the term is increasing with n and goes to infinity when n tends to infinity, n has to be small enough to satisfy the condition in proposition 3.3. From the definition of n, we observe that if the value of n will increase and then cannot be satisfied in most of the combinations of δ1, δ2 and δ3. This result provides some conditions of the parameters to prevent stem cell extinction.

the difference between a*/v and p*/v(1 − α2) becomes small, or

the difference between 1/(1 − α2) and 1/(1 − γ1) becomes large, or

the right-hand side becomes large for each m,

Let us discuss the above three possibilities. To reduce the difference between a*/v and p*/v(1 − α2), the ratio of cell cycle progression speed v to damage accumulation speed v should be large enough and close to p*/a*(1 − α2); to increase the difference between 1/(1 − α2) and 1/(1 − γ1), damage retention in asymmetric division should decrease, i.e. γ1 should be small; to increase the right-hand side for each m, damage distribution in self-renewal should become more symmetric, i.e. α1 should increase to close to 0.5. In conclusion, when v/v increases, γ1 decreases or α1 increases, it may provide a better condition to maintain the stem cell population. These results are supported by the numerical simulations shown in figure 3.

Figure 3.

The long-term stem cell population dynamics with different combinations of the division probabilities δ1 and δ3. The yellow regions represent the combinations of (δ1, δ3) whose corresponding population blow up or is eventually conserved, while the blue regions represent the situation where population goes to zero. The parameter values are set to be α1 = 0.3, γ1 = 0.1, p* = 1, a* = 1, v = 0.2 and v = 0.1, if not mentioned in the subfigures.

In figure 3, we study the long-term population dynamics with different combinations of the division probabilities δ1 and δ3. We uniformly generate 600 pairs of δ1 and δ3 with 0 < δ1 + δ3 ≤ 1 and 2δ1 + δ3 ≥ 1. We skip the region 2δ1 + δ3 < 1 as it implies stem cell extinction whatever other parameters are (by proposition 3.1). For each pair of δ1 and δ3, the system is solved by the numerical method described in appendix A 5. The blue regions in figure 3 represent the cases of stem cell extinction and the yellow regions represents the non-extinction cases (the stem cell population may tend to a constant non-zero value, approach infinity, or keep oscillating). When v/v increases (in figure 3a, v/v = 1.67; in figure 3b, v/v = 2), γ1 decreases (in figure 3c, γ1 = 0.3; in figure 3d, γ1 = 0.05) or α1 increases (in figure 3e, α1 = 0.1; in figure 3f, α1 = 0.45), the blue regions become smaller (the yellow regions become larger), which implies that there are more combinations of parameters (δ1, δ3) allowing stem cell survival.

For most of the combinations of parameters, the stem cell populations in the models without feedback regulation blow up to infinity or diminish to zero. Although the no-regulation assumption is not realistic, the simple model not only provides us a guidance on the selection of parameters used in the model with feedbacks but also reveals that population dynamics are results of all factors: the cell cycle progression of stem cell, damage accumulation, fractions of divisions and damage segregation rules. In the next section, we will consider the combinations of parameters that guarantee exponential growth in the stem cell population and study feedback regulations that could lead to non-zero steady-state population.

Model with feedback regulations

Biological evidence shows that some mammalian stem cells can switch between symmetric and asymmetric divisions in response to external and internal regulations [1,3]. For example, both epidermal [43] and neural [44] progenitors change from the primarily symmetric division that expands the stem-cell pool during embryonic development to primarily asymmetric in mid to late gestation. It is also observed that nervous [45] and haematopoietic [46] stem cells in adults can divide symmetrically to replace lost cells through injury, although they divide asymmetrically under steady-state conditions. In particular, two types of feedbacks have been proposed [34,47,48] (also see figure 4a): long-range feedbacks responding to the population of TD cells, and short-range feedbacks acting in an autocrine fashion from stem cells. These two types of feedbacks regulate stem cell population through controlling two types of parameters:

Figure 4.

Feedbacks in stem cell lineage. (a) Two types of feedback occur: the long-range feedback responding to the population of TD cells, and short-range feedback acting in an autocrine fashion in stem cells. (b) Two-compartment model with feedbacks from TD cells. (c) Two-compartment model with feedbacks from stem cells and TD cells.

the speed of stem cell division, V (or stem cell cycle progression),

the probabilities of three types of cell division, δ (differentiation versus renewal).

