Volume2
Nucleic Acids Research
number4April 1975
Modification of E. coli ribosomes and coliphage MS2 RNA by bisulfite: effects on ribosomal binding and protein synthesis
Barbara Braverman, Robert Shapiro, and Wlodzimierz Szer Department of Chemistry, New York University, New York, NY 10003 and Department of Biochemistry, NewYork University School of Medicine, New York, NY 10016, USA
Received 30 January 1975 ABSTRACT The reaction of E. coli 70S ribosomes with 0.2 M NaH35SO (pH 7.1 3.5 hrs, 370) led to the conversion of 4.5% of the uracil residues of the The modified ribosomes rRNA into 5,6-dihydrouracil-6-sulfonate residues. exhibited a ^ignificant decrease inliheir ability to bind [ C]-phenyland to incorporate [ C]-phenylalanine into protein in the alanyl-tRNA The ability of the modified ribosomes to presence of polyuridylic acid. form an initiation comp x as measure&b the A-U-G or coliphage MS2 RNA was also impaired, as was their dependent binding of ability to incorporate [ C]-lysine into 3rotein with MS2 RNA as messenger. Treatment of MS2 RNA with 0.2 M sodium [3 S] bisulfite, PH 7.0 at 250 C resulted in the substitution of 2.7% and 6.2% of the uracil residues by biImpairment of. respectively. sulfite after 1 and 3.5 hrs of reaction, functiQnof the MS2 RNA in both ini-ation complex formation and translation These reactions of the uracil residues of mRNA and assays was observed. rRNA may be a cause of biological damag-e inflicted by sodium bisulfite and sulfur dioxide.
[slQ-fMet-tRNA
INTRODUCTION The reaction of bisulfite ion with uracil and its derivatives to yield the corresponding 5,6-dihydrouracil-6-sulfonate (Scheme I) occurs rapidly under physiological conditions.1
HSO3-
p"H 71 37
+
OH(
o@LI ON
C
H
)
N 5033
90% intact and sedimented at 27S this suggests that bisulfite treatment did not as did untreated MS2 RNA; significantly alter the secondary structure of the RNA. RESULTS AND DISCUSSION The modified ribosomes were tested for impairment of function at various stages in protein synthesis using both a synthetic messenger, poly (U), and a natural messenger, coliphage MS2 RNA. As seen from Table I, control treatment of ribosomes with NaCl-Na2s%4has a negligible effect on their ability to function in all three assay systems with poly(U) as messenger. A 1.1% modification of the ribosomal RNA with bisulfite, however, resulted in a pronounced decrease in both the amount of [14C] phenylalanine incorporated into protein and the amount of [14C] phenylalanyl-t-RNAPhe Incubation bound to the treated ribosomes in the presence of poly(U). of the bisulfite modified ribosomes at 240C, pH 7.1,two hours, conditions which result in partial reversion of the uracil-bisulfite adduct to uracil1, restored significantly the activity of treated ribosomes in the tRNA binding assay. While the possibility exists that ribosomal proteins may also undergo reaction with bisulfite9, such reaction would not be reversible under the above conditions, and may therefore be ruled out Bias a major contributor to the observed loss of ribosomal function. ~14 sulfite modification has no effect on the binding of [ C] poly(U) to ribosomes (Table I, column 3). Table II reveals that "control" treatment of ribosomes has little or no effect on their ability to form iniation complexes with the trinucleoOnce again, bisulfite modification tide AUG or with MS2 RNA as messenger. results in a marked loss of tRNA binding capacity in both cases, with AUGdirected binding being somewhat more affected than MS2 RNA directed Since the ribosome-AUG- met-tRNA complex is less stable than binding. the complex with MS2 RNA, it is not surprising that this difference in In the stability is reflected to a greate? extent with damaged ribosomes. was fMet-tRNA RNA-directed of significantly activity MS2 binding, case restored after exposure of the modified ribosomes to conditions which reAll of the bound [14C] fmet-tRNA verse the bisulfite addition reaction. This means that bisulfite modifican react to form fMet-puromycin. fMet
503
Nucleic Acids Research
Table
I:
Effect of bisulfite treatment of ribosomes
tinding Pyly(U)-dependw, [
Poly4U)
of
C] phe-tRNA
Ribosomes
to bind and translate poly(U)
Untreated
1412
-
1045
-
Control treated
1353
100
1196
100
Bistulfite treated Control treated, reversed
823
42.1
570
Binding of [ 14 C]
ingorporation
dependent
of [ C] Phenylanine % Activity CP4
% Activity
CPM
their ability
on
1618
>100
-
1191
88
_
ribosoues
CP4
% Activity -
980
68.8
940
100
1030
>100
-
-
poly(U) to
-
Bisul fite treated, reversed
All values in net cpm a.
