Engineering Glucose Responsiveness Into InsulinNiels C. Kaarsholm, Songnian Lin, Lin Yan, Theresa Kelly, Margaret van Heek, James Mu, Margaret Wu,Ge Dai, Yan Cui, Yonghua Zhu, Ester Carballo-Jane, Vijay Reddy, Peter Zafian, Pei Huo, Shuai Shi,Valentyn Antochshuk, Aimie Ogawa, Franklin Liu, Sandra C. Souza, Wolfgang Seghezzi, Joseph L. Duffy,Mark Erion, Ravi P. Nargund, and David E. Kelley

Diabetes 2018;67:299–308 | https://doi.org/10.2337/db17-0577

Insulin has a narrow therapeutic index, reflected in a smallmargin between a dose that achieves good glycemic controland one that causes hypoglycemia. Once injected, theclearance of exogenous insulin is invariant regardless ofblood glucose, aggravating the potential to cause hypo-glycemia. We sought to create a “smart” insulin, one thatcan alter insulin clearance and hence insulin action in re-sponse to blood glucose, mitigating risk for hypoglyce-mia. The approach added saccharide units to insulin tocreate insulin analogs with affinity for both the insulin re-ceptor (IR) and mannose receptor C-type 1 (MR), whichfunctions to clear endogenous mannosylated proteins, aprinciple used to endow insulin analogs with glucoseresponsivity. Iteration of these efforts culminated in thediscovery of MK-2640, and its in vitro and in vivo preclinicalproperties are detailed in this report. In glucose clampexperiments conducted in healthy dogs, as plasma glu-cose was lowered stepwise from 280 mg/dL to 80 mg/dL,progressively more MK-2640 was cleared via MR, reduc-ing by ∼30% its availability for binding to the IR. In doseescalations studies in diabetic minipigs, a higher thera-peutic index for MK-2640 (threefold) was observed versusregular insulin (1.3-fold).

Nearly a century has passed since the discovery of insulin, yetthe remarkable success of insulin therapy has been temperedby the risk of iatrogenic hypoglycemia, which stands as amajor barrier to the achievement of tight glycemic control(1–4). To mitigate risk for hypoglycemia, insulin analogshave been developed with improvements in pharmaco-kinetics (PKs) (5–7). However, efforts to improve insulinPKs do not enable exogenous insulin to autonomously

modulate its action in the face of descending (or ascending)plasma glucose and thus do not change the intrinsicallynarrow therapeutic index of exogenously administered insulin.To address the latter, the notion of a glucose-responsiveinsulin (GRI) was proposed four decades ago (8). Attemptsat creating a GRI have sought to incorporate insulin withina subcutaneous depot matrix containing glucose-sensitivechemical “triggers” that release insulin (9–12), but there isa substantial challenge in sufficiently modulating insulinrelease across the narrow physiological range of plasma glu-cose. An alternative strategy for a GRI, to exploit endogenouslectin–based clearance, was reported by Zion and Lancaster(13), and is the focus of this report.

Lectins recognize and bind carbohydrate domains ofglycoproteins. Circulating and cell-based lectins function inimmune surveillance and as clearance of glycoproteins(14,15). Mannose receptor C-type 1 (MR) is the prototypicalmember of the mannose receptor family of transmembranelectins. Its main function is to recognize endogenous senes-cent proteins tagged for destruction and pathogens iden-tified by their surface glycan and to deliver these forlysosomal degradation without eliciting an immune responseor cytokine release (16–19). MR is an abundant lectin inhepatic sinusoidal endothelial cells and on certain macro-phages and dendritic cells, with a well-conserved homologyacross species (14). Glucose has a low affinity for MR butnevertheless can compete with binding of other ligands.

A concept of a GRI that could be engineered into aninsulin analog by targeting MR-based clearance is illustratedin Fig. 1. This concept involves creating a chimeric insulinanalog that undergoes a substantial fraction of its clearancethrough MR when plasma glucose is within the euglycemic

Merck Research Laboratories, Merck & Co., Inc., Kenilworth, NJ

Corresponding author: Ravi P. Nargund, ravi_nargund@merck.com, or David E.Kelley, kelleydek@gmail.com.

Received 29 May 2017 and accepted 30 October 2017.

This article contains Supplementary Data online at http://diabetes.diabetesjournals.org/lookup/suppl/doi:10.2337/db17-0577/-/DC1.

© 2017 by the American Diabetes Association. Readers may use this article aslong as the work is properly cited, the use is educational and not for profit, and thework is not altered. More information is available at http://www.diabetesjournals.org/content/license.

Diabetes Volume 67, February 2018 299

PHARMACOLOGYAND

THERAPEUTIC

S

or hypoglycemic range, thus lessening availability for interac-tion with the insulin receptor (IR), with the opposite direc-tion effect at a progressively higher ambient glucose. Wewill describe our initial steps in creating a GRI using lectin-based clearance (J. Mu et al., unpublished observations), whereasthe current report focuses on a single GRI, MK-2640. Wedescribe the preclinical development of MK-2640, includ-ing dog and minipig in vivo studies designed to assess glu-cose responsivity.

