The amino acid sequence R/K-R/K-N-S/T is the consensus PKA recognition site, and either S or T is the phosphorylation site. A putative PKA phosphorylation site containing threonine 341 (T341) is present in the S. cerevisiae eRF3 (Sup35p) sequence and is conserved among others yeasts and fungi. In drosophila, human and mouse the site is degenerated, but T341 still belongs to a phosphorylation site, recognized by the CK2 protein kinase (Fig. 1). In a sequence-based approach, the evolutionary conservation of all PKA consensus sites in the S. cerevisiae proteome was systematically assessed within a group of related yeasts 23. The study demonstrated that the likelihood of phosphorylation correlated well with the degree of conservation of the PKA consensus sites. The evolutionary preservation of the eRF3 phosphorylation site thus suggests that this site probably plays an important role in vivo. To test the functionality of the PKA phosphorylation site of Sup35p, we generated threonine to alanine (A) or aspartate (D) mutants. Alanine cannot be phosphorylated, whereas aspartate mimics the presence of phosphate groups.

His-Tag fusion polypeptides, with either the entire full-length wild-type Sup35p or the A and D mutants, were generated in E. coli and purified by Ni²⁺-nitrilotriacetic acid affinity chromatography. The purified proteins were incubated with the PKA catalytic subunit and [γ-³²P]ATP. Incubation of wild-type Sup35p resulted in the appearance of a ³²P-labeled species of 78 kDa (the expected apparent molecular mass of Sup35His) (Fig. 2a). Phosphorylation was completely inhibited by PKI, a specific inhibitor of PKA. Incubation of either mutant A or D with PKA and [γ-³²P]ATP resulted in a reduced phosphorylation. (Fig. 2b). Therefore, the T341 residue of Sup35p is phosphorylated in vitro by PKA, and the consensus phosphorylation site of Sup35p is not the only one to be phosphorylated by PKA in vitro. This is not surprising as kinases are known to also have non specific activity in in vitro conditions.

To verify the phosphorylation state of the T341 residue, in vitro phosphorylated wild-type, A and D proteins were submitted to western blot analysis using an antibody specific of substrates phosphorylated by PKA. As expected, only the phosphorylated wild-type protein was detected, the phosphorylated A and D proteins were not, and none of the non-phosphorylated proteins was revealed by the antibody (Fig. 2c).

The expression of full-length Sup35p was relatively poor in E. coli cells, while a much greater amount of protein could be recovered from constructs expressing the C-terminal domain alone. Furthermore, the purified full-length proteins were unstable and precipitated rapidly upon storage, probably due to the presence of the aggregating N-terminal domain. Consequently, the same experiments were performed with the C-terminal part of the wild-type, A or D mutant proteins purified as His-Tag fusion polypeptides from E. coli. We wondered if the Sup35pcter [C-terminus] alone could be phosphorylated by PKA in vitro. The results show that the truncated proteins behaved as did the corresponding full-length proteins. The C-terminal part of Sup35p was phosphorylated by PKA in vitro, and the A and D mutants were also phosphorylated but to a much lower extent (Fig. 2d).

To search for a difference in electrophoretic mobility between the in vitro phosphorylated and non-phosphorylated protein forms, we performed SDS-polyacrylamide gels in various conditions (see Materials and Methods), but no change in the electrophoretic migration was seen between the protein forms (not illustrated).

To check which residue was phosphorylated by PKA in vitro, the phosphorylated full-length His-Tag fusion Sup35p was subjected to mass spectrometry analysis. The MALDI mass spectrum of the unfractioned tryptic digest allowed us to obtain 70% sequence coverage (Fig. 3a). A peptide whose mass did not correspond to the mass of any predicted tryptic peptide was detected, but it was about 80 Da above the theoretical mass of the peptide [339–349]. The possible presence of one phosphate group was fully relevant, as long as the sequence of this tryptic fragment is RYTILDAPGHK, and contains 2 potential phosphorylation sites (Y340 and T341). This was confirmed by performing an on-target dephosphorylation experiment (Fig. 3a). As the signal of the alkaline phosphatase-treated sample was diminished by 80 Da compared to the untreated peptide, [339–349] was indeed a mono-phosphorylated species.

