Saturable transport of insulin from plasma into the central nervous system of dogs in vivo A mechanism for regulated insulin delivery to the brain

Saturable transport of insulin from plasma into

the central nervous system of dogs in vivo. A

mechanism for regulated insulin delivery to the

brain.

G D Baura, … , C Cobelli, M W Schwartz

J Clin Invest.

1993;

92(4)

:1824-1830.

https://doi.org/10.1172/JCI116773

.

By acting in the central nervous system, circulating insulin may regulate food intake and

body weight. We have previously shown that the kinetics of insulin uptake from plasma into

cerebrospinal fluid (CSF) can best be explained by passage through an intermediate

compartment. To determine if transport kinetics into this compartment were consistent with

an insulin receptor-mediated transport process, we subjected overnight fasted, anesthetized

dogs to euglycemic intravenous insulin infusions for 90 min over a wide range of plasma

insulin levels (69-5,064 microU/ml) (n = 10). Plasma and CSF samples were collected over

8 h for determination of immunoreactive insulin levels, and the kinetics of insulin uptake

from plasma into CSF were analyzed using a compartmental model with three components

(plasma-->intermediate compartment-->CSF). By sampling frequently during rapid changes

of plasma and CSF insulin levels, we were able to precisely estimate three parameters

(average standard deviation 14%) characterizing the uptake of insulin from plasma, through

the intermediate compartment and into CSF (k1k2); insulin entry into CSF and insulin

clearance from the intermediate compartment (k2 + k3); and insulin clearance from CSF

(k4). At physiologic plasma insulin levels (80 +/- 7.4 microU/ml), k1k2 was determined to be

10.7 x 10(-6) +/- 1.3 x 10(-6) min-2. With increasing plasma levels, however, k1k2

decreased progressively, being reduced sevenfold at supraphysiologic levels (5,064

microU/ml). The apparent […]

Research Article

Find the latest version:

Saturable

Transport

of Insulin from

Plasma

into the Central Nervous

System

of

Dogs In Vivo

A Mechanism for Regulated Insulin Deliverytothe Brain

Gail D.Baura, David M.Foster,DanielPorte, Jr.,StevenE.Kahn,RichardN.Bergman,* Claudio

Cobelli,*

and MichaelW.Schwartz

Departments ofBioengineeringandMedicine, University of Washingtonand Seattle VeteransAffairsMedicalCenter, Seattle, Washington98195, *Department ofPhysiologyandBiophysics, University ofSouthernCalifornia, LosAngeles,California 90033;and tDepartmentofElectronics andInformatics, Universita diPadova, 35131Padova,Italy

Abstract

Byactingin thecentralnervous system,circulatinginsulin may regulate food intake and body weight. We have previously

shown thatthekineticsof insulinuptakefromplasmainto

cere-brospinal fluid (CSF) can best be explained by passage

throughanintermediate compartment. Todetermine if

trans-port kinetics into this compartment wereconsistent with an

insulinreceptor-mediatedtransport process, wesubjected

over-night fasted, anesthetizeddogstoeuglycemic intravenous

insu-lin infusions for 90 min over a wide range of plasma insulin

levels(69-5,064,U/ml) (n =10).Plasma andCSF samples

were collected over 8 h for determination ofimmunoreactive

insulinlevels, andthe kinetics of insulin uptake from plasma into CSF were analyzed using a compartmental model with three components (plasma -- _{intermediate compartment}

-CSF).

By

sampling frequently during

rapid

changes of

plasma

andCSF insulin levels,we wereableto

precisely

estimatethree parameters (average standard deviation 14%) characterizing

the uptake of insulin from plasma, through the intermediate

compartment andintoCSF

(klk2);

insulinentryintoCSFand

insulin clearance fromtheintermediatecompartment(k2+k3);

and insulin clearance fromCSF

(k4).

At physiologic plasma insulinlevels

(80±7.4,gU/ml),

klk2wasdetermined to be10.7

X

10'±+1.3

X10-6

min2.

Withincreasing plasma levels, how-ever,

k1k2

decreasedprogressively, beingreducedsevenfoldat

supraphysiologiclevels(5,064

,U/ml).

The apparent KM of

this saturationcurvewas 742 uUU/ml ( - 5 nM).In contrast,

the rate constants for insulinremoval from the intermediate

compartment andfromCSF didnotvary with plasmainsulin

(k2

+k3=_{0.011±0.0019}

min'

and k4=0.046±0.021

min').

We conclude thatdelivery of plasma insulin intothe central nervoussystemissaturable, andislikely facilitatedbyan

insu-lin-receptormediatedtransport process.(J. Clin.Invest. 1993.

92:1824-1830.)Key words: central nervous system * compart-mental model *insulinaction * SAAM modeling program

Introduction

Several lines of

investigation

suggest that

circulating

insulin

levels within the physiologic range exert biologic effects on

functioning

ofthe central nervoussystem (CNS). For exam-Address reprint requests to Dr. Michael W. Schwartz, Metabolism

(151),VaMedical Center, 1660 South Columbian Way, Seattle, WA 98108.

Receivedfor publication 10 September 1992 and in revisedform 9 May 1993.

ple, experimental elevations of plasma insulin to postprandial values suppress feeding behavior (1, 2), an effect also observed

duringdirectinsulin infusion into the brain (3). That this

oc-curs despite the extensive network of endothelial tight

junc-tions in cerebral vasculature comprising the blood-brain barrier, which limits diffusion of peptides of this size, suggests that circulating insulin may enter the brain via a facilitated transport process (4). In vitro studies (5,6)demonstratingthat insulin transport across vascular endothelium isdependenton itsbinding to endothelial insulin receptors have led to the

hy-pothesisthatreceptor-mediatedinsulin transport may facilitate thedelivery ofcirculatinginsulin to target tissues in vivo. Moover, brain microvascular endothelium expresses insulin re-ceptors(7), and theuptakeofplasma insulin into cerebrospi-nal fluid (CSF)' demonstrates specificity when comparedto analogues with reducedaffinityfor theinsulin receptor, such as

proinsulin (8). These and other studies (9) providesupport for the hypothesis that the entry ofcirculating insulin into the brain isfacilitated by an insulin receptor-mediatedtransport process.However,quantitativemeasurementsof insulin trans-portkineticsinto theCNShave yet toaddressthepossibility of

saturable insulin transport.

