Mechanisms of Glucose Lowering ofDipeptidyl Peptidase-4 InhibitorSitagliptinWhen Used Alone orWithMetformin in Type 2 DiabetesA double-tracer study

CAROLINA SOLIS-HERRERA, MDCURTIS TRIPLITT, PHARMDJOSE DE JESÚS GARDUNO-GARCIA, MD

JOHN ADAMS, BSRALPH A. DEFRONZO, MDEUGENIO CERSOSIMO, MD, PHD

OBJECTIVEdTo assess glucose-lowering mechanisms of sitagliptin (S), metformin (M), andthe two combined (M+S).

RESEARCH DESIGN AND METHODSdWe randomized 16 patients with type 2 di-abetes mellitus (T2DM) to four 6-week treatments with placebo (P), M, S, and M+S. After eachperiod, subjects received a 6-h meal tolerance test (MTT) with [14C]glucose to calculate glucosekinetics. Fasting plasma glucose (FPG), fasting plasma insulin, C-peptide (insulin secretory rate[ISR]), fasting plasma glucagon, and bioactive glucagon-like peptide (GLP-1) and gastrointestinalinsulinotropic peptide (GIP) were measured.

RESULTSdFPG decreased from P, 1606 4 toM, 1506 4; S, 1546 4; andM+S, 1256 3mg/dL.Mean post-MTT plasma glucose decreased from P, 207 6 5 to M, 191 6 4; S, 1956 4; and M+S,1616 3mg/dL (P, 0.01). The increase in mean post-MTT plasma insulin and in ISR was similar inP,M, and S and slightly greater in M+S. Fasting plasma glucagon was equal (;65–75 pg/mL) withall treatments, but there was a significant drop during the initial 120 min with S 24% and M+S34% (both P, 0.05) vs. P 17% andM 16%. Fasting andmean post-MTT plasma bioactive GLP-1were higher (P, 0.01) after S andM+S vs. M and P. Basal endogenous glucose production (EGP)fell from P 2.06 0.1 to S 1.86 0.1 mg/kg zmin, M 1.86 0.2 mg/kg zmin (both P, 0.05 vs. P),and M+S 1.56 0.1 mg/kg zmin (P, 0.01 vs. P). Although the EGP slope of decline was faster inM and M+S vs. S, all had comparable greater post-MTT EGP inhibition vs. P (P , 0.05).

CONCLUSIONSdM+S combined produce additive effects to 1) reduce FPG and postmealplasma glucose, 2) augment GLP-1 secretion and b-cell function, 3) decrease plasma glucagon,and 4) inhibit fasting and postmeal EGP compared with M or S monotherapy.

Diabetes Care 36:2756–2762, 2013

In diet-treated patients with type 2diabetes mellitus (T2DM) and HbA1cof ;8.0%, sitagliptin (S) reducesHbA1c by 0.6–0.7% over a 6-month pe-riod (1). A slightly greater HbA1c decline(;0.8–0.9%) is observed when metfor-min (M) therapy is added to S (2). Dipe-ptidyl peptidase (DPP)-4 inhibitorspredominantly affect the postprandialplasma glucose excursion, but a significant,albeit modest, reduction in fasting plasma

glucose (FPG) also is observed (1–3). Themechanism of action of the DPP-4 inhib-itors has been well studied and includesincreased plasma glucagon-like peptide(GLP)-1 and gastrointestinal insulino-tropic peptide (GIP) levels, resulting inincreased insulin and reduced glucagonsecretion (4–6). The increase in plasmainsulin and the decline in glucagon in-hibit basal endogenous glucose produc-tion (EGP) and enhance the suppression

of EGP without affecting splanchnic (he-patic) glucose uptake or gastric emptying(6,7). Therapy with S and M combined(M+S) exerts an additive effect to reduceHgA1c; themechanismof action of this com-bination has yet to be examined. Severalstudies have demonstrated that M inhibitsDPP-4 activity, thus increasingplasma activeGLP-1 levels (8–10). There are also reportsindicating that the decline in plasma glucosewith M therapy can restore the b-cells’ sen-sitivity to the stimulatory effect of incretinson insulin secretion (11,12). Despite thesereports, we still do not fully understand thereasonswhy the use ofM in diabetic patientsis not accompanied by changes in insulinrelease. In the current study, we used thedouble-tracer technique (7) to examine themechanism(s) via which M+S and eachagent alone reduce the fasting and postmealplasma glucose concentration in T2DM.

