Targeted mass spectrometric analysis of N-terminally truncated isoforms generated via alternative translation initiation

FEBS Letters 583 (2009) 2441–2445

journal homepage: www.FEBSLetters.org

Targeted mass spectrometric analysis of N-terminally truncated isoforms generated via alternative translation initiation Ryuji Kobayashi a,*, Rebecca Patenia b, Satoshi Ashizawa b,1, Jody Vykoukal b a b

FTN Institute, 1147 Mariner Cove, Sugar Land, TX 77498, USA Department of Molecular Pathology, Unit 951, University of Texas M.D. Anderson Cancer Center, 7435 Fannin Street, Houston TX 77054, USA

a r t i c l e

i n f o

Article history: Received 19 March 2009 Revised 21 May 2009 Accepted 23 May 2009 Available online 28 May 2009 Edited by Michael Ibba Keywords: Targeted mass spectrometry Alternative translation initiation Dok-1 protein isoform Na-acetylation N-terminal truncation

a b s t r a c t Alternative translation initiation is a mechanism whereby functionally altered proteins are produced from a single mRNA. Internal initiation of translation generates N-terminally truncated protein isoforms, but such isoforms observed in immunoblot analysis are often overlooked or dismissed as degradation products. We identified an N-terminally truncated isoform of human Dok-1 with N-terminal acetylation as seen in the wild-type. This Dok-1 isoform exhibited distinct perinuclear localization whereas the wild-type protein was distributed throughout the cytoplasm. Targeted analysis of blocked N-terminal peptides provides rapid identification of protein isoforms and could be widely applied for the general evaluation of perplexing immunoblot bands. Ó 2009 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

1. Introduction Protein isoforms generated through alternative translation initiation have been known for more than two decades, and they appear to have important functions that are distinct from their more extensively characterized wild-type counterparts [1–6]. N-terminal truncation resulting from alternative translation initiation via a mechanism such as leaky scanning can influence protein subcellular localization and lead to formation of distinct protein complexes, or promote specific interactions with proteins or other molecules. Despite the recent dramatic increase in attention given to the field of proteomics following completion of the genome project and the development of mass spectrometry for protein characterization, these isoforms have received little attention in typical proteomic studies. Perhaps this is because the Nterminal peptides are not routinely detected in high throughput proteomic analysis without explicit targeting. These truncated isoforms are readily detectable as faster migrating bands in SDS–PAGE and possible isoforms are frequently observed in immunoblot analysis, but they have largely been dismissed as

* Corresponding author. Fax: +1 281 277 9440. E-mail address: [email protected] (R. Kobayashi). 1 Present address: Musahimurayama Hospital, Musahimurayama-shi, Tokyo 2080022, Japan.

degradation products of wild-type proteins. On the other hand, it is reported that at least 80% of eukaryotic cellular proteins are N-terminally blocked [7]. Additionally, the most commonly found N-terminal modification is cotranslational Na-acetylation of nascent polypeptides. Human p62dok (also called Dok-1) [8,9] was first identified as a tyrosine-phosphorylated 62 kDa protein in transformed and growth factor stimulated cells and is associated with RasGAP [8]. P62dok is constitutively tyrosine-phosphorylated in hematopoietic progenitor cells isolated from chronic myelogenous leukemia patients [9]. An isoform was found in T and B cells as C-terminally truncated protein, p22Dokdel that is generated via alternative splicing [10]. Another C-terminally truncated isoform discovered in a chronic lymphocytic leukemia patient was determined to be caused by a frameshift mutation [11]. In both cases, the PH domain of Dok-1 was retained. Here we report on an N-terminally truncated Dok-1 isoform that exhibits a unique subcellular localization. We also report for the first time the identification of an N-terminally truncated isoform generated via alternative translation initiation that receives the same protein modification as the wild-type. Targeting blocked N-terminal peptides for mass spectrometry proved a very efficient strategy for rapid determination of the translation initiation site in the investigation we report here, and we therefore suggest that the strategy could be applied to facilitate the study of other proteins, as well.