RESEARCH DESIGN AND METHODS

In Vitro Binding AssaysMK-2640 binding to IR was determined in a scintillationproximity assay with membranes prepared from CHO cellsoverexpressing human, minipig, or dog IR(B), and insulinas the competitive ligand (J. Mu et al., unpublishedobservations). MK-2640 direct binding to human or dogMR was assessed via surface plasmon resonance using His-tagged purified receptors immobilized onto a Biacore CM5chip. To appraise glucose inhibition of MR binding, thesurface plasmon resonance assay was modified to includevarying concentrations of glucose or a-methylmannose(a-MM) in the presence of a fixed concentration of MK-2640, approximating its Kd value, to determine the glucoseinhibitory IC50 for MK-2640 binding to human or dog MR.Competition binding assays for MR, DC-SIGN, and MBLwere conducted using mannosylated-BSA, labeled witheuropium for fluorescent detection, and varying concentra-tions of MK-2640 to determine affinity for the humanlectin receptors. An ex vivo assay of glucose inhibition ofMK-2640 binding in the liver was conducted by perfusingthe compound into the portal vein of mice with varyingconcentrations of glucose and then imaging liver sectionswith an anti-insulin antibody. Briefly, the portal vein

of mice was cannulated for perfusion with MK-2640,and varying concentrations of glucose were added tothe infusion buffer. Liver caudate lobe samples weretaken and fixed (10% buffered formalin), and immuno-histochemistry was performed with guinea pig anti-insulin antibody (Invitrogen #180067). Digital imageswere captured, and the percentage of immunoreactive cellswas determined via instrument software (Membrane v9,Aperio Technologies).

In Vivo Evaluations of MK-2640All animal procedures were reviewed and approved by theMerck Research Laboratories (Merck & Co., Kenilworth, NJ)Institutional Animal Care and Use Committee. The Guidefor the Care and Use of Laboratory Animals was followed inthe conduct of the animal studies. Veterinary care was givento any animals requiring medical attention.

Male Yucatan minipigs from the Sinclair Research Center,healthy (nondiabetic [ND]) or rendered type 1 diabetic (D)by alloxan injections, underwent placement of two jugularvenous vascular access ports for administration of MK-2640or recombinant human insulin (RHI; Humulin R, Lilly)intravenously (i.v.) and for blood sampling. The study usedminipigs rendered diabetic by alloxan that attained plasmaglucose levels of ;300 mg/dL. RHI or MK-2640 was coad-ministered with PBS or a-MM (9.4 mg/kg/min), the latter ahigh-affinity ligand for MR that can act as a chemicalblockade of MR (20). Insulin levels were measured by ELISA(RHI) or by liquid chromatography–mass spectrometryfor MK-2640. To elucidate glucose-responsiveness ofRHI or MK-2640, respective clearances were assessedin D minipigs and compared with corresponding valuesobtained in ND minipigs that had been coinfused withPBS or a-MM. To appraise therapeutic index, MK-2640 or

Figure 1—A concept for GRI is illustrated that is based on adding saccharide modifications to the insulin molecule. The resulting analog hasaffinity for the IR as well as for an MR, which clears insulin via endocytosis and lysosomal degradation. Glucose competes for binding to the MR.At high ambient glucose, less of the analog is cleared by MR and a larger proportion becomes available for IR pharmacology; conversely, at lowglucose, a proportionally larger fraction of the analog is cleared via MR, lowering availability for IR interaction. ICC, insulin-carbohydrateconjugate.

300 Glucose-Responsive Insulin Diabetes Volume 67, February 2018

RHI was administered subcutaneously (s.c.) to D minipigswith dose escalation (successive 30% dose increments),and plasma glucose monitoring for 8 h (AU480 ClinicalChemistry Analyzer, Beckman Coulter) until a dose thatcaused hypoglycemia was identified. Plasma epineph-rine was assessed by liquid chromatography–mass spec-trometry to monitor the hypoglycemic counterregulatoryresponse.

MK-2640 was examined in healthy 4- to 6-year-old malebeagle dogs with euglycemic and hyperglycemic clamps toexamine its PK and pharmacodynamic (PD) effects across arange of steady-state plasma glucose concentrations. Dogswere fasted overnight before a clamp and had previouslybeen prepared with dual-ported femoral artery and femoralvein cannulations, respectively, for sample collection andinfusions. Animals received an infusion of somatostatin(0.8 mg/kg/min) during 6 target glucose levels (80, 120,160, 200, 240, and 280 mg/dL) to suppress endogenousinsulin secretion. Each dog completed all clamps. RHI atan infusion rate of 3 pmol/kg/min, was used as a compar-ator to MK-2640. Pilot studies found an infusion rate forMK-2640 of 45 pmol/kg/min achieved PD equipoise withRHI when clamps were conducted at 200 mg/dL, and theserespective infusion rates were used in all clamps. PK read-outs were RHI and MK-2640 concentrations and clear-ance. A change in MK-2640 clearance was interpreted as

reflecting glucose responsive PKs. MK-2640 was tested foractivation of cytokine release in the plasma of dogs receiv-ing MK-2640 or RHI infusions. Cytokine levels (interleukin[IL]-6, IL-10, IL-8, IL-2, and tumor necrosis factor-a) wereevaluated by ELISA (Meso Scale Diagnostics, Rockville, MD),and dog blood exposed ex vivo to lipopolysaccharide(10 pg/mL) was a positive control.

