Vol. 168, No. 3

Control of metF Gene Expression in Maxicell Preparations ofEscherichia coli K-12: Reversible Action of the metJ Protein and

Effect of Vitamin B12MARK R. EMMETT AND JAMES R. JOHNSON*

Department of Microbiology, Miami University, Oxford, Ohio 45056

Received 31 January 1986/Accepted 8 September 1986

Expression of methionine regulon elements was controlled by the metJ protein gpMetJ. A maxicell systemwith cloned copies of the metF transcription unit allowed reversible action of gpMetJ. Expression of the metFtranscription 4nit in maxicells was reduced by exogenous vitamin B12 at concentrations of 0.5 nM or greater.

The methionine (met) regulon of Escherichia coli K-12consists of 11 genetic elements (1, 4, 8, 27). Eight genes havebeen isolated by recombinant DNA techniques, and thenucleotide sequences of six (metL, A, B, F, K, and J) havebeen reported (4, 6, 22, 23, 29, 30, 40). The metJBL genes ofSalmonella typhimurium LT2 were recently isolated, and themetJ nucleotide sequence was determined (36, 37). The metJprotein of both organisms, gpMetJ, regulates expression ofall met regulon elements (for reviews, see references 4, 8,27). Consequently, gpMetJ autoregulates its own expression

(15, 29, 36, 38). Genetic studies suggest that an effectormolecule controlling in vivo gpMetJ function is S-adenosylmethionine (Ado-met) (for reviews, see references 4, 8, 27).Recently, Shoeman et al. (32) demonstrated that purifiedgpMetJ and Ado-met blocked metF expression at the level ofRNA synthesis in an uncoupled DNA-directed protein syn-thesizing system. In the absence of Ado-met, high concen-

trations of gpMetJ also reduce in vitro metF expression. Aputative gpMetJ recognition sequence (the met box) hasbeen identified within the promoter region of each sequencedmet regulon transcription unit (6, 23, 29, 30). Purified gpMetJbinds a portion of the met box of the bipolar metJ-metBLelements, and Ado-met enhances its affinity for this se-

quence. The complete function(s) of the met box remainsundefined (33).A simple system with the maxicell procedure of Sancar et

al. (31) allows detection of gpMetJ activity in vivo. Appro-priate recA-carrying E. coli K-12 strains containing a wild-type or mutant metJ allele and copies of a met regulontranscription unit inserted in a ColEl-derived plasmid vectorare constructed. In this system, exogenous methionine pre-sumably serves as the source of in vivo effective concentra-tions of Ado-met. This paper reports the results of initialcharacterization of the ability of the system to regulateexpression of a cloned copy of the metF transcription unitwhich codes for the enzyme N5,N10-methylenetetrahydro-folate reductase (EC 1.1.1.68).

Standard media and culture conditions have been de-scribed (12). Minimal medium plus Casamino Acids (DifcoLaboratories) (DM+caa) contains 0.5% dextrose and 0.2%vitamin assay Casamino Acids (Difco) in Davis Mingioli salts(5). Chemicals were of reagent grade. A radioisotopicallylabeled L-amino acid mixture ([U-14C]methionine free, 0.1mCi/ml, approximately 284 mCi/mM) was purchased fromNew England Nuclear Corp. Recombinant plasmid con-

* Corresponding author.

struction was accomplished as described in reference 7. E.coli strains were transformed by the calcium shock proce-dure of Kushner (17). Plasmid identity was confirmed byrestriction endonuclease digestion and gel electrophoreticanalysis (10, 35). Purified plasmid samples were preparedfrom cell extracts by cesium chloride-ethidium bromideequilibrium density gradient centrifugation (11). Bacterio-phage transductions were performed as reported by Treat etal. (35). The recA58 allele of strain MS198 was crossed intorequired genetic backgrounds by conjugation based on thestrategy described in reference 21. Appropriate recA plas-mid-bearing strains were grown at 32°C to an optical densityat 550 nm of 0.41 in DM+caa medium plus the necessaryantibiotics. The cultures were harvested at room tempera-ture by low-speed centrifugation at 10,000 x g, washedtwice, and resuspended in minimal medium plus antibiotics(no Casamino acids). A 15-W germicidal lamp (modelG1ST8; General EFlectric Co.) was used to irradiate 10.5-mivolumes of these cultures with 90 J of UJV light per m2.Irradiated cultures were transferred to Raysorb flasks(American Scientific Products) or test tubes and maintainedat 320C. Samples were incubated in the presence or absenceof methionine or other effector molecules and receivedD-cycloserine (100 ,ug/ml, final concentration) at 30 minpostirradiation (31). Each sample was concentrated (10-fold)just before addition of 14C-labeled amino acids (6 ,XCi/ml). Inall cases, the radioisotope was incorporated for 30 min at32°C. Each maxicell sample was harvested, processed, re-

