S.N. Austad / Experimental Gerontology 36 (2001) 599±605

Experimental Gerontology 36 (2001) 599±605

599

www.elsevier.nl/locate/expgero

An experimental paradigm for the study of slowly
aging organisms
S.N. Austad*
Department of Biological Sciences, University of Idaho, P.O. Box 44-3051, Moscow, ID 83844-3051, USA

Abstract
An experimental paradigm for the study of mechanisms of resistance to aging in long-lived
organisms has been developed. The paradigm assumes, in concert with accumulating empirical
data, that resistance to the aging processes at the organismal level will be re¯ected in resistance
to various stressors at the cellular level. The advantage of this paradigm is that it requires neither the
long-term monitoring of individuals nor the use of exceptionally old individuals. The research
approach consists of: (1) verifying that primary cell cultures from the long-lived organism exhibit
better resistance to key stressors than cells from related, short-lived organisms; (2) assessing differences in gene-expression before and after stress exposure in cultured cells from the long- and shortlived species in order to identify key genes involved in the stress-resistance response; (3) transfecting
putative key genes from long-lived species into cells or cell lines of de®ned stress-resistance and
hope to observe that the stress-resistance phenotype has thereby been transferred with the gene(s);
(4) generating transgenic model animals containing the gene(s) of interest and look for extended life/
health span. q 2001 Elsevier Science Inc. All rights reserved.
Keywords: Cell stress; Long-lived organisms; Oxidative damage resistance; Bird longevity; Comparative geneexpression

1. Introduction
Traditional animal models used in aging research are chosen partially because of their
demonstrable failure to combat aging processes with much success. That is, they die
quickly over several weeks (C. elegans) or several months (Drosphila) or at the most
several years (mice and rats), even though living in a protected environment optimized for
their health. Their short life span is, in fact, a major contributor to their experimental utility
(Sprott and Austad, 1996). Short life allows animals to be followed for their entire lives
and mortality patterns to be recorded from whole experimental populations within a
* Tel.: 11-208-885-6598; fax: 11-208-885-7905.
E-mail address: austad@uidaho.edu (S.N. Austad).
0531-5565/01/$ - see front matter q 2001 Elsevier Science Inc. All rights reserved.
PII: S 0531-556 5(00)00229-1

600

S.N. Austad / Experimental Gerontology 36 (2001) 599±605

reasonable amount of time for a reasonable cost and facilitating comparisons among
populations which have been subjected to some environmental or genetic manipulation
that is suspected to in¯uence aging rate. The reasonable rationale behind using such shortlived animals in research, of hoped-for relevance to modulation of the much longer human
life span, is that fundamental aging processes will turn out to be much the same from
species to species. Therefore, learning the mechanisms by which these animals age, should
lead to a basic understanding of the aging processes. Such understanding might eventually
allow us to design interventions, to retard aging in any species.
By design and de®nition then, traditional laboratory animals have poorly developed
defenses against the destructive processes of aging. A tantalizingly attractive alternative
research approach might be to investigate the nature of some of the exceptionally effective
defenses against destructive aging processes that biological evolution has designed. This
strategy is in accord with what Francis Crick has called Orgel's Second Law, namely ªevolution is smarter than you areº (Dennett, 1996). If evolution has produced a model of successful
resistance to the damage of aging, we might learn a lot from investigation of that model.
Senescence, as measured by decline in physical performance or reproductive competence is clearly observable in humans no older than their early 40s, even though the rare
person (1/5,000±1/10,000) lives as long as 100 years in a protected modern environment.
Some animal species clearly age more slowly than this. For example, accumulating
evidence suggests that some species of bivalve mollusks, some ®shes, and possibly
even some mammals preserve physical and reproductive function for more than a century
and ultimately live in excess of 150 years, perhaps even in excess of 200 years in the wild
(Heller, 1990; Cailliet et al., 2001; George et al., 1999). Still other species of potential
gerontological interest include species such as many birds and bats, which, though they do
not live for centuries, survive several decades exhibiting few obvious physiological signs
of senescence, despite having exceptionally high metabolic rates and body temperatures,
factors which are commonly assumed to lead to accelerated aging. Two alluring questions
raised by these species are: (1) have they really managed to evolve exceptional cellular
defenses against the aging processes Ð defenses probably superior to those that humans
possess Ð or have they managed long life merely by slowing the rate of vital processes
and, (2) if they do exhibit specially effective adaptations against aging, what is (are) the
mechanistic nature of these defenses?
It is important to note that exceptional longevity, by itself, does not guarantee that
animals will possess exceptional anti-aging defenses. Some species may live long because
they do not suffer the degree of biological challenge faced by other species. For instance,
they may exhibit exceptionally chronic low metabolic rates or spend a substantial fraction
of their lives in a hypometabolic resting state like diapausing brine shrimp or tardigrades
or hibernating mammals. Indeed, one of the hypotheses for the longer lives of deepdwelling ®shes is that they exhibit reduced metabolism (Cailliet et al., 2001). It is well
known that metabolic rate reduction is suf®cient to achieve increased longevity (Lyman et
al., 1981), although it is also clear that it is not necessary (McCarter and Palmer, 1992).
The species of most interest to gerontologists interested in discovering how human aging
might be retarded, will be those which do demonstrate enhanced resistance to fundamental
aging processes, such as oxidative challenge and increasing glycational malfunction (see
Holmes et al., 2001).