Statistical AnalysisData analyses were performed in GraphPad Prism (GraphPadSoftware, San Diego, CA). Calculations of P values werebased on ANOVA and the unpaired Student t test, which-ever was applicable. Statistical significance was defined asP , 0.05.

RESULTS

Design and Synthesis of MK-2640Chemistry efforts focused on conjugating RHI at the aminotermini, including A1, B1, and the B29 Lys amino groups,individually or as bis-functionalization with linkers bearingmono-, di-, or trisaccharide units, as shown in Fig. 2A.Glucose, mannose, and fucose were selected for exploringthe identification of potential lead GRI candidates, andthrough studies of the structure activity relationship, alysine-based, nonsymmetric, mannose based–linker structurewas identified that began to demonstrate modest glucosemodulation of its PKs. This recognition also helped to define

Figure 2—A: Chemical ligation strategy to synthesize insulin oligosaccharide conjugates is outlined, together with the specific structure ofMK-2640. B: The structure of compound 2 (MK-2640 analog with A1 and B29 positions conjugated with moieties possessing non-MR bindingmorpholines instead of saccharide units). IUPAC, International Union of Pure and Applied Chemistry; NHS, N-hydroxysuccinimide.

diabetes.diabetesjournals.org Kaarsholm and Associates 301

an appropriate range for respective affinities for IR and MR.We subsequently observed that the use of fucose in place ofmannose increased MR binding affinity and with iterationsled to the identification of MK-2640 as an analog withpromising glucose responsiveness, and its structure is shownin Fig. 2A. Additional information on the chemistry effortcan be accessed in a recently published patent applicationand references therein (21). To support the mechanism ofaction studies in minipigs, compound 2 (Fig. 2B) was syn-thesized by modifications at both A1 and B29 positionswith morpholines replacing saccharide units that render itwith similar IR affinity but greatly reduced MR affinitycompared with MK-2640.

PK Properties of MK-2640The PK properties of MK-2640 were evaluated in beagledogs and Yucatan minipigs after i.v. and s.c. dosing. As apreliminary formulation effort, directed at achieving drugsolubility, MK-2640 was formulated in sodium phosphate(26.5 mmol/L) buffer (pH = 7.4) containing sodium chloride(87 mmol/L) and phenol (21 mmol/L). Dogs were dosedi.v. at 2.5 mg/kg and s.c. at 15 mg/kg. Minipigs were dosedi.v. at 5 mg/kg and s.c. at 45 mg/kg. After i.v. dosing,MK-2640 exhibited moderate to high systemic clearancein dogs and minipigs (8 and 39 mL/min/kg, respectively).The steady state volume of distribution was low (;0.04–0.2L/kg), and the elimination half-life was short (3–14 min) inthese species. After s.c. administration of MK-2640, absorp-tion was faster in the dogs than in the minipigs (tmax

18 min vs. 36 min), and the bioavailability varied between26% and 44%. The biodistribution of MK-2640 was exam-ined in healthy rats using s.c. dosing of [3H]MK-2640, andthis showed significant amounts of [3H]MK-2640–relatedradioactivity detected in liver, kidney, muscle, and fat.

Targeting of MRThe binding affinity of MK-2640 for the IR, as presented inTable 1, was substantially lower than RHI (;4% in com-parison), which suggests that the potential for full clinicaldevelopment of MK-2640 might be limited due to low po-tency. However, the potential for glucose responsiveness wasregarded as important to interrogate in vivo. In vitro, at thehighest concentrations of MK-2640 examined, the compoundcompletely blocked binding of native insulin. Binding to human

MR was similar to that observed for the minipig and dog MR,as reported in Table 2.

MK-2640 did not bind to MBL at the highest concen-tration tested (10 mmol/L) and had relatively weak affinityfor DC-SIGN, with an IC50 of 1,800 nmol/L. Consistent withthese findings, MK-2640 did not evoke cytokine releasefrom human differentiated macrophages (SupplementaryFig. 1). By using increasing concentrations of glucose, asshown in Fig. 3A, the ability of glucose to compete withthe binding of MK-2640 to MR was assessed in vitro, andan IC50 of 8 mmol/L glucose was determined; a correspond-ing value for a “high-affinity” MR–binding small molecule,a-MM, is 0.62 mmol/L. An ex vivo mouse liver perfusionmodel was used to assess whether glucose affected parti-tioning of selected analogs between IR and MR in the pres-ence of both receptors. As shown in Fig. 3B, a large fractionof MK-2640 was taken up by MR at a low glucose concen-tration, and this fraction progressively declined as theglucose concentration was raised to hyperglycemia, withan IC50 value of ;9 mmol/L glucose, quite similar to theIC50 estimated from in vitro binding experiments usingimmobilized MR (Fig. 3B).

