A New Variant in the Human Kv1.3 Gene Is Associated with Low Insulin Sensitivity and Impaired Glucose Tolerance

Context: The voltage-gated potassium channel Kv1.3 (KCNA3 )i s expressed in a variety of tissues including liver and skeletal muscle. In animal models, knockout of Kv1.3 has been found to improve insulin sensitivity and glucose tolerance. Objective:
of 5
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
   A New Variant in the Human Kv1.3 Gene Is Associated with Low Insulin Sensitivity andImpaired Glucose Tolerance Otto Tschritter, Fausto Machicao, Norbert Stefan, Silke Scha¨fer, Cora Weigert, Harald Staiger,Christian Spieth, Hans-Ulrich Ha¨ring, and Andreas Fritsche  Medizinische Klinik (O.T., F.M., N.S., S.S., C.W., H.S., H.-U.H., A.F.), Abteilung fu¨r Endokrinologie, Stoffwechsel und Pathobiochemie, Eberhard-Karls-Universita¨t, D-72076 Tu¨bingen, Germany; and Center of Bioinformatics (C.S.), Universityof Tu¨bingen, D-72076 Tu¨bingen, Germany Context:  The voltage-gated potassium channel Kv1.3 (  KCNA3 ) isexpressed in a variety of tissues including liver and skeletal muscle.In animal models, knockout of Kv1.3 has been found to improveinsulin sensitivity and glucose tolerance. Objective:  We examined whether mutations in the Kv1.3 gene existin humans and whether they are associated with alterations of glu-cose homeostasis. Design and Setting:  We conducted a genotype-phenotype associa-tion study at a university hospital. ParticipantsandMethods: In50nondiabeticsubjects,wescreenedapproximately4.5kbofchromosome1comprisingthesingleexon,thepromoter/5  -untranslated region, and the 3  -untranslated region of the human Kv1.3 gene for mutations by direct sequencing. Subse-quently, all identified single-nucleotide polymorphisms were ana-lyzed in 552 nondiabetic subjects who underwent an oral glucosetolerance test (OGTT). Of these, 304 had undergone an additionalhyperinsulinemic euglycemic clamp. Main Outcome Measures:  We assessed postprandial blood glucoseduring OGTT and insulin sensitivity measured by hyperinsulinemiceuglycemic clamp .Results:  We identified five single-nucleotide polymorphisms in thepromoter region (T-548C, G-697T, A-845G, T-1645C, and G-2069A)with allelic frequencies of the minor allele of 26, 23, 9, 41, and 16%,respectively. The  1645C allele was associated with higher plasmaglucose concentrations in the 2-h OGTT (  P  0.03) even after adjust-mentforsex,age,andbodymassindex(  P  0.002).Inaddition,itwasassociated with lower insulin sensitivity (  P  0.01, adjusted for sex,age, and body mass index). Functional  in vitro  analysis using EMSA showed differential transcription factor binding to the T-1645Cpolymorphism. Conclusions:  We show that a variant in the promoter of the Kv1.3gene is associated with impaired glucose tolerance and lower insulinsensitivity.Therefore,theKv1.3channelrepresentsacandidategenefor type 2 diabetes.  (  J Clin Endocrinol Metab  91: 654–658, 2006) I NRECENTYEARS,knowledgeontheinsulinsignaltrans-duction cascade has rapidly grown. However, the func-tional role of a variety of signaling molecules downstream of theinsulinreceptorhasnotbeenelucidatedyetindetail.Thevoltage-gated potassium channel Kv1.3, which is expressedin insulin-sensitive tissues such as skeletal muscle, adiposetissue, liver, and brain (1–6) and in olfactory bulb neurons,has been shown to be deactivated by insulin receptor kinasethroughphosphorylationofmultipletyrosineresidues(7–9).Targeted disruption of the Kv1.3 gene in mice resulted inlower body weight, higher insulin sensitivity, and lowerplasma glucose levels (10, 11). On a high-fat diet, these miceshowed lower weight gain than control mice (10). Insulinsensitivity was even higher in Kv1.3 knockout mice com-pared with weight-matched controls (11) and was preservedin diet-induced obesity (10). Moreover, pharmacological in-hibition of Kv1.3 