It is anticipated that gamma-secretase inhibitors (gamma-Sec-I) that modulate Notch processing will alter differentiation in tissues whose architecture is governed by Notch signaling. To explore this hypothesis, Han Wistar rats were dosed for up to 5 days with 10-100 micromol/kg b.i.d. gamma-Sec-I from three chemical series that inhibit Notch processing in vitro at various potencies (Notch IC(50)). These included an arylsulfonamide (AS) (142 nM), a dibenzazepine (DBZ) (1.7 nM), and a benzodiazepine (BZ) (2.2 nM). The DBZ and BZ caused dose-dependent intestinal goblet cell metaplasia. In contrast, the AS produced no detectable in vivo toxicity, despite higher exposure to free drug. In a time course using BZ, small intestinal crypt cell and large intestinal glandular cell epithelial apoptosis was observed on days 1-5, followed by goblet cell metaplasia on days 2-5 and crypt epithelial and glandular epithelial regenerative hyperplasia on days 4-5. Gene expression profiling of duodenal samples from BZ-dosed animals revealed significant time-dependent deregulation of mRNAs for various panendocrine, hormonal, and transcription factor genes. Somatostatin, secretin, mucin, CCK, and gastrin mRNAs were elevated twofold or more by day 2, and a number of candidate "early-predictive" genes were altered on days 1-2, remaining changed for 4-5 days; these included Delta1, NeuroD, Hes1-regulated adipsin, and the Hes-regulated transcriptional activator of gut secretory lineage differentiation, the rat homolog of Drosophila atonal, Rath1. Western blotting of fecal protein from BZ-and DBZ-dosed animals exhibited increased levels of both anti-Rath1 reactive protein and anti-adipsin reactive proteins, confirming their potential value as noninvasive biomarkers of intestinal goblet metaplasia.

Several cationic amphiphilic drugs cause local or systemic phospholipidosis (PLD) after chronic exposure in preclinical species. PLD is characterized by the accumulation of drug, phospholipid, and concentric lamellar bodies in cellular lysosomes. We have developed a fluorescence-based in vitro screen that is predictive of PLD using the Cellomics ArrayScan high-content screening platform, which captures and analyzes images from 96-well cell culture microtiter plates using multichannel fluorescence microscopy. I-13.35 adherent mouse spleen macrophage cells were cultured with drug and a fluorescently tagged phospholipid, N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)-1,2-dihexadecanoyl-sn-glycero-3-phosphoethanolamine (NBD-PE). Drug concentrations were used in a range from 1 to 100 micro mol/L. After 24 h incubations, the cells were fixed with formalin. NBD-PE uptake was quantified in controls and treated cells. Nuclei were identified by Hoechst 33258 staining and dead cells were identified using ethidium homodimer-2 incorporation. Thus, confounding accumulation of NBD-PE due to cytotoxicity that produces false-positive results at high concentrations was eliminated from quantitation by ethidium staining and employing cell gating (dead cell rejection). The assay was found to be both sensitive and selective in that 26 of 28 positive, phospholipidogenic controls and 8 of 8 negative, non-phospholipidogenic controls were correctly called.

Novel non-fluoroquinolone inhibitors of bacterial type II topoisomerases (DNA gyrase and topoisomerase IV) are of interest for the development of new antibacterial agents that are not impacted by target-mediated cross-resistance with fluoroquinolones. N-Linked amino piperidines, such as 7a, generally show potent antibacterial activity, including against quinolone-resistant isolates, but suffer from hERG inhibition (IC(50) = 44 μM for 7a) and QT prolongation in vivo. We now disclose the finding that new analogues of 7a with reduced pK(a) due to substitution with an electron-withdrawing substituent in the piperidine moiety, such as R,S-7c, retained the Gram-positive activity of 7a but showed significantly less hERG inhibition (IC(50) = 233 μM for R,S-7c). This compound exhibited moderate clearance in dog, promising efficacy against a MRSA strain in a mouse infection model, and an improved in vivo QT profile as measured in a guinea pig in vivo model. As a result of its promising activity, R,S-7c was advanced into phase I clinical studies.

Novel glutathione (GSH) analogs, previously shown to inhibit glutathione S-transferase (GST) activity at about 1 microM in vitro, were tested for their ability to potentiate the killing of cultured tumor cells by chemotherapeutic drugs. When tested at doses up to 200 microM, the analogs were neither toxic nor capable of potentiating drug toxicity unless the diethyl ester (DEE) form was used for treatment of the cells. HPLC analysis revealed rapid internalization of the DEE and intracellular conversion to a monoethyl ester form that accumulated in the cell, followed by a more gradual loss of the second ester to generate the active parent form. For the four GSH analogs tested, the ability of the DEE forms to potentiate chlorambucil (CMB) toxicity in HT-29 human colon adenocarcinoma cells strongly correlated with the in vitro ability of the parent form to inhibit recombinant human P1-1. This isozyme is the dominant form of GST present in HT-29 cells. Of the four analog DEEs tested, gamma-glutamyl-S-(benzyl)cysteinyl-R(-)-phenyl glycine (TER 117) DEE was the most effective in potentiating CMB toxicity in several cell lines: HT-29, HT4-1 (HT-29 subclone), SKOV-3 ovarian carcinoma, and SK VLB (vinblastine-resistant variant of SKOV-3) cells. gamma-Glutamyl-S-(octyl)cysteinyl-glycine (TER 143) DEE potentiated mitomycin C (MTC) toxicity in HT4-1 and SK VLB cells while TER 117 DEE did not. TER 117 DEE enhanced melphalan effects on xenografts of HT4-1 in mice to a similar extent as that achieved with the previously described nonspecific GST inhibitor, ethacrynic acid. Taken together, our results indicate that cell-permeable analogs of GSH can potentiate cytotoxicity of common chemotherapeutic drugs and this effect has a strong positive correlation with the ability of the analogs to inhibit specific GST isozymes.

Despite six decades of clinical experience with the polymyxin class of antibiotics, their dose-limiting nephrotoxicity remains difficult to predict due to a paucity of sensitive biomarkers. Here, we evaluate the performance of standard of care and next-generation biomarkers of renal injury in the detection and monitoring of polymyxin-induced acute kidney injury in male Han Wistar rats using colistin (polymyxin E) and a polymyxin B (PMB) derivative with reduced nephrotoxicity, PMB nonapeptide (PMBN). This study provides the first histopathological and biomarker analysis of PMBN, an important test of the hypothesis that fatty acid modifications and charge reductions in polymyxins can reduce their nephrotoxicity. The results indicate that alterations in a panel of urinary kidney injury biomarkers can be used to monitor histopathological injury, with Kim-1 and α-GST emerging as the most sensitive biomarkers outperforming clinical standards of care, serum or plasma creatinine and blood urea nitrogen. To enable the prediction of polymyxin-induced nephrotoxicity, an in vitro cytotoxicity assay was employed using human proximal tubule epithelial cells (HK-2). Cytotoxicity data in these HK-2 cells correlated with the renal toxicity detected via safety biomarker data and histopathological evaluation, suggesting that in vitro and in vivo methods can be incorporated within a screening cascade to prioritize polymyxin class analogs with more favorable renal toxicity profiles.