Glucose Responsiveness in MinipigsMK-2640 PK was evaluated in ND and D Yucatan minipigsafter i.v. dosing in the absence and presence of coinfuseda-MM. The i.v. administration of RHI led to a short-livedfall of plasma glucose; its PKs and PDs were unaffected bya-MM, as shown in Fig. 4A and C. The PKs and PDs ofMK-2640 in ND minipigs were strongly affected by a-MMcoinfusion. A dose of MK-2640 (0.35 nmol/kg) that hadrapid clearance and only a modest glucose-lowering effectin ND minipigs when administered alone produced transienthypoglycemia in the presence of a-MM (Fig. 4B), because itsclearance was markedly protracted (Fig. 4D). The MK-2640PK differences observed with and without a-MM servedto demarcate potential boundaries of a corresponding max-imal glucose responsiveness because these PK differencesdelineate the contribution of MR-mediated clearance (ateuglycemia). Glucose-responsive PKs were then exploredby giving RHI and MK-2640, each as an i.v. bolus to Dminipigs. MK-2640 has been shown to dose-dependentlylower glucose in D minipigs when administered i.v. (seeSupplementary Fig. 2), and a dose (0.35 nmol/kg) withminimal glucose-lowering was used to evaluate its PKs dur-ing stable hyperglycemia. As shown for RHI (Fig. 4C) andMK-2640 (Fig. 4D), hyperglycemia in D minipigs did notchange clearance of RHI versus ND, whereas the clearancefor MK-2640 in D minipigs was decreased. This differencein clearance of MK-2640 in D versus ND minipigs was inter-preted as highly similar to the PKs change observed withblockade of MR clearance by a-MM in ND minipigs. Ad-ditional studies demonstrated that a-MM–induced PK andPD changes for MK-2640 (Fig. 4B) cannot be ascribed to thereduced IR potency of MK-2640 per se because compound2 (0.69 nmol/L i.v. dose), with comparable IR potency asMK-2640 but minimal MR binding affinity, demonstrated

Table 1—Cell membrane IR-binding affinities

IR binding

Human Minipig Dog

IC50

SDIC50

SDIC50

SD(nmol/L) (nmol/L) (nmol/L)

RHI 0.48 1.5 0.39 1.8 0.44 1.3

MK-2640 7.0 2.1 7.4 1.2 8.5 1.1

Geometric mean IC50 and geometric SD are from four or moreindividual MK-2640 titrations for all species. Compound 2 is ananalog of MK-2640 wherein the saccharides are replaced bymorpholino groups. The IR human binding activity of compound2 was 5.7 nmol/L.

302 Glucose-Responsive Insulin Diabetes Volume 67, February 2018

negligible PK differences when dosed in ND minipigs withversus without a-MM.

Glucose ClampsTo assess glucose responsiveness of MK-2640, glucoseclamp studies were performed in healthy dogs clamped attarget plasma glucose concentrations of 80, 120, 160, 200,240, and 280 mg/dL, one level per study, with completecrossover among dogs. Doses of MK-2640 (45 pmol/kg/min)and RHI (3 pmol/kg/min) were used that required similarglucose infusion rate (GIR) during clamps conducted at200 mg/dL (11 6 1 vs. 10 6 1 mg/kg/min, respectively).These high rates of matched GIR, approximately fivefoldabove fasting rates of glucose flux, denote stimulation ofglucose utilization within peripheral tissues consistent withthe rat biodistribution data. Stepwise increases in plasma con-centration of MK-2640 and corresponding decreases in clear-ance were observed with increased plasma glucose duringclamps. The change in PKs of ;30% across this glycemicrange is shown in Fig. 5A. RHI plasma concentrationsremained unaffected by glycemic level, and RHI clearance(23 6 3 mL/kg/min) remained constant (Fig. 5B). ForMK-2640, the reduction of GIR required to maintain targetlevels of glycemia at less than 200 mg/dL declined to agreater extent than in the corresponding clamps conductedwith RHI. At a plasma glucose of 80 mg/dL, the steady stateGIR was 4 6 1 mg/kg/min for RHI and 2 6 1 mg/kg/minfor MK-2640 (P , 0.05). Two-hour infusions of MK-2640did not evoke changes in cytokines and were not differ-ent from the responses observed with RHI, as shown inSupplementary Fig. 3.

Partitioning of MK-2640 to Uptake by MR in DogsWhen compared during clamp studies conducted at a plasmaglucose of 200 mg/dL, the infusion rate of MK-2640(45 pmol/kg/min) needed to achieve PD equipoise with a3 pmol/kg/min infusion of RHI was 15-fold greater. Thisdifference can partly be attributed to the lower in vitropotency of MK-2640 for the IR. However, it is potentiallyimportant to take into account the percentage of MK-2640partitioned to degradation via MR-mediated uptake. Clampstudies were conducted in dogs with and without a con-comitant infusion of a-MM (9.65 mg/kg/min). At a plasmaglucose of 80 mg/dL, infusion of a-MM reduced the clear-ance of MK-2640 by 81%, delineating that a high fraction ofits clearance was via MR, leaving the remaining 19% to becleared by other pathways, presumably by IR. At a plasmaglucose of 300 mg/dL, the overall clearance of MK-2640 was

reduced by ;33% compared with euglycemic conditions,and with infusion of a-MM during hyperglycemia, clearanceby MR still accounted for a large fraction (;66%) of itsclearance.

Table 2—MR-binding affinities

MR binding (Biacore) MR binding (DELFIA)

Human Dog Human Minipig Dog

Kd (nmol/L) SD Kd (nmol/L) SD IC50 (nmol/L) SD IC50 (nmol/L) SD IC50 (nmol/L) SD

MK-2640 3.4 1.4 2.7 1.3 28 1.2 23 1.3 16 1.2

Geometric mean Kd and IC50 and geometric SD are from four or more individual MK-2640 titrations for all species. Compound 2 was inactivein the human IR binding assay. DELFIA, dissociation-enhanced lanthanide fluorescent immunoassay.

