Potential New Aging Model Schmidtea Mediterranea Undergoes Gradual Senescence, Experiences Accelerated Senescence at Elevated Temperature

David A Bertin

A Thesis presented to The University of Guelph

In partial fulfillment of requirements for the degree of Master of Science in Molecular and Cellular Biology

Guelph, Ontario, Canada

ABSTRACT

POTENTIAL NEW AGING MODEL SCHMIDTEA MEDITERRANEA UNDERGOES GRADUAL SENESCENCE, EXPERIENCES ACCELERATED SENESCENCE AT ELEVATED TEMPERATURE

David A Bertin Advisor:

University of Guelph, 2015 Dr. Ray Lu

Current model organisms for aging research, commonly C. elegans and D. melanogaster, possess significant limitations such as non-aging life stages, loss of 10-15% of genes compared to non-Ecydosozoans, and low or absent tissue repair by stem cells. An emerging model organism, Schmidtea mediterranea, does not have these limitations. This flatworm species may be a much better representation of human aging while still having the desirable traits of a simple model organism. This study characterized normal aging through demographic study illustrating that this flatworm species, from a demographic standpoint, is indeed a suitable model. A number of additional relationships were shown including the reduction of lifespan and reproductive output resulting from elevated growth temperature, and high reproductive output correlates with long life.

ACKNOWLEDGEMENTS

I would firstly like to express gratitude to my family for their love, advice, and support throughout these past years. Without them I wouldn’t have made it nearly this far. I would also like to thank past lab colleagues including Mandi Martyn, Cheryl Cragg, Adam McCluggage,

Talya Amor, Karin Olek, Wan Kong Yip, Minghua Zeng, and Alex Caruso for their training, advice, and shared knowledge. Finally, I wish to thank Dr. Ray Lu and my advisory committee members

Drs. Terry Van Raay, and Doug Fudge for helping me see the way, turning obstacles into avenues, and providing fresh perspective and indispensable advice.

iii

Table of Contents Acknowledgements ...... iii List of Figures ...... vi List of Tables ...... vii List of Abbreviations ...... viii 1. Introduction ...... 1 1.1 Evolutionary causes of aging...... 1 1.1.1 Mutation accumulation ...... 2 1.1.2 Antagonistic pleiotropy ...... 2 1.1.3 Disposable soma theory...... 3 1.1.4 Heat dissipation limit ...... 4 1.2 Aging mechanisms and interventions ...... 6 1.2.1 Insulin/insulin-like growth factor (IGF-1) signaling...... 6 1.2.2 Multiplex stress resistance ...... 9 1.2.3 Non-genetic interventions ...... 11 1.2.4 Germline ablation ...... 12 1.2.5 Caloric restriction ...... 13 1.3 Schmidtea mediterranea as a model organism for aging ...... 13 1.3.1 Planarians in research ...... 13 1.3.2 Replicative neoblast stem cell population ...... 15 1.3.3 Ecdysozoan gene loss ...... 16 1.3.4 Diapause and dauer states, larval instars ...... 18 2. Objectives ...... 20 3. Materials and Methods ...... 21 3.1 General culture conditions ...... 21 3.1.1 Growth media and feeding ...... 21 3.1.2 Age-synchronized colonies ...... 21 3.2 Demographic colony establishment ...... 22 3.2.1 Collection of anterior lines ...... 22 3.2.2 Demographic data curation ...... 23 3.2.3 Lifespan data analysis and model fitting ...... 23 3.3 Measurement of global protein carbonylation ...... 23 3.4 Measurement of available ATP ...... 24

3.5 Cloning of Smed-age-1 ...... 26 3.5.1 RNA isolation and first strand synthesis ...... 26 3.5.2 Cloning into L4440 dsRNA expression vector ...... 26 3.6 Antibody characterization...... 28 3.6.1 Antibody generation ...... 28 3.6.2 Western blotting ...... 28 3.6.3 Epitope preincubation ...... 28 3.7 siRNA knockdown ...... 29 3.7.1 In vitro expression of age-1 dsRNA ...... 29 3.7.2 Feeding protocol ...... 30 3.7.3 Quantitative reverse transcription PCR (qRT-PCR) ...... 30 4. Results ...... 31 4.1 S. mediterranea kept at an elevated growth temperature shows a reduced lifespan ...... 31 4.1.1 Survival data shows a significant difference between elevated and ambient temperature cohorts ...... 31 4.1.2 Model fitting shows that a Gompertz model best describes lifespan at elevated temperature ...... 34 4.2 Elevated growth temperature decreases reproductive rate at all lifespans ...... 37 4.3 Long-lived ambient temperature worms show a resurgence in weekly reproductive rate in late life ...... 39 4.4 Reproductive events become more frequent with each successive event when at elevated temperature but not at ambient temperature ...... 41 4.5 Global protein carbonylation does not significantly increase with advancing age ...... 43 4.6 ATP availability decreases with advancing age ...... 43 4.7 SMED-AGE-1 presents as a 140 kDa species by western blot ...... 46 4.8 siRNA knockdown of age-1 reduces expression levels to 20 % of endogenous levels and lasts between four and five weeks ...... 48 5. Discussion ...... 49 6. Conclusions and Future Directions ...... 54 7. References ...... 58 Appendix ...... 64

LIST OF FIGURES

Figure 1: Simplified diagram illustrating how IIS can affect multiple downstream signaling pathways

Figure 2: Overlap of IIS pathway and dauer formation pathway (DAF)

Figure 3: A partial cladogram showing the gene-loss event

Figure 4: Single measurements of standard concentrations and ratios of ATP and ADP

Figure 5: Diagram illustrating the cloning strategy for Smed-age-1 full-length coding sequence

Figure 6: Lifespan distributions for ambient (20 °C) and elevated (26 ± 0.5 °C) cohorts

Figure 7: Survivorship curve for ambient (20 °C) and elevated (26 ± 0.5 °C) cohorts

Figure 8: Total reproduction rate by life span with LOESS spline by treatment

Figure 9: Reproduction and survival rates for short-lived and long-lived groups of worms

Figure 10: Median time to next reproductive event plotted against current reproductive event

Figure 11: Two putative aging biomarkers were tested for use with S. mediterranea

Figure 12: Epitope preincubation of polyclonal rabbit α-AGE1 antibody (GenScript)

Figure 13: RNAi knockdown of Smed-age-1 duration and intensity

LIST OF TABLES

Table 1: Key data points extracted from demographic study of S. mediterranea including cohort size, average (avg), high, and low values for reproduction data

Table 2: Gompertz coefficients and mortality rate doubling time

Table 3: Oligonucleotide primers used for PCR amplification and cloning of age-1

vii

LIST OF ABBREVIATIONS ADP Adenosine diphosphate AKT/PKB Ak strain thymoma/protein kinase B AMP adenosine monophosphate AMPK AMP-activated protein kinase AP antagonistic pleiotropy ATP adenosine triphosphate BMR basal metabolic rate CA caloric restriction Ct cycle threshold DAF dauer formation DNP 2,4-dinitrophenol dsRNA double-stranded RNA dT deoxythymidine FOXO forkhead box class O Gapdh glyceraldehydes 3-phosphate dehydrogenase IGF-1 insulin-like growth factor 1 IIS insulin/IGF signaling kDa kilodalton LAM longevity assurance mechanism LC50 lethal concentration, 50% mortality LOESS local regression MA mutation accumulation MCS multiple cloning site MRDT mortality rate doubling time Nox NADPH oxidase PBS phosphate buffered saline PCR polymerase chain reaction PI3K phosphatidylinositol-3-kinase PIP2 phosphatidylinositol 4,5-bisphosphate PIP3 phosphatidylinositol (3,4,5)-trisphosphate PTEN phosphatase and tensin homolog PVDF polyvinylidene fluoride qRT-PCR quantitative reverse-transcription PCR RAS rat sarcoma family RLU relative light unit RNAi RNA interference ROS reactive oxygen species RT room temperature SDS sodium dodecyl sulfate siRNA small interfering RNA TOR target of rapamycin TORC target of rapamycin complex

viii

1. INTRODUCTION

1.1 Evolutionary causes of aging

To date, many hypotheses have been put forward to describe the aging process in terms of causes and effects involving a variety of intracellular and intercellular processes. There are a number of theories regarding the evolution of aging (Gavrilov & Gavrilova 2002; Hughes &

Reynolds 2005), as well as a great wealth of information regarding the mechanisms by which it is thought to occur (Hughes & Reynolds 2005; Liu & Rando 2011). There exists a large body of research on individual mechanisms impacting senescence, such as the involvement of reactive oxygen species (ROS), nutrient sensing and caloric restriction (CR), DNA damage and mutation accumulation, and telomere length, among many others (Medvedev 1990). From an evolutionary perspective, senescence is thought to be an inevitable consequence of natural selection (Hughes & Reynolds 2005). If we consider a population of a non-senescent organism, its lifespan must be determined by extrinsic (ie. environmental) causes such as predation and physical hazards. Therefore, there exists an expected lifespan for this species regardless of their immortality. Natural selection would tend to favour those individuals that invest more resources into reproduction in this expected lifespan at the expense of somatic maintenance in order to maximize genetic contribution to the next generation. This would drive the evolution of an intrinsic (ie. age-related) lifespan similar to or earlier than that specified by extrinsic causes. The review also provides experimental and survey data to support this hypothesis. For example, populations of a given species in high predation environments show earlier senescence patterns than those in low predation environments (Hughes & Reynolds 2005). As a

1 consequence of this selection process several phenomena are proposed to occur resulting in the complex aging process observed in contemporary organisms.

1.1.1 Mutation accumulation

Due to the asymmetric distribution of selection strength throughout lifespan, mutation accumulation (MA) was proposed as a specific evolutionary cause of senescence (Medawar

1952). This theory proposes that any mutations having deleterious effects late in life will have little or no selective pressure against them and may accumulate in the genome by genetic drift.

Since reproductive effort is primarily focused during early life, natural selection will act most strongly on genes with effects before or during this productive period. An example which fits this model is the mutant allele causing Huntington’s disease. Since the effects of the disease are post-reproductive there is little selection pressure and the allele persists in the population. This theory fits with regard to the aforementioned development of intrinsic mortality to optimize fitness, and also includes the special case of antagonistic pleiotropy.

1.1.2 Antagonistic pleiotropy

An additional level of complexity gives rise to the concept of antagonistic pleiotropy (AP)

(Williams 1957). Like MA, AP refers to the accumulation of late-acting deleterious genes during natural selection. The difference is that in AP the genes in question have beneficial early effects and detrimental late effects. An example of such a gene is any member of the Nox family. These genes produce physiological levels of ROS required for a number of cellular functions but also contribute to late-life pathologies such as hypertension and atherosclerosis (Lambeth 2007).