However, the underlying mechanisms are not fully understood. In addition, the effect of damage segregation rules on cell population is completely unknown. In the following, we use mathematical models with different feedbacks and damage segregation rules to study which types of feedback and damage segregation mechanisms are more plausible given the experimental observations.

For simplicity, we assume that damage segregation rules are fixed, i.e. α, β, γ are constants. Based on the results from the simple model in §3, to ensure that the stem cell population will blow up to infinity without any feedback regulation, additional assumptions are made:

Self-renewal fraction .

Damage accumulation speed V = v is slower than cell cycle progression of stem cells.

Feedback only from TD cells

The long-range feedback, which acts through signals sent by differentiated cells and inhibits stem cell division and self-renewal, has been biologically observed in numerous tissues, including muscle [49], bone [50], skin [51], nervous system [52] and haematopoietic systems [53]. Despite this wealth of data, there is less understanding of the exact mechanisms of feedback regulation. A significant number of mathematical models have been developed to explore the possible mechanisms behind feedback regulations [26,27,32,34-40,54,55]. The dynamics of signalling molecule s(t) can be described by a simple ODE as follows: where α is a constant synthesis rate, and the degradation is proportional to the level of s and affected by the cell population P of stem or TD cells. Since the dynamics of the signalling molecules take place on a faster time scale than the processes of cell proliferation and differentiation, we assume that the feedback signal can be approximated by a quasi-steady-state solution [25,27,32,34-40,54,55]. By properly rescaling s, the quasi-steady state of s is given by a Hill function where k is a regulation constant to account for the sensitivity to the cell population and m is the Hill exponent. The function in (4.2) reflects the assumption that the signal intensity achieves its maximum in the absence of cells and decreases asymptotically to zero if the number of cells increases. Note that Hill functions are widely used to describe ligand–receptor interactions, which also makes them natural choices to model the actions of secreted feedback factors [37].

According to biological evidence, we may model the feedback from TD cells in the following ways (figure 4b): when stem cell or TD cell population is small (in the early stage of development or with drastic loss due to injury), symmetric division predominates over asymmetric division; during the stable stage (mid and late gestation or tissue homeostasis), stem cells switch to asymmetric division. Such feedbacks may be added to the model by modifying division fractions δ as follows: where i ∈ {1, 2}, are basal division fractions, are regulation constants, m is the Hill exponent and P(t) is the TD cell population. The remaining division fraction δ3 is defined as 1 − δ1 − δ2.

Recent studies on cell cycles of neural stem cells have shown evidence that TD cells may be a source of signalling molecules that inhibits cell cycle progression of stem cells [56]. Therefore, besides the above regulation on cell fate decisions, negative feedback from TD cells can also be involved in stem cell proliferation V (cell cycle progression speed), i.e. excessive number of TD cell may slow down the proliferation of stem cell, and hence reduce the population of both stem cells and TD cells. Thus, V can be modified as where v is the initial cell cycle progression speed and is a regulation constant. For other velocities, we set V = v, U = u and U = u.

Among all parameters, we are most interested in the effect of segregation rules α, β, γ on the population dynamics. A set of reasonable estimations of the parameters other than α, β, γ are chosen based on appropriate biological and mathematical assumptions, and are shown in table 1. The detailed reasoning of the selection of parameters is provided in appendix A.3.1. In brief, we assume that initially, the stem cells are expanding, the cell cycle progression is fast, and the symmetric renewal predominates in three types of division. In the rest of this subsection, we will focus on the discussion of the effect of different combinations of segregation rules α, β, γ.

Table 1.

The choices of parameters, estimated based on biological evidence and previous modelling works, with details in appendix A.3.1.

parameters	meaning	value
p*	cell cycle threshold	1
a*	damage level threshold for stem cell	1
ac	damage level threshold for TD cell	1
vp	maximum cell cycle progression speed of stem cells	0.2
va	constant damage accumulation speed of stem cells	0.06
up	constant cell cycle progression speed of TD cells	0.02
ua	constant damage accumulation speed of TD cells	0.02
δ10	maximum fraction of symmetric renewal of stem cells	0.6
δ20	maximum fraction of symmetric differentiation of stem cells	0.3
δ30	maximum fraction of asymmetric division of stem cells	0.1
k1T	regulation constant	10−8
k2T	regulation constant	0.5 × 10−8
kvT	regulation constant	0.5 × 10−8
mT	Hill exponent	2

In the simulations, we consider the following situations: α1 varies from 0.1 to 0.5, indicating that the symmetry in damage segregation increases between the two stem daughter cells; β1 varies from 0.1 to 0.5, indicating that the symmetry in damage segregation increases between TD cells; γ1 varies from 0.1 to 0.9, indicating that the damage retention in stem cells increases in asymmetric division. Under different segregation rules, the following aspects are studied:

the population size of TD cells and the population ratio of TD cells to stem cells at steady state;

the death rate of stem cells and the fraction of three types of division at steady state; and

the dynamics of population evolution and the damage distribution in stem cell population at steady state.