In ourprevious work (I0),weprovidedevidence that insu-linuptakefromplasmainto CSFrequirespassagethroughan

intermediatecompartment, andreportedontheuseofa

com-partmentalmodel to facilitate theanalysisof thisuptake pro-cess. We now report thequantitationofthedose-response

re-lationship between the plasma insulin level and kinetics of

insulinuptakefromplasma,throughanintermediate

compart-ment,and intoCSF. Wehypothesizedthat ifinsulinreceptors,

localizedtoeither the blood-brainorblood-CSF barrier,

facili-tatethisuptakeprocess, then the rate constantcharacterizing CSF insulin uptakeshould besaturable,decreasing in magni-tude with increasing plasma insulin levels. Moreover, the

plasmainsulin concentrationatwhichhalf-maximaltransport occurs(KM)should becomparabletothedissociationconstant

(KD)

of the insulinreceptor.

Methods

Studyanimals and conditions. Six normal adult male mongrel dogs weighing 20-30 kg were studied after an overnight fast. Their care was

supervisedbyalicensed veterinarian. The dogs were housed in individ-ual cages thatincluded a 5X20ft area for exercise and were fed 0.3 kg/d of dry standard dog laboratory diet (Wayne Pro-Mix, Allied

The Journal of Clinical Investigation, Inc. Volume 92, October 1993, 1824-1830

1. Abbreviations used in this paper: CSF, cerebrospinal fluid; IRI, im-munoreactive insulin.

Mills, Inc., Memphis, TN) with unlimited access to water. For each study, eachdog was anesthetized with thiamylol (Surital, Parke-Davis, Morris Plains, NJ), 20 mg/kg i.v., and was placed on mechanical ven-tilation, with 1-2% halothane and 40% 02- Intravenous catheters were placed for both sampling and infusion, and the cisternum magnum was cannulatedwithano. 22gaugespinal needle for sampling of CSF, as reported previously (10). Body weight was measured immediately be-fore each study. All studies were approved by the Animal Care Com-mittee of the Seattle VA Medical Center.

Insulin infusion and sampling. The study protocol consisted of a

90-mm i.v.insulin infusion period with sampling of plasma and CSF for a total of 490 min. Intravenous insulin was administered as a 3-min primed infusion (8.5-50 mU/kg-min) commencing at t=0min, fol-lowed by a continuous infusion for 87 min at 20% of the primed infu-sion rate. Blood samples ( 1.8 ml) were obtained at t= -10, -5,1,2, 3, 4, 6, 8, 10, 13, 16, 20, 25, 30, 35, 40, 65, and 90min,at5-minintervals for 90<t< 150min,and at20-min intervals for 150<t<490 min. CSFsamples (0.4 ml) were obtained at t= -10, -5, 20, 40, 65, 90, 95, 100, 110, 120, 135, and 150 min and thereafter at 20-min intervals until 490 min. In three studies(P81, P520, and P1195), additional blood and CSFsamplesweretaken at20-min intervals for 510 < t

<590 min. Euglycemia was maintained in all studies by a variable-rate i.v. infusion of 50% dextrose with on-line monitoring of blood glucose levelsusing a hand-held, computerized glucose meter (Glucoscan, Life-scan, American Medical Systems,Cincinnati,OH). Blood and CSF samples were placedonicebeforeprocessing.After bloodseparation,

bothplasma and CSF were frozenat-20'Cuntil assay. Plasma levels of immunoreactive insulin (IRI)weremeasured induplicateby radio-immunoassay using amodification of the double-antibody method (11). CSFIRIwasmeasured in triplicate usingafurther modification of this method which enhances itssensitivity (8, 12).Plasmaglucose

was determined by the glucose oxidase method with a glucose autoana-lyzer (Beckman Instruments, Inc., Brea,CA).The averageplasma in-sulinconcentration duringinfusion, Pave,wasdeterminedby takingthe average ofplasmainsulin valuesduringtheinterval 1 . t<90 min. Twodogs were studied once at eitheralow orhighinsulin dose, and four were studiedtwice,atbothalowandhighinsulin dose.Thus,10 independent data sets were obtained.

Multicompartmentalmodel.Aschematic illustration of the multi-compartment model usedtointerpretthedatais shown inFig. 1; this modelwaspreviouslydescribed in Schwartzetal.(10).Because of the delay in the appearanceofinsulin in the CSF after intravenous insulin

infusion,insulin entry from plasma into CSF can best be explained by passage through an intermediate compartment, INS,(t). Insulin in plasma, INSp(t), enters CSF,INSc(t),afterpassing throughINSI(t),

and irreversible losscanoccurfrom bothINSc(t)andINSI(t).INSp(t)

providesafunctionaldescriptionof theplasmadataproviding inputto themodel. Byassumingthatflux of insulin from the CNS back into plasma has a negligible influence onthe plasma concentration, the needtopostulate a structure for the kinetics of insulin outside of the brain iseliminated.

Inthismodel,

k,

represents theuptake of plasmainsulin into the intermediate compartment, andk2represents themovementof insulin from the intermediate compartment into CSF. k3 represents insulin clearance from the intermediate compartment, andk4represents insu-lin clearance from CSF. Allrateconstants areexpressedinmin-'.