To further investigate the phosphorylation site of the peptide [339–349], we purified the tryptic digest by reversed phase HPLC separation. Each fraction collected was analysed by MALDI mass spectrometry prior to and after on-target dephosphorylation, and [339–349] was recovered in the presence of two other minor peptides (data not shown). Peptide [339–349] of the HPLC fraction was sequenced by PSD-MALDI TOF mass spectrometry. The PSD spectrum of the derivatised peptide [339–349] (m/z 1528.6565) is shown Fig. 3b. The β-elimination occurring (H₃PO₄ loss) suggested phosphorylation on the T341 residue rather than on Y340, because such β-elimination is observed for phosphorylated serine or threonine, but not for tyrosine. This was confirmed by the presence of a mass difference between two adjacent peaks corresponding to a dehydroamino-2-butyric acid product (i.e. 83 Da).

To determine whether Sup35p is phosphorylated in yeast cells, we labelled with [³²P]orthophosphate a culture of the SUP35-deleted strain that contains a plasmid-encoded SUP35 wild-type gene fused to a C-terminal HA-Tag, and immunoprecipitated Sup35p with an HA-specific antibody. Western blot analyses of the immunoprecipitated proteins and the immunoprecipitation supernatants showed that approximately one third of the Sup35 proteins was immunoprecipitated. The same experiments were performed using a strain deleted of PKA, and also in a SUP35-deleted strain expressing a plasmid coding for the A mutant protein fused to the HA-Tag and unable to be phosphorylated on T341. Labelling and immunoprecipitation were carried out on exponentially growing cells, as well as on stationary phase cells. In spite of numerous attempts, we did not observe a radioactive pool of Sup35p (data not shown). Since the FS1 strain carries a [PSI+] determinant, it seemed possible that the aggregation of Sup35p could alter protein phosphorylation, or decrease immunoprecipitation efficiency and as a result could reduce the immunoprecipitated sub-fraction of phosphorylated Sup35p. To test this possibility, similar experiments were performed in a SUP35-deleted strain carrying only the C-terminal domain of Sup35p, fused to a C-terminal HA-Tag and again, no radioactive pool of Sup35p was detected (data not shown).

Complementation assays of a SUP35 gene deletion were performed with the mutated alleles. A strain deleted for the endogenous SUP35 gene and saved by an URA3 plasmid-encoded wild-type SUP35 gene was transformed with a HIS3 vector carrying either the wild-type SUP35 gene, or one of the mutants, or an empty vector. Plasmid shuffling 24 was then carried out by plating the transformants onto medium supplemented with 5-fluoroorotic acid, an uracil analog that allows the formation of colonies only from cells that have lost the wild-type SUP35 plasmid with the URA3 marker. The ability to form colonies indicates that the Sup35p mutant alone can support viability, while the absence of colonies indicates that the mutant form of Sup35p is unable to support cell viability. The wild-type and the A mutant were able to grow but not the D mutant, suggesting that the T341 residue is subjected to functional constraints and that the introduction of a negative charge in this position alters cell viability. The A mutant behaved like the wild-type strain and did not show thermosensitivity at 37°C (data not shown). A glutamate mutant (T341E) was also constructed to mimic the presence of phosphate groups, and presented the same phenotypes as the D mutant. The essential function of Sup35p is located in the C-terminal domain, thus C-terminal-T341A and C-terminal-T341D mutants were also constructed and used in such a complementation assay as above. Surprisingly, both mutants were able to complement the loss of wild-type Sup35p (Fig. 4), indicating that the N-terminal part of Sup35p might be implicated in the lethality observed in presence of the full-length Sup35p D mutant.

To check if the lethality of the D mutant was due to the degradation of the protein, western blot analyses were carried out to monitor the steady-state levels of the mutant forms of Sup35p. Protein extracts were obtained from SUP35-deleted strains carrying either one of the mutated full-length alleles on a centromeric HIS3 plasmid (saved by the plasmid-encoded C-terminal part of the SUP35 wild-type gene). To permit valid comparisons, the amounts of total protein extract loaded per well were the same. The results indicate that each mutant form of Sup35p was present at a steady-state level similar to that of the wild-type Sup35p (Fig. 5). Thus the incapacity of the full-length D mutant to grow does not result from Sup35p degradation.

