u·m·i...estimation in manyfish species. theseconventional techniques areoften inadequate...
TRANSCRIPT
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may
be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of thecopy submitted. Broken or indistinct print, colored or poor qualityillustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adverselyaffect reproduction.
In the unlikely event that the author did not send UMI a complete
manuscript and there are missing pages, these will be noted. Also, ifunauthorized copyright material had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand corner and
continuing from left to right in equal sectionswith small overlaps. Each
original is also photographed in one exposure and is included in
reduced form at the back of the book.
Photographs included in the original manuscript have been reproducedxerographically in this copy. Higher quality 6" x 9" black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directly
to order.
U·M·IUniversity Microfilms International
A Bell & Howell Information Company300 North Zeeb Road. Ann Arbor. M148106-1346 USA
313/761-4700 800/521-0600
Order Number 9312191
Fluorescent age pigment accumulation as a determinant ofchronological age in aquatic organisms
Hill, Kevin Thomas, Ph.D.
University of Hawaii, 1992
V·M·I300N. ZeebRd.AnnArbor,MI48106
_.,- _.._--_._-_.....----_ ..__ ._ ..
FLUORESCENT AGE PIGMENT ACCUMULATION AS A DETERMINANT
OF CHRONOLOGICAL AGE IN AQUATIC ORGANISMS
A DISSERTATION SUBMITIED TO THE GRADUATE DIVISION OF THEUNIVERSITY OF HAWArI IN PARTIAL FULFILU"ffiNT OF THE
REQUIREMENTS FOR THE DEGREE OF
DOcrOR OF PmLOSOPHY
IN
ZOOLOGY
DECEMBER 1992
BY
Kevin T. Hill
Dissertation Committee:
Christopher Z. Womersley, ChairpersonGregory Ahearn
Fred I. KamemotoJames D. Parrish
Christopher L. BrownJack R. Davidson
ACKNOWLEDGMENTS
I am grateful to my mentor, Dr. Christopher Womersley, for inviting me into his
laboratory, showing interest in this project, and for providing enlightenment, insight, and
support through the duration of this dissertation project. I would also like to thank my
other committee members-- Drs. Gregory Ahearn, Christopher Brown, Jack Davidson,
Fred Kamemoto, and James Parrish. Their ideas, perspectives, and critical reviews
contributed greatly to the quality of this project.
I thank those individuals who aided in the culture and maintenance of fish used in
this study: Paul Shiota and Torn Kazama of the National Marine Fisheries Service,
Honolulu Laboratory, maintained the fish used in Chapter 3 of this study; Miles
Yamamoto andAdrian Franke helped maintainthe experimentalaquarium system in
Edmondson Hall during my absence; and Alane Gresham rescued hundreds of rosy barb
fry during an island-wide power outage.
Special thanks go out to past and present members of Chris Womersley's research
group for enriching my daily life in the laboratory, in particular, Lynne Riga who shared
her friendship and philosophies and Alane Gresham for sharing insights on the chemistry
of aging.
Marty Martinelliand Dr. Andrew Taylor provided much needed statisticaladvice
and computer assistance with SAS. I am also indebted to the support staff of the
Department of Zoology--Lori Yamamura, Sally Oshiro, and Audrey Shintani-- for their
patience and expertise in dealing with matterspertaining to red tape. illustrative skill of
Sue Monden was much appreciated for figures in Chapter 2.
The research was made possible through generous support provided by: the
University of Hawaii Sea Grant College Program; the Research Corporation of the
University of Hawaii; the Universityof Hawaii Center on Aging; Sigma Xi - The
Scientific Research Society; the ARCS Foundation of Hawaii; the University of Hawaii
iii
Office of Research Administration; and the Department of Zoology, University of
Hawaii. I am very grateful for all financial contributions that enabled me to complete this
research.
All of this work was inspired through the tolerance and affection of my wife,
Astrid Ronke, who instilled in me the persistence needed to complete this work. I also
thank my parents, Randall and Geraldine Hill, for their support and patience through my
years in graduate school.
iv
ABSTRACT
Age pigments have been intensivelystudied by gerontologists hoping to define the
biochemical processes involvedin aging. More recently,age pigment accumulation has
been studied by field biologistshoping to estimatechronologicalage of animals based on
quantitiesof pigments measured in postmitotic tissues. The present research was
conducted to evaluate methods used for extractionand measurementof fluorescent age
pigments (FAP), describe patterns of FAP accumulation in two teleost fish speciesand
develop multivariateage predictionmodelsbased on FAP content and other somatic
variables, and determine the effects of some environmentaland physiologicalvariableson
FAP variability. Evaluation of procedures for handling specimensand extracts revealed
increases in FAP-like fluorophores in vitro in Oreochrornis mossambicus brain, heart,
and muscle tissues and theirextracts with increasedstoragetemperature (-20°C and
above) and time, particularlyin the chloroform/methonal solvent system. Ultrasonication
of tissue homogenatesgreatlyenhanced this effect and generated other fluorophores in
solution. Fluorescence assay temperaturealso affectedexpression ofFAP. Some or all
of these effects of handlingproceduresmay have produced variabilityof results reported
from previous FAP studies.
NonpolarFAP extractedfrom o. mossambicus brain accumulated in an
increasingly rapid manner with age. Nonpolarand polar FAP from Puntius conchonius
brain leveled off with increasingage. FAP in £. conchonius heart increased linearly with
age. Multiple regressionanalysesdemonstratedthe age-predictive value of FAP data for
both species under controlledrearing conditions. FAP content and otolith dimension data
consistentlyprovided slightlystrongerpredictive models than somaticdata alone.
Temperatureand body weight affectedFAP content in £. conchonius brain and heart
tissues. Brain FAP was inversely related to body weight, especially at the lower
temperatures. Heart FAP was inverselyrelated to body weight at 19°C, but positively
v
--_._.._-- .-
related at higher temperatures. Increasing temperature decreased brain FAP, but
increased heart FAP. Different levels of saturation of cellular lipid constituents at
different temperatures may affect the potential for FAP formation via lipid peroxidation.
FAP differences detected between non-sibling fish indicated genetic variation in the ability
to control FAP genesis.
VI
TABLE OF CONTENTS
ACKNOWI...EDGMENTS , , , .. iii
ABSTRACT v
LIST OF TABLES ix
LIST OF FIGURES x
CHAPTER 1. INTRODUCTIONOverview 1
Cellular genesis ofFAP 2Quantification ofFAP 4Age pigments in aquatic animals , " 5
Research problem 8Study objectives 10Literature cited 11
CHAPTER 2. CRITICAL ASPECfS OF FLUORESCENT AGE PIG?vIENT?vIETHODOLOGIES: MODIFICATION FOR ACCURATE ANALYSIS AND
AGE ASSESS?vIENTS IN AQUATIC ANIMALSIntroduction " 29Materials and methods '" , . 32
Model animal/tissues 32Tissue storage temperature " . 32Unmodified extraction procedures 33Incubation temperature of sample extracts 33Effects of ultrasonication on fluorescence yield 34Confirmation and removal of interfering compounds 34Assay temperature 34Modified extraction and assay procedures 35
Results 36Tissue storage temperature 36Incubation temperature of sample extracts 37Effects of ultrasonication , , 38Fluorescent contaminants 38Assay temperature 39FAP accumulation with age 39
Discussion 40Literature cited 46
CHAPTER 3. FLUORESCENT AGE PIGMENT ACCUMULATION ANDTHE USE OF MULTIPLE REGRESSION MODELS FOR AGE ESTIMATION
IN TWO FISH SPECIESIntroduction " 70Materials and methods , , 73
Specimen culture and preparation 73FAP extraction and spectrofluorometric assay 74Multiple regression analysis 76
vii
T ABLE OF CONTENTS (continued)
Results 76Somatic and otolith growth 76Organ growth and FAP accumulation with age 77Multiple regression analyses 79
Oreochromis mossambicus 79Puntius conchonius 80
Discussion 82References 88
CHAPTER 4. INTERACTIVE EFFEcrS OF SOME ENVIRONMENTAL ANDPHYSIOLOGICAL VARIABLES ON FLUORESCENT AGE PIGMENT
ACCUMULATION IN BRAIN AND HEART TISSUES OF ANAQUATIC POIKILOTIffiRM
Introduction 112Materials and methods , 115
Model animal 115Aquarium system 115Experimental design 116
Temperature/ration experiment 116Temperature/photoperiod experiment 117
Oxygen consumption , " 117Specimen/tissue preparation 118FAP extraction and assay 118Statistical analysis 120
Results , 120Temperature/ration experiment 120Temperature/photoperiod experiment ;; 123
Discussion 124Literature cited 131
CHAPTER 5. SUMMARY AND CONCLUSIONS 158
viii
~ .. __ .._--_.._--.... ""--
LIST OF TABLES
Table Page
2-1 Previous studies of fluorescent age pigment accumulationin aquatic animals " , ,.. 54
3-1 Regression coefficients and associated statistics from multipleregression models of age for Oreochromis mossambicus 96
3-2 Regression coefficients and associated statistics from multipleregression models of age for Puntius conchonius 98
4-1 Results of ANCOVA model for testing effects of fish body weight,temperature, photoperiod, and clutch on polar FAP frombrain tissues of Puntius conchonius 140
4-2 Results of ANCOVA model for testing effects of fish body weight,temperature, photoperiod, and clutch on polar FAP fromhean tissues of Puntius conchonius 141
ix
Figure
1-1
1-2
1-3
1-4
2-1
2-2
2-3
2-4
2-5
2-6
2-7
3-1
3-2
3-3
LIST OF FIGURES
Page
Modulation of superoxide radicals produced through normalcellular metabolism 22
Free radical-mediated lipid peroxidation and malondialdehydeproduction " 24
Fluorescent crosslinkage reactions of malondialdehyde 26
Summary schematic ofprocesses leading to fluorescentage pigment formation , 28
Effect of storage temperature on fluorescent products extractedfrom skeletal muscle of Oreochromis mossambicus 57
Effects of incubation temperature on fluorescence ofunfractionated chloroform/methanol extracts of brain, heart, andmuscle tissues of Oreochromis mossambicus 59
Effects of incubation temperature on fluorescence of finalchloroform extracts of brain, heart and muscle tissues ofOreochromis mossambicus '" 61
Effects of ultrasonication on fluorescence of homogenates ofbrain, heart, and muscle tissues of Oreochromis mossambicus 63
Fluorescence spectra of fluorescent age pigments and flavincontaminants extracted from brain tissues ofOreochromismossambicus , '" " 65
Effects of assay temperature on expression of fluorescenceof quinine sulfate standard and fluorescent age pigments extractedfrom brain tissues of Oreochromis mossambicus 67
Fluorescent age pigment content and brain weight as functions ofchronological age in Oreochromis mossambicus 69
Body size and otolith weight as functions of chronological agein Oreochromis mossambicus , 101
Body size and otolith weight as functions of chronological agein Puntius conchonius l03
Brain weight and fluorescent age pigment level as functions ofchronological age in Oreochromis mossambicus , 105
x
Figure
3-4
3-5
3-6
4-1
4-2
4-3
4-4
4-5
4-6
4-7
4-8
-_.__..~------ ....._-
LIST OF FIGURES (continued)
Page
Emission spectra of fluorescent age pigments extracted frombrain and heart tissues of Puntius conchonius , 107
Brain weight and fluorescent age pigment level as functions ofchronologicalage in Puntius conchonius 109
Heart weight and fluorescentage pigment level as functions ofchronologicalage in Puntius conchonius.. , 111
Mean fish weights of Puntius conchonius reared in thetemperature/ration and temperature/photoperiod experiments 143
Effects ofdifferent temperatureand ration levels on FAP in braintissues of Puntius conchonius 145
Brain polar FAP as a function of fish weight for Puntiusconchoniusreared in the temperature/ration experiment. 147
Effects of different temperatureand ration levels on FAP in hearttissues of Puntius conchonius 149
Heart FAP as a function of fish weight for Puntiusconchoniusreared in the temperature/ration experiment 151
Weight-specificrate of oxygen consumption as a function of fishweight and temperature in Puntius conchonius 153
Effects of different temperature and photoperiodlevels on FAPin brain tissues of Puntius conchonius 155
Effects of different temperature and photoperiodlevels on FAPin heart tissues of Funtius conchonius , " .157
xi
CHAPTER 1
INTRODUCTION
Overview
Managers of living aquatic resources are becoming increasinglyconcernedabout
the welfare of fishery stocks under continuedexploitation. An understandingof the
sustainable yieldof a stockis essential for draftingand implementing effectivefisheries
management policies. Determination of age andgrowth rates is perhaps the most
fundamental componentof this type of assessment (Ricker 1979). Age data, in
conjunction with length and weight measurements, can give informationon life history
parameters (natality, age at recruitmentand sexualmaturity, longevity, and mortality)
necessary to evaluatethe overall conditionof thestock.
Traditional methodsof age determination in fish have included such indirect
measuresas length-frequency analysisor the quantification of rhythmic events in hard
tissues (BagenalI974). However, results from manylength-frequencystudies have been
inconclusive because: 1) there is difficultyin obtaining adequate, representative numbers
of specimensin a full range of life stages;2) sampling introduces gear bias; 3) modes for
older fish are difficult to distinguish due to asymptotic growth and variability within
cohorts; and 4) cohorts are ill-definedin fish species withprotractedrecruitmentperiods
(Neal & Maris, 1985).
Use of rhythmic events registered in hard tissues (e.g., bands in scales, otoliths,
fin spines,vertebrae) has been the most productive technique for chronologicalage
estimation in manyfish species. These conventional techniques are often inadequate
when applied to tropical fishes (Brothers 1987). Even with the discovery of daily growth
increments in theotoliths of juvenile fishes (Pannella 1980), Prince & Lee (1991) have
shown that thecorrespondence betweendays andincrements can break down in older
1
individuals. These problems have led researchers to the realization that alternative,
innovative methods of age determination are desirable to obtain the information required
to manage fish populations.
The quantitative assessment of cellular metabolites accumulated during the process
of physiological aging has been suggested as an alternative method for accurate estimation
of chronological age. One class of metabolites, commonly termed 'fluorescent age
pigments' (pAP), occurs in all animals studied to date, and is considered by many to be
the most clearly discernible evidence of aging in the cell (Sohal & Wolfe, 1986). Not
degradable by normal enzymatic processes, FAP polymers accumulate in cells over time,
and their quantity may provide an index of the relative age of the cell, tissue, and
ultimately the whole animal. Thus far, the study ofFAP has centered mostly on
gerontological research, but interest is growing among field biologists in the possible
application of these methodologies to study rates of aging, estimate chronological age,
and predict longevity in wild stocks. Early attempts to determine the age of animals in the
field using this approach have met with limited success because of high variability in FAP
levels and inadequate sampling. A satisfactory relationship between known chronological
age and FAP content has yet to be validated.
Cellular genesis ofFAP
Events leading to the production ofFAP are generally well known. The term
FAP is used to describe a heterogeneous family of compounds (pre-ceroids, ceroids, and
lipofuscins) which accumulate in the cell as a result of both normal metabolic activity and
general molecular 'wear and tear'. First described by Hannover (1842), age pigments
can be observed as yellowish-brown, polymorphic granules in the cytoplasm. Chemical
analysis has revealed that they are highly polymerized collections of lipids, protein
fragments, and acid-hydrolysis resistant residues (Bjorkerud 1964). They are believed to
2
collect as the result of numerous intracellular digestions by lysosomes of peroxidized
lipids and damaged proteins and organelles (Tsuchida et al. 1987).
Free radical production during cellular metabolism is considered the starting point
for age pigment formation (see reviews by Hannan 1984, Porta 1991). Superoxide
radicals produced in the respiratory chain are converted to hydrogen peroxide in the
presence of the superoxide dismutase (Fig. 1-1). Under homeostatic conditions,
hydrogen peroxide is subsequently converted to oxygen and water by catalase or
glutathione peroxidase. It is inevitable that some hydrogen peroxide will, in the presence
of divalent metals, dissociate into a hydroxide ion and a hydroxyl radical (Fig. 1-1).
Hydroxyl radicals are highly reactive, and there are no known protective mechanisms
against their attack on biomolecules.
Hydroxyl radicals may damage many types of biomolecules, but the unsaturated
fatty acids and membrane phospholipids are probably the most important to consider in
terms of FAP formation. In the example illustrated in Fig. 1-2, attack on linolenic acid by
a hydroxyl radical results in the formation of a lipid radical, which is oxidized in
subsequent steps to form an endoperoxide. Endoperoxides dissociate into two smaller
lipid radicals and malondialdehyde, a powerful bifunctional crosslinking agent in the
cytoplasm. Free radical-mediated lipid peroxidation is further propagated by the two new
lipid radicals unless controlled by antioxidants such as alpha-tocopherol (vitamin E).
Malondialdehyde, a by-product of lipid peroxidation, can crosslink with the amine
groups of any two biological compounds to form autofluorescent condensation products
such as amino-imino propene Schiff base fluorophores (Tappel, 1975), 1,4
Dihydropyridine-3,5-dicarbaldehydes (Kikugawa and Ido 1984), and
polymalondialdehyde (Gutteridge et al. 1977) (Fig. 1-3). Biomolecules crosslinked
through these reactions are inactivated, and the polymers are eventually engulfed by
autophagic lysosomes for intracellular digestion. Crosslinked regions are highly resistant
3
--_._ .._......
to degradation through normalenzymatic processes, and graduallyaccumulate in number
and size over time (Sohal and Wolfe 1986, Goebel 1987)as age pigment(pre-ceroid,
ceroid, and lipofuscin) granules (Fig. 1-4).
Manyfactors are thought to affect the rate ofFAP genesis. Theseinclude
metabolic rate (Sohal& Donato 1979), temperature (McArthur& Sohal 1982), degree of
fatty acid and membranephospholipid saturation (Sohalet al. 1984),production of
specificproteolytic enzymes (Katz& Shanker 1989), disease state (Hammer and Karakiri
1987,Koppang 1987,Hall et al. 1989), and in particular the efficiencyof antioxidant
defensemechanisms which intercept damaging freeradicals (Armstrong 1984,Brizzee et
al. 1984). Cell age may affect the abilityto preventfree radicaldamagethrough changes
in theexpression of antioxidant enzymes (Sohalet al. 1983); FAP may accumulate at a
faster rate in the tissues of senescent or diseasedanimals(Timiras 1974, Donato & Sohal
1978, Armstrong 1984, Nandy 1985,Porta 1987).
Quantification ofFAP
Agepigmentshave beenobserved at both the histological and biochemical levels.
Traditional histological methodshaveincluded directquantification of age pigment
granulesin stained tissues by lightmicroscopy (Strehler1964), epifluorescence
microscopy (Brizzee& Jirge 1981), and transmission electron microscopy (TEM)(Agius
& Agbede 1984). ProponentsofTEM methodshaveattemptedmorphometric
quantification throughdigitization of micrographs. Unfortunately, TEMmethods are time
consuming and expensive. More significantly, morphometric quantification is subject to
analytical biasesassociated with the selection of specific areas in the tissue and cell for
countinggranule numbersand for calculating percentvolumeof the celloccupied by the
granules.
An alternative method, introduced by Tappeland coworkers (Chio& Tappel
1969,Chio et al. 1969,Fletcher et al. 1973,TappeI1975), involves the measurement of
4
autofluorescent amino-imino propenes associated with age pigment polymers in whole
tissues. This method has gained widespread use in gerontological research on a wide
variety of animals, tissues, and model systems. It involves the homogenization,
extraction and fluorometric quantification of the soluble FAP present in the tissue. Early
studies used only the chloroform or lipid-soluble portion of these compounds, but it is
now known that some FAP's are soluble in more polar solvents as well (Tsuchida et al.
1987). Although the overall chemical composition of FAP products is diverse, the
presence of the autofluorescent crosslinks is universal. Thus, indirect measurement of
soluble FAP by fluorometry produces an index of the amount of all types of FAP present
in the tissue and provides a higher degree of overall replicability than light microscopic
and electron microscopic methods.
Eldred et al. (1982), Eldred (1987), and Porta (1987) have argued that Tappel's
fluorometric technique does not measure the fluorescence of whole lipofuscin granules,
but instead it measures direct products of lipid peroxidation not necessarily present in the
granules. This view has been based on comparison of the fluorescent properties of
solvent extracts with the properties of fluorescent granules visible by light microscopy.
The possibility exists that FAP measured fluorometrically represent more immediate
effects of lipid peroxidation associated with damaged membrane phopholipids (Tappe!
1975, Vladimirov 1986), damaged mitochondria (Dillard & TappeI1971, Tangeras
1983), or cross-linked proteins or oligonucleotides present in the cytosol, not yet
incorporated into large polymers in the lysosome (Kikugawa et al. 1989). Regardless of
the source of the measured fluorescence, Tappel's method has been successfully applied
to a large number of tissue systems over the past twenty years.
Age pigments in aquatic animals
Although age pigments have been described in a wide variety of animals, little
work has focused on aquatic animals such as fish and crustaceans. Some fishes are
5
potentially excellent experimental subjects for gerontological research (Comfort 1960,
1961, 1963; Woodhead 1979, 1980, 1985) because they have relatively short lifespans,
are easy to breed and maintain in captivity, and their metabolic rates can be manipulated.
Most researchers who have studyed FAP in fishes have used histological
procedures. Aboim (1946) reported age pigments in the interrenal cells of the
elasmobranch Raja clavata. Other studies have noted the presence of age pigment
granules in interrenal cells of Prionurus microlepidotus, Pterois lunulata, and Paralichthys
olivaceous (Oguri & Hibiya 1957), and more recently in the same tissues of Lophius
litulon and Triakis~ (Oguri 1986). Oguri (1986) commented on the abundant
occurrence of age pigment granules in fish specimens of large size and "advanced age".
Additional studies have focused on the accumulation of lipofuscin, haemosiderin, and
melanin pigments in macrophages, relating pigment content to fish health and age (Grove
1968; Agius 1980, 1981; Agius & Agbede 1984; Wolke et al. 1981; Brown & George
1985). Brown & George (1985) found that the number of macrophage age pigment
aggregates was more dependent on age than on the fish's health (as indicated by incidence
of parasitism and condition index). Aloj Totaro and coworkers have examined effects of
environmental pollutants on accumulation of lipofuscin granules in spinal ganglia of the
electric ray, Tomedo mannoratus, fmding higher levels in specimens from copper
polluted areas (Aloj Totaro et al. 1985; Aloj Totaro & Pisanti 1987). Unfortunately, no
consideration was given to the chronological age of their samples.
Fewer investigators have measured FAP in fish fluorometrically. Luo & Hultin
(1986) found that chloroform soluble FAP increased with age (estimated with unvalidated
scale annuli) in the skeletal muscle of winter flounder (pseudopleuronectes americanus).
Hammer & Madhusudhana-Rao (1987) found that presence or absence of dietary vitamin
E affected the final quantity of FAP in Oreochromis aurea brain, but had no effect on
levels in four other tissue types. Hammer & Karakiri (1988) measured FAP content in
6
field-captured Limanda limanda and found that individuals infected with lymphocystis
had significantly higher levels of the pigment Finally, Hammer (1988) tested effects of
temperature on FAP levels in larval Esox lucius, finding higher rates of accumulation in
larvae reared at suboptimal temperatures. Hill & Radtke (1988) measured FAP in
Dascyllus albisella brains, fmding a positive relationship between FAP content and
specimen size and age (estimated from unvalidated annuli in sagittae). Similar studies
have been conducted on dover sole and trout (Vemet et al, 1988) as well as larval
grunion, seabass, and halibut (Mullin & Brooks 1988) with variable results.
Ettershank (1983, 1984a, 1984b, 1985) was the first to apply FAP techniques to
crustaceans. He attempted to resolve age cohorts in Antarctic krill (Euphausia s\Werba),
which were previously thought to live three years based on length frequency
distributions. Ettershank found modal progressions of FAP content, assumed to be
annual, which led him to the conclusion that krill live up to six or seven years. Nicol
(1987) applied Ettershank's methods to field caught specimens of the euphausid,
Meganyctiphanes norvegica, and found a linear increase in FAP content with body
weight; however, fluorescence could not be compared to chronological age. More
significantly, Nicol pointed out a major flaw in Ettershank's use of formalin as a
specimen preservative by demonstrating enhanced fluorescence of samples after storage
in this powerful cross-linking agent. Subsequent investigators have committed similar
errors in the use of formalin to preserve spider crab larvae~ araneus; Hirche and
Anger 1987) and krill ill:. superba; Berman et al. 1989). Crossland et al. (1988) found an
increase in FAP in lobster (Panulirus cygnus) and scampi (Metanephrops andamanicus)
eye and tail muscle tissues with increasing size, however chronological age of their
specimens was unknown. Sheehy and Ettershank (1989) extracted FAP from whole
bodies ofDaphnia carinatl!, fmding an increase in absolute FAP content with increasing
age, but a decrease in the weight-specific content with increasing age. The decrease in
7
weight-specific FAP can be attributed to the dilutionof FAP by continuedtissuegrowth.
Most recently,Nicolet al. (1991)maintainedpopulations of E. superbain the laboratory
and measuredsignificant increases in absoluteand weight-specific FAP after one yearof
captivity.
PatternsofFAP accumulation have beenexaminedin only two molluscan species.
Nicol (1987)extractedFAP from mantle tissueof squid, nex illecebrosus, finding
increases in FAP contentwith increasingsize. Chronological age of his specimens was
not known. Clarkeet al. (1990)examinedFAP accumulation in field specimens of the
troehid gastropod Monodonta lineatawhose chronological age had beenestimatedfrom
shell annuli. Whole organFAP contentin brain and digestive gland increasedwith
estimated age, but in a variable manner. Weight-specific FAPvaluesdecreased with age,
a trend attributedto tissue growth.
Research problem
Little is yet understood about the relationship between FAP contentand
organismalage and growth in aquaticpoikilotherms. Even though the previously
mentioned studiesseempromising, results are preliminary at best. Each fluorescence
study has usedsomevariation of a technique originally developed. by Bligh & Dyer
(1959) for whole lipid extraction and later modified by Fletcheret al. (1973) for the
measurementofFAP. However,few of thesevariations have beenreplicable basedon
informationin the published reports of the studies. Many aspectsof the methodologies
have simplybeen assumed to workproperly, with no potentially detrimental effects to the
final expressionof fluorescence takeninto account. In addition, the manner in which
fluorescence results havebeenreported has beenobscure at best, with no standardization
or valid comparison possiblebetweenstudies.
8
As a result of the above limitations, it has yet to be demonstrated that FAP can be
used as an effective tool for predicting age of animals from the field. Only three
investigators (Vernet et al. 1988, Hammer 1988, and Mullin & Brooks 1988) have
examined FAP in fish of known age. Vernet et al. (1988) measured FAP content in three
year classes of laboratory reared Oncorhyncus mykiss (of mixed brood stock origin),
with highly variable results. Mullin & Brooks (1988) measured FAP in three larval fish
species (California grunion, Leuresthes~; white seabass, Atraetoscion nobilis; and
California halibut, Paralichthys californicus) up to 30 d in age, finding negative
correlations between age and FAP content. Hammer's (1988) use of methanol to
preserve larval pike specimens probably altered the final expression ofFAP fluorescence
in a fashion similar to Ettershank's krill samples (Nicol 1987). Even if a stable
relationship between age and FAP were verified, its usefulness for aging field specimens
would depend on a demonstration that information on FAP content can improve on the
accuracy of age estimates presently available for fish through other means.
Further - and thus far not considered in age estimations using FAP - a number of
environmental and physiological factors are likely to influence the rates ofFAP
accumulation in aquatic poikilotherms. Temperature, light, and diet are foremost among
the exogenous factors known to affect ontogenetic changes in growth (Kinne 1960;
Paloheimo & Dickie 1966; Brett 1979; Wurtsbaugh & Cech 1983; Cui & Wooton 1988).
Some of these factors have been implicated in determining the rate of FAP formation in
various poikilothermic (Sheldahl & Tappe11974; Ragland & Sohal 1975; Sohal & Donato
1979; Sohal et al. 1985) and homeothermic animals (porta 1987) and in liposomal model
systems (Petkau 1986). If these factors influence both growth and rates of FAP
accumulation in animals, they are likely to induce variability in this rate at a given age as
well as cwnulative amounts of FAP over longer periods of time. This is significant if
FAP quantities are to be used to predict the chronological ages of wild specimens.
9
Study objectives
In this dissertation, I study patterns of FAP accumulation and variation in a
freshwater <Puntiusconchonius Hamilton) and a euryhaline <Oreochromis rnossambicus
Peters) fish to evaluate the potential for accurate age estimation for management of
fisheries stocks. Specific objectives are: 1) to reassess aspects of FAP assay
methodologies and determine effects of these procedures on final FAP fluorescence
expression; 2) to establish the relationship between FAP content and known
chronological age in the selected species and to build age-predictive models using
multivariate analysis; and 3) to examine effects of specific factors (body size, ambient
temperature, food ration, and photoperiod) which may induce variability in the
relationship between age and FAP content.
Completing the first objective is critical to accomplishing the other two objectives.
In Chapter 2, various aspects of sample handling and extraction procedures are
investigated for their effects on final expression of FAP fluorescence in tissue samples.