Using this model,the ratesof change of exogenously administered insulin intheintermediate compartment,

INS,(t),

andof CSFinsulin,

INSc(t),aredescribed bythefollowingdifferentialequations:

d-INS(t)

=

kINSp(t)

-(k2+k3)INSI(t) (1)

dt

dINSc(t)

=_k2INSI(t)-

_k4INSc(t)

(2)

INSp(0)

=

_INSJ(0)

=_INSc(0)=0 (3)

All insulin variables,INSX(t),areexpressedinconcentration units of microunitspermilliliter.Toobtain the initialconditions of Eq.3, basal

plasmaand CSF levels were subtracted from all values of each data set beforemodeling,such that for modeling purposes the insulin concen-tration in each compartment was set equal to zero at t=0min. There-fore, only the kinetics of infused insulin were analyzed, and it was not necessary to assume that at the basal state all CSF IRI must be derived from plasma. Because the intermediate compartment was not directly sampled, only the rate constant characterizing insulin clearance from CSF,k4, can be identified independently, whereas

_ki,

k2, and k3 are not independently identifiable. Instead, the product

_klk2,

whichrepresents thethroughput of insulin from plasma, through the intermediate com-partment, and subsequently into CSF, and sum (k2+k3), which repre-sentsinsulin flux from the intermediate compartment into CSF as well asinsulin clearance from the intermediate compartment, were esti-mated.Adetailed derivation of unique parameter identification is pro-vided in the Appendix.

Parameteridentification.By using the mathematical modeling pro-gramSAAM ( 13) onamodel 486 computer (Gateway 2000, Gateway, SiouxCity, SD), the rate constants kk2(min2),(k2+k3) (min-'),

andk4(min-')were estimated from the plasma and CSF insulin data. To reduce the number of samples and therefore total fluid volumes required per study, only CSF insulin data points taken during t 2 90 min were used for modeling. It can be shown that as long as enough plasma and CSF samples are taken and steady state is reached, the same parameters will be obtained from CSF samples taken at 0<t<3, 3<t

<90,or t. 90 min ( 14). The intraassayerrorsfor the plasma and CSF insulinradioimmunoassays were assessed by measuring eight samples of various insulinconcentrations (data not shown). These error esti-mates(plasma insulin assay4%for concentrations<80

AU/ml,

12%

otherwise;CSF insulin assay 1 1%)wereusedby SAAMtodetermine theoptimalcurvefit andtoidentify the rate constant parameters. For each parameter, thecorrespondingcoefficient of variation(CV),also known asthe fractional standard deviation, was determined from the covariance matrix generated by SAAM.

IndependentestimationoftheCSF insulinclearanceparameter,ki4.

Toobtainan independent estimate of k4, the CSF insulin clearance parameter, insulin (750/Uin 750

_AI

ofsaline)wasinfusedover1 min into thecistemum magnum of three dogs which had been fasted over-night, anesthetized, and maintained on mechanical ventilation.

Imme-diately before theinfusion wasadministered, 750MlofCSFwas re-moved from the cisternum magnum, in ordertopreserve theoriginal

volume of CSF during the infusion. CSF samples (0.4 ml)were ob-tainedat t= _-10,-5,2, 5, 10, 15, 25,and 40 min, and thereafterat 20-min intervals till 240 min. The resulting CSF levelsdecayed

biex-ponentially,andwerefittedto atwo-compartment model. This two-compartment analysishaspreviously been usedto model transport kineticsofsubstances administeredintracerebroventricularly,withone compartmentrepresentingCSF and the otherrepresentingbrain inter-stitium (15, 16).TheCSF clearanceratewascalculatedbydividingthe estimated bolus concentration administered by the area under the curveof thetwoexponentials( 17).

EstimationofKM,theplasmainsulin concentrationatwhich

half-maximaltransport occurs. To estimateKM,theplasmainsulin concen-trationatwhich half-maximal transport occurs, the transportvelocity

for various plasma insulin concentrations wasrequired. Velocity is determined bymultiplyingki, therateconstantrepresenting uptakeof plasma insulin into the intermediate compartment,bythe

correspond-ing plasma insulin concentrationduring infusion,Pave.AskI couldnot

beuniquely identified,theproduct

klk2,

whichestimatedki,wasused for this calculation. Therefore,velocity*k2, which isdirectly propor-tionaltotherateof insulinuptakefromplasmaintoCSF,wasinstead calculated.Velocity*k2duetononspecificuptakeislinear,such thatat

high plasma insulinlevels, velocity*k2doesnotplateau.Asthis linear component is readily apparent at high plasma insulin levels,

ve-locity *2due tononspecific uptakewasestimatedthrough

graphical

analysisas2 X 10-7*Pave.Thisestimatewassubtracted from

veloc-ity*k2toisolatevelocity*k2duetosaturableuptake (18).

Velocity*

k2

wasthenplottedas afunction of Pave. KMwascalculated by usinga

Velocity*k2=

VMAxk2Pave

PaVe+KM

tofit theresultingcurve.Inthisequation, VMAX=maximumvelocity. Results

The90-minintravenous insulin infusionprotocolresulted in a

characteristic timecourse ofchangein CSF IRI levels,

illus-tratedinFig.2.Duringi.v.insulininfusion,CSF insulin levels rosegradually,reaching peak valuesat105min ( 15 minafter

discontinuationofinfusion),and fellexponentiallythereafter. The estimated parameters and corresponding coefficients of

variation for each studyaregivenin Table I. ThemeanCV for

allestimatedparameterswas 14%.

Thedata in Table I demonstrate that studies conducted at low plasma insulin levels are associated with relatively high valuesof

k1k2.

Thiswasevidentwhen

examining

all

dogs

asa

group, as well as whencomparing results obtained in separate

studiesat twodifferentplasmalevels.Examiningdata obtained

from two studies in individual dogs, the value of

k1k2

was

higher in studiesconductedatlowerplasmainsulin levels in all cases, whereas no consistent

relationship

between either

(k2

+k3)ork4 and theplasmalevelwasevident.

