Skip to Main Content
Skip Nav Destination
Open Access License / Drug Dosage / Disclaimer
This article is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND). Usage and distribution for commercial purposes as well as any distribution of modified material requires written permission. Drug Dosage: The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any changes in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug. Disclaimer: The statements, opinions and data contained in this publication are solely those of the individual authors and contributors and not of the publishers and the editor(s). The appearance of advertisements or/and product references in the publication is not a warranty, endorsement, or approval of the products or services advertised or of their effectiveness, quality or safety. The publisher and the editor(s) disclaim responsibility for any injury to persons or property resulting from any ideas, methods, instructions or products referred to in the content or advertisements.

Aging is a systemic process which progressively manifests itself at multiple levels of structural and functional organization from molecular reactions and cell-cell interactions in tissues to the physiology of an entire organ. There is ever increasing data on biomedical relevant network interactions for the aging process at different scales of time and space. To connect the aging process at different structural, temporal and spatial scales, extensive systems biological approaches need to be deployed. Systems biological approaches can not only systematically handle the large-scale datasets (like high-throughput data) and the complexity of interactions (feedback loops, cross talk), but also can delve into nonlinear behaviors exhibited by several biological processes which are beyond intuitive reasoning. Several public-funded agencies have identified the synergistic role of systems biology in aging research. Using one of the notable public-funded programs (GERONTOSYS), we discuss how systems biological approaches are helping the scientists to find new frontiers in aging research. We elaborate on some systems biological approaches deployed in one of the projects of the consortium (ROSage). The systems biology field in aging research is at its infancy. It is open to adapt existing systems biological methodologies from other research fields and devise new aging-specific systems biological methodologies.

Systems biology is a multidisciplinary approach that has been developed over the last 15 years, and which attempts to elucidate the structural and functional organization of biological systems at the intracellular level as well as the interaction between cells and their microenvironment [1]. What makes the methodology special is that questions concerning the structure and function of the biological systems are investigated using quantitative experimental data, which are analyzed through the use of sophisticated mathematical and computational tools, such as advanced statistics, data mining and mathematical modelling. The rationale for the methodology is to apprehend the complexity of biological systems, to derive nonintuitive hypotheses about their functioning, to help designing and formulating new experiments able to prove these hypotheses, and, ultimately, to develop computational tools with predictive ability in a biomedical focus.

In its original conception, the method implied iterative cycles of hypothesis formulation, design of experiments, and integration of the obtained data using advanced mathematical tools like modelling and simulation [1,2]. The method included the following steps (fig. 1):

Fig. 1

General structure of the canonical systems biology method. The procedure includes: construction of the regulatory map for the system under investigation using existing biomedical information; set up of mathematical model representing interesting part of the map; characterization of the model equations calibration using quantitative experimental data; model assessment and refinement; model validation through additional experiments, and use of the model to make predictions on the basic features of the system, which are tested with additional experiments.

Fig. 1

General structure of the canonical systems biology method. The procedure includes: construction of the regulatory map for the system under investigation using existing biomedical information; set up of mathematical model representing interesting part of the map; characterization of the model equations calibration using quantitative experimental data; model assessment and refinement; model validation through additional experiments, and use of the model to make predictions on the basic features of the system, which are tested with additional experiments.

Close modal

(a) Relevant biomedical knowledge is retrieved from publications, databases and public repositories of biological data using computational tools, text mining and manual curation. The information is used to construct a graphical depiction of the biochemical system (i.e. the network), its compounds (mRNAs, microRNAs, proteins, small molecules, protein complexes…), as well as relevant interactions. This graphical depiction is often called ‘regulatory map' of the system and contains the up-to-date biological, biomedical, and sometimes clinical, knowledge on the investigated system.

(b) The regulatory map itself is a resource that can be analyzed and mined using computational tools from network biology. This helps organizing the network in modules relevant to the investigated problem, or finding unexpected connections between pathways thought to be independent (i.e. cross talk).

(c) Critical parts of the regulatory network are translated into a mathematical model, consisting of nothing but a set of mathematical equations that encode the knowledge and hypothesis about the systems investigated. Under some circumstances, the analysis of this preliminary model can yield already useful information. But in most of the cases, the model has to be better characterized using quantitative data generated in customized experiments. Usually, one proceeds in iterative cycles of mathematical model derivation, integration of the experimental data in the model and reformulation of model equations in case the agreement between the experimental data and the model simulations is not satisfactory.

(d) The output of this process is a computational tool, the calibrated mathematical model, which has predictive abilities. This means that model simulations can be used to predict the behavior of the network under experimental scenarios not yet tested. In fact, many examples of validated mathematical model are available in the literature, for example to predict drug targets [3] or chemoresistance [4], and to identify diagnosis biomarkers [5] for cancer and other human diseases.

While in the last years several new communities of researchers (with very different scientific background) have joined the biomedical sciences in an attempt to break down prevalent diseases like cancer and diabetes, the philosophy behind the systems biology approach has been expanded, and other workflows can be found in the recent literature under this epigraph. The common point among all of them is the intensive use of advanced mathematical and computational tools to analyze, dissect, mine and make sense out of large sets of quantitative experimental data. We can distinguish between the omics approach, reconstruction and simulation of large biological networks, elucidation of the complexity emerging from nonlinear network motifs, and multi-scale data integration and modelling.

In an increasing amount of subfields, researchers face the problem of analyzing and integrating massive amounts of high-throughput biological or biomedical data. In this case, it has become very popular to make use of advanced statistical techniques in an attempt to identify global expression patterns. In cancer and other multifactorial diseases, large amounts of high-throughput and clinical data obtained from large cohorts of patients are analyzed using this strategy, aiming at ‘fishing' unknown and unexpected molecular mechanisms triggering the disease. Or even more important, obtaining reliable sets of biomarkers (the so-called gene signatures), able to accurately stratify populations of patients in a predictive fashion, making possible to make more precise the diagnosis and foresee the patient response to therapies [6].

In the last years, the certainty emerged that intracellular biochemical processes can rarely be compartmentalized into small pathways, especially when considering biomedical questions. The new pictures of many diseases show massive and complex biochemical networks, composed of coding genes, small and long noncoding RNAs, metabolites, small molecules and interacting proteins, all of them interconnected by a myriad of different kinds of interactions and molecular modifications [7]. The natural way to characterize these systems is the use of multi-level sets of in vitro or in vivo high-throughput data. Under these conditions, human intuition or basic data analysis techniques are clearly inadequate and limited tools, and hence advanced methods for data analysis, network reconstruction and simulation are required, which have been used with remarkable success in the last years [4,8].

