Glucose transport and metabolism in adipocytes from newly diagnosed untreated insulin dependent diabetics Severely impaired basal and postinsulin binding activities

Glucose transport and metabolism in

adipocytes from newly diagnosed untreated

insulin-dependent diabetics. Severely impaired

basal and postinsulin binding activities.

E Hjøllund, … , H Beck-Nielsen, N S Sørensen

J Clin Invest.

1985;

76(6)

:2091-2096.

https://doi.org/10.1172/JCI112213

.

Previous studies have shown cellular insulin resistance in conventionally treated

dependent diabetics. To determine whether insulin resistance is also present in

insulin-dependent diabetics before the commencement of insulin therapy, we studied nine newly

diagnosed untreated insulin-dependent diabetics and nine control subjects. Insulin binding

to adipocytes, monocytes, and erythrocytes was normal in the diabetic individuals. Basal

(noninsulin stimulated) glucose transport rate was normal, whereas the maximal insulin

responsiveness of glucose transport was severely impaired (P less than 0.02). Insulin

sensitivity as judged by left or rightward shifts in the insulin dose-response curves was

unchanged. Moreover, the basal lipogenesis rate measured at a glucose concentration of

0.5 mmol/liter was decreased in the diabetics (P less than 0.05), and the maximal insulin

responsiveness of lipogenesis was also reduced (P less than 0.05). We conclude that fat

cells from untreated insulin-deficient diabetics are insulin resistant. The major defects are

(1) reduced maximal insulin responsiveness of glucose transport and conversion to lipids

that are postbinding abnormalities, and (2) reduced basal glucose conversion to lipids.

Research Article

Find the latest version:

Glucose Transport and

Metabolism in

Adipocytes

from

Newly

Diagnosed

Untreated Insulin-dependent

Diabetics

Severely Impaired Basal and Postinsulin Binding Activities

Elisabeth Hjollund, OlufPedersen, Bjorn Richelsen, Henning Beck-Nielsen, and Niels Schwartz Sorensen

DivisionofEndocrinology and Metabolism,DepartmentofInternal Medicine and Clinical Chemistry, Aarhus Amtssygehus, Denmark

Abstract

Previous studies have shown cellular insulin resistance in

con-ventionally treated insulin-dependent diabetics. To determine whetherinsulin resistance is also present ininsulin-dependent diabetics before the commencementof insulin therapy, we studied

ninenewlydiagnoseduntreated insulin-dependentdiabetics and

nine control subjects. Insulin bindingto adipocytes, monocytes,

and erythrocytes wasnormal inthe diabetic individuals. Basal

(noninsulin stimulated) glucose transport rate was normal, whereas themaximal insulin responsiveness ofglucose transport wasseverely impaired (P <0.02). Insulin sensitivityasjudged

by leftorrightward shifts in the insulin dose-response curves wasunchanged. Moreover, the basallipogenesis rate measured at a glucose concentration of 0.5 mmol/liter was decreased in thediabetics (P<0.05), and the maximal insulin responsiveness of

lipogenesis

was

also

reduced (P<

0.05).

We conclude thatfat cells from untreated insulin-deficient diabeticsareinsulin resistant. The major defectsare(1) reduced maximal insulinresponsiveness ofglucose transport and

con-versiontolipidsthat are postbindingabnormalities,and(2) re-duced basalglucose conversionto lipids.

Introduction

Insulin-dependent diabetes mellitus is caused primarily by the lack

of

orseverely reduced

production

of insulin. However, re-centinvivo

studies

haveshown that

insulin-dependent

diabetic

subjects display

an

impaired

insulin effect on the

disposal

of glucose into

peripheral

tissues(1-7)aswellasdecreased

inhib-itory

effect of insulin

on

hepatic

glucose release (1,

5,

6).Invitro studies have shown reduced insulin receptor

binding

in

adipo-cytes from

conventionally

treated insulin-dependent diabetics

(8, 9).

In agreement with the

impaired

insulin

binding,

dose-response curvesfor the

insulin effect

on

glucose

transport

(8)

and

antilipolysis (8, 9)

wereshifted to the

right. However,

no

changes in

insulin

sensitivity

wereobservedon

glucose

oxidation andglucose conversionto

lipids

(8). Impaired

basal and maximal

insulin action on

adipocyte glucose metabolism

was foundas

well (8, 9). Ithas been suggested that the insulin resistance

of

insulin-treated

diabetic

patients might

bearesult of the insulin

treatment

itself,

and in

particular

thewayof administration of

insulin,

namely subcutaneous

injection,

which causes

high

pe-Addresscorrespondenceto Dr.Hjollund,Medical Department III,Aarhus

Amtssygehus,Tage Hansens Gade2,DK-8000 AarhusC, Denmark. Receivedfor publication 12February1985.

ripheral levelsof insulin asopposed to endogenously released

insulin (8,9).Alternatively, the insulin resistance might be due tothe

metabolic

decompensation ofthese patients.

The aim of the presentstudy was to determine whether a stateof cellular insulin resistance also exists innewlydiscovered, untreated patients with insulin-dependent diabetes. Hence, we

haveexamined adipocyte insulin binding and action in young untreated

insulin-dependent

diabeticsandina comparable group

of controls. Moreover, wehavealso measured insulin binding

toerythrocytes and monocytes in orderto make comparative studiesoninsulin bindingtodifferent celltypes.