This work includes evaluation of the effects of sample storage, extraction, and extract
storage procedures for a variety of post-mitotic tissues of the selected fish species. Once
procedures for extraction and measurement of the compounds have been tested and
defined, the methods are applied to tissues of the study species to determine the basic
relationship between chronological age and FAP content (Chapters 2 and 3). After the
patterns ofFAP accumulation have been established, I incorporate fish size, otolith size,
organ size, and FAP data into various multiple linear regression models (Boehlert 1985)
to examine the utility ofFAP data for chronological age prediction (Chapter 3). In
Chapter 4, experiments are conducted to determine the degree to which FAP content may
vary in P. conchonius tissues under different environmental and physiological conditions.
10
Literature cited
Aboim, A. N. 1946. L'Organe interrenal des cyclostomes et des poissons. Portugal.
Acta. Biol, Ser. A. 1: 353-383.
Agius, C. 1980. Phylogenetic development of melano-macrophage centres in fish. J.
Zool. 191: 11-31.
Agius, C. 1981. Preliminary studies on the ontogeny of the melano-macrophages of the
teleost hemopoietic tissues and age-related changes. Dev. Compo Anat. 5: 597
606.
Agius, C., and S. A. Agbede. 1984. An electron microscopical study on the genesis of
lipofuscin, melanin and haemosiderin in the haemopoietic tissues of fish. 1. Fish
BioI. 24: 471-488.
Aloj Totaro, E., F. A. Pisanti, and P. Glees. 1985. The role of copper level in the
formation of neuronal lipofuscin in the spinal ganglia of Tomedo m. Mar.
Environ. Res. 15: 153-163.
Aloj Totaro, E., and F. A. Pisanti. 1987. Lipofuscinogenesis and environmental stress:
study in Tomedo m. neurons. pp. 133-146 In: Aloj Totaro, E. and FA. Pisanti.
Advances in Age Pigments Research. Advances in the Biosciences. Pergamon
Press, N.Y.
Armstrong, D. 1984. Free radical involvement in the formation of lipopigments. pp. 129
141 In: Armstrong, D., R.S. Sohal, R.G. Cutler and T.F. Slater (Eds.) Free
Radicals in Molecular Biology, Aging, and Disease. Aging. Raven Press, N.Y.
Bagenal, T. B.(ed.) 1974. The ageing of fish. Unwin Brothers Ltd., England, 234p.
Berman, M. S., A. L. McVey, and G. Ettershank. 1989. Age determination of Antarctic
krill using fluorescence and image analysis of size. Polar Biol. 9: 267-271.
Bjorkerud, S. (1964). Isolated lipofuscin granules - survey of a new field. Adv.
Gerontol. Res. 1: 257-288
11
----_ .._-_ ..
Bligh, E. G., and W. 1. Dyer. 1959. A rapid method of total lipid extraction and
purification. Can. J. Biochem. and Physiol, 37: 911-917.
Boehlert, G. W. 1985. Using objective criteria and multiple regression models for age
determination in fishes. Fish. Bull., U.S. 83:103-117.
Brett, J. R. 1979. Environmental factors and growth. pp. 599-675 In: Hoar, W.S., DJ.
Randall and J.R. Brett, (Eds.) Fish Physiology. Academic Press, Inc.
Brizzee, K. R., D. E. Eddy, D. Harman, and 1. M. Ordy. 1984. Free radical theory of
aging: effect of dietary lipids on lipofuscin accumulation in the hippocampus of
rats. Age 7: 9-15.
Brizzee, K. R., and S. K. Jirge. 1981. Fluorescent microscope techniques for the
visualization and histological quantification of autofluorescent lipofuscin bodies in
brain tissues. Johnson, J. E. (Ed.) Current Trends in Morphological Techniques.
CRC Press.
Brothers, E. B. 1987. Methodological approaches to the examination of otoliths in aging
studies. pp.319-330 In: Summerfelt, R. C. and G. E. Hall (Eds.) The Age and
Growth of Fish. The Iowa State University Press, Ames, Iowa.
Brown, C. L., and C. 1. George. 1985. Age-dependent accumulation of macrophage
aggregates in the yellow perch, Perca flavescens (Mitchill). J. Fish Diseases 8:
135-138.
Chio, K. S., and A. L. Tappel, 1969. Synthesis and characterization of the fluorescent
products derived from malondialdehyde and amino acids. Biochem 8: 2821
2827.
Chio, K. S., U. Reiss, B. Fletcher, and A. L. Tappel, 1969. Peroxidation of subcellular
organelles: formation of lipofuscin-like fluorescent pigments. Science 166: 1535
1536.
12
Comfort, A. 1960. The effect of age on growth-resumption in fish CLebistes) checked by
food restriction. GerontoI. 4: 177-186.
Comfort, A. 1961. The longevity and mortality of a fish CLebistes reticulatus Peters) in
captivity. Geronto1.5: 209-222.
Comfort, A. 1963. Effect of delayed and resumed growth on the longevity of a fish
CLebistes reticulatus, Peters) in captivity. Gerontol. 8: 150-155.
Clarke, A., M. A. Kendall, and D. J. Gore. 1990. The accumulation of fluorescent age
pigments in the trochid gastropod Monodonta lineata. J. Exp. Mar. BioI. Ecol.
144: 185-204.
Crossland, C. J., G. Denby, B. F. Phillips, and R. Brown. 1987. The use of fluorescent
pigments (lipofuscin) for ageing western rock lobster (Panulirus cygnus) and
scampi CMetanephrops andamanicus): a preliminary assessment CSIRO Mar.
Lab. Rep. No. 195: 1-21.
Cui, Y., and R. J. Wootton. 1988. Bioenergetics of growth of a cyprinid, Phoxinus
phoxinus (L.): the effect of ration and temperature on growth rate and efficiency.
J. Fish BioI. 33: 763-774.
Dillard, C. J., and A. L. Tappel. 1971. Fluorescent products of lipid peroxidation of
mitochondria and microsomes. Lipids 6: 715-721.
Donato, H., and R. S. Sohal. 1978. Age-related changes in lipofuscin-associated
substances in the adult male housefly, Musca domestic;!. Exp. GerontoI. 13: 171
179.
Eldred, G. E. 1987. Questioning the nature of the fluorophores in age pigments. pp.23
36 In:. Aloj Totaro, E. and F. A. Pisanti (Eds.) Advances in Age Pigments
Research. Advances in the Biosciences. Pergamon Press, N.Y.
Eldred, G. E., G. V. Miller, W. S. Stark, and L. Feeney-Burns. 1982. Lipofuscin:
resolution of discrepant fluorescence data. Science 216: 757-759.
13
Ettershank, G. 1983. Age structure and cyclical annual size change in the Antarctic krill,
Euphausia superba. Polar BioI. 2: 189-193.
Ettershank, G. 1984a. A new approach to the assessment of longevity in the Antarctic
krill Euphausia superba. J. Crust. BioI. 4: 295-305.
Ettershank, G. 1984b. Methodology for age determination of Antarctic krill using the age
pigment lipofuscin. BIOMASS Handb. No. 26: 1-14.
Ettershank, G. 1985. Population age structure in males and juveniles of the Antarctic
krill, Euphausia superba Dana. Polar BioI. 4: 199-201.
Fletcher, B. L., C. J. Dillard, and A. L. TappeI. 1973. Measurement of fluorescent lipid
peroxidation products in biological systems and tissues. Anal. Chern. 52: 1-9.
Goebel, H. H. 1987. Ultrastructural features oflipopigments in neuronal ceroid
lipofuscinoses, especially in biopsies. pp. 249-275 In: Aloj Totaro, E., P. Glees,
and F. A Pisanti (Eels.) Advances in Age Pigments Research. Advances in the
Biosciences. Pergamon Press, N.Y.
Grove, J. H. 1968. Hemosiderin in bluegill spleens. Trans. Am. Fish. Soc. 97: 48-50.
Gutteridge, J. M. C. 1987. Oxygen radicals, transition metals and aging. pp. 1-22. In:
Aloj Totaro, E., P. Glees, and F. A Pisanti (Eds.) Advances in Age Pigments
Research. Advances in the Biosciences. Pergamon Press, N.Y.
Hall, N. A, B. D. Lake, D. N. Palmer, R. D. Jolly, and A D. Patrick. 1989.
Glycoconjugates in storage cytosomes from ceroid-lipofuscinosis (Batten's
disease) and in lipofuscin from old-age brain. pp. 225-242 In: Porta, E.A (ed),
Lipofuscin and Ceroid Pigments. Plenum Press, New York.
Hammer, C. 1988. Accumulation of fluorescent age pigments in embryos and larvae of
pike, Esox lucius (Esocidae, Teleostei) at different incubation temperatures.
Environ. BioI. Fish. 22:91-99.
14
Hammer, C. and M. Karakiri. 1988. Disease dependent accumulation of fluorescent age
pigments in the common dab <Limanda limanda, Teleostei). pp. 445-446 In: 2s.
Nagy, I. (00.) Lipofuscin - 1987: State of the Aart, Akademiai Kiad6, Budapest
and Elsevier Science Publishers, Amsterdam.
Hammer, C., and G. Madhusudhana-Rao. 1987. Individual and feed dependent
accumulation of fluorescent age pigments in five organs of tilapia (Oreochromis
~. Zeitschr. Angew. Z001. 74: 275-292.
Hannover, A. 1842. Mikroskopische Undersolgeser afNervensystemet. K. Dan.
Vidensk. Selsk. Nat. Math. Afb. 10: 1-112.
Harman, D. 1984. Free radicals and the origination, evolution, and present status of the
free radical theory of aging. pp. 1-12 in Armstrong, D., R. S. Sohal, R. G.
Cutler, and T. F. Slater (OOs.) Free Radicals in Molecular Biology, Aging, and
Disease. Raven Press, N.Y.
Hill, K. T., and R. L. Radtke. 1988. Gerontological studies of the damselfish, Dascyllus
albisella. Bull. Mar. Sci. 42: 424-434.
Hirche, H.-J., and K. Anger. 1987. The accumulation of age pigments during larval
development of the spider crab, Hyas araneus (Decapoda, Majidae). Compo
Biochem. Physio1. 88B: 777-782.
Katz, M. L., and M. J. Shanker. 1989. Development of lipofuscin-like fluorescence in
the retinal pigment epithelium in response to protease inhibitor treatment. Mech.
Age. Dev. 49: 23-40.
Kikugawa, K., Ido, Y. 1984. Studies on peroxidized lipids. V. Formation and
characterization of 1,4-dihydropyridine-3,5-dicarbaldehydes as model of
fluorescent components in lipofuscin. Lipids 19: 600-608
15
Kikugawa, K., T. Kato, and A. Iwata. 1989. A tetrameric dialdehyde formed in the
reaction of butyraldehyde and bebenzylamine: a possible intermediary component
for protein cross-linking induced by lipid oxidation. Lipids 24: 962-969.
Kinne, O. 1960. Growth, food intake, and food conversion in a euryplastic fish exposed
to different temperatures and salinities. Physiol. Zool. 33: 288-317.
Koppang, N. 1987. Hereditary ceroid lipofuscinogenesis in English setter dogs. A model
for human neuronallipofuscinogenesis and aging. pp. 185-195 In: Aloj Totaro,
E., P. Glees, and F. A. Pisanti (Eds.) Advances in Age Pigments Research.
Advances in the Biosciences. Pergamon Press, N.Y.
Luo, S.-W., and H. O. Hultin. 1986. Effect of age of winter flounder on some properties
of the sarcoplasmic reticulum. Mech. Age. Dev. 35: 275-289.
McArthur, M. C., and R. S. Sohal. 1982. Relationship between metabolic rate, aging,
lipid peroxidation, and fluorescent age pigment in milkweed bug, Oncopeltus
fasciatus (Hemiptera). J. Gerontol. 37: 268-274.
Mullin, M. M., and E. R. Brooks. 1988. Extractable lipofuscin in larval marine fish.
Fish. Bull., U.S. 86: 407-415.
Nandy, K. 1985. Lipofuscin as a marker of impaired homeostasis in aging organisms.
pp. 139-148 In: Davis, B.B. and W.G. Wood (Eds.) Homeostatic Function and
Aging. Raven Press, N.Y.
Neal, R. A. and R. C. Maris. 1985. Fisheries biology of shrimplike animals. pp.2-11O
In: Provezano, A.J., Jr. (Ed.), The Biology of Crustacea, Vol. 10, Economic
Aspects: Fisheries and Culture. Acad. Press, New York.
Nicol, S. 1987. Some limitations on the use of the lipofuscin ageing technique. Mar.
BioI. 93: 609-614.
16
Nicol, S., M. Stolp, and G. W. Hosie. 1991. Accumulation of fluorescent age pigments
in a laboratory population of Antarctic krill Euphausia superba Dana. J. Exp. Mar.
BioI. Ecol. 146: 153-161.
Oguri, M. 1986. On the yellowish brown pigment granules in the interrenal cells of
anglerfish and leopard shark. Bull. Jap. Soc. Sci. Fish. 52: 801-804.
Oguri, M., and T. Hibiya. 1957. Studies of the adrenal glands ofteleosts - II. On the
adrenal tissues of 15 species of fishes. Bull. Jap. Soc. Sci. Fish. 23: 144-149.
Paloheimo, J. E., and L. M. Dickie. 1966. Food and growth of fishes. II. Effects of
food and temperature on the relation between metabolism and body weight. J.
Fish. Res. Bd. Can. 23: 869-908.
Pannella, G. 1980. Growth patterns in fish sagittae. pp. 519-560 In: Rhoads, D.C. and
R.A. Lutz (Eels.). Skeletal Growth of Aquatic Organisms: Biological Records of
Environmental Change. Plenum Press, New York. 588p.
Petkau, A. 1986. Protection and repair of irradiated membranes. pp.481-508 In:
Johnson, J.E., R. Walford, D. Harman, and 1. Miguel (Eds.) Free Radicals,
Aging, and Degenerative Processes. Modern Aging Research. Alan R. Liss, Inc.,
New York.
Porta, E. A. 1987. Tissue lipoperoxidation and lipofuscin accumulation as influenced by
age, type of dietary fat and levels of vitamin E in rats. pp. 37-74 In: Aloj Totaro,
E. and F. A. Pisanti (Eds.) Advances in Age Pigments Research. Advances in
the Biosciences. Pergamon Press, N.Y.
Porta, E. A. 1991. Advances in age pigment research. Arch. Gerontol. Geriatr. 12: 303
320.
Prince, E. D., D. W. Lee, J. R. Zweifel, and E. B. Brothers. 1991. Estimating age and
growth of young Atlantic blue marlin Makaira nigricans from otolith
microstructure. Fish. Bull., U.S. 89: 441-459.
17
Ragland, S. S., and R. S. Sohal. 1975. Ambient temperature, physical activity and aging
in the housefly, Musca domestica. Exp. Gerontol. 10: 279-289.
Ricker, W. E. 1979. Growth rates and models. pp. 677-743 In: Hoar, W.S., D.J.
Randall and J.R. Brett (Eds.) Fish Physiology. Academic Press, Inc., New
York.
Sheehy, M. R. J., and G. Ettershank. 1989. Extractable age pigment-like
autofluorescence and its relationship to growth and age in the water-flea Daphnia
carlnata King (Crustacea: Cladocera). Aust. 1. Zool. 36: 611-625.
Sheldahl, J. A, and A. L. Tappe1. 1974. Fluorescent products from aging Drosophila
melanogaster: An indicator of free radical lipid peroxidation damage. Exp.
Gerontol. 9: 33-41.
Sohal, R. S., and H. Donato Jr. 1979. Effect of experimental prolongation of lifespan on
lipofuscin content and lysosomal enzyme activity in the brain of the housefly,
Musca domestica. 1. Gerontol. 34: 489-496.
Sohal, R. S., K. J. Farmer, R. G. Allen, and N. R. Cohen. 1983. Effect of age on
oxygen consumption, superoxide dismutase, catalase, glutathione, inorganic
peroxides and chloroform-soluble antioxidants in the adult male housefly, Musca
domestica. Mech. Age. Dev. 24: 185-195.
Sohal, R. S., R. G. Bridges, and E. A Howes. 1984. Relationship between lipofuscin
granules and polyunsaturated fatty acids in the housefly, Musca domestica. Mech.
Age. Dev. 25: 355-363.
Sohal, R. S., A MUller, B. Koletzko, and H. Sies. 1985. Effect of age and ambient
temperature on n-pentane production in the adult housefly, Musca domestica.
Mech. Age. Dev. 29: 317-326.
18
Sohal, R. S., and L. S. Wolfe. 1986. Lipofuscin: characteristics and significance. pp.
171-183 In: Swaab, D.F., E. Fliers, M. Mirmiran, W.A. Van Gool, and F. Van
Haaren (Eds.) Prog. Brain Res. Elsevier Science Publishers, Amsterdam.
Strehler, B. L. 1964. On the histochemistry and ultrastructure of age pigment. pp. 343
384 In: Strehler, B.L. (Ed.) Adv. Gerontol. Res .. Academic Press, N.Y.
Tangeras, A. 1983. Iron content and degree of lipid peroxidation in liver mitochondria
isolated from iron-loaded rats. Biochim. Biophys. Acta 757: 59-68.
Tappel, A. L. 1975. Lipid peroxidation and fluorescent molecular damage to membranes.
pp. 145-170 In: Trump, B. F. and A. V. Arstila (Eds.) Pathobiology of Cell
Membranes.
Timiras, P. S. 1974. Degenerative changes in cells and cell death. pp. 429-449 In:
Timiras, P.S. (Ed.) Developmental Physiology and Aging. MacMillan, N.Y.
Tsuchida, M., T. Miura, and K. Aibara. 1987. Lipofuscin and lipofuscin-like
substances. Chern. Phys. Lip. 44: 297-325.
Vernet, M., J. R. Hunter, and R. D. Vetter. 1988. Accumulation of age pigments
(lipofuscin) in two cold-water fishes. Fish. Bull.,U.S. 86: 401-407.
Vladimirov, Y. A. 1986. Free radical lipid peroxidation in biomembranes: mechanism,
regulation, and biological consequences. pp. 141-195 In: Johnson, T.E., R.
Walford, D. Harman and J. Miquel (eds.) Modern Aging Research: Free
Radicals, Aging, and Degenerative Processes. Alan R. Liss, Inc., New York.
Wolke, R. E., C. J. George, and V. S. Blazer. 1981. Pigmented macrophage
accumulations (MMC, PMB): possible monitors of fish health. Proceedings of
the US-USSR Symposium on Parasites and Pathogens of the World Oceans,
Leningrad
Woodhead, A. D. 1979. Senescence in fishes. Symp. Zool. Soc. Lond. 44: 179-205.
Woodhead, A. D. 1980. Fish in studies of aging. Exp. Gerontol. 13: 124-140.
19
Woodhead, A. D. 1985. Feral Fishes. In: Lints, F. A. (ed.). Non-mammalian models for
research on aging. Interdiscipl. Topics Geront. 21: 22-50. Karger, Basel.
Wurtsbaugh, W. A., and J. 1. Cech. 1983. Growth and activity of juvenile
mosquitofish: temperature and ration effects. Trans. Am. Fish. Soc. 112:
653-660.
20
Fig. 1-1. Modulation of superoxide radicals produced through normal cellular
metabolism. Adapted from Zs.-Nagy (1987).
21
_..__.._---------_._----_.- -.
Catalase
Glutathioneperoxidase
Superoxidedismutase
SuperoxideRadical
120~ I2H+
°2IH2021----->~H20 +~ O2
Fe (II)
Fe(III)
IOH"1Hydroxyl Radical
22
Fig. 1-2. Free radical-mediated lipid peroxidation and malondialdehyde production.
Adapted from Petkau (1986).
23
_•.__.._-_._-_ _-
-
Hydroxyl Radical Attack
OOH Linolenic Acid
OOH Lipid Radical
OOH Diene Conjugate
OOH Peroxy Radical
OOH Endoperoxide
OOH
0-0-
~ 0-0
Fe complexes ~
~ 0- 0-
~__~OOH
~rn Malondialdehydeo 0
24
Fig. 1-3. Fluorescent crosslinkage reactions of malondialdehyde. Adapted in part from
Gutteridge (1986).
25
MalondialdehydeOHC-CH 2 -CHO
OHi1;H0
NI
R
1,4-Dihydropyrldine3,5-dicarbaldehydes
(390/450nm)
II+[OHC-C-CH=ln
Polymalondialdehyde(370/480nm)
~N=CH-CH=CHOHr+ R,NH2
"N=CH-CH=CH-NHR 2
Amino-imino propeneConjugated Schiff Base Fluorophores
(EX 340·360nm. EM 410·470nm)
26
Fig. 1-4. Summary schematic of processes leading to fluorescent age pigment
formation.
27
Unsaturated Fatty Acids & Phospholipids
Free Radical Attack
Peroxidation
Malondialdehyde & OtherCarbonyl Compounds
Crosslinkages
Primary Amine Groups- phospholipids- proteins- free aminoacids- nucleicacids
Cross-Linked Macromolecules ContainingConjugated Schiff Base Fluorophores
-(N-C=C-C=N)-
fusion withIysosomes
Fluorescent Age Pigments(Pre-ceroid, Ceroid, Lipofuscin)
28
CHAPTER 2
CRITICAL ASPECTS OF FLUORESCENT AGE PIGMENTMETHODOLOGIES: MODIFICATION FOR ACCURATE ANALYSIS AND
AGE ASSESSMENTS IN AQUATIC ORGANISMS
Introduction
Age pigments, first described by Hannover (1842), are now recognized as being
one of the most clearly discernible markers of aging in cells (see review by Hammer and
Braum 1988). Age pigments form under normal and pathological conditions in cells as
the result of polymerization reactions between oxidized lipids and proteins (Bjorkerud
1964, Davies 1988, Katz 1990). The residues from these reactions have been variously
classified by their chemical and physical properties, solvent solubilities, fluorescent
properties, and stage of formation, being assigned names such as "lipofuscin" (Hueck
1912), "wear and tear pigment" (HamperlI934), and "ceroid" (Lillie et al. 1941).
Derived from a diverse array of sources, age pigments can now be characterized as
containing numerous autofluorescent components, ranging in emission color from blue
(Hyden and Lindstrom 1950, Barden 1980, Porta et al. 1988) to yellow (Eldred et al.
1982, Katz et al. 1984, Eldred and Katz 1988), many of which have been found to
accumulate with age (Shimasaki et al. 1980, Eldred et al. 1982, Sohal et al. 1983,
Csallany et al. 1984). Many attempts have been made to identify the fluorescent
components of age pigments, however, due to their complex chemical composition, this
topic is likely to remain controversial for some time. Possible sources of fluorescence
include conjugated Schiff bases (N,N' disubstituted-l-amino-3-iminopropenes; Chio and
Tappe11969, TappeI1975), polymalondialdehyde (Gutteridge et al. 1977), 1,4
dihydropyridine-3,5-dicarbaldehydes (Kikugawa and Ido 1984), and membrane bound
flavins (Kitani et al. 1988, Nokubo et al. 1988, 1989).
29
In 1973,Fletcher et aI. introduceda relatively simple spectrofluorometric assay
for the measurementof blue-emitting fluorophores, assumed to be related to age
pigments, from whole tissue samples. Numerous investigators have since adopted
Fletcher et al. 's method, demonstrating a relationship betweenorganismal age and
fluorescence of tissue extracts (Tappelet al, 1973,Miquel et aI. 1978, Bridges and Sohal
1980, McArthur and Sohal 1982, Ettershank et aI. 1983). The use of the terms
"lipofuscin" or "lipofuscin-like" to describesolvent-extractedfluorophores is currently
disputed (Eldred 1987,Eldred and Katz 1989)due to the differentialpropertiesof
fluorescence in tissue extracts comparedto histologically observed preparations. It
should at this point be recognized that Fletcher et al.'s method is probablynot specific to
a single type or class of age pigment; rather it is a more general assay for the blue-emitting
components of age-related fluorophores generated by lipid peroxidationprocesses in the
cell. Thus, for the purposes of this study, we will adopt the term 'fluorescent age
pigment', hereinafter referred to as 'FAP' (Sohal and Buchan 1981,Hammer and Braum
1988), to describe age-relatedfluorophores which have been measured
spectrofluorometrically. Use of more specific terms such as 'lipofuscin' or 'ceroid'
should be reserved for describing age pigments observed histologically.
Spectrofluorometric measurement of metabolically accumulatedFAP compounds
has recently been proposed as a means for estimatingage in wild specimens of aquatic
species. Ettershank (1983, 1984 a.b, 1985)first applied Fletcher et aI. 's (1973) FAP
assay to delineate age cohorts in Antarctickrill <Euphausia superba). Modifications of
Fletcher et al.'s (1973) and Ettershank's (1983, 1984 a.b) protocols have since been
applied to a number of teleost (Luo and Hultin 1986,Hammer and Madhusudhana-Rao
1987,Hammer 1988, Hammer and Karakiri 1988, Hill and Radtke 1988, Mullin and
Brooks 1988, Vernet et aI. 1988) and crustacean (Hirche and Anger 1987, Nicol 1987,
Berman et aI. 1989, Sheehy and Ettershank 1989) species.
30
.__.-----....._--
Results from these preliminarystudieshave been cursory and inconclusive with
regard to the applicationof this methodtowardage estimationfor population analyses.
For example, relativelyfew investigators have tested the method on animals of known
chronological age. Conversely, those studies using sample tissue of known age have
used samplingperiods insufficientto determine long-termontogeneticchanges in FAP
content. In addition, a survey of the methods thus far employed reveals a number of
unsubstantiated modifications to the extraction procedureand a lack of standardization for
reportingFAP data, thus negating the directcomparabilityof results between studies.
Using three marine invertebrate species, Nicol (1987) was the first to test specific
aspects (i.e., sonication and use of tissue preservatives) of Ettershank's (1983, 1984 a.b)
protocol. Of greatest significancewas his finding that tissue preservatives (formalin and
ethanol) caused in vitro elevationof samplefluorescence. These results were
corroborated by Mullin and Brooks (1988), who induced similareffects in formalin
preserved fish larvae. Both of these fmdings raised questions not only as to the source of
fluorescence (native FAP vs formalin induced)measured in Ettershank's (1983, 1984
a.b, 1985) and Berman et al.' s (1989) krill samples, but also in Hirche and Anger's
(1987) spider crab larvae, and Hammer's (1988) pike larvae. Further, additional
questions with regard to the stabilityof extractableFAP compounds under different
storage and handling conditions remainedunanswered, and in essence have largely been
ignored.
The purpose of the present study was two-fold. The first objective was to
evaluate various aspects (i.e., tissue storageand extract incubation temperature and time,
sonicationof tissue homogenates,assay temperature and the use of aqueous-methanol
extracts)of the methodologies currentlyadoptedfor the spectrofluorometric quantification
ofFAP and to determine their effects on the expression of and variation in sample
fluorescence. For this purpose we chose Oreochromis mossambicusas the model
31
animal, concentrating on postmitotic tissues only. The second objective involved the
development of modified handling, extraction, and assay procedures to negate in-sample
fluorescence variation and the application of these procedures to brain tissues of O.
mossambicus to determine whether a distinct relationship between FAP content and
chronological age exists.
Materials and methods
Model animal/tissues
Known-age specimens of Oreochromis mossambicus, stocked as fry, were reared
for up to 886 d at the Kewalo Basin Research Facility of the National Marine Fisheries
Service, Honolulu Laboratory. Fish were held in circular fiberglass holding tanks at
ambient temperature with daily water changes, and fed ad libitum rations of trout chow
twice daily. Subsampled specimens were transported live to the laboratory, sacrificed by
ice bath immersion, weighed to the nearest O.OIg, and standard and fork lengths
measured to the nearest 0.05 mm. Fish were bled from the caudal region before
removing tissues. A preliminary survey of numerous postmitotic tissues indicated that
brain, ventricle, and skeletal muscle extracts contained detectable levels ofFAP
fluorescence with the least amount of interfering compounds; therefore, these were
chosen as model tissues for the remainder of the study. Dissected tissues were placed in
Eppendorf vials and immediately frozen in liquid nitrogen until analysis.
Tissue storage temperature
Skeletal muscle tissues were placed at various short- and long-term storage
temperatures to detect changes in sample fluorescence over time. White skeletal muscle
was dissected from the left anterior epaxial region of three sibling fish, briefly minced
with a Virtis motorized blade, and stored in Eppendorf vials either at room (23°C) or
refrigerated (3°C) temperatures for 0.5, 1,2,4,8, and 12 h, or frozen at -20 or -80°C for
32
5, 15, 30, 60, 90, and 120 d. Sample vials were frozen in liquid nitrogen at the end of
each time point. Sample fluorescence was compared to that of absolutely fresh tissue.
Tissues were analyzed using the modified extraction procedure, and assayed as described
in later sections.
Unmodified extraction procedures
Extraction and assay procedures adopted for the first part of this study were based
on protocols outlined by Ettershank (1983, 1984 a.b). These included: maceration of
tissues in spectroscopic grade chloroform/methanol (CHCI3!MeOH, 2:1 v/v), sonication,
mixing with a magnesium chloride solution [100 roM in deionized glass distilled water
(GDW)], and centrifugation for 20 min at RCF =3000 x g. Fluorescence of the lower
CHCl3 layer was compared to a standard solution of quinine sulfate (0.1 mg/L IN
sulfuric acid). Further details on extraction procedures are provided, when appropriate,
in the following subsections.