RESEARCH DESIGN ANDMETHODSdSixteen patients withT2DM (9 male and 7 female; mean age47 6 3 years; body weight, 92.1 6 4.2kg; BMI, 33.5 6 2.0 kg/m2; A1C, 8.8 61.2%; FPG, 165 6 12 mg/dL; and diabe-tes duration, 1.5 6 0.5 years) were stud-ied. All subjects were in good health asdetermined by medical history, physicalexam, routine laboratory analysis, andelectrocardiogram. T2DM was controlledby diet (n = 7) or with a stable dose ($6months) of sulfonylurea (n = 6) or M (n =3). Three patients were on a “statin” stabledose. All studies were carried out at theClinical Research Center at the Texas Di-abetes Institute. The study was approvedby the Institutional Review Board of theUniversity of Texas Health Science Centerat San Antonio (UTHSCSA), and in-formed written consent was obtainedfrom each patient before participation.

The study used a double-blind pla-cebo (P)-controlled crossover design inwhich each subject randomly underwentfour 6-week treatment periods with Salone, M alone, M+S, and P with a 2-week

c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c

From the University of Texas Health Science Center, Texas Diabetes Institute, San Antonio, Texas.Corresponding author: Eugenio Cersosimo, cersosimo@uthscsa.edu.Received 10 October 2012 and accepted 13 February 2013.DOI: 10.2337/dc12-2072. Clinical trial reg. no. NCT00820573, clinicaltrials.gov.A slide set summarizing this article is available online.© 2013 by the American Diabetes Association. Readers may use this article as long as the work is properly

cited, the use is educational and not for profit, and thework is not altered. See http://creativecommons.org/licenses/by-nc-nd/3.0/ for details.

2756 DIABETES CARE, VOLUME 36, SEPTEMBER 2013 care.diabetesjournals.org

P a t h o p h y s i o l o g y / C o m p l i c a t i o n sO R I G I N A L A R T I C L E

mailto:cersosimo@uthscsa.eduhttp://creativecommons.org/licenses/by-nc-nd/3.0/http://creativecommons.org/licenses/by-nc-nd/3.0/

washout period between each treatmentperiod. At screening, all subjects metwith a dietitian and were instructed toconsume a weight-maintaining diet (55%carbohydrate, 20% protein, and 25% fat).Patients who were taking antidiabetesdrugs were asked to discontinue the med-ication for 6 weeks (washout prestudyperiod) and were monitored weekly todocument that the FPG changed by,10% from week 4 to 6 prior to studyinitiation. Eligible subjects then receivedP medication for 2 weeks to documentcompliance, after which they were ran-domized for “treatment sequence”using a computer-generated schedule toreceive P, M (1,000 mg b.i.d.) plus P, S(100 mg) plus P, or M+S (Fig. 1).

Meal tolerance test with double tracerMeal tolerance test (MTT) was performedafter the completion of each 6-weektreatment period for each allocation treat-ment. After a 10-h overnight fast, subjectswere admitted to the Clinical ResearchCenter at 0700 h and catheters wereplaced in each antecubital vein for bloodwithdrawal and infusion of test substan-ces. At 0730 h, a prime (25 mCi 3 FPG/100) continuous (0.25 mCi/min) infusionof [3-3H]glucose was started and contin-ued until study end at 1630 h. At 1030 h,patients ingested a standardized meal(one hard-boiled egg, 2 ounces of Munstercheese, and 75 g glucose containing100 mCi of 1-14C-glucose in 300 mLorange-flavored water). At 1000 h, bloodsamples were obtained every 10–15 minfor determination of plasma [14C]glucoseand [3H]glucose radioactivity and plasmaglucose, insulin, C-peptide, and bioactiveGLP-1/GIP concentrations. Urine was col-lected prior to themeal ingestion and from1030 to 1630 h for determination of glu-cose concentration. After completion ofeach MTT, subjects took no medicationfor 2 weeks (washout period).