Methods

Subjects. Nine insulin-dependent diabeticswerestudied withinthefirst

2dof admission withnewlydiagnosed diabetes mellitus beforethe com-mencement of dietandinsulintherapy. From the dayof admissionto theday ofstudy,thepatientsatethe normalhospital diet for nondiabetic subjects (8,500±1,000

Id,

with 40% carbohydrate, 41% fat, and 19%

protein).Allpatientswereinformedaboutthe nature andpossible risks of the studyaccordingtoHelsinki declaration II and written consent wasobtained.Patients in precoma or otherwise needing acute treatment andpatients withtemperature above37.50Candpatientsin whom di-abetesmellituswassecondary toother diseaseswerenotincludedin the

study. During admission (after 1to5d), aglucagontest wasperformed,

which ensured that the patients did have insulin-dependent diabetes mellitus (10) (TableI). Thepatientswerestudied in the morning after

anovernight fast.Afatbiopsywas takeninlocalanesthesiaaspreviously described(11), andabloodsamplewasdrawn for estimation of insulin

receptors onerythrocytes andmonocytes aswellasestimationof plasma hormonesandmetabolites.

Nineage- andsex-matched controlswerestudied inthe same way. 3 dbeforethefatbiopsythe volunteers madedietrecords.Theirdaily food intakewas8,900±1,500

Id,

with 39%carbohydrate, 43% fat,and

18%protein. The clinical and biochemicaldataofpatientsandnormal

controls aregiven inTable I.

Chemicals. Human albumin was obtained from Behring Werke,

Marburg, FederalRepublic of Germany. Collagenase fromclostridium

histolyticum, 213 U/mg,wasobtained fromWorthingtonBiochemical Corp.(Freehold, NJ).

'25I-Monoiodoinsulin

withthelabeled iodine in

tyrosine A14(specific activity,-250

,Ci/lg)

wasgenerouslydonatedby

NovoResearchInstitute, Copenhagen, Denmark

(12).

D-U-[

'4C]glucose

(specific activity,333mCi/mmol)wasobtained fromtheRadiochemical Centre, Amersham, England. Tissue andcellsweresuspendedinHepes

buffer

(100

mmol/liter)in thestudies ofmonocyte(13)anderythrocyte binding

(14)

and10mmol/literin thestudiesofadipocyteinsulinbinding

andaction(11, 15).ThepHwasadjustedto7.4at37°Cin thestudies

onthefatcells andmonocytes,andtopH7.8at37°Cin thestudieson theerythrocytes.

Insulinreceptorbinding studies. Adiposetissue(- 10g)wasobtained byopenbiopsyfrom the upper quarterofthe right gluteal

region

after

asquarefieldhad beenanesthetized withanepidermal

injection

of1%

lidocaine withoutepinephrine.Detailsaboutfatcellisolationaswellas

determination of fatcellsizeand number have beenpublished

previously

(I 1, 15). Insulinbindingtofatcells(_ _{105cells/mlofcell}

_suspension)

J. Clin. Invest.

©TheAmericanSociety for ClinicalInvestigation,Inc.

was measured ina Hepesbufferat370C after incubation for 60 min

with

_{tyrosine-A14-labeled}

'25I-insulinwithorwithoutincreasing

concen-trations of unlabeled insulin. In orderto compareinsulin bindingto

fatcells andblood cells at the sametemperature (insulin bindingto bloodcellsmustbemeasuredatsubphysiologictemperaturetoensure

steadystatespecific binding),wealso measured insulinbindingattracer

concentrationstofat cellsat 15C witha 120-min incubation period. Cell-associatedradioactivity inthe presenceof10umol/literunlabeled

insulin(nonspecific binding) averaged4%of totalbindingbothat370 and 15'C.Specific insulin bindingtoadipocyteswasexpressedper30

cm2of surfaceareapermilliliter. Insulindegradationin the mediumat

370Cwas<4%, measuredasTCAsolubilityafter 60 min.

Insulinreceptorbindingtoerythrocyteswasdeterminedasdescribed

(14)with thefollowing modification. Afterfractionatingthe bloodonce

on aFicoll-Isopaquegradient, theerythrocyteswerecollected from the bottom ofthetubes. Thecellswereresuspended 1:1 in 0.9% NaCl

con-taining 50 mg/mldextran 500T. The tubesweretilted 450 from the

verticalfor 15 min at370C.Theerythrocytesthen settledand the su-pernatantcontaining the granulocyteswasremoved.Inthisway,

gran-ulocyte contaminationwasreducedto<0.03/ 1,000erythrocytes. After washing,theerythrocytes(at avolumefraction of 0.45)wereincubated for210 min at15'C in 100mmol/liter Hepes buffer with

tyrosine-A,4-labeled'25I-insulinwithorwithout nativeinsulin (10gmol/liter)as de-scribed ( 14).Nonspecific binding averaged 12%of totalbinding.Specific

insulinbindingwasexpressedper5X109cells/ml.Puremonocyteswere

isolated fromthemononuclear cell layer obtained byfractionating the

blood on aFicoll-Isopaque gradient, basedon thefact thatmonocytes but notlymphocytes adheretoplastic surfacesat37°C and detachagain

in the cold(16).Inthisway, veryhomogeneoussuspensions ofmonocytes wereobtained(the percentage ofmonocytes was97+2,mean+I SD). Monocytes wereidentifiedbymorphologicalandcytochemicalcriteria (13,16).Calculation of insulin bindingto monocytes wasperformedas

previously described (13). Monocytes (2.5-8 X 106cells/ml) were in-cubatedfor 120min in 100mmol/liter Hepes bufferat 15°C with

ty-rosine-A,4-labeled '25I-insulinwithorwithoutnative insulin (10Mmol/

liter). Nonspecific bindingwas22%of totalbinding. Specific binding

wasexpressed per 5 X 106puremonocyte permilliliter.