1.1.3 Disposable soma theory

An organism’s lifespan may not be limited only by late-acting detrimental effects. It is thought that active, energy-intensive mechanisms exist for maintaining the body with age. Such a mechanism is known as a longevity assurance mechanism, or LAM (Speakman & Krol 2010).

The premise of the disposable soma theory (Kirkwood 1977) is that an individual organism has a limited amount of available energy. This energy must be shared between reproductive output and the LAMs, both of which are energy-intensive processes. This observation supports the theory that small animals tend to reproduce early and live short lives (Medawar 1952). Smaller animals tend to have a higher extrinsic mortality risk. As such, a small animal with a high rate of extrinsic mortality ought to invest more energy into reproduction before death occurs extrinsically. It follows that this high and early investment in reproduction comes at the cost of energy expenditure on LAMs. In larger organisms the risk of extrinsic mortality is low leaving ample time for reproduction while allotting more energy into LAMs resulting in a relatively longer lifespan and lower rate of aging. To clarify this idea we can consider the evolution of a large animal from a small ancestor (Speakman & Krol 2010). Initially extrinsic mortality is high, reproduction is early and in high numbers, late-acting detrimental mutations accumulate, and aging is rapid. As the size of the organism increases over generations, extrinsic mortality decreases and selection begins to favour lower intrinsic mortality while selecting against AP and

MA effects. All of these changes favour longer lifespan and reproductive investment is reduced.

While a decrease in reproductive output may seem counterintuitive from an evolutionary perspective, the large animal tends to have fewer offspring with a higher likelihood of survival compared to the many offspring at a lower likelihood of survival seen in the small animal. For

3 example, Asian elephants generally produce single offspring with 80-90 % surviving to adulthood (Mumby et al. 2013), and giant pacific octopuses produce around 56 000 offspring with less than 0.004% surviving to adulthood (Cosgrove & McDaniel 2009).

1.1.4 Heat dissipation limit

A novel theory contradicting the disposable soma theory is the heat dissipation limit proposed recently by Speakman and Krol (2010). These researchers reject the assumptions made by the disposable soma theory and instead argue that the limiting factor concerning reproductive output is the ability to dissipate waste heat. The rejection of this theory is primarily due to the failure to show high-energy requirements of supposed LAMs. Many cellular processes have been implicated in the modulation of aging, none of which when altered to prolong life show any increase in energy demand (van Voorhies et al. 2003). Further to this, larger animals tend to have a lower basal metabolic rate (BMR) per unit mass than smaller animals. This is contradictory to the idea that larger animals have active energy sink LAMs as is required by the disposable soma theory (Speakman & Krol 2010). The authors also argue that during times of reproduction energy reserves are often unlimited resulting in an energy surplus which can support both somatic maintenance and reproduction. Although some of these points are conjectural and lack significant support, the authors reference some convincing studies done on the subject. In a number of food supplementation studies less than a third reported increases in litter/clutch size, and these increases were minimal. This information does suggest that food energy is not a limiting resource in terms of reproductive output. The final premise offered to dismiss the idea of a limited amount of available energy concerns the winter coat seen in many endotherms. This coat serves to prevent heat loss and conserve energy during a

4 time of limited resources. The authors presume that if energy was limited year-round, the animals would maintain the coat year-round to conserve energy. This argument is weak in that the authors fail to consider that as weather warms a thick coat becomes an energy burden by requiring active cooling. Despite these cursory arguments against the disposable soma theory, the arguments given in favour of the heat dissipation limit are fairly strong. It is proposed that the low surface area to volume ratio of larger animals results in a lower capacity to dissipate heat which in turn would limit the amount of energy expended on a task like reproduction. As such, selection will not favour large animals with high investment in early reproduction. In studies on mice, there exists a maximum limit of lactation which cannot be exceeded by increasing food intake. This maximum was not exceeded by increasing litter size or forcing the mothers to expend energy by exercise, but was only exceeded by decreasing the environmental temperature. Mice kept at low temperatures produced more milk and larger offspring, and those at thermoneutral conditions produced less milk and smaller offspring, supporting the heat dissipation limit as the limiting factor in energy usage. Another key finding notes that metabolic rate scales to the surface area to volume ratio (Speakman & Krol 2010).

These theories and observations are best understood by again considering the evolution of a larger size in a small endotherm.. As the animal evolves a larger size the ability to dissipate heat is lessened. This results in a reduction in reproductive investment in order to reduce the production of heat since reproduction is energy intensive. Since reproductive investment is reduced or delayed there is now a selective pressure against the negative effects of AP and MA.

This selective pressure may also encourage LAMs provided that their energy requirement, and subsequent contributed heat generation, is low. The end result is that as organisms evolve into

5 larger forms, reproductive output is reduced and longevity is enhanced due to changing selection pressure on AP, MA, and LAM effects.

Without further testing it is premature accept the heat dissipation theory. This theory only pertains to endotherms and therefore provides only a partial explanation in the evolution of senescence. The lifespan of ectotherms also correlates size with extrinsic mortality risk, lifespan and reproductive output. Although heat dissipation ability appears to be a factor in endotherm life history evolution, it has much less relevance to ectotherms where similar phenomena are observed. This suggests that heat dissipation plays only a small part in lifespan determination.

1.2 Aging mechanisms and interventions

1.2.1 Insulin/insulin-like growth factor (IGF-1) signaling

Molecular mechanisms have been linked to the aging process and many pathways have been shown to affect lifespan (Friedman & Johnson 1988; Kenyon et al. 1993; Wei et al. 2008).

A key pathway with numerous regulatory components is the insulin/IGF signaling (IIS) pathway.

Multiple mutants have been identified that lack functional components of this pathway and subsequently show a change in longevity (Friedman & Johnson 1988; Gerisch et al. 2001; Wei et al. 2008). By way of the IIS and other converging signaling pathways a variety of cellular processes are modulated. These include general stress resistance, cell cycling, apoptotic control, proliferation, and translational control. As such, single changes in signaling can produce disparate cellular responses (Figure 1). Insulin-like signals trigger an activation of phosphatidylinositol-3-kinase (PI3K) of which AGE-1 makes up the catalytic subunit. PI3K

6 phosphorylates phosphatidylinositol 4,5-bisphosphate (PIP2) to form phosphatidylinositol

(3,4,5)-trisphosphate (PIP3). This action is opposed by PTEN which maintains a low PIP3 level in the absence of an insulin-like signal. PIP3 then activates AKT (also known as PKB) which in turn regulates multiple downstream cellular activities.

Figure 1: Simplified diagram illustrating how IIS can affect multiple downstream signaling pathways. Those shown (proliferation, cell cycling, apoptosis, translation) are modulated by AKT (also known as PKB). “CS” denotes catalytic subunit, “PH” denotes pleckstrin homology domain. Adapted from “Development_IGF-1 receptor signaling” by GeneGo, Inc. http://www.genego.com/maps/540_map.png.

1.2.2 Multiplex stress resistance

The most potent single gene influencing lifespan is known as age-1 which encodes a class-I PI3K catalytic subunit as seen in Figure 1 (Ayyadevara et al. 2008). The AGE-1 and DAF

(abnormal dauer formation) pathways converge at DAF-16 nuclear transport and subsequent activation (Figure 2). Consistent with the idea of cross-talk between intracellular signaling pathways presented above, C. elegans age-1 null mutants show resistance to a variety of cellular stresses in addition to the 10-fold lifespan extension. These null worms show significantly higher resistance to thermal stress, oxidative stress, and electrophilic stress relative to wild-type (Ayyadevara et al. 2008). Furthermore, the degree of resistance is often correlated with the degree of lifespan extension associated with the various age-1 alleles (Ayyadevara et al. 2008). Other studies have also shown that mutants of IIS components including age-1 and members of the daf gene family show resistance to UV exposure (Murakami & Johnson 1996) and have significantly higher lethal concentration (LC50) values for the heavy metals copper and cadmium (Barsyte et al. 2001).

This concept of lifespan extension via suppression of IIS is consistent with the disposable soma theory of aging. Under the assumptions of the theory, one would expect that the alteration of a nutrient sensing pathway could influence resource allocation, and thus lifespan.

A lack of functional AGE-1 or DAF-2 increases DAF-16 nuclear translocation resulting in transcriptional activation of target genes (Shmookler Reis et al. 2009). This DAF-16 activation also occurs during periods of stress. In effect, the age-1 or daf-2 mutant acts as if the cell is always under stress. Following the disposable soma theory, during times of stress resources will be shifted away from reproduction and towards somatic maintenance. In this case, insufficient

Figure 2: Overlap of IIS pathway and dauer formation pathway (DAF). A. IIS pathway contributing to DAF-16 activation and longevity. With permission from Neumann-Haefelin et al. (2008) B. Influence of DAF proteins on longevity and dauer formation. Also shown is AGE-1 which similarly to DAF-2 inhibits DAF-16 nuclear localization. Adapted with permission from Cold Spring Harbor C elegans II Section IV, Subsection B: Genetic Pathways, Figure 3, p. 754 (1997).

10 suppression of DAF-16 activity results in a shift towards somatic maintenance, and therefore, lifespan extension. It has also been observed that extended lifespan by age-1 mutation reduces fertility and hence reproductive output (Friedman & Johnson 1988).

Oxidative stress, among other cellular stresses, has major influence on some of the accumulative changes that are thought to contribute to aging (Sohal 2002). This type of stress leads to the accumulation of oxidized proteins in the form of carbonylated amino acid side chains (Stadtman 1993). These modified proteins have impaired function and contribute to physiological decline (Sohal 2002). The quantity of available metabolic energy also provides a measure for the activity of the IIS pathway. The relationship of energy balance and lifespan has been shown to be mediated through AAK-2 protein in C. elegans (Apfeld et al. 2004). Both of these aspects of cell physiology have successfully been used as biomarkers of aging in C. elegans, providing a good measure of true age.

1.2.3 Non-genetic interventions

A number of chemical compounds have been identified that affect aging in laboratory animals. Rapamycin is able to extend lifespan possibly via activation of FOXO (DAF-16) through inhibition of TORC2 complex (Shmookler Reis et al. 2009). Another more commonly known compound is resveratrol. Naturally found in red wine, resveratrol has been shown to promote longevity in C. elegans (Bass et al. 2007). Several other compounds have been shown to promote longevity in laboratory conditions including SRT1720 (Milne et al. 2007; Zarse et al.

