Vol. 176, No. 5JOURNAL OF BACFERIOLOGY, Mar. 1994, p. 1394-14040021-9193/94/$04.00+0Copyright X 1994, American Society for Microbiology

Escherichia coli-Salmonella typhimurium Hybrid nusA Genes:Identification of a Short Motif Required for Action of

the N Transcription Antitermination ProteinMARK G. CRAVEN,t ANDREW E. GRANSTON, ALAN T. SCHAUER,t CHUANHAI ZHENG,

TODD A. GRAY, AND DAVID I. FRIEDMAN*

Department of Microbiology and Immunology, University of MichiganMedical School, Ann Arbor, Michigan 48109

Received 20 July 1993/Accepted 4 December 1993

The Escherichia coli nusA gene, nusA,E, encodes an essential protein that influences transcription elongation.Derivatives of E. coli in which the SalmoneUla typhimurium nusA gene, nusAs, has replaced nusA$4 are viable.Thus, NusAst can substitute for NusAEz in supporting essential bacterial activities. However, hybrid E. colistrains with the nus4st substitution do not effectively support transcription antitermination mediated by theNgene product of phage X. We report the DNA sequence of nusAst, showing that the derived amino acid sequence

is 95% identical to the derived amino acid sequence of nus4,c. The alignment of the amino acid sequences

reveals scattered single amino acid differences and one region of significant heterogeneity. In this region, called449, NusAF, has four amino acids and NusAst has nine amino acids. Functional studies of hybrid ns&4 genes,constructed from nusAEc and nus4st, show that the 449 region of the NusAEc protein is important for XN-mediated transcription antitermination. A hybrid that has a substitution of the four E. coli codons for thenine S. typhimurium codons, but is otherwise nusAs, supports the action of the N antitermination protein. The449 region and, presumably, adjacent sequences appear to compose a functional domain of NusAE importantfor the action of the N transcription antitermination protein of phage X.

The nusA gene of Escherichia coli encodes an essential54,400-Mr protein that functions in transcription elongation(reviewed in references 20, 29, and 83) and termination (16, 37,49, 75, 81). Consistent with these roles, NusA has been shownto influence pausing ofRNA polymerase (16, 17, 47, 50, 75, 76,85). The demonstration of an association between NusA andRNA polymerase (31, 35, 43, 46, 74) suggests that NusA mayinfluence transcription by a direct interaction with polymerase.NusA was identified through its role in the regulated expres-

sion of phage X genes (reviewed in references 13, 20, 23, and67). Early A transcription initiating at promoters PL and PRpartially terminates at terminators tLl (15, 70) and tRl (10),respectively. The escaping transcripts in the latter case termi-nate completely at a collection of terminators in the nin region(8, 9, 11, 48, 52) (Fig. 1). The N gene product, encoded in thePL operon, acts with host Nus proteins to modify RNApolymerase into a termination-resistant form at nut sites (1, 14,32, 53-55, 64, 69). Mutations affecting N action have identifiedthe E. coli nus genes (reviewed in references 20, 22, 26, and67). For example, the nusAl mutation (19), a missense muta-tion resulting in a single base change (68), causes a failure inthe support of N-mediated antitermination (14, 24, 42, 82). Nprotein binds to NusA (34), and in vitro transcription studieshave confirmed that Nus proteins are required for N action(14, 42, 54, 82). When properly modified by N and Nusproteins, RNA polymerase is resistant to many terminationsignals and thus proceeds through downstream terminators. Inthe case ofPR, N-modified transcription passes through the nin

* Corresponding author. Phone: (313) 763-3142. Fax: (313) 764-3562. Electronic mail address: David.Friedman@um.cc.umich.edu.

t Present address: Roche Institute of Molecular Biology, Nutley, NJ07110-1199.

t Present address: Department of Microbiology, University ofTexas, Austin, TX 78712-1095.

terminators and proceeds into the Q gene. The Q geneproduct, which regulates late gene expression, is also anantitermination factor that requires NusA (67).

Isolation of nonsense (80) as well as thermo- and cold-sensitive (63, 73) mutations in nusA demonstrates that NusAprovides an essential bacterial function. Although the essentialrole(s) for NusA in cellular growth has not been identified,mutations in nusA reduce the effectiveness of some transcrip-tion terminators (62, 81).

In addition to RNA polymerase and N, NusA also has beenshown to bind the Rho termination protein (74) and RNA(79). The facts that it binds a number of factors and is arelatively large protein (36, 44, 68) suggest the possibility thatNusA has multiple functional domains. There must be func-tionally important regions of NusA located in the 5' two-thirdsof the protein, since a plasmid-based nusA gene with its3'-terminal one-third deleted (nusAA324) complements nusAmutations which fail to support either X N action or bacterialgrowth at low temperatures (73). However, the plasmid-basednusAA324 gene fails to complement the nusAII(Ts) mutation.Thus, the product of the nusAA&324 gene is unable to supportbacterial growth at high temperatures (61).

Experiments with a hybrid bacterium, whose genome isderived primarily from E. coli, provide additional evidencesupporting a role for NusA in X N action. In this hybrid (2), asmall exchange of DNA has resulted in the replacement ofnusAE,, the nusA gene of E. coli, with nusAst, the nusA genefrom the closely related enteric bacterium Salmonella typhi-murium (66, 72). In addition to the nusA gene, some additionaluncharacterized genetic information 5' of nusA has also beenreplaced. The viability of the hybrid bacterium, a priori,demonstrates that NusAst can substitute for NusAEc in sup-porting cellular processes in an E. coli genetic background.However, NusAst fails to support X N action (2, 21, 73).The failure of nusAst to support X N action, under condi-

1394

on October 26, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

HYBRID nusA GENES 1395

att int

ti

dll N

tLl nutLPL

d cro

nin5

cllOP ;I

nutR tRI

Qa I 4

42 43 %4 qut tRI

-N, nutor Nus

I

i I--

+N, nutand Nus

FIG. 1. Map of early region of the A genome. (Top line) Shown above are the locations of relevant genes as well as positions of the aut site andthe nin region, and shown below are the relevant transcription signals, including promoters (p), terminators (t), and antitermination signals (nutand qut). (Middle line) Patterns of transcription from PL and PR in the absence of N. (Bottom line) Patterns of transcription from PR and PLfollowing N modification of RNA polymerase.

tions in which nusAEc does, suggested that there are structuraldifferences between the two proteins that should be reflectedin differences in their primary amino acid sequences. Wedetermined the DNA sequence of nusA, to compare itsderived amino acid sequence with that derived from the DNAsequence of nusA,E (12, 45). This comparison, which revealedonly small differences between the two proteins, provided theinformation enabling us to construct a series of hybrid nusAgenes that have various amounts of genetic information fromE. coli and S. typhimurium. On the basis of studies with thesehybrid genes, we have identified a region of the nusA gene thatis required for X N action.

MATERIALS AND METHODS

Strains. Bacteria and phages are listed in Tables 1 and 2,respectively.

Media. All media have been described previously (56, 57).Bacterial strains with plasmids were cultivated in the presenceof the appropriate antibiotic, either ampicillin at 30 ,ug/ml ortetracycline at 15 ,ug/ml.

Plasmids. Plasmid manipulations were essentially as de-scribed by Sambrook et al. (71). Table 3 lists the plasmids usedin this study as well as their sources.

EOP. Methods for determining efficiency of plating (EOP)and for the qualitative assessment of phage growth have beendescribed previously (3, 59).

Genetic techniques. (i) Transfer of the nusAhy3O2 hybridgene from pNAX302 to the E. coli chromosome by homologousrecombination. A rec+ nusAlO(Cs) strain, K1456, containingpNAX302 was passed through four rounds of growth in LBbroth at 40°C in the absence of ampicillin. Colonies that had achromosomal nusAhy302 gene were selected by growth at 30°Con LB plates in the presence of XcI- and Xc -h80; a colonyformed by such a recombinant would be expected to be coldresistant and unable to support growth of X. Plasmid-freebacteria were identified by screening for ampicillin sensitivitywith Amp Screen (Bethesda Research Laboratories). P1 trans-duction (77) was used to transfer the nusAhy3O2 allele from acolony with the appropriate phenotype to K37, yielding strainK4092.

