Zurich Open Repository andArchiveUniversity of ZurichMain LibraryStrickhofstrasse 39CH-8057 Zurichwww.zora.uzh.ch
Year: 2019
Vascular density and distribution in neocortex
Schmid, Franca ; Barrett, Matthew J P ; Jenny, Patrick ; Weber, Bruno
Abstract: An amazingly wide range of complex behavior emerges from the cerebral cortex. Much ofthe information processing that leads to these behaviors is performed in neocortical circuits that spanthroughout the six layers of the cortex. Maintaining this circuit activity requires substantial quantities ofoxygen and energy substrates, which are delivered by the complex yet well-organized and tightly-regulatedvascular system. In this review, we provide a detailed characterization of the most relevant anatomicaland functional features of the cortical vasculature. This includes a compilation of the available data onlaminar variation of vascular density and the topological aspects of the microvascular system. We alsoreview the spatio-temporal dynamics of cortical blood flow regulation and oxygenation, many aspectsof which remain poorly understood. Finally, we discuss some of the important implications of vasculardensity, distribution, oxygenation and blood flow regulation for (laminar) fMRI.
DOI: https://doi.org/10.1016/j.neuroimage.2017.06.046
Posted at the Zurich Open Repository and Archive, University of ZurichZORA URL: https://doi.org/10.5167/uzh-146003Journal ArticleAccepted Version
The following work is licensed under a Creative Commons: Attribution-NonCommercial-NoDerivatives4.0 International (CC BY-NC-ND 4.0) License.
Originally published at:Schmid, Franca; Barrett, Matthew J P; Jenny, Patrick; Weber, Bruno (2019). Vascular density anddistribution in neocortex. NeuroImage, 197:792-805.DOI: https://doi.org/10.1016/j.neuroimage.2017.06.046
1
Vasculardensityanddistributioninneocortex
FrancaSchmid1,MatthewJ.P.Barrett2,3,PatrickJenny1,BrunoWeber2,3
1. InstituteofFluidDynamics,ETHZurich,Sonneggstrasse3,8092Zurich,Switzerland
2. InstituteofPharmacologyandToxicology,UniversityofZurich,Winterthurerstrasse
190,CH-8057Zurich,Switzerland
3. NeuroscienceCenter,UniversityandETHZurich,Winterthurerstrasse190,CH-8057
Zurich,Switzerland
Correspondingauthor:FrancaSchmid,[email protected]
2
Abstract
Anamazinglywiderangeofcomplexbehavioremergesfromthecerebralcortex.Muchof
theinformationprocessingthatleadstothesebehaviorsisperformedinneocorticalcircuits
thatspanthroughoutthesixlayersofthecortex.Maintainingthiscircuitactivityrequires
substantialquantitiesofoxygenandenergysubstrates,whicharedeliveredbythecomplex
yetwell-organizedandtightly-regulatedvascularsystem.Inthisreview,weprovidea
detailedcharacterizationofthemostrelevantanatomicalandfunctionalfeaturesofthe
corticalvasculature.Thisincludesacompilationoftheavailabledataonlaminarvariationof
vasculardensityandthetopologicalaspectsofthemicrovascularsystem.Wealsoreview
thespatio-temporaldynamicsofcorticalbloodflowregulationandoxygenation,many
aspectsofwhichremainpoorlyunderstood.Finally,wediscusssomeoftheimportant
implicationsofvasculardensity,distribution,oxygenationandbloodflowregulationfor
(laminar)fMRI.
Keywords:corticalmicrovasculature,vasculardensity,neurovascularcoupling,
hemodynamicresponse,cerebraloxygenation,laminarcharacteristics
1 Introduction
Thebrainconsumesapproximatelyaquarterofthebody’stotalglucoseandafifthofthe
oxygen,anorderofmagnitudemorethanwhatwouldbeexpectedonaweightbasis.This
remarkableenergydemandrequiresarobustenergysupplyviathebloodstream,anda
complexcerebrovascularsystemhasevolvedtomeetthisdemand.Adetailedknowledgeof
thecerebralvasculatureiscrucialtounderstandthebasicprinciplesofcerebralbloodflow
(CBF),itscouplingtoneuralprocessing,andalsotounderstandnon-invasivehumanbrain
imaging.Giventheextraordinaryimportanceofstructuralandfunctionalmagnetic
resonanceimaging(fMRI),itisquitesurprisingthatquantitativedataregardingcerebral
bloodvesselsaresparse.Inparticular,themicrovascularsystemissignificantly
understudiedbothintermsofitsstructuralandfunctionalcharacteristics.Theaimofthe
presentreviewistoprovideadetailedoverviewofthecerebralmicrovascularsystem,with
afocusontheneocortexanditsimportanceforlaminarMRI.
Thehumanbrainreceivesapproximately15-20%ofthetotalcardiacoutputandthisblood
istransportedfromthetrunkviafourlargevessels,theleftandrightinternalcarotid
arteriesandtheleftandrightvertebralarteries.Beforeramifyingintothelargefeeding
arteriesofthebrain,thevertebralarteriesjoinintothebasalarteryandtogetherwiththe
carotidvesselsformaring-likestructurecalledthecircleofWillis.Thisstructureintroduces
redundancyandservesasaprecautionarymeasure.Themiddlecerebralartery’scirculation
territoryisthelargestandcompriseslargepartsofthefrontal,temporalandparietalcortex.
Withoutbeingtoodetailed,theposteriorcerebralarteryfeedstheinferiorandmedial
surfaceoftheoccipitalandtemporalcortices,whereastheanteriorcerebralarteryis
responsiblefortheperfusionofmedialaspectsofthefrontalandparietalcortex.Whereas
thefeedingterritoriesofthelargecerebralarteriesplayanimportantroleinthediagnosis
ofneurologicaldisorders,theyhaveonlyaminorimportanceforcorticalMRI.Ofmuch
greaterimportanceisthecerebralmicrovasculatureandhencethisreviewwillbefocused
3
onthemammaliancorticalmicrovasculature(forarecentreview,pleasealsoseeHirschet
al.,2012).
Themaindutyofthecerebrovascularsystemistodeliveroxygenandglucosetothetissue.
Thehumanbrainmetabolizesabout31mmolglucoseper100gramsoftissueperminuteto
produceadenosine-triphosphate(ATP),whichistoalargeextentformedbyoxidative
phosphorylationinthetricarboxylicacidcycle.Thismetabolicpathwayrequiresalarge
amountofoxygen,whichdiffusesfrommicrovesselstomitochondriaofthebraincells.
Withinthebloodvessels,chemicallyboundoxygenmoleculesdissociatefromhemoglobin
anddissolvedoxygendiffusesthroughtheredbloodcellmembrane,bloodplasma,the
capillarywall,andinterstitialfluiduntilitenterstheneuronsandglialcells,whereitis
metabolized.Blood-borneglucosetravelsalongthesameroute;however,incontrastto
oxygen,whichcanfreelydiffuse,glucosemoleculesaretransportedacrossmembraneswith
thehelpofglucosetransporters.AmajorwasteproductofbrainenergymetabolismisCO2,
whichrapidlydiffusesoutofthenervoustissueandisclearedfromthebrainalsoviathe
vascularsystem.Thebrainvascularsystemisnotonlyresponsibleforsupplyandwaste
collection,butalsoactsasabarrierthatrestrictsthepassageofmoleculesfrombloodto
brain.Theendothelialcellsofthebrainvesselsarecoupledbytightjunctionsandformthe
blood-brainbarrier.
Itwasrecognizedasearlyasthe19thcenturythatthebrain’sbloodsupplyisdynamic,and
thatchangesinneuralactivityaremirroredbypreciselycontrolledchangesin
hemodynamics(Mosso,1881;RoyandSherrington,1890).Thisspatialandtemporal
neurovascularcouplinghasbeensystematicallyusedtogeneratedetailedmapsof
hemodynamicchangesthatareassumedtobesurrogatesoftheactualregionalneural
activation.Themostimportantapplicationofthisrelationshipistheso-calledblood
oxygenationlevel-dependent(BOLD)contrast(Kwongetal.,1992;Ogawaetal.,1990)used
infMRI.
Theneocortexisoneofthemostintriguingstructuresofthebrain.Itshowsan
extraordinaryflexibility,giventhelargerangeofbehaviorthatemergesdespiteonlylimited
variationinitsanatomicalorganizationacrossthedifferentcorticalareas(Douglasand
Martin,2004).Thelaminarorganizationofthecortexanditsimpactonstructuraland
functionalMRIisthetopicofthisspecialissue.Thesix-layeredstructure,withwell-defined
microcircuits(DouglasandMartin,2004),islargelypreservedacrossspecies.Similarly,the
vascularsystemoftheneocortexalsoshowsahighdegreeoforganizationthroughoutthe
corticaldepth,anorganizationthatisanalogousinallthestudiedanimalmodelsaswellas
inthehumancorticalmicrovasculature.
Thepresentreviewintroducesthevascularsystemoftheneocortexandisstructuredinthe
followingway.Afterashortexplanationofthemostimportantmethodsusedtoinvestigate
brainbloodvessels,weintroducethegrossorganizationfollowedbyadetailed
characterizationoflaminardensityvariation,topologicalaspects,andflowregulationofthe
microvascularsystem.Weendthereviewwithadiscussionoftherelevanceofthe
vasculaturesystemfor(laminar)fMRI.
2 Methodstostudybrainvasculature
ClassicalstudiesofthemicrovascularnetworkarebasedonIndianinkfillingsandscanning
electronmicroscopyofvascularcorrosioncasts(Duvernoyetal.,1981;Reina-DeLaTorreet
4
al.,1998;Weberetal.,2008).Thecombinationofimmunohistochemistryandstereology
alsoallowsestimatingvesseldiametersandvasculardensity(Weberetal.,2008).
However,thethree-dimensionalvascularnetworktopologyoftheneocortexhasnotbeen
studiedindetailuntilrecently,whennewtechnologiesbecameavailable.Inorderto
describethenetworkinthreedimensions,methodsarerequiredthathaveaspatial
resolutiontoresolvecapillariesinasufficientlylargefieldofviewtocoveranentirecortical
column.Inordertoestimatethediameterofcapillaries,submicronspatialresolutionis
required.DavidKleinfeldandcolleagueshavedevelopedtheso-calledall-opticalhistology
technique(Tsaietal.,2003;Tsaietal.,2009),wherebytwo-photonmicroscopyisusedto
imagethefluorescentlylabeledvasculature,andlaserablationisusedtoincreasethefield
ofview,whichwouldnormallybelimitedtoafewhundredmicrometers.Anotherpossible
approachistheuseofsynchrotronradiation-basedX-raymicroscopy,whereahighphoton
fluxisexploitedtoacquireabsorption-basedtomographicimagesofthevasculature
(Guibertetal.,2010;Heinzeretal.,2006;Plouraboueetal.,2004;Reicholdetal.,2009).
Recently,theclassicalIndianinkfillingmethodhasbeencombinedwithaserialsection
microscopictechnique,calledMicro-OpticalSectioningTomography(MOST),andwasused
toacquirethevascularsystemoftheentiremousebrain(Xueetal.,2014).
Withtheadventofadvancedtissueclearingmethods,selectiveplaneillumination
microscopy(SPIM)orultramicroscopy(Erturketal.,2012)isbecominganincreasingly
powerfulapproachtoacquireandreconstructfluorescentlylabeledvesselsinlargesamples.
Todate,thespatialresolutionandavaryingpointspreadfunctionacrossthefieldofview
aretechnicalchallengesthatneedtobeovercomeforthemethodtobecomeanovel
standard.
Allthehistologicaltechniquessharetheproblemoftissuedeformation,eitherdueto
fixationand/orclearingofthetissue.Therefore,measurementsofthemicrovascular
networkneedtobecorrectedforthesealterations.Onewayofdoingthisistomeasure
vasculardiametersinvivousingtwo-photonmicroscopyandtousethediameter
distributiontotransformtheexvivomeasurements(Tsaietal.,2009).Moreover,future
studiescoulddeploydeepinvivotwo-photonmicroscopytoacquirethecorticalvascular
systeminitsentiredepth.
3 Grossstructureofthecorticalvascularsystem
Thecerebralarteriesrunalongthesurfaceofthecortexandramifyintoacomplexnetwork
ofso-calledpialarteries.Thesearteriesaresituatedinthesubarachnoidspaceandare
surroundedbyapialcelllining.Thetopologyofthispialarterialnetworkensuresarobust
deliveryofbloodtothecerebralcortex(Section5.1).
Onceanarteryleavesthepialnetwork,itpenetratesperpendicularlyintothecortex(Figure
1).Thepialcellensheathmentofthearteriescontinuesintheformofaperivascularchannel
(Zhangetal.,1990).Thispialcelllayerisperforatedatthearteriolarlevelandiscompletely
absentoncapillariesandveins.Aperivascularspace,termedtheVirchow-Robinspace,
surroundsthecorticalarteriesandveinsinafunnel-shapedmanner.Aroundthelargepial
anddivingvessels,astrocytesdefinetheVirchow-Robinspace.Thisperivascularspaceisa
clearingrouteforinterstitialsolutes,muchasthelymphaticsystemisfortherestofthe
body(IliffandNedergaard,2013;Iliffetal.,2012;Iliffetal.,2013;Xieetal.,2013).Thisso-
calledglymphaticsystemconsistsofapara-arterialinfluxroute;apara-venousclearance
5
route;andatrans-parenchymalpathwaydependingontheaquaporin-4waterchannelin
astrocytes.
Thecorticalarteriessendoffcollateralsatdifferentcorticaldepths.Thisfactledtothe
introductionofaclassificationschemebyDuvernoyetal.(1981),inwhichavesseloftype1
wouldfeed/drainsuperficialcorticallayers,whereashigherordervesseltypeswould
feed/draindeeperlayers.Somecorticalarteriesevenpenetratetheentirecortexwithout
anycollateraluntilreachingwhitematter(Section5.2).Itisimportanttonotethatthese
classesarenotcompletelyseparable,butappearratherasacontinuum.Corticalarteries
divergeintoarterioles(smallarteries)andeventuallyendinthecapillarynetwork(Section
5.3).Thecapillarynetwork,wheremostoftheexchangeofenergysubstratesandoxygen
occursbetweenbloodandtissue,convergesintothevenoussystem.Again,principal
corticalveinsareorientedperpendicularlytothecorticalsurface(Section5.2).Thebasic
buildingprincipleofthecorticalvasculatureis,therefore,deliveryofbloodfromthecortical
surface,penetrationintothecortexinaperpendicularmannerand,afterthecapillary
passage,drainagebacktothesurface,wherethebloodiseventuallytransportedawayvia
thevenoussystemthatconvergesintolargesinuses.
Therearedistinctdifferencesintheultrastructurebetweenthedifferenttypesofcerebral
bloodvessels.Commontoarteries,capillariesandveinsaretheendothelialcelllayerand
thethinbasalmembrane.Endothelialcellsarecoupledbytightjunctionsandformthe
blood-brainbarrierthatrestrictsthepassageofmoleculesfrombloodtobrain.Whereas
capillariesonlyconsistofthesetwoelements,arteriesandtoamuchlesserextentalso
veinsarecoveredwithasmoothmusclesheath.Thesesmoothmusclesareresponsiblefor
regulatingthevascularresistancebychangingthevesseldiameter(Section6).Pericytesare
averyheterogeneouscelltypethathaveaclaw-likeappearanceandarelocatedonthe
abluminalsideofendothelialcells(Armuliketal.,2011).Theyalsohaveacontractile
capacity,buttheirroleasactiveregulatorsofvascularresistanceisdebated(Section6,
(Fernandez-Klettetal.,2010;Halletal.,2014;Hamiltonetal.,2010;Peppiattetal.,2006)).