Lipogenesis. Lipogenesis was measured by studies of the conversion of the

D-U-['4C]glucose

to

'4C

totallipidsasdescribed earlier (1 1). Isolated adipocyteswere prepared ina 10 mmol/liter Hepesbuffer containing 0.5mmol/liter glucose (volume fraction0.05).Thecellswere preincu-batedfor45minat37°C withorwithout insulin inincreasing

concen-trations. Then, 0.4

ACi

D-U-['4C]glucose

wasaddedtoeach tube(final

glucoseconcentration, 0.5mmol/liter)and theincubationwascontinued for90 min.Then, H2SO4wasadded,andaDoleextractionwasperformed

and asample forliquid scintillation countingwastakenfromthe upper phase(11).

'4C-radioactivity

was presentinan average amountof 21±6%

of noninsulin-stimulated lipogenesis when incubationswereperformed intheabsenceof fat cells (blankvalues). All valuesfor fatcell-produced totallipidswerecorrectedfor theindividual blank value.

Studiesofglucosetransport. Glucosetransport was measured as the conversion of

D-U-['4C]glucose

to totallipidsattracer glucose

concen-trations (5

tsmol/liter).

It hasbeen shown that glucosetransport isthe

rate-limitingstepforglucoseprocessingatvery low glucose concentrations (I 1, 17, 18).'Under theseconditions, >80% of the glucose is converted tolipids (17, 18).' Therefore, measurements of the conversion rate of D-glucose to total lipids at tracer glucose concentrations will yield an

indirect estimationof glucose transport rates(17, 18).'Glucose transport wasmeasured as described for lipogenesis with the following

modifica-tions: Thecells were preincubated in a glucose-free buffer. Then, 0.4MCi

DU-[U4Cjglucose

(finalconcentration,5,umol/liter)was added. The

in-cubationwasstoppedafter90min by theaddition of H2SO4.After this

aDoleextraction wasperformed. Blankvalues averaged 8±3% of the

noninsulin-stimulated lipogenesis under these conditions.

Analytic procedures. Plasma glucosewas analyzed with a glucose

1.

_Hjillund,

E., and0.Pedersen,submitted for publication.

dehydrogenasemethod(Merck enzymatic kit, Mannheim,Federal

Re-publicofGermany).Seruminsulin was measuredby radioimmunoassay (19). Plasmaacetoacetate andplasma 3-hydroxybutyrate (ketone bodies)

were measured separately byenzymatic micromethod(20).Inthe Results,

the sum isgivenof the substance concentrations of the two metabolites. Plasma FFA(aliphatic carboxylateC8toC18,nonesterified)wereassayed accordingtothe methodofItayaand Ui(21). Serum growthhormone wasestimated as describedby Orskovet al.(22). The cortisol concen-tration in urinewasdetermined accordingtoMurphy (23). C-peptide inserum was assayed by themethodof Heding (24).

Statistical methods.In thetextandtables, dataaregivenasmean±1 SD while the data inthefiguresrepresentthemean± ISEM.Anunpaired

ttestwasused forcomparison betweengroups.Linear regression analysis

was employedin correlation studies using the least-squares method. Ke-tonebodieswerelog distributedandhence were logtransformed before

statisticalanalysis.

Results

The biochemical data of the patientsand normal controls are

giveninTable I.Allpatients were nonobese. No significant dif-ferenceinfat cell diameterwasfound. Thefastingseruminsulin levels ofdiabeticswereonlyslightly decreased, whereas the basal

aswell as the glucagon-induced rise in serum C-peptide con-centration was reduced inall patients, indicating insulin defi-ciency (10).Asexpected, the patientshadsignificantlyelevated fasting plasma concentration of ketone bodies (P<0.001). However, noneof thepatients was acidotic. Fasting plasma

con-centrations of FFA weresignificantlyincreased in thediabetics (P<0.05). Two patients had elevated urinary cortisol excretion (>385 mmol/liier), whereas onlyonehadelevated fasting plasma growth hormone concentration (>6

_Ag/ml).

Table I. Clinical andBiochemicalData inHealthy Controls

andInsulin-dependent Diabetics (Mean±J SD)

Normals Diabetics

Sex 4956 495d

Age(yr) 39±9 38±9

Bodyweight(kg) 69±14 65±16

Fasting plasma glucose(mmol/

liter) 5.3±0.7 12.8±1.3

Fastingserum

insulin(gU/ml) 14±4 10±5

C-peptidebasal and

stimulated* 0.25±0.08

(nmol/liter) 0.42±0.14

Fasting plasma ketonebodies

(mmol/liter) 0.10±0.08 2.08±1.81i

Fasting plasmaFFA

(mmol/liter) 0.19±0.10 0.39±0.06§

Fatcelldiameter

(am) 93±15 87±12

* 6

rmin

after intravenous injection of1 mgglucagon.Alldiabetics

hadstimulated C-peptide concentrationsbelow0.6 nmol/liter (10), thusrenderingthepatients insulin requiring.

tP<0.001.

Fig. 1 depicts competitioncurvesforinsulin binding to fat

cellsmeasuredat37°C. When expressed in relationtocellsurface area concentration, no significant difference between insulin binding to fat cells fromnormals and newly diagnosed

insulin-dependent diabetics was observed. The same was the casewhen

binding was expressed in relation to cell numberconcentration.

