# Modelling the regulation of telomere length: the effects of telomerase and G-quadruplex stabilising drugs

- 2.1k Downloads
- 4 Citations

## Abstract

Telomeres are guanine-rich sequences at the end of chromosomes which shorten during each replication event and trigger cell cycle arrest and/or controlled death (apoptosis) when reaching a threshold length. The enzyme telomerase replenishes the ends of telomeres and thus prolongs the life span of cells, but also causes cellular immortalisation in human cancer. G-quadruplex (G4) stabilising drugs are a potential anticancer treatment which work by changing the molecular structure of telomeres to inhibit the activity of telomerase. We investigate the dynamics of telomere length in different conformational states, namely t-loops, G-quadruplex structures and those being elongated by telomerase. By formulating deterministic differential equation models we study the effects of various levels of both telomerase and concentrations of a G4-stabilising drug on the distribution of telomere lengths, and analyse how these effects evolve over large numbers of cell generations. As well as calculating numerical solutions, we use quasicontinuum methods to approximate the behaviour of the system over time, and predict the shape of the telomere length distribution. We find those telomerase and G4-concentrations where telomere length maintenance is successfully regulated. Excessively high levels of telomerase lead to continuous telomere lengthening, whereas large concentrations of the drug lead to progressive telomere erosion. Furthermore, our models predict a positively skewed distribution of telomere lengths, that is, telomeres accumulate over lengths shorter than the mean telomere length at equilibrium. Our model results for telomere length distributions of telomerase-positive cells in drug-free assays are in good agreement with the limited amount of experimental data available.

## Keywords

Mathematical model Telomeres Telomerase G-quadruplex T-loop RHPS4## Mathematics Subject Classification (2000)

92B05 37N25 34K07## 1 Introduction

Most normal cells cycle and divide a limited number of times, a discovery first made by Hayflick (1965), who grew normal human fibroblasts in culture and observed 60–80 population doublings before apoptosis. The limited number of divisions is referred to as the Hayflick limit (Hayflick 1979). On reaching the Hayflick limit, cells cease proliferation permanently and enter a state called replicative senescence in which they are still alive and functional but do not divide. The erosion of protective structures located at the ends of chromosomes, known as telomeres, during each replication is responsible for the limited lifespan of a cell, marking the ageing of cells and eventually triggering irreversible cell cycle exit and cell death when telomeres become critically short. Unprotected telomeres are detected by the DNA repair machinery and trigger a DNA damage response characterized by the formation of telomere dysfunction-induced foci at telomeres (Takai et al. 2003). Activation of the tumour suppressor gene p53 then triggers cell cycle arrest leading to apoptotic cell death or replicative senescence through the induction of p21. Alternatively, expression of p16 can induce senescence, although this is less well understood (Deng et al. 2008).

There have been several approaches based on mathematical modelling to understand telomere length dynamics of somatic and cancerous cells and how they contribute to chromosome stability and the initiation of senescence or apoptosis. The first papers on quantitative modelling of telomere dynamics describe the process of telomere shortening by simple deterministic (Levy et al. 1992) and probabilistic (Arino et al. 1995; Olofsson and Kimmel 1999) models, which only take account of losses due to a special phenomenon called the end-replication problem causing telomeres to shorten progressively with each cycle (see Sect. 2.1). Rubelj and Vondracek (1999) extended these previously established models of telomere shortening (‘gradual telomere shortening’) by the introduction of the possibility of ‘abrupt telomere shortening’ caused by DNA repair mechanisms due to accumulation of DNA damage, producing sudden, stochastic changes in telomere length, and becoming more frequent as telomere shortening advances. Similarly, Sozou and Kirkwood (2001) and Proctor and Kirkwood (2002) included environment-dependent components effecting telomere shortening, where oxidative stress in form of endogenous reactive oxygen species produced by mutant mitochondria is assumed to be the cause of substantial telomere loss.

Telomere length is maintained in most cancer cells by the enzyme telomerase capable of adding new telomeric DNA onto chromosome ends. Mechanisms contributing to telomere length equilibrium have been considered by Blagoev (2009), who proposed a model in which telomere extension by telomerase occurs more frequently at short telomeres than at long telomeres. A logistic function describes the probability of the occurrence of an extendible state of telomeres, opposed to a capped state, which was inspired by data from experiments in Teixeira et al. (2004) on telomere elongation in yeast cells. Other work on the dynamics of telomere length by Qi (2011) compares the effects of normal ageing, accelerated ageing of patients with Werner’s syndrome and the unlimited lifespan of telomerase-positive cells, where models involve telomere-length-dependent telomere loss and gain as well as telomere-length-dependent probabilities for cell division.

Mathematical models can be a useful means for integrating different types of experimental data to predict the mechanism of action of compounds (Wolkenhauer et al. 2009). In this paper, we aim to develop and analyse differential equation models of the effects of G-quadruplex ligands on telomere length regulation in telomerase-positive cancer cells, where we consider different time-scales, from one cell cycle to a large number of cell replications.

In Sect. 2 we explain the background biology. In Sect. 3 we develop and analyse a model describing the telomere length dynamics for telomerase-positive cells during one cell cycle and investigate how they respond to treatment with the G4-stabilising drug RHPS4. This model can also be used to describe telomere length dynamics over a small number of cell generations. In Sect. 4 we investigate for what levels of telomerase and RHPS4 the telomere length distributions stabilise over a large number of cell generations, and we predict the corresponding steady-state length distributions. Section 5 summarises the results and contains a concluding discussion.

## 2 Background biology

### 2.1 Telomeres

Mammalian telomeres are specialised nucleotide sequences that protect the end of chromosomes. Telomeres contain short tandem repeats—sequences of basepairs that are repeated numerous times. The sequence of the repeat unit is TTAGGG in mammalian cells and human telomeres contain around 10–15 kilobasepairs per chromosome end at birth. Telomeres of most somatic cells typically shorten in the synthesis (S) phase of the cell cycle during each replication due to the “end-replication problem”, that is, the inability to fully replicate the terminating DNA sequences (Levy et al. 1992). One shorter and one longer telomeric end are generated during replication, where the longer strand is generally rich in guanosine (G) and devoid of cytosine (C). The single-stranded protrusion of the longer strand is referred to as the *G-overhang*, which varies between 50–500 nucleotides in mammalian cells and is considerably shorter in most other eukaryotes. About 3 basepairs are lost from one DNA strand on each round of cell division due to the end-replication problem. However, human and mouse telomeres shorten by about 50–200 basepairs during each replication at both telomeric ends and the average telomere length in human cells decreases by roughly 2–4 kilobases during their lifetime. A more likely explanation of the intensive and double-sided telomere shortening is postreplicative processing by a nuclease (Palm and de Lange 2008; Makarov et al. 1997). Also, oxidative stress in the from of reactive oxygen species, which accumulates over the lifespan of a cell, causes accidental lesion in the DNA and is assumed to be the cause of substantial telomere loss (Richter and von Zglinicki 2007; von Zglinicki et al. 2005).