Furthermore,allmicrovesselsarealmostcompletelycoveredbyastrocyticendfeet
(Mathiisenetal.,2010).Thisperivascularastrocyticsheathisthoughttoplayapivotalrole
inthetranscellulartraffickingofmetabolitesaswellaswaterandionexchangeatthe
blood-braininterface.
1 mm
6
Figure1Grossstructureofthecorticalvasculature.Left:Scanningelectronmicrographofavascularcorrosioncastfrom
themonkeyvisualcortex(primaryvisualcortex).Arteriesareshadedinredandveinsareblue.FigurefromHirschetal.
(2012).Right:Schematicrepresentationofthecorticalvasculatureanditskeycomponents.
4 Densityofthecorticalmicrovascularsystem
Thediametersofcorticalvesselsrangefromaround4µm(capillaries)toafewtensof
microns(arteries/veins)(Blinderetal.,2013;Duvernoyetal.,1981;Gutiérrez-Jiménezetal.,
2016;Halletal.,2014;Tsaietal.,2009;Weberetal.,2008).Whenlookingatthe
distributionofthecalibers,itbecomesobviousthatthemostfrequentvesseltypeisthe
capillary,irrespectiveofthespecies(Weberetal.,2008).Thecapillarynetworkcanbe
regardedasaredundantnetworkwithameshwidthofapproximately50µm(Section5.3).
Thismeshwidthisprobablyadjustedtothediffusionconstantofoxygeninbraintissue.
Theoverallvascularvolumefractionrangesbetween1and3%ofthetotalbrainvolume,
dependingonthespeciesandtheappliedmethod(Blinderetal.,2013;Lauwersetal.,2008;
Risseretal.,2009;Tsaietal.,2009;Weberetal.,2008).Althoughsimilarvascularnetwork
characteristicscanbefoundthroughoutthecortex,therearevariations,bothacrossthe
corticallayersandbetweendifferentcorticalareas(Figure2).Intheprimarysensoryareas,
thehighestvasculardensitycanbefoundinlayerIV.Thisismostobviousintheprimate
primaryvisualcortex,wherelayerIVcβ displaysanincreaseinvasculardensitythatiseasily
detectable(FontaandImbert,2002;Weberetal.,2008).
Severalauthorshaveinvestigatedtherelationshipbetweenvascularandneuronaldensity,
buttheevidencesuggeststhattherelationshipisweak(Tsaietal.,2009;Weberetal.,
2008).Inthemacaquecortex,thiscorrelationwasparticularlyweakintheupperlayersof
cortex(Weberetal.,2008).However,amuchstrongercorrelationbetweenthepatternof
oxidativemetabolismandmicrovasculardensitywasobserved(Kelleretal.,2011;Weberet
al.,2008).
TheKleinfeldgrouphasstudiedthemousebarrelcortexandfoundthatthevariationin
cellulardensityacrossthecorticaldepthwasmorepronouncedthanthatofthevascular
density(Tsaietal.,2009).Moreover,thevascularradiusdistributionwasverysimilaracross
thecorticaldepth.Ifwelookatthemicrovasculardensityalongthecorticallayersinthe
primaryvisualcortex,thereisamarkedgradientwiththehighestdensityinlayerIVc-
β (Weberetal.,2008;Zhengetal.,1991).
Fromadevelopmentalperspective,FontaandImbert(2002)wereabletodemonstratethat
therelativevasculardensitydevelopedinparalleltocytochromeoxidaseactivityandwas
highestinlayerIVc-α inthefirstpostnatalmonth.(Cytochromeoxidaseactivityisa
histochemicalproceduretoassesstheoxidativemetabolicdemandofagivenbrainregion.)
Inthenextdevelopmentalphase,thevasculardensityandcytochromeoxidaseactivityin
thetwolayersaresimilar,beforelayerIVc-βeventuallybecomesthemostdensely
vascularizedlayer.Inthecat,asimilarchangeinsteady-statemetabolicdemandand
vasculardensitywasobserved.ItwasfoundthatlayerIVinthecatstriatearea17showed
thehighestrelativevasculardensityandrelativeglucoseutilization.However,thislaminar
differencecouldonlybefoundintheadultbutnotin5-week-oldkittens(Tiemanetal.,
2004).Changesinvascularorganizationcanalsobeobservedintheadultanimal.Whereas
acutehypoxiaiscompensatedwithanincreaseinCBF,prolongedhypoxialeadstoamarked
7
increaseincapillarydensity(DiemerandHenn,1965;Hariketal.,1994;LaMannaetal.,
1992;MillerandHale,1970;Opitz,1951).
Weberandcolleaguesstereologicallydeterminedvasculardensityvaluesonthebasisof
anti-collagenimmunohistochemistryfromalargenumberofsamplesfromthemacaque
visualcorticesV1,V2,V3andV4(Weberetal.,2008).Theauthorsfoundthattheoverall
vascularlengthdensityinvisualgraymatterlayaround478mm/mm3,whereasthevolume
fractionwasapproximately2.1%.Aselaboratedabove,thevascularnetworkofthestriate
cortexalsodisplaysaspeciallaminarorganization.Thevolumefractionandlengthdensityin
layerIVc-βwere2.70%and627.83mm/mm3respectively.Thelowestvascularlength
densitywasfoundinlayerI(408.44mm/mm3),whereaslayerIIshowedthelowestvolume
fraction(1.93%).Thesesuperficialdensitydatacoincidewiththerelativelylowcellbody
densityandhighproportionofmyelinatedandunmyelinatedfibersinlayerIandII.The
vasculardatafromV2,V3andV4wereverysimilar,withlayerIVshowingthehighest
density(volumefraction2.18%)andlayerVIthelowest(volumefraction2.01%).Itis
importanttonote,thatdifferentvesselcaliberscontributedifferentiallytothedifferent
vasculardensitymetrics.Whenfocusingonthemerelengthasinlengthdensity,itis
obviousthatbyfarthelargestcontributionisprovidedbythecapillaries.However,forother
quantitiesthecontributionofthenon-capillariesmaybehigher,accordingtotheir
respectivedependenceonthevesselradius.(Thelengthdensity’sdependenceonthevessel
diameterisd0,thatofthesurfacedensityisd1andthatofthevolumefractionisd2.)Dueto
thevolumefraction’squadraticdependenceonthediameter,thecontributionoflarge
vesselssurpassesthatofthecapillaries,despitetheirmuchlowerfrequency(Figure2).
Theprimaryvisualcortexseemstobeuniquewithrespecttothemicrovascularsystemas
well,sinceoverallvasculardensityisclearlyhigherinthestriatecortexthanintheextra-
striatecortices(FontaandImbert,2002;Tiemanetal.,2004;Zhengetal.,1991).However,
differencesbetweentheextra-striateareasaresmallorevennegligible(Weberetal.,2008).
Thevasculardensitydifferencesbetweentheprimaryandnon-primaryareaswerealso
foundinthesomatosensoryandauditorycortex,andthevasculardensitiesofthe
secondaryauditoryandsomatosensoryareaswerecomparabletothoseoftheextrastriate
cortex(ownunpublishedobservations).Takentogether,itseemsthatnon-primarycortical
areasshareasimilarmicrovasculararchitecture.Therefore,basedonthevascularstructure
alone,itappearsthatlargedifferencesinthehemodynamicresponseandlaminarMRIare
nottobeexpectedwithintheseareas.However,cautionisadvisedwhenhemodynamic
responsepatternsaredirectlycomparedbetweenprimaryandnon-primaryareas(Section
8).
Itissurprisingthatwedonotseesignificantdifferencesinthevasculardensityoflarge
vesselsovercorticaldepth,despitethefactthatDuvernoyetal.(1981)statesthatDAsand
AVsofgroup4(Table1)arethemostfrequentgroupofpenetratingtrees.Firstofall,it
shouldbenotedthatDuvernoy’sclassificationhasbeenderivedforthevasculatureofthe
humanbrainandthusspeciesdependentdifferencesmightexist.Furthermore,differences
invesseldiameterfordifferentgroupsofpenetratingtreesmightblurthosecharacteristics.
Inadditiontotheoverallvasculardensity,therelativevolumefractionoccupiedbyblood
vesselsofdifferentcategories,i.e.arteries,capillariesandveins,isrelevantformany
applications.Forexample,thefractionofbloodfoundinthedifferentvesseltypesisan
importantparameterincompartmentalbiophysicalmodels.
Themeasurementsfromthemacaquevisualcortexdiscussedaboveestimatedthat
capillaries(definedasvesselswithdiameter<8µm)madeupapproximately41%ofthe
8
totalvascularvolumefraction(Weberetal.,2008).Thisissimilartoanestimateof~48%
obtainedfromIndianinkinjectedsectionsofhumancortex,whichweretakenfromthe
fusiformandparahippocampalgyri(Lauwersetal.,2008).Whilethereisarangeof
microvasculardatacomparingtherelativenumberofdescendingarteries(DAs)and
ascendingveins(AVs,Section5.2),toourknowledgenosuchdataexistdescribingthe
distributionofbloodvolumebetweenarterialandvenouscompartments.However,itis
possibletousedatafromarangeoflowerresolutiontechniquessuchMRIandPET,and
basedonthisapproachBarrettetal.(2012)estimatedarteriesmakeup~29%oftotalblood
volume,andveinscontribute~27%,usingmostlyhumanandprimatedata.
Figure2Vasculardensityofthecorticaldepth.Vasculardensityovercorticaldepthfordifferentspecies,differentareas
anddifferentvesseltypeswithsimilar(A)anddifferent(B)x-axesscales.Datafrom:Lauwersetal.(2008)(Human);Weber
etal.(2008)(Macaque);Risseretal.(2009)(Marmoset);Tsaietal.(2009)(Mouse).Thecut-offdiametertodifferentiate
betweencapillariesandlargevesselsdiffersfordifferentspecies.Here,thefollowingvalueswereused:Human:10µm,
Macaque:6µm,Marmoset:11.2µm,Mouse6µm.
9
5 Topologicalaspectsofthecorticalvascularsystem
InSection3weintroducedthegrossanatomicalstructureofthecorticalvasculature
consistingofpialandpenetratingvesselsandthecapillarybed.Here,wefocusonspecific
topologicalcharacteristicsandontheroleofthethreevesseltypesinthedistributionof
blood.Aprofoundknowledgeofthevasculartopologyisrelevantnotonlyfor
understandingneurovascularcouplingbutalsotocommentontheseverityofvessel
occlusionatdifferentlocations.Unlessstatedotherwisethedescribedcharacteristicsare
validacrossspecies.
5.1 Thepialnetwork
Thepialnetworkisa2-D-planarnetworklocatedatthecorticalsurface.Thepialarteries
distributebloodfromthelargecerebralarteriestotheintracorticalvesselsandthepial
veinscollectit.
Asthepialarteriesareatthebeginningofthepathwayofbloodthroughthecortex,a
robustnetworktopology,whichguaranteesaconstantsupplyofblood,iscrucial.Thisis
achievedbyalargenumberofarterialanastomoses,whichonaveragecontainfouredges
(Blinderetal.,2010;Duvernoyetal.,1981;Schafferetal.,2006).Overall,thestructureof
thepialarterialnetworkiscomparabletothatofahoneycomb(Blinderetal.,2010).
Forathoroughanalysisoftheredundanciesatthepiallevel,Blinderetal.(2010)introduce
theconceptofthebackboneofthepialnetwork(Figure3A).Thebackbonespansthewhole
territoryofthepialnetwork,eventhoughitismadeupofonly~11%ofthepialarteries.In
addition,nearly75%ofalldescendingarteries(DAs)startatabackboneedge.Thisproperty
furtherincreasestherobustness,becausetwodifferentpialvesselscanfeedthosearteries.
Occlusionexperimentshavedemonstratedtherobustnessofthepialnetwork(Blinderet
al.,2010;Schafferetal.,2006).Schafferetal.(2006)showedthatsinglevesselocclusion
inducesaredistributionofflowwiththepositiveeffectthatallvesselsstayperfused(Figure
3F).Nonetheless,theflowratesinindividualvesselsaresignificantlyaffectedandflow
reversals,reductionsandevenincreasesareobserved(Schafferetal.,2006).Blinderetal.
(2010)notedthatinlowfluxDAstheflowispreservedinresponsetopialarteryocclusion,
whileitdecreasesinhighfluxDAs.ThissuggeststhathighfluxDAscontainbloodreserves
thatcanberedistributedincaseofocclusion.
Thenetworktopologyofthepialveinsissignificantlylessstudied.Theprevailingviewisthat
ithasfeweranastomosesthanitsarterialcounterpartandisrathera“drainagesystemlike
ariverwatershed”(Adamsetal.,2014).Generally,pialveinsarelargerindiameterthanpial
arterioles.Duvernoyetal.(1981)observethatinthehumanbrainlargepialveinstendto
surpassthesulciandremainatthecorticalsurface.Thisisincontrasttopialarterieswhere
themaintrunkisoftenlocatedwithinthesulci.Furthermore,pialarteriesnormallyrun
abovepialveins(Duvernoyetal.,1981).
5.2 Thepenetratingvessels
TheDAsdeliverbloodtothecapillarybedovertheentiredepthofthecortex.Afterthe
bloodpassesthecapillarynetworkitiscollectedintheascendingveins(AVs)andreturns
towardsthecorticalsurface.ItiswellestablishedthatDAs,aswellasAVs,haveatree-like
10
structureanddiffersignificantlyintheirpenetrationdepth(Blinderetal.,2013;Cassotet
al.,2009;Duvernoyetal.,1981;Guibertetal.,2012;Hirschetal.,2012;Lauwersetal.,
2008;Reina-DeLaTorreetal.,1998).Themostwidespreadclassificationforpenetrating
vesselsisbyDuvernoyetal.(1981),andissummarizedinTable1.Theauthorsproposed
thatdifferentgroupsofDAsareresponsibleforfeedingdifferentcorticallayers.Guibertet
al.(2012)providedsomeevidenceforthishypothesisintheirnumericalwork,andshowed
thatthedepthofthefeedingregionstronglycorrelateswiththepenetrationdepthofthe
DA.
VesselGroup 1 2 3 4 5 6
Penetration
depth(cortical
layer)
I-II IIIa IIIc-Va VI downto
WM
downtoWM
without
branching
Table1ClassificationofpenetratingvesselsbasedontheirpenetrationdepthbyDuvernoyetal.(1981).Theclassification
isvalidforDAsaswellasAVs(withtheexceptionofgroup6whichonlyexistsforDAs).Allpenetratingtrees,with
exceptionofgroup6,haveoffshootsalongtheirdepth.WM:whitematter.
TheDAsbelongingtogroup4arethemostnumerous(Duvernoyetal.,1981),andthe
numberofbranchesfeedingthecapillarybedpeaksatcorticallayerIV(Figure3E)(Blinder
etal.,2013;Schmidetal.,2017).WhilethisevidencesuggeststhattheDAtopologyis
designedwiththemajorpurposeofsupplyingbloodtolayerIV,topologicalcharacteristics
alonedonotpredictflow.Indeed,Schmidetal.(2017)showedthat,incontrasttothe
numberofcapillarystartingpoints,theRBCinfluxismaximalclosetothecorticalsurface
(Figure3E).
