Eur J Biochem 271, 863–874 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2004.03991.x Vertical-scanning mutagenesis of amino acids in a model N-myristoylation motif reveals the major amino-terminal sequence requirements for protein N-myristoylation Toshihiko Utsumi, Kengo Nakano, Takeshi Funakoshi, Yoshiyuki Kayano, Sayaka Nakao, Nagisa Sakurai, Hiroyuki Iwata1 and Rumi Ishisaka Department of Biological Chemistry and 1Department of Veterinary Medicine, Faculty of Agriculture, Yamaguchi University, Yamaguchi, Japan In order to determine the amino-terminal sequence requirements for protein N-myristoylation, site-directed mutagenesis of the N-terminal region was performed using tumor necrosis factor (TNF) mutants as model substrate proteins Subsequently, the susceptibility of these mutants to protein N-myristoylation was evaluated by metabolic labeling in an in vitro translation system using rabbit reticulocyte lysate A TNF mutant having the sequence MGAAAAA AAA at its N-terminus was used as the starting sequence to identify elements critical for protein N-myristoylation Sequential vertical-scanning mutagenesis of amino acids at a distinct position in this model N-terminal sequence revealed the major sequence requirements for protein N-myristoylation: the combination of amino acids at position and constitutes a major determinant for the susceptibility to protein N-myristoylation When Ser was located at position 6, 11 amino acids (Gly, Ala, Ser, Cys, Thr, Val, Asn, Leu, Ile, Gln, His) were permitted at position to direct efficient protein N-myristoylation In this case, the presence of Lys at position was found to affect the amino acid requirement at position and Lys became permitted at this position When Ser was not located at position 6, only amino acids (Ala, Asn, Gln) were permitted at position to direct efficient protein N-myristoylation The amino acid requirements found in this study were fully consistent with the N-terminal sequence of 78 N-myristoylated proteins in which N-myristoylation was experimentally verified These observations strongly indicate that the combination of amino acids at position 3, and is a major determinant for protein Nmyristoylation A number of eukaryotic cellular proteins are found to be covalently modified with the 14-carbon saturated fatty acid, myristic acid [1–5] Many of the N-myristoylated proteins play key roles in regulating cellular structure and function They include proteins involved in a wide variety of cellular signal transduction pathways In general, protein N-myristoylation is the result of cotranslational addition of myristic acid to a Gly residue at the extreme N-terminus after removal of the initiating Met A stable amide bond links myristic acid irreversibly to proteins N-Myristoylation can also occur post-translationally, as in the case of the pro-apoptotic protein BID and cytoskeletal actin, where proteolytic cleavage by caspase reveals a ÔhiddenÕ myristoylation motif [6,7] N-Myristoylation is catalyzed by N-myristoyltransferase (NMT), a member of the GCN5 acetyltransferase (GNAT) superfamily of proteins [8] NMT has been purified and cloned from many organisms [9–12] and its substrate specificities have been characterized In general, Ser or Thr is preferred at position 6, and the N-terminal consensus motif Me-GlyX-X-X-Ser/Thr that directs protein N-myristoylation has been defined [13] Saccharomyces cerevisiae NMT (NMT1p) is the best studied of the known NMTs The precise substrate specificity of this enzyme has been characterized using purified enzyme and synthetic peptides derived from the N-terminal sequences of known N-myristoylated proteins [1,14,15] These studies have produced a set of empiric rules for amino acid requirements at distinct positions in the N-myristoylation motif as follows: (a) the requirement for Gly at position is absolute; (b) charged residues, aromatics and Pro are not allowed at position 3; (c) all amino acids are allowed at positions and 5; (d) Ser, Thr, Ala, Gly, Cys, or Asn are permitted at position 6; (e) all but Pro are allowed at position [5] Thus, it is well accepted that in addition to Gly at position 2, the amino acids at positions 3, 6, and play an important role in substrate recognition by NMT In fact, it was demonstrated that the difference in the substrate specificity of NMT in different species depends mainly on the difference in the permitted amino acid residues in these three positions [16,17] However, the Correspondence to T Utsumi, Department of Biological Chemistry, Faculty of Agriculture, Yamaguchi University, Yamaguchi 753-8515, Japan Fax: + 81 83 933 5820, Tel.: + 81 83 933 5856, E-mail: utsumi@yamaguchi-u.ac.jp Abbreviations: DMEM, Dulbecco modified Eagle’s medium; DPBS, Dulbecco’s phosphate-buffered saline; NMT, N-myristoyltransferase; TNF, tumor necrosis factor (Received November 2003, revised 25 December 2003, accepted 13 January 2004) Keywords: N-myristoylation motif; N-myristoyltransferase; protein N-myristoylation; substrate specificity; vertical scanning mutagenesis Ó FEBS 2004 864 T Utsumi et al (Eur J Biochem 271) relative roles of these residues in substrate recognition or the relationship between the amino acids that reside at these three positions have not been well characterized so far Proteins destined to become N-myristoylated begin with the sequence Met-Gly However, proteins having the MetGly sequence at their N-terminus may also be subjected to another cotranslational protein modification, N-acetylation In fact, many proteins having an N-terminal MetGly sequence, such as ovalbumin [18], cytochrome c [19], actin [20], and 20S proteasome a3 subunit [21] have been found to be N-acetylated N-Acetyltransferases that catalyze cotranslational protein N-acetylation also have a restricted number of substrates [22–24] However, the differences in the N-terminal sequence requirements for protein N-myristoylation and protein N-acetylation have not been fully characterized yet In a previous report, we showed that metabolic labeling in an in vitro translation system is an effective strategy to characterize cotranslational N-terminal protein modifications [25,26] As the in vitro translation system using rabbit reticulocyte lysate contains all the components involved in cotranslational protein N-myristoylation and N-acetylation [18,20,27], the use of this system to study cotranslational protein N-myristoylation seems to be appropriate Using this assay system, we demonstrated previously that the amino acid residue at position strongly affects protein Nmyristoylation, and the amino acid requirements at this position are significantly affected by the amino acid at position [25] These results suggested that the combination of amino acids at positions and might be a critical determinant