Correction for Tee et al., Tuberous sclerosis complex-1 and -2 gene products function together to inhibit mammalian target of rapamycin (mTOR)-mediated downstream signaling

CELL BIOLOGY Correction for “Tuberous sclerosis complex-1 and -2 gene products function together to inhibit mammalian target of rapamycin (mTOR)-mediated downstream signaling,” by Andrew R. Tee, Diane C. Fingar, Brendan D. Manning, David J. Kwiatkowski, Lewis C. Cantley, and John Blenis, which was first published October 15, 2002; 10.1073/pnas.202476899 (Proc. Natl. Acad. Sci. U.S.A. 99, 13571–13576).

The authors wish to note the following:

“We recently learned of concerns related to Figs. 1 and 2C in our article.

Fig. 1.

Hamartin and tuberin inhibits PI3K-dependent 4E-BP1 phosphorylation. HEK293E cells coexpressing Flag-tagged hamartin (Ham) and tuberin (Tub), where indicated, with HA-tagged 4E-BP1 were serum-starved and pretreated with 20 nM rapamycin (rap) for 30 min before being stimulated with insulin (100 nM) or PMA (100 ng/mL) for 30 min, where indicated. (A) The levels and Thr-308 phosphorylation of Akt and the levels and phosphorylation of the MAPK isoforms (p44 and p42) were determined. (B) Hamartin and tuberin protein levels were accessed from the cell lysates (Sol) and the insoluble fraction (Non-Sol) as described in Materials and Methods by using the anti-Flag antibody. Thr-1462 tuberin phosphorylation was analyzed by using a tuberin Thr-1462 phospho-specific antibody. (C) The phosphorylation of exogenous 4E-BP1 was determined with an anti-HA antibody and phospho-specific antibodies for 4E-BP1 at Thr-37 and/or 46, Ser-65, and Thr-70, as indicated. The α-, β-, and γ-species of 4E-BP1 are labeled accordingly. (D) Cell extracts were subjected to affinity chromatography on m⁷GTP-Sepharose, as described in Materials and Methods. The levels of eIF4E and exogenous 4E-BP1 that was copurified were determined.

Fig. 2.

Hamartin and tuberin inhibit 4E-BP1 phosphorylation and S6K1 activity within proliferating cells. U20S cells overexpressing 4E-BP1 (A) or S6K1 (B) with or without hamartin (Ham) and tuberin (Tub), where indicated, were grown in serum and treated with 20 nM of rapamycin (Rap) for 30 min, as indicated. Hamartin, tuberin, and MAPK isoform (as a loading control) protein levels are shown. The α-, β-, and γ-species of 4E-BP1 are labeled accordingly. S6K1 kinase assays were carried out as described in Materials and Methods. The total levels of S6K1 are shown. Incorporation of ³²P label into GST-S6 was assessed, and an autoradiograph of the gel is presented. The ratios of ³²P label incorporated into GST-S6 were normalized against the empty vector (pRK7). The data presented are representative of at least three experiments. (C) HEK293E cells overexpressing 4E-BP1 with or without hamartin (Ham), tuberin (Tub), and the tuberin K599M mutant [Tub(K599M)], where indicated, were serum-starved and then stimulated with 100 nM insulin for 30 min, where indicated. Hamartin and tuberin expression and the extent of phosphorylation of 4E-BP1 was analyzed as for Fig. 1C.

“The goals of the experiments shown in Figs. 1 and 2 were to provide evidence that hamartin (Tsc1) and tuberin (Tsc2) function together to inhibit PI3K-dependent 4E-BP1 phosphorylation as well as 4E-BP1 phosphorylation in serum-starved cells that were then stimulated with insulin. These experiments were representative of several reproduced, large experiments that originally involved using both insulin or PMA agonists. We observed that insulin was utilizing Akt upstream of mTORC1 signaling whereas PMA was not using Akt, but ERK-mediated signaling (see Fig. 1, lane 9). As it was unknown at the time how PMA activated mTORC1 signaling, we made the decision to focus on how Tsc1 and Tsc2 were suppressing insulin and Akt-dependent regulation of 4E-BP1 phosphorylation. Thus, to clarify the presentation, we removed the PMA lanes from the original autoradiogram in preparing the final figures (Figs. 1 and 2C).

“Our intention was to keep the PNAS paper specifically focused on insulin and Akt signaling. In assembling these figures there was no intention to deceive the PNAS readership, and the removal of the PMA data in no way compromises the conclusions of the paper.

“As these experiments were completed more than 18 y ago, the original autoradiographs have not been located. Thus, to support the original data in Figs. 1 and 2C, we have carried out two new replicates for each figure using the same HEK293 cell lines, DNA vectors, and methodologies, as previously used. Representative figures of the new data presented below show that the original data are reproducible and support the original observations in full. The new figures follow current best-practice procedures for figure assembly. As the experiments were carried out exactly the same as the original, the figure legends for Figs. 1 and 2 are unchanged.”

The corrected Figs. 1 and 2 appear below with their legends.