OffurtherinterestisthedistributionofDAswithrespecttoeachotherandtotheAVs.The
ratioofDAstoAVsishighlyspeciesdependentandwhileforprimatestherearemoreDAs
thanAVs,theoppositetrendisobservedforrodents(Table2).Theevolutionarybasisfor
thesedifferencesremainsunclear.
Species Human Monkey Rat Mouse
RatioDA:AV 2.2:1 2.1:1 1:1.8 1:3.0
DAspermm2 1.0 7.9 8.3 3.9
AVspermm2 0.5 3.6 10.3 -
Table2AverageratiosofDAstoAVs,averagenumberofDAsandAVs2fordifferentspecies.References:Human:(Cassot
etal.,2009;Lauwersetal.,2008);Monkey:(Adamsetal.,2014;Guibertetal.,2010;Risseretal.,2009;Weberetal.,
2008);Rat:(Blinderetal.,2010;Nguyenetal.,2011);Mouse:(Blinderetal.,2010;Blinderetal.,2013).
IthasbeenhypothesizedthatthedistributionofDAscorrelateswiththelocationof
functionalneuronalunits,suchasbarrels(somatosensorycortex)orblobs(visualcortex).
Forthebarrelcortexnosuchcorrelationcouldbedetected(Figure3B;Blinderetal.(2013)).
Fortheblobsthematterremainscontroversial.Kelleretal.(2011)observedanincreased
DAdensitybetweentheblobs.However,thisisincontrasttoresultsfromAdamsetal.
(2014),whodidnotobserveanydifferencesintheDAdensity.
Althoughthereisnoclearevidencesupportingastrongcorrelationbetweenfunctional
unitsandthedistributionofDAs,itseemslikelythateachDAisresponsibleforfeedinga
11
specifictissuevolume.Shihetal.(2013)occludedDAsintheratcortextoestimatethe
feedingregionofindividualDAs.Theystatethattheinfarctvolumeisproportionaltothe
baselinefluxintheoccludedDAandonaverageaffectsacylindricalvolumewitharadiusof
460µmandadepthof1.17mm.
Guibertetal.(2012)determinedthefeedingvolumebasedonnumericalbloodflow
simulationsinthemarmosetcortex.Theyobtainedafeedingvolumewitharadiusof386
µmandadepthof2mm.Consideringthelargesizeofthebrainthisvaluemightseem
comparablysmall.However,itshouldbekeptinmindthattheDA:AVratiois>1for
primateswhileitis<1forrodents.Furthermore,thevolumemeasurementpresentedby
Guibertetal.(2012)isaconservativeestimatebecauseitisdefinedasthevolumewhichis
exclusivelyfedbyoneDA,andnotthewholevolumeaffectedbyoneDA.Accurate
comparisonsarealsocomplexbecausetheDAdensityincreaseswiththedistancetothe
originofthemiddlecerebralartery(MCA),andconsequentlythefeedingvolumeislikelyto
decrease(Blinderetal.,2010).Additionally,thedensityofpenetratingvesselsdiffers
dependingoncorticalarea(Risseretal.,2009).
ThestudiesbyNishimuraetal.(2007)andBlinderetal.(2013)furtherunderlinethecrucial
roleofDAsinthesupplyofblood(Figure3F).TheeffectofaDAocclusionisapparentupto
tensegmentsfromthesiteofocclusion(Blinderetal.,2013).Thereisstillincreased
variabilityinRBCvelocitiesevenafteradistanceof350µm,wherethemeanRBCspeed
maintainsitsbaselinevalue(Nishimuraetal.,2007).
EventhoughevidencesuggeststhatanAVocclusionisascriticalasaDAocclusion(Shihet
al.,2013),theAVshavereceivedsignificantlylessattentionthantheDAs.Nguyenetal.
(2011)hypothesizedthatiftheratioofDAstoAVsislargerthan1anAVocclusionismore
severeandviceversa.Forthemarmoset,macaqueandhuman,wheretheratioofDA:AVis
greaterthan1,thedrainingregionofoneAVissignificantlylargerthanthefeedingregionof
oneDA(Guibertetal.,2012;Lorthoisetal.,2011,Weberetal.,2008).Furthermore,
Lorthoisetal.(2011)showedthatthedrainagevolumeincreaseswiththediameterofthe
AV.
Thediameterofthepenetratingvesselsisstronglyspeciesdependent.Blinderetal.(2013)
statethatinthemousethemeandiameteris11µmand9µmforDAsandAVs,
respectively.Opposingtrendshavebeenrecordedinthehumanbrain,herethevessel
diameterofAVsislargerthantheoneofDAs(65µmvs.35µmforpenetratingvesselsof
group4)(Duvernoyetal.,1981).Moreover,themeanvesseldiameterincreaseswiththe
penetrationdepthoftheDA/AV.Interestingly,incontrasttotheDAs,thenumberofAV
offshootsdecreaseswithcorticaldepth(Blinderetal.,2013).
RelativelylittleisknownabouttherelativeplacementofDAsandAVs.Basedontheanalysis
ofmicrovascularnetworkfromthemousecortexBlinderetal.(2013)suggestedarhombic
latticewhereoneDAisencircledbysixAVs.Acomparableconceptwasproposedby
Duvernoyetal.(1981)whostatethata“vascularunit”inthehumanbrainconsistsofanAV
surroundedbyseveralDAs(PleasebearinmindthatinthehumanbraintheDA:AVratio
approximately2).FromasolefluiddynamicalpointofviewitisplausiblethattheAVs/DAs
arefedequallybythecentralDAs/AVs.However,opposingtrendshavebeenobservedin
numericalworksfromGuibertetal.(2012)andSchmidetal.(2017).Guibertetal.(2012)
statethat61.5%oftheflowofoneDAisdrainedintooneAV.AdditionallySchmidetal.
(2017)observethat72%ofRBCsdrainedbyanAVoriginatefromoneDA.Those
observationscanonlybeexplainedbyaspecificmicrovasculaturetopologyorflowpattern
thatfavorsthedirectionalflowtowardsoneAV.Nonetheless,ithastobekeptinmindthat
12
althoughthereseemstobeapreferentialAVforeveryDA,theDAsandAVsarehighly
interconnected.Indeed,Guibertetal.(2012)showedthatinthemarmosetoneDAis
connectedto22AVs.
Allinall,multipleexperimentalaswellasnumericalstudiesdemonstratethatthe
penetratingvesselsarecrucialtomaintainasufficientsupplyofbloodtothetissue.
However,manyopenquestionsconcerningtheirdistributionandlayer-specificfeeding
remaintobeanswered.
5.3 Thecapillarybed
Thecapillarybedhasamesh-likestructure,generallydescribedashomogeneousandhighly
interconnected(Blinderetal.,2013),andhenceitstopologyissignificantlymoredifficultto
analyzethanthepialvessels’orthepenetratingtrees’(Blinderetal.,2013;Cassotetal.,
2009;Hirschetal.,2012;Lauwersetal.,2008).Nonetheless,theflowfieldinthecapillary
bedisveryheterogeneouswithalargerangeofRBCvelocitiesandahighcapillarytransit
timeheterogeneity(JespersenandØstergaard,2012).
Thisisalsoreflectedinthenumberofavailablepathwaysthroughthecapillarybedandthe
frequencieswithwhichthesearechosen.FiveexemplaryRBCpathwaysthroughthecortical
microvasculatureareillustratedinFigure3D.Werecentlyshowedthatforeachcapillary
startingpointthereareonaverage8differentRBCpathwaysleadingfromDAtoAV.
However,formorethan50%ofallcapillarystartingpointsthereisapreferentialpathwhich
ischosenwithafrequency>50%(Schmidetal.,2017).Furthermore,ourresultsrevealed
thatthecapillarystartingpointandthecapillaryendpointarestronglycorrelated(Figure
3C)andhence,theRBCtendtomove“in-plane”throughthecapillarybed(Schmidetal.,
2017).Itseemslikelythatthecapillarybedisdesignedtoencouragethein-plane(i.e.
paralleltothesurface)motionofRBCs.
Allinall,eventhoughthestructureofthecapillarybediscommonlydescribedasbeing
homogeneousitsflowfieldishighlyheterogeneous.Furtherinvestigationsarenecessaryto
thoroughlyexplaintheobservedcharacteristics.
Similarlytothepialandpenetratingvessels,occlusionexperimentshavebeenperformedto
assesstheoverallrobustnessofthecapillarybedinthedistributionofflow(Figure3F;
(Nishimuraetal.,2007;Shihetal.,2013)).Nishimuraetal.(2007)andShihetal.(2013)
showedthattheimpactofamicrovesselocclusionisminimalandthattheRBCflux
recoveredto45%ofitsbaselinevaluebythreebranchesdownstreamoftheocclusion.The
robustnessofthecapillarybedcanbeexplainedbyitsmesh-likestructure,whichis
beneficialforanefficientredistributionofflow.Furthermore,Guibertetal.(2012)show
that63%ofallcapillariesarefedbymorethanoneDAandhencearedundancytowardsDA
occlusionpersistsaswell.Interestingly,therobustnessofthecapillarybedtowardsDA
occlusionincreaseswithdepth(Figure3G;Guibertetal.(2012)).Ithasalsobeensuggested
thatcapillarybeddensitymightcorrelatewithneuronalfunctionalunits.Forthebarrel
cortexnosuchcorrelationwasobserved(Blinderetal.,2013).Similarlytothepenetrating
vessels,theresultsofKelleretal.(2011)andAdamsetal.(2014)divergefortheblobsinthe
visualcortex,andonlyKelleretal.(2011)notedaslightlyincreasedvesseldensitywithinthe
blobs.
13
Figure3Topologicalcharacteristicsofthecerebralvasculature.(A)Exemplarypialnetwork.Thebackboneofthepial
networkishighlightedinblack(FigurefromBlinderetal.(2010)).(B)Feedingregionsofdescendingarteries(DAs)andtheir
distributionwithrespecttowhiskerbarrels(“goldenbands”).ThecolorsindicatefeedingregionsfordifferentDAs.(Figure
fromBlinderetal.(2013).ReprintedbypermissionfromMacmillanPublishersLtd:NatureNeuroscience,2013).(C)Cortical
depth(CD)ofcapillarystartandcapillaryendpointstoillustratethein-planemotionofRBCsinthecapillarybed(CB)
(FigurefromSchmidetal.(2017)).(D)ExemplaryRBCtrajectoriesfordifferentdepthsofcapillarystartingpoints(Figure
fromSchmidetal.(2017)).(E)Layer-specificfeedingcharacteristicsofthecapillarybed(CB)(FigurefromSchmidetal.
(2017)).(F)Impactofocclusionfordifferentvesseltypes(DA:descendingarteries,C:capillaries,PA:pialarteries).The
impactismeasuredbasedonthefractionalRBCspeedinvesselsegmentsdownstreamthesiteofocclusion(D1:one
segmentdownstream,D2:twosegmentsdownstream,D3&D4:threeandfoursegmentsdownstream)(Figureadapted
from:Nishimuraetal.(2007)).(G)Robustnessindexofthecapillarybed(CB)fordifferentcorticaldepths.Therobustness
indexisdefinedasthenumberofdescendingarteriesfeedingacapillarysegment(Figureadaptedfrom:Guibertetal.
(2012)).
6 Flowregulation
Intheprevioussectionwedescribedstaticcharacteristicsofthecerebralvasculature,yet
thevasculatureisconstantlyadaptingtothemetabolicneedsoftheparenchyma.The
resultinghemodynamicchangesarethebasisofamultitudeofimagingtechniques.
Understandingthevasculartopologygoeshandinhandwithunderstandingthevascular
responsetoneuronalactivation.Hence,inthefollowingwesummarizethecurrent
knowledgeonvascularregulationmechanisms.Webeginbydescribingthecandidate
locationsforregulationandthecontractilecellsintheirsurroundings.Subsequently,we
commentonspatio-temporalandlayer-specificdynamicsofregulation.Theneurovascular
signalingpathwaysarenotaddressedinthismanuscript.Agoodoverviewisgiveninthe
reviewsbyIadecolaandNedergaard(2007),Attwelletal.(2010)andHillman(2014).
6.1 Contractilecellsofthecerebralvasculature
(A) Structure of the pial network (B) Distribution of DAs with respect
to barrels
(C) In-plane motion of RBCs
through the CB
(D) Exemplary RBC trajectories
Analy
sis
Layer
1
2
3
4
5
(E) Layer-specific
feeding of CB
(F) Vessel specific
occlusion effects
(G) Robustness index of
CB over depth
14
Twocelltypeshavebeenreportedtomodulatevasculardiameterinvivo:smoothmuscle
cellsandpericytes.Whiletheinvolvementofsmoothmusclecellsinneurovascularcoupling
iswellestablished,thecontractilityofpericytesinvivohasbeenshownonlyrecentlyand
theirroleinneurovascularcouplingisamatterofongoingdebate(Attwelletal.,2010;
Attwelletal.,2015;Fernandez-Klettetal.,2010;Fernández-KlettandPriller,2015;Hallet
al.,2014;Hilletal.,2015;Hillman,2014;ItohandSuzuki,2012).
Arterialsmoothmusclecells(SMCs)aretightlywrappedaroundarteriesandhenceperfectly
positionedtocontrolvesseldiameters(Figure4A).Pericytesdiffersignificantlyintheir
shapeandintheexpressionof𝛼smoothmuscleactin(𝛼-SMA).Whileinthesmooth
musclecellsaroundDAs𝛼-SMAisomnipresentinpericytesitonlypersistsclosetotheDA
(Figure4A;(Attwelletal.,2015;Hartmannetal.,2015;Hilletal.,2015)).However,it
remainsunclearif𝛼-SMAisaprerequisiteforthecontractilityofpericytesorifother,asyet
unknownmechanismsforalteringthediameterofcapillariesexist(Fernández-Klettand
Priller,2015).
6.2 Vascularresponseofdifferentvesseltypestoneuronalactivation
Theoverallhemodynamicresponsetoneuronalactivationresultsfromaninterplayof
differentvascularresponses(Hillman,2014).Inthissubsection,wediscusstheresponseand
spatio-temporaldynamicsofindividualvesseltypes,namely:(1)pialanddescending
arteries,(2)pre-capillaryarteriesandcapillaries,and(3)venulesandveins.Table3
summarizesthekeycharacteristicsofthevascularresponseofdifferentvesseltypestobrief
stimulation(<5s).
Ingeneralthelocationwherethevascularresponseisinitiateddependsonthecorticalarea
underinvestigationandthetypeofstimulusapplied.Fortheratprimarysomatosensory
cortexandthestimulationoftheforepawtheprevailingviewisthattheresponsestartsas
deepas0.6-0.9mminthecortexandspreadstowardsthesurfacewithapropagationspeed
of~0.9mm/s(Tianetal.,2010;Uhlirovaetal.,2016b).Thevasculatureitselfmaybe
responsibleforthepropagationofthedilatorysignal(Chenetal.,2011;Chenetal.,2014;
Hillman,2014;Iadecolaetal.,1997;Uhlirovaetal.,2016b),butitremainsunclearif
vasodilationistriggeredataspecificvesseltypeonly.
