fMRI Localization of Spinal Cord Processing Underlying Female Sexual Arousal

12 Pages·2015·0.13 MB·English

JOURNALOFSEX&MARITALTHERAPY,42(1),36–47,2016
(cid:2)
Copyright C Taylor&FrancisGroup,LLC
ISSN:0092-623Xprint /1521-0715online
DOI:10.1080/0092623X.2015.1010674
fMRI Localization of Spinal Cord Processing Underlying
Female Sexual Arousal
MarcaleeS.Alexander
DepartmentofPhysicalMedicineandRehabilitation,UniversityofAlabamaatBirmingham,
Birmingham,Alabama,USA
NatalieKozyrev,RachaelL.Bosma,andChaseR.Figley
6
1 CentreforNeuroscienceStudies,Queen’sUniversity,Kingston,Ontario,Canada
0
2
y
ar J.ScottRichards
u
n
a DepartmentofPhysicalMedicineandRehabilitation,UniversityofAlabamaatBirmingham,
J
5 Birmingham,Alabama,USA
1
7
3
6: PatrickW.Stroman
1
at CentreforNeuroscienceStudies,Queen’sUniversity,Kingston,Ontario,Canada
]
s
e
ri
a
r
b
Li Usingfunctionalmagneticresonanceimaging,theauthorsaimedtodeterminetherolesofthehuman
U
spinal cord in mediating sexual responses in women. Functional magnetic resonance imaging of
H
J theentirelowerthoracic,lumbar,andsacralspinalcordwasperformedusingasexualstimulation
[
y paradigm designed to elicit psychological and physical components of sexual arousal. Responses
b
d weremeasuredin9healthyadultwomenduring3consecutiveconditions:(a)eroticaudiovisual,(b)
e
d manualclitoral,and(c)audiovisualplusmanualstimulation.Functionalmagneticresonanceimaging
a
nlo results in healthy subjects demonstrate that this method is sensitive for mapping sexual function
w in the spinal cord, and identify several key regions involved in human sexual response, including
o
D theintermediolateralcellcolumn,thedorsalcommissuralnucleus,andthesacralparasympathetic
nucleus.Usingspinalfunctionalmagneticresonanceimaging,thisstudyidentifiedmanyofthespinal
cord regions involved in female sexual responses. Results from audiovisual and manual clitoral
stimulationcorrespondwithpreviousdataregardinglumbarandsacralneurologicchanges during
sexualarousal.Thisstudyprovidesthefirstcharacterizationofneuralactivityinthehumanspinal
cordunderlyinghealthyfemalesexualresponsesandsetsafoundationforfuturestudiesaimedat
mappingchangesthatresultfromsexualdysfunction,spinalcordtraumaordisease.
AddresscorrespondencetoDr.MarcaleeS.Alexander,700South19thStreet,Birmingham,AL35233,USA.E-mail:
[email protected]
Colorversionsofoneormoreofthefiguresinthisarticlecanbefoundonlineatwww.tandfonline.com/usmt.
FMRILOCALIZATIONOFSPINALCORDPROCESSING 37
INTRODUCTION
Untilthedevelopmentoffunctionalmagneticresonanceimaging(fMRI)anditsspecificadapta-
tion for use in the spinal cord (spinal fMRI), noninvasive in vivo studies of human spinal cord
activity were not possible (Leitch, Figley, & Stroman, 2010; Stroman, 2005; Stroman et al.,
2014).Therefore,toobtaininformationaboutspinalcordresponses,researchersreliedonanimal
experiments (Cai, Alexander, & Marson, 2008; Marson & McKenna, 1990; Truitt & Coolen,
2002)orclinicalstudiesoftheeffectsoftrauma(Courtois,Goulet,Charvier,&Leriche,1999;
Sipski, Alexander, & Rosen, 1995, 2001). While these studies have helped shape our current
understanding about the neural underpinnings of sexual responses, our knowledge of human
spinalcordfunctionduringfemalesexualstimulationisbasedonverylittledirectevidencefrom
neurologicallyintacthumans(Giuliano,Rampin,&Allard,2002;McKenna,2002).