To further characterize these

relationships,

klk2,

(k2+k3), andk4wereplottedas afunction of the averageplasmainsulin

concentration

during infusion,

Pave, to determine dose-re-sponse

relationships.

Asshown in

Fig. 3,

theparameter

charac-terizing

insulinuptakefromplasmainto CSF variedinversely with plasma insulinlevels,

being

reduced sevenfoldfrom

physi-ologic to

supraphysiologic

levels. The apparent KM for this transport process wasdetermined from a Michaelis-Menten

dose-responsecurvebasedonthedata in Table I. As shown in

Fig. 4, the apparent KMwas

approximated

tobe 742

,uU/ml

(

- ₅

_nM).

_{Incontrast,the}meanrate constantvaluesfor

insu-lin removal from the intermediatecompartment

(k2

+

_k3)

and

insulinclearance from CSF

(k4)

didnotappeartovary with plasma insulin

(k2

+

k3

= 0.011±0.0019

min-'

and k4

=0.046±0.021

min-') (Fig. 5).

Therefore, delivery

of

circu-lating insulin into the

intermediate

compartment and

subse-quently

into CSF demonstrates saturable

kinetics,

whereas

in-sulin clearance from the intermediatecompartment doesnot.

Data are insufficientto assessthe

possibility

thata saturable

processalsocontributestoCSFinsulinclearance.

Tovalidate the

model-derived

parameter

representing

CSF insulin clearance, the meanCSFclearance rate constant was

INS(t)

k1 _(t)kk

Figure 1. Schematic illustration of the pharmacokinetic model used to analyze the kinetics of insulin uptake from plasma into CSF in the

dog.Inthismodel,

k1

represents theuptake of plasma insulin into the intermediate compartment, hypothesized to occur across the blood-brainbarrier,andk2 represents the transport of insulin from the in-termediate compartment into CSF. k3 represents insulin clearance from the intermediate compartment, and k4 represents insulin clear-ancefrom CSF. All rate constants are expressed inmin-1.Modified from Schwartz et al.(10).

TableI.

Identified

Parametersfrom Optimal CurveFits

ofDatafrom Nine studies in Dogs of Various Insulin Doses

Study P.". k,k2 k2+k3 k4

UU/mI min-2 min-'

P69 69 10.2e-6 0.0086 0.060

(14%) (11%) (16%)

P81 81 9.1e-6 0.014 0.043

(14%) (28%) (16%)

P83 83 11.5e-6 0.011 0.10

(16%) (8%) (19%)

P86 86 12.0e-6 0.012 0.047

(10%) (19%) (23%)

P520 520 6.7e-6 0.011 0.026

(4%)

(5%)

(8%)

P989 989 7.6e-6 0.0095 0.035

(12%) (18%) (22%)

P1195 1195 3.8e-6 0.011 0.034

(16%) (24%) (22%)

P1203 1203 3.1e-6 0.012 0.032

(10%) (11%) (19%)

P2076 2076 3.4e-6 0.011 0.042

(13%) (12%) (18%)

P5064 5064 1.6e-6 0.0073 0.042

(8%) (6%) (12%)

Instudies P81 andP1194,plasmainsulin samples were not taken att

= 1,2, 3, 4, 6, 8, 10, 13, and 16min,which resulted in higher coeffi-cients of variation than in studies in which these samples were ob-tained.Studies conducted in the same dog were P69 and P989, P81 andP1203,P520andP1194,and P86and P5064.

alsodetermined independentlybymodeling datafrom intra-cisternal insulin injection (n = 3).Theintracisternal insulin

injection protocol resulted in a characteristic time course of

changeinCSFIRIlevels,illustrated in Fig.6. Themean CSF

insulinclearance ratedetermined from these studies was 0.028

±0.0080

min-',

which is withinthe errorofthe meanvalue of

k4

determined during i.v. insulin infusion (Fig. 7). These

re-sultsvalidate the CSFinsulin clearance parameter identified

during

intravenous insulin infusion.

Discussion

Inthe currentstudies,weanalyzedthe kinetics of CSFinsulin uptake from plasmausingapreviouslypresented

multicom-partmental model ( 10). Our results indicatethat theefficiency

with which circulatinginsulin enters CSF via passage through anintermediatecompartmentvaries inverselywith theplasma

insulinconcentration. These data support the hypothesis that

insulinentryintothe CNSfromplasma is facilitated by a satu-rable transport process. As the efficiency of uptake into the

CNSdecreases sevenfold with increasingplasmainsulin

con-centration,weestimatethat 85%ofthisuptake ismediated by a

saturable process atphysiologicplasma insulin levels (< 100

,uU/ml).

The apparentKM of this transport process was 742

/gU/ml (- 5 nM), which is similartothe KDofthe insulin

receptor identified in brain microvasculature (2 nM) (19).

Therefore,data areconsistent with the hypothesis that insulin receptorsfacilitatetransportofplasma insulininto thecentral nervoussystem.

c

=v83PU/mi

E

3EXL

>

Ic

('₀₎ _ ₁₍₎₍₎M _ _2(X()W __.300(t _ 4(X) _5(X)_ _

time(min)

._ 0).