solved by sodium dodecyl sulfate-polyacrylamide gel elec-trophoresis (SDS-PAGE) and analyzed by Coomassie blue(18) or silver (39) staining plus autoradiography of theresulting electrophoretogram. Samples were electropho-resed on 18% polyacrylamide slab gels (16 by 16 cm by 1.5mm) by using the buffers and protocol of Laemmli (18).Autoradiographic techniques were adapted from those ofLaskey and Mills (19). In most cases, Autofluor (NationalDiagnostics, Inc.) was used to enhance the exposure of thegels. Autoradiograms were obtained by using Kodak X-Omat RP or Kodak X-Omat AR films (Eastman Kodak Co.).The plasmid pEJ3-lB was constructed by inserting a

6.7-kilobase-pair BamHI fragment from X dmet-128::TnS or

Tn3-1 (35) into the single BamHI site of pBR322 (3, 34). Thefragment carries metF and the left half of TnS including neo,the gene coding for neomycin phosphotransferase (gpNeo)(2) plus 1.5 kilobase pairs of bacterial DNA which liesclockwise of metF on the E. coli K-12 chromosome. Thefollowing strains were constructed: (i) JJ122R[EJ3-lB]

1491

JOURNAL OF BACTERIOLOGY, Dec. 1986, p. 1491-14940021-9193/86/121491-04$02.00/0Copyright © 1986, American Society for Microbiology

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/j

b on

23

Febr

uary

202

2 by

218

.101

.136

.145

.

1492 NOTES

1234 12 34

A B~~~~

FIG. 1. SDS-PAGE electrophoretogram (A) and autoradiogram(B) depicting the reversal of the gpMetJ function through manipu-lation of the exogenous methionine concentration before, '4C-aminoacid labeling of samples of a JJ122R[EJ3-lB] maxicell preparation(see text for procedure). Lanes: 1, JJ122R[EJ3-lB], no methionine(--;2, JJ122R[EJ3-lB], no methionine for 3.5 hfollowed by 200

p.M exogenous methionine for 1 h (-1+); 3, JJ122R[EJ3-lB]', 10 p.Mexogenous methionine for 3.5 h followed by no methionine for 1 h(+-;4, JJ122R[EJ3-lB], 10 p.M exogenous methionine for, 3.5 h

followed by 200 p.M exogenous methionine for 1 h (++.The filledarrow indicates the position of the gpMetF band, and the unfilledarrow indicates the position of the gpNeo band. Protein standardsare to the left of lane 1, panel A. Top to bottom: phosphorylase B,bovine serum albumin, ovalbumin, carbanic anhydrase, soybeantrypsin inhibitor, and lysozyme (egg white).

