J. Biochem. 119, 1062-1069 (1996)

The Virion Proteins Encoded by Bacteriophage $K and Its Host-RangeMutant ^KhT: Host-Range Determination and DNA BindingProperties1

Ken-Ichi Kodaira,** Masaya Oki,* Makiko Kakikawa,* Hisashi Kimoto,* andAkira Taketo*

'Molecular Biology Group, Chemical and Biochemical Engineering, Faculty of Engineering, Toyama University,3190 Gofuku, Toyama, Toyama 930; and'Department of Biochemistry I, Fukui Medical School, Matsuoka, Fukui910-11

Received for publication, December 15, 1995

The microvirid phage ^K, specific for Escherichia coli K12, contains a circular single-stranded (SS) DNA in the icosahedral virion, which comprises four phage gene products, F(capsid), G (major spike), H (minor spike), and J (core). ^KhT, a host-range mutant of ^K,can grow on E. coli C and B, besides K12, and is more thermosensitive than the parentalphage ^K. Sequencing analysis revealed that the genome of ^K and ^KhT consists of 6,089nucleotides (nt), and codes for eleven genes, whose sequences are similar to those of a3,^X174, and G4 infective to strain C. In #KhT, two nt had changed: one is in the gene G,resulting in replacement of the 75th codon Ala with Ser, and the other is at 67th codon of thegene H: Val to Ala. Chemically synthesized gene J protein composed of 23 amino acids (aa)binds to ^K SS DNA more tightly than and preferentially over the host E. coli SS-DNA-binding protein (SSB). These results indicate that the two spike proteins G and H areinvolved in the determination of ^K host-range, and support a model in which the gene Jprotein functions in packaging the viral SS DNA into the virion vesicle.

Key words: bacteriophage, DNA-binding protein, host-range, £K, virion proteins.

Various strains of microvirid (isometric) phages of Escheri-chia coli have been extensively used as probes for elucida-tion of various biochemical and genetic problems, includinggene structure, DNA replication, transcription, and host-parasite relationship {1, 2). These phages contain a circularSS DNA in the icosahedral virion, which is composed offour phage gene products, F (capsid), G (major spike), H(minor spike), and J (core), and are classified into threemajor groups: (i) the first group consists of #X174, S13,and G6, (ii) the second contains G4, <£C, and U3, and (iii)the third includes a3, 4>K, and St-1 (2, 3). ^X174 and G4can grow only on E. coli C, and a 3 infects E. coli B as wellas E. coli C, whereas <f> K and St-1 propagate on E. coli K12(3). Total genome sequencing has been accomplished withthree species <£X174 (4), G4 (5), and a 3 (6). Comparativestudies have indicated that the three phages closely resem-ble each other, and code for the same series of eleven genesA, A% B, C, K, D, E, J, F, G, andH. However, the detailsof the mechanism of the host-range determination are stillunknown (2): no host-range mutants have so far beenisolated from the phages <^X174, G4, and a3 (2), and it isunclear which of virional protein(s) F, G, H, and/or J isdeterminant for the host-range (or adsorption).

As in <£X174, G4, and «3 (2), the life cycle of d>Kconsists of three successive stages (3, 7-9): (i) synthesis ofparental double-stranded (RF) DNA (stage I), (ii) replica-tion of progeny RF DNA by a rolling circle (RC) mechanism

1 Nudeotide sequence data have been registered with EMBL DataLibrary (accession number: X60323).

(stage II), and (iii) RC-type synthesis of progeny viral SSDNA driven by packaging into a virional vesicle composedof <̂ K gene products F, G, and H (stage HI or mor-phogenesis). The stages I and II depend upon host E. coliSSB (7, 10), whereas the stage III is thought to rely on cf>Kgene J protein, instead of SSB (11, 12). However, thefunction of the gene J protein remains to be clarified indetail: e.g., whether or not the gene J protein binds to SSDNA more preferentially over SSB.

