JOURNAL OF BACTERIOLOGY, May 1976, p. 1009-1011 Copyright © 1976 American Society for Microbiology
Vol. 126, No. 2 Printed in U.S.A.
Selective Methylation:
an
Incorrect Hypothesis
LEE SHUGART Biology Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37830
Received for publication 5 February 1976
"Selective methylation," a hypothesis proposed to explain the discrepancy found in the degree of methyl deficiency of transfer ribonucleic acid, cannot be explained on the basis of some biological phenomenon. I have reported (7) that transfer ribonucleic acid (tRNA) obtained from an rel - mutant of Escherichia coli, cultured under conditions of methionine starvation, is not completely undermethylated. This observation was based on a comparison of the methylated nucleoside content of tRNA obtained from rel- cells grown under normal and methionine-starvation culture conditions, showing that 7-methylguanine (7MeGua) occurs in quantities larger than predicted in the latter form of tRNA; the findings were interpreted to mean that some type of "selective methylation" of tRNA is operative during the latter culture conditions. Selective methylation was postulated to occur during methionine-starvation culture conditions as a result of (i) loss by the microorganism of certain tRNA methyltransferase activities, and/or (ii) difference in affinity of these enzymes for the precursors of transmethylation reaction that are present in the cell at the onset of methionine starvation. Coupled with these possibilities was the suspected accumulation in the cell of S-adenosylhomocysteine (SAdoHcy), a potent, differential competitive inhibitor of the enzymes of transmethylation. Recently Davis and Nierlich (2) reported no significant preferential synthesis of 7MeGua during methionine starvation. Results presented in this paper support their observation rather than the proposed hypothesis of selective methylation. tRNA methyltransferases. A comparison of several tRNA methyltransferase activities present in cell extracts from E. coli cells grown under normal and methionine-starved culture conditions was made by determining the distribution of enzymatically incorporated [14C]CH3 groups into various nucleosides and bases of methyl-deficient tRNA (9). The data in Table 1 show that tRNA(adenine-N6)-, tRNA(guanine-1)-, tRNA(guanine7)-, and tRNA(uracil-5)-methyltransferase activities (EC 2.1.1.30, 2.1.1.31, 2.1.1.33, and 2.1.1.35, respectively) are present in both cell
extracts and account for approximately 80 to 90% of the total amount of tRNA transmethylation. Therefore, if selective methylation was occurring (as previously reported [7]) at positions such as N6-methyladenine (N6MeAde) and 7MeGua of tRNA during methionine starvation of E. coli, it could not be due to the loss of tRNA methyltransferase activities specific for other sites. In fact, the apparent decrease in tRNA(guanine-7)-methyltransferase activity in the cell extract from methionine-starved cells is the opposite of what was predicted. Because the cell extracts were dialyzed before conducting these experiments, it cannot be concluded from these data whether or not inhibition in vivo of these activities occurs during methionine-starvation culture conditions as a result of the presence of S-AdoHcy. S-AdoHcy. Attempts to detect S-AdoHcy in the cell extracts obtained from either normal or methionine-starved cells were unsuccessful. The presence of a specific hydrolytic nucleosidase (3) would account for this observation. Therefore, the following experiments were performed in an effort to understand the fate of SAdoHcy in E. coli during methionine-starvation culture conditions. At various intervals after the commencement of logarithmic growth, samples of cells were removed and were assayed for their S-AdoHcy nucleosidase activity. The data (summarized in Table 2) show that the cells proceed through a normal logarithmic phase of growth which is terminated by the depletion of methionine from the culture medium (8). During methionine-starvation culture conditions, methyl-deficient tRNA is synthesized. S-AdoHcy nucleosidase activity is shown to be present in the cells before and after methionine depletion from the culture medium. Further, the demonstration of S-AdoHcy nucleosidase activity in viable cells made permeable by cold shock (1) indicates that the enzyme activity measured in the cell extracts is probably not a result of an "unmasking" phenomenon due to disruption of cells by sonic