This progressive telomere erosion has been designated as the reason why normal mammalian somatic cells only divide a finite number of times in vitro, before undergoing permanent growth arrest. It is, however, not yet clear whether it is the average telomere length (Martens et al. 2000) or the length of the shortest telomere (Hemann et al. 2001) that is critical for the onset of cell cycle arrest in a cell. Looking at the distribution of telomere lengths can help determine whether a percentage of short telomeres or the mean telomere length is a trigger for the onset of replicative senescence. Telomeres that become critically short, and thus unprotected, are recognised by the cell as DNA double-strand breaks, inducing senescence and apoptosis, which is dependent on the expression of the oncosuppressor gene p53. Those cells which pass this point in cell replication through inactivation of p53 continue dividing, lose all their protective telomeric DNA and enter a state called crisis, causing end-to-end joining of chromosomes and other forms of enormous genomic instability, carcinogenesis and eventually cell death (see Greenberg 2005 for a review).

### 2.2 Telomerase

The enzyme telomerase can antagonise telomere shortening by association with the telomeric end, where it progressively synthesises telomeric repeat sequences at the single-stranded overhang of the telomere, thus inhibiting telomere uncapping which occurs when telomeres become too short. It has been suggested that human telomerase acts rapidly on most (\(\sim \)70–100 %) telomeres following replication (Wu and de Lange 2009) and requires the telomeric G-overhang for telomere elongation. In HeLa cells that were synchronised at the \({\mathrm{{G}}}_1/\mathrm{S}\) transition, the total overhang length gradually increased over the next 6–7.5 h (Zhao et al. 2009; Dai et al. 2010), indicating a phase of increased telomerase activity, and that telomeric sequences are replenished each time a cell divides. Telomerase consists of TERC (Telomerase RNA Component) with a template region for copying telomeric repeat sequences, and the catalytic protein TERT (Telomerase Reverse Transcriptase), which catalyses the G-rich extension of linear chromosomes. TERC is generally highly expressed in all cells, and independently of telomerase activity, whereas the concentration of TERT is estimated at less than 50 copies per cell. In normal somatic cells the catalytic subunit TERT is repressed, but it is upregulated in immortal cells, suggesting it is the major determinant for telomerase activity.

The majority of cancer cells express telomerase continually; they possess altered telomeres and have the potential for unlimited replication. Telomerase was found to be present in 85–90 % of cancerous cells and it is believed that its specific role is to immortalise these cells (Kim et al. 1994). Most of the remaining 10–15 % of cancer cells, in contrast, can maintain their telomeres by a telomerase-independent pathway called alternative lengthening of telomeres (ALT); see Cesare and Reddel (2010).

### 2.3 Telomere structure

The extendible, open form of telomeres is presumably the most likely form occurring during telomere synthesis. Telomeres, however, can loop back and tuck their single-stranded end into the duplex DNA of telomeric sequences to form a *t-loop* (reviewed by Blackburn 2001; de Lange 2004, 2009), where a specific protein complex named shelterin (or telosome) (de Lange 2005; Palm and de Lange 2008) is involved in protecting chromosome ends from DNA degradation and DNA damage responses. The t-loop might dissolve during DNA replication; however, it is not yet known whether t-loops switch into an open state during the S phase or persist throughout the cell cycle. In our model (Sect. 3) we will assume that t-loops also function as telomerase inhibitors, as they hide the telomeric G-rich end from access by telomerase, and structural rearrangements between t-loops and the open form of telomeres allow telomerase to establish telomere length homeostasis.

Alternatively, telomeric ends can spontaneously fold into guanine-rich structures called G-quadruplexes (G4), discovered by Henderson et al. (1987), which are supported by monovalent cations such as potassium (\({\mathrm{K}}^+\)) in the nucleus. G4 structures form in vivo and probably unfold during telomere replication (Schaffitzel et al. 2001). When G-quadruplexes are located at the very end of the telomeric G-overhang, which has been shown to be their preferred location (Tang et al. 2008), the enzyme telomerase is inhibited by the folding of the G-rich end (Zahler et al. 1991). For reviews of G-quadruplex structures in vitro and in vivo, see Lipps and Rhodes (2009) and Knig et al. (2010). More general reviews on telomere structures and their function in chromosome-end protection can be found in Oganesian and Karlseder (2009) and Xu (2011).

Optimal telomerase activity seems to require the unfolded single-stranded form of terminal telomere sequences. Despite the length variation of individual telomeres within a cell or an organism, average telomere length is maintained within a narrow range that is specific for each species. Studies of sperm (Allsopp et al. 1992), for instance, suggest 8–20 kb in human, where germ-line cells generally express telomerase (except from mature sperm and oocytes, see Wright et al. 1996). Furthermore, data on telomere length in the telomerase-positive HeLa and MCF-7 human breast cancer cell lines (Canela et al. 2007) indicate a coefficient of variation of, respectively, 0.23 and 0.11. The stability of telomere length suggests the thesis that telomerase-positive cells establish an equilibrium between telomere attrition and elongation. However, Cristofari and Lingner (2006) found that HeLa telomeres, which were observed over 56 population doublings (PD), elongated at a constant rate of 415–635 bp/PD upon overexpression of the main functional subunits of the enzyme telomerase, the catalytic protein TERT and the telomerase RNA component (TERC). This massive telomerase activity is referred to as super-telomerase, and long telomeres did not change into a permanently non-extendible state in super-telomerase cells.

Investigation of Martens et al. (2000) into normal human fibroblasts (telomerase-negative) having a limited lifespan showed that short telomeres increasingly accumulate in cells and the length distribution of telomeres becomes positively skewed close to senescence. Proctor and Kirkwood (2003) considered the uncapping of telomeres by the opening of t-loops as a trigger for replicative senescence to account for the experimental results found by Martens et al. (2000).

### 2.4 Quadruplex-stabilising drugs

On the other hand, stabilisation of G-quadruplexes by specific ligands can limit telomerase activity and alter telomere function in cancer cells. Anti-cancer researchers are now trying to design G-quadruplex ligands that will mimic the effect of the metal ions and inhibit telomerase, with the aim of achieving antitumour activity through the effective stabilisation of G-quadruplexes (Monchaud and Teulade-Fichou 2008). The G-quadruplex ligands RHPS4 (3,11-difluoro-6,8,13-trimethyl-8H-quino[4,3,2-kl]acridinium methosulphate) (Heald et al. 2002) together with the 3,6,9-trisubstituted acridine compound BRACO-19 (Gunaratnam et al. 2007) and telomestatin (Tauchi et al. 2006), are promising compounds among the cancer inhibitor agents and have come close to clinical testing (Bilsland et al. 2011). These compounds all inhibit telomerase activity, limiting long term proliferation of cancer cells, and directly target components of the protective cap of telomeres, leading to immediate effects on cancer cell proliferation (Neidle 2010).