Figure 3—Glucose modulation of MK-2640 binding to MR. A: Datafrom an MR Biacore binding assay are shown in which titratingglucose concentrations are used in the constant presence of4 nmol/L MK-2640 (;Kd concentration added to the running buffer).A glucose IC50 of ;8 mmol/L was determined (compared with IC50

;0.62 mmol/L for a-MM). B: Ex vivo mouse liver perfusion assaywhere an increasing concentration of glucose was added to the buffercontaining a fixed concentration of MK-2640. The MK-2640 taken upby the liver was quantified by immunohistochemistry using an anti-insulin antibody capable of detecting MK-2640. IOC, insulin oligosac-charide conjugate.

diabetes.diabetesjournals.org Kaarsholm and Associates 303

Dose Escalation to HypoglycemiaDose escalations (30% successive increases) with s.c. ad-ministration of MK-2640 and RHI in D minipigs wereperformed to determine the effect of upward titrations ofdose to lower plasma glucose and to thereby identify a dosethat caused hypoglycemia. To estimate the therapeuticindex of RHI and MK-2640, the ratio of a dose that causedhypoglycemia (defined as a value of 55 mg/dL), relative tothe minimally efficacious dose (MED), was calculated; MEDwas defined as the dose at which at least 50% of the ani-mals reach glucose levels,120 mg/dL. As shown in Fig. 6A,s.c. administration of RHI led to a rapid decrease in plasmaglucose levels, with a MED of 0.41 nmol/kg. Furtherincreases above the MED for RHI resulted in a fall of bloodglucose levels into the hypoglycemic range. At a dose of RHIof 0.53 nmol/kg, a robust counterregulatory epinephrineresponse occurred, with a 150% increase over baseline val-ues observed at hypoglycemia (Fig. 6C); this RHI dose wasonly a 1.3-fold increment over the MED. MK-2640 caused adose-dependent decrease in plasma glucose with an MED of2.1 nmol/kg (Fig. 6B), and hypoglycemia caused by MK-2640was not observed until the 8.0 nmol/kg dose, a nearly four-fold therapeutic index. A counterregulatory response of epi-nephrine was, however, observed with an MK-2640 dose of6.1 nmol/kg and was more pronounced at the 8.0 nmol/kg

dose (Fig. 6D). By these criteria, the therapeutic index forMK-2640 can be estimated as 3:1 relative to its MED.

Subchronic Dosing of MK-2640 in D MinipigsProof of concept that sustained insulin therapy canmaintain reduction of hyperglycemia is well established.However, given the novelty of a mannosylated insulin suchas MK-2640, subchronic studies in D minipigs were under-taken to assess stability of glycemic control. Thrice-dailyinjections of MK-2640 or RHI were administered to Dminipigs for 2 weeks after animals were withdrawn fromtheir usual long-acting insulin therapy. Repeated dosing ofMK-2640 achieved control of hyperglycemia during the2 weeks of dosing that was comparable to that attainedwith RHI, as shown in Supplementary Fig. 4. Repeateddosing of MK-2640 was well tolerated, did not cause localreactions at injection sites, and did not elicit the generationof anti-drug antibodies or anti-insulin neutralizing anti-bodies (Supplementary Table 1).

Formulation Efforts to Achieve a Basal Insulin ProfileThe reduced in vitro potency of MK-2640 and its consequentrelatively slow onset of action, even when administeredi.v., fit a profile that seemed better suited as a long-acting basal insulin, rather than as a rapid-acting insulin,because the latter requires high in vitro potency. However,

Figure 4—PDs and PKs after an i.v. bolus of RHI (0.17 nmol/kg) (A and C) and MK-2640 (0.35 nmol/kg) (B and D) given to ND and D minipigswith and without coinfusion of a-MM (ND only), which provides a short-term chemical blockade of MK-2640 clearance by MR. A and B: Theeffect of the two insulins is shown in ND minipigs with and without a-MM. C and D: The PK differences and the effects of a-MM in ND and Dminipigs are compared. For panels A–D, data is presented as mean 6 SEM. Areas under the curve0-∞ for the different MK-2640 groups areindicated in panel D (as mean 6 SD).

304 Glucose-Responsive Insulin Diabetes Volume 67, February 2018

to achieve a sustained PK profile, it was necessary to advancebeyond the preliminary formulation used in the PK and PDstudies described above and explore development of a formu-lation of MK-2640 that could manifest slow absorptionkinetics over hours rather than minutes. Exploratory formu-lation work was begun to assess whether slow absorptionkinetics could be achieved for MK-2640. The preliminaryresults are shown in Supplementary Fig. 5, suggesting thatwith further refinement, a basal insulin preparation couldbe achieved for MK-2640 or other similar GRI candidates.