1.0
per
A260
d
unit of rihosomes.
According to Nirenberg and Leder, blank.s without poly(U)
b. As previously reported2', blanks without poly(U) c.
As previously reported
2,13
d. The deviation in percent activity for identical averaged ab'l't 3.0%.
30-40
were
blanks without ribosomes
were
messenger'
CPM
%
Activity
CPM
Untreated
2174
-
1124
1959
100
1147
Bisulfite treated
23.8
467 -
[14C]
14
%
Activity
[14C] phe-tRNAPhe=700 cpm/pmole
fpt-uroycn
1452
directed binding) CPM Activity
CPM
-
3057
1812
100
1912
1026
48.3
92.5
711
61.1
1026
56.6
[
C]
lysine)
1784
554
-'W-sRNALIe
(incoj-
RNA
and of
Translation
poration of
100
1061
cpm/p.ole
phenylalanine=5
C] poly(U)=10,000 cpm/A260 unit.
MS2 RNA as messenger formation
Control treated
Control treated, reversed
cpm;
their ability to form initiation complexes
on
AfNet
AUG as
cpm;
25 cpm; [
Bindingof [ 14C], fMet ;14 of Binding [ttRC]
Ribosomes
60-80
employing similar preparations of modified ribosomes
assays
Table II: Effect of bisulfite modification of ribosomes translate MS2 RNA
were
%
Activity
CPM
%
Activit
680 100
53.6
670 515
100
76.9
f
Bisulfite treated, reversed
-
_
All values in net cpm per 1.0 A260 unit of
_
ribosomesf
a. Blanks without initiation factors were about 120
cpm;
_
_
Assays as recently reported15'16'17 14 [
C] fMet-tRNA
fIlet
=350 cpm/pmole.
b. Blanks without initiation factors were 15-25 clpm. c. Blanks without initiation factors were about 200
cpm. 14
d. Blanks without initiation e.
Factors were 200-400
Blanks without ribosomes were 20-25 cpm;
[
cpm; [ C] lysine=50 cpm/pmole.
H] MS2 RNA
=
4500 cpm/0.32 A260 units.
f. A similar preparation of ribosomes reverted to 87.4% of the control value after 2 weeks at 5°C. g.
The deviation in percent activity for identical assays employing similar preparations of modified ribosomes averaged about 3.0%.