2010). Both resveratrol and SRT1720 exert this effect by activation of sirtuins. Sirtuins have been shown to extend lifespan in yeast, possibly through gene silencing via histone deacetylation, and increased sirtuin levels are seen in calorie restricted mice (David Adams &

Klaidman 2007). Although chemical modifiers of aging look very alluring, findings have been mixed overall. Resveratrol has been shown to extend lifespan in some studies (Zarse et al.

2010), but has also been shown to have no effect, or a negative effect in others (Bass et al.

2007). Another issue regarding the use of chemical interventions is the physiologically relevant dosage. In many cases the compounds were studied at doses too high for practical therapeutic use (Zarse et al. 2010). In the case of resveratrol, replication of dosage in humans would require drinking over 12 000 glasses of red wine (David Adams & Klaidman 2007). Resveratrol has more recently been thought to target a variety of other signaling molecules apart from sirtuins. One of these targets is AMP-activated protein kinase (AMPK) which acts to regulate energy balance via carbon utilization and ATP production (Kaeberlein 2010). Interestingly though, these compounds have been shown to augment lifespan through IIS or by mimicking the effect of caloric restriction (Wood et al. 2004; Kaeberlein 2010). The fact that these compounds interact with those pathways already shown to alter lifespan lends credence to the claim of life- extending pharmaceuticals, but many obstacles exist which currently limit clinical applicability.

1.2.4 Germline ablation

Due to the integrative nature of the molecular pathways influencing aging, it comes as little surprise that the germline is intimately related to the aging process. The original age-1 mutant strain showed lifespan extension as well as a marked decrease in fertility (Friedman &

Johnson 1988). In a reverse cause and effect fashion, ablation of the germ-line can lead to lifespan extension (Berman & Kenyon 2006). The results of these studies are consistent with the disposable soma theory and the division of resources; extending lifespan requires energy normally allotted to reproduction. By ablating the germline, the energy is made available for

12 somatic maintenance and long life. Likewise, extending lifespan robs the germline of energy and results in reduced fertility.

1.2.5 Caloric restriction

Dietary restriction has long been known to extend lifespan (McCay et al. 1935). In more recent studies particular mechanisms of action have been proposed. When grown in calorie limiting media, C. elegans larvae were shown to induce a stress response involving DAF-16.

Similar to age-1 null mutants, which exhibit constitutive nuclear localization of DAF-16, starvation induces an acute stress response which also causes nuclear DAF-16 localization

(Weinkove et al. 2006). As such, caloric restriction results in increased longevity and stress resistance by invoking the same DAF-16 transcription factor and stress response as seen in IIS.

A more recent study has implicated the Ras, Tor, and Sch9 signaling pathways in the mediation of the effects of caloric restriction (Wei et al. 2008). These pathways act to integrate nutrient sensing pathways like the IIS with cellular growth and division. Although currently very complex and convoluted, the specific contributions of pathways and their individual protein components are being probed, and their involvement in stress resistance and lifespan extension is being uncovered.

1.3 Schmidtea mediterranea as a model organism for aging

1.3.1 Planarians in research

A range of planarian species have been studied for their many remarkable properties throughout the past two centuries. Planarian regenerative capacity was first studied by zoologist George Shaw in 1790, and was thoroughly evaluated by T. H. Morgan in 1898.

Although planarians have classically been studied for their regenerative properties, there has been a more recent surge in experimental usage due to interest in stem cells. Adult planarians maintain a population of pluripotent stem cells known as neoblasts throughout the majority of their body tissues which orchestrate regeneration and other processes (Baguñà et al. 1989).

Currently, major planarian research is being performed focusing on parasitology (Collins III &

Newmark 2013), regeneration (King & Newmark 2012), apoptosis (Bender et al. 2012), and development (Molina et al. 2011; Auletta et al. 2012).

Schmidtea mediterranea has a fairly small (4.8 x 108 bp), stable, diploid genome with few transposable elements (Saló 2006). Other species of this phylum are known to be mixoploid with varying ploidy states for somatic and germ cells (Saló 2006). S. mediterranea is also available in sexual or asexual strains as determined by a simple Robertsonian translocation (De

Vries et al. 1984). The asexual strain is particularly useful in that the generation of clonal lines can be simple and rapid. These genetically uniform lines can be produced by simply allowing an individual worm to asexually bud or by manually cutting and allowing the worm to regenerate.

Planarians possess many other more typical traits seen in model organisms: they are small, kept in a simple low salt solution, reproduce at high rates, only need to be fed weekly, and flourish at room temperature. Recently, a multitude of molecular interventions have been established for this species including an optimized whole mount in situ hybridization protocol, simple RNA interference using an expression vector system with application by feeding, and specialized protocols for other mainstream procedures like protein/DNA/RNA isolation from tissue

(Sanchez Alvarado & Newmark 1999; Pearson et al. 2009). The genome sequencing project and

14 corresponding database is also a highly valuable tool for study design or in silico study (Robb et al. 2008).

The study of aging has heretofore been performed primarily using C. elegans and D. melanogaster, and also to some extent using mice. Although effective initial choices, each model organism possesses noteworthy shortcomings which can impede longevity studies.

Although mice have value in being more closely related to humans, colony maintenance is expensive, lifespan is relatively long, and experimental manipulations can be complex. The flies and nematodes, both of the superphylum Ecdysozoa, have more desirable traits such as short lifespan, simple colony maintenance, and ease in molecular interventions, but also have their own serious shortcomings. The Ecdysozoans have life history states, dauer and diapause in addition to metamorphosis in Drosophila, which have no parallel in the mammalian life history, and somatic cells are largely, or entirely post-mitotic. Furthermore, it has been shown that C. elegans and Drosophila underwent a significant gene loss event wherein at least 7-9% of genes were lost compared to the chordates (Technau et al. 2005). Taking these characteristics into consideration, S. mediterranea is proposed as a new model organism for aging study which may provide information about the role of unique, physiologically relevant processes on senescence.

1.3.2 Replicative neoblast stem cell population

The primary feature making the planarian (S. mediterranea; terms will herein be used interchangeably) a preferred model organism for aging studies is the presence of pluripotent neoblasts. In C. elegans, age-1 null alleles affect AKT signaling which is required for proper cell replication (Shmookler Reis et al. 2009). In the entirely post-mitotic C. elegans this does not pose a problem, but may be complicated when actively dividing cells are involved. Planarians

15 have an actively replicating cell population. Since organismal aging is thought to occur as a result of the accumulation of senescent cells in tissues (Campisi 2005), the role of tissue regeneration by the action of stem cells must also be important.

1.3.3 Ecdysozoan gene loss

As mentioned, planarians have not suffered the gene loss observed in the Ecdysozoans,

C. elegans and Drosophila (Technau et al. 2005) which occurred more recently than the most common ancestor of the Ecdysozoans and the Lophotrochozoans. Some of these lost genes may be involved in the aging process or in stress resistance. Future genetic screens would benefit greatly by having this body of genes common to both humans and planarians, but not

Ecdysozoans. By using a model organism that underwent this gene loss, we exclude any pertinent genes that may be included in this set. The gene loss is illustrated in Figure 3.

Figure 3: A partial cladogram showing the gene-loss event. Relatedness between the Ecdysozoa (C. elegans and D. melanogaster), the Chordata (H. sapiens), Lophotrochozoa (planarians; S. mediterranea), and the outgroup Cnidaria (Hydra spp) is shown. The proposed gene loss event occurred after the most recent common ancestor between Lophotrochozoa and Ecdysozoa indicated by the asterisk. Cladogram was compiled from Nielsen (2001) and Maddison and Schulz (2007).

1.3.4 Diapause and dauer states, larval instars

Diapause in flies and the dauer state in nematodes are both stages induced during unfavourable environmental conditions wherein the organism appears not to age (Austad

2009). In C. elegans, the organism can remain in dauer state for several months (Weinkove et al. 2006) whereas the normal, non-dauer lifespan of the worm is only about fifteen days

(Ayyadevara et al. 2008). After a return to favourable conditions these organisms are able to exit from these stages and resume their lifespan; That is, their biological clock resumes ticking upon reintroduction from dauer/diapause.

Similarly, the flies and nematodes both have discrete developmental stages demarcated by molting during growth. Drosophila undergo three larval stages, pupate, and then eclose as an adult fly, while nematodes undergo four larval stages which may or may not include dauer formation depending on available resources and temperature.

Planaria have no periods of growth arrest as the Ecdysozoan model organisms do, nor do they develop through discrete life stages. These stages have no human equivalent and as such may present as confounding factors. For example, the original age-1 null allele is considered a dauer constitutive mutation; that is, unless grown at low temperature all mutants remain permanently in the dauer stage (Ayyadevara et al. 2008). In fact, experimenters must raise the mutants at low temperature until all larval stages are complete. Only then may the worms be brought to normal growth temperatures and exhibit long lifespan. Thus the observed lifespan extension could be due to dauer-type stress resistance mechanisms being in effect while the worm is not physiologically or behaviourally in the dauer state. If this is the case, would this mutation have the same effect in an organism without such growth arrested states?

In fact, some life span extension has been seen in higher eukaryotes but with less potency by knocking out the mammalian homolog of igf-1 receptor daf-2 (Holzenberger et al. 2003). It appears that some inference to mammalian systems can be made from studies done in C. elegans, but there must be some additional processes that make up the difference between the

25% lifespan increase seen in mice and the 1000% increase seen in C. elegans. We propose that these processes exist in S. mediterranea and are awaiting discovery.

2. OBJECTIVES

In order to make full use of S. mediterranea as a model organism for aging studies, some groundwork must be completed. We must first characterize normal aging in this species; for example, even the most basic information such as the lifespan of the species is unknown. In fact, many researchers consider the worms immortal based on their endless regeneration capacity (Torre & Ghigo 2015). This study aimed to characterize normal aging using a demographic study based on previous work performed by Mouton et al. (2009). Furthermore, reproductive data was collected and described for this population as reproductive rate has long been shown to correlate with lifespan (Chen et al. 2007; Mukhopadhyay & Tissenbaum 2007).

The study also aimed to adapt or create aging biomarkers to be used as a tool to measure the impact of experimentation on lifespan.