for protein N-myristoylation In this study, to examine the effect of the combination of amino acids at positions and on protein N-myristoylation, sequential vertical-scanning mutagenesis of the amino acids at positions and in a model N-terminal sequence was performed and the susceptibility of these mutants to protein N-myristoylation was evaluated by metabolic labeling in an in vitro translation system using rabbit reticulocyte lysate Experimental procedures Materials Restriction endonucleases, DNA-modifying enzymes, RNase inhibitor, and Taq DNA polymerase were purchased from Takara Shuzo (Japan) The mCAP RNA capping kit and proteinase K were from Stratagene RNase was purchased from Boehringer-Mannheim (Germany) Rabbit reticulocyte lysate was from Promega [3H]leucine, [3H]myristic acid, [35S]methionine and Amplify were from Amersham (UK) The Dye Terminator Cycle Sequencing kit was from Applied Biosystems Anti-human TNF polyclonal Ig was purchased from R & D systems Protein G Sepharose was from Pharmacia Biotech Other reagents purchased from Wako Pure Chemical, Daiichi Pure Chemicals, and Seikagaku Kogyo (Japan) were of analytical or DNA grade Plasmid construction Plasmid pBluescript II SK(+) lacking ApaI and HinDIII sites was constructed as described previously [28], and designated pB Plasmid pBDpro-TNF, which contains a cDNA coding for the mature domain of TNF, was constructed as described [28,29] Plasmid pBMA(9)-TNF was constructed by utilizing PCR For this procedure, pBDpro-TNF served as a template, and two oligonucleotides [MA(9), B1] as primers (Table 1) After digestion with BamHI and PstI, the amplified product was subcloned into pB at the BamHI and PstI sites Plasmids pBMGA(8) and pBMG6S were constructed by a method similar to that used to construct pBMA(9)-TNF using two primers [MGA(8) and MG6S, respectively] as mutagenic primers (Table 1) The cDNAs coding for MG6X-TNF, in which Ser at position in MG6S-TNF was replaced with each of the 19 other amino acids, were constructed by using a degenerated primer, MG6X, as mutagenic primer After digestion with BamHI and PstI, the amplified product was subcloned into pB at the BamHI and PstI sites The DNA Table Nucleotide sequences of oligonucleotides used for the construction of mutant TNF cDNAs N, A + C + G + T; K, T + G Primer Sequence (5¢fi3¢) MA(9) MGA(8) MG6S MG6X MG3X XHO-TNF13 6S-XHO 6T-XHO 6F-XHO C3K-A7K HC-K7A MG3K6S MG3K6S7K T3 B1 ATATGGATCCATGGCTGCGGCAGCAGCGGCAGCAGCAGCAGACAAGCCTGTAGCC ATATGGATCCATGGGCGCGGCAGCAGCGGCAGCAGCAGCAGACAAGCCTGTAGCC GCCGGGATCCATGGGCGCAGCAGCATCTGCAGCAGCAGCAGACAAGCCTGTAGCC GCCGGGATCCATGGGCGCAGCAGCANNKGCAGCAGCAGCAGAC ATATGGATCCATGGGCNNKGCAGCAGCGGCAGCAGCAGCAGAC GCCGCTCGAGCCTGTAGCCCATGTT AATTCTCGAGTGCTGCTGCTGCCGATGCTGC AATTCTCGAGTGCTGCTGCTGCCGTTGCTGC AATTCTCGAGTGCTGCTGCTGCGAATGCTGC GCCGGGATCCATGGGCAAAACGCTGAGCAAAGAGGACAAGCTCGAG GCCGGGATCCATGGGCAAGCAGAATAGCGCACTGCGGCCAGACAAG GCCGGGATCCATGGGCAAGGCAGCATCTGCAGCAGCAGCAGACAAGCCTGTAGCC GCCGGGATCCATGGGCAAGGCAGCATCTAAGGCAGCAGCAGACAAGCCTGTAGCC AATTAACCCTCACTAAAGGG GCCGGGATCCTAGGGCGAATTGGGTACC Ó FEBS 2004 Analysis of the N-myristoylation consensus motif (Eur J Biochem 271) 865 sequences of the obtained plasmids were determined by the dideoxynucleotide chain termination method and plasmids having a distinct triplet codon corresponding to each of the 19 amino acids at position were obtained The cDNAs coding for MG3X-TNF, in which the Ala at position in MGA(8)-TNF was replaced with each of the 19 other amino acids, were constructed by a method similar to that used to construct MG6X-TNF using a degenerated primer, MG3X, as a mutagenic primer pBGi1a-, pBGi1a-C3K- and pBhippocalcin-TNF were constructed as described previously [25] The cDNAs coding for MG3X6S-, MG3X6Tand MG3X6F-TNF, in which the Ala at position in MG3X-TNF was replaced with Ser, Thr and Phe, respectively, were constructed as follows pB Gi1a-D1-12-TNF, in which the DNA sequence encoding the N-terminal 12 amino acids of the mature domain of TNF was deleted from pBGi1a-TNF, was first generated from pBGi1a-TNF by PCR For this procedure, pBDpro-TNF served as a template and two oligonucleotides (XHO-TNF13, B1) as primers After digestion with XhoI and PstI, the amplified product was subcloned into pBGi1a-TNF at the XhoI and PstI sites DNA fragments coding for the N-terminal 10 residues of MG3X6S-, MG3X6T-, and MG3X6F-TNF were amplified by PCR In this case, pBMG3X-TNF served as a template and two oligonucleotides (T3 plus 6SXHO, T3 plus 6T-XHO, or T3 plus 6F-XHO, respectively) as primers After digestion with SacI and XhoI, the amplified product was subcloned into pB Gi1a-D1-12TNF at the SacI and XhoI sites In these three sets of TNF mutants, the amino acids at positions 11 and 12 were changed from Asp-Lys to Leu-Glu because of the insertion of the Xho I-linker sequence pBGi1a-C3K-A7K-TNF, in which the Ala at position in pBGi1a-C3K-TNF was replaced with Lys, was generated from pBGi1a-C3K-TNF by PCR In this case, pBGi1a-C3K-TNF served as a template and two oligonucleotides (C3K-A7K, B1) as primers After digestion with BamHI and PstI, the amplified products were subcloned into pB at the BamHI and PstI sites pBhippocalcin-K7A-TNF, pBMG3K6S-TNF, and pBMG3K6S7K-TNF were constructed by a method similar to that used to construct pBGi1a-C3K-A7K-TNF The mutagenic primers used to construct these three mutants were HC-K7A, MG3K6S, and MG3K6S7K, respectively (Table 1) The DNA sequences of these recombinant cDNAs were confirmed by the dideoxynucleotide chain termination method [30] or [3H]acetyl-CoA (2 lCi) and 3.0 lL of mRNA) was incubated at 30 °C for 90 Transfection of COS-1 cells and determination of N-myristoylated proteins The simian virus 40-transformed African Green monkey kidney cell line, COS-1, was maintained in Dulbecco modified Eagle’s medium (DMEM, Gibco BRL) supplemented with 10% fetal bovine serum (Gibco BRL) Cells (2 · 105) were plated onto 35 mm-diameter dishes day before transfection The pcDNA3 construct (2 lg; Invitrogen) containing mutant TNF cDNA was used to transfect each plate of COS-1 cells along with lL of LipofectAmine (2 mgỈmL)1; Gibco BRL) in mL of serum-free medium After incubation for h at 37 °C, the cells were re-fed with serum-containing medium and incubated again at 37 °C for 24 h The cells were then washed twice with mL of serumfree DMEM and incubated for h in mL of DMEM with 2% fetal bovine serum containing [3H]myristic acid (100 lCiỈmL)1) Subsequently, the cells were washed three times with Dulbecco’s phosphate-buffered saline (DPBS) and collected with cell scrapers, and then lysed with 200 lL of RIPA buffer [50 mM Tris/HCl (pH 7.5), 150 mM NaCl, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, proteinase inhibitors] on ice for 20 The cell lysates were centrifuged at 21 000 g at °C for 15 