0014-5793/$36.00 Ó 2009 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2009.05.042

2442

R. Kobayashi et al. / FEBS Letters 583 (2009) 2441–2445

2. Materials and methods

2.4. Isolation of the blocked N-terminal peptide of p44dok

2.1. Cell lines

The 44 kDa proteins in cell lysates of the C-myc dok-1 Mo7 cells were enriched on immobilized anti-myc tag antibody (ProFound cMyc TagIP/Co-IP kit, Pierce) and eluted with myc-tag peptide (100 lg/ml) in 25 mM Tris–HCl containing 150 mM NaCl (pH 7.2). The eluted proteins were denatured for 10 min at 80 °C in 50 lL of the digestion buffer prior to protease treatment. The digestion was performed using either Lys-N (Sigma) or modified trypsin overnight at 37 °C. The digest was then mixed with isothiocyanato-glass and incubated for 3 h at 50 °C. The reaction mixture was filtered with an ultrafreeMC filter (0.22 lm, Millipore). The flow-through fraction from the isothiocyanato-glass bead was desalted using a ZipTip-C18 (Millipore) and analyzed by nanoLCMS/MS.

Leukemia cell line (K562), pancreatic cancer cell lines (MIAPaCa-2, SU86.86, PANC-1, AsPC-1, BxPC-3, Capan-1), colon cancer cell lines (RKO), malignant glioma cell line (U87-MG), and fibroblasts (IMR-90, WI-38) were obtained from ATCC. Malignant glioma cell line U373-MG was obtained from Sigma. Pancreatic cancer cell lines (PANC-3, PANC-28, and PANC-48) were gifted by Dr. Li and lung cancer cell line (H226) and its brain metastasis in mouse (H226Br) were gifted by Dr. Herbst, both of MD Anderson Cancer Center. The human Mo7 megakaryoblastic leukemic cell line previously described [9] was maintained in our lab in RPMI1640 containing 10 ng/ml IL-3 (PeproTech EC). P210Bcr-Abl expressing Mo7 cell line was established in our lab via retroviral transduction as also previously described [12].

2.5. Immunofluorescence study Three cell lines, Mo7p210Bcr-Abl, Mo7p210Bcr-Ablp62dok, and Mo7p210Bcr-Ablp44dok, were cultured in RPMI1640 containing 10% fetal bovine serum (Hyclone). Approximately 5  104 cells were washed with PBS and applied to slides using a Cytospin (Model7620 Cytocentrifuge, Wesco). The cells were fixed in cold methanol for 10 min at 20 °C and permeabilized for 10 min at room temperature with 0.2% TritonX-100. After blocking with 5% bovine serum albumin (Sigma) in PBS for 30 min, the samples were probed with AlexaFluor488 conjugated anti-myc-tag monoclonal antibody (9B11, Cell Signaling) at room temperature for 2 h. The slides were washed with PBS and water, and mounted with Vectashield (Vector Laboratories). Images were taken using a confocal microscopy (Nikon D-Eclipse C1) under 60 magnification and analyzed using imaging software EZ-C1 3.40 (Nikon).

2.2. Immunoblot analysis and antibodies Whole cell extracts were prepared using lysis buffer (50 mM Tris–HCl, pH 7.6, 150 mM NaCl, 5 mM EDTA, 1 mM EGTA, 1%Triton X-100, 0.1 lg/ml aprotinin, 0.5 lg/ml leupeptin, 100 lg/ml PMSF, 1 mM Na3VO4, 1 mM NaF and 100 lM PAO). Cellular debris was removed by centrifugation at 14,000g for 30 min at 4 °C. Samples were mixed with Laemmli’s loading buffer, boiled for 5 min, and subjected to SDS–PAGE. 50 lg of protein from each cell line was separated by SDS–PAGE (Criterion 4–15% gradient gel, BioRad). Proteins were then transferred to PVDF membrane (Immobilon-P, Millipore) and probed with specific antibodies. Anti-Dok-1 polyclonal antibody [9], anti-myc-tag monoclonal antibody (4A6, Millipore), anti-pTyr monoclonal antibody (4G10, Millipore) and antibeta-actin (A-5441, Sigma) were used for detection. SuperSignal West Pico chemiluminescent substrate (Pierce) was used for immunoblot analysis.