Fat cellinsulin binding measured at insulin tracer concentration at 15°C as well as insulin binding to erythrocytes and pure

monocytes also measured at 15°C was similar in normals and diabetics (Table II).

When adipocyte insulin binding was expressed in relation tocellsurface area, no correlations between insulin binding to monocytes and adipocytes or monocytes and erythrocyteswere

found

(Fig.

2). However, a significant inverse correlation existed betweenerythrocytes and adipocytes (r=-0.70,P <0.05, Fig. 2). When insulin binding was expressed in relation to cell number concentration, the negative correlation between erythrocytes and adipocytes still existed (r=-0.58, 0.1 >P >0.05). Onthe other

hand, no correlations between individual values of cell insulin

binding

were found when data obtained in the

37°C

studies of fat cell insulin binding were used (data not shown). No

corre-lations between insulin binding to fatorblood cells and fasting plasmaconcentrations of insulin, ketone bodies, or FFA were

found.

Thebasal (noninsulin-stimulated) glucose transport rate into fat cells was identical in diabetics and normal controls

(Fig.

3). However, the insulin response above basal levels wasseverely reduced in fat cells from the diabetic patients as compared with normals(41±25% vs. 101±69%, P < 0.02). No difference in insulin

sensitivity

asexpressed as leftor

rightward

shifts in insulin dose-responsecurves wasobserved(ED50was58±25pmol/liter in normals and 65±30

pmol/liter

in the diabetic

patients).

In

the diabetic

patients,

no correlation was found between per-centage insulin response above the basallevel ofglucose transport

and fasting insulin levels (r= 0.05, NS), whereas plasmaFFA

concentrations

correlated

negatively

to percentage insulin

re-sponse(r =

-0.72,

P<

0.05).

Nocorrelations between

degree

of ketosis and glucose transport rates were found. In normal

controls,no

correlations

betweeninsulinemiaorketonemiaor

plasmaFFA levelsandglucose transportrates were observed.

3-0 10l _l5

Insulin(pmol/1)

Figure1.Insulinbindingtoadipocytesfromninenewly

diagnosed

in-sulin-dependent diabeticpatients (o)andninecontrol

subjects (-).

Adipocyteswereincubated with 15pmol/liter

'251I-insulin

at37°Cfor 1 hin the absenceorpresenceofunlabeledinsulininincreasing

con-centrations(mean± 1 SEM).

Table II. Insulin Bindingto Adipocytes, Monocytes,

and Erythrocytes at150C in Healthy

Controls andInsulin-dependent Diabetics*

Normals Diabetics

Adipocytes

(30cm2/ml) 4.87±1.70 4.59±1.07

Monocytes

(5- 106/ml) 2.66±0.86 3.09±1.10

Erythrocytes

(5*109/ml) 5.69±1.08 5.58±1.70

* Insulintracerconcentration, 15 pmol/liter. Specific insulin bound

fractions times 102(mean±1 SD).

Glucose conversion to total lipids measured at a glucose concentration of 0.5 mmol/liter, where steps distal to glucose transport are rate limiting, is depicted in Fig. 4. A significant decrease in noninsulin-stimulated lipogenesis was found in fat cellsfrom the diabetic patients (P<0.05). Moreover, the insulin response above basal levels was very low (35±23% vs. 72±41% in normals, P < 0.05). Actually, in two patients, no insulin re-sponse was detectable. The absolute lipogenesis rate in maximally insulin-stimulated cells was thus markedly reduced (to 40% of that in normals,Fig. 4). Due to the low insulin responses, changes ininsulin sensitivity could not be estimated in these studies. No correlations between basal or maximally insulin-stimulated li-pogenesis rates and fasting concentrations of insulin or ketone bodyor FFAwerefound in any of the groups.

Discussion

Normal insulin

binding

in patients with untreated

insulin-de-pendent diabetes

mellitus. The present study demonstrates that insulin binding to fat cells from insulin-requiring diabetics before theinstitution of insulin therapy is normal. The unaltered insulin

binding

inthis study as opposed to the reduced binding in insulin-treateddiabetics (8, 9) supports the hypothesis that the reduced insulin binding in the latter studies is an acquired defect, probably secondary tosubcutaneous insulin therapy, which causes high

peripheral

levels of insulin (8). The insulin binding to erythro-cytes and monoerythro-cyteswasalso unaltered in the untreated, insulin-dependent diabetic patients corroborating previous studies (25). Monocytes and erythrocytes have been widely usedastools in clinical studies of insulin receptors, and receptor data from

A

6'

_A

: B

7s,21 ,.,*8-, *

2

2 4 6 2 4 6 2 4 6

Specific adipocyte Specific adipocyte Specific monocyte bound fraction -102 bound fraction-102 boundfraction-102 Figure 2.Interrelationshipsbetweeninsulinbindingtoadipocytes,

monocytes, anderythrocytesfromnine newly diagnosed insulin-de-pendent diabetic patients. Cellswereincubated with 15pmol/liter

ra-dioactive insulinat 15'C. Insulinbindingtoadipocytes,monocytes, anderythrocyteswasexpressedto30cm2surfacearea/ml,5X 106

monocytes/ml,and 5X 109erythrocytes/ml,respectively.

x

T\

_\6

?"

_It

_\\

,,6

t

i

Z.,e

1("11

..9 _,-4 .-A

2-Ew

°E

E

ok --Ff

CC

L~

-"I 0s

> 10

0 10

Insulin(pmol/lI

Figure

3. Glucose transportratesin

_adipocytes

from nine diabetic pa-tients

_(o)

andnine control

_subjects

_(.).