[recA58 metJ+ metF(pEJ3-1B)]; (ii) JJ127R[EJ3-lB] [recA58metJJ8S(Am) metF(pEJ3-1B)]; and (iii) JJ127RFP[EJ3-lB][recA58 metJJ8S(Am) metW 4)80 psu3+ lysogen(pEJ3-1B)].The JJ127R[EJ3-lB] strain was lysogenized with phage 4)80psu3+ to yield isolate JJ127RFP[EJ3-lB]. The metJJ85(Am)allele resulted from a single-point mutation at the secondposition (G C = A T) of the fourth codon (Trp) of themetJ gene (20, 29). The lysogen, JJ127RFP[EJ3-lB], has aMetJW phenotype due to the suppressor action of the supFtRNA (tRNA Tyr) encoded by the 4)80 prophage (26, 28),whereas the nonlysogen, JJ127R[EJ3-lB], has a MetjY phe-notype. The strains were cultured, harvested, and UVirradiated as described above. Samples were incubated inthe presence or absence of 200 p.M methionine for 4 h andlabeled with 14-amino acids for 30 min. The Coomassieblue-stained SDS-PAGE profiles of the JJ122R(EJ3-lB) sam-ples revealed that a protein band with a molecular weightcomparable to that of gpMetF (N5,N10-methylenetetra-hydrofolate reductase) (32,500, molecular weight; 30) waspresent only in the sample incubated in the absence ofmethionine (data not shown). This result confirmed theability of 200 p.M exogenous methionin'e to induce regulationof metW expression. An autoradiogram of the electrophoreto-gram indicates that exogenous methionine reduced incorpo-ration of the isotopically labeled amino acids only intogpMetF and not into gpNeo and gpBla (data not shown).Only in the presence of methionine did the 4)80 psu3+lysogen, JJ127RFP[EJ3-lB], regulate metW expression. TheCoomassie blue-stained protein profiles of the nonlysogensamples indicate that an absence of functional gpMetJ re-sulted in significant overproduction of gpMetF by bothsamples of JJ127R[EJ3-lB] (data not shown).

To assess gpMetJ action, an irradiated sample of theJJ122R[EJ3-lB] strain was divided into two portions, +methionine (10 ,.M exogenous methionine) and - methio-nine, and incubated for 3.5 h at 320C. Each sample waswashed by centrifugation-resuspension and subdivided. Thefollowing nomenclature is used: -/-, preincubation minusmethionine, labeling minus methionine; -/+, preincubationminus methionine, labeling in presence of methionine (200jig/ml); +/-, preincubation plus methionine, labeling minusmethionine; +/+ preincubation plus methionine, labeling inpresence of methionine (200 ,.g/ml). The four samples wereincubated for 30 min and labeled with '4C-amino acids. Theresulting electrophoretogram and autoradiogram are shownin Fig. 1A and B, respectively. The intensity of the gpMetFband on the autoradiogram indicates that metF expressionwas constitutive in the -/- sample and abolished in the -/+sample (Fig. 1B, lanes 1 and 2). The opposite pattern ofexpression was observed in the +/- and +/+ samples (Fig.1B, lanes 3 and 4). The silver-stained electrophoretogramdemonstrates that the -/- and -/+ samples containedequivalent amounts of gpMetF, whereas the +/- and +/+samples contained reduced amounts of the protein. Thisresult is consistent since both the -/- and -/+ samplessynthesized gpMetF for 3.5 h before the described manipu-lations. Expression of the plasmid-borne neo gene was notsignificantly affected by the presence or absence of methio-nine (Fig. 1A and B).

In E. coli K-12, metF expression is also regulated byvitamin B12 and gpMetH (the B12-dependent methyl trans-ferase) (16, 24, 25). To test the ability of the maxicell systemto reproduce this control, samples of an irradiated culture ofJJ122R[EJ3-lB] were incubated with vitamin B12, labeledwith 14C-amino acids, and processed for SDS-PAGE analy-sis. As shown by the resulting electrophoretogram andautoradiogram (Fig. 2A and B, lanes 5 to 8), exogenous

1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8

_ . ._ _ ..

low

FIG. 2. SDS-PAGE electrophoretogram (A) and autoradiogram(B) of extracts of "4C-amino acid-labeled maxicell preparations ofJJ122R[EJ3-lB] incubated in the presence or absence of exogenousvitamin B12 (see text for procedure). Lanes: 1, protein standards(same as for Fig. 1); 2, JJ122R(EJ3-1B), no vitamin B12; 3,JJ122R[EJ3-lB], 0.005 nM vitamin B12; 4, JJ122R[EJ3-lB], 0.050nM vitamin B12; 5, JJ122R[EJ3-lB], 0.500 nM vitamin B12; 6,JJ122R[EJ3-lB], 5 nM vitamin B12; 7, JJ122R[EJ3-lB], 50 nMvitamin B12; 8, JJ122R[EJ3-lB], 500 nM vitamin B12. The filledarrow indicates the position of the gpMetF band, and the unfilledarrow indicates the position of the gpNeo band.

J. BACTERIOL.

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/j

b on

23

Febr

uary

202

2 by

218

.101

.136

.145

.