(ii) Creation of the chromosomal nusAhy3O2-449 allele.Plasmid pSES28, a derivative of pUC18 with an 840-bp frag-ment composed of an internal portion of nusAs, in which thenucleotides encoding the four amino acids of the nus4EC 449sequence have replaced the nucleotides encoding the nineamino acids of the nusAst 449 sequence, was transformed intostrain K6600. K6600 has the nusAhy3O2 gene and a defective

TABLE 1. Bacterial strains

Straina Relevant genotype Parent Source and/or reference

K37 Wild type This laboratory (19)K95 nusAI K37 This laboratory (19)K1102 nusAhyllO2 L. S. Baron (21)K1456 argG::Tn5 K37 This laboratory (19, 73)K1914 argG::Tn5 nusAlO(Cs) K1456 This laboratoryK2415 recA nusAlO(Cs) K1914 L. BaronOR1150 XcI857pO-IS2-gal S. Adhya (65)K4012 nusAll4(Ts) K37 68K4092 argG::Tn5 nusAhy3O2 K37 This paper

K6600 argG::Tn5 nusAhy3O2-449 XcI857pR-IS2-gal This paperK6906 argG::TnS nus4hy3O2-449 K37 This paperK7346 nusA134(Am) recA::TnlO Y. Nakamura (80)LT2 Wild type G. Jones

a All strains are E. coli except LT2, which is S. typhimurium.

'A~ ~~I

s 0I 0xs a* a

VOL. 176, 1994

I

on October 26, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

1396 CRAVEN ET AL.

TABLE 2. Phages

Phage Relevant genotype Source or

and/or phenotype reference

AcI857 clts NIHa collectionXcI60 cI- NIH collectionAcI857bio256sex clts, altered N NIH collectionXcI -h80 cI-, adsorption specificity NIH collection

of 48OXninS N independent NIH collectionM13mp8 NAb 84M13mp9 NA 84Pi NA NIH collection

a NIH, National Institutes of Health.b NA, not applicable.

XcI857 prophage fused to the bacterial gal operon so that thegal operon is expressed from the 'APR promoter. The presenceof intervening termination signals between PR and gal as wellas expression of X repressor from a cI gene with the cI857tsallele results in expression of the gal operon being dependentupon both a functional phage N antitermination system to readthrough these terminators and growth at a high temperature toremove repression. Since the nusAhy302 allele fails to supportN-mediated transcription antitermination, K6600 is Gal - athigh temperatures. Gal' segregants were isolated, and platingof appropriate X derivatives identified those with the samedistinctive phenotype as the nusAhy449 gene on a plasmid.One derivative was chosen for further study. The putativenusAhy3O2-449 allele was transduced to K37, creating strainK6906, and DNA sequencing of a PCR-amplified fragmentconfirmed its structure.

(iii) Isolation of the nusAst gene. A X1059-Salmonella clonebank, constructed by R. Mauer and kindly provided by P.Youderian, was used to isolate the nusAs, gene according to amethod suggested by P. Youderian. The underlying assump-tion was that the arrangement of the S. typhimurium genome inthe nusA region would be the same as that found in the E. coligenome; i.e., nusA would be adjacent to argG. A culture of E.coli K-12 strain K1456 (K37 argG::TnS) was mixed withdifferent dilutions of the phage bank, and the infected bacteriawere plated in 2.5 ml of top agar with M9 minimal glucosemedium supplemented with 0.1 ml of LB broth on M9 minimal

TABLE 3. Plasmids

Plasmid Relevant marker(s) Parent Source orreference

pBR322 NA" NA 5pACYC184 NA NA 7pUC19 NA NA 84pWR324 nusAA324 pBR322 73pNAG15 nusAEc pBR322 This workpNAG2010 nusAEc pBR322 This workpNAS1000 nusAs, pACYC184 This workpNAS2001 nusA s pBR322 This workpN2/EuS nusAA50L-nusA,s pBR322 This workpNAX2 nusAhy2 pBR322 This workpNAX3 nusAhy3 pBR322 This workpNAX302 nusAhy3O2 pBR322 This workpSES28 840-bp fragment of nusA,s pUC18 This work

with 449 region ofnusAEc

pNAX449 nusAhy449 pBR322 This work

a NA, not applicable.

glucose plates. After 2 days, large haloed plaques could beobserved on the thinly growing bacterial lawn. The bacteriacould grow because the infecting phage carried and expressedthe argG gene of S. typhimurium. The centers of these plaqueswere picked, and the phage were purified on bacterial lawnsformed under the conditions of the original isolation. Phageobtained from plaques identified in this second isolation wereused to make lysates. The resulting lysates were screened fornusAst by testing for complementation of both the argG::TnSand nusAlO(Cs) mutations. The nusAst gene on the X1059derivative was isolated from a partial Sau3AI digestion ofthe phage DNA that was cloned into the BamHI site ofpACYC184. A plasmid that complemented the nusAJO(Cs)mutation was isolated; this plasmid, pNAS1000, contains a2.9-kb insert that includes nusA5s,DNA sequencing. The dideoxy chain termination method

described by Biggin et al. (4) was employed. Primers forsequencing were obtained from the University of MichiganOligonucleotide Synthesis Facility and Bethesda ResearchLaboratories.The nusA,st gene was sequenced by using a collection of

overlapping subclones of pNAS1000 constructed by ligatingsonication-generated DNA fragments into M13mp8. In thoseregions where overlapping subclones were not obtained, thesequence was determined directly from pNAS1000 by usingnusA-specific primers whose sequences were derived from theM13mp8 clones.PCR. Reactions were carried out in a thermocycler for 30

cycles. Cycle durations were 1 min at 94°C, 1 min at 44°C, and1 min at 72°C. The reaction sample consisted of 0.4 ,ug of theDNA template (pNAS1000) and 80 pmol of the primers.

Oligonucleotides used for priming PCRs were prepared bythe University of Michigan Oligonucleotide Synthesis Facility.The sequences of the primers and the nusAs, nucleotides towhich they bind are as follows: SaI128, 5' TCGATGTTCGTGTAGAA 3' (binds to nucleotides 128 to 144 of the bottomstrand); Las971, 5' ACGTTCTGACCATTACG 3' (binds tonucleotides 971 to 955 of the top strand); SaI483, 5' GGATQTGGGCAACAACGCTGAAGCGGTGATTC[G 3' (binds tonucleotides 504 to 484 of the bottom strand); and Las457, 5'GITGCCCAGATCCAGAGAGATA1TGTCGCGGTTC 3'(binds to nucleotides 457 to 435 of the top strand). PrimersSa1128 and Las971 perfectly match the nusAs, sequences.Primers Sal483 and Las457 were designed so that 13 bases attheir 5' ends were complementary to each other and thus canhybridize. Eleven (shown underlined) of the 13 bases corre-spond to the nusA,E nucleotides that replace nucleotides 458through 483 of the nusAst DNA sequence.

Construction of hybrid nusA genes. The nusAhy449 gene wascreated by PCR with the "gene splicing by overlap extension"technique (41). The first two rounds of PCR used, in separatereactions, primers Sal128 and Las457 and primers Sal483 andLas971 to amplify a portion of nusAst DNA, with pNAS1000 asthe template. The two fragments were gel purified and com-bined with primers Sal128 and Las971, the primers in the thirdround of PCR. Since the two fragments overlap by 13 nucle-otides, they could, at some frequency, anneal to each other.The annealed strands acted as templates for polymerase andgenerated the spliced recombinant DNA. The presence ofSal128 and Las971 allowed for amplification of the recombi-nant in subsequent cycles. The 840-bp fragment in which the 26bp of S. typhimurium DNA was replaced with 11 bp of E. coliDNA was digested with PvuI and used to replace the analo-gous fragment of the nusAst gene in pNAS1000. The resultingplasmid, pHSE19, was sequenced between the PvuI restrictionsites of the DNA insert to confirm that the inserted fragment

J. BACT1E1RIOL.

on October 26, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

HYBRID nusA GENES 1397

nusA

nusAEC

v- ~~~~~~~~~~-4

8

* ********(4/9) * * * *** * **

nusAhyllO2

nusAhy2 9

nusAhy3 9

nusAhy3O2 9

nusAhy3O2-449

nusAhy449 4li

nusA A324

FIG. 2. Structures of nusA genes. The bar at the top shows therelative positions of relevant restriction sites shared by the nusAFs andnusAs, genes. The site marked 4/9 is the position of the 449 region ofmajor heterology, and the asterisks indicate other regions where thetwo open reading frames encode different amino acids. Aligned beloware schematic representations of hybrid genes constructed fromnusAEc (solid bars) and nusAs, (hatched bars). A 4 or a 9 in the boxedarea indicates the derivation of the heterologous region: a 4 indicatesthe four codons from E. coli, and a 9 indicates the nine codons from S.typhimurium. The four amino acids found at this position in theNusAE, protein are DLGN, and the nine amino acids in the NusAs,protein are EIKSEGMAG. The small region with parallel horizontallines in the representation of nusAhy1lO2 indicates the region betweenthe AvaI and BglII restriction sites in nusAst where one of therecombination events is assumed to have occurred.

had the desired deletion-substitution and that no additionalchanges were introduced by the PCR and cloning steps.