Lindvereetal.(2013)werethefirsttomonitordiameterchangessimultaneouslyina
microvascularnetworkspanning0.5x0.5x0.6mm.Theyshowthatevenalongindividual
vesselstheresponseisheterogeneous,whichaddstothecomplexityofthevascular
response.Inordertoanalyzechangesinthewholenetwork,theyintroducetheconceptof
earlyandlaterespondingvessels.Whiletheyobserveanupwardpropagationofearly
dilationsandlateconstrictions,theearlyconstrictionsandlatedilationsspreadwith
increasingdepth.
Thestimuluslengthaswellastheusedanesthesiacanplayaroleinthevascularresponse.
Drewetal.(2011)showedthatthelongerthestimulusthelargerthevasodilationuntila
plateaudilationisreached.TheeffectsofanesthesiahavebeeninvestigatedbyLyonsetal.
(2016)andMasamotoetal.(2009).Theirworksshowthatthelevelofanesthesiaaffectsthe
increaseincerebralbloodflowaswellasthetissueoxygenpartialpressure.Furthermore,
onehastobearinmindthatbaselinefluctuationsindiameterpersist,whichforthepial
vesselshavethesameorderofmagnitudeasfunctionaldiameterchanges(Drewetal.,
15
2011).Allinall,thespatio-temporaldynamicsofneurovascularcouplingarehighlycomplex
anddifficulttoelucidate,partlyduetotheirmutualinterdependence.
Vesseltype Maximum
dilation
Averagepeak
dilation
Onsettime Timetopeak
Pialarteries 31.6±4.1% 11.0±3.5% 1.0±0.1s 2.9±1.1s
Descending
arteries
31.0±5.0% 9.7±2.2% 0.8±0.2s 2.4±0.4s
Capillaries* 40% 13.3±2.0% 1.7±1.0s 2.8s
Table3:Keycharacteristicsofthevascularresponseofdifferentvesseltypes.Thegivenvaluesaretheaverage
andthestandarddeviationofallvaluesfoundinliteratureforshortsensorystimulation(<5s).Capillaries*:
Dataoninvivocapillarydilationisverysparse.Thus,wealsoconsideredthemeasurementsbyHalletal.
(2014)althoughthestimulusdurationis15s.Furthermore,forcapillariesitwasnotalwaysclearifthegiven
valuesareforactiveorpassivedilation.Maximumdilation:maximumdilationthathasbeenmeasuredin
individualvessels.Averagepeakdilation:averageoverallmeasureddilations.Onsettime:timeafterstartof
stimulusuntildilationisinitiated.Timetopeak:timeafterbeginofthestimulusuntilthemaximumdilationis
reached.Referenceliterature:Pialarteries:(Devoretal.,2008;Devoretal.,2007;Drewetal.,2011;Hillmanet
al.,2007;Sekiguchietal.,2014;Tianetal.,2010;Uhlirovaetal.,2016b).Descendingarteries:(Sekiguchietal.,
2014;Tianetal.,2010;Uhlirovaetal.,2016b).Capillaries:(Halletal.,2014;Tianetal.,2010).
Pialanddescendingarteries:
AsmentionedinSection5.1thepialnetworkisresponsibleforarobustsupplyofbloodto
differenttissueregions.Furthermore,ithasbeenhypothesizedthatitredistributesblood
duringneuronalactivation(Devoretal.,2007;Shihetal.,2015).
Asthepialvasculatureiscomparablyeasytoaccess,manyworkshaverecordedvascular
changesatthepiallevel(Chenetal.,2014;Devoretal.,2008;Devoretal.,2007;Drewet
al.,2011;Hillmanetal.,2007;Ngaietal.,1988;NgaiandWinn,2002;Sekiguchietal.,2014;
Tianetal.,2010;Uhlirovaetal.,2016b).Despitedifferencesinmethodologyand/orspecies,
allstudiesagreethatmultiplepialarteriesaltertheirdiameterinresponsetoneuronal
activation.
Thelargestchangesarelocatedclosetothecenterofactivation(CoA),andtheamplitudeof
therelativediameterchangedecreaseswiththedistancefromtheCoA(Figure4B).Positive
diameterchangeshavebeenobservedupto3mmawayfromtheCoA;however,for
distances>2mmthevascularresponseispredominantlynegative(i.e.moreconstrictions
thandilations(Devoretal.,2008;Devoretal.,2007)).
Formanypialarteriesthereisaperiodofconstrictionafterstimuluscessation,forreasons
notcurrentlyunderstood(Figure4B).However,theamplitudeofconstrictionissignificantly
smallerthanthatfordilation(Devoretal.,2007;Drewetal.,2011;Hillmanetal.,2007;
Uhlirovaetal.,2016b).
TheDAsfeedthecapillarybedwithbloodfromthecorticalsurface.Astheyposethe
“bottleneckofperfusion”(Nishimuraetal.,2007)itseemslikelythattheyarealsoideally
placedforalocalizedincreaseinbloodflow,andthereisalargebodyofevidenceshowing
thedilationofDAsduringactivation(Attwelletal.,2010;Halletal.,2014;Hillman,2014;
IadecolaandNedergaard,2007;Lindvereetal.,2013;Tianetal.,2010;Uhlirovaetal.,
2016b).BasedontheresultsofHalletal.(2014)~50%ofthemonitoredDAsrespondedto
16
neuronalactivation.However,itremainsunknownhowthosearedistributedwithrespect
totheCoA.
RecentstudieshavereportedthatthedilationofDAsisinitiateddeepinthecortex
(measurementsupto0.9mm)andpropagatestowardsthecorticalsurface(Figure4C,
(Lindvereetal.,2013;Tianetal.,2010;Uhlirovaetal.,2016b)).
Uhlirovaetal.(2016b)showedthatapproximately50%ofallrespondingDAsexperiencea
post-stimulusconstrictionphase.Asforthepialvessels,theamplitudeofconstrictionis
significantlylowerthanfordilation.
Itisstilldebatediftheamplitudeofdilationisafunctionofcorticaldepth,andresults
concerningthismatterhavebeendiverging(Figure4C)(Lindvereetal.,2013;Sekiguchiet
al.,2014;Tianetal.,2010;Uhlirovaetal.,2016b).EventhoughitiswellknownthatDAs
dilateduringactivationthedetailedpatternsoftheirresponsestillhavetobeelucidated.
Althoughlackingtheresolutiontodistinguishbetweenchangesoriginatingfrompialand
descendingarteries,arangeofMRItechniqueshavealsoreportedsubstantialincreasesin
arterialbloodvolumeuponstimulation,generallyconsistentwiththediameter
measurementsdiscussedabove(Hoetal.,2011;Kimetal.,2007;KimandKim,2010).
Afurtherinterestingaspectofthevascularresponseisthereturntobaseline.However,this
aspectislessstudiedthantheonsetofdilationandstronglydependsonthedurationof
stimuli.Ithasbeenhypothesizedthatthereisadistinctregulationmechanism,whichis
responsibleforthedecayphase(Chenetal.,2011).Approximately4safterstimulus
cessation,thepialvesselsreachtheirbaselinediameteratwhichtheyremainorcontinueto
constrict(Devoretal.,2008;Sekiguchietal.,2014;Tianetal.,2010).Thereturntobaseline
ofpialvesselsdoesnotdependonthedistancetotheneuronalCoA(Devoretal.,2007).
Pre-capillariesandcapillaries:
Overthepastyearsmoreevidencefortherelevanceofcapillariesforneurovascular
couplinghasemerged(Halletal.,2014;Tianetal.,2010).Ascapillariesarethevesselsmost
proximaltothelargestpartoftissueitseemsplausiblethattheymayplayaroleintheup-
regulationofflow,oxygenandenergysubstratedelivery.However,thereisrelativelylittle
dataavailable,anditischallengingtodifferentiatebetweeneffectsresultingfromarteriole
andcapillarydilationinvivo.Inthefollowing,wecommentondirectandindirectevidence
thatsupportsthehypothesisofanactiveinvolvementofcapillariesinneurovascular
coupling.
AcrucialargumentforactivecapillarydilationhasbeenprovidedbyHalletal.(2014).They
demonstratethatcapillariesdilateonaverage1.4spriortoarterioles,whicheliminatesthe
possibilityofapurelypassiveresponseofcapillaries.Additionally,theyreportthat
capillariesaremorelikelytorespondinthevicinityofpericytes(50%vs.22%response
frequency)andthatthefrequencyofrespondingcapillariesdecreaseswithbranchingorder.
Thisobservationisinlinewithadecreasein𝛼-SMAforpericyteslocatedathigher
branchingorders(Attwelletal.,2015;Hartmannetal.,2015;Hilletal.,2015).Similarlyto
DAs,capillarieslocateddeepinthecortexdilateearlierthantheonesclosetothecortical
surface(Tianetal.,2010).Whetherornottheresistanceatthelevelofsmallvesselsis
regulatedbypericytesorsmoothmusclecellsremainsamatterofongoingdebate(Attwell
etal.,2015;Hartmannetal.,2015;Hilletal.,2015),whichmaybecausedinpartby
differentdefinitionsofpericytesand/orcapillaries.
However,theobservationsbyHalletal.(2014)areincontrasttotheresultsofTianetal.
(2010)whostatethatthedilationspreadsfromtheDAstothecapillarybed.Theworkby
17
Chenetal.(2011)supportsanonsetofvasodilationatthecapillarylevel,becausethey
reportedanincreaseintotalhemoglobinintheparenchymapriortotheincreaseinthe
arterioles.Theoriginofthosedifferencesisnotyetclear.
Anumberofstudiesprovidefurtherindirectevidenceforanactiveregulationmechanismat
thecapillarylevelleadingtoahomogenizationofflow(Gutiérrez-Jiménezetal.,2016;Lee
etal.,2014;Leeetal.,2015;Stefanovicetal.,2008).Leeetal.(2015)measuredtheRBCflux
in~200capillariessimultaneously,duringbaselineandactivation,andshowedthatthe
standarddeviationofRBCfluxdecreased2spriortotheincreaseinthemean.Thisagrees
withtheobservationsofStefanovicetal.(2008)andGutiérrez-Jiménezetal.(2016),who
reportedthatlowbaselinefluxcapillariesexperiencealargerresponsetostimulation.
Furthermoreduringactivation,thecapillarytransittimeheterogeneityisreduced
(Gutiérrez-Jiménezetal.,2016).ThealteredRBCvelocitiesandthehomogenizationofflow
followingneuronalactivationmayalsopartlyresultfromapO2-dependentincreaseinRBC
flexibility(Weietal.,2016).
Itshouldalsobekeptinmindthatcapillarydilationcanbeaneffectivemeanstoalterthe
distributionofRBCs(Schmidetal.,2015),andthatflowhomogenizationduringneural
activationmayplayanimportantroleinoxygendelivery(BarrettandSuresh,2013;
JespersenandØstergaard,2012;Vazquezetal.,2008).
VenulesandVeins
Noactivedilationofvenulesandveinshasbeenreported;however,thereareconflicting
reportsaboutthepresenceand/orsignificanceofpassivedilationofvenulesandveins.For
example,directopticalmeasurementsofvenousdiameterhaveshowneithernoincrease
(Hillmanetal.,2007)orverysmallincreases(Drewetal.,2011),whereasMRI-based
approacheshavereportedconsiderableincreasesinvenousbloodvolume(ChenandPike,
2009;ChenandPike,2010).However,usingabiophysicalmodel,Barrettetal.(2012)
demonstratedthatthesediscrepanciescouldbeexplainedbydifferencesinstimulation
length(typicallyshortinopticalimagingexperiments,butlonginMRIexperiments)andthe
factthatevenrelativelysmallchangesincapillaryand/orvenousdiameterscanleadtolarge
changesinvenousCBV.Takenalongsidetheexperimentalevidence,thisresultsuggests
that,forbriefstimulation,dilationofarteriesandarteriolescontributesthemajorityof
bloodvolumeincreases;however,dilationofpost-arteriolarvesselsisrelevantduring
longerstimulation(>10s).
6.3 Layer-specificregulation
Inthecortexthedistributionofneuronsandconsequentlythemetabolicneedsvaryover
depth.Wehavealreadybrieflycommentedonthepropagationofvasodilationoverdepth.
Yetthequestionremainswhetherthosedifferencesarealsoanindicationforlayer-specific
regulationmechanisms.Here,wesummarizetheavailableevidencefortheexistenceof
layer-specificregulation,andwediscusswhylayer-specificregulationseemsplausiblewith
respecttotheflowfieldfromafluiddynamicspointofview.
Directevidenceforlayer-specificregulationmechanismsisstillrelativelysparse.To
investigatewhethertheresponsetostimulationvariesoverdepth,Gutiérrez-Jiménezetal.
(2016)measuredRBCvelocity,RBCfluxandcapillarytransittimewithtwo-photon
18
microscopy.ArangeofhighresolutionMRIstudieshavealsoobservedlayer-dependent
changesinCBFand/orCBV(Goenseetal.,2012;Hiranoetal.,2011;Hoetal.,2011;Huber
etal.,2015;Huberetal.,2016;KimandKim,2010,2011;Zhaoetal.,2006).Both
approachesagreethatthehemodynamicresponsevariesoverdepth,whichpointstowards
layer-specificregulationmechanisms.However,thesedifferencescouldalsoresultpassively
fromlaminarvariationsinvasculartopology.
Inthecerebralmicrovasculaturedepth-dependentflowandpressurecharacteristicspersist
(Figure4D)andprovidefurtherevidencewhylayer-specificregulationmightbebeneficial
(Gutiérrez-Jiménezetal.,2016;Kleinfeldetal.,1998;Leeetal.,2014;Schmidetal.,2017).
WerecentlyanalyzedthepressuredropalongthepathwayofindividualRBCsandshowed
thatthedeepertheRBCentersthecapillarybedthelargerthepressuredropintheDAand
thesmallerinthecapillarybed(Figure4E,Schmidetal.(2017)).Thepressuredrophasa
strongimpactontheincreaseinflowrateresultingfromdilation.Hence,ourresultssupport
thehypothesisthatlayerspecificregulationcouldbeadvantageous.
Additionally,variousexperimentalaswellasnumericalworksobservethattheRBCvelocity
decreasesoverdepthandconsequentlythecapillarytransittimeincreases(Figure4D)
(Gutiérrez-Jiménezetal.,2016;Kleinfeldetal.,1998;Leeetal.,2014;Schmidetal.,2017).
ThesefactorsstronglyimpacttheamountofoxygendischargedfromRBCsandthereforeit
seemslikelythattheoxygendischargealsovariesoverdepth(Section7.3).
6.4 Dimensionoftheareaaffectedbyneuronalactivation
Inordertodiscussthedimensionsoftheareaaffectedbyneuronalactivationitisimportant
todistinguishbetween(1)theareainwhichthevasculaturerespondstostimulationand(2)
theareathatisaffectedbythealteredvesseldiameters.
Relativelylittleisknownabouttheprecisespatialpatternofthevascularresponse.Atthe
pialleveldiameterchangeshavebeenrecordedupto3mmapartfromtheCoA(Devoret
al.,2008;Devoretal.,2007).Basedontheconceptofretrogradepropagationitseemslikely
thatthosevesselsarelocatedupstreamoftheneuronalCoA(ErinjeriandWoolsey,2002;
Iadecolaetal.,1997).However,theresultsfromChenetal.(2011)suggestthatthe
vasodilationspreads“spatiallyoutwards”andisindependentoftheflowdirection.Evenif
theprecisepatternofvesselrecruitmentisnotyetfullyunderstood,aselectiverecruitment
ofvesselsseemslikely(ErinjeriandWoolsey,2002;Hillman,2014).Abetterknowledgeof
thesignalingpathwaysandmeasurementsofdiameterchangeswithhightemporaland
spatialresolutionarenecessarytoadvanceourunderstandingofthevascularresponse
patterns.