1009
1010
NOTES
J. BACTERIOL.
TABLE 1. Analysis of the distribution of enzymatically incorporated methyl groups into methyl-deficient tRNAa Source of enzyme preparation
Method of tRNA hydrol-
ysis
Distribution of methyl groups (% total methyl groups incorporated)
5MeUra
N6MeAde
lMeGua
7MeGua
Normal cells Enzymatic 62.1 5.4 4.3 Normal cells Formic acid 68.3 3.8 12.6 Methionine-starved cells Enzymatic 54.9 9.1 8.9 Methionine-starved cells Formic acid 7.7 61.4 2.8 a The in vitro methylation of methyl-deficient tRNA with [14C]CH3-S-adenosylmethionine and its subsequent recovery was as described (9). Analyses of the distribution of methyl groups into tRNA were by two methods: (i) enzymatic digestion of tRNA to nucleosides (9), and (ii) chemical hydrolysis of tRNA to its bases with 88% formic acid (6). Separation and detection of these constituents was by cation-exchange column chromatography (9, 10). 5MeUra, 5-methyluracil; N6MeAde, N6-methyladenine. TABLE 2. Analysis of S-adenosylhomocysteine nucleosidase activity and tRNA acceptance during normal and methionine-starved culture conditions of E. coli Growtha S-AdoHcy nucleosidase activity9 Acceptancec (nmol/A2w) Time (h)
Absorbancy (A470)
Cell count (x 10)
Cold-shocked
Cell extract
(nmol/h per (nmol/h per mg) Methyl groups cellA A470)
Leucine
0.00 0.012 0.75 0.053 1.50 0.091 0.2 2.25 0.325 0.7 715 5.00 0.901 2.2 0.000 10.3 750 0.073 6.00 0.470 1.510 820 0.076 8.00 1.490 4.0 800 11.00 1.500 850 12.0 1.086 0.060 " Two liters of defined culture medium containing 5 ,ug of methionine per ml were inoculated with relmutant cells of E. coli. At various time intervals after appearance of growth (zero time), the absorbancy at 470 nm (A47(1) and number of cells per ml of culture medium were measured. b Samples of cells equal to 200 ml of culture medium were recovered by centrifugation, washed with cold, sterile 1% saline solution, and suspended in 2 ml of 50 mM tris(hydroxymethyl)aminomethane buffer (pH 7.5) in 10 mM MgCl2. Before measuring S-AdoHcy nucleosidase activity (3), the cells were made permeable (1) and/or disrupted by sonic oscillation. Protein determinations of the 150,000 x g of cell extract was performed by the method of Lowry et al. (5). ' tRNA was extracted from the cell extracts, and the methyl group and leucine acceptance was determined as described (8, 9).
oscillation. These data suggest that the inability to detect S-AdoHcy in cell extracts of E. coli is a valid observation which is due to the presence of an active nucleosidase in these cells. Therefore, the postulated differential inhibitory effect of S-AdoHcy on the tRNA methyltransferases in the cell during methionine-starvation culture conditions cannot be invoked as a mechanism for selective methylation. 5-Methyluracil and 7MeGua. The basis for the hypothesis of selective methylation was an analysis (7) of the quantitative data (11) obtained on the methylated nucleoside content of E. coli tRNA. Because methylation at the 5 position of uracil and the 7 position of guanine demonstrated the greatest divergency, their contents in normal and methyl-deficient tRNA's were redetermined by using formic acid hydrol-
ysis (6) and cation-exchange column chromatography (9, 10). These techniques result in the minimum of chemical degradation. The data obtained on the content of these methylated bases in both mixed normal and methyl-deficient tRNA preparations, as well as two purified phenylalanine tRNA species, are compared in Table 3. The ratio of 5-methyluracil content in the mixed normal tRNA preparation to the content in the mixed methyl-deficient tRNA preparation is essentially the same as the ratio of the 7MeGua content in the two preparations. In addition, these two bases are absent in a purified methyl-deficient phenylalanine tRNA (4), which are the anticipated results if selective methylation doesn't occur. It is concluded, therefore, that both positions reflect the same degree of methyl deficiency.