in a microcentrifuge (HITACHI-CF15D2) and supernatants were collected After immunoprecipitation with anti-TNF Ig, the samples were analyzed by SDS/PAGE and fluorography Western blotting TNF samples immunoprecipitated from in vitro translation products or total cell lysates of each group of transfected cells were resolved by 12.5% SDS/PAGE and then transferred to an Immobilon-P transfer membrane (Millipore) After blocking with nonfat milk, the membrane was probed with a specific goat anti-hTNF Ig as described previously [31] Immunoreactive proteins were specifically detected by incubation with horseradish peroxidase-conjugated anti-goat IgG (Santa Cruz) The membrane was developed with ECL Western blotting reagent (Amersham Corp.) and exposed to an X-ray film (Kodak) Quantitative analysis of immunoreactive proteins was carried out by scanning the X-ray film using an imaging densitometer (Bio-Rad GS-700) In vitro transcription and translation Methods essentially identical to those described previously were employed [28] T3 polymerase was used to obtain transcripts of these cDNAs subcloned into pB vector These transcripts were purified by phenol/chloroform extraction and ethanol precipitation prior to use Subsequently, the translation reaction was carried out using the rabbit reticulocyte lysate (Promega) in the presence of [3H]leucine, [35S]methionine or [3H]myristic acid under conditions recommended by the manufacturer The mixture (composed of 17.5 lL of rabbit reticulocyte lysate, 0.5 lL of mM leucine- or methionine-free amino acid mixture, or mM complete amino acid mixture, 4.0 lL of [3H]leucine (5 lCi), [35S]methionine (1 lCi), [3H]myristic acid (25 lCi) Immunoprecipitation Samples containing TNF mutants were immunoprecipitated with a specific goat anti-hTNF polyclonal Ig (R & D systems) as described [28] SDS/PAGE and fluorography Samples were denatured by boiling for in SDS/ sample buffer and then analyzed by SDS/PAGE on a 12.5% gel Thereafter, the gel was fixed and soaked in AmplifyTM (Amersham) for 30 The gel was dried under vacuum and exposed to X-ray film (Kodak) for an appropriate period Quantitative analysis of the labeled 866 T Utsumi et al (Eur J Biochem 271) Ó FEBS 2004 proteins was carried out by scanning the fluorogram using an imaging densitometer (Bio-Rad GS-700) Results Effect of the amino acid residue at position in the N-myristoylation consensus motif on the efficiency of the cotranslational N-myristoylation reaction To determine the amino-terminal sequence requirements for protein N-myristoylation, the relative roles of amino acids in the N-myristoylation consensus motif, especially those at positions and 6, in protein N-myristoylation were evaluated by metabolic labeling of model substrate proteins in an in vitro translation system In this case, to avoid the effects of amino acids at other positions in the N-terminal region on protein N-myristoylation, MA(9)-TNF, a TNF mutant in which amino acids following the initiating Met were changed to Ala, was used as the starting sequence to evaluate the roles of the amino acids at distinct positions in protein N-myristoylation (Fig 1) Fig MGA(8)-TNF with an N-terminal sequence MGAAAAAAAA is N-myristoylated The mRNAs encoding Gi1a-, MA(9)-, MGA(8)-, and MG6S-TNF were translated in vitro in the presence of [3H]leucine, [35S]methionine or [3H]myristic acid using rabbit reticulocyte lysate Following immunoprecipitation with anti-TNF Ig, the labeled translation products were analyzed by SDS/PAGE and fluorography (A) The cDNAs encoding Gi1a-, MA(9)-, MGA(8)-, and MG6S-TNF were transfected into COS-1 cells, and their expression and N-myristoylation were evaluated by Western blotting analysis and [3H]myristic acid-labeling, respectively (B) Fig Schematic representation of generation of MA(9)-, MGA(8)-, and MG6S-TNF cDNA coding for Dpro-TNF, which contains the mature domain of TNF, was first generated from pro-TNF cDNA by deleting the nucleotide sequence encoding the propeptide region of pro-TNF Subsequently, cDNAs of MA(9)-, MGA(8)-, and MG6STNF were generated from Dpro-TNF cDNA by site-directed mutagenesis As shown in Fig 2A (lane 2), translation of an mRNA coding for MA(9)-TNF in the presence of [3H]leucine gave rise to two translation products; one was the major product with an expected molecular mass (17 kDa) and the other was a fainter band with a molecular mass  kDa larger than expected No incorporation of [3H]myristic acid was detected in these translation products, as shown in Fig 2A (lane 10) [35S]Met labeling of this mutant revealed that [35S]Met was specifically incorporated into the upper of the two bands detected with [3H]Leu (lane 6) As there is no Met residue in the mature domain of TNF, this result indicates that the upper band corresponds to the protein species retaining the initiating Met residue and the lower band to the one lacking this residue When Ala at position was changed to Gly, the obtained mutant [MGA(8)-TNF] was efficiently N-myristoylated, as shown in lanes and 11 Ó FEBS 2004 Analysis of the N-myristoylation consensus motif (Eur J Biochem 271) 867 The levels of incorporation of [3H]Leu and [3H]myristic acid into the lower band of the expressed MGA(8)-TNF were comparable with those into Gi1a-TNF [32], in which the N-terminal 10 residues of the Gi1a protein were linked to the N-terminus of the mature domain of TNF (lanes 1, 3, and 11) These results revealed that MGA(8)-TNF is efficiently N-myristoylated, similarly to a protein having a natural N-myristoylation motif When Ala at position in MGA(8)-TNF was changed to Ser, a similar level of [3H]myristic acid incorporation was observed with the obtained mutant (MG6S-TNF), as shown in lanes and 12 The results obtained with MGA(8)- and MG6S-TNF clearly indicate that Ser at position is not critical for protein N-myristoylation When these four mutants were expressed in COS-1 cells and their susceptibility to protein N-myristoylation was evaluated by in vivo metabolic labeling with [3H]myristic acid, efficient protein N-myristoylation was detected equally with Gi1a-, MGA(8)- and MG6S-TNF, as also observed in an in vitro translation system (Fig 2B) In this case, the upper bands detected in the in vitro translation system were not detected in the Western blotting analysis of the expressed proteins in COS-1 cells These results suggest that the incorporation of [3H]myristic acid into the major protein band expressed in the in vitro translation system fully reflect the in vivo protein N-myristoylation that occurs in intact cells To determine the effect of the amino acid residue at position in the N-myristoylation consensus motif on protein N-myristoylation, vertical scanning mutagenesis of the amino acid at position in MG6S-TNF was performed; a series of TNF mutants