If the size of biomedical relevant networks and the multiplicity of high-throughput data sources necessary to characterize them was already a challenge big enough, we have found that these networks are extensively enriched in nonlinear network motifs. We mean systems of interactions between network compounds displaying the structure, and therefore the behavior, of structures like positive and negative feedback loops, but also coherent and incoherent feed-forward loops, or network hubs (see fig. 2) [7,9,10]. Even more challenging, rather than isolated, these motifs very often overlap and cross talk, resulting in a structural and functional complexity that is far beyond the understanding of our mind [11]. However, feedback loops have been extensively investigated for decades by engineers and physicists, who have developed a vast framework of analysis based on the use of mathematical modelling. This framework has been largely exploited in systems biology [12,13].

Fig. 2

Biochemical networks are enriched in nonlinear motifs, like positive feedback loops (left), a regulatory structure in which the activation of a signaling event positively regulates a signaling process upstream of the pathway. Positive feedback loops can induce signal amplification, ultrasensitivity to input signal, but also in some special cases bistability (right). a Ultrasensitivity enables a cellular system to transform graded input signals at a certain threshold into discrete all-or-none output. b In systems showing bistability, for some given experimental regime, small perturbations in the setup may induce totally different fates for the system, inducing for example quick and irreversible signal termination or persistent activation.

Fig. 2

Biochemical networks are enriched in nonlinear motifs, like positive feedback loops (left), a regulatory structure in which the activation of a signaling event positively regulates a signaling process upstream of the pathway. Positive feedback loops can induce signal amplification, ultrasensitivity to input signal, but also in some special cases bistability (right). a Ultrasensitivity enables a cellular system to transform graded input signals at a certain threshold into discrete all-or-none output. b In systems showing bistability, for some given experimental regime, small perturbations in the setup may induce totally different fates for the system, inducing for example quick and irreversible signal termination or persistent activation.

Close modal

We know now that in order to apprehend the complexity of many biological phenomena, we have to go beyond the scale of the intracellular biochemical processes, and consider the communication between cells and their microenvironment, the dynamics of tissue organization and replenishment, and even the influence of the whole-body processes. When integrating quantitative biological data with different structural, spatial and temporal scales of these biological phenomena (e.g. intracellular-omics data, live microscopy and systemic blood markers), the so-called multi-scale mathematical modelling has emerged as a tool to make sense out of this complexity. This approach has been especially successful at investigating complex spatial-temporal phenomena occurring in solid tumors and leukemia, including angiogenesis, or systemic response after chemotherapy [14,15].

If we integrate the definition of systems biology and the different subfields discussed in previous sections with aging research, then we can conclude that aging research appears to be an ideal research field to deploy extensive systems biology efforts.

Contrary to most of the other fields in biomedicine, the notion of biochemical networks and their dysfunction as drivers for the emergence of aging has been firmly established for decades. Back in 1994, Kowald and Kirkwood hypothesized that multiple biological processes, some of them involving the progressive dysfunction of critical metabolic and repair pathways, work in parallel and probably cross talk during the emergence of aging [16,17]. To test the consequences of their hypothesis, they derived, simulated and analyzed a mathematical model, which integrates several existing theories about the triggers of aging at the cellular level, including the so-called free radical theory and the protein error theory. The model results were in agreement with the caloric restriction hypothesis and its role in reducing free radical production, protein damage and hence extending life span.

Franceschi and coworkers further extended the network theory of aging by including the role of the immune system and the inflammatory signals in it [18]. In their opinion, an age-related decline in the ability of the tissues to mitigate the effects of environmental and internal stress, probably connected to the progressive dysfunction of the damage repair systems hypothesized by Kirkwood and Kowald, triggers a progressive systemic proinflammatory phenotype. This proinflammatory phenotype negatively influences body performance and contributes to the emergence of many aging-associated diseases like Alzheimer's disease, osteoporosis and diabetes.

To a certain extent, Franceschi's hypothesis of the so-called ‘inflamm-aging' supports the notion that aging is the consequence of the progressive deregulation of a ‘network of networks'. In line with this, De Magalhaes and Toussaint developed the first curated database of genes related to human aging [19]. The information contained in the database was further used to construct the first large proteomic network map of human aging. Interestingly, when they analyzed the obtained network, they found that many of the genes in their aging network are commonly associated with the genetics of development. This work has been continued by other groups who have worked in establishing the topology of this aging ‘network of networks'. To mention an additional example, Xue and collaborators combined information on protein-protein interactions and gene expression data during fruit fly life span to construct and modularize a network of interacting proteins playing a role in aging [20]. Interestingly, their results pinpoint to a reduced number of network modules as critical mediators of aging. But they also suggest that the structure of the obtained network was such that dysfunction in few critical regulatory nodes, which connect some of these aging modules, may be behind what they call ‘the molecular basis for the stochastic nature of aging'.

The aging network is enriched in motifs that govern the emergence of aging in a timely and spatially nonlinear fashion. Further investigation into the network and the pathways guilty of mediating aging emergence found that, to the surprise of many researchers, they are enriched in motifs like feedback and -forward loops [21,22]. Thus, their likely nonlinear dynamics are in clear contradiction to the vision of aging as a progressive process. Furthermore, some results support the idea that at least some of the critical events triggering aging emergence may not behave linearly [18]. As Kitano, Kirkwood and others mention, the motivation behind the existence of large biochemical networks enriched in nonlinear motifs may be to provide robustness against perturbation, and the loss of this robust performance due to pathway dysfunction may be the basis of several mechanisms involved in aging [23,24,25].

Further examples illustrate the effect of nonlinear regulatory circuits involved in aging-related phenotypes. Passos and collaborators found a delayed feedback loop system involving a long signaling circuit, by which long-term activation of p21 and reactive oxygen species (ROS) production and subsequent cell senescence can be activated upon a sufficiently long-lasting DNA damage response [26]. In line with this, Lai and collaborators show that p21 can be regulated in a cooperative manner by multiple miRNAs, which allows for a context-dependent regulation of p21 signaling in several aging-associated contexts. Kriete and collaborators constructed a regulatory network, which included most of the central cellular mechanisms involved in aging, and showed that most of them were connected and cross-talked via a number of overlapping positive and negative feedback loops [22]. When translated into a mathematical model, their model simulations showed that dysfunction of key, positive-feedback loop-regulated processes in the cellular energy metabolism may lead to enhanced and accelerated cellular damage. Furthermore, they showed that a number of negative-feedback loop systems involved in the NFκB and mTOR signaling have the ability to alleviate cellular damage by regulating processes that are essential for cell survival, such as mitochondrial respiration or biosynthesis, and their deregulation may enhance aging phenotypes.

Aging is not a cellular phenotype, but it manifests at multiple levels of organization in the human body. It is even intuitive to say that aging as a phenomenon evades the frontiers of the intracellular machinery, and involves complex interactions between the cell and its microenvironment, the role of cell dynamics in tissue organization and replenishment, but also higher-scale processes connecting the functioning of organs with systemic, whole-body like processes. The emerging multi-scale system (cell damage → tissue/organ dysfunction → whole body deregulation) is enriched in equally multi-scale feedback loops, which connect the progressive cellular dysfunction with systemic stress signals, and the way back (systemic stress signals → tissue/organ dysfunction → cell damage [18,24]. In line with this perception, the ultimate aim of aging research is to create a framework to connect experimental evidence of aging progression emerging at different structural, spatial and temporal scales, similar to what has been done in cancer [14].