In order to obtain further insight into role of the virion-forming proteins F, G, H, and J in determination ofhost-range and interaction with DNA, total genomes of </>Kand its host-range mutant $KhT were sequenced, and a23-aa <£K gene J protein was chemically synthesized. Theresults obtained in this study clearly indicate that the <j>Khost-range is determined by the major (G) and minor (H)spike proteins, and that the synthetic J proteins bind to theviral SS DNA more tightly than and preferentially overSSB in vitro.

MATERIALS AND METHODS

Phages and Bacterial Strains—Bacteriophage <f> K and E.coli K12 W3110, B, C, and their lipopolysaccharide-defective (LPS) mutants (Table I) were from our labora-tory stock (3). <f,K was multiplied in E. coli K12 W3110(7). <f> KhT, which infects E. coli C and B as well as K12 (seebelow), is a host-range mutant spontaneously derived from4>K, and was propagated on E. coli C or K12 W3110. Otherphages, E. coli strains, and plasmids used in this study are

1062 J. Biochem,

Structure and Function of the <pK Viral Gene Products 1063

TABLE I. Strains of E. coli, phages, and plasmids.

Strain

E. coli bacteriaK12 W3110CC61C71C23BBBB1BB5XLl-blue

JM109

PhageflSKtfKhT

M13mpl8M13mpl9M13KO7

PlasmidpUC18pUC118pUC19pUCH9

Properties

Delivative of K12Wild-typeLPS mutant of CLPS mutant of CLPS mutant of CWild-typeLPS mutant of BBLPS mutant of BB8upE44,hsdR17,recAl,endAl,gyrA46,

thi,relAl,lac-,F'[proAB+,lacI'>,lacZ4M15,TnlO(tetT)]

recAl,supE44,endAl,hsdR17,gyrA96,relAl,thi,d(lac-proAB),F'[traD36,proAB+,lacI",lacZAM15]

Wild-typeHost-range mutant of <̂ KWild-typeM13-derived vector for sequencingM13-derived vector for sequencingM13-derived helper phage for

sequencing

2.7-kb, E. coli vector, Apr, lacZ3.2-kb, E. colt vector, Apr, lacZ2.7-kb, E. coli vector, Apr, lacZ3.2-kb, E. coli vector, Apr, lacZ

Source/Reference

Lab. stockLab. stockLab. stockLab. stockLab. stockLab. stockLab. stockLab. stock(19)

(20)

Lab. stockThis studyLab. stock(21)(21)(22)

(21)(21)(21)(21)

LPS, lipopolysaccharide.

summarized in Table I.Enzymes and Biochemicals—Restriction enzymes, phage

T4 DNA ligase, E. coli DNA polymerase I (Klenow frag-ment), and reagents used for dideoxy sequencing werepurchased from CalBiochem. (USA), New England Biolabs(USA), Takara Shuzo (Kyoto), and Nippon Gene (Toyama).Buffers for the enzymes were as recommended by themanufacturers. [a-35S]dCTP was from NEN (USA). Allother materials used in this study were prepared asdescribed previously (6, 13, 14).

DNA Sequencing Analysis—Cloning and sequencing of<f>K DNA was carried out essentially as described byKodaira et al. (6). Restriction enzyme fragments of the <f>KRF DNA were cloned into sequencing vectors (Ml3mpl8,M13mpl9, pUCH8, or pUC119) using E. coli strainXLl-blue or JM109 (Table I). SS and RF DNAs as a directtemplate for sequencing were prepared from these clonesby standard methods (25). Sequencing was performed bythe chain termination method {16) with universal andreverse vector primers.

Biological Assay—Biological properties of phages wereexamined as reported previously (13). For determinationof heat stability, <f> KhT (or <f> K) propagated on E. coli C (orW3110) were incubated at 60'C in dilution fluid (3), and thesurviving fractions were periodically titrated, using C orW3110 as the indicator. For eclipse assay, cells of E. coli C(or W3110) exponentially growing at 3TC in LB broth(Afto'. 0.25) were mixed with chloramphenicol (50 //g/ml).After lOmin, phage was added (input multiplicity: 0.1)with CaCl2 (final concentration: 5mM). At intervals, analiquot was removed, then treated with chloroform, and theremaining free phage was titrated on host strain C orW3110.