Thedimensionoftheareathatisaffectedbythealteredvesseldiameterstronglydepends
onthelocationofthevesselalongthevascularpathway:thefurtherupstreamthelarger
theareaaffected.IllustrativeproofisgiveninthenumericalworkbyReicholdetal.(2009).
TheyshowedthatifthesiteofdilationalongtheDAisclosetothecorticalsurfaceitsarea
ofinfluenceislargerthanifitislocatedfurtherdownstream.
Toinvestigatethisinvivoisdifficult,becausechangesinalargevolumehavetobe
monitoredsimultaneously.Frequently,techniqueswhichmeasure2Dprojectionsofthe
flowfieldareapplied.Forexample,Dunnetal.(2005)usemulti-wavelengthreflectance
imagingtoestimatethespatialextentofhemodynamicchangesduringfunctional
19
activation.Theyshowthatthesurfaceareawhereachangeintotalhemoglobinisnoted
differsforforepawandwhiskerstimulation:~2.0mm2and~1.6mm2,respectively.
Figure4Vascularresponsetoneuronalactivationandflowcharacteristics.(A)Schematicrepresentationofsmoothmuscle
cellandpericytesinthecerebralvasculature(Figurefrom:Hartmannetal.(2015)).(B)B.1:Responseofpialarteriesto
stimulationasafunctionoftimefordifferentdistancestotheneuronalcenterofactivation(CoA).Distances:red:0-0.5
mm,darkred:0.5-1.5mm,orange:1.5-2.5mm.B.2:Maximumdilation/constrictionforpialarteriesasafunctionofthe
distancetotheCoA(Figureadaptedfrom:Devoretal.(2007).Pialvasculatureininsertadaptedfrom:Schafferetal.
(2006)).(C)C.1:Responseofdescendingarteriestostimulationasafunctionoftimefordifferentdepths.C.2:Onsettime
andtimetopeakfordescendingarteriesasafunctionofcorticaldepth(Figureadaptedfrom:Uhlirovaetal.(2016b).
Schematicdrawingofadescendingarteryadaptedfrom:Duvernoyetal.(1981)).(D)RBCvelocityinthecapillarybedover
depth(Figurefrom:Schmidetal.(2017)).(E)PressuredropalongRBCtrajectoriesforanalysislayer1(AL1:0–200µm
corticaldepth)andanalysislayer5(AL5:800–1000µmcorticaldepth)fordifferentvesseltypes.Theanalysislayersare
200µmthickslicesforwhichdifferentflowcharacteristicsareanalyzed.(Figurefrom:Schmidetal.(2017)).
7 Oxygenation
Impairmentstocerebralbloodflowontheorderofafewminutesaresufficienttocause
irreversiblehypoxicischemicinjury(Jonesetal.,1981;Moskowitzetal.,2010).Therefore
supplyingoxygen,alongwithmetabolicsubstrates,isthemostcriticalroleofthebrain
vasculature.Inthissectionweprovideabriefoverviewofcerebraloxygenation,especially
asitrelatestolaminarfMRI.
7.1 Baselineoxygenation
Inthetraditionalviewofoxygenationinthebrain,arteriesprovideconstantsupplyof
oxygenatedblood,capillariesarethesiteofoxygendelivery,andveinsdraindeoxygenated
blood.However,evidencehasemergedthatchallengesthisconcept.Forexample,several
studiesreportedthatbloodincorticalarteriesandparticularlyarteriolesisnotcompletely
(A) Contractile cells
(B) Response of pial arterioles
20
oxygenated(Devoretal.,2011;Lyonsetal.,2016;Sakadžićetal.,2014;Vazquezetal.,
2010;Vovenko,1999;Yaseenetal.,2011).Inaddition,directmeasurements(Sakadžićetal.,
2014)andcombinedmorphologicalandfunctionaldata(Kasischkeetal.,2011)suggestthat
arteriolessupplyasignificantfractionofthetotaloxygentothetissueunderbaseline
conditions.Furthermore,thereisconflictingevidenceaboutthepresenceand/or
significanceofoxygengradientssurroundingvenulesandveins(Devoretal.,2011;Vazquez
etal.,2010;Vovenko,1999),andoxygenshuntsfromarterialtovenousvesselswerealso
directlyobservedinvivo(Lecoqetal.,2011),consistentwithanumberofstudiesthathave
reportedincreasesinvenouspO2(Sakadžićetal.,2014;Vazquezetal.,2010;Vovenko,
1999;Yaseenetal.,2011).Nonetheless,inrodents,pO2valuesinarteriesandarterioles
typicallyfallbetween60and110mmHg,valuesincapillariesvaryfrom20-60mmHg,and
valuesinveinsandvenulesrangefrom30to60mmHg(Lyonsetal.,2016;Parpaleixetal.,
2013;Sakadžićetal.,2014;Vazquezetal.,2010;Vovenko,1999;Yaseenetal.,2011).
Itisimportanttonotethatdifferencesinoxygenationbetweenawakeandanaesthetized
animalshavebeenreported,andalthoughoverallcapillarypO2appearedsimilarinthe
mouseolfactorybulbandsomatosensorycortex,thismaynotbetrueforallregions,
particularlysincemeasurementsofRBCfluxandlineardensitydiddifferbetweenthetwo
regions(Lyonsetal.,2016).Inaddition,sinceexperimentalandmodelingevidencesuggests
theexistenceoferythrocyte-associatedtransients(EATs;(GolubandPittman,2005;
Hellums,1977;Lecoqetal.,2011;Lückeretal.,2015;Parpaleixetal.,2013)),theremaybe
substantialdiscrepanciesbetweentruehemoglobinsaturationandthatestimatedbasedon
averagebloodpO2(Lyonsetal.,2016).Theoppositecaseisalsorelevant,sinceoxygen
saturationisoftenmeasuredusingtechniqueslikeintrinsicimagingofopticalsignals(Dunn
etal.,2003)orphotoacoustictomography(PAM;(Yaoetal.,2015))andthemeasured
saturationmaythereforeoverestimatemeanbloodpO2.
7.2 Activation-inducedchangesinoxygenation
Inresponsetoactivation,dynamicpO2increaseshavebeenreportedinallvesseltypes
(Lecoqetal.,2011;Parpaleixetal.,2013;Vazquezetal.,2010;Yaseenetal.,2011).Priorto
themainpositiveresponse,abrief,relativelysmalldecreaseinpO2hasalsobeenobserved
incapillariesandparenchymalregionsclosetocapillaries(Lecoqetal.,2011;Parpaleixet
al.,2013;Weietal.,2016).Thisearlyresponse,oftencalledthe‘initialdip’,hasbeen
controversialintheopticalimagingandfMRIcommunitiesduetoconflictingreports
(Buxton,2001;HuandYacoub,2012).Somereportssuggestthattheinitialdipmaybemore
spatiallyspecificthanthesubsequentincreaseinsignal(Vazquezetal.,2010).
ThereisalsosomeuncertaintyregardingthemagnitudeandimportanceofpO2increasesin
thetissuefollowingactivation.ManystudieshaveobservedrobustincreasesintissuepO2
(Lecoqetal.,2011;Parpaleixetal.,2013;Thompsonetal.,2003;Vazquezetal.,2010)and,
usingadrugthatpre-dilatedarteriestopreventfurtheractivation-inducedCBFincreases,
Masamotoetal.(2008)showedthatthesetissuepO2increasesundernormalconditions
occurredinspiteofincreasedoxygendemandinthetissue.Incontrast,Devoretal.(2011)
reportedthat,duringsustainedstimulation,therewasnopO2increaseinregionsofthe
tissuewithlowbaselinepO2.Theauthorsproposedthatactivation-inducedincreasesinCBF
mayoccurtopreventdangerouslevelsofhypoxiaintheseregions;however,itisunclear
21
whetherchangesinCBFwithoutanyincreaseinCMRO2wouldleadtopO2increasesin
theseregions,oriftheyweresimplytoofarfromthevasculaturetobenoticeablyaffected.
7.3 Laminaroxygenation
Tothebestofourknowledge,thereareonlyveryfewstudiesthathavereporteddirect
laminarmeasurementsofoxygenationinthebrain,partlybecauseofthetechnical
difficultiesinobtainingsuchdata.Forexample,electrodemeasurementsareinvasive,
especiallyatdepth,andcurrentopticalapproacheshavelimiteddepthpenetration(Lecoq
etal.,2011;Sakadžićetal.,2010).Furthermore,thestudiesthatdoexistaresomewhat
contradictory.Usingtwo-photonphosphorescencelifetimemicroscopy(2PLM)inalpha-
chloraloseanesthetizedrats,Devoretal.(2011)reportedasubstantialdecreasein
estimatesofarterialoxygenationfromthesurfaceto~200µmintothecortex;however,
Lyonsetal.(2016),usingthesametechniqueinawakemice,didnotobserveanysignificant
depth-dependenceinarterialorvenouspO2intheupper400µmofthecortex.Devoretal.
(2011)alsoobservednoticeabledecreasesintissuepO2intheupper~300µmofthecortex,
whichisconsistentwithpreviousmeasurementsusingoxygensensitivemicroelectrodes
(Masamotoetal.,2003).Masamotoetal.(2003)wereabletomeasurethroughoutthe
corticalgreymatter,deeperintothecortex.Allthreesomatosensoryregionsconsidered
showedasubstantialpO2decreasefromthesurfacetolayerII,buttheforelimband
hindlimbareasshowedfurthervariation,peakingaroundlayerV,whilethetrunkregion
remainedfairlyconstant.Notethatinallofthesestudies,pO2intheuppercorticallayers
mayhavebeeninfluencedbythesurgicalpreparation,includingremovaloftheskull.
Althoughtheirmeasurementtechnique(constantpotentialamperometry)wasnotableto
generatemeasurementsofbaselinepO2,Lietal.(2011)reportedsimultaneous,high
temporalresolutionmeasurementsofchangesoftissuepO2atmultiplecorticaldepthsin
responsetoelectricalstimulationoftheratwhiskerpad.Theauthorsobservedagenerally
biphasicresponse,wherebythepO2initiallydecreased,moststronglyinlayerIV,and
subsequentlyincreased,withthelargestincreasesoccurringintheuppercorticallayers.
Applyinganitricoxidesynthaseinhibitortoreduceactivation-inducedCBFincreasesmade
thepO2responsesmorenegativeatalldepths.Whilethisstudyoffersaninterestinginitial
viewintolaminardifferencesinoxygenation,furtherstudies,ideallyusingcomplementary
techniquessuchas2PLMorPAM,areneededtoinvestigatethesecomplexchanges.
8 Relevanceofvasculaturefor(laminar)fMRI
ThevasculatureisrelevantforanumberofMRItechniques,includingthediffusionand
perfusionweightedimagingapproachescommonlyusedinclinicaldiagnosisandtreatment
ofneuropathologiessuchastumors,stroke,andtransientischemicattack(Finketal.,2015;
Souillard-Scemamaetal.,2015).However,inthisreviewwefocusontherelevanceofthe
vasculatureforBOLD-fMRI.
AlthoughthedetailsoftheunderlyingphysicsofthefMRIsignalarebeyondthescopeof
thisreview,theequationfromBuxton(2013)listedbelow(Equation1)servestosummarize
thewaysinwhichthevasculaturecaninfluencetheBOLDsignal.(Readersinterestedinthe
derivationofthisequationoramorerigorousintroductiontotherelevantphysicsarehighly
recommendedtoconsultBuxton(2013)).Wefocushereonthemorecommonlyused
22
gradientecho(GE)technique,butnotethatthereareimportantdifferencesbetween
gradientechoandspinecho(SE)sequences.Forexample,GEsequencesaretypicallymore
sensitivetolargervessels,particularlyveins,thanSE(Boxermanetal.,1995b;Menon,
2012).
Briefly,theBOLDsignalchangenormalizedtobaseline,∆𝑆 𝑆!,canbedescribedsuchthat
∆𝑺 𝑺𝟎 = 𝒌 ∙ 𝑻𝑬 ∙ 𝑽𝟎 ∙ 𝑫𝟎
𝜷∙ 𝟏 − 𝒗 ∙ 𝒅
𝜷 , (1)
where𝑉!isthebaselinebloodvolume,𝐷!isthebaselineconcentrationof
deoxyhemoglobin,𝑣and𝑑arethedynamicbloodvolumeanddeoxyhemoglobin
concentrationnormalizedtotheirrespectivebaselines,𝑘isaconstantrelatedtotheMRI
fieldstrength,𝑇𝐸istheechotime,and𝛽isaconstantdescribingtheeffectof
deoxyhemoglobinontherelaxationrate.Whilethebloodvolumetermsarerelativelyself-
explanatory,itisworthemphasizingthatlocaldeoxyhemoglobinconcentrationreflectsthe
balanceoftwocompetingprocesses:oxygendelivery(viaCBF),andoxygenconsumption
(CMRO2).SinceCBFgenerallyincreasesmorethanCMRO2duringactivity(FoxandRaichle,
1986),theconcentrationofdeoxyhemoglobindecreases,andtheBOLDsignalincreases
(Ogawaetal.,1990).
AlthoughfMRIhasprimarilybeenusedtolocalizeneuronalactivity,thereisanincreasing
desiretouseitasaquantitativetooltomeasureCMRO2changes(Buxton,2013).Earlier
studiessuggestedthatCMRO2changesprimarilyreflectedenergyuseassociatedwith
neuronalsignaling(AttwellandLaughlin,2001).Assuch,alongstandingaimofthe
neuroimagingcommunityhasbeento‘unmix’or‘deconvolve’thevascularcomponentfrom
themeasuredfMRIresponse,inordertoisolatethemetaboliccontributiontothesignal.
However,amorerecenthypothesisproposesthatdifferenttypesofneuronalactivity,e.g.
excitatoryvsinhibitorysignaling,mayhavedistincteffectsonchangesinCBF,CMRO2,and
electricalactivity(Buxtonetal.,2014;Uhlirovaetal.,2016a).Inthefollowingsections,we
provideanoverviewofthewaysinwhichthevasculatureinfluencestheBOLDresponse,
includingthoseaspectsparticularlyrelevanttolaminarfMRI.
8.1 Baselinebloodvolumeanddeoxyhemoglobinconcentration
AsshowninEquation(1),thebaselinevaluesofCBVanddHbconcentrationacttoscalethe
magnitudeoftheBOLDsignalchangeforagivenchangeinCBVanddHb.Thismeansthat,
dependingonthebaselinevaluesofCBVanddHb,differentchangesinCBVanddHbcould
resultinthesameBOLDsignalchangeand,inversely,thatdifferentBOLDsignalchanges
couldresultfromthesamechangesinCBVanddHb.GiventhatCBVandpO2(likely
reflectingdifferencesindHb)varybetweendifferentcorticalregions,andindeedlayers
(Section4andSection7),thiseffectisimportanttoconsiderwhencomparingBOLD
responsesfromdifferentlocations.