_

-().

c 8(1 B. c

4_,)

c

-._

E EL

c _

._ _

V, _4

c;

6 .4

.2

0

X] O)

I(X) 200 (X) 400

time(min)

V

.hd 4₇

5(X)

Pave=520ilU/ml

OF ^- - - _^^ be-- _

^_-0 100 200 300 400 500 600

time(min) 6

4 L 2

0 1(X) 200 300 400 500 600

time(min)

In

0

0 1000 2000 3000 4000 5000 Average Plasma InsulinDuring Infusion (iU/ml)

6000

Figure3. Dose-responsecurveillustratingtherelationship between

therateofuptakeofplasmainsulin into the intermediate

compart-mentand subsequentlyinto CSF.

kjk2

decreases sevenfold with in-creasing plasmainsulin levels from thephysiologicto supraphysio-logicrangeduringintravenous insulin infusion. Eacherrorbar repre-sentsthe fractional standard deviationassociatedwith theidentified

parametervalue.

c

,_1.2(X)

C. S*E 8(X)

:D

-c

r__

C

P?==1195pU/ml

t

) I(X) 200 30) 4(X) (X) time(min)

8

[

6[

4

(mm)

()

I(X) 2(X 3() 4X) S(X) time(min)

the choroidplexus. One hypothesisto accountfor this observa-tionis that insulinentering the choroid plexus is sequestered by choroidal insulinreceptors, whichareexpressedathigh

con-centrations, preventing its diffusion into CSF, andarepossibly

6(X) responsiblefordegrading insulinfollowing binding,asoccurs

in the placenta (20, 21 ).

The intermediate compartment through which insulin

passes en route toCSF likely represents either brain ISFor

choroidplexus epithelium. Although the identity of this

com-partmentcannotbeestablished without undertaking the diffi-60( cult task of sampling thecompartmentitself, dataareavailable

which indirectly addressthesetwopossibilities. The observa-Figure2.Representativestudies in which the kinetics of CSFuptake

ofintravenouslyinfused insulinweredetermined. Insulinwasinfused during0 t 90min.Studies shownarerepresentativeof low(A), medium(B), andhigh (C)insulin doses.Upper panels:timecourse

ofplasmainsulin. Lowerpanels:timecourseof CSF insulin and

op-timalcurvefit,from whichrate constant parameterswerederived.

Two barrier systems exist in the central nervous system

whichregulate solutemovementbetween the blood and CNS. The first oftheseis the blood-brain barrier, which arises from epithelial-like tight junctionsthat virtuallycement adjoining capillary endothelium together in the brain microvasculature. Similarly, the blood-CSF barrier arises from tight junctions

be-tweenadjacentepithelial cells investing the choroid plexus, the primary site of CSF formation. Ingeneral, mostsolutesgain

access from blood to CSF via the blood-CSF barrier at the choroid plexus. Transport of solutes such asNa' occurs

di-rectly from bloodtoCSF, and doesnotnecessitate postulating

transport throughan intermediatecompartment(16). Thus, while insulin maygain accessto CSFvia either of the CNS barrier systems, the observation that plasma insulin passes

throughanintermediatecompartmentenroutetoCSFattests notonlytothe uniqueness with which insulinenterstheCSF,

but suggeststhatabarriertodiffusion ofinsulin existsacross

tion that insulinconcentrations inratbrainextractsare -

30-0.008

ll.l .5

E E

=L

1-1

eq

19 >1

8

lu 0.006

0.004-

0.002-a

a

0 1000 2000 3000 4000 5(000

Average Plasma Insulin During Infusion (pU/ml)

Figure4.Michaelis-Mentencurve,basedoncalculatedkk2,therate

of insulinuptakefromplasmaintointermediatecompartment and subsequentlyintoCSF,for various insulin doses. The apparentKM, theconcentrationduringwhich halfmaximal transportoccurs,is

742_,U/ml(-. 5nM).

SaturableTransport ofPlasma Insulin into theCentralNervousSystem 1827 A.

.kw

By

a

Velocity*_k2 =0.0074P

P +74

ave

a

Pave 42

estimated initial insulinconc.=₆₁

_ptU/ml

100o

ok2 +

_k3

Ak

a

i

?

o

Q9

0

=._

cw

ii'

UA

CSF

clearance rate

=0.036min1 (3%)

_ ft

.

_

10o

100

0 O 1000 2000 3000 4000 5000

AveragePlasma Insulin During Infusion_(pU/ml) 6000

10 Figure5.Dose-responsecurvesillustratinghowratesof insulin

trans-portinto CSF and insulinclearance from intermediatecompartment (k2+k3) andrateof CSFinsulin clearance(k4)varywithaverage plasmainsulin levelduringinfusion. Eacherrorbarrepresentsthe fractionalstandard deviation associatedwith the identifiedparameter value.

U

.O

-3

co

I L. . .

o 50 100 150

time(min) 200 250

estimated initial insulinconc.=69

gU/ml

CSFclearancerate=0.027mind (3%)

A

-AA

0 50 100 150

time(min)

200 250

40% of plasma levels (22), whereas CSF levelsare < 5% of

plasma levels, for example,suggests that insulinuptakeoccurs

preferentially across the blood-brain barrier, rather than the

blood-CSF barrier. Moreover, blood-brain barrier

endothe-liumexpressesinsulinreceptors(4, 23),and afterintracarotid injection of radiolabeled insulin, penetration of insulintracer

into brain parenchyma was inhibited by coadministration of excess insulinin newborn rabbits (9), suggesting asaturable

transport process within the blood-brain barrier. Taken

to-gether, these datasupport thehypothesisthatasaturable trans-portsystemoperatesatthe level of the blood-brain barrierto promotethe delivery of circulating insulintotargetsites in the brain. Analogous blood-brain barrier transport systems have

been described for glucose (24),aminoacids (25), plasma

pro-teins suchastransferrin (26), and peptides suchasenkephalins

(27). Iftransportacrossthe blood-brain barrierrepresentsthe majorroute by which circulating insulinenters the CNS, as

suggested by these observations, then the intermediate

com-partmentidentified inourstudiesmayrepresentbrain ISF. Alternatively, the intermediatecompartmentcould

repre-senttheintracellularspaceofthe choroid plexus epithelium. As

notedabove, the choroid plexusexpressesinsulinreceptors at

high levels, and likelyconcentratescirculating insulin within the blood-CSF barrier. Having entered the intracellularspace

of the choroidal epithelium, insulin in thiscompartmentcould subsequently be released into CSF. However, the dynamics of CSF insulin uptakearequalitatively similar whethersampled

from lumbarsacinman(28)orcisternamagnaindogs (8, 10),

suggesting that the choroid is unlikelyto representthemajor

sourceofinsulin in CSF. Further, the teleologic rationalefora specializedtransportsystemforinsulinacrossthechoroid into