Chemical Synthesis of Protein—Gene J protein of <f>Kwas chemically synthesized using an Applied Biosystems

-Peptide Synthesizer {Model-43OA) with a standard-solid-phase peptide technique, and then purified by high-per-formance liquid chromatography. The 23-aa alignmentsynthesized in this study was KKARRSPSRRKGAR-LWYVGGSQF (Fig. 3A). The synthetic protein is solublein H2 O or Tris - HC1 buffer (pH 7.3), and can be stored stablyat 4"C.

RESULTS AND DISCUSSION

Properties of the Host-Range Mutant <f>KhT—<pKinfects E. coli K12 derivatives but not strain C, differing inhost-range from the related phages <pX174, G4, and aZ.Intact particles of <f>K were unable to be adsorbed onto cellsof E. coli C, but the DNAs (SS and RF) efficiently transfectthis strain upon Ca2+-treatment or electroporation (datanot shown). A host-range mutant <f>KhT was spontaneouslyobtained from <£K at approximate frequency of 10~8 on E.coli C. Unlike the parental strain, <j>KhT can infect E. coliB besides C and K12 at almost the same efficiency; the rateof eclipse of <£KhT on K12 W3110 was identical to that of<£K (Fig. la). As to host-range, ^KhT resembles aZ (3),and can be adsorbed onto LPS mutants of E. coli C {e.g.,C61 and C23) and B (BB1 and BB5). The plating efficienciesof <pKhT on these LPS strains are shown in Table II,compared with those of <pK and aZ.

<j>KhT is somewhat unstable; the phage particles weremore sensitive to heat treatment at 60'C in dilution fluid(3) than those of <f>K (Fig. lb). The virion particles of<pKhT are composed of four major proteins F, G, H, and J.By SDS polyacrylamide gel electrophoresis (PAGE), theirmolecular weights (MW) were estimated to be approxi-mately 52.3, 35.8, 21.8, and 2.8 kDa, respectively. Thesevalues are identical to those of the parent ^K (14).

Genome Structures of <f>K and 4>KhT—To identifystructural genes of the four virion-forming proteins (F toJ), the total genomes of <pK and (£KhT were sequenced.After establishment of restriction enzyme cleavage maps(data not shown), appropriate restriction fragments of <f>K(<£KhT) were cloned into the sequencing vector using E.coli strain XLl-blue or JM109 (Table I). SS and RF DNAswere prepared from the resulting recombinant clones, andtheir sequences were determined from both strands of theviral and complementary DNA by the chain terminationmethod (16).

The circular SS DNA of <£K (<£KhT) consists of 6,089 nt(EMBL Data Library accession number X60323), and is 2,53, and 703 nt longer than the genome of «3 (6), G4 (5),and <£X174 (4), respectively. As in the three phages, <f>Khas eleven homologous genes (A, A*, B, C, K, D, E, J, F, G,and H), each of which is encoded in the viral strand DNA,preceded by a possible ribosome-binding sequence (17).The gene organization of <f>K thus established is illustratedin Fig. 2, compared with that of aZ, <£X174, and G4. As awhole, <pK shows an extensive sequence homology (89.3%)with a3, and moderately resembles <^X174 and G4 (ap-proximately 50% homologies). In Fig. 3, the deduced aaalignments of <j>K virion-forming proteins, F (capsid), G(major spike), H (minor spike), and J (core), are shown incomparison with those of aZ, <£X174, and G4.

Properties of the Virional Proteins F, G, H, and J—Like

Vol. 119, No. 6, 1996

1064 K.-I. Kodaira et al.

a3 (6), <f>K gene F codes for a 431-aa protein, which is 4residues longer than that of c£X174 (4) and G4 (5). Thecapsid (F) protein is highly conserved among the fourmicrovirid species (Fig. 3B): the degree of aa similarity is91.2% between <f>K and aZ, 70.8% between <pK and<£X174, and 63.1% between <f>K and G4 (66.0% between<^X174 and G4). Despite the similarity, the three phagesa3, <£X174, and G4 can not substitute <£K gene F proteinfor their own protein in vivo {6).

