Molecular Basis of Nutrition and Aging: A Volume in the Molecular Nutrition Series

Molecular Basis of Nutrition and Aging: A Volume in the Molecular Nutrition Series focuses on the nutritional issues associated with aging and the important metabolic consequences of diet, nutrition, and health. The book is subdivided into four parts that reflect the impact of nutrition from a biomolecular level to individual health.

In Part One, chapters explore the general aspects of aging, aging phenotypes, and relevant aspects of nutrition related to the elderly and healthy aging. Part Two includes molecular and cellular targets of nutrition in aging, with chapters exploring lipid peroxidation, inflammaging, anabolic and catabolic signaling, epigenetics, DNA damage and repair, redox homeostasis, and insulin sensitivity, among others.

Part Three looks at system-level and organ targets of nutrition in aging, including a variety of tissues, systems, and diseases, such as immune function, the cardiovascular system, the brain and dementia, muscle, bone, lung, and many others. Finally, Part Four focuses on the health effects of specific dietary compounds and dietary interventions in aging, including vitamin D, retinol, curcumin, folate, iron, potassium, calcium, magnesium, zinc, copper, selenium, iodine, vitamin B, fish oil, vitamin E, resveratrol, polyphenols, vegetables, and fruit, as well as the current nutritional recommendations.

• Offers updated information and a perspectives on important future developments to different professionals involved in the basic and clinical research on all major nutritional aspects of aging
• Explores how nutritional factors are involved in the pathogenesis of aging across body systems
• Investigates the molecular and genetic basis of aging and cellular senescence through the lens of the rapidly evolving field of molecular nutrition

• Language: English
• Publisher: Academic Press
• Release date: Apr 15, 2016
• ISBN: 9780128018279

Book preview

possible.

Part I

Introductory Aspects on Aging and Nutrition

Outline

Chapter 1 Molecular and Cellular Basis of Aging

Chapter 2 Unraveling Stochastic Aging Processes in Mouse Liver: Dissecting Biological from Chronological Age

Chapter 3 Nutrigenomics and Nutrigenetics: The Basis of Molecular Nutrition

Chapter 4 Diet and Longevity Phenotype

Chapter 5 Nutrition in the Elderly: General Aspects

Chapter 6 Nutrition in the Hospitalized Elderly

Chapter 7 Drug–Nutrient Interactions in the Elderly

Chapter 8 Nutritional Biomarkers of Aging

Chapter 9 Food Preferences in the Elderly: Molecular Basis

Chapter 1

Molecular and Cellular Basis of Aging

Suresh I.S. Rattan,    Laboratory of Cellular Ageing, Department of Molecular Biology and Genetics, Aarhus University, Aarhus, Denmark

Abstract

Aging at the molecular and cellular levels is characterized by the occurrence and accumulation of damage in DNA, RNA, proteins, and other macromolecules. Increased molecular heterogeneity is the fundamental basis for the cellular and physiological changes that happen during aging. Specifically, imperfections of the maintenance and repair systems that comprise the homeodynamic space lead to a progressive failure of homeodynamics. Major age-related changes include genomic instability, mutations, dysregulated gene expression, cellular senescence, impaired intercellular communication, tissue disorganization, organ dysfunctions, increased vulnerability to stress, and reduced ability to adapt and remodel. Various approaches for intervention, prevention, and modulation of aging incorporate means to minimize the occurrence and accumulation of molecular damage. Mild stress-induced hormesis caused by physical, biological, and nutritional hormetins is a promising holistic strategy for strengthening the homeodynamics. Food components, which induce one or more pathways of stress response in cells and organisms, are potential nutritional hormetins, and can have health- and longevity-promoting effects.

Keywords

Health; longevity; stress; homeostasis; homeodynamics; repair; food; nutrition

Key Facts

• Signs of biological aging appear progressively and exponentially during the period of survival beyond the ELS of a species.

• There are no gerontogenes evolved with a specific function of causing aging and eventual death.

• The role of genes in aging and longevity is mainly at the level of longevity-assurance in evolutionary terms.

• The phenotype of aging is highly differential and heterogeneous at all levels of biological organization.

• Aging is characterized by a stochastic occurrence, accumulation, and heterogeneity of damage in macromolecules.

• Mild stress-induced activation of defense and repair processes helps to maintain health and prolong longevity.

Dictionary of Terms

• Essential lifespan (ELS): optimal duration of life as required by the evolutionary life history of a species. ELS of a species is different from both the average lifespan (ALS) of a cohort of a population, and the maximum lifespan (MLS) recorded for an individual within a species. For example, ELS for Homo sapiens is considered to be about 45 years, whereas the present ALS for the populations of industrially developed countries is about 80 years, and the MLS recorded so far for human beings is 122 years.

• Homeodynamics: in contrast to the machine-based conceptual model homeostasis, which means the same state, the term homeodynamics incorporates the dynamic nature of the living systems, which is not static but constantly changing, remodeling, and adapting.

• Homeodynamic space: a conceptual term to describe the survival ability or the buffering capacity of a biological system; it is comprised of three main categories of biological processes—stress response, damage control, and continuous remodeling.

• Hormesis: biphasic dose response in which the negative or toxic consequences of exposure to high levels of a stressor are observed to be reversed (positive or beneficial) at low levels. Moderate physical exercise is the paradigm for physiological hormesis. The science and study of hormesis is known as Hormetics.

• Hormetin: a condition that induces hormesis; three main types of hormetins are: physical hormetins (temperature, irradiation, mechanical tension); nutritional hormetins (spices and other NNFC, calorie restriction, fasting); and mental hormetins (psychological challenge, meditation).

Introduction

Improving human health and longevity through nutrition is one of the longest running themes in history. While dreams of a perfect food for eternal youth and immortality may still occupy the minds of some, modern scientific knowledge has opened up novel approaches toward understanding and utilizing nutrition in a more realistic and rational way. However, in order to fully appreciate and evaluate the possible approaches toward modulating aging, it can be useful to have an overview and understanding of the current status of biogerontology—the study of the biological basis of aging.

This chapter aims to provide a general review of the molecular and cellular basis of aging, mechanistic theories of aging, homeodynamic mechanisms of survival, maintenance, and repair, followed by a discussion of nutrition-based aging interventions, especially the nutritional hormetins that bring about their health beneficial effects by stress-induced hormesis.

Biological Principles of Aging

Modern biogerontology can be considered to originate in the second half of the 20th century with the writings and experimental findings of Peter Medawar [1], Denham Harman [2], and Leonard Hayflick [3,4]. It can be safely said that the biological bases of aging are now well understood and a distinctive framework has been established [5–7]. This framework has been developed from numerous theoretical analyses, and hundreds of descriptive and interventional experimental studies performed on a wide variety of biological systems with a range of life histories and traits. Four main biological principles can be derived from these, which cover evolutionary, genetic, differential, and molecular aspects of aging and longevity (Table 1.1).

Table 1.1

Principles of Aging and Longevity

Thus, aging is an emergent and epigenetic metaphenomenon, which is neither determined by any gerontogenes, nor is it controlled by a single mechanism. Furthermore, individually no tissue, organ, or system becomes functionally exhausted even in very old organisms; and it is their interconnectedness, interaction, and interdependence that determine the survival of the whole. Various ideas have been put forward to explain the mechanistic basis of aging, and generally all of them incorporate, in one or the other way, molecular damage, molecular heterogeneity, and metabolic imbalance as the cause of aging. These ideas include virtual gerontogenes [12], system failure [13], unregulated growth-related quasiprograms [14], and metabolic instability [15,16]. Most importantly, almost all these views directly or indirectly reject the notion of the evolution of any specific and real genes for aging.