All other hybrid genes were constructed by one of two invitro methods outlined below; both methods exploit restrictionenzyme recognition sequences that occupy identical locationsin the reading frames of nusAEc and nusAst. The structures ofthe hybrid genes are shown in Fig. 2 and described below (theplasmids containing these genes are described in Table 3).Method one, depicted in Fig. 3, permits isolation of recom-

binants without employing selective pressure for a functionalnusA gene. The recipient plasmid pNAS2001 has a tetR geneadjacent to the cloned nusA gene. A portion of this nusA geneis resected and replaced with a fragment from the donorplasmid that contains the analogous portion of the heterolo-gous nusA gene and downstream DNA that has replaced thetetR gene up to the BamHI site. This cloning not only createsa hybrid nusA gene but also results in the recombinant plasmid,like the donor, containing only a 3' portion of the tetR gene.

The hybrids constructed by this method are nusAhy2 andnusAhy3O2.Method two, depicted in Fig. 4, relies on selection of

functional hybrid nusA genes by complementation ofnusAlO(Cs), a conditionally lethal mutation (73). Plasmid

* * * * pN2/EuS has a tandem arrangement of the nusA4Ec and nusAs,genes. Since recognition sites for restriction enzyme PvuI areat the same positions in the two nusA genes, cleavage with thisenzyme could be employed to delete the DNA between the twoPvuI sites. Ligation of the remaining plasmid DNA generatesa hybrid nusA gene. The nusAhy3 gene was constructed by thismethod.

Southern and Western blot analyses. Southern blots wererun essentially as outlined by Sambrook et al. (71). Westernblots (immunoblots) were run essentially according to themethod employed by Schauer et al. (73), except antibody wasidentified with the ECL kit from Amersham.

Nucleotide sequence accession number. The DNA sequenceof nusA,st has been deposited in GenBank and given accessionnumber M61008.

RESULTS

Nucleotide sequence of nusAst. The DNA sequence ofnusAs, was determined as described in Materials and Methods.The nusAEc sequence used for comparison is based on theoriginal sequencing data as well as subsequent corrections (12,44, 45, 68). The DNA sequences of the two nusA genes areextremely similar, varying at only 156 of 1,488 nucleotides(data not shown). The predicted primary sequences of theNusAs5 and NusAEc proteins are compared in Fig. 5A. Asexpected from the comparison of the DNA sequences, the twoproteins are also very similar, varying at only 27 of 495 aminoacid residues.The alignment of the derived amino acid sequences of the

two NusA proteins, shown in Fig. 5A, reveals one region ofsignificant heterogeneity that is located between amino acids152 and 162 of the nusA,s sequence. In this region, the NusAstsequence has nine amino acids, EIKSEGMAG, while thecorresponding site in the NusAEc sequence has four differentamino acids, DLGN. This small region of NusA amino acidmismatch is referred to as the 449 region to reflect the aminoacid heterogeneity. A second distinguishing feature, a set oftandem repeats at the carboxy ends of both NusA proteins, wasrevealed by a dot plot comparison of the two amino acidsequences. The locations of the repeats are shown in Fig. 5A,and the alignment of the repeats is shown in Fig. SB. We havenot identified any functional role for this repeat. However, arepeat of an amino acid sequence has been implicated inTATA recognition by the eukaryotic transcription factorTFIID (40, 60).To determine whether amino acid sequence differences

relate to functional differences of the two NusA proteins, wecreated a series of hybrid genes from nusAEc and nusA,s thatexpress hybrid NusA proteins. The constructs used in this studyhave sufficient genetic material of the nusA genes to include allof the known promoters of the nusAEc operon or thoseidentified by sequence analysis for the nusA,s operon. Com-parison of the DNA sequence we obtained for the regionupstream of nusA4s with that obtained for the analogous regionupstream of nusAEc reveals a high degree of homology; the S.typhimurium genome also has, upstream of nusA, an openreading frame corresponding to a protein of 15 kDa, anfMet-tRNA, and a similar set of promoter sequences (32a, 33,38, 72a). In each of the constructions, the DNA upstream ofnusA derives from the same source as the 5' end of the nusA

_S__

__

__b > ' ,VOL. 176, 1994

on October 26, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

1398 CRAVEN ET AL.

EcoRIBomHI

BomHI

Stul

NcoI -

Eco RI

x Nco Ix Bam HI Eco RI

i

purify fragment -,combine + ligate

FIG. 3. Construction of nusAhy2. Plasmid pNAG15 (containing nusAE) was digested with restriction enzymes BamHI and NcoI. The fragmentcontaining the 3' end of nusAE, was isolated and ligated with linearized pNAS2001 that had been completely digested with BamHI but onlypartially digested with NcoI. This construction created a hybrid nusA gene. Like all of the constructs employed in this study, the cloned nusA geneincluded the associated promoter of the nusA operon. See Materials and Methods for details.

hybrid; e.g., the upstream region derives from S. typhimuriumwhen the 5' end of nusA derives from S. typhimurium.

Structures of the hybrid nus4 genes. In our initial studies,hybrid nusA genes were constructed according to either of twostrategies that relied on in vitro splicing, as detailed in Mate-rials and Methods. The structures of the hybrid genes areshown in Fig. 2 and described below (the plasmids containingthese genes are detailed in Table 3).The nusAhy2 gene contains nusAst sequences in the 5'

(N-terminal) two-thirds, and contains nusAEc sequences in the3' (C-terminal) one-third, of its reading frame. The nusAhy3gene is virtually identical to nusAst, since the nusAEc DNA atits 5' end encodes a sequence with single amino acid differencefrom the amino acid sequence derived for nusA5s in thatregion. However, the derivation of the associated 5' regulatoryregion in the case of nusAhy3 is from E. coli, while that fornusAst is, obviously, from S. typhimurium.The nusAhy3O2 gene has nusA,E sequences at its 5' and 3'

ends but contains an internal segment that is derived fromnusAs5. Because it has homologies with the E. coli chromosomeon either side of the S. typhimurium sequences, nusAhy3O2could be recombined into the E. coli chromosome. The result-ing hybrid chromosome was E. coli except for the smallreplacement of nuSAEc with nusAst sequences. Thus, we wereable to unambiguously distinguish effects due to the nusAstreading frame from any effects on expression that might be dueto upstream or downstream elements from the S. typhimuriumgenome.Complementation of conditionally defective nusA mutations

by the hybrid nus4 genes. Either the nusA,E or nusAst geneexpressed from plasmids complements two different condition-ally defective nusA mutations in E. coli, one cold sensitive[nusAlO(Cs)] and the other thermosensitive [nusAll(Ts)](data not shown). Therefore, like their chromosomal progen-itors, these cloned genes express products that are functionalin supporting essential bacterial processes requiring NusA

action. All of the plasmid-cloned hybrid nusA genes testedsupported the growth of bacteria with either thermosensitiveor cold-sensitive conditionally lethal nusA alleles (data notshown). Therefore, the resulting hybrid NusA proteins alsosupport essential bacterial functions that require NusA action.Complementation of the nusAl mutation by the hybrid nusA

genes. To determine whether a specific region of the NusAEcprotein is necessary for N-mediated antitermination, the plas-mid-based hybrid nusA genes were tested for support of Naction in a nusAl mutant host. The complementation testemployed was based on two observations. First, at highertemperatures wild-type X grows poorly in an E. coli derivativecarrying the nusAl mutation, because of a failure in N-mediated antitermination (19). Second, the wild-type allele isdominant in a nusA1/nusA+ heterozygote (21).

Derivatives of X with mutations in various genes and nucleicacid sites involved in termination and antitermination oftranscription provide a means for assaying host support of Naction (28). Three X derivatives that exhibit different degreesof dependence on host-encoded Nus factors were employed toassess N-mediated antitermination: (i) Abio256sex, which failsto form plaques on lawns formed from nus mutants at low aswell as high temperatures and exhibits maximal dependence(73); (ii) wild-type X, which forms plaques on such lawns at lowbut not at high temperatures and thus is considered to havenormal dependence (21); and (iii) Xnin, which forms plaqueson lawns formed from nus mutants at both low and hightemperatures and therefore, in terms of N action, exhibits Nusindependence. On the basis of these considerations, this groupof phages was employed to test the abilities of nusA genescloned on plasmids to support N-mediated antitermination ina host with a chromosomal nusAl allele. The growth of thetester phage in the heterozygote can be assessed at 40°C, atwhich the chromosomally encoded mutant NusA fails tosupport N action. We find that the restriction on X growthoccurs at lower temperatures (40 rather than 42°C) if the

Nco I

BomHI

x NcoIx Bom HI

J. BAC-FERIOL.

on October 26, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

HYBRID nusA GENES 1399

A

½-4P-f

> > 4-..

