Control of basal-level codon misreading in Escherichia coli

Vol. 121, No. 2, 1984 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

June 15, 1984 Pages 487-492

CONTROL OF BASAL-LEVEL CODON MISREADING IN ESCHERICHIA COLI

Jack Parker and Greg Holtz

Department of Microbiology, Southern Illinois University, Carbondale, Illinois, 62901

Received April 24, 1984

SUMMARY: Basal-level misreading of asparagine codons was examined in a number of Escherichia coli strains. Lysine substitutions were measured by quantitating the amount of charge heterogeneity in MS2 coat protein. In most strains the heterogeneity was consistent with misreading of AAU codons at a frequency of 3-6 x 10-3. Strains with streptomycin resistance mutations (*) have reduced levels of misreading. There is no significant difference in the frequency of basal-level errors in stringent (relti) and relaxed (relA) strains, even during starvation for amino acids unrelated to the substitution being studied.

flissense errors in codon reading result in specific amino acid

substitutions. Only relatively recently have in vivo measurements been made --

of the frequency, specificity, and control of such errors in E. coli. --

Cysteine is misincorporated into flagellin at a frequency of 6 x 10 -4 ,

probably as the result of a first position error in reading the arginine

codons CGU or CGC (1). I f codon use in the gene for flagellin is typical of

e. coli genes (2) the error frequency per codon may be less than 10 -4 (3).

The single CGU codon in =mRNA is misread at a frequency of 10 -3 (4).

The difference between these measurements likely involves codon context

effects (5,6,7). I f all CGU codons were misread at 10 -3 , then many 5 @&.

proteins should show charge heterogeneity--which they do not (8). Cysteine is

misincorporated for tryptophan (UGG), a third position error, at a frequency

of 3-4 x 10 -3 in HmRNA (4). We have shown that the asparagine codon AAU

is misread as lysine, a third position error, at a frequency of 2-5 x 10e3,

six times more frequently than AAC (9). We believe this high frequency may be

typical of AAU codons.

Abbreviations used: ppGpp, guanosine-3'-diphosphate-5'-diphosphate; pL, lambda promoter left.

0006-291X/84 $1.50

487 Copyright 0 1984 by Academic Press, Inc.

All rights of reproduction in any form reserved.


The above errors can be increased or decreased predictably by mutations

affecting ribosomes (1,4,9). They can also be increased by specific amino acid

starvations, i.e. asparagine starvation increases lysine incorporation. Amino

acid starvation increases codon misreading most dramatically in relaxed

(relA) mutants of 5. coli (see 10). Presumably in stringent (relA+)

strains the production of ppGpp during amino acid starvation prevents the

increase in codon misreading (11).

The existence of easily identifiable products of codon misreading,

proteins with altered isoelectric points, allows one to test the effects of

mutations and growth conditions on mistranslation. In this paper we show that

production of ppGpp has little, if any, direct effect on basal-level errors in

asparagine codon misreading.

MATERIALS AND METHODS

Strains. All strains used were derivatives of 5 coli K-12, except NC3 which is a derivative of B/r, and contained both the MS2 encoding plasmid pPLaACR26 and the ~I857 encoding plasmid pRK248At2 (12). These plasmids confer kanamycin and tetracycline resistance, respectively. The congenic strain set used was constructed in JKlOO (13), relevant markers are given in Table 1. The genotype of JKlSlB is Hfr, lacI, lacZ, lysA, SPOT and I-. --

Media. The media used was based on the morpholine propanesulfonic acid buffered medium of Neidhardt et al. (14) with 0.4% (wt/vol) glucose. This -- medium was supplemented with 1% Methionine Assay Medium (Difco) for labelling with [35S]-L-methionine or a supplement based on Methionine Assay Medium but, in addition, lacking proline, valine, leucine and isoleucine for starvation experiments. Tetracycline was added to 10 pg/ml and kanamycin to 50 pg/ml to maintain the plasmids.

Growth,. Cultures were grown at 30°C with rotary shaking. Growth was monitored by measuring absorbance at 420 nm. The induction of synthesis of MS2 proteins encoded under the the control of lambda pL on pPLaACR26 was accomplished by shifting cultures to 42'C to inactivate the lambda repressor encoded on pRK248At2. Cells were labelled by transferring 3 ml of culture to a tube containing 10 nCi of [35S]-methionine, and incubated at 42'C from 20 to 40 min post-induction. Cells were starved for isoleucine by the addition of valine to 500 rig/ml.