Occurrence, Accumulation, and Consequences of Molecular Damage

As discussed in detail previously [9,17], molecular damages within a cell arise constantly mainly from the following sources: (i) reactive oxygen species (ROS) and other free radicals (FR) formed by the action of external inducers of damage (eg, UV-rays), and as a consequence of intrinsic cellular metabolism involving oxygen, metals, and other metabolites; (ii) nutritional glucose and its metabolites, and their biochemical interactions with ROS and FR; and (iii) spontaneous errors in biochemical processes, such as DNA duplication, transcription, posttranscriptional processing, translation, and posttranslational modifications. Occurrence of molecular damage has led to the formulation of at least two mechanistic theories of biological aging, which have been the basis of most of the experimental aging research during the last 50 years [9].

The first one of these is the so-called free radical theory of aging (FRTA), which arose from the premise that a single common biochemical process may be responsible for the aging and death of all living beings [18,19]. There is abundant evidence to show that a variety of ROS and other FR are indeed involved in the occurrence of molecular damage that can then lead to structural and functional disorders, diseases, and death. The chemistry and biochemistry of FR are very well worked out, and the cellular and organismic consequences are also well documented [20]. However, the main criticism raised against FRTA is with respect to its lack of incorporation of the essential and beneficial role of FR in the normal functioning and survival of biological systems [21,22]. Furthermore, FRTA presents FR as the universal cause of damage without taking into account the differences in the wide range of FR-counteracting mechanisms in different species, which effectively determine the extent of damage occurrence and accumulation. Additionally, a large body of data that shows the contrary and/or lack of predictable and expected beneficial results of antioxidant and FR-scavenging therapies has restricted the application of FRTA [22–25].

The second major mechanistic theory that incorporates the crucial role of macromolecular damage is the so-called protein error theory of aging (PETA). The history of PETA, also known as the error catastrophe theory, is often marked with controversy [9,17,26]. Since the spontaneous error frequency in protein synthesis is generally several orders of magnitude higher than that in nucleic acid synthesis, the role of protein errors and their feedback in biochemical pathways has been considered to be a crucial one with respect to aging. Several attempts have been made to determine the accuracy of translation in cell-free extracts, and most of the studies show that there is an age-related increase in the misincorporation of nucleotides and amino acids [26–30]. It has also been shown that there is an age-related accumulation of aberrant DNA polymerases and other components of the transcriptional and translational machinery [31,32].

Further evidence in support of PETA comes from experiments which showed that an induction and increase in protein errors can accelerate aging in human cells and bacteria [26,33,34]. Similarly, an increase in the accuracy of protein synthesis can slow aging and increase the lifespan in fungi [35–37]. Therefore, it is not ruled out that several kinds of errors in various components of the protein synthetic machinery and in mitochondria do have long term effects on cellular stability and survival [29,30]. However, almost all these methods have relied on indirect in vitro assays, and so far direct, realistic, and accurate estimates of age-related changes in errors in cytoplasmic and mitochondrial proteins, and their biological relevance, have not been made. Similarly, applying methods such as two-dimensional gel electrophoresis, which can resolve only some kinds of misincorporations, have so far remained insensitive and inconclusive [26,38,39].

Both the FRTA and PETA provide molecular mechanisms for the occurrence of molecular damage. Additionally, nutritional components, especially the sugars and metal-based micronutrients, can induce, enhance, and amplify the molecular damage either independently or in combination with other inducers of damage. It is important to point out that although the action of the damaging agents is mainly stochastic, the result of whether a specific macromolecule will become damaged and whether damage can persist depends both on its structure, localization, and interactions with other macromolecules, and on the activity and efficiency of a complex series of maintenance and repair pathways, discussed below [9,17]. Understanding the quantitative and qualitative aspects of molecular damage in terms of their biological relevance is one of the most challenging aspects of the present biogerontological research.

Whatever the reason for the occurrence of molecular damage, accumulation of damage in DNA, RNA, proteins, and other macromolecules is a well-established molecular phenotype of aging. Since there is an extremely low probability that any two molecules become damaged in exactly the same way and to the same extent, an increase in molecular heterogeneity is inevitable. Increased molecular heterogeneity is the fundamental basis for the molecular, biochemical, cellular, and physiological changes happening during aging. Such age-related changes include genomic instability, mutations, dysregulated gene expression, cellular senescence, cell death, impaired intercellular communication, tissue disorganization, organ dysfunctions, increased vulnerability to stress, reduced ability to adapt and remodel, and increased chances of the emergence of age-related diseases [7,40,41].

Homeodynamics and the Homeodynamic Space

Another way to understand aging is by understanding the processes of life and their intrinsic limitations. Survival of an organism is a dynamic tug between the occurrence of damage and the processes of maintenance and repair systems (MARS). The main MARS that comprise the longevity-assurance processes are listed in Table 1.2.

Table 1.2

Main MARS in a Biological System

Another way of conceptualizing MARS is the idea of homeodynamic space, which may also be considered as the survival ability or the buffering capacity of a biological system [10]. The term homeodynamics, meaning the same dynamics, is distinct from the classical term homeostasis, that means the same state, which ignores the reality of ever-dynamic, ever-changing and yet appearing to remain the same, dynamic living systems [42]. Biological systems—cells, tissues, organs, organisms, and populations—are never static, and therefore the most commonly used term homeostasis is wrong for living systems.

Three main characteristics of the homeodynamic space are the abilities to control the levels of molecular damage, to respond to external and internal stress, and to constantly remodel and adapt in dynamic interactions. A large number of molecular, cellular, and physiological pathways and their interconnected networks, including MARS listed in Table 1.2, determine the nature and extent of the homeodynamic space of an individual.

At the species level, biological evolutionary processes have assured the essential lifespan (ELS) of a species by optimizing for homeodynamic space through MARS, which are also the main target of evolutionary investment, stability, and selection [16,43–47]. However, the period of survival beyond ELS is characterized by the progressive shrinkage of the homeodynamic space characterized by reduced ability to tolerate stress, to control molecular damage, and to adapt and remodel. Shrinkage of the homeodynamic space leads to an increase in the zone of vulnerability, reduced buffering capacity, and increased probabilities for the onset and emergence of chronic diseases [10]. Major chronic conditions, for example, metabolic disorders, depression, dementia, malnutrition, and several types of age-related cancers, are mostly due to the generalized failure and dysregulation of processes of life and their interactive networks, and not due to any specific cause(s) [48–51]. Thus, aging in itself is not a disease, but is a condition that allows the emergence of one or more diseases in some, but not all, old people.

Nutrition and Food for Aging Interventions

There is a lot of scientific and social interest in the real and potential power of food in improving health, preventing diseases, and extending the lifespan [52–55]. However, in scientific research and experimentation, often little or no distinction is made between nutrition and food, which is a gross omission in a social context. As discussed elsewhere [56], nutrition is the amalgamation of various components, such as proteins, carbohydrates, fats, and minerals, which are needed for the survival, growth, and development of a biological system. However, food is what, why, and how we eat for survival, health, and longevity. This distinction between nutrition and food is a very important variable for humans, and may be equally important for other animal models used in research, where the appearance, the smell, the texture, and the taste of the food matter. None of the nutritional components is by itself either good or bad, and none of the foods is either healthy or unhealthy. Nutrition can lead to either good effects or bad effects; and the food can have consequences making us either healthy or unhealthy. It is the quantity, quality, frequency, and emotional satisfaction that determine whether any particular food can help us achieve the aim of maintaining and improving health, and delaying, preventing, or treating a disease [56].