504
Nucleic Acids Research cation does not affect the peptidyl transferase reaction (Table II, column 3). In contrast to the translation of poly (U), control treatment of ribosomes with NaC1-'Na2S4 results in a marked decrease of MS2 RNA transBisulfite treated ribosomes were about half lation (Table II, column 4). as active as control ones, indicating significant impairment due to modification. In contrast to the results obtained with poly(U) as messenger, bisulfite modification has a moderate but significant effect on the ability of the treated ribosomes to bind I 3 H] MS2 RNA. This result is not unexpected since recent evidence indicates a possible role for the 16S RNA in the recognition of initiation binding sites on natural messenger 10,1 An eight nucleotide segment at the 3'-terminal end of 16S RNA which is complementary to the cistron initiation sites on MS2 RNA contains three uri1 dylic acid residues Two preparations of bisulfite-treated MS2 RNA (0.7% and 1.6% modification of total bases) were tested for their ability to function as messengers in the E. coli system, and to mediate the binding of the initiator tRNA to ribosomes. Control treatment of MS2 RNA with a mixture of sodium sulfate and sodium chloride did not affect its messenger properties (not shown). As seen from Table III, bisulfite modification results in a progressive loss of messenger activity in translation with increasing per cent saturation. Apparently, as in the case of poly(U)2, each uracil residue
Table III: Messenger activity of bisulfite-modified MS2 RNA
MS2 RNA
Bind iMtof [ C] fMettRNA % Activity CPM
Untreated
1222
100
Translatioe4of MS2 RNA (incorporation of [ C] lysine) % Activity CPM 100 9840
Bisulfite treated (2.71% saturation)
8
treated100 (6.18% saturation)100
8.3476 8.3476
Bisulfite
80.2
7413
75
unit of ribosomes. Assays as in All values in net cpm per 1.0 A Table II. Each value represents an aIv8age of two determinations employing similar but separate preparations of MS2 RNA.
Nucleic Acids Research saturated by bisulfite represents a block to translation. Binding of 14C] was reduced with-the 0.7% modified preparation as comf?4et--tRNA pared to untreated MS2 RNA but further modification appeared to have no inhibitory effect; indeed, a slight increase in binding was noted with the 1.6% modified RNA. This is most probably explained by partial denaturation of the RNA and the exposure of new non-physiological fMet-tRNA binding sites upon more extensive bisulfite treatment. It was observed that mild treatment of coliphage f2 RNA with formaldehyde also exposes abnormal sites for initiation 12' We had previously shown that partial saturation of the synthetic messenger, poly(U), by bisulfite disrupted the ability of poly(U) to participate in several of the steps in protein synthesis in the E. coli cell free system 2,' The same conclusion can now be applied to bisulfite modification of the natural messenger.MS2 RNA and of the ribosomes. In all three systems, the extent of depression of amino acids incorporation per amount of bisulfite saturation of the total RNA is roughly comparable. Thus, in 2.6% saturated poly(U), amino acid incorporation was depressed 54%2, in 1.7% saturated MS2 RNA, incorporation was down by 64%, in 0.7% saturated MS2 RNA, incorporation was down by 25% and in 1.1% saturated ribosomes, amino acid incorporation was reduced by 31%. The microbial growth-inhibiting properties of sodium bisulfite have 18 long been known . Epidemiological evidence has implicated sulfur dioxide (which is converted to bisulfite in aqueous solution) as a factor in a number of adverse effects of air pollution on the health of human populations 19. We believe that disruption of cellular protein synthesis is a plausible mechanism for these effects. The effect of the messenger RNA reaction, particularly, would be expected to be more serious in vivo than in our in vitro studies. In the assay for incorporation of amino acids In vivo into protein, relatively short peptides are scored positively. such fragments would be inactive. As the equilibrium constant for saturation of uracil by bisulfite is high, these effects should operate even at low concentrations of bisulfite. We are presently trying to verify our hypothesis in a bacterial system.
ACKNOWLEDGEMENTS This research was supported by grants from the National Institutes of Health (ES-18583 and GM-18589) (to R.S.) and from the National Institutes of Health (AI-11517 and CA-16239) (to W.S.). One of us (R.S.) is the holder of a Public Health Service Career Development Award (GM-50188) pro-
506
Nucleic Acids Research vided by the National Institute of General Medical Sciences, National Institutes of Health. REFERENCES 1 Shapiro, R., Servis, R.E. and Welcher, M. (1970) J. Am. Chem. Soc. 92, 422-424 2 Shapiro,R. and Braverman, B. (1972) Biochim. Biophys. Res. Commun. 47, 544-550
3 Shapiro, R.,Braverman, B. and Szer, W. (1973) Mol. Biol. Reports 1, 123-127
4 Stanley, W.M. Jr. and Bock, R.M.(1965) Biochemistry 4, 1302-1311 5 Tissieres, A., Watson, J.D., Schlessinger, D. and Hollingworth, B.R. (1959) J. Mol. Biol. 1, 221-233
6 Strauss, J.H. Jr. and Sinsheimer, R.L. (1963) J. Mol. Biol. 7, 43-54 7 Min Jou, W., Haegeman, G., Ysebaert, M. and Fiers, W. (1972) Nature 237, 82-88 8 Shapiro, R., Braverman, B., Louis, J.B. and Servis, R.E. (1973) J. Biol. Chem. 248, 4060-4064 9
Gunnison,
A.F. and Palmes, E.D. (1973) Toxicol. and Applied Pharmacol.
24, 266-278 10 Held, W.A., Gette, W.R. and Nomura, M. (1974) Biochemistry 13, 2115-2122
11 Shine, J. and Dalgarno, L. (1974) Proc. Natl. Acad. Sci. 71, 1342-1346 12 Lodish, H.F. (1970) J. Mol. Biol. 50, 689-702 13 Nirenberg, M. and Leder, P. (1964) Science 145, 1399-1407 14 Szer, W. (1970) Biochim. Biophys. Acta 213, 159-170 15 Leffler, S. and Szer, W. (1974) J. Biol. Chem. 249, 1458-1464
16 Leffler, S. and Szer, W. (1974) J. Biol. Chem. 249, 1465-1468 17 Szer, W. and Leffer, S. (1974) Proc. Natl. Acad. Sci. 71, 3611-3615
18 Chichester, D.F. and Tanner, F.W. Jr. (1972) Handbook of Food Additives, 2nd Ed., 142-147, CRC Press, Cleveland 19 Air Quality Criteria for Sulfur Oxides, Natl. Air Pollution Control Administration Publication No. AP-50 (1969) Ch. 9, Wash., D.C.
507
Nucleic Acids Research
number4April 1975
Modification of E. coli ribosomes and coliphage MS2 RNA by bisulfite: effects on ribosomal binding and protein synthesis
Barbara Braverman, Robert Shapiro, and Wlodzimierz Szer Department of Chemistry, New York University, New York, NY 10003 and Department of Biochemistry, NewYork University School of Medicine, New York, NY 10016, USA
Received 30 January 1975 ABSTRACT The reaction of E. coli 70S ribosomes with 0.2 M NaH35SO (pH 7.1 3.5 hrs, 370) led to the conversion of 4.5% of the uracil residues of the The modified ribosomes rRNA into 5,6-dihydrouracil-6-sulfonate residues. exhibited a ^ignificant decrease inliheir ability to bind [ C]-phenyland to incorporate [ C]-phenylalanine into protein in the alanyl-tRNA The ability of the modified ribosomes to presence of polyuridylic acid. form an initiation comp x as measure&b the A-U-G or coliphage MS2 RNA was also impaired, as was their dependent binding of ability to incorporate [ C]-lysine into 3rotein with MS2 RNA as messenger. Treatment of MS2 RNA with 0.2 M sodium [3 S] bisulfite, PH 7.0 at 250 C resulted in the substitution of 2.7% and 6.2% of the uracil residues by biImpairment of. respectively. sulfite after 1 and 3.5 hrs of reaction, functiQnof the MS2 RNA in both ini-ation complex formation and translation These reactions of the uracil residues of mRNA and assays was observed. rRNA may be a cause of biological damag-e inflicted by sodium bisulfite and sulfur dioxide.