acutely elevated insulin sensitivity in nor-mal and genetically obese (ob/ob and db/db) mice (11).In humans, the role of Kv1.3 in insulin sensitivity, glucosetolerance, and obesity is not yet known. However, geneticvariants affecting the activity of this channel might play ananalogous role in humans as it was shown in the animalmodel. Therefore, we sequenced the promoter/5  -untrans-lated region (UTR) and the coding region of the Kv1.3 genein 50 subjects to identify polymorphisms and to test theirfunctionalrelevanceinalargecohortofnondiabeticsubjects. Subjects and Methods  Subjects The promoter, 5  -UTR, 3  -UTR, and the full length of the codingregion of the Kv1.3 gene were sequenced in 50 unrelated nondiabeticindividuals. Polymorphisms identified in this step were determined bysequencing their locus in the DNA in 552 nondiabetic (fasting glucose,  7 m m;  2-h glucose,  11.1 m m ) unrelated participants in the ongoingTu¨bingen Family Study for type 2 diabetes. Primarily, subjects wererecruited by asking first-degree relatives of type 2 diabetic patients totake part in the study. The study protocol was approved by the EthicalCommitteeoftheUniversityofTu¨bingen,andinformedwrittenconsenthad been obtained before the studies. According to World Health Or-ganizationcriteria(12),478subjectshadnormalglucosetolerance(NGT)(2-hglucose,  7.7m m )and74hadimpairedglucosetolerance(IGT)(2-hglucose,  7.7 and  11.1 m m ). The participants did not take any med- First Published Online November 29, 2005 Abbreviations:BMI,Bodymassindex;FCS,fetalcalfserum;FFA,freefatty acids; IGT, impaired glucose tolerance; NGT, normal glucose tol-erance; OGTT, oral glucose tolerance test; SNP, single-nucleotide poly-morphism; UTR, untranslated region.  JCEM is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the en-docrine community. 0021-972X/06/$15.00/0 The Journal of Clinical Endocrinology & Metabolism 91(2):654–658  Printed in U.S.A.  Copyright © 2006 by The Endocrine Societydoi: 10.1210/jc.2005-0725 654  ication known to affect glucose tolerance, insulin sensitivity, or insulinsecretion.Inasubgroupof304(274NGTand30IGT)subjects,datafroma hyperinsulinemic euglycemic clamp were available. Oral glucose tolerance test (OGTT) Aftera10-hovernightfast,thesubjectsingestedasolutioncontaining75 g dextrose, and venous blood samples were obtained at 0, 30, 60, 90,and 120 min for determination of plasma glucose, plasma insulin, andplasma free fatty acids (FFA). FFA data were available in 514 subjects.  Hyperinsulinemic euglycemic clamp After the baseline period, subjects received a bolus-primed insulininfusionatarateof1.0mU/kg/minfor2haspreviouslydescribed(13).Blood was drawn every 5–10 min for determination of plasma glucose,and the infusion rate of exogenous glucose was adjusted appropriatelyto maintain the baseline glucose level. Plasma insulin levels were mea-sured at baseline and in the steady state of the clamp.  Body composition and body fat distribution Bodycompositionwasmeasuredbybioelectricalimpedanceanalysisand expressed as percent body fat. Body mass index (BMI) was calcu-lated as weight divided by the square of height (kg/m 2 ). Waist and hipcircumferences were measured in the supine position, and waist-to-hipratio was calculated as an index of body fat distribution.  Analytical procedures Plasma glucose was determined using a bedside glucose analyzer(glucose-oxidase method; Yellow Springs Instruments, Yellow Springs,CO). Plasma insulin was determined by microparticle enzyme immu-noassay (Abbott Laboratories, Tokyo, Japan) and serum FFA concen-trations with an enzymatic method (WAKO Chemicals, Neuss,Germany).  