Mutations in the gene encoding eRF3 have frequently been reported to exhibit an omnipotent suppressor phenotype, where readthrough is thought to increase at all stop codons 1625. We examined how mutation of T341 affected the efficiency of translation termination at each of the three stop codons by using a dual reporter system 26. The dual reporter plasmids contain an upstream lacZ gene and a downstream firefly luciferase gene separated by different in-frame stop codons in the Tobacco Mosaic Virus readthrough context. These constructs made it possible to monitor the translation readthrough of different termination signals in different genetic contexts by measuring firefly luciferase activity, while the β-galactosidase activity served as an internal normalization control.

The SUP35-deleted strain saved by either a plasmid-encoded C-terminal wild-type, C-terminal A mutant or C-terminal D mutant gene was transformed with the dual reporter constructs, readthrough activities were measured and statistical analyses using the Mann-Whitney non parametric test were performed 27. Yeast strains expressing the C-terminal A or D Sup35p as the sole source of Sup35 protein exhibited a significantly increased readthrough level compared to cells expressing wild-type Sup35p (Fig. 6). The readthrough efficiency of the A mutant was 1.4 to 1.8 fold higher than the wild-type (P value TAG = 0.00333, TGA = 0.00177, TAA = 0.00177). The D mutant showed a stronger increase in readthrough with increase factors ranging from 3.4 for the TAG codon (P value = 0.00214) and 3.6 for the TAA codon (P value = 0.00044), to 4.2 (P value = 0.00044) for the TGA codon compared to the wild-type C-terminal Sup35p.

The Sup45p- and ribosome-dependent GTPase activity of the A and D mutant Sup35pcter was checked in vitro by accumulation of [³²P]P_i using the modified charcoal precipitation assay as described 7. The D mutant was practically inactive in the GTPase assay whereas the A mutant exhibited residual GTPase activity (6% compared to the wild-type Sup35pcter) (Table 1).

Translation termination is triggered by a complex consisting of Sup45p and Sup35p 9. We analyzed the capacity of the A and D mutant proteins to interact with the Sup45p partner by a two-hydrid approach. Interactions were visualized by growth on a medium lacking histidine with 10 mM 3-aminotriazol added. The two mutant Sup35 proteins could still interact with Sup45p, but with various efficiencies. The results indicated strong interaction between the wild-type proteins, decreased interaction between Sup45p and the Sup35p A mutant, and the weakest between Sup45p and the Sup35p D mutant (Fig. 7). The frequency of diploid cells interacting with each others was estimated for these three diploids. Three independent cultures were prepared in minimal medium lacking uracile and leucine for each diploid, and plated onto either a minimal medium lacking histidine, uracile and leucine to count the cells presenting a Sup35p and Sup45p interaction, or a minimal medium lacking uracile and leucine to count diploid cells. The results indicated an interaction in 64% of the cells between the wild-type proteins, in 22% of the cells between the Sup35p A mutant and Sup45p, and in 15% of the cells between the Sup35p D mutant and Sup45p.

A similar picture of various interaction efficiencies of the A and D mutant Sup35pcter with Sup45p was also observed in another in vitro assay. It was shown that Sup35p can stimulate the RF activity of Sup45p in an in vitro system at low stop codon concentration while Sup45p alone at the same concentration of stop codon is inactive 828. The stimulation of Sup45p activity in the presence of Sup35p might be caused by mutual interaction of these proteins. The RF stimulating activity of Sup35pcter towards Sup45p was measured as described 8. The stimulating RF activities of the A and D mutant Sup35pcter towards Sup45p were 60 and 35%, respectively compared to the wild-type Sup35pcter activity (not shown).

The S. cerevisiae strains used here were FS1 (MAT α ade2-592 (frameshift) lys2Δ201 leu2-3, 112 his3Δ-200 ura3-52) which is derived from Y349 38, and pJ69-4A for the two-hybrid analysis (MAT a and α trp1-901 leu2-3, 112 ura3-52 his3Δ-200 gal4Δ gal80Δ GAL2-ADE2 LYS2::GAL1-HIS3 met2::GAL7-lacZ) (described in 39). Standard rich (YPD) or synthetic (SC) culture medium was prepared as described 40. The E. coli strain DH5α (supE44 Δlac U169 (Φ80 lacZΔM15) hsdR17 recA1 endA1 gyrA96 thi-1 relA1) was used for cloning experiments, and was cultivated in LB medium 41. Appropriate amounts of antibiotics, amino acids and bases were added when necessary. Yeast transformation was performed by the lithium acetate method as described previously 42. E. coli cells were transformed as indicated 43. For plasmid shuffling 44, a selective medium containing 0,25% 5-fluoroorotic acid (5-FOA, Fermentas) was used.

