Dual IRE1 RNase functions dictate glioblastoma development

Abstract

Correction to: EMBO Mol Med (2018) 10: e7929. DOI: 10.15252/emmm.201707929 ¦ Published online 8 January 2018

This corrigendum replaces the editorial note from January 2022. The journal contacted the authors after becoming aware of potential image aberrations in their 2018 paper. Based on the exchange with the authors and the institute, the journal has agreed to correct the following figures and publish the related source data with this notice.

Figure 2A

There had been a previously undeclared splice in the control panel. The figure is herewith modified.

Figure 2A. Original.

Figure 2A. Corrected.

Figure 6D–E and Figure S7E

The authors state that they had analyzed the same dataset in both figures. The authors have provided a reference to the dataset with the accession information and modified the figure legends to reflect (text in bold) that the curves were obtained from the same dataset (TCGA—The Cancer Genome Atlas, dbgap, V1. Phs000178.v11.p8).

The legend to Fig. 6D‐E has been updated from

Kaplan–Meier survival curves of XBP1+/RIDD− (red) and XBP1−/RIDD+ (blue) GBM patients of the TCGA microarray cohort (D) and TCGA RNAseq cohort (E). The Student's t‐test was used.

To (new text in bold).

Kaplan–Meier survival curves for the XBP1+/RIDD+ and XBP1−/RIDD− can be seen in Fig. S7E for the same cohorts. Data were extracted from the TCGA‐GBM datasets obtained either using microarray (D) or RNAseq (E).

The legend to Appendix Fig S7E has been updated from

Kaplan–Meier survival curves of XPB1s^high/RIDD^low (red), XPB1s^high/RIDD^high (gray), XPB1s^low/RIDD^low (green), and XPB1s^low/RIDD^high (blue) GBM tumor patients of both TCGA cohorts microarrays (left) and RNAseq (right).

To (new text in bold).

Kaplan–Meier curves obtained for XPB1s ^high /RIDD ^low (red) and XPB1s ^low /RIDD ^high (blue) groups were presented in Fig 6D and E and serve as references for XPB1s ^high /RIDD ^high (gray) and XPB1s ^low /RIDD ^low (green) groups. Data were extracted from the TCGA‐GBM datasets obtained either using microarray or RNAseq.

Figure 8A

Figure 8A. Original.

In the original paper, the figure contained undeclared splice sites and duplicated actin controls. The journal could not resolve the aberrations in the actin controls, and in the revised figure, the authors display nonspecific bands as a loading control in place of actin. In the original figure, the authors had displayed a long exposure for the RNS130 and RNS96 samples to highlight the IRE1‐truncated form produced by the Q780* mutant. The revised figure removes the longer exposure for clarity. The long and short exposure blots are provided as source data.

Figure 8A. Corrected.

The authors had run the same samples for Fig 8A in a similar experiment with the actin blot. This experiment is provided as a new Appendix Fig S8C.

Figure S4A. Original.

Figure S4A. Corrected.

The legend to Fig 8A has been updated from

“Western blot analysis of the expression of WT or Q780* IRE1 in RNS85, RNS87, RNS96, and RNS130 primary GBM lines.”

To (new text in bold).

“Western blot analysis of the expression of WT or Q780* IRE1 in RNS85, RNS96, and RNS130 primary GBM lines. Arrow shows the full‐length IRE, and * indicates the IRE1 truncated form produced by the Q780* mutant. Protein loading was evaluated using a nonspecific band on the blot (NS).”

Legend to the new Appendix Fig S8C (new text in bold).

Western blot analysis of the expression of WT or Q780* IRE1 in RNS85, RNS87, RNS96, and RNS130 primary GBM lines. Fl shows the full‐length IRE1, and Q indicates the IRE1‐truncated form produced by the Q780* mutant. Protein loading was evaluated using actin as a housekeeping protein. Samples on these blots are the same as those ran in Fig 8A, but only 20 μg of proteins was separated by SDS–PAGE prior to immunoblotting.

Added reference to Appendix Fig S8C.

The reference has been added onto the following text.

“We then further tested whether altering IRE1 activity in those cells could impact on their tumoral properties. To this end, we overexpressed IRE1 WT and the Q780* mutant (known to impair XBP1s [Fig 2B]) in four primary lines, namely RNS85, RNS87, RNS96, and RNS130.”

The reference to Appendix Fig S8C has been added to and updated from

“The lines were selected on the basis of IRE1 mRNA expression (Appendix Fig S8A) and then analyzed by Western blot with anti‐IRE1 antibodies (Fig 8A). As expected, IRE1 expression was higher upon overexpression on IRE1 WT and the shorter form of IRE1 observed upon expression of the Q780* variant.”

To (Added and updated text in bold).

Two clones per line were selected on the basis of IRE1 mRNA expression (Appendix Fig S8A and B). Subsequently, Western blot using IRE1 antibodies was performed on the corresponding lysates to look for the presence of (A) an increased intensity band in the case of the WT overexpression and the presence of an additional band in the case of Q780* overexpression. We chose for downstream experimentation one clone per construct and per line, which either displayed the highest band intensity in the case of WT overexpression or the clearest additional band in the case of Q780* mutant overexpression (Fig 8A and Appendix Fig S8C).

Figure S4A

There had been a previously undeclared splice in the control panel. The figure is herewith modified.

Figure S8. Original.

Figure S8C. New panel.

The authors apologize for these errors and any confusion they may have caused; no response could be obtained from JO, FJ, ML, and P‐JLR.

Supporting information

Appendix S1

Source Data for Appendix

Source Data for Figure 2

Source Data for Figure 8

EMBO Mol Med (2023) 15: e16731