Cheng et al. (2008) compared relative quadruplex and duplex binding affinity constants of different quaternary polycyclic acridinium salts and found that quaternised quino[4,3,2-kl]-acridinium salts, such as RHPS4, selectively bind and stabilise quadruplex DNA. Also, quadruplex DNA binding affinity correlated strongly with telomerase-inhibitory activity data for these G4 ligands. Cookson et al. (2005) showed a notable reduction in telomere length of MCF-7 breast cancer cells when treated with subtoxic doses (\({<}1\,\upmu \mathrm{M}\)) of RHPS4. RHPS4 treatment of human melanoma lines possessing relatively long telomeres resulted in a dose-dependent decrease in cell replication and accumulation of cells in the \(\mathrm{{S}}\)-\(\mathrm{{G}}_2/\mathrm{M}\) phase of the cell cycle (Leonetti et al. 2004). Furthermore, RHPS4 induces a marked decrease of cell growth in human cell lines such as the 21NT breast cancer cells and A431 vulval carcinoma cells after 15 days and for concentrations lower than the level of acute cytotoxicity (Gowan et al. 2001). It also rapidly induces telomere dysfunction by telomere uncapping, which leads to short-term cell death through usage of higher doses. Integrative approaches to investigate the effects of RHPS4 experimentally and by means of mathematical modelling were proposed by Johnson et al. (2011) and Hirt et al. (2012). The precise cell-cycle specific behaviour of RHPS4 and its mechanism of action in cancer cells, however, are still to be elucidated.

### 2.5 Mathematical modelling

Golubev et al. (2003) used mathematical modelling to investigate possible causes for the observed positive skewness of telomere length distribution, such as DNA damage caused by free radicals. Similarly, Grasman et al. (2011) characterise the dynamics of telomere shortening by the property that longer telomeres are more vulnerable to oxidative stress, as they are larger targets. The enzyme telomerase is also active at a low level in some somatic cells (Masutomi et al. 2003), such as human fibroblasts, and telomerase-dependent shortening, leading to positively skewed telomere length distributions, has been explained by op den Buijs et al. (2004). Another mechanism yielding skewed telomere length distributions has been described by Itzkovitz et al. (2008), who developed a population mixture model with a re-populating pool of stem cells of constant telomere length and a derived pool which experiences constant decrease in telomere length, where one daughter cell of the repopulating pool stays and one transfers to the derived pool after cell replication.

Furthermore, there are a number of models of telomere-length maintenance in telomerase-positive cells. For example, Kowald (1997) developed a mathematical model involving the concentration of a capping protein, which can bind the G-overhang once it is sufficiently long, and which inhibits telomerase and facilitates DNA replication by maintenance of the single-stranded overhang it is bound to, but is released after telomere replication. To account for the assumption that the functional state of the telomere rather than its length determines the fate of a cell, Arkus (2005) considered the binding and dissociation of a telomeric protein, TRF2, to telomeric repeat sequences, assuming that TRF2 caps telomeres and inhibits telomerase. Rodriguez-Brenes and Peskin (2010) proposed another approach of modelling telomere length maintenance processes based on the biophysics of t-loop formation, which is assumed to determine the state of a telomere and also the cell’s fate. They assumed that the longer telomeres are, the more frequently telomere ends come into close proximity of internal positions of the telomere, and hence the more likely are invasions of double-stranded DNA by the G-overhang, which can be facilitated by TRF2 and results in the formation of displacement loops together with t-loops. The dynamics of t-loop formation were described by a worm-like chain model and an algorithm was developed to sample telomeric chromatin chains (modelled as semi-flexible polymer chains) at thermodynamic equilibrium. An ODE model and a stochastic model describe shortening by the end-replication problem, C-strand processing and telomerase-induced telomere elongation. In addition to telomere length maintenance, Kowald (1997) modelled the increase in telomere length when oligonucleotides are added to cell culture. On the other hand, Sidorov et al. (2003) investigated the impact of telomerase inhibition on the growth of tumours possessing either homogeneous or heterogeneous telomere length distributions.

Telomerase-independent pathways of telomere length maintenance are considered by Olofsson and Bertuch (2010), who capture the mechanisms of survivorship of individual budding yeast cells. Another approach to understanding telomere length maintenance by means of mathematical modelling is presented by Antal et al. (2007), who superimpose stochastic telomere length variations upon the systematic decrease in telomere length in ALT cells.

## 3 Model for discrete generation numbers

*U*ncapped/

*U*nfolded) form, those

*B*ound to telomerase (T), those in a

*G*-quadruplex formation and those forming a

*C*omplex with the drug RHPS4 (R). The model is illustrated in Fig. 2. We refer to B, C and the t-loop in the model as the capped states, to G as the folded form of telomeres and to U as the state of telomeres that are neither capped nor folded, that is in the open form. After telomere duplication in the S phase telomeres are introduced into system at rate \(k_e\) in the open form (U) state and then switch between the open and G4 forms (G4 folding rate \(k_f\) and G4 unfolding rate \(k_u\)), where telomeres in the open form bind to free telomerase molecules \(T\), with association rate \(k_\mathrm{on }\) and dissociation rate \(k_\mathrm{off }\), which synthesise nucleotides with rate \(\rho \) at the telomere end. Telomeric intramolecular G4 structures do not allow telomerase association with the G-overhang, and can be stabilised by free G4 ligand molecules \(R\), where the association and dissociation rates of RHPS4 are \(k_{s}\) and \(k_{d}\), respectively. We assume that one telomerase molecule binds one telomere to elongate the telomeric end, and one G4 ligand molecule is sufficient to stabilise a G4 form. Furthermore, all kinetic rates are assumed to be constant and non-negative.

### 3.1 Steady state

Parameter estimates for the discrete-generation model

Parameter | Description | Value |
---|---|---|

\(k_\mathrm{on }\) | Telomerase binding rate | \(2.6\times 10^{-7}\, \mathrm{s}^{-1}\) |

\(k_\mathrm{off }\) | Telomerase dissociation rate | \(2.2\times 10^{-4}\, \mathrm{s}^{-1}\) |

\(\rho \) | Rate of nucleotide addition | \(6.287\times 10^{-2}\, \mathrm{{nt}}\,\mathrm{s}^{-1}\) |

\(k_{f}\) | G4 folding rate | \(1.6\times 10^{-2}\, \mathrm{s}^{-1}\) |

\(k_{u}\) | G4 unfolding rate | \(3.8\times 10^{-3}\, \mathrm{s}^{-1}\) |

\(K\) | RHPS4 equilibrium binding constant | \(1.4\times 10^{-5}\) |

\(k_{s}\) | RHPS4 binding rate | \(10^{-7}~\mathrm{s}^{-1}\) |

\(k_{d}\) | RHPS4 dissociation rate | \(7.1\times 10^{-3}~\mathrm{s}^{-1}\) |

\(k_e\) | Telomere influx rate | \(1.5\times 10^{-2}\, \mathrm{s}^{-1}\) |

\(k_r\) | Telomere loss rate | \(5\times 10^{-6}\, \mathrm{s}^{-1}\) |

\(\delta \) | Maximum formation rate of t-loops | \(5\times 10^{-5}~\mathrm{s}^{-1}\) |

\(\alpha \) | Parameter describing the value \(x\) at which \(k_c(x)=\frac{\delta }{2}\) | 1,775 bp |

\(\beta \) | Parameter describing the range of \(x\) over which the transition \(k_c(x)\) occurs | 300 bp |

\(L_0\) | Mean length of HeLa telomeres | 3,440 bp |

\(\sigma _0\) | Standard deviation of HeLa telomere length | 800 bp |

\(\mu \) | Average telomere loss per cell replication | 45 bp |

### 3.2 Gamma-distributed input

### 3.3 Gaussian-distributed input

We now return to the more general case, where \(x<\alpha \) is permitted, and impose the boundary condition \(B(-\infty )=0\) or \(B(+\infty )=0\) on (8). This boundary condition is chosen such that to avoid negativity in \(B(x)\) (imposing \(B(0)=0\) leads to a sign change of \(B(x)\) in \(x=0\) due to \(p(x)>0\) for all \(x\in \mathbb{R }\)).