[slQ-fMet-tRNA
INTRODUCTION The reaction of bisulfite ion with uracil and its derivatives to yield the corresponding 5,6-dihydrouracil-6-sulfonate (Scheme I) occurs rapidly under physiological conditions.1
HSO3-
p"H 71 37
+
OH(
o@LI ON
C
H
)
N 5033
90% intact and sedimented at 27S this suggests that bisulfite treatment did not as did untreated MS2 RNA; significantly alter the secondary structure of the RNA. RESULTS AND DISCUSSION The modified ribosomes were tested for impairment of function at various stages in protein synthesis using both a synthetic messenger, poly (U), and a natural messenger, coliphage MS2 RNA. As seen from Table I, control treatment of ribosomes with NaCl-Na2s%4has a negligible effect on their ability to function in all three assay systems with poly(U) as messenger. A 1.1% modification of the ribosomal RNA with bisulfite, however, resulted in a pronounced decrease in both the amount of [14C] phenylalanine incorporated into protein and the amount of [14C] phenylalanyl-t-RNAPhe Incubation bound to the treated ribosomes in the presence of poly(U). of the bisulfite modified ribosomes at 240C, pH 7.1,two hours, conditions which result in partial reversion of the uracil-bisulfite adduct to uracil1, restored significantly the activity of treated ribosomes in the tRNA binding assay. While the possibility exists that ribosomal proteins may also undergo reaction with bisulfite9, such reaction would not be reversible under the above conditions, and may therefore be ruled out Bias a major contributor to the observed loss of ribosomal function. ~14 sulfite modification has no effect on the binding of [ C] poly(U) to ribosomes (Table I, column 3). Table II reveals that "control" treatment of ribosomes has little or no effect on their ability to form iniation complexes with the trinucleoOnce again, bisulfite modification tide AUG or with MS2 RNA as messenger. results in a marked loss of tRNA binding capacity in both cases, with AUGdirected binding being somewhat more affected than MS2 RNA directed Since the ribosome-AUG- met-tRNA complex is less stable than binding. the complex with MS2 RNA, it is not surprising that this difference in In the stability is reflected to a greate? extent with damaged ribosomes. was fMet-tRNA RNA-directed of significantly activity MS2 binding, case restored after exposure of the modified ribosomes to conditions which reAll of the bound [14C] fmet-tRNA verse the bisulfite addition reaction. This means that bisulfite modifican react to form fMet-puromycin. fMet
503
Nucleic Acids Research
Table
I:
Effect of bisulfite treatment of ribosomes
tinding Pyly(U)-dependw, [
Poly4U)
of
C] phe-tRNA
Ribosomes
to bind and translate poly(U)
Untreated
1412
-
1045
-
Control treated
1353
100
1196
100
Bistulfite treated Control treated, reversed
823
42.1
570
Binding of [ 14 C]
ingorporation
dependent
of [ C] Phenylanine % Activity CP4
% Activity
CPM
their ability
on
1618
>100
-
1191
88
_
ribosoues
CP4
% Activity -
980
68.8
940
100
1030
>100
-
-
poly(U) to
-
Bisul fite treated, reversed
All values in net cpm a.
1.0
per
A260
d
unit of rihosomes.
According to Nirenberg and Leder, blank.s without poly(U)
b. As previously reported2', blanks without poly(U) c.
As previously reported
2,13
d. The deviation in percent activity for identical averaged ab'l't 3.0%.
30-40
were
blanks without ribosomes
were
messenger'
CPM
%
Activity
CPM
Untreated
2174
-
1124
1959
100
1147
Bisulfite treated
23.8
467 -
[14C]
14
%
Activity
[14C] phe-tRNAPhe=700 cpm/pmole
fpt-uroycn
1452
directed binding) CPM Activity
CPM
-
3057
1812
100
1912
1026
48.3
92.5
711
61.1
1026
56.6
[
C]
lysine)
1784
554
-'W-sRNALIe
(incoj-
RNA
and of
Translation
poration of
100
1061
cpm/p.ole
phenylalanine=5
C] poly(U)=10,000 cpm/A260 unit.