The FS1 sup35Δ strain was generated by a one-step gene replacement method. The SUP35 gene was disrupted by the removal of the entire open reading frame and the insertion of the ADE2 gene. Y349 genomic DNA was used as template to generate the PCR fragment carrying the ADE2 gene plus the promoter with the sense (5'-TTCTTCATCGACTTGCTCGGAATAACATCTATATCTGCCCACTAGCAACAAACACC AACATAACACTGACATC) and reverse (5'-CTTTATGATCGGTATTATTGTGTTTGCATTTACTTATGTTTGCAAGAAATGGACAC CTGTAAGCGTTGATTTC) primers, which contain extensions corresponding to the flanking upstream and downstream sequences of the SUP35 open reading frame, respectively. The upstream (sense 5'-CGAGAAGATATCCATCATATTACC and reverse 5'-TGTTGCTAGTGGGCAGATATAG primers) and downstream (sense 5'-ATTTCTTGCAAACATAAGTAAATG and reverse 5'-GCTTCTGAAGATAGATGAGCGG primers) regions of the SUP35 gene were amplified by PCR from FS1 genomic DNA. The three PCR fragments were then joined by running a PCR reaction with each of them in equimolar amounts, resulting in the final disruption fragment. The yeast FS1 strain was transformed with the disruption fragment and the centromeric URA3 plasmid pRS316-SUP35. Ura⁺ Ade⁺ transformants (white colonies) were tested for the ability to evict the URA3 plasmid on 5-FOA plates. Clones which were unable to grow on 5-FOA medium had integrated the ADE2 gene on the chromosome, and not on the pRS316-SUP35 plasmid. The correct genomic integration event was then verified by PCR on the genomic DNA.

Centromere- and 2μ DNA-based plasmids were used to express wild-type and mutant forms of SUP35 in yeast. The centromeric URA3 plasmid pRS316-SUP35 was kindly provided by Dr. M.D. Ter-Avanesyan (Institute of Experimental Cardiology, Moscow). The pRS316-SUP35A, pRS316-SUP35D, and pRS316-SUP35E constructs with the mutant T341A (ACC to GCC change using the sense 5'-AAAAAGGCGTTAT

GCC

GGC

To express the entire wild-type, A and D Sup35p HA proteins in yeast, the corresponding open reading frames (without the stop codon) were amplified using the direct (5'-CACTA

GAATTC

GGATCC

TGGCCA

TCTAGA

pET-SUP35 for the expression of His₆-tagged Sup35p in E. coli was produced by inserting a PCR fragment amplified from pRS316-SUP35 with the direct (5'-AGATATA

CATATG

GGATCC

The plasmid for expression of Sup45p in E. coli was created on the basis pET30a and kindly provided by V.N. Urakov (Institute of Experimental Cardiology, Moscow).

The 2μ-based pGAD-C3 (LEU2 marker) and pGBDU-C3 (URA3 marker) vectors 39 were used to generate plasmids for the two-hybrid analysis. The pET-SUP35 vector was digested by NdeI, then Klenow-filled, and digested by BamH1. The resulting fragment carrying the entire SUP35 open reading frame was cloned into the SmaI-BamH1 sites of the pGAD-C3 and pGBDU-C3 plasmids. The Gal4 activating and binding domain fusions with the A and D Sup35 mutants were obtained by replacing the wild-type NcoI-StuI fragments by the mutated fragments digested from pRS313-SUP35A and pRS313-SUP35D. For SUP45, a PCR fragment amplified from FS1 genomic DNA with the direct (5'-CC

ATCGAT

CTGCAG

Sup35 and Sup35cter proteins were purified as follows. Expression was performed at 37°C using the transformed E. coli Rosetta strains (Novagen) and LB medium supplemented with 17 μg/ml chloramphenicol and 50 μg/ml kanamycin. When the cell culture reached an OD_{600 nm} of 0.5, protein expression was induced with 0.5 mM IPTG (Sigma) and the cells were grown for a further 3 h. The cells were harvested by centrifugation and stored at -20°C. The cell pellet was resuspended in 40 ml of 20 mM Tris-HCl pH 7.5, 150 mM NaCl, 5 mM β-mercaptoethanol, and protease inhibitors. Cell lysis was completed by sonication. The His-tagged protein was purified on an Ni-NTA column (Qiagen Inc.), eluted with imidazole, and loaded onto a SuperdexTM200 column (Amersham Pharmacia Biotech) equilibrated with 20 mM Tris-HCl pH 7.5, 150 mM NaCl, and 10 mM β-mercaptoethanol. The homogeneity of the proteins was checked by SDS-PAGE.

The Sup45p with 6 His-residues on the N terminus was expressed in E. coli and purified on Ni-NTA resin (Qiagen).

Aliquots (2 μg) of purified N-terminally His₆-tagged Sup35 and Sup35cter proteins were added to a PKA reaction buffer containing 25 mM Tris pH 7.5, 10 mM MgCl₂, and 2 μCi of [γ-³²P]ATP (3000 Ci/mmol), and incubated with 2.5 U of the PKA catalytic subunit (Sigma) at 30°C for 30 min in a total volume of 40 μl. To test for the specificity of the reaction for PKA, 30 μg of PKA inhibitor (PKI) (Sigma) were added to the reaction buffer to specifically inhibit the activity of PKA. For mass spectrometry analyses, Sup35p (100 μg) was phosphorylated by 75 U of PKA with 750 μM unlabelled ATP in 1.2 ml. Phosphorylated proteins were then TCA precipitated and loaded onto 10% SDS-polyacrylamide gels, which were then autoradiographed. For mass spectrometry analyses, the gels were stained with Coomassie Blue, and the bands of interest were excised for further analysis.

To investigate differences in electrophoretic mobility between the in vitro phosphorylated and non-phosphorylated protein forms, samples were loaded onto various SDS-polyacrylamide gels and revealed by Coomassie staining: 4–12% gradient polyacrylamide gel, or 10% acrylamide/0.1% bis-acrylamide gel, or 10% ProSieve 50 gel, which is a modified acrylamide formulation developed for high performance protein gel electrophoresis (TEBU).

Protein samples were separated on SDS-polyacrylamide gels and electrophoretically transferred to nitrocellulose membranes 46. Western blots were probed with either affinity purified polyclonal rabbit antibodies against the peptide 137–151 of Sup35p (a kind gift from Dr. S. Lindquist), or a rabbit monoclonal antibody against phospho-PKA substrates (detecting R-R-X-S*/T* motifs, Cell Signaling). The bound antibodies were detected using the Amersham ECL system.

Yeast cells were grown in a low-phosphate YPD medium (LP-YPD Broth, BIO 101 Systems) to the appropriate OD₆₀₀ (1.5 for the exponential phase, and 6 for the stationary phase). Cells in 1 ml growth medium, harvested from 4 ml for the exponential and 1 ml for the stationary phases were labeled by adding 250 μCi of [³²P] orthophosphate (Amersham) for 1 h at 30°C. The cells were lysed by incubation for 10 min on ice with 100 μl of 1.85 M NaOH plus 1% β-mercaptoethanol. Proteins were then precipitated by adding 100 μl of 50% TCA. The precipitate was centrifuged at 15,000 g for 5 min, washed with 1 M Tris base, resuspended in 50 μl of SDS-PAGE loading buffer without β-mercaptoethanol 47, and dissociated by heating for 10 min at 95°C. TNET buffer (0.6 ml; TNET = 50 mM Tris-HCl pH 7.4, 150 mM NaCl, 5 mM EDTA, 1% Triton X-100, plus 1 mM phenylmethylsulphonyl fluoride (PMSF) and protease inhibitors) was added, and the insoluble material was removed by centrifugation for 30 min at 12,000 g. The supernatant was incubated overnight at 4°C with 50 μl rat monoclonal anti-HA antibody (clone 3F10, Roche Diagnostics), plus 100 μl of 50% protein G-Agarose beads (Roche Diagnostics). Supernatants were 10-fold concentrated using Vivaspin 500 ultrafiltration spin columns (10,000 MWCO PES, Sartorius AG), and aliquots were loaded on 10% SDS-PAGE gels. Immunoprecipitates were washed twice with TNET buffer, twice with 100 mM Tris-HCl pH 7.5, 300 mM NaCl, and once with 25 mM Tris-HCl pH 7.5. The proteins were eluted by incubation for 5 min at 95°C in 100 μl of 2-fold concentrated SDS-PAGE loading buffer. Samples were resolved by 10% SDS-polyacrylamide gels and autoradiographed.

