effects of dietary restriction and antioxidants on presbyacusis

The LaryngoscopeLippincott Williams & Wilkins, Inc., Philadelphia© 2000 The American Laryngological,Rhinological and Otological Society, Inc.

Effects of Dietary Restriction andAntioxidants on Presbyacusis

Michael D. Seidman, MD

Objectives/Hypothesis: The premise of this studyis that the membrane hypothesis of aging, also knownas the mitochondrial clock theory of aging, is the ba-sis for presbyacusis. Furthermore, it is proposed thattreatment with antioxidants or dietary restrictioncan attenuate age-related hearing loss. Many studieshave demonstrated a reduction in blood flow to spe-cific tissues, including the cochlea, with aging. Hypo-perfusion leads to the formation of reactive oxygenmetabolites (ROM). ROM are highly toxic moleculesthat directly affect tissues including inner ear struc-tures. In addition, ROM can damage mitochondrialDNA (mtDNA), resulting in the production of specificmtDNA deletions (mtDNA del4977 [human] or mtDNAdel4834 [rat]; also known as the common aging dele-tion]. Previous corroborating data suggest that thecommon aging deletion mtDNA4834 may be associatednot only with aging but also with presbyacusis, thusfurther strengthening the basis of the current studies.In this study, experiments provide compelling evi-dence that long-term treatment with compounds thatblock or scavenge reactive oxygen metabolites atten-uate age-related hearing loss and reduce the impactof associated deleterious changes at the molecularlevel. Study Design: Prospective randomized study.Methods: One hundred thirty rats were randomly as-signed to one of six groups with appropriate controls.Animals were divided into the following treatmentarms: group 1, 30% caloric restriction; group 2, vita-min E oversupplementation; group 3, vitamin C over-supplementation; group 4, melatonin treatment;group 5, lazaroid treatment; and group 6, placebo. Inaddition, 10 animals were used to determine the ap-propriate caloric restriction. All subjects underwentbaseline and every-3-month testing until their health

failed (range, 18–28 mo; average, 25 mo). This testingincluded auditory sensitivity studies using auditorybrainstem response (ABR) testing, as well as tissueanalysis for mtDNA deletions using molecular biolog-ical techniques. At the conclusion of the study, ani-mals underwent a final ABR test and were tested formtDNA deletions in brain and inner ear tissues, andthe opposite ear was used for histological analysis.Results: Results indicated that the 30%-caloric-restricted group maintained the most acute auditorysensitivities, the lowest quantity of mtDNA deletions,and the least amount of outer hair cell loss. Theantioxidant-treated subjects had improved auditorysensitivities, and a trend for fewer mtDNA deletionswas observed compared with the placebo subjects.The placebo subjects had the poorest auditory sensi-tivity, the most mtDNA deletions, and the greatestdegree of outer hair cell loss. Conclusions: Interven-tion designed to reduce reactive oxygen metabolitedamage appears to protect against age-related hear-ing loss specifically and aging in general. This is re-flected by an overall reduction in mtDNA deletions.These data also suggest that the common aging dele-tion appears to be associated with presbyacusis, asdemonstrated by an increased frequency of themtDNA del4834 in the cochleae with the most signifi-cant hearing loss. Nutritional and pharmacologicalstrategies may very well provide rational treatmentoptions that would limit the age-associated increasein ROM generation, reduce mtDNA damage, and re-duce the degree of hearing loss as the organism ad-vances in age. Key Words: Antioxidants, reactive oxy-gen metabolites, anti-aging, presbyacusis.

Laryngoscope, 110:727–738, 2000

INTRODUCTIONThe premise of this study is that the membrane hy-

pothesis of aging, also known as the mitochondrial clocktheory of aging, is the basis for presbyacusis. This studyexpands on the fact that aged animals have a reduction inblood supply to the cochlea. It is hypothesized that thisgeneralized hypoperfusion leads to the generation of reac-tive oxygen metabolites (ROM). The ROM damage mito-chondria, manifesting as a reduction in mitochondrialmembrane potentials and an increase in mitochondrial

Presented as a Candidate’s Thesis to the American Laryngological,Rhinological and Otological Society, Inc. Recipient of the Fowler Award.

From the Department of Otolaryngology—Head and Neck Surgery,Henry Ford Health System, West Bloomfield, Michigan.

Supported by NIDCD grant No. DC00101-01-05.Editor’s Note: This Manuscript was accepted for publication January

5, 2000.Send Correspondence to Michael D. Seidman, MD, Department of

Otolaryngology—Head and Neck Surgery, Henry Ford Health System,6777 West Maple Road, West Bloomfield, MI 48323, U.S.A.

Laryngoscope 110: May 2000 Seidman: Presbyacusis

727

DNA (mtDNA) deletions with concomitant hearing loss.This study also demonstrate that long-term treatmentwith compounds that either block or scavenge reactiveoxygen metabolites have a protective effect on age-relatedhearing loss and the associated alterations at the genomiclevel.

This study investigates new hypotheses designed toenhance the understanding of the molecular and biologicalmechanisms of presbyacusis. The primary purpose ofthese experiments was to determine whether the mem-brane hypothesis of aging has a fundamental role in age-associated hearing loss. Furthermore, these experimentswere designed to determine whether diet restriction andtreatment with compounds known to reduce damage fromROM will affect the aging process in general, and presby-acusis in particular.

The process of aging is associated with many molec-ular, biochemical, and physiological changes including in-creases in DNA damage, reduction in mitochondrial func-tion, decreases in cellular water concentration, ionicchanges, vascular insufficiencies, and decreased elasticityof cellular membranes. One contributing factor to thisprocess is altered vascular characteristics, such as re-duced erythrocyte velocity and vascular plasticity, as wellas increased vascular permeability.1,2 These age-relatedchanges result in reductions in oxygen and nutrient deliv-ery, as well as waste elimination.3,4 Such functional alter-ations favor the production of ROM. In addition, there issupport in the literature for an age-associated reduction inendogenous enzymes that protect from ROM damage, in-cluding superoxide dismutase, catalase, and glutathioneperoxidase.5 Collectively, these changes enhance the gen-eration of ROM. ROM are responsible for producing mito-chondrial DNA damage by causing mutations in the mi-tochondrial genome. Specific mutations are known to bedirectly proportional to aging, such as the common agingdeletion in humans, which is 4,977 base pairs (bp) inlength, and in rats (4,834 bp). When mtDNA deletionsreach a certain level, the mitochondria become bioener-getically inefficient. Investigators have shown that agedrats, monkeys, and humans have reductions in mitochon-drial function and increases in mtDNA deletions in heart,brain, liver, and skeletal muscle.6–8 To develop the basisfor this study, the following topics are discussed: presby-acusis, reactive oxygen metabolites, mitochondrial func-tion and mitochondrial DNA deletions, aging, and antioxi-dants.

HISTORICAL PERSPECTIVES AND REVIEW OFTHE LITERATURE

Presbyacusis“Biological aging appears to be a physiological

process engendered by the genotype and the adap-tive norm of the species. Superimposed on the inher-ited processes seems to be the gradual accumulationof errors in DNA, which is probably related to adecline in normal mechanisms of repair.”

Dr. Schuknecht displayed tremendous insight whenhe wrote these lines some two decades ago. That the

concepts he had outlined remain relevant today is anaffirmation of his stature as a scientist and clinician.Aging research has fortified these precepts as an accurateand meaningful perspective on senescence and its manymanifestations, such as presbyacusis.

Presbyacusis is characterized by the progressive de-terioration of hearing associated with aging and is themost common cause of adult auditory deficiency in theUnited States, affecting approximately 23% of the popu-lation between 65 and 75 years of age and 40% of thepopulation older than 75 years of age.9 Thus the socioeco-nomic ramifications are staggering. While presbyacusis iscommon in industrialized societies, age-related hearingloss is less pronounced in other societies. This discrepancyhas been attributed to many factors, including genetics,diet, and socioeconomic and environmental factors.10,11

Reactive Oxygen Metabolites or Free OxygenRadicals

Reactive oxygen metabolites are species that containan unpaired number of electrons making them chemicallyreactive and extremely toxic to cellular and subcellularstructures. It has been speculated that ROM are involvedin more than 100 clinical conditions.12 They are producedin vivo during mitochondrial respiration, as well as viaauto-oxidation of chemical and biological molecules. ROMare also environmental contaminants, can be formed fromionizing and ultraviolet radiation, and occur in response tohypoperfusion or ischemia followed by reperfusion.

