Biochemistry. In the article “KSR stimulates Raf-1 activity in a kinase-independent manner” by Neil R. Michaud, Marc Therrien, Angela Cacace, Lisa C. Edsall, Sarah Spiegel, Gerald M. Rubin, and Deborah K. Morrison, which appeared in number 24, November 25, 1997, of Proc. Natl. Acad. Sci. USA (94, 12792–12796), the following correction should be noted. Due to a printer’s error, background was incorrectly added to Fig. 2 A and B on page 12793, Fig. 3B on page 12794, and Fig. 4 A–C on page 12795 so that multiple panels from different gels or Western blots appear as one continuous panel. Correct versions of Figs. 2, 3, and 4 and each figure legend are reproduced here and on the opposite page.

Figure 2.

The mKSR1 CA3 domain is required for the Rasdependent membrane localization of mKSR1. (A) 293 cells transiently expressing WT and Myr mKSR1 or coexpressing WT and Ras^V12 were fractionated into membrane and cytosolic fractions. The mKSR1 proteins were immunoprecipitated and examined by immunoblot analysis using αPyo antibody. (B) 293 cells coexpressing Ras^V12 and either WT or CRM mKSR1 proteins were analyzed as in A.

Figure 3.

The mKSR1 CA3 domain augments Raf-1 activity in a detergent-sensitive manner. Xenopus oocytes expressing Raf-1 alone (−) or coexpressing Raf-1 and the mKSR1 CA3 domain (+) were injected with Ras^V12 RNA. Immediately after (0 min) or 150 min after Ras^V12 injection, oocytes were lysed in hypotonic buffer and membranes were isolated. (A) Raf-1 proteins were immunoprecipitated from membrane fractions resuspended in RIPA buffer (+ detergent) or phosphate-buffered saline (− detergent), and in vitro kinase assays were performed using kinase-inactive MEK as a substrate (1). Phosphorylation of MEK1 on Ser-218 and Ser-222 was determined by tryptic peptide mapping analysis. (B) Ras^V12 and CA3 proteins were immunoprecipitated from membrane (P100) and cytosolic fractions (S100) isolated at 150 min after injection with Ras^V12 RNA and were examined by immunoblot analysis using Ras and Pyo antibody, respectively. The migration of processed (Pro) and unprocessed (UnPro) Ras proteins is indicated.

Figure 4.

Augmentation of Ras signaling by mKSR1 does not involve ceramide. (A) Cos cells were transiently transfected with constructs encoding wild-type (WT) or kinase-inactive (RM) mKSR1. At 60 hr posttransfection, serum starved cells were left untreated or were stimulated with 20 μM C2 ceramide for 5 or 10 min, 100 milliunits/ml sphingomyelinase (SMase) for 20 min or 10 nM tumor necrosis factor α (TNFα) for 20 min. KSR proteins were immunoprecipitated using αPyo antibody, and mKSR1 immune complex kinase assays were performed in vitro as described by Zhang et al. (ref. 19; Top). Immunoprecipitated mKSR1 was detected by immunoblot analysis (Middle). To observe phosphorylation of Raf-1 or modulation of Raf-1 activity, purified activated Raf-1, coexpressed in Sf9 cells in the presence of RasV12 and v-src, and kinase-inactive MEK1 were added to the mKSR1 immune complex kinase assays previously described (ref. 19; Bottom). (B) Cos cells were treated as in A and endogenous ceramide levels, JNK activity, and MAPK activity were determined. Ceramide levels were normalized to the untreated control. C2-ceramide levels were elevated 2.9- and 3.7-fold at 5 and 10 min, respectively, and long-chain ceramide levels were elevated 12-fold by SMase and 1.9-fold by TNFα. (C) Purified brain ceramide (100 nM) (+) or diluent (−) was added in vitro to mKSR proteins immunoprecipitated from transfected Cos cells and immune complex kinase assays performed in the presence of activated Raf-1 and kinase-inactive MEK1 as previously described (ref. 19; Top). Immunoprecipitated mKSR1 was detected by immunoblot analysis (Bottom). (D) Oocytes preinjected with buffer or RNA encoding WT and CRM mKSR1 constructs were treated with 250 milliunits sphingomyelinase. GVBD was then scored 6 and 10.5 hr after treatment.