ThebaselinevalueofdHbconcentrationalsoimposesatheoreticallimitonthemaximum
achievableBOLDincrease,sincedHbconcentrationcanonlydecreasefromitsbaseline
valuetozero,andnotbelow.Inaddition,itisalsoimportanttonotethat,despiteprevious
assumptionstothecontrary(Davisetal.,1998;Dunnetal.,2005;Hogeetal.,1999;
Mayhewetal.,2000),capillariesandevenarteriolescontainanon-negligibleamountof
deoxyhemoglobin(Section7;(Lyonsetal.,2016;Sakadžićetal.,2014;Yaseenetal.,2011)).
Evidencefromarecentmodelingstudysuggeststhatassumingarterialhemoglobinis
23
completelysaturatedcanleadtoerrorswhenestimatingCMRO2(BarrettandSuresh,
2015).
8.2 Activation-inducedchangesinbloodvolumeanddeoxyhemoglobin
DeoxyhemoglobinconcentrationreflectsthedynamicbalanceofchangesinCBFand
CMRO2,soitisimportanttoconsiderthespatiotemporalinterplayofthesetwoprocesses.
IntermsofCBF,someauthorssuggestthatthespacingbetweenpenetratingvessels
representsalimitontheminimalachievablepointspreadfunction(PSF)ofBOLD-fMRI
measurements(Turner,2016;UludağandBlinder,2017).Thiswouldbeconcerningforthe
utilityofhighresolutionandlaminarfMRI,particularlyconsideringthatcombined
anatomicalandfunctionaldatafromthemousebarrelcortexshowedthatthelocationof
descendingarteriesandascendingveinsdidnotcorrelatewithwhiskerbarrelregions
(Blinderetal.,2013).Inaddition,diameterchangesindescendingarteriesinthecatvisual
cortexwererecentlyshowntobelessselectiveforstimulusorientationthancalciumsignals
intheneuronssurroundingthem(O’Herronetal.,2016).However,theconclusionthat
penetratingvesselgeometrylimitsthePSFofBOLD-fMRImaybepremature,fortwo
reasons.
First,asdiscussedinSection6,flowregulationmayalsooccuratsitesdownstreamof
descendingarteries,suchaspre-capillaryarterioles,capillaries,andperhapsevenviared
bloodcellsthemselves(Weietal.,2016).ThiswouldallowCBFchangestoberegulatedover
amuchfinerspatialscalethantheterritoryofasingledescendingartery.
Secondly,changesinCBFrepresentonlyoneoftheeffectsleadingtothechangesindHb
thatdrivetheBOLDsignal;changesinCMRO2alsoplayavitalrole.Therefore,evenifCBF
increasesoccuroveralargerordifferentareathantheregionofneuronalactivity,orwith
differentkinetics,modelingapproachesthatmakeuseofmultimodaldatamaybeableto
decoupletheeffectsoflayer-dependentchangesinCBFandCMRO2ondHb(Gagnonetal.,
2015;Heinzleetal.,2016;Markuerkiagaetal.,2016).AnydifferencesbetweentheCBFand
CMRO2responsemayalsohelptoinferthenatureoftheunderlyingneuralactivity
(Uhlirovaetal.,2016a).
TheinfluenceofCBVchangesontheBOLDsignalarecomplex.Partofthismayberelated
todifficultiesindistinguishingbetweendirecteffectsofvolumechangesonthesignal,and
indirecteffectsthatrelatetovolumechanges.AsperEquation(1),earlysimulationsand
experimentsshowedthatpureincreasesinCBVtendtoreducetheBOLDsignalchange
(Boxermanetal.,1995a;Ogawaetal.,1993;YablonskiyandHaacke,1994).However,in
practice,anincreaseinarterialbloodvolumemaysomewhatincreasethesignal,by
exchangingvolumewiththeextravascularfluid,whichhasaweakersignal(Buxton,2013).
Furthermore,increasesinvenousbloodvolumewouldalsotypicallyincreasethe
concentrationofdHbinavoxel,leadingtoreducedBOLDsignal.
Severalrecentstudies,usingelegantapproachestoproducespatiallyconfinedresponses,
haveobservedthatactivation-inducedchangesinCBVappearedmorespecificandlocalized
toneuronalactivitythanBOLDresponses(Moonetal.,2013;Poplawskyetal.,2015).While
theseareinterestingresults,itisimportanttonotethatchangesinCBVareapurely
vascularresponse,andsodonotcontainanyinformationaboutthechangesinCMRO2
thoughttoderivemoredirectlyfromneuronalactivity.Nonetheless,combininghigh
resolutionCBVimagingwithCBFandBOLDmeasurementsintoadetailedbiophysicalmodel
24
wouldbeaparticularlypowerfulapproachforprobingneuronalactivitynon-invasively
(Uhlirovaetal.,2016a).
8.3 DirectMReffects
ThevascularstructureitselfcanalsohaveadirectinfluenceofontheBOLDsignal.Using
detailedreconstructionsofthemicrovasculaturecoupledwithhighresolutionfunctional
measurements,Gagnonetal.(2015)recentlydevelopeda‘bottomup’modeloftheBOLD
signalwhichpredictedthatactivation-inducedsignalchangeswouldvarybyupto40%,
dependingontheorientationofthecortextothescanner’sprimarymagneticfield.The
effectwaspresentonlywhenusinggradientecho,ratherthanspinecho,pulsesequences,
andderivesfromthefactthattheorientationoflargerveinsisnotisotropic,sincetheyare
predominantlyalignedeitherinparallelwithorperpendiculartothecorticalsurface
(Gagnonetal.,2015).Thisorientationdependenceisparticularlyrelevantfordatafrom
humanandprimatebrains,sincethefoldingpatternofgyriandsulciproducesasignificant
variationinlocalsurfaceorientation(Cohen-Adadetal.,2012),whichisnotpresentin
lissencephalicanimals.AlthoughthemodeldevelopedbyGagnonetal.(2015)usedthe
vascularstructurefrommice,ratherthanprimates,andfunctionaldatafromrats,
predictionsfromtheirmodelagreedverywellwithhumandata.However,giventhatthe
imagingvoxelsizewas3.4x3.4x6mm,itremainstobeseenwhethersuchorientation
dependenceexiststhroughthecortex,atthehigherspatialresolutiontypicalforlaminar
fMRI.
9 Outlook
Wehavetriedtocompilethecurrentstatusofresearchonthecorticalvascularsystemin
thepresentreview.Asmentioned,thevascularsystemisatthebasisformanyimportant
topicsinneuroscience,bothinhealthanddisease.Manyaspectsremaininsufficiently
understood.Firstandforemost,thehumancorticalvascularsystemhasnotbeenstudied
quantitatively,andmanyofourcurrentconceptsrelyonrodentdata.Itistherefore
importantthatnoveltechnicalapproachesemergethatallowvascularlabellingandimaging
ofhumanpost-mortemtissue.Thiswilldirectlyleadustoaddressingrelevantquestions
regardingtheinvolvementof(micro-)vascularnetworkalterationsinneurodegenerative
diseases.Anotherimportantfutureresearchdirectionconcernsthesizeofthe
reconstructednetworks.Muchofwhatweknowtodayreliesonrelativelysmall(fewcubic
millimeters)networks.Itisdesirablethatwholebrainvascularsystemreconstructions
becomeavailable,whichseemsfeasibleatleastinthemouse.Thiswouldsignificantly
reducetheproblemoftheboundaryconditionsformodelingbloodflowdynamics.
Furthermore,bloodflowdynamicsmusteventuallybestudiedinvivo,andmethodsare
neededthatcancoverentiremicrovascularnetworkswithsufficientspatialandtemporal
resolutiontocapturedynamicsatthesingleredbloodcelllevel.
10 Acknowledgements
25
FundingforthisworkwasprovidedbytheUniversityandETHZurichandtheSwissNational
ScienceFoundationGrantNo.140660.BWisamemberoftheClinicalResearchPriority
ProgramoftheUniversityofZurichonMolecularImaging.MBissupportedbythe
ForschungskreditoftheUniversityofZurich.
References
Adams, D.L., Piserchia, V., Economides, J.R., Horton, J.C., 2014. Vascular supply of the
cerebralcortexisspecializedforcelllayersbutnotcolumns.CerebralCortex,bhu221.
Armulik, A., Genove, G., Betsholtz, C., 2011. Pericytes: developmental, physiological, and
pathologicalperspectives,problems,andpromises.DevCell21,193-215.
Attwell,D.,Buchan,A.M.,Charpak,S., Lauritzen,M.,MacVicar,B.A.,Newman,E.A.,2010.
Glialandneuronalcontrolofbrainbloodflow.Nature468,232-243.
Attwell,D., Laughlin,S.B.,2001.AnEnergyBudget forSignaling in theGreyMatterof the
Brain.JournalofCerebralBloodFlow&Metabolism21,1133-1145.
Attwell, D., Mishra, A., Hall, C.N., O'Farrell, F.M., Dalkara, T., 2015. What is a pericyte?
JournalofCerebralBloodFlow&Metabolism,0271678X15610340.
Barrett, M.J., Suresh, V., 2013. Extra permeability is required to model dynamic oxygen
measurements: evidence for functional recruitment? Journal of Cerebral Blood Flow &
Metabolism33,1402-1411.
Barrett, M.J.P., Suresh, V., 2015. Improving estimates of the cerebral metabolic rate of
oxygenfromopticalimagingdata.NeuroImage106,101-110.
Barrett,M.J.P., Tawhai,M.H., Suresh,V., 2012.Arteries dominate volume changesduring
brief functionalhyperemia: evidence frommathematicalmodelling.NeuroImage62,482–
492.
Blinder, P., Shih, A.Y., Rafie, C., Kleinfeld, D., 2010. Topological basis for the robust
distributionofbloodtorodentneocortex.ProceedingsoftheNationalAcademyofSciences
107,12670-12675.
Blinder,P.,Tsai,P.S.,Kaufhold,J.P.,Knutsen,P.M.,Suhl,H.,Kleinfeld,D.,2013.Thecortical
angiome: an interconnected vascular network with noncolumnar patterns of blood flow.
NatureNeurosci16,889-897.
Boxerman,J.L.,Bandettini,P.A.,Kwong,K.K.,Baker,J.R.,Davis,T.L.,Rosen,B.R.,Weisskoff,
R.M.,1995a.Theintravascularcontributiontofmrisignalchange:montecarlomodelingand
diffusion-weightedstudiesinvivo.MagneticResonanceinMedicine34,4-10.
Boxerman, J.L., Hamberg, L.M., Rosen, B.R.,Weisskoff, R.M., 1995b.MR contrast due to
intravascularmagneticsusceptibilityperturbations.MagnResonMed34,555-566.
Buxton,R.B.,2001.TheElusiveInitialDip.NeuroImage13,953-958.
Buxton,R.B.,2013.Thephysicsoffunctionalmagneticresonanceimaging(fMRI).Reportson
ProgressinPhysics76,096601.
Buxton,R.B.,Griffeth,V.E.M., Simon,A.B.,Moradi, F., 2014.Variabilityof the couplingof
bloodflowandoxygenmetabolismresponsesinthebrain:aproblemforinterpretingBOLD
studies but potentially a new window on the underlying neural activity. Brain Imaging
Methods8,139.
Cassot, F., Frederic, L., Sylvie, L.,Prasanna,P.,Henri,D.,2009. Scaling Laws forBranching
VesselsofHumanCerebralCortex.Microcirculation16,331-344.
Chen,B.R., Bouchard,M.B.,McCaslin,A.F., Burgess, S.A.,Hillman, E.M., 2011.High-speed
vasculardynamicsofthehemodynamicresponse.NeuroImage54,1021-1030.
26
Chen,B.R.,Kozberg,M.G.,Bouchard,M.B.,Shaik,M.A.,Hillman,E.M.,2014.Acriticalrole
for the vascular endothelium in functionalneurovascular coupling in thebrain. Journalof
theAmericanHeartAssociation3,e000787.
Chen, J.J., Pike, G.B., 2009. BOLD-specific cerebral blood volume and blood flow changes
duringneuronalactivationinhumans.NMRinBiomedicine22,1054–1062.
Chen,J.J.,Pike,G.B.,2010.MRImeasurementoftheBOLD-specificflow-volumerelationship
duringhypercapniaandhypocapniainhumans.NeuroImage53,383-391.
Cohen-Adad,J.,Polimeni,J.R.,Helmer,K.G.,Benner,T.,McNab,J.A.,Wald,L.L.,Rosen,B.R.,
Mainero, C., 2012. T2*mapping and B0 orientation-dependence at 7 T reveal cyto- and
myeloarchitectureorganizationofthehumancortex.NeuroImage60,1006-1014.
Davis, T.L., Kwong, K.K., Weisskoff, R.M., Rosen, B.R., 1998. Calibrated functional MRI:
Mapping the dynamics of oxidativemetabolism. Proceedings of theNational Academy of
SciencesoftheUnitedStatesofAmerica95,1834-1839.
Devor,A.,Hillman,E.M.,Tian,P.,Waeber,C.,Teng,I.C.,Ruvinskaya,L.,Shalinsky,M.H.,Zhu,
H.,Haslinger,R.H.,Narayanan,S.N.,others,2008.Stimulus-inducedchanges inbloodflow
and 2-deoxyglucose uptake dissociate in ipsilateral somatosensory cortex. Journal of
Neuroscience28,14347-14357.
Devor,A.,Sakadzic,S.,Saisan,P.A.,Yaseen,M.A.,Roussakis,E.,Srinivasan,V.J.,Vinogradov,
S.A.,Rosen,B.R.,Buxton,R.B.,Dale,A.M.,Boas,D.A.,2011."Overshoot"ofO-2IsRequired
toMaintain Baseline Tissue Oxygenation at Locations Distal to Blood Vessels. Journal of
Neuroscience31,13676-13681.
Devor,A.,Tian,P.,Nishimura,N.,Teng, I.C.,Hillman,E.M.,Narayanan,S.,Ulbert, I.,Boas,
D.A.,Kleinfeld,D.,Dale,A.M.,2007.Suppressedneuronalactivityandconcurrentarteriolar
vasoconstrictionmayexplainnegativebloodoxygenationlevel-dependentsignal.Journalof
Neuroscience27,4452-4459.
Diemer,K.,Henn,R.,1965.KapillarvermehrunginderHirnrindeder Rratte unter
chronischemSauerstoffmangel.DieNaturwissenschafter52,135-136.
Douglas,R.J.,Martin,K.A.,2004.Neuronalcircuitsoftheneocortex.AnnuRevNeurosci27,
419-451.
Drew,P.J.,Shih,A.Y.,Kleinfeld,D.,2011.Fluctuatingandsensory-inducedvasodynamicsin
rodent cortexextendarteriole capacity.Proceedingsof theNationalAcademyofSciences
108,8473-8478.
Dunn,A.K.,Devor,A.,Bolay,H.,Andermann,M.L.,Moskowitz,M.A.,Dale,A.M.,Boas,D.A.,
2003.Simultaneous imagingof totalcerebralhemoglobinconcentration,oxygenation,and
bloodflowduringfunctionalactivation.OpticsLetters28,28-30.
Dunn,A.K.,Devor,A.,Dale,A.M.,Boas,D.A.,2005.Spatialextentofoxygenmetabolismand
hemodynamic changes during functional activation of the rat somatosensory cortex.
NeuroImage27,279-290.