in which the Ser at position in MG6S-TNF was changed to each of the 19 other amino acids was generated Subsequently, the susceptibility of these mutants to cotranslational protein N-myristoylation was evaluated by using the in vitro translation system The results for the 20 amino acids are arranged according to their radius of gyration All of these mutants, except for a mutant having a Cys residue at position 6, were efficiently expressed as determined by the incorporation of [3H]Leu, as shown in the upper panels of Fig 3A The labeling with [3H]myristic acid revealed that in addition to Ser and Thr, five other amino acids (Gly, Ala, Leu, Ile and Phe) were permitted at position to direct efficient protein N-myristoylation (Fig 3B) In these mutants, a low level Fig Effect of the amino acid residue at position in N-myristoylation consensus motif on the efficiency of cotranslational N-myristoylation reaction The mRNAs encoding MG6X-TNF were translated in vitro in the presence of [3H]leucine or [3H]myristic acid using rabbit reticulocyte lysate Following immunoprecipitation with anti-TNF Ig, the labeled translation products were analyzed by SDS/PAGE and fluorography Results for the 20 amino acids were arranged according to their radius of gyration Three independent experiments showed similar labeling patterns (A) The efficiency of protein N-myristoylation ([3H]myristic acid incorporation/[3H]leucine incorporation) of MG6X-TNF was compared by quantitative analysis of the fluorograms of [3H]myristic acid- and [3H]leucine-labeled proteins shown in the lower and upper panels of (A) Relative N-myristoylation efficiency of each MG6X-TNF was expressed as the percentage of the N-myristoylation efficiency of MG6L-TNF Results for the 20 amino acids were arranged according to their radius of gyration (B) ND, not determined 868 T Utsumi et al (Eur J Biochem 271) Ó FEBS 2004 positions, thereby favoring the susceptibility to protein N-myristoylation Effect of the amino acid residue at position in the N-myristoylation consensus motif on the amino acid requirement at position for cotranslational protein N-myristoylation Fig Amino acid residues at position in naturally occurring N-myristoylation motif The numbers of each amino acid residue located at position in 78 N-myristoylated proteins in which N-myristoylation was experimentally verified listed in a recent report [33] were counted and arranged according to their radius of gyration of [3H]myristic acid incorporation was detected with mutants having Pro, Asn, Gln, Glu, His, Met, Tyr and Trp at this position It is generally accepted that Ser or Thr is preferred at position for protein N-myristoylation In fact, when the number of each amino acid residues located at position in 78 N-myristoylated proteins in which N-myristoylation was experimentally verified listed in a recent report [33] were counted, 74% (58 of 78) of these proteins had a Ser residue and 13% (10 of 78) had a Thr residue at position (Fig 4) The number of N-myristoylated proteins having other amino acids at position accounted for only 13% (10 of 78) of the total N-myristoylated proteins These observations, taken together with the fact that five other amino acids could be permitted at this position in the model substrate protein, suggest that the presence of Ser or Thr at position might affect the amino acid requirements at other In a previous report, we demonstrated that the amino acid at position strongly affected protein N-myristoylation, and the amino acid requirements at this position were significantly affected by the amino acid at position [25] Therefore, we next determined the effect of the amino acid at position on the amino acid requirements at position for protein N-myristoylation by vertical scanning mutagenesis We first determined the effect of the amino acid residue at position in MGA(8)-TNF on protein N-myristoylation A series of TNF mutants (MG3X6A-TNF) in which Ala at position in MGA(8)-TNF was changed to 19 other amino acids were generated and their susceptibility to cotranslational protein N-myristoylation was evaluated in the in vitro translation system The results revealed that the amino acid at position in MGA(8)-TNF strongly affected protein N-myristoylation and only three amino acids (Ala, Asn and Gln) could direct efficient protein N-myristoylation, as shown in Fig 5B In these mutants, a low level of [3H]myristic acid incorporation was detected with mutants having Ser, Cys, Val or Ile at this position Metabolic labeling of the same set of mutants with [3H]acetyl CoA revealed that efficient protein N-acetylation was detected in mutants having Ser, Thr, Asp, Glu or Met at position 3, as shown in Fig 5C These results indicate that the experimental results obtained by metabolic labeling with [3H]myristic acid in the in vitro translation system not reflect a simple enzyme reaction mediated by NMT, but reflect the result of the overall reaction involving a set of cotranslational N-terminal modifications When the Ala at position in MG3X6 A-TNF was changed to Ser to generate MG3X6S-TNF, a dramatic change in the amino acid requirement at position was Fig Effect of the amino acid residue at position on the protein N-myristoylation and N-acetylation of MG3X6A-TNF The mRNAs encoding MG3X6A-TNF were translated in vitro in the presence of [3H]leucine (A), [3H]myristic acid (B) or [3H]acetyl CoA (C) using rabbit reticulocyte lysate Following immunoprecipitation with anti-TNF Ig, the labeled translation products were analyzed by SDS/PAGE and fluorography Results for the 20 amino acids were arranged according to their radius of gyration Ó FEBS 2004 Analysis of the N-myristoylation consensus motif (Eur J Biochem 271) 869 Fig Effect of the amino acid residue at position in N-myristoylation consensus motif on the amino acid requirements at position for cotranslational protein N-myristoylation The mRNAs encoding MG3X6S-, MG3X6T-, and MG3X6F-TNF were translated in vitro in the presence of [3H]leucine or [3H]myristic acid using rabbit reticulocyte lysate Following immunoprecipitation with anti-TNF Ig, the labeled translation products were analyzed by SDS/PAGE and fluorography Results for the 20 amino acids were arranged according to their radius of gyration A, B and C show results with MG3X6S-, MG3X6T-, and MG3X6F-TNF, respectively observed: 11 amino acids (Gly, Ala, Ser, Cys, Thr, Val, Asn, Leu, Ile, Gln, His) were permitted at position to direct efficient protein N-myristoylation, as shown in Fig 6A A low level of [3H]myristic acid incorporation was also detected with mutants having Pro, Asp, Glu and Met at this position To determine whether the remarkable change in the amino acid requirement at position was specific for Ser or not, the Ala at position in MG3X6 A-TNF was changed to