Two-dimensional gel electrophoresis. Extracts were prepared and samples separated by nonequilibrium pH gradient electrophoresis (15) as described (16). The radioactivity in a protein was determined by digestion and counting (16).

RESULTS

Quantitation of the Basic Satellite Spot. Coat protein mRNA from the

coliphage MS2 contains four AAU and six AAC codons (17). Basal-level

misreading of these codons results in production of a basic derivative of coat

488


A A BS N

Fig. 1. Autoradiograma of two-dimensional gels containigg coat protein m$de in plasmid bearing cells. Strain JKlOOB was grown at 30 C, shifted to 42 C and labeled with [35S]-methionine. Cells were harvested and gels run as described in MATERIALS AND METHODS. The gel is oriented with basic proteins to the left and larger proteins to the top. Panel A is an autoradiogram of the entire gel; panel B is a close-up of the relevant portion. Native coat protein (N) and the Basic Satellite (BS) are indicated.

protein which is readily observable on two-dimensional gels (9, and Fig. 1).

This satellite has been confirmed as a derivative of coat protein by peptide

mapping (9) and by immunoblotting with specific antibody (results not shown).

The availability of a plasmid containing this RNA as a DNA copy (12) allows

one to measure coat protein heterogeneity in a large number of strains. Table

1 shows the results obtained for a set of congenic strains. There are

variations from experiment to experiment by as much as two fold. As can be

seen, basal-level misreading is not affected by the state of the relA gene,

but is affected by mutations in ribosomal protein genes. Basal-level

489


Table 1. Heterogeneity of Coat Protein in Unstarved Cells

relA Strain'

cpm c allele ribosome b satellite total coat X satellite

JKlOOB + 16,800 442,000 3.8

JKlZOB + + 15,800 424,000 3.7

JK299B rpsE2123 11,800 263,000 4.5

JK305 rpsW 1,800 276,000 0.65

JK306 rpsL125 600 80,000 0.75

GH2 + rpsL125 2,000 229,000 0.87

a All strains were constructed in JKlOO or JKl20 by Pl transduction and transformation. In addition to the markers listed the strains are spoT

b mutants. c A + indicates wild-type riboeomal protein genes.

The cpm are recoverable counts of [35S]-methionine. Data are sums of as many as three Independent samples.

mistranslation in this system may be somewhat higher than observed in

MSZ-infected cells (18). However the latter utilized cells grown in a minimal

medium and at 37'C.

Using these plasmids a large number of other strains were analyzed,

including derivatives of many of the primary E. - coli K-12 lines (see 19) and

g. coli B/r. All strains with wild-type ribosomes synthesize coat protein

with a basic satellite which averages 2.5% of the total (results not shown),

indicating that the high level of basal-level of misreading observed is not

the result of a peculiar genetk background.

Starvation and Basal-Level Codon Misreading. Starvation for asparagine

increases the frequency of asparagine substitutions by over two orders of

magnitude in relA mutants. In relA+ strains a ten fold increase is seen (18).

This increase is easily measured by the increase in charge heterogeneity.

However, not all amino acid substitutions yield a charge change, i.e.

starvation for phenylalanine may greatly increase leucine for phenylalanine

substitution, but the resultant mistranslated protein will co-migrate with

native protein on the gel.

5. coli K-12 can be easily starved for isoleucine by the addition of

valine. Isoleucine starvation does not lead to an increase in protein charge

490


Table 2. Heterogeneity of Coat Protein in Isoleucine Starved Cells

Strain

JRlZOB

JRlCOB

JK151B

relA

+

+

X satellite unstarved starved

1.9 1.7

4.3 3.5

2.3 1.7

heterogeneity. In relA+ strains isoleucine starvation does lead to a great

increase in the level of ppGpp. In order to test the effects of ppGpp on

basal-level lysine for asparagine substitution, coat protein was induced by

shifting to 42'C and the cells then starved for isoleucine. Results from

typical experiments are shown in Table 2. Although we see a slight reduction

in heterogeneity during starvation, the reduction is within experimental error

and is not confined to relA+ strains.

