Diabetes due to secretion of a structurally
abnormal insulin (insulin Wakayama). Clinical
and functional characteristics of [LeuA3]
insulin.
K Nanjo, … , K Miyamura, B D Given
J Clin Invest.
1986;
77(2)
:514-519.
https://doi.org/10.1172/JCI112331
.
We have identified a non-insulin-dependent diabetic patient with fasting hyperinsulinemia
(90 microU/ml), an elevated insulin:C-peptide molar ratio (1.68; normal, 0.05-0.20), normal
insulin counterregulatory hormone levels, and an adequate response to exogenously
administered insulin. Insulin-binding antibodies were absent from serum, erythrocyte insulin
receptor binding was normal, and greater than 90% of circulating immunoreactive insulin
coeluted with 125I-labeled insulin on gel filtration. The patient's insulin diluted in parallel
with a human standard in the insulin radioimmunoassay, confirming close molecular
similarity. The patient's insulin was purified from serum and shown to possess both reduced
binding and ability to stimulate glucose uptake and oxidation in vitro. Analysis of the
patient's insulin by high-performance liquid chromatography (HPLC) revealed two products:
7.3% of insulin immunoreactivity coeluted with the human standard, while the remaining
92.7% eluted as a single peak with increased hydrophobicity. Family studies confirmed the
presence of hyperinsulinemia in four of five relatives in three generations, with secretion of
an abnormal insulin documented by HPLC in the three tested. Leukocyte DNA was
harvested from the propositus and the insulin gene cloned. One allele was normal, but the
other displayed a thymine for guanine substitution at nucleotide position 1298 from the
putative cap site, resulting in a leucine for valine substitution at position 3 of the insulin A
chain. Insulin Wakayama is therefore identified as [LeuA3] […]
Research Article
Find the latest version:
http://jci.me/112331/pdf
Diabetes Due
to
Secretion of
aStructurally
Abnormal Insulin
(Insulin Wakayama)
Clinical andFunctional Characteristics of[LeuA3]
InsulinK. Nanjo, T.Sanke, M. Miyano, K.Okai,R.Sowa,M.Kondo, S.Nishimura,K.Iwo,and K.Miyamura
DepartmentofMedicine, Wakayama UniversityofMedical Science, Wakayama, Japan B. D.Given, S. J.Chan, H.S. Tager, D. F. Steiner, and A. H. Rubenstein
DepartmentsofMedicine, Biochemistry, and Molecular Biology, University of Chicago, Chicago, Illinois 60637
Abstract
We haveidentifiedanon-insulin-dependent diabetic patientwith
fastinghyperinsulinemia (90 ,uU/ml),anelevated insulin:C-pep-tide molar ratio(1.68; normal, 0.05-0.20),normal insulin
coun-terregulatory hormone levels,andanadequateresponseto
ex-ogenously administeredinsulin.Insulin-bindingantibodieswere
absent from serum, erythrocyte insulin receptor binding was
normal, and >90% of circulating immunoreactiveinsulincoeluted with 1251-labeled insulin ongel filtration. The patient's insulin
dilutedinparallel withahumanstandard in the insulin radioim-munoassay,confirming close molecular similarity.
Thepatient's insulinwaspurified fromserumandshownto possess bothreduced bindingand abilityto stimulate glucose uptake and oxidationin vitro.Analysis ofthepatient's insulin by high-performance liquid chromatography (HPLC)revealed
two products: 7.3% of insulin immunoreactivity coeluted with the humanstandard, while theremaining92.7% elutedas asingle
peak with increased hydrophobicity. Familystudiesconfirmed thepresenceofhyperinsulinemiainfourof five relatives in three
generations, with secretion ofanabnormalinsulin documented by HPLC inthe three tested. LeukocyteDNA washarvested
from thepropositus and the insulingenecloned. One allelewas
normal,but theotherdisplayedathymine forguaninesubstitution
atnucleotideposition 1298 from theputativecapsite, resulting
inaleucine forvaline substitution atposition 3 of the insulinA chain. InsulinWakayama is thereforeidentifiedas
[Leu'J
insulin.Introduction
Hyperinsulinemia isafrequentoccurrence innonketotic Type
IIdiabetes, especially in the obese (1, 2). Itis generally
accom-panied by elevated C-peptide levels andareduction inthe
hy-poglycemicresponsetoexogenousinsulin (3, 4). Recently, several
patients with hyperinsulinemia, relatively normal C-peptide levelsresulting in elevatedinsulin:C-peptide molarratios, and withanormalresponsetoexogenousinsulin have been identified.
Thesepatients have been proven to secrete a structurally
ab-normalinsulin(5-8). Two of these patients have had the specific abnormality localizedby comparison with synthesized insulin
Addresscorrespondence toDr. Rubenstein,Departmentof Medicine, TheUniversity of Chicago, 5841 S.Maryland Ave.,Chicago,IL 60637.