DNA analyses Genomic DNA was isolated from whole blood with a commercialDNA isolation kit (Nucleospin; Macherey-Nagel, Du¨ren, Germany). Toamplify the coding region and the promoter of Kv1.3, oligonucleotideprimers were designed that revealed amplification products of approx-imately 230–600 bp. By 10 overlapping reactions, a region comprisingapproximately 4.5 kb from the promoter/5  -UTR, the single exon andthe complete 3  -UTR were amplified by PCR (chromosome 1, position110927146–110931517). The PCR products generated from these primersets were investigated by direct sequencing. PCR products were se-quenced bidirectionally to avoid sequencing artifacts with an ABI Prismdyeterminatorcyclesequencingreadyreactionkit(AppliedBiosystems,Foster City, CA) and analyzed on an automated sequencer (ABI model310). Genotyping of the newly detected single-nucleotide polymor-phisms (SNPs) in 552 DNA samples was done using the TaqMan assay(Applied Biosystems). The TaqMan genotyping reaction was amplifiedon a GeneAmp PCR system 7000, and fluorescence was detected on anABI PRISM 7000 sequence detector (Applied Biosystems).  EMSA Primary human skeletal muscle cells were obtained from needle biopsies of the vastus lateralis muscle, grown and differentiated aspreviously described (14). Nuclear extracts of C2C12 cells and humanmyotubes were prepared as described previously (15). Before the prep-aration of nuclear extracts, human myotubes were held in four differentconditions: fasted, 2 h stimulation with 20 n m  insulin, 2 h stimulationwith 20 n m  IGF-I, and 2 h stimulation with 20% fetal calf serum (FCS).C2C12 cells were held fasted, with 2-h 100 n m  insulin stimulation, andwith2-h10%FCSstimulation.Syntheticoligonucleotidescontainingthehuman  KCNA3  promoter sequence   1658 to   1627 were end labeledwith [  - 32 P]dATP (3000 Ci/mmol) and Klenow enzyme and were in-cubated with 8   g nuclear protein in 20   l 7 m m  HEPES-KOH (pH 7.9),100 m m  KCl, 3.6 m m  MgCl 2 , and 10% glycerol on ice for 20 min, and 0.05mg/ml poly[d(I-C)] was added as nonspecific competitor. The sampleswere run on a 5% nondenaturing polyacrylamide gel in a buffer con-taining 25 m m  Tris-HCl (pH 8.0), 190 m m  glycine, and 1 m m  EDTA. Gelswere dried and analyzed by autoradiography. Calculations The insulin sensitivity index (  mol  kg  1  min  1  p m  1 ) for systemicglucose uptake was calculated as mean infusion rate of exogenous glu-cose necessary to maintain euglycemia during the last 60 min of theeuglycemic clamp divided by the steady-state insulin concentration.First-andsecond-phaseinsulinsecretionwasestimatedfromOGTTdatausing indexes described by Stumvoll  et al.  (16): first phase    1283   1.829  Ins 30  138.7  Gluc 30  3.772  Ins 0 ; second phase  287  0.4164  Ins 30  26.07  Gluc 30  0.9226  Ins 0 .Haplotype analyses were performed using the THESIAS program(17). Haplotype effects were tested for all haplotypes with a haplotypefrequency greater than 5% in an additive model and are shown asdifference from the most common haplotype.  Statistical analyses Unless otherwise stated, data are given as mean    sem . Statisticalcomparison of normally distributed parameters between two groupswas performed using Student’s  t  test. Distribution was tested for nor-mality using Shapiro-Wilk W test. For all analyses, nonnormally dis-tributed parameters were logarithmically transformed to approximate anormal distribution. To adjust the effects of covariates and identifyindependent relationships, multivariate linear regression analyses wereperformed. The phenotype was treated as dependent variable, whereasthe genotype was treated as a nominal/independent variable. Compar-isonsoftwonominalparametersweredoneinacontingencytableusingthe    2 test on likelihood ratios. A  P  value of   0.05 was considered to bestatistically significant. The statistical software package JMP (SAS In-stitute Inc, Cary, NC) was used. In case of low allelic frequencies, het-erozygous and homozygous carriers of the rare allele were combinedassuming a dominant model. Otherwise recessive, dominant, and ad-ditive models were