*vice versa*for \(x>\alpha \). The parameter \(b_2\) determines how rapidly the telomere length distribution approaches zero for increasing telomere lengths, \(x\), with smaller values of \(b_2\) increasing the positive skewness of the distribution. Hence, decreasing the rate of t-loop formation (by lowering \(\delta \)) increases the positive skewness of \(B(x)\), and so does increasing the number \(T_0\) of telomerase molecules (or the rate, \(\rho \), of telomere elongation), for example.

The parameters \(k_\mathrm{off }\) and \(k_e\) function as scaling factors that determine the contribution of \(B(x)\) and \({\tilde{p}}(x)\) to the telomere length distribution \(U(x)\). For small \(k_\mathrm{off }\), when \(B(x)\) is increasingly positively skewed, we expect \(B(x)\) to have a larger tail than \(U(x)\) due to the respectively decreasing and increasing contributions of \(B(x)\) and \({\tilde{p}}(x)\) to \(U(x)\). We note that \(U(x)\) is independent of the rates \(k_f\) and \(k_u\) of G4 folding and unfolding, respectively, for control cells (\(R_0=0\)). The distributions of \(G\) and \(C\) are of the same shape as \(U(x)\), where larger \(R_0\) increases the ratio of telomere numbers \(C/G\). The ratio of telomere numbers \(C/U\) increases with increasing \(R_0\) in a nonlinear fashion, and tends to \(k_f/k_r\) for large \(R_0\). On the other hand, the ratio of telomere numbers \(G/U\) decreases with increasing \(R_0\) in an inverse fashion, and is equal to \(k_f/k_u\) for \(R_0=0\). For larger \(T_0\), the distributions become increasingly positively skewed.

The expression \(\widehat{L}\) is independent of \(k_e\), and in the limiting case \(R_0=0, \widehat{L}\) is also independent of \(k_f, k_u, k_s, k_d\) and \(k_r\). An increase in the parameter values \(T_0\) (or \(\rho \)) leads to a linear increase in (25), becoming nonlinear for positive values of \(R_0\). An increase in \(\sigma \) leads to an increase in \(\widehat{L}\) for \(R_0=0\).

### 3.4 Numerical solutions

We aim to show numerical results of telomere length distributions over a few generations. To derive steady state distributions at the end of each replication, we initially assume the length distribution of telomeres before the first replication event to be Gaussian \(p_0(x)\) as in (16) with \(L=L_0\) and \(\sigma =\sigma _0\). Telomeres shorten at an average amount \(\mu \) due to the end-replication problem and postreplicative processing, and consequently enter the system with length distribution \(p(x)=p_0(x+\mu )\). By numerically integrating (8) and using (23) we simulate the telomere length probability density function of telomeres leaving the system, \(p_1(x)=(k_c(x)U(x)+k_r C(x))/k_e\), at steady state and compare it to the distribution \(p_0(x)\) of telomeres entering the system before telomere shortening takes place. Assuming telomere length does not change between the \({{\mathrm{{S/G}}}}_2\) phases of subsequent cell divisions, we treat the output telomere lengths of one cycle as the input telomere length for the next generation, \(p=p_1(x+\mu )\). In particular, we use the steady state distribution of telomeres leaving the system at the end of the initial generation \(i=0\) as input (of rate \(k_e\)) into the system at the beginning of generation \(i=1\) and after telomere loss of amount \(\mu \) occurred, that is \((k_c(x+\mu )U(x+\mu )+ k_r C(x+\mu ))/k_e=p_1(x+\mu )\) replacing \(p(x)\) in Eq. (2), and derive steady state distributions \(p_i(x)\) for higher generations \(i\) in this fashion.

In general, there is only very little data on the rate parameters available in the literature. We estimate \(k_e=1.5\times 10^{-2}\, \mathrm{s}^{-1}\) according to an approximate influx of \(2\times 2\times 82\) telomeres [after telomere duplication and considering HeLa cells with a modal chromosome number of 82, as determined by Macville et al. (1999)] into the system per 6 h, that is the time during which telomere extension occurs (Zhao et al. 2009), and estimate, within biologically feasible ranges, the values \(\alpha , \beta \) and \(\delta \) for the sigmoid function \(k_c(x)\) (compare Fig. 1b) and the loss rate \(k_r\), as shown in Table 1. In order to adopt appropriate values for the remaining parameters, given in Appendix, we employ the fact that a concentration of 1 nM corresponds to approximately \(10^{-9} N_A V_n\,\mathrm{{mol}}\,\mathrm{l}^{-1} \approx 415\) molecules per HeLa nucleus, where \(N_A=6.022\times 10^{23}~\mathrm{{mol}}^{-1}\) is the Avogadro constant and \(V_n=6.9\times 10^{-13}\) l is the nuclear volume of a HeLa cell. The BioNumbers data base of Milo et al. (2010) provides us with this average volume for HeLa nuclei, which is taken from Monier et al. (2000). Using standard conversion factors, we obtain the rate parameters shown in Table 1. Note that \(k_\mathrm{on }\ll k_\mathrm{off }\) and \(k_d\gg k_s\) as required by the earlier assumption that \(T(t)\approx T_0\) and \(R(t)\approx R_0\) for the numbers of bound telomerase and RHPS4 molecules, respectively. The exact values for each of the rates of RHPS4 binding, \(k_{s}\), and RHPS4 dissociation, \(k_d\), are not relevant for the steady state of the system, so we set \(k_{s}=10^{-7}~\mathrm{s}^{-1}\) (following the size of the other molecular binding kinetic parameter, \(k_\mathrm{on }\)), for example, and derive \(k_d=k_s K^{-1}\) from the RHPS4 equilibrium binding constant, \(K\).

### 3.5 Summary

We developed and analysed a model of telomere length dynamics for a single division event, which describes the length regulation by telomerase and a G4-stabilising drug. We solved the model numerically and iterated the replication process over few generations. We now investigate the long term evolution over many generations.

## 4 Model with continuous time

We are interested in analysing the system’s steady-state telomere length distributions and predicting how telomerase and/or RHPS4 affect telomere length distributions over large numbers of cell divisions.

### 4.1 Steady states

### 4.2 Parameter range of steady state solutions

Having constructed a steady-state approximation for the solution (49), and noted at the start of Sect. 4.1 that presuming the existence of a steady state is a significant assumption, we now analyse the conditions under which such a solution may be expected to be relevant. Necessary conditions for a steady solution, \(U(x)\), of the continuum model (27)–(30) to be a distribution are that \(U\) must have a maximum at a point \({\widehat{x}}\) where \(U^{\prime }(\,{\widehat{x}}\,)=0\) and \(U(x)\ge 0\) for all \(x\in \mathbb{R }\). From (46) it follows that \(g(\,{\widehat{x}}\,)=c_T\) must be satisfied for the relevant values of \(T_0\ge 0\) and \(R_0\ge 0\) in order for \(U\) to be physical. We note that solutions \(U(x)>0\) for \(x<0\), representing a positive probability of telomeres with a negative length, should be regarded as unphysical. The probability of telomeres with length \(x<0\), however, is usually very small in our simulations and we interpret the occurrence of larger proportions of telomeres with negative length to reflect the presence of telomeres that have lost all their telomeric sequences and are no longer functional. Such telomeres would typically cause a cell to become senescent or undergo apoptosis. In the following, we aim to find conditions on \(T_0\) and \(R_0\) that must be satisfied to yield solutions with \(U\gneqq 0\), and we require \({\widehat{x}}>0\) for such solutions to be physical.

