Coffeehouse notes on: The Multistress Resistance Theory of Aging

It took me forever to publish another blog post, even though I wrote this one over just a few days. The problem is, every time I have a new idea for a post I start writing and end up having 50 drafts of unfinished posts, so from now on I'll try to focus. This post is a primer on the wide topic of ROS, other stressors and their impact on aging, with a focus on mitochondria.

Thanks to comparative and interventional biologists the moribund oxidative stress theory has been revitalized and transformed into what we can call the multi-stress resistance theory. Simply put, stress resistance is one of several contributors to long lifespans. Oxidative stress is just one of the many noxious agents and antioxidants are just one way to counteract them, and perhaps the least efficient one! Taken together with Täuber’s paradox it seems trivial to explain why almost all antioxidant studies failed to extend mouse lifespan (Täuber’s in a nutshell: decades ago epidemiologists were able to show empirically that one would have to modulate all causes of aging to produce large gains in life expectancy).

Stress resistance is now considered a composite of: low generation of reactive species, interception by antioxidants, prevention of damage amplification and protection of vulnerable sites, and of course robust sensing of damage/ROS is necessary to increase resistance on-demand. Importantly, all these mechanisms are just the tip of the iceberg. Damage is inevitable as ROS come in contact with biomolecules, but these may be resistant to damage or functional impairment (i.e. cope well with damage). Immediate repair is also common, especially of pre-mutagenic DNA lesions. There can be other mechanisms of “repair” or “coping” like the killing of dysfunctional cells, degradation of whole mitochondrial genomes or the autophagic consumption of damaged biomolecules and so much more.

Overview of multistress resistance (could fit on a napkin)

ROS production
Hydrogen peroxide production at mitochondrial complex I is thought to be lower in long-lived species, but the number of species that was used to show this was rather small (n~13; Lambert 2007) and this limits the statistical power after the recommended statistical adjustment for bodyweight and phylogeny was performed. Another study using MitoSOX Red as a readout for mitochondrial ROS had a similar problem with sample size (Csiszar 2012). Hopefully, new methods - like the mitochondria-targeted mass spectrometry probe MitoB - will enable larger and more thorough as well as artifact-free studies. For instance, the study design by Lambert is very laborious and it is unclear if the condition of reverse electron flow occurrs in vivo.

Without minimizing the controversy over stringent adjustment of small comparative datasets, we can say that ROS production appears to be lower in long-lived species. In the future, it would be insightful to look at other sources of radical oxygen or nitrogen species in the cell e.g. dopamine metabolism, xanthine oxidase, phase I metabolism, etc.

Experimental support: technically difficult hence limited. CR might change mtROS production (controversial; see Walsh 2014).

Antioxidants, stress sensing and damage amplification - the antioxidant controversy continues
Barja writes about the studies of Pamplona et al.: “Among a total of 78 correlations between endogenous tissue antioxidants and longevity, 72 were negative, 6 did not show significant differences, and only a single one was positive.” On the other hand, Nrf2, one master regulator of the stress response, is consistently more active in long-lived species (Lewis 2013). Nrf2 has emerged as one of the most important leads because of good experimental support. In mice Nrf2 KO attenuates the cancer preventative effect of CR against induced and endogenous cancers, though, curiously not the lifespan benefit of CR.

It is not immediately clear why it should be so. As we will see later, almost all mechanisms of stress protection are superior in long-lived species (low damage generation, resistant biomolecules, damage repair, etc.) and obviously the same could have been true for the antioxidant system. Barja argues that all the other mechanisms are more “efficient”. Due to site-specific ROS production and three-dimensional diffusion it would be exceedingly hard to intercept these species, but intuitively this answer doesn’t satisfy me. In this context it is even more paradoxical that Nrf2 is an important regulator of GSH synthesis, but GSH has a negative relationship with lifespan!

Either way, not all protective systems (“antioxidants”) have been studied exhaustively. If we now use the exclusionary principle and combine the list of known Nrf2 targets with the data of Pamplona et al. 2011 we can find a set of candidate genes (and antioxidants) that could correlate with lifespan, among those ferritin, for instance. Ferritin seems promising because it fits the bill of damage resistance. Ferritin does not intercede damage, instead modulating generation of ROS by Fenton-type reactions.