6 On the basis of animal models and patients with specific neurologic injuries, we know that
1
0 thehumansexualresponseismediatedbyintegratedfunctionswithpsychogenic,reflexogenic,
2
y andsomaticcomponents(Giulianoetal.,2002;Levin&Riley,2007;Neuhuber,1982;Schroder,
r
ua 1980).Ingeneral,psychogenicsexualresponsesinhumansarecontrolledbythethoracolumbar
n
a segments, T11 to L2, whereas reflexogenic aspects are mediated by the sacral segments, S2 to
J
5 S5.Thepsychogeniccomponentismediatedbysympatheticpreganglionicneuronsinthedorsal
1
7 commissural nucleus and the intermediolateral cell column (Giuliano et al., 2002). In contrast,
3
6: the reflexogenic component is thought to be controlled by preganglionic neurons in the sacral
1
at parasympathetic nucleus and by pelvic and pudendal nerve afferents from the clitoris, which
s] terminate in the sacral parasympathetic nucleus, dorsal gray commissure, and the dorsal horn
e
ri (Marson & Murphy, 2006; McKenna, 2000, 2002; McKenna & Marson, 1997; McKenna &
a
r
b Nadelhaft, 1986). The somatic component of the sexual response is found in Onuf’s nucleus,
Li
U which controls muscle and sphincter contraction during penile erection and ejaculation in men
H (Schroder,1981;Xuetal.,2007),butispoorlyunderstoodinwomen’ssexualfunction.
J
y [ We recently published a spinal cord functional MRI study in able-bodied men (Kozyrev,
d b Figley, et al., 2012), which confirmed advanced spinal fMRI acquisition and analysis methods
de canbeusednoninvasivelytoelucidateneuralcorrelatesofhumanspinalsexualresponsesinmen.
a
o Therefore,thepurposeofthepresentstudywastousesimilarmethodstocharacterizespinalcord
nl
w activityinresponsetosexualstimulationinneurologicallyintactwomen.Weaimedtodetermine
o
D whichregionsofthethoracolumbarandsacralhumanspinalcordplayaroleinmediatingsexual
responses in healthy women, allowing us to validate, refute, or refine, long-held assumptions
frompreviousanimalandclinicalstudies.
METHOD
Participants
Spinal cord responses in the lower thoracic, lumbar, and sacral areas were measured in nine
healthyfemalevolunteers(meanage=20.7±2.3years).Allunderwentscreeningbeforetesting
toensurethattheyfounderoticasexuallyarousingandwerecomfortablewithgenitalstimulation.
TheQueen’sUniversityHealthScienceResearchEthicsBoardapprovedthestudyandallsubjects
providedwritteninformedconsentbeforeparticipating.
38 ALEXANDERETAL.
ArousalParadigms
Similar to our previous study of male sexual responses (Kozyrev, Figley, et al., 2012), this
study consisted of three conditions designed to elicit neural responses from psychological
and/or physical sexual activity. On the basis of previous studies (Sipski, Alexander, & Rosen,
1997; Sipski et al., 2001), conditions included audiovisual stimulation (AV) to elicit psy-
chological arousal/psychogenic responses; manual stimulation (MAN) to elicit physiological
arousal/reflexogenic responses; and combined audiovisual plus manual stimulation (COM) to
simultaneouslyelicitpsychogenicandreflexogenicresponses.
During the AV (i.e., psychogenic) condition, participants viewed erotic films in two 5-min
blocksseparatedbya3-minbaselinecondition(blankscreen)andprecededandfollowedby1.5-
minbaselineconditions.DuringtheMAN(i.e.,reflexogenic)condition,participantsperformed
6 self-stimulation of the clitoris using a small, custom-made, MRI-compatible vibrator without
1
0 eroticvisualstimuli(onlywritteninstructionsthatindicatedto“stimulate”or“rest”).Inthisway,
2
y MANwasperformedintwo1.5-minblocksofstimulationthatwereseparatedbya1.5-minrest
r
ua condition,andprecededandfollowedby1.5-minrestconditions.Thefinalstimulationparadigm
n
a (COM) involved a combination of audiovisual and manual genital self-stimulation, in which,
J
5 after an initial 1.5-min baseline condition, participants viewed erotic films and self-stimulated
1
7 with the vibrator for 20 min or until orgasm (which was verbally reported by subjects via the
3
6: in-scannerintercom).
1
at Toimposeconsistencyacrosssubjects,thesameeroticfilmswereshown,inthesameorder,
s] to all study participants during the first and third (AV and COM) study conditions. The erotic
e
ri filmswerechosenbecausetheyreceivedhighsubjectiverankings(greaterthan8outof10)by
a
r
b agroupofindependentviewerssimilartoKozyrev,Figley,andcolleagues(2012).Becauseour
Li
U goalwastohavesubjectsachievemaximumarousal,butnotorgasm,duringeachofthefirsttwo
H stimulation paradigms, the second part of the study (MAN) was shortened compared with the
J
y [ audiovisualcondition.