3. Results 3.1. Immunoblot analysis of human cell lines using anti-Dok-1 antibody

2.3. LC-MS/MS analysis

As shown in Fig. 1, immunoblot analysis using Dok-1 specific antibody revealed 44 kDa major bands in addition to the band for wild-type p62dok for the bulk of the cell lines. We refer to this protein isoform as p44dok. 3.2. Sequence analysis of the lower molecular weight protein observed in immunoblot analysis

WI-38

IMR-90

U373-MG

U87-MG

RKO

Capan-1

BxPC-3

AsPC -1

Extensive LC-MS/MS analysis of the 44 kDa band revealed that it is essentially the C-terminal half region of p62dok (Fig. 2). However, Edman degradation (Procise474, Applied Biosystems) of the

PANC-48

PANC-28

PANC-3

PANC-1

SU.86.86

MIA PaCa-2

H226/Br

H226

blot:

K562

Mass spectrometric analysis was performed using an electrospray ion-trap mass spectrometer (LTQ, ThermoElectron). Solvents used for LC-MS/MS analysis were all HPLC grade from Burdick & Jackson. Modified trypsin (Promega) was used for in-gel digestion. The peptides were separated by nano-LC (UltiMate, LC-PackingsDionex) using a C18 column (PepMap 75 lm  150 mm, LC-Packings) and eluted into the LTQ by an increasing gradient of acetonitrile containing 0.01%TFA. The mass spectra were searched against the NCBI database using Mascot (ver. 2.1.03) or SEQUEST (Bioworks 3.1).

p62dok anti-Dok-1

p44dok

β-actin lane:

1

2

3

4

5

6

7

8

9

10

11

12

13

14 15

16

17

Fig. 1. Immunoblot analysis of human cell lines using anti-Dok1 antibody. Lane 1 is a leukemia line; lanes 2 and 3 are respectively a lung cancer line and its brain metastasis in mouse; lanes 4–12 are pancreatic cancer lines; lane 13 is a colon cancer line; lanes 14 and 15 are malignant glioma lines, and lanes 16 and 17 are fibroblasts. P62dok and p44dok bands are indicated. Antibody epitope is illustrated in Fig. 2.

R. Kobayashi et al. / FEBS Letters 583 (2009) 2441–2445

2443

Fig. 2. LC-MS/MS of p44dok and schematic diagram for comparison of p62dok and p44dok. The amino acid sequence of Dok-1 is shown (top panel) and matching peptide hits from LC-MS/MS analysis of p44dok are indicated in red. The p44dok sequence begins at Met-140 of the Dok-1 sequence. The hooked arrows indicate translation initiation sites for p62dok and p44dok. All Met residues are underlined. The blocked N-terminal peptides identified in LC-MS/MS are indicated by the blue underline. Epitope sequence recognized by Dok-1 antibody is boxed and indicated by an arrow. N-termini of both p62dok and p44dok are acetylated. The PH and PTB domains are indicated (bottom panel), as well as residue numbers. The Dok-1 antibody recognition site is indicated by the inverted Y symbol.

44 kDa protein was unsuccessful. Furthermore, we were unable to detect (data not shown) the missing N-terminal domain fragment by immunoblot analysis using a Dok-1 PH domain specific antibody (kindly gifted from Dr. Van Aelst, Cold Spring Harbor Laboratory). This result led us to speculate that this protein was perhaps not a degradation product and might also be N-terminally blocked. We therefore isolated the N-terminal peptides of the potentially blocked p44dok protein for mass spectrometric analysis using our previously described method [13]. We were able to identify the acetylated N-terminal peptide of p62dok (Supplementary Fig. 1A); however, the blocked N-terminal peptide of p44dok was not detectable using the Lys-N digestion approach. Subdigestion of the N-terminal fraction using endoproteinase Glu-C indicated a potential Nterminal fragment but did not provide a definitive answer. We were finally able to identify acetylated-MLENSLYSPTWEGSQFWVTVQR as an N-terminal fragment of the 44 kDa protein using trypsin digest (Supplementary Fig. 1B). Like the N-terminus of p62dok, the N-terminal end of p44dok starts at an acetylated Met; however, p44dok starts at the third methionine (Met-140) of Dok-1 (Fig. 2). LC-MS/MS analysis of p44dok covered 88.9% of the C-terminal region of the Dok-1 sequence from M-140 to T-481. Because the first Lys located at downstream of M-140 is Lys-213, the Lys-N digest of p44dok produced a 74 amino acid-long peptide. This peptide was too large for MS/MS analysis. We also found that p44dok lacked the Pleckstrin homology (PH) domain. 3.3. Insertion of myc-tag sequence after the first AUG at 50 -end interferes with p44dok production To test if a single cDNA produces the two forms of protein, we expressed either N-myc-dok-1 or C-myc-dok-1 in Mo7p210Bcr-Abl