VOL. 168, 1986

vitamin B12 concentrations between 0.5 and 500 nM weresimilar in their inhibition of metF expression but exerted lesseffect on neo expression. Between 0.05 and 0.5 nM,exogenous vitamin B12 inhibition of metF expression waslost (Fig. 2A and B, lanes 4 and 5).The lanes of a second SDS-PAGE electrophoretogram of

the vitamin B12 experiment were cut into 1-mm slices andprocessed for liquid scintillation spectrophotometry (9; datanot shown). No significant effect was detected in the profilesof 14C-amino acid incorporation for the samples containing0.005 and 0.05 nM vitamin B12, but for the sample containing0.5 nM vitamin B12, incorporation into gpMetF was reducedby 50%. At vitamin B12 concentrations of 5, 50, and 500 nM,a generalized inhibitory effect on gene expression was de-tected due to a reduction in 14C-amino acid incorporationinto gpNeo. When the samples were corrected for thisreduction, each exhibited approximately 50% inhibition ofmetF expression. Mulligan et al. (25) reported that 74 nMvitamin B12 (0.1 ,ug/ml) reduced expression of a metF pro-moter-lacZ fusion to 33% of the constitutive level. Perhaps aportion of this inhibition was due to the relatively highexogenous vitamin B12 concentration.The results indicate that the described maxicell system

provides enough methionine to sustain protein synthesis.However, the system depends on exogenous methionine toprovide a sufficient concentration of an effector required forthe function of the met regulon aporepressor, gpMetJ (Fig.2A and B). Ado-met is the most probable candidate for thiseffector molecule (32, 33). Therefore, the predicted de-creases in the Ado-met and methionine pools, which resultfrom the removal of the exogenous methionine supplement(Fig. 1), would cause the observed induction of metF ex-pression. In the 10 ,LM concentration range, exogenousmethionine is transported into the cell by the high-affinitymethionine transport system (Km = 0.13 p,M; 13). Thelow-affinity system, which has a Km of 20 puM, would beeffective only at higher (200 ,uM) methionine concentrations(13, 14).These results (-/+ methionine experiment; Fig. 1) dem-

onstrate that the action of gpMetJ is reversible in a shorttime period (maximum of 30 min). Because the action isreversible, the function of gpMetJ is probably stoichiometricrather than catalytic. This conclusion is consistent with thedata of Shoeman et al. (32) and Smith et al. (33) whichdescribe the interaction of gpMetJ with met regulon DNAsequences in cell-free systems. Neither report demonstratesreversible functional interaction of gpMetJ with the de-scribed DNA templates.

Recently Kirby et al. (15) identified three E. coli metJpromoters and demonstrated that two are under autoregula-tory control of gpMetJ. Presumably, during the cellular-growth phase of maxicell strains bearing plasmid pEJ3-1B,expression of the single metJ allele is adjusted by thisautoregulatory mechanism. Consequently, the amount ofgpMetJ present at the time of UV irradiation remains suffi-cient to control multiple copies of the plasmid-borne metFtranscription unit for at least 6 to 8 h postirradiation.Experiments are in progress to determine the metF mRNA

levels and the intracellular concentrations of methionineand Ado-met associated with the regulatory events de-scribed in this paper. The respective functional roles ofgpMetH and gpMetJ and their cognate effector molecules arebeing evaluated.

Portions of this project were completed by M.R.E. at Texas A &M University as partial fulfillment of the M.S. degree. Helpfuldiscussions with D. R. Lueking and R. C. Burghardt are acknowl-

edged. W. Rickoll and D. Storts provided initial protocols for silverstaining SDS-polyacrylamide gels.

This project was supported by Public Health Service researchgrants GM 28214 and 35065 to J.R.J. from the National Institutesof Health.

LITERATURE CITED1. Bachmann, B. J. 1983. Linkage map of Escherichia coli K-12,

edition 7. Microbiol. Rev. 47:180-230.2. Beck, E., G. Ludwig, E. A. Auerswald, B. Reiss, and H. Schaller.

1982. Nucleotide sequence and exact localization of theneomycin phosphotransferase gene from transposon Tn5. Gene19:327-336.

3. Bolivar, F., R. L. Rodriguez, P. J. Greene, M. C. Betlach, H. L.Heyneker, and H. W. Boyer. 1977. Construction and character-ization of new cloning vehicles. II. A multipurpose cloningsystem. Gene 2:95-113.

4. Cohen, G. N. 1983. The common pathway to lysine, methionineand threonine, p. 147-169. In K. M. Herrman and R. L.Somerville (ed.), Amino acids biosynthesis and genetic regula-tion. Addison-Wesley Publishing Co., Inc., Reading, Mass.

5. Davis, B. D., and E. S. Mingioli. 1950. Mutants of Escherichiacoli requiring methionine or vitamin B12. J. Bacteriol. 60:17-28.

6. Duchange, N., M. M. Zakin, P. Ferrara, I. Saint-Girons, I. Park,S. V. Tran, M. Py, and G. N. Cohen. 1983. Structure of themetJBLF cluster in Escherichia coli K-12: sequence of the metBstructural gene and of the 5' and 3' flanking regions of the metBLoperon. J. Biol. Chem. 258:14868-14871.

7. Dugaiczyk, A., H. W. Boyer, and H. M. Goodman. 1975.Ligation of EcoRI endonuclease generated fragments into linearand circular structures. J. Mol. Biol. 96:171-184.

8. Flavin, M. 1975. Methionine biosynthesis, p. 457-503. In D. M.Greenberg (ed.), Metabolic pathways, vol. 7. Academic Press,Inc., New York.

9. Fraley, R. T., D. R. Lueking, and S. Kaplan. 1977. Intracyto-plasmic membrane synthesis in asynchronous cell populationsof Rhodopseudomonas sphaeroides. J. Biol. Chem. 253:458-466.

10. Holmes, D. S., and M. Quigley. 1981. A rapid boiling method forpreparation of bacterial plasmids. Anal. Biochem. 114:193-197.

11. Humphreys, G. O., G. W. Wilshaw, and E. S. Anderson. 1975. Asimple method for the preparation of bacterial plasmids.Biochim. Biophys. Acta 383:457-463.

12. Johnson, J. R., R. C. Greene, and J. H. Krueger. 1977. Isolationand characterization of specialized lambda transducingbacteriophage carrying the metBJF methionine gene cluster. J.Bacteriol. 131:795-800.

13. Kadner, R. J. 1977. Transport and utilization of D-methionineand other methionine sources in Escherichia coli. J. Bacteriol.129:207-216.

14. Kadner, R. J., and W. J. Watson. 1974. Methionine transport inEscherichia coli: physiological and genetic evidence for twouptake systems. J. Bacteriol. 119:401-409.

15. Kirby, T. W., B. R. Hindenach, and R. C. Greene. 1986.Regulation of in vivo transcription of the Escherichia coli K-12metJBLF gene cluster. J. Bacteriol. 165:671-677.

16. Kung, H., C. Spears, R. C. Greene, and H. Weissbach. 1972.Regulation of the terminal reactions in methionine biosynthesisby vitamin B12 and methionine. Arch. Biochem. Biophys.150:23-31.

17. Kushner, S. A. 1978. An improved method for transformation ofEscherichia coli with col El derived plasmids, p. 17-23. InH. W. Boyer and S. Nicosia (ed.) Genetic engineering.Elsevier/North-Holland Publishing Co., London.

18. Laemmli, U. K. 1970. Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature (London)227:680-685.

19. Laskey, R. A., and A. D. Mills. 1975. Quantitative film detectionof 3H and "C in polyacrylamide gels by fluorography. Eur. J.Biochem. 56:335-341.

20. Liljestrand-Golden, C. A., and J. R. Johnson. 1984. Physicalorganization of the metJB component of the Escherichia coliK-12 metJBLF gene cluster. J. Bacteriol. 157:413-419.

NOTES 1493

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/j

b on

23

Febr

uary

202

2 by

218

.101

.136

.145

.

1494 NOTES

21. Lorence, M. C., and C. S. Rupert. 1983. Convenient construc-tion of recA deletion derivatives of Escherichia coli. J. Bacte-riol. 156:458-459.

22. Markham, G. D., J. DeParasis, and J. Gatmaitan. 1984. Thesequence of metK, the structural gene for S-adenosyl methio-nine synthetase in Escherichia coli. J. Biol. Chem. 259:14505-14507.