Among the four virion-forming proteins, the major spikeprotein (G) is most divergent (Fig. 3C). The 4>K gene Gprotein consists of 187 aa, like that of aZ (6), and is 12 and10 residues longer than that of <£X174 (4) and G4 (5),respectively. Concerning aa arrangement, <£K is closelyrelated to a-3 with 88.2% similarity, and is relativelyremote from ^X174 and G4 with 31.6 and 28.3% coinci-dence, respectively (42.0% between <£X174 and G4). The<f>K gene G product functions in a3 in vivo, but not in<£X174 or G4 (6).

<f>K codes for a 332-aa gene H protein, which is 2 and 4residues longer than that of a3 (6) and <^X174 (4),respectively, and 5 residues shorter than that of G4 (5).The aa conservation among the four minor spike proteins is

TABLE n. Plating efficiency of #KhT. Cells of E. coli wereinfected with #KhT, <£K, or a3, and plated at 37'C. E. coli strainsused were as indicated in Table I. Relative plaque yield is presented.

O t _:_ P W «O i l " • ••

E. coli K12W3110

E. coli CCC61C71C23

E. coli BBBBB1BB5

1.0

6.3x10"<1.0xl0"<1.0xl0-

1.1x10'

4.3x10-1.1 x 10-

< 1.0x10"

^KhT

8.2x10"'

1.01.0

-ci.oxio-'6.7x10-'

2.3x10"'2.4X10"'3.0x10"'

<i.oxio-'

1.06.8x10"

<1.0xl0"7.6x10"

8.9x10"<i.oxio-

1.1 x 10"'

as follows (Fig. 3D): 97.3% between ^K and a3, 77.6%between <f>K and <£X174, and 71.7% between <£K and G4(65.5% between ^X174 and G4). Complementation of thegene H proteins has not been detected between $K (a-3)and (£X174 (G4)(6).

In order to elucidate the molecular basis of host-rangedetermination, the total genome sequence of $KhT wascompared with that of ̂ K; c^KhT has two nt changes (Fig.4). One is from G (<£K) to T (<£KhT), located at the 223rdnt in the gene G. Consequently, the 75th aa of the

>

I/)

1

V?

\

\

-

1 1a

\s

\ ' ' '\ ' b

\ " '

-

1 1 1

20 40 0 10 20 30

Time (min)Fig. 1. Eclipse (a) and heat stability (B) of jJKhT mutant, (a) E.coli C (or W3110) cells exponentially growing at 37*C in LB brothwere mixed with chloramphenicol (50^g/ml). After 10 min, phagewas added at a multiplicity of infection of 0.1, together with CaCl2(final concentration: 5 mM). At the indicated time, an aliquot wasremoved and treated with chloroform, and the remaining free phagewas titrated on C (or W3110). O, ^KhT on C; • , tfKhT on W3110;A, wt on W3110. (b) phage particles propagated on E. coli C wereincubated at 60"C in dilution fluid and the surviving fractions wereperiodically determined, using C and K12 W3110 as the indicator. O,<)JKhT on C; • , 0KhT on W3110; A, wt on W3110.

0.3 A(A«) I d D

- O r i

G • H

0 K A ( A M U L 1

-Ori

G M

• O r i

-Ori

7A • ; A(A*>

-Ori

0 1,1 F G H

1000 0( b a s e s )

Fig. 2. Genetic maps of the genomes of<z3, ^K, G4, and #X174. The genetic mapsof a3, <j>K, <f>X174, and G4 are drown toscale from their nucleotide sequences: aZfrom Kodaira et al. (6), <£K from this study,G4 from Godson et al. (5), and ^X174 fromSanger et aL (4). The two replication originsfor viral and complementary strand areshown as + ori and —ori, respectively (6).The dotted lines in gene A of G4 (5) andi#X174 (4) indicate the putative startingpositions for gene A*.