For \(\tilde{T}_\mathrm{min }<T_0<T_\mathrm{max }\) we expect steady state solutions, for \(T_0<\tilde{T}_\mathrm{min }\) the amount of telomerase is insufficient and the telomere length decays causing the cell to become senescent or undergo apoptosis. For \(T_0>T_\mathrm{max }\) telomere length grows without limit.

Mean telomere lengths, \(\widehat{L}\), with their respective standard deviations, \(\widehat{\sigma }\), of telomeres leaving the system at steady state. Data are shown for the continuous-time model and two different values of telomerase molecule numbers, \(T_0\)

\(T_0\) | RHPS4 (nM) | \(\widehat{L}\) (nt) | \(\widehat{\sigma }\) (nt) |
---|---|---|---|

25 | 0 | 3,440 | 1,517 |

50 | 2,706 | 966 | |

100 | 2,303 | 789 | |

150 | 1,975 | 768 | |

30 | 0 | 12,201 | 6,549 |

50 | 4,265 | 2,363 | |

100 | 3,017 | 1,304 | |

150 | 2,477 | 988 |

The value \(\rho =6.287\times 10^{-2}\, \mathrm{{nt}}\,\mathrm{s}^{-1}\) for the lengthening of telomeres by telomerase has been chosen in the simulations, as the telomere length distribution for telomeres leaving the system for \(T_0=25\) is in good agreement (\(\widehat{L} = 3{,}440\pm 1{,}517\) nt) with experimental data from HeLa cells (compare Figs. 1a, 7). The choice of a smaller value of \(\rho \) would yield much larger values for the number of telomerase molecules, that is \(T_0 \approx 400\) for \(\bar{\rho } = 4.5\times 10^{-3}\, \mathrm{{nt}}\,\mathrm{s}^{-1}\) (mean value from experimental measurements, see Appendix) at telomere length equilibrium. Slightly smaller values of \(\rho \) with corresponding larger values of \(T_0\) can produce telomere length distributions being similar to each other. Having more knowledge of the values \(\rho \), therefore, will help us determine the number of telomerase molecules in the nucleus more accurately. The small value of \(T_0=25\) is, however, consistent with the average value of about 20–50 telomerase molecules per HEK-293 (human embryonic kidney) nuclei measured by Cohen et al. (2007), which is the only quantitative measurement of telomerase levels in cells we are aware of in the literature.

## 5 Discussion and conclusions

We have investigated the effects of G-quadruplex interactive agents, such as RHPS4, on telomere erosion in telomerase-positive cells, and developed mathematical models of telomere length dynamics. In particular, we considered telomere length dynamics during the \(\mathrm{{S/G}}_2\) phases, when telomerase replenishes telomeric sequences of open t-loops but not the G-quadruplex structure at each cell division. We determined steady-state length distributions, over small and large numbers of cell generations, with and without treatment by RHPS4.

We derived approximate analytic expressions, and simulated numerically, steady-states of the length distributions of telomeres, as cancer cells exit the cell cycle. We analysed the effects of different levels of telomerase and various concentrations of RHPS4 on telomere length during the \(\mathrm{{S/G}}_2\) phases and considered how these effects evolve over large numbers of generations. Our models predict a positively skewed length distribution of telomeres and are in good agreement with experimental observations of HeLa cells. Moreover, our predicted value of the number of telomerase molecules in the nucleus is consistent with experimental findings in cancer cells.

In our models we assumed that t-loops and G4 structures are the key inhibitors of telomerase, and telomerase-induced telomere elongation. Constant telomere shortening due to the end-replication problem and C-strand resection, described by a constant loss term, \(\mu \), in our model for a population of continuously cycling cells, that is (3)–(5) and (26), complete the mechanisms that determine the shape of telomere length distributions. We modelled telomere elongation as a length-dependent process in that shorter telomeres are more likely to remain in an open state and be extendible by telomerase than longer telomeres, as longer telomeres have a higher tendency to coil up and form t-loops.

We further assumed that telomeres only exit the system by t-loop formation, unless G4 structures are locked by RHPS4 binding, when telomeres leave the \(\mathrm{{G}}_2\) phase of the cell cycle in the complex state and may unfold either at a later stage or at the beginning of the next cell cycle. We also supposed that the concentrations of telomerase and RHPS4 do not change over cell generations.

We have estimated most of the kinetic parameters in each system by using experimental results from the literature, and estimated the remaining parameters in order to reproduce the experimental results of HeLa cells obtained by Canela et al. (2007). Our model results agree well with the experimental telomere length distribution of HeLa cells and suggests a low concentration of about \(T_0 = 25\) telomerase molecules per nucleus. In the literature the telomerase processivity parameter \(\rho \) is given in the range \(1.2-7.7\times 10^{-3}\, \mathrm{{nt}}\,\,\mathrm{{s}}^{-1}\). We investigated the sensitivity of the results to \(\rho \) within this range. For small \(\rho \) we obtained narrower distributions of telomere lengths (and larger values of \(T_0\)) than in experimental data. We therefore fitted \(\rho \) using a larger value, \(\rho = 6.287\times 10^{-2}\, \mathrm{{nt}}\,\mathrm{s}^{-1}\), to describe the experimental data. There is almost no visible change in the telomere length distribution for smaller numbers of \(T_0\) (\(\le \!100\)) during one \(\mathrm{{S/G}}_2\) phase as simulated by the model for a population of telomeres undergoing a single replication event, described by Eqs. (2)–(5), but visible changes occur after 5–30 generations, as shown in Fig. 3 with \(T_0=25\) and \(T_0=50\).

Our simulations of conditions on \(T_0\) and \(R_0\) for physically relevant steady solutions showed that the range of values \(T_0\ge 25\) reproduces well the experimental results in the literature, and telomeres grow unboundedly in length for values of \(T_0\) larger than 30. In contrast, telomeres shorten below physical lengths (or indefinitely) if we use drug concentrations larger than \({\sim }100\) nM of RHPS4 (see Fig. 7). Hence, the equilibrium state is rather sensitive to the amount of telomerase and RHPS4 in the system: larger doses of RHPS4 lead to continuous telomere shortening, whilst telomerase overexpression, on the other hand, induces continuous telomere lengthening. The steady telomere length increase we found for large \(T_0\) is consistent with the findings of Cristofari and Lingner (2006), who observed elongation of telomeres at a constant rate in super-telomerase HeLa cells for over 50 population doublings. Hence, telomere length homeostasis cannot be established with telomerase overexpression.

Our model suggests two different effects of the treatment with RHPS4 which are dependent on the drug concentration used: low concentrations reduce telomere length, but do not impair the ability of the system to maintain an equilibrium, and high concentrations destabilise the system leading to chromosome degradation and senescence and/or cell death. Additionally, overexpression of telomerase can counteract telomere degradation; however, telomerase addition should be carefully regulated to maintain the system in equilibrium and not trigger unlimited telomere elongation.