S.N. Austad / Experimental Gerontology 36 (2001) 599±605

601

The chief reason that exceptionally long-lived species have not been a focus of more
intense study is that the standard research paradigm of aging studies involves longitudinal
observations of animals across the life span. This is clearly impossible with long-lived
animals. However, in working with long-lived birds over the past few years, my collaborators and I have attempted to circumvent this problem by developing an experimental
paradigm which does not necessitate long-term monitoring of individuals or require the
use of those rare individuals that have already reached an exceptionally old age. Our
research paradigm may be easily applied to virtually any species, including many of the
long-lived species in this special issue. Furthermore, it offers a straightforward approach
towards the identi®cation of genes and gene products associated with exceptional
resistance to aging.
2. Experimental paradigm
The fundamental idea behind this research paradigm is that differences between species
in organismic aging rate will often translate into differences observable on the cellular
level. That is, primary cell cultures from long-lived species will have de®nable cellular
phenotypes, particularly resistant to oxidative and possibly nonoxidative stresses, that
differ from short-lived species. This makes intuitive sense, in that if tissues and organs
are resistant to aging, presumably the cells of which they are composed also will be. A
similar assumption has underlain a great deal of research on cellular replicative
senescence, in which species clearly differ in cellular replicative potential in parallel
È
with differences in longevity (Rohme, 1981). Similarly, cellular stress-resistance (as
measured by cell survival) of primary dermal ®broblasts and lymphocytes to both oxidative and nonoxidative stressors has been shown to correlate with maximum longevity
among eight species of mammals (Kapahi et al., 1999). Additionally, work by my collaborators and I on birds has also shown that primary cultures of either kidney epithelial cells
or embryonic ®broblasts from long-lived mammals (Ogburn et al., 1998) or short-lived
birds (Ogburn et al., unpublished data). For instance, 12 h exposure to 20 mM hydrogen
peroxide has virtually no effect on embryonic ®broblasts from long-lived budgerigars
(20 1 years maximum longevity), whereas 60% of similar cells from short-lived Japanese
quail (5 1 years maximum longevity) are killed by this exposure. Finally, Campisi (2001)
notes that mouse cells are much more sensitive to oxidative insult than are human cells.
Therefore, the ®rst step of this paradigm is to determine whether cells grown in primary
culture from the long-lived species are more resistant to a speci®c stressor than from a
related short-lived species. For instance, an intriguing comparison might be among rock®shes of the genus Sebastes, in which the maximum longevity apparently ranges between
12 and 205 years for different species (Cailliet et al., 2001). Ideally, one might assess also,
how the cells of interest survive, compared with analogous human cells, assuming that
culture conditions can be made suf®ciently similar so that human cells could be directly
compared with the long-lived species of interest.
Assuming that the cells from the long-lived species are indeed more stress-resistant than
from the short-lived species, the next procedural step is to use one or more of the available
varieties of comparative gene-expression analysis to relate gene-expression patterns to the

602

S.N. Austad / Experimental Gerontology 36 (2001) 599±605

Table 1
Selected methods of comparative gene-expression analysis
Method

Comments

Differential display

Identi®es differentially expressed
genes without requiring a priori
DNA sequencing. Requires only
a small amount of total mRNA.
Can analyze several hundred
genes in one experiment.
Requires RNA of extreme purity

Representational
difference analysis

Identi®es differentially expressed
genes without requiring a priori
DNA sequencing. Speci®c
ampli®cation of cDNA
fragments of potential interest

Serial analysis of
gene-expression
(SAGE)

Identi®es expressed mRNAs in
approximate relation to their
abundance. Can analyze
expression of many genes
simultaneously. Computationally
challenging. Identi®cation of
genes requires presence of
suf®ciently similar genes in
existing gene banks