considered. Results Genetic variants in the Kv1.3 gene IntheSNPdatabase(http://www.ncbi.nlm.nih.gov/SNP/),two SNPs in the exon, six SNPs in the promoter/5  -UTR, andseven SNPs in or near the 3  -UTR were reported in the humanKv1.3 gene. Three of the SNPs near the 3  -UTR are locateddownstreamofthepoly-Asignalandwerenotanalyzedinthisstudy. Because the transcription start is unknown so far, thepromoter/5  -UTR is referred to as promoter and the positionsof all SNPs are given as nucleotide count from the ATG start.In the 50 subjects in whom we sequenced the whole codingregion, 3  -UTR, and the promoter, we detected five SNPs thatwere all located in the promoter. The other seven SNPs (onepromoter SNP, both exon SNPs, and four 3  -UTR SNPs) werenot found in these individuals and therefore seem to be rare(allelicfrequency  5%).Accordingly,intheSNPdatabase,theallelic frequencies were approximately 1%, if quoted.The promoter SNPs, which we found in the 50 subjects in-vestigatedfirst,weredeterminedinacohortof552nondiabeticsubjects. The polymorphisms that we found were T-548C(rs2840381; allelic frequency of the minor allele, 26%), G-697T(rs2821555, 23%), A-845G (rs7528937, 9%), T-1645C (rs2821557,41%), and G-2069A (rs3762379, 16%). These polymorphismswere all in Hardy-Weinberg equilibrium and in linkage dis-equilibrium (D     0.51;  P    0.002). Determination of the ge-notype failed in two subjects at position  845, in one subject atposition  2069, and in 25 subjects in the positions  548 and  697.Insubsequentanalysesaddressingthesepolymorphisms Tschritter  et al.  • Kv1.3 in Glucose Metabolism J Clin Endocrinol Metab, February 2006, 91(2):654–658  655  andinthehaplotypeanalysis,thecorrespondingsubjectswereexcluded. Gene effects Allmeasuresforobesity(BMI,waist-to-hipratio,bodyfat)did not differ between the genotype groups (Table 1 andsupplemental Table 1, published as supplemental data onThe Endocrine Society’s Journals Online web site at http:// jcem.endojournals.org).Four of the polymorphisms (T-548C, G-697T, A-845G, andG-2069A) showed no effect on relevant metabolic parameters.There was a minor association of the T-548C with age andfasting plasma glucose (supplemental Table 1). However, afteradjustment for the age difference, the association with glucoselevelsdisappeared.TheA-845Gpolymorphismwasassociatedwithlowerinsulinlevelsat120minintheOGTT(supplementalTable 1). This difference was still significant in a multivariateregression analysis with sex, BMI, and age as covariates ( P  0.02). However, insulin sensitivity was not significantly higherin this group, and there was no difference in glucose tolerance,fasting glucose, and fasting insulin.The C allele at position  1645 was associated with lowerinsulinsensitivityinthehyperinsulinemiceuglycemicclampand a higher glucose level at 120 min in the OGTT (Table 1).Higher glucose levels at 120 min were also evident in theclamp subgroup [wild type (TT), 5.55  0.17; heterozygouscarriers of the polymorphism (TC), 5.46  0.11; homozygouscarriersoftheCallele(CC),6.27  0.22; P  0.003(TX vs. CC, P    0.0007)]. These effects were detectable in the homozy-gous carriers of the C allele (CC), whereas heterozygouscarriers of the polymorphism (TC) did not differ from wild-type individuals (TT). Moreover, in multivariate regressionanalysis, both glucose tolerance and insulin sensitivity werereduced in subjects with the CC genotype independent fromsex, BMI, and age ( P    0.002 and  P    0.01, respectively).Moreover, in the CC genotype group, we found significantlymore subjects with impaired glucose tolerance (Table 1).None of the polymorphisms were associated with insulinsecretion(Table1andsupplementalTable1).Byexclusionof IGT subjects, the results on insulin secretion were not af-fected (data not shown). However, the effect of the T-1645Con insulin sensitivity was no longer significant ( P  0.25).  