Adaptive translational reprogramming of metabolism limits the response to targeted therapy in BRAFV600 melanoma

Despite the success of therapies targeting oncogenes in cancer, clinical outcomes are limited by residual disease that ultimately results in relapse. This residual disease is often characterized by non-genetic adaptive resistance, that in melanoma is characterised by altered metabolism. Here, we examine how targeted therapy reprograms metabolism in BRAF-mutant melanoma cells using a genome-wide RNA interference (RNAi) screen and global gene expression profiling. Using this systematic approach we demonstrate post-transcriptional regulation of metabolism following BRAF inhibition, involving selective mRNA transport and translation. As proof of concept we demonstrate the RNA processing kinase U2AF homology motif kinase 1 (UHMK1) associates with mRNAs encoding metabolism proteins and selectively controls their transport and translation during adaptation to BRAF-targeted therapy. UHMK1 inactivation induces cell death by disrupting therapy induced metabolic reprogramming, and importantly, delays resistance to BRAF and MEK combination therapy in multiple in vivo models. We propose selective mRNA processing and translation by UHMK1 constitutes a mechanism of non-genetic resistance to targeted therapy in melanoma by controlling metabolic plasticity induced by therapy.


nature research | reporting summary
April 2020 Field-specific reporting Please select the one below that is the best fit for your research. If you are not sure, read the appropriate sections before making your selection.

Life sciences Behavioural & social sciences Ecological, evolutionary & environmental sciences
For a reference copy of the document with all sections, see nature.com/documents/nr-reporting-summary-flat.pdf

Life sciences study design
All studies must disclose on these points even when the disclosure is negative.  (Figure 2 and S3) and polysome protein precipitation experiments ( Figure 6). Each independent experiment contained appropriate technical replication, as determined by the standard deviation across technical replicates. We included more technical replicates to determine the mean from assays with higher intrinsic variation, for example Seahorse assays which assess metabolism in live cells, in real time, and electron acceptor rescue studies.
The following data exclusions were made: -One out four biological replicates for the polysome profiling analysis was excluded due to high variability between this sample and all other biological replicate samples of the same condition. Inclusion of this data point would have lead to data misinterpretation and was therefore removed from the analysis completely. The data and all analyses presented in the manuscript is from 3 independent biological replicates thus still provides adequate statistical power for analysis and ensures reproducibility of the observations across independent measurements. Exclusion criteria was not predetermined, but assessed from MDS and PCA plots.
-Seahorse assays were performed as 4-5 technical replicates in each independent biological replicate experiment. In some experiments, technical replicates were omitted from the analysis if they were extreme outliers due to the nature of these experiments -dynamic metabolic measurements in live cells, in real time.
We have used biological (not technical) replicates in this study to demonstrate statistical significance (by a defined methodology). We have made efforts to clearly describe in the manuscript text, relevant Methods sections and Figure Legends precisely how we chose biological replicates (i.e. animal numbers or independent repeats of in vitro experiments). For example, for the western blot assays and proliferation assays, we have shown representative data in the main figure but we have performed at least three independent experiments, unless otherwise stated.
For mouse experiments, once tumours reached an average volume of 100mm3, mice were randomized into groups of 9 for therapy studies.
For the genome wide screen, controls were randomized across alternating wells, on 2 different regions of each plate to minimize plate location effects. For tissue culture experiments, treatments were not randomized but performed in predetermined wells/dishes following a treatment layout. For these experiments, randomization was not applicable and predetermined plate layouts were essential to appropriately perform the experiments.
For all tissue culture experiments, investigators were not blinded. Due to complex siRNA and CRISPR genetic modifications, together with drug treatments, a single investigator was responsible for all steps of the experiment and therefore could not be blind to the various conditions in each well. However, for in vivo experiments, technicians responsible for tumour measurements and subsequent statistical analyses were blinded to the nature of the experimental groups. Only the chief investigator, Dr. Lorey Smith, who did not participate in the experiments beyond generation of cell lines, knew the identity of the experimental groups.