DNA microarrays

Requires more a priori DNA
sequencing. Can analyze
expression patterns of thousands
of genes simultaneously

stress-resistance pattern (Table 1). It is beyond the scope of this paper to detail each of the
possible techniques, although they have been reviewed elsewhere (Kozian and
Kirschbaum, 1999). Much of the discussion that follows is drawn from Kozian and
Kirschbaum's review. Because most of the long-lived species of interest will have had
little, if any, previous genetic characterization, methods in which candidate genes of
particular interest could be identi®ed prior to expensive and time consuming semirandom
DNA sequencing, would be particularly useful.
One such method is differential display of expressed mRNAs (Liang and Pardee, 1998;
Ito and Sakaki, 1997), in which total mRNA from two samples is reverse transcribed and
the resultant cDNAs ampli®ed with random oligonucleotide primers, then analyzed on a
sequencing gel. Differentially expressed genes are detectable, as bands present in only one
lane of the gel and these products can subsequently be cut out of the gel and sequenced.
Redundancy in the ampli®ed cDNAs can be minimized by selective ampli®cation of 3 0 end restriction fragments of the differentially expressed cDNAs (Prashar and Weissman,
1996). Besides avoiding a lot of a priori DNA sequencing, the additional advantage of this
method is that expression pattern of several hundred genes can be assessed in a single
experiment and cDNA probes become immediately available.

S.N. Austad / Experimental Gerontology 36 (2001) 599±605

603

A second method, which identi®es genes of potential interest prior to sequencing is
representational difference analysis (Lisitsyn et al., 1993), which utilizes subtractive
hybridization and selective ampli®cation of cDNA restriction fragments that are only
present in one of the two samples. A major advantage of this approach is the speci®c
ampli®cation of differentially expressed mRNAs. Improved ef®ciency of enrichment in
DNAs of interest can be achieved by multiple rounds of subtraction.
The advantage of the two previous methods is that they do not require an enormous DNA
sequencing effort and do not rely on gene identi®cation from existing databases. For some
long-lived species of interest, there may already exist substantial enough genetic characterization of related species of suf®cient gene sequence similarity that existing databases will
prove helpful or that hybridization between homologous cDNAs is feasible. If so, this opens
the possibility of implementing other techniques such as serial analysis of gene-expression
(SAGE) or DNA microarray technology to hunt for genes which are differentially
expressed. For instance, in our work using long-lived budgerigars (Melopsittacus undulatus) and short-lived Japanese quail (Coturnix japonicus), we found that more than 90%
of cDNAs from both species, hybridized well enough with chicken cDNAs to give meaningful results using a chicken DNA microarray developed by Paul Niemann (Carlberg et
al., unpublished data). This was particularly surprising in the budgerigar, which unlike the
Japanese quail, comes from a different avian order to that of the chicken.
The experimental power of utilizing cells from both long-lived, stress-resistant and
short-lived, stress-prone species in this paradigm is that expression patterns before and
after exposure to the stressor may be compared within and between species. The withinspecies comparison reveals how the expression of speci®c genes is induced or suppressed
by the stressor used. In the stress-resistant species, this change in expression pattern will
likely include genes, critical to the stress-resistant phenotype as well as genes that are not
critical. The between-species comparison, helps categorize expression changes. For
instance, if the expression level of gene x doubles after exposure to the stressor in both
species, it is not likely to be a gene critical to the stress-resistant phenotype unless the
protein product is differentially active Ð a possibility that should not be dismissed. On the
other hand if the expression level of gene y does not increase, or increases marginally, in
the short-lived, stress-prone species, but increases tenfold in the long-lived, stress-resistant
species, it suggests that this is a gene of exceptional interest, which may be critical to the
stress-resistance response. Indeed, our preliminary research utilizing subtractive hybridization suggested that such expression differences could be found (Oshima et al.,
unpublished data).
The next step to con®rm that genes identi®ed by the procedures above are indeed critical
to the stress-resistance response, will be to transfect the candidate genes into cell lines of
quantitatively characterized resistance to the stressor invoked in the original experiments.
The hoped-for result, of course, is that some or all of the stress-resistant phenotypes will be
observable in the transfected cells. Even after con®rming this, however, it will still be
necessary to link the cell resistance phenotype to longer life and health. This can be done
by transfecting the genes of interest into traditional animal models of aging, to observe
whether their life- and/or health-span have been extended. Excellent methods of controlling for potentially confounding positional effects are now available (Sun and Tower,
1999).