There is extensive support in the literature for theprotective effects of antioxidants and reactive oxygen me-tabolite scavengers. For example, tocopherols (vitamin E)decrease atherosclerosis and delay death from myocardialinfarction, presumably by inhibiting lipid peroxidation.13

Carotenoids, such as b-carotene and other plant pigments,may also have preventive effects against cancer and car-diovascular disease.14,15 Available data also suggest thatmany of the pathological correlates of Alzheimer’s diseaseare precipitated by free radical–induced and oxidativestress–induced mechanisms.16 Similarly, Parkinson’s dis-ease has also been associated with oxidative stress, in-creased lipid peroxidation, reduced levels of glutathione,high concentrations of iron, and free radical generationvia autocatalytic mechanisms within neuromelanin-containing catecholaminergic neurons.17

Experimentally and clinically, it is well known that aprimary source of ROM generation is through oxidativephosphorylation, ischemia/reperfusion, or prolonged hypo-perfusion, such as is seen in myocardial infarction, cere-brovascular accidents, aging, and, possibly, sudden senso-rineural hearing loss and presbyacusis. It is clear that inthe aging cochlea there is a significant reduction of bloodsupply2 and the ongoing need for energy generationthrough oxidative phosphorylation. Thus these two pro-cesses allow for the generation of ROM within the cochlea.

Vascular Changes Associated With AgingThe aging process is clearly linked with alterations in

circulatory function. Studies have demonstrated markedlydecreased flow within the circulatory system in the elderlypopulation.3,4 Prolonged periods of reduced blood flow


728

(hypoperfusion) may lead to decreased oxygen tensionswithin tissues culminating in the generation of damagingROM.

Various lines of research have centered on the con-cept that altered blood flow and/or oxygen delivery resultsin inner ear hair cell damage and hearing loss. In a humancadaveric study of cochlear vessels, an age-related, grad-ual loss of capillaries in the spiral ligament of the scalavestibuli was observed, with an increase in intravascularstrands and avascular channels. Similar but less pro-nounced findings were noted in the spiral ligament in thescala tympani.18 Another human cadaver study, using amicrosphere technique to quantify blood flow, found di-minished flow in morphometrically normal–appearingbasal turn capillaries.19 Changes in whole blood viscosityand red blood cell rigidity have been correlated with high-frequency hearing loss in elderly humans.20 Furthermore,a series of in vivo experiments using intravital microscopyof the cochlear microvasculature demonstrated age-dependent, statistically significant reductions in meanerythrocyte velocity and significant increases in capillarypermeability.2 In contrast, Axelsson21 noted no significantdifferences in cochlear blood vessels of young and oldguinea pigs. However, there were more failures in con-trast filling of the older animals’ vessels, because of morefrequent ruptures of the vessels.21 It is possible that thismay in part be due to increased vascular fragility that is acommon feature of aged vasculature, a speculation en-dorsed by Dr. Axelsson. Furthermore, these findings mayalso be secondary to increased vascular permeability,which is a common sequela of ischemic injury.

Effects of Hypoxia on the Inner EarThe effects of cochlear hypoxia on auditory function

have been extensively studied. There is a direct correla-tion between hypoxia and decreased hearing sensitivity.Studies inducing hypoxia by inspiration of 100% nitro-gen,22 tracheal clamping and cessation of respiration,23

and vessel occlusion24 have all shown significant alter-ations of auditory sensitivity as measured by compoundaction potentials, summating potentials, cochlear micro-phonics, and otoacoustic emissions.25,26

Decreased blood flow from a variety of causes includ-ing hypertension, atherosclerosis, noise exposure, or agingpromotes hypoxia with formation of ROM and resultanttissue/organ damage.27,28

Mitochondrial DNA DeletionsMitochondria are unique organelles possessing their

own DNA as well as their own enzymatic constituents toallow for transcription and translation of genetic informa-tion into proteins. Each mtDNA codes for a complete set ofribosomal (rRNA) and transfer RNA (tRNA). In addition,mtDNA codes for 13 of the approximately 60 polypeptidesnecessary for oxidative phosphorylation. These include 7of the 25 subunits of respiratory complex I (ND1–ND4, 4L,5, and 6) (reduced nicotinamide adenine dinucleotide[NADH]-ubiquinol-oxidoreductase), 1 of the approxi-mately 9 subunits (cyt b) of respiratory complex III(ubiquinol-cytochrome c oxidoreductase), 3 of the 13 sub-units of respiratory complex IV (COI, COII, and COIII)

(cytochrome c oxidase), and 2 of the 12 subunits of respi-ratory complex V (ATP6 and ATP8) (ATP synthase). Theremaining subunits of these complexes are encoded bynuclear DNA.29

It has been proposed that mitochondrial genomic mu-tations may be a major cause of human diseases. The firstpatient reported with a mtDNA deletion as a cause of illnesswas reported in 1959.30 Luft is often credited with demon-strating the importance of mitochondrial medicine.31 Mito-chondrial mutations followed by cytoplasmic segregation hasbeen shown to contribute to neuromuscular disorders suchas Kearns-Sayre/chronic external ophthalmoplegia plus syn-drome,32 mitochondrial encephalomyopathy, lactic acidosis,and stroke-like episodes (MELAS),33 subacute necrotizingencephalomyopathies, SNE (Leigh syndrome) and progres-sive neuronal degeneration of childhood (Alpers syn-drome),34 and myoclonic epilepsy associated with ragged-redmuscle fibers (MERRF).29 Interestingly, as many as 67% ofpatients with mtDNA disorders are known to manifest sen-sorineural hearing loss.35

Experiments performed since 1989 have demon-strated that mtDNA deletion with gene induction is alsoassociated with hypoxia. Corral-Debrinski et al.36 havefound a specific mtDNA deletion occurring at increasedfrequency in ischemic hearts. They concluded that isch-emic hearts have increased mtDNA damage and oxidativephosphorylation gene expression that may be associatedwith oxidative phosphorylation deficiency. In addition,there is evidence suggesting 5- to 12-fold elevations inmtDNA deletions in the brain associated with chronichypoxia.37

Deletions of mtDNA with subsequent segregationand enrichment of the deletions throughout life are con-tributing factors in the aging process.6 It has been shownthat mtDNA deletions in the human heart are age depen-dent.6 Mitochondrial mutations occur and accumulatecontinuously until death. Cells that accumulate largenumbers of mitochondrial mutations become bioenergeti-cally deficient, which explains their important role inmany human diseases.38

Mitochondrial and nuclear DNA are exposed to asignificant amount of oxidative damage. Studies haveshown that increasing amounts of ROM occur as a func-tion of age, which leads to an increase in membrane per-oxide content and rapidly exceeds the capacity of homeo-static protection.39 Thus there is extensive support thatmtDNA deletions accumulate with age and disease.40,41

Deafness has also been shown to have an associationwith mtDNA deletions. It has been suggested that mito-chondrial diseases should be considered in cases of pro-gressive sensorineural hearing loss, especially with thecoexistence of multisystem involvement.42 A 10.4-kbmtDNA deletion has been identified in association withmaternally transmitted diabetes and deafness, withoutophthalmoplegia or mitochondrial myopathies, which wasthe hallmark of mtDNA deletion syndromes.43 Other stud-ies have identified mutations in the tRNA(Leu)(UUR)gene in a large pedigree with maternally inherited dia-betes mellitus type II and deafness.44 A 3,243-point mu-tation (A3G) has been demonstrated in a patientwith sensorineural deafness without diabetes.45 Several


729

human studies have demonstrated an association of mito-chondrial DNA mutations and presbyacusis, including astudy demonstrating that older patients with presbyacu-sis had a higher frequency of the common aging deletion(4,977 bp) compared with similar-aged patients withoutpresbyacusis.46 A significant difference in mtDNA dele-tion was noted in human archival temporal bones. Specif-ically, 14 of 17 aged patients with presbyacusis had the4,977-bp deletion compared with 8 of 17 control patientswith normal hearing.47

AgingAging is the progressive accumulation of metabolic

and physiological changes associated with an increasingsusceptibility to disease. There are many hypotheses inthe current literature providing explanations for senes-cence. Three of the most convincing theories are the te-lomerase theory of aging, the “dysdifferentiation” hypoth-esis of aging, and the membrane hypothesis of aging (alsoreferred to as the mitochondrial clock theory of aging).

