Hypoxia inhibits protein synthesis through a 4E-BP1 and elongation factor 2 kinase pathway controlled by mTOR and uncoupled in breast cancer cells

Abstract

Hypoxia is a state of low oxygen availability that limits tumor growth. The mechanism of protein synthesis inhibition by hypoxia and its circumvention by transformation are not well understood. Hypoxic breast epithelial cells are shown to downregulate protein synthesis by inhibition of the kinase mTOR, which suppresses mRNA translation through a novel mechanism mitigated in transformed cells: disruption of proteasome-targeted degradation of eukaryotic elongation factor 2 (eEF2) kinase and activation of the regulatory protein 4E-BP1. In transformed breast epithelial cells under hypoxia, the mTOR and S6 kinases are constitutively activated and the mTOR negative regulator tuberous sclerosis complex 2 (TSC2) protein fails to function. Gene silencing of 4E-BP1 and eEF2 kinase or TSC2 confers resistance to hypoxia inhibition of protein synthesis in immortalized breast epithelial cells. Breast cancer cells therefore acquire resistance to hypoxia by uncoupling oxygen-responsive signaling pathways from mTOR function, eliminating inhibition of protein synthesis mediated by 4E-BP1 and eEF2.

FIG. 1.

(A) Diagrammatic representation of the mTOR signaling pathway involved in translational control. Arrows indicate activation, bars indicate inhibition, and dotted lines indicate uncertainty as a primary or established mechanism. Figure based on references and . (B) Protein synthetic rates following 24 h of hypoxia. Cells were grown for 24 h under atmospheric oxygen (normoxic) or hypoxic (0.5% O₂) conditions. Cultures were labeled for 1 h with [³⁵S]methionine, lysates were prepared, and rates of protein synthesis were determined by protein-specific activity derived by trichloroacetic acid precipitation and scintillation counting of samples containing equal amounts of protein. The results are the means with standard deviations derived from at least three independent experiments performed in duplicate. Data were normalized to the mean value of normoxic HTB20 cells.

FIG. 2.

Effect of hypoxia on abundance and phosphorylation of key translation and regulatory factors. MCF10A, CRL2324, and HTB20 cells were subjected to hypoxia (0.5% O₂) for 24 h (H) or in parallel were grown under normoxia (N) as a control. Cells were collected and lysed in 0.5% NP-40 buffer, normalized for soluble protein content, and resolved by SDS-PAGE, and proteins were identified by immunoblot analysis with specific antisera as shown. Representative immunoblots are shown. (A) Immunoblot analysis of equal amounts of total protein (50 μg) of eIF4A, eIF4E, and high-resolution separation of hyperphosphorylated (hyper-P) and hypophosphorylated (hypo-P) forms of 4E-BP1, resolved by SDS-15% PAGE. (B) Low-resolution SDS-8% PAGE analysis of 30 μg of cell lysates showing total eIF4G and 4E-BP1 levels, using extracts prepared as described above. (C) Immunoblot analysis of total eEF2K, Thr56 phosphorylated eEF2 (eIF2-P), and total eEF2 protein levels in 100 μg of protein lysates, prepared as described above. (D) Immunoblot analysis of total and phosphorylated forms of eEF2, total EF2K, eIF4A, and p27 cell cycle regulator as a control from 30 μg of lysate, prepared as described above, under hypoxic and normoxic conditions, with and without prior treatment of cells with proteasome inhibitor MG132 (MG) or control (C) vehicle. (E) Immunoblot analysis of serine 51 phosphorylated eIF2α and total eIF2α protein levels, prepared as described above. (F) Immunoblot analysis of total eIF2B protein levels in 30 μg lysate: only samples from normoxia are shown. Data were quantified by densitometry of autoradiograms from at least three independent experiments; representative results are shown.

FIG. 3.

Analysis of eIF4E interaction with 4E-BP1 and eIF4G during hypoxia. (A) Equal amounts of NP-40 lysates (300 μg) from MCF10A, CRL2324, and HTB20 cells under normoxic (N) or hypoxic (H) conditions were subjected to m⁷GTP-Sepharose cap-chromatography, recovered by elution with m⁷GTP, and resolved by SDS-15% PAGE, followed by immunoblot analysis with antisera as indicated. (B) Equal amounts of NP-40 lysates (300 μg) from MCF10A, CRL2324, and HTB20 cells cultured under normoxia (N) or hypoxia (H) were subjected to immunoprecipitation with anti-human eIF4GI antibodies. Immunoprecipitates were resolved by SDS-15% PAGE, and proteins were detected by immunoblot analysis as indicated. Data are representative of three independent experiments, which were quantified by densitometry of autoradiograms.

FIG. 4.

Inhibition of eIF2α Ser51 phosphorylation in MCF10A cells has no effect on translation inhibition during hypoxia. (A) Immunoblot analysis of protein extracts from MCF10A cells expressing the FLAG-GADD34 C-terminal fragment, which constitutively dephosphorylates eIF2α, pBABEpuro vector control, or parental MCF10A cells. Antibodies specific for total and Ser51 phosphorylated eIF2α were used. A cross-reactive nonspecific protein is identified. Non-trans., nontransfected.(B) Relative protein synthesis activity during hypoxia was determined by [³⁵S]methionine labeling cells for 1 h, followed by trichloroacetic acid precipitation, determination of specific activity per mg of protein, and scintillation counting of samples. Samples were normalized to normoxic vector control set at 100%.

FIG. 5.

Selective silencing of 4E-BP1 by siRNA in immortalized MCF10A cells partially prevents protein synthesis inhibition during hypoxia. MCF10A cells were transfected four times with either 4E-BP1 siRNA (4E-BP1i) or a nonsilencing (NS) control siRNA. Forty-eight hours following the last transfection, cells were cultured for 24 h under either normoxic (N) or hypoxic (H) (0.5% O₂) conditions. (A) Equal amounts of protein lysates from cells were resolved by SDS-PAGE, and immunoblot analysis was carried out with antibodies specific for 4E-BP1 or eIF4A. eIF4A was used as a loading control. (B) Total protein synthesis was determined by [³⁵S]methionine labeling cells for 1 h, followed by trichloroacetic acid precipitation of equal amounts of protein and scintillation counting of samples to determine protein specific activities. Results represent an average of three independent experiments, normalized to the normoxic control. (C) m⁷GTP (cap) chromatography was carried out using equal amounts (300 μg) of protein extracts from cells transfected with control nonsilencing or specific siRNA for eEF2K or 4E-BP1; bound proteins were eluted and compared by immunoblot analysis to unbound 4E-BP1. The 4E-BP1 blots were overexposed to visualize the low levels of protein remaining following knockdown.

FIG. 6.

RNAi-mediated knockdown of eEF2K and 4E-BP1, or TSC2, in immortalized MCF10A cells confers resistance to hypoxia-mediated protein synthesis inhibition. (A) MCF10A cells were transfected four times with siRNAs as indicated or a nonsilencing (NS) control siRNA, and protein levels were determined by immunoblot analysis for eEF2K, eEF2, eIF4A, and Thr56 phospho-eEF2. (B) Total protein synthesis activity during hypoxia and normoxia was determined by [³⁵S]methionine incorporation in vivo for 1 h, followed by trichloroacetic acid precipitation, scintillation counting of samples, and determination of specific activity of incorporation into protein. Data were derived from triplicate studies. Values were normalized to the normoxic control at 100%. (C) shRNA lentivirus vectors were developed to specifically knock down expression of TSC2, eEF2K, mTOR, and 4E-BP1 proteins or to express a nonspecific control siRNA. Cells were stably transformed with vectors and subjected to normoxia or hypoxia, and equal amounts of protein lysates were analyzed by immunoblot analysis as shown. (D) Total protein synthesis activity during hypoxia and normoxia in shRNA lentivirus-transformed MCF10A cells was determined by [³⁵S]methionine incorporation, as described above.

FIG. 7.

Levels and phosphorylation state of oxygen signaling pathway proteins in normoxic and hypoxic cells. MCF10A, CRL2324, and HTB20 cells were cultured for 24 h under either normoxic (N) or hypoxic (H) (0.5% O₂) conditions. (A) Equal amounts of protein lysates from cells were resolved by SDS-PAGE, and immunoblot analysis was carried out with antibodies specific for total proteins and phospho-specific forms as shown. (B) Selective silencing of TSC2 or mTOR was carried out using shRNA lentivirus vectors under normoxic or hypoxic conditions in MCF10A cells. Equal amounts of protein lysates were analyzed as shown by immunoblot analysis using protein-specific and protein phospho-specific antibodies. (C) m⁷GTP-Sepharose cap chromatography was carried out using equal amounts of protein lysates from normoxic (0 h hypoxic) and hypoxic MCF10A cells that have undergone lentivirus-mediated gene silencing for TSC2, mTOR, or control nonsilencing shRNA. Proteins were eluted and identified by immunoblot analysis.