QL DLU)I i I I

I I.I5n;vsAA50L*w-

a E c-2 C._ 0 aK >

IIYlIIO-QTITTIT

/ /HIxfsI

-4

0-

F-4

Eam

////

FIG. 4. Construction of nusAhy3. Plasmid pN2/EuS contains one copy each of a carboxy-truncated nusAEc gene and a complete nusAs, geneoriented in the same direction and separated by a short intervening sequence. Cleavage at the PvuI sites (indicated by the arrows) followed byligation generates a hybrid nusA gene. Following partial digestion with PvuI and ligation, functional hybrids were selected by complementation ofthe nusAlO(Cs) mutation. See Materials and Methods for details of the construction.

nusAl mutant host carries a pBR322-based plasmid (Table 3and data not shown). Therefore, the restrictive temperatureused in this study was 40°C.The growth of X derivatives was quantitatively assayed in

nusAl/nusAhy heterozygotes by measuring EOP at 40°C (Table4). Although a nusA+ bacterium supports growth of all threephages, only Xnin5 is able to grow at 40°C in a nusAl host thatcarries pBR322, the vector employed in the cloning of thevarious nusA genes. A plasmid containing nusAEc fully com-

plements the nusAl mutation; all phages tested grow as well ina nusAl/nusAEc heterozygote as in the nusA+ bacterium. Aplasmid containing nusAst, however, fails to show complemen-tation for X growth.The plasmid with the truncated nusAA324 gene shows

complementation when assessed for growth of the X derivativewith wild-type Nus factor requirements, showing the impor-tance of the 5' region of nusA for N action. However, unlikethe case of the complete nusA gene, when the plasmid has thetruncated gene, the strain that is heterozygotic at the nusA

locus fails to support growth of Xbio256sex. Plasmids with thecloned hybrid genes nusAhy2, nusAhy3, and nusAhy3O2 do notcomplement nusAl (as shown in Table 4 for plasmids pNAX2,pNAX3, and pNAX302). Growth of the tester phages in theheterozygotes with these plasmids is indistinguishable fromthat observed in the nusAlInusAst heterozygote. The centralregions of these genes are derived from nusAst (Fig. 2).

Levels of expression by plasmid-based nusA genes. The

failure to support N-mediated antitermination by E. coliderivatives containing some of the plasmid-based hybrid nusAgenes could be explained by expression of low levels of NusAprotein. To assess levels of NusA expression from the varioushybrid nusA constructs, we measured steady-state levels ofNusA protein by employing immunoblots. Since E. coli re-

quires NusA for growth and wild-type NusA expressed fromthe chromosome would interfere with measuring NusA ex-

pressed from plasmids, we employed a derivative of E. coli,K7346, with the nusA134(Am) allele (80). In the absence of anamber suppressor, this strain expresses a truncated NusAprotein that supports bacterial growth at low temperatures.Thus, K7346 provides a means for identifying NusA proteinexpressed from a plasmid, provided that the plasmid-encodedNusA differs in size from the NusA amber fragment. Theimmunoblot in Fig. 6 shows that most of the NusA protein inK7346 runs lower in the gel than the wild-type protein at theposition expected for that amber fragment (lane 1). Therefore,the protein identified by the anti-NusA antibody observed atthe position of the wild-type NusA protein in the other lanesidentifies protein expressed from the plasmids. All of theplasmid-based hybrid genes express roughly equal levels oftheir respective NusA proteins. Therefore, the functionaldifferences observed between the different nusA-bearing plas-mids are not likely to be due to differences in the levels ofexpression of their respective NusA proteins.Chromosomal hybrid nusA genes. Previous studies with

P'-4

0C.)

Lii

I

P-4 -4

= C> Co

III I

VOL. 176, 1994

I -

on October 26, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

1400 CRAVEN ET AL.

A

S.t. MNKEILAVVEAVSNEKALPREKIFEALESALATATKKKYEQEIDVRVEID 5011111111111111111111111111111111111111111111111:11

E. c. MNKEILAVVEAVSNEKALPREKIFEALESALATATKKKYEQEIDVRVQID 50

S.t. RKSGDFDTFRRWLIVEEVTMPTKEITLEAARFEDESLNVGDYVEDQIESV 1001111111111111:1:111 11111111111:111111:11111111111

E. c. RKSGDFDTFRRWLVVDEVTQPTKEITLEAARYEDESLNLGDYVEDQIESV 100

S.t. TFDRITTQTAKQVIVQKVREAERAMVVDQFRDQEGEIVTGVVKKVNRDNI 150l l l ll Ill 11l11llll ll ll ll ll l: : ll l:1l

E. c. TFDRITTQTAKQVIVQKVREAERAMVVDQFREHEGEIITGVVKKVNRDNI 150

NineS.t. SLEIKSEGMAGNAEAVILREDMLPRENFRPGDRIRGVLYAVRPEARGAQL 200

I I: : : . I II II II II II I:1I I.1I1 IE. c. SL _ jNAEAVILREDMLPRENFRPGDRVRGVLYSVRPEARGAQL 195

Four InusAlR

S.t. FVTRSKPEMLIELFRIEVPEIGEEVIEIKAAARDPGSRAKIAVKTNDKRI 25011I111111111111111111111111111111111111111111111111

E. c. FVTRSKPEMLIELFRIEVPEIGEEVIEIKAAARDPGSRAKIAVKTNDKRI 245

S.t. DPVGACVGMRRARVQAVSTELGGERIDIVLWDDNPAQFVINAMAPADVAS 300111111I1111 111111111111111111111111111111111111111

E. c. DPVGACVGMRGARVQAVSTELGGERIDIVLWDDNPAQFVINAMAPADVAS 295

S.t. IVVDEDKHTMDIAVEAGNLAQAIGRNGQNVRLASQLSGWELNVMTVDDLQ 35011111111111111111111111111111111111111111111111111

E. c. IVVDEDKHTMDIAVEAGNLAQAIGRNGQNVRLASQLSGWELNVMTVDDLQ 345

S.t. AKHQAEAHAAIEIFTKYLDIDEEFATVLVEEGFSTLEELAYVPMKELLEI 40011111111111:. 111111111:111111111111111111 111111111

E. c. AKHQAEAHAAIDTFTKYLDIDEDFATVLVEEGFSTLEELAYVPMKELLEI 395

V 4S.t. DGLDEPTVEALRERAKNALATLAQDQEASLGDNKPADDLLNLEGVDRDMA 450

:11111111111111111111:11.11.11111111111111111111:1E. c. EGLDEPTVEALRERAKNALATIAQAQEESLGDNKPADDLLNLEGVDRDLA 445

VS.t. FKLAARGVCTLEDLAQQGIDHLADIEGLTDEKAGALIIAARNICWFGDEA 500

111111111II111111:1111.1111111111111111:111111111111E. c. FKLAARGVCTLEDLAEQGIDDLADIEGLTDEKAGALIMAARNICWFGDEA 495

BEc 364-414 DIDEDFATVLVEEGFSTLEELAYVPMKELLEIEGLDEPTVEALRERAKNAL

:1 11 1 1 111:11 : :1 :1111: : 11 I IEc 439-489 GVDRDLAFKLAARGVCTLEDLAEQGIDDLADIEGLTDEKAGALIMAARNIC

:1: 1 1 111:11 : : :1:11 : : 11St 369-489 DIDEEFATVLVEEGFSTLEELAYVPMKELLEIDGLDEPTVEALRERAKNAL

:1: 1 1 1 111:11 : :1:11: : 11 I ISt 444-494 GVDRDNAFKLAARGVCTLEDLAQQGIDHLADIEGLTDEKAGALIIAARNIC

I D D I E D E D AConsensus ID-I-A--L---G--TLEILA---I-IL-JIIGL-I--l-AL---A-N--

V E E M D E D E V

FIG. 5. Analysis of NusA proteins. Single-letter abbreviations areused for amino acids. (A) Comparison of the S. typhimurium and E. coliNusA protein sequences. The optimal alignment requires one gap of fiveamino acids between residues 153 and 154 of the E. coli sequence. Thetop line is the deduced S. typhimurium (S.t.) NusA protein sequence,and the bottom line is the deduced E. coli (E.c.) NusA protein sequence.The symbols placed between the two sequences indicate the relatednessof the amino acids: identical amino acids are indicated by a solid line,chemically similar amino acids are indicated by either two dots (verysimilar) or one dot (similar), and unrelated amino acids are indicated bya blank space. The 449 region is labeled, and its location is shown bylines above and below the sequence. The amino acid change of thenusAl mutation (68) is shown downstream of the 449 region. Thepositions of the repeated sequence are indicated by horizontal arrows attheir starts and triangles at their endpoints. The junctions of the hybridgenes created with restriction enzyme sites shared by the two nusA genesare indicated by vertical arrows ( T, PvuI; T T, NcoI; and T t I ,BglII). (B) Alignment of the partial repeats at the carboxy ends of thetwo NusA proteins. Related amino acids are indicated by the samesymbols described above. The consensus sequence derived from thisalignment is shown in the bottom line. Positions with alternative butrelated as well as unrelated (-) amino acids are indicated.