VOL. 126, 1976
NOTES
TABLE 3. Analysis of the distribution of 5methyluracil (5MeUra) and 7-methylguanine in E. coli rel - tRNA after formic acid hydrolysis Amount of methylated bases in tRNA
tRNA preparation
preparationa (mol/ mol of tRNA)
5MeUra
7MeGua
0.52 0.92 Mixed normal 0.16 0.31 Mixed methyl deficient 0.95 1.00 Normal phenylalanine 0.00 Methyl-deficient phenylalanine 0.00 a Approximately 4.5 absorbancy units at 260 nm of each tRNA preparation were chemically hydrolyzed with 88% formic acid (6). 5MeUra and 7MeGua were separated and quantitatively determined by cation-exchange column chromatography (9, 10). The data represents the average of three separate determinations for each of the mixed tRNA preparations and two separate determinations for each of the purified phenylalanine tRNA species.
In this paper I report data which indicate that selective methylation cannot be attributed either to the absence of certain tRNA methyltransferase activities or the presence of SAdoHcy in methionine-starved cells. An attempt to verify the data on the 7MeGua content of methyl-deficient tRNA (11) was unsuccessful. Presumably the chemical instability of this nucleoside to the analytical procedures used resulted in erroneous data. These results suggest that selective methylation of E. coli tRNA during methionine-starvation culture conditions probably doesn't occur. This research was sponsored by the U.S. Energy Re-
1011
search and Development Administration under contract with the Union Carbide Corporation. LITERATURE CITED 1. Atherly, A. G. 1974. Ribonucleic acid regulation in permeabilized cells of E. coli capable of ribonucleic acid and protein synthesis. J. Bacteriol. 118:1186-1189. 2. Davis, A. R., and D. P. Nierlich. 1974. The methylation of transfer RNA in Escherichia coli. Biochim. Biophys. Acta 374:22-37. 3. Duerre, J. 1962. A hydrolytic nucleosidase acting on Sadenosylhomocysteine and on methylthioadenosine. J. Biol. Chem. 237:3737-3741. 4. Isham, K. R., and M. P. Stulberg. 1974. Modified nucleosides in undermethylated phenylalanine transfer RNA from Escherichia coli. Biochim. Biophys. Acta 340:177-182. 5. Lowry, 0. H., N. J. Rosenbrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275. 6. Munns, T. W., K. C. Podratz, and P. A. Katzman. 1974. A method for determining the methylated constituents of tRNA. Biochemistry 13:4409-4416. 7. Shugart, L. 1973. Selective methylation of newly synthesized tRNA from rel - mutant of E. coli during methionine starvation. Biochem. Biophys. Res. Commun. 53:1200-1204. 8. Shugart, L., G. D. Novelli, and M. P. Stulberg. 1968. Isolation and properties of undermethylated phenylalanine tRNA from a relaxed mutant of E. coli. Biochim. Biophys. Acta 157:83-90. 9. Shugart, L., and M. P. Stulberg. 1974. Isolation, purification, and methylation of undermethylated tRNAPhe from an RC'*' mutant of E. coli. Methods Enzymol. 29:492-502. 10. Uziel, M., C. K. Koh, and W. E. Cohn. 1968. Rapid ionexchange chromatographic microanalysis of ultraviolet-absorbing materials and its application to nucleosides. Anal. Biochem. 25:77-98. 11. Waters, L., L. Shugart, W. K. Yang, and A. Best. 1973. Some physical and biological properties of 4-thiouridine- and dihydrouridine-deficient tRNA from chloramphenicol-treated E. coli. Arch. Biochem. Biophys. 156:780-793.