b
d
de SubjectiveMeasuresofSexualArousal
a
o
nl
w Before the study, subjects were informed that they would be asked throughout the study about
o
D their levels of mental and physical sexual arousal. For these purposes, mental sexual arousal
(MSA)wasdefinedasfeelingsandmentalimageryassociatedwiththedesireandmotivationto
engageinsexualbehaviors,whilephysicalsexualarousal(PSA)wasdefinedasphysicalchanges
suchasvaginallubrication.Subjectswereaskedtoratethesemeasuresonascalerangingfrom1
(verylittleornosexualarousal)to10(greatestpossiblesexualarousal)atthebeginningofthe
experiment,aswellasatendofeachofthethreestimulusconditions.Subjectswerealsoasked
toindicateverballyifandwhenanorgasmoccurred.
fMRIDataAcquisitionandAnalysis
Data acquisition and analysis procedures used in the present study were replicated from our
previousstudyonmalesubjects(Kozyrev,Figley,etal.,2012).Imagingwasbrieflyperformed
with a 3T whole-body MRI system. Participants were positioned on a phased-array spine coil,
alignedandcenteredattheT12vertebrallevel.Anopticalsensorwasattachedtotheseconddigit
FMRILOCALIZATIONOFSPINALCORDPROCESSING 39
ofthenonstimulatinghandtorecordtheperipheralpulsethroughoutthestudy.Initialthree-plane
localizer images were acquired, and were used for subsequent slice positioning. An optimized
spinalfMRIsequence(Stroman,Kornelsen,&Lawrence,2005)wasthenusedtomeasuretask
related signal changes based on a combination of BOLD and SEEP contrast (for review, see
Figley,Leitch,&Stroman,2010).Thiswasachievedbyacquiringpredominantlyproton-density-
weightedimagesfromninecontiguous,2-mm-thicksagittalslices—spanningthethoracolumbar
(T7–L2) and lumbo-sacral (L3–S4) spinal cord regions—using a half-Fourier single-shot fast
spin-echo(HASTE)MRIsequence,withthefollowingparameters:TE=38msec;TR=9sec
(i.e., 1 sec per slice); FOV = 280 mm × 140 mm, spatial resolution = 1.5 mm × 1.5 mm ×
2mm.Spatialsaturationpulseswereappliedtosuppresssignalsoriginatingfromanteriortothe
spineandoutsideoftheimagingfield-of-view,andflow-compensationgradientswereappliedin
thehead-footdirectiontominimizepotentialartifactsfromcerebrospinalfluidflow.Withthese
6 acquisitionparameters,each16-minparadigmfortheAVconditionspanned107fMRIvolumes,
1
0 the7.5-minparadigmforMANconditionspanned50volumes,andtheCOMconditionspanned
2
y upto147volumesoruntilthesubjectachievedorgasm.
r
ua Spinal fMRI data processing and statistical analyses were carried out using custom-made
n
a softwarewritteninMATLAB.Eachparticipant’sspinalfMRIdataweresubjectedtoafirst-level
J
5 analysis,whereregionsofspinalcordactivityrelatedtoeachtaskwereidentifiedusingamass
1
7 univariateapproach.Thetime-courseofeachvoxelwasfitwithagenerallinearmodeltoidentify
3
6: regionsthatweretemporallycorrelatedwitheachofthetaskparadigms.Thecomponentsofthe
1
at general linear model consisted of the task time-courses, a constant function, a linear ramp (to
s] account for scanner drift over the course of each scan), and three slice-specific physiological
e
ri motion regressors, as previously described (Figley & Stroman, 2009). Following the first-level
a
r
b analyses, each participant’s data were spatially normalized to a standard spinal cord template
Li
U (Stroman, Figley, & Cahill, 2008) to enable second-level (i.e., across-subject) random effects
H analysesforeachtaskcondition.Thus,activitymapswerecreatedtoidentifyregionsofconsistent
y [J activityduringAV,MAN,andCOMconditionswithuser-definedstatisticalthresholds(T>2.5
d b orT<–2.5,p<.008).
de Given the nature of spinal fMRI (and this study in particular), several precautions were
a
o takeninanattempttoreduceandeliminatepossiblemotionartifacts.First,subjectsweregiven
nl
w detailed information about the consequences of bulk motion in MRI data, and were instructed
o
D to “remain as still as possible throughout the experiment.” Second, masking tape was placed
across each subject’s torso and abdomen during the initial setup to increase awareness of bulk
motion.Third,spatialsaturationpulsesandflow-compensationgradientswereusedtominimize
signalsoriginatingfrombeyondthespinalcord,andduetocerebrospinalfluidflow,respectively.