CSF is unclear, since this uptakeresultsinrelatively low CSF

insulin levels, andrecentstudies have establishedthatinfusion

ofinsulininto CSF isaninefficientmeansfor itsdelivery into the brain. For example, 100-1,000-foldgreaterCSFinfusion

ratesarerequiredtoachieveeffectsonfood intake comparable

tothose resulting from directintrahypothalamicinsulin

deliv-ery(29). For thesereasons, wefavor the hypothesis that brain

0.1

0.011

a _{estimated initialinsulin}conc.=0.83 uU/mI ~~~~CSFclearancerate:=

a

=0.020

min'I

(6%)

0 50 100 150 200 250

time(min)

Figure6. Studies in which the CSF clearance of intracerebroventricu-larly injectedinsulinwasdetermined. Insulinwasadministeredasa

1-mininfusionatt =0min. Eachpercentage inparentheses is the fractional standarddeviation associated with the identifiedparameter value. TheaverageCSF insulin clearanceratefor these studieswas

0.028±0.0080min-'.

ISFrepresentsthe primarycompartmentintowhich

circulat-ing insulinenterswithin the centralnervoussystem,with sub-sequentmovementinto CSF. Furtherstudiesare requiredto

confirm this possibility.

Insulinexertsavariety of actions within the CNS (see refer-ence23forreview), including behavioralresponses(i.e.,

sup-0.10. 0.08-

0.06-k4

(mind) 0.04-

0.02-o

A

AT

A~L a

i

+

from directly modeling measured

studies

Figure7. Range of

esti-matedk4,theCSF

insu-linclearancerate

con-stant,bytwomethods:

derived frommodeling CSF insulin changes in

responsetoan

intrave-nousinsulininfusion

(meanvalue 0.046±0.021 min'), anddirectlyestimated froman

intracerebro-ventricular insulin bolus injection (meanvalue 0.028±0.0080 min-').Eacherrorbarrepresentsthefractional

stan-darddeviation associated with themeanparametervalue.

1828 Bauraetal.

S

0.10.

0.08.

0.06.

0.04-

0.02-fi

1-.s

0

u

84

IL

0.12-vu

II

.-pression of food intake),regulation of neuropeptide

biosynthe-sis and catecholamine metabolism,andactivation of the

sym-pathetic nervous system. Collectively, these central actions promote a state of negative energy balance, favoring weight loss. These observations have led to the hypothesis that

circu-lating insulinacts in the brain as a negative feedback signal

proportionalto body adipose mass in the process of long-term

body weight regulation(30, 31). From the available data, we

believe thatplasmainsulin is deliveredto brain interstitial fluid at least inpart viaan insulin receptor-mediated transcytotic process. As mostinsulin-receptor mediatedactions are subject to avariety of regulatory influences, it is possiblethat this

insu-lin transport process may also be regulated, which would be

unlikely were transport dependent solely on nonsaturable

mechanisms such as diffusion. Theoretically, regulation of

CNS insulin transport could alter the efficiency with which circulating insulin gainsaccess to targetsites inthebrain,and

could thus represent an important determinant ofthe level

aboutwhich body weight is maintained.

Receptor-mediated insulin transport from vasculature to

tissue interstitium may regulate processes other than body

weight.The movementof insulin frombloodintothe

intersti-tiumof tissues isanimportant determinant of therateof insu-linactiononcarbohydrate utilization in vivo.Yang et al. (32)

have shownastrongcorrelationbetweenthedynamics of

tho-racic duct lymph insulin, which is derived from interstitial

fluid,

andinsulinmediated glucose utilization.Further, lymph

insulinis

proportional

toglucose utilization within the physio-logic but notpharmacologic range of insulin, indicatingthat

transcapillary insulin transport israte-limiting for insulin

ac-tioninthisrange(33). Thus,the movementof insulin from

vasculature tointerstitium isacriticalstepindetermining

tis-sue

sensitivity

toinsulin,andit is

possible

thatdefects in trans-port maycontributetothe

pathogenesis

of insulin resistance in conditions such as non-insulin-dependent diabetes mellitus

and

obesity.

Itshould bestressed, however, that whilewehave demonstrated saturable

uptake

of insulin into

CSF,

studies

have yet toestablish asaturable mechanism of insulin trans-portintoperipheral interstitial fluid. Itis

possible

that

mecha-nismsoperating within brain vasculature differ from those of othertissueswhereinsulin uptake is concerned.

It should be noted that the estimated

k1k2

values in the

supraphysiologic

rangeof

plasma

insulin levelsin the current

studiesareabouthalfthosecalculated

previously (10).

These differencesare

probably

attributabletomodified

plasma

and

CSF

sampling schedules,

whichhave

improved

parameter

esti-mates.

Specifically,

thenumber of

samples

taken

during

rapid

changes ofplasma and CSF insulin levelswasincreased. This change of

sampling

times reduced the

variability

ofestimated

parameters,

compared

tothose

generated using

conventional

sampling

procedures

(34).

In our

previous work,

estimated

parameters often had associated coefficients of variation

> 100%. Incontrast, themeancoefficient of variation for 30

identified parameters in thecurrent studies is 14%. In both cases, thetrueparameter value waslocated withinthe coeffi-cient ofvariation ofeachestimate.

However,

because the coeffi-cients of variationhave beenreducedin thecurrent

studies,

the

mean estimatesare nowmuch closerto thetrue parameters than in the

previous

work.

