GPi688

• Molecular Formula: C19H18ClN3O4S
• Molecular Weight: 419.9 g/mol
• InChI: InChI=1S/C19H18ClN3O4S/c20-16-7-11-6-13(22-18(11)28-16)17(26)21-14-5-10-3-1-2-4-15(10)23(19(14)27)8-12(25)9-24/h1-4,6-7,12,14,22,24-25H,5,8-9H2,(H,21,26)
• InChIKey: UICNBXVDHCBKCE-UHFFFAOYSA-N

GPi688: Indole-Site Glycogen Phosphorylase Inhibitor

GPi688 (CAS 918902-32-6) is a chiral, allosteric glycogen phosphorylase (GP; EC 2.4.1.1) inhibitor that binds at the indole-site dimer interface of the enzyme [1]. It exhibits low-nanomolar potency against recombinant human liver, rat liver, and human skeletal muscle GPa isoforms with IC50 values of 19 nM, 61 nM, and 12 nM, respectively, in the presence of physiological glucose concentrations [1]. The compound was developed by AstraZeneca and is licensed for research use . GPi688 is orally bioactive, demonstrates glucagon-mediated glucose output inhibition in primary rat hepatocytes, and suppresses glucagon-driven hyperglycemia in both lean Wistar (65%) and obese Zucker (100%) rat models with a long duration of action [2].

• [1] Freeman S, Bartlett JB, Convey G, et al. Sensitivity of glycogen phosphorylase isoforms to indole site inhibitors is markedly dependent on the activation state of the enzyme. Br J Pharmacol. 2006;149(6):775-785. doi:10.1038/sj.bjp.0706925 View Source
• [2] Poucher SM, Freeman S, Loxham SJG, et al. An assessment of the in vivo efficacy of the glycogen phosphorylase inhibitor GPi688 in rat models of hyperglycaemia. Br J Pharmacol. 2007;152(8):1239-1247. doi:10.1038/sj.bjp.0707502 View Source

Why GPi688 Is Unique Among GP Inhibitors

Glycogen phosphorylase inhibitors are not pharmacologically interchangeable because they target distinct allosteric sites (indole site, AMP site, catalytic site) and exhibit fundamentally different isoform selectivity profiles that are modulated by the enzyme's activation state (T vs. R state) [1]. Even within the indole-site inhibitor subclass, subtle structural differences between compounds such as GPi688 and GPi819 produce divergent absolute potencies and differential sensitivity to AMP-driven shifts in IC50 across liver and skeletal muscle isoforms [1]. Moreover, indole-site inhibitors couple GP inhibition to glycogen synthase (GS) activation — a functional property absent in catalytic-site inhibitors such as DAB — making the choice of tool compound critical for interpreting downstream metabolic readouts [2]. The quantitative evidence below establishes exactly where GPi688 differentiates from its closest analogs and from mechanistically distinct comparator classes.

• [1] Freeman S, Bartlett JB, Convey G, et al. Sensitivity of glycogen phosphorylase isoforms to indole site inhibitors is markedly dependent on the activation state of the enzyme. Br J Pharmacol. 2006;149(6):775-785. doi:10.1038/sj.bjp.0706925 View Source
• [2] Poucher SM, Freeman S, Loxham SJG, et al. An assessment of the in vivo efficacy of the glycogen phosphorylase inhibitor GPi688 in rat models of hyperglycaemia. Br J Pharmacol. 2007;152(8):1239-1247. doi:10.1038/sj.bjp.0707502 View Source

GPi688 Quantitative Evidence Guide

Human Liver GPa Potency: GPi688 vs GPi819

In a direct head-to-head assay using purified recombinant human liver glycogen phosphorylase a (huLGPa) at 8 mM glucose without AMP, GPi688 (IC50 = 0.019 μM; 95% CI 0.011–0.034) was 1.8-fold more potent than GPi819 (IC50 = 0.034 μM; 95% CI 0.024–0.047) [1]. Against human skeletal muscle GPa (huSMGPa) at 15 mM glucose without AMP, GPi688 (IC50 = 0.012 μM; 95% CI 0.007–0.021) was 1.4-fold more potent than GPi819 (IC50 = 0.017 μM; 95% CI 0.007–0.044) [1]. Both compounds were evaluated in the same experimental system (fluorometric coupled-enzyme assay, glycogenolytic direction, 30 °C, pH 7.2), enabling direct potency comparison without cross-study normalization [1].

Evidence Dimension	Inhibitory potency (IC50) against recombinant human liver GPa (huLGPa) at 8 mM glucose, no AMP
Target Compound Data	GPi688 IC50 = 0.019 μM (95% CI: 0.011–0.034 μM)
Comparator Or Baseline	GPi819 IC50 = 0.034 μM (95% CI: 0.024–0.047 μM)
Quantified Difference	1.8-fold higher potency for GPi688 vs. GPi819 against huLGPa
Conditions	Purified recombinant human liver GPa; 8 mM glucose; 0 mM AMP; fluorometric coupled-enzyme assay at 30 °C, pH 7.2

Why This Matters

Researchers targeting hepatic glucose production in type 2 diabetes models may require the higher liver isoform potency of GPi688 to achieve target engagement at lower compound concentrations, reducing the risk of confounding off-target effects relative to GPi819.

• [1] Freeman S, Bartlett JB, Convey G, et al. Sensitivity of glycogen phosphorylase isoforms to indole site inhibitors is markedly dependent on the activation state of the enzyme. Br J Pharmacol. 2006;149(6):775-785. Table 1. doi:10.1038/sj.bjp.0706925 View Source

AMP-Induced Selectivity: Liver vs Muscle GPa

GPi688 exhibits a structure-dependent differential sensitivity to AMP-driven potency shifts across GPa isoforms. At 0.02 mM AMP (sub-maximal activation), the IC50 of GPi688 increased 6-fold for skeletal muscle GPa but only 2-fold for liver GPa — a 3:1 differential in isoform sensitivity [1]. At 0.2 mM AMP (maximal activation), the IC50 increased 28-fold for skeletal muscle vs. 7-fold for liver GPa — a 4:1 differential [1]. Consequently, at 0.2 mM AMP, GPi688 was ~2.5-fold selective for liver over skeletal muscle GPa [1]. GPi819 exhibited a qualitatively similar pattern but with distinct absolute IC50 values [1]. The closely related analog CP-316819 (GPi819), when tested in vivo after 5 days of dosing in Zucker fa/fa rats, produced 1.5-fold greater glycogen elevation in soleus muscle than in liver (P < 0.05), confirming that the in vitro selectivity profile does not fully translate to in vivo tissue targeting [1].

Evidence Dimension	Ratio of IC50 shift (fold increase) under AMP activation: skeletal muscle GPa vs. liver GPa
Target Compound Data	GPi688: 0.02 mM AMP → 6-fold (SM) vs. 2-fold (liver); 0.2 mM AMP → 28-fold (SM) vs. 7-fold (liver); net liver selectivity factor ~2.5 at 0.2 mM AMP
Comparator Or Baseline	GPi819: qualitatively similar AMP-induced shift pattern; in vivo glycogen elevation 1.5-fold greater in soleus muscle than liver (P < 0.05) after 5-day dosing at 37.5 μmol·kg⁻¹
Quantified Difference	GPi688 liver-to-SM selectivity differential = 3:1 (0.02 mM AMP) and 4:1 (0.2 mM AMP); GPi819 in vivo data shows divergent muscle-predominant glycogen deposition
Conditions	Purified recombinant human liver and skeletal muscle GPa; 8 mM glucose (liver) / 15 mM glucose (SM); varying AMP (0–0.2 mM); fluorometric assay

Why This Matters

The graded, AMP-dependent selectivity shift of GPi688 enables researchers to model liver-predominant GP inhibition under defined activation states, a critical parameter when designing experiments that must distinguish hepatic from skeletal muscle glycogenolysis — especially relevant for safety pharmacology where excessive muscle glycogen accumulation is undesirable.

• [1] Freeman S, Bartlett JB, Convey G, et al. Sensitivity of glycogen phosphorylase isoforms to indole site inhibitors is markedly dependent on the activation state of the enzyme. Br J Pharmacol. 2006;149(6):775-785. Figures 5a-b; Table 1. doi:10.1038/sj.bjp.0706925 View Source

In Vivo Efficacy: Glucagon Challenge vs OGTT

GPi688 was evaluated against two distinct hyperglycemic challenges in vivo, revealing a clear fasting-over-prandial efficacy profile. In a glucagon-mediated hyperglycemia model, GPi688 inhibited the glucagon challenge by 65% in Wistar rats and 100% in obese Zucker rats, with a long duration of action observed in the Zucker model [1]. By contrast, in an oral glucose tolerance test (OGTT) in Zucker rats, GPi688 reduced glucose excursion by only 7%, compared to a 22% reduction achieved by a GSK-3 inhibitor under the same protocol [1]. This 3.1-fold difference in OGTT efficacy (7% vs. 22%) occurred despite GPi688 producing a measurable 45% activation of glycogen synthase in vivo [1]. Additionally, GPi688 reduced fasting blood glucose in obese Zucker rats after a 7 h fast by 23% [1]. These data collectively demonstrate that GPi688 preferentially suppresses hepatic glucose production (glucagon-driven) rather than enhancing peripheral glucose disposal (insulin/glucose-driven), representing a distinct in vivo pharmacological signature.

Evidence Dimension	In vivo glycemic efficacy: glucagon challenge vs. oral glucose tolerance test in Zucker fa/fa rats
Target Compound Data	GPi688: 65% (Wistar) and 100% (Zucker) inhibition of glucagon challenge; 23% fasting glucose reduction (7 h); 7% OGTT glucose reduction; 45% GS activation in vivo
Comparator Or Baseline	GSK-3 inhibitor: 22% OGTT glucose reduction (same Zucker rat model)
Quantified Difference	3.1-fold lower OGTT efficacy vs. GSK-3 inhibitor (7% vs. 22%); contrasting near-complete glucagon challenge suppression (100%) vs. minimal OGTT effect (7%) — a 14.3:1 within-compound efficacy ratio
Conditions	Glucagon challenge model: Wistar and obese Zucker rats, compound dosed orally; OGTT: Zucker fa/fa rats, oral glucose load; 7 h fast for fasting glucose measurement

Why This Matters

This differential efficacy profile establishes GPi688 as the preferred tool compound for research focused specifically on fasting hyperglycemia and hepatic glucose overproduction — rather than post-prandial glucose disposal — enabling investigators to select the correct pharmacological tool for their specific metabolic research question.

• [1] Poucher SM, Freeman S, Loxham SJG, et al. An assessment of the in vivo efficacy of the glycogen phosphorylase inhibitor GPi688 in rat models of hyperglycaemia. Br J Pharmacol. 2007;152(8):1239-1247. doi:10.1038/sj.bjp.0707502 View Source

Glycogen Synthase Activation: Indole-Site vs Catalytic-Site

GPi688 treatment of rat primary hepatocytes inhibited glucagon-mediated glucose output in a concentration-dependent manner while simultaneously reducing GP activity and producing a seven-fold increase in glycogen synthase (GS) activity [1]. This functional coupling — GP inhibition paired with GS activation — is a hallmark of indole-site allosteric inhibitors [1]. In contrast, catalytic-site GP inhibitors such as the imino sugar DAB (1,4-dideoxy-1,4-imino-D-arabinitol) do not demonstrate direct or indirect inhibition of gluconeogenesis in hepatocytes, perfused rat liver, whole rats, or conscious dogs, and lack evidence of GS activation [1]. This mechanistic bifurcation between allosteric (indole-site) and orthosteric (catalytic-site) GP inhibitors has direct consequences for experimental interpretation: indole-site compounds like GPi688 produce a dual metabolic signal (reduced glycogenolysis + enhanced glycogen synthesis), whereas catalytic-site inhibitors produce only a unidirectional signal (reduced glycogenolysis alone).

Evidence Dimension	Glycogen synthase (GS) activity fold-change in rat primary hepatocytes upon GP inhibitor treatment
Target Compound Data	GPi688 (indole-site inhibitor): 7-fold increase in GS activity, concomitant with GP inhibition and suppression of glucagon-mediated glucose output
Comparator Or Baseline	DAB / catalytic-site GP inhibitors: no detectable GS activation; no evidence of gluconeogenesis inhibition in hepatocytes, perfused liver, or in vivo models (literature-based class property)
Quantified Difference	Qualitative mechanistic difference: dual GP inhibition + GS activation (indole-site class) vs. GP inhibition only (catalytic-site class). Quantitative GS activation: 7-fold (GPi688) vs. no measurable activation (DAB class).
Conditions	Rat primary hepatocytes; glucagon-stimulated glucose output assay; GP and GS activity measured enzymatically

Why This Matters

Investigators studying the interplay between glycogenolysis and glycogen synthesis must select an indole-site inhibitor such as GPi688 to capture the full metabolic response; using a catalytic-site inhibitor would miss the GS activation component and could lead to incorrect conclusions about the net effect of GP inhibition on hepatic glycogen flux.

• [1] Poucher SM, Freeman S, Loxham SJG, et al. An assessment of the in vivo efficacy of the glycogen phosphorylase inhibitor GPi688 in rat models of hyperglycaemia. Br J Pharmacol. 2007;152(8):1239-1247. doi:10.1038/sj.bjp.0707502 (GS activation data and DAB comparison cited within.) View Source

PK/PD Model: Free Concentration and Glucagon Inhibition

GPi688 is one of the few glycogen phosphorylase inhibitors for which a validated, quantitative PK/PD relationship has been established linking in vitro potency to in vivo efficacy. The equation: % glucagon inhibition = 56.9 + 34.3 × log([free plasma GPi688] / rat liver GPa IC50) yielded a correlation coefficient of r = 0.921 across multiple dose levels and time points in the Zucker rat glucagon challenge model [1]. This log-linear model enables researchers to prospectively estimate the in vivo glucagon-suppression efficacy of GPi688 from measured or predicted free plasma concentrations and the rat liver GPa IC50 (61 nM), facilitating rational dose selection for in vivo metabolic studies [1]. Comparable validated PK/PD models are not publicly available for many other indole-site GP inhibitors, making GPi688 a uniquely predictable tool compound for translational pharmacology experiments.

Evidence Dimension	Validated PK/PD model for in vivo glucagon challenge efficacy prediction
Target Compound Data	GPi688: % glucagon inhibition = 56.9 + 34.3 × log([free plasma] / 61 nM); r = 0.921 (Zucker rat model)
Comparator Or Baseline	No equivalent validated PK/PD model publicly reported for GPi819, DAB, or other GP inhibitor classes in the glucagon challenge model (class-level absence of data)
Quantified Difference	Model r² = 0.848, indicating that ~85% of the variance in in vivo efficacy is explained by free plasma concentration and rat IC50 alone
Conditions	Obese Zucker fa/fa rats; glucagon challenge model; free plasma GPi688 measured by LC-MS; rat liver GPa IC50 = 61 nM

Why This Matters

The availability of a robust PK/PD model makes GPi688 the most predictable glycogen phosphorylase inhibitor for in vivo study design, reducing the number of pilot experiments needed to establish effective dosing regimens and enabling more efficient use of animal resources.