Transport

wasmeasuredas

li-pogenesis

at tracer

glucose

concentration

(5

umol/liter). Adipocytes

were

preincubated

ina

_glucose-free

_Hepes

bufferat

370C

withoutor

with insulin intheindicated concentrations for45 min.Thenlabeled

glucose

was

_added,

andtheincubationwascontinued for 90min

(mean±l

SEM).

blood cells have been

_extrapolated

totarget cells for

insulin.

The

validity

of this

extrapolation

has been

said

tobe

supported

by

severalstudies

_(26,

_{27). However,}

Solletal.

(26) merely

dem-onstrated that

_lymphocytes

and

adipocytes

from obese mice showed lower insulin

_binding

values than did

_lymphocytes

and

adipocytes

from

normalmice.

_Olefsky

etal.

(27)

found

decreased

mean

_binding

tomonocytes and

adipocytes

from

obese

subjects.

However,

a

_positive

correlation was

found

_only

after

pooling

data from obese and normal

subjects.

In

_previous

studies

of normal

_subjects

_(I

₁₎

or

long-term

insulin-treated diabetics

(8)

we

found

no

consistent correlations

betweeninsulin

binding

to

erythrocytes,

monocytes, and

adipocytes.

Taylor

etal.

(28)

found no

_relationship

_{between monocyte and}

_adipocyte

_insulin

_binding

ina

_variety

of clinical

situations

including

normal

subjects,

in-sulin-treated

-

diabetics,

and

patients

with cirrhosis

of the

liver.

=- To

WI= j

:° Q C

-a An

Eo

C)

4-3-~~~~~~~~~~~~~~-,

2- ; ; i

1]

'96e-Ifr-9---9

-9

IL'

10 102 0

Insulin(pmol/l)

Figure

4.

_Lipogenesis

in

_adipocytes

from nine diabetic

patients (o)

and

ninecontrolsubjects

_(-).

_Lipogenesis

wasmeasuredasdescribed in the

legend

to

_Fig.

3ina

Hepes

buffer

_containing

0.5

mmol/liter

glu-cose

(mean±

₁

_SEM).

Recently,

in a

study

of

newly diagnosed noninsulin-dependent

diabetics,

wefailedtofind

significant

correlations between blood cells and fat cells

(29).

In the present

study,

monocyte

and adi-pocyte

binding

didnot

correlate,

but

erythrocyte

and

adipocyte

insulin

binding

was

_{significantly negatively}

correlated. Taken

together,

this indicates that the

extrapolation

ofindividual

in-sulin-binding

values from blood cells to fat cells should be avoided.

Impaired

maximal insulin

responsiveness ofglucose

transport

and metabolism in

_{adipocytesfrom}

untreateddiabetics. The most

striking

defect in fat cells from

insulin-dependent

diabetics before thestartof insulintreatmentisa

severely

blunted orabolished

responsiveness

toinsulin of

glucose

transportand metabolism. Theseare

postbinding phenomena, probably

duetothe

depletion

of

glucose

carriers

(30-32)

andalack ofenzyme

responsiveness

inthe

glycolytic pathways.

Since such defects are reversed

by

insulintreatment

(8),

theseare

acquired

defects.

They

may

rep-resent

adaptive changes

in the cells to the catabolic situation withinsulin

deficiency.

Thesimilaritiestothe results in

fasting

human subjects

(33),

in whom the same reduced insulin

re-sponsiveness

was

found,

should be

emphasized.

Comparisons

toin vivo studies

of

insulin action in

patients

with

newly diagnosed insulin-dependent diabetes

mellitus.

Adi-posetissueis

only responsible

foraminor fraction of total

body

glucose

uptake

(34).

Therefore,

ourresults should be

compared

with appropriate in vivo studies. Recently,twogroupshave

in-vestigated

in vivo insulin action in newly diagnosed diabetics

using

the

euglycemic clamp technique.

Nankervis et al.

(35)

found that nontreated

insulin-requiring

diabeticswere charac-terized

by

elevated basal

hepatic glucose production

and

hepatic

insensitivity

to

insulin,

aswell as markedly

impaired

glucose

disposal

toperipheral tissues. Yki-Jarvinenetal. (36), who stud-ied newly diagnosed diabetics after 2 wk of insulin therapy, found similar defects.

Thus,

ourfindings of decreased maximal insulin responsiveness in fat cellsarecompatible with these reports.

Possible

factors responsible for

the

insulin

resistance in pa-tients with untreated

insulin-dependent diabetes

mellitus. Ke-toacidosis isawell knowncause of insulin resistance

clinically

aswell as

experimentally (37, 38). Although

our

patients

had

significantly higher

levels of ketone bodies than did

normals,

the absolute level of ketonemiawasnothigh, and noneofour

patients was acidotic. No relationship between the levels of plasma ketones and insulin bindingoractionwasobserved.

Plasma counterregulatory hormoneswerealsomeasuredin ordertoevaluate their role in theeventofinsulin resistance

(39-41). However, only two ofour patients had elevated urinary cortisol excretions, andonehadelevated fasting plasma growth hormone concentrations. These patientswerecomparabletothe

restof thegroupinall aspects.Therefore, itseemsunlikely that

theinsulin unresponsiveness found in fatcellsfromour

patients

is caused primarily by these hormones. Catecholaminesarealso abletoinduce insulin resistance in vivo inman (42) and invitro inisolated adipocytes(43).Actually,wedidnotmeasureserum

catecholamines inourdiabetics, but previous studies(44)have shown that theserumlevels of catecholamines parallel thedegree

of metabolic derangement. Sincenoneofourpatientswas

aci-dotic orseverelyketotic, it is unlikely that catecholamineswere

the majorcause of insulin resistance in the diabetics for the

presentstudy.

