ZEB1 loss increases glioma stem cell tumorigenicity and resistance to chemoradiation

OBJECTIVE

Glioblastoma has been known to be resistant to chemotherapy and radiation, whereas the underlying mechanisms of resistance have not been fully elucidated. The authors studied the role of the transcription factor ZEB1 (zinc finger E-box-binding homeobox 1 protein), which is associated with epithelial-mesenchymal transition (EMT) and is central to the stemness of glioblastoma, to determine its role in therapeutic resistance to radiation and chemotherapy. The authors previously demonstrated that ZEB1 is deleted in a majority of glioblastomas.

METHODS

The authors explored resistance to therapy in the context of ZEB1 loss and overexpression in glioma stem cells (GSCs) and in patient data.

RESULTS

Patients with ZEB1 loss had a shorter survival time than patients with wild-type ZEB1 in both the high- and low-MGMT groups. Consistent with the clinical data, mice implanted with ZEB1 knockdown GSCs showed shortened survival compared with mice inoculated with nonsilencing control (NS) short-hairpin RNA (shRNA) GSC glioblastoma. ZEB1-deleted GSCs demonstrated increased tumorigenicity with regard to proliferation and invasion. Importantly, GSCs that lose ZEB1 expression develop enhanced resistance to chemotherapy, radiotherapy, and combined chemoradiation. ZEB1 loss may lead to increased HER3 expression through the HER3/Akt pathway associated with this chemoresistance. Conversely, overexpression of ZEB1 in GSCs that are ZEB1 null leads to increased sensitivity to chemoradiation.

CONCLUSIONS

The study results indicate that ZEB1 loss in cancer stem cells confers resistance to chemoradiation and uncovers a potentially targetable cell surface receptor in these resistant cells.

ZEB1 Is a Transcription Factor That Is Prognostic and Predictive in Diffuse Gliomas

Objective: To address the unmet medical need to better prognosticate patients with diffuse gliomas and to predict responses to chemotherapy regimens.

Methods: ZEB1 alterations were retrospectively identified from a cohort of 1,160 diffuse glioma patients. Epigenome-wide association scans (EWAS) were performed on available data. We determined the utility of ZEB1 as a prognostic indicator of patient survival in diffuse gliomas and assessed the value of ZEB1 to predict the efficacy of treating diffuse glioma patients with procarbazine, CCNU, and vincristine along with radiation at diagnosis. Decision curve analysis (DCA) was used to determine if ZEB1 added benefit to clinical decision-making over and above conventional methods.

Results: Fifteen percent of diffuse glioma patients had a ZEB1 deletion. ZEB1 deletion was associated with poor overall survival (OS) with and without adjustment for age and tumor grade (adjusted HR: 4.25; 95% CI: 2.35 to 7.66; P < 0.001). Decision curve analysis confirmed that ZEB1 status with or without IDH1 was more beneficial to clinical decision making than conventional information such as age and tumor grade. We showed that ZEB1 regulates TERT expression, and patients with ZEB1 deletions likely subsume patients with mutant TERT expression in diffuse gliomas. ZEB1 influenced clinical decision making to initiate procarbazine, CCNU, and vincristine treatment.

Conclusion: We demonstrate the prognostic value of ZEB1 in diffuse glioma patients. We further determine ZEB1 to be a vital and influential molecular marker in clinical decisions that exceed conventional methods regarding whether to treat or not treat patients with diffuse glioma.

The Curious Case of ZEB1

The Zinc Finger E-box binding homeobox (ZEB1/TCF8 or DeltaEF1) is at the forefront of transcription factors involved in controlling epithelial-to-mesenchymal transitions (EMT). Essentially, EMT allows for the reorganization of epithelial cells to become migratory cells with a mesenchymal phenotype.  In addition to ZEB1 being involved in embryonic development, ZEB1 has also been linked to processes involving micro-RNAs, long non-coding RNAs and stem cells. In recent years there has been an accumulation of evidence with regard to ZEB1 in various cancers. Although increased ZEB1 expression has largely been associated with EMT, cancer invasion, and tumorigenicity, there have been some episodic reports that have gone against the traditional reporting of the role of ZEB1. Indicating that the function of ZEB1 and the mechanisms by which ZEB1 facilitates its activities is more complex than was once appreciated. This complexity is further exacerbated by the notion that ZEB1 can act not only as a transcriptional repressor but a transcriptional activator as well. This review seeks to shed light on the complexity of ZEB1 with respect to cancer.

ZEB1 regulates glioma stemness through LIF repression

The identification of a stem cell regulatory gene which is aberrantly expressed in glioma and associated with patient survival would increase the understanding of the role of glioma cancer stem cells (GCSCs) in the virulence of gliomas. Interrogating the genomes of over 4000 brain cancers we identified ZEB1 deletion in ~15% (grade II and III) and 50% of glioblastomas. Meta-analysis of ZEB1 copy number status in 2,988 cases of glioma revealed disruptive ZEB1 deletions associated with decreased survival. We identified ZEB1 binding sites within the LIF (stemness factor) promoter region, and demonstrate LIF repression by ZEB1. ZEB1 knockdown in GCSCs caused LIF induction commensurate with GCSC self-renewal and inhibition of differentiation. IFN-γ treatment to GCSCs induced ZEB1 expression, attenuating LIF activities. These findings implicate ZEB1 as a stem cell regulator in glioma which when deleted leads to increased stemness, tumorigenicity and shortened patient survival.

Comparison of the effects of sevoflurane and propofol on cortical somatosensory evoked potentials

BACKGROUND

Propofol (P) and sevoflurane (S) are potential anaesthetic agents if electrophysiological monitoring is required during spinal surgery. They allow rapid recovery and do not depress cortical somatosensory evoked potentials (SSEP) as much as other agents. The effects of these agents on SSEP have not been compared before.

METHODS

Twenty-four patients were allocated randomly to receive either S (n = 12) or P (n = 12). SSEP evoked by electrical stimulation of the posterior tibial nerve at the ankle were recorded before anaesthesia. The cortical potential P40 was recorded (latency P40 and amplitudes N29P40 and P40N50). The anaesthetic concentration was adjusted gradually to obtain three predetermined ranges of values of bispectral index (BIS): 45-55, 35-45 and 25-35. For each range, a stable state was maintained for 10 min and SSEP were recorded.

RESULTS

For the BIS 45-55 range, compared with preoperative values, P40 latency increased during S [mean change +2 (SD 0.6) ms] but not during P [+0.4 (0.2) ms (P = 0.12)] and both amplitudes (N29P40 and P40N50) decreased with S. Increasing S concentration caused a dose-dependent depression of SSEP. P did not have a statistically significant effect on the recordings and the signals remained stable in each BIS range.

CONCLUSION

Sevoflurane affected SSEP recordings in a dose-dependent fashion. Propofol had a minimal effect on SSEP recordings.

[Amelioration of perioperative somatosensory evoked potentials in the surgery of injuries of the spine]

Somatosensory evoked potentials were performed in 38 patients with traumatic spine pre, per and postoperatively. In 3 cases, the S.E.Ps altered before surgery, were improved after decompression and reduction. This S.E.Ps improvement was associated to clinical improvement. Even though this study was carried out on few cases, it shows prognosis and diagnosis value of intraoperative S.E.Ps in traumatic spine surgery.