Fourth,althoughspinalcordmotionisattenuatedinthelower-thoracic,lumbar,andsacralregions
comparedwiththecervicalandupper-thoraciclevels(Figley&Stroman,2007;Figley,Yau,&
Stroman,2008),time-seriesspinalfMRIdatawereanalyzedusingamotion-compensatinggeneral
linearmodelanalysistoaccountforcardiac-relatedphysiologicnoise(Figley&Stroman,2009).
Becauseofthepotentialforinvoluntarymusclecontractionsandbulkmotion,imagesacquiredat
thetimeofsubject-reportedorgasmwereexcludedfromthefMRIanalysis.Eachvolumeofthe
time-seriesdatawasalsocomparedwiththefirstvolumeoftheseries,andifthespatialcorrelation
dropped below a threshold of R = 0.85, the volume was not used in the general linear model
analysis.Thisaccommodatedbriefbulkbodydisplacements,anduseofsubsequentvolumesin
theanalysis,ifthepersonthenreturnedtotheiroriginalposition.
40 ALEXANDERETAL.
6
1
0 FIGURE1 Regionsofthespinalcordthatcontrolsexualresponsesinninehealthyable-bodiedwomen.(a)Histological
2
y referenceofspinalcordregions.(b)CombinedspinalfMRIdatasetsfromninehealthywomenforallthreeexperimental
r
a conditions:audio-visual(AV)stimulation,clitorisself-stimulation(SS),GenitalStimulation,andcombinedaudiovisual
u
n andclitorisself-stimulation(AV&SS).Foreachdataset,functionalactivityisoverlaidonhighresolutionanatomical
a
5 J imagesofthespinalcord(T/L=thoracolumbar,S=sacral).Thecolorbarontherightquantifiespositive(red)and
1 negative(blue)signalintensitychangescorrespondingtoincreasedanddecreasedactivityrespectively.
7
3
6:
1
at
] RESULTS
s
e
ri
a
br To ensure that subjects were sexually aroused during the procedures, each participant ver-
Li bally indicated their level of MSA and PSA at the end of each study epoch (Table 1). MSA
U
H was significantly higher during the CAM condition compared with either audiovisual or MAN
[J (p = .003). In contrast, PSA significantly increased during each successive phase of the study
y
b (p<.001).ResultsrevealthatMSAandPSAwerecloselylinkedduringtheAVandMANstudy
d
e phases,buttherewasnotasignificantdifferencebetweenMSAandPSAattheendoftheCAM
d
oa period.
wnl In the present study, fMRI results are thought to demonstrate the anatomical locations of
o responding spinal cord regions and the relative changes in (presynaptic) neural input to these
D
regions(Figleyetal.,2010;Logothetis,Pauls,Augath,Trinath,&Oeltermann,2001).Consistent
withtherodentliterature,theAVconditionelicitedspinalfMRIresponsesthroughoutthethora-
columbarspinalcordsofoursubjects,particularlyinregionsoftheintermediolateralcellcolumn
and the dorsal commissural nucleus (Figure 1). During AV stimulation, spinal fMRI responses
werealsoobservedintheleftsacralparasympatheticnucleusanddorsalhornofthesacralspinal
cord.Comparedwithbaseline,theMANconditionproduceddecreasedfMRIsignalintheright
ventral horn and increased fMRI responses in a large area to the right of the central canal in
thelumbarspinalcord,insegmentsL1–L4.DuringtheCAMcondition,negativethoracolumbar
responses were observed in the left ventral horn, while positive thoracolumbar signal changes
wereobservedintherightintermediolateralcellcolumn,rightventralhornandrightdorsalhorn,
and positive sacral cord responses were observed in the right sacral parasympathetic nucleus.
SimilarlytotheMANcondition,increasedsignalchangeswereobservedinalargeregiontothe
rightofthecentralcanalinthelumbarspinalcord,approximatelyinsegmentsL2-L4.
alone. alone
s] at 16:37 15 January 2016 1ExperimentalConditions Audiovisualandmanualstimulation SignificanceMSD 7.671.94Mentalarousalduringaudiovisualplusmanualstimulationwassignificantlyhigherthanmentalarousalduringaudiovisualormanualstimulation8.221.86Physicalarousalduringaudiovisualplusmanualstimulationwassignificantlygreaterthanduringmanualstimulationaloneandmanualstimulationwassignificantlygreaterthanduringaudiovisualstimulationalone.