Resistant biomolecules
Unsaturated fatty acids are considerably more reactive than mono- and saturated fatty acids. This is one well studied example of damage resistance and long-lived species have indeed fewer highly unsaturated fatty acids in their membranes compared to their less fortunate brothers and sisters. Recent studies indicate the same is true for mitochondrial lipids (e.g. Cortie 2015). While the link with membrane saturation has been known for a while, and it is again true that the connection is attenuated after adjustment (Valencak et al. 2007), we now have some elegant evidence in favour of causality, because unsaturated dietary fatty acids can negate the benefits of CR (López-Domínguez 2015).

What about our beloved, precious mitochondrial DNA?

Pamplona 2011 on mtDNA resistance. Not conclusive but certainly suggestive:
...the free energy (a physical property of the double-stranded DNA molecule related to the binding energy between the two DNA strands; the more negative the free energy is, the less likely is the spontaneous separation of the two strands through thermal fluctuations, conferring to the mtDNA a greater structural stability and lesser susceptibility to damage)....analyzing the lineage-specific mitochondrial mutation rate across 1,696 mammalian species and comparing it with the nuclear rate, is reported a selected decrease of substitution rate in long-lived species, in agreement with the evidence for a causal role of mtDNA mutations in aging...

All in all, this hypothesis is doing very well. As predicted by Barja and Pamplona, the concept seems to apply not only to lipids and possibly nucleic acids, but also protein resistance to unfolding. Here, GAPDH resistance to urea has emerged as one correlate for proteome stability (Treaster 2015).

Mitochondrial damage - not clear at all
Given these findings we would expect less damage to mtDNA and indeed the pre-mutagenic 8oxodG lesion correlates with species lifespan. What we do have to keep in mind is that not all damage leads to mutations and other pathologic changes. Different types of oxidative lesions are clearly correlated with LS (Bokov 2004). However, the safety of our precious mtDNA is a little more complicated than that.

Consistent with our simplistic damage theory (Kauppila 2016): “The amount of mtDNA mutations is known to increase with age in multiple human tissues such as brain, heart, colon, and skeletal muscle (Bua et al., 2006; Cortopassi and Arnheim, 1990; Greaves et al., 2014; Kennedy et al., 2013). Both mtDNA point mutations and deletions have been shown to accumulate with age (Bua et al., 2006; Cortopassi and Arnheim, 1990; Greaves et al., 2014; Kennedy et al., 2013).”
However, while mutations accumulate, the source is not necessarily ROS. Kennedy et al. (2013) have made a strong case against GT transversions arising form 8oxodG in the prefrontal cortex. This brings us back to the introductory paragraph: ROS are but one player. As far as I understand, only a select few tissues might be susceptible to mitochondrial ROS (dopamine metabolism goes where? Exactly: basal ganglia. Perhaps we need to look there?)

DNA damage repair - mitochondria remain understudied
The evidence that some - and curiously only some - types of DNA damage repair correlate with lifespan is now quite strong. It is too early to speak of much experimental support, but see Liao & Kennedy 2014 for interesting mouse KO data, for instance, hMTH1-Tg mice are long-lived (Luca 2013).
Mitochondria, on the other hand, are not well studied when it comes to cross-species comparisons DNA damage repair. What we know paints a picture that is confusing at first: “Surprisingly, Ogg1 and Mutyh double-knockout mice do not accumulate mutations of mtDNA, although they display a cancer-prone phenotype caused by deficient BER in the nucleus (Halsne et al., 2012)”

This is not really unexpected, though, if nuclear BER doesn’t correlate with lifespan to begin with, why should mitochondrial BER? So the mitochondrial hypothesis is not quite dead: we know that ROS can induce double strand breaks (DSB), some tissues produce orders of magnitude more ROS and that repair of ROS/DSB leads to a specific mutational signature (Lakshmipathy and Campbell 1999; Moraes data) that is very similar to age-related large, direct-repeat flanked, deletions.

Coping, cell death, (delayed) repair and so on
HSP proteins are higher in long-lived species (Stuart 2013), but proteasome activity is no different in the liver of long-lived species (Salway 2015). On the other hand, HSPs, autophagy and proteasome activity are activated in cells isolated from long-lived species (Pride 2015).