23. Michaeli, S., M. Mevarech, and E. Z. Ron. 1984. Regulatoryregion of the metA gene of Escherichia coli K-12. J. Bacteriol.160:1158-1162.

24. Milner, L., C. Whitfield, and H. Weisbach. 1969. Effect ofL-methionine and vitamin B12 on methionine biosynthesis inEscherichia coli. Arch. Biochem. Biophys. 133:413-419.

25. Mulligan, J. T., W. Margolin, J. H. Krueger, and G. C. Walker.1982. Mutations affecting regulation of methionine biosyntheticgenes isolated by use of met-lac fusions. J. Bacteriol. 151:609-619.

26. Rossi, J. J., and A. Landy. 1979. Structure and organization ofthe two tRNATYr gene clusters on the E. coli chromosome. Cell16:523-534.

27. Rowbury, R. J. 1983. Methionine biosynthesis and its regula-tion, p. 191-211. In K. M. Herrman and R. L. Somerville (ed.),Amino acids biosynthesis and genetic regulation. Addison-Wesley Publishing Co., Inc., Reading, Mass.

28. Russel, R. L., J. N. Abelson, A. Landy, M. L. Gelfter, S.Brenner, and J. D. Smith. 1970. Duplicate genes for tyrosinetransfer RNA in Escherichia coli. J. Mol. Biol. 47:1-13.

29. Saint-Girons, I., N. Duchange, G. N. Cohen, and M. M. Zakin.1984. Structure and autoregulation of the metJ regulatory genein Escherichia coli. J. Biol. Chem. 259:14282-14285.

30. Saint-Girons, I., N. Duchange, M. Zakin, I. Park, D. Margarita,P. Ferrara, and G. Cohen. 1983. Nucleotide sequence of metF,the E. coli structural gene of 5-10 methylene tetrahydrofolatereductase and of its control region. Nucleic Acids Res. 11:6723-6732.

31. Sancar, A., A. M. Hack, and W. D. Rupp. 1979. Simple methodfor identification of plasmid-coded proteins. J. Bacteriol. 137:692-693.

32. Shoeman, R., B. Redfield, T. Coleman, R. C. Greene, A. A.Smith, N. Brot, and H. Weisbach. 1985. Regulation of methio-nine synthesis in Escherichia coli: effect of metJ gene productand S-adenosyl methionine on the expression of the metF gene.Proc. Natl. Acad. Sci. USA 82:3601-3605.

33. Smith, A. A., R. C. Greene, T. W. Kirby, and B. R. Hindenach.1984. Isolation and characterization of the product of themethionine-regulatory gene metJ of Escherichia coli K-12.Proc. Natl. Acad. Sci. USA 82:6104-6108.

34. Sutcliffe, J. G. 1978. Complete nucleotide sequence of theEscherichia coli plasmid pBR322. Cold Spring Harbor Symp.Quant. Biol. 48:77-90.

35. Treat, M. L., M. L. Weaver, M. R. Emmett, and J. R. Johnson.1984. Mutagenesis of the metJBLF gene cluster with transposonTn5: localization of the metF transcription unit. Mol. Gen.Genet. 193:370-375.

36. Urbanowski, M. L., and G. V. Stauffer. 1985. Nucleotidesequence and biochemical characterization of the metJ genefrom Salmonella typhimurium LT 2. Nucleic Acids Res. 13:673-685.

37. Urbanowski, M. L., and G. V. Stauffer. 1985. Cloning and initialcharacterization of the metJ and metB genes from Salmonellatyphimurium LT2. Gene 35:187-197.

38. Urbanowski, M. L., and G. V. Stauffer. 1986. Autoregulation bytandem promoters of the Salmonella typhimurium LT2 metJgene. J. Bacteriol. 165:740-745.

39. Wray, W., T. Boulikas, V. P. Wray, and R. Hancock. 1981.Silver staining of proteins in polyacrylamide gels. Anal.Biochem. 118:197-203.

40. Zakin, M. M., N. Duchange, P. Ferrara, and G. N. Cohen. 1983.Nucleotide sequence of the metL gene of Escherichia coli. J.Biol. Chem. 258:3028-3031.

J. BACTERIOL.

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/j

b on

23

Febr

uary

202

2 by

218

.101

.136

.145

.