Haplotype analysis Of 32 possible haplotypes, 14 were detected in our studypopulation. Only four haplotypes (TGATG, CTACG,TGACA, and TGGTG) had a haplotype frequency greaterthan 5% (Table 2). The most common haplotype (TGATG),which served as reference haplotype in our analyses, in-cluded 83% of the T alleles in position  1645. The majorityof C alleles (95%) in position   1645 were spread to twocommon haplotypes (CTACG, 57%; TGACA, 38%). None of the haplotypes with a haplotype frequency greater than 5%was associated with obesity, insulin sensitivity, or glucosetolerance (Table 2).  Alteration of the DNA sequence by the T-1645C polymorphism and estimation of transcription factorbinding by EMSAs Interestingly, the wild-type promoter contains a CCAAT box at position  1645. By the replacement of T by C in themutant promoter, CCAAT is changed to CCAAC. To eval-uate the putative functional relevance of the loss of theCCAATboxbythisSNP,wedeterminedwhethertheCalleleat position  1645 affected binding of nuclear factors to thispromoter region. EMSAs were performed with nuclear ex-tracts of C2C12 cells as well as human myotubes and thecomplementary oligonucleotides 5  -AGAGTAGGTC-CTAGCCAAT/CTTATATTTCTAGC-3  containing a T or Cat position   1645 (changed bases are  underlined ). Nuclear TABLE 1.  Effects of the T-1645C polymorphism Position (from ATG),  1645  P  (ANOVA)  P  (TX   vs.  CC)TT (n  193) TC (n  268) CC (n  91) NGT/IGT (n/n) 164/29(85%/15%)243/25(91%/9%)71/20(78%/22%)0.008 a 0.010 a Sex (M/F) 84/109 109/159 39/52 0.83 1.00BMI (kg/m 2 ) 26.4  0.4 26.7  0.4 26.9  0.7 0.85 0.74Body fat (%) 27.8  0.8 27.5  0.6 27.6  1.1 0.99 0.98 Age (yr) 38.2  0.9 36.8  0.8 36.4  1.4 0.21 0.33Waist-to-hip ratio 0.86  0.006 0.86  0.005 0.86  0.01 0.83 0.66Fasting plasma glucose (mmol/liter) 5.00  0.04 4.94  0.04 4.94  0.05 0.51 0.74Plasma glucose 120 min (mmol/liter, OGTT) 5.81  0.12 5.72  0.09 6.26  0.18 0.030 0.008Fasting plasma insulin (pmol/liter) 54  3 52  3 53  3 0.33 0.30Plasma insulin 120 min (pmol/liter, OGTT) 351  26 309  16 367  33 0.25 0.12Fasting plasma FFA (  mol/liter) b 505  17 522  15 526  27 0.53 0.59Plasma FFA 120 min (  mol/liter, OGTT) b 73  4 73  5 73  5 0.57 0.49Insulin sensitivity (  mol  kg   1  min  1  p M  1 , clamp) c 0.099  0.005 0.102  0.005 0.084  0.007 0.057 0.017First-phase insulin secretion (p M ) d 1089  42 1108  41 1155  64 0.35 0.22Second-phase insulin secretion (p M ) d 291  10 294  9 306  15 0.49 0.24Results are shown as means  SEM . TT, Wild-type individuals; TC, heterozygous carriers of the polymorphism; CC, homozygous carriers of the C allele. a   2 test. b N  514. c n  304 (92 TT, 153 TC, and 59 CC). d Estimated from OGTT. 656  J Clin Endocrinol Metab, February 2006, 91(2):654–658 Tschritter  et al.  • Kv1.3 in Glucose Metabolism  extracts of both human myotubes and C2C12 cells wereobtained from fasted cells and after stimulation with insulin,IGF-I, or FCS. EMSA revealed specific binding of nuclearproteins to the oligonucleotide containing the C allele asverified by competition with 30-fold molar excess of thecorresponding unlabeled oligonucleotide. Such a complexwas not found with the oligonucleotide containing the Tallele. This suggests a qualitative difference in the binding of transcription factors to this promoter sequence. This specific binding was reproducible in both cell types after stimulationwith either insulin or FCS. Figure 1 shows a representativegel. Discussion In this study, we were investigating the role of the humanKv1.3geneinglucosemetabolism.Wedescribefivecommonpolymorphisms in the promoter of the human Kv1.3 gene.Although in linkage disequilibrium, these polymorphismswere different with regard to their metabolic effects. TheG-2069A and the G-697T polymorphism were not associatedwith any metabolic parameter involved in the pathogenesisof type 2 diabetes. In univariate