Incubation temperature of sample extracts
The stability ofFAP products in solvent was examined in intermediate-stage
CHCI3!MeOH (2:1 v/v) and final CHCl3 fractions of tissue extracts at room temperature
(23°C) and -20°C for up to 8 h. Lyophilized brain, heart, and muscle tissues from
siblings were pooled by pulverizing and mixing dried tissues in 15 ml glass centrifuge
tubes with a teflon pestle. Stability of intermediate stage extracts was tested by extracting
triplicate aliquots of 20 mg of dry tissue in 2 ml CHCI3/MeOH, centrifuging the
homogenate for 20 min at a relative centrifugal force (RCF) of 3000 x g and 0 to 5°C, and
collecting and storing the CHCI3/MeOH supernatant in amber glass vials at the defined
incubation temperatures. Final stage CHC13 extracts were prepared by adding an equal
volume of 100 roM magnesium chloride solution in GDW to the CHC13!MeOH
homogenates, centrifuging as defined above, and transferring the CHCl3 hypophase to
amber glass vials for incubation.
33
----------- -
Effects of ultrasonication on fluorescence yield
Ultrasonication of tissue homogenates (widely adopted for the disruption of
biological membranes and thus increased exttaetion efficiency) was tested to determine
the overall effects on fluorescence yield ofCHCl3-extractable FAP compounds. Eighty
mg aliquots of lyophilized and pooled brain, heart, or muscle tissues were glass
homogenized in 8 ml CHCI3!MeOH (2:1), transferred to a 15 ml glass centrifuge tube,
and sonicated with a Sonics & Materials VibraCeli model VC300 microtip probe sonicator
for up to 35 min. Sonication output was 25 W at a 50% duty cycle with sample tubes
held in ice and water to absorb heat. 1 ml aliquots of CHCI3!MeOH homogenate were
removed at 5 min intervals, vortex mixed with an equal volume of the 100 mM MgCI
solution, centrifuged, and the CHCl3 layer assayed for FAP content.
Confirmation and removal of interfering compounds
Fletcher et al. (1973) mentioned the importance of removing interfering
fluorescent compounds from CHCI3!MeOH tissue extracts. Retinol fluorescence
(excitation 325 to 340 nm, emission 475 nm), present in the final CHCl3 extract, could
be removed by a brief (30 s) exposure of the solvent to high intensity ultraviolet light.
Flavin compounds fluoresce at 510 to 520 nm with excitation maxima at 280,350, and
450 nm and are present in the CHCI3/MeOH homogenate, removable by a water wash
step. In the present study, complete excitation and emission fluorescence spectra were
obtained before all sample assays to detect the presence of fluorescent contaminants. In
addition, the fluorescent properties of all aqueous/methanol phase extracts were checked
for the presence of these contaminants.
Assay temperature
Temperature dependence of sample and standard fluorescence has been
established (Guilbault 1973), however, most researchers have conducted assays at
ambient temperature. The importance of assay temperature control was tested on
34
Oreochromis mossambicus brain, heart, and muscle extracts by measuring sample
luminescence at temperatures ranging from 0 to 42°C at 3°C intervals. Temperature of the
sample solvent was maintained using a thermostatically controlled turret microcell holder
connected to a Brinkmann Lauda Model RM6 refrigerated circulating water bath.
Resulting changes in the calculated relative fluorescence intensity values were noted.
Modified extraction and assay procedures
Basefl on the results from the above experimental analyses a modified procedure
for the extraction and assay of FAP was implemented. Tissues were prepared for
extraction by immersion in liquid nitrogen and lyophilization to obtain the most accurate
sample weight. Dry tissues were weighed on a Sartorius Model 4503 MP6 electronic
microbalance to the nearest microgram, placed in glass tubes, and returned to the
lyophilizer until extraction. Tissues were macerated at room temperature for 4 min in a
Potter-Elvehjem all-glass tissue homogenizer using at least 1 ml of CHCl3 per 10 mg dry
tissue. Approximately one-third of the solvent typically evaporated during
homogenization. Because of this the initial meniscus was marked and the homogenate
adjusted back to its original volume after treatment. Homogenates were then transferred
to a glass centrifuge tube and vortex mixed for 2 min with an equal volume of a solution
containing 100 roM MgCI in GDW and methanol at a ratio of 3:1. Samples were
centrifuged at D-5°C for 20 min at RCF =3000x g. The lower CHCl3 layer was
immediately removed and assayed for FAP content.
Chloroform extracts were placed in matched quartz silicate microcuvettes with a 4
mm pathlength, sealed with a teflon stopper, and assayed at 18°C in a Perkin Elmer model
LS-5 fluorescence spectrophotometer equipped with a Xenon light source and standard S
5 photomultiplier tube. Excitation and emission slits were both set at 10 nm for standard
assays, and luminescence readings were at a fixed scale setting of 1.000. Excitation and
emission maxima were determined using PreScan mode, and the maximum luminescence
35
at these settings was recorded. Fluorescence spectra reported herein were not corrected
for the spectral emission characteristics of the light source, nor the spectral efficiencies of
the monochromators, optics, and the photomultiplier tube. Quinine sulfate at a
concentration of 0.1 mg/11H sulfuric acid (H2S04) was used as a fluorescence standard,
and the sample readings were expressed as a percentage of the standard luminescence at
an excitation setting of 345 om and emission setting of 450 om. Sample FAP
concentrations are expressed either as relative fluorescence intensity (RFI), where:
sample luminescence solvent volume (ml)RFI = X X 100
standard luminescence sample dry weight (mg)
or as whole organ %fluorescence (%FL), where:
sample luminescenceWhole Organ %FL = d d I' X solvent volume (ml) X 100
stan ar ummescence
Sample luminescence spectral scans were plotted for all samples with a Perkin
Elmer model R100A chart recorder. Spectral scans presented herein were produced by
recording luminescence at 1 nm intervals with the complementary excitation/emission
wavelengths at a fixed setting.
Results
Tissue storage temperature
Chloroform extracts from fresh (t=O) pooled Oreochromis mossambicus muscle
had an average RFI value of 0.4412 (±O.0275) at excitation maxima of 350 to 355 nm
and emission maxima of 425 to 430 nm. Muscle samples stored at elevated (>O°C)
temperatures exhibited linear increases in fluorescence after a matter of hours with no
apparent shifts in excitation/emission maxima Muscle samples held at room temperature
(23°C) had elevated levels of fluorescence after only I h. A mean increase (.1RFI) of
9.3% was evident after only 2 h of incubation, and this increased to as much as 87%
36
-----------
~RFI at the end of 12 h (Fig. 2-1 A) (%dRFI = -1.754 + 6.303 (Hours), r2 = 0.96).
Although refrigeration at 3°C slowed the rate of fluorophore accumulation, a mean
increase of 9.1% ~RFI was measured after 8 h incubation (%~RFI=0.093 + 0.722
(Hours), r2 =0.56). Tissues stored for longer periods at -20°C also exhibited linear
increases in fluorescence, with a mean ~RFI of 9.7% after 15 d (Fig. 2-1 B) (%dRFI =
-0.087 + 0.638 (Days), r2 =0.98). No significant changes in fluorescence were
observed in muscle samples stored at -80°C after 120 d (%8RFI =0.621 + 0.0029
(Days), r2 = 0.003).
Incubation temperature of sample extracts
Fluorescence of CHCI3!MeOH- and CHCl3-soluble extracts of Oreochrornis
mossambicus brain, heart, and muscle increased over time at both room (23°C) and
reduced (-20°C) temperatures (Figs. 2-2 and 2-3). Increases in CHCI3!MeOH extract
fluorescence were accompanied by shifts in the excitation/emission maxima from
approximately 355/430 to 380/455 nm, while fluorescence maxima of CHCl3 extracts
remained unchanged. The change in CHCI3!MeOH extract fluorescence wavelengths
occurred by 3 h at 23°C and by 4 h at -20°C. Chloroform/methanol extracts incubated at
23°C exhibited the most rapid rises in fluorescence, with heart extracts increasing by an
average of 147% ~RFI after 3 h and as much as 880% ~RFI at the end of 8 h (Fig. 2-2
A). Final water-washed CHCl3 extracts of all three tissues increased in fluorescence on
the average of 10 to 16% 8RFI after only 2 h incubation at 23°C with heart extracts again
increasing more than brain or skeletal muscle (Fig. 2-3 A). Incubation at lowered
temperature (-20°C) slowed in vitro changes in sample fluorescence. Even so, increases
were observed within 2 to 4 h of incubation (Figs. 2-2 Band 2-3 B). Brain CHCl3
extracts were the least stable at -20°C, with an average dRFI of7.2% after 4 h and up to
18.5% after 8 h.
37
Effects of ultrasonication
Sonication of CHCI3!MeOH homogenates at 25 W of output caused dramatic
increases in fluorescence for all three tissues, with mean LlRFI values of 200, 119, and
83% for brain, heart, and muscle, respectively, after only 5 min sonication (Fig. 2-4 A).
Continued sonication for up to 30 min caused further increases in fluorescence, after
which levels decreased or stabilized. Of the tissues studied, brain homogenates had the
greatest change with an average LlRFI of 902% after 30 min (Fig. 2-4 A). Following a
sonication period of 5 min, the uncorrected fluorescence spectra of all tissue homogenates
also revealed shifts in excitation/emission maxima similar to those described for
CHCI3!MeOH extracts in the previous subsection. Results are presented for brain tissue
only (Fig. 2-4 B).
Fluorescent contaminants
Uncorrected fluorescence spectra from some brain, heart, and skeletal muscle
extracts showed the presence of a second emission peak in the 510 to 520 nm range in the
CHCI3/MeOH of the intermediate extraction stage as well as in the aqueous phase final
wash. Excitation scans of these samples at an emission setting of 520 nm revealed
excitation maxima at 286,351, and 447 nm (Fig. 2-5 B), confirming the presence of
fluorescent flavin contaminants in the CHCI3!MeOH and aqueous phase solvents. The
second 510 to 520 nm emission peak was not obvious in all CHCI3!MeOH and aqueous
extracts (Fig. 2-5 A). However, excitation scans of all samples at emission =520 nm
confirmed the presence of the three excitation maxima (Fig. 2-5 B). Flavin contaminants
were not present at detectable levels in the final CHCl3 sample solvent, and a second
water wash of the CHCl3 layer contained no flavin or FAP fluorescence. Thus, a single
water wash was sufficient to remove all flavin contaminants. There was no evidence of
retinol contaminants in any of the three chosen tissues.
38
Assay temperature
Absolute luminescence values of both standards and samples decreased with
increasing assay temperature (Fig. 2-6 A). The relationship between assay temperature
(T in "C) and brain extract luminescence (BL, as raw luminescence values) was linear,
where:
BL =0.61109 - 0.01021 (T), r2=0.969
and the relationship between assay temperature (T) and quinine sulfate luminescence (QL)
was best described as:
QL =1.6532 - 0.00467 (T), r2=0.886
The effect of temperature on fluorescence data is better observed when RFI values are
considered (Fig. 2-6 B). A three-fold decrease in RFI occurred over the temperature
range of 0 to 42°C, and inflections were apparent in this relationship (Fig. 2-6 B) within
the working range of room temperatures (20 to 24°C).
FAP accumulation with age
The extraction and assay of whole brain FAP content (% FL) using our modified
procedures resulted in a positive correlation with chronological age (Fig. 2-7 A). This
increase was evident both in immature fish and in mature male and female specimens, and
for the portion of the lifespan examined, it was well described by second order
polynomial regressions (Fig. 2-7 A) (Immature + males: %FL =7.407 - 4.586*10-2
(Days) + 2.33*10-4 (Days)2, r2 =0.926, P< 0.0001, n =101; Immature + females: %FL
=6.347 - 1.41*1O-2(Days) + 1.074*10-4 (Days)2, r2 =0.936, P< 0.0001, n =99). FAP
content (% FL) accumulated gradually during the first 370 d, after which it increased
rapidly and resulted in a marked difference between males and females.
Relative fluorescence intensity values decreased rapidly between 32 and 92 d,
remained low until approximately 330 d, and gradually increased thereafter, also
differentiated by sex (Fig. 2-7 B). After about the first 80 d, relationships between age
39
and RFI were well described using second order polynomial equations (Fig. 2-7 B)
(Immature + males: RFI = -0.366 + 2.033*10-3 (Days) + 3.843*10-6 (Days)2, r2 =
0.921, P< 0.0001, n = 101; Immature + females: RFI = 0.866 - 2.678*10-3 (Days) +
5.872*10-6 (Days)2, r2 =0.935, P< 0.0001, n =99).
Brain growth with age was most rapid initially, slowing after 300 d, and mirrored
the relationship between age and RFI. Relationships between age and brain dry weight
(BW in mg) were well described by logarithmic equations (Fig. 2-7 C) (Immature +
males: BW =-36.488 + 23.333 * loglQ(Days), r2 =0.943, P< 0.0001, n = 123;
Immature + females: BW =-32.101 + 21.067 * 10glQ(Days), r2 =0.914, P< 0.0001, n =121).
Discussion
It is apparent from the results of this study that the temperatures at which tissues
and extracts are handled and the time frame involved markedly affect expression of FAP
fluorescence and can introduce considerable error unless controlled. Similarly,
application of ultrasonication to tissue homogenates drastically alters the properties and
levels of sample fluorescence. These observed changes in tissue and extract fluorescence
suggest the in vitro formation ofFAP-like substances caused by lipid peroxidation (Luo
and Hultin 1986), the production of cross-linking agents (e.g., malondialdehyde and
formaldehyde), and the subsequent genesis of fluorophores in the sample solvent.
The effects of storage, handling, and assay temperatures on the accurate analysis
ofFAP are not surprising. The chemical instability of fish tissues under different
freezing conditions (as related to food quality) has been well documented. Changes
occurring during freezing include "browning" reactions between free amino acids and
sugars (Jones 1959), denaturation of proteins (Sawant and Magar 1961), dissociation of
trimethylamine to dirnethylamine and formaldehyde (Amano and Yamada 1964,
40
Tokunaga 1964, Castell et al. 1971), loss of membrane phospholipid through
autoxidation (Hardy et al. 1979), and increases in muscle fluorescence (Davis 1982).
Davis (1982) used fluorometric methods to document effects of freezing (-5 to -30°C) on
muscle tissues. In a companion paper, increases were attributed to formaldehyde
production and browning reactions between fructose, ribose, and free amino groups
(Davis and Reece 1982). The results reported for skeletal muscle of Oreochromis
mossambicus in the present work indicate similar changes affecting levels of apparent
FAP content, and demonstrate the necessity for proper freezing temperatures from the
moment the animal is sacrificed (i.e., immediate placement and storage in liquid nitrogen,
or at least at temperatures of -80aC or lower).
Because temperature would seem to be a critical factor affecting the native FAP
content of tissues, the increases observed with time in this research suggest that they are
probably due to the in vitro formation of other chemical complexes. We suspect that
these complexes are similar to those described in studies seeking to determine the
chemical nature of FAP fluorophores (Chio and Tappel 1969). Their formation is
presumably initiated by lipid peroxidation, which has been well defmed for muscle
microsome and sarcoplasmic reticulum (SR) model systems of fish. Lipid peroxidation,
measured by malondialdehyde (MDA) production, has been demonstrated for muscle
microsomal fractions from red hake (McDonald et al. 1979) and herring (Slabyj and
Hultin 1982). MDA production was dependent upon the levels of peroxidizing cofactors
(NADH, ATP, and Fe3+) and incubation temperature, and occurred under all conditions
over short periods of time (0 to 60 min) in both studies (McDonald et al. 1979, Slabyj
and Hultin 1982). Related studies using SR model systems have found similar cofactor
effects (Luo and Hultin 1986) as well as inhibition of peroxidation by phospholipase A2
(Shewfelt and Hultin 1983). Luo and Hultin (1986) incubated SR from winter flounder,
documenting increases in both MDA and fluorescence after only 1 and 3 h, respectively.
41
Once formation of MDA has occurred, a cascade of chemical reactions has been proposed
that eventually results in the genesis of blue and blue-green emitting fluorophores.
Chio and Tappel (1969) first described the cross-linking reaction between MDA
and amino acids to form N,N'-disubstituted l-amino-3-iminopropenes which fluoresce at
wavelengths between 400and 550 nm. Similar cross-linking reactions have now been
established between MDA and the primary amine groups of phospholipids (Dillard and
TappeI1971, Bidlack and Tappe11973, Trombly and TappeI1975), proteins (Shimasaki
et al. 1982, Fukuzawa et al. 1985, Kikugawa and Beppu 1987), and nucleic acids (Nair
et al. 1986). Lipid-soluble extracts used for FAP studies represent complex mixtures of
phospholipids, fatty acids, non-polar amino acids, proteins, and many other biological
compounds which are highly susceptible to peroxidation and polymerization reactions.
Our results support the conclusion that temperature and time are important in the
accelerated formation of fluorescent complexes. To eliminate further synthesis of
fluorescent adducts, tissue extracts should be handled expediently within a critical
window of time (from beginning of extraction through fluorescence assay) and assayed at
controlled temperature.
The fluorescence maxima shifts observed in CHCI3/MeOH extracts (whether or
not ultrasonication was applied to the homogenate) suggest the formation of compounds
similar to 1,4-dihydropyridine-3,5-dicarbaldehydes first isolated by Kikugawa and Ido
(1984) through the reaction of primary amines with the hydrolysate of
tetramethoxypropane; however, this process has yet to be substantiated. Regardless of
the source of change, CHCI3/MeOH tissue homogenates are highly unstable under
normal extraction conditions, and ultrasonication of homogenates greatly enhanced the
rates of fluorophore production. This may be due either to generation of free radicals in
sample solution through cavitation or to the localized generation of heat within
42
homogenates. Our results indicate that ultrasonication should be avoided because of the
rapid alteration of sample fluorescence properties.
Overall, our data raise questions as to the nature of fluorescence described in
previous investigations on FAP in aquatic animals (Table 1). A survey of the methods
adopted in these studies emphasizes the fact that few used fresh tissues or stored tissues
at temperatures adequate for stabilization of cellular constituents, most did not mention the
time elapsed between freezing and analysis, and none took extract incubation time and
temperature into full consideration. Additionally, ultrasonication was applied to tissue
homogenates in at least half of these studies (Table 1), and whereas the majority
employed sonication for less than 5 min, most applied a power output four times greater
than that found to generate additional fluorophores under our experimental procedures.
Conversely, Nicol (1987) found no significant differences between sonified and non
sonified Meganyctiphanes norvegica individuals. This was probably because the samples
were preserved in formalin and had already increased in fluorescence prior to extraction.
Moreover, several of the studies cited in Table 1 (i.e., Hammer and Madhusudhana-Rao
1987, Hirche and Anger 1987, Hammer 1988, Hammer and Karakiri 1988) used the
polar extract phase to quantify FAP but failed to verify the absence of flavin contaminants
- which we found to comprise as much as half of the total sample fluorescence. Sheehy
and Ettershank (1989) also reported considerable quantities of flavin-like fluorescence in
CHCI3/MeOH and polar phase extracts from whole Daphnia carlnata. Ettershank (1983,
1984 a.b, 1985) was the only investigator to control temperature during fluorometric
analysis; the others conducted assays at room temperature.
When all these critical aspects ofFAP analysis are considered, it is not surprising
that the data from previous investigations have been too inconsistent for direct
comparison. More importantly, the main goal of most previously published research has
been to establish a relationship between FAP content and chronological age; however,
43
becauseof the highdegreeof variability and thefact that the techniques wereapplied to
animalsof known age in only a few studies (Table1), the relationships reportedare open
to question. In all probability, theyreflect fluorescence modifiedby analytical artifacts
rather than native tissue fluorescence. Of those studieson fishes that have taken
chronological age into account, the most complete is probably that of Vernet et al. (1988),
who measuredFAP content in adultOncorhynchus mykiss of known age (3 mo to 3 yr).
However, they sampledonly 21 individuals derivedfrom different broodstockanddid
not take into accountextractincubation or assaytemperature and time.
In contrast to theseprevious investigations, our study not only took into account
sourcesof variabilityin most aspectsof the extraction and assay procedure,but also
provided a chronological range of samples (from33 to 886 d), with more stringent
samplingprocedures. The presentresearchis the first to describe a clear relationship
betweenblue-emitting FAP accumulation and chronological age in an aquaticorganism.
However,other sourcesof variability have been subsequently exposed that require
further research, in order that the technique can be validated for accurate assessment of
age. For example, initialdecreasesin the weight-specific FAP concentrations in juvenile
Oreochromis mossambicus appear to correspondto a period of rapid brain growth,
suggestingthat althoughFAP accumulates in this tissue, growth of the whole organ
exceeds the rate of accumulation at this developmental stage. Siakotoset al. (1977) found
similarpatternsof accumulation of humanbrain lipofuscin which wasextractedand
purified by differential centrifugation, and speculated that the largepool of lipofuscin
found in youngindividuals may be related to the lysosomesderivedfrom the remodeling
of cellular structures duringearlydevelopmental stages. The markeddifferences in FAP
accumulation between male and female Q. mossambicus have yet to be explained, but will
have direct bearingon the applicability ofFAP analysis for accurateage assessment of
wild stocks.
44
An additional problem must be overcome before spectrofluorometric assays for
FAP can be applied successfully to crustaceans. Many crustacean species contain
considerable quantities of carotenoid compounds such as astaxanthin and canthaxanthin
(Thommen and WackernageI1964, Herring 1968 a.b, Goodwin 1984), that have light
absorptive properties within the emission range ofFAP compounds (Zeller et al. 1959,
Cooper et al. 1975) and are highly soluble in nonpolar solvents such as chloroform.
Thus, these carotenoid compounds may cause internal quenching of FAP fluorescence.
Preliminary investigations of FAP quantification in Penaeus vannamei (Hill, Moss, and
Womersley unpublished data) and Heterocarpus laevigatus (Hill, Shiota, and Womersley
unpublished data) have revealed large quantities of orange colored pigments in nonpolar
extract fractions which have UV/visual absorption maxima similar to those reported for
astaxanthin and canthaxanthin, Increases in orange pigment concentrations were
accompanied by decreases in FAP concentrations. If the observations indicate quenching
phenomena, they may explain the negative correlations between age and weight-specific
FAP fluorescence described by Hirche and Anger (1987) for Hyas araneus and by
Sheehy and Ettershank (1989) for Daphnia Carinata.
Clearly, the fluorometric assay for FAP holds a great deal of promise. However,
further investigation is necessary to determine the source in variations in FAP content
between sexes and the effects of environmental factors (e.g., temperature, light, food
ration) on tissue fluorescence. These and other problems must be resolved before
measurement ofFAP can become a standard method for aging aquatic animals.
45
Literature cited
Amano, K, Yamada, K (1964). A biological formation offonnaldehyde in the muscle
tissue of gadoid fish. Bull. Jap. Soc. scient. Fish. 30: 430-435
Barden, H. (1980). Interference filter microfluorimetry of neuromelanin and lipofuscin in
human brain. J. Neuropathol. expo Neurol. 39: 598-605
Berman, M. S., McVey, A. L., Ettershank, G. (1989). Age determination of Antarctic
krill using fluorescence and image analysis of size. Polar BioI. 9: 267-271
Bidlack, W. R., Tappel, A. L. (1973). Peroxidation of subcellular organelles: formation
of lipofuscin-like pigments. Lipids 8: 203-207
Bjorkerud, S. (1964). Isolated lipofuscin granules - survey of a new field. Adv. geront.
Res. 1: 257-288
Bridges, R. G., Sohal, R. S. (1980). Relationship between age-associated fluorescence
and linoleic acid in the housefly Musca dornestica. Insect Biochem. 10: 557-562
Castell, C. H., Smith, B., Neal, W. (1971). Production of dimethylamine in muscle of
several species of gadoid fish during frozen storage, especially in relation to
presence of dark muscle. J. Fish. Res. Bd. Can. 28: 1-5
Chio, K S., Tappel, A. L. (1969). Synthesis and characterization of the fluorescent
products derived from malondialdehyde and amino acids. Biochemistry (Am.
chern. Soc.) Easton, Pa. 8: 2821-2827
Cooper, R. D. G., Davis, J. B., Leftwick, A. P., Price, C., Weedon, B. C. L. (1975).
Carotenoids and related compounds. Part XXXII. Synthesis of astaxanthin,
phoenicoxanthin, hydroxyechinenone, and the corresponding diosphenols. 1.
chern. Soc. Perkin Trans. (I: Org. bio-org. Chern.) 21: 2195-2204
Csallany, A. S., Ayaz, K L., Menken, B. Z. (1984). Effects of dietary vitamin E upon
fluorescent compounds of the rat uterus. Lipids 19: 911-915
46
Davies, K. J. A. (1988). Protein oxidation, protein cross-linking, and proteolysis in the
formation of lipofuscin. In: Zs.-Nagy, I. (ed.) Lipofuscin - 1987: state of the art.
Akademiai Kiad6, Budapest and Elsevier Science Publishers, Amsterdam, p. 24
25
Davis, H. K. (1982). Fluorescence of fish muscle: description and measurement of
changes occurring during frozen storage. J. Sci. Fd Agric. 33: 1135-1142
Davis, H. K., Reece, P. (1982). Fluorescence of fish muscle: causes of change occurring
during frozen storage. 1. Sci. Fd Agric. 33: 1143-1151
Dillard, C. J., Tappel, A. L. (1971). Fluorescent products of lipid peroxidation of
mitochondria and microsomes. Lipids 6: 715-721
Eldred, G. E. (1987). Questioning the nature of fluorophores in age pigments. In: Aloj
Totaro, E., Glees, P., Pisanti, F. A. (eds.) Advances in age pigments research.
Pergamon Press, Oxford, p. 23-36
Eldred, G. E., Katz, M. L. (1988). Fluorophores of the human retinal pigment
epithelium: separation and spectral characterization. ExpI. Eye Res. 47: 71-86
Eldred, G. E., Katz, M. L. (1989). The autofluorescent products of lipid peroxidation
may not be lipofuscin-like. Free radical BioI. Med. 7: 157-163
Eldred, G. E., Miller, G. V., Stark, W. S., Feeney-Burns, L. (1982). Lipofuscin:
resolution of discrepant fluorescence data. Science, N.Y. 216: 757-759
Ettershank, G. (1983). Age structure and cyclical annual size change in the Antarctic krill,
Euphausia superba. Polar BioI. 2: 189-193
Ettershank, G. (1984a). A new approach to the assessment of longevity in the Antarctic
krill Euphausia superba. 1. Crustacean BioI. 4 (Spec. No.1): 295-305
Ettershank, G. (1984b). Methodology for age determination of Antarctic krill using the
age pigment lipofuscin. BIOMASS Handb. Ser. 26: 1-14
47
Ettershank, G. (1985). Population age structure in males and juveniles of the Antarctic
krill, Euphausia superba Dana. Polar BioI. 4: 199-201
Ettershank, G., Macdonnell, 1., Croft, R. (1983). The accumulation of age pigment by
the fleshfly Sarcophaga bullata Parker (Diptera: Sarcophagidae). Aust. J. Zool.
31: 131-138
Fletcher, B. L., Dillard, C. J., Tappel, A. L. (1973). Measurement of fluorescent lipid
peroxidation products in biological systems and tissues. Analyt. Biochem. 52: 1-9
Fukuzawa, K., Kishikawa, K., Tokumura, A., Tsukatani, H., Shibuya, M. (1985).
Fluorescent pigments by covalent binding of lipid peroxidation by-products to
protein and amino acids. Lipids 20: 854-861
Goodwin, T. W. (1984). The biochemistry of the carotenoids, Vol. 2, Animals.
Chapman and Hall, London
Guilbault, G. G. (1973). Practical fluorescence: theory, methods and techniques. Marcel
Dekker, New York
Gutteridge, J. M. C., Heys, A. D., Lunec, 1. (1977). Fluorescent malondialdehyde
polymers from hydrolysed 1,1,3,3-tetramethoxypropane. Analytica chim. Acta
94: 209-211
Hammer, C. (1988). Accumulation of fluorescent age pigments in embryos and larvae of
pike, Esox lucius (Esocidae, Teleostei) at different incubation temperatures.
Envir. BioI. Fish. 22: 91-99
Hammer, C., Braum, E. (1988). Quantification of age pigments (lipofuscin). Compo
Biochem. Physiol. 90B: 7-17
Hammer, C., Karakiri, M. (1988). Disease dependent accumulation of fluorescent age
pigments in the common dab CLimandalimanda, Teleostei). In: Zs.-Nagy, I. (ed.)
Lipofuscin - 1987: state of the art. Akademiai Kiad6, Budapest and Elsevier
Science Publishers, Amsterdam, p. 445-446.
48
_..__.._-_._-_ .. -------_._•..- -.