DISCUSSION

The synthesis of MK-2640 was undertaken in an effortto create an analog with a capacity to alter its PKs in a

glucose-responsive manner and hence its availability forinsulin action. The innovation was based on creating aninsulin chimera by attaching saccharides to native insulinthat would imbue capacity to bind to the lectin receptor MRyet retain capacity for alternatively binding to the IR. Uponbinding to MR, the insulin analog MK-2640, like otherligands for MR, is taken up and degraded in lysosomes. Thisfate precludes interaction with the IR. Glucose responsive-ness does not ensue per se from a capacity for binding toMR; rather, MR binding opens a pathway of insulin egressthat does not evoke insulin action. Creating glucose respon-siveness required that the interaction with MR could inturn be modulated by the level of plasma glucose. MK-2640binding by MR in vitro is increased at low concentrations ofglucose and vice versa. Liver perfusion studies also revealedthat uptake by the MR pathway is modulated by ambientglucose in a linear response and with an IC50 of;9 mmol/Lglucose very similar to that identified from in vitro bindingstudies. The results of these assays suggested that the clear-ance and degradation of MK-2640 by the MR pathwaycould have potential to be the greatest when ambient glu-cose is reduced, when it will be desirable to lower circulatinginsulin and thereby to hopefully mitigate risk for insulin-induced hypoglycemia.

From the outset of the protein engineering effort, thebalance of the affinity for IR relative to MR would clearly bean essential parameter guiding construction of GRI candi-date molecules. Attachment of saccharides to the insulinbackbone enabled interaction with MR, and with iterationsof saccharide selection, this interaction with MR could betuned to respond to ambient glucose. Attachment ofsaccharides to the insulin backbone, the site of attachments(e.g., A1, B1, and/or B29 of insulin), and the length andcomposition of the linker joining a saccharide to insulin alsoaffected analog potency on the IR (21). MK-2640 was iden-tified as an analog that might have a desirable balance ofrespective binding affinities for IR and MR, and the glucoseresponsiveness of MK-2640 derives from modulation of itsPKs because changes in ambient glucose do not modify theintrinsic potency of MK-2640 for the IR. The diminishedin vitro potency of MK-2640 for the IR likely would posi-tion this or similar GRI analogs to find usefulness as a basalrather than as a rapid-acting insulin because the latterneeds full in vitro potency so that its onset of action israpid. Although the reduced IR potency of MK-2640 posi-tions it as a potential basal insulin, additional effort toachieve a formulation that would slow its absorption kinet-ics would ultimately need to be achieved. In the currentreport, preliminary formulations efforts with MK-2640 in-dicated that slowing its absorption kinetics might be feasi-ble, but the in vivo data presented for MK-2640 in dogs andminipigs, whether administered s.c. or by i.v. injection, useda simpler “fit for purpose” formulation that achieved goodsolubility of the analog and that was associated with rela-tively prompt absorption kinetics after s.c. administration.

The hypothesis that MK-2640 might have a desirablebalance of respective binding affinities for IR and MR

Figure 5—Plasma drug concentrations for infused insulins weremeasured during steady-state conditions of a glucose clamp pro-cedure performed in healthy dogs using a constant rate of infusionof 45 pmol/kg/min for MK-2640 (A) and an infusion of 3 pmol/kg/minfor RHI (B). The target plasma glucose concentrations for each of theclamp studies are shown on the x-axis. A crossover study designwas used, by which dogs participated in each of the clamps, and aglucose-dependent increase in MK-2640 plasma concentration wasobserved, consistent with a glucose-responsive modulation of itsclearance, whereas no effect on the clearance of RHI was observed.Data are mean 6 SEM.

diabetes.diabetesjournals.org Kaarsholm and Associates 305

was tested in the dog and minipig, two preclinical speciescommonly used in the development of novel insulins. Inboth species, MK-2640 was shown to demonstrate glucose-responsive PKs and PDs. The MK-2640 infusion studiesconducted in healthy dogs, using glucose clamp methodsconducted across a range of target glucose concentrations,focused on PK changes and demonstrated that ambientglucose modulates in a linear manner the clearance ofMK-2640, increasing its clearance at lower plasma glucose.Collateral studies conducted during coinfusion of a-MM, ahigh-affinity ligand for MR that provides chemical blockadeof MK-2640 clearance by MR, revealed that at euglycemia,;80% of MK-2640 clearance is accounted for by MR-basedclearance. Thus, a substantial portion of the MK-2640 dosewas cleared via MR in the euglycemic clamp. On one hand,the considerable flux of MK-2640 through MR offers thepotential for meaningful plasma PK shifts upon glucosemodulation of MR binding. On the other hand, whetherit is practical to have an insulin analog that undergoesthis proportion of degradation independent of evoking in-sulin action and whether it is desirable to chronically engagethe MR pathway to this extent by a drug administered dailyclearly will require more investigation.

Glucose responsiveness was also clearly revealed inthe studies using MK-2640 in ND and alloxan-induced D

minipigs. The clearance of MK-2640 in ND minipigs wasgreatly impeded by an infusion of a-MM, a finding recapit-ulated in the setting of marked hyperglycemia of D mini-pigs. Importantly, dose escalation studies in D minipigssuggested a safer therapeutic index for MK-2640 comparedwith regular insulin. The working definition of therapeuticindex for insulin used in our studies was simple: it was tocompare the dose that induced hypoglycemia relative to onethat reduced fasting hyperglycemia to a desirable target of120 mg/dL. In the comparison of MK-2640 to regular insulin,the MK-2640 dose could be increased by threefold above theMED before hypoglycemia ensued or hypoglycemic counter-regulation was evoked. For regular insulin, the dose escalationwas an order of magnitude smaller, a 30% increase above theMED caused hypoglycemia and a counterregulatory response.