The use of this approach in aging is still in its infancy, but we still can find some notable examples. Hoehme and collaborators derived and characterized using multiple sources of data a multi-scale mathematical model accounting for the liver regeneration after damage in mice [27]. Remarkably, the model was able to predict novel essential mechanisms for liver regeneration, a process that must be critical to impede aging-associated deterioration of the liver. Van Leeuwen and coworkers made use of mathematical modelling to interconnect biological processes happening at different organizational scales within the body, which connect caloric restriction and the dynamics of oxidative cellular damage, metabolism, body weight change and ultimately life span extension [28,29]. The obtained mathematical model was able to reproduce growth and survival data on mice exposed to different food levels and supported the caloric restriction hypothesis. McAuley and collaborators developed a whole-body mathematical model of age-related deregulation of the cholesterol metabolism [30]; for its characterization, they used multifactorial data, accounting for the dynamics of the system from both the molecular and the physiological perspectives. Model simulations and analysis were used to investigate the actual relevance of known molecular mechanisms in the age-associated deregulation of cholesterol.

When thinking about fighting aging, an engineering approach can be used. This idea is behind one of the most controversial theories in aging research, the so-called SENS hypothesis postulated by Aubrey de Grey [31,32]. This hypothesis assumes that aging emerges due to the progressive accumulation of different kinds of damage during life span, which act in a potentially synergistic manner. These sources of systemic damage include mutations and epigenetic modifications in critical genes, stress-induced mutations in the mitochondrial genome, as well as intracellular and extracellular accumulation of different sorts of cellular debris, which are the basis of many aging-linked diseases. The concept is quite in agreement with Kowald and Kirkwood's Network Theory of Aging. However, the disputed point of de Grey's theory is that it proposes an engineering-like approach to fight the emergence of aging, which neglects the deep understanding of aging's origin and focuses on the design of biomedical therapies aimed at mitigating, preventing or repairing the sources of these cumulative organic damage that defines aging.

If at some point in the near future, the idea behind the SENS hypothesis becomes prevalent, then systems biology-like approaches would become an even more integrative part of aging research: system biology is in fact inspired by the methodology used by physicists and engineers to design and optimize complex technological devices, whose behavior is governed by nested regulatory circuits, especially feedback loops, as it is in the case of the network of networks whose dysfunction causes the emergence of aging.

Having in mind the motivations discussed in the previous section, it is not a surprise that worldwide several public funding agencies have identified the synergistic role of systems biology in aging research and promoted in the very last years a number of projects making use of systems biology in aging research. A remarkable one is the Gerontosys initiative funded by the German Federal Ministry of Education and Research, which is being funded for the period 2009-2016, with a global budget of about EUR 32 million. Figure 3 summarizes the majority of projects funded in the second call. In the following, we examine some of these projects to illustrate the overall structure of an aging systems biology project and how most of them were conceived as rather large interdisciplinary research networks. Finally, we will focus on our own project, the ROSAge project, to provide an example for the initial conception, the work plan for the integration of different disciplines, and its development, but also the particular difficulties that one faces when approaching aging from a systemic view.

Fig. 3

Intracellular biochemical systems, tissues and phenotypes related to aging emergence and some systems biology projects addressing them. Legend of projects: A = SyStaR, B = MAGE, C = Gerontoshield and Primage, D = SyBACol, E = NephAge, F = GerontoMitosys, G = JenAge, H = ROSage.

Fig. 3

Intracellular biochemical systems, tissues and phenotypes related to aging emergence and some systems biology projects addressing them. Legend of projects: A = SyStaR, B = MAGE, C = Gerontoshield and Primage, D = SyBACol, E = NephAge, F = GerontoMitosys, G = JenAge, H = ROSage.

Close modal

Several projects in the Gerontosys initiative focus on aging in the hematopoietic system. Some of them examine the aging effects in hematopoietic stem cells (HSC), whereas others focus on aging-associated defects of the immune system. SyStaR (Molecular Systems Biology of Impaired Stem Cell Function in Regeneration during Aging; A in fig. 3) and MAGE (Model of the Aging Epigenome; B in fig. 3) are the projects with stem cell focus. Aging of the immune system is covered by Primage (Protective Immunity in Aging) and GerontoShield (C infig. 3).

To mention some of the insights emerging from the systems biology approach used in these projects, the researchers involved in the SyStaR project could prove that HSC differentiation in response to DNA damage is a protection strategy which operated through the BATF (basic leucine zipper transcription factor-ATF like) pathway, and can reduce the burden of mutations in the stem cell niche (A1 in fig. 3). Interestingly, Wang et al. (2012) [33] corroborated the experimental findings theoretically by encoding the cellular processes of proliferation, differentiation and apoptosis into a mathematical model. The model simulations ruled out the role of apoptosis and strengthened the conclusion that differentiation is the driving force behind the reduction of the HSC pool in their niche, and not apoptosis. Moreover, the modelling and experimental results can explain the uneven differentiation levels between lymphoid and myeloid competent HSCs. The mechanistic interpretation emerging out of the systems biology analysis performed is that the increase in the BATF level stimulates predominantly the differentiation of lymphoid-competent HSCs, thereby draining their stem cell pool [33].

It is known that epigenetic histone modifications have a big impact on stem-cell-based organ regeneration and hence on the emergence of aging-associated organ deterioration. The team of the SyStar project could experimentally show epigenetic histone deacetylation via activation of Wnt5a and CDC42 signaling, which ultimately contributed to the cellular stem cell aging (A2 in fig. 3). In line with this, the effects of similar epigenetic changes on aging were investigated via mathematical modelling and simulation in the MAGE project [34,35] (B infig. 3). Interestingly, epigenetic modifications are processes inviting systems biological interpretations as they comprise several nonlinear network motifs like feedback loops [36], which are an integral part of engineering approaches, as discussed in the section Systems Biology Comes in Several ‘Flavors'. Precisely, a positive feedback loop is generated when modified histones stimulate recruitment of modification enzymes and can lead either to a rapid increase of histone modifications (H+), or to a second steady state with low modifications (H⁰; see fig. 2) [36]. Modelling results indicate that this switch from high to low modifications can be triggered by the dilution of histone modifications during DNA replication [34]. Ultimately, this process leads to impaired stem-cell function, loss of stem cell pools and hence the emergence of aging-associated phenotypes.