Hammer, C., Madhusudhana-Rao, G. (1987). Individual and feed dependent
accumulation of fluorescent age pigments in five organs of tilapia (Oreochromis
aurea). Zeitschr. angew. Zool, 74: 275-292
Hamperl, H. (1934). Die Fluoreszenzmikroskopie menschlicher Gewebe. Virchows
Arch. path. Anat. Physiol, 292: 1-51
Hannover, A. (1842). Mikroskopische Undersolgeser af Nervensystemet. K. danske
Vidensk. Selsk. Skr. (Nat. Math. Afh.) 10: 1-112
Hardy, R., McGill, A. S., Gunstone, F. D. (1979). Lipid and autoxidative changes in
cold stored cod (Gadus morhua). 1. Sci. Fd Agric. 30: 999-1006
Herring, P. J. (1968a). The carotenoid pigments of Daphnia magma Strauss - I. The
pigments of animals fed Chiorella pyrenoidosa and pure carotenoids. Compo
Biochem. Physiol. 24: 187-203
Herring, P. J. (1968b). The carotenoid pigments of Daphnia magna Strauss - II. Aspects
of pigmentary metabolism. Compo Biochem. Physiol. 24: 205-221
Hill, K. T., Radtke, R. L. (1988). Gerontological studies of the damselfish, Dascyllus
albisella. Bull. mar. Sci. 42: 424-434
Hirche, H.-J., Anger, K. (1987). The accumulation of age pigments during larval
development of the spider crab, Hyas araneus (Decapoda, Majidae). Compo
Biochem. Physiol, 88B: 777-782
Hueck, W. (1912). Pigmentstudien. Beitr. path. Anat. 54: 68-232
Hyden, H., Lindstrom, B. (1950). Microspectrographic studies on the yellow pigment in
nerve cells. Discuss. Faraday Soc. 9: 436-441
Jones, N. R. (1959). "Browning" reactions and the loss of free amino acid and sugar
from lyophilized muscle extractives of fresh and chill-stored codling (Gadus
callarias). Fd Res. 24: 704-710
49
Katz, M. L. (1990). Incomplete proteolysis may contribute to lipofuscin accumulation in
the retinal pigment epithelium. In: Porta, E. A. (ed.) Lipofuscin and ceroid
pigments. Plenum Press, New York, p. 109-118
Katz, M. L., Robison, W. G., Herrmann, R. K., Groome, A. B., Bieri, J. G. (1984).
Lipofuscin accumulation resulting from senescence and vitamin E deficiency:
spectral properties and tissue distribution. Mechanisms Ageing Dev. 25: 149-159
Kikugawa, K., Beppu, M. (1987). Involvement of lipid oxidation products in the
formation of fluorescent and cross linked proteins. Cherny Phys. Lipids 44: 277
296
Kikugawa, K., Ido, Y. (1984). Studies on peroxidized lipids. V. Formation and
characterization of 1,4-dihydropyridine-3,5-dicarbaldehydes as model of
fluorescent components in lipofuscin. Lipids 19: 600-608
Kitani, K., Nokubo, M., Ohta, M., Zs.-Nagy, I. (1988). Characterization of peroxide
induced autofluorescence of hepatocyte plasma membrane proteins in relation to
age pigments in rats. In: Zs.-Nagy, I. (ed.) Lipofuscin - 1987: state of the art.
Akademiai Kiad6, Budapest and Elsevier Science Publishers, Amsterdam, p.
159-182
Lillie, R. D., Daft, F. S., Sebrell, W. H. (1941). Cirrhosis of the liver in rats on a
deficient diet and the effect of alcohol. Pub. Hlth Rep., Wash. 56: 1255-1258
Luo, S.-W., Hultin, H. O. (1986). Effect of age of winter flounder on some properties
of the sarcoplasmic reticulum. Mechanisms Ageing Dev. 35: 275-289
McArthur, M. C., Sohal, R. S. (1982). Relationship between metabolic rate, aging, lipid
peroxidation, and fluorescent age pigment in milkweed bug, Oncopeltus fasciatus
(Hemiptera). J. Geront. 37: 268-274
McDonald, R. E., Kelleher, S. D., Hultin, H. O. (1979). Membrane lipid oxidation in a
mucrosomal (sic) fraction of red hake muscle. J. Fd Biochem. 3: 125-134
50
Miquel, J., Lundgren, P. R, Johnson, J. E. (1978). Spectrophotofluorometric and
electron microscopic study of lipofuscin accumulation in the testis of aging mice.
J. Geront. 33: 5-19
Mullin, M. M., Brooks, E. R. (1988). Extractable lipofuscin in larval marine fish. Fish.
Bull. U.S. 86: 407-415
Nair, V., Cooper, C. S., Vietti, D. E., Turner, G. A. (1986). The chemistry of lipid
peroxidation metabolites: cross-linking reactions of malondialdehyde. Lipids 21:
6-10
Nicol, S. (1987). Some limitations on the use of the lipofuscin ageing technique. Mar.
BioI. 93: 609-614
Nokubo, M., Ohta, M., Kitani, K., Zs.-Nagy, I. (1989). Identification of protein-bound
riboflavin in rat hepatocyte plasma membrane as a source of autofluorescence.
Biochim. biophys. Acta 981: 303-308
Nokubo, M., Zs.-Nagy, I., Kitani, K., Ohta, M. (1988). Characterization of the
autofluorescence of rat liver plasma membranes. Biochim. biophys. Acta 939:
441-448
Porta, E. A., Mower, H. F., Moroye, M., Lee, Ch., Palumbo, N. E. (1988).
Differential features between lipofuscin (age pigment) and various experimentally
produced "ceroid pigments". In: Zs.-Nagy, I. (ed.) Lipofuscin - 1987: state of the
art. Akademiai Kiado, Budapest and Elsevier Science Publishers, Amsterdam, p.
341-374
Sawant, P. L., Magar, N. G. (1961). Studies on frozen fish. II. Some chemical changes
occurring during frozen storage. Fd Technol., Lond. 15: 347-350
Sheehy, M. R. J., Ettershank, G. (1989). Solvent extractable age pigment-like
autofluorescence and its relationship to growth and age in the water flea, Daphnia
carlnata. Aust. J. Zoo!. 36: 611-625
51
Shewfelt, R. L., Hultin, H. O. (1983). Inhibition of enzymic and non-enzymic lipid
peroxidation of flounder muscle sarcoplasmic reticulum by pretreatment with
phospholipase A2. Biochim. biophys. Acta 751: 432-438
Shimasaki, H., Veta, N., Privett, O. S. (1980). Isolation and analysis of age-related
fluorescent substances in rat testes. Lipids 15: 236-241
Shimasaki, H., Veta, N., Privett, O. S. (1982). Covalent binding of peroxidized linoleic
acid to protein and amino acids as models for lipofuscin formation. Lipids 17:
878-883
Siakotos, A N., Armstrong, D., Koppang, N., Muller, J. (1977). Biochemical
significance of age pigment in neurones. In: Nandy, K., Sherwin, I. (eds.) The
aging brain and senile dementia. Plenum Press, p. 99-118
Slabyj, B. M., Hultin, H. O. (1982). Lipid peroxidation by microsomal fractions isolated
from light and dark muscles of herring (Clupea harengus). J. Fd Sci. 47: 1395
1398
Sohal, R. S., Buchan, P. B. (1981). Relationship between fluorescent age pigment,
physiological age and physical activity in the housefly, Musca domestica.
Mechanisms Ageing Dev. 15: 243-249
Sohal, R. S., Ragland, S. S., Allen, R. G., Farmer, K. (1983). Correlation between
oxygen consumption, lipofuscin, fluorescent age pigment, lipid peroxidation
potential and life span in insects. In: Greenwald, R. A, Cohen, G. (eds.) Oxy
radicals and their scavenger systems. Vol. II: Cellular and medical aspects.
Elsevier Science Publishing Co., Inc., New York, p. 361-364
Tappel, A L. (1975). Lipid peroxidation and fluorescent molecular damage to
membranes. In: Trump, B. F., Arstila, A V. (eds.) Pathobiology of cell
membranes. Vol. I. Academic Press, New York, p. 145-170
52
Tappel, A. L., Fletcher, B., Deamer, D. (1973). Effect of antioxidants and nutrients on
lipid peroxidation fluorescent products and aging parameters in the mouse. J.
Geront. 28: 415-424
Thommen, H., Wackernagel, H. (1964). Zum Vorkommen von Keto-Carotinoiden in
Crustaceen. Naturwissenschaften 51: 87-88
Tokunaga, T. (1964). Studies on the development of dimethylamine and formaldehyde in
Alaska pollack muscle during frozen storage. Bull. Hokkaido reg. Fish. Res.
Lab. 29: 90-97
Trombly, R., Tappel, A. (1975). Fractionation and analysis of fluorescent products of
lipid peroxidation. Lipids 10: 441-447
Vernet, M., Hunter, 1. R., Vetter, R. D. (1988). Accumulation of age pigments
(lipofuscin) in two cold-water fishes. Fish. Bull. U.S. 86: 401-407
Zeller, P., Bader, E, Lindlar, H., Montavon, M., MUller, P., RUegg, R., Ryser, o.,
Sancy, a., Schaeren, S. F., Schwieter, U., Stricker, K., Tamm, R., ZUrcher,
P., Isler, O. (1959). Synthesen in der Carotinoid-Reihe. 13. Mitteilung. Synthese
von Canthaxanthin. Helv. chim. Acta 42: 841-847
53
-_.-.._-_ ..
Table2-1.Previous studiesof spectrofluorometrically measured fluorescent agepigmentaccumulation in aquatic animals. Tissues:b =brain;ceph.=cephalothorax; h =heart; 1=liver; m =muscle; s =spleen. Storage: FA, preservedin formalin; FNS, preservedin formalin then transferred to storagesolution; refrig.,refrigerated. Solvent: C, chloroform; M, methanol. Sonic.: ultrasonication.rt: assayedat room temperature. -: no information available basedon the described methods. Abbreviations for speciesnamesare:E.. superba =Euphausia £,., fa. americanus =Pseudopleuronectes a., Q.. aureus =Oreochromis .lL., H.. araneus =.lhas..iL" .c...hyperbQreus =Calanus. h., M.. norvegica =Meganyctiphanes n...., 1.illecebrosus =m I, E...~ =fu.Qx 1.., L...limanda =Limanda1, .Q.. albisella =Dascyllus ,ih, L.. Imlli£ =Leuresthes 1.u &. nobilis =Atractoscion n, fa. californicus =Paralichthys~,Q..mykiss=Oncorhynchus IIh, M.. pacificus=Microstomus p,.., .Q.. carinate=Daphnia £...
Species Tissue Age Tissue storage Extractincub. Assay Source
known method time temp. time Sonic. Solv. temp.
E. superba ceph. no 5%FNS 4wk/- WOC - yes C WOC Ettershank (1983)VI~
E. superba ceph. no 5%FA/S 4wk/- 2QoC - yes C WOC Ettershank (1984a)
E. superba ceph. no 5% FAIS 4wk/- WOC - yes C WOC Ettershank (l984b)
E. superba ceph. no 5%FA/S 4wk/- 2Q°C - yes C WOC Ettershank (1985)
P. americanus m no fresh - - - no C - Luo and Hultin(1986)
O. aureus b.h.l.s.m yes fresh - - - yes C+M rt Hammer& M-Rao (1987)
H. araneus larvae yes 4% FA - - - no C+M rt Hirche& Anger (1987)
C. hyperboreus body no frozen(?°C) - - - no C rt Nicol (1987)
M. norvesica body no frozen(?°C) - - - yes C rt Nicol (1987)
1 illecebrosus mantle no frozen(?°C) - - - no C rt Nicol (1987)
Table 2-1. (Continued) Previous studies of spectrofluorometrically measured fluorescent age pigment accumulation in aquaticanimals. Tissues: b =brain; ceph. = cephalothorax; h = heart; I =liver; m =muscle; s =spleen. Storage: FA, preserved informalin; FA/S, preserved in formalin then transferred to storage solution; refrig., refrigerated. Solvent: C, chloroform; M,methanol. Sonic.: ultrasonication. rt: assayed at room temperature. -: no information available based on the described methods.Abbreviations for species names are: E... superba =Euphausia s.., ~ americanus =Pseudopleuronectes iL., Q.. aureus =Oreochromis a, lL. araneus =~ a, !: hyperboreus =Calanus h.., M.. norve(:ica =Me(:anyctiphanes 11&, 1.illecebrosus =I!gi.., Ii~=~1,:b limanda =Limanda 1,Ik albisella =Dascyllus a.., L...~=Leuresthes L,~ nobilis =Atractoscionn..., .e californicus =Paralichthys £..,!1 mykiss =Oncorhynchus m..,~ pacificus =Microstomus lk Ik carl nata =Daphnia £...
Species Tissue Age Tissue storage Extract incub. Assay Source
known method time temp. time Sonic. Solv. temp.
klimanda b no - - - - yes C+M rt Hammer & Karakiri (1988)VIVI
D. albisella b no -WOC lOmo WOC 24h yes C rt Hill & Radtke (1988)
L. tenuis larvae yes -15 & -70°C - 4°C 3-4 h no C rt Mullin & Brooks (1988)
A. nobilis larvae yes -15 & -70°C - 4°C 3-4 h no C rt Mullin & Brooks (1988)
P. califomicus larvae yes -15 & -70°C - 4°C 3-4 h no C rt Mullin & Brooks (1988)
O. mykiss b.h.l yes -800C - refrig. - no C rt Vernet et al. (1988)
M. pacificus b no -800C - refrig. - no C rt Vernet et al, (1988)
E. superba ceph. no 5%FNS 4wk/- - - yes C rt Berman et al. (1989)
D. carinata body yes fresh - - - no C+M rt Sheehy & Ettershank (1989)
Fig. 2-1. Oreochromis mossambicus: Effect of storage temperature on relative
fluorescence intensity of skeletal muscle. (A) short-term storage of tissues at 23 and 3°e;
(B) long-term storage of tissues at -20 and -80°C. Uncorrected excitation and emission
maxima were 350 to 355 nm and 425 to 430 nm, respectively. Maxima did not shift as
intensity increased. Data presented are means ± standard deviations for triplicate
analyses.
56
-'- ......- ..._--~---_ ..- -....~----_...-., ...~.....
90A
80
70
60
50
40>.
30 ..-enc:Q) 20- 3°Cc:Q) 10 i/! !0c:Q) 0 •.j:.-. •0enQ) 0 1 2 3 4 5 6 7 8 9 10 11 12...0:::::l 90 Time (hours)u, SQ) 80 -20°C>- 70 ~!CI:lQ)
a: 60 .>'<I~ 50
40
~!30
20,/!10 -80°C
~/! !0 .~.-t-I !
-100 10 20 30 40 50 60 70 80 90 100 110 120
Time (days)
57
Fig. 2-2. Oreochromis mossambicus: Effect of incubation temperature on relative
fluorescence intensities of unfractionated chloroform/methanol-soluble extracts of brain,
heart, and muscle tissues. Extracts incubated at 23°C (A) and -20°C (B). Initial
uncorrected excitation and emission maxima were 355 and 430 nm, respectively, shifting
to 380 and 455 nm after 2 and 3 h of incubation at either temperature. Data presented are
means ± standard deviations for triplicate analyses.
58
--_._-_..---------------
873 4 5
Incubation T' 6rrne (hours)
21
o Brain
• Heart
• Muscle
A
4
2
0r---~-O)~-=;--~-
6
400
200
~ 300'enc(1)-c
59
Fig. 2-3. Oreochromis mossarnbicus: Effect of incubation temperature on relative
fluorescence intensities of final chloroform-soluble extracts of brain, heart, and muscle
tissues. Extracts incubated at 23°C (A) and -20°C (B). Uncorrected excitation and
emission maxima were 350 to 355 nm and 425 to 430 nm, respectively. Maxima did not
shift as intensity increased. Data presented are means ± standard deviations for triplicate
analyses.
60
_.._-_..._------_.- •.._---_ .._-- _.
8
-Q-r
765432
o Brain
• Heart
• Muscle
A
o
/1/ I
/1 ,A~·2 _
o~--_.,----I----I----I
120
>- 80-C/lc:Q)-c: 40Q)
oc:Q)
0oC/lQ)
""a.2 Bu,Q)
>-ellQ)
a:<l<fi
10
8
6
Incubation Time (hours)
61
Fig. 2-4. - Oreochromis mossambicus. Effects of ultrasonication at 25 W of output on
the expression of sample fluorescence. (A) change in relative fluorescence intensities of
sonified homogenates of brain, heart, and muscle, (B) uncorrected fluorescence spectra
of brain extracts before sonication (excitation/emission maxima =355/440 nm,
respectively) and after 5 min sonication (excitation/emission maxima =383/453 nm,
respectively). Data in (A) are means ± standard deviations for triplicate analyses.
62
353010 15 20 25Sonication Time (minutes)
5
o Brain
• Heart
A Muscle
A1000
>.-C/J 900c:2 800c:
Q) 700o:ii 600o~ 500...o::::l 400u,
Q) 300>-; 200Q)
a: 100<J~ 011II:;;,,--""---....1.---........----'-----1..__--"__---1
o
12B
550500450400350300250OL--.-.-:=...!....__...I..-__...L-__.......__....J-__--L.__-J
200
10
2
Q) 8oc:Q)oC/J 6Q)c:E::::l 4
..J
Excitation/Emission Wavelength (nm)
63
Fig. 2-5. - Oreochromis mossambicus. Uncorrected fluorescence spectra of
aqueous/methanol phase extracts of brain tissue. (A) excitation spectrum at emission =
455 om and emission spectrum at excitation =338 om; (B) excitation spectrum of same
sample at emission =520 nm. Excitation peaks at 286, 351, and 447 om indicate
contaminating flavin compounds.
64
1312 A 338 455
1 110
9- 8en- 7c:::l 6>.... 5~...
4-.c... 3~- 2Q)
o 1c:Q) 0oen 351Q) 7c: BE 6::l
5..J
43210250 300 350 400 450 500 550
Excitation/Emission Wavelength (nrn)
65
Fig. 2-6. - Oreochromis mossarnbicus. (A) Relationship between assay temperature and
luminescence of quinine sulfate standard (0.1 mg/11N H2S04) and chloroform extracts
from brain, and (B) resulting relative fluorescence intensity values from same sample
data.
66
Quinine Sulfa te Standard
12 18 24 30
( Oe )Temperature
6
0-0_0_0 Brain Extract-0-0-0_ _
o 0-0-0-0_0_0-0-
o
~~-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
4
2
~...L...-1....-.~.L.-....L..--.L....J-....L-...---'---0.0
5
2.0A-en-c: •::l
>. 1.5~
~~-.0~
~
1.0-Q)0c:Q)0en
0.5Q)c:
E::l..J
>.-"Ciic:Q)-c:
Q)oc:Q)oQ)en...o::l
IJ..Q)
>-~Q)
a:
67
-._-..~ .._,~--_.---~ ..
Fig. 2-7. - Qreochrornis mossambicus. Fluorescent age pigment (FAP) accumulation
and brain weight as functions of chronological age. (A) Total FAP content of an entire
brain; (B) weight-specific FAP (relative fluorescence intensity) in brain; (C) brain dry
weight.
68
....;,.; ....... .:...--._"""-_--.4...:....:...... _
A :::
'" B
/Co ~ !'
....•. ..........•. ..... ..'.....:::l:::o =.•
/,' .......•" ",,0 ::: I ... ·
't ~ - ~~""a :J.,••••
a~ ~·~..?-:~:~·i ••l~-~-~·~·='Ii ...
160
~ 140Q)(J
c 120Q)(Jenell 1000::Ju, 80c:ell
cD 60Q)
0 40.J:
~20
a14
.:. Immature
o Male
• Female
- Male 1" Immature
.......Female 1" Immature
:::
:J
:J
:J
>- 12'iiic:Q)
.= 10
8
6
•
c
Q)(Jeell(Jenell'-o::J
u:ell>(;;Q)
a:
0
40
35
OJ 30.5:c 25OJQ)
~ 20>-'-0 15eell
cD 10
5
OL....E!._~--''----'---'_--I._--I._--'-_.........- .......o 100 200 300 400 500 600 700 800 900
Age (days)
69
CHAPTER 3
FLUORESCENT AGE PIGMENT ACCUMULATION AND THE USE OFMULTIPLE REGRESSION MODELS FOR AGE ESTIMATION IN TWO
FISH SPECIES
Introduction
For the past decade, aquatic biologists have attempted to establish whether a
relationship exists between chronological age and the accumulationof soluble fluorescent
age pigments (pAP) extracted from postmitotic tissues, with the ultimate goal of applying
this method to age prediction in population studies. To date, results from FAP studies
have been published for crustacean (Ettershank, 1983, 1984a,b, 1985; Hirche & Anger,
1987; Nicol, 1987; Crossland et aI., 1988; Berman et aI., 1989; Sheehy & Ettershank,
1989; Nicol et al., 1991), mollusk (Nicol, 1987; Clarke et al., 1990), and fish (Luo &
Hultin, 1986; Hammer & Madhusudhana-Rao, 1987; Hammer, 1988; Hammer &
Karakiri, 1988; Hill & Radtke, 1988; Mullin & Brooks, 1988; Vemet et aI., 1988; Hill
&Womersley, 1991) species. Results from some of these preliminary studies have been
variable, leading to skepticism about the applicability ofFAP assays for age estimation;
however, variability of results from some ofthese early studies has been explained in part
by the application of methodological procedures leading to inaccuracies in the FAP assay.
Nicol's (1987) discovery that formalin fixation of samples artificially induced FAP-like
fluorescence in the tissue essentially invalidated the results of numerous studies
(Ettershank 1983, 1984a,b, 1985; Hirche & Anger, 1987; Hammer, 1988; Berman et aI.,
1989). Furthermore, Hill &Womersley (1991; see Chapter 2) demonstrated that other
factors related to sample storage, extraction techniqueand handling time may have
affected variability of the FAP assay in numerousother studies.
An additional problem that has led some investigators to question the FAP assay
has been the presumption that weight-specific concentrationsof FAP (commonly referred
70
to as "relative fluorescence intensity", or RFI) should show some positive correlation
with age. On the contrary, most studies have shown that RFI values are either negatively
correlated with age (Mullin & Brooks, 1988;Vernet et al., 1988; Crossland et al., 1988;
Sheehy & Ettershank, 1989; Clarke et al., 1990),or decrease initially, increasing later
with advancing age (Clarke et al., 1990; Hill & Womersley, 1991). Only one study
reports consistent age-related increases in weight-specificFAP concentrations in an
aquatic species (Luo & Hultin, 1986). Earlier reports describe such a trend in other
vertebrates (e.g. - Tappel et al., 1973; Shimasaki et aI., 1977; Miquel et al., 1978;
Shimasaki et al., 1980). As many investigators have pointed out, any relationship
between tissue FAP concentration (RFI) and age will be inherently confused as long as
the rate of FAP genesis differs from the rate of accumulation of tissue mass (Hammer&
Madhusudhana-Rao, 1987; Vemet et aI., 1988; Hill & Womersley, 1991; Sheehy &
Roberts, 1991). Thus, RFI values are only useful for comparing FAP from areas within
the same tissue, FAP in different tissues from the same organism, FAP between different
animals of identical age, or for examining trends in FAP accumulationin animals whose
somatic or organ growth has approached some asymptote.
Questions have also been raised as to the vaildity of the FAP technique originally
developed by Fletcher et al. (1973). This controversy has centered mainly on
discrepancies in the autofluorescenceproperties of histologicallyobserved age pigment
granules (lipofuscins and ceroids) having emission maxima in the 500 to 630 nm
wavelength range, and soluble age-relatedfluorophores (FAP) which emit light in the
blue (400-490 nm) range. Eldred et al. (1982) suggested that this problem was related to
differential sensitivityof fluorescence instrumentation which biased fluorescencereadings
in the blue-green range unless corrected; however, Eldred & Katz (1989) later
demonstrated that this difference remained even after instrumentalbiases were corrected.
Numerous attempts have been made to characterize age pigment fluorescenceby
71
reproducing crosslinking reactions involved in the pathology of aging, and all have
resulted in modelfluorophores withautofluorescence in the 400-490 om range (Chio&
Tappel, 1969; Malshetet al., 1974; Gutteridge et al., 1977; Kikugawa & Ido, 1984;
Fukuzawa et al., 1985; Kikugawaet al., 1989; Yin & Brunk, 1991b). Despite this
controversy, a numberof studies haveclearlydemonstrated that solubleFAP is presentin
the tissuesof aquatic animals and that it accumulates in a mannerthat correlates with
chronological age (Luo & Hultin, 1986; Mullin& Brooks, 1988; Vemet et al., 1988; Hill
& Womersley, 1991; Nicolet al., 1991). Subsequently, Yin & Brunk (1991a) provided
data to support all FAP studies,and thus the FAP assay, by demonstrating that model
FAP fluorophores undergo a metachromatic shiftin fluorescence emissionfrom the
orangeand yellow rangeat highconcentration to the blue range at dilutedconcentrations.
This 'inner-filter' effect goes a long way towardexplaining the differencesbetween
fluorescence spectral properties of intactage pigmentgranules that are dense, light
absorbing bodies (Katz et al., 1984; Docchioet al., 1991)and those that have been
isolated and solvent-extracted (Hendleyet al., 1963; Siakotos& Koppang, 1973).
The question of primary importance to mostfield biologists interested in the FAP
methodis whetherFAP content,which is a function of physiological agingprocesses,
can be used to predict chronological age. The presentstudy was designed to address this
issueby examining age-related patterns of FAP accumulation in laboratory-reared
populationsof a freshwater (Puntiusconchonius Hamilton) and a euryhaline
(Qreochromis mossambicus Peters) teleost The problemof age prediction was
approached by developing multiple linearregression models which include information
availablefrom somaticgrowth, otolithmorphometries, and FAP content, to establish
whetheradditional information available from theFAP methodcan improve the ability to
predictchronological age.
72
------- . "'--- ---- ..
Materials and methods
Specimenculture and preparation
Known-age specimens of O. mossarnbicus were reared for up to 886 d and sampled
under conditions described previously (Hill & Womersley, 1991). P. conchonius fry
were hatched at the Kewalo Basin ResearchFacilityof the NationalMarine Fisheries
Service, Honolulu, and reared for up to 538 d in cylindrical fiberglass tanks. Fish were
reared at room temperature (22 to 24°C), with an approximate 12 h photoperiod, bidaily
water changes, and fed ground trout chow twice daily. Specimens remaining after 538 d
were subsequently transferred to wet room facilities at the Department of Zoology,
Universityof Hawaii, and held under similar conditions until sacrifice at 1517 d.
Sampled P. conchonius individuals were transported live to the laboratory,
anesthetized by ice-bath immersion, weighed to the nearest 0.01 g, and standard length
measured to the nearest 0.05 mm using calipers. Specimens were labeled, wrapped in
aluminumfoil, frozen in liquid nitrogen, and either stored in liquid nitrogenor at -80°C
until dissection. Specimens were dissected while still in semi-frozen condition. Skull
tops were removed and the region surrounding the brain thoroughly rinsed with Ringer's
saline solution to remove blood and lipid deposits. Brain tissues removed included the
cerebral, inferior, optic, and cerebellar lobes with the medulla being severed posterior to
the tenth cranial nerve. Cranial nerves were severedat their point of attachment to the
brain. Hearts were excised and ventricle myocardial tissues separated and saved. Brain
and myocardial tissues were briefly rinsed in Ringer's solution, placed in Eppendorfvials
and immediately frozen in liquid nitrogenuntil analysis.
Sagittal otoliths from both species were removedfrom the otic cavity, cleaned of
extraneous tissue, rinsed in distilled water, soaked in 95% ethanol, and stored dry. O.
mossarnbicus otoliths were weighed to the nearest 0.001 mg using a Sartorius Model
4503 MP6 electronic microbalance, and otoliths from P. conchonius were weighed to the
73
nearest 0.0001 mg using a Cahn Model C-32 electronic microbalance. Otolith lengths
and widths for both species were measured to the nearest 0.01 mm using a stereo
dissecting microscope with an ocular micrometer.
FAP extraction and spectrofluorometric assay
Soluble FAP materials from brain and myocardial tissues were quantified following
the modified extraction and assay protocols outlined by Hill & Womersley (1991).
Tissues were lyophilized and their dry weight measured to the nearest 0.1 ug on a Cahn
Model C-32 electronic microbalance. Tissues were macerated at room temperature for 4
min in a Potter-Elvehjem glass tissue homogenizer using at least 1 ml of spectroscopic
grade chloroform (CHCI3) per 10 mg dry tissue. The initial meniscus was marked and
the volume reconstituted after homogenization. Homogenates were transferred to glass
centrifuge tubes and vortex-mixed for 2 min with an equal volume of a solution
containing 100 mM magnesium chloride in glass distilled water and spectroscopic grade
methanol at a ratio of 3:1. Samples were centrifuged at 0-5°C for 10 min at RCF =3000
x g. Nonpolar and polar fractions were immediately removed and assayed for FAP
content Sample extractions were performed singly and rapidly to prevent in vitro
production ofFAP-like fluorophores in the sample solvent (Hill & Womersley, 1991).