TABLE 4. Complementation of the nusAI mutation by hybridnusA genes

Plasmid Episomal EOPanusA gene A Anin5 Xbio256sex

pBR322 None 10-6 -1 <lo-8pNAG2010 nusAE,, -1 -1 -1pNAS2001 nusAst <10-7 -1 <10-8pWR324 nusAA324 -1 -1 <10-8pNAX2 nusAhy2 <10-7 -1 <10-8pNAX3 nusAhy3 <10-7 -1 <10-8pNAX302 nusAhy3O2 <10-7 -1 <10-8pNAX449 nusAhy449 -1 -1 <10-6

a Calculated by dividing the phage titer at 40°C with the indicated strain as alawn by the titer of the same phage with the permissive strain K37 carryingpBR322 as a lawn.

nusAs, employed a hybrid bacterium, Ki 102, defective forsupport of N action as assayed by X growth (21, 73). Thegenome of K1102, derived primarily from E. coli, has a smallsubstitution of S. typhimurium genetic information that wasthought to include the entire nusAst gene (2, 21). However,there was no direct evidence as to the structure of the nusAgene in this bacterium. Therefore, we employed Southernanalysis of chromosomal DNA digested with restriction en-zymes to determine the structure of this nusA gene. Thesestudies (data not shown) indicate that a crossover eventgenerating one of the hybrid junctions must have occurredwithin the 3' segments of the nusA genes (Fig. 2). The positionof the crossover is located between sites for AvaI (nucleotide1172 of nusAst), which is present in the nusA gene from K1102,and BglI (nucleotide 1280 of nusAs,), which is absent in the

1 2 3 4 5 6 7

84kd -

Full sized NusA

58kd -

48.5kd -Amber fragment of NusA 40Mq*M :to

gz*

36.5 kd -

FIG. 6. Western blot analysis to assess NusA expression fromplasmids. Proteins were isolated and analyzed essentially as outlined bySchauer et al. (73). The E. coli strain used in this study was K7346,which has the nusA134(Am) mutation and expresses a truncated NusAprotein. Positions of molecular mass markers, full-sized NusA protein,and truncated NusA amber fragment are indicated to the left. Each ofthe lanes (numbered at the top) was loaded with a lysate preparedfrom K7430 or one of its derivatives containing the following pBR322derivative, with the nusA gene listed in parentheses: 1, no plasmid(control); 2, pNAX2010 (nusAE); 3, pNAS2001 (nusAsj); 4, pNAX449(nusAhy302-449); 5, pNAX2 (nusAhy2); 6, pNAX3 (nusAhy3); 7,pNAX302 (nusAhy302). Slight differences reflect differences in totalprotein loaded.

J. BAC-rERIOL.

on October 26, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

HYBRID nusA GENES 1401

nusA gene from K1102. The other crossover creating thishybrid gene occurred somewhere 5' to the nusA gene. Becauseof the position of the crossover within the nusA gene, we havenamed this allele nusAhyllO2.A second chromosomal hybrid nusA gene was created by

crossing the nusAhy302 allele from plasmid pNAX302 to thechromosome (see Materials and Methods for details) to gen-erate K4091. The portion of the chromosome of K4091 that isderived from S. typhimunium sequences can extend no furtherthan from nucleotide 148 to 926, as this is the extent of S.typhimurium DNA in nusAhy302. The chromosomalnusAhy302 allele, like the nusAhyllO2 allele, fails to supportN-mediated antitermination (data not shown).

Defining the region of NusA required for N action. ThenusAEc-nusAst hybrids that support action of N have nUSAEcsequences that include the 449 region, which encodes aminoacids at positions 153 to 156. In this region there are fouramino acids in NusAE, (DLGN) and nine amino acids inNusAst (EIKSEGMAG).To assess the functional significance of the 449 region, we

created a hybrid gene that is primarily nusAs, but has asubstitution in the 449 region; the nucleotides encoding thefour amino acids of the nusAE, 449 region replaced thenucleotides encoding the nine amino acids of NusAst. Thehybrid gene was named nusAhy449 to reflect this switch.The effectiveness of the nusAhy449 gene product in support

of the X N-mediated antitermination reaction was assessed byusing the previously described complementation assay. For thisstudy we employed K6210, which has the chromosomal nusAlallele and a pBR322 derivative with the nusAhy449 allele(pNAX449). A plates with an EOP of 1 on this heterozygoticstrain at 40°C (Table 4). Bearing in mind that a host with thenusAl allele fails to support A growth at 40°C specificallybecause of a failure in N action, this result establishes that theproduct of the plasmid-based nusAhy449 gene supports Naction. However, the low EOP of Xbio256sex on this diploid at40°C (less than 10-6) indicates that although the potency ofthe nusAhy449 allele in support of N action is significantlygreater than that of nusAst, it is less than that of wild-typenusAEc.We next determined whether expression of nusAhy449 in the

homozygous state is sufficient to support N action. This wasaccomplished by creating a chromosomal gene with the struc-ture of nusAhy449. To this end, the central region ofnusAhy449 was crossed to a chromosomally located nusAhy3O2allele (see Materials and Methods). As previously discussed,nusAhy3O2, like nusAst, fails to support X N action. Theresulting hybrid gene, nusAhy302-449, differs from nusAhy449,because the former has E. coli sequences on the 5' and 3' ends.However, EOP studies showed that the chromosomally basednusAhy3O2-449 allele, like the plasmid-based nusAhy449 allele,supports growth of X with a wild-type N but not growth ofXbio256sex (data not shown).

Effect of the nusA allele on bacterial growth. The observa-tion that X N functions with NusAEc but not with NusAstindicates that the two proteins differ in at least one functionaldomain. Such a domain might facilitate NusA interaction(s)with a bacterial factor(s), and thus there could conceivably beconditions in which NusAEc functions more effectively in E.coli than NusAst. To assess the relative physiological effective-ness of the two nusA gene products in an E. coli background,we compared the growth of two isogenic strains of E. coli thatdiffer only in their nusA genes. These strains are K1456, whichhas nusAEc, and K4092, a derivative of K1456 which has thenusAhy3O2 allele. Studies monitoring the growth of cultures ofthe two strains failed to reveal any significant differences under

a wide variety of conditions, which included temperaturesranging between 16 and 42°C in both defined minimal and richmedia as well as anaerobiosis in rich medium and growth in themouse gut (data not shown).

DISCUSSION

Variations between closely related species have long beenused as tools to assess the functional consequences of subtlegenetic differences (58). In the present study, we have exam-ined the structural basis of differences in the actions of nusAgene products from two closely related species of entericbacteria, E. coli (nusA,) and S. typhimurium (nusAst) (66, 72).The demonstration that an E. coli hybrid with a small substi-tution of S. typhimurium DNA in the nusA region failed tosupport N-mediated antitermination (21) suggested that if S.typhimurium has a nusA gene, its product is unable to supportN action. Since NusA provides an essential function, it seemedlikely that S. typhimurium has such a nusA gene. The basis forthe current study was the cloning and sequencing of the nusA,sgene as well as the comparison of the derived amino acidsequence with that of nusAEc. This comparison revealed twofeatures that we consider significant: (i) the two sequenceshave extensive homologies, and (ii) alignment of the two aminoacid sequences reveals one region in which there is significantvariation between the two NusA proteins.

Since NusAst functions in the genetic background of E. coliK-12, it was not surprising to find that the derived amino acidsequences of the two nusA genes have extensive homologies.Both in vivo and in vitro recombination were employed tocreate a series of hybrid nusA genes. These hybrid genes wereused to relate structure and function by examining theireffectiveness in supporting bacterial growth and N-mediatedantitermination.

All of the hybrids used in this study were constructed so asto include the upstream promoter regions. The observationthat all of the plasmid constructs with hybrid nusA genescomplemented nusA mutations for bacterial growth demon-strates that these plasmid-based hybrid genes must be ex-pressed. Moreover, Western blot analysis shows that all of theconstructs express similar levels of NusA protein.

Support of N action. Hybrid nusA genes with the proximalpart of the central portion from nusAst fail to support XN-mediated antitermination. This portion of NusA, not sur-prisingly, includes the 449 region that was identified by thealignment of the two NusA amino acid sequences. In thisregion NusAst has nine amino acids covering a region in whichNusAEc has four different amino acids. To test the functionalsignificance of this heterology, we created a hybrid nusA genewhose encoded protein is NusAst except for the region ofheterogeneity, where the four amino acids in NusAEc havebeen substituted for the nine in NusAst.

This substitution of 449 regions, creating nusAhy449, wassufficient to create a hybrid protein that supports N action. Thediploid with a chromosomal nusA1 allele and a plasmid withnusA449 supports growth of a X derivative with wild-type Nusdependence under conditions in which the nusAl haploidparent fails to support growth of this phage. However, theheterozygote does not support growth of Abio256sex, a phagethat expresses low levels of an altered N protein. It has beenproposed that because it has a significant impairment in Nactivity, this phage requires more-effective interaction of theother components that make up the antitermination complex(73). Thus, the inability of the nusAhy449 gene product tosupport growth of Abio256sex indicates that this NusA protein

VOL. 176, 1994

on October 26, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

1402 CRAVEN ET AL.

is not as effective in support ofN action as the wild-type nus4E",gene product.Lambdoid phages 21 and P22 have antitermination systems

analogous to the N system of X, with similarly located N genesand nut target sites (18, 30, 39, 51). What is curious, at least forphage 21, an E. coli phage, is that these N products, unlike thatof X, function with nusAs5. Thus, the N product of A uniquelyrequires the 449 region from E. coli. However, the requirementof the X N for a NusA protein with the E. coli 449 region is notimmutable; a X derivative with mutations in the N gene as wellas in the nutR signal will grow in a hybrid E. coli strain withnusAhyllO2 (25). The N mutations, punAl and punA133,change, respectively, amino acids 45 (Lys to Arg) and 50 (Serto Asp) of the N protein (73). The nut mutations, either box,A4or boxA(Con), are changes in base pairs that create boxAsequences in nutR that conform more closely to the consensusbox;A sequence (27).

Considering these findings in light of the in vitro studiesdemonstrating N binding to NusA (34), it is plausible toassume that the interaction with X N is influenced by the aminoacids in the 449 region. Perhaps the 449 region of NusAErpositions X N so that proper contacts are made with NusAand/or other proteins or nucleic acid sites. This may ensure theformation of other contacts necessary to establish a function-ing antitermination complex. Our finding that the 449 regionfrom NusAE, is not sufficient for full X N action is consistentwith the argument of additional contacts within NusA. Asso-ciated with the 449 region is a run of amino acids immediatelydownstream that appears to be functionally important. Within26 amino acids of the 449 region are found mutations influ-encing both NusA support of N action (nusAl) and bacterialgrowth [nusAll(Ts)] phenotypes (63, 73). Moreover, thisregion in both nusA Ec and nusAs, is unusually rich in arginines;there are 8 conserved arginines in a run of 35 amino acids.Since arginines have been associated with regions of proteinsthat interact with RNA (6,51, 78), it is conceivable that the 449region and this associated arginine-rich region represent suchan active domain for NusA (79). The nusAl mutation results ina change from a leucine to an arginine in the NusA protein.This extra arginine residue in the run of arginines might causeNusA to bind N or another Nus factor less effectively orposition it unfavorably in the antitermination complex.Chromosomally located nusA hybrids. The phenotype of

nusAst in single copy was studied by using chromosomal hybridgenes nusAhyllO2 and nusAhy3O2. Although these hybridgenes have different proportions of nusAEr and nusAst, theyboth fail to support N action. The nusAhyllO2 allele had beenassumed to be a complete nusAst gene (21, 73). Our analysisdemonstrates that one of the crossover events that substitutedS. typhimurium DNA for E. coli DNA in K1102 occurredwithin the 3' regions of the nusA genes. The nusAst portion ofnusAhyl1O2, as well as that of nusAhy3O2, includes the 449region. Therefore, we think it logical to assume that studies ofN action with either of these chromosomally based hybridgenes should reflect the phenotype expected of the completenusAst gene.

Finally, we focus on nusAhy3O2. Since this hybrid is nusA,Econ both the 5' and 3' ends, the only S. typhimurium DNA in thebacterium is internal to the nusA gene. Therefore, all regula-tory signals (transcriptional and translational) that are 5' aswell as 3' of the nusA gene are from E. coli. The functionalfeatures distinguishing nusAst from nusAEc should thereforeprimarily reflect structural differences. The chromosomallylocated nusAhy302 allele, like the plasmid-based hybrid, fails tosupport N action. However, a derivative of nusAhy3O2 that hasthe four NusAEC amino acids instead of the nine NusA5t amino

acids in the 449 region supports wild-type N action. Thus, thenusAhy3O2-449 allele is active in single copy, and its phenotypedoes not vary from that observed when it is in multiple copies;e.g., it supports growth of X but not growth of Xbio256sex.

Conclusion. Our studies identify the 449 region as importantfor NusA support of A N action but fail to provide sufficientinformation to elucidate the specific role of that region in Naction or to determine whether it has importance for NusAinteractions in E. coli. Since the NusA structure probably didnot evolve to foster an interaction with X N, it is likely that the449 region serves to enhance interactions within the bacteriumthat are yet to be identified. Perhaps results of current studieson the effects of altering amino acids in the 449 region andsurrounding regions will provide the information to formulatea definitive and unified model.

ACKNOWLEDGMENTS

We thank the following colleagues: J. L. Claflin for suggesting theuse of the "gene splicing by overlap extension" technique, JuliaGeorge for help in mastering PCR technology, Rolf Freter for the useof the anaerobic chamber and help with the mouse experiments, P.Youderian for helpful suggestions and the S. typhimurium library,Brian Haarer and Tim Weber for technical help, and Victor DiRitaand David Engelke for critical reading of the manuscript.

This work was supported by Public Health Research Grant A1l1459-10. M.G.C., A.E.G., and T.A.G. were supported, in part, by NIHtraining grant 5 T32 GM07315 13. DNA sequence analysis wasperformed in part with computer programs made available throughGCRC grant M01-RR00042.

REFERENCES1. Barik, S., B. Ghosh, W. Whalen, D. Lazinski, and A. Das. 1987. An

antitermination protein engages the elongating transcription ap-paratus at a promoter-proximal recognition site. Cell 50:.885-899.

2. Baron, L S., E. Penido, I. R. Ryman, and S. Falkow. 1970.Behavior of coliphage lambda in hybrids between Escherichia coliand Salmonella. J. Bacteriol. 102:221-233.

3. Bear, S. E., D. L. Court, and D. L. Friedman. 1984. An accessoryrole for Escherichia coli integration host factor: characterization ofa lambda mutant dependent upon integration host factor for DNApackaging. J. Virol. 52:966-972.

4. Biggin, M. D., T. J. Gibson, and G. F. Hong. J983. Buffer gradientgels and 35S label as an aid to rapid DNA sequence determina-tion. Proc. Natl. Acad. Sci. USA 80:3963-3965.

5. Bolivar, F. 1978. Construction and characterization of new cloningvehicles. III. Derivatives of plasmid pBR322 carrying unique EcoRI sites for selection of Eco RI generated recombinant DNAmolecules. Gene 4:121-136.

6. Calnan, B. J., B. Tidor, S. Biancalana, D. Hudson, and A. D.Frankel. 1991. Arginine-mediated RNA recognition: the argininefork. Science 252:1167-1171. (Erratum, 255:665, 1992.)

7. Chang, A. C., and S. N. Cohen. 1978. Construction and character-ization of amplifiable multicopy DNA cloning vehicles derivedfrom the P1SA cryptic miniplasmid. J. Bacteriol. 134:1141-1156.

8. Cheng, S. W., E. C. Lynch, K. R. Leason, D. L. Court, B. A.Shapiro, and D. I. Friedman. 1991. Functional importance ofsequence in the stem-loop of a transcription terminator. Science254:1205-1207.

9. Costantino, N., M. Zuber, and D. Court. 1990. Analysis ofmutations in the ninR region of bacteriophage lambda that bypassa requirement for lambda N antitermination. J. Bacteriol. 172:4610-4615.