Given the general importance of histone modifications in the aging process of HSCs, Przybilla and collaborators further performed an in silico simulation of the effect of histone and DNA modification rates on transcriptional activities of quiescent cells in the stem cell niches. They found that fluctuations in this DNA modification process could rejuvenate aged HSC and further activate their proliferation and differentiation capacity [35]. Genes of HSC become silenced during proliferation because of the dilution of histone methylation, and this silencing induces age-related phenotypes and reduces the overall the capacity of proliferation and differentiation. Because of the reduced proliferation, histone modification can reactivate silenced genes leading to stem cell rejuvenation. In line with this, Przybilla and coworkers' simulations indicate that an efficient DNA demethylation activity is necessary to enable sufficient plasticity of histone modifications [35]. Overall, the results emerging from the use of a systems biology workflow in the MAGE project suggest that aged cells with low proliferation rate accumulate in the HSC niche, and young active cells are pushed to differentiate, thus keeping the age of differentiated cells young.

mTOR is a central signal transducer of the nutritional cell status and growth signals [37]. In particular, organs with high metabolic activity, like liver, muscles and adipose tissue, and kidney are sensitive to mTOR signaling [37,38]. Kidney diseases are among the leading ailments in the elderly, and the NephAge project was conceived to investigate the involvement of aging in them via a systemic approach (E in fig. 3). One of the starting points is that dynamics of the activation of the protein complex mTORC2 is insufficiently understood. Thus, the NephAge consortium ran an iterative cycle of hypothesis formulation, mathematical modelling analysis and experimental validation in order to understand the mTORC2 activation. Modeling enabled hypothesizing a new mechanism, which could reproduce all the findings. The new mechanism was then validated in subsequent experiments [39].

The GerontoMitoSys project tracks the eluding principles of aging by examining the consequences of protein quality and ROS in fungus, mouse, and man (F in fig. 3). Advanced statistical techniques were employed on transcriptome data of the fungus Podospora anserina in an attempt to identify global expression patterns. The data analysis indicated that the mTOR signaling pathway turned out to be important during aging, which provided protein quality control through autophagy [40].

The GerontoMitoSys project finds that in addition to autophagy, protein quality control is also important to delay detrimental effects of aging [40]. Among the principal agents of protein damage are ROS. In order to further elucidate the role of ROS in the emergence of aging-associated phenotypes, a mathematical model was generated using a set of differential equations describing the time course of ROS, its scavengers like cellular antioxidants (SOD) and a protease involved in protein quality control (F2 in fig. 3). Model simulations indicated that the mechanism by which ROS mediates cellular damage is far more complicated than originally stated by the free radical theory of aging. In the paper, the authors suggest that a systems biology approach, similar to the one deployed in the paper, is the one required for an effective integration of the various pathways known to be involved in the control of biological aging.

Although higher ROS levels contribute to accumulation of damage with age, ROS at lower levels trigger defense mechanisms and even provide life extension [41]. These dose-dependent antagonistic actions of ROS are particularly interesting from a systems biology perspective, though explicit systemic studies have yet to be published. In line with this, two separate endogenous mechanisms of ROS production in Caenorhabditis elegans were recently described by the JenAge initiative (G). One mechanism commences with impaired insulin and IGF-1 signaling [42], the other is rooted in the deacetylation reaction catalyzed by the sirtuin Sir-2.1 [43].

Taken together, we have shown that several projects within the Gerontosys initiative have proved to make significant contributions in aging research applying systems biology approaches that rely on either advanced analysis of high-throughput data or mathematical modelling of aging-relevant signaling pathways. We think that systems biology in aging is particularly challenging because of the multiple aging phenotypes that exist, the large and often poorly characterized regulatory networks involved and the multiple levels of organization cross talking in the emergence of aging. To make the situation more cumbersome, even for aging processes where the involved pathways are fairly well characterized, experimentally measured output may lead to ambiguous results, because of unexpected interferences by other pathways. It is therefore challenging to devise and plan interdisciplinary research projects in the field of aging. To substantiate this claim, in the next section we discuss the concept, development, and challenges of implementing systems biological approaches in a real case study, a project dealing with the interplay between oxidative phosphorylation, ROS production and aging effects in mice that we have developed over the last 4 years.

The ROSage project has been also part of the Gerontosys Initiative. The acronym of the project stands for ‘Reactive Oxygen Species and the Dynamics of Ageing'. The underlying hypothesis of the project, similar to others already mentioned, was that mitochondrial dysfunction and subsequent ROS production are crucial players in the emergence of aging phenotypes via the induction of DNA damage, inflammation and, ultimately, cell senescence and tissue dysfunction, in a probably tissue-dependent manner. The project put emphasis on the role of mutations of the proteins integrated in the mitochondrial respiratory chain complexes as drivers of age-associated dysfunction. Published results indicate that each complex causes diverse but still unique phenotypes in an organ-specific manner, but also different organs seem to have different energy requirements and ROS sensitivities (see table 1).

Table 1

Proteins and protein complexes in the focus of the ROSage research initiative. ROSage used defined mutations of components of the oxidative phosphorylation with phenotypes of clinical significance

Proteins and protein complexes in the focus of the ROSage research initiative. ROSage used defined mutations of components of the oxidative phosphorylation with phenotypes of clinical significance
Proteins and protein complexes in the focus of the ROSage research initiative. ROSage used defined mutations of components of the oxidative phosphorylation with phenotypes of clinical significance

To further substantiate our hypothesis, we conceived a complex systems biology workflow, which combines systematic measurements of intracellular signals and tissue markers of organ function during the life span of mice, advanced data analysis and mathematical modelling of selected signaling pathways.

As experimental system, we used several conplastic mouse strains, which hold defined stable mtDNA mutations in several critical proteins involved in the respiratory chain complexes [44]. Since the mitochondrial respiratory chain complexes carry out oxidative phosphorylation, and its malfunctioning can generate excessive ROS, we expected to generate valuable data on the relation between the overproduction of ROS, the activation of cell damage signaling pathways and the triggering of cell senescence and certain tissue-specific aging phenotypes. To this end and for different organs whose dysfunction is associated with aging, we have measured over the life span of the mice: (a) molecular and functional markers of the mitochondrial function, including ROS production, oxygen consumption, quantification of OXPHOS components, intracellular calcium dynamics, mitochondrial dynamics; (b) ROS-mediated activation of signaling pathways involved in damage response like p53, NFκB and JNK; (c) markers for the triggering of phenotypes of cell cycle arrest, senescence and apoptosis, and (d) tissue-specific markers of tissue function, for example neuronal plasticity and learning behavior to account for the phenotypic effects of mitochondria dysfunction in the brain (fig. 4). Five model systems of organ aging were considered: brain function (neurodegeneration), liver- and pancreatic function (metabolic syndrome), pancreatic function (inflammation), hematopoietic system (autoimmune processes) and skin.

Fig. 4

Regulatory map of the modelling strategy proposed for the ROSage project. The model investigates how enhanced ROS production due to impairment of respiratory chain complexes (I, II, III, IV, V) affects energy homeostasis (a), activates inflammatory responses leading to cellular dysfunction and senescence (b), and induces further impairment of mitochondrial function (c). The interplay between these molecular mechanisms inducing metabolism and inflammation is being further characterized and refined in modules a-c.