Extract solvents were placed in quartz-silicate microcuvettes with a 4 mm pathlength
and sealed with a teflon stopper to prevent evaporation. Temperature of the sample
solvent was maintained at 18°C using a thermostatically controlled turret microcell holder
connected to a Brinkmann Lauda Model RM6 refrigerated circulating water bath. A
Perkin Elmer LS-5B luminescence spectrometer, Acer computer, and Perkin Elmer CFS
PC-control software (version 3.0) were used for the fluorometric assay. The LS-5B
luminescence spectrometer was equipped with a Xenon light source and a Hamamatsu
Type R928 red-sensitive photomultiplier tube to reduce possible bias in the 500-700 nm
region (Eldred et al., 1982). Excitation and emission slits were set at 5 and 20 nm,
74
------------ .._----- ----_ ... -.
respectively. Fluorescence spectra reported herein were corrected for the spectral
emission characteristics of the source lamp, the spectral efficiencies of the excitation and
emission monochromators, optics, and the photomultiplier tube using the method of
Rhys-Williams et al. (1983) and the emission correction routine in the CFS PC-control
software. All sample readings were expressed as a percentage of Quinine sulfate (0.1
mg/11N sulfuric acid) standard luminescence at an excitation setting of 345 nm and an
emission setting of 450 om. Sample FAP concentrations were expressed as whole-organ
%fluorescence (%FL), where:
sample luminescence%FL = x solvent volume (ml) x 100
standard luminescence
Only brain tissues were available for Q. mossarobicus, and only their chloroform
soluble (nonpolar) FAP was quantified due to high levels of interfering flavin compounds
in the methanol/water (polar) fraction (Hill & Womersley, 1991).
Preliminary assays of P. conchonius brain and heart tissues revealed large quantities
ofFAP in the polar extract fraction. Excitation scans (250 to 500 om) at an emission
setting of 520 om (Fletcher et al., 1973) revealed low quantities of flavin compounds in
the polar fraction which, in most cases, contributed less than 5% of the total luminescence
at the FAP maxima settings. A simple algorithm was applied to polar fraction FAP data
to subtract the luminescence contributed by flavin compounds. Flavin luminescence
contribution at the FAP emission maxima was approximated and corrected by; 1)
assuming that the flavin luminescence spectrum is symmetrical in form, 2) measuring
luminescence at a point equidistant from the wavelength difference between the FAP
emission maximum and the flavin emission maximum (520 nm), and 3) subtraction of
this luminescence from the FAP raw luminescence value. For example, if the FAP
emission maximum was at 425 nm, then luminescence was also measured at 615 nm (=
520-425+520 nm) and subtracted from the FAP raw luminescence at 425 nm.
75
-'_ ......._.-----~---_ ...._---_ •.._...-... ... -"
Multiple regression analysis
Multiple regression models for fish, otolith, and FAP measurements were fitted to
the following form:
loglO (Age) =a + bj X; + b2X2 + b3X3 +...+ bnXn
where age (days) is known, a is the intercept, bi =regression coefficients, Xi =loglO
independent variables. Regressions were fitted in a stepwise manner with the inclusion
level for all variables set at £ =0.05. Independent variables included fish size [fish
weight (W) and standard length (SL)], otolith size [otolith weight (OWt), otolith length
(OL), and otolith width (OW)], brain and heart dry weight (BW and HW) and FAP
content (%FL). All variables were log transformed to meet the asswnptions of normality
and homogeneous variances. Four model types were developed in order to compare the
usefulness of information available from combinations of fish size, otolith size, and FAP
data. These models were developed for males and females separately, with data from
immature specimens included in both.
Results
Somatic and otolith growth
Somatic growth of O. mossambicus differed by sex, with males reaching a mean
standard length (SL) of 246.3 ± 15.3 mm and a mean weight (W) of 462.6 ± 101.9 g at
an age of 886 d (Fig. 3-1a,b). The largest male weighed 576.97 g with a SL of 264 mm.
Females reached a mean 185 ± 10.8 mm SL and 212.1 ± 44.3 g W by 886 d, with the
largest specimen being 253.85 g at a SL of 194 mm. Quadratic growth models
determined by regression and some associated statistics areprovided in Fig. 3-1. Growth
in SL was asymptotic in nature up to 886 d (Fig. 3-1a) for both males and females,
whereas the growth rate ofW continued to increase with age (Fig. 3-1b).
76
O. mossambicus males had larger otoliths than females (Fig. 3-lc). The best
quadratic models describing the relationship of otolith weight (OWt) to age for both males
and females are shown in Fig. 3-1c. The rate of growth in otolith weight did not decrease
significantly during this experiment. At 886 d, males had a mean OWt of 49.149 ± 1.299
mg and females had a mean OWt of 33.955 ± 3.661 mg.
Somatic growth in P. conchonius differed little between sexes for the first 538 d of
age (Fig. 3-2a,b). Logarithmic models described the relationship between SL and age
reasonably well up to 538 d. Growth was asymptotic in nature for both males and
females (Fig. 3-2a). Growth in W, also modeled using logarithmic equations (Fig. 3
2b), was most rapid between 60 and 300 d, leveling off between 300 and 538 d.
Individuals held until 1517 d continued to grow, with females attaining a mean W of
16.59 ± 3.56 g (max. =18.86 g) and males having mean W of 7.43 ± 1.93 g (max. =9.05 g). Otolith growth (OWt) in P. conchonius was well described by logarithmic
regressions for both males and females and did not vary greatly between the sexes (Fig.
3-2c).
Organ growth and FAP accumulation with age
Spectral characteristics of nonpolar FAP extracted from O. mossambicus brain
tissues have been previously reported (Hill & Womersley, 1991). Brain growth and
whole-brain FAP (%FL) has also been previously described (Hill & Womersley, 1991).
Regression models describing brain growth and nonpolar FAP accumulation (%FL) with
age are presented again for comparative purposes (Fig. 3-3). In brief, whole brain
nonpolar FAP (%FL) accumulated gradually during the first 370 d, after which it
accumulated more rapidly, with males (mean %FL =140.6 ±37.3) having almost twice
as much FAP as females (mean %FL of 74.3 ± 8.51) by 886 d (Fig. 3-3b).
Characteristics of corrected emission (EM) spectra of FAP extracted from P.
conchonius brain and heart tissues are presented in Fig. 3-4. Brain tissues had far greater
77
proportions of polar to nonpolar FAP than did heart tissues. Polar FAP extracted from
both tissue types had similar corrected emission maxima, which ranged from 415 to 420
nm when excited (EX) at 340 nm wavelength. There was no obvious evidence of
fluorescence from interfering compounds (e.g., retinols and flavins), however, excitation
scans of polar extracts at EM =520 revealed minor levels of flavin fluorescence, and
steps were taken to account for this contribution to the total fluorescence at the EX and
EM maxima of the FAP (see Methods section).
Brain growth and FAP accumulation with age in P. conchonius were expressed with
logarithmic models (Fig. 3-5). Brain growth was most rapid during the first year of life,
and was similar for males and females (Fig. 3-5a). At 1517 d, males had a mean brain
dry weight (BW) of 9.730 ± 1.259 mg, and females had a mean BW of 8.236 ± 2.050
mg.
Soluble brain FAP in P. conchonius was detected in both the nonpolar and polar
solvent fractions; levels were up to 100-fold greater in the polar fraction (Fig. 3-5b,c).
Nonpolar and polar FAP were detected in immature fish sampled at 60 d and gradually
increased in quantity with age in both sexes. The relationship of brain nonpolar FAP to
age, modeled up to 538 d by logarithmic regression, was variable for both males (r2 =
0.615, P < 0.0001) and females (r2 =0.682, P < 0.0001) at each age, and the rate of
accumulation declined with increasing age (Fig. 3-5b). The highest mean values of
nonpolar FAP were present in the oldest individuals sampled. Brain polar FAP was also
present in large quantity in brain tissues at 60 d, and they slowly increased with time.
Regression models were highly variable for both males (r2 =0.312, P < 0.0001) and
females (r2 =00405, P < 0.0(01).
Relationships of heart weight and FAP content to age in P. conchonius were
modeled using logarithmic regression (Fig. 3-6). Heart weight increased continuously
but became more variable with increasing age (Fig. 3-6a).
78
FAP extracted from the ventricle myocardium was also more soluble in polar than
nonpolar solvents (Fig. 3-6b,c), but the difference was not as great as was found for
brain tissues. Both nonpolar and polar FAP were detectable in heart tissues from the
youngest individuals (60d) which had an average ventricle dry weight of 0.147 mg.
Nonpolar heart FAP increased linearly with increasing age (Fig. 3-6b), but with
considerable variability in the data (male r2 =0.605, P < 0.0001; female r2 =0.517, P <
0.0001). Similar trends were apparent for polar heart FAP (Fig. 3-6c), which was
approximately four times as abundant as nonpolar FAP extracted from the same tissues.
Multiple regression analyses
Oreochromis mossambicus
Multiple linear regression models using various combinations of independent
variables from fish size, FAP, and otolith morphometric data provided strong predictions
of age for O. mossambicus (Table 3-1). Four model types were developed in order to
compare the relative usefulness of information available from fish size, FAP, and otolith
data. For the first type considered, data included fish size (SL and W) and otolith
morphometries; for the second type, fish size and organ weight and FAP data were
considered; for the third type, all variables except FAP were considered; for the fourth
type, all these independent variables were considered.
Stepwise regression analysis of fish size and otolith data (organ size and FAP data
excluded) demonstrated that fish weight and the three otolith measurements (OWt, OL,
and OW) were the most important variables in predicting age for both males and females
(Table 3-1). All of the correlations were highly significant, and the included variables
explained 97.99% of the variation in age for males and 97.20% for females, as measured
by the adjusted r2 values.
Stepwise regression analysis of fish size, organ size and FAP data (otolith data
excluded) demonstrated that fish weight, brain dry weight, and nonpolar FAP (%FL)
79
-------._---- --
were the best predictorsof age for bothsexes (Table3-1). Again,coefficients for each
variablewere highlysignificant, and coefficients of determination were only slightly
higher for both males (r2 =0.9820)andfemales (r2 =0.9830) than in the model that
excluded brain and FAP information. Thus, in the absenceof information on otolith size,
a model includingFAP contentprovided more information on age than did bodysize
alone.
When all independent variables exceptFAP wereconsidered, it was determined that
fish weight, brain weight, otolithweight and otolithlength were the best predictor
variables for males (r2 =0.9815), and that brain weight, otolith weightand otolith length
were best for females (r2 =0.9798). The fit of these models was slightly poorer than
those includingsomaticandFAP data,and slightlybetter than those including only
somaticand otolithdata.
When information from all independent variables was includedin the stepwise
multiple regression analysis, it was determined that brain weight, nonpolarFAP (%FL),
and otolith weightwere the best variables for predictingage of males,and that these same
variablesalong with otolith lengthwere bestfor ageprediction in females. Coefficients
of determination were marginally highest for these models, explaining 98.43%of the
variation in age for males and 98.60% for females.
Puntius conchonius
The four multipleregression model typesdevelopedfor P. conchonius were similar
to thosedescribed for O. mossambicus, with the additionof information on heart weight
as well as polar and nonpolarFAP (%FL) data from both brain and heart. As for O.
mossambicus, fish size, otolith morphometries, organ weight andFAP data all proved
useful to variousdegrees for the prediction of chronological age (Table3-2).
Through stepwiseregression analysis of fish size and otolithdata (minusorgan and
FAP data) it was determined that fish weight and otolithweightwere the most important
80
variables for age prediction in male P. conchonius, as were fish standard length, otolith
weight, and otolith width for females (Table 3-2). These variables explained 95.35% of
the variation of age for males, and 97.12% for females, and the coefficients of all the
variables were significant (Table 3-2).
Stepwise multiple regression of fish size, organ weight, and FAP data revealed that
brain weight and heart nonpolar and polar FAP (%FL) were the best predictors of age in
males (r2 =0.9406) and that fish weight, brain weight, and brain and heart nonpolar FAP
(%FL) were the best variables for females (r2 =0.9402) (Table 3-2). Coefficients of
determination were slightly less than for the fish size/otolith size model, however
information on FAP content strengthened the prediction of age over that of somatic
variables alone.
When all independent variables except FAP were considered, brain weight, heart
weight, and otolith weight were the best predictor variables for males (r2 =0.9604), and
fish standard length, otolith weight, and otolith width were best for females (r2 =
0.9712). These models provided equal or better age prediction than those including
somatic and otolith data or somatic and FAP data.
As was the case for O. mossambicus, the strongest models for age prediction in P.
conchonius were the ones including some information from each of the data types (Table
3-2). Stepwise regression analysis of all independent variables determined that brain
weight, heart nonpolar and polar FAP, and otolith weight were the most important
variables for age prediction in males (r2 =0.9633). For females, the best variables for
age prediction included fish standard length, heart polar FAP, otolith weight, and otolith
width (r2 =0.9725).
81
Discussion
Results from this study provideadditionalevidence that FAP accumulatesin brain
and heart tissues offish over time and, more importantly, that informationon soluble
FAP content can, at the very least, strengthen the prediction of chronologicalage. This
result is particularlyrelevant to researchers wishing to apply an age pigment technique,
eitherhistologicalor biochemical, as an alternativeor supplement to standardmethodsof
agedetermination. Patterns ofFAP accumulationreported in this researchclearly show
that rates of FAP genesis can be both species- and tissue-specific. NonpolarFAP
fractions extracted from brain tissue of Q. mossambicusand P. conchoniusincreased
with age, but the rate of FAP genesisapparently began to slowfairly early in life in P.
conchonius. The rate of accumulation was increasinglyrapid to the maximum age
sampledin Q. mossambicus. FAP accumulation differed in patternbetween brain and
heart tissues of£.. conchonius,with the rate of accumulationof heart FAP remaining
more constant with age. Within a given tissue type, considerablesimilaritywas evident
in the pattern of accumulationof nonpolar and polar FAP in £. conchonius,despite vast
differencesin the relative solubilities. Resultsof this nature emphasize the importanceof
examiningmore than one type of tissue when initiating studieson FAP accumulationfor
age prediction.
The decline in rate of accumulation ofFAP observed in brainsofP. conchonius with
age appears inconsistentwith the assumption that age pigmentsaccumulate in a linear
fashion throughout an animals' lifespan (Strehler et al., 1959); however, the assumption
of linearage pigment accumulation may not be entirelyvalid ifontogenetic changes in
metabolicrate are involved. While informationon such changes is presentlyunavailable
for P. conchonius, studies on other fish species have demonstratedage-relateddeclines in
weight-specific oxygen consumption, a trendrelated to the decreasedratio of area to
volume at greater age (De Silva et al., 1986; Oikawa et al., 1991). In one of the few
82
studies on FAP accumulation in aquatic species, Vemet et al. (1988) also reported a
leveling-off in the apparent rate offormation ofFAP in brain tissue of Dover sole
(Microstomus pacificus). Miquel et al. (1978) reported a similar decline in FAP extracted
from mouse testes.
A surprising result of the present study was the extreme ratio (ca. 100: 1) of polar to
nonpolar FAP extracted from P. conchonius brain. This is much higher than the ratios
reported in brain tissues of either Oreochromis aureus (Hammer & Madhusudhana Rao,
1987) or Q.. mossambicus (Hill & Womersley, 1991). Polar FAP was also higher in P.
conchonius heart extracts, but only by a ratio of ca. 4:1. We were only able to provide
these data for ~ conchonius because of relatively low levels of flavins in tissue samples.
Such was not the case for Q.. mossambicus, where high levels of contaminating flavin
compounds are present in brain tissue (Hill & Womersley, 1991). Clearly, flavin
contamination prevents useful measurement of polar FAP in some cases, but where it is
not masked by such interference, the polar fraction should not be ignored.
The extraction of age-related FAP which is more polar in solubility is not new (Desai
et aI., 1975; Taubold et al., 1976; Klass 1977; Davis et al., 1982), but the subject
involves some controversy. For example, Sheehy & Roberts (1991), working
exclusively with fluorescent compounds extracted from insect tissues, asserted that most
of the nonpolar fluorescence measured in other studies is in fact due to polar pteridine
contamination and not to the presence ofFAP. This does not appear to be the case for P.
conchonius, however, because the ratios of polar:nonpolar fluorescence were so vastly
different between tissue types. Ifnonpolar fluorescence were merely an artifact of polar
contaminants, consistent ratios between fraction quantities would be expected. Sheehy &
Roberts (1991) also assumed that all lipofuscin compounds are lipid in nature and that
FAP should therefore be soluble only in nonpolar solvents; however, this is not the case.
Detailed studies on the biochemical composition of isolated lipofuscin and ceroid granules
83
have revealed that the relative protein and lipid composition of these residues can vary
greatly, with protein fragments comprising up to 70% of the total mass (Porta, 1991).
Furthermore, most investigators using malondialdehyde (a breakdown product of lipid
peroxidation) to synthesize crosslinked products with FAP-like fluorescence, have done
so in polar solvents using amino acids and proteins as substrates (Chio & Tappel, 1969;
Shimasaki et al., 1982; Kikugawa & Ido, 1984). Thus, it is not surprising that much of
the FAP extracted from P. conchonius brain and heart, and the tissues of other animals,
(excluding insects) is polar in solubility.
The application of multiple linear regression analysis for chronological age prediction
is a recent development in fisheries biology (Boehlert 1985). This statistical methodology
appears to be effective for examining the ultility of FAP for age assessment Previous
studies have included variables obtained from measurement of otoliths (Boehlert, 1985)
or a combination of somatic and otolith data (Radtke et al., 1989; Radtke 1990; Beckman
et aI., 1991). Our multiple regression analyses show improvement in the prediction of
chronological age in both O. mossambicus and P. conchonius when FAP data are added.
This is the first time that such a utility has been demonstrated. FAP variables entered all
fitted models in the analysis when included in the initial variable list, however the fits (as
measured by r2) were never more than slightly better than models which included only
size, otolith, and organ weight variables.
The strongest model for Q.. mossambicus males and females was that which
incorporated all variables, with brain dry weight, brain nonpolar FAP (%FL), and otolith
weight (plus otolith length for females) being most important When otolith information
was arbitrarily excluded from male and female models, brain dry weight and FAP (%FL)
remained in the model with the addition of fish weight, to produce an age prediction
model with only slightly smaller coefficients of determination. The model with the
smallest coefficients of determination was the one that included only somatic and otolith
84
--. __.. _-----_ ..._-_ .. _..__ .- ..
data. In the absence of FAP information, fish weight and otolith weight and dimensions
were the most important age prediction variables, with fish standard length consistently
rejected from the model. Thus, for O. mossambicus, it would appear that FAP
information combined with fish and brain weight can provide a more reliable prediction of
age than otolith and somatic data alone, but that brain FAP and otolith weight data are the
best overall predictor variables.
The best age prediction models developed for f.. conchonius also included FAP data.
The most important variables for age prediction in males were heart nonpolar and polar
FAP, brain weight, and otolith weight Fish standard length, heart polar FAP, and
otolith weight and width were best for predicting female age. However. unlike O.
mossambicus, the model including only somatic and FAP data did not provide as strong a
prediction of age as the model including only somatic and otolith data. Nevertheless, the
model excluding otolith variables presented strong evidence that FAP information can
provide a more reliable prediction of age than somatic data alone. This was certainly the
case for P. conchonius males in which only brain weight, and heart nonpolar and polar
FAP data were included in the best predictive model.
From the above, it is clear that whole-organ FAP data can increase the accuracy of
age prediction in multivariate analyses of carefully standardized specimens. The use of
two teleosts species, with calcified structures traditionally used for aging, to demonstrate
the importance of FAP was an important aspect of the experimental design. This allowed
us to assess by comparison the importance of FAP in relation to calcified structures.
Thus, it was of particular interest when a combination of otolith and FAP information
provided the best predictive model. This held true even when the patterns ofFAP
accumulation were highly variable, as was the case for P. conchonius brains and hearts.
The retention of variables such as fish weight, otolith weight, organ weight, and
some FAP variables in the multiple regression models was probably due to their more
85
consistent patterns of increase with age as compared to some other variables. For
example, fish standard length was consistently excluded from multiple regression models
for O. mossambicus, whereas fish weight and otolith weight were, in most cases,
included in the models. Weight increases with age were consistent and fairly rapid for O.
mossambicus males and females, with sizes-at-age comparable to those reported by
Hodgkiss & Man (1977) for this species. The continuous pattern of growth in somatic
and otolith weight should not be surprising. O. mossambicus in this study were sampled
through less than half of their lifespan (Bruton & Allanson, 1974), and animals had only
reached 20% of their potential maximum weight (2953 g; Jubb, 1967).
Many animals of interest do not contain calcified skeletal structures suitable for
aging. Thus, it is encouraging that results of our models that excluded otolith data
demonstrated that FAP data can at least supplement, if not replace, somatic data for
predicting age. This analytical tool is especially pertinent to the problem of aging
commercially important fisheries species (e.g.- hagfish, lobster, shrimp, and squid) for
which tagging and size-based methods are currently the only alternatives.
With regard to field populations, further questions obviously must be addressed
before FAP assays or histological lipofuscin methods (Sheehy 1990b) can be used as
tools for age estimation. Age-related pigments (pAP and lipofuscin) are thought to form
as products of physiological aging processes related to metabolic rate. The rates of these
processes can vary greatly in poikilothermic animals, and little is known of the possible
effects of temperature or other environmental variables on the aging process and thus on
the rate of FAP genesis. Hammer (1988) tested the effect oftemperature on FAP
accumulation in pike (Esox lucius) larvae, but these results have been questioned due to
the use of formalin for specimen preservation. Sheehy's (1990a) histological study of
age pigment accumulation in crayfish <!'Jlerax cuspidatus) brain demonstrated a direct
effect of temperature on lipofuscin volume fraction in the olfactory lobe region. Studies
86
of temperature effects on "FAP" accumulation have been conducted on insects (Ragland
& Sohal, 1975; Sohal et al., 1981; McArthur & Sohal, 1982), but the fluorescence
extracted from insects is now suspected to be pteridine in origin (Lehane & Mail, 1985;
Sheehy & Roberts, 1991).
Application of FAP content as an aging tool for field populations will ultimately be
contingent on obtaining baseline information on age, growth, and accumulation patterns,
together with concurrent data on environmental and physiological variables that affect the
rate of FAP accumulation. The animals used for the present study were reared under
strictly controlled conditions in the laboratory. The resulting data represent a near
optimum which is probably unattainable with wild specimens. Sources ofFAP
variability must be determined and taken into account before multiple regression models
can be used for age prediction in wild stocks. A combination of mark-recapture and
laboratory growth studies will probably be required over relevant portions of the animal's
lifespan.
87
--- ------ . -_._------_ .
Literature cited
Beckman, D. W., A L. Stanley, J. H. Render & C. A Wilson, 1991. Age and growth
rate estimation of sheepshead Archosargus probatocephalus in Louisiana waters
using otoliths. Fish. Bull., U.£., Vol. 89, pp. 1-8.
Berman, M. S., A L. McVey & G. Ettershank, 1989. Age determination of Antarctic
krill using fluorescence and image analysis of size. Polar BioI., Vol. 9, pp. 267
271.
Boehlert, G. W., 1985. Using objective criteria and multiple regression models for age
determination in fishes. Fish. Bull., us, Vol. 83, pp. 103-117.
Bruton, M. N. & B. R. Allanson, 1974. The growth of Tila,pia mossambica Peters
(Pisces: Cichlidae) in Lake Sibaya, South Africa. I. Fish Biol., Vol. 6, pp. 701
715.
Chio, K. S. & A L. Tappel, 1969. Synthesis and characterization of the fluorescent
products derived from malondialdehyde and amino acids. Biochem., Vol. 8, pp.
2821-2827.
Clarke, A, M. A Kendall & D. J. Gore, 1990. The accumulation of fluorescent age
pigments in the trochid gastropod Monodonta lineata. I. Exp. Mar. BioI. Ecol.,
Vol. 144, pp. 185-204.
Crossland, C. 1., G. Denby, B. F. Phillips & R. Brown, 1987. The use of fluorescent
pigments (lipofuscin) for ageing western rock lobster (Panulirus cygnus) and
scampi (Metanephrops andamanicus): a preliminary assessment. CSIRO Mar.
Lab. Rep., No. 195, pp. 1-21.
Davis, B. 0., G. L. Anderson & D. B. Dusenbery, 1982. Total luminescence
spectroscopy of fluorescence changes during aging in Caenorhabditis elegans.
Biochem., Vol. 21, pp. 4089-4095.
88
Desai, 1. D., B. L. Fletcher & A. L. Tappel, 1975. Fluorescent pigments from uterus of
vitamin E-deficient rats. Lipids, Vol. 10, pp. 307-309.
De Silva, C. D., S. Premawansa & C. N. Keembiyahetty, 1986. Oxygen consumption in
Oreochromis niloticus (L.) in relation to development, salinity, temperature and
time of day. I. Fish BioI., Vol. 29, pp. 267-277.
Docchio, E, M. Boulton, R. Cebeddu, R. Ramponi & P. D. Barker, 1991. Age-related
changes in the fluorescence of melanin and lipofuscin granules of the retinal
pigment epithelium: a time-resolved fluorescence microscopy study. Photochem.
Photobiol., Vol. 54, pp. 247-253.
Eldred, G. E. & M. L. Katz, 1989. The autofluorescent products of lipid peroxidation
may not be lipofuscin-like. Free Rad. BioI. Med., Vol. 7, pp. 157-163.
Eldred, G. E., G. V. Miller, W. S. Stark & L. Feeney-Bums, 1982. Lipofuscin:
resolution of discrepant fluorescence data. Science, Vol. 216, pp. 757-759.
Ettershank, G., 1983. Age structure and cyclical annual size change in the Antarctic krill,
Euphausia superba. Polar BioI., Vol. 2, pp. 189-193.
Ettershank, G., 1984a. A new approach to the assessment of longevity in the Antarctic
krill Euphausia superba. I. Crust. BioI., Vol. 4, pp. 295-305.
Ettershank, G., 1984b. Methodology for age determination of Antarctic krill using the
age pigment lipofuscin. BIOMASS Handb., No. 26, pp. 1-14.
Ettershank, G., 1985. Population age structure in males and juveniles of the Antarctic
krill, Euphausia superba Dana. Polar BioI., Vol. 4, pp. 199-201.
Fletcher, B. L., C. J. Dillard & A. L. Tappel, 1973. Measurement of fluorescent lipid
peroxidation products in biological systems and tissues. Anal. Chern., Vol. 52,
pp. 1-9.
89
Fukuzawa, K, K. Kishikawa, A. Tokumura, H. Tsukatani & M. Shibuya, 1985.
Fluorescent pigments by covalent binding of lipid peroxidation by-products to
protein and amino acids. Lipids, Vol. 20, pp. 854-861.
Gutteridge, J. M. c., A. D. Heys, & J. Lunec, 1977. Fluorescent malondialdehyde
polymers from hydrolyzed 1,1,3,3,-tetramethoxypropoane. Analytica Chim.
Acta, Vol. 94, pp. 209-211.
Hammer, C., 1988. Accumulation of fluorescent age pigments in embryos and larvae of
pike, Esox lucius (Esocidae, Teleostei) at different incubation temperatures.
Environ. BioI. Fish., Vol. 22, pp. 91-99.
Hammer, C. & M. Karakiri, 1988. Disease dependent accumulation of fluorescent age
pigments in the common dab CLimandalimanda, Teleostei). In, Lipofuscin
.l281:~ of~ art, edited by I. Zs.-Nagy, Akademiai Kiad6, Budapest and
Elsevier Science Publishers, Amsterdam, pp. 445-446.
Hammer, C. & G. Madhusudhana-Rao, 1987. Individual and feed dependent
accumulation of fluorescent of fluorescent age pigments in five organs of tilapia
(Qreochromis~). Zeitschr. Angew. ZQQ!., Vol. 74, pp. 275-292.
Hendley, D. D., A. S. Mildvan, M. C. Reporter & B. L. Strehler, 1963. The properties
of isolated human cardiac age pigment I. Preparation and physical properties. I.
Gerontol., Vol. 18, pp. 144-150.
Hill, K T. & R. L. Radtke, 1988. Gerontological studies of the damselfish, Dascyllus
albisella. Bull. Mar. Sci., Vol. 42, pp. 424-434.
Hill, K T. & C. Womersley, 1991. Critical aspects of fluorescent age pigment
methodologies: modification for accurate analysis and age assessments in aquatic
organisms. Mar. BioI., Vol. 109, pp. 1-11.
90
Hirche, H.-J. & K Anger, 1987. The accumulation of age pigments during larval
development of the spider crab, Hyas araneus (Decapoda, Majidae). Compo
Biochem. Physiol., Vol. 88B, pp. 777-782.
Hodgkiss, J. I. & H. S. H. Man. 1977. Age composition, growth and body condition of
the introduced Sarotherodon mossambicus (Cichlidae) in Plover Cove Reservoir,
Hong Kong. Environ. BioI. Fish., Vol. 2, pp. 35-44.
Ikeda, T. & P. Dixon, 1982. Body shrinkage: a possible over-wintering strategy of the
Antarctic krill (Euphausia superba Dana). J:. Exp. Mar. BioI. Ecol., Vol. 62, pp.
143-151.
Ikeda, T. & P. G. Thomas, 1987. Longevity of the Antarctic krill (Euphausia superba
Dana) based on a laboratory experiment.Proc, Natl.!n.s.t. Polar Res. ~.,
Polar Bio!., Vol. 1, pp. 56-62.
Jubb, R. A., 1967. Freshwater Fishes QfSouthern Africa. Cape Town: A. A. Balkema,
Rotterdam.