Noninsulin-stimulated

(basal) glucose transport and metab-olism in fat

cells

from

patients

with

insulin-dependent

diabetes

mellitus.

Nochange in basal glucose transport rate in adipocytes from ourdiabetic patients was found, whereas basal glucose

me-tabolism was severely depressed in the samecells. The

interpre-tation of the depressed basal lipogenesis raterelative to the

nor-mal basal transport rate in fat cells from the untreated diabetics depends on whether transportor posttransport steps are regarded as being rate limiting for glucose metabolism. In our hands, transport is not rate limiting for basal glucose metabolism even in adipocytes from normal subjects (1

1).'

This is discussed in detail

elsewhere.'

Shortly, it is based on the following observa-tions: When medium glucose concentration is increased from tracer only (5

gmol/liter

where transport is rate limiting [11, 18]) to 0.5 mmol/liter by addition of unlabeled glucose, the cel-lularuptake of radioactive glucose decreases to about one-third, whereas the transport rate is only slightly decreased

(K.

for

glu-cose transport is 7 mmol/liter). The same decrease in cellular uptake is found if glucose is replaced by 2-deoxyglucose (which is phosphorylated but not further metabolized) at the same con-centration, and/or 2-deoxyglucose is used as tracer. Thus, the decreasein cellular hexose uptake must be caused by competition

atthelevel

of phosphorylation

and notby escape of transported glucose converted to metabolic products. Moreover, total lipids comprise >80% of the intracellular radioactivityattracer

con-centration (1 1, 18)aswell asat aglucose concentration of 0.5

mmol/liter

(1 1)1 in human fat cells. The decreased basal

lipo-genesis

ratein fat cells fromuntreated diabetics at thisglucose

concentration

is therefore causedby atrue decrease in

metab-olismrateandnotbyashiftinrate-limiting stepsfromtransport

toposttransport steps.

Although glucose conversion to total lipids is the most im-portant metabolic pathway

for

glucose at this glucose

concen-tration,

other pathways such as lactate production may become ofrelativemore

importance

at

higher

glucose concentra-tions (18).

The depressed basal rate of lipogenesis is notaninsulin

re-ceptoror apostreceptorphenomenon, since it is presentincells

not acutely exposed to insulin. Yet, the

possibility

exists that the

impaired

basal metabolic

activity

isareflection of the insulin-deficient state,

especially

the metabolic deterioration and

cata-bolic

changes that follow insulin

deficiency.

This

interpretation

agrees with the

findings

in

insulin-deficient

streptozotocin-treated

rats(45). Recently, however, itwasshown thatacute

hyperin-sulinism in normal

subjects,

induced

by

6hofinsulin

infusion,

also leadsto an

impaired

basal

lipogenesis

rate in

adipocytes

(46).

Also,

after

chronic

insulin

therapy

for several years, the basal

glucose metabolism

offat cells from

conventionally

treated diabetics is severely

depressed (8).

Ithas been

hypothesized

that

this

depression

may be induced

by

the chronic

peripheral

hy-perinsulinemia

(8).

This idea is

supported

by

studies of the

long-term

effects of continuous insulin

infusion

therapy

on

adipocyte

metabolism in insulin-dependent diabetics where a

significant

aggravation

of basal

glucose

conversionto

lipids

has beenfound (47). Thus,

experimental

evidenceexists thatsuggeststhat both

hypo- and

hyperinsulinemia

may induce

impaired

basal

glucose

metabolism inhuman

adipocytes.

Weconcludethatinfat cells from untreated

insulin-depen-dent

diabetic

subjects,

insulin

binding

is

normal,

whereas the insulin

responsiveness

is

markedly

reduced at both

glucose

transportand

glucose

metabolism level. Thebasal

glucose

me-tabolism rate isalso reduced. The observed abnormalitiesare

most probably caused by insulin deficiency and accompanying

catabolic deterioration.

Acknowledgments

Theauthors thank Pernille Sonne, ToveSkrumsager, Lisbeth Blak, and JytteSoholt for their expert technical assistance, and Conni Mohl for skillfulpreparation of the manuscript. We are indebted to Novo Research Institute, Copenhagen, Denmark, for its generous donation ofA,4-labeled

'I-insulin.

Thefollowingfoundationssupported this study: the Danish Medical Research Council, Landsforeningen for Sukkersyges Fond, Kong Chris-tian X's Fond, Aarhus UniversitetsForskningsfond,and Nordisk Insulin Fond.

References

1. DeFronzo, R. A., R. Hendler, and D. Simonson. 1982. Insulin resistance is a prominent feature ofinsulin-dependentdiabetes. Diabetes. 31:795-801.

2. Lager, I., P. Lonnroth, H. Von Schenck, and U. Smith. 1983. Reversal of insulin resistance in type I diabetes after treatment with continuous subcutaneous insulin infusion. Br. Med.J. 287:1661-1664.

3.Beck-Nielsen, H., B. Richelsen, C. Hasling,0.H.Nielsen, L. Hed-ing, and N. S. Sorensen. 1984. Improved in vivo insulin effect during continuous subcutaneous insulin infusion in patients with IDDM.