Analytical determinationsPlasma glucose concentration was deter-mined by glucose oxidase method (AnaloxGlucose Analyzer; Analox Instruments,Lunenburg,MA). Plasma insulin, C-peptide,and glucagon concentrations were de-termined by radioimmunoassay (Diagnos-tic Products, Los Angeles, CA). Plasma [3H]glucose and [14C]glucose radioactivitieswere determined on barium hydroxide/zincsulfate–precipitated plasma extracts as pre-viously described (6,7). Plasma bioactiveGLP-1 concentration was measured byELISAusingNH2-terminal–specific antibody

(BD Diagnostics, St. Charles, MO) andplasma intact GIP by radioimmunoassaywith an antibody specific for the NH2-terminus (BD Diagnostics).

CalculationsThe basal rate of EGP was calculatedas the [3-3H]glucose infusion (dpm/min)divided by the steady-state plasma[3-3H]glucose specific activity (dpm/mg). After glucose ingestion, non–steadystate conditions prevail and total Ra (RaT)and Rd from the systemic circulation werecomputed from the [3-3H]glucose datausing the Steele equation as previouslyreported (7). Total tissue glucose disposalwas calculated by Rd (integrated over thetime period 0–360 min) minus the mea-sured urinary glucose loss over the sametime period. The [1-14C]glucose datawere used to calculate the Ra of oral glu-cose (RaO), and after the meal EGP wascalculated as the difference between RaTand RaO (7). Splanchnic (hepatic) glucoseuptake was calculated as follows: 75 g 2RaO from 0 to 360 min. Insulin secretoryrate (ISR) was calculated from deconvo-lution of plasma C-peptide curve (13) us-ing ISEC (14). Indices of insulin secretionwere calculated as DI/DG and DISR/DGand of b-cell function as DI/DG 3 MIand DISR/DG3MI, where MI is Matsudaindex of insulin sensitivity: a compositeindex of hepatic insulin sensitivity cal-culated using baseline and changes inplasma insulin and glucose concentra-tions as previously described (15).

Statistical methodsAll 16 patients completed all four treat-ment periods and were included in theanalysis. ANOVA with fixed terms forperiod and treatment and a random termfor subject was used to compare treat-ment groups. A general unstructuredvariance-covariance matrix was appliedfor measurements at different periods.Between-group comparisons after 6weeks of treatment were assessed usingthe ANOVAmodel at a = 0.05 (two sided)and an appropriate contrast statement.Power calculations indicated that with16 patients, the study would provide;90% power to detect a difference inEGP of 0.28 mg z kg21 z min21 betweentreatment groups with a half-width of the95% CI of 0.17 mg z kg21 z min21. Thiscalculation used a within-subject SD esti-mate of 0.24 mg z kg21 z min21 with a =0.05 and a two-sided test. The treatmenteffect of 0.28 mg z kg21 z min21 was thedifference observed between single-dose

vildagliptin and P in a two-period cross-over study (6). Correlation coefficients andmultiple regression analyses were derivedand used as a measure of association.

RESULTSdAll subjects tolerated theantidiabetes medications well, and therewere no adverse events. Body weight(92.1 6 4.2 vs. 91.8 6 3.8 kg) did notchange, and mean HbA1c decreased to7.8 6 1.0% at study end. FPG (Fig. 2A)decreased from 160 6 4 mg/dL (P) to150 6 4 mg/dL (M), to 154 6 4 (S)(both P , 0.05 vs. P), and to 125 6 3mg/dL (M+S) (P , 0.01 vs. M and S).Mean post-MTT plasma glucose de-creased from 207 6 5 mg/dL (P) to191 6 4 mg/dL (M), to 195 6 4 mg/dL(S) (both P, 0.01 vs. P), and to 1616 3mg/dL (M+S) (P , 0.01 vs. M and S).