Katz, M. L., W. G. Robison Jr., R. K Herrmann, A. B. Groome, & J. G. Bieri, 1984.
Lipofuscin accumulation resulting from senescence and vitamin E deficiency:
spectral properties and tissue distribution. Mech. Age. Dev., Vol. 25, pp. 149
159.
Kikugawa, K. & Y. Ido, 1984. Studies on peroxidized lipids. V. Formation and
characterization of 1,4-dihydropyridine-3,5-dicarbaldehydes as model of
fluorescent components in lipofuscin. Lipids 19, pp. 600-608.
Kikugawa, K, T. Kato, M. Beppu & A. Hayasaka, 1989. Fluorescent and cross-linked
proteins formed by free radical and aldehyde species generated during lipid
oxidation. In, Lipofuscin m:ill~ pigments, edited by E. A. Porta, Plenum
Press, New York, pp. 345-357.
91
Klass, M. R., 1977. Aging in the nematode Caenorhabditis elegans: major biological and
elemental factors influencing lifespan. Mech.~. Dev., Vol. 6, pp. 413-429.
Lehane, M. 1. & T. S. Mail, 1985. Determining the age of adult male and female
Glossina morsitans morsitans using a new technique. Ecol. Entomol., Vol. 10,
pp. 219-224.
Luo, S.-W. & H. O. Hultin, 1986. Effect of age of winter flounder on some properties
of the sarcoplasmic reticulum. Mech. Age. Dev., Vol. 35, pp. 275-289.
Malshet, V. G., A. L. Tappel & V. M. Burns, 1974. Fluorescent products of lipid
peroxidation: II. Methods for analysis and characterization. Lipids 9, pp. 328
332.
McArthur, M. C. & R. S. Sohal, 1982. Relationship between metabolic rate, aging, lipid
peroxidation, and fluorescent age pigment in milkweed bug, Oncopeltus fasciatus
(Hemiptera). 1.. Gerontol., Vol. 37, pp. 268-274.
Miquel, J., D. R. Lundgren & J. Johnson, 1978. Spectrophotofluorometric and electron
microscopic study of lipofuscin accumulation in the testis of aging mice. ,I.
Gerontol., Vol. 33, pp. 5-19.
Mullin, M. M. & E. R. Brooks, 1988. Extractable lipofuscin in larval marine fish. Fish.
Bull., us, Vol. 86, pp. 407-415.
Nicol, S., 1987. Some limitations on the use of the lipofuscin ageing technique. Mar.
BioI., Vol. 93, pp. 609-614.
Nicol, S., M. Stolp & G. W. Hosie, 1991. Accumulation of fluorescent age pigments in
a laboratory population of Antarctic krill Euphausia superba Dana. J:. Exp. Mar.
Biol. Ecol., Vol. 146, pp. 153-161.
Oikawa, S., Y. Itazawa & M. Gotoh, 1991. Ontogenetic change in the relationship
between metabolic rate and body weight in a sea bream~ major (Temminck
.& Schlegel). 1. Fish BioI., Vol. 38, pp. 483-496.
92
Porta, E.A., 1991. Advances in age pigment research. Arch. Gerontol. Geriatr., Vol. 12,
pp. 303-320.
Radtke, R. L., 1990. Age determination of the Antarctic fishes Champsocephalus gunnari
and Notothenia rossii marmorata from South Georgia. Polar BioI., Vol. 10, pp.
321-327.
Radtke, R. L., T. E. Targett, A. Kellermann, J. L. Bell & K. T. Hill, 1989. Antarctic
fish growth: profile of Trematomus newnesi. Mm:. Ecol. frQg. ~., Vol. 57, pp.
103-117.
Ragland, S. S. & R. S. Sohal, 1975. Ambient temperature, physical activity and aging in
the housefly, Musca domestica. Exp. Gerontol., Vol. 10, pp. 279-289.
Rhys-Williams, A. T., S. A. Winfield & 1. N. Miller, 1983. Relative fluorescence
quantum yields using a computer-controlled luminescence spectrometer. Analyst,
Vol. 108, pp. 1067-1071.
Sheehy, M. R. J., 1990a. Individual variation in, and the effect of rearing temperature
and body size on, the concentration of fluorescent morphological lipofuscin in the
brains of freshwater crayfish, Cherax cuspidatus (Crustacea: Parastacidae).
Qmm. Biochem. Physiol., Vol. 96A, pp. 281-286.
Sheehy, M. R. J., 1990b. The potential of morphological lipofuscin age pigment as an
index of crustacean age. Mar. BioI., Vol. 107, pp. 439-442.
Sheehy, M. R. J. & G. Ettershank, 1989. Extractable age pigment-like autofluorescence
and its relationship to growth and age in the water-flea Daphnia carinata King
(Crustacea: Cladocera). Aust. I. Zool., Vol. 36, pp. 611-625.
Sheehy, M. R. 1. & B. E. Roberts, 1991. An alternative explanation for anomalies in
"soluble lipofuscin" fluorescence data from insects, crustaceans, and other aquatic
species. Exp. Gerontol., Vol. 26, pp. 495-509.
93
Shimasaki, H., T. Nozawa, O. S. Privett & W. R. Anderson, 1977. Detection of age
related fluorescent substances in rat tissues. Arch. Biochem. Biophys., Vol. 183,
pp. 443-451.
Shimasaki, H., N. Ueta & O. S. Privett, 1980. Isolation and analysis of age-related
fluorescent substances in rat testes. Lipids, Vol. 15, pp. 236-241.
Shimasaki, H., N. Ueta & O. S. Privett, 1982. Covalent binding of peroxidized linoleic
acid to protein and amino acids as models for lipofuscin formation. Lipids, Vol.
17, pp. 878-883.
Siakotos, AN. & N. Koppang, 1973. Procedures for the isolation of lipopigments from
brain, heart and liver, and their properties: A review. Mech. Age. Dev., Vol. 2,
pp. 177-200.
Sohal, R. S., H. Donato & E. R. Biehl, 1981. Effect of age and metabolic rate on lipid
peroxidation in the housefly, Musca domestica L. Mech. ~. Dev., Vol. 16, pp.
159-167.
Strehler, B. L., D. D. Mark, AS. Mildvan & M. V. Gee, 1959. Rate and magnitude of
age pigment accumulation in the human myocardium. I. Gerontol., Vol. 14, pp.
430-439.
Tappel, A, B. Fletcher, & D. Deamer, 1973. Effect of antioxidants and nutrients on lipid
peroxidation fluorescent products and aging parameters in the mouse. 1
Gerontol., Vol. 28, pp. 415-424.
Taubold, R. D., A. N. Siakotos & E. G. Perkins, 1974. Studies on chemical nature of
lipofuscin (age pigment) isolated from normal human brain. Lipids, Vol. 10, pp.
383-390.
Vernet, M., J. R. Hunter & R. D. Vetter, 1988. Accumulation of age pigments
(lipofuscin) in two cold-water fishes. Fish. BulL, us, Vol. 86, pp. 401-407.
94
Yin, D. & U. T. Brunk, 1991a. Microfluorometric and fluorometric lipofuscin spectral
discrepancies: a concentration-dependent metachromatic effect? Mech. Age. Dev.,
Vol. 59, pp. 95-109.
Yin, D. & U. T. Brunk, 1991b. Oxidized ascorbic acid and reaction products between
ascorbic acid and amino acids might constitute part of age pigments. Mech. ~.
Dev., Vol. 61, pp. 99-112.
95
._-_.._-_ _.__._ _-_ .
Table 3-1. - Coefficients and associated statistics from multiple regressions ofmultivariate linear models of age (logro) for Oreochromis mossambicus. Models werefitted in a stepwise manner using combinations of the independent variables including fishsize and otolith data, fish size, organ size, FAP data, and all variables combined. Datafrom immature specimens were included in both male and female analyses. All variableswere loglO transformed for analyses. Intercepts, coefficients and their standard errors(SE) are for log transformed data.
Independent ModelVariable Coefficient SE P Adj. r2
Males (n =92)
(fish size model) 0.9740Intercept (a) =1.729Fish weight 0.413 0.007 <0.0001
(fish size and otolith model) 0.9799Intercept (a) =2.153Fish weight 0.391 0.084 <0.0001Otolith weight 0.911 0.194 <0.0001Otolith length -1.165 0.282 <0.0001Otolith width -1.179 0.381 0.0027
(fish size and FAP model) 0.9820Intercept (a) =1.696Fish weight 0.425 0.023 <0.0001Brain dry weight -0.104 0.031 0.0012Brain nonpolar FAP 0.111 0.028 <0.0001
(all variables w/o FAP) 0.9815Intercept (a) =2.062Fish weight 0.262 0.080 0.0014Brain dry weight -0.148 0.035 <0.0001Otolith weight 0.591 0.156 0.0003Otolith length -0.745 0.291 0.0122
(all variables model) 0.9843Intercept (a) =1.886Brain dry weight -0.153 0.031 <0.0001Brain nonpolar FAP 0.147 0.025 <0.0001Otolith weight 0.566 0.028 <0.0001
96
Table 3-1. (Continued) - Coefficients and associated statistics from multiple regressionsof multivariate linear models of age (loglO) for Oreochromis mossambicus. Models werefitted in a stepwise manner using combinations of the independent variables including fishsize and otolith data, fish size, organ size, FAP data, and all variables combined. Datafrom immature specimens were included in both male and female analyses. All variableswere loglO transformed for analyses. Intercepts, coefficients and their standard errors(SE) are for log transformed data.
Independent ModelVariable Coefficient SE P Adj. r2
Females (n =90)
(fish size model) 0.9640Intercept (a) =1.724Fish weight 0.441 0.009 <0.0001
(fish size and otolith model) 0.9720Intercept (a) =2.069Fish weight 0.392 0.121 0.0017Otolith weight 0.929 0.214 <0.0001Otolith length -0.455 0.148 0.0028Otolith width -1.937 0.512 0.0003
(fish size and FAP model) 0.9830Intercept (a) =1.680Fish weight 0.492 0.022 <0.0001Brain dry weight -0.188 0.030 <0.0001Brain nonpolar FAP 0.168 0.030 <0.0001
(all variables wloFAP) 0.9798Intercept (a) =2.158Brain dry weight -0.259 0.034 <0.0001Otolith weight 0.882 0.049 <0.0001Otolith length -0.374 0.125 0.0037
(all variables model) 0.9860Intercept (a) =1.985Brain dry weight -0.213 0.029 <0.0001Brain nonpolar FAP 0.172 0.026 <0.0001Otolith weight 0.759 0.045 <0.0001Otolith length -0.313 0.103 0.0032
97
Table 3-2. - Coefficients and associated statistics from multiple regressions ofmultivariate linear models of age (loglO) for Puntius conchonius. Models were fitted in astepwise manner using combinations of the independent variables including fish size andotolith data, fish size, organ size, FAP data, and all variables combined. Data fromimmature specimens were included in both male and female analyses. Animals sampledat 1517 d were excluded from the analysis. All variables were loglO transformed foranalyses. Intercepts, coefficients and their standard errors (SE) are for log transformeddata.
Independent ModelVariable Coefficient SE P Adj. r2
Males (n = 90)
(fish size model) 0.8980Intercept (a) =2.213Fish weight 0.774 0.028 <0.0001
(fish size and otolith model) 0.9535Intercept (a) =2.870Fish weight 0.143 0.064 0.0280Otolith weight 0.772 0.075 <0.0001
(fish size and FAP model) 0.9406Intercept (a) =1.776Brain dry weight 0.718 0.064 <0.0001Heart nonpolar FAP 0.201 0.060 0.0013Heart polar FAP 0.152 0.046 0.0014
(all variables w/o FAP) 0.9604Intercept (a) =2.672Brain dry weight 0.246 0.085 0.0047Heart weight 0.151 0.066 0.0251Otolith weight 0.580 0.079 <0.0001
(all variables model) 0.9633Intercept (a) =2.500Brain dry weight 0.246 0.082 0.0035Heart nonpolar FAP 0.136 0.048 0.0060Heart polar FAP 0.079 0.038 0.0381Otolith weight 0.554 0.075 <0.0001
98
--_._-- ...._-_ .._._- ....
Table 3-2. (Continued) - Coefficients and associated statistics from multiple regressionsof multivariate linear models of age (loglO) for Puntius conchonius. Models were fittedin a stepwise manner using combinations of the independent variables including fish sizeand otolith data, fish size, organ size, FAP data, and all variables combined. Data fromimmature specimens were included in both male and female analyses. Animals sampledat 1517 d were excluded from the analysis. All variables were loglO transformed foranalyses. Intercepts, coefficients and their standard errors (S£) are for log transformeddata.
Independent ModelVariable Coefficient SE P Adj. r2
Females (n =89)
(fish size model) 0.9040Intercept (a) =1.100Fish weight 0.034 0.001 <0.0001
(fish size and otolith model) 0.9712Intercept (a) =3.774Standard length -0.464 0.195 0.0193Otolith weight 0.748 0.094 <0.0001Otolith width 1.048 0.242 <0.0001
(fish size and FAP model) 0.9402Intercept (a) =1.925Fish weight 0.211 0.092 0.0242Brain dry weight 0.515 0.119 <0.0001Brain nonpolar FAP 0.144 0.066 0.0311Heart nonpolar FAP 0.137 0.057 0.0191
(all variables w/o FAP) 0.9712Intercept (a) =3.774Standard length -0.464 0.195 0.0193Otolith weight 0.748 0.094 <0.0001Otolith width 1.048 0.242 <0.0001
(all variables model) 0.9725Intercept (a) =3.730Standard length -0.493 0.190 0.0114Heart polar FAP 0.082 0.036 0.0252Otolith weight 0.722 0.092 <0.0001Otolith width 1.059 0.236 <0.0001
99
-------._..-- ...
Figure 3-1. - Oreochromis mossambicus. Body size and otolith weight as functions of
chronological age (Days). Quadratic growth regressions for immature + males (n =101)
and immature + females (n =99) all have P < 0.0001. (A) Standard length (SL) as a
function of age for immature + males: SL =3.3813 + 0.51878(Days) - 2.8061*10
4(Days)2, r2 = 0.976; and for immature + females: SL = 8.0858 + 0.45049(Days)
2.8413*10-4(Days)2, r2 =0.983. (B) Fish wet weight (W) as a function of age for
immature + males: W =-15.692 + 0.21892(Days) + 3.8805*10-4(Days)2, r2 = 0.929;
and for immature + females: W =-17.042 + 0.26895(Days) + 3.2033* 10-5(Days)2, r2 =
0.940. (C) Otolith weight (OWt) as a function of age for immature + males: OWt =-2.5115 + 4.997*1O-2(Days) + 9.6809*1O-6(Days)2, r2 =0.950; and for immature +
females: OWt =-2.1908 + 4.7062*1O-2(Days) - 5.3785*1O-6(Days)2, r2 =0.973.
100
300 A•- 250 .sE
E--.c 200
o ~-0>c: 150J!2"Eca 100"Cc )( Immatureca- 50 • MaleCJ)
o Female
0
600 B •500 d'
-0> 400---.c0>
300'CD;: ~.c 200eniI
100
0
55 C if50
- 450>E 40-- 35-..c:C) 30
'CD25;:
..c: 20
.'!:0 15-0 10
5
00 100 200 300 400 500 600 700 800 900
Age (days)
101
Figure 3-2. - Puntius conchonius. Body size and otolith weight as functions of
chronological age (Days). Logarithmic growth regressions for immature + males (n =
90) and immature + females (n =89) all have P < 0.0001. Animals sampled at 1517 d
were excluded from the regression analyses. (A) Standard length (SL) as a function of
age for immature + males: SL = -23.418 + 25.536 * loglO(Days), r2 = 0.905; and for
immature + females: SL =-26.240 + 26.997 * loglQ(Days), r2 =0.906. (B) Fish wet
weight (W) as a function of age for immature + males: W =-5.6956 + 3.3363 *loglQ(Days), r2 =0.785; and for immature + females: W = -7.0110 + 4.0120 *loglQ(Days), r2 =0.828. (C) Otolith weight (OWt) as a function of age for immature +
males: OWt =-0.70798 + 0.41255 * loglQ(Days), r2 =0.898; and for immature +
females: OWt =-0.75479 + 0.43729 >Ie loglO(Days), r2 =0.908.
102
70 A0- 60E 0
<i •E 0- 50..c •-C> ifc 40~
"EtiS 30
"'Cc:: x ImmaturetiS- 20 • Maleen
o Female
10 ~
20 B18 8
16- 14C>-- 120
s:C>
10'~ •
8 •..cen 6 9u: • •
4 a2 i 'if0 ~
1.2 C •0- 1.0
C>E- 0.8 •-..cC>
'0) 0.6 •~0
•..c 0.40 0
,~
'0 I •-0 0.2
0.0 ~0 100 200 300 400 500 1517
Age (days)
103
Figure 3-3. - Oreochromis mossambicus. Brain weight and fluorescent age pigment
(FAP) level as functions of chronological age (Days). Logarithmic and quadratic
regressions for immature + males (n = 101) and immature + females (n = 99) all have P <
0.0001. (A) Brain dry weight (BW) as a function of age for immature + males: BW =
-36.488 + 23.333 * logIQ(Days), r2 =0.943; and for immature + females: BW =-32.101
+ 21.067 * logIQ(Days), r2 =0.914. (B) Brain nonpolar FAP (%FL) as a function of
age for immature + males: %FL =7.4078 - 4.5872*1O-2(Days) + 2.3313*1O-4(Days)2,
r2 = 0.926; and for immature + females: %FL = 6.3477 - 1.4104*1O-2(Days) +
1.0736*1O-4(Days)2, r2 =0.936.
104
-_._----_._-----_._----_._-- .
40 A •
- 35 6'0> •E 30 ~-..... e.!: 25 0.0> 0
Q)203:
~ 15-cc:: 10·co x Immature~
CO 5 - Maleo Female
0
180~ B- -6'-J 160u..~ 1400-Q. 120 •«u.. 100 -~
(\180 ~0c.
c:: 600c:: 40c::
"(\1 20~
CO0
0 100 200 300 400 500 600 700 800 900Age (days)
105
-----------------------
Figure 3-4. - Puntius conchonius. Corrected emission spectra for soluble fluorescent age
pigments (FAP) extracted from brain and heart tissues. (A) Brain polar and nonpolar
FAP corrected emission maxima were 415 and 420 nm when excited at 340 and 345 nm,
respectively. (B) Heart polar and nonpolar FAP corrected emission maxima were 415
and 425 nm when excited at 340 and 345 nm, respectively.
106
5.5 A5.0
4.5
4.0
3.5 Brain FAP
3.0-en 2.5.1::C::J 2.0
e- 1.5~"- 1.0+-' nonpolar fraction:c
0.5 ~"-~- 0.0(1)CJC 0.35 8(1)CJen(1) 0.30C
E Heart FAP::J
0.25-.J
0.20
0.15
0.10
0.05
0.00350 400 450 500 550
Emission wavelength (nm)
107
Figure 3-5. - Puntius conchonius. Brain weight and fluorescent age pigment (pAP) level
as functions of chronological age (Days). Logarithmic regressions for immature + males
(n = 90) and immature + females (n = 89) all have P < 0.0001. Animals sampled at 1517
d were excluded from the regression analyses. (A) Brain dry weight (BW) as a function
of age for immature + males: BW = -10.337 + 6.2202 * loglO(Days), r2 = 0.919; and for
immature + females: BW = -9.3086 + 5.6154 * loglO(Days), r2 = 0.926. (B) Brain
nonpolar FAP (%FL) as a function of age for immature + males: %FL =-4.5132 +
3.8772 * loglO(Days), r2 =0.615; and for immature + females: %FL =-5.6471 +
4.3583 * logIQ(Days), r2 = 0.682. (C) Brain-polar FAP (%FL) as a function of age for
immature + males: %FL = -234.80 + 200.43 * loglO(Days), r2 =0.312; and for
immature + females: %FL =-348.42 + 258.05 * logIQ(Days), r2 =00405.
108
---_ .._---.
12 A •- 100
C)
E •-- •.... 8 0
.l::C)
"056
0
:=~"0 4c:.~
2x Immature
CD • Male
o Female
0 ~
14 B--I 0
U- 12~ 00--a.. 10<C 0 0
U- 8 I S?•
I •~ 0 88as0
0 •6a. I • d'c • I la0 4 aO
c:c
"as 2~
CO0 ~
1200 C •--I 1000U-~0-- 800a.. 0<C •U- 600 •~
0 0as • I00
S? a0 400 x I I " s· 0
a. • 8 Ic: 0
I a. d'"as 200 i e I ~ §~
g •CO a 8 • a •~
100 200 300 400 500 1517
Age (days)
109
Figure 3-6. - Puntius conchonius. Heart weight and fluorescent age pigment (pAP) level
as functions of chronological age (Days). Linear and logarithmic regressions for
immature + males (n = 90) and immature + females (n = 89) all have P < 0.0001.
Animals sampled at 1517 d were excluded from the regression analyses. (A) Heart dry
weight (HW) as a function of age for immature + males: HW =-1.6957 + 1.0180 *
loglO(Days), r2 =0.574; and for immature + females: HW =-1.2374 + 0.77254 *
loglO(Days), r2 =0.566. (B) Heart nonpolar FAP (%FL) as a function of age for
immature + males: %FL = -1.0720*10-2 + 9.2545*10-3 (Days), r2 = 0.605; and for
immature + females: %FL = 0.16355 + 7.8694*10-3 (Days), r2 = 0.517. (C) Heart
polar FAP (%FL) as a function of age for immature + males: %FL =4.0984 +
1.9406*10-2 (Days), r2 =0.313; and for immature + females: %FL =4.1528 +
1.6479*10-2 (Days), r2 =0.368.
110
.....;.; .• l-._... _....:...~_. .•• ~. . •. _.-.....-....0.-" _ .•~ .. ".+.
5 A 0
- )( Immature0> 4E • Male- o Female •- •..c !0> 3
"Q53: 0
e- 2 •"'C I if1:: 0 IIca •Q)
I0
900
0 ~
14 B-....J 00
LL. 12~ I •0- 0
a.. 10« •LL. 8 0 0'-
ifca 0
"0 6 •C.c:0 4c:1::ca 2Q)
I0
60 C •-....J 50LL. 0~0- 40a..« •LL. •30 •'- • .~as
! 0
"0 20 • ifc. 0 0 Ii 80
• I 0t:: " 0 •as10 I e 9Q) I ••I
0 ~
0 100 200 300 400 500 1517Age (days)
111
-_.__.._---_._--_....._-_. _..__ .- ..
CHAPTER 4
INTERACTIVE EFFECTS OF SOME ENVIRONMENTAL ANDPHYSIOLOGICAL VARIABLES ON FLUORESCENT AGE PIGMENT
ACCUMULATION IN BRAIN AND HEART TISSUES OF AN AQUATICPOIKILOTHERM
Introduction
The accumulation of age pigment complexes related to physiological aging is
considered one of the most consistent events in all senescing, nonmitotic cells (see
reviews by Sohal and Wolfe 1986, Tsuchida et al. 1987, Hammer and Braum 1988,
Porta 1991). Age pigments have been variously classified as pre-ceroid, ceroid,
lipofuscin, and fluorescent age pigment, depending upon their physical and chemical
properties determined by either histologic or biochemical methods. The age-related
accumulation of these pigments has been of interest to gerontologists who hope to define
biochemical aspects of aging and ultimately slow or arrest this process. More recently,
age pigment studies have been initiated by field biologists wishing to develop predictive
models, based on pigment content, for the estimation of chronological age for population
studies (Ettershank 1983, 1984 a.b, 1985; Brown and George 1985, Hirche and Anger
1987, Hill and Radtke 1988, Hammer 1988, Mullin and Brooks 1988, Vernet et al.
1988, Berman et al. 1989, Clarke et aI. 1990, Sheehy 1990a, Hill and Womersley 1991,
Nicol et al. 1991, Hill and Womersley, in review).
The majority of these field studies have examined soluble fluorescent age
pigments (FAP) using the extraction methods and spectrofluorometric assay described by
Fletcher et al. (1973). This method of age pigment quantification has recently been
criticized because of discrepancies between fluorescent emission properties of solvent-
extracted pigments (blue emission at 400 to 490 nm; Siakotos and Koppang 1973, Tappel
1975) and those observed in situ using fluorescence microscopy (green to yellow
112
emission at 500 to 630 nm; Eldredet al. 1982,Eldred and Katz 1989). This issue has
been partly resolved by Yin and Brunk(1991),who provided strong evidence for
concentration-dependent metachromatic shifts in fluorophore emission from yellow at
high density to blue under dilutedconditions. In addition to this controversy, Nicol
(1987)and Hill and Womersley (1991) defined flaws in the sample handling procedures
that most likely affected variability of FAP results in many of the early field studies. In
spiteof theseproblems,FAP has been shownto accumulateover time in nonmitotic
tissues and organs of numerous aquatic poikilotherms(Luo and Hultin 1986,Mullinand
Brooks 1988, Vernet et al. 1988,Hill and Womersley 1991,Nicol et al. 1991,Hill and
Womersley, in review) as well as mammalian species traditionally studied in
gerontological research (Tappel et al. 1973,Shimasaki et al. 1977, Miquel et al. 1978,
Shimasaki et al. 1980).
It is of primary importance to bothgerontologistsand field fish biologists that the
factors that lead to age pigment genesis and variabilitybe defined. Biochemicalevents
involvedwith age pigment formation are consideredto be stochasticin nature,determined
in part by rates of cellular metabolism and the generationof superoxideand hydroxyl
radicals in organelles and the cytoplasm (Harman 1984). Free radicals, in the presenceof
metalsand divalentoxygen, can initiatea cascadeof lipid peroxidation events leading to
malondialdehyde and other carbonylcross-linkingreactants (Gutteridge 1987). Amine
containingbiomolecules (amino andnucleic acids, proteins, phospholipids,etc.) that
have been cross-linkedbecomedeactivated and eventuallydigestedin the lysosomal
recycling system (Harman 1989). Cross-linked portions that are nondigestable remainin
the cytoplasm as age pigment residuesand have autofluorescentproperties. Thus, there
are numerous criticalfactors that maydetermine the rate of age pigment genesis at the
cellular level, including: 1) the rate of free radical productionthrough oxidative
metabolism (Fridovich 1976,Armstrong 1984,Freeman 1984),2) the efficiency of
113
antioxidant mechanisms responsible for free radical control (Sohal et al. 1990, Cutler
1991), 3) the susceptibility of cellular lipid components to free radical-mediated oxidation
(Packer et al. 1967, Tappe11975, Mead 1976),4) presence of catalytic factors such as
oxygen (Sohal and Brunk 1989, Sohal et al. 1989) and metals (Aloj Totaro et al. 1985,
Gutteridge 1985), and 5) the efficiency oflysosomal digestive enzymes (Davies 1988,
Marzabadi et al. 1991)
Relatively few studies have examined the effects of metabolism on age pigment
formation. Rearing DrosQphila melano~ster at different temperatures, Sheldahl and
Tappel (1974) first demonstrated a direct effect on the quantity of fluorescent products
extracted from the flies. Other studies on FAP in insects followed, testing the effects of
oxygen consumption ill. melanogaster, Miquel et al. 1976), physical activity level
(Musca domestica, Sohal and Donato 1979, Sohal and Buchan 1981), and ambient
temperature (Oncweltus fasciatus, McArthur and Sohal 1982). Questions have now been
raised as to whether the fluorescent products extracted in these studies represent FAP or
pteridine metabolites that are ubiquitous in insects (Ziegler and Harmsen 1969, Lehane et
al. 1986, Sheehy and Roberts 1991). Nonetheless, Sohal and Donato (1979) and Sohal
(1981) confirmed the positive effect of activity level on lipofuscin accumulation in M.
domestica using electron microscopy. Activity level has also been found to affect rates of
age pigment accumulation in rats (Basson et al, 1982) and hamsters (papafrangos and
Lyman 1982). Most recently, Sheehy (l990b) reared crayfish (Cherax cuspidatus) at
three temperatures, finding greater quantities of lipofuscin in the olfactory lobe cell mass
of animals reared at higher temperature.
The following investigation was conducted to address additional questions related
to metabolic and growth-related effects on FAP genesis in nonmitotic tissues, using a
vertebrate aquatic poikilotherm as the study organism. Our primary objectives were: 1) to
determine the degree to which FAP content may vary within and between sibling and
114
non-sibling fish of identical age, and 2) to determine whether body weight, rearing
temperature, ration level, photoperiod, or a combination of these factors, can influence
FAP genesis, ultimately affecting the ability to predict chronological age.
Materials and methods
Model animal
We chose the tropical freshwater fish Puntius conchonius (Cyprinidae), the rosy
barb, as our study organism. ~. conchonius is small, easy to breed, and tolerant to a
wide range of physiological temperatures (15 to 33° C), lending itself well to
physiological studies requiring large numbers of individuals in limited laboratory space.
Clutches used for the following experiments were obtained by breeding adult pairs in 19 I
all-glass aquaria for 8 to 12 h at 25° C. Adults were removed immediately following the
spawning event Larvae hatched within 3 to 4 d after fertilization, and yolk sack
absorption was complete after 3 to 4 d, at which time the larvae were fed a diet of 50-150
um diameter Hatchfry Encapsulon (Argent Chemical Laboratories) for 4 d. Fry were
subsequently fed a combination of larval Anemia salina and crushed flake food until
transfer to experimental aquaria at approximately 45 d of age.