Duvernoy,H.M.,Delon,S.,Vannson, J.L.,1981.Corticalbloodvesselsof thehumanbrain.
BrainResBull7,519-579.
Erinjeri,J.P.,Woolsey,T.A.,2002.Spatialintegrationofvascularchangeswithneuralactivity
inmousecortex.JournalofCerebralBloodFlow&Metabolism22,353-360.
Erturk,A.,Becker,K.,Jahrling,N.,Mauch,C.P.,Hojer,C.D.,Egen,J.G.,Hellal,F.,Bradke,F.,
Sheng, M., Dodt, H.U., 2012. Three-dimensional imaging of solvent-cleared organs using
3DISCO.NatProtoc7,1983-1995.
27
Fernandez-Klett,F.,Offenhauser,N.,Dirnagl,U.,Priller, J., Lindauer,U.,2010.Pericytes in
capillariesarecontractileinvivo,butarteriolesmediatefunctionalhyperemiainthemouse
brain.ProcNatlAcadSciUSA107,22290-22295.
Fernández-Klett, F., Priller, J., 2015. Diverse functions of pericytes in cerebral blood flow
regulationandischemia.JournalofCerebralBloodFlow&Metabolism35,883-887.
Fink, J.R.,Muzi,M., Peck,M., Krohn, K.A., 2015.Multimodality Brain Tumor Imaging:MR
Imaging,PET,andPET/MRImaging.JournalofNuclearMedicine56,1554-1561.
Fonta, C., Imbert, M., 2002. Vascularization in the primate visual cortex during
development.CerebralCortex12,199-211.
Fox, P.T., Raichle, M.E., 1986. Focal physiological uncoupling of cerebral blood flow and
oxidativemetabolismduringsomatosensorystimulationinhumansubjects.Proceedingsof
theNationalAcademyofSciencesoftheUnitedStatesofAmerica83,1140-1144.
Gagnon,L.,Sakadžić,S.,Lesage,F.,Musacchia,J.J.,Lefebvre,J.,Fang,Q.,Yücel,M.A.,Evans,
K.C.,Mandeville,E.T.,Cohen-Adad, J.,Polimeni, J.a.R.,Yaseen,M.A., Lo,E.H.,Greve,D.N.,
Buxton,R.B.,Dale,A.M.,Devor,A.,Boas,D.A.,2015.QuantifyingtheMicrovascularOriginof
BOLD-fMRI from First Principles with Two-Photon Microscopy and an Oxygen-Sensitive
Nanoprobe.TheJournalofNeuroscience35,3663-3675.
Goense, J., Merkle, H., Logothetis, N.K., 2012. High-resolution fMRI reveals laminar
differences in neurovascular coupling between positive and negative BOLD responses.
Neuron76,629-639.
Golub, A.S., Pittman, R.N., 2005. Erythrocyte-associated transients in Po2 revealed in
capillaries of rat mesentery. American Journal of Physiology - Heart and Circulatory
Physiology288,H2735-H2743.
Guibert,R.,Fonta,C.,Plouraboue,F.,2010.Cerebralbloodflowmodelinginprimatecortex.
JCerebBloodFlowMetab30,1860-1873.
Guibert, R., Fonta, C., Risser, L., Plouraboué, F., 2012. Coupling and robustness of intra-
corticalvascularterritories.NeuroImage62,408-417.
Gutiérrez-Jiménez,E.,Cai,C.,Mikkelsen, I.K.a.,Rasmussen,P.M.,Angleys,H.,Merrild,M.,
Mouridsen, K., Jespersen, S.N., Lee, J., Iversen, N.K., others, 2016. Effect of electrical
forepawstimulationoncapillarytransit-timeheterogeneity(CTH).JournalofCerebralBlood
Flow&Metabolism,0271678X16631560.
Hall, C.N., Reynell, C., Gesslein, B., Hamilton, N.B.,Mishra, A., Sutherland, B.A., O'Farrell,
F.M., Buchan, A.M., Lauritzen,M., Attwell, D., 2014. Capillary pericytes regulate cerebral
bloodflowinhealthanddisease.Nature508,55-60.
Hamilton, N.B., Attwell, D., Hall, C.N., 2010. Pericyte-mediated regulation of capillary
diameter: a component of neurovascular coupling in health and disease. Front
Neuroenergetics2,102-115.
Harik, S.I., Behmand, R.A., LaManna, J.C., 1994. Hypoxia increases glucose transport at
blood-brainbarrierinrats.JApplPhysiol77,896-901.
Hartmann, D.A., Underly, R.G., Grant, R.I., Watson, A.N., Lindner, V., Shih, A.Y., 2015.
Pericyte structure and distribution in the cerebral cortex revealed by high-resolution
imagingoftransgenicmice.Neurophotonics2,041402-041402.
Heinzer, S., Krucker, T., Stampanoni,M.,Abela,R.,Meyer, E.P., Schuler,A., Schneider, P.,
Muller, R., 2006. Hierarchical microimaging for multiscale analysis of large vascular
networks.NeuroImage32,626-636.
Heinzle, J., Koopmans, P.J., den Ouden, H.E.M., Raman, S., Stephan, K.E., 2016. A
hemodynamicmodelforlayeredBOLDsignals.NeuroImage125,556-570.
28
Hellums,J.D.,1977.Theresistancetooxygentransport inthecapillariesrelativetothat in
thesurroundingtissue.MicrovascRes13,131–136.
Hill,R.A.,Tong,L.,Yuan,P.,Murikinati,S.,Gupta,S.,Grutzendler, J.,2015.Regionalblood
flow in the normal and ischemic brain is controlled by arteriolar smooth muscle cell
contractilityandnotbycapillarypericytes.Neuron87,95-110.
Hillman, E.M., 2014. Coupling mechanism and significance of the BOLD signal: a status
report.Annualreviewofneuroscience37,161-181.
Hillman,E.M.C.,Devor,A.,Bouchard,M.B.,Dunn,A.K.,Krauss,G.W.,Skoch,J.,Bacskai,B.J.,
Dale, A.M., Boas, D.A., 2007. Depth-resolved optical imaging andmicroscopy of vascular
compartmentdynamicsduringsomatosensorystimulation.NeuroImage35,89-104.
Hirano, Y., Stefanovic, B., Silva, A.C., 2011. Spatiotemporal evolution of the functional
magnetic resonance imaging response to ultrashort stimuli. Journal of Neuroscience 31,
1440-1447.
Hirsch, S., Reichold, J., Schneider, M., Szekely, G., Weber, B., 2012. Topology and
hemodynamicsofthecorticalcerebrovascularsystem.JCBFM32,952-967.
Ho,Y.-C.L.,Petersen,E.T.,Zimine,I.,Golay,X.,2011.SimilaritiesandDifferencesinArterial
Responses to Hypercapnia and Visual Stimulation. Journal of Cerebral Blood Flow &
Metabolism31,560-571.
Hoge,R.D.,Atkinson, J.,Gill,B.,Crelier,G.R.,Marrett,S.,Pike,G.B.,1999. Investigationof
BOLD signal dependence on cerebral blood flow and oxygen consumption: The
deoxyhemoglobindilutionmodel.MagneticResonanceinMedicine42,849–863.
Hu,X.,Yacoub,E.,2012.ThestoryoftheinitialdipinfMRI.NeuroImage62,1103-1108.
Huber,L.,Goense,J.,Kennerley,A.J.,Trampel,R.,Guidi,M.,Reimer,E.,Ivanov,D.,Neef,N.,
Gauthier, C.J., Turner, R., Möller, H.E., 2015. Cortical lamina-dependent blood volume
changesinhumanbrainat7T.NeuroImage107,23-33.
Huber,L., Ivanov,D.,Handwerker,D.A.,Marrett,S.,Guidi,M.,Uludağ,K.,Bandettini,P.A.,
Poser, B.A., 2016. Techniques for blood volume fMRI with VASO: From low-resolution
mappingtowardssub-millimeterlayer-dependentapplications.NeuroImage.
Iadecola,C.,Nedergaard,M.,2007.Glialregulationofthecerebralmicrovasculature.Nature
neuroscience10,1369-1376.
Iadecola,C.,Yang,G.,Ebner,T.J.,Chen,G.,1997.Localandpropagatedvascularresponses
evokedby focal synapticactivity in cerebellar cortex. Journalofneurophysiology78,651-
659.
Iliff,J.J.,Nedergaard,M.,2013.IsThereaCerebralLymphaticSystem?Stroke44,S93-S95.
Iliff, J.J.,Wang,M.H., Liao, Y.H., Plogg, B.A., Peng,W.G.,Gundersen,G.A., Benveniste,H.,
Vates,G.E.,Deane,R.,Goldman,S.A.,Nagelhus,E.A.,Nedergaard,M.,2012.AParavascular
PathwayFacilitatesCSFFlowThroughtheBrainParenchymaandtheClearanceofInterstitial
Solutes,IncludingAmyloidbeta.ScienceTranslationalMedicine4.
Iliff,J.J.,Wang,M.H.,Zeppenfeld,D.M.,Venkataraman,A.,Plog,B.A.,Liao,Y.H.,Deane,R.,
Nedergaard,M.,2013.CerebralArterialPulsationDrivesParavascularCSF-Interstitial Fluid
ExchangeintheMurineBrain.JournalofNeuroscience33,18190-18199.
Itoh, Y., Suzuki, N., 2012. Control of brain capillary blood flow. Journal of Cerebral Blood
Flow&Metabolism32,1167-1176.
Jespersen,S.N.,Østergaard,L.,2012.Therolesofcerebralbloodflow,capillarytransittime
heterogeneity, and oxygen tension in brain oxygenation and metabolism. Journal of
CerebralBloodFlowandMetabolism32,264–277.
29
Jones,T.H.,Morawetz,R.B.,Crowell,R.M.,Marcoux,F.W.,FitzGibbon,S.J.,DeGirolami,U.,
Ojemann,R.G., 1981. Thresholdsof focal cerebral ischemia in awakemonkeys. Journalof
Neurosurgery54,773-782.
Kasischke,K.A.,Lambert,E.M.,Panepento,B.,Sun,A.,Gelbard,H.A.,Burgess,R.W.,Foster,
T.H., Nedergaard, M., 2011. Two-photon NADH imaging exposes boundaries of oxygen
diffusionincorticalvascularsupplyregions.JournalofCerebralBloodFlowandMetabolism
31,68–81.
Keller, A.L., Schüz, A., Logothetis, N.K., Weber, B., 2011. Vascularization of cytochrome
oxidase-rich blobs in the primary visual cortex of squirrel and macaque monkeys. The
JournalofNeuroscience31,1246-1253.
Kim,T.,Hendrich,K.S.,Masamoto,K.,Kim,S.-G.,2007.Arterialversus totalbloodvolume
changes during neural activity-induced cerebral blood flow change: implication for BOLD
fMRI.JournalofCerebralBloodFlowandMetabolism27,1235–1247.
Kim,T.,Kim,S.-G.,2010.Corticallayer-dependentarterialbloodvolumechanges:Improved
spatialspecificityrelativetoBOLDfMRI.NeuroImage49,1340-1349.
Kim,T.,Kim,S.-G.,2011.Temporaldynamicsandspatial specificityofarterialandvenous
bloodvolumechangesduringvisualstimulation:implicationforBOLDquantification.Journal
ofCerebralBloodFlowandMetabolism31,1211–1222.
Kleinfeld,D.,Mitra,P.P.,Helmchen,F.,Denk,W.,1998.Fluctuationsandstimulus-induced
changes in blood flow observed in individual capillaries in layers 2 through 4 of rat
neocortex.ProceedingsoftheNationalAcademyofSciences95,15741-15746.
Kwong, K.K., Belliveau, J.W., Chesler, D.A., Goldberg, I.E.,Weisskoff, R.M., Poncelet, B.P.,
Kennedy, D.N., Hoppel, B.E., Cohen, M.S., Turner, R., et al., 1992. Dynamic magnetic
resonance imaging of human brain activity during primary sensory stimulation. Proc Natl
AcadSciUSA89,5675-5679.
LaManna, J.C., Vendel, L.M., Farrell, R.M., 1992. Brain adaptation to chronic hypobaric
hypoxiainrats.JApplPhysiol72,2238-2243.
Lauwers, F., Cassot, F., Lauwers-Cances, V., Puwanarajah, P., Duvernoy, H., 2008.
Morphometry of the human cerebral cortexmicrocirculation: General characteristics and
space-relatedprofiles.NeuroImage39,936-948.
Lecoq,J.,Parpaleix,A.,Roussakis,E.,Ducros,M.,Houssen,Y.G.,Vinogradov,S.A.,Charpak,
S., 2011. Simultaneous two-photon imaging of oxygen and blood flow in deep cerebral
vessels.Naturemedicine17,893–898.
Lee,J.,Jiang,J.Y.,Wu,W.,Lesage,F.,Boas,D.A.,2014.Statisticalintensityvariationanalysis
for rapid volumetric imagingof capillary network flux. Biomedical optics express 5, 1160-
1172.
Lee,J.,Wu,W.,Boas,D.A.,2015.Earlycapillaryfluxhomogenizationinresponsetoneural
activation.JournalofCerebralBloodFlow&Metabolism,0271678X15605851.
Li, J., Bravo,D.S., LouiseUpton,A.,Gilmour,G., Tricklebank,M.D., Fillenz,M.,Martin, C.,
Lowry, J.P., Bannerman, D.M., McHugh, S.B., 2011. Close temporal coupling of neuronal
activityand tissueoxygen responses in rodentwhiskerbarrel cortex. European Journalof
Neuroscience34,1983-1996.
Lindvere, L., Janik, R., Dorr, A., Chartash, D., Sahota, B., Sled, J.G., Stefanovic, B., 2013.
Cerebral microvascular network geometry changes in response to functional stimulation.
NeuroImage71,248-259.
30
Lorthois,S.,Cassot,F.,Lauwers,F.,2011.Simulationstudyofbrainbloodflowregulationby
intra-cortical arterioles in an anatomically accurate large human vascular network: Part I:
Methodologyandbaselineflow.NeuroImage54,1031-1042.
Lücker,A.,Weber,B.,Jenny,P.,2015.Adynamicmodelofoxygentransportfromcapillaries
totissuewithmovingredbloodcells.AmericanJournalofPhysiology-HeartandCirculatory
Physiology308,H206-H216.
Lyons,D.G.,Parpaleix,A.,Roche,M.,Charpak,S.,2016.Mappingoxygenconcentration in
theawakemousebrain.Elife5,e12024.
Markuerkiaga,I.,Barth,M.,Norris,D.G.,2016.Acorticalvascularmodelforexaminingthe
specificityofthelaminarBOLDsignal.NeuroImage132,491-498.
Masamoto, K., Fukuda, M., Vazquez, A., Kim, S.-G., 2009. Dose-dependent effect of
isoflurane on neurovascular coupling in rat cerebral cortex. European Journal of
Neuroscience30,242-250.
Masamoto,K., Takizawa,N.,Kobayashi,H.,Oka,K., Tanishita,K.,2003.Dual responsesof
tissuepartialpressureofoxygenafter functional stimulation in rat somatosensorycortex.
Brainresearch979,104-113.
Masamoto, K., Vazquez, A.,Wang, P., Kim, S.-G., 2008. Trial-by-trial relationship between
neuralactivity,oxygenconsumption,andbloodflowresponses.NeuroImage40,442-450.