Fasting plasma insulin did not changesignificantly with any treatment (Fig. 2B).Mean post-MTT insulin concentrationafter M (55 6 2 mU/mL) and S (53 62 mU/mL) were similar to P (50 6 3 L)but increased to 646 4 mU/mL after M+S(P, 0.05 vs. P,M, and S). Fasting ISRwasequivalent with all treatments (Fig. 2C).Mean post-MTT ISRs were similar after M(7.3 6 0.2 pmol/kg z min), S (7.1 60.3 pmol/kg z min), and P (6.8 60.4 pmol/kg z min) but increased signifi-cantly to 7.8 6 0.4 pmol/kg z min afterM+S (P , 0.05 vs. P, M, and S). Fastingplasma glucagon was comparable with alltreatments (Fig. 2D). During the initial120 min of the MTT, mean plasma gluca-gon concentration after S (716 5 pg/mL)and M+S (72 6 6 pg/mL) was reducedcompared with both P (84 6 5 pg/mL)and M (88 6 7 pg/mL).The baselineplasma bioactive GLP-1 concentration(Fig. 2E) was higher after S (10.1 6 1.2pg/mL) and M+S (16.9 6 2.2 pg/mL)compared with M (5.5 6 0.5 pg/mL) orP (6.9 6 1.0 pg/mL). Mean post-MTTplasma bioactive GLP-1 concentrationwas higher after S (26.1 6 2.2 pg/mL)and M+S (34.9 6 2.0 pg/mL) comparedwith M (14.26 1.3 pg/mL) and P (13.161.0 pg/mL) (both P , 0.01). Baselineplasma GIP concentrations were similar(;44 pg/mL) in all four treatmentgroups and increased similarly inP (186 6 8 pg/mL), M (171 6 7 pg/mL),S (173 6 6 pg/mL), and M+S (183 6 9pg/mL).

Under fasting conditions, EGP (Fig. 3A)in P (1.95 6 0.06 mg/kg z min) decreasedmodestly afterM (1.8460.08mg/kg zmin)and S (1.826 0.09 mg/kg zmin) (both P,0.05 vs. P) and decreased further after M+S

care.diabetesjournals.org DIABETES CARE, VOLUME 36, SEPTEMBER 2013 2757

Solis-Herrera and Associates

(1.496 0.05mg/kg zmin) (P, 0.01 vs. P).During the MTT, the mean RaT was 2.7860.18 mg/kg z min after P, decreased to2.59 6 0.17 mg/kg z min after M and to2.56 6 0.11 mg/kg z min after S (bothP , 0.05 vs. P), and decreased further to2.23 6 0.12 mg/kg z min after M+S (P ,0.01 vs. S and M). During the MTT, EGP(RaT2RaO)was suppressed to 0.7360.14mg/kg z min after P (Fig. 3B). Treatmentswith M (0.33 6 0.12 mg/kg z min), S(0.516 0.09 mg/kg z min), and M+S com-bined (0.25 6 0.10 mg/kg z min) furtherenhanced the suppression of EGP duringthe MMT (all P , 0.05 vs. P). The meanrate of oral glucose appearance (Fig. 3C)was similar after all four treatment regimens.The total amount of glucose that appearedin the peripheral circulation over the post-meal period of 360 min ranged between 62and 68 g, i.e., 82–91% of the ingested glu-cose load of 75 g. Thus, the splanchnic glu-cose uptake was similar (7–13 g) afterall four treatments. The mean tissue Rd(Fig. 3D) during the MTT were comparableafter treatmentwith P (2.566 0.12mg/kg zmin), M (2.516 0.13 mg/kg zmin), and S(2.41 6 0.10 mg/kg z min) and after M+S(2.38 6 0.11 mg/kg z min).

Insulin sensitivity and insulinogenicindices were calculated from 0 to 180minfor each treatment period, adapted from aprevious publication (15). These indiceswere calculated between 0 and 180 min

because most hormonal changes occurduring this postmeal period. The MI ofinsulin sensitivity was similar after eachtherapy, while hepatic insulin sensitivityimproved significantly with M and M+S(13.26 2.1 and 12.66 1.5 mU/mL zmg/kg zmin) but not with S alone (18.06 2.7mU/mL z mg/kg z min) compared with P(20.86 4.2 mU/mL z mg/kg z min). Insu-lin secretion measured both as (DI/DGand DISR/DG) increased mildly withS alone and M alone (both P = NS andfurther after M+S treatment, P , 0.05).The calculated b-cell function index(DI/DG3MI) paralleled insulin secretion.There was close concordance between thechanges in insulinogenic indices calculatedover the entire 360-min period of the MTTand the initial 180 min of the MTT for allfour treatment periods.