Aquarium system
The experimental system consisted of 18 118 I all-glass closed-system aquaria.
Water quality was maintained using air-driven undergravel filters with 25% of the water
exchanged biweekly. Dissolved oxygen and pH were monitored weekly and temperature
was monitored daily. Aquaria were cooled to 19° C using a thermostatically controlled
refrigerated water bath which circulated chilled water through coiled glass tubing in each
aquarium. All 25° and 31° C aquaria were held at temperature using 100 W thermostatic
heaters. In addition, all 25° C aquaria were equipped with glass coils cooled by tap water
(22 to 23°C) to assure proper temperature on days when room temperature exceeded 25°
115
C. Photoperiod was controlled by enclosingeach aquariumin heavy gauge black plastic
sheeting to excludeexternallight. All aquaria were coveredwith full hoods equipped
with 15 W fluorescentbulbs controlled by timer boxes.
Experimental design
Two independent experimentswere conducted to assess the interactive effectsof
temperature, ration level,photoperiod,clutch, and individualsizeon FAP accumulation
in brain and heart tissues of f. conchonius. In the first, temperature and ration level were
varied while photoperiod was held constant In the secondexperiment, temperature and
photoperiod were variedwhile ration level was held constant. Possible interactions
between photoperiod. andration level were not addressedin this study.
Temperature/ration experiment: This experimentwas designedto test for
interactiveeffects of temperature, ration level, clutch, and individual body size on
variabilityof FAP. At an age of 45 d, three uniparentalclutchesof fry (referred to as CI,
C2, and C3) were randomlydivided into experimentalaquariaat a stockingdensity of 24
fish per aquarium. Temperaturelevels were 19°C± 0.2°C, 25°C± 0.2°C,and 31°C±
D.2°C. Within each temperature, fish were fed commercialflake food (Wardley'sBasic
Flake, The Wardley Corporation) either ill! libitum twice daily (highration, or Hi), or ad
libitum once on alternating days (low ration, or 1..0). This three-temperature, two-ration
design was repeated for each of the three clutches,which were acquiredfrom different
parental sources. All aquariawere held at a constant 12 h light: 12 h dark photoperiod.
Individualswere sacrificed for FAP measurementafter beingheld underexperimental
conditions for 8 mo, at an age of 280 to 290 d. The effects of temperature and body size
on metabolicrate weremeasuredon individualsfrom each of the six treatments in this
experiment Methodsfor oxygen consumptionmeasurement are describedin a later
subsection.
116
Temperature/photoperiod experiment: Interactive effects of temperature,
photoperiod, clutch, and body size on FAP content were tested using a design similar to
the temperature/ration experiment. Individuals were divided into aquaria at 45 d with 24
fish per aquarium. Treatments included three temperature levels (19, 25, and 31°C) and
photoperiod was held at two levels within each temperature; 18 h light: 6 h dark (referred
to as l8L), or 6 h light: 18 h dark (referred to as 6L). All treatments were fed high ration
levels. Three uniparental clutches, referred to as C4, C5, and C6 to avoid confusion with
the previous experiment, were reared under each of the six treatment conditions.
Individuals were held under experimental conditions for 8 mo, and sacrificed at 280 to
290 dofage.
Oxygen consumption
Laboratory determinations of oxygen consumption were made on three randomly
selected f. conchonius from each of the 6 temperature/ration treatments (C3 only)
approximately one month prior to the end of the experiment. Respiration was determined
by measuring changes in oxygen tension in the water medium using a polarographic
electrode. The respiration chamber consisted of a one liter specimen flask connected in
series to another one liter flask containing oxygen and temperature electrodes and a
mixing bar driven by a magnetic stirrer. Oxygen tension was measured directly as mg 02
per liter using an Omega Model PHH-71 dissolved oxygen/temperature probe. Water
medium was recirculated through the two flasks in series using a motorized pump. The
two flasks were held in a water bath whose temperature was maintained using a
submerged glass coil connected to a thermostatically-controlled Lauda Model RM5 water
circulator. Temperature was held at either 19,25, or 31°C ± 0.2°C throughout each
observation.
Triplicate oxygen consumption measurements were made for each individual,
with measurements lasting between 20 and 40 min, depending upon the rate of
117
consumption. Oxygen measurements were only taken within the range of 80 to 100% of
air saturation (Belman and Childress, 1973), and the medium was resaturated between
observations using an air stone. New water medium, sterilized by UV irradiation, was
used for each specimen, and background oxygen depletion due to microbial respiration
was negligible. Observations were made between 0900 and 1500 hrs to account for daily
changes in metabolic rhythms. Specimens were starved for at least 16 h prior to the
observations to minimize specific dynamic action (Saunder, 1963). Fish wet weight and
chamber water volume were measured at the end of the three replicate observations.
Specimen/tissue preparation
Puntius conchonius individuals were anesthetized by ice-bath immersion, and
weighed to the nearest 0.01 g. Upon removal of cranial bones, the brain cavity was
thoroughly rinsed with Ringer's saline solution to remove blood and lipid deposits.
Brain tissues removed for FAP assay included the cerebral, inferior, optic, and cerebellar
lobes, with the medulla being severed posterior to the tenth cranial nerve. Cranial nerves
were severed at their point of attachment to the brain. Hearts were excised and ventricle
myocardial tissues separated and saved for FAP analysis. Brain and myocardial tissues
were briefly rinsed in Ringer's solution, placed in Eppendorf vials, and immediately
frozen in liquid nitrogen until analysis.
FAP extraction and assay
Soluble FAP materials in tissues were quantified following the modified protocols
outlined by Hill and Womersley (1991). Tissues were lyophilized and their dry weights
measured to the nearest 0.0001 mg on a Cahn Model C-32 electronic microbalance.
Tissues were macerated at room temperature for 4 min in a Potter-Elvehjem glass tissue
homogenizer using at least 1 m1of spectroscopic grade chloroform (CHCI3) per 10 mg
dry tissue. The initial meniscus was marked and the volume reconstituted after
homogenization. Homogenates were transferred to glass centrifuge tubes and vortex
118
mixed for 2 min with an equal volume of a solution containing 100 roM magnesium
chloride (MgCI) in glass distilled water and spectroscopic grade methanol at a ratio of 3: 1.
Samples were centrifuged at 0-5°C for 10 min at RCF =3000 x g. Nonpolar and polar
fractions were immediately removed and assayed for FAP content Sample extractions
were performed singly and rapidly to prevent in vitro production ofFAP-like
fluorophores in the sample solvent (Hill and Womersley, 1991).
Extract solvents were placed in quartz-silicate microcuvenes with a 4 mm pathlength
and sealed with a teflon stopper to prevent evaporation. Temperature of the sample
solvent was maintained at 18°C using a thermostatically controlled turret microcell holder
connected to a Brinkmann Lauda Model RM6 refrigerated circulating water bath. A
Perkin Elmer LS-5B luminescence spectrometer, Acer computer, and Perkin Elmer CFS
PC-control software (version 3.0) were used for the fluorometric assay. The LS-5B
luminescence spectrometer was equipped with a Xenon light source and a Hamamatsu
Type R928 red-sensitive photomultiplier tube to reduce possible bias in the 500-700 nm
region (Eldred et al., 1982). Excitation and emission slits were set at 5 and 20 nm,
respectively. Fluorescence spectra reported herein were corrected for the spectral
emission characteristics of the source lamp, the spectral efficiencies of the excitation and
emission monochromators, optics, and the photomultiplier tube using the method of
Rhys-Williarns et aI. (1983) and the emission correction routine in the CPS PC-control
software. All sample readings were expressed as a percentage of Quinine sulfate (0.1
mglllN sulfuric acid) standard luminescence at an excitation setting of 345 nm and an
emission setting of 450 om. Sample FAP concentrations were expressed as whole-organ
%fluorescence (%FL), where:
sample luminescence%FL = x solvent volume (ml) x 100
standard luminescence
119
Statistical analysis
Unpaired two-tailed t-tests were applied to data within treatments to test for
differences in FAP content between sexes. Preliminary examination of the data revealed a
body weight effect on FAP content, thus it was necessary to standardize this effect by
applying Analysis of Covariance (ANCOVA) to the data. ANCOVA was used, whenever
possible, to evaluate the effects and interactions of temperature, ration level or
photoperiod, and clutch on FAP extracted from brain and heart tissues, with body weight
regressed as a covariate. Temperature, ration, photoperiod, and clutch were treated as
class variables, with clutch being random. All FAP data were log-transformed to
homogenize the variances. Computations were made using GLM procedures of the SAS
statistical software (SAS, 1985).
Results
Temperature/ration experiment
Both temperature and ration level had considerable effects on body weight in this
experiment by 280 to 290 d of age (Fig. 4-1a). At low ration level, mean (± 95%
confidence limits) fish weight ranged from 0.79 ± 0.18 g (CI-31°-Lo) to 1.33 ± 0.15 g
(C3-19°-Lo). High ration fish ranged in weight from 1.72 ±0.19 g (CI-19°-Hi) to 3.97
± 0.73 g (C2-31-Hi). Within temperatures, weight was greater in fish fed high rations,
and this effect was accentuated with increasing temperature (Fig. 4-1a). Mean weight
ratios between high and low ration treatments were approximately 2:1 at 19°C, 3:1 at
25°C, and 4: 1 at 31°C. Thus, under low ration conditions, temperature had little or no
effect on final size, whereas temperature had a positive effect on weight at high ration
levels.
Polar and nonpolar FAP extracted from P. conchonius brain tissues had
excitation and emission characteristics similar to those previously reported for this species
120
(Hill and Womersley, 1992). There were no apparent treatment effects on excitation and
emission maxima or the overall shape of the spectra. FAP extracted from brain was much
more polar in solubility with ratios of polar:nonpolar FAP varying great!y (ca. 6:1 to
50:1) within and between clutches and treatments (Fig. 4-2).
Brain polar FAP content varied greatly between all experimental treatments with
mean %FL values ranging from 52.1 ± 12.5 (CI-31°-Hi) to 541.7 ± 98.1 (C2-19°-Lo)
(Fig. 4-2a). The most notable trend was the inverse relationship between rearing
temperature and polar FAP content This trend held for all three clutches and for both
ration levels. There was a tendency for low ration treatments to have higher polar FAP
values than high ration treatments at both 19°C and 31°C in all three clutches (Fig. 4-2a).
The above trends were not apparent for nonpolar FAP extracted from the same brain
tissues (Fig. 4-2b). Brain nonpolar FAP, present in much lower quantity than polar
FAP, varied between clutches both within and between treatments. Within clutches,
nonpolar FAP tended to be higher in both 25°C treatments.
An attempt was made to apply ANCOVA models to the above brain FAP data
The data failed to meet one of the most basic assumptions of ANCOVA, which is a
parallel relationship between the dependent variable (in this case, brain FAP) and the
covariate (fish weight) under the different treatment conditions. Scatter plots of brain
polar FAP as a function of fish weight revealed a negative relationship between the two
variables, and this was evident for most treatments (Fig. 4-3); however, the nature of this
response varied under the different treatment conditions. For example, body weight
appears to have a far greater effect on brain polar FAP at 19°C (Fig. 4-3a) than for
animals of similar size at 31°C (Fig. 4-3c). Because of this, it was not possible to apply
further statistical analyses to these data to determine possible interactive effects of
temperature, ration level, or clutch on brain FAP. While ANCOVA was not feasible, it is
121
clear that brain polar FAP values are inversely correlated with temperature in fish of
similar size.
FAP extracted from heart tissues also had excitation and emission properties
similar to those previously reported for this species (Hill and Womersley, 1992).
Rearing conditions in this experiment did not affect the shape of the luminescence spectra
of either soluble fraction. The solubility of heart FAP in polar and nonpolar solvent
fractions varied between clutches, but was approximately equal or slightly greater in the
polar fraction.
Heart polar FAP varied between treatments and clutches, with %FL values
ranging from 2.44 ± 0.34 (C2-19°-Lo) to 11.77 ± 2.08 (C2-31°-Hi) (Fig. 4-4a). Over all
temperatures, except for C1-19° and C3-19°, FAP values were higher in fish fed high
ration levels. Excluding C3-19°-Lo, temperature had no clear effect on heart polar FAP in
low ration treatments. Temperature did, however, have a positive effect on polar FAP in
high ration animals, especially between 25° and 31°C (Fig. 4-4a). Heart nonpolar FAP
were variable between clutches both within and between treatments with no obvious
consistent trends related to rearing temperature or ration level (Fig. 4-4b).
Attempts to apply ANCOVA to heart FAP data failed for reasons similar to those
stated above for brain data. Plots of heart total FAP as a function of fish weight (Fig. 4
5) revealed that FAP was negatively correlated with fish weight at 19°C, and positively
correlated at 25° and 31°C.
Metabolic rate, measured as oxygen consumption, varied with both fish weight
and temperature. Within the three temperatures, the weight-specific rate of oxygen
consumption was inversely related to fish weight, presumably due to the greater ratio of
surface area to volume at smaller body sizes (Fig. 4-6). The effect of body weight on
oxygen consumption was stronger at 19°C and 25°C than at 31°C. Temperature showed
the expected positive effect on metabolic rate.
122
TemperaturelphotQperlod experiment
Fish weight was not greatly affected by temperature, photoperiod, or clutch in this
experiment (Fig. 4-1b). Mean (± 95% confidence limits) fish weights ranged from 2.57
±0.28 g (C6-19°-6L) to 4.27 ±0.70 g (C6-31°-18L). There were no apparent
differences in weight between photoperiods within temperatures, and size differences
between temperatures were not as obvious as those for high ration treatments in the
temperature/ration experiment
Brain polar FAP content was inversely related to rearing temperature for all three
clutches, with mean %FL values ranging from 275.1 ±84.6 (C4-19°-6L) to 50.5 ± 16.6
(C5-31°-6L) (Fig. 4-7a). The relation of polar FAP to temperature was similar to that
reported for high ration animals in the temperature/ration experiment (Fig. 4-2a).
Photoperiod did not appear to have a consistent influence on polar FAP content.
Nonpolar FAP, present in much lower quantities than the polar fraction, did not appear to
be affected by either rearing temperature or photoperiod in C5 or C6, but was lower at 19°
and 25°C in C4 (Fig. 4-7b).
Results of the ANCOVA model developed for brain polar FAP data are presented
in Table 4-1. The ability to meet ANCOVA criteria was probably due to the smaller
magnitude of size differences resulting from the experimental treatments. The ANCOVA
test revealed that both temperature (F 2,219 =96.54, P < 0.0001) and clutch (F 2,219 =
203.96, P < 0.00(1), as well as their interactive term (F 4,219 = 147.96, P < 0.0001),
were significant factors affecting polar FAP content in P. conchonius brain (Table 4-1).
Neither fish weight or photoperiod alone significantly affected polar FAP. The interactive
term of photoperiod x clutch was significant (F 2,219 = 7.92, P = 0.0005). Brain
nonpolar FAP data failed to meet ANCOVA assumptions, and a model was not applied.
Heart polar FAP appeared to bepositively correlated to rearing temperature, with
mean %FL values ranging from 7.80 ±1.99 (C6-19°-18L) to 12.33 ± 4.75 (C5-31°-18L)
123
-_.__ ._-_._-_._ ..._--_._ ..__ .- ...
(Fig. 4-8a). Application of an ANCOVA model to the data revealed that fish weight (F
1,219 = 30.60, P < 0.00(1), temperature (F 2,219 = 20.05, P < 0.0001), and clutch (F
2,219 =12.29, P < 0.00(1) all significantly affected polar FAP content, and that
photoperiod had no general effect (F 1,219 =0.01, P =0.9202) (Table 4-2). Significant
interaction terms included temperature x photoperiod, temperature x clutch, and
temperature x photoperiod x clutch, indicating a highly mixed response of heart polar
FAP to each these factors. Application of ANCOVA procedures to heart nonpolar FAP
was not possible, however the most generalized trend was for %FL values to be lowest at
25°C (both photoperiods and all clutches), slightly higher at 19°C, and highest at 31°C
(Fig. 4-8b).
Discussion
The results of the present study provide strong evidence for a relationship
between temperature, body weight, metabolic rate, and FAP genesis in ~. conchonius.
Further, the differences observed between clutches suggest that this is not merely a
physiological effect, but that genetic factors are integrally involved. This is the first study
to demonstrate a temperature effect on FAP accumulation in a vertebrate organism,
confirming the usefulness of fish and other higher order poikilotherms for aging research
(Comfort 1963, Woodhead 1980).
Both temperature and body size showed differential effects on FAP accumulation
in brain and heart tissues. The most striking pattern, repeated for all clutches in both
experiments, was the inverse relationship between ambient temperature and brain polar
FAP. This result is entirely contrary to patterns reported by Sheehy (1990b) for
lipofuscin detected in the olfactory lobe cell mass of crayfish brains. Moreover, our brain
polar FAP data are contradictory to the precept that metabolic rate, free radical production,
and age pigment genesis should be positively correlated with temperature (Sohal 1984).
124
WhileMcArthur and Sohal (1982) havecautioned againstassuming that oxygen
consumption is strictly temperature dependent, our own data clearlyshow thatE.
conchonius has higher ratesof oxygen consumption at higher temperatures. In
comparison, heartdata fromthe sameanimals demonstrates a significant positive
relationship of FAP to ambient temperature, a result that agrees with previously published
age pigmentresearch (Sheehy 1990b).
The fact that differentpatterns of FAP accumulation occurin E. conchoniusbrain
and heart tissueshave led us to suspect the involvement of other unknown
physiologicallbiochemical factors in the genesisofFAP. Thesedifferences are better
understood when the determinants of age pigmentgenesisare considered in terms of the
adaptive physiologyof the organism. It is well established that poikilothermic animals
acclimate to colder ambient temperatures by increasing thedegreeof unsaturation of the
free fattyacids and fatty acidmoieties of phospholipids to maintain membrane
permeability and fluidity (Hazeland Prosser 1974,Cossins et al. 1977, Cossins and
MacDonald 1989,Hazel andWilliams, 1990). Homeoviscous adaptation could
theoretically predispose cold-adapted tissues to be more susceptible to freeradical attack
and peroxidative breakdown, leading to the increasedproduction of cross-linking
reactants in the cytoplasm. In theevent that increased lipid radical generation is not
compensated for by increases in non-enzymatic antioxidants (e.g., beta carotene, alpha
tocopherol, ascorbic acid), higherrates of age pigment formation would result. Thus,
temperature may affectage pigmentgenesis in twodifferentmanners: through increased
metabolism and free radicalproduction at higher temperatures, and by increased lipid
peroxidation potentialat lower temperatures.
Thoughspeculative, theabove theory may partiallyexplain the differences in FAP
accumulation observed in E. conchonius brain and heart tissuesunder the various
experimental conditions. IfFAP genesis is indeeddependenton some balancebetween
125
metabolism and lipid peroxidative potential, then it would be expected that brain (a lipid
rich tissue) would be more susceptible to lipid peroxidation events than heart tissues at
lower temperature. Consequently, if the potential for lipid peroxidation is great enough,
it could mask any positive temperature effects related to metabolism. The converse may
hold true for myocardial tissues, which possess a lower lipid content and whose
metabolic activity is directly affected by temperature (Bowler and Tirri 1990).
An additional factor important to the above considerations is the dependence of
metabolic rate and FAP genesis on body size. Both brain and heart FAP contents varied
with fish weight in this study. The nature of this relationship varied between tissues as
well as between temperatures within tissue type. Respiration measurements also
demonstrated that fish of smaller body size had higher rates of oxygen consumption per
unit of body weight at any given temperature. Higher weight-specific metabolism in
smaller specimens has been widely described in animals including fish (Wieser 1984,
Schmidt-Nielson 1984, Taylor 1987). Using freshwater carp (Cyprinus carpio), Itazawa
and Oikawa (1986) demonstrated a decrease in weight-specific metabolic rate with
increasing size that was due to a relative increase in quantity of tissue of low metabolic
activity (bone and white muscle). Based on this widely accepted relationship between
size and metabolism, it could be predicted that smaller fish would accumulate higher
quantities of FAP than larger fish of identical age, particularly smaller fish living at lower
temperatures where the potential for lipid peroxidation is higher. This is indeed the case
for both brain and heart tissues of f. conchonius. While no previous study has
considered weight-specific metabolic rate as a factor affecting age pigment accumulation
within a given species, comparisons have been made between species. Dogs, which have
a fivefold greater weight-specific rate of oxygen consumption than humans, also
accumulate myocardial lipofuscin five times faster than humans (Munnel and Getty 1968,
Strehler et al. 1959).
126
Restriction of caloric intake has been shown to increase mean life span and
maximum lifespan potential in numerous invertebrate and vertebrate animals (see reviews
by Weindruch 1984, and Finch 1990). There are many potential mechanisms by which
caloric restriction might slow the aging process, yet the precise manner in which it occurs
is still unknown. The broadest interpretation of most caloric restriction studies is that
reduced caloric intake acts to decrease oxidative metabolism, reducing free radical
production and associated molecular damage (Weindruch 1984). In addition, diet
restriction induces synthesis of catalase (Koizumi et al. 1987, Semsei et al. 1989, Rao et
al. 1990), free glutathione (Lang et al, 1990), glutathione peroxidase (Rao et al. 1990)
and superoxide dismutase (Semsei et al. 1989, Rao et al. 1990) in livers of diet-restricted
rodents. Superoxide dismutase catalyzes conversion of the superoxide radical to
hydrogen peroxide, while catalase and glutathione peroxidase are responsible for
converting hydrogen peroxide into oxygen and water, preventing its dissociation into
hydroxide and the highly-damaging hydroxyl radical.
We found no evidence that diet restriction lowers oxygen consumption in E.
conchonius. On the contrary, fish fed low rations had an overall smaller body size and
higher weight-specific rates of oxygen consumption. While fish held on low ration
attained less weight than their high ration counterparts, they were by no means
undernourished or starved. All low ration fish had well developed gonads and abundant
visceral fat deposits by the end of the experiment Thus, even though ANCOVA analysis
of FAP data was not possible for the temperature/ration experiment, we feel that any FAP
differences observed between dietary treatments were due mainly to the differential effects
of body size and the lipid peroxidative potential of the respective tissues rather than
malnourishment stress.
Photoperiod - frequently a determinant of growth and activity level in fishes
appears to have no significant effect on FAP accumulation in E. conchonius. However,
127
_. __..•_------.. --"'---
our experiments were conducted only through the first quarter of the potential lifespan of
P. conchonius; longer-term effects may occur. This is the first study to test for effects of
photoperiod on accumulation of age pigments, but it is not the first to question the role of
the level of physical activity. Physical activity level has been shown to directly affect age
pigment levels in flies (Sohal and Donato 1979, Sohal 1981) and rodents (Basson et al.
1982, Papafrangos and Lyman 1982). Papafrangos and Lyman (1982) induced
hibernation in Turkish hamsters, slowing the rate of lipofuscin accumulation in the heart
and brain and prolonging their lifespan.
The significant differences in FAP observed between £. conchonius non-siblings
of identical age suggest that genetic factors may be involved in their formation.
Differential expression of genes related to metabolism, production of antioxidant
enzymes, or lysosomal digestive enzymes could all potentially influence the rate of age
pigment production. This is highly speculative with regard to £. conchonius, but the
topic deserves further investigation.
With respect to aging research, the results of this study provide further support
for Strehler's (1959) suggestion that age pigments represent physiological events
associated with the aging process. We have provided evidence that FAP production is
dependent upon temperature, body size, and oxygen consumption, and that the manner in
which these factors influence FAP differs between tissue types. While the
interdependence of these relationships is not completely understood, our results provide
indirect evidence that homeothermic adaptation may affect the susceptibility of cellular
lipid components to peroxidative decomposition, a process which may itself lead to
increased FAP production. This possibility has not previously been considered in studies
on temperature and FAP accumulation in poikilotherms.
The results reported in the present work are also relevant to biologists hoping to
apply FAP methodologies to models for chronological age estimation in wild stocks of
128
-------- -------- ----- --
aquaticpoikilotherms. Hill and Womersley (inreview) examinedthe timecourseofFAP
accumulation in brain and heart tissuesof P. conchonius rangingfrom 60 to 538 d of age.
Brain and heartFAP accumulated with time, but brainFAP was highly variable at any
given age, and less useful than heart FAP for purposes of age prediction. Hill and
Womersley's findings were based on animals that had beenreared in the laboratory under
stable temperature and dietaryregimes. In the presentstudy, the degree of brain and heart
FAP variationmeasured between temperatures, body sizes, and clutches in P. conchonius
was inducedin approximately half the timeframe of Hill and Womersley's (in review)
study and over a relatively short portionof the animals' potentiallifespan (ca.4 yr).
Presumably, thesevariations would increasewith time,and this significantvariability
must be taken into accountin considering the useof any age-predictive model basedon
FAP content The largeeffectsof interactions betweenvariables reportedin the present
studyindicatethat it will be difficult to account for individual effectsof variables and
accurately calibrate age againstFAP contentin wildspecimens. Althoughmanymarine
animalslive underrelatively stable thermal conditions, manyother speciesexperience
ranges of temperature that, according to our results, wouldproducemajor variationsin
FAP content The lifetimehistoryof exposure to temperature or other variables that affect
FAP content is usually unknown for a wildspecimen at the time of aging. Even without
temperature variability, sizevariabilitywithinage classesmay pose a problemif the rate
of FAP accumulation is dependentupon the weight-specific rate of oxygen consumption.
Variability in FAP accumulation mayoccuras a directmetabolic resultof variability in
ration level; suchan effect could not be separated from the effectof growth in our study.
FAP variability may be accentuated in freshwater species in temperate regions, or in
nearshore or migratory marinespecies thatexperience widefluctuations in temperature,
salinity, food availability, pollution (AlojTotaroet al, 1985), or other environmental or
physiological variables. All these factorsmust be given high priority if FAP
129
-_.__.._------_ .. -._-_._ ..__ ...
methodologies are to be used for chronologically aging natural populations of aquatic
animals. Based on the amount of variability we have encountered, the number of
variables involved, and the complexity of the interactions, successful estimation of age of
wild specimens is likely to be very difficult.
130
Literature cited
Aloj Totaro, E., F. A. Pisanti, and P. Glees. 1985. The role of copper level in the
formation of neuronal lipofuscin in the spinal ganglia of Tor,pedo m. Mar.
Environ. Res. 15: 153-163.
Armstrong, D. 1984. Free radical involvement in the formation of lipopigments. pp. 129
141 in Armstrong, D., R. S. Sohal, R. G. Cutler, and T. F. Slater (eds.) Free
Radicals in Molecular Biology, Aging, and Disease. Raven Press, N.Y.
Basson, A. B. K., S. E. Terblanche, and W. Oelofsen. 1982. A comparative study on
the effects of aging and training on the levels of lipofuscin in various tissues of
the rat. Compo Biochem. PhysioI. 71A: 369-374.
Belman, B. W., and J. J. Childress. 1973. Oxygen consumption of the larvae of the
lobster Panulirus interruptus (Randall) and the crab Cancer productus (Randall).
Compo Biochem. PhysioI. 44A: 821-828.
Berman, M. S., A. L. McVey, and G. Ettershank. 1989. Age determination of Antarctic
krill using fluorescence and image analysis of size. Polar BioI. 9: 267-271.
Bowler, K., and R. Tirri. 1990. Temperature dependence of the heart isolated from the
cold or warm acclimated perch~ fluviatilis). Compo Biochem. PhysioI. 96A:
177-180.
Brown, C. L., and C. 1. George. 1985. Age-dependent accumulation of macrophage
aggregates in the yellow perch, Perea flavescens (Mitchill). J. Fish Diseases 8:
135-138.
Clarke, A., M. A. Kendall, and D. J. Gore. 1990. The accumulation of fluorescent age
pigments in the trochid gastropod Monodonta lineata. J. Exp. Mar. BioI. EcoI.
144: 185-204.
Comfort, A. 1963. Effect of delayed and resumed growth on the longevity of a fish
<Lebistes reticulatus, Peters) in captivity. Gerontol. 8: lSD-ISS.
131
Cossins, A. R., M. J. Friedlander, and C. L. Prosser. 1977. Correlations between
behavioral temperature adaptations of goldfish and the viscosity and fatty acid
composition of their synaptic membranes. J. Compo Physiol, 120: 109-121.
Cossins, A. R., and A. G. Macdonald. 1989. The adaptation of biological membranes to
temperature and pressure: fish from the deep and cold. J. Bioenerg. Biomem. 21:
115-135.
Cutler, R. G. 1991. Antioxidants and aging. Am. J. Clin. Nutrit. 53: 373-378.
Davies, K. J. A. 1988. Protein oxidation, protein cross-linking, and proteolysis in the
formation of lipofuscin. pp. 24-25 in Zs.-Nagy, I. (ed.). Lipofuscin - 1987: State
of the Art. Elsevier Science Publishers, Amsterdam.
Eldred, G. E., and M. L. Katz. 1989. The autofluorescent products of lipid peroxidation
may not be lipofuscin-like. Free Rad. BioI. Med. 7: 157-163.