Receivedfor publication10June 1985.
analogsusing high-performanceliquid chromatography (HPLC)' following localization of the defect by Mbo IIrestrictionenzyme analysis (7, 9, 10) and subsequently by detailed analysisat the genomic level (1 1, 12). The study of these mutant insulins has improved our understanding ofstructural determinants ofinsulin receptor binding and function.
Thepropositusof the current study was shown to secrete an abnormal insulin of increased hydrophobicity on HPLC (7). However, structural differences could not be identified by re-striction enzyme analysis. The current study extends considerably ourunderstanding ofthisfamilial mutant insulin. Affectedfamily
members possess a normal insulin allele and one which codes for a leucine for valine substitutionatposition 3 of theinsulin Achain. The clinical andbiological impact of this autosomal, dominantlyinherited defect isdescribed.
Methods
Clinicalinformation.Theproposituswas a56-yr-old,non-obese(height, 148cm;weight, 43 kg) Japanesewomanwhopresented in 1980 with polyuria and weightloss. Hyperglycemia (273 mg/dl at fasting) with glucosuria and ketonuriawerepresent.Physical examinationwas un-remarkable, as wereroutine laboratoryandradiographictests.Highly purified porkorbeef insulinwasadministered for2 mo,butwas dis-continuedbecauseofallergic reactionstobothinsulinsattheinjection site. Subsequently, shehas beentreatedwith oralhypoglycemicagents
withadequate control. She had nohistoryof obesityor hypoglycemic symptoms. Her youngerbrother subsequentlydevelopeddiabetes mellitus followingafebrile illness in 1982atage52, and hasalsobeen treated withoralhypoglycemicagents.
Circulating insulin counterregulatory hormoneswerewithin the
nor-malrange(cortisol, 24.5gg/dl;growth hormone, 1.6 ng/ml;glucagon,
56pg/ml;triiodothyronine, 0.9 ng/ml; thyroxine, 7.9 ug/dl),and anti-insulin, anti-receptor,andisletcellsurface antibodieswere absent.
Thy-roid anti-microsomal antibodieswerepositive(1:102,400)andthyroid biopsy confirmed chronic lymphocytic thyroiditis. Thyroid functiontests,
however,werenormal.
Clinical studies. Thepropositusand five of her firstdegree relatives agreedtobestudied. Fasting levels of plasmaglucose,C-peptide,and seruminsulinwere determined inall. The glycemic response to exogenous purepork insulin (Actrapid; Novo Research Institute, Copenhagen, Denmark),0.1U/kgadministeredintravenously, was determined in the
proposituson two occasions. Theglucose, insulin, and C-peptide re-sponsesto75 g oralglucoseweredeterminedin thepropositusand three otherfamilymembers.
Characterization ofcirculating insulin. The percent of fasting insulin immunoreactivity correspondingto circulating proinsulin was determined
by gel filtrationonBiogel P-30 (Bio-rad Laboratories, Richmond, CA) forthepropositusand allfivefamilymembers.
1.Abbreviations used in thispaper:ACN, acetonitrile;HPLC, high-per-formance liquid chromatography; IRI, immunoreactive insulin; TFA,
trifluoroaceticacid.
514 Nanjoetal.
J.Clin.Invest.
©TheAmericanSociety for Clinical Investigation,Inc.
Insulin forreceptor binding, biological activity studies, and HPLC waspurifiedusing octadecyl silica cartridges(Sep-Pak; Waters Associates, MilliporeCorp.,Milford,MA)(Cohen, R. M., B. D. Given, J.
Licino-Paixao,S. A. Provow, P. A. Rue, B. H. Frank, M. A.Root,K. S. Polonsky, H.S. Tager, and A. H. Rubenstein, manuscript submitted for publication).
Cartridgeswereprepared bywashing with 5 ml of acetonitrile (ACN, HPLC grade, Fisher Scientific Co., Fairlawn, NJ) followed by 5 ml of water. Plasma drawn into a disposable plastic syringe was loaded undiluted ontothecartridgeat a rate of 20 ml/h using a Harvard pump (Harvard Apparatus Co., Inc., S. Natick, MA), with a0.45-umfilter(Millex HA;
MilliporeCorp.,Bedford,MA)interposedbetween the syringe and
car-tridge.Thesyringewaswashedwith5 mlofwater pumped through the
filterand cartridge. Thefilterand syringe were then discarded and all
subsequentsteps wereperformed usingglasssyringesatthe same pump
flowrate.
Thecartridge was washedsequentiallywith threedifferent solvent
mixtures.Thefirstwash consistedof5 ml 20% acetonitrile (ACN): 79.2% water0.8% trifluoroaceticacid (TFA, Aldrich Chemical Co., Milwaukee, WI),followedby 5 mlofa 99%water: 1%TFA mixture, and 5 ml of an
80% ACN: 20%methylenechloride (Aldrich Chemical Co.) mixture.3 mlof ACNwereappliedtothecartridgetoremoveresidualmethylene
chloride.