Di-abetes. 33:832-837.

4. Proietto, J., A. Nankervis, P. Aitken, G. Caruso, and F. Alford. 1983.Glucose utilization in type I(insulin-dependent)diabetes: evidence

for a defect not reversible by acute elevations of insulin. Diabetologia. 25:331-335.

5. Del Prato, S., R. Nosadini, A. Tiengo, P. Tessari, A. Avogaro, R. Trevisan, A. Valerio, M. Muggeo, C. Cobelli, and G. Toffolo. 1983.

Insulin-mediatedglucose disposal in type I diabetes: evidence for insulin

resistance. J. Clin. Endocrinol. Metab. 57:904-910.

6.DeFronzo, R. A., D. Simonson, and E. Ferrannini. 1982. Hepatic and peripheral insulin resistance: a common feature of type 2 (non-insulin-dependent) and type 1 ((non-insulin-dependent) diabetes mellitus.

Diabetologia. 23:313-319.

7.Pernet, A., E. R. Trimble, F. Kuntschen, P. Damoiseaux, J.-Ph. Assal, C. Hahn, and A. E. Renold. 1984. Insulin resistance in type 1

(insulin-dependent)diabetes: dependence on plasma insulin

concentra-tion. Diabetologia. 26:255-260.

8.Pedersen,O.,and E.Hjollund. 1982. Insulin receptor binding and insulin action in fat cells from insulin-dependent diabetics. Diabetes.

31:706-715.

9. Lonnroth, P., G. Blohme, I. Lager, L. E. Tisell, and U. Smith. 1983. Insulin resistance in fat cells from insulin-treated type 1 diabetic

individuals. Diabetes Care. 6:586-590.

10. Madsbad, S., T. Krarup, P. Mc Nair, C. Christiansen, 0. K. Faber,I. Transbol,and C. Binder. Practical clinical value of the C-peptide response to glucagon stimulation in the choice of treatment in diabetes

mellitus.Acta Med. Scand. 210:153-156.

11.Pedersen, O., E. Hjollund, and H.0. Lindskov. 1982. Insulin

bindingand action infatcells from young healthy females and males.

Am.J. Physiol. 243:E158-E167.

12.Jorgensen, K. H., and K. D. Larsen. 1980. Homogenous

mono-'25I-insulins.Preparation andcharacterizationofmono-'25I(4tyrA,4)-and

mono-'251-(tyr

_{A,q)-insulin.}

Diabetologia. 19:546-554.

13.Beck-Nielsen,H.,0.Pedersen, K.Kragballe,and N. S. Sorensen. 1977. The monocyte as amodel forthe study ofinsulin receptors in

man. Diabetologia. 13:563-569.

14.Pedersen, O., H.Beck-Nielsen,and L. Heding. 1980. Increased

insulinreceptorsafterexerciseinpatients withinsulin-dependentdiabetes

mellitus.N.Engl.J.Med. 302:886-892.

Sonne, and J. Gliemann. 1981. Insulin receptor binding and receptor

mediated insulin degradation in human adipocytes. Diabetologia. 20: 636-641.

16.Kragballe, K., J. Ellegaard, and T. Herlin. 1980.

Antibody-de-pendent monocyte-mediated cytotoxicity. Isolation and identification byesterase-staining of monocytes detached from monolayers by lowering

the temperature. Scand.J.Haematol.24:399-403.

17.Hj0llund, E., 0.Pedersen, B.Richelsen,H. Beck-Nielsen,and N. S.S0rensen. 1983. Increasedinsulin bindingtoadipocytesand

mono-cytes andincreased insulin sensitivityofglucosetransport and metabolism

inadipocytes from noninsulin-dependent diabetics aftera low-fat/high-starch/high-fiberdiet. Metab. Clin. Exp. 32:1067-1075.

18.Kashiwagi,A., M. A.Verso,J.Andrews,B.Vasquez,G.Reaven,

and J. E. Foley. 1983. Invitroinsulin resistanceof humanadipocytes isolated from subjects with noninsulin-dependent diabetes mellitus.J.

Clin.Invest.72:1246-1254.

19. Heding, L. G. 1966. A simplified insulin radioimmunoassay

method. In LabelledProteins inTracerStudies. L.Donato,editor. Eu-ropeanAtomicEnergyCommunity,Brusells. 345-350.

20.Wildenhoff, K. E.1970. Amicro-method for the enzymatic de-termination ofacetoacetateand3-hydroxybutyrateinblood andurine. Scand.J. Clin. Lab. Invest. 25:171-175.

21. Itaya, K., and M.Ui. 1965.Colorimetric determination of free fatty acids in biological fluids.J.LipidRes.6:16-22.

22.0rskov,H., H.G. Thomsen, andH. Yde. 1968.Wick chroma-tography forrapidandreliableimmunoassay of insulin, glucagonand

growth hormone. Nature (Lond.). 219:193-195.

23. Murphy, B. E. P. 1979.Methodologicalproblems incompetitive protein binding techniques,the useofSephadexcolumnchromatography

toseparate steroids.ActaEndocrinol. (Copenh.). 147:37-55.

24.Heding,L.G. 1975.Radioimmunological determination of

hu-manC-peptideinserum.Diabetologia. 11:541-548.

25.Pedersen,O.,H.Beck-Nielsen,N.S.Sorensen,and A.Svejgaard. 1982. Heterogeneity of insulinreceptorsin patients withuntreated insulin-dependent diabetes mellitus.J.Clin. Endocrinol. Metab. 55:30-39.