The telomerase theory of aging suggests that there isa reduction in telomere length over time. The end of achromosome is made up of a structure called the telosome.The tip of the telosome is a region of DNA repeat se-quences and associated proteins called the telomere. It ishypothesized that DNA transcription and replication areaffected by position effects mediated by the telomere. Re-duction in the length of the telomere and alterations in itschromatin assembly may explain the instability that oc-curs during senescence, as well as the immortalizationprocess in vitro.48 Although many aspects of telomeraseactivity remain undefined, it has been hypothesized thatthe balance between telomere shortening and telomeraseactivity underlies cellular aging processes.

The dysdifferentiation theory suggests that aging is acontinuum of programmed differentiation leading to ei-ther a cessation of normal gene activity or a systematicactivation of genes whose effects are deleterious to cellularfunction. Support for this theory is provided by apoptosis(programmed cell death) studies in hermaphroditicworms. These experiments elegantly elaborated the ge-netic mechanisms responsible for controlling cell death.The maintenance of homeostasis for cellular metabolismand function consumes a large fraction of total body en-ergy expenditure. This is engineered by the delicate bal-ance between cellular proliferation and death. For exam-ple, the Bcl-2 gene appears to prevent oxidative damage tocellular organelles and lipid membranes.49,50 Another pro-tein that appears to operate as an accessory to Bcl-2 is a21-kD protein referred to as Bax. The ratio between Bcl-2and Bax appears to determine survival or death followingan apoptotic stimulus. Specifically, elevated expression ofBcl-2 appears to be preventive, while that of Bax favorsthe apoptotic process.49

The membrane hypothesis of aging states that agingis related to decreasing effectiveness of cellular protectiveand reparative mechanisms. This yields biochemical andmetabolic errors, which progressively accumulate, result-ing in cell death.51 The hypothesis further postulates thatcellular senescence is attributable to the cross-linking ac-tion of ROM within the cellular membrane. In addition,

ROM lead to lipid peroxidation, polysaccharide depoly-merization, nucleic acid disruption, and oxidation of sulf-hydryl groups leading to enzyme inactivation.52 Thereforethe membrane hypothesis of aging suggests that ROM-induced cell membrane structural damage is the primarymediator in cellular aging.53

Careful analysis of the above mechanisms suggeststhat certain aspects of the three leading theories of agingmay be interrelated. That is, free radical species lead togenetic and cellular alterations resulting in cellular dys-function and, consequently, senescence. It is even moreintriguing to realize that a trigger for the Bax gene isROM.54 Thus a critical analysis of the prominent hypoth-eses of aging suggests that aspects of all three theories arelikely to apply. Specifically, the generation of ROM dam-ages cellular integrity, which may lead to alterations ingene expression, including telomere shortening and acti-vation of Bax genes resulting in aging, apoptosis, and,ultimately, death.

Primary Treatment InterventionsThe primary treatment interventions used in these

experiments were the following:Diet restriction. It is well known that a 30% caloric

dietary restriction can enhance longevity by 30% to 50%.The putative mechanism is a reduction in metabolism andoxidation reactions and hence, a significant reduction inthe generation of ROM.

Vitamin E. Vitamin E (tocopherol) is a potent fat-soluble antioxidant. The most active naturally occurringcompound is RRR-a-tocopherol. Polyunsaturated fatty ac-ids (PUFAs) are liable to auto-oxidation. Each fatty acidfree radical that is oxidized damages about three otherPUFA molecules, thus producing a chain reaction thatexpands geometrically. Vitamin E can readily donate hy-drogen atoms to terminate this chain reaction. The vita-min also appears to be essential for the protection ofcirculating lipoproteins and the correct functioning andstability of cell membranes.

Vitamin C. Vitamin C or ascorbic acid is a water-soluble vitamin. It cannot be endogenously synthesized incertain species, including humans; therefore its require-ment must entirely be met by dietary intake. It is aneffective scavenger for superoxide (O2

.-), thiol (sulfur-centered), singlet O2, and hydroxyl radicals. It has beenshown to recycle a-tocopherol in lipid membranes, thusdelaying the onset of peroxidation in isolated human low-density lipoproteins (LDLs).

Melatonin. Melatonin (N-acetyl-5-methoxytryptamine)is an indole hormone primarily secreted by the pineal gland.Although primarily known for its effect on entrainmentpathways of the biological clock and actions on photoperiodictime, melatonin is also an important neural antioxidant andfree radical scavenger. It is believed to work via electrondonation to directly detoxify free radicals such as the hy-droxyl radical. In addition, melatonin has been found toprotect cells (in vivo as well as in vitro) against oxidativedamage induced by a variety of free radical–generatingagents and processes. It is also effective in protecting nuclearDNA, membrane lipids, and, possibly, cytosolic proteinsfrom oxidative damage. In addition, evidence suggests that it


730

enhances the defense capacity of many antioxidative en-zymes such as superoxide dismutase, glutathione peroxi-dase, and nitric oxide synthase.

Lazaroids. Lazaroids are 21-aminosteroids with noglucocorticoid, mineralocorticoid, or other hormonal prop-erties. They are multimechanistic inhibitors of lipid per-oxidation. Lazaroids scavenge free radicals and stabilizecell membranes by decreasing fluidity, preserving vitaminE content in these membranes, and increasing surfaceviscosity. They have demonstrated efficacy in improvingneurological outcome after central nervous systemtrauma, subarachnoid hemorrhage, and ischemia.

MATERIALS AND METHODSFischer 344 rats were obtained from Harlan Inc. (Indianap-

olis, IN). A total of 130 2-month-old rats were used for theseexperiments. All procedures and protocols were reviewed andapproved by the institution’s Care for Experimental AnimalsCommittee. All animal experiments were performed in a humanemanner according to standards established by the National In-stitutes of Health (Bethesda, MD).

Subjects were maintained in individual cages at 21°C to22°C with a 12:12-hour light-dark cycle initiated at 7:00 A.M. Foodand water were available ad libitum to all groups with the excep-tion of diet-restricted subjects.

To determine the appropriate dietary restriction, a group of10 animals was originally studied for a period of 1 month. These10 animals were placed in metabolic cages where strict caloricintakes and urinary and stool outputs were recorded. The ani-mals ingested 16 g 6 3 g of a standard rodent diet formulation perday. Thus the appropriate caloric restriction was calculated to be11.2 g (30% restriction) of feed per day for the caloric-restrictedgroup.

The remaining 120 rats were randomly assigned to one ofthe following six groups: group 1: 30% caloric restriction (n 5 20);group 2: vitamin E oversupplementation (2.475 mg/d) (n 5 20);group 3: vitamin C oversupplementation (4.44 mg/d) (n 5 20); group4: melatonin treatment (0.1 mg/d) (n 5 20); group 5: lazaroid treat-ment (1.25 mg/d) (n 5 20); and group 6: placebo (matrix pill)(n 5 20).

The subjects were initially anesthetized using ketamine andxylazine (100 mg/kg and 15 mg/kg, respectively, intramuscularly)and supplemented with ketamine as required.

To simplify identification of each individual rat and to elim-inate the possibility of inadvertently misidentifying the subject,an electronic identification (ID) microchip (Avid, Norco, CA) wasinserted into each animal and read with a scanner before anytesting or treatment. In brief, the posterior nuchal region wasprepared and draped in a sterile manner. A 3-mm incision wasmade posteriorly, and the ID microchip inserted. The wound wasclosed with an interrupted 4-0 Vicryl suture.