Insulin waseluted from the cartridge with 5 ml 40% ACN: 59.4% water: 0.6% TFA and collected into asiliconizedscintillation vial. The
ACNwasevaporated under reduced pressure,afterwhich the remaining
solutionwasquick-frozen in dry iceacetone andlyophilized. The
ly-ophilized powderwascompletelysoluble inaslittleas0.1 mlofneutral
bufferordilute acetic acid. Yields by this methodwere70-90%.
Invitroreceptorbinding of insulinextractedfrom propositusserum wasstudiedusingratadipocytes(13).Thebiological activity of her insulin wascompared with a semisynthetic human insulin standard (Novo Re-searchInstitute) and with insulin extracted fromanormalcontrol for its abilitytostimulate uptakeof2-deoxy-D-[l-'4Cjglucose (sp act, 50
mCi/mM; New England Nuclear, Boston, MA) and oxidation of ['4C]glucose (spact,250mCi/mM, NewEngland Nuclear) in isolated
ratadipocytes (14, 15). Insulin extracted fromtheplasma ofthepropositus
andthree family members was separated by reverse-phase HPLC (7) usingaC- 18 column(Altex, Inc., Berkeley, CA) andamobilephase of 30%ACN and 70%TEAPbuffer (20mMtriethylamine, 50mMsodium perchlorate, 100mMphosphoric acid broughttopH 3.0 with NaOH; allreagents were HPLCgrade). Thecolumn eluantwascollected and assayed for insulin immunoreactivityasdescribed (7). Recoveries of in-sulin immunoreactive materialinjectedontothecolumnwere70-90%. Characterizationofinsulingene structure. GenomicDNA wasisolated fromcirculating leukocytes obtained from thepropositusasdescribed previously (9).Thetwoallelesof the insulingenewerethenclonedusing
amodification of thexVX systemdeveloped by Seed (16) and will be described in detail elsewhere(Chan, S.J.,Y.Lu, K.Nanjo, T.Sanke,
M. Miyano,A.H.Rubenstein, andD. F.Steiner, manuscriptin
prep-aration). Briefly,plasmidirHInswasconstructed from-rVX. Itcontains
aunique 1,400basepairEco RI-Xho I DNAfragmentderived from the
5'flanking regionofthehuman insulingeneaswellasthe Escherichia colisuppressor F geneandwastransformed intoE.colistrainMC1061 (p3) (16). Subsequently, high molecularweight leukocyteDNAisolated from the proposituswasdigestedtocompletionwith restriction nucleases
Eco RI andSalI,ligatedtoXgtWES-Eco RI arms,andpackagedinto recombinant phage in vitro. This recombinantphage "library"wasthen amplifiedonE.coli MC1061(p3, irHIns).During amplification, however, onlytherecombinant phagecontainingthe human gene sequencewould acquirethe suppressor F gene duetohomologousrecombinationin vivo
withirHIns.Theseplaquesweresubsequently isolatedbygrowthon a suppressor minus E.coli strainLG75 andpositivelyidentifiedby plaque hybridizationto32P-labeledcloned humaninsulingene DNAfragment availablein thelaboratory(12).Toobtaincloned DNAforeachallelic insulingene, advantage wastaken ofthe finding that thetwo alleles
containedaslightdifference in thesize ofthe 5'flanking polymorphic region (17). Restriction mapping andDNA sequence analysisofthe
insulingeneswereperformedasdescribedpreviously (12).
Assay techniques. Plasma glucose was measured by the glucose oxidase
method, using a glucose analyzer (Model 23, Yellow Springs Instrument Co., Yellow Springs, OH). Serum insulin (18), growth hormone (19), and plasma C-peptide (20) and glucagon (21) concentrations were mea-sured byradioimmunoassay, andcortisol levels were determined by a competitive-binding assay (22). Insulin antibodies were measured as pre-viouslydescribed (23). Insulin binding to erythrocyte insulinreceptors from the propositus was examined using the method of Gambhir et al.
(24). Data is expressed as mean±standard error of the mean. All studies wereapproved by the institutional review board and written informed consent was obtained.
Results
Screeningfor insulin resistance. At the time of presentation, the propositus was found to have hyperinsulinemia (Fig. 1). This apparent insulin resistance was not due to elevated levels of counterregulatory hormonesor the presenceof circulating anti-insulin oranti-insulin receptor antibodies.The patient's eryth-rocyteinsulin receptors, when compared with 20 normal con-trols, displayed similar specific binding(8.85 vs. 8.21±0.2%) and dissociationconstants(KD) (0.82 X 109M vs.0.92±0.05 X 10-9
M) at the high affinity site. The fact that she was not severely insulin-resistant was confirmed by demonstratinga 39% fallin plasma glucose in responseto0.1 U/kg intravenous insulin (from
185 mg/dlbaseline to 116 mg/dl at 60min).