Fig. 4

Regulatory map of the modelling strategy proposed for the ROSage project. The model investigates how enhanced ROS production due to impairment of respiratory chain complexes (I, II, III, IV, V) affects energy homeostasis (a), activates inflammatory responses leading to cellular dysfunction and senescence (b), and induces further impairment of mitochondrial function (c). The interplay between these molecular mechanisms inducing metabolism and inflammation is being further characterized and refined in modules a-c.

Close modal

To analyze these data, we designed a systems biology workflow that combined advanced data analysis techniques and mathematical modelling. Concerning data analysis, we have been using unsupervised clustering and machine learning techniques to dissect the three-dimensional data grid, which emerges when measuring the indicated markers over the life span for the different organs and developmental stages in the mice strains with different mtDNA mutations (fig. 5). The aim is to find systemic patterns relating ROS generation across time, organ and mitochondrial gene mutations, as well as their relation to organ function/viability. This analysis was complemented with the derivation and characterization of mathematical models accounting for critical parts of the ROS-related network proposed.

Fig. 5

Systematic measurement of experimental parameters across the dimensions of mitochondrial mutations, organ and developmental stage (age) proposed for the Gerontosys project ROSage. Each dot in the grid represents a set of biomedical insights; the size of the dot corresponds to organ function/viability measured independently for the same mutation, organ and age. Given a similarity measure, a clustering based on modeling insights yields further understanding.

Fig. 5

Systematic measurement of experimental parameters across the dimensions of mitochondrial mutations, organ and developmental stage (age) proposed for the Gerontosys project ROSage. Each dot in the grid represents a set of biomedical insights; the size of the dot corresponds to organ function/viability measured independently for the same mutation, organ and age. Given a similarity measure, a clustering based on modeling insights yields further understanding.

Close modal

Precisely, the network was organized into interconnected modules, accounting for mitochondrial performance, metabolism and inflammation (fig. 4a-c, respectively).

The project is still in progress, and most of the data related to life span-related changes in the mice strains are still to be processed. Nevertheless, preliminary results confer to mutations in complexes I, IV, and V, known to be important contributors to a healthy physiology, clear tissue-dependent effects; for example, only the central neural system (CNS) differences between control mouse strains and those mice carrying a mutation in complex I (NADH:ubiquinone oxidoreductase). The differences are a higher ROS level in early life, which reduced to normal level from middle age onwards. Also during middle age, mutant mouse strains displayed reduced plasticity and learning performance.

Mutations in complex IV (cytochrome C oxidase) reduced learning performance in the CNS as well, but the effects were only apparent late in life (24 months). However, no differences in ROS levels were detectable. In addition to the central nervous system, complex IV profoundly affected the liver. By the age of 6 months, mitochondria start to aggregate and ROS as well as antioxidants increase.

Mutations in complex V (ATP synthase) affected the hematopoietic system and the endocrine pancreas. The ROS level was increased in the β-cells of the endocrine pancreas, whereas the ATP level was decreased, although the ATP/ADP ratio was comparable to the wild type [45]. In a high-fat diet condition, the mitochondria grow in size in β-cells of the wild type, but not in complex V mutants. The mutants were also subjected to reductions in insulin secretion and insulin sensitivity along with higher serum insulin concentrations. To counterbalance the loss in insulin secretion due to the lower absolute ATP level, complex V mutants experienced an increase in the total β-cell mass [45]. In the hematopoietic systems of aged mice, the effect of mutations in complex V was surprisingly different - ROS was reduced and ATP increased. ROS reduction and ATP increase could be the result of a metabolic switch to energy conservation in a background with ATP synthase mutations.

Taken together, our preliminary results indicate that (a) overall, organs and tissues are surprisingly robust regarding the mtDNA mutations investigated, and (b) the effect of mutations is tissue-dependent. The phenotypes of mutants are often comparable to the wild type in conventional rearing conditions over most of the mouse life span. This resilience to genetic perturbation highlights the extraordinary capacity of the OXPHOS system and the metabolism in general to balance energy demands. In particular, the mutations of complex III and UCP2 are buffered well because for conventional treatment, physiological conditions and ROS level were comparable in the four examined organs for mutant and wild-type strains.

As mentioned before, in the project, the network was organized into interconnected modules, accounting for mitochondrial performance, metabolism and inflammation (fig. 4). We have derived mathematical models for these modules and used them to investigate critical features of the modules associated with aging. For example, combining modeling and experiments, we demonstrated that miRNAs, whose deregulation has been related to aging [46], can modulate, alone or synergistically, the activation of p21 by signals activating the DNA damage response like ROS [9,47]. Our model simulations indicate that miRNAs can posttranscriptionally regulate p21 basal levels in a biological context-dependent manner. Since p21 has been recently involved together with ROS in a feedback loop system able to trigger cell senescence upon a sufficiently long-lasting DNA damage response [21], one can expect miRNA regulation to play a role in this process.

Although this is a promising result, most of the modeling effort in the project has focused on the relation between ROS and metabolic performance (fig. 4b). Despite glucose homeostasis being so crucial for long-term metabolic consequences, there are only a few models which deal with the long-term changes during the life span [48]. The paucity of long-term models is due to several challenges. One of the formidable challenges is that glucose homeostasis involves multiple levels of structural and functional organization, from molecular reactions and cell-cell interactions in tissues like pancreas and liver to the physiology of the entire organs. Figure 6 depicts the involvement of several tissues and organs in carrying out the glucose insulin metabolism in response to nutrient stress, during the life span of a mouse. Although disperse, unconnected mathematical models exist in the literature at all levels of organization [49]; the real challenge here is to integrate multiple levels of data, models and knowledge into a comprehensive multi-level model able to interpret data on the evolution of glucose homeostasis over life span. Multi-levelness and integration of knowledge can be addressed through multi-scale modeling, a topic strongly developed in the last years in cancer biology, but still in its infancy for most of the other biomedical fields, including aging research.

Fig. 6

Multiple levels of structural and functional organization of nutrient-induced glucose metabolism during aging. Glucose is produced (mainly by liver), distributed and utilized by pancreas-induced insulin and adipose tissue-induced insulin and leptin. High-fat nutrient results in higher energy intake, which increases fat accumulation in adipose tissues. Increased fat mass is sensed by the brain, which then signals an increase in leptin and insulin levels in blood for a more rigorous utilization of glucose by other tissues so as to bring glucose levels to levels prior to perturbation. With our aging mice strains, we are deriving and characterizing a mathematical model of this system, which describes the role played by systemic signals like leptin and insulin in modulating cross talk between the organs involved in glucose metabolism and the overall regulation of the glucose metabolism. Interestingly, the system is enriched in insulin- and leptin-mediated feedback loops, which we think control the nonlinear behavior in the blood glucose levels of several strains of mice used in our project (negative feedback loops: glucose → insulin ┤ glucose; glucose → leptin ┤ glucose).