J. Biochem.

Structure and Function of the <f>K Viral Gene Products 1065

(A)

* Ka 35» XG 4

J-chei

HNHK'—'

KKSII

KKKKf

A-BA-BGKKSIRA-S

RSSSBSRSSS

P SBBP SBB

GABPGBPQPLBGTKGKBG- GKS— P- Stl

KGABLVYVGGKGARLVYVGGKGARLVYVGGKGABLWYVGGIGAILVYVGG

(B)

* Ka 3* XG A

HSHHSNHSNHSN

QTQTQTQT

SSGS

A]EA EA EA|D

BBBB

EIVEIVHPHVPH

DLSHLDLSHLDLSHL01.Sill.

FFFF

DCDCLAEA

GGGG

HIHI.QIKI

GRLGRI.GRLGRL

KKIK

TTTT

VV11

ssss

VVTV

TPTPTPTP

AGDSFEAGDSFEAGDSFEAGDSFE

VGAVGAVGAVGA

LLL1

BLSPLBBGLABLSPLBBGLARLSPLBBGLABLSPLBBGLA

PHBHPHRIIPHBHPHBH

VV11

YGYGY(iYG

DDKQ

QVIQWIQVIQVI

QQKN

FHFHFHFH

BRItK

DGVDGVDGVDGV

DNNN

AAAA

SQTS

PI.PPl.PPI.PPLP

SSTP

VVVV

_-N-

TTTT

T--TKYPC--NBYPT--G-YICSSG-V-

DDDD

DIIIIS

AAAA

GYVGYVAFI.AYL

GTIGTIGTIGTI

* Ka 3* XG 4

pppp

K-SNRIA-NNBID-TNK1SSTLKV

PKPKPKPR

FFHF

I.1,1.L

HHFH

QQQQ

SSGG

YLNIYNNYFYLNIYNNYFYLNIYNNYFYLNIYNNYF

BBKK

APVAPVAPVPPV

HPEBMPERHPDRSDDL

TTTT

F.FFY

ANPANPANPANP

SS

S

NNNN

LDEI.NE

— ELNQHPSE---

D1)DD

DSBDABDAR

YYYY

--GF--BF--GFKVGV

RRBB

CCHcr.HCCHVAN

I.KI.KI.KLK

TNNS

* K Va 3 V* X VG 4 V

ssTT

APLPPAPLPPAPLPPAPLPP

EEED

TTTT

KLAEQKLAEEELSKQBTSEN

HMHH

GIESN-G1ESN-TT-STTTT-GT-

SSSS

IDIHGLQAAYA[D1HGLQAAYA[DIHGLQAAYAIDIHGLQAAYA

QQNK

I.HTI.HTI.HTLHT

EK0E

QF.BQEBQF.BQEB

TTDD

YFHYFHYFHYFH

QQQT

BYBYBYBY

BBHB

0DDD

VVV-

IIII

ss-ss-ss-HKE

FGGFGGFGGFGG

SsKH

TSYTSYTSYTSY

HTDFHTDFBSNLBSEF

VASGYDVDGTDQVASGYDVDGTOQVASGYDVDGTOQVASGYDVDGTDQ

SSTS

SLGQFSGBVQQTSLGQFSGBVQQTSLGQFSGSVQQTSLGQFSGRVQQT

VPBFVPBFVPBFVPBF

FFFY

VPEHVPEHVPEHVPEH

GV-GV-GT-BVI

HHHH

NHF-

TLTLTLTL

NLV-ALI-ALV-A-VT

EFPPEFPPRFPPRFPP

ISISTATH

* Ka 3* XG 4

* Ka 3* XG 4

PI.?\.TKEH

F.f.FE

miHHIQHH

Yl.YI.YlYL

-VGBNN-AGKSQNA-KGAVG-KEN

LTYTDLTYTDLTYTDLTYTD

1,1,II

AAAA

GGfiC

DPDPDPDP

AAVA

1.I.1,L

IGIGYGHA

NLPPBENLPPBENLPPBENLPPBE

II1V

SSSS

YQDI.YBDLHKDVLKEF

FFFF

BDG-BPGIRDG-RSGIBSG-DSSKHSSPDSA-

KItKK

I1FF

KKKK

VV1I

AEAEAKAE

SISIGQGQ

VYRVYRVYBVYR

THTHYATQ

PPPP

SN-YKN-FKS-PAAFP-

GFPFGFPFGFPFGFPF

,DI)A.DDAIQF.PYSAL

PIGTTP GT-p ...P ~