Some food components in the diet of human beings do not have any nutritional value in the normal sense of providing material for the structure, function, and energy requirements of the body [57]. Such nonnutritional food components (NNFC) usually come from spices, herbs, and the so-called vegetables and fruits, for example, onion, garlic, ginger, shallot, chive, and chilies [58]. Different combinations of NNFC are integral parts of different food cultures in different social setups, and carry a wide range of claims made for their health beneficial and longevity promoting effects. Not all such claims for NNFC have been scientifically tested and confirmed, and often very little is known about their biochemical mode of action. However, recent research in the field of hormesis is unraveling some of the mechanistic basis for the effects of NNFC [59].

Hormesis is the positive relationship between low-level stress and health [59–61]. Whereas uncontrolled, severe, and chronic stress is recognized as being harmful for health, single- or multiple-exposures to mild stress are generally health beneficial. Moderate exercise is the best example of such a phenomenon of mild stress-induced physiological hormesis. Exercise initially increases the production of FR, acids, and other potentially harmful biochemicals in the body, but the cellular responses to stress, in increasing defense and repair processes, protect and strengthen the body. Such conditions, which induce hormesis, are called hormetins, and are categorized as physical, mental, and nutritional hormetins [62,63].

Nutritional Hormetins

Among different types of hormetins, nutritional hormetins, especially those derived from plant sources, have generated much scientific interest for their potential health beneficial effects. This is because of the realization that not all chemicals found in plants are beneficial for animals in a direct manner, but rather they cause molecular damage by virtue of their electrochemical properties [64]. Several NNFC and their constituent chemical entities, such as flavonoids or bioflavonoids, are nutritional hormetins. This is because they directly or indirectly induce one or more stress responses, such as Nrf2 activation, heat shock response (HSR), unfolded protein response, and sirtuin response [63,65]. After the initial recognition of disturbance or damage caused by a stressor, numerous downstream biochemical processes come into play, including the synthesis and activation of chaperones, stimulation of protein turnover, induction of autophagy, and an increase in antioxidant enzymes [65].

Several NNFC have been shown to achieve the antioxidant effects by the activation of Nrf2 transcription factor. This activation generally happens following the electrophilic modification/damage of its inhibitor protein Keap1, which then leads to the accumulation, heterodimerization, nuclear translocation, and DNA binding of Nrf2 at the antioxidant response element, resulting in the downstream expression of a large number of the so-called antioxidant genes, such as heme oxygenase HO-1, superoxide dismutase, glutathione, and catalase [64,66,67]. Some well-known phytochemicals and plant extracts which strongly induce Nrf2-mediated hormetic response include curcumin, quercetin, genistein, eugenol coffee, turmeric, rosemary, broccoli, thyme, clove, and oregano [64,68].

Another stress response pathway that has been studied in detail and can be the basis for identifying novel nutritional hormetins is the HSR. Induction of proteotoxic stress, such as protein misfolding and denaturation, initiates HSR by the intracellular release of the heat shock transcription factor from their captor-proteins, followed by its nuclear translocation, trimerization, and DNA-binding for the expression of several heat shock proteins (HSP) [69,70]. A wide range of biological effects then occur which involve HSP, such as protein repair, refolding, and selective degradation of abnormal proteins leading to the cleaning up and an overall improvement in the structure and function of the cells. Various phytochemicals and nutritional components have been shown to induce HSR and have health beneficial effects including antiaging and longevity promoting effects. Some examples of nutritional hormetins involving HSR are phenolic acids, polyphenols, flavonoids, ferulic acid [71,72], geranylgeranyl, rosmarinic acid, kinetin, zinc [72–74], and the extracts of tea, dark chocolate, saffron, and spinach [75]. Further screening of animal and plant components for their ability to induce HSR can identify other potential nutritional hormetins.

Other pathways of stress response, which are involved in initiating hormetic effects of nutritional components are the NFkB, FOXO, sirtuins, DNA repair response, and autophagy pathways. Resveratrol and some other mimetics of calorie restriction work by the induction of one or more of these pathways [74,76]. Discovering novel nutritional hormetins by putting potential candidates through a screening process for their ability to induce one or more stress pathways in cells and organisms can be a promising strategy [63].

Conclusions

The molecular and cellular bases of aging lie in the progressive failure of MARS that leads to the emergence of the senescent phenotype. There are no gerontogenes with the specific evolutionary function to cause aging and death of an individual. The concept of homeodynamic space can be a useful one in order to identify a set of measurable, evidence-based, and demonstratable parameters of health, robustness, and resilience. Age-related health problems, for which there are no clear-cut causative agents, may be better tackled by focusing on health mechanisms and their maintenance, rather than disease management and treatment. Biogerontological and other research on life processes and lifestyle-related diseases have shown that the issues of aging, quality of life, and longevity need to be approached with health-oriented paradigms.

Summary Points

• Molecular and cellular bases of aging lie in the occurrence and accumulation of damage.

• Imperfections of the MARS that comprise the homeodynamic space for survival lead to a progressive failure of homeodynamics.

• Impaired and dysregulated function, increased vulnerability to stress, and reduced ability to adapt and remodel are the major signs of aging.

• Aging is a continuum of life-history in which some changes can lead to the clinical diagnosis as emergence of one or more diseases.

• Approaches for intervention, prevention, and modulation of aging require means to minimize the occurrence and accumulation of molecular damage.

• Mild stress-induced hormesis caused by physical, biological, and nutritional hormetins is a promising holistic strategy for strengthening the homeodynamics.

• Some food components, which induce one or more pathways of stress response, are potential nutritional hormetins, and can have health- and longevity-promoting effects.

References

1. Medawar PB. An unsolved problem in biology. London H.K. Lewis 1952.

2. Harman D. Aging: a theory based on free radical and radiation chemistry. J Gerontol. 1956;11(3):298–300.

3. Hayflick L, Moorhead PS. The serial cultivation of human diploid strains. Exp Cell Res. 1961;25:585–621.

4. Hayflick L. The limited in vitro lifetime of human diploid cell strains. Exp Cell Res. 1965;37:614–636.

5. Holliday R. Aging is no longer an unsolved problem in biology. Ann N Y Acad Sci. 2006;1067:1–9.

6. Hayflick L. Biological aging is no longer an unsolved problem. Ann N Y Acad Sci. 2007;1100:1–13.

7. Rattan SIS. Biogerontology: from here to where? The Lord Cohen Medal Lecture-2011. Biogerontology. 2012;13(1):83–91.

8. Rattan SIS. Biogerontology: the next step. Ann N Y Acad Sci. 2000;908:282–290.

9. Rattan SIS. Theories of biological aging: genes, proteins and free radicals. Free Rad Res. 2006;40:1230–1238.

10. Rattan SIS. Homeostasis, homeodynamics, and aging. In: Birren J, ed. Encyclopedia of gerontology. 2nd ed. London: Elsevier Inc; 2007;696–699.

11. Holliday R, Rattan SIS. Longevity mutants do not establish any new science of ageing. Biogerontology. 2010;11(4):507–511.