Since proinsulin and its 9,000-mol-wt intermediates cross-react in theinsulin assay (25) and elevated serum immunoreac-tive insulin (IRI) can arise from elevations in proinsulin (26, 27), serum from the propositus and her five available family members was subjected to size exclusion gel chromatography. The peak coeluting with 1251-insulin accounted for >90% of serum IRI in all six subjects (data not shown).
Clinical studies. To determine if hyperinsulinemia was pres-entin otherfamily members, fasting insulin and C-peptide levels weredetermined in the five additional family members willing tobe studied. As shown in Fig. 1, hyperinsulinemia was present
6ev
D-T
Subject Age/Sex Plasma Glucose (mg/d1)
56F 2 49M
3 54F 4 28 M 5 4M 6 2M
IRI# CPR
(pmol/ml) (pmol/ml)
272 0.62(90) 81 0.36(52)
93 0.50(73) 69 0.55 (80) 72 0.23(33) 70 0.02( 3)
0.37 0.38 0.49
0.66 0.29 0.28
RI CPR
ratio 1.68 0.95 1.02 0.83 0.79 0.07
Figure1.
Pedigree
ofthepropositus (subject
1)and herfamily.
Num-beredsubjects
werestudied.Family
members withfasting
hyperinsulin-emia
(diagonal lines)
and/ordiabetes(*)
areindicated.CPR,C-pep-tide
immunoreactivity. f,
deceased; #,uU/mlinparentheses.
infourofthefive,althoughonly thepropositus (subject 1) dis-playedfasting hyperglycemia. Oral glucose tolerance tests per-formed afteranovernight fastareshowninFig. 2for the pro-positus and three others.Ascan be seen, in additiontothe pro-positus, the brother (subject 2) and sister (subject 3) both met thecriteriafor the diagnosis of diabetes.Notably,while the pro-positus and her brother, who had the worstglucose tolerance, had onlymarginal increasesin seruminsulininresponseto oral glucose, the son (subject 4) and sister had substantialincreases in serum insulin and relatively normal glucose tolerance
(Fig. 2).
Characterization ofcirculating insulin. Following the initial identification of hyperinsulinemia in thepropositus in 1982, HPLCanalysis wasperformed.Itidentified the presence inserum of an abnormal insulin species of increased hydrophobicity (7). Tocharacterize thispatient'scirculatinginsulinfurther,a num-berof studies wereperformed.
Seruminsulin from the proposituswasassayedinmultiple dilutions byradioimmunoassay. AsshowninFig. 3,serum in-sulindilutedin parallel withsemisynthetichumaninsulin stan-dard,indicatingcloseimmunologic similarity.However,insulin purified from propositusserum by Sep-pakextractionshowed reduced binding to rat adipocytes. As shown in Fig. 4, the pa-tient'sinsulin competed with
'251I-labeled
insulin for binding toadipocytereceptorsonly 19.8% as well as thesemisynthetic hu-maninsulin standard. This difference was not an artifact of the extraction procedure, as demonstrated by the normal binding of insulin extracted from the serum of a normal subject, using the same procedure (Fig. 4).
Aswould beanticipated from the receptor binding studies, thepatient's insulin demonstrated decreasedbiological activity
invitroaswell.Fig. 5Adocuments decreaseduptake of
2-deoxy-D-[1-4Cjglucose
in rat adipocytes in response tothe patient's insulin. Glucose oxidation stimulated by the patient's insulinwas alsoreduced asmeasured by themetabolismof
['4C]glucose
to
"'CO2
(Fig.5B).Again,insulin extracted fromanormalsub-ject bythesameprocedure stimulatedglucose uptakeand me-tabolism normally (Fig. 5). Insufficient material was available to constructfullbindingordose-responsecurvesfor insulin from thepropositusornormalsubject.
400-C
300-0
, _
, -0
O'c
200-vE
E 00
co
100-I E
0
_
- a
2
50-50
02
I,
.,
If'/8 1/4 1/2 '/I
0 0.1 0.5 1 2.5 5 0 20 Human Insulin[ng]v- .
Figure3.Radioimmunoassay dilutioncurvesof thepropositus'serum
insulin (o) andsemisynthetichumaninsulin (-). The dilutions
per-formed for propositus'serum areindicated in the figure.
Having demonstrated hyperinsulinemiain several firstdegree relatives, serum insulin from the propositus and three hyper-insulinemic family members was subjectedto HPLC. As shown inFig. 6, 7.3% ofthe propositus' insulin coelutedwith the human insulin standard (peak binFig. 6). However,theremaining92.7% eluted as a single peakofsignificantly greater hydrophobicity (peak dinFig. 6). Thisfindingis inagreementwith ouroriginal report (7). The hyperinsulinemia inthe family members was alsodue tohighcirculating levels ofthe mutant insulin(Fig. 6). Importantly, humaninsulin, when added to a patient sample, elutedatitsexpectedposition,confirming the absence ofa sep-aration artifact (datanotshown).
