Toward a Systems Biology Framework for Understanding Aging and Health Span

Abstract

It is argued that aging research is at a stage where it could benefit greatly from a more intense engagement with the perspectives emphasized by systems biology and complexity science. A more integrated, systematic approach is needed if we are ever to have a fully developed, fundamental understanding of aging, longevity, and their relationship to health. A broader, deeper, more quantitative, and predictive conceptual framework can lead to theoretical approaches and realistic models that can be quantitatively confronted with data and, perhaps more importantly, stimulate novel questions and novel experiments. Integral to this is the search for underlying causal multilevel mechanisms and principles that can be quantified and developed into a serious predictive theoretical framework, providing a point of departure for framing a more integrated research agenda.

THE aging of an organism is both a manifestation and a result of complex changes in structure and function across all levels of biological organization from molecules and cells to tissues and whole body systems. However, almost all investigations have typically been carried out at a specific single level of organization and often in isolation with a relatively narrow focus. Indeed, this strategy reflects the more highly focused, reductionistic methodology and culture that dominates biological and biomedical research across the entire spectrum of research activities. Although this traditional strategy has unquestionably led to significant progress and remarkable insights, it is nevertheless becoming clearer to many researchers that a more integrated approach is needed if we are ever to have a fully developed, fundamental understanding of aging and longevity and their relationship to health (1–6).

A shift in focus from a relatively narrow perspective to a broader vision has been gaining recognition in many areas of biology over the past several years. The call has come from many quarters for bigger picture, more systematic, and systemic approaches that integrate across multiple scales of structure and organization. A prominent manifestation of this trend is the emergence of the burgeoning field of systems biology (7–9). To a large degree, this was developed as a response to the extraordinary proliferation of multi-omics data sets. As genome projects and other users of high-throughput technologies began to generate data at a dizzying pace, researchers have sought out computational tools for organizing, sifting through, and presenting large amounts of biological information (4,10). Although this is clearly of great importance, it represents a narrow view of what the vision of a systems biology approach can, and hopefully will, bring to the quest for a deeper understanding of biological phenomena and their relationship to health and medicine.

Although the amassing of large databases manifesting ill-understood phenomena at multiple scales was a major driving force for establishing systems biology, many researchers are beginning to view systems biology as a paradigm for developing a broader, deeper conceptual framework for integrating the proliferation of such multi-omics data across multiple scales. This view recognizes that the multiple levels of organization ranging from molecular and subcellular scales through the organismal, ecological, and societal to the evolutionary are not independent, decoupled systems and that there is a compelling need to provide a systematic framework for understanding the underlying science connecting various levels (11–14). Implicit in this is the challenge of developing a quantitative, analytic, predictive approach for understanding many of the fundamental problems in biology. A broader, deeper conceptual framework can lead to theoretical approaches and realistic models that can be quantitatively confronted with data and, perhaps more importantly, stimulate novel questions and novel experiments. Integral to this is the search for underlying causal mechanisms and principles that can be quantified and developed into a serious predictive theoretical framework, providing a point of departure for framing a more integrated research agenda (15,16).

Given that aging and mortality are ubiquitously manifested across the entire spectrum of life from the cellular level to the ecosystem, it is both natural and compelling that many of the ideas, concepts, techniques, and ways of thinking that fall under the umbrella of systems biology and complexity science should begin to play an important role in developing a serious, predictive theory of aging. There are few more fundamental and critical problems in biology that are better positioned to benefit from this than the challenge of aging and death. Furthermore, without such a theoretical framework it will be difficult to develop broad, credible strategies for addressing the multiple challenges of human life span and health span and their societal consequences.

An important component of this new paradigm is the recognition that developing such a framework requires a more multi- and transdisciplinary approach where ideas, concepts, and ways of thinking derived from nontraditional areas can potentially have a major impact on thinking about aging. It seems clear that aging research can benefit from ideas in computer science for organizing large databases and for extracting underlying commonalities and phenomenological “laws.” Similarly, ideas from mathematics, physics, and engineering regarding the robustness and resilience of networks, questions of vulnerability and adaptation, dissipation of energy and resources and their manifestation in wear and tear phenomena, repair mechanisms, and error correction strategies all have potentially important consequences for understanding healthy aging.

We, as organisms, are complex systems that operate far from equilibrium using energy and resources in a highly efficient way to combat the concomitant fight against the inevitable increase in entropy. As we begin to lose the multiple localized battles against entropy we age, and when we ultimately lose the war we die (17). It is therefore reasonable to conjecture that the highly developed sciences of thermodynamics, statistical physics, and information theory will provide part of the underlying framework for understanding aging. Biologically, this means that a fundamental theory of aging will very likely incorporate the integration of the dynamics of metabolic networks with those of genomic networks. As such, network theory, which has been a fast developing field in recent years, will play a central role (16,18,19). Understanding the robustness of such networks and where the weakest nodes and links reside could be key to providing the framework for healthy aging.

More generally, much can be potentially learned from the study of complex adaptive systems, which have received a great deal of attention in recent years. Indeed, systems biology has much in common with complexity science. For example, systems biology aims at understanding how the higher level behavior of complex biological systems emerge from interactions among individual components across multiple temporal, spatial, and functional scales in their genetic, epigenetic, and environmental contexts. As with any complex system, it recognizes that the behavior of the system as a whole is not simply the sum of its individual components.