analysis, the T-548C poly-morphism was accompanied by reduced fasting glucose lev-els. However, after adjustment for the observed age differ-ence, there was no significant effect. The association of theA-845G polymorphism with reduced insulin levels after anoral glucose load was not accompanied by significantchanges in insulin sensitivity, insulin secretion, and glucosetolerance. Moreover, a confounding role of Kv1.3 in insulinsecretion seems to be unlikely, because expression of thisgene has not been found in human pancreatic islets (18).Taken together, we do not think that these four polymor-phisms (T-548C, G-697T, A-845G, and G-2069A) have a rel-evant impact on diabetes risk.Our main finding was that the T-1645C polymorphism isassociated with reduced insulin sensitivity and impairedglucose tolerance. These effects occurred only in homozy-gous carriers of the  1645C allele and were not attributableto a distinct haplotype. However, because in the haplotypeanalysis two common haplotypes were found in the rareallele group, the failure to detect a haplotype effect mightalso be a consequence of reduced statistical power.To provide evidence for alteration of transcriptional activity by the T-1645C polymorphism, we analyzed the functionaleffect of the T/C exchange. The wild-type promoter contains aCCAATboxatthispositionthatislostandchangedtoCCAACin the mutant promoter. The CCAAT box is a very common bindingsiteformanytranscriptionfactorslikeC/EBPs,CUTL1,and nuclear factor Y. Therefore, the C allele might be accom-panied by a reduced affinity to such factors or enable bindingof other transcription factors. The EMSA revealed a specifictranscription factor binding to the synthetic oligonucleotide TABLE 2.  Haplotype analysis Haplotype Frequency Effect BMI a  P  ISI b  P  Glucose 120 c  P TGATG 0.498 Intercept 13.10  0.21 0.024  0.003 2.62  0.06CTACG 0.229 Difference 0.02  0.49 0.59   0.006  0.007 0.08 0.14  0.14 0.40TGACA 0.155 Difference   0.25  0.54 0.23 0.006  0.008 0.97 0.14  0.13 0.65TGGTG 0.076 Difference   0.71  0.77 0.79 0.006  0.011 0.07 0.11  0.17 0.63CGGTG 0.011 Not estimatedCGACG 0.007 Not estimatedTGACG 0.006 Not estimatedCGATG 0.006 Not estimatedTGGCA 0.004 Not estimatedTTACG 0.003 Not estimatedTGATA 0.002 Not estimatedCGACA 0.001 Not estimatedCTATG 0.001 Not estimatedTTATG   0.001 Not estimatedThe order of polymorphisms in the haplotype designation follows their distance to the ATG start, beginning with the nearest (T-548C). Inthe analysis, each haplotype accounts for half the phenotype. The most frequent haplotype serves as a reference. Effects of other haplotypesare shown as difference to the reference haplotype. Results are means   SEM . ISI, Insulin sensitivity index. a  Adjusted for sex and age. b n  304, adjusted for sex, age, and BMI. c  Adjusted for sex, age, and BMI.F IG . 1. Nuclear proteins of C2C12 cells were incubated with 50,000cpm of radiolabeled 5  -AGAGTAGGTCCTAGCCAA  C TTATATT-TCTAGC-3   (C, lanes 1–3) or 5  -AGAGTAGGTCCTAGCCAA  T  T-TATATTTCTAGC-3  (T, lanes 4–6) in the presence of 30-fold molarexcess of unlabeled C (lanes 2 and 5) or of unlabeled T (lanes 3 and6). Specific protein-DNA complexes are indicated by the  arrow  on the left . Similar results were obtained with nuclear extracts of humanmyotubes. Tschritter  et al.  • Kv1.3 in Glucose Metabolism J Clin Endocrinol Metab, February 2006, 91(2):654–658  657  containing the C allele at the corresponding position. This sug-gests that this polymorphism influences transcription of theKv1.3 gene and therefore alters the function of the channel.Mice with Kv1.3 deficiency have higher insulin sensitivityand lower plasma glucose levels (10, 11). Additionally, phar-macological inhibition of Kv1.3 by margatoxin (a selectivelyacting scorpion toxin) lowered plasma glucose levels and im-provedinsulinsensitivitywithin2h(11).Thisindicatesthatthefunction of the voltage-gated potassium channel Kv1.3 is di-rectly involved in acute changes of insulin sensitivity and thatimproved insulin sensitivity in Kv1.3 KO mice is not only aconsequence of lower body weight and reduced fat mass.Theexactmechanismbywhichthefunctionofvoltage-gatedpotassium channels influences glucose metabolism is not clearyet. Because Kv1.3 is abundantly expressed in liver, skeletalmuscle, adipose tissue, and brain, there is a great variety of possible direct and indirect mechanisms being involved inmodulation of insulin sensitivity by this channel. It has beenpreviously shown that Kv1.3 is inactivated by insulin-depen-dent phosphorylation on tyrosine residues resulting in a re-duction of potassium current (7–9, 19), which causes depolar-ization of the cell membrane and may mediate insulin effects.Both Kv1.3 deficiency and margatoxin treatment enhancedtranslocation of glucose transporter 4 to the plasma membranesimilarly to insulin (11). This might explain insulin-sensitizingeffects of Kv1.3 deficiency. Vice versa, overexpression of Kv1.3might enhance the number of active Kv1.3 channels at the cellmembrane and, therefore, deteriorate glucose transporter 4translocation under insulin stimulation. This could lead to areduction of insulin-stimulated glucose uptake. Therefore, apromoter polymorphism leading to alteration of transcriptionfactorbindinghasthepotentialtomodulateinsulin-stimulatedglucose uptake.In addition to the metabolic consequences of Kv1.3 disrup-tion, Kv1.3 knockout mice showed a super-smeller phenotype,characterized by a markedly reduced threshold for perceptionof odors as well as an increased ability to discriminate similarodors (20). It would be worthwhile to know whether the met-abolic effects depend on the olfactory effects and whether inhumans Kv1.3 also plays an important role in the olfactorysense.Onewouldexpectthatagain-of-functionmutationintheKv1.3 gene could deteriorate the ability to detect and discrim-inate odors. Such effects might influence eating behavior andenergy homeostasis. Data on olfactory sense is not available inour study population. However, the polymorphism had noeffect on body weight and BMI.Inconclusion,wefoundfivecommonpolymorphismsinthepromoter of the human Kv1.3 gene. One of these polymor-phisms was associated with reduced insulin sensitivity andimpaired glucose tolerance. Although the exact mechanism bywhich voltage-gated potassium channels control glucose me-tabolism is not yet clear, we could detect in humans a pheno-type with reduced insulin sensitivity and impaired glucosetolerance as would be expected from animal data. Therefore,this polymorphism is likely to be involved in the pathogenesisof type 2 diabetes.  Acknowledgments We thank all the research volunteers for their participation. We gratefullyacknowledge the superb technical assistance of Katharina Kienzle, AnnaTeigeler, Heike Luz, Alke Guirguis, Claudia Peterfi, and Melanie Weisser.Received April 4, 2005. Accepted November 21, 2005.Address all correspondence and requests for reprints to: Dr. AndreasFritsche, Medizinische Universita¨tsklinik, Otfried-Mu¨ller-Strasse 10, D-72076Tu¨bingen, Germany. E-mail: andreas.fritsche@med.uni-tuebingen.de.This study was in part supported by a grant of the Deutsche For-schungsgemeinschaft DFG (KFO 114). References 1.  Jacob A, Hurley IR, Goodwin LO, Cooper GW, Benoff S  2000 Molecularcharacterization of a voltage-gated potassium channel expressed in rat testis.Mol Hum Reprod 6:303–3132.  Chandy KG, Williams CB, Spencer RH, Aguilar BA, Ghanshani S, TempelBL, Gutman GA  1990 A family of three mouse potassium channel genes withintronless coding regions. Science 247:973–9753.  Mourre C, Chernova MN, Martin-Eauclaire MF, Bessone R, Jacquet G, Gola M,AlperSL,CrestM 1999Distributioninratbrainofbindingsitesofkaliotoxin,ablockerof Kv1.1 and Kv1.3   -subunits. J Pharmacol Exp Ther 291:943–9524.  