Characterizationofgenomic DNA. Theabilitytoclone and sequenceleukocyteDNAhas allowed the molecularabnormality
in twoprevious cases (insulinsChicagoand LosAngeles)tobe confirmedatthegenomiclevel.
Both alleles of theinsulin gene of the proposituswerecloned andsequenced. One allele wasentirelynormal. However, in the other allele, athymine forguanine substitution was found in the codonforposition 3 of theA chain, resultingin a leucine for valine substitution (Table I). The cloning procedure will be
4
- IF F F FI F IF- ITF~
-
6-4 I--
-2-
-'-
,-0d--I
.6 60 o120 60 1200 6C
TIME (minutes)
D1200 60 120
Figure2.Changes in the levels of plasma glucose (top), IRI, and C-peptide (CPR) (bottom) after administration of 75 g of glucose orallytothepropositus(1), brother (2), sister (3), and son (4).
ae 0
m
I-m
Insulin[ng]
Figure 4.Bindingof'25I-insulinto ratadipose tissue cells in the
pres-enceofvaryingamountsofsemisynthetichumaninsulin standard (-)
and IRIisolated from theserumof thepropositus (o) andanormal subject(A). Results are expressed as the percentageof'25I-insulin
spe-cificallybound per 1 X 105 ratadipocytes.
r-I c
0 2
I v E 2.0-o-
-I co
a @
1.5-0
* .0- 10
In 0
X CW 0
1.0-cmj E _
o
0T
c
0
x
-0
° 8( o°° 6(
elnE 4C
C-Discussion
A
a
S
---T
2 4 6 8
Insulin [ng]
B
0-~~~~~~~~
2~~~~~
D=
0
0 0. 1.0 1.5 2.0
Insulin
[ng]
2.5
Figure5.Stimulation of
2-deoxy-D-[1-'4Clglucose
uptake(A)and ['4C]glucoseoxidation (B) in rat adipocytes by varying amountsof hu-man insulin standard(-) and insulins purified fromserum of the pro-positus (o) and a normal subject (A).described in detail elsewhere (Chan, S. J., Y. Lu, K. Nanjo, T. Sanke, M. Miyano, A. H.Rubenstein, and D. F. Steiner, manu-scriptin preparation).
a b c S.
2-~~~~~~~
X
0-0 2
0-4
0-50 60 70 80 90 100
Fraction Number
Figure6.HPLCanalysisofseruminsulin from thepropositus (1), brother(2),sister(3),andson(4).Chromatographyofamixture of bovine(a),human(b),andporcine(c)insulins is shown in theupper
panel (S).
Theproposituspresentedwith glucose intolerance, hyperinsu-linemia,andanelevatedinsulin:C-peptide molar ratio.Obesity
and counterregulatory hormoneexcess wereabsent,aswereother less likelycauses ofinsulin resistance: circulating anti-insulin
oranti-insulin receptor antibodies, or alterations in receptor binding. Since thispresentation resembled the two
previously
reported cases manifesting structurally abnormal insulin
mol-ecules,exogenousinsulinwasadministeredto testin vivo insulin
sensitivity.
Theadequate dropinblood glucosenotedsupportedthelikelihoodthatthecirculatingIRI was notcomprisedentirely ofnormalinsulin.Gelfiltration ofserumexcluded hyperproin-sulinemiaand documented that thecirculatingIRIinthis patient (and five ofherfamilymembers) coelutedwith'25I-insulin.Based
onthese
findings,
HPLCanalysis
ofpropositus
insulinwas per-formedthat documented amolecular abnormality(7). Further characterization of thismutantinsulin species followed.The patient's circulating IRI diluted in parallel with the standard in the insulin assay(Fig. 3), again demonstrating the inability of polyclonal antisera to detect moleculardifferences acrossinsulin species (mutantoranimal) (6). However, funda-mental differences in bioactivity were apparent. The
patient's
insulincompetedwith'25I-labeledinsulinfor bindingto rat adi-pocytes only about 20% as well as a human insulin standard (Fig. 4). The patient's insulin also showed markedly reduced potencyinits abilitytostimulate glucose uptake and oxidation inrat adipocytes (Fig. 5). Insufficient materialwasavailable to
produce a complete dose-response curve, which made exact quantitation
difficult,
and therelativecontributionsof the normal and mutant insulins present in propositus serum with respect tobindingand biological activity could notbeassessed.TableI. Gene SequenceofInsulin Wakayama
Achainposition
1 2 3 4 5 6 7 8 9 10
Normal Gly Ile Val Glu Gin Cys Cys Thr Ser Ile
GGC ATT GTG GAA CAA TGC TGT ACC AGC ATC
Mutant Gly Ile Leu Glu Gln Cys Cys Thr Ser Ile
GGC ATT TTG GAA CAA TGC TGT ACC AGC ATC
T T T
1,292 1,300 1,320
Comparison of nucleotide and amino acidsequencesof thetwoalleles in theregionofthepointmutation.Numbers abovethe amino acid se-quenceindicate position of amino acid residues of the insulinAchain.Numbers below thenucleotidesequenceindicate thepositionof nucleo-tides from theputativecapsite.AsingleGtoTtransversionatthe codon forvalineatpositionA3 resulted inaleucine for valine substitution.