Ssss

-GDDTGDN-GD--TE-

1.1.1.L

QKAIQEAIQFRVKDRV

1.1.1.L

IDHNDVBHQDIBHHDVNTNN

YYYY

NAHDACDQCDEI

FQSFQSFQSFQS

QQVH

QI.QLQLQL

I.QI.QI.QAH

VNVNVNVN

NKSH

QQQQ

ARYARYVKFTKF

NNNN

VNVSVTIN

* Ka 3*> XG 4

VYRVYRVYRVYR

HIHHNl.HH

PTPTPTPT

VVTT

BDSIHTSBDSIHTSBDSIHTSBDSIHTS

Fig. 3 (Continued on next page)

minor spike protein is altered to Ser from Ala in <£K; theother one, from T («£K) to C (<^KhT), is at the 200th nt inthe gene H: the aa replacement of minor spike protein (H)is from Val (<£K) to Cys (^KhT). In the capsid (F) and core(J) proteins, c£KhT has no such aa changes. These sequenc-ing results clearly demonstrate that ^KhT is a doublemutant bearing two missense mutations in genes G and H,suggesting that, at least in <f> K, the two spike proteins G andH are involved in the host-range determination. In <ar3 aswell (13), comparable results have been obtained; threemissense strains of gene H (from He to Val at 56th aa, fromGly to Cys at 69th aa, and from Leu to Phe at 71st aa)showed altered heat sensitivities and eclipse rates. Thesefacts suggest that the N-terminal domain of the gene Hprotein (see Fig. 3D) is important for functional and/or

structural interaction among virional proteins F, G, H, andJ (see below). Certain host-range mutants of <£X174 alsohave alterations in genes G and H (18).

In order to discern the contributions of gene G and H tothe altered host range and heat stability, phage strainshaving single missense mutations were constructed byexchanging a PpuMl restriction fragment of 876 bp (fromnucleotide 4059 to 4934, containing the mutated region ofgene G), between ^KhT and wild-type <£K. Thus, each RFIDNA was cut into two fragments and purified by repeatedagarose gel electrophoresis. Ligation of the wild-type largefragment with the <£KhT G segment resulted in an RFImolecule having a single missense G mutation. Conversely,ligation of the <£KhT large fragment with the wild-type Gsegment yielded an RF molecule with a single missense H

Vol. 119, No. 6, 1996

1066 K.-I. Kodaira et al.

(C)