Vol. 126, No. 2 Printed in U.S.A.
Selective Methylation:
an
Incorrect Hypothesis
LEE SHUGART Biology Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37830
Received for publication 5 February 1976
"Selective methylation," a hypothesis proposed to explain the discrepancy found in the degree of methyl deficiency of transfer ribonucleic acid, cannot be explained on the basis of some biological phenomenon. I have reported (7) that transfer ribonucleic acid (tRNA) obtained from an rel - mutant of Escherichia coli, cultured under conditions of methionine starvation, is not completely undermethylated. This observation was based on a comparison of the methylated nucleoside content of tRNA obtained from rel- cells grown under normal and methionine-starvation culture conditions, showing that 7-methylguanine (7MeGua) occurs in quantities larger than predicted in the latter form of tRNA; the findings were interpreted to mean that some type of "selective methylation" of tRNA is operative during the latter culture conditions. Selective methylation was postulated to occur during methionine-starvation culture conditions as a result of (i) loss by the microorganism of certain tRNA methyltransferase activities, and/or (ii) difference in affinity of these enzymes for the precursors of transmethylation reaction that are present in the cell at the onset of methionine starvation. Coupled with these possibilities was the suspected accumulation in the cell of S-adenosylhomocysteine (SAdoHcy), a potent, differential competitive inhibitor of the enzymes of transmethylation. Recently Davis and Nierlich (2) reported no significant preferential synthesis of 7MeGua during methionine starvation. Results presented in this paper support their observation rather than the proposed hypothesis of selective methylation. tRNA methyltransferases. A comparison of several tRNA methyltransferase activities present in cell extracts from E. coli cells grown under normal and methionine-starved culture conditions was made by determining the distribution of enzymatically incorporated [14C]CH3 groups into various nucleosides and bases of methyl-deficient tRNA (9). The data in Table 1 show that tRNA(adenine-N6)-, tRNA(guanine-1)-, tRNA(guanine7)-, and tRNA(uracil-5)-methyltransferase activities (EC 2.1.1.30, 2.1.1.31, 2.1.1.33, and 2.1.1.35, respectively) are present in both cell
extracts and account for approximately 80 to 90% of the total amount of tRNA transmethylation. Therefore, if selective methylation was occurring (as previously reported [7]) at positions such as N6-methyladenine (N6MeAde) and 7MeGua of tRNA during methionine starvation of E. coli, it could not be due to the loss of tRNA methyltransferase activities specific for other sites. In fact, the apparent decrease in tRNA(guanine-7)-methyltransferase activity in the cell extract from methionine-starved cells is the opposite of what was predicted. Because the cell extracts were dialyzed before conducting these experiments, it cannot be concluded from these data whether or not inhibition in vivo of these activities occurs during methionine-starvation culture conditions as a result of the presence of S-AdoHcy. S-AdoHcy. Attempts to detect S-AdoHcy in the cell extracts obtained from either normal or methionine-starved cells were unsuccessful. The presence of a specific hydrolytic nucleosidase (3) would account for this observation. Therefore, the following experiments were performed in an effort to understand the fate of SAdoHcy in E. coli during methionine-starvation culture conditions. At various intervals after the commencement of logarithmic growth, samples of cells were removed and were assayed for their S-AdoHcy nucleosidase activity. The data (summarized in Table 2) show that the cells proceed through a normal logarithmic phase of growth which is terminated by the depletion of methionine from the culture medium (8). During methionine-starvation culture conditions, methyl-deficient tRNA is synthesized. S-AdoHcy nucleosidase activity is shown to be present in the cells before and after methionine depletion from the culture medium. Further, the demonstration of S-AdoHcy nucleosidase activity in viable cells made permeable by cold shock (1) indicates that the enzyme activity measured in the cell extracts is probably not a result of an "unmasking" phenomenon due to disruption of cells by sonic
1009
1010
NOTES
J. BACTERIOL.
TABLE 1. Analysis of the distribution of enzymatically incorporated methyl groups into methyl-deficient tRNAa Source of enzyme preparation
Method of tRNA hydrol-
ysis
Distribution of methyl groups (% total methyl groups incorporated)
5MeUra
N6MeAde
lMeGua
7MeGua
Normal cells Enzymatic 62.1 5.4 4.3 Normal cells Formic acid 68.3 3.8 12.6 Methionine-starved cells Enzymatic 54.9 9.1 8.9 Methionine-starved cells Formic acid 7.7 61.4 2.8 a The in vitro methylation of methyl-deficient tRNA with [14C]CH3-S-adenosylmethionine and its subsequent recovery was as described (9). Analyses of the distribution of methyl groups into tRNA were by two methods: (i) enzymatic digestion of tRNA to nucleosides (9), and (ii) chemical hydrolysis of tRNA to its bases with 88% formic acid (6). Separation and detection of these constituents was by cation-exchange column chromatography (9, 10). 5MeUra, 5-methyluracil; N6MeAde, N6-methyladenine. TABLE 2. Analysis of S-adenosylhomocysteine nucleosidase activity and tRNA acceptance during normal and methionine-starved culture conditions of E. coli Growtha S-AdoHcy nucleosidase activity9 Acceptancec (nmol/A2w) Time (h)
Absorbancy (A470)
Cell count (x 10)
Cold-shocked
Cell extract
(nmol/h per (nmol/h per mg) Methyl groups cellA A470)
Leucine
0.00 0.012 0.75 0.053 1.50 0.091 0.2 2.25 0.325 0.7 715 5.00 0.901 2.2 0.000 10.3 750 0.073 6.00 0.470 1.510 820 0.076 8.00 1.490 4.0 800 11.00 1.500 850 12.0 1.086 0.060 " Two liters of defined culture medium containing 5 ,ug of methionine per ml were inoculated with relmutant cells of E. coli. At various time intervals after appearance of growth (zero time), the absorbancy at 470 nm (A47(1) and number of cells per ml of culture medium were measured. b Samples of cells equal to 200 ml of culture medium were recovered by centrifugation, washed with cold, sterile 1% saline solution, and suspended in 2 ml of 50 mM tris(hydroxymethyl)aminomethane buffer (pH 7.5) in 10 mM MgCl2. Before measuring S-AdoHcy nucleosidase activity (3), the cells were made permeable (1) and/or disrupted by sonic oscillation. Protein determinations of the 150,000 x g of cell extract was performed by the method of Lowry et al. (5). ' tRNA was extracted from the cell extracts, and the methyl group and leucine acceptance was determined as described (8, 9).
oscillation. These data suggest that the inability to detect S-AdoHcy in cell extracts of E. coli is a valid observation which is due to the presence of an active nucleosidase in these cells. Therefore, the postulated differential inhibitory effect of S-AdoHcy on the tRNA methyltransferases in the cell during methionine-starvation culture conditions cannot be invoked as a mechanism for selective methylation. 5-Methyluracil and 7MeGua. The basis for the hypothesis of selective methylation was an analysis (7) of the quantitative data (11) obtained on the methylated nucleoside content of E. coli tRNA. Because methylation at the 5 position of uracil and the 7 position of guanine demonstrated the greatest divergency, their contents in normal and methyl-deficient tRNA's were redetermined by using formic acid hydrol-
ysis (6) and cation-exchange column chromatography (9, 10). These techniques result in the minimum of chemical degradation. The data obtained on the content of these methylated bases in both mixed normal and methyl-deficient tRNA preparations, as well as two purified phenylalanine tRNA species, are compared in Table 3. The ratio of 5-methyluracil content in the mixed normal tRNA preparation to the content in the mixed methyl-deficient tRNA preparation is essentially the same as the ratio of the 7MeGua content in the two preparations. In addition, these two bases are absent in a purified methyl-deficient phenylalanine tRNA (4), which are the anticipated results if selective methylation doesn't occur. It is concluded, therefore, that both positions reflect the same degree of methyl deficiency.