10. Court, D., C. Brady, M. Rosenberg, D. L Wulf, M. Behr, M.Mahoney, and S. U. Izumi. 1980. Control of transcription termi-nation: a rho-dependent termination site in bacteriophage lambda.J. Mol. Biol. 138:231-254.

11. Court, D., and K. Sato. 1969. Studies of novel transducing variantsof lambda: dispensability of genes N and Q. Virology 39:348-352.

12. Craven, M. G., and D. L. Friedman. 1991. Analysis of the Esche-richia coli nusAIO(Cs) allele: relating nucleotide changes to phe-

J. BACTERIOL.

on October 26, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

HYBRID nusA GENES 1403

notypes. J. Bacteriol. 173:1485-1491.13. Das, A. 1992. How the phage lambda N gene product suppresses

transcription termination: communication of RNA polymerasewith regulatory proteins mediated by signals in nascent RNA. J.Bacteriol. 174:6711-6716.

14. Das, A., and K. Wolska. 1984. Transcription antitermination invitro by lambda N gene product: requirement for a phage nut siteand the products of host nusA, nusB, and nusE genes. Cell38:165-173.

15. Drahos, D., and W. Szybalski. 1981. Antitermination and termi-nation functions of the cloned nutL, N, and tLl modules ofcoliphage lambda. Gene 16:261-274.

16. Farnham, P. J., J. Greenblatt, and T. Platt. 1982. Effects of NusAprotein on transcription termination in the tryptophan operon ofEscherichia coli. Cell 29:945-951.

17. Faus, I., C. Y. Chen, and J. P. Richardson. 1988. Sequences in the5' proximal segment of the paused transcript affect NusA-medi-ated enhancement of transcriptional pausing. J. Biol. Chem.263:10830-10835.

18. Franklin, N. C. 1985. Conservation of genome form but notsequence in the transcription antitermination determinants ofbacteriophages lambda, phi 21 and P22. J. Mol. Biol. 181:75-84.

19. Friedman, D. I. 1971. A bacterial mutant affecting lambda devel-opment, p. 733-738. In A. D. Hershey (ed.), The bacteriophagelambda. Cold Spring Harbor Laboratory, Cold Spring Harbor,N.Y.

20. Friedman, D. I. 1988. Regulation of phage gene expression bytermination and antitermination of transcription, p. 263-319. In R.Calendar (ed.), The bacteriophages 2. Plenum Press, New York.

21. Friedman, D. I., and L. S. Baron. 1974. Genetic characterization ofa bacterial locus involved in the activity of the N function of phagelambda. Virology 58:141-148.

22. Friedman, D. I., and M. Gottesman. 1983. Lytic mode of lambdadevelopment, p. 21-51. In R. W. Hendrix, J. W. Roberts, F. W.Stahl, and R. A. Weisberg (ed.), Lambda II. Cold Spring HarborLaboratory, Cold Spring Harbor, N.Y.

23. Friedman, D. I., M. J. Imperiale, and S. L. Adhya. 1987. RNA 3'end formation in the control of gene expression. Annu. Rev.Genet. 21:453-488.

24. Friedman, D. I., C. T. Jolly, and R. J. Mural. 1973. Interferencewith the expression of the N gene function of phage lambda in a

mutant of Escherichia coli. Virology 51:216-226.25. Friedman, D. I., and E. R. Olson. 1983. Evidence that a nucleotide

sequence, "boxA," is involved in the action of the NusA protein.Cell 34:143-149.

26. Friedman, D. I., E. R. Olson, C. Georgopoulos, K. Tilly, I.

Herskowitz, and F. Banuett. 1984. Interactions of bacteriophageand host macromolecules in the growth of bacteriophage lambda.Microbiol. Rev. 48:299-325.

27. Friedman, D. I., E. R. Olson, L. L. Johnson, D. Alessi, and M. G.Craven. 1990. Transcription-dependent competition for a hostfactor: the function and optimal sequence of the phage lambdaboxA transcription antitermination signal. Genes Dev. 4:2210-2222.

28. Friedman, D. I., A. T. Schauer, M. R. Baumann, L. S. Baron, and

S. L. Adhya. 1981. Evidence that ribosomal protein S10 partici-pates in control of transcription termination. Proc. Natl. Acad. Sci.USA 78:1115-1118.

29. Friedman, D. I., A. T. Schauer, E. J. Mashni, E. R. Olson, andM. F. Baumann. 1983. Escherichia coli factors involved in theaction of the k gene N antitermination function, p. 39-42. In D.Schlessinger (ed.), Microbiology-1983. American Society forMicrobiology, Washington, D.C.

30. Friedman, D. I., G. S. Wilgus, and R J. Mural. 1973. Gene Nregulator function of phage lambda immun2l: evidence that a siteof N action differs from a site of N recognition. J. Mol. Biol.81:505-516.

31. Gill, S. C., S. E. Weitzel, and P. H. von Hippel. 1991. Escherichiacoli sigma 70 and NusA proteins. I. Binding interactions with core

RNA polymerase in solution and within the transcription complex.J. Mol. Biol. 220:307-324.

32. Goda, Y., and J. Greenblatt. 1985. Efficient modification of E. coliRNA polymerase in vitro by the N gene transcription antitermi-

nation protein of bacteriophage lambda. Nucleic Acids Res.13:2569-2582.

32a.Granston, A. E. 1988. Ph.D. thesis. University of Michigan, AnnArbor.

33. Granston, A. E., D. L. Thompson, and D. I. Friedman. 1990.Identification of a second promoter for the metY-nusA-infBoperon of Escherichia coli. J. Bacteriol. 172:2336-2342.

34. Greenblatt, J., and J. Li. 1981. The nusA gene protein ofEscherichia coli. Its identification and a demonstration that itinteracts with the gene N transcription anti-termination protein ofbacteriophage lambda. J. Mol. Biol. 147:11-23.

35. Greenblatt, J., and J. Li. 1981. Interaction of the sigma factor andthe nusA gene protein of E. coli with RNA polymerase in theinitiation-termination cycle of transcription. Cell 24:421-428.

36. Greenblatt, J., and J. Li. 1982. Properties of the N gene transcrip-tion antitermination protein of bacteriophage lambda. J. Biol.Chem. 257:362-365.

37. Greenblatt, J., M. McLimont, and S. Hanly. 1981. Termination oftranscription by nusA gene protein of Escherichia coli. Nature(London) 292:215-220.

38. Grunberg-Manago, M. 1987. Regulation of the expression ofaminoacyl-tRNA synthetases and translation factors, p. 1836-1409. In F. C. Neidhardt, J. L. Ingraham, K. B. Low, B. Magasanik,M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli andSalmonella typhimurium: cellular and molecular biology. Ameri-can Society for Microbiology, Washington, D.C.

39. Hilliker, S., and D. Botstein. 1976. Specificity of genetic elementscontrolling regulation of early functions in temperate bacterio-phages. J. Mol. Biol. 106:537-566.

40. Hoeumakers, J. H. J. 1990. Letter to the editor. Nature (London)343:417-418.

41. Horton, R. M., H. D. Hunt, S. N. Ho, J. K. Pullen, and L. R. Pease.1989. Engineering hybrid genes without the use of restrictionenzymes: gene splicing by overlap extension. Gene 77:61-68.

42. Horwitz, R. J., J. Li, and J. Greenblatt. 1987. An elongationcontrol particle containing the N gene transcriptional antitermi-nation protein of bacteriophage lambda. Cell 51:631-641.

43. Ishihama, A., M. Kajitani, M. Enami, H. Nagasawa, and R.Fukuda. 1983. Transcriptional apparatus of Escherichia coli: RNApolymerase and its accessory proteins, p. 4-6. In D. Schlessinger(ed.), Microbiology-1983. American Society for Microbiology,Washington, D.C.

44. Ishii, S., M. Ihara, T. Maekawa, Y. Nakamura, H. Uchida, and F.Imamoto. 1984. The nucleotide sequence of the cloned nusA geneand its flanking region of Escherichia coli. Nucleic Acids Res.12:3333-3342.

45. Ito, K, K. Egawa, and Y. Nakamura. 1991. Genetic interactionbetween the P' subunit of RNA polymerase and the arginine-richdomain of Escherichia coli NusA protein. J. Bacteriol. 173:1492-1501.

46. Jin, D. J., M. Cashel, D. I. Friedman, Y. Nakamura, W. A. Walter,and C. A. Gross. 1988. Effects of rifampicin resistant rpoBmutations on antitermination and interaction with nusA in Esch-erichia coli. J. Mol. Biol. 204:247-261.