After the microchip insertion, a separate incision was madeimmediately cephalad to the microchip incision and the appropri-ate pellet (per previous randomization) was inserted into thesubcutaneous tissues. This wound was closed with one 4-0 Vicrylsuture. The pellets were purchased from Innovative Research(Sarasota, FL). The pellets have been tested extensively to verifystated timed-release parameters, appropriate dosing, and appro-priate blood levels. The pellets selected for these experimentswere 3-month pellets and were guaranteed to provide basal levelsof the particular drug or vitamin throughout the entire timecourse.

Drug DosagesDosages of drugs used were as follows: vitamin E, one pellet

implanted with 222.75 mg vitamin E, each delivering 2.475 mg/d;

vitamin C, two pellets implanted with 200 mg vitamin C, eachdelivering 4.44 mg/d; melatonin, one pellet implanted with 9 mgmelatonin, each delivering 0.1 mg/d; and lazaroid, one pelletimplanted with 112.5 mg drug, each delivering 1.25 mg/d.

In addition, random assays were performed on animals todetermine blood levels of vitamin E and melatonin groups atthree separate time points during the life of the pellet. Thetechnique to perform the assays used high-pressure liquid chro-matography.

Auditory Brainstem Response MeasurementsAuditory brainstem response (ABR) measurements were

collected every 3 months from 2 months of age until the death ofthe animal. Each ear was studied for auditory sensitivity. Sub-jects were anesthetized using ketamine and Rompun (100 mg/kgand 15 mg/kg, respectively, intramuscularly) and supplementedwith ketamine as required. The subject’s temperature was main-tained at 38°C 6 1°C with a heating blanket and monitored witha rectal probe.

Signal generation. The test stimulus consisted of tone-bursts with a rise/fall time of 1 millisecond, a duration of 15milliseconds, and a period of 100 milliseconds. Intensity serieswere obtained at 3, 6, 9, 12, and 18 kHz, ranging from 0 dB soundpressure level (SPL) to 100 dB SPL. These stimuli were generatedusing a digital-to-analogue converter (model DA3–2, TuckerDavis Technologies, Gainesville, FL). The output of the digital-to-analogue converter was connected to a programmable attenu-ator (model PA4,Tucker Davis Technologies), a weighted summer(model SM3, Tucker Davis Technologies), a headphone buffer(model HB6, Tucker Davis Technologies), and an earphone (mod-el DT-48, Beyer Dynamic, Heilbronn, Germany) that was placedin close approximation to the tympanic membrane of the animal.A sampling frequency of 100 kHz was used to generate the signal.The signals were calibrated at the tympanic membrane using aprobe microphone (model ER-7C, Etymotic Research, Elk GroveVillage, IL). The output of the microphone was connected to ananalogue-to-digital converter (model AD2, Tucker Davis Technol-ogies) and a Pentium computer. An automatic calibration routinewas used for on-line calibration.

Recording. Sterile platinum needle electrodes (model E2,Grass Instruments, Quincy, MA) were placed under each pinnaand at the vertex of the subjects (under anesthesia). The elec-trodes were connected to a biological amplifier (model P5 series,Grass Instruments) with the gain setting at 3 100,000. Theresponse was filtered between 0.3 and 3 kHz. The output of theamplifier was connected to an analogue-to-digital converter (mod-el AD2, Tucker Davis) and a Pentium computer. The responseswere digitized with a sampling frequency of 50 kHz. For eachrecording, a 20-millisecond neural response was averaged 1,024times. Thresholds were identified when an N1 response 1 mVabove baseline was observed, using a standard protocol.

Isolation of Intact Mitochondria From MuscleSamples

Isolation of mitochondria from muscle was performed aspreviously described.55 In brief, the muscle specimen, weighingapproximately 1 g, was freed of any fat or connective tissue andrinsed in cold, nonionic isolation buffer containing 210 mmol/Lmannitol, 70 mmol/L sucrose, 1 mmol/L ethyleneglycol-bis-(beta-aminoethylether)-N,N,N9,N9-tetra-acetic acid (EGTA), 0.5% BSA,and 5 mmol/L N-2-hydroxyethylpiperazine-N-2-ethanesulfonicacid (HEPES), pH 7.2. The specimen was processed in approxi-mately 0.5-g batches with a Thomas (Thomas Scientific, Swedes-boro, NJ) tissue slicer. The thinly sliced muscle was collected intoa 50-mL tube, suspended in 10 mL isolation buffer/g muscletissue. The muscle tissue was homogenized using 10 passes in a


731

Teflon-glass homogenizer with moderate rotation speed. The re-sulting sample was centrifuged at 1500 3 g for 5 minutes at 4°C.The supernatant was decanted into a fresh tube, the pellet wasdiscarded, and the sample underwent repeat centrifugation. Thesupernatant was decanted and centrifuged at 8000 3 g for 15minutes. The mitochondrial pellet was suspended in 30 mL iso-lation buffer and centrifuged again at 8000 3 g for 15 minutes.The washed mitochondrial pellet was resuspended in 0.1 mL ofthe isolation buffer for each gram of muscle used. This samplecould then be stored at 270°C. For protein estimation, a 30-mLaliquot was removed and centrifuged at 8000 3 g for 15 minutes,and the pellet was resuspended in isolation buffer without BSA.

Tissue SourcesThe tissues harvested from the animal subjects included

brain, auditory nerve, and stria vascularis. All tissues were usedimmediately or stored at 270°C for later use.

DNA ExtractionAnimal tissues were harvested and stored at 270°C, until

the time of DNA extraction. The samples were homogenized in 10mmol/L Tris (pH 8) containing 1 mmol/L EDTA buffer and incu-bated overnight at 56°C with 15 mL Proteinase-K (10 mg/mL) in0.5 mL digestion buffer consisting of 10 mmol/L Tris (pH 8), 10mmol/L EDTA, 50 mmol/L NaCl, and 2% sodium-dodecyl sulfate.Standard extraction protocols for DNA were used with phenol,chloroform, and isoamyl alcohol. The proteins were removed fromthe sample solution with phenol and chloroform, (25:24) both ofwhich served as separate organic solvents and hence deprotein-ized more efficiently. The tissue extracts were centrifuged at10,000 3 g at room temperature to separate mtDNA from cellulardebris and protein. The supernatant was drawn off, and theresidual phenol removed with equal volumes of chloroform andisoamyl alcohol (24:1). This subsequent extraction with chloro-form removed the remaining traces of phenol from the prepara-tion. A 1/10 volume of 3 mol/L NaOAc and 1/100 volume of 1 mol/LMgCl2 were added, and mtDNA was recovered by precipitationwith 2 volumes of cold ethanol. This preparation was stored at270°C for 60 minutes, and the precipitate was recovered bycentrifugation at 12,000 rpm for 30 minutes (4°C). The superna-tant was removed, and the pellet washed with 70% ethanol,air-dried, and redissolved in Tris-Edta buffer at the desired con-centration. DNA concentrations were determined spectrophoto-metrically using optical densities of 260 and 280 nm, and aliquotswere used for polymerase chain reaction (PCR).

There are technical differences in isolating DNA from skel-etal muscle compared with other tissues. The methodology neces-sitates special attention for the isolation of tightly coupled mito-chondria. Our laboratory has used the modifications aspreviously described.56

Polymerase Chain ReactionOligonucleotide primers were designed in our laboratory

and synthesized by Fisher Biotech (Pittsburgh, PA) to amplifyseveral distinct regions of the rat, mouse, and human mtDNAgenome. Of particular interest is the cytochrome b gene, the ND1segment of the 16S rRNA region and a 4,834-bp deletion in rats.PCR is used to amplify the base-pair products that correspond tothese regions. Appropriate positive and negative controls wererun and then sequenced to authenticate the PCR products. ThePCR reaction contained 100 to 200 mg of test sample, 200 mmol/Lof each deoxynucleotide triphosphate, 50 mmol/L KCl, 10 mmol/LTris-HCl (pH 8.3), 1.5 mmol/L MgCl2, 0.01% (wt/vol) gelatin, 1mmol/L of each primer, and 5 U of Taq polymerase in a finalvolume of 100 mL. The thermal cycling parameters were as fol-lows: initial denaturation at 94°C for 3 minutes, followed by 30

cycles of denaturation at 94°C (for 30 s), annealing at 56°C (for30 s), and extension at 72°C (for 1 min).