insulin molecule (28-32). It is therefore likely that this neutral for neutral amino acid substitution produced subtle confor-mational changes. These changeswereinsufficienttoalter im-munogenicity(Fig. 3), but markedly affected receptorbinding,
biological activity (Figs. 4, 5), and surface hydrophobicity as assessedby HPLC(Fig. 6).Knowledge of the structureofthis mutant insulin will permit its synthesis. Availability of larger
quantitiesof this insulin variant will then allowmoredefinitive studies of itsbindingandglucose utilization characteristics to becarriedoutin the presence and absence of normal insulin.
Finally,returningtotheclinical syndrome, all ofthe affected
family members studieddisplayed fasting hyperinsulinemiain the presenceof normalormildlyelevatedC-peptidelevels, re-sulting in an elevated insulin:C-peptide molarratio. Only the propositus underwent aninsulintolerance test,witharesultant modestfall in blood sugar. A second insulin tolerance test,at 2 yrafterinitiationof treatment with oralhypoglycemic agents, wasnormalwith a 78%fall in plasma glucose at 45 min. Thus,
althoughanormal responsetoexogenous insulin or anormal euglycemic insulin clamp has been proposedas oneof the di-agnostic criteria forpatientswith this syndrome (8),somedegree of secondary insulin resistance in thosepatients with marked
glucoseintolerance may beseen.It hasbeenproposedthat insulin
sensitivity improveswithnormalization of blood glucose in di-abeticpatients(33), and this mayexplain the improvement in thepropositus' response to exogenousinsulin. Thus, a greater than 50% fall inplasmaglucose in response to0.1 U/kg of in-travenous insulin may not always be seen in this syndrome. Anothervariable inthis syndrome is the presence or absence of glucoseintolerance. Fig. 2 shows the results of 2-h oral glucose tolerance tests on severalfamily members. While fasting hyper-glycemia was present only in the propositus, several members
displayedpeak plasma glucose levels above 200mg/dl.Inanother
family that was well characterized with oral glucose tolerance testsandwith glucose and insulin clamp studies in the propositus (8), variable glucose intolerance was also found and it was hy-pothesized that diabetes may occur when there is an additional defectinbeta cell secretion, such that the sum of normal and
abnormalinsulin molecules secreted is inadequate to maintain euglycemia (8, 34). Nevertheless, it is obvious that further studies will be required to determine why individuals who secrete the mutantinsulin show varying patterns of glucose tolerance. This will include detailed quantitative studies ofbasal and stimulated levels of thecirculating normal and abnormal insulin species,
thebiological activity of the mixture of circulating insulins, and the peripheral sensitivitytoinsulin in vivo in each affected family member. These investigationsare nowunder way.
Since it is likely thatmost orallpatients withastructurally abnormal insulin molecule will be heterozygous, withonenormal insulin allele, and since removal of up to 90% ofthe pancreas can still result in normal glucosetolerance (35), it seemsthat forglucoseintolerance todevelopin thesepatients,anadditional diabetogenic insult will needtobepresent(6, 34). An alternative explanation would involve repression of the normal allele, as suggestedby therecent report oftwo additional cases ofabnormal insulin secretion by Seinoet al.(36). Nonormal human insulin wasdemonstrated in serum in either of their cases on HPLC analysis. However,normalinsulin was not added to theirputative abnormal insulin samples, and the exact relationship oftheir findingstothosewehavedescribedisuncertain.
Thisrepresents the second abnormal insulin in which the familial mode of inheritance could be examined. In theinitial family studied, the father of the proposituswasthe earliest iden-tifiablecase.Fiveof sixsubsequent offspringswereaffected (8). In the family reported here, only 6 of 14 potentially affected members could be screened. However, five of those secrete the abnormal species. While adominant mode of transmission is obvious inboth cases,it seemsnecessarytoproposesomepositive genetic pressure leadingto agreater than expectedoccurrence of the mutantgeneinaffected families.