Mathiisen,T.M.,Lehre,K.P.,Danbolt,N.C.,Ottersen,O.P.,2010.Theperivascularastroglial
sheathprovidesacompletecoveringofthebrainmicrovessels:anelectronmicroscopic3D
reconstruction.Glia58,1094-1103.
Mayhew, J., Johnston,D.,Berwick, J., Jones,M.,Coffey,P.,Zheng,Y.,2000.Spectroscopic
AnalysisofNeuralActivityinBrain:IncreasedOxygenConsumptionFollowingActivationof
BarrelCortex.NeuroImage12,664–675.
Menon,R.S.,2012.Thegreatbrainversusveindebate.Neuroimage62,970-974.
Miller,A.T.,Jr.,Hale,D.M.,1970.Increasedvascularityofbrain,heart,andskeletalmuscle
ofpolycythemicrats.AmJPhysiol219,702-704.
Moon, C.H., Fukuda, M., Kim, S.-G., 2013. Spatiotemporal characteristics and vascular
sources of neural-specific and -nonspecific fMRI signals at submillimeter columnar
resolution.NeuroImage64,91-103.
Moskowitz,M.A.,Lo,E.H.,Iadecola,C.,2010.Thescienceofstroke:mechanismsinsearchof
treatments.Neuron67,181–198.
Mosso,A.,1881.UeberdenKreislaufdesBlutesimMenschlichenGehirn,Leipzig.
Ngai, A.C., Ko, K.R.,Morii, S.,Winn,H.R., 1988. Effect of sciatic nerve stimulation on pial
arterioles in rats. American Journal of Physiology-Heart and Circulatory Physiology 254,
H133-H139.
Ngai,A.C.,Winn,H.R.,2002.Pialarterioledilationduringsomatosensorystimulationisnot
mediated by an increase in CSF metabolites. American Journal of Physiology-Heart and
CirculatoryPhysiology51,H902.
Nguyen, J., Nishimura, N., Fetcho, R.N., Iadecola, C., Schaffer, C.B., 2011. Occlusion of
cortical ascending venules causes blood flow decreases, reversals in flow direction, and
vessel dilation in upstream capillaries. Journal of Cerebral Blood Flow &Metabolism 31,
2243-2254.
Nishimura, N., Schaffer, C.B., Friedman, B., Lyden, P.D., Kleinfeld, D., 2007. Penetrating
arterioles are a bottleneck in the perfusion of neocortex. Proceedings of the National
AcademyofSciences104,365-370.
31
O’Herron,P.,Chhatbar,P.Y.,Levy,M.,Shen,Z.,Schramm,A.E.,Lu,Z.,Kara,P.,2016.Neural
correlates of single-vessel haemodynamic responses in vivo. Nature advance online
publication.
Ogawa,S., Lee,T.M.,Kay,A.R.,Tank,D.W.,1990.Brainmagnetic resonance imagingwith
contrastdependentonbloodoxygenation.ProceedingsoftheNationalAcademyofSciences
oftheUnitedStatesofAmerica87,9868-9872.
Ogawa,S.,Menon,R.S.,Tank,D.W.,Kim,S.G.,Merkle,H.,Ellermann,J.M.,Ugurbil,K.,1993.
Functional brain mapping by blood oxygenation level-dependent contrast magnetic
resonance imaging. A comparison of signal characteristics with a biophysical model.
BiophysicalJournal64,803–812.
Opitz, E., 1951. Increased vascularization of the tissue due to acclimati- zation to high
altitudeanditssignificancefortheoxygentransport.ExpMedSurg9,389-403.
Parpaleix, A., Goulam Houssen, Y., Charpak, S., 2013. Imaging local neuronal activity by
monitoringPO2transientsincapillaries.Naturemedicine19,241–246.
Peppiatt, C.M., Howarth, C., Mobbs, P., Attwell, D., 2006. Bidirectional control of CNS
capillarydiameterbypericytes.Nature443,700-704.
Plouraboue,F.,Cloetens,P.,Fonta,C.,Steyer,A.,Lauwers,F.,Marc-Vergnes,J.P.,2004.X-
rayhigh-resolutionvascularnetworkimaging.JMicrosc215,139-148.
Poplawsky,A.J.,Fukuda,M.,Murphy,M.,Kim,S.-G.,2015.Layer-SpecificfMRIResponsesto
Excitatory and Inhibitory Neuronal Activities in the Olfactory Bulb. The Journal of
Neuroscience35,15263-15275.
Reichold,J.,Stampanoni,M.,LenaKeller,A.,Buck,A.,Jenny,P.,Weber,B.,2009.Vascular
graph model to simulate the cerebral blood flow in realistic vascular networks. J Cereb
BloodFlowMetab29,1429-1443.
Reina-De La Torre, F., Rodriguez-Baeza, A., Sahuquillo-Barris, J., 1998. Morphological
characteristics anddistributionpatternof thearterial vessels inhuman cerebral cortex: a
scanningelectronmicroscopestudy.AnatRec251,87-96.
Risser, L., Plouraboue, F., Cloetens, P., Fonta, C., 2009. A 3D-investigation shows that
angiogenesisinprimatecerebralcortexmainlyoccursatcapillarylevel.InternationalJournal
ofDevelopmentalNeuroscience27,185-196.
Roy, C.S., Sherrington, M.B., 1890. On the regulation of the blood-supply of the brain.
JournalofPhysiologyLondon11,85-108.
Sakadžić, S., Mandeville, E.T., Gagnon, L., Musacchia, J.J., Yaseen, M.A., Yucel, M.A.,
Lefebvre,J.,Lesage,F.,Dale,A.M.,Eikermann-Haerter,K.,Ayata,C.,Srinivasan,V.J.,Lo,E.H.,
Devor, A., Boas,D.A., 2014. Large arteriolar component of oxygendelivery implies a safe
marginofoxygensupplytocerebraltissue.Naturecommunications5.
Sakadžić, S., Roussakis, E., Yaseen, M.A., Mandeville, E.T., Srinivasan, V.J., Arai, K.,
Ruvinskaya, S., Devor, A., Lo, E.H., Vinogradov, S.A., others, 2010. Two-photon high-
resolutionmeasurement of partial pressure of oxygen in cerebral vasculature and tissue.
NatureMethods7,755-759.
Schaffer,C.B.,Friedman,B.,Nishimura,N.,Schroeder,L.F.,Tsai,P.S.,Ebner,F.F.,Lyden,P.D.,
Kleinfeld, D., 2006. Two-photon imaging of cortical surfacemicrovessels reveals a robust
redistributioninbloodflowaftervascularocclusion.PLoSBiol4,e22.
Schmid,F.,Reichold, J.,Weber,B., Jenny,P.,2015.The impactofcapillarydilationonthe
distributionof redblood cells in artificial networks.American Journal of Physiology-Heart
andCirculatoryPhysiology308,H733-H742.
32
Schmid, F., Tsai,P.S.,Kleinfeld,D., Jenny,P.,Weber,B.,2017.Depth-dependent flowand
pressurecharacteristicsincorticalmicrovascularnetworks.PLOSComputationalBiology13,
e1005392.
Sekiguchi,Y.,Takuwa,H.,Kawaguchi,H.,Kikuchi,T.,Okada,E.,Kanno,I.,Ito,H.,Tomita,Y.,
Itoh,Y.,Suzuki,N.,others,2014.Pialarteriesrespondearlierthanpenetratingarteriolesto
neuralactivationinthesomatosensorycortexinawakemiceexposedtochronichypoxia:an
additionalmechanismtoproximal integrationsignaling? JournalofCerebralBloodFlow&
Metabolism34,1761-1770.
Shih,A.Y.,Blinder,P., Tsai,P.S., Friedman,B., Stanley,G., Lyden,P.D.,Kleinfeld,D.,2013.
Thesmalleststroke:occlusionofonepenetratingvesselleadstoinfarctionandacognitive
deficit.Natureneuroscience16,55-63.
Shih, A.Y., Rühlmann, C., Blinder, P., Devor, A., Drew, P.J., Friedman, B., Knutsen, P.M.,
Lyden, P.D.,Mateo, C.,Mellander, L., others, 2015. Robust and fragile aspects of cortical
bloodflowinrelationtotheunderlyingangioarchitecture.Microcirculation22,204-218.
Souillard-Scemama,R.,Tisserand,M.,Calvet,D.,Jumadilova,D.,Lion,S.,Turc,G.,Edjlali,M.,
Mellerio,C.,Lamy,C.,Naggara,O.,Meder,J.-F.,Oppenheim,C.,2015.Anupdateonbrain
imagingintransientischemicattack.JournalofNeuroradiology42,3-11.
Stefanovic,B.,Hutchinson,E.,Yakovleva,V.,Schram,V.,Russell,J.T.,Belluscio,L.,Koretsky,
A.P.,Silva,A.C.,2008.Functionalreactivityofcerebralcapillaries.JCerebBloodFlowMetab
28,961-972.
Thompson, J.K., Peterson, M.R., Freeman, R.D., 2003. Single-Neuron Activity and Tissue
OxygenationintheCerebralCortex.Science299,1070-1072.
Tian,P.,Teng,I.C.,May,L.D.,Kurz,R.,Lu,K.,Scadeng,M.,Hillman,E.M.,DeCrespigny,A.J.,
D'Arceuil,H.E.,Mandeville,J.B.,others,2010.Corticaldepth-specificmicrovasculardilation
underlies laminardifferences inbloodoxygenation level-dependent functionalMRI signal.
ProceedingsoftheNationalAcademyofSciences107,15246-15251.
Tieman,S.B.,Mollers,S.,Tieman,D.G.,White,J.,2004.Thebloodsupplyofthecat'svisual
cortexanditspostnataldevelopment.BrainRes998,100-112.
Tsai,P.S.,Friedman,B.,Ifarraguerri,A.I.,Thompson,B.D.,Lev-Ram,V.,Schaffer,C.B.,Xiong,
Q., Tsien, R.Y., Squier, J.A., Kleinfeld, D., 2003. All-optical histology using ultrashort laser
pulses.Neuron39,27-41.
Tsai, P.S., Kaufhold, J.P., Blinder, P., Friedman, B., Drew, P.J., Karten, H.J., Lyden, P.D.,
Kleinfeld,D., 2009. Correlationsof neuronal andmicrovascular densities inmurine cortex
revealed by direct counting and colocalization of nuclei and vessels. The Journal of
Neuroscience29,14553-14570.
Turner, R., 2016. Uses, misuses, new uses and fundamental limitations of magnetic
resonanceimagingincognitivescience.Phil.Trans.R.Soc.B371,20150349.
Uhlirova, H., Kılıç, K., Tian, P., Sakadžić, S., Gagnon, L., Thunemann, M., Desjardins, M.,
Saisan, P.A., Nizar, K., Yaseen, M.A., Hagler, D.J., Vandenberghe, M., Djurovic, S.,
Andreassen,O.A.,Silva,G.A.,Masliah,E.,Kleinfeld,D.,Vinogradov,S.,Buxton,R.B.,Einevoll,
G.T., Boas, D.A., Dale, A.M., Devor, A., 2016a. The roadmap for estimation of cell-type-
specific neuronal activity from non-invasive measurements. Phil. Trans. R. Soc. B 371,
20150356.
Uhlirova, H., Kılıç, K., Tian, P., Thunemann, M., Desjardins, M., Saisan, P.A., Sakadžić, S.,
Ness, T.V., Mateo, C., Cheng, Q., others, 2016b. Cell type specificity of neurovascular
couplingincerebralcortex.Elife5,e14315.
33
Uludağ,K.,Blinder,P.,2017.Linkingbrainvascularphysiologytohemodynamicresponsein
ultra-highfieldMRI.NeuroImage.
Vazquez,A.L.,Fukuda,M.,Tasker,M.L.,Masamoto,K.,Kim,S.-G.,2010.Changesincerebral
arterial, tissue and venous oxygenation with evoked neural stimulation: implications for
hemoglobin-basedfunctionalneuroimaging.JournalofCerebralBloodFlowandMetabolism
30,428–439.
Vazquez, A.L., Masamoto, K., Kim, S.-G., 2008. Dynamics of oxygen delivery and
consumption during evoked neural stimulation using a compartmentmodel and CBF and
tissuePO2measurements.NeuroImage42,49–59.
Vovenko,E.,1999.Distributionofoxygentensiononthesurfaceofarterioles,capillariesand
venulesofbraincortexand in tissue innormoxia:anexperimental studyon rats.Pflugers
Archiv(EuropeanJournalofPhysiology)437,617–623.
Weber,B.,Keller,A.L.,Reichold,J.,Logothetis,N.K.,2008.Themicrovascularsystemofthe
striateandextrastriatevisualcortexofthemacaque.CerebralCortex18,2318-2330.
Wei, HelenS., Kang, H., Rasheed, I.-YarD., Zhou, S., Lou, N., Gershteyn, A., McConnell,
EvanD.,Wang,Y.,Richardson,KristopherE.,Palmer,AndreF.,Xu,C.,Wan,J.,Nedergaard,
M., 2016. Erythrocytes Are Oxygen-Sensing Regulators of the Cerebral Microcirculation.
Neuron91,851-862.
Xie, L.L., Kang, H.Y., Xu, Q.W., Chen, M.J., Liao, Y.H., Thiyagarajan, M., O'Donnell, J.,
Christensen,D.J.,Nicholson,C.,Iliff,J.J.,Takano,T.,Deane,R.,Nedergaard,M.,2013.Sleep
DrivesMetaboliteClearancefromtheAdultBrain.Science342,373-377.
Xue, S., Gong, H., Jiang, T., Luo, W., Meng, Y., Liu, Q., Chen, S., Li, A., 2014. Indian-ink
perfusionbasedmethod for reconstructingcontinuousvascularnetworks inwholemouse
brain.PLoSOne9,e88067.
Yablonskiy, D.A., Haacke, E.M., 1994. Theory of NMR signal behavior in magnetically
inhomogeneoustissues:Thestaticdephasingregime.MagneticResonanceinMedicine32,
749-763.
Yao,J.,Wang,L.,Yang,J.-M.,Maslov,K.I.,Wong,T.T.W.,Li,L.,Huang,C.-H.,Zou,J.,Wang,
L.V., 2015. High-speed label-free functional photoacoustic microscopy of mouse brain in
action.NatureMethodsadvanceonlinepublication.
Yaseen, M.A., Srinivasan, V.J., Sakadzic, S., Radhakrishnan, H., Gorczynska, I., Wu, W.,
Fujimoto, J.G.,Boas,D.A., 2011.Microvascularoxygen tensionand flowmeasurements in
rodent cerebral cortex during baseline conditions and functional activation. Journal of
CerebralBloodFlowandMetabolism31,1051–1063.
Zhang, E.T., Inman,C.B.E.,Weller, R.O., 1990. Interrelationshipsof thePiaMater and the
Perivascular(Virchow-Robin)SpacesintheHumanCerebrum.JournalofAnatomy170,111-
123.
Zhao,F.,Wang,P.,Hendrich,K.,Ugurbil,K.,Kim,S.-G.,2006.Corticallayer-dependentBOLD
and CBV responses measured by spin-echo and gradient-echo fMRI: insights into
hemodynamicregulation.NeuroImage30,1149-1160.
Zheng,D.,LaMantia,A.S.,Purves,D.,1991.Specializedvascularizationoftheprimatevisual
cortex.JNeurosci11,2622-2629.