To some extent cross-disciplinary collaborations already take place in many areas of biology, and in particular in several areas of aging research where investigators use tools and models that originate in other fields. Going further afield, in dealing with highly complex phenotypes, basic aging research also touches upon aspects of demography, the behavioral sciences, and the economic sciences. However, these kinds of collaborations tend to use borrowed techniques to solve very specific, highly localized problems, rather than in the more expansive way implied by bigger picture thinking.

The more quantitative, conceptual framework we are seeking can be illustrated by asking questions not only about mechanisms of biological aging that ultimately determine life span, but to add to that questions like can we explain, for example, why the scale of human life is of the order of 100 years rather than thousands of years or just a few months, and why mice live for only 2–3 years, even though they are made of essentially the same basic cell types and same tissue? Where in the complex molecular structures of genes and respiratory enzymes, which operate at microscopic timescales, can we locate timescales of tens of years? Why is it that life span and the rate of aging, like the timing of almost all life-history events, scale with the quarter power of body mass? Even more intriguing, why is it that both the total lifetime energy needed to sustain a unit mass of tissue and the total number of heartbeats in a lifetime are approximate invariants (20)? More fundamentally, why is it that the number of turnovers of cytochrome oxidase molecules in mitochondria during a lifetime is also an approximate invariant across all aerobic organisms (15)? Such questions and “coarse-grained” observations provide encouraging hints for believing that a generic fundamental dynamic theory of longevity and senescence can be constructed.

Because age-related alterations progress in a system-wide fashion, stronger consideration must be given to the networks involved. This view shifts the focus from studying single genes and proteins toward gene regulation pathways and protein–protein networks as a whole. The exploration of such networks is the primary goal of computational systems biology by developing mechanistic models of gene regulatory, metabolic, signaling, or DNA repair pathways. Rather than describing properties of a list of individual parts, model computer simulations permit investigations of the dynamics and properties of these networks. Moreover, distinct processes across levels of biological organization can be studied using multiscale models. Such models, which describe specific phenomena at an appropriate resolution in space and time, can be integrated together to investigate more complex phenomena (8,13,21).

Building a conceptual framework for developing systems-level modeling depends upon the resolution, or granularity, appropriate for the kinds of questions being asked and the kinds of experiments that can be conducted. Furthermore, the modeling process can be either bottom-up or top-down. Bottom-up models require precise knowledge of the individual parts that will be linked together in network representations such as pathways. The top-down view aims to dissect available observations into meaningful modular entities. Because a current challenge for aging research is to integrate knowledge from fundamental physiological properties of different species as they age with data from molecular and cellular investigation, a more comprehensive framework that accommodates both approaches to study the complex processes of aging will be needed (5).

This approach is effective in associating genotypes with phenotypes. As already indicated, the rise of systems biology can be viewed as a response to the challenge that came into view through systematic biology. Namely, to provide a mechanistic, causal, and thus explanatory linkage between genotypes and the associated phenotypes by studying the structure and dynamics of entangled molecular interaction networks that operate cells and tissues in development and adulthood. Systems biology aims at developing and integrating experimental and mathematical techniques in the pursuit of principles that would make the nature of cellular phenotypes more intelligible and their control more deliberate. This pursuit is driven by the practical need to cure disease. However, it also reflects a desire for a theoretical perspective needed to disentangle the complexity of the cell and the organism. Aging research has made great strides with genetic screens and with bringing molecular cell biology to bear on specific processes and events implicated in aging. As such, aging research could now benefit from a more intense engagement with the perspectives emphasized by systems biology, specifically:

• Understanding how aging manifests itself in and/or is causally connected to dynamic aspects of molecular networks underlying metabolism, signaling, cellular organization, and fundamental integrated physiological processes.

• Cultivating mathematical approaches that inject model-based reasoning about complex systems into the daily practice of aging research. Regardless of whether models are stylized heuristic vignettes or representations attentive to empirical detail, models are never right or wrong, but only useful or useless in making reasoning both explicit and quantitative, and therefore open to constructive criticism.

Successful results using a systems biological approach to elucidate human longevity, which may lead to healthy aging, have already been achieved. For example, in a recent study, an unrealized potential to understand the genetic basis of aging in humans was explored. In this study, the investigators considered the immense survival advantage of the rare individuals who live 100 years or more. This Longevity Gene Study was initiated in 1998 at the Albert Einstein College of Medicine to investigate longevity genes in a selected population: the “oldest old” Ashkenazi Jews, 95 years of age and older, and their children. The study proved the principle that some of these participants are endowed with longevity-promoting genotypes. The investigators reasoned that some of the favorable genotypes act as mechanisms that buffer the deleterious effect of age-related disease genes (22). As a result, the frequency of deleterious genotypes may increase among individuals with extreme life span because their protective genotype allows disease-related genes to accumulate. Thus, studies of genotypic frequencies among different age groups can elucidate the genetic determinants and pathways responsible for longevity (23,24).

Borrowing from evolutionary theory and using systems biological analysis, the investigators argued that differential survival via buffering mechanisms can reveal target age-related disease genes as well as buffering longevity genes. Using more than 1,200 participants between the 6th and 11th decade of life (at least 140 participants in each group), these hypotheses were experimentally corroborated. Genome-wide, high-throughput hypothesis-free analyses are now being utilized to elucidate unknown genetic pathways in many model organisms, linking observed phenotypes to their underlying genetic mechanisms. The longevity phenotype and its genetic mechanisms, such as the buffering hypothesis, are similar; thus the experimental corroboration of these hypotheses demonstrate the utility of high-throughput methods and systems-level approaches for elucidating such complex mechanisms. It also provides a framework for developing strategies to prevent some age-related diseases by intervention at the appropriate level.