Eldred, G. E., G. V. Miller, W. S. Stark, and L. Feeney-Burns. 1982. Lipofuscin:
resolution of discrepant fluorescence data. Science 216: 757-759.
Ettershank, G. 1983. Age structure and cyclical annual size change in the Antarctic krill,
Euphausia superba. Polar BioI. 2: 189-193.
Ettershank, G. 1984a Methodology for age determination of Antarctic krill using the age
pigment lipofuscin. BIOMASS Handb. No. 26. 14 pp.
Ettershank, G. 1984b. A new approach to the assessment of longevity in the Antarctic
krill Euphausia superba. J. Crust. BioI. 4: 295-305.
Ettershank, G. 1985. Population age structure in males and juveniles of the Antarctic
krill, Euphausill superba Dana. Polar BioI. 4: 199-201.
Finch, C. E. 1990. Longevity, senescence, and the genome. The University of Chicago
Press, Chicago. 922 pp.
Fletcher, B. L., C. J. Dillard, and A. L. Tappel. 1973. Measurement of fluorescent lipid
. peroxidation products in biological systems and tissues. Anal. Chern. 52: 1-9.
132
Freeman, B. A. 1984. Biological sites and mechanisms of free radical production. pp.
43-52 in Armstrong, D., R. S. Sohal, R. G. Cutler, and T. F. Slater (OOs.) Free
Radicals in Molecular Biology, Aging, and Disease. Raven Press, N.Y.
Fridovich.L 1976. Oxygen radicals, hydrogen peroxide, and oxygen toxicity. pp. 239
277 in Pryor, W. A. (00.) Free Radicals in Biology. Academic Press, N.Y.
Gutteridge, J. M. C. 1985. Age pigments and free radicals: fluorescent lipid complexes
formed by iron- and copper-containing proteins. Biochim. Biophys. Acta 834:
144-148.
Gutteridge, J. M. C. 1987. Oxygen radicals, transition metals and aging. pp. 1-22 in Aloj
Totaro, E., P. Glees, and F. A. Pisanti (eds.) Advances in Age Pigments
Research. Advances in the Biosciences. Pergamon Press, N.Y.
Hammer, C. 1988. Accumulation of fluorescent age pigments in embryos and larvae of
pike, Esox lucius (Esocidae, Teleostei) at different incubation temperatures.
Environ. BioI. Fish. 22: 91-99.
Hammer, C., and E. Braum. 1988. Quantification of age pigments (lipofuscin): Mini
Review. Compo Biochem. PhysioI. 90B: 7-17.
Harman, D. 1984. Free radicals and the origination, evolution, and present status of the
free radical theory of aging. pp. 1-12 in Armstrong, D., R. S. Sohal, R. G.
Cutler, and T. F. Slater (eds.) Free Radicals in Molecular Biology, Aging, and
Disease. Raven Press, N.Y.
Harman, D. 1989. Lipofuscin and ceroid formation: the cellular recycling system. pp. 3
15 in Porta, E. A. (ed.). Lipofuscin and Ceroid Pigments. Plenum Press, N.Y.
Hazel, 1. R., and C. L. Prosser. 1974. Molecular mechanisms of temperature
compensation in poikilothenns. Physiol. Rev. 54: 620-677.
Hill, K. T., and R. L. Radtke. 1988. Gerontological studies of the damselfish, Dascyllus
albisella. Bull. Mar. Sci. 42: 424-434.
133
Hill, K. T., and C. Womersley. 1991. Critical aspects of fluorescent age-pigment
methodologies: modification for accurate analysis and age assessments in aquatic
organisms. Mar. BioI. 109: 1-11.
Hill, K. T., and C. Z. Womersley. Fluorescent age pigment accumulation and the use of
multiple regression models for age estimation in two fish species. J. Exp. Mar.
Biol, Ecol. (in review).
Hirche, H.-J., and K. Anger. 1987. The accumulation of age pigments during larval
development of the spider crab, Hyas araneus (Decapoda, Majidae). Compo
Biochem. Physiol. 88B: 777-782.
Itazawa, Y and S. Oikawa. 1986. A quantitative interpretation of the metabolism-size
relationship in animals. Experentia 42: 152-153.
Koizumi, A, R. Weindruch, and R. L. Walford. 1987. Influences of dietary restriction
and age on liver enzyme activities and lipid peroxidation in mice. 1.Nutrit. 117:
361-367.
Lang, C. A, T. S. Chen, and B. J. Mills. 1990. Glutathione status and longevity are
enhanced by dietary restriction. FASEB J. 4: abstract 601.
Lehane, M. J., J. Chadwick, M. A Howe, and T. S. Mail. 1986. Improvements in the
pteridine method for determining age in adult male and female Stomoxys
calcitrans (Diptera: Muscidae). 1. Econ. Entomol. 79: 1714-1719.
Luo, S.-W., and H. O. Hultin. 1986. Effect of age of winter flounder on some properties
of the sarcoplasmic reticulum. Mech. Age. Dev. 35: 275-289.
Marzabadi, M. R., R. S. Sohal, and U. T. Brunk. 1991. Mechanisms of
lipofuscinogenesis: effect of the inhibition of lysosomal proteases and lipases
under varying conditions of ambient oxygen in cultured rat neonatal myocardial
cells. APMIS 99: 416-426.
134
_.._- ._._-_._--_...--_ .. "_.'- - .
McArthur, M. C., and R. S. Sohal. 1982. Relationship between metabolic rate, aging,
lipid peroxidation, and fluorescent age pigment in milkweed bug, Oncopeltus
fasciatus (Hemiptera). J. Gerontol, 37: 268-274.
Mead, J. F. 1976. Free radical mechanisms of lipid damage and consequences for cellular
membranes. pp. 51-68 in Pryor, W.A. (ed.) Free Radicals in Biology. Academic
Press, N.Y.
Miquel, 1., R. R. Lundgren, K. G. Bensch, and H. H. Atlan. 1976. Effects of
temperature on life span, vitality and fine structure of Drosophila melanogaster.
Mech. Ageing Dev. 5: 347-370.
Miquel, J., D. R. Lundgren, and J. Johnson. 1978. Specrophotofluorometric and
electron microscopic study of lipofuscin accumulation in the testis of aging mice.
J. Gerontol, 33: 5-19.
Mullin, M. M., and E. R. Brooks. 1988. Extractable lipofuscin in larval marine fish.
Fish. BulL,U.S. 86: 407-415.
Munnel, J., and R. Getty. 1968. Rate of accumulation of cardiac lipofuscin in the aging
canine. J. Gerontol, 23: 154-158.
Nicol, S. 1987. Some limitations on the use of the lipofuscin ageing technique. Mar.
BioI. 93: 609-614.
Nicol, S., M. Stolp, and G. W. Hosie. 1991. Accumulation of fluorescent age pigments
in a laboratory population of Antarctic krill Euphausia superba Dana. 1.Exp. Mar.
BioI. £Col. 146: 153-161.
Packer, L., D. W. Deamer, and R. L. Heath. 1967. Regulation and deterioration of
structure in membranes. pp. 77-120 in Strehler, B. L. (ed.) Advances in
Gerontological Research. Academic Press, N.Y.
Papafrangos, E. D., and C. P. Lyman. 1982. Lipofuscin accumulation and hibernation in
Turkish hamster, Mesocricetus brandti. J. Gerontol, 37: 417-421.
135
Porta, E. A. 1991. Advances in age pigment research. Arch. Gerontol. Geriatr. 12: 303
320.
Rao, G., E. Xia, and A. Richardson. 1990. Effect of dietary restriction and aging on the
expression of antioxidant enzymes. pp. 391-404 in Finch, C. E., and T. E.
Johnson (eds.). Molecular Biology of Aging. Alan R. Liss, Inc. N.Y.
Rhys Williams, A. T., S. A. Winfield, and J. N. Miller. 1983. Relative fluorescence
quantum yields using a computer-controlled luminescence spectrometer. Analyst
108: 1067-1071.
SAS (Statistical Analysis System). 1985. Version 6. A series of user's manuals,
statistical procedures and software for the personal computer. SAS Inst. Inc.,
Cary, N.C.
Saunder, R. L. 1963. Respiration of the Atlantic cod. J. Fish. Res. Bd. Can. 20: 373
386.
Schmidt-Nielson, K. 1984. Scaling: why is animal size so important? Cambridge:
Cambridge University Press.
Semsei, I., G. Rao, and A. Richardson. 1989. Changes in the expression of superoxide
dismutase and catalase as a function of age and dietary restriction. Biochem.
Biophys. Res. Comm. 164: 620-625.
Sheehy, M. R. J. 1990a. The potential of morphological lipofuscin age pigment as an
index of crustacean age. Mar. BioI. 107: 439-442.
Sheehy, M. R. J. 1990b. Individual variation in, and the effect of rearing temperature and
body size on, the concentration of fluorescent morphological lipofuscin in the
brains of freshwater crayfish, Cherax cuspidatus (Crustacea: Parastacidae).
Compo Biochem. Physioi. 96A: 281-286.
136
._--_._----
Sheehy, M. R. J., and B. E. Roberts. 1991. An alternative explanation for anomalies in
"soluble lipofuscin" fluorescence data from insects, crustaceans, and other aquatic
species. Exp. Gerontol. 26: 495-509.
Sheldahl, J. A., and A. L. Tappel. 1974. Fluorescent products from aging Drosophila
melano~aster:An indicator of free radical lipid peroxidation damage. Exp.
Gerontol. 9: 33-41.
Shimasaki, R., T. Nozawa, O. S. Privett, and W. R. Anderson. 1977. Detection of age
related fluorescent substances in rat tissues. Arch. Biochem. Biophys. 183: 443
451.
Shimasaki, R., N. Ueta, and O. S. Privett. 1980. Isolation and analysis of age-related
fluorescent substances in rat testes. Lipids 15: 236-241.
Siakotos, A. N., and N. Koppang. 1973. Procedures for the isolation of lipopigments
from brain, heart and liver, and their properties: A review. Mech. Age. Dev. 2:
177-200.
Sohal, R. S. 1981. Relationship between metabolic rate, lipofuscin accumulation and
lysosomal enzyme activity during aging in the adult housefly, Musca domestica.
Exp. Gerontol. 16: 347-355.
Sohal, R. S. 1984. Metabolic rate, free radicals, and aging. pp. 119-127 in Armstrong,
D., R. S. Sohal, R. G. Cutler, and T. F. Slater (eds.) Free Radicals in Molecular
Biology, Aging, and Disease. Raven Press, N.Y.
Sohal, R. S., and U. T. Brunk. 1989. Lipofuscin as an indicator of oxidative stress and
aging. pp. 17-29 in Porta, E. A. (ed.) Lipofuscin and Ceroid Pigments. Advances
in Experimental Medicine and Biology. Plenum Press, N.Y.
Sohal, R. S., and P. B. Buchan. 1981. Relationship between fluorescent age pigment,
physiological age and physical activity in the housefly, Musca domestica Mech.
.Age. Dev. 15: 243-249.
137
_..__.._------_ .... _---_ ..__ ..
Sohal, R. S., and H. Donato Jr. 1979. Effect of experimental prolongation of life span
on lipofuscin content and lysosomal enzyme activity in the brain of the housefly,
Musca domestica. J. GerontoI. 34: 489-496.
Sohal, R. S., M. R. Marzabadi, D. Galarls, and U. T. Brunk. 1989. Effect of oxygen
concentration on lipofuscin accumulation in cultured rat heart myocytes - a novel
in vitro model of lipofuscinogenesis. Free Rad. BioI. Med. 6: 23-30.
Sohal, R. S., B. H. Sohal, and U. T. Brunk. 1990. Relationship between antioxidant
defenses and longevity in different mammalian species. Mech. Age. Dev. 53:
217-227.
Sohal, R. S., and L. S. Wolfe. 1986. Lipofuscin: characteristics and significance. pp.
171-183 in Swaab, D. F., E. Fliers, M. Mirmiran, W. A. Van GooI, and F. Van
Haaren (eds.) Prog. Brain Res.. Elsevier Science Publishers, Amsterdam.
Strehler, B. L., D. D. Mark, A. S. Mildvan, and M. V. Gee. 1959. Rate and magnitude
of age pigment accumulation in the human myocardium. J. GerontoI. 14: 430
439.
Tappel, A. L., B. Fletcher, and D. Deamer. 1973. Effect of antioxidants and nutrients on
lipid peroxidation fluorescent products and aging parameters in the mouse. J.
Gerontol. 28: 415-424.
Tappel, A. L. 1975. Lipid peroxidation and fluorescent molecular damage to membranes.
pp. 145-170 in Trump, B. F., and A. V. Arstila (eds.) Pathobiology of Cell
Membranes. Vol. I. Academic Press, N.Y.
Taylor, C. R. 1987. Structural and functional limits to oxidative metabolism: Insights
from scaling. Ann. Rev. Physiol. 49: 135-146.
Tsuchida, M., T. Miura, and K. Aibara. 1987. Lipofuscin and lipofuscin-like
substances. Chem. Phys. Lip. 44: 297-325.
138
Vemet, M., J. R. Hunter, and R. D. Vetter. 1988. Accumulation of age pigments
(lipofuscin) in two cold-water fishes. Fish. Bull.,U.S. 86: 401-407.
Weinclruch, R. 1984. Dietary restriction and the aging process. pp. 181-202 in
Armstrong, D., R. S. Sohal, R. G. Cutler, and T. F. Slater (OOs.) Free Radicals
in Molecular Biology, Aging, and Disease. Raven Press, N.Y.
Wieser, W. 1984. A distinction must be made between the ontogeny and the phylogeny
of metabolism in order to understand the mass exponent of energy metabolism.
Respir. Physiol. 55: 1-9.
Woodhead, A. D. 1980. Fish in studies of aging. Exp. Gerontol. 13: 124-140.
Yin, D., and U. T. Brunk. 1991. Microfluorometric and fluorometric lipofuscin spectral
discrepancies: a concentration-dependent metachromatic effect? Mech. Age. Dev.
59: 95-109.
Ziegler, I., and R. Harmsen. 1969. The biology of pteridines in insects. pp. 139-203 in
Beament, J. W. L., J. E. Treherne, and V. B. Wigglesworth (eds.). Advances in
Insect Physiology. Academic Press, Inc., London.
139
--------------- --------_.-..__ .....
Table 4-1. Results of the ANCOVA model testing the effects of fish weight (ascovariate), temperature (Temp), photoperiod (Phot), and clutch (Clutch) on polar FAPextracted from Puntius conchonius brain tissues. FAP data were log-transformed tostabilize variances. Model r 2 (0.895) significant at P < 0.0001.
Dependentvariable Source df SS MS F p
Brain polar FAP Fish weight 1 0.0002 0.0002 0.04 0.8434
Temp 2 0.8879 0.4440 96.54 <0.0001
Phot 1 0.0017 0.0017 0.37 0.5444
Clutch 2 1.8758 0.9379 203.96 <0.0001
TempxPhot 2 0.0155 0.0078 1.69 0.1869
Temp x Clutch 4 2.7213 0.6803 147.96 <0.0001
Phot x Clutch 2 0.0728 0.0364 7.92 0.0005
Temp x Phot x Clutch 4 0.0171 0.0043 0.93 0.4881
Error 219 1.0071 0.0046
140
Table 4-2. Results of the ANCOVA model testing the effects of fish weight (ascovariate), temperature (Temp), photoperiod (Phot), and clutch (Clutch) on polar FAPextracted from Puntius conchonius hean tissues. FAP data were log-transformed tostabilize variances. Model r 2 (0.532) significant at P < 0.0001.
Dependentvariable Source df SS MS F P
Heart polar FAP Fish weight 1 0.3623 0.3623 30.60 <0.0001
Temp 2 0.4747 0.2373 20.05 <0.0001
Phot 1 0.0001 0.0001 0.01 0.9202
Clutch 2 0.2909 0.1455 12.29 <0.0001
TempxPhot 2 0.0844 0.0422 3.56 0.0300
Temp x Clutch 4 0.2870 0.0718 6.06 <0.0001
Phot x Clutch 2 0.0066 0.0033 0.28 0.7580
Temp x Phot x Clutch 4 0.1864 0.0466 3.94 0.0042
Error 219 2.5926 0.0118
141
--_.._---_ __ _ .
Fig. 4-1. - Puntius conchonius. Mean fish weight ±95% confidence limits at 280 to 290
d of age as affected by experimental treatments: (A) temperature/ration experiment with
low and high rations at each temperature (QC), and (B) temperature/photoperiod
experiment with 6 hand 18 h light (6L and 18L) per 24 h period at each temperature eC).
142
6L 18L31°
6L 18L25°
Treatment
• Clutch 1~ Clutch 2
II1iI Clutch 34
2
o
3
5 A
1
3
2
...-..C)
1'-"+-'..cC) 0-- Low High Low High Low HighQ)
~ 19o 25° 31°5 8 • Clutch 4..c
(/) ~ Clutch 5--LL 4 II Clutch 6
143
Fig. 4-2. - Puntius conchonius. Effects of temperature eC) and ration level (low and
high) on brain FAP (%FL) soluble in (A) polar and (B) nonpolar solvent fractions. Bars
represent means ± 95% confidence limits for Clutches 1, 2, and 3.
144
600 • Clutch 1 A-....J ~ Clutch 2LL 500~ II Clutch 30-o, 400«LL
300"-ro-0 200a.c
"(ij 100"-
CD0
Low High Low High Low High19' 25° 31°
- 80 B....JLL~0- 60o,<LL"-
40ro0a.c0 20ccOro"-co 0
Low High Low High Low High19' 25° 31°
Treatment
145
Fig. 4-3. - Puntius conchonius. Brain polar FAP (%FL) as a function of fish weight for
the temperature/ration experiment, Clutch 1 animals. Solid and open circles represent
fish fed low and high ration levels, respectively.
146
147
Fig. 4-4. - Pontius conchonius. Effects of temperature (0C) and ration level (low and
high) on heart FAP (%FL) soluble in (A) polar and (B) nonpolar solvent fractions. Bars
represent means ±95% confidence limits for Clutches 1, 2, and 3.
148
.-- ._-_ .....__ ..•.._.....
14 A---I 12u,"::R. 100-o,« 8LL~ 6CO-0c.. 41::CO
2Q)
:c0
Low High Low High Low High1~ 25° 31 0
22- B • Clutch 1--I 20u, FIa Clutch 2"::R. 180II Clutch 3- 16o,
« 14u,
12~
CO 10-0c.. 8c:
60c:
41::eu 2Q):c 0
Low High Low High Low High1~ 25° 31°
Treatment
149
Fig. 4-5. - Puntius conchonius. Heart total FAP (%FL) as a function of fish weight for
the temperature/ration experiment, Clutch 1 animals. Solid and open circles represent
fish fed low and high ration levels, respectively.
150
35A 19°C
30 • o High Ration
25 • Low Ration
20 • 0..,..15 ••• • 0 010 ...-. ~ 005 ••• 0 0000 0
CDO 0
0
- 35 8 25°C.....Ju, 0
~30 • 0
0 0 0- 25 0
a. 0 0
« • 0<020 • 0u, 0 0
ctS 15 .: 0cP 0
+-' • 0 00 10 •••• 0+-'
.1 G--.t::= 0
ctS 5 •Q)
I 0
35 C 31°C30 0
25 00
020 0 0
0 0
1500 o 0
• 0 o 0 0
• 010 • • °r) •5 • • •
00 2 3 4 5
Fish weight (9)
151
Fig. 4-6. - Puntius conchonius. Weight-specific rate of oxygen consumption as a
function of fish weight and ambient temperature. Data are means ± 95% confidence
intervals for triplicate determinations on individuals subsampled from Clutch 3,
temperature/ration experiment.
152
.....................:~.........•...... ..·..·..·..·..·..·..·..·!···I 31°I'·'.".{.••......·····
1'119°
3.5
-.c:~
3.0..c.2>Q)
3:.c: 2.5en'+=~0>
2.0E-c:0
:;::::::1.5a.
E::senc: 1.00oc:Q)0> 0.5>0-x0
4.03.51.5 2.0 2.5 3.0
Fish weight (g)1.0
0.0 -I-----..---.-------.-----r---....----.......,.....----.0.5
153
Fig. 4-7. - Puntius conchonius. Effects of temperature (0C) and photoperiod (6L and
18L) on brain FAP (%FL) soluble in (A) polar and (B) nonpolar solvent fractions. Bars
represent means ± 95% confidence limits for Clutches 4, 5, and 6.
154
-_._---------- -------
400 A• Clutch 4--J ~ Clutch 5LL~ 300 II Clutch 60-a..«LL 200~
ctS0a.
100c:"ffi~
CD0
6L 18L 6L 18L 6L 18L1S' 25° 31°
12 B--JLL 10~0-a.. 8<LL~ 6ctS0a. 4c:0c:c: 2
"as~
CD 06L 18L 6L 18L 6L 18L
1S' 25° 31°Treatment
155
Fig. 4-8. - Fundus conchonius. Effects of temperature (0C) and photoperiod (6L and
18L) on heart FAP (%FL) soluble in (A) polar and (B) nonpolar solvent fractions. Bars
represent means ± 95% confidence limits for Clutches 4, 5, and 6.
156
18 A • Clutch 4- 16....J~ Clutch 5LL.
~ 14 II Clutch 60-o, 12
~ 10~
8ctS0 6a.t 4ctSQ)
2J:0
6L 18L 6L 18L 6L 18L1~ 25° 31°
- B....J 6LL.~0 5-o,« 4LL~
ctS 3-0a.c: 20c:t 1ctSQ)
:::c 06L 18L 6L 18L 6L 18L
19" 25° 31°Treatment
157
-----------
CHAPTER 5
SUMMARY AND CONCLUSIONS
This dissertation has examinedthree important aspectsof FAP complexes formed
in postrnitotic tissuesof aquaticpoikilothenns. It providesnew insights relative to
applications in fisheries biology and biomedical gerontology. The primaryobjective of
thisresearch was to critically evaluate the potential for usingFAP methodologies for
accurate estimation of age in fish (andultimately in otheraquaticanimals suchas
mollusksand crustaceans). This goal was accomplished by 1) reassessing various
aspectsof methodologies used to quantifyFAP in animal tissues; 2) establishing the
natureof the relationship between measured levels of FAP and chronological age in two
fish species,and 3) examiningthe effects of several environmental and physiological
variables that can alter the rate ofFAP accumulation and inducevariability in the
relationship to chronological age.
In the first part of this research (Chapter2), it was found that the temperature and
time involvedin storage, incubationand analysis of tissuesand their solventextracts can
criticallyaffectFAP expression throughthe in vitroproduction of additionalFAP-like
fluorophores in tissuesof Oreochromismossambicus. Suchproduction was drastically
increasedif samplehomogenates were ultrasonicated. The finding that FAP-like
fluorophores can form in dead tissuesor samplesolvent is not surprisingin view of the
biochemical processesleading to fluorophore formation. The complex and unstable
biochemical composition of dead tissuesand theirextracts predisposes themto lipid
peroxidation, malondialdehyde production,and fluorophore formation as describedin
Chapters 1 and2. Ulttasonication, commonly applied to disintegratecells, generates free
radicals in solution and expedites these reactions. The fmdingsdescribed in Chapter2
can have profoundimplications in termsof the sources of fluorescencedescribed in all
158
previous FAP studies. In most FAP studieson aquatic animals (Table 2-1), some or all
of these critical sourcesof error have been ignored. This has probably led to artifactual
fluorescenceand high variability in the results of some of these early studies, as well as
erroneous conclusionsas to the usefulnessof the technique. My study did not attempt to
address such problems for the hundreds of pathologicaland gerontological studies
employing FAP methods over the past twenty years, but similaroversights have probably
occurred.
An additionalproblem, common to most of the preliminarystudies of FAP in
aquatic species, has been the use of animals whose chronologicalage was estimated, but
not validated by other methods. In some cases, age was not even estimated, and FAP
content was only compared with body size. Several studieshave attempted to examine
FAP accumulation in whole animalswithout regard for differential rates of accumulation
in different tissue types. The sampling frequency and time scale of many studies have
been inadequatefor accuratelydescribingontogeneticpatternsof FAP accumulationin the
study animals.
In the second phase of this dissertationresearch (Chapter3), the above problems
were addressedby examining patterns of FAP accumulation for extended periodsof time
in specimens of two fish species of known chronologicalage. Nonpolar FAP
accumulated in brain tissues of Oreochromismossambicusduring the first 2.5 years of
life, and it is assumed that accumulationwouldcontinue throughthe latter half of the
lifespan as well. FAP also accumulated in brain and heart tissues of Funtius conchonius
over the first 1.5 years, and the highest values in both tissues were found in isolated
measurementsof the oldest animals sampled (> 4 yearsold). Patterns of accumulation
were different in the two tissue types over the first 1.5 years; brain tissues had greater
and more variablequantitiesof FAP than heart tissues. Brain FAP tended to level off
with increasingage, whereas heart FAP increased in a more linear pattern. Reasons for
159
these differences are not clear; it is assumed that the rates of accumulation are affected by
ontogenetic changes in tissue metabolism as well as changes in the activity of the
antioxidant enzymes responsible for the control of free radicals. Differences were also
found in the relative solubilities of FAP in polar and nonpolar solvents within and
between tissue types, emphasizing the importance of examining both fractions whenever
possible.
This is the first study to incorporate FAP data into multiple regression models for
prediction of chronological age. Of the several types of multiple regression models
developed in Chapter 3, the best models for age prediction in both species were those that
included all types of data (body size, otolith size, organ weight, and FAP). However,
these best models provided only slightly stronger age prediction than similar models
constructed without FAP data. This result raises the question of whether obtaining the
additional information available from FAP measurement would be a cost effective
addition to gathering the fish size, otolith size, and organ weight data; however, a major
goal of FAP research is to use FAP to age animals lacking skeletal structures such as
otoliths. When otolith data were excluded from the multiple regression models in this
study, inclusion ofFAP still provided a better age-predictive model than somatic data
alone. This was especially evident for male P. conchonius, for which organ weight and
FAP data were the only significant variables remaining in the final fitted regression
model. The relationship between FAP content and chronological age was variable in this
study, but FAP variables consistently remained in the [mal multiple regression models.
The animals used for these multiple regression studies were reared under closely similar,
strictly controlled conditions in the laboratory, so the resulting data represent a near
optimum which is probably unattainable with field specimens.
The final goal of this dissertation was to determine the possible effects of some
environmental (temperature, photoperiod) and physiological (body weight, weight-
160
specific metabolic rate, ration level, genome) variables on FAP accumulation in brain and
heart tissues of P. conchonius (Chapter 4). These effects were tested in two experiments:
one tested for effects of temperature, ration level, and clutch; the other tested for effects
of temperature, photoperiod, and clutch. Temperature and body weight affected FAP
content in brain and heart in both experiments, but in a different manner in each organ.
Brain FAP was inversely related to body weight in both experiments, especially at the
lowest rearing temperature (19°C). Heart FAP was inversely related to body weight at
19°C, but positively related to weight at higher temperatures. Increasing temperature
decreased brain FAP in both experiments, but increased heart FAP content Ration level
appeared to affect FAP level by modulating body size, however statistical analyses and
separation of possible direct effects were not possible. Photoperiod had no significant
effect on FAP in either tissue. Significant differences in FAP were detected between
non-sibling fish, implicating genomic variation in the ability to control FAP genesis. The
mixed effects of body size, temperature, and weight-specific oxygen consumption for
each tissue indicated involvement of factors in addition to metabolic rate in the formation
of FAP products. Homeothermic adaptation via modification of cellular lipid constituents
may predispose lipid-rich brain tissues to lipid peroxidation at lower temperatures. In
contrast, heart tissues, which are lower in lipid content, are more susceptible to the
positive effects of temperature on oxygen consumption, and thus have higher FAP values
at higher temperatures. This is the first study to demonstrate these environmental and
physiological effects on FAP accumulation in a vertebrate poikilotherm.
Biologists hoping to apply FAP methodologies for age estimation offield
populations will be faced with a number of serious complications. The ability to apply
FAP as an aging tool will ultimately be contingent on obtaining experimental baseline
information on age, growth, and FAP accumulation patterns which can be related to
animals collected from the natural populations. Measuring FAP fluorescence in a tissue
161
-------_.-- -
is a relatively simple procedure to perform, however interpretation of the data for age
prediction is not nearly as straightforward a procedure as counting growth increments in
skeletal parts. Age pigment content of a tissue, whether measured as extractable FAP or
quantified using histological! morphometric methods, will always be dependent upon the
physiological and environmental life history of the animal. Physiological aging
processes are correlated only roughly with chronological age. Thus, investigators
wishing to apply FAP methods to natural populations will be required to conduct baseline
research defining the relationship between FAP accumulation and chronological age in
combined laboratory growth and mark-recapture studies, through relevant portions of the
animal's lifespan. Laboratory growth studies should take into account variation in FAP
induced by temperature, body size, nutrition and genetic factors, which were shown to be
important in this study. Other factors that may be important to consider (but not studied
here) include disease history and exposure to salinity fluctuations and radiation,
xenobiotics such as heavy metals, or other environmental extremes. Based on the amount
of variability and the complexity of the interactions encountered in this study, successful
estimation of age of wild specimens using FAP is likely to be difficult
162