cells from which we originally purified p62dok [9]. Because a single dok-1 cDNA generated both p62dok and p44dok, the possibility of alternative splicing as a mechanism of p44dok production was eliminated. Interestingly, when N-myc-dok-1 was expressed, the 44 kDa protein band was discernable than that of p62dok, whereas when C-myc-dok-1 was expressed, the 44 kDa band appeared much stronger than that of p62dok (Fig. 3). This finding suggests that p44dok is a product of alternative translation initiation, presumably by leaky scanning. We did not observe p44dok in immunoblot analysis of N-terminally myc-tagged Dok-1 expression using antibody specific for the Dok-1 C-terminus epitope indicated in Fig. 2, nor did immunoblot analysis using Dok-1 PH domain specific antibody detect any plausible bands for fragments containing the N-terminal domain. We therefore eliminated proteolysis of p62dok as a likely mechanism for production of p44dok. Additionally, in vitro transcription/translation of Dok-1 cDNA using a rabbit reticulocyte lysate cell free expression system (which does not likely possess the same proteolytic pathways as intact human cells) yields a pattern for production of p62dok and p44dok (Supplementary Fig. 2) that is similar to that shown in Fig. 3. 3.4. Nucleotide sequence surrounding Met-140 has a much weaker Kozak consensus sequence than Met-1 Alignment of nucleotide sequences surrounding all AUG codons in dok-1 with the Kozak consensus sequence [14] is shown in Table 1. Selection of the initiation codon is generally determined by the optimal Kozak consensus sequence GCC(A/G)CCAUGG with the 3 and +4 positions being the most influential. Interestingly, the sequence surrounding AUG for Met-1 is an especially strong Kozak consensus sequence match with G at the 3 and +4 positions. On

2444

R. Kobayashi et al. / FEBS Letters 583 (2009) 2441–2445

C-myc

N-myc

Vector

C-myc

N-myc

Vector

blot:

myc-p44dok to analyze the subcellular localization patterns of the different protein isoforms. Very interestingly, p44dok exhibited perinuclear localization, whereas p62dok was more broadly distributed throughout the cytoplasm (Fig. 4).

Mo7 p210Bcr-Abl

Mo7

p62dok anti-Dok-1 -actin p62dok anti-myc-tag

p44dok

-actin lane:

4. Discussion

p44dok

1

2

3

4

5

6

Fig. 3. Location of myc-tag influences production of p44dok. P44dok was absent (or substantially weaker) when myc-tag was added to N-terminus (N-myc) in immunoblot analysis using anti-Dok-1 antibody (lanes 2 and 5). In contrast, when myc-tag was added at the C-terminus (C-myc), the p44dok proteins were expressed much more abundantly than p62dok (lanes 3 and 6) while p62dok production was reduced. Vector refers to the vector only.

Table 1 Alignment of nucleotides surrounding all AUG codons in dok-1 with the Kozak consensus sequence. The most influential positions, 3 and +4 are highlighted in boldface. Kozak: Met-1 Met-6 Met-140 Met-215