The most important model system for “coping” is certainly the naked mole rat (NMR), this long-lived rodent has high levels of oxidative damage, like protein carbonyls or urinary isoprostanes, and their antioxidant system is not more potent compared to the house mouse (Lewis et al., 2013). Other than that, the NMR combines most of the tools we already discussed: high autophagy, proteasome activity, DNA repair, and quite interestingly high ribosomal fidelity (=damage generation) short telomeres and low telomerase activity. Their proteasome is very efficient and resists inactivation (Rodriguez 2014):

Clearly, the preternaturally long-lived naked mole-rats have evolved certain molecular mechanisms that contribute to their ability to prolong good health and attenuate the aging process [1]. The high proteasome content coupled with its distinctive composition in naked mole-rats, that we previously described, may play an integral role in this regard [18]. However, we describe here another important, complementary mechanism. We report here for the first time that naked mole-rats express high levels of key chaperones, HSP72, HSP40, and HSP25, even in untreated tissues when compared to those of the mouse. Further we present evidence suggesting the presence of a novel cytosolic factor that contains two of these chaperones, and that not only protects proteasome function against cell stressors but also enhances proteasome performance.

While reduced telomerase activity can help to cope with DNA damage that has initiated cancer by limiting cancer progression. In theory, cancer cells might die from multiple rounds of replication and telomeric attrition before they are able to re-activate telomerase.


Kauppila, T. E., Kauppila, J. H., & Larsson, N. G. (2016). Mammalian Mitochondria and Aging: An Update. Cell Metabolism.

Lambert A. Boysen H. Buckingham JA. Yang T. Podlutsky A. Austad SN. Kunz TH. Buffenstein R. Brand D. Low rates of hydrogen peroxide production by isolated heart mitochondria associate with long maximum lifespan in vertebrate homeotherms. Aging Cell. 2007;6:607–618.

Csiszar, A., Podlutsky, A., Podlutskaya, N., Sonntag, W. E., Merlin, S. Z., Philipp, E. E., ... & Pinto, J. T. (2012). Testing the oxidative stress hypothesis of aging in primate fibroblasts: is there a correlation between species longevity and cellular ROS production?. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 67(8), 841-852.

The rate of free radical production as a determinant of the rate of aging: evidence from the comparative approach. Perez-Campo R, López-Torres M, Cadenas S, Rojas C, Barja G. J Comp Physiol B. 1998 Apr; 168(3):149-58.

Molecular and structural antioxidant defenses against oxidative stress in animals. Pamplona R, Costantini D Am J Physiol Regul Integr Comp Physiol. 2011 Oct; 301(4):R843-63.

Cortie, C. H., Hulbert, A. J., Hancock, S. E., Mitchell, T. W., McAndrew, D., & Else, P. L. (2015). Of mice, pigs and humans: An analysis of mitochondrial phospholipids from mammals with very different maximal lifespans. Experimental gerontology, 70, 135-143.

Valencak, T. G., & Ruf, T. (2007). N− 3 polyunsaturated fatty acids impair lifespan but have no role for metabolism. Aging cell, 6(1), 15-25.

Treaster, S. B., Chaudhuri, A. R., & Austad, S. N. (2015). Longevity and GAPDH Stability in Bivalves and Mammals: A Convenient Marker for Comparative Gerontology and Proteostasis. PloS one, 10(11), e0143680.

Aging Cell. 2013 Aug;12(4):695-705. doi: 10.1111/acel.12094. Epub 2013 May 30. Prolonged lifespan with enhanced exploratory behavior in mice overexpressing the oxidized nucleoside triphosphatase hMTH1. De Luca et al.

Age (Dordr). 2011 Mar;33(1):33-47. doi: 10.1007/s11357-010-9157-5. Enhanced protein repair and recycling are not correlated with longevity in 15 vertebrate endotherm species. Salway KD1, Page MM, Faure PA, Burness G, Stuart JA.

Biochem Biophys Res Commun. 2015 Feb 20;457(4):669-75. doi: 10.1016/j.bbrc.2015.01.046. Epub 2015 Jan 21. Long-lived species have improved proteostasis compared to phylogenetically-related shorter-lived species. Pride H1, Yu Z1, Sunchu B1, Mochnick J1, Coles A2, Zhang Y3, Buffenstein R3, Hornsby PJ4, Austad SN5, Pérez VI6.

Antioxid Redox Signal. 2013 Oct 20;19(12):1388-99. doi: 10.1089/ars.2012.4911. Epub 2012 Dec 7. The naked mole-rat response to oxidative stress: just deal with it. Lewis et al.


Walsh 2014: