Inhibition of translation by cytotrienin A--a member of the ansamycin family

Abstract

The ansamycins are a diverse and often physiologically active group of compounds that include geldanamycin and rifamycin, inhibitors of heat shock protein 90 and prokaryotic DNA-dependent RNA synthesis, respectively. Cytotrienin A is an ansamycin-type small molecule with potent antiproliferative and proapoptotic properties. Here, we report that this compound inhibits eukaryotic protein synthesis by targeting translation elongation and interfering with eukaryotic elongation factor 1A function. We also find that cytotrienin A prevents HUVEC tube formation and diminishes microvessel formation in the chorioallantoic membrane assay. These results provide a molecular understanding into cytotrienin A's previously reported properties as an anticancer apoptosis-inducing drug.

FIGURE 1.

Cyt A inhibits eukaryotic translation. (A) Chemical structure of Cyt A. (B) Cyt A inhibits both cap-dependent and HCV IRES-driven translation in Krebs-2 extracts. In vitro translations were performed in the presence of [³⁵S]methionine and programmed with FF/HCV/Ren. DMSO, anisomycin (aniso), or Cyt A (lanes 3–9) were added to Krebs-2 extracts at the indicated concentrations. Proteins were separated by SDS-PAGE and visualized by autoradiography. The arrow and arrowhead denote Firefly and Renilla luciferase, respectively. (C) Luciferase activity from translations performed in Krebs-2 extracts programmed with FF/HCV/Ren shown in B. Light units were set relative to the values obtained in the presence of vehicle (DMSO). The average of three measurements is shown with the SEM represented by error bars.

FIGURE 2.

Cyt A inhibits translation elongation. (A) Cyt A does not inhibit translation initiation. Ribosome bindings were performed in RRL using ³²P-labeled CAT RNA. Reactions were separated by centrifugation on 10%–30% glycerol gradients and fractions quantitated by scintillation counting. (Left) Ribosome bindings were performed in the presence of 50 μM Cyt A alone, preincubated with 50 μM hippuristanol (Hipp), followed by addition of 50 μM Cyt A, or preincubated with 50 μM Cyt A, followed by the addition of 50 μM Hipp. (Right) Ribosome bindings were performed in the presence of 0.6 mM CHX or 50 μM Hipp. Both panels are part of the same experiment, but were separated for clarity. (B) Cyt A inhibits translation elongation. In vitro translations in RRL supplemented with [³H]phenylalanine and programmed with poly(U) RNA. Polypeptides were TCA precipitated and quantitated by scintillation counting. Counts were set relative to DMSO levels. The average of four measurements is shown with the SEM. (C) Cyt A does not permit ribosome run-off. In vitro translation reactions of Krebs-2 extracts were allowed to proceed in the absence of compound for 5 min, after which time DMSO, HHT (200 μM), CHX (50 μM), or Cyt A (20 μM) were added. Aliquots were taken at the indicated times, TCA precipitated, and quantitated by scintillation counting. The average of three measurements is shown with the SEM. The downward arrow indicates the point of addition of compound or vehicle.

FIGURE 3.

The effect of Cyt A on the steps of translation elongation. (A) Cyt A does not inhibit peptidyl transferase activity. [³⁵S]Methionine-puromycin formation was monitored in the presence of purified 40S and 60S ribosomes using [³⁵S]Met-tRNA_i and ribosomal high-salt wash from RRL. (Left) Aliquots of samples were taken at the indicated time points and separated by TLC. The position of migration of [³⁵S]Met-puro, [³⁵S]Met, and [³⁵S]Met-tRNA_i is indicated to the left. The addition of 50 μM Cyt A, 40 μM HHT, or the absence of puromycin (−Puro) or ribosomes (−Rib) is indicated at top. (Right) Quantitation of [³⁵S]Met-puro production. The average of four experiments relative to the DMSO control at 30 min is shown. Note that values obtained from the reaction in the absence of puromycin were subtracted as background. The SEM is represented using error bars. (B) Cyt A does not inhibit eEF1A-independent [¹⁴C]Phe-tRNA^Phe binding to 80S ribosomes. Filter binding of [¹⁴C]Phe-tRNA^Phe was performed with purified 80S ribosomes, 0.4 mg/mL poly(U) RNA, and either DMSO, 50 μM Cyt A, or 50 μM CHX. The average of four experiments is shown with the SEM indicated by error bars. (C) Cyt A modulates eEF1A-dependent [¹⁴C]Phe-tRNA^Phe binding to 80S ribosomes. Filter binding of [¹⁴C]Phe-tRNA^Phe with purified 80S ribosomes, and 0.4 μg/mL poly(U) RNA in the presence of either DMSO, 50 μM Cyt A, 50 μM DidB, or 50 μM CHX. The presence of eEF1A and nucleotide is indicated. The average of three to six measurements is shown with SEM indicated by error bars. (D) eEF2-dependent translocation of [¹⁴C]Phe-tRNA^Phe is inhibited by Cyt A only when aminoacyl-tRNA is loaded in an eEF1A-dependent manner. Following nonenzymatic or eEF1A-dependent tRNA binding (as described in B and C with GTP, respectively), eEF2 was added to the reaction with puromycin. The amount of puromycin-active [¹⁴C]Phe-tRNA^Phe was extracted with ethyl aceteate and quantitated by scintillation counting. tRNA already bound to the P-site was subtracted from these values (see Materials and Methods) and set relative to the DMSO control. The average of two to four experiments is shown with the SD.

