SUMMARY Efforts to identify and target glioblastoma (GBM) drivers have primarily focused on receptor tyrosine kinases (RTKs). Clinical benefits, however, have been elusive. Here, we identify a SRY-related box 2 (SOX2) transcriptional regulatory network that is independent of upstream RTKs and is capable of driving glioma initiating cells. We identified oligodendrocyte lineage transcription factor 2 (OLIG2) and zinc finger E-box binding homeobox 1 (ZEB1) as potential SOX2 targets, which are frequently co-expressed irrespective of driver mutations. In murine glioma models, we show that different combinations of tumor suppressor and oncogene mutations can activate Sox2, Olig2, and Zeb1 expression. We demonstrate that ectopic co-expression of the three transcription factors can transform tumor suppressor deficient astrocytes into glioma initiating cells in the absence of an upstream RTK oncogene. Finally, we demonstrate that the transcriptional inhibitor mithramycin downregulates SOX2 and its target genes, resulting in markedly reduced proliferation of GBM cells in vivo.

No abstract

We examined inhibitory effects of ofloxacin (OFX), orbifloxacin (OBFX), ciprofloxacin (CFX), enrofloxacin (EFX) and norfloxacin (NFX) on cytochrome P-450 1A (CYP1A) activities using hepatic microsomes from four beagle dogs. Ethoxyresorufin O-de-ethylation was referred as CYP1A activities. All the fluoroquinolones inhibited the reaction in a noncompetitive manner. The determined inhibitory constants were the followings; 10.1 +/- 3.8 mM for OFX, 6.43 +/- 2.01 mM for OBFX, 0.726 +/- 0.134 mM for CFX, 4.06 +/- 1.19 mM for EFX and 4.75 +/- 1.63 mM for NFX respectively. As these values are>100-fold of plasma concentrations after a clinical single dose of the fluoroquinolones, it is suggested that the inhibitory effect on CYP1A activities is not so high to elicit drug-drug interaction with CYP1A substrates, when these fluoroquinolones are co-administered. Mechanism based inhibition was also examined in this study. Of the five fluoroquinolones examined, OFX, OBFX and CFX had this inhibition manner. As this inhibition is irreversible, inhibitory effects of the three fluoroquinolones may accumulate, when they are repeatedly administered. Therefore, OFX, OBFX and CFX may result in substantial drug-drug interaction with a CYP1A substrate even in clinical states. As EFX is metabolized to CFX in the body, it may also have the same possibility.

In this study we examined activities of cytochrome P450 (CYP)1A, 2C, 2D and 3A using hepatic microsomes from five male and five female cats. CYP1A, 2C, 2D and 3A activities were referred by ethoxyresorufin O-deethylation (EROD), tolbutamide hydroxylation (TBH), bufuralol 1'-hydroxylation (BLH) and midazolam 1'- and 4-hydroxylation respectively. The anti-rat CYP1A2 and CYP3A2 serum significantly inhibited EROD and midazolam 1'- and 4-hydroxylation, suggesting that EROD and midazolam 1'- and 4-hydroxylation were catalysed by CYP1A and 3A in cats respectively. Quinidine inhibited BLH in cats microsomes at quite low concentrations, suggesting that BLH was catalysed by CYP2D in cats. Tolbutamide hydroxylation activities were negligible in hepatic microsomes from both male and female cats, suggesting CYP2C activities of cats are extremely low. This suggests that CYP2C substrates should be carefully administered to cats. Although there is no sexual difference in CYP1A activities, there are differences in CYP2D and 3A activities of cats. CYP2D activities were higher (3-fold), but CYP3A activities were lower (one-fifth) in female cats. These results might suggest that CYP2D and 3A substrates should be prescribed for male and female cats using different dosage regimen.

Inhibitory effects of several fluoroquinolones (FQs) on liver CYP3A activities were examined by in vitro and in vivo tests in dogs. Midazolam (MDZ) hydroxylation rate was used to determine the CYP3A activities in liver microsomes. Enrofloxacin (EFX), ofloxacin (OFX) orbifloxacin (OBFX) and ciprofloxacin (CFX) were tested. None of the FQs changed Vmax, Km or intrinsic clearance (Vmax/Km) of MDZ. For in vivo test, we examined the effects of oral administration of EFX and OFX on the pharmacokinetics of quinidine (QN), a CYP3A substrate. EFX or OFX (10 mg/kg) was administered once a day for 3 days. QN (2 mg/kg) was intravenously injected at 2 h after the final dose of FQs administration. The same dose of QN was intravenously injected 3 weeks before the start of FQs administration for control. Neither EFX nor OFX changed the pharmacokinetic parameters of QN. These in vitro and in vivo consisted results suggest that these FQs lack the inhibitory effects on CYP3A activities in dogs. Hence, given these results, the risk of drug-drug interaction is unlikely to occur between FQs and CYP3A substrates in clinical situation in dogs.