12. Rattan SIS. Gerontogenes: real or virtual? FASEB J. 1995;9:284–286.

13. Gavrilov LA, Gavrilova NS. The reliability theory of aging and longevity. J Theor Biol. 2001;213:527–545.

14. Blagosklonny MV. Cell cycle arrest is not yet senescence, which is not just cell cycle arrest: terminology for TOR-driven aging. Aging (Albany, NY). 2012;4(3):159–165.

15. Demetrius L. Calorie restriction, metabolic rate and entropy. J Gerontol Biol Sci. 2004;59A(9):902–915.

16. Demetrius L. Boltzmann, Darwin and directionality theory. Phys Rep. 2013;530:1–85.

17. Rattan SIS. Increased molecular damage and heterogeneity as the basis of aging. Biol Chem. 2008;389(3):267–272.

18. Harman D. Free radical theory of aging: an update. Ann N Y Acad Sci. 2006;1067:10–21.

19. Harman D. Origin and evolution of the free radical theory of aging: a brief personal history, 1954–2009. Biogerontology. 2009;10(6):773–781.

20. Kalyanaraman B. Teaching the basics of redox biology to medical and graduate students: oxidants, antioxidants and disease mechanisms. Redox Biol. 2013;1(1):244–257.

21. Linnane AW, Eastwood H. Cellular redox regulation and prooxidant signaling systems A new perspective on the free radical theory of aging. Ann N Y Acad Sci. 2006;1067:47–55.

22. Linnane AW, Kios M, Vitetta L. Healthy aging: regulation of the metabolome by cellular redox modulation and prooxidant signaling systems: the essential roles of superoxide anion and hydrogen peroxide. Biogerontology. 2007;8(5):445–467.

23. Le Bourg E, Fournier D. Is lifespan extension accompanied by improved antioxidant defences? A study of superoxide dismutase and catalase in Drosophila melanogaster flies that lived in hypergravity at young age. Biogerontology. 2004;5(4):261–264.

24. Le Bourg E. Antioxidants and aging in human beings. In: Rattan SIS, ed. Aging interventions and therapies. Singapore: World Scientific Publishers; 2005;85–107.

25. Howes RM. The free radical fantasy: a panoply of paradoxes. Ann N Y Acad Sci. 2006;1067:22–26.

26. Holliday R. The current status of the protein error theory of aging. Exp Gerontol. 1996;31(4):449–452.

27. Rattan SIS. Synthesis, modifications and turnover of proteins during aging. Exp Gerontol. 1996;31(1–2):33–47.

28. Rattan SIS. Transcriptional and translational dysregulation during aging. In: von Zglinicki T, ed. Aging at the molecular level. Dordrecht: Kluwer Academic Publishers; 2003;179–191.

29. Holliday R. Streptomycin, errors in mitochondria and ageing. Biogerontology. 2005;6(6):431–432.

30. Hipkiss A. Errors, mitochondrial dysfunction and ageing. Biogerontology. 2003;4:397–400.

31. Srivastava VK, Miller S, Schroeder M, Crouch E, Busbee D. Activity of DNA polymerase alpha in aging human fibroblasts. Biogerontology. 2000;1(3):201–216.

32. Srivastava VK, Busbee DL. Replicative enzymes and ageing: importance of DNA polymerase alpha function to the events of cellular ageing. Ageing Res Rev. 2002;1:443–463.

33. Nyström T. Translational fidelity, protein oxidation, and senescence: lessons from bacteria. Ageing Res Rev. 2002;1:693–703.

34. Nyström T. Aging in bacteria. Curr Opin Microbiol. 2002;5:596–601.

35. Silar P, Picard M. Increased longevity of EF-1a high-fidelity mutants in Podospora anserina. J Mol Biol. 1994;235:231–236.

36. Silar P, Rossignol M, Haedens V, Derhy Z, Mazabraud A. Deletion and dosage modulation of the eEF1A gene in Podospora anserina: effect on the life cycle. Biogerontology. 2000;1(1):47–54.

37. Holbrook MA, Menninger JR. Erythromycin slows aging of Saccharomyces cerevisiae. J Gerontol Biol Sci. 2002;57A(1):B29–B36.

38. Hipkiss A. Accumulation of altered proteins and ageing: causes and effects. Exp Gerontol. 2006;41:464–473.

39. Kirkwood TBL. A systematic look at an old problem. Nature. 2008;451(7 Feb):644–647.

40. Rattan SIS. Aging is not a disease: implications for intervention. Aging Dis. 2014;5(3):196–202.

41. Lopez-Otin C, Blasco MA, Partridge L, Serrano M, Kroemer G. The hallmarks of aging. Cell. 2013;153(6):1194–1217.

42. Yates FE. Order and complexity in dynamical systems: homeodynamics as a generalized mechanics for biology. Math Comput Model. 1994;19:49–74.

43. Kirkwood TBL. Time of our lives London: Weidenfeld & Nicolson; 1999.

44. Holliday R. Ageing: the paradox of life Dordrecht: Springer; 2007.

45. Martin GM. Modalities of gene action predicted by the classical evolutionary theory of aging. Ann N Y Acad Sci. 2007;1100:14–20.

46. Gladyshev VN. The origin of aging: imperfectness-driven nonrandom damage defines the aging process and control of lifespan. Trends Genet. 2013;29(9):506–512.

47. Wensink MJ, van Heemst D, Rozing MP, Westendorp RG. The maintenance gap: a new theoretical perspective on the evolution of aging. Biogerontology. 2012;13(2):197–201.

48. Tacutu R, Budovsky A, Yanai H, Fraifeld VE. Molecular links between cellular senescence, longevity and age-related diseases—a systems biology perspective. Aging (Albany NY). 2011;3(12):1178–1191.

49. Walter S, Atzmon G, Demerath EW, et al. A genome-wide association study of aging. Neurobiol Aging. 2011;32 2109 e15–2109 e28.

50. O’Neill D. 2012—That was the year that was. Age Ageing. 2013;42(2):140–144.

51. Watson J. Oxidants, antioxidants and the current incurability of metastatic cancers. Open Biol. 2013;3(1):120144.

52. Ferrari CKB. Functional foods, herbs and neutraceuticals: towards biochemical mechanisms of healthy aging. Biogerontology. 2004;5(5):275–289.

53. Mutch DM, Wahli W, Williamson G. Nutrigenomics and nutrigenetics: the emerging faces of nutrition. FASEB J. 2005;19:1602–1616.

54. Mattson MP. Dietary factors, hormesis and health. Ageing Res Rev. 2008;7:43–48.

55. Naik SR, Thakare VN, Joshi FP. Functional foods and herbs as potential immunoadjuvants and medicines in maintaining healthy immune system: a commentary. J Compl Integr Med. 2010;7 Article 46.

56. Rattan SIS. Nutrition and food for health and longevity. Int J Nutr Pharm Neur Dis. 2015;5(2):45.

57. Gutteridge JM, Halliwell B. Antioxidants: molecules, medicines, and myths. Biochem Biophys Res Commun. 2010;393(4):561–564.

58. Aggarwal BB, Kunnumakkara AB, Harikumar KB, Tharakan ST, Sung B, Anand P. Potential of spice-derived phytochemicals for cancer prevention. Planta Med. 2008;74(13):1560–1569.

59. Rattan SIS, Le Bourg E, eds. Hormesis in health and disease. Boca Raton, FL: CRC Press; 2014.

60. Calabrese EJ, Bachmann KA, Bailer AJ, et al. Biological stress response terminology: integrating the concepts of adaptive response and preconditioning stress within a hormetic dose-response framework. Toxicol Appl Pharmacol. 2007;222(1):122–128.