Inthe current

studies,

wemeasured

k4,

therateof insulin

clearancefromCSF,tobe0.046±0.021 min-'. Thisclearance

rate was

independently

verified by intracerebroventricular

ad-ministration of insulin, and is much higher than the rate of bulkflow CSFclearance in the dog(0.0025±0.0003 min '),as

determined from previous inulin studies (10, 15, 16). This large disparity between CSF clearance rates suggests that

mech-anisms other than bulk flow contribute to insulin removal from CSF. One possibility is that insulinreceptorsonthe cho-roid plexus bind and remove/degrade insulin in CSF, as sug-gested by Manin et al. (35). While this would imply a saturable

mechanism forCSFinsulin removal, saturability wouldnot be

demonstrableat such low CSF insulin levels as occurred in our

studies(< 10

_,U/ml).

Thus,while CSF insulin removal may involve a saturable mechanism, further study is required to address this possibility.

In summary,we havedemonstratedthat uptake of plasma

insulininto the central nervous system is saturable. This

mecha-nism is consistent with insulin binding to blood-brain barrier and/or blood-CSF insulin receptors and subsequent

transcyto-sis intothe central nervous system.

Appendix

To determine if model parameters are uniquely identifiable, thesystem impulseresponse, that is, the cause-effect

relation-shipbetweeninputandoutput, is analyzed. For this system, the

CSFimpulseresponseis:

INSc(t) = 1c- _c-

kw [e-w-

-e-

(k2+k3)t]I(5)

Fortheoretical, or a priori, identifiability, all the rate constants must be uniquely identified from Eq. 4. For the function

D(e-'

-end), D, a, and 3can be determined from a semilog plot of the natural log of this function versus time. Therefore,

k4can be identified, as can the quantity (k2 + k3). These three constants make up thedenominator of the terms multiplying theexponentials, so the numerator (klk2) can also be

identi-fied. Inorder to uniquely identify k, k2, and k3, more infor-mation isneeded (36). Therefore, this model is not theoreti-callyidentifiable.

Eqs. 1 and 2 can be rearranged to simplify parameter

identi-ficationbydividingbothsides ofEq. 1 by

_ki,

subtractingand

addingthe same term to Eq. 1, and rewriting Eq. 2 in terms of

INSI(t)/kl:

d

_INS,

(t)

=INS~()

INS-(t)

(k2+ k3- k~k2-Fk1k2)

(6)

dt

_k,

k

d

_INSc(t)

=

k2l-

k

INSI(t)

-_k4INSc(t)

dt ' _,

(7).

Theseequationsarerealizedin the rearrangedmodel of Fig. 8.

1

_ki

_k2k4

INSp(t) I _{(t) $INS (t)} k2+k3 -

_ki

k2

Figure 8. Schematic illustration of therearranged pharmacokinetic

modelused toanalyze the kinetics of insulin uptakefromplasma

intoCSF.

theCentral

System

1829

The parametersidentified with this model arek,k2, (k2 + k3

-klk2), andk4. Once klk2and(k2+k3-klk2)aredetermined,

(k2

+

_k3)

is alsoknown.

Acknowledgments

Wegratefully acknowledgecriticalassistancewithstudy design pro-vided by Dr.Alfredo Ruggieri. Expert technical assistance was pro-vided by Rix Kuester, Wendy Hamer, and Dave DeGroot in conduct-ing thesestudies.Insulinassayswereperformed byRichard Chan and HongNguyen,andglucose assays by RuthHollingworth.

This workwassupported by grants RRO2176, DK-12829, DK-17047, andDK-29867from the National Institutes ofHealth, byan Associate Investigator Award and Research Associate Award of the Dept.ofVeteransAffairs,andbyanNIHPhysicianScientist Award.

References

1.Woods, S.C.,D.Porte, Jr., E.Bobbioni,E.Ionescu,J.Sauter,F. Rohner-Jeanrenaud, and B. Jenrenaud. 1985.Insulin:itsrelationshiptothe central ner-voussystem andthe controloffood intake and body weight. Am.J. Clin.Nutr. 42:1063-1071.

2.Vanderweele, D. A., E.Haraczkiewicz,and T. B. VanItallie. 1982. Elevated

insulinandsatietyin obese and normalweightrats.Appetite.3:99-109. 3. Woods, S. C., E. C. Lotter, L. D. McKay, and D. Porte, Jr. 1979. Chronic intracerebroventricular infusion of insulin reduces food intake and body weight of baboons. Nature(Lond.).282:503-505.

4.Pardridge, W. M. 1987. Receptor-mediated transport through the

blood-brainbarrier.Endocr.Rev.7:314-330.

5.Bar, R. S., M. Boes, and A. Sandra. 1985. Vascular transport of insulin to

ratcardiac muscle. J. Clin. Invest. 81:1225-1233.

6. King,G. L., and S. M. Johnson. 1985. Receptor-mediated transport of insulin across endothelial cells. Science(Wash.DC). 277:1583-1586.

7. Frank, H. J. L., and W. M.Pardridge. 1983. Insulinbindingtobrain

microvessels. Adv.Metab. Disord. 10:291-302.

8.Schwartz,M.W.,A.J.Sipols,S. E.Kahn,D. F.Lattemann,G. J.Taborsky,

Jr., R. N.Bergman, and D. Porte, Jr. 1990. Kinetics and specificity of insulin uptakefromplasmainto cerebrospinal fluid.Am.J.Physiol.259 (Endocr. Me-tab.22):E378-E383.

9.Duffy,K.R., and W. M.Pardridge. 1987. Blood-brain barrier transcytosis

of insulinindeveloping rabbits.Brain Res. 420:32-38.