Insulin-induced hypoglycemia is the most serious acutecomplication of using this medicine (2,3). Innovations ininsulin chemistry have yielded basal insulin preparationswith more consistent duration of the absorptive phaseand flatter PK profiles, while efforts on prandial insulincontinue to strive for rapid onset of action (6,7). Yet, theseinsulin analogs continue to manifest a narrow therapeuticindex and do not per se contain a capacity to modify insulinaction once the compound has been administered. The as-piration of a GRI is to engineer into an insulin preparation a

Figure 6—Plasma glucose levels (A and B) and changes in plasma epinephrine levels (C and D) in D minipigs after s.c. administration of RHI(A and C) and MK-2640 (B and D) in escalating doses of a 30% increment over the preceding dose until the dose yielding a hypoglycemicresponse was observed. The shadowed area in A and B (55–120 mg/dL) defines the euglycemia range for the minipigs. The shadowed areas inC and D represent the normal range of variance for plasma concentrations of epinephrine under resting conditions in the D minipigs. Data aremean 6 SEM.

306 Glucose-Responsive Insulin Diabetes Volume 67, February 2018

“postinjection” flexibility that can modulate insulin avail-ability based on changing needs, as dictated by changes inblood glucose. Some of the approaches that have been ex-plored or are still being explored for engineering GRI centeron introducing glucose-dependent flexibility in the releaseof insulin from its injected depot (9,10,12). The approachwe have undertaken with the compound MK-2640 is tointroduce a capacity for glucose responsiveness onto thebackbone of native insulin itself to enable its uptake anddegradation by the MR pathway. In this specific regard, it isinteresting to consider that a major site for expression ofMR is on hepatic sinusoidal endothelial cells that are closelyadjacent to underlying hepatocytes, cell types separatednarrowly by the space of Disse (14,15). More studies areneeded to understand not only the systemic changes inthe PKs and PDs of a glucose-responsive analog such asMK-2640 but also the tissue- and organ-specific contribu-tions to glucose responsiveness, notably the contribution ofthe hepatic bed. Although a key site of drug disposal forMK-2640 within hepatic sinusoidal endothelial cells is likelydue to the high expression of MR in these cells, biodistri-bution studies and relatively high rates of GIR stimulatedduring dog clamp studies indicate that MK-2640 is not a“liver-selective” insulin analog per se.

The intent of the reported studies is to focus on proof ofpharmacology that engineering a novel GRI analog using astrategy of lectin-based glucose tunable clearance is feasible.Across the range of plasma glucose from 80 to 300 mg/dL,the clearance of MK-2640 changed by 30%, which is un-precedented, but how much glucose-based PK modula-tion of a GRI analog must be achieved to attenuate riskfor hypoglycemia is unknown. In the D minipig studies ofescalating dose until hypoglycemia ensued, a substantiallygreater therapeutic index was observed for MK-2640 thanfor RHI (threefold vs. 1.3-fold, respectively). In general, thegreater the amount of glucose-responsive PK change andthe greater the therapeutic index, the greater would seema potential to mitigate the risk for hypoglycemia. Whether thedegree of glucose responsiveness achieved by MK-2640 inpreclinical studies has sufficient potential to attain a meaning-ful reduction risk for hypoglycemia remains unknown.

In summary, proof of pharmacology preclinical studies,together with relevant in vitro assays, are presented thatdemonstrate preliminary feasibility in engineering glucoseresponsivity onto native insulin. The PKs of MK-2640 canchange by nearly one-third across a physiologically rele-vant range of plasma glucose values that would be experi-enced by an individual with insulin-requiring diabetes. Theglucose-dependent decrease in PKs at euglycemia comparedwith hyperglycemia was mediated by increases in theclearance of MK-2640 by MR and was associated withreduced insulin action on glucose metabolism. The firstlocus of control in insulin therapy is cognitive and re-sides with a physician and patient selecting an appropriatedosage, and the second locus of control with currentlyavailable insulin resides in its absorption kinetics becauseelimination kinetics are invariant. We hope that by adding

to these, a third modulus of control, that of glucose-responsive variance of insulin elimination kinetics once aGRI is already circulating in plasma, that a potential forincreased efficacy and safety can be introduced into insulintherapy.