Fig. 6

Multiple levels of structural and functional organization of nutrient-induced glucose metabolism during aging. Glucose is produced (mainly by liver), distributed and utilized by pancreas-induced insulin and adipose tissue-induced insulin and leptin. High-fat nutrient results in higher energy intake, which increases fat accumulation in adipose tissues. Increased fat mass is sensed by the brain, which then signals an increase in leptin and insulin levels in blood for a more rigorous utilization of glucose by other tissues so as to bring glucose levels to levels prior to perturbation. With our aging mice strains, we are deriving and characterizing a mathematical model of this system, which describes the role played by systemic signals like leptin and insulin in modulating cross talk between the organs involved in glucose metabolism and the overall regulation of the glucose metabolism. Interestingly, the system is enriched in insulin- and leptin-mediated feedback loops, which we think control the nonlinear behavior in the blood glucose levels of several strains of mice used in our project (negative feedback loops: glucose → insulin ┤ glucose; glucose → leptin ┤ glucose).

Close modal

In line with the idea and in the context of the ROSage project, we have been analyzing the relation between glucose metabolism and mitochondria performance in aging by deploying a cascade of models describing different levels of organization. At the physiological level (fig. 6), we have developed a phenomenological mesoscopic mathematical model, which accounts for the interplay between different critical organs (liver, pancreas, adipose tissue, brain) via molecules that act in this context as systemic signals, secreted by these organs and used to mediate their cross talk and the overall regulation of the glucose metabolism (insulin, leptin). The mathematical modelling has been used to establish the role played by several insulin- and leptin-meditated feedback loops in shaping nonlinear behavior in the glucose blood levels of several of our aging mouse strains.

In parallel to this, we built upon one of the assumptions of damage accumulation in the mtDNA during the life span and their effect on the mitochondrial performance through its effect in the balance between the antagonizing forces of mitochondrial fission and fusion [50] Mitochondria organize themselves as dynamic populations within a cell, by undergoing continuous cycles of fission and fusion. In several neurodegenerative and metabolic diseases, the dynamic balance of fission and fusion is disturbed. A diversified mitochondrial population is generated by the heterogeneous availability of nutrients and signals in cellular cross-section [51]. Segregation through fission increases local sensitivity of mitochondria towards signals, whereas fusion interconnects mitochondrial networks from one location to another, hence increasing their global sensitivity. One can say that global coupling of excitatory and restoring mitochondrial subpopulations in a heterogeneous environment enables the system to selectively adapt the response at one location, while amplifying the response at another location.

We were interested in establishing how the equilibrium between fission and fusion is altered over the life span of the investigated mouse strains. To this end, we established a mathematical model of antagonistic fission and fusion subpopulations of mitochondria by adapting the model proposed by Wallach and coworkers to investigate spatially heterogeneous neuronal systems [52].

Fission and fusion subpopulations encounter heterogeneous signals in the peripheral (high frequency signal) and perinuclear (low frequency signal) cellular locations [51]. In our mathematical model, we divide the mitochondrial subpopulations into three: peripheral fission, perinuclear fission and global fusion subpopulations (see fig. 7a for graphical depiction of the model and explanation of the mathematical equations). Two fission subpopulations interact antagonistically with the global fusion subpopulation (fig. 7a). Model simulations and analysis indicate that global coupling of mitochondrial subpopulations in a heterogeneous environment enables the system to selectively adapt the response at one location while amplifying the response at another location (fig. 7b). In figure 7c, a model simulation is shown. Here, we can see that adaptation emerges in the response of the system to low scenarios of variability between the frequency of peripheral and perinuclear stimuli driving fission-fusion cycles (δ = 1-3). In contrast, amplification appears for scenarios of high variability (δ = 5-10).

Fig. 7

a Our model of mitochondria performance includes two antagonistic subpopulations of dividing (fission) and fusing mitochondria, which are globally coupled through fusion in the heterogeneous cellular environment. At the peripheral locations, high-frequency stimuli are prevalent, while at the perinuclear location low-frequency stimuli are prevalent. In our mathematical model, we divide the mitochondrial subpopulations into three: peripheral fission, perinuclear fission and global fusion subpopulations. Two fission subpopulations interact antagonistically with global fusion subpopulation. The three subpopulations are represented by three equations of following type: P denotes the availability of respective subpopulation: peripheral fission, perinuclear fission or global fusion. The availability of a subpopulation depends on the time an average mitochondrial subpopulation takes to recover, and on the depletion of resources in maintaining the activity (Ψ) of respective average subpopulation. Activity is a function of availability of subpopulations (P) and external stimulation (f), such that fission subpopulations are antagonistic to the global fusion subpopulation. τ represents the recovery time constant. Next, mitochondrial response is defined as the ratio of activity upon stimulation with a certain frequency (f) and activity without stimulation (0). In the model, we introduce a parameter δ to describe the degree of variability between the peripheral (fhigh) and perinuclear (flow). δ is defined by a ratio of global frequency fglobal (fhigh + flow) and flow. fglobal is the frequency with which fusion population is stimulated as a result of global coupling of stimuli: b Schematic representation of two types of stimuli (flow, fhigh) and their responses. c Model simulations. Mitochondrial Ca2+ levels adapt (b and δ = 1-3 in c) to high-frequency stimuli, whereas in response to low-frequency stimulation mitochondrial Ca2+ levels amplify (b and δ = 5-10 in c).

Fig. 7

a Our model of mitochondria performance includes two antagonistic subpopulations of dividing (fission) and fusing mitochondria, which are globally coupled through fusion in the heterogeneous cellular environment. At the peripheral locations, high-frequency stimuli are prevalent, while at the perinuclear location low-frequency stimuli are prevalent. In our mathematical model, we divide the mitochondrial subpopulations into three: peripheral fission, perinuclear fission and global fusion subpopulations. Two fission subpopulations interact antagonistically with global fusion subpopulation. The three subpopulations are represented by three equations of following type: P denotes the availability of respective subpopulation: peripheral fission, perinuclear fission or global fusion. The availability of a subpopulation depends on the time an average mitochondrial subpopulation takes to recover, and on the depletion of resources in maintaining the activity (Ψ) of respective average subpopulation. Activity is a function of availability of subpopulations (P) and external stimulation (f), such that fission subpopulations are antagonistic to the global fusion subpopulation. τ represents the recovery time constant. Next, mitochondrial response is defined as the ratio of activity upon stimulation with a certain frequency (f) and activity without stimulation (0). In the model, we introduce a parameter δ to describe the degree of variability between the peripheral (fhigh) and perinuclear (flow). δ is defined by a ratio of global frequency fglobal (fhigh + flow) and flow. fglobal is the frequency with which fusion population is stimulated as a result of global coupling of stimuli: b Schematic representation of two types of stimuli (flow, fhigh) and their responses. c Model simulations. Mitochondrial Ca2+ levels adapt (b and δ = 1-3 in c) to high-frequency stimuli, whereas in response to low-frequency stimulation mitochondrial Ca2+ levels amplify (b and δ = 5-10 in c).