10.Schwartz, M. W., R. N. Bergman, S. E. Kahn, G. J. Taborsky, Jr., L. D. Fisher, A. J. Sipols, S. C. Woods, G. M. Steil, and D. Porte, Jr. 1991. Evidence for entryofplasmainsulin into cerebrospinalfluidthrough anintermediate com-partment indogs. J. Clin. Invest. 88:1272-1281.

11. Morgan, C.R., and A. Lazarow. 1963. Immunoassay of insulin: two

antibodysystem.Diabetes. 12:115-126.

12.Baskin, D. G., S. C. Woods, D. B. West, M. van Houten, B.I.Posner, D. M. Dorsa, andD.Porte, Jr.1983. Immunocytochemical detection of insulin inrathypothalamus and its possible uptake from cerebrospinal fluid.

Endocrinol-ogy. 113:1818-1825.

13. Berman, M., and M. F. Weiss. 1978. The SAAM Manual. DHEW Publi-cation No. (NIH) 78-180. U.S. Government Printing Office, Washington, DC.

14.Gabel,R. A., and R. A. Roberts. 1980. Signals and Linear Systems. 2nd edition. JohnWiley&Sons, Inc., New York.

15.Reed, D. J., and D. M. Woodbury. 1963. Kinetics ofmovement of iodide,

sucrose, inulin and radio-iodinatedserumalbumin in the centralnervoussystem andcerebrospinalfluid of therat.J.Physiol. (Lond.).169:816-850.

16.Davson,H.1970.Physiologyof theCerebrospinalFluid.Churchill, Lon-don.

17.Gilman,A.G.,T. W.Rall,A. S.Nies,and P. Taylor,editors. 1990. Goodman and Gilman's ThePharmacologicalBasis ofTherapeutics.8thedition.

PergamonPress,New York. 21.

18.Tallarida,R. J., and L. S. Jacob. 1979. The Dose-Response Relation in

Pharmacology.Springer-Verlag, Inc.,New York. 50.

19. Frank, H. J. L., and W. M.Pardridge.1981.Adirect in vitro demonstra-tion of insulinbindingtoisolated brain microvessels. Diabetes. 30:757-761.

20. Haugaul, F., and V. Desmaizieres. 1986. The effect of insulin onglucose uptakeandmetabolism in the humanplacenta. J. Clin.Endocrinol. Metab. 62:803-807.

21.Keller, J. M., and J. S. Krohmer. 1968. Insulin transfer in the isolated humanplacenta.Obstet. Gynecol.32:77-80.

22. Yalow, R.S.,and J.Eng. 1983. Insulin in the central nervous system. Adv.

Metab.Disord. 10:341-354.

23.Schwartz,M.W.,D. P.Figlewicz,D.G.Baskin,S. C.Woods,and D. Porte, Jr. 1992. Insulin in the brain: a hormonal regulator of energy balance.

Endocr.Rev. 13:387-414.

24. Crone, C. 1965. Facilitated transfer of glucose from blood into brain tissue. J.Physiol.(Lond.). 181:103-1 13.

25.Cutler,R. W. P.1980.Neurochemical aspects of blood-brain

cerebrospi-nalfluid barriers. In Neurobiology of Cerebrospinal Fluid. J. H. Wood, editor. PlenumPress, New York. p. 43.

26.Fishman,J.B., J. B.Rubin, J. V. Handrahan, J. R. Connor, and R. E. Fine.1987. Receptor-mediated transcytosis of transferrin across the blood-brain

barrier.J. Neurosci. Res.18:299-304.

27.Zlokovic, B. V., M. N.Lipovac,D. J.Begley, H. Davson, and L. Rakie. 1987. Transport of leucine-enkephalin across the blood-brain barrier in the

per-fusedguinea pigbrain. J. Neurochem. 49:310-315.

28. Wallum, B. J., G. J. Taborsky, Jr., D. Porte, Jr., D. P. Figlewicz, L. Jacobson, J. C. Beard, W. K. Ward, and D. Dorsa. 1987. Cerebrospinal fluid insulin levels increase during intravenous insulin infusions in man. J. Clin. Endo-crinol. Metab. 64:190-194.

29. McGowan, M. K., K. M. Andrews, J. Kelly, and S. P. Grossman. 1990. Effects of chronic intrahypothalamic infusion of insulin on food intake and diur-nal mealpatterning in the rat. Behav. Neurosci. 104:373-385.

30.Woods,S.C.,and D.Porte, Jr. 1976. Insulin and the set-point regulation of body weight. In Hunger: Basic Mechanisms and Clinical Implications. D. Novin,G. A. Bray, and W. Wyrwichka, editors. Raven Press, New York. 273-280.

31. Woods, S. C., D. P. Figlewicz Lattemann, M. W. Schwartz, and D. Porte, Jr. 1990. A re-assessment of the regulation of adiposity and appetite by the brain insulin system. Int. J. Obesity. 14 (Suppl. 3):1063-1071.

32.Yang, Y. J., I. D. Hope, M. Ader, and R. N. Bergman. 1989. Insulin transport across capillaries is rate limiting for insulin action in dogs. J. Clin. Invest.84:1620-28.

33. Ader, M., R. A. Poulin, Y. J. Yang, and R. N.Bergman. 1992. Dose-re-sponse relationship between lymph insulin and glucose uptake reveals enhanced insulin sensitivity of peripheral tissues. Diabetes. 41:241-53.

34. D'Argenio, D. Z. 1981. Optimal sampling times for pharmacokinetic experiments. J. Pharmacokin. Biopharm. 9:739-756.

35.Manin, M., Y. Broer, M. Balage, W. Rostene, and J. Grizard. 1990. Meta-bolic clearance of insulin from the cerebrospinal fluid in the anesthetized rat. Peptides. 11:5-12.

36.Carson, E. R., C. Cobelli, and L. Finkelstein. 1983. The Mathematical Modeling of Metabolic and Endocrine Systems. John Wiley & Sons, Inc., New York.