3. MATERIALS AND METHODS

3.1 General culture conditions

3.1.1 Growth media and feeding

All planarians were initially maintained in a 1 x Montjuïch solution (1.6 mM NaCl, 1.0 mM CaCl2, 1.0 mM MgSO4, 0.1 mM MgCl2, 0.1 mM KCl, 1.2 mM NaHCO3, pH 7.0) as specified by

Cebria and Newmark (2005). The demographic colony responded well to this medium and it was used for the duration of the study. The mixed age source colony and age-synchronized colonies experienced periodic bacterial or protozoal infection and in addition to antibiotic treatment (gentamycin, 50 μg/mL) were switched to a 0.5 g/L Sea Salt Mixture (Instant Ocean,

United Pet Group, Blacksburg, VA). Planarians were kept entirely in the dark when not being handled.

Planarians were fed a weekly diet of 1-2 hours ad libitum feeding on pureed veal liver.

The liver was dissected and major connective tissue and blood vessels were removed. The remaining tissue was blended in a Waring blender and frozen in 5 cm Petri plates at -80 °C.

After 1-2 hours of feeding excess liver was discarded and the worms transferred to fresh growth medium and clean containers.

3.1.2 Age-synchronized colonies

In order to test biomarkers of aging, groups of age-synchronized worms were established. Initially, before the lifespan of the species was known, age-synchronized colonies were established every two months. As the demographic study progressed and the lifespan appeared to be much longer than initially predicted, this collection was changed to once every

21 four months. These colonies were maintained in ~175 mL cups with population sizes beginning at around 100 individuals. In order to preserve the uniform age of each colony, fission products

(buds) were removed weekly and recorded for future analysis and/or tracking of the colony health.

3.2 Demographic colony establishment

The demographic study was conducted by following a previous study of the marine planarian, Macrostomum lignano (Mouton et al. 2009). Individual posterior fragments were selected from the population at large and were kept, individually, in 3 mL of media in a well of a

12-well tissue culture plate. Each week the worms were fed 6 μL liver puree for 1-2 hours, any buds were removed and recorded, and any dead worms were assessed for cause of death and also recorded. Occasionally worms would die due to physical injury during handling, desiccation, infection, etc. These individuals were censored and their lifespan and reproductive data were not used in the final analysis.

3.2.1 Collection of anterior lines

Based on the results of classic and modern studies, two hypotheses were made about S. mediterranea. Firstly, this organism undergoes senescence as observed in other asexual metazoans (Martinez & Levinton 1992). Secondly, the posterior fragment of transverse fission is immortal while the anterior fragment senesces (Sonneborn 1930; Newmark & Alvarado 2002).

It is obvious that one of the two fragments must be immortal, otherwise the species would die out within one lifespan. All age-synchronized colonies used in this study were established through the collection of anterior fission products which were assigned an age of zero.

3.2.2 Demographic data curation

Lifespan and reproduction data were noted weekly during feedings. The data were entered into a database program written in Pelles C IDE (Version 6.00.4, Copyright Pelle Orinius

1999-2009). The data were stored in a Microsoft Excel file for export and further analysis.

3.2.3 Lifespan data analysis and model fitting

Basic lifespan data from the elevated temperature cohort were analyzed using mathematical model fitting procedures described by Broadbent (1958) and Pletcher (1999). The standard temperature cohort has not reached 90th percentile in terms of lifespan and therefore cannot be processed by this method. Additional statistical analyses were conducted by Dr.

Daniel Gillis from the Department of Mathematics and Statistics, University of Guelph. Analyses were conducted using the freeware statistical program R (v3.1.2) (R Core Team 2014). Packages downloaded for analyses and data visualization included ggplot2 (v1.0.0) (Wickham & Chang

2013), survival (v2.37-7) (Therneau 2013), multcomp (v1.3-7) (Hothorn et al. 2008), gmodels

(v2.15.4.1) (Warnes 2012), and contrast (v0.19) (Kuhn et al. 2013). All plots were produced using ggplot2, with the exception of Figure 7, which also required the ggsurv function (code freely available at the R-statistics blog [url: http://www.r-statistics.com/2013/07/creating- good-looking-survival-curves-the-ggsurv-function/]).

3.3 Measurement of global protein carbonylation

Lysates were collected by homogenizing whole worms in 100 μL modified Orii buffer

(62.5 mM Tris-HCl buffer pH 6.8, 2% SDS, 10% glycerol, 2% β-mercaptoethanol) (Orii et al. 2005) per mg of worm tissue. The lysate was then treated with OxyBlot™ Kit (EMD Millipore, Billerica,

MA, USA) components to convert carbonyl groups into 2,4-dinitrophenylhydrazone (DNP- hydrazone). The samples were then slot blotted onto nitrocellulose membranes using a 24 slot vacuum blotter (Whatman, GE Healthcare, Baie d’Urfe, QC). The membranes were blocked and probed with an α-DNP antibody (in kit) at 1:150 dilution in block or a monoclonal mouse α-β- actin antibody (Sigma-Aldrich, St. Louis, MO, USA) at 1:5000 dilution in block. Membranes were treated with secondary antibody, 1:300 kit secondary or 1:10 000 HRP conjugated α-mouse

(Thermo Fisher Scientific, Waltham, MA) for DNP and β-actin respectively. Membranes were treated with ECL Plus (GE Healthcare, Baie d’Urfe, QC) and signal was detected with Molecular

Imager® ChemiDoc™ XRS System (Bio-Rad Laboratories, Hercules, CA).

3.4 Measurement of available ATP

Lysates were produced by rapidly homogenizing whole worms in 10 μL ice cold PBS per mg of tissue with a pellet pestle (Sigma-Aldrich, St. Louis, MO, USA). Large cellular debris was removed by centrifugation. A small amount of supernatant was snap frozen in liquid nitrogen where it remained frozen until immediately before assay. Using the EnzyLight™ ADP/ATP Ratio

Assay Kit (BioAssay Systems, Hayward, CA, USA), ATP concentration was measured relative to

ADP concentration using a Turner TD-20e Luminometer (Promega Corporation, Fitchburg, WI).

Initially, a standard curve was performed to confirm that the concentrations were within range of detection (Figure 4).

Figure 4: Single measurements of standard concentrations and ratios of ATP and ADP. A) ADP:ATP ratio measured for three concentrations. B) RLU measurements for ADP and ATP independently, normalized to the 1x sample. 1x – 1 μM ATP and 0.2 μM ADP, 10x – 10 μM ATP and 2 μM ADP, 25x – 25 μM ATP and 5 μM ADP, 100x – 100 μM ATP and 20 μM ADP. The 100x sample saturated the luminometer and is not shown. The expected measurement based on concentration or ratio is shown.

3.5 Cloning of Smed-age-1

3.5.1 RNA isolation and first strand synthesis

Total RNA was isolated from whole organisms using TRIzol reagent (Thermo Fisher

Scientific, Waltham, MA). The mRNA was reverse-transcribed using a SuperScript® III Reverse

Transcriptase (Thermo Fisher Scientific, Waltham, MA) and oligo-dT primers (Sigma-Aldrich, St.

Louis, MO, USA).

3.5.2 Cloning into L4440 dsRNA expression vector

Figure 5 depicts the cloning process with a list of primers provided in Table 3. The age-1 cDNA was amplified using three sets of overlapping primers, each encompassing around 1000 base pairs referred to as fragments A, B and C. Each fragment was amplified using the respective primer sets with the addition of a 5’ HindIII recognition sequence on fragment C and a 3’ NheI recognition site on fragment A. After silica column purification (Bio Basic Canada Inc.,

Markham, ON) fragment B and C were joined by a single polymerase reaction utilizing the forward primer for fragment B, the reverse primer for fragment C, and the overlapping region between them. Fragment BC was then subcloned into the multiple cloning site (MCS) of the

L4440 plasmid making use of an internal XbaI site and the added HindIII site. Finally, fragment A was subcloned into the L4440-BC plasmid using the same internal XbaI site. NheI and XbaI produce compatible ends (5’ CTAG) which allowed the insertion of fragment A which was then screened for proper orientation. L4440 was a gift from Andrew Fire (Addgene plasmid # 1654).

Figure 5: Diagram illustrating the cloning strategy for Smed-age-1 full-length coding sequence.

3.6 Antibody characterization

3.6.1 Antibody generation

Rabbit polyclonal antibodies were raised using peptides NEPEGPTETDQSLSC for Smed-

WI-1, a neoblast marker, and CNNRKKKFGVNRERV for Smed-AGE-1. Both antibodies were produced and affinity purified by Genscript (Piscataway, NJ). Both antibodies were affinity purified.

3.6.2 Western blotting

Whole planarians were homogenized in a modified Laemmli buffer (62.5 mM Tris-HCl pH 6.8, 10% glycerol, 2% SDS, 10% β-mercaptoethanol) previously used by Orii et al. (2005).

Tissues were homogenized using a pellet pestle (Sigma-Aldrich, St. Louis, MO, USA) boiled for

10 minutes and centrifuged to pellet larger debris. Samples were separated by electrophoresis on an 8% SDS-PAGE gel and wet transferred to PVDF membrane. Membranes were blocked in

5% skim milk powder in PBS-Tween (140 mM NaCl, 3 mM KCl, 5 mM NaHPO4, 2 mM KH2PO4,

0.1% Tween-20, pH 7.4) for 1 hour at room temperature (RT), incubated in primary antibody diluted in blocking solution 1 hour to overnight, and secondary antibody in blocking solution for

1 hour. Blots were treated with ECL Plus (GE Healthcare, Baie d’Urfe, QC) and exposed to

Hyperfilm ECL (GE Healthcare, Baie d’Urfe, QC) and developed.

3.6.3 Epitope preincubation

Initial tests of the α-AGE-1 antibody showed a large number of distinct bands. In order to distinguish the non-specific bands from those identifying the protein of interest, epitope preincubation was used. The primary antibody solution was treated with an excess of the same

28 peptide used for antibody generation, ie. the epitope. This should block the antibody from binding the AGE-1 protein and therefore the change in banding pattern will identify the specific target polypeptide.

3.7 siRNA knockdown

A simple system for knocking down gene expression was developed for use in C. elegans

(Timmons & Fire 1998; Timmons et al. 2001), and has since been adapted for use in planarians

(Sanchez Alvarado & Newmark 1999). The L4440 vector was used in this study with the gene of interest inserted in the multiple cloning site set between flanking T7 promoters. These promoters provide locations to initiate transcription both upstream and downstream of the gene thereby producing two fully complementary RNA molecules which then anneal to form a duplex. This transcription may occur in vitro as done in this study, or may be produced by

RNase-deficient E. coli strain HT115. This dsRNA is fed to the subject inducing RNA interference leading to reduction of mRNA and protein levels by siRNA.