MS2 RNA as messenger formation
Control treated
Control treated, reversed
cpm;
their ability to form initiation complexes
on
AfNet
AUG as
cpm;
25 cpm; [
Bindingof [ 14C], fMet ;14 of Binding [ttRC]
Ribosomes
60-80
employing similar preparations of modified ribosomes
assays
Table II: Effect of bisulfite modification of ribosomes translate MS2 RNA
were
%
Activity
CPM
%
Activit
680 100
53.6
670 515
100
76.9
f
Bisulfite treated, reversed
-
_
All values in net cpm per 1.0 A260 unit of
_
ribosomesf
a. Blanks without initiation factors were about 120
cpm;
_
_
Assays as recently reported15'16'17 14 [
C] fMet-tRNA
fIlet
=350 cpm/pmole.
b. Blanks without initiation factors were 15-25 clpm. c. Blanks without initiation factors were about 200
cpm. 14
d. Blanks without initiation e.
Factors were 200-400
Blanks without ribosomes were 20-25 cpm;
[
cpm; [ C] lysine=50 cpm/pmole.
H] MS2 RNA
=
4500 cpm/0.32 A260 units.
f. A similar preparation of ribosomes reverted to 87.4% of the control value after 2 weeks at 5°C. g.
The deviation in percent activity for identical assays employing similar preparations of modified ribosomes averaged about 3.0%.
504
Nucleic Acids Research cation does not affect the peptidyl transferase reaction (Table II, column 3). In contrast to the translation of poly (U), control treatment of ribosomes with NaC1-'Na2S4 results in a marked decrease of MS2 RNA transBisulfite treated ribosomes were about half lation (Table II, column 4). as active as control ones, indicating significant impairment due to modification. In contrast to the results obtained with poly(U) as messenger, bisulfite modification has a moderate but significant effect on the ability of the treated ribosomes to bind I 3 H] MS2 RNA. This result is not unexpected since recent evidence indicates a possible role for the 16S RNA in the recognition of initiation binding sites on natural messenger 10,1 An eight nucleotide segment at the 3'-terminal end of 16S RNA which is complementary to the cistron initiation sites on MS2 RNA contains three uri1 dylic acid residues Two preparations of bisulfite-treated MS2 RNA (0.7% and 1.6% modification of total bases) were tested for their ability to function as messengers in the E. coli system, and to mediate the binding of the initiator tRNA to ribosomes. Control treatment of MS2 RNA with a mixture of sodium sulfate and sodium chloride did not affect its messenger properties (not shown). As seen from Table III, bisulfite modification results in a progressive loss of messenger activity in translation with increasing per cent saturation. Apparently, as in the case of poly(U)2, each uracil residue
Table III: Messenger activity of bisulfite-modified MS2 RNA
MS2 RNA
Bind iMtof [ C] fMettRNA % Activity CPM
Untreated
1222
100
Translatioe4of MS2 RNA (incorporation of [ C] lysine) % Activity CPM 100 9840
Bisulfite treated (2.71% saturation)
8
treated100 (6.18% saturation)100
8.3476 8.3476
Bisulfite
80.2
7413
75
unit of ribosomes. Assays as in All values in net cpm per 1.0 A Table II. Each value represents an aIv8age of two determinations employing similar but separate preparations of MS2 RNA.