61. Le Bourg E, Rattan SIS, eds. Mild stress and healthy aging: applying hormesis in aging research and interventions. Dordrecht: Springer; 2008.

62. Rattan SIS. Hormesis in aging. Ageing Res Rev. 2008;7:63–78.

63. Rattan SIS. Rationale and methods of discovering hormetins as drugs for healthy ageing. Expert Opin Drug Discov. 2012;7(5):439–448.

64. Balstad TR, Carlsen H, Myhrstad MC, et al. Coffee, broccoli and spices are strong inducers of electrophile response element-dependent transcription in vitro and in vivo—studies in electrophile response element transgenic mice. Mol Nutr Food Res. 2011;55(2):185–197.

65. Demirovic D, Rattan SI. Establishing cellular stress response profiles as biomarkers of homeodynamics, health and hormesis. Exp Gerontol. 2013;48(1):94–98.

66. Calabrese V, Cornelius C, Dinkova-Kostova AT, Calabrese EJ, Mattson MP. Cellular stress responses, the hormesis paradigm, and vitagenes: novel targets for therapeutic intervention in neurodegenerative disorders. Antioxid Redox Signal. 2010;13(11):1763–1811.

67. Calabrese V, Cornelius C, Mancuso C, et al. Cellular stress response: a novel target for chemoprevention and nutritional neuroprotection in aging, neurodegenerative disorders and longevity. Neurochem Res. 2008;33(12):2444–2471.

68. Lima CF, Pereira-Wilson C, Rattan SI. Curcumin induces heme oxygenase-1 in normal human skin fibroblasts through redox signaling: relevance for anti-aging intervention. Mol Nutr Food Res. 2011;55(3):430–442.

69. Verbeke P, Fonager J, Clark BFC, Rattan SIS. Heat shock response and ageing: mechanisms and applications. Cell Biol Int. 2001;25(9):845–857.

70. Liberek K, Lewandowska A, Zietkiewicz S. Chaperones in control of protein disaggregation. EMBO J. 2008;27:328–335.

71. Barone E, Calabrese V, Mancuso C. Ferulic acid and its therapeutic potential as a hormetin for age-related diseases. Biogerontology. 2009;10(2):97–108.

72. Son TG, Camandola S, Mattson MP. Hormetic dietary phytochemicals. Neuromol Med. 2008;10(4):236–246.

73. Berge U, Kristensen P, Rattan SIS. Hormetic modulation of differentiation of normal human epidermal keratinocytes undergoing replicative senescence in vitro. Exp Gerontol. 2008;43:658–662.

74. Sonneborn JS. Mimetics of hormetic agents: stress-resistance triggers. Dose Response. 2010;8(1):97–121.

75. Wieten L, van der Zee R, Goedemans R, et al. Hsp70 expression and induction as a readout for detection of immune modulatory components in food. Cell Stress Chaperones. 2010;15(1):25–37.

76. Longo VD. Linking sirtuins, IGF-I signaling, and starvation. Exp Gerontol. 2009;44(1–2):70–74.

Chapter 2

Unraveling Stochastic Aging Processes in Mouse Liver

Dissecting Biological from Chronological Age

L.W.M. van Kerkhof¹, J.L.A. Pennings¹, T. Guichelaar², R.V. Kuiper³, M.E.T. Dollé¹ and H. van Steeg¹,⁴,    ¹Centre for Health Protection, National Institute for Public Health and the Environment (RIVM), Bilthoven, The Netherlands,    ²Centre for Infectious Disease Control, National Institute for Public Health and the Environment (RIVM), Bilthoven, The Netherlands,    ³Department of Laboratory Medicine, Karolinska Institutet, Stockholm, Sweden,    ⁴Department of Human Genetics, Leiden University Medical Center, Leiden, The Netherlands

Abstract

Cellular damage accumulation is a central feature of aging, resulting in functional decline and increased vulnerability to pathology and disease. This accumulation occurs over time, but is not exclusively time-dependent. Recent studies showed that, concerning biomedical issues, age may rather be defined as biological age, expressed as the cumulative change of physiological or functional parameters, in contrast to chronological age. We describe a method that goes beyond chronological age as the primary determinant for aging in the investigations of mechanisms underlying aging. We show in mice that the use of biological phenotypes reveals processes, genes, and pathways related to aging that would not have been revealed when using chronological age as the main determinant of aging. Using determinants of biological aging will improve our knowledge regarding the mechanisms underlying aging and may lead to the discovery of new biomarkers of frailty valuable for predicting health risks.

Keywords

Chronological age; biological age; aging; gene expression profiles; pathology; immune system

Key Facts

• Accumulation of cellular damage is a central feature of aging, resulting in functional decline and increased vulnerability to pathology and disease. This accumulation occurs over time, but is not exclusively time-dependent.

• Concerning biomedical issues, age may rather be predicted or defined as biological age, in contrast to chronological age.

• Examples of determinants of biological age are functional and physiological parameters, including pathological parameters.

• Age-related pathological phenotypes accumulate at different rates per individual and independent of other pathological endpoints within the same tissue, or similar endpoints across tissues.

• Within one individual, severe scorings of pathological parameters are rarely present for multiple parameters, that is, individuals score old on one parameter but not on the others, within and between organs.

• Using age-related phenotypes, gene-expression profiles and biological pathways are identified which are different from the profiles and pathways identified when using chronological age as the sole determiner of age.

• As all age-related phenotypes show some correlation with age, this new approach gives insight in phenotype-specific age markers that could be used to stratify individuals in their personal aging trajectories.

• Appropriate weighing factors between phenotype-specific markers may ultimately lead to the derivation of a biological age score for the entire individual organism.

Dictionary of Terms

• Chronological age: age expressed as the amount of time a person has lived in years, weeks, or days (eg, passport age).

• Biological age: age expressed as a measurement of biologically relevant parameters such as vulnerability to death and/or disease, risk of functional decline, or frailty index.

• Gene expression profiles/whole transcriptome analysis: profiles of the amount of mRNA expression in a certain set of samples. Whole transcriptome analysis refers to the ability to analyze a large set of genes in a single sample.

• Age-related phenotypes: phenotype is a term used to parse a disease or set of symptoms in more stable components. In our case, phenotypes are used to parse symptoms related to aging into measurable parameters which are (partially) independent of each other.

• Markers of frailty: frailty refers to the age-related decline in physical, mental, and/or social functioning and as such is a concept of unhealthy aging. Markers of frailty can be used to identify people at risk for accumulated aging phenotypes.

• Lipofuscin: pigment granules that are mainly composed of lipid-containing residues of lysosomal digestion. Accumulation of lipofuscin is associated with aging and can be observed in several organs (such as liver, nerve cells, kidney).

• Karyomegaly: refers to the presence of enlarged cell nuclei, and is associated with aging.

• Liver vacuolization: vacuolization potentially caused by increased glycogen storage, fat storage, or cellular swelling and may represent a (reversible) degenerative stage.