In each of the four treatment proto-cols, the decrement in EGP correlatedwith the decrement in FPG concentration(r = 0.54, P , 0.03). The increment inplasma insulin concentration and the dec-rement in plasma glucagon concentrationwere correlated with the decrement inEGP (r = 0.45, P , 0.01, and r = 0.35,P , 0.05, respectively) after the M+Scombination. The decrement in plasmaglucagon concentration also correlatedwith the decrement in EGP (r = 0.35,P , 0.05) after the S monotherapy. Theincrement in bioactive GLP-1 plasma

concentration correlated with both the in-crement in plasma insulin concentrationand the decrement in plasma glucagonconcentration (r = 0.38, P , 0.05, andr = 0.22, P , 0.05, respectively) aftercombined M+S therapy. Multiple regres-sion analyses performed for all of these pa-rameters further indicate that FPGconcentration is correlated with changesin post-MTT glucose, EGP, insulin, gluca-gon, and bioactive GLP-1 (r2 = 0.16, P =0.05).

There was no correlation between theincrement in plasma GIP concentrationand either the increment in plasma in-sulin concentration or the decrement inplasma glucagon concentration after anyof the four treatments.

CONCLUSIONSdCombination ther-apy with M+S is an effective treatmentoption for patients with T2DM (2). How-ever, the mechanism(s) underlying the“additive” effect of M+S have yet to beestablished. Our findings confirm that,when used as monotherapy, both S andM reduce postprandial glycemic excur-sion to a comparable degree, while com-bination of the two promotes greaterattenuation of the postprandial hypergly-cemia. The postprandial effect of M mon-otherapy appears to be secondary to adirect suppression of endogenous (he-patic) glucose production, since neither

Figure 1dSchematic representation of the study design. A random sequence was assigned for each subject to receive 6 weeks of treatment with P, M,S, and M+S combination. A 2-week washout period was observed between treatment periods.

2758 DIABETES CARE, VOLUME 36, SEPTEMBER 2013 care.diabetesjournals.org

DPP-4 inhibitors, metformin, and glucose lowering

plasma insulin nor glucagon levelschanged significantly, while that of S ismediated primarily by the suppressionof glucagon secretion. The direct effectof M on the liver presumably is hormoneindependent, whereas the reduction inEGP after S therapy most likely is second-ary to alterations in the portal insulin-to-glucagon ratio resulting from the increasein plasma GLP-1. The steep slope of de-cline in EGP after M compared with

the slower decrease in EGP after S mono-therapy also suggests that different mech-anisms are involved in the enhancementof EGP suppression. Themarked decreasein basal EGP and the steep slope withgreater suppression of EGP after combi-nation M+S also are consistent with in-dependent and additive effects of the twoantidiabetes agents. Comparable EGPsuppression has been reported in normalsubjects after a glucose challenge, which

also indicated that most of the glucosewas disposed peripherally (16).

Neither plasma insulin nor ISRs weresignificantly enhanced after the meal witheither M or S monotherapy. In contrast,M+S combination was accompanied by aclear augmentation of insulin secretionand potentiating of plasma glucagonsuppression, resulting in the greatest sup-pression of EGP. Of note, plasma gluca-gon suppression after M+S therapy

Figure 2dA: Changes in plasma glucose concentration during MTT. B: Changes in plasma insulin concentration during MTT. C: Changes in ISR,calculated form the plasma C-peptide deconvolution curve, during MTT. D: Changes in plasma glucagon concentration during MTT. E: Plasmabioactive GLP-1 concentration during MTT.

care.diabetesjournals.org DIABETES CARE, VOLUME 36, SEPTEMBER 2013 2759

Solis-Herrera and Associates

correlated with the quick surge (initial120 min) in plasma bioactive GLP-1. Wedid not observe a significant change inpostprandial insulin secretion after 6weeks of S monotherapy despite themodest increase in plasma bioactiveGLP-1 concentration. However, equiva-lent plasma insulin and C-peptide levels(i.e., increased DI/DG and DISR/DG) inthe face of a lower plasma glucose concen-tration after S (comparedwith P) indicatesthat b-cell function improved. Thesefindings are consistent with a previous re-port (6) showing that the immediate (mi-nutes to hours) absolute increase inplasma insulin response induced by vilda-gliptin dissipates overtime. Unlike insu-lin, suppression of plasma glucagon wasdetected during the initial 120 min of theMTT after treatment with S alone and thecombined M+S. These observationssuggest a differential sensitivity of pancre-atic a- and b-cells to GLP-1 (17,18).

