PERSPECTIVE
NATUREGENETICS|VOLUME40|NUMBER3|MARCH2008 275
WhatcausesmitochondrialDNAdeletionsin
humancells?
KimJKrishnan,AmyKReeve,DavidCSamuels,PatrickFChinnery,JohnKBlackwood,RobertWTaylor,
SjoerdWanrooij,JohannesNSpelbrink,RobertNLightowlers&DougMTurnbull
MitochondrialDNA(mtDNA)deletionsareaprimarycauseof
mitochondrialdiseaseandarelikelytohaveacentralrolein
theagingofpostmitotictissues.Understandingthemechanism
oftheformationandsubsequentclonalexpansionofthese
mtDNAdeletionsisanessentialfirststepintryingtoprevent
theiroccurrence.Wereviewthepreviousliteratureandrecent
resultsfromourownlaboratories,andconcludethatmtDNA
deletionsaremostlikelytooccurduringrepairofdamaged
mtDNAratherthanduringreplication.Thisconclusionhas
importantimplicationsforpreventionofmtDNAdiseaseand,
potentially,forourunderstandingoftheagingprocess.
MitochondrialDNA(mtDNA)fascinatesbothbiologistsandclinicians
becauseofitsuniquegeneticsanditsroleinhumandisease.Themost
common genetic defects seen in individuals with mtDNA-associated
diseasesaredeletions
1,2
orpointmutations
3,4
.MtDNAdeletions,mol-
eculesthathavelostlargesectionsofthemtDNAgenome,havebeen
showntohaveanimportantroleinhumanpathologyinthreedifferent
clinicalscenarios.First,singlemtDNAdeletionsareacommoncauseof
sporadicmitochondrialdisease;inthesecases,anidenticalmtDNAdele-
tionisdetectedinallcellswithinanaffectedtissue
5
.Second,inanother
largegroupofindividualswithmitochondrialdisease,therearemultiple
mtDNAdeletionsinaffectedtissues,particularlyinthemuscleandthe
centralnervoussystem
6
.Theprimarygeneticdefectintheseindividuals
involvesnucleargenesencodingproteinsinvolvedineithermitochondrial
nucleotidemetabolism(TPand SLC25A4,thegeneencodingtheheart-
musclespecificisoformofANT1)ormtDNAmaintenance(C10orf2 (also
knownas PEO1), POLGand POLG2)
7
.Third,therearenumerousreports
ofmtDNAdeletionsinagedpostmitotictissuesandindividualswith
neurodegenerativediseases
8,9
.TheextentofmtDNAdeletionsinthese
casesisoftenmuchlowerthanthatseeninindividualswithmitochon-
drialdisease,butinsubstantianigraneurons,forexample,approximately
50%ofallmtDNAmoleculescontainanmtDNAdeletion
8,9
.
The ratio of mutated versus wild-type mtDNA is a crucial fac-
tor in determining whether there will be a biochemical defect
6
. As
mtDNAmutationsarefunctionallyrecessive,usually>60%ofmtDNA
moleculescontainadeletionbeforeabiochemicaldefectisobserved
(detectedbytheabsenceofcytochromecoxidase(COX)activity)
10
.In
individualswithsomaticmtDNAmutations,therearedifferentmtDNA
deletionsindifferentcells,suggestingthatclonalexpansioncanreach
levelsthatcauseCOXdeficiency
8,9,11
.
Tounderstandthemechanismsinvolvedintheformationandclonal
expansion of mtDNA deletions, one must take into consideration
theverydifferentclinicalscenariosinwhichthesemtDNAdeletionsare
observed.Thus,anyhypothesismustexplainarangeoffindings,from
mtDNA-deletionformationinoocytestothatinneuronsfromelderly
subjects,forexample.
CommonfeaturesofmtDNAdeletions
MostmtDNAdeletionssharesimilarcharacteristics.Mostarelocated
in the major arc between two proposed origins of replication (O
H
andO
L
;Mitomap),andarepredominantly(~85%)flankedbyshort
directrepeats
11,12
.WehaverecentlystudiedthetypesofmtDNAdele-
tionsdetectedinneuronsfromaginghumanbrain,individualswith
Parkinson’sdiseaseandanindividualwithamultiplemtDNAdeletion
disorder,andcomparedthesewiththosedescribedinindividualswith
single,large-scalemtDNAdeletions
13
.Theresultsfromthisstudyshow
thatmtDNA-deletioncharacteristicsaresimilarunderallcircumstances,
withnonotabledifferencesinsize,presenceorabsenceofflankingrepeat
sequences,orlengthofrepeats.Thisfindingsuggeststhatasimilarmech-
anismisgeneratingthemtDNAdeletionsinallthesedifferentclinical
situations.
AremtDNAdeletionsgeneratedbyreplicationerrors?
Currently,mostresearchersconsiderthatreplicationisthelikelymecha-
nismbehinddeletionformation.Atpresent,therearetwomainmodelsof
mtDNAreplication:astrand-asynchronous(strand-displacement)mode
ofreplication
14,15
,andamoreconventionalcoupledleading-lagging-
strandreplication(Fig.1)
16
.TheobservationthatmostreportedmtDNA
deletionsoccurinthemajorarchasledtotheproposalthatmtDNA
KimJ.Krishnan,AmyK.Reeve,PatrickF.Chinnery,JohnK.Blackwood,
RobertW.Taylor,RobertN.LightowlersandDougM.Turnbullareatthe
MitochondrialResearchGroup,TheMedicalSchool,NewcastleUniversity,
NewcastleuponTyne,NE24HH,UK.KimJ.Krishnanand
DougM.TurnbullarealsoattheInstituteforAgeingandHealth,
NewcastleUniversity,NewcastleuponTyne,NE46BE,UK.DavidC.
SamuelsisattheVirginiaBioinformaticsInstitute,VirginiaPolytechnic
InstituteandStateUniversity,Blacksburg,Virginia24061,USA.Sjoerd
WanrooijandJohannesN.SpelbrinkareattheInstituteofMedical
TechnologyandTampereUniversityHospital,Tampere33014,Finland.
SjoerdWanroojisalsoattheKarolinskaInstitute,Departmentof
MetabolicDiseaseNovum,SE-14186,Huddinge,Sweden.
e-mail:d.m.turnbull@ncl.ac.uk
Publishedonline27February2008;doi:10.1038/ng.f.94
© 2008 Nature Publishing Group http://www.nature.com/naturegenetics