Note that the upper limit for \(R_0\), when telomere equilibrium can still be maintained, is probably lower than we predicted. The threshold value for telomere length as determined by the Hayflick limit (Hayflick 1979), triggering senescense or apoptosis pathways in a cell terminating cell proliferation, is likely to be based on the shortest telomere in the cell, not the average length (Hemann et al. 2001; Abdallah et al. 2009). As far as we are aware, telomere length distributions for cells exposed to different drug concentrations of RHPS4 have not yet been measured experimentally. Measurements of telomere length distributions of telomerase-positive cells at different population doublings and for varying concentrations of the drug RHPS4 or different concentrations of telomerase will be useful in comparing our model predictions to experimental results.

We found that higher concentrations of telomerase can lead to ongoing telomere lengthening, which is consistent with observations from experiments with telomerase-positive cells. We derived regions of different telomerase and RHPS4 levels that provide physically plausible solutions to our model of telomere length dynamics over large numbers of generations and showed that telomerase expression must be strictly regulated for telomere length maintenance. Too high concentrations of RHPS4 can lead to progressive telomere erosion; we estimate that drug concentrations larger than \(\approx \)100 nM impair the equilibrium of the system leading to continuous telomere shortening and triggering senescence and apoptosis.

In summary, the main results of this paper are showing how telomerase acts as a regulator for telomere length, and how RHPS4 can disrupt this regulation as illustrated by Figs. 6b and 9. At small concentrations, RHPS4 has little effect, but there is a critical concentration above which telomerase is unable to maintain a steady state, and rapid shortening occurs. The region of telomerase-RHPS4 parameter space where steady states exist has also been determined and is illustrated in Fig. 5.

Promising directions for future work include modelling other players involved in telomere maintenance. For instance, the shelterin protein POT1 is involved in several processes at the telomere end, which might be an interesting avenue to follow in future work. For example, two distinct functions of the protein have been identified, depending on the position of POT1 at the 3’-overhang: if POT1 is bound at the very end of the overhang (leaving less than 8 nt free at the 3’ end), telomerase cannot extend the telomere (Lei et al. 2005). On the other hand, the formation of G-quadruplex structures require that POT1 is not bound to the terminal four telomeric sequences involved in G-quadruplex formation. Since G-quadruplexes form spontaneously at the end of telomeres and are in a dynamic equilibrium with unfolded or partially unfolded forms, POT1 binding of unfolded structures may trap telomeres in the open form (Zaug et al. 2005). It may be interesting in future studies to analyse the effects of different levels of POT1 binding, in particular POT1 being overexpressed or suppressed in telomerase-positive cells.

## Notes

### Acknowledgments

We would like to thank Charles Laughton for useful conversations. B.V. Hirt expresses his thanks for funding to the School of Mathematical Sciences, University of Nottingham.