Introduction

Aging can be broadly defined as the functional decline occurring in organisms in a time-dependent manner. Aging is a complex process comprising a wide variety of interconnected features [1–7]. A central feature is the accumulation of cellular damage which results in functional decline and increased vulnerability to pathology and disease [1,2]. This accumulation occurs over time, but is not exclusively time-dependent. Human lifespan varies from less than 10 years for the severe progeria patients to over 100 years for centenarians [8–10]. This variation in human lifespan is partly related to genetic variation [8–10]. However, even in genetically identical animals and monozygotic twins lifespan, fitness, and biological functions of, for example, the immune system vary substantially [11–14], indicating that other factors than time and genetic variation are important determinants of biological aging as well. It is likely that the heterogeneity in lifespan observed within and between species is determined by a balance between damaging exposures and resilience to this damage [15]. It is important to note that biological aging is a stochastically driven process; there are components of coincidence involved in the net damage that occurs. Hence, variations in lifespan and fitness at older age arise from differences in exposure to damaging properties from the environment, differences in the characteristics of the damage done, and the body’s innate ability to repair and compensate for this damage.

The functional decline associated with progressing chronological age has been proven difficult to mechanistically dissect or translate into consistent biomarkers of this decline. Likely, this is partly due to the complex interconnection of involved mechanisms that are difficult to disentangle and partly due to the use of chronological age (time) as the most important determinant in defining young and old animals or humans. Each individual organism follows its own path of aging, and, as a result, generalization of groups based on chronological age results in a heterogeneous set of aging processes, particularly in older groups. Consequently, this hampers investigations into the mechanisms underlying aging. Instead of using chronological age as the most important determinant, we propose to take biological age-ranks for a better assessment of the underlying mechanisms.

The idea of investigating biological age was proposed in 1969 by Alex Comfort [16] and has received attention ever since (eg, see Refs. [2,12,15,17–21]). Biological age refers to the age of an organism expressed in terms of biological fitness based on parameters that relate to the functional decline or vulnerability to death, in contrast to the chronological age, which depends exclusively on time. Hence, for a parameter to be informative of biological age it should associate with functional decline, vulnerability to death, or lifespan, and not necessarily with chronological age.

Several researchers have investigated whether physiological and/or functional parameters were associated with vulnerability to death or lifespan and, as such, whether these parameters were suitable markers for biological age. For example, Levine investigated several methods to use a set of physiological factors (eg, C-reactive protein, white blood cell count, serum urea nitrogen) to predict biological age [15]. The value of the predictions was tested by applying them to a large cohort of subjects (30–75 years) of which the physiological parameters and lifespan were known. He concluded that the Klemera and Doubal method, in which biological age is calculated based on a combined score of multiple aging dependent fitness parameters (including, eg, C-reactive protein, cholesterol levels, forced expiratory volume, and blood pressure) is better at predicting mortality than chronological age [15]. A slightly different approach was recently used by Tomas-Loba et al., who related metabolic profiles to chronological age and consecutively investigated if these could predict biological age in mice known to have a short or long lifespan [19]. The metabolic signature associated with chronological aging was able to predict aging produced by telomere-shortening. However, in this study the researchers did not investigate if the metabolic profile was related to actual lifespan or functional decline.

A study by Holly et al. identified a set of six genes from human peripheral blood leukocytes that was predicative for chronological age [17]. This set was used in a second set of subjects to investigate if they could identify subjects with a younger predicted biological age (ie, healthier phenotype) compared to their chronological age. Subsequently, this group of younger subjects was compared to the remainder of the group based on several functional and physiological parameters associated with aging. They observed that the predicted biologically younger group scored better on the functional and physiological parameters (such as muscle strength, c-reactive protein, and several others), indicating that their biological age was lower than their chronological age.

In summary, these studies indicate that physiological and functional parameters can be used to predict or determine biological age. For example, it has been proposed that predicted biological age can be used to estimate an individual’s chance of success when undergoing a complex surgery [22]. Another important functionality that using biological age enables is the investigation of mechanisms underlying aging and the search for pathology specific markers or frailty markers. In this chapter, we describe a method that takes into account a measure of biological aging, histopathological parameters in the liver, instead of chronological age as a mere determinant for age, with the aim of investigating mechanisms underlying aging.

Pathological Parameters are only Partially Associated with Chronological Age

To investigate biological aging-related processes, different pathological parameters were scored at six chronological ages (13, 26, 52, 78, 104, 130 weeks) in female C57BL/6J mice [12,18]. Pathology scores were used to rerank the animals, independent of their chronological age, to derive a biological age ranking for each specific phenotype (Fig. 2.1). In Fig. 2.1, animals with the same rank are separated by chronological age for visualization purposes only, analyses were performed with original rankings.

Figure 2.1 Ranking of mice based on chronological age (A) and pathological endpoint (B). This figure shows that pathology scores are partially associated with chronological age. Color indicates chronological age. Blue, 13 weeks; light blue, 26 weeks; green, 52 weeks; yellow, 78 weeks; orange, 104 weeks; red, 130 weeks. Black cross (panel A) indicates average values per group. Arrows indicate an example of a chronologically old mouse (130 weeks, red arrow) with a relatively low lipofuscin index (48.67, ranking 29/51) and a chronologically younger mouse (52 weeks, green arrow) with a relatively high lipofuscin index (97.3, ranking 40.5/51), see also section Intraorgan Specific Biological Phenotypes.

We focused on three different liver pathologies (lipofuscin index, karyomegaly, and liver vacuolization) and to allow comparisons between organs we determined the severity of lipofuscin accumulation in the brain (brain lipofuscinosis). This focus on the liver was chosen considering the importance of metabolic processes of the liver in aging [2]. These three different pathological endpoints were selected from a set of seven initially analyzed endpoints [12], based on their scoring: higher dynamic scores are favorable in correlation analyses. Lipofuscin refers to the pigment granules that are mainly composed of lipid-containing residues of lysosomal digestion. Accumulation of lipofuscin is associated with chronological aging and can be observed in several tissues (such as liver, brain, nerve cells, kidney). Lipofuscin accumulation in the liver is expressed as the lipofuscin index (spot count × spot size × spot intensity, for all parameters the average of three fields is used). In the brain, lipofuscin accumulation is scored as stages of lipofuscinosis, which refers to the accumulation of lipofuscin in neurons. Karyomegaly refers to the presence of enlarged cell nuclei, associated with aging, and liver vacuolization is likely a consequence of increased glycogen storage, fat storage, or cellular swelling and may represent a (reversible) degenerative stage. The liver vacuolization was not further specified. For some animals certain pathology scores were not assessed. Therefore, the number of animals is unequal: liver lipofuscin 51 animals, liver karyomegaly 49 animals, liver vacuolization 49 animals, and brain lipofuscinosis 46 animals.

The pathology parameters partially correlate with chronological age (Table 2.1 and Fig. 2.1(A)). The strongest correlations are observed for lipofuscin accumulation in liver (R=0.85) and brain (R=0.79). The other two pathological parameters measured in liver correlate less strongly with chronological age: karyomegaly R=0.54 and liver vacuolization R=0.52 (Table 2.1). In Fig. 2.1(B), individual mice are ranked based on the separate pathology scores and colors indicate chronological age. Here the partial correlations observed between pathological parameters and chronological age are visualized, which exemplifies the concept of biological aging being partially distinct from chronological aging and emphasizes the importance of using biological endpoints. For example, based on lipofuscin scoring, a liver sample of a 2-year-old mouse could be considered younger than the liver of a 1-year-old mouse (mice are indicated with arrows in Fig. 2.1, red arrow=130 weeks of age, green arrow=52 weeks of age).

Table 2.1

Correlation Matrix of Chronological Age and Pathology Parameters: Values Represent the Spearman Rank Correlation

van Kerkhof et al.: Expression profiles of aging-associated liver pathology phenotypes: unraveling stochastic aging processes.