The prevalence of genetic mutationsaffectingthe structure ofthe insulin molecule in the general population is unknown. Upto thepresent, only thosepatients manifesting the mutant insulin syndrome (5-8, 36) with unusual or familial Type II diabetes have been screened and discovered. Thus, mutant in-sulin species with normal or relatively well-preserved binding and biological activity characteristics, and therefore normal metabolic clearances, are unlikely to be discovered by this ap-proach sincehyperinsulinemia will be absent or subtle. Future screening techniques using HPLC and/or gene analyses are thus being developed.
References
1.Glass, A. R., K. D. Burman, W. T. Dahms, and T. M. Boehm. 1981.Endocrine functioninhuman obesity. Metab. Clin. Exp. 30:89-104.
onsetdiabetesonsplanchnicglucosebalanceafteroral glucose ingestion.
Diabetes. 27:121-131.
3. Reaven, G. M., R.Bernstein,B.Davis, and J. M.Olefsky. 1976.
Nonketotic diabetes mellitus: insulin deficiency or insulin resistance. Am.J.Med.60:80-88.
4.Kuzuya, T., A.Matsuda,T.Saito, and S. Yoshida. 1976. Human
C-peptide immunoreactivity(CPR)inblood and urine: evaluation of a
radioimmunoassaymethod and its clinical applications.Diabetologia.
12:511-518.
5.Tager, H., B.Given,D.Baldwin,M. Mako, J. Markese, A.
Rub-enstein,J.Olefsky,M.Kobayashi, 0. Kolterman, andR.Poucher. 1979.
Astructurally abnormal insulin causing human diabetes. Nature(Lond.). 281:122-125.
6.Given,B. D., M. E. Mako, H. S. Tager, D.Baldwin,J. Markese, A. H. Rubenstein, J.Olefsky, M. Kobayashi, 0. Kolterman, and R. Poucher. 1980. Diabetes due to secretionofan abnormalinsulin. N.
Engl.J.Med. 302:129-135.
7.Shoelson, S.,M.Haneda,P.Blix,K.Nanjo,T.Sanke,K.Inouye,
D. Steiner,A.Rubenstein,and H. Tager. 1983. Three mutantinsulins inman. Nature(Lond.). 302:540-543.
8.Haneda, M., K.S. Polonsky,R. M.Bergenstal,J. B.Jaspan,S.E.
Shoelson, P. M. Blix,W. B. Wishner,J. M. Olefsky, G. Freidenberg,
H.S. Tager,D. F. Steiner, andA. H. Rubenstein. 1984. Familial hy-perinsulinemiadue to astructurally abnormal insulin: definition ofan
emergingclinical syndrome.N.Engl.J. Med. 310:1288-1294.
9. Kwok, S.C. M., S.J.Chan,A.H.Rubenstein,R.Poucher, and
D. F.Steiner. 1981.Lossofanendonuclease cleavage site in thegeneof
astructurally abnormal human insulin. Biochem.Biophys.Res.Commun. 98:844-849.
10. Shoelson,S., M. Fickova, M. Haneda, A. Nahum, G. Musso,
E. T.Kaiser,A. H.Rubenstein, andH.Tager. 1983. Identification ofa mutanthumaninsulin predictedtocontainaserumforphenylalanine substitution.Proc.Natl.Acad. Sci. USA. 80:7390-7394.
11.Kwok, S. C. M., D.F.Steiner,A.H.Rubenstein, andH.S. Tager. 1983.Identification ofapoint mutation in the human insulingenegiving riseto astructurally abnormal insulin (Insulin Chicago). Diabetes. 32: 872-875.
12. Haneda,M.,S.J.Chan, S. C.M. Kwok, A. H.Rubenstein, and
D. F.Steiner. 1983.Studiesonmutanthumaninsulingenes:identification
and sequenceanalysis ofageneencoding Ser B24 insulin. Proc. Nati. Acad. Sci. USA. 80:6366-6370.
13.Rodbell,M. 1964. Metabolism of isolated fat cells. I. Effects of
hormones onglucose metabolismandlipolysis.J.Biol.Chem. 239:375-380.
14.Olefsky, J.M. 1975.Effect of dexamethasoneoninsulinbinding, glucose transport,andglucose oxidation of isolatedrat adipocytes.J. Clin.Invest. 56:1499-1508.
15. Ueda,Y. 1977. Specificity of insulin action onfat cell of the Zuckergenetically obese rat "Fatty". [II]. Folia Endocrinol. Jpn. 53: 679-689.
16.Seed,B.1983. Purification ofgenomicsequencesfrom bacterio-phage libraries by recombinationandselectioninvivo. Nucleic Acids
Res. 11:2427-2445.
17.Bell, G. I.,J.H.Karam, andW.J. Rutter. 1981.Apolymorphic
DNAregion adacenttothe 5' endofthehuman insulingene.Proc.Nati. Acad. Sci. USA. 78:5759-5763.