The present results demonstrate thatM, S, and the combination M+S therapydid not have any effects on the Ra of in-gested glucose in the peripheral circula-tion, on splanchnic glucose utilization, or

on the rate of tissue glucose disposal.However, we cannot entirely rule outsmall differences among treatments in ei-ther splanchnic glucose uptake or tissueglucose disposal, since these could haveescaped our detection. These results indi-cate that the reduction in EGP was theprimary factor responsible for the attenu-ation in the postmeal plasma glucoseconcentration after each of the three ther-apeutic interventions. Thus, unlike theGLP-1 analogs (19–21) S has no effecton either gastric emptying or splanchnicglucose utilization. The failure to observeany increase peripheral glucose disposalis not surprising for two reasons: 1) nei-ther M, in the absence of weight loss(22,23), nor S (1–3) is known to enhancemuscle insulin sensitivity and 2) despitethe increase in plasma insulin responsewith the M+S treatment, any incrementin insulin-stimulated glucose disposalwould have been offset by the large reduc-tion in plasma glucose concentration.

In this study, we show a two- tothreefold increase in basal plasma bioactiveGLP-1 concentration in T2DM patientsafter 6 weeks of S used in combination

with M. Within 120 min after meal in-gestion, bioactive GLP-1 concentration in-creased with S treatment, and this elevationwas much greater with the combined M+Streatment, even though there was no sig-nificant rise with M treatment. This obser-vation is consistent with previous reports(8–12) and underscores the clinical rele-vance of the additive, even synergistic, ef-fect of M when used together with S toincrease plasma GLP-1 levels. However,this effect of M apparently relies on simul-taneous improvement in plasma glucoseconcentration, perhaps leading to aug-mented insulin secretion and enhancedsuppression of glucagon secretion. Anovel finding was the fourfold increasein plasma GIP levels observed with Mmonotherapy. However, since basal andpostmeal plasma GIP levels were compa-rable in all four treatment periods, the en-hanced stimulation of insulin secretionand suppression of glucagon secretion inT2DM patients treated with S alone or incombination with M must have resultedfrom the increase in bioactive plasmaGLP-1 levels. Our results are consistentwith those of others (24,25) and indicate

Figure 3dA: Changes in RaT during MTT. B: Changes in EGP during MTT. C: RaO during MTT. D: Rd during MTT.

2760 DIABETES CARE, VOLUME 36, SEPTEMBER 2013 care.diabetesjournals.org

DPP-4 inhibitors, metformin, and glucose lowering

that GIP does not potentiate the antidiabeteseffect of GLP-1 in T2DM patients. Consis-tentwith this, there is no known role forGIPon glucagon suppression and the effect ofGIP on insulin stimulation overlap with theGLP-1 intracellular pathways (17).

Whether the magnified GLP-1 re-sponse to a standard meal after DPP-4treatment in combination with M is aconsequence of the effect of S plus betterglycemic control, superimposed on Mstimulation of endogenous GLP-1 secre-tion, cannot be determined from ourstudy. M, as well as some other antidia-betes agents (8–10,24), has been sug-gested to directly enhance GLP-1 releasefrom the intestinal L cells. However, thepossibility that this is a nonspecific effectrelated to improved glycemic control hasyet to be excluded (26). Because Mmono-therapy was not accompanied by any sig-nificant change in plasma GLP-1 in ourstudies either at baseline or afterthe meal, our data support the notionthat the magnified GLP-1 response isprimarily explained by better glycemiccontrol. These observations are of consi-derable clinical significance, since thecombination of M with a DPP-4 inhibitorprovides a unique mechanism of glucoselowering with additive effect to reduceplasma glucose concentration in T2DMpatients.

In summary, the current study dem-onstrates that S and M monotherapyreduces glycemic excursions via indepen-dent and additive mechanisms. The in-hibition of endogenous (hepatic) glucoseproduction by M most likely resultedfrom a direct effect on the liver, whereasthe effect of S on EGP was indirect andassociated with elevated plasma GLP-1levels with subsequent glucagon suppres-sion. Combination M+S therapy furtherreduced postmeal plasma glucose con-centrations subsequent to a greater in-crease in fasting and postmeal GLP-1levels, accompanied by a further en-hancement in insulin secretion and glu-cagon suppression. There were nochanges in splanchnic glucose uptake,gastric emptying, or insulin-mediatedglucose disposal with any treatment regi-mens, and the attenuation of the post-prandial hyperglycemia with all therapieswas entirely attributed to the reductionin EGP.

AcknowledgmentsdFunding for this studywas provided in part by the University of TexasHealth Science Center.