rie LEus
Libra TABVario ation SD 0.83 1.54
oaded by [JHU usalLevelsDuring Manualstimul M 5.78 6.89 Nosignificantdifference
Downl Aro ualstimulation SD 1.90 1.82 Physicalarousalwassignificantlygreaterthanmentalarousal
s
ovi fi-
Audi M 5.11 4.89 Nosignicantdiffer-ence
e
c
n
a
c
Typeofarousal Mentalarousal Physicalarousal Statisticalsignifi
41
42 ALEXANDERETAL.
DISCUSSION
During this study, we assessed participants’ subjective reports of sexual arousal in comparison
withneuralactivitydetectedbymeansofspinalcordfMRI.Therewerenosignificantdifferences
inlevelsofreportedMSAorPSAduringtheAVstimulationparadigm.FunctionalMRIresults
obtained during AV stimulation revealed significant responses within the thoracolumbar cord.
Becausethesubjectswerenotperforminganyphysicalstimulationormovementduringthistime
andbecauseconditionsofincreasedsexualarousalwereconfirmedbyself-report,theseresults
provide further support for the theory that psychogenic sexual responses are controlled by the
sympatheticnervoussystem.Previousresearchinable-bodiedwomenandthosewithspinalcord
injuries,hasdemonstratedactivationofthesympatheticnervoussystemfacilitatesfemalesexual
arousal (Lorenz, Harte, Hamilton, & Meston, 2012; Meston & Gorzalka, 1996a, 1996b; Sipski
6 etal.,2001).Moreover,thereisaneffectofthedegreeofsympatheticstimulationinthatlowor
1
0 very high levels of sympathetic nervous system stimulation result in decreased genital arousal
2
y whereasmoderatelevelsofstimulationareassociatedwithhighergenitalarousal(Lorenzetal.,
r
ua 2012).DuringtheAVstimulationtherewerefMRIsignalchangesnotedinseveralareasofthe
n
a thoracic,lumbarandsacralcord.Thesefindingsareconsistentwithsubjects’self-reportsofPSA
J
5 in this paradigm. Furthermore the sacral spinal cord responses are likely also associated with
1
7 increased genital vasocongestion as a result of MSA as genital sexual arousal is reported to be
3
6: controlledbytheparasympatheticnervoussysteminanimalmodels(Caietal.,2008).
1
at During the MAN condition, subjects’ level of PSA was significantly greater than their level
s] ofMSA.Inaddition,PSAlevelsweresignificantlyhigherduringtheMANcomparedwiththe
e
ri AVcondition.FunctionalMRIresultsrevealeddiminishedactivityintheareaoftherightventral
a
r
b hornofthethoracolumbarspinalcord.Theobserveddecreasedactivityinthelumbarregionof
Li
U theventralhornmaysignifyadecreaseinneuronalinhibitoryinputfromsupraspinalregions.The
H nucleusparagigantocellularisinthebrainstemisknowntobeinvolvedintheinhibitionofspinal
J
y [ sexualreflexesandsendsdirectneuronalprojectionstotheventralhornandinterneuronsinthe
d b lumbosacralspinalcord(Marson,List,&McKenna,1992;Marson&McKenna,1992),which,
de in turn, project to nuclei in the lumbosacral spinal cord, including the sacral parasympathetic
a
o nucleus (Nadelhaft & Booth, 1984). Thus, the attenuation in descending inhibitory input may
nl
w accountfortheobserveddecreaseinactivityinthelumbosacralspinalcordinconjunctionwith
o
D subjects’reportsofincreasedPSA.
Subjects’reportedlevelsofMSAandPSAweresignificantlyhigherduringtheCOMcondition
compared with AV or MAN conditions. However, there was no significant difference between
subjects’reportedlevelsofMSAversusPSAduringtheCOMcondition.Inthisfinalparadigm,
fMRIresultsdemonstratedincreasedactivityinregionsoftherightintermediolateralcellcolumn,
ventralhornanddorsalhorninthethoracolumbarspinalcord.Inaddition,therewasincreased
activityintherightsacralparasympatheticnucleusandventralhornalongsidedecreasedactivity
intheleftventralhorninthelumbosacralspinalcord.Theseresultssupportthehypothesisofthe
reciprocalnatureofphysicalandmentalsexualarousalasaresultofpsychogenicorreflexogenic
sexualstimulationinhealthy,able-bodiedwomenandareconsistentwithpreviousfindingsthat
indicatemultipleneurologicinputsareinvolvedintriggeringgenitalsexualresponses(Caietal.,
2008;Giulianoetal.,2010;McKenna,1997;Sipski,2001;Sipski,Alexander,&Rosen,1999).
Moreover,theincreasedactivityobservedintheventralhornofthelumbosacralregionsduring
thecombinedstimulationconditionmaybeindicativeofincreasedneuronalinputtothepudendal
FMRILOCALIZATIONOFSPINALCORDPROCESSING 43
motorneurons immediately prior to the onset of orgasm in the female subjects. This finding is
consistentbothwiththerhythmicmotorcontractionsofpudendalinnervatedmusclesthatoccur
during orgasm and with published literature showing pharmacological stimulation of pudendal
motorneuronstriggersejaculatoryreflexesinmalerats(Staudtetal.,2011).
DuringboththeMANandCOMstimulationinthepresentstudy,increasedactivitywasob-
servedalongtherightsideofthelumbarspinalcordbetweentheL1–L4spinalsegments.There-
fore,givenitslocation,wesuspectthatthisactivitylikelyreflectsincreasedlumbarspinothalamic
cellactivity,whichhasbeenpreviouslybeenshowntocoordinateautonomicresponses,sensory
inputs and motor outflow to trigger ejaculation in male rats (Allard et al., 2005; Coolen et al.,
2004;Kozyrev,Lehman,&Coolen,2012;Staudtetal.,2011;Truitt&Coolen,2002).Thelumbar
spinothalamiccellsarelocatedinlumbarlevelsL3–L4surroundingthecentralcanalinlamina
X and the medial portion of lamina VII, and are activated specifically with ejaculation in male
6 rats.Inaddition,targetedlesionsofthelumbarspinothalamiccellsabolishesejaculationbutnot
1
0 othercomponentsofmaleratsexualbehavior(Coolenetal.,2004;Staudtetal.,2011;Truitt&
2
y Coolen, 2002; Truitt, Shipley, Veening, & Coolen, 2003). In this study, increased activity was
r
ua observedinthevicinityofthelumbarspinothalamiccellsduringthemanualstimulationandthe
n
a combinedstimulationconditions.Thesefindingsareincontrasttoapreviousstudythatreported
J
5 an absence of lumbar spinothalamic cell activation, as visualized with Fos-immunoreactivity,
1
7 followingvaginocervicalstimulationinfemalerats(Truittetal.,2003).Thisdissonancemaybe
3
6: associated with interspecies differences (i.e., humans in the present study vs. rats), the method
1
at of stimulation (i.e., clitoral stimulation in present study vs. vaginocervical stimulation), or the
s] sensitivity/specificityofthequantitativemeasure(fMRIinthepresentstudyvs.Fosstaining).1
e
ri Evidencesuggeststhattheclitorisformsfromthesamestructuresasthepenisduringembryonic
a
r
b development and that they are homologous organs (Baskin et al., 1999; O’Connell, Sanjeevan,
Li
U &Hutson,2005;Shih,Cold,&Yang,2013).Furthermore,arecentstudyreportedthepresence
H of the spinal pattern generator involved in the control of ejaculatory pattern in the female rat.
J
y [ Specifically, mechanical stimulation of the urethra, vagina and clitoris triggered an ejaculatory
d b motorpattern—asvisualizedfromrecordingsoftheurethralismusclecontractionsinthefemale
de rat (Carro-Juarez & Rodriguez-Manzo, 2006)—similar to the motor pattern observed in male
a
o ratsduringcontractionsofthemuscleofthebulbocavernosus(Carro-Juarez,Cruz,&Rodriguez-
nl
w Manzo, 2003; Carro-Juarez & Rodriguez-Manzo, 2008; Coolen, Allard, Truitt, & McKenna,
o
D 2004; Kozyrev, Lehman, & Coolen, 2012; Staudt, de Oliveira, Lehman, McKenna, & Coolen,
2010;Staudtetal.,2011;Truitt&Coolen,2002),amarkeroftheexpulsioncomponentofejacula-
tioninrats(Coolenetal.,2004)andmen(Gerstenberg,Levin,&Wagner,1990).Furthermore,in
apilotstudy,electricalstimulationattheL1–L2vertebrallevelhasbeenreportedtoelicitpleasur-
ablegenitalandorgasm-likesensationsinwomenwithorgasmicdysfunction;andtheorgasmic
dysfunctionreturneduponremovaloftheelectricalnervestimulation(Meloy&Southern,2006).
Inlightofthepreviousstudies,activityobservedintheregionofthelumbarspinothalamiccells
during the MAN and COM stimulation paradigms suggests that a spinal pattern generator for
ejaculationmayexistinwomenandmaybeinvolvedintheintegrationofautonomicoutflow(as
1NotethatfMRIresponsesarethoughttoreflectchangesinthemeansynapticinputtoagivenarea,asopposedto
thespikingactivity(output)oftheneuronslocatedthere(Figley&Stroman,2011;Logothetisetal.,2001),whereas
c-Fosstainingcodesforatranscriptionfactorthatisthoughttomediatelong-termchangesinneuralfunctioning(Kovacs,
2008).