## References

- Abdallah P, Luciano P, Runge KW, Lisby M, Geli V, Gilson E, Teixeira MT (2009) A two-step model for senescence triggered by a single critically short telomere. Nat Cell Biol 11:988–993CrossRefGoogle Scholar
- Allsopp RC, Vaziri H, Patterson C, Goldstein S, Younglai EV, Futcher AB, Greidert CW, Harley CB (1992) Telomere length predicts replicative capacity of human fibroblasts. Proc Natl Acad Sci USA 89:10114–10118CrossRefGoogle Scholar
- Antal T, Blagoev KB, Trugman SA, Redner S (2007) Aging and immortality in a cell proliferation model. J Theor Biol 248:411–417CrossRefMathSciNetGoogle Scholar
- Arino O, Kimmel M, Webb GF (1995) Mathematical-modeling of the loss of telomere sequences. J Theor Biol 177:45–57CrossRefGoogle Scholar
- Arkus N (2005) A mathematical model of cellular apoptosis and senescence through the dynamics of telomere loss. J Theor Biol 235:13–32CrossRefMathSciNetGoogle Scholar
- Bilsland AE, Cairney CJ, Keith WN (2011) Targeting the telomere and shelterin complex for cancer therapy: current views and future perspectives. J Cell Mol Med 15:179–186CrossRefGoogle Scholar
- Blackburn EH (2001) Switching and signaling at the telomere. Cell 106:661–673CrossRefGoogle Scholar
- Blagoev KB (2009) Cell proliferation in the presence of telomerase. PLoS One 4:e4622CrossRefGoogle Scholar
- Canela A, Vera E, Klatt P, Blasco MA (2007) High-throughput telomere length quantification by FISH and its application to human population studies. Proc Natl Acad Sci USA 104:5300–5305CrossRefGoogle Scholar
- Cesare AJ, Reddel RR (2010) Alternative lengthening of telomeres: models, mechanisms and implications. Nat Rev Genet 11:319–330CrossRefGoogle Scholar
- Cheng MK, Modi C, Cookson JC, Hutchinson I, Heald RA, McCarroll AJ, Missailidis S, Tanious F, Wilson WD, Mergny JL, Laughton CA, Stevens MFG (2008) Antitumor polycyclic acridines. 20. Search for DNA quadruplex binding selectivity in a series of 8,13-dimethylquino 4,3,2-kl acridinium salts: Telomere-targeted agents. J Med Chem 51:963–975CrossRefGoogle Scholar
- Cohen SB, Graham ME, Lovrecz GO, Bache N, Robinson PJ, Reddel RR (2007) Protein composition of catalytically active human telomerase from immortal cells. Science 315:1850–1853CrossRefGoogle Scholar
- Collins MA (1981) A quasicontinuum approximation for solitons in an atomic chain. Chem Phys Lett 77:342–347CrossRefMathSciNetGoogle Scholar
- Cookson JC, Dai FP, Smith V, Heald RA, Laughton CA, Stevens MFG, Burger AM (2005) Pharmacodynamics of the G-quadruplex-stabilizing telomerase inhibitor 3,11-difluoro-6,8,13-trimethyl-8H-quino 4,3,2-kl acridinium methosulfate (RHPS4) in vitro: activity in human tumor cells correlates with telomere length and can be enhanced, or antagonized, with cytotoxic agents. Mol Pharmacol 68:1551–1558Google Scholar
- Cristofari G, Lingner J (2006) Telomere length homeostasis requires that telomerase levels are limiting. EMBO J 25:565–574CrossRefGoogle Scholar
- Dai XY, Huang CH, Bhusari A, Sampathi S, Schubert K, Chai WH (2010) Molecular steps of G-overhang generation at human telomeres and its function in chromosome end protection. EMBO J 29:2788–2801CrossRefGoogle Scholar
- de Lange T (2004) T-loops and the origin of telomeres. Nat Rev Mol Cell Biol 5:323–329CrossRefGoogle Scholar
- de Lange T (2005) Shelterin: the protein complex that shapes and safeguards human telomeres. Gene Dev 19:2100–2110CrossRefGoogle Scholar
- de Lange T (2009) How telomeres solve the end-protection problem. Science 326:948–952CrossRefGoogle Scholar
- Deng YB, Chan SS, Chang S (2008) Telomere dysfunction and tumour suppression: the senescence connection. Nat Rev Cancer 8:450–458Google Scholar
- Dmitriev PV, Vassetzky YS (2009) Analysis of telomeric DNA: current approaches and methods. Russ J Dev Biol 40:125–144CrossRefGoogle Scholar
- Golubev A, Khrustalev S, Butov A (2003) An in silico investigation into the causes of telomere length heterogeneity and its implications for the Hayflick limit. J Theor Biol 225:153–170CrossRefMathSciNetGoogle Scholar
- Gowan SM, Heald R, Stevens MFG, Kelland LR (2001) Potent inhibition of telomerase by small-molecule pentacyclic acridines capable of interacting with G-quadruplexes. Mol Pharmacol 60:981–988Google Scholar
- Grasman J, Salomons HM, Verhulst S (2011) Stochastic modeling of length-dependent telomere shortening in corvus monedula. J Theor Biol 282:1–6CrossRefGoogle Scholar
- Greenberg RA (2005) Telomeres, crisis and cancer. Curr Mol Med 5:213–218CrossRefGoogle Scholar
- Griffith JD, Comeau L, Rosenfield S, Stansel RM, Bianchi A, Moss H, de Lange T (1999) Mammalian telomeres end in a large duplex loop. Cell 97:503–514CrossRefGoogle Scholar
- Gunaratnam M, Greciano O, Martins C, Reszka AP, Schultes CM, Morjani H, Riou JF, Neidle S (2007) Mechanism of acridine-based telomerase inhibition and telomere shortening. Biochem Pharmacol 74:679–689CrossRefGoogle Scholar
- Hayflick L (1965) The limited in vitro lifetime of human diploid cell strains. Exp Cell Res 37:614–636CrossRefGoogle Scholar
- Hayflick L (1979) Cell biology of aging. J Investig Dermatol 73:8–14CrossRefGoogle Scholar
- Heald RA, Modi C, Cookson JC, Hutchinson I, Laughton CA, Gowan SM, Kelland LR, Stevens MFG (2002) Antitumor polycyclic acridines. 8. Synthesis and telomerase-inhibitory activity of methylated pentacyclic acridinium salts. J Med Chem 45:590–597CrossRefGoogle Scholar
- Hemann MT, Strong MA, Hao LY, Greider CW (2001) The shortest telomere, not average telomere length, is critical for cell viability and chromosome stability. Cell 107:67–77CrossRefGoogle Scholar
- Henderson E, Hardin CC, Walk SK, Tinoco I, Blackburn EH (1987) Telomeric DNA oligonucleotides form novel intramolecular structures containing guanine guanine base pairs. Cell 51:899–908CrossRefGoogle Scholar
- Hirt BV (2012) Mathematical modelling of cell cycle and telomere dynamics. PhD thesis, University of Nottingham. http://etheses.nottingham.ac.uk
- Hirt BV, Wattis JAD, Preston SP, Laughton CA (2012) The effects of a telomere destabilizing agent on cancer cell-cycle dynamics: integrated modelling and experiments. J Theor Biol 295:9–22CrossRefMathSciNetGoogle Scholar
- Itzkovitz S, Shlush LI, Gluck D, Skorecki K (2008) Population mixture model for nonlinear telomere dynamics. Phys Rev E 78(6):060902Google Scholar
- Johnson LA, Byrne HM, Willis AE, Laughton CA (2011) An integrative biological approach to the analysis of tissue culture data: application to the antitumour agent RHPS4. Integr Biol 3:843–849CrossRefGoogle Scholar
- Kim NW, Piatyszek MA, Prowse KR, Harley CB, West MD, Ho PLC, Coviello GM, Wright WE, Weinrich SL, Shay JW (1994) Specific association of human telomerase activity with immortal cells and cancer. Science 266:2011–2015CrossRefGoogle Scholar
- Kimura M, Stone RC, Hunt SC, Skurnick J, Lu X, Cao X, Harley CB, Aviv A (2010) Measurement of telomere length by the southern blot analysis of terminal restriction fragment lengths. Nat Protoc 5:1596–1607CrossRefGoogle Scholar
- Knig SLB, Evans AC, Huppert JL (2010) Seven essential questions on G-quadruplexes. BioMol Concepts 1:197–213Google Scholar
- Kowald A (1997) Possible mechanisms for the regulation of telomere length. J Mol Biol 273:814–825CrossRefGoogle Scholar
- Lei M, Zaug AJ, Podell ER, Cech TR (2005) Switching human telomerase on and off with hPOT1 protein in vitro. J Biol Chem 280:20449–20456CrossRefGoogle Scholar
- Leonetti C, Amodei S, D’Angelo C, Rizzo A, Benassi B, Antonelli A, Elli R, Stevens MFG, D’Incalci M, Zupi G, Biroccio A (2004) Biological activity of the G-quadruplex ligand RHPS4 (3,11-difluoro-6,8,13-trimethyl-8H-quino 4,3,2-kl acridinium methosulfate) is associated with telomere capping alteration. Mol Pharmacol 66:1138–1146CrossRefGoogle Scholar