other amino acids, and their susceptibility to protein N-myristoylation was evaluated In this case, we chose Thr and Phe to further analyze the effect of the amino acid residue at position on the amino acid requirement at position because of the presence of these amino acids at position in naturally observed N-myristoylated proteins (Fig 4) The results revealed that amino acid requirements at position very similar to those of MG3X6A-TNF were observed with both of these two series of mutants (MG3X6T- and MG3X6F-TNF) as shown in Fig 6B,C In these mutants, a low level of [3H]myristic acid incorporation was detected with several amino acids: Ser, Thr, Val, Ile in MG3X6T-TNF and Thr, Val, Ile in MG3X6F-TNF Thus, it was concluded from these observations that the combination of amino acids at positions and constitutes a major determinant for the susceptibility to protein N-myristoylation When Ser was located at position 6, 11 amino acids (Gly, Ala, Ser, Cys, Thr, Val, Asn, Leu, Ile, Gln, His) were permitted at position to direct protein Nmyristoylation When Ser was not located at position 6, only amino acids (Ala, Asn, Gln) were permitted at position to direct efficient protein N-myristoylation The presence of a Lys residue at position in the N-myristoylation consensus motif affects the amino acid requirement at position and Lys becomes permitted at this position We next determined whether the effect of the amino acid at position on the amino acid requirements at position found in the model substrate proteins were applicable to naturally N-myristoylated proteins or not The numbers of each amino acid residue located at position in 74 naturally N-myristoylated proteins having Ala, Ser, Thr or Phe at position listed in a recent report [33] were counted and are summarized in Fig As shown in the figure, 95 per cent (70 out of 74) of these proteins had amino acid residues at position that were consistent with the amino acid requirements at position for protein N-myristoylation found in this study All of the proteins in which the amino Ó FEBS 2004 870 T Utsumi et al (Eur J Biochem 271) Table N-terminal sequence of N-myristoylated proteins having Lys residue at position Amino acids at positions 3, and are in bold type Protein Ca2+ binding/EF hand proteins Aplycalcin Hippocalcin Neurocalcin Rem-1 Visinin-like protein ADP-ribosylation factor Arf-6 Fig The combination of the amino acid residues at position and in naturally occurring N-myristoylation motif The numbers of each amino acid residue located at position in 74 naturally N-myristoylated proteins having Ala, Ser, Thr or Phe at position listed in recent report [33] were counted and arranged according to their radius of gyration A, B, C and D show results with N-myristoylated proteins having Ala, Ser, Thr and Phe at position 6, respectively Filled bars, amino acid residue consistent with the amino acid requirements at position found in this study; striped bars, amino acid residue inconsistent with the amino acid requirements at position found in this study acid at position is inconsistent with our present results have a Lys residue at this position These observations suggest that an N-myristoylation motif having a Lys residue at position might have other specific structural determinants that permit the Lys residue at position while still directing protein N-myristoylation When the N-terminal sequences of five N-myristoylated proteins having a Lys residue at position listed in a recent review [4] were compared, a striking similarity was observed; the amino acid at position was in all cases Lys (Table 2) It was speculated from these observations that the specific determinant that permits the Lys residue at position might N-terminal sequence (M)GKRASKLKPEEVEEL (M)GKQNSKLRPEMLQDL (M)GKQNSKLRPEVMQDL (M)GKQNSKLRPEVLQDL (M)GKQNSKLRPEVLQDL (M)GKVLSKIFGNKEMRI be the Lys residue at position To test this possibility, the effect of a Lys residue at position on the amino acid requirement at position was evaluated by using several TNF mutants When the Cys residue at position in Gi1a-TNF, which has a natural N-myristoylation motif at the N-terminus, was changed to Lys, N-myristoylation was significantly reduced, as shown in Fig 8A lanes and However, when the Ala residue at position of this mutant (Gi1a-C3K-TNF) was changed to Lys, efficient N-myristoylation was observed with the obtained mutant (Gi1a-C3K-A7K-TNF), as shown in lane In contrast, when the Lys residue at position in hippocalcin-TNF, which has Lys residues at positions and 7, was replaced with Ala, N-myristoylation was completely inhibited, as shown in lanes and These results clearly suggest that the specific determinant that permits the Lys residue at position is the Lys residue at position To further confirm this idea, the effect of the Lys residue at position on the amino acid requirement at position was evaluated by using MG6S-TNF as a model substrate When the Ala residue at position in MG6S-TNF was changed to Lys, N-myristoylation was completely inhibited, as shown in Fig 8B lanes 1, 2, and However, when the Ala residue at position of this mutant (MG3K6S-TNF) was changed to Lys, efficient N-myristoylation was observed with the obtained mutant (MG3K6S-7K-TNF), as shown in lanes and These results strongly support the idea that the specific determinant that permits the Lys residue at position is the Lys residue at position Discussion Protein N-myristoylation is a cotranslational protein modification catalyzed by an enzyme, N-myristoyl transferase (NMT) NMT is a member of the GCN5 acetyltransferase (GNAT) superfamily All family members catalyze the transfer of an acyl group from CoA to a primary amino group NMT can be distinguished from other GNAT family members on the basis of the remarkable diversity of its protein substrates For example, it was reported recently that the Arabidopsis thaliana genome encodes 437 known or putative NMT substrates, accounting for 1.7% of all proteins [17] S cerevisiae Nmt1p is the best studied of the known NMTs The X-ray structure of a binary complex of Nmt1p with bound myristoyl-CoA has been determined [34] A structure of a ternary complex of Nmt1p with a bound Ó FEBS 2004 Analysis of the N-myristoylation consensus motif (Eur J Biochem 271) 871 Fig The presence of a Lys residue at position affects the amino acid requirement at position and allows Lys to occur at this position mRNAs encoding Gi1a-, Gi1a-C3K-, Gi1a-C3K-A7K-, Hippocalcin-, Hippocalcin-K7A-, MG6S-, MG3K6S-, MG3K6S-7K-TNF were translated in vitro in the presence of [3H]leucine or [3H]myristic acid using rabbit reticulocyte lysate Following immunoprecipitation with anti-TNF Ig, the labeled translation products were analyzed by SDS/PAGE and fluorography (A) Results with Gi1a-, Gi1a-C3K-, Gi1a-C3K-A7K-, Hippocalcin-, Hippocalcin-K7A-TNF (B) Results with MG6S-, MG3K6S-, MG3K6S-7K-TNF nonhydrolyzable myristoyl-CoA analogue [S-(2-oxo)pentadecyl-CoA] and an octapeptide substrate has also been defined [34] The Nmt1p fold consists of a saddle-shaped b-sheet flanked by a helices There is pseudo-2-fold symmetry The N-terminal half forms the myristoyl-CoAbinding site The C-terminal half forms the bulk of the peptide-binding site Each half has a fold similar to the core structure of GNAT superfamily members [8] Proteins destined to become N-myristoylated begin with the sequence Met-Gly The initiating Met is removed cotranslationally by methionine aminopeptidase and then myristic acid is linked to Gly-2 via an amide bond by NMT However, not all proteins with an N-terminal glycine are N-myristoylated and the ability to be recognized by NMT depends on the downstream amino acid sequence In addition, proteins with an N-terminal glycine may also be subjected to another cotranslational modification, N-acetylation The precise substrate specificity of S cerevisiae Nmt1p has been characterized mainly by using purified enzyme and synthetic peptides derived from the N-terminal sequences of known N-myristoylated proteins [1,14,15] Some amino acid preferences have been observed at distinct positions downstream of the N-terminal glycine [1,13,16] In general, Ser or Thr is preferred at position 6, and an N-terminal consensus motif such as Met-Gly-X-X-X-Ser/Thr- [13] has been defined In addition to the preference for Ser/Thr residues at position 6, positively charged residues (Lys or Arg) are known to be preferred at position [1,16] Amino acid preference was also observed at position 3: charged residues, aromatic residues and Pro are not allowed at this position [5] These amino acid preferences were confirmed by recent studies on the NMT1p structure as determined by X-ray crystallography [34,35] In these studies, the structure of a ternary complex of Nmt1p with a bound nonhydrolyzable myristoyl-CoA analogue [S-(2-oxo)pentadecyl-CoA] and an Arf2p-derived octapeptide substrate, GLYASKLA, ˚ has been defined to 2.5 A resolution The determined structure allows identification of specific residues within NMT that account for the amino acid preference at positions 3, and of the peptide substrate Ser6 (Ser5 in peptide GLYASKLA), which is greatly preferred in N-myristoylated proteins, is H-bonded to the side chain of His221 in NMT1p, as well as the backbone amides of Asp417 and Gly418 in NMT1p Lys7 (Lys6 in peptide GLYASKLA), also preferred in N-myristoylated proteins, is H-bonded to the side chains of Asp417 and Gly418 in NMT As for Leu3 (Leu2 in peptide GLYASKLA), it was shown that contacts between the side chain of Leu3 and pantetheine of myristoyl-CoA complete formation of the 872 T Utsumi et al (Eur J Biochem 271) peptide-binding site and at the same time generate a 90° bend in the peptide backbone, turning it away from myristoyl-CoA and toward a peptide-binding groove Thus, the amino acid at position is important for positioning the substrate peptide in the peptide-binding site The residues described above in NMT1p that interact with GLYASKLA are highly conserved in NMTs derived from other species These results indicate that in addition to the Gly at position 2, the amino acids at positions 3, 6, and in the substrate protein play important roles in substrate recognition by NMT These findings were further confirmed by an Alascanning mutagenesis study designed to define the extent to which residues at positions 2, 3, 5, and of GLYASKLA contribute to proper placement of the N-terminal Gly in the active site [35] In these experiments, a panel of GLYASKLA derivatives with single Ala substitutions at these positions was produced and presteady-state kinetic analysis was performed The results revealed that Ala substitution for Leu2, Ser5, or Lys6 produced a 12–18-fold reduction in the burst rate Based on these results, it was postulated that differences in the efficiency of N-myristoylation of various cellular proteins may arise in part because of differences in the presentation of Gly2 dictated by interactions among the residues at positions 3, 6, and of the substrate and elements in the enzyme’s peptide-binding site Thus, it is well established that in addition to the Gly at position 2, amino acids at positions 3, 6, and play important roles in substrate recognition by NMT However, the relative role of these residues in substrate recognition, the relationship between amino acids that reside in these three distinct positions, or favorable amino acid combinations in these positions are not yet well characterized In a previous report, we showed that metabolic labeling in an in vitro translation system is an effective strategy to characterize the N-terminal sequence requirements for cotranslational protein N-myristoylation Using this assay system, we demonstrated that the amino acid residue at position strongly affects protein N-myristoylation, and the amino acid requirements at this position were significantly affected by the amino acid at position [25] These results suggest that the combination of amino acids at positions and might be a critical determinant for protein Nmyristoylation In the present study, to examine the effect of the combination of amino acids at positions and on protein N-myristoylation, sequential vertical-scanning mutagenesis of the amino acids at positions and in a model substrate protein having a sequence MGAAAAAAAA at its N-terminus was performed and the susceptibility of these mutants to protein N-myristoylation was evaluated by metabolic labeling in an in vitro translation system using rabbit reticulocyte lysate The results revealed that the combination of amino acids at positions and strongly affected the susceptibility of the protein to protein N-myristoylation When Ser was located at position 6, 11 amino acids (Gly, Ala, Ser, Cys, Thr, Val, Asn, Leu, Ile, Gln, His) were permitted at position to direct efficient protein N-myristoylation In contrast, when Ala, Thr or Phe was located at position 6, only amino acids (Ala, Asn, Gln) were permitted at position to direct efficient modification These results clearly indicate that the amino acid residues permitted at position are affected by the Ó FEBS 2004 amino acid residue reside at position The fact that the increase in the number of permitted amino acid residues at position in response to varying the amino acid at position was specific for the Ser residue well explains the fact that Ser-6 is frequently observed in the naturally observed N-myristoylated proteins The mechanism by which the Ser6 of the substrate affects the amino acid residue permitted at position is not clear It is possible to speculate that this phenomenon is mediated by the specific