Interestingly, the variation observed within the age groups increases with chronological age for lipofuscin accumulation in liver and brain, with low levels of variation at a young age and increased variation at older ages (Fig. 2.1(A)). This is a phenomenon described for other factors as well: for example, variation of immune parameters increases with chronological age [23]. For liver karyomegaly and liver vacuolization, an increase in variation with age is not clearly observed (Fig. 2.1). This might be caused by the characteristics of these parameters. For example, for karyomegaly, large variation is already present at a young age. In addition, ceiling effects might occur, when pathology is abundantly present and is scored as the highest level, further increases in pathology do not result in a higher score. For liver vacuolization, scores might be underestimated when the total number of young small nuclei are decreased, since then size variation might be less evident and scoring might be affected. In addition, for liver vacuolization, a selection process might occur during aging in which animals that develop a more severe pathology score have a poorer survival, resulting in higher numbers of animals with a low pathology score in the older age groups (ie, the animals that do survive).

In summary, pathology parameters only partially correlate with chronological age, that is, within chronological age groups variation in pathology scores is clearly observed. This illustrates the concept of biological aging being partially distinct from chronological aging and emphasizes the importance of using biologically based endpoints.

Intraorgan Specific Biological Phenotypes

Fig. 2.2 visualizes in 3D the correlation observed between liver pathology parameters, of which the correlation coefficients are presented in Table 2.1. These results show that the severity of the lipofuscin index, karyomegaly, and hepatocellular vacuolization differs within one individual. The lipofuscin index correlates weakly with karyomegaly (R=0.47) and with hepatocellular vacuolization (R=0.49) (Table 2.1 and Fig. 2.2). Interestingly, only a few animals score old on all three parameters. For example, when the criterion for an old mouse is set to having for each parameter a score that is the oldest approximately 10%: lipofuscin index ≥ 130 (5 out of 51 mice), karyomegaly score ≥ 5 (3 out of 49 mice), and liver vacuolization score ≥ 4 (9 out of 49 mice), only one mouse meets this criterion (age=104 weeks, indicated with orange arrow and number 1 in Fig. 2.2). When the criterion is set to having the approximately 40% oldest scores on liver pathology: lipofuscin index ≥ 90 (14 out of 51), karyomegaly score ≥ 4 (23 out of 49), and liver vacuolization score ≥ 3 (20 out of 49), only five mice meet these criteria. These results indicate that there is a low level of overlap in individual animals in intraorgan pathology endpoints. The chronological age of the 40% oldest five mice was 104 weeks for two mice and 78 weeks for three mice. Interestingly, none of the chronologically oldest mice (130 weeks) have an old pathology score on all three liver pathology parameters. An example of a 130-week-old mouse is indicated with the red arrow in Fig. 2.2, modest lipofuscin index (48.67, ranking 29/51), high karyomegaly score (5, ranking 48/48), and a low vacuolization score (0, ranking 2/45). Keeping in mind that the median survival of these mice is around 103 weeks of age [12], the mice in the oldest age group result from a strong selection bias for successful aging. Although it is not yet clear if the investigated pathological phenotypes directly relate to survival and fitness, these data suggest that having multiple severe pathology scores is related to poor survival. Hence, it appears that a selection process occurs during aging, resulting in the finding that some animals have a more beneficial or compensatory aging scenario and, therefore, survive better (eg, the 130-week-old animals which are not in the top 40% biologically oldest group). These findings strengthen the importance of using biological phenotypes and indicate that within an organ pathological processes occur largely independent of each other, indicating the need for intraorgan specific biological phenotypes.

Figure 2.2 Pathology ranking for multiple endpoints in the liver. Mice are arranged by liver lipofuscin (y-axis), liver vacuolization (x-axis), and liver karyomegaly (z-axis). This figure visualizes the weak correlation between the different liver pathological parameters. Panels represent different viewpoints to the 3D matrix. Bead size reflects distance from sight to position of beads in the 3D matrix. Color indicates chronological age. Blue, 13 weeks; light blue, 26 weeks; green, 52 weeks; yellow, 78 weeks; orange, 104 weeks; red, 130 weeks. Arrows indicate examples of mice: orange arrow (1) indicates a mouse with a high score (top 10%, see section Tissue-Specific Biological Phenotypes) on all pathology parameters: lipofuscin index=232, ranking 50/51, karyomegaly score=4, ranking 36.4/48, vacuolization score=4, ranking 45/45, age=104 weeks; blue arrow (2) indicates a mouse with low pathology scores: lipofuscin index=1, ranking 10/51, karyomegaly score=0, ranking 1.5/48, vacuolization score=1, ranking 10.5/45, age=13 weeks; red arrow (3) indicates a mouse with high variation among pathology scores: lipofuscin index=48.67, ranking 29/51, karyomegaly score=5, ranking 48/48, vacuolization score=0, ranking 2/45, age=130 weeks.

Tissue-Specific Biological Phenotypes

For the use of pathological endpoints as determinants of biological age, the question arises whether these processes occur in multiple organs in a similar fashion. Therefore, the severity of lipofuscin accumulation in the liver and brain were compared. Fig. 2.3 shows that there is only partial correlation between the severity of lipofuscin accumulation in the liver and brain (R=0.61) (Table 2.1), although it is the highest correlation observed between the pathological parameters included. For example, the animal with the highest liver lipofuscin score has a relatively low brain lipofuscin score (Fig. 2.3, right most animal indicated with a black arrow, age 130 weeks). This indicates that within one individual tissue-specific aging processes occur, which should be considered as (partially) separate phenotypes, underlying the stochastic nature of aging processes.

Figure 2.3 Pathology ranking for lipofuscin accumulation in liver and brain. Mice are arranged by liver lipofuscin ranking (x-axis) and brain lipofuscinosis ranking (y-axis) to allow comparison of both parameters. This figure visualizes the partial overlap between lipofuscin scores in liver and brain. Color indicates chronological age. Blue, 13 weeks; light blue, 26 weeks; green, 52 weeks; yellow, 78 weeks; orange, 104 weeks; red, 130 weeks. Arrow indicates an animal with high liver lipofuscin ranking, but only modest brain lipofuscinosis ranking (see also section Gene Expression Profiles Related to Pathological Aging Parameters).

Gene Expression Profiles Related to Pathological Aging Parameters

The scoring of the different pathological parameters was used to investigate gene expression changes associated with these parameters. Genes that alter expression in a manner that follows the kinetics of a certain pathological parameter might represent processes related to that biological endpoint’s specific aging course. Gene expression is involved in and affected by most cellular processes, and therefore, whole transcriptome analysis allows investigation of several of the different aging processes simultaneously. For methodology of the whole transcriptome analysis, see Refs. [12,18]. For each of the endpoints, we identified genes that have an R²>0.5 (R>0.707 or R<−0.707; at least 50% of the variation in ranking is related to the pathological endpoint), using the Spearman rank correlation analysis. The number of genes with significant correlation to a pathological endpoint was highest for liver lipofuscin (287), followed by karyomegaly (225), brain lipofuscinosis (186), and liver vacuolization (111). Heat maps for the most significantly correlating genes (|R|>0.8) are presented in Fig. 2.4(A–D). These expression patterns are not very consistent across the various pathological endpoints, which is in line with our previous findings that there are different genes associated with the different pathological endpoints [12,18].

Figure 2.4 Heat maps of genes correlating with pathological parameters: (A) liver lipofuscin, (B) liver karyomegaly, (C) liver vacuolization, and (D) brain lipofuscinosis. The genes correlating most significantly |R|>0.8) are shown. Gene names and correlation coefficient (in brackets) are shown on the y-axis. Age of the animals is shown on the x-axis in weeks.

Functional annotation analysis indicated that genes positively associated with liver lipofuscin are mainly associated with immunological processes, such as immune response (20 upregulated genes), inflammatory response (9 genes), and oxidative stress response (3 genes), all indicating that stress at the cellular or tissue level occurs. Additionally, these genes were enriched for terms such as phagocytosis, vesicle mediated transport, and lysosome, all of which point to uptake and degradation of unwanted materials or waste, such as done by macrophages. Among the genes downregulated in relation to liver lipofuscin, a notable number of mitochondria-associated genes were detected. Mitochondrial dysfunction has previously been associated with chronological aging [2,24]. Genes associated with karyomegaly were also, albeit less clearly, enriched for genes involved in immune response- or lysosome-related processes as well as oxidative stress. In addition, there was enrichment for apoptosis-associated genes. In contrast, no significant functional enrichment was found among the genes associated with liver vacuolization, nor with brain lipofuscin. This, again, indicates that there are different functional processes involved in the various pathological endpoints. Interestingly, several of the pathways, such as the immune-related pathways, differ from the pathways that are detected when using chronological age as the sole determinant of age [12,18], indicating the value of using pathology-related endpoints.

With respect to the genes that are involved in the inflammatory response it should be noted that chronological aging and many aging-related chronic diseases are accompanied by alteration and decline of immune functions, such as the increased occurrence of chronic inflammation, a process known as inflammaging [25]. Moreover, in line with the increased variation found for biological endpoints at higher chronological age presented in this chapter (Fig. 2.1), also variation of immune parameters increase with chronological age [23]. In summary, these results indicate that the process of inflammaging should be taken into account in our search to further unravel the process of biological aging and how biological aging may be functionally related to immunology.

Gene Expression Profiles Correlating with Pathological Parameters are Largely Specific to the Pathological Parameters

As described in sections Tissue-Specific Biological Phenotypes and Gene Expression Profiles Related to Pathological Aging Parameters, there are large intraindividual differences in the severity scores of different pathologies, however, there is some correlation between the liver pathologies as well (Table 2.1 and Fig. 2.2). This indicates that there are animals with similar severity levels of two (or three) liver pathologies (Fig. 2.2), which might result in some contamination of the gene sets, that is, some of the genes associated with one process are being detected in the analysis of the other process as well due to the ranking approach. To determine this possible bias in the gene sets, we investigated the overlap in these gene sets. For most endpoints very little overlap is observed (Fig. 2.5), only between liver lipofuscin and karyomegaly a substantial overlap (62 genes) was observed (overlap is 22% of lipofuscin correlating genes and 28% of karyomegaly correlating genes). Therefore, some bias might be present in these results.

Figure 2.5 Venn diagram of genes correlating with the liver pathological endpoints (R²>0.5, R>0.707, or R<−0.707). Within the liver gene set, a significant correlation was observed for 287 genes, for 225 genes with liver karyomegaly, and for 111 with liver vacuolization. This figure shows the overlap among these genes. Of the gene set, 20,427 genes did not show a correlation with any pathological parameter.

For lipofuscin accumulation in liver and brain, some overlap in genes correlating with these endpoints might be expected, since these phenotypes have similarities. However, of the genes correlating with lipofuscin accumulation in liver and brain only four genes are overlapping (of 287 genes associated with liver lipofuscin and 186 genes associated with brain lipofuscinosis). To determine if the nonoverlapping genes were mainly tissue-specific genes, we used a mouse tissue data set [26] (www.biogps.org, top 1% tissue-specific genes). Interestingly, only low levels of tissue-specific genes were observed in the set of lipofuscin associated genes in liver (10 genes) and brain (3 genes). Taken together, these results might indicate that lipofuscinosis in liver and brain are different. Possibly, cellular processes that underlie lipofuscinosis in the liver and brain are different, or alternatively, but not mutually exclusive, liver and brain might to respond in a different manner to similar types of cellular damage.

Interestingly, in the set of 283 genes correlating only with liver lipofuscin, some genes belong to the top 1% of immune cell specific genes. These are mainly macrophages and cell types related to these phagocytes of the immune system (26–30 genes). These results indicate that macrophages, or macrophage-like cells, such as Kupffer cells of the liver, are likely involved with liver lipofuscin accumulation, which is in line with the results reported in the section Gene Expression Profiles Related to Pathological Aging Parameters.

Future Perspectives

We have shown that ranking complex data sets (ie, gene expression data) according to lesion specific severity, results in the detection of unique markers sets for each of these endpoints. These markers are lost in the background noise when a chronological time ranking is used. As all markers show some correlation with age, this new approach gives insight in phenotype-specific age markers that may be used to stratify individuals during their respective aging trajectories. This type of approach can be applied in large scale molecular epidemiology studies since it can be performed with any phenotype related to biological aging, such as functional measurements (eg, grip strength) and physiological measurements (eg, blood pressure) or combinations of these (eg, a frailty index) and can be applied to a variety of complex data sets, such as metabolomics and epigenomics data sets. Application of this approach in multiple studies on different phenotypes and different data sets will greatly enhance our understanding of the processes underlying aging and will lead to the discovery of new biomarkers of unhealthy aging, that is, biomarkers of frailty. These biomarkers of frailty will aid in the early detection of unhealthy aging. In addition, using appropriate weighing factors between phenotype-specific markers may ultimately lead to the derivation of a biological age score for the entire individual organism and to the derivation of a personalized biological age score for individuals within the population comprising our aging society.

Conclusion

We have shown that pathological endpoints (as phenotypes of biological aging) accumulate at different rates per individual and independently of other pathological endpoints within the same tissue, or similar endpoints across tissues. Furthermore, by ranking complex data sets (ie, gene expression data) according to lesion specific severity, unique sets of markers are found for each of these endpoints. For example, reranking of mice according to their lipofuscin score revealed the involvement of immune processes, such as inflammation, indicating that the process of inflammaging should be taken into account in our search to further unravel the process of biological aging. These genes are lost in the background noise, that is, are not identified, when a chronological time ranking is used. As all phenotypes show some correlation with age, this new approach gives insight in phenotype-specific age markers that may be used to stratify individuals during their personal aging trajectories. This approach will lead to more insight into the mechanisms underlying aging. Appropriate weighting factors between phenotype-specific markers may ultimately lead to the derivation of a biological age score for the entire individual organism.

References

1. Garinis GA, van der Horst GT, Vijg J, Hoeijmakers JH. DNA damage and ageing: new-age ideas for an age-old problem. Nat Cell Biol. 2008;10(11):1241–1247.

2. Lopez-Otin C, Blasco MA, Partridge L, Serrano M, Kroemer G. The hallmarks of aging. Cell. 2013;153(6):1194–1217.

3. Mocchegiani E, Malavolta M. Possible new antiaging strategies related to neuroendocrine-immune interactions. Neuroimmunomodulation. 2008;15(4–6):344–350.

4. Morimoto RI, Cuervo AM. Proteostasis and the aging proteome in health and disease. J Gerontol A Biol Sci Med Sci. 2014;69(Suppl. 1):S33–S38.

5. Oh J, Lee YD, Wagers AJ. Stem cell aging: mechanisms, regulators and therapeutic opportunities. Nat Med. 2014;20(8):870–880.

6. Xi H, Li C, Ren F, Zhang H, Zhang L. Telomere, aging and age-related diseases. Aging Clin Exp