Population size of TD cells and population ratio of TD cells to stem cells

Figure 5 shows that the damage distribution between TD cells in symmetric differentiation (β1 and β2) does not affect population size at steady state. When the parameter β1, which determines how damage is distributed between two TD cells in the symmetric differentiation of stem cells, varies from 0.1 to 0.5, neither stem cell population nor TD cell population has a big change. This result is biologically meaningful, since the replenishment of TD cells is efficient compared to the loss of TD cells due to damage accumulation. In the following part, we will focus on the case with β1 = 0.5.

Figure 5.

Model with feedback only from TD cells: TD cell population and population ratio at steady state for different combinations of α1, β1 and γ1. The parameters used in these simulations are shown in table 1.

Figure 5 also shows that as damage segregation α between stem cells becomes more symmetric, TD cell populations increase, while the ratios P/P are almost constant with a slight decrease. For fixed α1, as damage retention γ1 in asymmetric division increases, both TD cell populations and the population ratios decrease. Interestingly, when α1 = 0.5, although the populations of TD cells are very close for the cases γ1 = 0.1, 0.3 and 0.5, the population ratios differ dramatically. Compared to γ1 = 0.1, one needs a larger stem cell pool to generate a similar number of TD cells when damage retention is relative higher, i.e. γ1 = 0.5. Moreover, it would be very difficult to find explicit formulae for the populations, since the population sizes at steady state are affected by all parameters as observed in the simulations. However, when steady state exists, we can give an upper-bound estimate of the populations of stem cells and TD cells, with details given in appendix A 4.

Death rate of stem cells and fraction of three types of division

Other than studying the population size, we also compare the death rate of stem cells and the fraction of three types of division at steady state. In our model, the death rates of stem cell (r) and TD cell (r) are measured in the following way: where the numerators are population out-fluxes due to death and the denominators are the total populations. Figure 6 shows that more equal distribution of damage in symmetric renewal and less damage retention in asymmetric division result in a lower death rate of stem cells and less symmetric division at steady state. According to our simulations, the segregation rules do not have much impact on the death rate of TD cells, which ranges from 2.130 × 10−4 to 3.345 × 10−4. However, the death rate of stem cells changes dramatically as we vary the segregation rules. From figure 6a, we can see that the death rate of stem cells increases as α1 decreases or γ1 increases. The lowest death rate of stem cells is attained when damage is equally distributed between stem cells in symmetric renewal and damage retention is minimal in asymmetric division. To maintain the steady stem cell population and to replenish the short-lived TD cells, a higher death rate should be associated with a fast turnover of stem cells. Indeed, when we examine the fractions of three types of division, we find that the higher death rate of stem cells is always associated with the higher fraction of symmetric renewal (figure 6b). These results suggest that to maintain stabilized populations, tissues may have different mechanisms that involve different damage segregation rules and division rules.

Figure 6.

Results of the death rates of stem cell and the fractions of symmetric division in steady state for the model with feedbacks from TD cells under different damage segregation rules. (a) Death rate of stem cell at steady state. The definition of death rate can be found in equation (4.5). (b) Symmetric renewal fraction δ1 at steady state. The parameters used in these simulations are shown in table 1 and β1 = 0.5.

Dynamics of population evolution and damage distribution of stem cell population

The comparison of population dynamics among different combinations of α1 and γ1 (figure 7a) shows that higher damage retention results in oscillations in population evolution. Oscillations start to appear as we increase the damage retention γ1 in stem cells during asymmetric division (γ1 = 0.7 in figure 7a). Also, the oscillations become more severe when the damage distribution among stem cells in symmetric renewal becomes more symmetric (α1 = 0.5 and γ1 = 0.7 in figure 7a). Moreover, population overshoots before steady state are observed in all cases with α1 ≠ 0.5.

Figure 7.

Dynamics of population evolution and damage distribution of stem cell population at steady state for the model with feedbacks from TD cells under different damage segregation rules. (a) Sample population dynamics under different combinations of α1 and γ1. The red curves stand for TD cell population and the blue curves are for stem cell population. (b) Population density and damage distribution for stem cells. In these three cases, we set γ1 = 0.3. The other parameters used in these simulations are listed in table 1 and β1 = 0.5.