GCCACCAUGG G GGGGCCAUGG GCAGUGAUGG CUGGAGAUGC AAGGUCAUGU

the other hand, the sequence surrounding Met-140 is suboptimal. As is evident in Fig. 3, p44dok is produced in much greater abundance relative to p62dok, contrasting the expression levels that would be expected based on the comparative strength of the associated Kozak sequences. 3.5. Subcellular localization Since it has been reported that the PH domain of recombinant p62dok is necessary for localization to membranes [15], we expected that p44dok would likely not migrate to the plasma membrane. We used cell lines expressing either N-myc-Dok-1 or N-

a

blot: anti-myc beta-actin

Protein isoforms produced through alternative translation initiation are probably identified much less frequently than their actual occurrence in nature despite their potential functional importance and influence in biological systems. It is well known that most human cellular proteins are naturally N-terminally blocked. In this study, we identified a new p62dokisoform, p44dok that was produced via alternative translation initiation. The N-terminal truncation that yields p44dok results in specific localization of the isoform within the perinuclear region. In immunoblot analysis, the isoform p44dok appeared to be ubiquitously expressed (Fig. 1); however, the role of the isoform is currently unknown. As we demonstrate in this report, targeting blocked N-terminal peptide provides a straightforward approach for identifying previously overlooked protein isoforms. We also found that p44dok and p62dok have the same posttranslational modification – acetylation – at their respective N-termini. Although some proteins are subject to regulated degradation through the N-end rule pathway [16], it is unlikely that such processing leads to the formation of p44dok, since the truncated isoform is produced both in vivo as well as in the in vitro cell free transcription/translation system used in this study. In contrast, the acetylated N-terminal Met of p44dok (and p62dok) is consistent with known cotranslational N-terminal processing that is conserved in eukaryotes [17] wherein N-terminal methionines remain if the penultimate residue is Asp or Leu (as is the case for p62dok and p44dok, respectively). Na-acetyltransferase anchors to the ribosome and interacts with nascent polypeptides while they are within 20–50 amino acids in length [17,18]. N-terminal acetylation is almost exclusively found to occur cotranslationally, and Na-acetylation following translation has been reported only in the particular case of short peptides such as a-melanocyte-stimulating hormone (13 residues) [19], leading us to postulate that acetylation of p44dok most likely occurs cotranslationally. The Na-acetylation of N-terminally truncated proteins could be common feature to isoforms generated via alternative translation initiation – and possibly alternative splicing, as well. This appears to be the first direct evidence of this phenomenon.

b

1

2

c

4

3 p62dok p44dok

5

6

blot: anti-Dok-1 beta- actin

Fig. 4. P44dok has distinct subcellular localization compared to p62dok. (A) Immunofluorescent study. N-myc-p44dok (a) or N-myc-p62dok (b) was visualized using anti-myc tag antibody with AlexaFluor488 in Mo7p210Bcr-Abl cells. P44dok exhibited marked perinuclear localization (a), whereas the wild-type was distributed throughout the cytoplasm (b). Mo7p210Bcr-Abl cells (vector only) were used as a control for background fluorescence (c). Similar results were obtained in at least three separate experiments. (B) Immunoblot analysis of the cell lines used for immunofluorescent study. The left panel shows immunoblot using anti-myc antibody and the right panel shows that with anti-Dok-1 antibody. Lanes 1 and 4 are N-myc-p44dok expressing Mo7p210Bcr-Abl cells, lanes 2 and 5 are N-myc-p62dok expressing Mo7p210Bcr-Abl cells, and lanes 3 and 6 are the vector control. Only myc-tagged protein bands appear in lanes 1 and 2 whereas endogenous p62dok and p44dok are also detected in lanes 4–6 in addition to the myctagged proteins detected in lanes 1 and 2.