VOL. 126, 1976
NOTES
TABLE 3. Analysis of the distribution of 5methyluracil (5MeUra) and 7-methylguanine in E. coli rel - tRNA after formic acid hydrolysis Amount of methylated bases in tRNA
tRNA preparation
preparationa (mol/ mol of tRNA)
5MeUra
7MeGua
0.52 0.92 Mixed normal 0.16 0.31 Mixed methyl deficient 0.95 1.00 Normal phenylalanine 0.00 Methyl-deficient phenylalanine 0.00 a Approximately 4.5 absorbancy units at 260 nm of each tRNA preparation were chemically hydrolyzed with 88% formic acid (6). 5MeUra and 7MeGua were separated and quantitatively determined by cation-exchange column chromatography (9, 10). The data represents the average of three separate determinations for each of the mixed tRNA preparations and two separate determinations for each of the purified phenylalanine tRNA species.
In this paper I report data which indicate that selective methylation cannot be attributed either to the absence of certain tRNA methyltransferase activities or the presence of SAdoHcy in methionine-starved cells. An attempt to verify the data on the 7MeGua content of methyl-deficient tRNA (11) was unsuccessful. Presumably the chemical instability of this nucleoside to the analytical procedures used resulted in erroneous data. These results suggest that selective methylation of E. coli tRNA during methionine-starvation culture conditions probably doesn't occur. This research was sponsored by the U.S. Energy Re-
1011
search and Development Administration under contract with the Union Carbide Corporation. LITERATURE CITED 1. Atherly, A. G. 1974. Ribonucleic acid regulation in permeabilized cells of E. coli capable of ribonucleic acid and protein synthesis. J. Bacteriol. 118:1186-1189. 2. Davis, A. R., and D. P. Nierlich. 1974. The methylation of transfer RNA in Escherichia coli. Biochim. Biophys. Acta 374:22-37. 3. Duerre, J. 1962. A hydrolytic nucleosidase acting on Sadenosylhomocysteine and on methylthioadenosine. J. Biol. Chem. 237:3737-3741. 4. Isham, K. R., and M. P. Stulberg. 1974. Modified nucleosides in undermethylated phenylalanine transfer RNA from Escherichia coli. Biochim. Biophys. Acta 340:177-182. 5. Lowry, 0. H., N. J. Rosenbrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275. 6. Munns, T. W., K. C. Podratz, and P. A. Katzman. 1974. A method for determining the methylated constituents of tRNA. Biochemistry 13:4409-4416. 7. Shugart, L. 1973. Selective methylation of newly synthesized tRNA from rel - mutant of E. coli during methionine starvation. Biochem. Biophys. Res. Commun. 53:1200-1204. 8. Shugart, L., G. D. Novelli, and M. P. Stulberg. 1968. Isolation and properties of undermethylated phenylalanine tRNA from a relaxed mutant of E. coli. Biochim. Biophys. Acta 157:83-90. 9. Shugart, L., and M. P. Stulberg. 1974. Isolation, purification, and methylation of undermethylated tRNAPhe from an RC'*' mutant of E. coli. Methods Enzymol. 29:492-502. 10. Uziel, M., C. K. Koh, and W. E. Cohn. 1968. Rapid ionexchange chromatographic microanalysis of ultraviolet-absorbing materials and its application to nucleosides. Anal. Biochem. 25:77-98. 11. Waters, L., L. Shugart, W. K. Yang, and A. Best. 1973. Some physical and biological properties of 4-thiouridine- and dihydrouridine-deficient tRNA from chloramphenicol-treated E. coli. Arch. Biochem. Biophys. 156:780-793.