Acknowledgments. The authors thank many Merck colleagues for theircontributions to this project, including Raul Camacho, Carlos G. Rodriguez, andJoel Mane for technical and scientific support during in vivo studies, XiaopingZhang and Steven Williams for IR-binding data, and Xun Shen for theMR-binding data. The authors wish to acknowledge Hsuan-shen Chen, HuaibingHe, Tina Santos, Ling Xu, Xiaofang Li, Xinchun Tong, and Bernard Choi, of theDiscovery Bioanalytics Group, Department of Pharmacokinetics, Pharmacody-namics and Drug Metabolism, Merck Research Laboratories, Rahway, NJ, forthe quantitative analysis of MK-2640, insulin, and epinephrine. The authors alsoacknowledge the valuable contributions of Merck colleagues Nancy Thornberry,Natarajan Sethumaran, Michael Meehl, and Gus Gustafson for early contribu-tions to the project, Paul Carrington for editorial comments, and Alan Cherrington,of Vanderbilt University, for insightful discussions throughout the course of thisproject.Duality of Interest. All of the authors were full-time employees of MerckSharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, and heldstock or stock options in the company during the conduct of these studies. MerckResearch Laboratories was the sponsor of this research. No other potential conflictsof interest relevant to this article were reported.Author Contributions. N.C.K., S.L., T.K., M.v.H., J.M., V.R., M.E., R.P.N.,and D.E.K. interpreted the data. N.C.K., T.K., M.v.H., J.M., E.C.-J., V.R., S.S., andD.E.K. designed the studies conducted at Merck Research Laboratories. N.C.K., L.Y.,T.K., M.v.H., J.M., M.W., G.D., Y.C., Y.Z., E.C.-J., V.R., P.Z., S.S., A.O., F.L., S.C.S.,and W.S. researched data. N.C.K., J.M., R.P.N., and D.E.K. wrote the manuscript.S.L., P.H., S.S., V.A., J.L.D., and R.P.N. led the medicinal chemistry research thatdiscovered and synthesized MK-2640. R.P.N. and D.E.K. are the guarantors of thiswork and as such had full access to all the data in the study and take responsibilityfor the integrity of the data and the accuracy of the data analysis.Prior Presentation. Parts of this study were presented in abstract form at the25th American Peptide Symposium, Whistler, BC, Canada, 17–22 June 2017.

References1. Frier BM. Hypoglycaemia in diabetes mellitus: epidemiology and clinical im-plications. Nat Rev Endocrinol 2014;10:711–7222. The Diabetes Control and Complications Trial Research Group. Hypoglycemia inthe Diabetes Control and Complications Trial. Diabetes 1997;46:271–2863. UK Hypoglycaemia Study Group. Risk of hypoglycaemia in types 1 and 2 di-abetes: effects of treatment modalities and their duration. Diabetologia 2007;50:1140–11474. Seaquist ER, Anderson J, Childs B, et al. Hypoglycemia and diabetes: a reportof a workgroup of the American Diabetes Association and the Endocrine Society.Diabetes Care 2013;36:1384–13955. Bolli GB, DeVries JH. New long-acting insulin analogs: from clamp studies toclinical practice. Diabetes Care 2015;38:541–5436. Heinemann L, Muchmore DB. Ultrafast-acting insulins: state of the art. J Di-abetes Sci Technol 2012;6:728–7427. Zaykov AN, Mayer JP, DiMarchi RD. Pursuit of a perfect insulin. Nat Rev DrugDiscov 2016;15:425–4398. Brownlee M, Cerami A. A glucose-controlled insulin-delivery system: semi-synthetic insulin bound to lectin. Science 1979;206:1190–11919. Chou DH, Webber MJ, Tang BC, et al. Glucose-responsive insulin activity bycovalent modification with aliphatic phenylboronic acid conjugates. Proc Natl AcadSci U S A 2015;112:2401–2406

diabetes.diabetesjournals.org Kaarsholm and Associates 307

10. Guo J, Sun H, Alt K, et al. Boronate-phenolic network capsules with dual re-sponse to acidic pH and cis-diols. Adv Healthc Mater 2015;4:1796–180111. Hu X, Yu J, Qian C, et al. H2O2-responsive vesicles integrated with trans-cutaneous patches for glucose-mediated insulin delivery. ACS Nano 2017;11:613–62012. Yu J, Zhang Y, Ye Y, et al. Microneedle-array patches loaded with hypoxia-sensitive vesicles provide fast glucose-responsive insulin delivery. Proc Natl Acad SciU S A 2015;112:8260–826513. Zion TC, Lancaster TL. Conjugate based systems for controlled drug delivery.Patent application WO2010088294 A1. 5 August 201014. Drickamer K, Taylor ME. Recent insights into structures and functions of C-typelectins in the immune system. Curr Opin Struct Biol 2015;34:26–3415. Taylor ME, Drickamer K. Convergent and divergent mechanisms of sugarrecognition across kingdoms. Curr Opin Struct Biol 2014;28:14–22

16. Kerrigan AM, Brown GD. C-type lectins and phagocytosis. Immunobiology2009;214:562–57517. Lee RT, Hsu TL, Huang SK, Hsieh SL, Wong CH, Lee YC. Survey of immune-related, mannose/fucose-binding C-type lectin receptors reveals widely divergentsugar-binding specificities. Glycobiology 2011;21:512–52018. Lee SJ, Evers S, Roeder D, et al. Mannose receptor-mediated regulation ofserum glycoprotein homeostasis. Science 2002;295:1898–190119. Martinez-Pomares L. The mannose receptor. J Leukoc Biol 2012;92:1177–118620. Firon N, Ashkenazi S, Mirelman D, Ofek I, Sharon N. Aromatic alpha-glycosidesof mannose are powerful inhibitors of the adherence of type 1 fimbriated Escherichiacoli to yeast and intestinal epithelial cells. Infect Immun 1987;55:472–47621. Lin S, Yan L, Kekec A, et al. Glucose-responsive insulin conjugates. PatentWO/2015/051052 A3. 2 July 2015.

308 Glucose-Responsive Insulin Diabetes Volume 67, February 2018