The results above suggest that more equal distribution of damage in symmetric renewal and less damage retention in asymmetric division are favourable segregation rules in four aspects: populations converge to steady states faster without oscillations or severe population overshoot; the death rate of stem cells is much smaller than that of TD cells; asymmetric division predominates in three types of division at steady state; not only the size of TD cell population is larger, the population ratio P/P is also larger.

However, the asymmetric damage segregation between stem cells in self-renewal may have benefits that cannot be shown from the analysis of total populations as integrated results. Our PDE modelling approach allows us to investigate more details of population density and damage distribution of stem cells. Figure 7b gives the stem cell population densities corresponding to different segregation rules. As we studied in §3, damage segregation rules determine the limit damage band after a sufficiently long time. In particular, the less symmetric the segregation rules are, the wider the limit damage bands will be. Although asymmetric segregation rules may result in a greater percentage of stem cells inheriting more damage and accelerated death, it also leads to a higher percentage of healthier stem cells with less damage. This can be observed in the samples shown in figure 7b. When the symmetry of damage segregation in stem cell population increases, the damage distributions at the end of cell cycle for stem cells become more concentrated, as such symmetry increases. The concentrated damage distribution could be a disadvantage, since it is less resistant to external perturbations. In this sense, we think that some stem cells may sacrifice the lower death rate for more diversified damage distribution, in order to protect the stem cell pool from a possible unfavourable situation that may lead to a sudden increase in damage.

Feedback from both TD and stem cells

The maintenance of the stem cell pool is not only affected by signalling from mature cells, but also by the stem cell pool itself. Stem cells reside in the so-called stem cell niche, where both cellular and non-cellular components interact in order to control stem cell proliferation and differentiation [57-60]. The regulations from the stem cell population include two aspects: a negative feedback control of stem cell proliferation as a result of contact inhibition; and a self-renewal inhibition factor secreted by the stem cell niche, in which case the more stem cells there are, the less likely stem cells will divide symmetrically. Experiments also show that cells inheriting the majority of damage protein aggregates during asymmetric division have an increased cell cycle length and tendency to differentiate [13,14,18].

In addition to feedback regulations from differentiated cells, we are also interested in how the dynamics of populations would change if regulations from stem cells are included (figure 4c). Many modelling works (e.g. [34,40]) only consider the feedbacks from stem cell populations, due to the simplicity of their models. However, we include damage as a state variable in our model and could simulate feedbacks from both the stem cell population and the stem cell cellular damage level.

Thus, in our model, we assume that the excessive stem cell population and elevated stem cell cellular damage slow down stem cell proliferation and modify V in (4.4) as where is a decreasing function of a with sigmoid shape, in which , , a1 and b1 are constants.

We also assume that the excessive stem cell population inhibits stem cell symmetric division, and that the elevated stem cell cellular damage promotes stem cell differentiation via decreasing stem cell self-renewal. As a result, we modify δ1 and δ2 in (4.3) in the same way as above and where is another function of a sigmoid shape with constants , , a2 and b2.

As a continuation of §4.1, our analysis still focuses on the discussion of segregation rules. In addition to the parameters in table 1, the values of more parameters are given in table 2, for which the detailed reasoning can be found in appendix A.3.2.

Table 2.

The choices of additional parameters, estimated based on biological evidence and previous modelling works, with details in appendix A.3.2.

parameters	meaning	value
k1S	regulation constant	10−7
k2S	regulation constant	0.25 × 10−7
kvS	regulation constant	0.5 × 10−7
mS	Hill exponent	2
a1, a2	constant for sigmoid regulation	1.1
b1, b2	constant for sigmoid regulation	− 0.7
k1a, k2a	stiffness parameter	20
a10, a20	damage threshold for sigmoid regulation	0.75

To examine the change in population dynamics after introducing feedbacks from stem cells, the same combinations of segregation rules as in §4.1 are considered: α1 varies from 0.1 to 0.5, β1 varies from 0.1 to 0.5, and γ1 varies from 0.1 to 0.9. Here, we study the effect of the feedbacks from stem cells on the following two aspects:

the population size of TD cells and the population ratio of TD cells to stem cells at steady state; and

the dynamics of population evolution and the damage distribution of stem cell population at steady state.