604

S.N. Austad / Experimental Gerontology 36 (2001) 599±605

The experimental paradigm speci®ed above assumes that reasonably closely related
species differences in cellular stress-resistance are largely due to differences in geneexpression level as opposed to differences in the speed of the expression response or
differential ef®ciency of stress-resistance by speci®c proteins in different species. The
®rst complication (differential speed of response) can be addressed by a similar paradigm,
only now, the timing of expression pattern differences become a focus of the research. The
second problem (differential stress-resistance due to more effective protein design) will
require a somewhat different approach. However, there does seem to be reason to expect
that the phenotype of long-life and resistance to fundamental aging processes will be
recapitulated on the cellular level. It is at a minimum, worth investigating more fully,
as it would open up a world of new approaches to understanding why some animal species
can live so long without appreciably aging.
Acknowledgements
The research from which these ideas developed was supported by the National Institute
on Aging (Grant No. AG-01751) and the Ellison Medical Foundation. I would like to
particularly thank my collaborators in the avian oxidative damage resistance project,
George M. Martin, Junko Oshima, Charles Ogburn, Donna Holmes, and Kristen Carlberg
for helping to generate these ideas.
References
Cailliet, G.M., Andrews, A.H., Burton, E.J., Watters, D.L., Kline, D.E., Ferry-Graham, L.A., 2001. Age determination and validation studies of marine ®shes: do deep-dwellers live longer?. Exp. Gerontol. 36, 739±764.
Campisi, J., 2001. From cells to organisms: can we learn about aging from cells in culture?. Exp. Gerontol. 36,
607±618.
Dennett, D.C., 1996. Darwin's Dangerous Idea: Evolution and Meanings of Life. Touchstone, New York.
George, J.C., Bada, J., Zeh, J., Scott, L., Brown, S.E., O'Hara, T., Suydam, R., 1999. Age and growth estimates of
bowhead whales (Balaena mysticetus) via aspartic acid racemization. Can. J. Zool. 77, 571±580.
Heller, J., 1990. Longevity in mollusks. Malacologia 31, 259±295.
È
Holmes, D.J., Fluckiger, R., Austad, S.N., 2001. Comparative biology of birds: an update. Exp. Gerontol. 36,
869±883.
Ito, T., Sakaki, Y., 1997. Fluorescent differential display. Methods Mol. Biol. 85, 37±44.
Kapahi, P., Boulton, M.E., Kirkwood, T.B.L., 1999. Positive correlation between mammalian life span and
cellular resistance to stress. Free Radical Biol. Med. 26, 495±500.
Kozian, D.H., Kirschbaum, B.J., 1999. Comparative gene-expression analysis. Trends Biotechnol. 17, 73±78.
Liang, P., Pardee, A.B., 1998. Differential display. A general protocol. Mol. Biotechnol. 10, 261±267.
Lisitsyn, N., Lisitsyn, N., Wigler, M., 1993. Cloning the differences between two complex genomes. Science 259,
946±951.
Lyman, C.P., O'Brien, R.C., Greene, G.C., Papagrangos, E.D., 1981. Hibernational longevity in the Turkish
hamster Mesocricetus brandti. Science 212, 668±670.
McCarter, R.J., Palmer, J., 1992. Energy metabolism and aging: a lifelong study in Fischer 344 rats. Am. J.
Physiol. 263, E448±E452.
Ogburn, C.E., Austad, S.N., Holmes, D.J., Kiklevich, J.V., Gollahon, K., Rabinovitch, P.S., Martin, G.M., 1998.
Cultured renal epithelial cells from birds and mice: enhanced resistance of avian cells to oxidative stress and
DNA damage. J. Gerontol. Biol. Sci. 53A, B287±B292.

S.N. Austad / Experimental Gerontology 36 (2001) 599±605

605

Prashar, Y., Weissman, S.M., 1996. Analysis of differential gene-expression by display of 3 0 end restriction
fragments of cDNAs. Proc. Natl. Acad. Sci. USA 93, 659±663.
È
Rohme, D., 1981. Evidence for a relationship between longevity and mammalian species and lifespans of normal
®broblasts in vitro and erythrocytes in vivo. Proc. Natl. Acad. Sci. USA 78, 5009±5013.
Sprott, R.L., Austad, S.N., 1996. Animal models for aging research. In: Schneider, E., Rowe, J.W. (Eds.).
Handbook of the Biology of Aging. 4th ed.. Academic Press, San Diego, CA, pp. 3±23.
Sun, J., Tower, J., 1999. FLP recombinase-mediated induction of Cu/Zn-superoxide dismutase transgene-expression can extend the life span of adult Drosophila melanogaster ¯ies. Mol. Cell. Biol. 19, 216±228.