3.7.1 In vitro expression of age-1 dsRNA

Large amounts of starting material are produced by PCR amplification of the gene of interest using primers designed against the T7 promoter. This DNA is purified and then transcribed in vitro using T7 polymerase. The original DNA template is degraded using DNaseI treatment. The final dsRNA is purified and analyzed for quality by spectrophotometry

(NanoDrop 8000, Thermo Fisher Scientific, Waltham, MA) and by non-denaturing agarose gel.

3.7.2 Feeding protocol

A mixture of liver puree and dsRNA is made fresh at a concentration of 100 ng/mL of liver plus 1 % food dye. This mixture is fed to a number of worms for 1-2 h and worms which have eaten become stained and may be identified by colour and retained while those that have not eaten are discarded.

3.7.3 Quantitative reverse transcription PCR (qRT-PCR)

The amount of mRNA was measured by qRT-PCR using a SYBR based system. Total RNA was reverse transcribed using qScript supermix (Quanta BioSciences, Inc., Gaithersburg, MD) and qPCR was used to determine relative abundance of mRNA using PerfeCTa® SYBR® Green

FastMix® (Quanta BioSciences, Inc., Gaithersburg, MD) and an Applied Biosystems 7300 Real-

Time PCR System (Waltham, MA). Data were analyzed using the ΔΔCt method (Livak &

Schmittgen 2001).

4. RESULTS

4.1 S. mediterranea kept at an elevated growth temperature shows a reduced lifespan

4.1.1 Survival data shows a significant difference between elevated and ambient temperature cohorts

In addition to basic characterization of flatworm lifespan, the elevated temperature cohort (26 ± 0.5 °C) was kept in an attempt to shorten lifespan as observed previously for ectotherms (Ragland & Sohal 1975; Miquel et al. 1976; Klass 1977). Worms were kept individually and age at death was recorded. These data are presented in Figure 6 as a lifespan distribution and some key statistics are presented in Table 1. A Wilcoxon Signed-Rank test verifies a non-zero shift between the curves with a p-value of 7.4 x 10-10. This statistical test is a non-parametric test used to compare samples which are not assumed to be normally distributed. A non-zero shift means that the data distributions are significantly different.

Survival curves for each cohort are also presented in Figure 7. A Cox proportional hazards model yields a parameter level of -0.812 at a p-value of 1.2 x 10-8, which indicates that the worms at elevated temperature die at approximately 2.3x the rate of the ambient worms. This statistical model relates the amount of time passing (lifespan) before an event occurs (death) in relation to one or more covariates (temperature). The parameter level returned describes the relative difference in the amount of time (lifespan) between the two groups where the fold difference is a function of the parameter level (= eparameter level).

Table 1: Key data points extracted from demographic study of S. mediterranea including cohort size, average (avg), high, and low values for reproduction data.

Figure 6: Lifespan distributions for ambient (20 °C) and elevated (26 ± 0.5 °C) cohorts. Density specifies the proportion of all worms falling into each age range. The median lifespan group for the ambient cohort is 90-120 weeks, and for the elevated temperature cohort is 90 weeks.

4.1.2 Model fitting shows that a Gompertz model best describes lifespan at elevated temperature

The aging process is defined by characteristic physiological changes occurring over the lifespan of an organism eventually ending with death. At every point in the lifespan of a population there exists a mortality rate which is a reflection of these physiological changes. The mortality rate is a measure of normal aging for a given population and is represented by a mortality distribution. For current model organisms used to study aging this information is available and represents normal aging for the species in question. This information describes the changes in mortality rate that occur in an aging population. To develop S. mediterranea as a model for aging research, this demographic information needs to be developed. In order to collect these mortality data and infer important parameters such as initial mortality rate and mortality rate doubling time, a longitudinal study was performed based on a previous design with modification (Mouton et al. 2009). Two large cohorts were established, one at ambient laboratory temperature (20 °C) and another at an elevated temperature (26 ± 0.5 °C). Deaths were recorded and mortality models were fitted to the data (Broadbent 1958; Wilson 1994).

The mortality curve generated (Figure 7) shows a typical sigmoid shape consistent with the commonly observed exponential increase in mortality following reproductive age and the long tail characteristic of exceptionally long-lived individuals. The ambient/room temperature (RT) cohort study is in progress as of the writing of this manuscript whereas the elevated temperature cohort has surpassed the 90th percentile. As such, the mortality data of the elevated temperature cohort was fit to a Gompertz distribution, a probability distribution commonly applied to lifespan, with the coefficients of the fit presented in Table 2. A Gompertz

34 coefficient (b) of 0.145 gives a mortality rate doubling time (MRDT) of 4.79 time intervals (r) which equates to 0.46 years. This means that the likelihood of death at any given moment doubles every 0.46 years. This value falls within the range indicative of gradual senescence

(Finch 1990).

Table 2: Gompertz coefficients and mortality rate doubling time. Ambient cohort remains to be determined (TBD) due to exceptionally long lived individuals.

Figure 7: Survivorship curve for ambient (20 °C) and elevated (26 ± 0.5 °C) cohorts. The curves were generated using lifespan data and a Cox Proportional Hazards Model. Median survival is indicated for each cohort, ambient being 81 weeks and elevated 55 weeks.

4.2 Elevated growth temperature decreases reproductive rate at all lifespans

The effect of temperature on lifespan, mortality rate and distribution is displayed in

Figures 6 and 7. Since lifespan and reproduction are closely linked, we sought to examine the effect of temperature and lifespan on reproductive rate by plotting total reproductive rate against lifespan for each growth temperature (Figure 8). A LOESS curve, a non-parametric technique combining multiple regression models, was fit to the data to show trends. Our first observation is that each cohort has a unique profile but it is clear that total reproduction rate is higher for all lifespans at ambient temperature compared to the elevated temperature cohort.

Another observation is that in the ambient cohort, peak total reproductive rate occurs near 50 weeks of age but remains relatively stable for all lifespans in the elevated temperature cohort.

Figure 8: Total reproduction rate by life span with LOESS spline by treatment. Treatments include ambient (20 °C) and elevated (26 ± 0.5 °C) temperatures. Each data point represents an individual planarian and its lifelong reproductive rate. Note that for the elevated temperature cohort, there were 13 individuals who lived ~52 weeks or less that had no reproduction.

4.3 Long-lived ambient temperature worms show a resurgence in weekly reproductive rate in late life

In addition to examining the relationship between lifespan and total reproductive output, a more detailed analysis was performed by subdividing the ambient temperature cohort into short-lived (bottom 25th percentile) and long-lived (top 25th percentile) groups. Both weekly survival rate and weekly reproduction rate were plotted (Figure 9). The short-lived group shows a steady and rapid decline in both reproduction rate and survival rate with increasing age. The long-lived group shows a much slower fall in survival rate and in late life shows a resurgence in reproductive rate.

Figure 9: Reproduction and survival rates for short-lived and long-lived groups of worms. Short- lived (SL: bottom 25th percentile of lifespan) and long-lived (LL: top 25th percentile of lifespan) ambient temperature worms were analyzed for reproductive rate and survival rate through rolling intervals of 11 weeks beginning at week 3. The reproductive rate and survival rate for each interval are plotted. Confidence intervals are shaded for each LOESS curve. A Poisson regression was applied and tested by Tukey’s post hoc analysis showing a significant difference between short-lived and long-lived reproductive rate. In addition, long-lived reproduction rate began increasing after reaching 100 weeks of age.

4.4 Reproductive events become more frequent with each successive event when at elevated temperature but not at ambient temperature

In addition to analyzing the frequency and magnitude of reproduction, we also wanted to determine the interval between reproductive events for each cohort. We hypothesized that planaria at higher temperatures would have a longer overall time interval between reproductive events given the overall lower lifelong reproductive rate, and a shorter time interval between successive reproductive events in order to explain the increase in total reproduction at longer lifespans shown in Figure 8. Figure 10 shows the median time between reproductive events for each cohort. In support of our hypothesis, reproductive events become gradually more frequent with advancing age at elevated temperatures contrary to the relatively constant interval observed in the ambient cohort

Figure 10: Median time to next reproductive event plotted against current reproductive event. Numbers below each line indicate the number of individuals used to calculate each median. Treatments include ambient (20 °C; gray) and elevated (26 ± 0.5 °C; red) temperatures. Note: At the beginning of the experiment, 100 planaria at the elevated temperature had approximately 9 weeks between reproductive events, while 96 planaria at the ambient temperature cohort had 3 weeks between reproductive events. Also note that the longest lived planarian at the elevated temperature reproduces once every 6 weeks.

4.5 Global protein carbonylation does not significantly increase with advancing age

We have shown by demographic analysis that this planarian species undergoes senescence. Several established biomarkers of aging in other species were examined for applicability in this species. Such biomarkers will be useful as an instantaneously measurable metric for cellular senescence, and will also provide insight into the underlying mechanisms of senescence. In a number of species cellular proteins undergo irreversible oxidative damage and accumulate with advancing age (Levine 2002; Nystrom 2005) typically forming aldehyde or ketone groups on side chains of lysine, arginine, proline and threonine residues (Stadtman

1993). Although the cause of this accumulation is currently unknown, elevated levels of protein carbonyl content are observed in a number of studies occurring in a number of tissues and organisms (Levine 2002). To examine this biomarker, carbonylated proteins were quantified and normalized (Figure 11A) using a kit (Protein Carbonyl Content Assay Kit, Abcam). No significant difference was observed between young (4-5 week old) and old (122 week old) samples.

4.6 ATP availability decreases with advancing age

Another well established biomarker concerns available metabolic energy as measured by ATP concentration (Collins et al. 2008). The concentration of ATP is a product of catabolic processes and represents the available energy content of the cell which itself represents the metabolic status of the cell. An age-dependent decline in ATP concentration was observed in C. elegans and this decline can be blocked by interfering with daf family genes (Braeckman et al.

2002). ATP content was measured in four age groups using a luminescent assay. Since the luminometer returns relative light units ADP content was also measured to normalize ATP

43 content against in order to compare different samples. A statistically significant increase in the ratio, which implies a decrease in ATP concentration, was observed over the course of approximately 125 weeks but there was no significant difference between any lesser time span

(Figure 11B).

Figure 11: Two putative aging biomarkers were tested for use with S. mediterranea. (A) Global protein carbonylation was quantified using the OxyBlot™ Kit and (B) ADP:ATP ratio was measured using EnzyLight™ ADP/ATP Ratio Assay Kit. There is a significant difference in ADP:ATP ratio between the youngest (2-5 weeks old) and oldest (127-131 weeks old) animals.

4.7 SMED-AGE-1 presents as a 140 kDa species by western blot

The Smed-age-1 coding sequence is 3081 base pairs corresponding to a 1026 amino acid polypeptide. This polypeptide is predicted to have a molecular weight of 119 kDa by Compute pI/Mw tool hosted by ExPASy resource portal (Gasteiger et al. 2003). Western blotting using our polyclonal α-AGE-1 antibody results in a characteristic banding pattern of three, always present, major bands and up to six minor bands which may or not be visible depending on total protein concentration and antibody dilution. The largest of the three major bands is eliminated through pre-incubation of peptide epitope, and is not eliminated in the non-specific peptide

(WI1) negative control (Figure 12). This band presents as a ~140 kDa species, running ~20 kDa larger than expected, which may be a result of post-translational modification or differences in buffer between the ladder and the lysates.

Figure 12: Epitope preincubation of polyclonal rabbit α-AGE1 antibody (GenScript). An arrow indicates which band corresponds to the protein of interest. Primary antibody dilution in blocking solution is given plus the corresponding peptide epitope added. A size standard is given on the left with sizes expressed in kDa.

4.8 siRNA knockdown of age-1 reduces expression levels to 20 % of endogenous levels and lasts between four and five weeks

To assess the effects of genetic knockdowns on aging it is useful to know how long a single treatment will last. Worms were fed dsRNA representing a large portion of the age-1 coding sequence. Individuals were harvested weekly and mRNA levels measured by qRT-PCR.

Knockdown of age-1 was reduced to approximately 20 % of normal levels and persisted for at least four weeks (Figure 13).

Figure 13: RNAi knockdown of Smed-age-1 duration and intensity. qRT-PCR measurements of age-1 mRNA quantity were taken weekly and normalized to gapdh after RNAi feeding.

Expression drops by 80% for 4-5 weeks before returning to or exceeding wildtype levels.

5. DISCUSSION

When modeling human aging it is advantageous to use a model system with a similar aging rate; we have shown that this is true for S. mediterranea. The MRDT, the length of time required for mortality rate to double, of 0.46 years is desirable in that, while considered to describe gradual senescence (between 0.3 to 10 year MRDT), the organism still has a high mortality rate compared to humans with an MRDT of around 8 years (Finch 1990; Wilson 1994).

This organism permits the study of aging including the effects of stem cells on tissue replacement and repair while still senescing at a high rate compared to humans. As such, S. mediterranea could provide novel insight into mammalian aging more quickly and easily than using mammalian models. The results of this work represent the first demographic characterization for this species, and only the second for all flatworm species. The resultant data provide a framework that future investigators can use to design and execute aging studies.

The mortality distributions presented in Figure 7 illustrate the typical sigmoid shape observed for other gradually senescing species including rodents, birds, and domestic dogs

(Finch & Pike 1996), showing a decline in mortality rate at advanced age represented as a long tail. This sigmoid shape is also observed in populations of C. elegans (Stroustrup et al. 2013) and

Drosophila (Linford et al. 2013), which are both categorized as rapidly senescing species. Based on the Gompertz coefficients produced by the model fitting procedure for the elevated temperature cohort, S. mediterranea exhibits the characteristics of a good model for aging research since its mortality rate doubling time falls in the same category as humans (Finch et al.

1990). In addition, the elevated temperature cohort illustrates the effect of temperature on senescence as was previously observed in other ectotherms both in wild and laboratory

49 conditions (Miquel et al. 1976; Klass 1977). Despite their exceptional regenerative capability and ability to maintain telomere ends (Tan et al. 2012), this study shows that planarians do undergo senescence. In contrast, a lack of senescence has been demonstrated in hydra, a genus also characterized by asexually reproducing individuals with a high degree of stem cell activity

(Martinez 1998). Future research could focus on differences in these organisms that may cause flatworms to age while hydra appear non-senescent.

The use of S. mediterranea as a model organism for aging research is largely predicated on the presence of actively dividing totipotent stem cells known as neoblasts. These cells represent approximately 30 % of total cell population in these planarians (Baguñà et al. 1989).

The importance of stem cell activity and tissue replacement in organismal aging is suggested by a number of studies in a few species of planaria (Lange 1968; Nimeth et al. 2002; De Mulder et al. 2010) in which ablation of all stem cells is accomplished by RNAi, by γ-irradiation, or chemical inhibition of the cell cycle. Upon removal of the stem cell population lifespan is drastically shortened. Senescent cells exhibit characteristic changes in physiology including adapting a secretory phenotype and exhibiting changes in cell morphology. There also exist changes in gene expression patterns (Coppe et al. 2008). Targeted apoptosis of senescent cells in mice has been shown to extend lifespan (Baker et al. 2011). These findings may also indicate cellular replacement and tissue repair act as an opposing force to cellular senescence and therefore organismal senescence. The exceptional lifespan of S. mediterranea, especially when compared to other simple ectotherms such as C. elegans, is consistent with this proposed involvement of stem cells in aging. This discrepancy in lifespan between S. mediterranea and the closely related species P. velata, with a lifespan of one to two months (Austad 2009), is

50 worth further investigation. Any differences in stem cell number or function could further implicate their involvement in aging.

Kirkwood’s disposable soma theory (Kirkwood 1977) proposes a trade-off between reproduction and somatic maintenance as a result of limiting resources. The analysis of reproduction and lifespan presented here is in concordance with this concept. Figure 8 illustrates an overall fall in total reproductive rate with advancing age, although when grouped by lifespan (Figure 9) we see long-lived individuals have a resurgence in weekly reproductive rate in late life. The fact that this late-life increase in weekly reproductive rate in the long-lived group has little effect on total reproductive rate in all groups (Figure 8) could indicate a distinct subpopulation experiencing this phenomenon. In addition to the overall declining trend in total reproduction rate, each cohort displays a unique profile in terms of the shape of each LOESS spline. The ambient cohort shows a prominent peak around the 50 week lifespan point. This means that individuals with this lifespan tend to have the highest total reproductive rate. This is consistent with the “live fast, die young” concept proposed by Kirkwood. In contrast, the elevated temperature cohort appears to peak later and much more modestly at around 100 weeks. This peak is also preceded by a trough around the 50 week lifespan. This trough and subsequent delay in peak reproductive rate may be due to an adjustment period associated with the elevated growth condition. When this colony was established, new offspring were taken from an ambient colony and placed for growth at the elevated temperature. Under new growth conditions, the worms may have undergone physiological or epigenetic changes as acclimation. Further analysis could confirm the existence of this adjustment period by

51 comparing newly introduced elevated temperature individuals with those that have been maintained at elevated temperature for a number of generations.

Time between reproductive events is displayed in Figure 10. Despite the overall lower lifelong reproductive rate at elevated temperatures, this cohort shows a decreasing trend in timing between reproductive events. With each reproductive event the time until each successive event lessens. At first this appears to contradict the general trend of decreasing reproduction with age, but this decreasing rate actually supports the upward trend observed in

Figure 8 after week 50. This analysis also only includes lifespan indirectly and only looks at the timing between events. For example, the third reproductive event for one individual may occur at age 5 weeks while occurring at age 40 weeks for another worm. This trend is not observed in the ambient cohort, which may in fact show a slight increase in time between events. It is possible that the downward trend in the elevated cohort reveals the existence of a distinct subpopulation of high output, high frequency reproductive worms. Such a population would be obscured by lower output worms until their reproductive period ceases. This phenomenon may also be present in the ambient cohort but is obscured by the rest of the population. Perhaps these subpopulations could be identified and isolated by observing individual reproductive rates and then those groups with low reproductive rates could be compared to those with high reproductive rates..

In order to make full use of S. mediterranea as an aging model, biomarkers of aging will need to be confirmed or discovered. A number of biomarkers that correlate with senescence have already been established (Dimri et al. 1995; Krishnamurthy et al. 2004; Collins et al. 2008;

Park et al. 2009). These biomarkers represent consistent changes with age involving behaviour,

52 tissue organization, metabolism, gene expression and many other facets of physiology (Collins et al. 2008). This planarian species is a relatively long-lived species, which presents a challenge since longitudinal studies of lifespan-altering interventions are impractical. Biomarkers present an alternative metric for assessing the effect of such interventions. We investigated ATP availability and global protein carbonylation, two established biomarkers in other aging models, as possible biomarkers for aging in S. mediterranea (Dimri et al. 1995; Krishnamurthy et al.

2004; Collins et al. 2008; Park et al. 2009). The results of these studies show that neither is an effective biomarker in this species (Figure 11). Global protein carbonylation shows no change between two extremes of age and ATP availability as assessed by ADP:ATP ratio only shows a modest change between two extremes in age. An effective biomarker would serve to measure fine distinctions between ages on a practical timescale on the order of several weeks or months at the most. This is necessary to measure the effect of various interventions on the aging process in a reasonable amount of time. One feature of note is that in both assays the older samples tend to have a smaller standard error. In addition to the aforementioned data concerning total reproductive rate, this could also suggest the existence of subpopulations of worms that age at a different rate than the rest of the population. This would interfere with the function of these aging biomarkers and may explain the lack of observable differences. Since the planarian strain used is highly inbred to the point of possibly being clonal, epigenetic changes may well be responsible for the variation seen in this species.

6. CONCLUSIONS AND FUTURE DIRECTIONS

This study intended to complete the first steps required to establish S. mediterranea as a model organism for aging research. Demographic analysis represents an essential part of this process. We have shown that this species falls into the same category of senescence as humans and has provided a thorough understanding of the senescent profile, including MRDT, maximum and median lifespan. For future study this work will serve as an important reference tool for experimental design. In addition, the change and general decline in reproductive output is consistent with a senescing species.

The two putative biomarkers, global protein carbonylation and ATP availability, showed limited utility. The small amount of change seen in ADP:ATP ratio was over a period of two and a half years which is too long to be practical. It is likely that the cellular changes that occur in other organisms are obscured by the stem cell activity of these worms. It has been shown that telomere length is maintained and telomerase activity is high in these worms (Tan et al. 2012), so perhaps other processes are at work in neoblasts reducing or preventing characteristic accumulative changes. To address the role of neoblasts in aging, a few approaches could be taken. One could impair neoblast function by siRNA against cell cycle genes, or by γ-irradiation followed by measurement of putative biomarkers which have not yet been established for S. mediterranea such as fluorescent pigment accumulation, morphological changes, accumulation of DNA damage, etc. (Collins et al. 2008).

Several additional biomarkers were tested in this study. A number of physical and chemical stressors were implemented and measured by physiological and behavioural changes.