Gel ElectrophoresisThe amplified PCR products were separated by electro-

phoresis on 1.5% agarose gel containing ethidium bromide. Thegels were read under ultraviolet light and imaged. Agarose gelconcentrations vary depending on fragment length. Thus forlarger fragments a 1% agarose gel was used.

DNA SequencingDetails of sequencing have been described previously.57 The

age and, possibly, presbyacusis-dependent mtDNA region wasselectively amplified, and the resulting fragment was gel purified.Asymmetric PCR was performed with primers at a 1:100 ratio. Adirect sequencing of the single-stranded mtDNA was made byasymmetric PCR using the “fmol” DNA Sequencing System (Pro-mega Co., Madison, WI).

HistologySubjects were deeply anesthetized. Cardiac puncture was

performed, and the animal was perfused with 2.5% glutaralde-hyde solution in a veronal acetate buffer freshly normalized to apH of 7.4. The bullae were rapidly removed, and the cochleaewere reperfused through the oval and round windows. The co-chleae were stored at 4°C for 48 hours, then perfused with 1%osmium tetroxide in veronal acetate buffer normalized to a pH of7.4 for 30 minutes. After perfusion, the organs were dehydratedusing ethanol of ascending strength from 30% to 70%. The co-chleae were decalcified overnight in 0.35 mol/L of EDTA in vero-nal acetate buffer normalized to a pH of 7.4 at room temperature.After decalcification, the organ of Corti was dissected in 70%ethanol. Each turn was mounted with glycerol and examinedunder an optical microscope. The hair cells were counted at orig-inal magnification 3 630, using a Carl Zeiss GFL (Carl Zeiss Inc.,Goettingen, Germany) differential interference contrast micro-scope. The data were used to generate cytocochleograms as afrequency position map based on the following mathematicalderivation:

F 5 A(10ax 2 k),

where F stands for frequency, A 5 0.64, a 5 2.1, x is expressed asa proportion of basilar length, and k 5 0.85.

Statistical MethodologyOutcome variables. The hearing levels for five frequencies

(3, 6, 9, 12, and 18 kHz) were tested separately.Multiple comparisons. A Bonferroni adjusted P value of

.01 (P 5 .05/5) was used to determine statistical significance foreach overall test for a treatment group effect for the five frequen-cies. If the overall test was not significant, tests between individ-ual treatment groups were not performed. If the overall test wassignificant, comparisons between treatment groups, two at atime, were performed and a P value of .05 or less was consideredstatistically significant.

Analysis methods. To compare the baseline, pretreatmentvalues, one-way ANOVA was performed. To make use of serialobservations, repeated-measures analysis was performed usingthe SAS PROC MIXED58 procedure. The model assumed a corre-lation between observations that decreased with the time (i.e., alower correlation between the observations at months 15 and 25compared with those between months 15 and 20 or betweenmonths 20 and 25). The PROC MIXED procedure allows subjectsto be included even if observations are not available for all times.


732

The model included an adjustment for the average levels ob-served at each month.

Four analyses were performed. These were restricted to alldata available at 25 months, all data from 20 months onward, alldata from 15 months onward, and all data from 12 months on-ward. The data from the oldest animals were of greatest interest,but earlier data were also included.

RESULTSThis randomized prospective study was completed

over a 3-year period. A total of 120 subjects were initiallytested, and all subjects were studied throughout their lifespan. At the conclusion of the experiments, there were fiveanimals available for analysis in the diet-restricted group(mean age, 27 mo), four in the vitamin E group (mean age,26 mo), four in the vitamin C group (mean age, 24.5 mo),three in the melatonin group (mean age, 25 mo), three inthe lazaroid group (mean age, 26 mo), and three in theplacebo group (mean age, 26).

Blood Levels of Vitamin E and MelatoninTo verify the appropriate functioning of the matrix

drug delivery system, serum for vitamin E and melatonintesting was obtained. Other assays are available for theremaining treatment options, but they are considered un-reliable and extremely expensive. Testing was performedat four separate time points: at baseline, 1 week afterimplantation of the 90-day pellet, 6 weeks after implan-tation, and just before the pellet was due to be changed(85–89 d). Levels of the supplemented vitamin or hormonewere in the appropriate range.

The vitamin E group baseline levels were 9 6 0.5nmol/mL. The levels at 1 week, 6 weeks, and just beforepellet change were 16.4 nmol/mL 6 1.7 nmol/mL. Melato-nin levels are subject to diurnal variation. Testing wasperformed at approximately 10:00 A.M. and revealed abaseline level of 59.5 pg/mL 6 5 pg/mL. The levels at 1week, 6 weeks, and just before pellet change were 1,446pg/mL 6 145 pg/mL.

Auditory Sensitivity ResultsBaseline auditory thresholds for all subjects at 2 to 4

months of age were in the range of previous studies in ourlaboratory as well as other laboratories, and were between15 and 35 dB in the frequencies tested. Over the span of 24to 27 months the thresholds shifted the least in the diet-restricted groups and the treated subjects. The thresholdshifted the most in the placebo group as tested at 26months. The difference between the restricted-diet groupand the placebo group was statistically significant (P ,.01). The antioxidant-treated groups had a statisticallysignificant protective effect when compared with the pla-cebo group, but this effect was frequency specific (P , .05).In other words, not all frequencies showed a statisticallysignificant response. Specifically, the vitamin E group ex-perienced statistically better auditory thresholds at 6, 9,and 12 kHz (P , .05), but not at 3 or 18 kHz. VitaminC–treated animals showed a statistically significant lowerthreshold compared with the placebo group at all frequen-cies except 3 kHz. Melatonin results were statistically

better than placebo results at all frequencies except 12kHz. Lazaroid was better than placebo only at 12 kHz.

There was a progressive reduction of auditory sensi-tivity which began at approximately 6 to 9 months of agein all groups. The greatest amount of hearing loss oc-curred over the last 12 months of life. The least amount ofpresbyacusis was observed in the diet-restricted subjects.The other treated animals experienced better hearing sen-sitivities than the control group, and this was statisticallysignificant for all groups (P , .01) with the exception ofthe lazaroid group, and the overall effect varied with fre-quency. Figure 1 is a graph depicting cumulative audio-logical data of all groups.

Table I summarizes the tests for overall treatmenteffects for each of the four analyses. The first four frequen-cies showed significant treatment effects in all of the anal-yses. A treatment effect could not be confirmed for a fre-quency of 18 kHz for the two analyses restricted toanimals observed for 20 months or more.

The means for the various groups and the results ofthe comparisons between groups within the repeated-measures analysis are available in the detailed thesis(unpublished data).

Mitochondrial DNA DeletionsTo study changes in mitochondrial DNA, we evalu-

ated brain, auditory nerve, and stria vascularis for cyto-chrome b, the ND-1 16S rRNA region of the mitochondrialgenome and the common aging deletion (mtDNA4834). Cy-tochrome b and ND-1 16S rRNA are used as controls toverify the presence of mitochondrial DNA while themtDNA4834 is the test of interest that provides a molecu-lar marker of aging. All specimens were positive for cyto-chrome b and ND-1 16S rRNA, verifying the presence ofmtDNA. A small piece of muscle was harvested at every3-month testing interval, thus allowing the opportunity todetermine when the deletion was first apparent. The com-mon aging deletion gradually appeared at 9 months of agein all groups except the restricted-diet group. This groupdid not show evidence of the common aging deletion until16 months of age. Quantification studies revealed an age-appropriate increase in the mtDNA4834, and DNA se-quencing verified the authenticity of the PCR products.The most rapid accumulation of the mtDNA4834 was evi-dent in the ad libidum diet (placebo); the other treatedgroups had a slower accumulation of the deletion (Figs.2A–C).