47. Kingston, R. E., and M. J. Chamberlin. 1981. Pausing andattenuation of in vitro transcription in the rrnB operon of E. coli.Cell 27:523-531.

48. Kroger, M., and G. Hobom. 1982. A chain of interlinked genes inthe ninR region of bacteriophage lambda. Gene 20:25-38.

49. Kung, H., C. Spears, and H. Weissbach. 1975. Purification andproperties of a soluble factor required for the deoxyribonucleicacid-directed in vitro synthesis of beta-galactosidase. J. Biol.Chem. 250:1556-1562.

50. Lau, L. F., J. W. Roberts, and R. Wu. 1983. RNA polymerasepausing and transcript release at the lambda tRI terminator invitro. J. Biol. Chem. 258:9391-9397.

51. Lazinski, D., E. Grzadzielska, and A. Das. 1989. Sequence-specificrecognition of RNA hairpins by bacteriophage antiterminatorsrequires a conserved arginine-rich motif. Cell 59:207-218.

52. Leason, K. R., and D. I. Friedman. 1988. Analysis of transcriptiontermination signals in the nin region of bacteriophage lambda: theroc deletion. J. Bacteriol. 170:5051-5058.

53. Li, J., R. Horwitz, S. McCracken, and J. Greenblatt. 1992. NusG,

VOL. 176, 1994

on October 26, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

1404 CRAVEN ET AL.

a new Escherichia coli elongation factor involved in transcriptionalantitermination by the N protein of phage lambda. J. Biol. Chem.267:6012-6019.

54. Mason, S. W., and J. Greenblatt. 1991. Assembly of transcriptionelongation complexes containing the N protein of phage lambdaand the Escherichia coli elongation factors NusA, NusB, NusG,and S10. Genes Dev. 5:1504-1512.

55. Mason, S. W., J. Li, and J. Greenblatt. 1992. Host factor require-ments for processive antitermination of transcription and suppres-

sion of pausing by the N protein of bacteriophage lambda. J. Biol.Chem. 267:19418-19426.

56. Miller, H. I., and D. L. Friedman. 1980. An E. coli gene productrequired for lambda site-specific recombination. Cell 20:711-719.

57. Miller, H. I., M. A. Mozola, and D. I. Friedman. 1980. int-h: an intmutation of phage lambda that enhances site-specific recombina-tion. Cell 20:721-729.

58. Montgomery, T. H., Jr. 1901. A study of the chromosomes of thegerm-cells of Metazoa. Trans. Am. Philosophical Soc. 20:154-236.

59. Mozola, M. A., D. I. Friedman, C. L. Crawford, D. L Wulff, H.Shimatake, and M. Rosenberg. 1979. Mutations reducing theactivity of c17, a promoter of phage lambda formed by a tandemduplication. Proc. Natl. Acad. Sci. USA 76:1122-1125.

60. Nagai, K. 1990. Letter to the editor. Nature (London) 343:418.61. Nakamura, Y., and S. Mizusawa. 1985. In vivo evidence that the

nusA and infB genes of E. coli are part of the same multi-geneoperon which encodes at least four proteins. EMBO J. 4:527-532.

62. Nakamura, Y., S. Mizusawa, D. L Court, and A. Tsugawa. 1986.Regulatory defects of a conditionally lethal nusAts mutant ofEscherichia coli. Positive and negative modulator roles of NusAprotein in vivo. J. Mol. Biol. 189:103-111.

63. Nakamura, Y., S. Mizusawa, A. Tsugawa, and M. Imai. 1986.Conditionally lethal nusAts mutation of Escherichia coli reducestranscription termination but does not affect antitermination ofbacteriophage lambda. Mol. Gen. Genet. 204:24-28.

64. Nodwell, J. R., and J. Greenblatt. 1991. The nut site of bacterio-phage lambda is made of RNA and is bound by transcriptionantitermination factors on the surface ofRNA polymerase. GenesDev. 5:2141-2151.

65. Reyes, O., M. Gottesman, and S. Adhya. 1979. Formation oflambda lysogens by IS2 recombination: gal operon-lambda pRpromoter fusions. Virology 94:400-408.

66. Riley, M., and A. Anilionis. 1978. Evolution of the bacterialgenome. Annu. Rev. Microbiol. 32:519-560.

67. Roberts, J. W. 1988. Phage lambda and the regulation of transcrip-tion termination. Cell 52:5-6.

68. Saito, M., A. Tsugawa, K. Egawa, and Y. Nakamura. 1986. Revisedsequence of the nusA gene of Escherichia coli and identification ofnusAl 1 (ts) and nusAl mutations which cause changes in a

hydrophobic amino acid cluster. Mol. Gen. Genet. 205:380-382.69. Salstrom, J. S., and W. Szybalski. 1978. Coliphage lambda nutL: a

unique class of mutants defective in the site of gene N productutilization for antitermination of leftward transcription. J. Mol.Biol. 124:195-221.

70. Salstrom, J. S., and W. Szybalski. 1978. Transcription termination

sites in the major leftward operon of coliphage lambda. Virology88:252-262.

71. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecularcloning: a laboratory manual, 2nd ed. Cold Spring Harbor Labo-ratory, Cold Spring Harbor, N.Y.

72. Sanderson, K. E. 1976. Genetic relatedness in the family Entero-bacteriaceae. Annu. Rev. Microbiol. 30:.327-349.

72a.Schauer, A. T. 1985. Ph.D. thesis. University of Michigan, AnnArbor.

73. Schauer, A. T., D. L. Carver, B. Bigelow, L. S. Baron, and D. I.Friedman. 1987. Lambda N antitermination system: functionalanalysis of phage interactions with the host NusA protein. J. Mol.Biol. 194:679-690.

74. Schmidt, M. C., and M. J. Chamberlin. 1984. Binding of rho factorto Escherichia coli RNA polymerase mediated by nusA protein. J.Biol. Chem. 259.15000-15002.

75. Schmidt, M. C., and M. J. Chamberlin. 1987. nusA protein ofEscherichia coli is an efficient transcription termination factor forcertain terminator sites. J. Mol. Biol. 195:809-818.

76. Sigmund, C. D., and E. A. Morgan. 1988. Nus A protein affectstranscriptional pausing and termination in vitro by binding todifferent sites on the transcription complex. Biochemistry. 27:5622-5627.

77. Sternberg, N. L., and R Maurer. 1991. Bacteriophage-mediatedgeneralized transduction in Escherichia coli and Salmonella typhi-murium. Methods Enzymol. 204:18-43.

78. Tao, J., and A. D. Frankel. 1992. Specific binding of arginine toTAR RNA. Proc. Natl. Acad. Sci. USA 89:.2723-2726.

79. Tsugawa, A., T. Kurihara, M. Zuber, D. L Court, and Y. Naka-mura. 1985. E. coli NusA protein binds in vitro to an RNAsequence immediately upstream of the boxA signal of bacterio-phage lambda. EMBO J. 4:2337-2342.

80. Tsugawa, A., M. Saito, D. L. Court, and Y. Nakamura. 1988. nusAamber mutation that causes temperature-sensitive growth ofEsch-erichia coli. J. Bacteriol. 170.908-915.

81. Ward, D. F., and M. E. Gottesman. 1981. The nus mutations affecttranscription termination in Escherichia coli. Nature (London)292:212-215.

82. Whalen, W., B. Ghosh, and A. Das. 1988. NusA protein isnecessary and sufficient in vitro for phage lambda N gene productto suppress a rho-independent terminator placed downstream ofnutL. Proc. Natl. Acad. Sci. USA 85:2494-2498.

83. Yager, T. D., and P. H. Von Hippel. 1987. Transcript elongationand termination in Escherichia coli, p. 1241-1275. In F. C.Neidhardt, J. L. Ingraham, K. B. Low, B. Magasanik, M. Schaech-ter, and H. E. Umbarger (ed.), Escherichia coli and Salmonellatyphimurium: cellular and molecular biology. American Society forMicrobiology, Washington, D.C.

84. Yanisch-Perron, C., J. Vieira, and J. Messing. 1985. ImprovedM13 phage cloning vectors and host strains: nucleotide sequencesof the M13mpl8 and pUC19 vectors. Gene 33:103-119.

85. Zengel, J. M., and L. Lindahl. 1992. Ribosomal protein LA andtranscription factor NusA have separable roles in mediatingtermination of transcription within the leader of the S10 operon ofEscherichia coli. Genes Dev. 6:2655-2662.

J. BACTERIOL.

on October 26, 2020 by guest

http://jb.asm.org/

Dow

nloaded from