Correction to: Nature Immunology https://doi.org/10.1038/ni907, published 3 March 2003.

Figure 1a is intended to show by northern blotting that only IL-4 among a set of six cytokines tested induces expression of Xbp1 mRNA. A PubPeer posting alleges that the bands representing Xbp1 mRNA from splenic B cells treated with IL-2 and IL-6 may be identical. It further alleges that the bands representing the Actg mRNA loading controls for IL-2 and IL-5 treatment may be identical. Because we cannot retrieve the original source data for this figure, we cannot definitively exclude the possibility of image duplication. However, Fig. 1b and Fig. 1c confirm that IL-4 induces XBP-1 expression in splenic B cells, which is the main message of Fig. 1a. The paper offers no independent demonstration of the specificity of IL-4’s effect among this set of cytokines, which is the secondary message of Fig. 1a. Therefore, although the Fig. 1a portrayal of IL-4’s specificity must be interpreted with caution, its effect on XBP-1 induction in splenic B cells is confirmed in other panels of Fig. 1, while the overall scientific conclusions of the paper would not be altered by the omission of Fig. 1a.

Figure 1f is intended to show by northern blotting that IL-4 fails to induce the expression of two XBP-1 target genes, Grp78 and Chop, in splenic B cells from STAT6–/– mice, supporting the idea that IL-4 induces active XBP-1 expression via the IL-4 receptor’s known signal transduction pathways. A PubPeer posting alleges that the bands representing the Actg loading controls for Grp78 expression in untreated wild-type and STAT6–/– cells are duplicates of the bands for these loading controls from IL-4-treated cells. Because we cannot retrieve the original source data for this figure, we cannot definitively exclude the possibility of image duplication. However, there is no dispute about the other component of this figure, which demonstrates STAT6-dependent induction of Chop expression in response to IL-4. Thus, the figure still makes the point that an XBP-1 target gene requires STAT6 for its induction in response to IL-4. Consequently, while the Grp78 component of Fig. 1f must be interpreted with caution, the overall scientific conclusions of the paper would not be altered by its omission.