Funding was also provided in part byMerck. C.T. has received honoraria from thespeakers bureau at Boehringer-Ingelheim andAmylin Pharmaceuticals. E.C. has receivedhonoraria from the speakers bureau at Sanofi,Amylin Pharmaceuticals, and Boehringer-Ingelheim–Eli Lilly Alliance. R.A.D. hasreceived honoraria as a member of the advisoryboard of Amylin Pharmaceuticals, NovoNordisk, Boehringer-Ingelheim, Bristol-MyersSquibb, and GlaxoSmithKline. No other po-tential conflicts of interest relevant to this articlewere reported.C.S.-H. executed the research procedures,

sample collection, laboratory analyses, and datainterpretation. C.T. designed the study; super-vised research procedures, clinical manage-ment, and laboratory analyses; interpreted data;contributed to the discussion; and reviewed andedited the manuscript. J.d.J.G.-G. executedthe research procedures, sample collection,laboratory analyses, and data interpretation.J.A. provided laboratory analyses, data in-terpretation, and technical assistance. R.A.D.and E.C. designed the study; supervised re-search procedures, clinical management, andlaboratory analyses; interpreted data; contrib-uted to the discussion; and reviewed and editedthe manuscript. E.C. is the guarantor of thiswork and, as such, had full access to all the datain the study and takes responsibility for theintegrity of the data and the accuracy of the dataanalysis.Parts of this studywere presented in abstract

form at the 72nd Scientific Sessions of theAmerican Diabetes Association, Philadelphia,Pennsylvania, 8–12 June 2012.The authors are thankful for the support of

the staff and faculty of the Texas Diabetes In-stitute Research Center and R. Pesek, RN, NurseCoordinator, Division of Diabetes, Departmentof Medicine, UTHSCSA, for her invaluablecontribution in the care of the study subjects.The authors are also grateful to L. Albarado,Administrative Assistant, Division of Diabetes,Department of Medicine, UTHSCSA, for helpwith the preparation of the manuscript.

References1. Aschner P, Kipnes MS, Lunceford JK,

Sanchez M, Mickel C, Williams-HermanDE; Sitagliptin Study 021 Group. Effect ofthe dipeptidyl peptidase-4 inhibitor sita-gliptin as monotherapy on glycemic con-trol in patients with type 2 diabetes.Diabetes Care 2006;29:2632–2637

2. Goldstein BJ, Feinglos MN, Lunceford JK,Johnson J, Williams-Herman DE; Sita-gliptin 036 Study Group. Effect of initialcombination therapy with sitagliptin,a dipeptidyl peptidase-4 inhibitor, andmetformin on glycemic control in patientswith type 2 diabetes. Diabetes Care 2007;30:1979–1987

3. Nauck MA, Meininger G, Sheng D,Terranella L, Stein PP; Sitagliptin Study024 Group. Efficacy and safety of the

dipeptidyl peptidase-4 inhibitor, sita-gliptin, compared with the sulfonylurea,glipizide, in patients with type 2 diabetesinadequately controlled on metforminalone: a randomized, double-blind, non-inferiority trial. Diabetes Obes Metab2007;9:194–205

4. Deacon CF. Therapeutic strategies basedon glucagon-like peptide 1. Diabetes2004;53:2181–2189

5. Ahrén B. Vildagliptin: an inhibitor of di-peptidyl peptidase-4 with antidiabeticproperties. Expert Opin Investig Drugs2006;15:431–442

6. Balas B, Baig MR, Watson C, et al. Thedipeptidyl peptidase IV inhibitor vilda-gliptin suppresses endogenous glucoseproduction and enhances islet functionafter single-dose administration in type 2diabetic patients. J Clin Endocrinol Metab2007;92:1249–1255

7. Steele R, Bjerknes C, Rathgeb I, AltszulerN. Glucose uptake and production duringthe oral glucose tolerance test. Diabetes1968;17:415–421

8. Mannucci E, Ognibene A, Cremasco F,et al. Effect of metformin on glucagon-likepeptide 1 (GLP-1) and leptin levels inobese nondiabetic subjects. Diabetes Care2001;24:489–494

9. Lindsay JR, Duffy NA, McKillop AM, et al.Inhibition of dipeptidyl peptidase IV ac-tivity by oral metformin in type 2 diabetes.Diabet Med 2005;22:654–657