44 ALEXANDERETAL.
indicatedbyactivitypresentintheintermediolateralcellcolumnandsacralparasympatheticnu-
cleus)withspinalactivityassociatedwiththeculminationofsexualactivity(indicatedbyactivity
inthedorsalhorninthelumbosacralspinalcord).Althoughithasbeenpreviouslyreportedthat
lumbarspinothalamiccellsarespecificallyactivatedwithejaculationandnotothercomponents
ofsexualbehavior(Staudtetal.,2011;Truitt&Coolen,2002;Truittetal.,2003),weobserved
increased spinal fMRI activity in the vicinity of the lumbar spinothalamic cells before orgasm,
as verbally reported by the female volunteers. This suggests that this region of the spinal cord
may receive inputs from the clitoris and supraspinal regions during sexual activity and begin
the process of integrating these inputs with autonomic outflow from the intermediolateral cell
columnandsacralparasympatheticnucleusbeforetheonsetandpossiblyaspartofatriggerfor
ejaculationand/ororgasm.
This study is unique in a number of ways. The present study is the first to report activity in
6 specificregionsofthethoracolumbarandlumbosacralspinalcordduringdifferentcomponentsof
1
0 sexualbehaviorinwomen.Inaddition,thepresentstudyusedfunctionalMRItovisualizeactivity
2
y in the spinal cord, which is believed to be related to neuronal input (Figley & Stroman, 2011;
r
ua Logothetisetal.,2001),whereaspreviousstudiesreportedneuronalactivationvisualizedwithC-
n
a FosandC-FosmRNAthatareindirectmarkersofrecentneuronalactivity(Kovacs,1998,2008)
J
5 andphosphorylationofextracellularsignal-kinasesthatareinvolvedincellsignalingpathways
1
7 (Murphy,Smith,Chen,Fingar,&Blenis,2002).Therefore,neuronalactivationinspecificregions
3
6: ofthespinalcordduringsexualactivityasvisualizedwithC-FosandpERKmaynotrepresent
1
at analogousregionsofthespinalcordreceivingneuronalinputsduringsexualactivityasvisualized
s] withfunctionalMRI.
e
ri Taken together, our findings provide support for the involvement of the thoracolumbar and
a
r
b lumbosacralspinalcordinfemalesexualfunctionandareconsistentwithfindingsinwomenwith
Li
U spinal cord injuries that demonstrate an intact sacral reflex arc is necessary to achieve orgasm
H (Sipski et al., 2001). The present results are also consistent with the finding of climactic-like
J
y [ pudendal motor neuron firing in response to pudendal sensory nerve stimulation (Cai et al.,
d b 2008).Wehypothesizethattheactivationofpudendalmotorneurons(detectedasactivityinthe
de ventralhorninthepresentstudy)isrelatedtotherhythmicmuscularcontractionsduringorgasm
a
o and the lumbar spinothalamic cells (corresponding to activity in the lumbar spinal cord to the
nl
w rightofthecentralcanal)areevidenceofheightenedsympatheticactivityatthemomentorgasm
o
D is triggered; however the specific role of this neurologic activity during female spinal sexual
reflexesatpresentremainsinconclusive.Thepresentreportservesasthefirstpieceofevidence
tosupporttheinvolvementofthelumbarspinothalamiccellsandpudendalmotorneuronsinthe
regulationofgenitalresponsesleadinguptoorgasminwomen.
ThelateralityoftheobservedfMRIresponsesraisestheissueofthehandednessofsubjects
and whether they preferentially stimulated different sides of the genitals. Unfortunately during
this study, we did not record handedness; however, data show that approximately 10% of the
general population is left-handed (Holder, 1997). In light of this, we hypothesize that subjects
preferentiallystimulatedtherightsideoftheirgenitals,whichmayhaveincreasedneuralactivity
on this side. Moreover, neuroplastic changes from historic repetitive genital stimulation could
have resulted in increased activity on the right side of their spinal cord. Further studies are
warranted to determine whether this laterality of fMRI changes is reproducible. In addition,
futurestudiesarewarrantedtodeterminewhetherchangesinfMRIsignalsinthethoracolumbar
andlumbosacralspinalcordcanbeusedtodocumentneuroplasticityasthereasonforefficacy
FMRILOCALIZATIONOFSPINALCORDPROCESSING 45
oftherapiessuchasclitoralvacuumsuction(Schroderetal.,2005),whichispurportedtowork
byattemptingtoretrainsacralreflexresponses.
