Targeting N-myristoyl transferase in cancer using peptidearrays
Emmanuelle Thinon1, David Mann2, Edward W. Tate1
1Department of Chemistry or 2Department of Biochemistry, Imperial College London, Exhibition Road, SW7 2AZ London, UK
E-mail:
Introduction
Protein N-myristoylation is the irreversible attachment of a C14 saturated fatty acid to the N-terminal glycine of a target protein. This modification is catalysed by myristoyl CoA: protein N-myristoyl transferase (NMT) [1] (Fig. 1.).
Fig.1. Myristoylation of a protein substrate
In humans, two distinct NMT enzymes, NMT1 and NMT2, have been identified. It has been suggested that they possess distinct but overlapping substrate specificity and function. It has been reported that NMT is upregulated in several cancers [1].Preliminary experiments with a non selective inhibitor in cancer cells showed that NMT inhibition led to cell growth arrest after 1 day and cell death after 7 days [2]. Previous studies based on RNA interference experiments on a mouse model suggested that isoform-specific inhibitors against NMT1 might be effective anti-cancer agents as a knockdown of NMT1 inhibits the tumour growth, whereas a knockdown of NMT2 has no effect [3]. If residual NMT2 activity can compensate for loss of NMT function in healthy cells, potential toxicity may also be minimised. We herepresent a method to identify peptide substrates of NMT1 and/or NMT2. Peptide libraries can be prepared on a cellulose membrane and screen for activity using a labeling technology. Peptide substrates could then be develop as peptidomimetic inhibitors and used in cancer cell linesto validate NMT1 and/or NMT2 as a new therapeutic target for cancer.
Results and Discussion
Peptides made of the first 15 amino acids at the N-terminus of known or predicted myristoylated proteins were synthesized on an amino-functionalised cellulose membraneusing standard solid phase peptide synthesis.
Fig. 2. Detection of NMT1/2 peptide substrate on a cellulose membrane
Peptides were screened for activity using the labeling technology developed in our group to detect NMT substrates [4]. Peptides were exposed to recombinant NMT1 and/or NMT2 and an alkyne-tagged analogue of myristoyl CoA (Fig. 2.). Only peptide substrates should be modified by the enzyme and bear an alkyne tag. Subsequent azide-alkyne “click” cycloaddition allows visualisation of the myristoylated substrates in fluorescence or chemiluminescence, using a fluorescent or a biotin moiety on the capture reagent (Fig. 3.).
Attempts to visualise the spots in fluorescence were unsuccessful due to high background. Membranes were therefore blocked with BSA, probed with Neutravidin HRP and visualised using a chemiluminescent HRP substrate. Peptides which were reported, predicted or reported not to be myristoylated weresynthesised on a cellulose membrane toFig. 3. Capture reagent [4] validate our method (Fig. 4.). No spot were detected for the peptidescorresponding to proteins that have been shown not to be myristoylated, showing that the method does not give unspecific background.Few spots were detectedfor peptides which werereportedor predictedFig.4. Peptide arrays screened against to be substrates of NMT1/2.
NMT1 or NMT2.Spots correspond to NMTInterestingly a different pattern wassubstrates. obtained for NMT1 and NMT2. Peptides were re-synthesised and their activity tested using a fluorescent assay developed in our group [5]. It allowed us to confirm that the peptides were substrates or not of NMT1/2.
We have shown that it is possible to myristoylate and detect NMT1/NMT2 peptide substrates on a cellulose membrane. However, the use of cellulose membranes presents some limitations. Fluorescence, which is less time consuming than chemiluminescence cannot be used to detect the substrates on a cellulose membrane. Moreover, peptides which were reported to be myristoylated were not all detected using this method, probably due to synthesis issues on the membrane. Future work will focus on the preparation and screening of peptide libraries functionalised with a biotin moiety at the C-terminus and immobilised on an avidin-functionalised glass plate.
Acknowledgments
This work was supported by Cancer Research UK.
References
[1]Wright, M.H.; Heal, W.P.; Mann, D.J.; Tate, E.W. J. Chem. Biol.2010, 3, 19-35.
[2]Manuscript in preparation
[3]Ducker, C.E.;Upson, J.J.;French, K.J.;Smith, C.D. Mol. Cancer Res. 2005, 3, 463-76.
[4]Heal, W.P.; Wright, M.H.; Thinon, E.; Tate, E.W.Nat. Protoc.2011,1,105-17.
[5]Goncalves, V.; Brannigan, J.A.; Thinon, E.; Olaleye, T.O.; Serwa, R.; Lanzarone, S.; Wilkinson, A.J.; Tate, E.W.; Leatherbarrow, R.J.Anal Biochem.2012, 421, 342-4