10. Green BD, Irwin N, Duffy NA, Gault VA,O’harte FP, Flatt PR. Inhibition of di-peptidyl peptidase-IV activity by metfor-min enhances the antidiabetic effects ofglucagon-like peptide-1. Eur J Pharmacol2006;547:192–199

11. Mannucci E, Pala L, Ciani S, et al. Hy-perglycaemia increases dipeptidyl pepti-dase IV activity in diabetes mellitus.Diabetologia 2005;48:1168–1172

12. Fadini GP, Albiero M, Menegazzo L,de Kreutzenberg SV, Avogaro A. Theincreased dipeptidyl peptidase-4 activityis not counteracted by optimized glucosecontrol in type 2 diabetes, but is lower inmetformin-treated patients. DiabetesObes Metab 2012;14:518–522

13. Van Cauter E, Mestrez F, Sturis J,Polonsky KS. Estimation of insulin se-cretion rates from C-peptide levels.Comparison of individual and standardkinetic parameters for C-peptide clear-ance. Diabetes 1992;41:368–377

14. Hovorka R, Soons PA, Young MA. ISEC:a program to calculate insulin secretion.ComputMethods Programs Biomed 1996;50:253–264

15. Matsuda M, DeFronzo RA. Insulin sensi-tivity indices obtained from oral glucosetolerance testing: comparison with theeuglycemic insulin clamp. Diabetes Care1999;22:1462–1470

16. Jackson RA, Roshania RD, Hawa MI, SimBM, DiSilvio L. Impact of glucose ingestion

care.diabetesjournals.org DIABETES CARE, VOLUME 36, SEPTEMBER 2013 2761

Solis-Herrera and Associates

on hepatic and peripheral glucose metab-olism in man: an analysis based on simul-taneous use of the forearm and doubleisotope techniques. J Clin EndocrinolMetab 1986;63:541–549

17. Baggio LL, Drucker DJ. Biology of in-cretins: GLP-1 and GIP. Gastroenterology2007;132:2131–2157

18. Carr RD, Larsen MO, Jelic K, et al. Secre-tion and dipeptidyl peptidase-4-mediatedmetabolism of incretin hormones after amixed meal or glucose ingestion in obesecompared to lean, nondiabetic men. J ClinEndocrinol Metab 2010;95:872–878

19. Cervera A, Wajcberg E, Sriwijitkamol A,et al. Mechanism of action of exenatide toreduce postprandial hyperglycemia intype 2 diabetes. Am J Physiol EndocrinolMetab 2008;294:E846–E852

20. Cersosimo E, Gastaldelli A, Cervera A,et al. Effect of exenatide on splanchnic andperipheral glucose metabolism in Type 2Diabetic Subjects. J Clin EndocrinolMetab 2011;96:1763–1770

21. Mudaliar S, Henry RR. The incretin hor-mones: from scientific discovery to prac-tical therapeutics. Diabetologia 2012;55:1865–1868

22. Cusi K, Consoli A, DeFronzo RA. Meta-bolic effects of metformin on glucose andlactate metabolism in noninsulin-dependentdiabetes mellitus. J Clin Endocrinol Metab1996;81:4059–4067

23. Hirshman MF, Goodyear LJ, Braun B, et al.Combining short-term metformin treat-ment and one bout of exercise does notincrease insulin action in insulin-resistant

individuals. Am J Physiol EndocrinolMetab2010;298:E815–E823

24. Mentis N, Vardarli I, Köthe LD, et al. GIPdoesnot potentiate the antidiabetic effects ofGLP-1 in hyperglycemic patients with type2 diabetes. Diabetes 2011;60:1270–1276

25. Deacon CF, Nauck MA, Meier J, H€uckingK, Holst JJ. Degradation of endogenousand exogenous gastric inhibitory poly-peptide in healthy and in type 2 diabeticsubjects as revealed using a new assay forthe intact peptide. J Clin EndocrinolMetab 2000;85:3575–3581

26. Lenhard JM, Croom DK, Minnick DT.Reduced serum dipeptidyl peptidase-IVafter metformin and pioglitazone treat-ments. Biochem Biophys Res Commun2004;324:92–97

2762 DIABETES CARE, VOLUME 36, SEPTEMBER 2013 care.diabetesjournals.org

DPP-4 inhibitors, metformin, and glucose lowering