Close modal

graphic

graphic

graphic

At the moment, we are elaborating a strategy to design experiments able to test our model predictions. Furthermore, we are expanding our model to account for the connection between the dynamics of mitochondrial subpopulations and the response of critical cell/tissues to environmental perturbations on the longer timescales that correspond to the emergence of aging phenotypes.

In this chapter, we support the idea that boosting aging research will require the use of systems biology-inspired approaches because (a) the dysfunction of large, interconnected biochemical networks is in the origin of aging-associated phenotypes; (b) these networks are enriched in nonlinear regulatory motifs like positive and negative feedback loops; (c) aging manifests at multiple, interconnected and interdependent levels of organization in the body, from the intracellular machinery to the dynamics of tissue organization and beyond, and (d) the optimal design of biomedical strategies to counteract aging-associated pathologies will require the use of tools and strategies adapted from engineering.

The specific problems that one will face when designing a systems biology project in aging are not minor. The difficulty to generate reliable experimental data for the investigation of aging phenotypes increases when thinking from a systemic view because one will require more systematic and frequent and better quantifiable experimental measurements, but also an increase in the number of experimental replicates. The strategy to overcome the difficulties of integrating in a coherent manner quantitative data produced at multiple levels of organization to measure the emergence of aging phenotypes is still an open question. In line with this, the strategy to adapt the multi-scale methodologies used in other biomedical fields like cancer is still to be established. The question of establishing reliable methodologies for actual quantification of aging phenotypes at the intracellular, tissue and organ levels is open to debate. Finally, large and long-lasting collaborative systems biology projects like those addressing the molecular basis of aging can only prosper when a good data management strategy is designed prior to project initiation.

This work was supported by the German Federal Ministry of Education and Research as part of the project Gerontosys-ROSage (0315892A).