Nucleic Acids Research saturated by bisulfite represents a block to translation. Binding of 14C] was reduced with-the 0.7% modified preparation as comf?4et--tRNA pared to untreated MS2 RNA but further modification appeared to have no inhibitory effect; indeed, a slight increase in binding was noted with the 1.6% modified RNA. This is most probably explained by partial denaturation of the RNA and the exposure of new non-physiological fMet-tRNA binding sites upon more extensive bisulfite treatment. It was observed that mild treatment of coliphage f2 RNA with formaldehyde also exposes abnormal sites for initiation 12' We had previously shown that partial saturation of the synthetic messenger, poly(U), by bisulfite disrupted the ability of poly(U) to participate in several of the steps in protein synthesis in the E. coli cell free system 2,' The same conclusion can now be applied to bisulfite modification of the natural messenger.MS2 RNA and of the ribosomes. In all three systems, the extent of depression of amino acids incorporation per amount of bisulfite saturation of the total RNA is roughly comparable. Thus, in 2.6% saturated poly(U), amino acid incorporation was depressed 54%2, in 1.7% saturated MS2 RNA, incorporation was down by 64%, in 0.7% saturated MS2 RNA, incorporation was down by 25% and in 1.1% saturated ribosomes, amino acid incorporation was reduced by 31%. The microbial growth-inhibiting properties of sodium bisulfite have 18 long been known . Epidemiological evidence has implicated sulfur dioxide (which is converted to bisulfite in aqueous solution) as a factor in a number of adverse effects of air pollution on the health of human populations 19. We believe that disruption of cellular protein synthesis is a plausible mechanism for these effects. The effect of the messenger RNA reaction, particularly, would be expected to be more serious in vivo than in our in vitro studies. In the assay for incorporation of amino acids In vivo into protein, relatively short peptides are scored positively. such fragments would be inactive. As the equilibrium constant for saturation of uracil by bisulfite is high, these effects should operate even at low concentrations of bisulfite. We are presently trying to verify our hypothesis in a bacterial system.
ACKNOWLEDGEMENTS This research was supported by grants from the National Institutes of Health (ES-18583 and GM-18589) (to R.S.) and from the National Institutes of Health (AI-11517 and CA-16239) (to W.S.). One of us (R.S.) is the holder of a Public Health Service Career Development Award (GM-50188) pro-
506
Nucleic Acids Research vided by the National Institute of General Medical Sciences, National Institutes of Health. REFERENCES 1 Shapiro, R., Servis, R.E. and Welcher, M. (1970) J. Am. Chem. Soc. 92, 422-424 2 Shapiro,R. and Braverman, B. (1972) Biochim. Biophys. Res. Commun. 47, 544-550
3 Shapiro, R.,Braverman, B. and Szer, W. (1973) Mol. Biol. Reports 1, 123-127
4 Stanley, W.M. Jr. and Bock, R.M.(1965) Biochemistry 4, 1302-1311 5 Tissieres, A., Watson, J.D., Schlessinger, D. and Hollingworth, B.R. (1959) J. Mol. Biol. 1, 221-233
6 Strauss, J.H. Jr. and Sinsheimer, R.L. (1963) J. Mol. Biol. 7, 43-54 7 Min Jou, W., Haegeman, G., Ysebaert, M. and Fiers, W. (1972) Nature 237, 82-88 8 Shapiro, R., Braverman, B., Louis, J.B. and Servis, R.E. (1973) J. Biol. Chem. 248, 4060-4064 9
Gunnison,
A.F. and Palmes, E.D. (1973) Toxicol. and Applied Pharmacol.
24, 266-278 10 Held, W.A., Gette, W.R. and Nomura, M. (1974) Biochemistry 13, 2115-2122
11 Shine, J. and Dalgarno, L. (1974) Proc. Natl. Acad. Sci. 71, 1342-1346 12 Lodish, H.F. (1970) J. Mol. Biol. 50, 689-702 13 Nirenberg, M. and Leder, P. (1964) Science 145, 1399-1407 14 Szer, W. (1970) Biochim. Biophys. Acta 213, 159-170 15 Leffler, S. and Szer, W. (1974) J. Biol. Chem. 249, 1458-1464
16 Leffler, S. and Szer, W. (1974) J. Biol. Chem. 249, 1465-1468 17 Szer, W. and Leffer, S. (1974) Proc. Natl. Acad. Sci. 71, 3611-3615
18 Chichester, D.F. and Tanner, F.W. Jr. (1972) Handbook of Food Additives, 2nd Ed., 142-147, CRC Press, Cleveland 19 Air Quality Criteria for Sulfur Oxides, Natl. Air Pollution Control Administration Publication No. AP-50 (1969) Ch. 9, Wash., D.C.
507