Yeast strains were transformed with pAC99 reporter family plasmids. In each case, at least five independent transformants, cultivated in the same conditions, were assayed. Cells were broken using acid-washed glass beads, as described 26. Luciferase and β-galactosidase activities were assayed in the same crude extract. Readthrough efficiency is estimated by the ratio of luciferase activity to β-galactosidase activity. To establish the relative activities of β-galactosidase and luciferase when expressed in equimolar amounts, the ratio of luciferase activity to β-galactosidase from an in-frame control plasmid was taken as a reference. Readthrough frequency, expressed as percentage, was calculated by dividing the luciferase/β-galactosidase ratio obtained from each test construct by the same ratio obtained with the in-frame control construct 24. This allowed us to determine the percent readthrough while controlling differences in mRNA abundance or efficiency of translation initiation.

The GAL4-based two-hybrid system 39 was used. pGAD plasmids (empty pGAD, pGAD-SUP35, pGAD-SUP35A, pGAD-SUP35D and pGAD-SUP45) were transformed into the MATα pJ69-4A strain, and pGBDU plasmids (empty pGBDU, pGBDU-SUP35, pGBDU-SUP35A, pGBDU-SUP35D and pGBDU-SUP45) into the MATa pJ69-4A strain. These transformants were mated in different pairwise combinations overnight at 30°C on YPD plates, and diploids selected on uracil and leucine omission medium. The two-hybrid interactions were assessed by the ability to transactivate HIS3 or ADE2 reporter genes (to confer growth either in the absence of adenine, or in the absence of histidine and in presence of different concentrations of 3-amino-1,2,4-triazole (Sigma)). Interactions were visualized on histidine, uracil and leucine omission medium containing either 5 mM, 10 mM, or 25 mM 3-amino-1,2,4-triazole, and also on the more selective adenine, uracil and leucine omission medium. Similar interactions were detected on each medium when Sup35 was fused to the Gal4 activating domain (pGAD vectors) and Sup45 to the binding domain (pGBDU vectors). Lighter interactions were seen in the opposite combinations (Sup35 into pGBDU vectors and Sup45 into pGAD vectors) on the histidine omission medium, and no interaction at all on the adenine omission medium. As negative control, diploids bearing the empty vector and fusion constructs were used, and the HIS3 or ADE2 reporter genes were not activated. Quantitative β-galactosidase assays were undertaken on lysates prepared from each diploid strain in order to estimate the interaction strength, but no coherent results were obtained because of the leakiness of the promoter controlling the expression of the lacZ gene (same level of β-galactosidase activity in the diploid carrying at least one empty vector).

The incubation mixture (12.5 μl) contained 2 μM [γ-³²P] GTP (sp. activity 4000–5000 cpm/pmole), 20 mM Tris-HCl, pH 7.5, 30 mM NH₄Cl, 15 mM MgCl₂, 0.2 μM 80S rabbit ribosomes, 0.32 μM Sup45p and 0.4 μM Sup35pCter. The reaction was performed at 25°C for 30 min, stopped by adding 750 μl of 5% charcoal in 50 mM NaH₂PO₄ on ice. The mixture was vortexed and centrifuged at 10,000 rpm for 10 min at 4°C. The [³²P]P_i released into 500 μl of supernatant was quantitated by liquid scintillation counting. [³²P]P_i release in the absence of one out of the three protein components of the incubation mixture was measured and the average value (8–10%) was subtracted for all samples.

The RF stimulating activity of Sup35pcter towards Sup45p was measured as described 8. Stop-codon-dependent hydrolysis of f[³⁵S]Met-tRNA^fMet associated with the AUG-80S rabbit ribosome complex took place at 5 μM stop codon (UAAA) that corresponded to ~10% of the saturation level needed for complete fMet release with Sup45p but without Sup35pcter. The RF activity was calculated as the amount of f[³⁵S]Met released in the presence of stop codon; the value of f[³⁵S]Met released in the absence of stop codon was subtracted from all values.