18. Starr,J.I.,D. L.Horwitz,A.H.Rubenstein,andM.E.Mako. 1979. Insulin,proinsulinandC-peptide. InMethodsofHormone
Ra-dioimmunoassay. 2nded. B. M.Jaffeand H. L.Bhrman,editors.
Aca-demic Press,NewYork. 613-642.
19. Odell, W. D., P. L. Rayford, and G. T. Ross. 1967. Simplified,
particallyautomated methodfor radioimmunoassayof humanthyroid.
Stimulating, growth, leutinizingandfollicle stimulating hormones. J. Lab. Clin. Med.70:973-980.
20. Faber, 0. K., C. Binder, J. Markussen, L. G. Heding, V. K.
Naithani,H.Kuzuya, P. Blix, D. L. Horwitz, and A. H. Rubenstein. 1978. Characterization of seven C-peptide antisera. Diabetes. 27:170-177.
21. Kondo, M., K. Nanjo,T. Sanke, H. Koike, Y. Moriyama, K. Iwo,and K.Miyamura. 1980. Clinical applicationoftwo glucagon RIA kitsforglucagonassay.J.Jpn.Diab.Soc. 23:803-808.
22.Murphy, B. E. P. 1967. Somestudies oftheprotein bindingof steroidsandtheirapplicationto theroutine microand ultramicro mea-surements of various steroids in body fluids bycompetitive protein-binding radioassay. J. Clin. Endocrinol. Metab. 27:973-990.
23. Goldman, J., D.Baldwin,W. Pugh, and A. H.Rubenstein. 1978.
Equilibrium bindingassay andkineticcharacteristicsofinsulinantibodies. Diabetes. 27:653-660.
24. Gambhir, K. K., J. A.Archer, and C. J. Bradley. 1978. Char-acteristics ofhuman erythrocyte insulin receptors. Diabetes. 27:701-708.
25. Robbins,D. C., H. D. Tager, and A. H.Rubenstein. 1984. Bio-logical and clinical importance of proinsulin.N.Engl. J. Med. 310:1165-1175.
26.Gabbay, K. H., K. Deluca, J. N.Fisher,Jr., M. E. Mako, and A.H.Rubenstein. 1976. Familialhyperproinsulinemia: anautosomal
dominantdefect.N.Engl.J. Med. 294:911-915.
27.Robbins,D. C., P. M.Blix,A. H.Rubenstein,Y. Kanazawa, R. Kosaka, and H.S.Tager. 1981. A humanproinsulin variantatarginine 65.Nature(Lond.). 291:679-681.
28. Blundell,T.L., D.Phil,J. F.Cutfield, S.M.Cutfield,E. J.Dodson, G.G.Dodson,D.C.Hodgkin, andD. A.Mercola.1972. Three-dimen-sionalatomicstructureof insulin and itsrelationshiptoactivity. Diabetes. 21:492-505.
29. Zahn, H.,D. Brandenburg, and H.G. Gattner. 1972. Molecular basisof insulin action: contributions of chemical modifications and
syn-thetic approaches. Diabetes. 21:468-475.
30. Pullen, R. A., D. G. Lindsay, S.P. Wood, I. J. Tickle, T. L. Blundell, A. Wollmer,G.Krail,D.Brandenberg, H.Zahn,J.Gliemann, andS. Gammeltoft. 1976. Receptor-binding regionof insulin.Nature
(Lond.). 259:369-373.
31.DeMyets, P., E. V.Obberghen, andJ.Roth. 1978.Mapping of
theresiduesresponsible forthenegative cooperativity of the
receptor-binding region of insulin.Nature(Lond.).273:504-509.
32.Gliemann,J.,0. Sonne, S. Linde, andB. Mansen.1979.Biological
potencyandbindingaffinity of monoiodoinsulin with iodine andtyrosine
A14 ortyrosine A19. Biochem.Biophys.Res.Commun.87:1183-1190. 33.Scarlett,J. A., R.S.Gray,J.Griffin, J. M.Olefsky,and0. G. Kolterman. 1982. Insulintreatment reversesthe insulin resistance of TypeIIdiabetesmellitus. Diabetes Care. 5:353-363.
34.Shoelson, S. E.,K.Polonsky,A.Zeidler,A. H.Rubenstein,and
H.S. Tager. 1984.HumaninsulinB24(Phe- Ser):secretion and
met-abolic clearance ofthe abnormalinsulin inmanandinadogmodel.J.
Clin.Invest.73:1351-1358.
35.Constam,G. R. 1978. The clinical features of diabetes mellitus.
InClinicalEndocrinology. Alexis Labhart, editor.Springer-Verlag,New
York. 770-810.
36.Seino, S.,A.Funakoshi,Z.Z.Fu,andA.Vinik. 1985. Identifi-cation of insulin variantsinpatientswithhyperinsulinemiaby reversed-phase high-performance liquid chromatography. Diabetes. 34:1-7.