DISCUSSION

If an mRNA contained 4 AAU codons then a misreading frequency of 5 X 10 -3

would yield a satellite containing 2% of the total protein. The four AAU's in

the coat protein are misread at about this frequency in all strains with

wild-type ribosomes. Most chromosomally encoded abundant proteins do not show

this charge heterogeneity because AAU is infrequently used (2,9).

As expected (1,18), the relA gene had no significant affect on basal-

level mistranslation in unstarved cells. However, our finding that the same

was true in starved cells was unexpected. During asparagine starvation JR120

shows much less misreading of asparagine codons than does .7X100 (13),

presumably because of the production of ppGpp in the former. However, during

isoleucine starvation (Table 2) or serine starvation (results not shown) the

relA+ allele had little if any effect on misreading of asparagine codons.

We also saw no difference in coat protein heterogeneity between relA and

rel$ strains after relief of amino acid starvation (results not shown).

Since the strains examined are &mutants , ppGpp should persist after

refeeding (20). Apparently then ppGpp does not enhance accuracy per se, at

491


least under these conditions. This has also been observed in an unstarved, in -

vitro protein synthesizing system (21). During amino acid starvation ppGpp

seems to specifically affect the misreading of cognate codons. It could do

this by differentially altering the amount (see 22) or the flow of uncharged

cognate tRNA.

It is conceivable that the shift to 42'C, which is required for pL

induction, altered the frequency and/or control of mistranslation. Work is now

in progress to test this possibility.

Acknowledgements. We would like to thank Barbara Bachmann for the many strains she sent and Jon Gallant for discussions and encouragement. This work was supported by grant GM25855 from the National Institutes of Health.

REFERENCES

1. Edelmann, P., and Gallant, J. (1977) Cell 10, 131-137. 2. Gouy, M. and Gautier, C. (1982) Nucleic Acids Res. 10, 7055-7074. 3. Silverman, M. and Simon, M. (1972) J. Bacterial. 112, 986-993. 4. Bouadloun, F., Donner, D. and Kurland, C.G. (1983) Eur. Mol. Biol.

2, 1351-1356. 5. Bossi, L. and Roth, J.R. (1980) Nature 286, 123-127. 6. Bossi, L. (1983) J. Mol. Biof. 164, 73-87. 7. Miller, J.H. and Albertini, A.M. (1983) J. Mol. Biol. 164, 59-71. a. Ellis, N. and Gallant, J. (1982) Mol. Gen. Genet. 188, 169-172. 9. Parker, J., Johnston, T.C., Borgia, P.T., Holtz, G., Remaut, E. and

Piers, W. (1983) J. Biol. Chem. 258, 10007-10012. 10. Gallant, J.A. (1979) Ann. Rev. Genet. 13, 393-415. 11. O'Farrell, P.H. (1978) Cell 14, 545-557. 12. Remaut, E., De Waele, P., Marmenout, A., Stanssens, P. and Fiers, W.

(1982) Eur. Mol. Biol. 1, 205-209. 13. Parker, J. and Friesen, J.D. (1980) Molec. Gen. Genet. 177, 439-445. 14. Neidhardt, F.C., Bloch, P.L. and Smith, D.F. (1974) J. Bacterial.

119, 736-747. 15. O'Farrell, P.Z., Goodman, H.M. and O'Farrell, P.H. (1977) Cell 12,

1133-1142. 16. Pedersen, S., Reeh, S.V., Parker, J. Watson, R-J., Friesen, J.D. and

Fiil, N.P. (1976) Mol. Gen. Genet. 144, 339-343. 17. Min Jou, W., Haegeman, G., Ysebaert, M. and Fiers, W. (1972) Nature

237, 82-88. 18. Parker, J., Johnston, T.C. and Borgia, P.T. (1980) Mol. Gen. Genet.

180, 275-281. 19. Bachmann, B.J. (1972) Bacterial. Rev. 36, 525-557. 20. Laffler, T. and Gallant, J. (1974) Cell 1, 27-30. 21. Wagner, E.G.H. and Kurland, C.G. (1980) 180, 139-145. 22. Wagner, E.G.H., Ehrenberg, M. and Kurland, C.G. (1982) Mol. Gen.

Genet. 185, 269-274.

492

Documents

Control of basal-level codon misreading in Escherichia coli