We show that in the absence of 4E-BP1 overexpression or eIF4E downregulation, the efficient pharmacological inhibition of mTOR pathway does not change, in short term, the efficiency of the translational apparatus. These data indicate that part of the regulation of translation rate, at least in REN cells, may occur transcriptionally through the abundance of eIF4E. In addition, rapamycin does not depress translation, in spite of reducing cap complex formation. It follows that, in vivo, at steady-state recycling of eIF4E from a processed mRNA to a new preinitiation complex is either kinetically or sterically favored respect to binding of eIF4E to 4E-BP.
Translational control is becoming an attractive target in cancer cells 
. However, we need to further understand the complex ways by which the process of transformation can perturb the translational machinery. REN mesothelioma cells, here characterized, are able to survive and proliferate under growth factor and matrix deprivation, giving rise to metastatic cancer in experimental models 
. We characterized translational control in REN mesothelioma cells in the absence of exogenous stimulation, in order to establish the relevance of endogenous pathways converging on translation. Malignant mesothelioma is a slow growing tumor resistant to chemotherapy 
. Here, we report that the metabolic rate of REN cells is not only independent from exogenous growth factor activation, but also largely insensitive to the inhibition of both mTOR and MAPK pathways, suggesting that it is controlled by transcriptional activation of the ribosomal machinery. We observed similar results in NCI-H28, but not in mesMM98 mesothelioma cells.
The transcriptional regulation of initiation factors is an emerging concept, that is becoming relevant in cancer research 
. Myc is known to regulate the translational machinery 
and, specifically, it regulates eIF4E and eIF4G levels 
. Myc also stimulates the output of Pol I 
and Pol III 
, resulting in the increase of rRNA and tRNA expression. It remains to be established if also 4E-BP1 is under the control of Myc oncogene. REN cells express more c-myc than mesMM98 in line with their rapamycin insensitivity. It has to be established with future work if a causative relation between c-myc expression and eIF4E exists in mesothelioma cells. P53 has been reported to regulate eIF4G and 4E-BP1 
. We could not establish a relationship between p53 and rapamycin sensitivity. A more intriguing possibility exists: it was recently shown that mTOR inactivation in the heart led to 4E-BP upregulation 
. If this is a general case, one contribution of the constitutively active PI3K-mTOR pathway on the translational control, is the stimulation of eIF4F formation by the transcriptional repression of one key player, 4E-BPs, through yet unknown mechanisms. Thus, in cancer cells, cap complex activation may result from three separate routes: inactivation of 4E-BPs by phosphorylation, upregulation of eIF4E and transcriptional downregulation of 4E-BP1. These three elements may not be exclusive.
The marginal effect of the inhibition of signaling pathways on the translational machinery can reflect the fact that we employed short-term treatments. This led to an interesting observation: the temporal disjunction between the efficient inhibition of the pathway and the effect on translational machinery. First, we found that in REN cells rapamycin treatment blocked mTOR dependent assembly of cap complex, without a significative reduction of metabolic rate. Moreover, when we blocked TOR kinase activity with PP242, leading to a complete inhibition of both mTORc1 and mTORc2, even at 100 µM, where PP242 has been reported to aspecifically block PI3K/Akt pathway 
, the global rate of translation was not affected, after short term treatment. We previously found that the sensitivity of the translational apparatus to pharmacological inhibition with Everolimus, a rapamycin analogue, was linked to mutations in the Ras pathway 
. The MAPK-ERK signaling pathway, downstream of Ras, acts on the initiation of translation, at least at three levels: Mnk1/2 phosphorylate eIF4E at Ser 209, and this event contributes to malignancy 
; rpS6 is phosphorylated by p90 Ribosomal S6 Kinases 
; 4E-BP1 has rapamycin insensitive, serum dependent phosphorylation sites 
. Even if ERK1/2 was constitutively activated in REN cells, the inhibition of either this pathway trough U0126 drug or Mnk1 inhibition, had a modest role on the general level of translation. In addition, the combination of multiple inhibiting drugs was not effective. Finally, when we performed long-term treatment with rapamycin, we were able to reduce cap dependent translation (Grosso, unpublished data). Taken together, these observations are consistent with two interpretations, which are not mutually exclusive: 1) the kinetics of inhibition of eIF4E by 4E-BP1 in vivo
, is delayed, indicating that eIF4E actively engaged in eIF4F, recycles to the mRNA, rather than binding to free 4E-BPs. This interpretation is backed up by the observation that increasing 4E-BP1 or lowering eIF4E results in a more rapid inhibition of translation by rapamycin; 2) in principle, oncogenic activation by Ras, PI3K or Myc may alter by a transcriptional mechanism the sensitivity of the translational machinery to pharmacological inhibition.
An early event observed after rapamycin treatment is the decrease of translation of TOP mRNAs. This indicates that TOP mRNA translation is highly sensitive to cap complex inhibition as shown in several laboratories (reviewed in 
). The fact that at the same time the metabolic rate is not yet altered can be due to two reasons: 1) the reduction in TOP mRNA translation is marginal with respect to global protein synthesis and below the sensistivity of our assay, 2) in early phases of inhibition, TOP mRNAs are competitively replaced by other classes of mRNAs that have a reduced requirement for the cap complex. It remains to be established which are the mRNAs that replace TOP on the translating ribosomes. Attractive candidates are IRES containing mRNA that can be translated independently from eIF4F complex 
. Whatever the mechanism at play, these data strengthen the importance of the regulation of the translational machinery in cell cycle progression, since a continuous feed from the growth factor signaling pathway is required for the synthesis of new ribosomes.
In consideration of the data that we presented, showing that inhibition of signaling pathways may not always affect translation of cancer cells, it may be more fruitful to directly inhibit the association of initiation factors. Recently, it was reported that 4EGI-1, a compound that disrupts eIF4E/eIF4G association, inhibits cap-dependent translation 
. In this context, also hippuristanol has been reported to block eIF4A helicase activity reducing the translation of structured mRNAs 
, and ribavirin affects the eIF4E localization in the cytoplasm 
, without acting as mimic of mRNA cap structure 
. It will be interesting to investigate whether REN cells are also insensitive to these compounds.