HistologyThe histological results demonstrated loss of outer

hair cells that is typically observed with aging. The differ-ences between the groups showed less hair cell loss in therestricted-diet group and a slight reduction in hair cellloss in the treated animals as compared with the controlsubjects. Fig. 3A and B shows cochlear histograms fromthe diet-restricted and placebo groups, respectively.

DISCUSSIONThe results of this prospective study demonstrate

that with the normal aging process (the placebo group)there is a progressive reduction in auditory sensitivity


733

similar to presbyacusis observed in humans. This reduc-tion in hearing accelerates during the last 12 months oflife. Furthermore, there is an age-appropriate accumula-tion of the common mtDNA deletion. This progression ofage-related hearing loss and accumulation of mtDNA de-letions can be attenuated by long-term treatment withantioxidants and scavengers of ROM. The findings ob-served in the histological results are a bit more difficult tointerpret, but overall, they support these conclusions.

Presbyacusis or age-related hearing loss is the lead-ing cause of hearing loss in the world, creating a signifi-

cant burden not only for the sufferers, but also for thosewho communicate with them. The medical and socioeco-nomic costs are staggering, and with the expansion of theworld population and the numbers of elderly individualsexpecting to more than double by the year 2030, thisproblem is escalating.

There is significant evidence supporting the conceptthat altered blood flow and/or oxygen delivery results inhair cell damage. Atrophy of striae has been reported inelderly human cadavers, even in the absence of classicatherosclerotic changes.18 Similar changes have beenidentified in the stria vascularis of gerbils.59 Prazma etal.19 have observed diminished flow in capillaries of thebasal turn of the cochlea, and blood viscosity and red bloodcell rigidity appear to correlate with high-frequency hear-ing loss in elderly human subjects.20 A relationship be-tween vascular atrophy and presbyacusis has been pro-posed.60 In addition, studies have demonstrated asignificant reduction in erythrocyte velocity, increasedvascular permeability, a trend for reduced capillary diam-eters, and a decrease in auditory sensitivity with aging.2

Thus genetics, dietary factors, and environment have beensuspected to play a pivotal, yet poorly defined, role. Thecurrent experiments support previous literature and ex-pand on earlier findings by observing the progression of

Fig. 1. Auditory thresholds of the tested groups at the last auditory brainstem response (ABR) recording across all five frequencies. The X-axisplots frequency; the Y-axis plots decibels.

TABLE I.Overall Treatment Effect for Adjusted Averages Over Various

Observation Intervals.

Frequency(Hz)

Month25 Only(n 5 25)

Month20–25

(n 5 59–66)

Month15–25

(n 5 108–116)

Month12–25

(n 5 180–198)

3,000 .005 .009 ,.001 ,.001

6,000 .007 ,.001 ,.001 ,.001

9,000 .001 ,.001 ,.001 ,.001

12,000 .007 ,.001 ,.001 ,.001

18,000 .208 .114 .005 ,.001


734

age-related hearing loss in a prospective randomized trial.Furthermore, the identification of significant protection ofauditory thresholds in the diet-restricted groups and thepartial protection observed in the treated subjects in com-

parison with the placebo group support the hypothesis ofthis proposal; specifically, diet restriction, which reducesthe formation of ROM, and antioxidants, which attenuateage-associated hearing loss.

Fig. 2. A. Electrophoretic gel depicting the mitochondrial DNA data from stria vascularis. Similar data (not shown) exist for brain and auditorynerve. The gel provides the data for identification of cytochrome b from stria vascularis. This represents a 365-base-pair (bp) product andverifies the integrity of the polymerase chain reaction (PCR) and the presence of mitochondrial DNA. B. Data for identification of ND1–16S rRNAfrom stria vascularis. This represents a 601-bp product and also verifies the integrity of the PCR and the presence of mitochondrial DNA. C.Data identifying the 4,834-bp deletion (common aging deletion) from stria vascularis. This represents a 597-bp product. Lane 1 5 negativecontrol; lane 2 5 placebo; lane 3 5 diet restriction; lane 4 5 vitamin C; lane 5 5 vitamin E; lane 6 5 melatonin; lane 7 5 lazaroid; lane 8 5molecular weight standard curve. DNA concentrations were normalized to 200 ng/50 mL of PCR reactant.

Fig. 3. A. Representative animal cochlea from a diet-restricted subject (original magnification 3 400) . Note the three outer hair cell rows (1,2, and 3). There are some hair cells missing, which correlates well with the moderate degree of hearing loss noted on ABR testing. B.Representative animal cochlea from a placebo subject (original magnification 3 400). Note the three outer hair cell rows (1, 2, 3). There aremany hair cells missing, which correlates well with the severe-profound hearing loss noted on ABR testing.


735

This report represents the first prospective random-ized study designed to investigate the effect of antioxi-dants, nutritional supplementation, and dietary restric-tion on hearing loss specifically and on aging in general.Furthermore, these studies propose that the mechanisminvolving the generation of ROM and resultant damage tomitochondria may be responsible in part for presbyacusis.

The observations from this study that dietary restric-tion and other nutrient supplementation reduce the pro-gression of age-related hearing loss is an important initialstep toward outlining strategies that may slow the pro-gression of presbyacusis. All of the interventions usedaffect the generation of ROM.

There is an increasing body of evidence implicatingROM in the damage associated with cochlear ischemia,noise trauma, aging, presbyacusis, and ototoxicity. Specif-ically, localized inner ear ischemia and hypoxia inducedvia selectively clamping the anterior inferior cerebellarartery normally abolishes the compound action potentialwithin seconds and the effect becomes permanent after 8minutes of ischemia. When subjects were pretreated withallopurinol or superoxide dismutase followed by ischemia,compound action potential thresholds were maintained.61

In addition, these experiments were extended to studynoise- induced hearing loss, which, like aging, has alsobeen shown to cause vascular perturbations.

Despite the potential for prevention of some of thesedisorders using antioxidants, there are other studieswhich have not demonstrated a benefit of treatment usingantioxidant compounds such as superoxide dismutase(SOD)62,63 and a-tocopherol.64 Critical analysis of thesenegative studies reveals several flaws ranging from insuf-ficient statistical power to lack of appropriate controls, useof the wrong conjugates of SOD (i.e., those with a shorthalf-life of 1–4 s), and design difficulties. For example, onestudy gave SOD 8 hours after closed-head injury, whenthe majority of the damage had already occurred.62

There is a significant body of literature providingevidence that oxidative damage is important in the agingprocess. It has been demonstrated that life span differ-ences in certain rodent and primate species depend in parton free radical scavenging systems. Investigators ob-served that animals with longer life spans had higherlevels of SOD activity.65 Many studies have investigateddietary increases in antioxidants and have demonstratedlife span increases of 13% to 30%.66,67

Metabolic rate is also closely linked to the agingprocess. Many investigators have shown an inverse rela-tionship between metabolic rate and life span.68,69 Therelationship between aging and metabolic rate is clearbecause the production of toxic oxygen radicals is propor-tional to the rate of oxygen utilization (i.e., the metabolicrate), and this has a close direct relationship with dietaryintake. Collectively, these studies provide both indirectand direct evidence that ROM are responsible for manyclinical diseases including cochlear damage from isch-emia, noise, and, possibly, aging.

The results of the mitochondrial experiments clearlydemonstrate the highest amount of the common agingdeletion in the placebo subjects with the least numbers ofdeletions observed in the diet-restricted group. There was

a trend showing fewer deletions in the treated groupscompared with placebo. Given the long-term effect of dam-age to cells and tissues exposed to the normal amounts ofROM, the findings of protection in auditory sensitivity andmtDNA deletions are encouraging.