Population size of TD cells and the population ratio of TD cells to stem cells

Similar to the model in §4.1, damage distribution β between TD cells does not affect the results much. Since the regulation is stronger when we consider feedbacks from both stem cells and TD cells, the stabilized population of TD cells is smaller than that in §4.1 under the same parameters. It may not be meaningful to directly compare the size of populations in different models, since we do not have real experimental data. However, similar trends are observed in the model with the feedbacks from stem cells (figure 8): for fixed α1, as damage retention γ1 in stem cell increases, both the TD cell population and the population ratio P/P decrease; for fixed γ1, as damage segregation between stem cells becomes more symmetric, the TD cell population increases, but the population ratio is almost unchanged. It is worth mentioning that the population ratio increases a little bit for each damage segregation configuration, after adding feedbacks from stem cells. This means that the regulation is more efficient, since one needs a smaller stem cell pool to generate the same number of TD cells.

Figure 8.

TD cell population and population ratio in steady state for different combinations of α1, β1 and γ1 for the model with feedbacks from stem cells and TD cells. The parameters used in these simulations are shown in tables 1 and 2, and β1 = 0.5.

A notable consequence of adding feedbacks from stem cells is the reduction of the oscillations in population dynamics. Comparing figure 7a with figure 9a, after considering the feedbacks from stem cells, the oscillations in population dynamics disappear even when γ1 is large, and that the population overshoot problem resolves when α1 is small.

Figure 9.

Dynamics of population evolution and damage distribution of stem cell population at steady state for the model with feedbacks from stem cells and TD cells under different segregation rules. (a) Sample population dynamics under different combinations of α1 and γ1. The red curves stand for TD cell population and the blue curves are for stem cell population. (b) Population density and damage distribution for stem cells. In these three cases, we set γ1 = 0.3. The other parameters used in these simulations are listed in tables 1 and 2, and β1 = 0.5.

Since the feedbacks from stem cells described by (4.7) and (4.6) also depend on stem cell cellular damage level, we want to examine the effect of such regulations on damage distribution at the end of cell cycle by comparing the models with/without the feedbacks from stem cells (figures 7b and 9b). Similar to §4.1, the damage distributions at the end of cell cycle become more concentrated, as the degree of symmetry in damage segregation increases. However, compared to the results in §4.1 (figure 7b), the model with the feedback from stem cells (figure 9b) provides better fitness, since under the same segregation rules, there are more cells with less damage at the end of cell cycle at steady state. This effect becomes most significant when the segregation rule is symmetric. This result suggests that slowing down the cell cycle progression of stem cells with a high level of damage and promoting such stem cells to differentiate can indeed improve the overall health of the stem cell population.

Conclusion

In this research, a novel model was developed to integrate stem cell proliferation and differentiation with damage accumulation in stem cell ageing process. A system of two structured PDEs are used to model stem cells (including all multiple progenitors) and TD cells. In our model, cell cycle progression and damage accumulation are continuous while division is discrete, and the damage segregation takes place at each division. Ageing effect is included through the inhibition from damage accumulation on stem cell proliferation and self-renewal. Regulations from TD cell and stem cell populations are incorporated through negative feedbacks on stem cell proliferation and symmetric division.

Our results showed that more equal distribution of damage between stem cells in symmetric renewal and less damage retention in stem cell in asymmetric division are still favourable segregation rules resulting in higher population size and greater population ratio; asymmetric damage segregation in stem cells leads to less concentrated damage distribution in stem cells population, which may be more robust to sudden increase in damage. These two results provide an answer for why some types of stem cells can switch between symmetric and asymmetric divisions in response to external and internal regulations [1,3].

Other than the feedbacks solely from TD cells, adding feedback regulations from stem cells can reduce oscillations and population overshoot in some situations with unfavourable damage segregation rules, say, in the situation where α1 is small and γ1 is large. Moreover, the stem cell feedback regulation can slow down the proliferation of stem cells with high level of damage to increase their tendency to differentiate. Overall, the feedback regulation system can improve the fitness of stem cells by increasing the percentage of stem cells with low level of damage. This result provides a new aspect to understand the role of the self-regulations in stem cell lineage [32,47,52,61].

In this study, our model provides a framework to study how the damage segregation rules at division affect the population dynamics. Moreover, our model is an excellent tool to understand the data which may result from observing the dynamics of regeneration with damage distribution in stem cell population after a knock-out process. In the future work, we can extend our study to the system with more than two stages, including progenitor cells [40,61] to get a better understanding for the system of stem cell lineage.