The stressors tested included heat, ultraviolet light, oxidative stress caused by hydrogen

54 peroxide exposure, and electrophilic stress caused by N-ethylmaleimide which represents the end result of biological lipid peroxidation. These treatments were tested at a variety of concentrations and exposure times, and the effects were measured by behavioural observations. General observations were made and it was decided that self-righting behaviour would serve as an appropriate means of measurement. In an unstressed condition each worm will instantaneously right itself when inverted. As applied stress is intensified this reaction time is reduced until reaching unresponsiveness. Although the general trend is present, there was immense variability within the control groups at a given stress dosage. This variability also occurred in the time elapsed from treatment until death. As such, no experimental manipulation was performed.

An important conclusion of this work is that S. mediterranea experiences senescence including reproductive decline despite a lack of measured biochemical changes with age, most notably telomere length (Tan et al. 2012). Many have considered telomere shortening, the agent specifying the Hayflick limit, a key feature limiting lifespan. These conflicting results raise the question, what causes aging in these worms if not these well-established accumulative or degenerative changes? This seems to indicate that a major piece of the puzzle is still missing.

This species reproduces by transverse fission producing asymmetrical products. By necessity, one of these fragments must be immortal in order for the species to survive, and because we see senescence, the other fragment must be mortal. The differences between these two fragments could provide an enormous amount of insight into senescence.

Furthermore, if the anterior fragment ages and the posterior does not, we encounter an interesting problem: After transverse fission the posterior fragment must undergo some sort of

55 rejuvenating process. Such processes may also be unique to this organism and anti-aging in nature. If we can discover the processes by which this organism can rejuvenate, we can infer that the systems or pathways acted upon must be progeric and lead to senescence.

Furthermore, we can look for similar processes in mammals or possibly use their mechanism in combating human age-related disease.

There has been a strong association with stress response and aging (Minois 2000). As such, it would be beneficial to also gain an understanding of the planarian stress response.

Once again, the stress response has not been studied in planarians and a great deal of work must be done to understand how these worms respond to various types of stress before any real experimentation can be done. A good starting point would be a comparison of heat shock response with or without age-1 knockdown since the C. elegans age-1 knockout showed multiplex stress resistance (Kenyon et al. 1993).

Further biomarker testing is also warranted. This study only considered two biochemical attributes, both of which have close ties to metabolism: carbonylation occurs as a result of metabolic ROS byproducts and ADP:ATP ratio represents available cellular energy. A number of other biomarkers have been shown to be relevant in C. elegans and it would be well worth testing and confirming the effectiveness of several more. We have shown that this species does senesce, so it must be true that some aspects of cell physiology are changing and would serve as biomarkers of aging. Whether testing biomarkers already demonstrated in other organisms or searching for novel markers, it is important to continue the search. These markers will have great value in quantifying the effect of age-related interventions and would be great tools for

56 screening subpopulations to distinguish the short-lived from the long-lived either independently, or in concert with the aforementioned reproductive rate assessment.

Some fundamental tools for the study of aging using this model system have been established by this study. Future longitudinal studies may rely on these demographic data to determine appropriate sample sizes and experimental timelines. Several biomarkers have been tested and ruled out for high resolution age-group discrimination. This may inform future experimenters of potential pitfalls and provide guidelines for design. This collection of findings provides a firm starting point for aging studies in a unique organism with features relevant to human aging and possibly leading to valuable novel discoveries in this field of study.

7. REFERENCES Apfeld J, O'Connor G, McDonagh T, DiStefano PS, Curtis R (2004). The AMP-activated protein kinase AAK-2 links energy levels and insulin-like signals to lifespan in C. elegans. Genes & development. 18, 3004-3009. Auletta G, Adell T, Colagè I, D’Ambrosio P, Salò E (2012). Space Research Program on Planarian Schmidtea Mediterranea’s Establishment of the Anterior-Posterior Axis in Altered Gravity Conditions. Microgravity Science and Technology. 24, 419-425. Austad SN (2009). Is there a role for new invertebrate models for aging research? The journals of gerontology. Series A, Biological sciences and medical sciences. 64, 192-194. Ayyadevara S, Alla R, Thaden JJ, Shmookler Reis RJ (2008). Remarkable longevity and stress resistance of nematode PI3K-null mutants. Aging cell. 7, 13-22. Baguñà J, Salo E, Auladell C (1989). Regeneration and pattern formation in planarians. III. Evidence that neoblasts are totipotent stem cells and the source of blastema cells. Development. 107, 77-86. Baker DJ, Wijshake T, Tchkonia T, LeBrasseur NK, Childs BG, van de Sluis B, Kirkland JL, van Deursen JM (2011). Clearance of p16Ink4a-positive senescent cells delays ageing- associated disorders. Nature. 479, 232-236. Barsyte D, Lovejoy DA, Lithgow GJ (2001). Longevity and heavy metal resistance in daf-2 and age-1 long-lived mutants of Caenorhabditis elegans. FASEB J. 15, 627-634. Bass TM, Weinkove D, Houthoofd K, Gems D, Partridge L (2007). Effects of resveratrol on lifespan in Drosophila melanogaster and Caenorhabditis elegans. Mechanisms of ageing and development. 128, 546-552. Bender CE, Fitzgerald P, Tait SW, Llambi F, McStay GP, Tupper DO, Pellettieri J, Alvarado AS, Salvesen GS, Green DR (2012). Mitochondrial pathway of apoptosis is ancestral in metazoans. Proceedings of the National Academy of Sciences. 109, 4904-4909. Berman JR, Kenyon C (2006). Germ-cell loss extends C. elegans life span through regulation of DAF-16 by kri-1 and lipophilic-hormone signaling. Cell. 124, 1055-1068. Braeckman BP, Houthoofd K, De Vreese A, Vanfleteren JR (2002). Assaying metabolic activity in ageing Caenorhabditis elegans. Mechanisms of ageing and development. 123, 105-119. Broadbent S (1958). Simple Mortality Rates. Roy Stat Soc C-App. 7, 86-95. Campisi J (2005). Senescent cells, tumor suppression, and organismal aging: good citizens, bad neighbors. Cell. 120, 513-522. Cebria F, Newmark PA (2005). Planarian homologs of netrin and netrin receptor are required for proper regeneration of the central nervous system and the maintenance of nervous system architecture. Development. 132, 3691-3703. Chen J, Senturk D, Wang JL, Muller HG, Carey JR, Caswell H, Caswell-Chen EP (2007). A demographic analysis of the fitness cost of extended longevity in Caenorhabditis elegans. The journals of gerontology. Series A, Biological sciences and medical sciences. 62, 126-135. Collins III JJ, Newmark PA (2013). It's no fluke: the planarian as a model for understanding schistosomes. PLoS pathogens. 9, e1003396.

Collins JJ, Huang C, Hughes S, Kornfeld K (2008). The measurement and analysis of age-related changes in Caenorhabditis elegans. WormBook : the online review of C. elegans biology, 1-21. Coppe JP, Patil CK, Rodier F, Sun Y, Munoz DP, Goldstein J, Nelson PS, Desprez PY, Campisi J (2008). Senescence-associated secretory phenotypes reveal cell-nonautonomous functions of oncogenic RAS and the p53 tumor suppressor. PLoS biology. 6, 2853-2868. Cosgrove JA, McDaniel NG (2009). Super suckers : the giant Pacific octopus and other cephalopods of the Pacific coast. Madeira Park, BC: Harbour Pub. David Adams J, Klaidman LK (2007). Sirtuins, nicotinamide and aging: a critical review. Letters in Drug Design & Discovery. 4, 44-48. De Mulder K, Kuales G, Pfister D, Egger B, Seppi T, Eichberger P, Borgonie G, Ladurner P (2010). Potential of Macrostomum lignano to recover from gamma-ray irradiation. Cell and tissue research. 339, 527-542. De Vries E, Baguñà J, Ball I (1984). Chromosomal polymorphism in planarians (Turbellaria, Tricladida) and the plate tectonics of the western Mediterranean. Genetica. 62, 187-191. Dimri GP, Lee X, Basile G, Acosta M, Scott G, Roskelley C, Medrano EE, Linskens M, Rubelj I, Pereira-Smith O, et al. (1995). A biomarker that identifies senescent human cells in culture and in aging skin in vivo. Proceedings of the National Academy of Sciences of the United States of America. 92, 9363-9367. Finch CE (1990). Longevity, Senescence, and the Genome. Chicago: The University of Chicago Press. Finch CE, Pike MC (1996). Maximum life span predictions from the Gompertz mortality model. The journals of gerontology. Series A, Biological sciences and medical sciences. 51, B183- 194. Finch CE, Pike MC, Witten M (1990). Slow mortality rate accelerations during aging in some animals approximate that of humans. Science. 249, 902-905. Friedman DB, Johnson TE (1988). A mutation in the age-1 gene in Caenorhabditis elegans lengthens life and reduces hermaphrodite fertility. Genetics. 118, 75-86. Gasteiger E, Gattiker A, Hoogland C, Ivanyi I, Appel RD, Bairoch A (2003). ExPASy: The proteomics server for in-depth protein knowledge and analysis. Nucleic acids research. 31, 3784-3788. Gavrilov LA, Gavrilova NS (2002). Evolutionary theories of aging and longevity. TheScientificWorldJournal. 2, 339-356. Gerisch B, Weitzel C, Kober-Eisermann C, Rottiers V, Antebi A (2001). A hormonal signaling pathway influencing C. elegans metabolism, reproductive development, and life span. Developmental cell. 1, 841-851. Holzenberger M, Dupont J, Ducos B, Leneuve P, Geloen A, Even PC, Cervera P, Le Bouc Y (2003). IGF-1 receptor regulates lifespan and resistance to oxidative stress in mice. Nature. 421, 182-187. Hothorn T, Bretz F, Westfall P (2008). Simultaneous inference in general parametric models. Biometrical Journal. 50, 346-363. Hughes KA, Reynolds RM (2005). Evolutionary and mechanistic theories of aging. Annual review of entomology. 50, 421-445.