StuhmerW,RuppersbergJP,SchroterKH,SakmannB,StockerM,GieseKP,PerschkeA,BaumannA,PongsO 1989Molecularbasisoffunctionaldiversityofvoltage-gatedpotassiumchannelsinmammalianbrain.EMBOJ8:3235–32445.  Grande M, Suarez E, Vicente R, Canto C, Coma M, Tamkun MM, ZorzanoA, Guma A, Felipe A  2003 Voltage-dependent K  channel   -subunits inmuscle:differentialregulationduringpostnataldevelopmentandmyogenesis. J Cell Physiol 195:187–1936.  Yao X, Chang AY, Boulpaep EL, Segal AS, Desir GV  1996 Molecular cloningof a glibenclamide-sensitive, voltage-gated potassium channel expressed inrabbit kidney. J Clin Invest 97:2525–25337.  Fadool DA, Tucker K, Phillips JJ, Simmen JA  2000 Brain insulin receptorcauses activity-dependent current suppression in the olfactory bulb throughmultiple phosphorylation of Kv1.3. J Neurophysiol 83:2332–23488.  Fadool DA, Levitan IB  1998 Modulation of olfactory bulb neuron potassiumcurrent by tyrosine phosphorylation. J Neurosci 18:6126–61379.  Fadool DA, Holmes TC, Berman K, Dagan D, Levitan IB  1997 Tyrosinephosphorylation modulates current amplitude and kinetics of a neuronalvoltage-gated potassium channel. J Neurophysiol 78:1563–157310.  Xu J, Koni PA, Wang P, Li G, Kaczmarek L, Wu Y, Li Y, Flavell RA, DesirGV  2003 The voltage-gated potassium channel Kv1.3 regulates energy ho-meostasis and body weight. Hum Mol Genet 12:551–55911.  Xu J, Wang P, Li Y, Li G, Kaczmarek LK, Wu Y, Koni PA, Flavell RA, DesirGV  2004 The voltage-gated potassium channel Kv1.3 regulates peripheralinsulin sensitivity. Proc Natl Acad Sci USA 101:3112–311712.  World Health Organization Expert Committee  1980 Second report on dia- betes mellitus. Geneva: World Health Organization; Tech Rep Ser 646:A-8013.  Koch M, Rett K, Volk A, Maerker E, Haist K, Deninger M, Renn W, HaringHU  1999 Amino acid polymorphism Gly 972 Arg in IRS-1 is not associated tolower clamp-derived insulin sensitivity in young healthy first degree relativesof patients with type 2 diabetes. Exp Clin Endocrinol Diabetes 107:318–32214.  Krutzfeldt J, Kausch C, Volk A, Klein HH, Rett K, Haring HU, Stumvoll M 2000 Insulin signaling and action in cultured skeletal muscle cells from leanhealthy humans with high and low insulin sensitivity. Diabetes 49:992–99815.  Weigert C, Brodbeck K, Sawadogo M, Haring HU, Schleicher ED  2004Upstream stimulatory factor (USF) proteins induce human TGF-  1 gene ac-tivation via the glucose-response element-1013/-1002 in mesangial cells: up-regulation of USF activity by the hexosamine biosynthetic pathway. J BiolChem 279:15908–1591516.  Stumvoll M, Mitrakou A, Pimenta W, Jenssen T, Yki-Jarvinen H, VanHaeften T, Renn W, Gerich J  2000 Use of the oral glucose tolerance test toassess insulin release and insulin sensitivity. Diabetes Care 23:295–30117.  Tregouet DA, Escolano S, Tiret L, Mallet A, Golmard JL  2004 A new algo-rithm for haplotype-based association analysis: the Stochastic-EM algorithm.Ann Hum Genet 68:165–17718.  Yan L, Figueroa DJ, Austin CP, Liu Y, Bugianesi RM, Slaughter RS, Kac-zorowski GJ, Kohler MG  2004 Expression of voltage-gated potassium chan-nels in human and rhesus pancreatic islets. Diabetes 53:597–60719.  BowlbyMR,FadoolDA,HolmesTC,LevitanIB 1997ModulationoftheKv1.3potassium channel by receptor tyrosine kinases. J Gen Physiol 110:601–61020.  Fadool DA, Tucker K, Perkins R, Fasciani G, Thompson RN, Parsons AD,Overton JM, Koni PA, Flavell RA, Kaczmarek LK  2004 Kv1.3 channel gene-targeted deletion produces “Super-Smeller Mice” with altered glomeruli, in-teracting scaffolding proteins, and biophysics. Neuron 41:389–404  JCEM is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving theendocrine community. 658  J Clin Endocrinol Metab, February 2006, 91(2):654–658 Tschritter  et al.  • Kv1.3 in Glucose Metabolism
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks