A gut-restricted glutamate carboxypeptidase II inhibitor reduces monocytic inflammation and improves preclinical colitis

Abstract

There is an urgent need to develop therapeutics for inflammatory bowel disease (IBD) because up to 40% of patients with moderate-to-severe IBD are not adequately controlled with existing drugs. Glutamate carboxypeptidase II (GCPII) has emerged as a promising therapeutic target. This enzyme is minimally expressed in normal ileum and colon, but it is markedly up-regulated in biopsies from patients with IBD and preclinical colitis models. Here, we generated a class of GCPII inhibitors designed to be gut-restricted for oral administration, and we interrogated efficacy and mechanism using in vitro and in vivo models. The lead inhibitor, (S)-IBD3540, was potent (half maximal inhibitory concentration = 4 nanomolar), selective, gut-restricted (AUC_colon/plasma > 50 in mice with colitis), and efficacious in acute and chronic rodent colitis models. In dextran sulfate sodium-induced colitis, oral (S)-IBD3540 inhibited >75% of colon GCPII activity, dose-dependently improved gross and histologic disease, and markedly attenuated monocytic inflammation. In spontaneous colitis in interleukin-10 (IL-10) knockout mice, once-daily oral (S)-IBD3540 initiated after disease onset improved disease, normalized colon histology, and attenuated inflammation as evidenced by reduced fecal lipocalin 2 and colon pro-inflammatory cytokines/chemokines, including tumor necrosis factor-α and IL-17. Using primary human colon epithelial air-liquid interface monolayers to interrogate the mechanism, we further found that (S)-IBD3540 protected against submersion-induced oxidative stress injury by decreasing barrier permeability, normalizing tight junction protein expression, and reducing procaspase-3 activation. Together, this work demonstrated that local inhibition of dysregulated gastrointestinal GCPII using the gut-restricted, orally active, small-molecule (S)-IBD3540 is a promising approach for IBD treatment.

Fig. 1.. Synthesis of SBA-GCPII inhibitor conjugates and efficacy assessment in DSS colitis.

(A) Synthetic scheme for generation of GCPII inhibitor, 2-PMPA, and LCA (8a), DCA (8b), or UDCA (8c) conjugates. (B to D) Efficacy evaluation in the 2.5% DSS colitis mouse model at a 10 mg/kg of 2-PMPA molar equivalent (eq.) dose. Body weight, stool consistency, and rectal bleeding were scored daily and summated to calculate the disease activity index (DAI). DCA–2-PMPA (IBD3540) was identified as an active compound and was evaluated head to head versus its constituents, (E) 2-PMPA and DCA, and clinical agents (F) sulfasalazine and (G) tofacitinib, at doses indicated. All test articles were administered once daily by oral gavage beginning on study day 0, except for tofacitinib, which was administered twice daily. (B to G) n = 5 to 15 per group; (B to D and G) repeated measures (RM) two-way ANOVA with post hoc Bonferroni’s multiple comparisons test; (E and F) mixed effects model (REML, restricted maximum likelihood) with post hoc Bonferroni’s multiple comparisons test; *P < 0.05, **P < 0.01, and ***P < 0.001. Studies were performed in duplicate with representative data shown. RT, room temperature; DCM, dichloromethane; DMF, N,N-dimethylformamide; THF, tetrahydrofuran.

Fig. 2.. (S)-IBD3540 is 150-fold more potent than (R)-IBD3540.

(A) Separation of (R/S) enantiomers of intermediate 6 by chiral chromatography, followed by conjugation to DCA and deprotection of benzyl groups yielding (S)-IBD3540 8A and (R)-IBD3540 8B. (B and C) Recombinant human GCPII inhibition curves for purified diastereoisomers used to calculate IC₅₀ values [(S)-IBD3540 IC₅₀ = 4 ± 0.1 nM; (R)-IBD3540 IC₅₀ = 600 ± 30 nM].

Fig. 3.. (S)-IBD3540 is gut-restricted under both normal and DSS-permeabilized GI conditions.

Colon and plasma were collected at multiple time points after a single oral administration of 100 mg/kg eq. (S)-IBD3540 to male C57BL/6NHsd mice pretreated with freshwater (intact GI barrier, blue) or 4% DSS (compromised GI barrier, red) for 5 days, followed by LC-MS/MS–based compound quantification. Pharmacokinetic time courses of (A) parent (S)-IBD3540 and (B) metabolite 2-PMPA were used to calculate (C) colon and plasma C_max, T_max, AUC_0–t, and colon:plasma AUC ratios. Both (S)-IBD3540 and 2-PMPA liberated from (S)-IBD3540 were gut-restricted as demonstrated by colon:plasma AUC ratios of >16 for all conditions. (A and B) n = 3 per time point. LLOQ, lower limit of quantification.

Fig. 4.. Oral (S)-IBD3540 displays dose-dependent activity in the DSS colitis mouse model and has superior activity to constituents 2-PMPA and DCA.

(A to D) To assess dose dependency of anti-colitis activity, vehicle and 1, 10, or 100 mg/kg eq. (S)-IBD3540 were administered once daily by oral gavage in C57BL/6NHsd mice exposed to 4% DSS in drinking water. (A) DAI, composed of (B) stool consistency score, (C) rectal bleeding score, and (D) body weight score, was monitored daily. On study day 6, mice were euthanized 4 hours after the final dose. (E) Colon length was measured and (F) GCPII enzymatic activity in the colon was quantified using a radiosubstrate assay. (G and H) To confirm the minimum effective dose of (S)-IBD3540 in this model and to examine comparative efficacy, a follow-up study was performed where 10 mg/kg eq. (S)-IBD3540, 10 mg/kg eq. DCA, or 10 mg/kg 2-PMPA was administered once daily by oral gavage with measurement of (G) DAI and (H) terminal colon length. (A to E) n = 13 to 15 per group; (F) n = 8 per group; (G and H) n = 10 to 15 per group; (A to D, G) two-way ANOVA with post hoc Bonferroni’s multiple comparisons test; (E and F) Kruskal-Wallis with post hoc Dunn’s multiple comparisons test; (H) ordinary one-way ANOVA with post hoc Bonferroni’s multiple comparisons test; *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 versus vehicle.

Fig. 5.. Oral (S)-IBD3540 is anti-inflammatory in DSS colitis mouse models.

Anti-inflammatory activity was assessed in acute severe (4% DSS, male C57BL/6NHsd) and acute mild/moderate (2.5% DSS, female C57BL/6NCrl) colitis. (A to C) Daily oral 100 mg/kg eq. (S)-IBD3540, beginning on study day 0, improved (A) blinded histopathology scores for ulceration and inflammation on study day 6 as depicted in (B and C) representative images of hematoxylin and eosin (H&E)–stained colon. Scale bars, 100 μm. (D and E) Daily oral 100 mg/kg eq. (S)-IBD3540, initiated after disease onset on study day 2, improved (D) blinded histopathology scores on study day 7 and (E) reduced pro-inflammatory cytokine interleukin-1β (IL-1β) as measured by Luminex multiplex immunoassay. (F to L) Flow cytometry of colon lamina propria mononuclear cells isolated on 4% DSS, study day 5, shown as % of total viable cells for (F) CD3⁺ T lymphocytes, (G) CD11b⁺ myeloid cells, and CD11b⁺ subpopulations including (H) neutrophils, (I) circulating monocytes, (J) transitioning monocytes, (K) dendritic cells, and (L) resident macrophages. (A) n = 10 per group; Mann-Whitney U test comparing sum histology scores; (D) n = 10 per group; two-tailed t test comparing sum histology scores; (E) n = 5 to 10 per group; ordinary one-way ANOVAwith post hoc Bonferroni’s multiple comparisons test; (F to L) n = 4 to 8 per group; ordinary two-way ANOVA with post hoc Tukey’s honestly significant difference test; *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. ns, not significant.

Fig. 6.. Oral (S)-IBD3540 attenuates established colitis in IL-10^−/− mice.

(A) Stool consistency was monitored in IL-10^−/− mice before treatment (black) or after randomization to once-daily vehicle PO (red) or 100 mg/kg eq. (S)-IBD3540 PO (blue) treatment groups. (B) Quantitative ELISA for GI inflammatory biomarker lipocalin 2 (LCN2). (C) Gross colon weight-to-length ratios. (D) Blinded histopathological scores of location-matched distal colon H&E sections. (E) Results of a 31-plex cytokine/chemokine assay performed on location-matched mid-colon segments. (A) n = 7 to 10 per group; two-way ANOVA with Fisher’s least significant difference test; (B) n = 4 or 5 per group; Grubb’s outlier test followed by two-way ANOVA with post hoc Bonferroni’s multiple comparisons test; (C) n = 4 to 6 per group; unpaired two-tailed t test; (D) n = 4 to 7 per group; normalcy determined by Shapiro-Wilk test; TNF-α, IL-12p40, IL-17, LIF, LIX, IP-10, MIP-2 one-way ANOVA with post hoc Bonferroni’s multiple comparisons test; VEGF Kruskal-Wallis with post hoc Dunn’s multiple comparisons test; *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Fig. 7.. GCPII is up-regulated in the gastrointestinal epithelium of patients with CD and UC.

GCPII IHC was performed on commercially acquired full-thickness biopsies of patients with CD or UC and non-IBD controls. GCPII expression (brown) in (A) non-IBD ileum, (B) CD ileum, (C) non-IBD colon, and (D) UC colon. Representative images were shown from both architecturally intact and lesional areas. Arrows point to GCPII-positive epithelial cells and asterisks depict GCPII in epithelial crypt cells. Scale bars, 50 μM. (E) Scoring of GCPII expression by two trained investigators blinded to diagnosis, who separately scored membrane-associated and cytoplasmic signal, using a scale of 0 = no signal, 1 = positive signal in <10% of epithelial cells, and 2 = positive signal in >10% of epithelial cells, which were summated, yielding a maximum score of 4 (n = 11 to 16; two-tailed Mann-Whitney U test).

Fig. 8.. (S)-IBD3540 and 2-PMPA protect against media submersion–induced oxidative stress injury in human colon epithelial ALI monolayers.

(A) Monolayers before (ALI) or after 8-hour resubmersion with media containing vehicle (DMSO) or (S)-IBD3540 (1 to 100 μM). Left: H&E stain. Right: Immunofluorescence for ZO-1 (red), E-cadherin (green), and 4′,6-diamidino-2-phenylindole (DAPI)/nuclei (blue); n = 7. (B) Monolayer height relative to unsubmerged monolayers (ALI) (n = 6; *P < 0.05, **P < 0.01, ****P < 0.0001 versus vehicle). (C) Paracellular permeability of 4-kDa fluorescein isothiocyanate (FITC)–dextran and 70-kDa rhodamine-dextran 8 hours after resubmersion relative to vehicle (n = 3; *P < 0.05, **P < 0.01 versus 4 kDa vehicle; ++P < 0.01, +++P < 0.001 versus 70 kDa vehicle). (D) ALI monolayer immunoblots of cleaved caspase-3; n = 3. (E) Monolayers before (ALI) or after 8-hour resubmersion with media containing vehicle (DMSO) or 10 μM (S)-IBD3540, 2-PMPA, or DCA. Left: H&E stain. Right: Immunofluorescence for ZO-1 (red), E-cadherin (green), and DAPI/nuclei (blue); n = 5. (F) Monolayer height, relative to unsubmerged monolayers (ALI). (n = 3; *P < 0.05 versus vehicle). (G) Dextran paracellular permeability as above (n = 3; *P < 0.05 versus 4-kDa vehicle, **P < 0.01 versus 4-kDa vehicle, ++P < 0.01 versus 70-kDa vehicle, +++P < 0.001 versus 70-kDa vehicle). (H) ALI monolayer immunoblots of caspase-3 activation (n = 4; *P < 0.05, **P < 0.01 versus vehicle). Statistical significance is determined by RM two-way ANOVA with Dunnett’s multiple comparisons test (B, C, F, and G), Friedman test with Dunn’s multiple comparisons test (D), or one-way ANOVA with Dunnett’s multiple comparisons test (H).