The results of this study clearly demonstrate a pro-tective effect of dietary restriction and nutritional over-supplementation on age-related hearing loss and mtDNAdeletions. This beneficial effect most likely represents aslowing of the mitochondrial clock theory of aging. Specif-ically, dietary restriction and other mechanisms aimed atreducing the overall burden of ROM provide an internalmilieu that is more favorable to the organism. The find-ings support that the other treatment arms were protec-tive overall. Vitamins E and C, melatonin, and lazaroidappear to offer at least a partial protective effect. Thiseffect is not as evident as with dietary restriction. Thislimited benefit may be explained in part by the fact thateach animal was oversupplemented with individual com-pounds. There is compelling evidence that supplementingwith only one antioxidant may not be as effective as sup-plementing with multiple antioxidants. This makes tele-ological sense when considering that nutritional antioxi-dants work by a variety of mechanisms. Some stabilize cellmembranes, others scavenge ROM molecules, yet othersenhance the role of other antioxidants. However, in exper-imental design, it is crucial to test supplements individu-ally, first, to elucidate specific properties and effects.Based on such experimentation, studies subsequently canbe designed to observe the effect of supplements in com-bination. Indeed, experiments to investigate such a com-bined effect have clearly shown a synergistic positive ef-fect on generalized system function.70 Furthermore,ongoing studies in our laboratory are addressing this spe-cific issue using a combination of vitamins and compoundsthat can enhance mitochondrial function.

CONCLUSIONThis study presents empirical evidence in support of

an integrated hypothesis of aging and presbyacusis. Inaddition, presented data provide a framework for the con-tinuing identification and evolution of pharmacologicaland nutritional strategies designed to attenuate age-associated hearing loss. Areas of scientific inquiry havealso been identified that require further investigationwith the ultimate goal of enhancing our understanding ofthe mechanisms that surround senescence, both as a phe-nomenon and as an intrinsic biological property.

ACKNOWLEDGMENTMy gratitude is extended to Drs. Wayne Quirk, Uma

Bai, Mumtaz Khan, Wen Xue Tang, and Najeeb Shirwany,who assisted over the years, and Mr. Joseph Henig, andMrs. Cristy McAuley. Additional gratitude is expressedfor the clinical insight and support of Drs. Richard Ni-chols, Malcolm Graham, and Herbert Silverstein.

BIBLIOGRAPHY1. Prazma J, Carrasco VN, Butler B, Waters G, Anderson T,

Pillsbury HC. Cochlear microcirculation in young and old


736

gerbils. Arch Otolaryngol Head Neck Surg 1990;116(8):932–936.

2. Seidman MD, Khan MJ, Dolan DF, Quirk WS. Age-relateddifferences in cochlear microcirculation and auditory brainstem response. Arch Otolaryngol Head Neck Surg 1996;122(11):1221–1226.

3. Gacek MR, Schuknecht HF. Pathology of presbycusis. IntAudiol 1969;8:199–209.

4. Rosenhall U, Pederson K, Dotevall M. Effects of presbycusisand other types of hearing loss on auditory brainstemresponses. Scand Audiol 1986;15(4):179–185.

5. Semsei I, Roa G, Richardson A. Changes in the expression ofsuperoxide dismutase and catalase as a function of age anddietary restriction. Biochem Biophys Res Commun 1989;164(2):620–625.

6. Hattori K, Tanaka M, Sugiyama S, et al. Age-dependentincrease in deleted mitochondrial DNA in the humanheart: possible contributory factor to presbycardia. AmHeart J 1991;121(6, Part 1):1735–1742.

7. Wallace DC. Diseases of the mitochondrial DNA. Annu RevBiochem 1992;61:1175–1212.

8. Cortopassi GA, Arnheim N. Detection of a specific mitochon-drial DNA deletion in tissues of older humans. NucleicAcids Res 1990;18(23):6927–6933.

9. Johns M, Brackmann DE, Kimmelman C, Papsidero MJ,Koopman CF Jr, Loury M. Goals and mechanisms fortraining otolaryngologists in the area of geriatric medicine.Otolaryngol Head Neck Surg 1989;100(4):262–265.

10. Gates GA, Caspery DM, Clark W, Pillsbury HC III, BrownSC, Dobie RA. Presbyacusis. Otolaryngol Head Neck Surg1986;100(4):266–271.

11. Sixt E, Rosenhall U. Presbyacusis related to socioeconomicfactors and state of health. Scand Audiol 1997;26(3):133–140.

12. Halliwell B, Gutteridge JM, Cross CE. Free radicals, antioxi-dants, and human disease: where are we now? J Lab ClinMed 1992;119(6):598–620.

13. Byers T, Bowman B. Vitamin E supplements and coronaryheart disease. Nutr Rev 1993;51(11):333–336.

14. Ziegler RG. Vegetables, fruits and carotenoids and the risk ofcancer. Am J Clin Nutr 1991;53(Suppl 1):251S-259S.

15. Comstock GW, Helzlsouer KJ, Bush TL. Prediagnostic serumlevels of carotenoids and vitamin E as related to subse-quent cancer in Washington County, Maryland. Am J ClinNutr 1991;53(Suppl 1):260S-264S.

16. Smith MA, Perry G. Free radical damage, iron and Alzhei-mer’s disease. J Neurol Sci 1995;134(Suppl):92–94.

17. Jenner P, Olanow CW. Oxidative stress and the pathogenesisof Parkinson’s disease. Neurol 1996;47(6 [Suppl 3]):S161–S170.

18. Johnsson LG, Hawkins JE Jr. Vascular changes in the hu-man inner ear associated with aging. Ann Otol RhinolLaryngol 1972;81(3):364–376.

19. Prazma J, Butler B, Waters G, Anderson TL, Pillsbury HC.Proceedings of the 71st Annual Meeting of the Federation ofAmerican Societies for Experimental Biology [abstract1538]. Washington, DC. March 29–April 2, 1987;46(4):1538.

20. Gatehouse S, Lowe GD. Whole blood viscosity and red cellfilterability as factors in sensorineural hearing impair-ment in the elderly. Acta Otolaryngol Suppl (Stockh) 1990;476:37–43.

21. Axelsson A. The cochlear blood vessels in guinea pigs ofdifferent ages. Acta Otolaryngol (Stockh) 1971;72(3):172–181.

22. Wing KG. Studies of basic cochlear physiology and the energymetabolism of the cochlear response in the rat. Acta Oto-laryngol Suppl (Stockh) 1959;148:1–97.

23. Fernandez C, Alzate R. Modifications of cochlear responses tooxygen deprivation. Arch Otolaryngol 1959;69(1):82–94.

24. Kimura R, Perlman HB. Experimental obstruction of thevenous drainage and arterial supply of the inner ear. AnnOtol Rhinol Laryngol 1956;65(2):332–350.

25. Honrubia V, Johnston BM, Butler RA. Maintenance of co-

chlear potentials during asphyxia. Acta Otolaryngol(Stockh) 1965;60:105–112.

26. Woolf NK, Ryan AF, Harris JP. Development of mammalianendocochlear potential: normal ontogeny and effects of an-oxia. Am J Physiol 1986;250:493–498.

27. Demopuolos HB, Flamm ES, Pietronigro DD, Seligman ML.The free radical pathology and the microcirculation in themajor central nervous system disorders. Acta PhysiolScand Suppl 1980;492:91–119.

28. Raichle ME. The pathophysiology of brain ischemia. AnnNeurol 1983;13(1):2–10.

29. Wallace DC, Zheng XX, Lott MT, et al. Familial mitochon-drial encephalopathy (MERFF): genetic, pathophysiologi-cal, and biochemical characterization of a mitochondrialDNA disease. Cell 1988;55(4):601–610.

30. Ernster L, Ikkos D, Luft R. Enzyme activities of humanskeletal muscle mitochondria: a tool in clinical metabolicresearch. Nature 1959;184:1851–1854.

31. Luft R, Ikkos D, Palmieri G, Ernster L, Afzelius B. A case ofsevere hypermetabolism of nonthyroid origin with a defectin the maintenance of mitochondrial respiratory control: acorrelated clinical, biochemical, and morphological study.J Clin Invest 1962;41:1776–1804.