The results of this study represent the first in vivo glimpse of human sexual responses in
theintactfemalespinalcordbymeansofspinalfMRI.Specifically,wehaveidentifiedthekey
regionsinvolvedinthecentralnervoussystemcontrolofsexualarousalinwomen,whichinclude
the intermediolateral cell column, sacral parasympathetic nucleus, and the dorsal commissural
nucleus. Although fMRI may not afford the spatial resolution to specify the involvement of
specificnuclei,ourresultsareconsistentwithregionsshowntobeinvolvedinvariousaspectsof
sexualfunctioninanimalstudies.ThepresentstudyclearlydemonstratesthatspinalfMRIcanbe
usedtomaphumanfemalesexualresponsesandholdspromisetoprovideessentialinformation
fortreatmentplanningandoutcomemonitoringforwomenwithsexualdysfunction.Furthermore,
useofsequentialfMRIstudiesmaybeusefultoelucidateneuroplasticityassociatedwithvarious
6 treatments(e.g.,clitoralvacuumsuction)forfemalesexualdysfunction.
1
0
2
y
ar FUNDING
u
n
a
J
5 FundingforthisresearchwasprovidedbytheCraigNeilsenFoundation(Grant#2786).
1
7
3
6:
1 REFERENCES
at
]
s
e Allard,J.,Truitt,W.A.,McKenna,K.E.,&Coolen,L.M.(2005).Spinalcordcontrolofejaculation.WorldJournalof
ri
a Urology,23,119–126.
r
Lib Baskin,L.S.,Erol,A.,Li,Y.W.,Liu,W.H.,Kurzrock,E.,&Cunha,G.R.(1999).Anatomicalstudiesofthehuman
U clitoris.JournalofUrology,162,1015–1020.
H Cai,R.S.,Alexander,M.S.,&Marson,L.(2008).Activationofsomatosensoryafferentselicitchangesinvaginalblood
[J flowandtheurethrogenitalreflexviaautonomicefferents.JournalofUrology,180,1167–1172.
y Carro-Juarez,M.,Cruz,S.L.,&Rodriguez-Manzo,G.(2003).Evidencefortheinvolvementofaspinalpatterngenerator
b
d inthecontrolofthegenitalmotorpatternofejaculation.BrainResearch,975,222–228.
e
d Carro-Juarez,M.,&Rodriguez-Manzo,G.(2006).Evidenceforthepresenceofthespinalpatterngeneratorinvolvedin
a
o thecontrolofthegenitalejaculatorypatterninthefemalerat.BrainResearch,1084,54–60.
wnl Carro-Juarez,M.,&Rodriguez-Manzo,G.(2008).Thespinalpatterngeneratorforejaculation.BrainResearchReviews,
o 58,106–1020.
D
Coolen,L.M.(2005).Neuralcontrolofejaculation.JournalofComparativeNeurology,493,39–45.
Coolen,L.M.,Allard,J.,Truitt,W.A.,&McKenna,K.E.(2004).Centralregulationofejaculation.Physiologyand
Behavior,83,203–215.
Courtois,F.J.,Goulet,M.C.,Charvier,K.F.,&Leriche,A.(1999).Posttraumaticerectilepotentialofspinalcordinjured
men:Howphysiologicrecordingssupplementsubjectivereports.ArchivesofPhysicalMedicineandRehabilitation,
80,1268–1272.
Figley, C. R., Leitch, J. K., & Stroman, P. W. (2010). In contrast to BOLD: Signal enhancement by extravascular
waterprotonsasanalternativemechanismofendogenousfMRIsignalchange.MagneticResonanceImaging,28,
1234–1243.
Figley, C. R., & Stroman, P. W., (2007). Investigation of human cervical and upper thoracic spinal cord motion:
Implicationsforimagingspinalcordstructureandfunction.MagneticResonanceinMedicine,58,185–189.
Figley, C. R., & Stroman, P. W. (2009). Development and validation of Retrospective Spinal Cord Motion Time-
CourseEstimates(RESPITE)forspin-echospinalfMRI:Improvedsensitivityandspecificitybymeansofamotion-
compensatinggenerallinearmodelanalysis.Neuroimage.44,421–427.
Figley,C.R.,&Stroman,P.W.(2011).Therole(s)ofastrocytesandastrocyteactivityinneurometabolism,neurovascular
coupling,andtheproductionoffunctionalneuroimagingsignals.EuropeanJournalofNeuroscience,33,577–588.