1.
Vera J, Wolkenhauer O: A system biology approach to understand functional activity of cell communication systems. Methods Cell Biol 2008;90:399-415.
2.
Vera J, Rath O, Balsa-Canto E, Banga JR, Kolch W, Wolkenhauer O: Investigating dynamics of inhibitory and feedback loops in ERK signalling using power-law models. Mol Biosyst 2010;6:2174-2191.
3.
Schoeberl B, Pace EA, Fitzgerald JB, Harms BD, Xu L, Nie L, Linggi B, Kalra A, Paragas V, Bukhalid R, Grantcharova V, Kohli N, West KA, Leszczyniecka M, Feldhaus MJ, Kudla AJ, Nielsen UB: Therapeutically targeting ErbB3: a key node in ligand-induced activation of the ErbB receptor-PI3K axis. Sci Signal 2009;2:ra31.
4.
Vera J, Schmitz U, Lai X, Engelmann D, Khan FM, Wolkenhauer O, Pützer BM: Kinetic modeling-based detection of genetic signatures that provide chemoresistance via the E2F1-p73/DNp73-miR-205 network. Cancer Res 2013;73:3511-3524.
5.
Ptitsyn AA, Weil MM, Thamm DH: Systems biology approach to identification of biomarkers for metastatic progression in cancer. BMC Bioinformatics 2008;9(suppl 9):S8.
6.
Sara H, Kallioniemi O, Nees M: A decade of cancer gene profiling: from molecular portraits to molecular function. Methods Mol Biol 2010;576:61-87.
7.
Vera J, Lai X, Schmitz U, Wolkenhauer O: MicroRNA-regulated networks: the perfect storm for classical molecular biology, the ideal scenario for systems biology. Adv Exp Med Biol 2013;774:55-76.
8.
Meyer P, Cokelaer T, Chandran D, Kim KH, Loh P-R, Tucker G, Lipson M, Berger B, Kreutz C, Raue A, Steiert B, Timmer J, Bilal E, Sauro HM, Stolovitzky G, Saez-Rodriguez J: Network topology and parameter estimation: from experimental design methods to gene regulatory network kinetics using a community based approach. BMC Systems Biol 2014;8:13.
9.
Lai X, Schmitz U, Gupta SK, Bhattacharya A, Kunz M, Wolkenhauer O, Vera J: Computational analysis of target hub gene repression regulated by multiple and cooperative miRNAs. Nucleic Acids Res 2012;40:8818-8834.
10.
Tyson JJ, Chen KC, Novak B: Sniffers, buzzers, toggles and blinkers: dynamics of regulatory and signaling pathways in the cell. Curr Opin Cell Biol 2003;15:221-231.
11.
Nikolov S, Wolkenhauer O, Vera J: Tumors as chaotic attractors. Mol Biosyst 2014;10:172-179.
12.
Aldridge BB, Burke JM, Lauffenburger DA, Sorger PK: Physicochemical modelling of cell signalling pathways. Nat Cell Biol 2006;8:1195-1203.
13.
Kreeger PK, Lauffenburger DA: Cancer systems biology: a network modeling perspective. Carcinogenesis 2010;31:2-8.
14.
Byrne HM: Dissecting cancer through mathematics: from the cell to the animal model. Nat Rev Cancer 2010;10:221-230.
15.
Engel C, Scholz M, Loeffler M: A computational model of human granulopoiesis to simulate the hematotoxic effects of multicycle polychemotherapy. Blood 2004;104:2323-2331.
16.
Kirkwood TB, Kowald A: Network theory of aging. Exp Gerontol 1997;32:395-399.
17.
Kowald A, Kirkwood TB: Towards a network theory of ageing: a model combining the free radical theory and the protein error theory. J Theor Biol 1994;168:75-94.
18.
Franceschi C, Bonafè M, Valensin S, Olivieri F, De Luca M, Ottaviani E, De Benedictis G: Inflamm-aging. An evolutionary perspective on immunosenescence. Ann N Y Acad Sci 2000;908:244-254.
19.
de Magalhães JP, Toussaint O: GenAge: a genomic and proteomic network map of human ageing. FEBS Lett 2004;571:243-247.
20.
Xue H, Xian B, Dong D, Xia K, Zhu S, Zhang Z, Hou L, Zhang Q, Zhang Y, Han J-DJ: A modular network model of aging. Mol Syst Biol 2007;3:147.
21.
Passos JF, Nelson G, Wang C, Richter T, Simillion C, Proctor CJ, Miwa S, Olijslagers S, Hallinan J, Wipat A, Saretzki G, Rudolph KL, Kirkwood TB, von Zglinicki T: Feedback between p21 and reactive oxygen production is necessary for cell senescence. Mol Syst Biol 2010;6:347.
22.
Kriete A, Bosl WJ, Booker G: Rule-based cell systems model of aging using feedback loop motifs mediated by stress responses. PLoS Comput Biol 2010;6:e1000820.
23.
Kitano H: Biological robustness. Nat Rev Genet 2004;5:826-837.
24.
Kirkwood TB: Understanding the odd science of aging. Cell 2005;120:437-447.
25.
Kirkwood TB: A systematic look at an old problem. Nature 2008;451:644-647.
26.
Passarino G, Rose G, Bellizzi D: Mitochondrial function, mitochondrial DNA and ageing: a reappraisal.' Biogerontology 2010;11:575-588.
27.
Hoehme S, Brulport M, Bauer A, Bedawy E, Schormann W, Hermes M, Puppe V, Gebhardt R, Zellmer S, Schwarz M, Bockamp E, Timmel T, Hengstler JG, Drasdo D: Prediction and validation of cell alignment along microvessels as order principle to restore tissue architecture in liver regeneration. Proc Natl Acad Sci USA 2010;107:10371-10376.
28.
van Leeuwen IM, Vera J, Wolkenhauer O: Dynamic energy budget approaches for modelling organismal ageing. Philos Trans R Soc Lond B Biol Sci 2010;365:3443-3454.
29.
van Leeuwen IM, Kelpin FD, Kooijman SA: A mathematical model that accounts for the effects of caloric restriction on body weight and longevity. Biogerontology 2002;3:373-381.
30.
McAuley MT, Wilkinson DJ, Jones JJ, Kirkwood TB: A whole-body mathematical model of cholesterol metabolism and its age-associated dysregulation. BMC Syst Biol 2012;6:130.
31.
De Grey AD, Rae M: Ending Aging: The Rejuvenation Breakthroughs That Could Reverse Human Aging in Our Lifetime. New York, St. Martin's Griffin, 2008.
32.
De Grey AD, Ames BN, Andersen JK, Bartke A, Campisi J, Heward CB, McCarter RJM, Stock G: Time to talk SENS: critiquing the immutability of human aging. Ann N Y Acad Sci 2002;959:452-462.
33.
Wang J, Sun Q, Morita Y, Jiang H, Gross A, Lechel A, Hildner K, Guachalla LM, Gompf A, Hartmann D, Schambach A, Wuestefeld T, Dauch D, Schrezenmeier H, Hofmann W-K, Nakauchi H, Ju Z, Kestler HA, Zender L, Rudolph KL: A differentiation checkpoint limits hematopoietic stem cell self-renewal in response to DNA damage. Cell 2012;148:1001-1014.
34.
Przybilla J, Galle J, Rohlf T: Is adult stem cell aging driven by conflicting modes of chromatin remodeling? Bioessays 2012;34:841-848.
35.
Przybilla J, Rohlf T, Loeffler M, Galle J: Understanding epigenetic changes in aging stem cells - a computational model approach. Aging Cell 2014;13:320-328.
36.
Dodd IB, Micheelsen MA, Sneppen K, Thon G: Theoretical analysis of epigenetic cell memory by nucleosome modification. Cell 2007;129:813-822.
37.
Cornu M, Albert V, Hall MN: mTOR in aging, metabolism, and cancer. Curr Opin Genet Dev 2013;23:53-62.
38.
Grahammer F, Wanner N, Huber TB: mTOR controls kidney epithelia in health and disease. Nephrol Dial Transplant 2014;29(suppl 1):i9-i18.
39.
Dalle Pezze P, Sonntag AG, Thien A, Prentzell MT, Gödel M, Fischer S, Neumann-Haefelin E, Huber TB, Baumeister R, Shanley DP, Thedieck K: A dynamic network model of mTOR signaling reveals TSC-independent mTORC2 regulation. Sci Signal 2012;5:ra25.
40.
Philipp O, Hamann A, Servos J, Werner A, Koch I, Osiewacz HD: A genome-wide longitudinal transcriptome analysis of the aging model podospora anserine. PLoS One 2013;8:e83109.
41.
Wink DA, Hines HB, Cheng RYS, Switzer CH, Flores-Santana W, Vitek MP, Ridnour LA, Colton CA: Nitric oxide and redox mechanisms in the immune response. J Leukoc Biol 2011;89:873-891.
42.
Zarse K, Schmeisser S, Groth M, Priebe S, Beuster G, Kuhlow D, Guthke R, Platzer M, Kahn CR, Ristow M: Impaired insulin/IGF1 signaling extends life span by promoting mitochondrial L-proline catabolism to induce a transient ROS signal. Cell Metab 2012;15:451-465.
43.
Schmeisser K, Mansfeld J, Kuhlow D, Weimer S, Priebe S, Heiland I, Birringer M, Groth M, Segref A, Kanfi Y, Price NL, Schmeisser S, Schuster S, Pfeiffer AFH, Guthke R, Platzer M, Hoppe T, Cohen HY, Zarse K, Sinclair DA, Ristow M: Role of sirtuins in lifespan regulation is linked to methylation of nicotinamide. Nat Chem Biol 2013;9:693-700.
44.
Yu X, Gimsa U, Wester-Rosenlof L, Kanitz E, Otten W, Kunz M, Ibrahim SM: Dissecting the effects of mtDNA variations on complex traits using mouse conplastic strains. Genome Res 2008;19:159-165.
45.
Weiss H, Wester-Rosenloef L, Koch C, Koch F, Baltrusch S, Tiedge M, Ibrahim S: The mitochondrial Atp8 mutation induces mitochondrial ROS generation, secretory dysfunction, and β-cell mass adaptation in conplastic B6-mtFVB mice. Endocrinology 2012;153:4666-4676.
46.
Smith-Vikos T, Slack FJ: MicroRNAs and their roles in aging. J Cell Sci 2012;125:7-17.
47.
Lai X, Bhattacharya A, Schmitz U, Kunz M, Vera J, Wolkenhauer O: A systems biology approach to study microRNA-mediated gene regulatory networks. Biomed Res Int 2013;703849:1-15.
48.
Goldbeter A: A model for the dynamics of human weight cycling. J Biosci 2006;31:129-136.
49.
Wolkenhauer O: The role of theory and modeling in medical research. Front Physiol 2013;4:377.
50.
Chauhan A, Vera J, Wolkenhauer O: The systems biology of mitochondrial fission and fusion and implications for disease and aging. Biogerontology 2014;15:1-12.
51.
Collins TJ, Berridge MJ, Lipp P, Bootman MD: Mitochondria are morphologically and functionally heterogeneous within cells. EMBO J 2002;21:1616-1627.
52.
Wallach A, Eytan D, Marom S, Meir R: Selective adaptation in networks of heterogeneous populations: model, simulation, and experiment. PLoS Comput Biol 2008;4:e29.

A.C., U.W.L. and J.V. contributed equally to this work.

Send Email

Recipient(s) will receive an email with a link to 'Aging and Health - A Systems Biology Perspective > 155 - 176: Systems Biology Approaches in Aging Research' and will not need an account to access the content.

Subject: Aging and Health - A Systems Biology Perspective > 155 - 176: Systems Biology Approaches in Aging Research

(Optional message may have a maximum of 1000 characters.)

×
Close Modal

or Create an Account

Close Modal
Close Modal