32. Shoffner JM, Lott MT, Voljavec AS, et al. SpontaneousKearns-Sayre/chronic external ophthalmoplegia plus syn-drome associated with a mitochondrial DNA deletion: aslip-replication model and metabolic therapy. Proc NatlAcad Sci U S A 1989;86(20):7952–7956.

33. Pavlakis SG, Phillips PC, DiMauro S, DeVivo DC, RowlandLP. Mitochondrial myopathy, encephalopathy, lactic acido-sis and strokelike episodes: a distinctive clinical syndrome.Ann Neurol 1984;16(4):481–488.

34. DiMauro S, Bonilla E, Lombes A, Shanske S, Minetti C,Moraes CT. Mitochondrial encephalomyopathies. NeurolClin 1990;8(3):483–506.

35. Gold M, Rapin I, Shanske S. Mitochondrial inheritance ofacquired deafness. Ann N Y Acad Sci 1991;630:301–302.

36. Corral-Debrinski M, Stepien G, Shoffner JM, et al. Hypox-emia is associated with mitochondrial DNA damage andgene induction: implications for cardiac disease. JAMA1991;266(13):1812–1816.

37. Merril CR, Zullo S, Ghanbari H, et al. Possible relationshipbetween conditions associated with chronic hypoxia andbrain mitochondrial DNA deletions. Arch Biochem Biophys1996;326(1):172–177.

38. Trounce I, Byrne E, Marzuki S. Decline in skeletal musclemitochondrial respiratory chain function: possible factor inaging [comments]. Lancet 1989;1(8639):637–639.

39. Nohl H, Hegner D. Do mitochondria produce oxygen radicalsin vivo? Eur J Biochem 1978;82(2):563–567.

40. Bandy B, Davison AJ. Mitochondrial mutations may increaseoxidative stress: implications for carcinogenesis and aging?Free Radical Biol Med 1990;8(6):523–539.

41. Piko L, Houghman AJ, Bulpitt KJ. Studies of sequence het-erogeneity of mitochondrial DNA from rat and mouse tis-sues: evidence for an increased frequency of deletions/additions with aging. Mech Ageing Dev 1988;43(3):279–293.

42. Miyabayashi S, Hanamizu H, Endo H, Tada K, Horai S. Anew type of mitochondrial DNA deletion in patients withencephalomyopathy. J Inherit Metab Dis 1991;14(5):805–812.

43. Ballinger SW, Shoffner JM, Hedaya EV, et al. Maternallytransmitted diabetes and deafness associated with a 10.4kb mitochondrial DNA deletion. Nat Genet 1992;1(1):11–15.

44. van den Ouweland JM, Lemkes HH, Ruitenbeek W, Sand-kuijl LA. Mutation in mitochondrial tRNA(Leu)(UUR)gene in a large pedigree with maternally transmitted typeII diabetes mellitus and deafness. Nat Genet 1992;1(5):368–371.

45. Sawada S, Takeda T, Kakigi A, et al. Audiological findings ofsensorineural deafness associated with a mutation in themitochondrial DNA. Am J Otol 1997;18:332–335.


737

46. Ueda N, Oshima T, Ikeda K, Takasaka T. Mitochondrial DNAmutations and presbycusis. Presented at the MolecularBiology of Hearing and Deafness meeting. Bethesda, MD,October 6–8, 1995.

47. Bai U, Seidman MD, Hinojosa R, Quirk WS. MitochondrialDNA deletions associated with aging and possibly presby-acusis: a human archival temporal bone study. Am J Otol1997;18:449–453.

48. Pommier JP, Lebeau J, Ducray C, et al. Chromosomal insta-bility and alteration of telomere repeat sequences. Bio-chimie 1995;77(10):817–825.

49. Hockenberry DM, Oltvai ZN, Yin XM, Milliman CL, Kors-meyer SJ. Bcl-2 functions in an antioxidant pathway toprevent apoptosis. Cell 1993;75(2):241–252.

50. Kane DJ, Sarafian TA, Anton R, et al. Bcl-2 inhibition ofneural cell death: decreased generation of reactive oxygenspecies. Science 1993;262(5137):1274–1277.

51. Sohal RS, Allen RG. Relationship between metabolic rate,free radicals, differentiation and aging: a unified theory.Basic Life Sci 1985;35:75–104.

52. Southorn PA, Powis G. Free radicals in medicine, I: chemicalnature and biologic reactions. Mayo Clin Proc 1988;63(4):381–389.

53. Zs-Nagy I, Semsei I. Centrophenoxine increases the rates oftotal and mRNA synthesis in the brain cortex of old rats:an explanation of its action in terms of the membranehypothesis of aging. Exp Gerontol 1984;19(3):171–178.

54. Oltvai ZN, Milliman CL, Korsmeyer SJ. Bcl-2 heterodimer-izes in vivo with a conserved homolog, Bax, that acceler-ates programmed cell death. Cell 1993;74(4):609–619.

55. Trounce IA, Kim YL, Jun AS, Wallace DC. Assessment ofmitochondrial oxidative phosphorylation in patient musclebiopsies, lymphoblasts, and transmitochondrial cell lines.Methods Enzymol 1996;264:484–509.

56. Ernster L, Nordenbrand K. Skeletal muscle mitochondria.Methods Enzymol 1967;10:86–94.

57. McCabe PC. A Guide to Methods and Applications in PCRProtocols. New York: Academic Press, 1990:76–83.

58. SAS Institute, Inc. SAS/STAT Software: Changes and En-hancements Through Release 6.11. Cary, NC: SAS Insti-tute, Inc., 1996:31–656.

59. Gratton MA, Schulte BA. Alterations in microvasculature are

associated with atrophy of the stria vascularis in quiet-aged gerbils. Hear Res 1995;82(1):44–52.

60. Pujol R, Guy R, Puel J, Lenoir M, Eybalin M, Recasens M.Glutamate neurotoxicity in the cochlea: a possible conse-quence of ischaemic or anoxic conditions occurring in ag-ing. Acta Otolaryngol Suppl (Stockh) 1990;476:32–36.

61. Seidman MD, Quirk WS, Nuttall AL, Schweitzer VG. Theprotective effects of allopurinol and SOD-PEG on ischemicinduced cochlear damage. Otolaryngol Head Neck Surg1991;105(3):457–463.

62. Bolli R. Superoxide dismutase 10 years later: a drug in searchof a use. J Am Coll Cardiol 1991;18(1):231–233.

63. Young B, Runge JW, Waxman KS, et al. Effects of pegorgot-ein on neurologic outcome of patients with severe headinjury. JAMA 1996;276(7):538–543.

64. Riemersma RA, Wood DA, Macintyre CC, Elton RA, Gey KF,Oliver MF. Risk of angina pectoris and plasma concentra-tions of vitamin A, C, and E, and carotene. Lancet 1991;337(8732):1–5.

65. Tolmasoff JM, Ono T, Cutler RG. Superoxide dismutase: cor-relation with life-span and specific metabolic rate in pri-mate species. Proc Natl Acad Sci U S A 1980;77(5):2777–2781.

66. Harman D. Free radical theory of aging: effect of free radicalreaction inhibitors on the mortality rate of male LAF mice.J Gerontol 1968;23(4):476–482.

67. Heidrick ML, Hendricks LC, Cook DE. Effect of dietary2-mercaptoethanol on the life span, immune system, tumorincidence and lipid peroxidation damage in spleen lympho-cytes of aging BC3F1 mice. Mech Ageing Dev 1984;27(3):341–358.

68. Cutler RG. Redundancy of information content in the genomeof mammalian species as a protective mechanism deter-mining aging rate. Mech Ageing Dev 1973;2(6):381–408.

69. Sacher. Longevity in vertebrates: a further comment. FedProc 1976;35:1112–1113.

70. Chen H, Tappel AL. Protection of vitamin E, selenium, TroloxC, ascorbic acid palmitate, acetylcysteine, coenzyme Q-0,coenzyme Q-10, beta carotene, canthaxanthin, and (1)-catechin against oxidative damage to rat blood and tissuesin vivo. Free Rad Biol Med 1995;18(5):949–953.


738

effects of dietary restriction and antioxidants on presbyacusis

Documents