FIGURE 4.

Cyt A does not affect ternary formation. (A) Cyt A does not inhibit GTP binding to eEF1A. Purified eEF1A (1 μg) was UV cross-linked to [α³²P]GTP in the presence (lanes 4–7) or absence (lanes 1–3) of Phe-tRNA^Phe and 50 μM Cyt A or 1 mM unlabeled GTP. Reactions were treated with RNase A, separated by SDS-PAGE, and visualized by autoradiography. (B) Cyt A does not affect [¹⁴C]Phe-tRNA^Phe binding to eEF1A. Increasing amounts of eEF1A were incubated with [¹⁴C]Phe-tRNA^Phe in the presence of DMSO, 50 μM Cyt A, or unlabeled Phe-tRNA^Phe competitor. EMSAs were performed on 6% polyacrylamide gels and visualized by autoradiography. The position of migration of free [¹⁴C]Phe-tRNA^Phe and complexes are indicated to the left. (C) Cyt A does not affect the GTPase activity of eEF1A. eEF1A and [γ-³²P]GTP were incubated with 40S and 60S ribosomes and Phe-tRNA^Phe in the presence of 50 μM Cyt A or DMSO. GTPase activity was also measured in the absence of eEF1A or without ribosomes or Phe-tRNA^Phe. The average of three to four measurements is shown with the SEM represented as error bars.

FIGURE 5.

Cyt A reversibly inhibits translation in cell culture. (A) Consequences of Cyt A exposure on DNA, RNA, and protein synthesis in HeLa cells. Cyt A was added to cell medium for 1 h and [6-³H]thymidine, [5-³H]uridine, or [³⁵S]methionine/cysteine was present during the last 20 min of incubation. Counts from TCA-precipitated material were normalized to total protein content and set relative to the DMSO control. The average of four data points is shown with the SEM indicated by error bars. (B) Inhibition of translation by Cyt A is reversible. HeLa cells were incubated in 2 μM Cyt A for 1 h, after which fresh medium lacking Cyt A was added. Twenty minutes before lysis, [³⁵S]methionine/cysteine was added. Normalization was performed to total protein concentration and set relative to the DMSO control. The average of four measurements is shown with the SEM represented by error bars. (C) Cyt A does not allow ribosome run-off in cell culture. Polysome formation in HeLa cells exposed to 2 μM Cyt A for 1 h and/or 5 μM Hipp for 30 min or 0.5 μM HHT for 1 h. Panels are from the same experiment and were separated for clarity.

FIGURE 6.

Cyt A inhibits angiogenesis. (A) Photomicrographs of HUVEC tube formation at different concentrations of Cyt A or silvestrol (Silv). Scale bar, 0.1 mm. (B) Quantitation of tube formation in HUVECs. Each well was photographed in seven fields, and the average number of tubes formed was counted. The average of four experiments is shown. Error bars represent the SEM. (C) Cyt A inhibits protein synthesis without inducing apoptosis in HUVECs. Following a 24-h exposure to Cyt A or DMSO, HUVECs were labeled for 20 min with [³⁵S]methionine/cysteine or monitored for apoptosis. For the translation assays, TCA-precipitable material was normalized to total protein content and set relative to the DMSO control. The average of four measurements is shown with the SEM represented by error bars. Cell viability was judged by the relative percent of Annexin-FITC or propidium iodide staining compared with DMSO controls. The average of five data points is shown with the SEM represented by error bars. (D) Cyt A inhibits angiogenesis in the CAM assay. Values presented represent the average number of vessels per cm² area for three samples with the SEM; **P < 0.01 (vs. vehicle); ***P < 0.001 (vs. vehicle).