Kaeberlein M (2010). Resveratrol and rapamycin: are they anti-aging drugs? BioEssays : news and reviews in molecular, cellular and developmental biology. 32, 96-99. Kenyon C, Chang J, Gensch E, Rudner A, Tabtiang R (1993). A C. elegans mutant that lives twice as long as wild type. Nature. 366, 461-464. King RS, Newmark PA (2012). The cell biology of regeneration. The Journal of cell biology. 196, 553-562. Kirkwood TB (1977). Evolution of ageing. Nature. 270, 301-304. Klass MR (1977). Aging in the nematode Caenorhabditis elegans: major biological and environmental factors influencing life span. Mechanisms of ageing and development. 6, 413-429. Krishnamurthy J, Torrice C, Ramsey MR, Kovalev GI, Al-Regaiey K, Su L, Sharpless NE (2004). Ink4a/Arf expression is a biomarker of aging. The Journal of clinical investigation. 114, 1299-1307. Kuhn M, Weston S, Wing J, Forester J, Thaler T (2013). Contrast: A collection of contrast methods. R package version 0.19. Lambeth JD (2007). Nox enzymes, ROS, and chronic disease: an example of antagonistic pleiotropy. Free radical biology & medicine. 43, 332-347. Lange CS (1968). A possible explanation in cellular terms of the physiological ageing of the planarian. Experimental gerontology. 3, 219-230. Levine RL (2002). Carbonyl modified proteins in cellular regulation, aging, and disease. Free radical biology & medicine. 32, 790-796. Linford NJ, Bilgir C, Ro J, Pletcher SD (2013). Measurement of lifespan in Drosophila melanogaster. Journal of visualized experiments : JoVE. Liu L, Rando TA (2011). Manifestations and mechanisms of stem cell aging. The Journal of cell biology. 193, 257-266. Livak KJ, Schmittgen TD (2001). Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods. 25, 402-408. Martinez DE (1998). Mortality patterns suggest lack of senescence in hydra. Experimental gerontology. 33, 217-225. Martinez DE, Levinton JS (1992). Asexual metazoans undergo senescence. Proceedings of the National Academy of Sciences of the United States of America. 89, 9920-9923. McCay C, Crowell MF, Maynard L (1935). The effect of retarded growth upon the length of life span and upon the ultimate body size. J nutr. 10, 63-79. Medawar PB (1952). An unsolved problem of biology: an inaugural lecture delivered at University College, London, 6 December, 1951: HK Lewis and Company. Medvedev ZA (1990). An attempt at a rational classification of theories of ageing. Biological Reviews. 65, 375-398. Milne JC, Lambert PD, Schenk S, Carney DP, Smith JJ, Gagne DJ, Jin L, Boss O, Perni RB, Vu CB, Bemis JE, Xie R, Disch JS, Ng PY, Nunes JJ, Lynch AV, Yang H, Galonek H, Israelian K, Choy W, Iffland A, Lavu S, Medvedik O, Sinclair DA, Olefsky JM, Jirousek MR, Elliott PJ, Westphal CH (2007). Small molecule activators of SIRT1 as therapeutics for the treatment of type 2 diabetes. Nature. 450, 712-716. Minois N (2000). Longevity and aging: beneficial effects of exposure to mild stress. Biogerontology. 1, 15-29.

Miquel J, Lundgren PR, Bensch KG, Atlan H (1976). Effects of temperature on the life span, vitality and fine structure of Drosophila melanogaster. Mechanisms of ageing and development. 5, 347-370. Molina MD, Neto A, Maeso I, Gómez-Skarmeta JL, Saló E, Cebrià F (2011). Noggin and noggin- like genes control dorsoventral axis regeneration in planarians. Current Biology. 21, 300- 305. Mouton S, Willems M, Back P, Braeckman BP, Borgonie G (2009). Demographic analysis reveals gradual senescence in the flatworm Macrostomum lignano. Frontiers in zoology. 6, 15. Mukhopadhyay A, Tissenbaum HA (2007). Reproduction and longevity: secrets revealed by C. elegans. Trends in cell biology. 17, 65-71. Mumby HS, Courtiol A, Mar KU, Lummaa V (2013). Birth seasonality and calf mortality in a large population of Asian elephants. Ecology and evolution. 3, 3794-3803. Murakami S, Johnson TE (1996). A genetic pathway conferring life extension and resistance to UV stress in Caenorhabditis elegans. Genetics. 143, 1207-1218. Neumann-Haefelin E, Qi W, Finkbeiner E, Walz G, Baumeister R, Hertweck M (2008). SHC- 1/p52Shc targets the insulin/IGF-1 and JNK signaling pathways to modulate life span and stress response in C. elegans. Genes & development. 22, 2721-2735. Newmark PA, Alvarado AS (2002). Not your father's planarian: a classic model enters the era of functional genomics. Nat Rev Genet. 3, 210-219. Nimeth K, Ladurner P, Gschwentner R, Salvenmoser W, Rieger R (2002). Cell renewal and apoptosis in macrostomum sp. [Lignano]. Cell biology international. 26, 801-815. Nystrom T (2005). Role of oxidative carbonylation in protein quality control and senescence. The EMBO journal. 24, 1311-1317. Orii H, Sakurai T, Watanabe K (2005). Distribution of the stem cells (neoblasts) in the planarian Dugesia japonica. Development genes and evolution. 215, 143-157. Park SK, Kim K, Page GP, Allison DB, Weindruch R, Prolla TA (2009). Gene expression profiling of aging in multiple mouse strains: identification of aging biomarkers and impact of dietary antioxidants. Aging cell. 8, 484-495. Pearson BJ, Eisenhoffer GT, Gurley KA, Rink JC, Miller DE, Sanchez Alvarado A (2009). Formaldehyde-based whole-mount in situ hybridization method for planarians. Developmental dynamics : an official publication of the American Association of Anatomists. 238, 443-450. Pletcher SD (1999). Model fitting and hypothesis testing for age-specific mortality data. Journal of Evolutionary Biology. 12, 430-439. R Core Team (2014). R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. URL http://www.R-project.org/. Ragland SS, Sohal RS (1975). Ambient temperature, physical activity and aging in the housefly, Musca domestica. Experimental gerontology. 10, 279-289. Riddle DL (1997). C. elegans II. Plainview, N.Y.: Cold Spring Harbor Laboratory Press. Robb SM, Ross E, Sanchez Alvarado A (2008). SmedGD: the Schmidtea mediterranea genome database. Nucleic acids research. 36, D599-606. Saló E (2006). The power of regeneration and the stem‐cell kingdom: freshwater planarians (Platyhelminthes). BioEssays : news and reviews in molecular, cellular and developmental biology. 28, 546-559.

Sanchez Alvarado A, Newmark PA (1999). Double-stranded RNA specifically disrupts gene expression during planarian regeneration. Proceedings of the National Academy of Sciences of the United States of America. 96, 5049-5054. Shmookler Reis RJ, Bharill P, Tazearslan C, Ayyadevara S (2009). Extreme-longevity mutations orchestrate silencing of multiple signaling pathways. Biochimica et biophysica acta. 1790, 1075-1083. Sohal RS (2002). Role of oxidative stress and protein oxidation in the aging process. Free radical biology & medicine. 33, 37-44. Sonneborn TM (1930). Genetic studies on Stenostomum incaudatum (nov. spec.). I. The nature and origin of differences among individuals formed during vegetative reproduction. Journal of Experimental Zoology. 57, 57-108. Speakman JR, Krol E (2010). The heat dissipation limit theory and evolution of life histories in endotherms--time to dispose of the disposable soma theory? Integrative and comparative biology. 50, 793-807. Stadtman ER (1993). Oxidation of free amino acids and amino acid residues in proteins by radiolysis and by metal-catalyzed reactions. Annual review of biochemistry. 62, 797-821. Stroustrup N, Ulmschneider BE, Nash ZM, Lopez-Moyado IF, Apfeld J, Fontana W (2013). The Caenorhabditis elegans Lifespan Machine. Nature methods. 10, 665-670. Tan TC, Rahman R, Jaber-Hijazi F, Felix DA, Chen C, Louis EJ, Aboobaker A (2012). Telomere maintenance and telomerase activity are differentially regulated in asexual and sexual worms. Proceedings of the National Academy of Sciences of the United States of America. 109, 4209-4214. Technau U, Rudd S, Maxwell P, Gordon PM, Saina M, Grasso LC, Hayward DC, Sensen CW, Saint R, Holstein TW, Ball EE, Miller DJ (2005). Maintenance of ancestral complexity and non- metazoan genes in two basal cnidarians. Trends in genetics : TIG. 21, 633-639. Therneau T (2013). A package for survival analysis in S. R package version 2.37-4. URL http://CRAN. R-project. org/package= survival. Box. 980032, 23298-20032. Timmons L, Court DL, Fire A (2001). Ingestion of bacterially expressed dsRNAs can produce specific and potent genetic interference in Caenorhabditis elegans. Gene. 263, 103-112. Timmons L, Fire A (1998). Specific interference by ingested dsRNA. Nature. 395, 854. Torre C, Ghigo E (2015). [Planaria: an immortal worm to clarify human immune response]. Medecine sciences : M/S. 31, 20-22. van Voorhies WA, Khazaeli AA, Curtsinger JW (2003). Selected contribution: long-lived Drosophila melanogaster lines exhibit normal metabolic rates. Journal of applied physiology. 95, 2605-2613; discussion 2604. Warnes GR (2012). gmodels: Various R programming tools for model fitting. R package version 2.15.4.1. Wei M, Fabrizio P, Hu J, Ge H, Cheng C, Li L, Longo VD (2008). Life span extension by calorie restriction depends on Rim15 and transcription factors downstream of Ras/PKA, Tor, and Sch9. PLoS genetics. 4, e13. Weinkove D, Halstead JR, Gems D, Divecha N (2006). Long-term starvation and ageing induce AGE-1/PI 3-kinase-dependent translocation of DAF-16/FOXO to the cytoplasm. BMC Biol. 4, 1-1.

Wickham H, Chang W (2013). An implementation of the Grammar of Graphics. R package version 0.9. 3. Williams GC (1957). Pleiotropy, Natural Selection, and the Evolution of Senescence. Evolution; international journal of organic evolution. 11, 398-411. Wilson DL (1994). The analysis of survival (mortality) data: fitting Gompertz, Weibull, and logistic functions. Mechanisms of ageing and development. 74, 15-33. Wood JG, Rogina B, Lavu S, Howitz K, Helfand SL, Tatar M, Sinclair D (2004). Sirtuin activators mimic caloric restriction and delay ageing in metazoans. Nature. 430, 686-689. Zarse K, Schmeisser S, Birringer M, Falk E, Schmoll D, Ristow M (2010). Differential effects of resveratrol and SRT1720 on lifespan of adult Caenorhabditis elegans. Horm Metab Res. 42, 837-839.

APPENDIX Table 3: Oligonucleotide primers used for PCR amplification and cloning of age-1. Bolded text represents the addition of a restriction site.