R. Kobayashi et al. / FEBS Letters 583 (2009) 2441–2445

Insertion of the N-terminal myc-tag sequence following the first AUG appears to interfere with the ribosomal scanning, resulting in attenuation of translation initiation at the third AUG for Met-140 (Fig. 3), and a concomitant reduction in the production of p44dok. A possible explanation for this is that the mechanism of alternative initiation for p44dok production is leaky scanning caused by the second AUG (Met-6) being located in close proximity (12 nucleotides) to the first AUG. Considering that a single ribosome contacts approximately 30 nucleotides of the mRNA during translation, it is conceivable that these proximate AUGs cause ribosome wobbling that impedes normal translation, but that ribosomal subunits that continue scanning downstream reach third AUG are able to successfully initiate translation. Insertion of the myc-tag sequence (33 nucleotides) between the first two AUGs creates additional space for assembly of the initiation complex at the first AUG and could thus stabilize recognition of the initiation site, resulting in a preclusion of downstream alternative initiation that leads to p44dok production. In other studies, more stable initiation at a first AUG was reported [20,21] when it was moved away from a nearby second AUG. In contrast to previously discovered Dok-1 isoforms, p44dok lacks a PH domain but preserves the RasGAP binding sites [22]. It has been reported that Dok-1 inhibits PDGF-induced mitogenesis and has a role as a tumor suppressor [23]. Recently, Dok-1 was found to be involved in diet-induced adipocyte hypertrophy and obesity [24]. Dok-1 appears to be involved in a number of other diverse biological functions. N-terminally truncated isoforms can be generated through mechanisms including alternative translation initiation or alternative splicing. Like alternative splicing [25], alternative translation initiation may play a vital role in regulation and control by yielding protein variants that are linked to a distinct biological functionality and that help to maintain the fidelity of biological systems. Acknowledgements We are grateful to Bih-Fang Pan for her excellent technical assistance and David Hawke in the MD Anderson Cancer Center supported Proteomics Facility for mass spectrometry. We are also grateful to Arne Stenlund, Dan Jones, Marcus T. Kuo, and Rolf Sternglanz for valuable discussion, and special thanks to Daynene Vykoukal for valuable advice throughout the course of this study and in helping to prepare this report of our findings. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.febslet.2009.05.042. References [1] Strubin, M., Long, E.O. and Mach, B. (1986) Two forms of the Ia antigenassociated invariant chain result from alternative initiations at two in-phase AUGs. Cell 47, 619–625. [2] Spence, A.M., Sheppard, P.C., Davie, J.R., Matuo, Y., Nishi, N., McKeehan, W.L., Dodd, J.G. and Matusik, R.J. (1989) Regulation of a bifunctional mRNA results in synthesis of secreted and nuclear probasin. Proc. Natl. Acad. Sci. USA 86, 7843–7847.