interaction between Ser6 of the substrate and elements in the enzyme’s peptide-binding site This specific interaction probably induces the changes in the structure of the peptide-binding site that cause the alterations in the permitted amino acid at position Unfortunately, the only reported structural information about interactions between an NMT and its peptide substrates comes from the ternary structure with bound Arf2p-derived GLYASKLA, which has a Ser residue at position Therefore, if we could obtain the X-ray structure of NMT bound to a peptide which does not have a Ser residue at position 6, it might be possible to elucidate the mechanism by which the Ser at position significantly affects the amino acid residue permitted at position In addition to the combination of amino acids at positions and 6, it was demonstrated that the combination of amino acids at positions and also affects the susceptibility of the protein to protein N-myristoylation In this case, the presence of Lys at position was found to affect the amino acid requirement at position 3, and allowed Lys to occur at this position This finding demonstrates again that the specific interaction between an amino acid at a distinct position in the substrate protein and elements in the enzyme’s peptide-binding site will induce changes in the structure of the peptide-binding site that cause an alteration in the permitted amino acid at an other position In the present study, we focused our attention on the effect of amino acid residue at positions and of substrate protein on the amino acid requirement at position for protein Nmyristoylation, and did not study the effect of amino acids at other positions Recent studies revealed that, for complete substrate protein, at least the N-terminal 17 residues of the substrate protein experience amino acid type variability restrictions for protein N-myristoylation [33] Therefore, it might be possible that the amino acids located beyond position would also affect the amino acid requirement at position Further studies are required to fully characterize the favorable combinations of amino acids at distinct positions in the substrate protein It is very important to determine whether the restrictions on the amino acid combinations at positions 3, and in the substrate protein for protein N-myristoylation found in this study are applicable to NMTs derived from different species As described previously, the residues in NMT1p that interact with the amino acids at positions 2, and in the Arf2p-derived peptide GLYASKLA are highly conserved in NMTs derived from other species Therefore, favorable combinations of amino acids at positions 3, and in substrate protein for protein N-myristoylation might be similar in many of the NMTs in different species In fact, the amino acid requirements found in this study were fully consistent with the N-terminal sequence of 78 N-myristoylated proteins derived from various species in which N-myristoylation was experimentally verified Ó FEBS 2004 Analysis of the N-myristoylation consensus motif (Eur J Biochem 271) 873 It is well known that differences in the amino acid requirements for protein N-myristoylation are observed in different species The difference in the substrate specificity of NMT in different species has been attributed mainly to the difference in the permitted amino acid residues at positions 3, 6, and For example, for the amino acid at position 6, amino acid residues Leu, Ile, and Phe are known not to be permitted for S cerevisiae NMT1p [5], but were found to be permitted for rabbit NMT (Fig 3) For the amino acid at position 3, His is not permitted for S cerevisiae NMT1p [5], but was permitted for rabbit NMT (Fig 5A) Therefore, it seems likely that the restrictions on the amino acid combinations at positions 3, and are partly different for each NMT derived from different species In order to clarify the difference in the favorable amino acid combinations at positions 3, and in different species, experiments similar to those in the present study should be performed on each of the NMT species Thus, the present study revealed that in addition to the permitted amino acids at positions 3, 6, and in the N-terminal sequence, the amino acid combinations in these positions were major determinants for the susceptibility of the protein to protein N-myristoylation For postgenomic studies, reliable tools for the prediction of co- and posttranslational modifications would be valuable for functional assignments of functionally unknown proteins In fact, a sophisticated program for automated prediction of protein N-myristoylation from the substrate protein sequence has been developed recently and is available via public access web-server [36] However, as for the amino acid requirements at distinct positions in the N-terminal sequence, only the permitted amino acids in each position have been taken into consideration in the prediction program This might lead to the failure in the accurate prediction of the N-myristoylated proteins In fact, we have recently revealed that the C-terminal 15 kDa fragment of cytoskeletal actin is post-translationally N-myristoylated upon caspase-mediated cleavage during apoptosis [7] In this case, however, the N-terminal 17 amino acid sequence of this fragment was not predicted to be N-myristoylated by the prediction program The amino acids at positions and of this fragment are Gln and Thr, respectively, and are in agreement with the favorable amino acid combinations at these positions found in this study These observations indicated that the reliability of the prediction program might be significantly improved by the consideration of the amino acid combinations at positions 3, and in the N-terminal sequence Acknowledgements Part of this work was supported by a Grant-in-Aid for Scientific Research (No 12660080, no 15580080) from the Ministry of Education, Science and Culture of Japan References Towler, D.A., Gordon, J.I., Adams, S.P & Glaser, L (1988) The biology and enzymology of eukaryotic protein acylation Annu Rev Biochem 57, 69–99 Spiegel, A.M., Backlund, P.S., Butrynski, J.E., Jones, T.L.Z & Simonds, W.F (1991) The G protein connection: molecular basis of membrane association Trends Biochem Sci 16, 338–341 Boutin, J.A (1997) Myristoylation Cell Signal 9, 15–35 Resh, M.D (1999) Fatty acylation of proteins: new insights into membrane targeting of myristoylated and palmitoylated proteins Biochem Biophys Acta 1451, 1–16 Farazi, T.A., Waksman, G & Gordon, J.I (2001) The biology and enzymology of protein N-myristoylation J Biol Chem 276, 39501–39504 Zha, J., Weiler, S., Oh, K.