• [1] Poucher SM, Freeman S, Loxham SJG, et al. An assessment of the in vivo efficacy of the glycogen phosphorylase inhibitor GPi688 in rat models of hyperglycaemia. Br J Pharmacol. 2007;152(8):1239-1247. doi:10.1038/sj.bjp.0707502 View Source

GPi688 Research and Industrial Applications

Fasting Hyperglycemia in Type 2 Diabetes Models

GPi688 is the tool compound of choice for studies targeting fasting hyperglycemia driven by excessive hepatic glucose production. The compound suppressed glucagon-mediated hyperglycemia by 100% in obese Zucker rats and reduced 7 h fasting blood glucose by 23%, while showing only 7% efficacy in an oral glucose tolerance test [1]. This fasting-over-prandial efficacy profile makes GPi688 ideal for experimental protocols focused on overnight or prolonged fasting metabolic states, where hepatic glycogenolysis is the dominant source of circulating glucose. Researchers studying the contribution of hepatic glycogen phosphorylase to fasting glycemia in genetic (Zucker fa/fa, ob/ob) or diet-induced obesity models will benefit from GPi688's validated in vivo efficacy and established PK/PD model [1].

Glucagon Signaling and GP-GS Coupling

GPi688 is uniquely suited for experiments investigating the coordinated regulation of glycogen phosphorylase and glycogen synthase in hepatocytes. The compound simultaneously inhibits GP activity and produces a seven-fold increase in GS activity, a dual functional effect that catalytic-site GP inhibitors such as DAB do not exhibit [1]. This makes GPi688 the appropriate selection for studies requiring concomitant modulation of both arms of glycogen metabolism — glycogen breakdown and synthesis — including investigations of glucagon-to-insulin signaling crosstalk, hepatic glucose flux partitioning, and the role of the GP-GS regulatory phosphatase PP1 in metabolic control [1]. The validated in vitro potency data (human liver GPa IC50 = 19 nM) also supports precise concentration-response experimental designs [2].

Translational PK/PD Modeling

GPi688 is the glycogen phosphorylase inhibitor of choice for pharmacokinetic-pharmacodynamic (PK/PD) modeling studies, as it is supported by a published, high-confidence quantitative relationship between free plasma drug concentration and in vivo glucagon suppression efficacy (r = 0.921) [1]. This enables researchers to prospectively calculate target plasma exposures needed to achieve a desired degree of glucagon challenge inhibition, using the equation: % inhibition = 56.9 + 34.3 × log([free plasma] / rat IC50). No comparable PK/PD model has been publicly validated for other indole-site or catalytic-site GP inhibitors, making GPi688 essential for any study where quantitative predictability of in vivo pharmacology is a priority [1].

Isoform-Selective Inhibition: Liver vs Muscle Glycogenolysis

GPi688 provides a quantifiable, AMP-dependent liver-over-muscle selectivity profile that can be exploited in experiments requiring differential suppression of hepatic vs. skeletal muscle glycogen phosphorylase. At 0.2 mM AMP, GPi688 is ~2.5-fold more potent against liver GPa than skeletal muscle GPa [2]. The graded loss of potency with increasing AMP concentration (28-fold IC50 increase for skeletal muscle vs. 7-fold for liver at 0.2 mM AMP) allows researchers to titrate isoform selectivity by modulating the activation state of the enzyme preparation through AMP or glucose concentration [2]. This property is particularly valuable for safety pharmacology studies examining the risk of skeletal muscle glycogen accumulation, as observed in vivo with the related inhibitor GPi819 (1.5-fold greater muscle glycogen elevation vs. liver; P < 0.05) [2].

• [1] Poucher SM, Freeman S, Loxham SJG, et al. An assessment of the in vivo efficacy of the glycogen phosphorylase inhibitor GPi688 in rat models of hyperglycaemia. Br J Pharmacol. 2007;152(8):1239-1247. doi:10.1038/sj.bjp.0707502 View Source
• [2] Freeman S, Bartlett JB, Convey G, et al. Sensitivity of glycogen phosphorylase isoforms to indole site inhibitors is markedly dependent on the activation state of the enzyme. Br J Pharmacol. 2006;149(6):775-785. doi:10.1038/sj.bjp.0706925 View Source