26. Soll, A. H., I. D. Goldfine, J. Roth, C. R. Kahn, and D. M.

Neville. 1974.Thymic lymphocytes in obese (ob/ob) mice. Amirrorof theinsulin receptordefect in liver and fat. J.Biol. Chem. 249:4127-4131.

27.Olefsky,J. M. 1976. Decreased insulin binding to adipocytes and

circulatingmonocytesfromobesesubjects.J. Clin.Invest. 57:1165-1172.

28. Taylor,R., S. J. Proctor,0. James,F. Clark, and K. G. M. M.

Alberti. 1984.Therelationshipbetween human adipocyte and monocyte

insulinbinding. Clin. Sci. 67:139-142.

29. Hjollund, E., H.Beck-Nielsen, 0.Pedersen, B. Richelsen, and N. S. Sorensen. Defective non-insulin-mediated and insulin-mediated glucose transport andmetabolism in adipocytes from obese and lean

patients with untreated NIDDM. Diabetic Medicine. In press. 30.Karnieli,E., P.Hissin, I.A.Simpson,L. Salans, and S. Cushman. 1981. Apossiblemechanism of insulin resistance in the rat adipose cell instreptozotozin-induced diabetes mellitus. Depletion of intracellular

glucose transportsystems. J. Clin. Invest. 68:811-814.

31. Suzuki, K., and T. Kono. 1980. Evidence that insulin causes

translocation of glucosetransportactivitytotheplasmamembrane from

anintracellular storagesite. Proc. Nati. Acad. Sci. USA. 77:2542-2545. 32.Oka, Y., and M. Czech. 1984. Photoaffinity labelling of

insulin-sensitive hexosetransporters in intactratadipocytes.J.Biol.Chem. 259:

8125-8133.

33. Pedersen, O., E. Hj0llund, and N. S. S0rensen. 1982. Insulin receptorbinding and insulinaction in human fat cells: effects ofobesity andfasting.Metab. Clin. Exp. 31:884-895.

34. DeFronzo, R. A., E. Jacot, E. Jequier, E. Maeder, J. Wahren,

andJ. P.Felber. 1981.The effect of insulinonthedisposalof intravenous glucose. Results from indirect calorimetry and hepatic and femoral venous

catheterization. Diabetes. 30:1000-1007.

35. Nankervis, A., J.Proietto, P. Aitken, and F. Alford. 1984.

Im-paired insulinaction innewly diagnosedtype 1 (insulin-dependent)

di-abetes mellitus. Diabetologia.27:497-503.

36. Yki-Jarvinen, H.,and V. A. Koivisto. 1984. Insulinsensitivity

in newlydiagnosedtype 1 diabetics after ketoacidosis and after three

monthsof insulin therapy. J. Clin. Endocrinol. Metab. 59:371-378.

37. Barret, E. J., R. A.DeFronzo,S.Bovilacqua,and E. Ferrannini.

1982.Insulin resistance in diabetic ketoacidosis.Diabetes. 31:923-928.

38.Hidaka, H.,B.V.Howard,F.Ishibashi,F. C.Kosmakos,J. W.

Craig,P. H.Bennett,and J. Larner. 1981. Effect ofpHand 3-hydroxy-butyrateon insulinbindingandaction in cultured human fibroblasts.

Diabetes. 30:402-406.

39.Bratusch-Marrain,P. R., D.Smith, andR. A. DeFronzo. 1982. Theeffectof growthhormone onglucose metabolism andinsulin

secre-tion inman.J. Clin. Endocrinol. Metab. 55:973-982.

40.Rizza,R. A., L. J.Mandarino,and J. E.Gerich. 1982.Effects of growth hormone on insulin action inman. Mechanismsof insulin

re-sistance, impaired suppression of glucose production, and impaired

stimulation of glucose utilization. Diabetes. 31:663-669.

41.Cigolini,M.,and U.Smith. 1979.Humanadipose tissue inculture. VIII.Studies on theinsulin-antagonistic effect ofglucocorticoids. Metab. Clin. Exp. 28:502-5 10.

42.Deibert,D.C., and R. A. DeFronzo. 1980.Epinephrine-induced insulin resistancein man.J.Clin. Invest. 65:717-721.

43.Smith, U.,M. Kuroda,andI.A.Simpson. 1984. Counter-reg-ulation of insulin-stimulatedglucose transport bycatecholamines inthe

isolated rat adipose cell. J.Biol. Chem. 259:8758-8763.

44.Christensen,N. J.1974. Plasmanorepinephrineandepinephrine in untreateddiabetics during fastingand after insulinadministration.

Diabetes. 23:1-8.

45.Kobayashi, M., andJ. M.Olefsky. 1979.Effects of streptozotocin-induceddiabetesoninsulin binding,glucose transport andintracellular glucose metabolism in isolatedratadipocytes.Diabetes. 28:87-95.

46. Mandarino, L., B. Baker, R. Rizza,J. Genest,and J. Gerich. 1984.Infusion of insulin impairshumanadipocyteglucose metabolism

in vitro without decreasing adipocyte insulinreceptor binding.

Diabe-tologia. 27:358-363.

47.Pedersen,O., E. Hjollund,H.Beck-Nielsen, B. Richelsen, and N.S.S0rensen. Continuoussubcutaneousinsulin infusionfails to correct

impaired basalglucose metabolismandimpaired insulin sensitivity of

adipocytes from patientswith type 1 (insulin-dependent) diabetes.