2445

[3] Descombes, P. and Schibler, U. (1991) A liver-enriched transcriptional activator protein, LAP, and a transcriptional inhibitory protein, LIP, are translated from the same mRNA. Cell 67, 569–579. [4] Ossipow, V., Descombes, P. and Schibler, U. (1993) CCAAT/enhancer-binding protein mRNA is translated into multiple proteins with different transcription activation potentials. Proc. Natl. Acad. Sci. USA 90, 8219–8223. [5] Vanhoutte, P., Nissen, J.L., Brugg, B., Gaspera, B.D., Besson, M.-J., Hipskind, R.A. and Caboche, J. (2001) Opposing roles of Elk-1 and its brain-specific isoform, short elk-1, in nerve growth factor-induced PC12 differentiation. J. Biol. Chem. 276, 5189–5196. [6] Courtois, S., Verhaegh, G., North, S., Luciani, M.G., Lassus, P., Hibner, U., Oren, M. and Hainaut, P. (2002) DeltaN-p53, a natural isoform of p53 lacking the first transactivation domain, counteracts growth suppression by wild-type p53. Oncogene 21, 6722–6728. [7] Brown, J.L. and Roberts, W.K. (1976) Evidence that approximately eighty per cent of the soluble proteins from Ehrlich ascites cells are Na-acetylated. J. Biol. Chem. 251, 1009–1014. [8] Ellis, C., Moran, M., McCormick, F. and Pawson, T. (1990) Phosphorylation of GAP and GAP-associated proteins by transforming and mitogenic tyrosine kinases. Nature 343, 37781. [9] Carpino, N., Wisniewski, D., Strife, A., Marshak, D., Kobayashi, R., Stillman, B. and Clarkson, B. (1997) p62dok: A constitutively tyrosine-phosphorylated, GAP-associated protein in chronic myelogenous leukemia progenitor cells. Cell 88, 197–204. [10] Hubert, P., Ferreira, V., Debre, P. and Bismuth, G. (2000) Molecular cloning of a truncated p62Dok1 isoform, p22Dokdel. Eur. J. Immunogenet. 27, 145–148. [11] Lee, S., Roy, F., Galmarini, C.M., Accardi, R., Michelon, J., Viller, A., Cros, E., Dumontet, C. and Sylla, B.S. (2004) Frameshift mutation in the Dok1 gene in chronic lymphocytic leukemia. Oncogene 23, 2287–2297. [12] Fearnhead, H.O., Rodriguez, J., Govek, E.E., Guo, W., Kobayashi, R., Hannon, G. and Lazebnik, Y.A. (1998) Oncogene-dependent apoptosis is mediated by caspase-9. Proc. Natl. Acad. Sci. USA 95, 13664–13669. [13] Coussot, G., Hawke, D.H., Mularz, A., Koomen, J.M. and Kobayashi, R. (2007) A method for the isolation of blocked N-terminal peptides. Anal. Biochem. 361, 302–304. [14] Kozak, M. (1991) Structural features in eukaryotic mRNAs that modulate the initiation of translation. J. Biol. Chem. 266, 19867–19870. [15] Zhao, M., Schmitz, A.A., Qin, Y., Di Cristofano, A., Pandolfi, P.P. and Van Aelst, L. (2001) Phosphoinositide 3-kinase-dependent membrane recruitment of p62dok is essential for its negative effect on mitogen-activated protein (MAP) kinase activation. J. Exp. Med. 194, 265–274. [16] Varshavsky, A. (2005) Regulated protein degradation. Trends Biochem. Sci. 30, 283–286. [17] Polevoda, B. and Sherman, F. (2000) Nalpha -terminal acetylation of eukaryotic proteins. J. Biol. Chem. 275, 36479–36482. [18] Gautschi, M., Just, S., Mun, A., Ross, S., Rucknagel, P., Dubaquie, Y., EhrenhoferMurray, A. and Rospert, S. (2003) The yeast N(alpha)-acetyltransferase NatA is quantitatively anchored to the ribosome and interacts with nascent polypeptides. Mol. Cell. Biol. 23, 7403–7414. [19] Granger, M., Tesser, G.I., De Jong, W.W. and Bloemendal, H. (1976) Model studies of enzymatic NH2-terminal acetylation of porteins with des-Nalpha1acetyl-alpha-melanotropin as a substrate. Proc. Natl. Acad. Sci. USA 73, 3010– 3014. [20] Williams, M.A. and Lamb, R.A. (1989) Effect of mutations and deletions in a bicistronic mRNA on the synthesis of influenza B virus NB and NA glycoproteins. J. Virol. 63, 28–35. [21] Matsuda, D. and Dreher, T.W. (2006) Close spacing of AUG initiation codons confers dicistronic character on a eukaryotic mRNA. RNA 12, 1338–1349. [22] Kashige, N., Carpino, N. and Kobayashi, R. (2000) Tyrosine phosphorylation of p62dok by p210bcr-abl inhibits RasGAP activity. Proc. Natl. Acad. Sci. USA 97, 2093–2098. [23] Zhao, M., Janas, J.A., Niki, M., Pandolfi, P.P. and Van Aelst, L. (2006) Dok-1 independently attenuates Ras/mitogen-activated protein kinase and Src/cmyc pathways to inhibit platelet-derived growth factor-induced mitogenesis. Mol. Cell. Biol. 26, 2479–2489. [24] Hosooka, T., Noguchi, T., Kotani, K., Nakamura, T., Sakaue, H., Inoue, H., Ogawa, W., Tobimatsu, K., Takazawa, K., Sakai, M., Matsuki, Y., Hiramatsu, R., Yasuda, T., Lazar, M.A., Yamanashi, Y. and Kasuga, M. (2008) Dok1 mediates high-fat diet-induced adipocyte hypertrophy and obesity through modulation of PPARgamma phosphorylation. Nat. Med. 14, 188–193. [25] Wakasugi, K., Slike, B.M., Hood, J., Otani, A., Ewalt, K.L., Friedlander, M., Cheresh, D.A. and Schimmel, P. (2002) A human aminoacyl-tRNA synthetase as a regulator of angiogenesis. Proc. Natl. Acad. Sci. USA 99, 173–177.