- Levy MZ, Allsopp RC, Futcher AB, Greider CW, Harley CB (1992) Telomere end-replication problem and cell aging. J Mol Biol 225:951–960CrossRefGoogle Scholar
- Lipps HJ, Rhodes D (2009) G-quadruplex structures: in vivo evidence and function. Trends Cell Biol 19:414–422CrossRefGoogle Scholar
- Macville M, Schröck E, Padilla-Nash H, Keck C, Ghadimi BM, Zimonjic D, Popescu N, Ried T (1999) Comprehensive and definitive molecular cytogenetic characterization of HeLa cells by spectral karyotyping. Cancer Res 59:141–150Google Scholar
- Makarov VL, Hirose Y, Langmore JP (1997) Long G tails at both ends of human chromosomes suggest a C strand degradation mechanism for telomere shortening. Cell 88:657–666CrossRefGoogle Scholar
- Martens UM, Chavez EA, Poon SSS, Schmoor C, Landsdorp PM (2000) Accumulation of short telomeres in human fibroblasts prior to replicative senescence. Exp Cell Res 256:291–299CrossRefGoogle Scholar
- Masutomi K, Yu EY, Khurts S, Ben-Porath I, Currier JL, Metz GB, Brooks MW, Kaneko S, Murakami S, DeCaprio JA, Weinberg RA, Stewart SA, Hahn WC (2003) Telomerase maintains telomere structure in normal human cells. Cell 114:241–253CrossRefGoogle Scholar
- Milo R, Jorgensen P, Moran U, Weber G, Springer M (2010) Bionumbers-the database of key numbers in molecular and cell biology. Nucleic Acids Res 38:D750–D753CrossRefGoogle Scholar
- Monchaud D, Teulade-Fichou MP (2008) A hitchhiker’s guide to G-quadruplex ligands. Org Biomol Chem 6:627–636CrossRefGoogle Scholar
- Monier K, Armas JCG, Etteldorf S, Ghazal P, Sullivan KF (2000) Annexation of the interchromosomal space during viral infection. Nat Cell Biol 2:661–665CrossRefGoogle Scholar
- Neidle S (2010) Human telomeric G-quadruplex: The current status of telomeric G-quadruplexes as therapeutic targets in human cancer. FEBS J 277:1118–1125CrossRefGoogle Scholar
- Oganesian L, Karlseder J (2009) Telomeric armor: the layers of end protection. J Cell Sci 122:4013–4025CrossRefGoogle Scholar
- Olofsson P, Bertuch AA (2010) Modeling growth and telomere dynamics in saccharomyces cerevisiae. J Theor Biol 263:353–359CrossRefMathSciNetGoogle Scholar
- Olofsson P, Kimmel M (1999) Stochastic models of telomere shortening. Math Biosci 158:75–92CrossRefzbMATHMathSciNetGoogle Scholar
- op den Buijs J, van den Bosch PPJ, Musters MWJM, van Riel NAW (2004) Mathematical modeling confirms the length-dependency of telomere shortening. Mech Ageing Dev 125:437–444Google Scholar
- Palm W, de Lange T (2008) How shelterin protects mammalian telomeres. Annu Rev Genet 42:301–334CrossRefGoogle Scholar
- Poon SSS, Martens UM, Ward RK, Lansdorp PM (1999) Telomere length measurements using digital fluorescence microscopy. Cytometry 36:267–278CrossRefGoogle Scholar
- Proctor CJ, Kirkwood TBL (2002) Modelling telomere shortening and the role of oxidative stress. Mech Ageing Dev 123:351–363CrossRefGoogle Scholar
- Proctor CJ, Kirkwood TBL (2003) Modelling cellular senescence as a result of telomere state. Aging Cell 2:151–157CrossRefGoogle Scholar
- Qi Q (2011) Mathematical modelling of telomere dynamics. PhD thesis, University of Nottingham. http://etheses.nottingham.ac.uk/2258
- Richter T, von Zglinicki T (2007) A continuous correlation between oxidative stress and telomere shortening in fibroblasts. Exp Gerontol 42:1039–1042CrossRefGoogle Scholar
- Rodriguez-Brenes IA, Peskin CS (2010) Quantitative theory of telomere length regulation and cellular senescence. Proc Natl Acad Sci USA 107:5387–5392CrossRefGoogle Scholar
- Rosenau P (1986) Dynamics of nonlinear mass-spring chains near the continuum-limit. Phys Lett A 118:222–227CrossRefMathSciNetGoogle Scholar
- Rubelj I, Vondracek Z (1999) Stochastic mechanism of cellular aging—abrupt telomere shortening as a model for stochastic nature of cellular aging. J Theor Biol 197:425–438CrossRefGoogle Scholar
- Saldanha SN, Andrews LG, Tollefsbol TO (2003) Assessment of telomere length and factors that contribute to its stability. Eur J Biochem 270:389–403CrossRefGoogle Scholar
- Samassekou O, Gadji M, Drouin R, Yan J (2010) Sizing the ends. Normal length of human telomeres. Ann Anat 192:284–291CrossRefGoogle Scholar
- Schaffitzel C, Berger I, Postberg J, Hanes J, Lipps HJ, Pluckthun A (2001) In vitro generated antibodies specific for telomeric guanine-quadruplex DNA react with Stylonychia lemnae macronuclei. Proc Natl Acad Sci USA 98:8572–8577CrossRefGoogle Scholar
- Sidorov IA, Hirsch KS, Harley CB, Dimitrov DS (2003) Cancer treatment by telomerase inhibitors: predictions by a kinetic model. Math Biosci 181:209–221CrossRefzbMATHMathSciNetGoogle Scholar
- Sozou PD, Kirkwood TBL (2001) A stochastic model of cell replicative senescence based on telomere shortening, oxidative stress, and somatic mutations in nuclear and mitochondrial DNA. J Theor Biol 213:573–586CrossRefGoogle Scholar
- Takai H, Smogorzewska A, de Lange T (2003) DNA damage foci at dysfunctional telomeres. Curr Biol 13:1549–1556CrossRefGoogle Scholar
- Tang J, Zy Kan, Yao Y, Wang Q, Hao Yh, Tan Z (2008) G-quadruplex preferentially forms at the very 3’ end of vertebrate telomeric DNA. Nucleic Acids Res 36:1200–1208CrossRefGoogle Scholar
- Tauchi T, Shin-ya K, Sashida G, Sumi M, Okabe S, Ohyashiki JH, Ohyashiki K (2006) Telomerase inhibition with a novel G-quadruplex-interactive agent, telomestatin: in vitro and in vivo studies in acute leukemia. Oncogene 25:5719–5725CrossRefGoogle Scholar
- Teixeira MT, Arneric M, Sperisen P, Lingner J (2004) Telomere length homeostasis is achieved via a switch between telomerase-extendible and -nonextendible states. Cell 117:323–335CrossRefGoogle Scholar
- von Zglinicki T, Martin-Ruiz CM, Saretzki G (2005) Telomeres, cell senescence and human ageing. Signal Transduct 3:103–114CrossRefGoogle Scholar
- Wallweber G, Gryaznov S, Pongracz K, Pruzan R (2003) Interaction of human telomerase with its primer substrate. Biochemistry 42:589–600CrossRefGoogle Scholar
- Wang F, Podell ER, Zaug AJ, Yang Y, Baciu P, Cech TR, Lei M (2007) The POT1-TPP1 telomere complex is a telomerase processivity factor. Nature 445:506–510CrossRefGoogle Scholar
- Wattis JAD (1996) Approximations to solitary waves on lattices. 3. The monatomic lattice with second-neighbour interactions. J Phys A: Math Gen 29:8139–8157CrossRefzbMATHMathSciNetGoogle Scholar
- Wolkenhauer O, Fell D, De Meyts P, Bluthgen N, Herzel H, Le Novere N, Hofer T, Schurrle K, van Leeuwen I (2009) Sysbiomed report: Advancing systems biology for medical applications. IET Syst Biol 3:131–136CrossRefGoogle Scholar
- Wright WE, Piatyszek MA, Rainey WE, Byrd W, Shay JW (1996) Telomerase activity in human germline and embryonic tissues and cells. Dev Genet 18:173–179CrossRefGoogle Scholar
- Wu P, de Lange T (2009) Human telomerase caught in the act. Cell 138:432–434CrossRefGoogle Scholar
- Xu Y (2011) Chemistry in human telomere biology: structure, function and targeting of telomere DNA/RNA. Chem Soc Rev 40:2719–2740CrossRefGoogle Scholar
- Zahler AM, Williamson JR, Cech TR, Prescott DM (1991) Inhibition of telomerase by G-quartet DNA structures. Nature 350:718–720CrossRefGoogle Scholar
- Zaug AJ, Podell ER, Cech TR (2005) Human POT1 disrupts telomeric G-quadruplexes allowing telomerase extension in vitro. Proc Natl Acad Sci USA 102:10864–10869CrossRefGoogle Scholar
- Zhao Y, Kan ZY, Zeng ZX, Hao YH, Chen H, Tan Z (2004) Determining the folding and unfolding rate constants of nucleic acids by biosensor. Application to telomere G-quadruplex. J Am Chem Soc 126:13255–13264CrossRefGoogle Scholar
- Zhao Y, Sfeir AJ, Zou Y, Buseman CM, Chow TT, Shay JW, Wright WE (2009) Telomere extension occurs at most chromosome ends and is uncoupled from fill-in in human cancer cells. Cell 138:463–475CrossRefGoogle Scholar

## Copyright information

**Open Access**This article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.