-J., Wei, M.C & Korsmeyer, S.J (2000) Posttranslational N-myristoylation of BID as a molecular switch for targeting mitochondria and apoptosis Science 290, 1761–1765 Utsumi, T., Sakurai, N., Nakano, K & Ishisaka, R (2003) C-terminal 15 kDa fragment of cytoskeletal actin is posttranslationally N-myristoylated upon caspase-mediated cleavage and targeted to mitochondria FEBS Lett 539, 37–44 Dyda, F., Klein, D.C & Hickman, A.B (2000) GCN5-related N-acetyltransferases: a structural overview Annu Rev Biophys Biomol Struct 29, 81–103 Towler, D.A., Adams, S.P., Eubanks, S.R., Towery, D.S., Jackson-Machelski, E., Glaser, L & Gordon, J.I (1987) Purification and characterization of yeast myristoyl CoA: protein N-myristoyltransferase Proc Natl Acad Sci USA 84, 2708–2712 10 Duronio, R.J., Reed, S.I & Gordon, J.I (1992) Mutations of human myristoyl-CoA: protein N-myristoyltransferase cause temperature-sensitive myristic acid auxotrophy in Saccharomyces cerevisiae Proc Natl Acad Sci USA 89, 4129–4133 11 Lodge, J.K., Johnson, R.L., Weinberg, R.A & Gordon, J.I (1994) Comparison of myristoyl-CoA: protein N-myristoyltransferases from three pathogenic fungi: Cryptococcus neoformans, Histoplasma capsulatum, and Candida albicans J Biol Chem 269, 2996–3009 12 Giang, D.K & Cravatt, B.F (1998) A second mammalian N-myristoyltransferase J Biol Chem 273, 6595–6598 13 Johnson, D.R., Bhatnagar, R.S., Knoll, L.J & Gordon, J.I (1994) Genetic and biochemical studies of protein N-myristoylation Annu Rev Biochem 63, 869–914 14 Towler, D.A., Adams, S.P., Eubanks, S.R., Towery, D.S., Jackson-Machelski, E., Glaser, L & Gordon, J.I (1988) Myristoyl CoA: protein N-myristoyltransferase activities from rat liver and yeast possess overlapping yet distinct peptide substrate specificities J Biol Chem 263, 1784–1790 15 Rocque, W.J., McWherter, C.A., Wood, D.C & Gordon, J.I (1993) A comparative analysis of the kinetic mechanism and peptide substrate specificity of human and Saccharomyces cerevisiae myristoyl-CoA: protein N-myristoyltransferase J Biol Chem 268, 9964–9971 16 McWherter, C.A., Rocque, W.J., Zupec, M.E., Freeman, S.K., Brown, D.L., Devadas, B., Getman, D.P., Sikorski, J.A & Gordon, J.I (1997) Scanning alanine mutagenesis and depeptidization of a Candida albicans myristoyl-CoA: protein N-myristoyltransferase octapeptide substrate reveals three elements critical for molecular recognition J Biol Chem 272, 11874–11880 17 Boisson, B., Giglione, C & Meinnel, T (2003) Unexpected protein families including cell defense components feature in the N-myristoylome of a higher eukaryote J Biol Chem 278, 43418–43429 18 Palmiter, R.D., Gagnon, J & Walsh, K.A (1978) Ovalbumin: a secreted protein without a transient hydrophobic leader sequence Proc Natl Acad Sci USA 75, 94–98 19 Dailey, T.A & Dailey, H.A (1996) Human protoporphyrinogen oxidase: expression, purification, and characterization of the cloned enzyme Protein Sci 5, 98–105 20 Mayer, A., Siegel, N.R., Schwartz, A.L & Ciechanover, A (1989) Degradation of proteins with acetylated amino termini by the ubiquitin system Science 244, 1480–1483 21 Kimura, Y., Takaoka, M., Tanaka, S., Sassa, H., Tanaka, K., Polevoda, B., Sherman, F & Hirano, H (2000) N (alpha)-acety- Ó FEBS 2004 874 T Utsumi et al (Eur J Biochem 271) 22 23 24 25 26 27 28 29 lation and proteolytic activity of the yeast 20 S proteasome J Biol Chem 275, 4635–4639 Lee, F.J., Lin, L.W & Smith, J.A (1990) Model peptides reveal specificity of N alpha-acetyltransferase from Saccharomyces cerevisiae J Biol Chem 265, 11576–11580 Bradshaw, R., Brickey, W & Walker, K (1998) N-terminal processing: the methionine aminopeptidase and N alpha-acetyltransferase families Trends Biochem Sci 23, 263–267 Polevoda, B & Sherman, F (2003) N-Terminal acetyltransferases and sequence requirements for N-terminal acetylation J Mol Biol 325, 595–622 Utsumi, T., Sato, M., Nakano, K., Takemura, D., Iwata, H & Ishisaka, R (2001) Amino acid residue penultimate to the aminoterminal Gly residue strongly affects two cotranslational protein modifications, N-myristoylation and N-acetylation J Biol Chem 276, 10505–10513 Umigai, N., Sato, Y., Mizutani, A., Utsumi, T., Sakaguchi, M & Uozumi, N (2003) Topogenesis of two transmembrane type K+ channels, Kir 2.1 and KcsA J Biol Chem 278, 40373–40384 Deichaite, I., Casson, L.P., Ling, H.P & Resh, M.D (1988) In vitro synthesis of pp60v-src: myristoylation in a cell-free system Mol Cell Biol 8, 4295–4301 Utsumi, T., Akimaru, K., Kawabata, Z., Levitan, A., Tokunaga, T., Tang, P., Ide, A., Hung, M.-C & Klostergaad, J (1995) Human pro-tumor necrosis factor: molecular determinants of membrane translocation, sorting, and maturation Mol Cell Biol 15, 6389–6405 Utsumi, T., Kuranami, J., Tou, E., Ide, A., Akimaru, K., Hung, M.C & Klostergaard, J (1996) In vitro synthesis of an 30 31 32 33 34 35 36 N-myristoylated fusion protein that binds to the liposomal surface Arch Biochem Biophys 326, 179–184 Sanger, F., Nicklen, S & Coulson, A.R (1977) DNA sequencing with chain-terminating inhibitors Proc Natl Acad Sci USA 74, 5463–5467 Utsumi, T., Levitan, A., Hung, M.-C & Klostergaad, J (1993) Effects of truncation of human pro-tumor necrosis factor transmembrane domain on cellular targeting J Biol Chem 268, 9511–9516 Utsumi, T., Tou, E., Takemura, D., Ishisaka, R., Yabuki, M & Iwata, H (1998) Met-Gly-Cys motif from G-protein alpha subunit cannot direct palmitoylation when fused to heterologous protein Arch Biochem Biophys 349, 216–224 Maurer-Stroh, S., Eisenhaber, B & Eisenhaber, F (2002) N-terminal N-myristoylation of proteins: refinement of the sequence motif and its taxon-specific differences J Mol Biol 317, 523–540 Farazi, T.A., Waksman, G & Gordon, J.I (2001) Structures of Saccharomyces cerevisiae N-myristoyltransferase with bound myristoylCoA and peptide provide insights about substrate recognition and catalysis Biochemistry 40, 6335–6343 Farazi, T.A., Manchester, J.K., Waksman, G & Gordon, J.I (2001) Pre-steady-state kinetic studies of Saccharomyces cerevisiae myristoyl-CoA: protein N-myristoyltransferase mutants identify residues involved in catalysis Biochemistry 40, 9177–9186 Maurer-Stroh, S., Eisenhaber, B & Eisenhaber, F (2002) N-terminal N-myristoylation of proteins: prediction of substrate proteins from amino acid sequence J Mol Biol 317, 541–557  ... GCCGGGATCCATGGGCAAAACGCTGAGCAAAGAGGACAAGCTCGAG GCCGGGATCCATGGGCAAGCAGAATAGCGCACTGCGGCCAGACAAG GCCGGGATCCATGGGCAAGGCAGCATCTGCAGCAGCAGCAGACAAGCCTGTAGCC GCCGGGATCCATGGGCAAGGCAGCATCTAAGGCAGCAGCAGACAAGCCTGTAGCC AATTAACCCTCACTAAAGGG... ATATGGATCCATGGCTGCGGCAGCAGCGGCAGCAGCAGCAGACAAGCCTGTAGCC ATATGGATCCATGGGCGCGGCAGCAGCGGCAGCAGCAGCAGACAAGCCTGTAGCC GCCGGGATCCATGGGCGCAGCAGCATCTGCAGCAGCAGCAGACAAGCCTGTAGCC GCCGGGATCCATGGGCGCAGCAGCANNKGCAGCAGCAGCAGAC... effect of the combination of amino acids at positions and on protein N-myristoylation, sequential vertical-scanning mutagenesis of the amino acids at positions and in a model N-terminal sequence was