* Ka 3* XG 4

* Ka 3* XQ 4

NMnN

YYFF

gggg

iNTK

VTKHDTAVTKHDTA

R F-8 F-RHNS—NFFKHNAPIN —

SSS

---V--_-V-DKLVLTQLAA

TTTT

GS-ITGNVI-S-VT-K--T-

PPPP

V —A —ASSAVA

APAPAPAP

— TGNITGNI

VLQT---VLSV- —

pppp

VINTGNIVINGGSIKATS-S-NL-SRS-

SLYANLTAGTSSDGSF—NLYANMNVSTSSDGSF—N--AG-NG G-FL-HC--T--TH-SG LCHVVR

IIII

VAHKVVAMKVQ-H--

DDDD

--E

TTTT

sSP-/»—D—PNCVISP-T--D---PNCVISV-N—A-—ANgVVNPTNHHALSIAGSL-

-AGVN-LS-AGVN-LS-VGAD-IANVPADHIA

Til

-AGTSYPIVG-I —-AGTSYPIVG-I —DAD---PKFF-ACL-AI- —R-FEVADG

71-AEHAV —-T -AERAV---T -LYFDSLT-T ILINATTTA

RFESRFESRFES

a 3* XG 4

ASDQASEQSSV-— V-

PTPTPTPT

SI-SI—TLPTAVP-

AAAA

-GSgiEII-GSEVEH-YD-V--LYD-V--

YPYPYPYP

1 — ENSVGS--I —EHSVGS—L---N-—GRHDIETFN- — N—-

G---VCSARG —-VCSARGYYTVK- —KAISFK

0DDD

cvCACVAV

TTTT

VVII

DDDD

IHIHVLSN

PRPRPRPR

AFTSTPTV

GNGNGNGN

NNND

VVVV

FVFVYVYA

GGGG

— VIC--VICFHVW-IHLW-

SSSS

* Ka 3* XG 4

s—s—1K-TNAWT

AAAA

KWRW_-S-

TTTT

SGRVLSG8VIKCR--IS—

GGr,G

TI-ATTTI-ATTLV-SLN-VLSVN

gvgvgvQV

IRINIKNR

EEF,E

YavYgvIICATV

LQPLKLQPLKLQPLKLQPLK

(D)

* Ka 3* XG 4

MMNH

I.I.FF

GfiGG

SIIAWAIASIA

GfilGGIGGIGGI

GAAA

SSRS

SAAA

1.1,1,L

LGASAGAG

GGGG

IAS—AASKLFAHSKI.FLMGKLF

GGGGGGGG

ISSLLNKHFSKNPEHAA-SQRGV — S V H Q Q G A E

G gKAA-G gSAD-

ssss

SA-A-G-T

Q

IQGNVlL

Gg-GTIGMDTGG-GAISHDNNTVG---HG-NVVG —-AN-

* Ka 3* XG 4

A

AA

GIGIGIGI

QgKK

SAIflGSSAIQGSSA1BGSSAIQGS

NNNT

Aty?VPVPVP

-p - -- p - -NPOF.NSQE

AAAA

GQLGOI.A--A —

PPPP

AK--

SSss

-NTQN-FV-Al-

SGSGSGSG

VIIVHAHIL

AAAA

DADAKADS

GGGG

NHIRNAGRANMLRNAGKS—K--G--- —K--H-A

LLLL

__----SS

LLLL

DGDGEG--

TTTT

IQIQLQ-N

AGAGAGAG

-SDKV---SDKV--TSA-VSD--A — N-

ICKKK

* Ka 3* XG 4

EAI.IDKLVQALIDKI.II.-L-D-LVL-N-E-AV

GGGG

-GND--GND-LGGKSI.S-KS

AAAA

KKAS

0DDD

SRKK

GKGKGKGK

AADD

TTTT

DYLAAAFPELNDYLAAAFPELNDYLAAAFPELNOYLAAAFPELN

PPAP

WERAGAWEEAGAWEEAGAWERAGA

GGDG

ASASASAS

TATASASP

GGGG

LESSLESSNYD-MQD-

AAAA

Q—Q--GFEGFQ

NQNgNQN«

ag

-

KE1H[LKE M LKE - LKE - L

* Ra 3* XG 4

KHQLDNQKKHgLDNQKRHQI.DNQKKHflLDNQK

IAIAIAIA

KKF.K

HQHQHQHQ

HNHD----

NNNN

NL QNL QET QET 0

KEKE

IAGIAGIAGLAG

1![L

QSQSQSQS

TSRTSRTSRTSR

ggQE

NTKDNTKDNTKDNTKD

SSQT

VYAgNEHLVYAgNEHLVYAgNEHLVYAgNEHL

* Ka 3* XG 4

11II

LALAHELE

NTNTNTNT

DDNS

1.1.1.1.

TTTTS-T-

KQKQKQKQ

ATH-ATH-QQVSQQTS

EIKEEIHKEIHRF.IHR

_QQ

HHHH

AA--

I.TI.TLTLT

RRQQ

AQAQAgAg

EE

-

TTTT

_AA

GQGgegeg

HIHI.YFYF

TNTNTNTN

SSDD

gigiQIQl

--HALE--HALEKEHTH-KELTR-

KKKK

KK--

VVVV

YYSG

AAAA

EIGKIEIGKIEVDLVDIDAVRANTA

9* Ka 3* XG 4 RT

HIIHII

QDQDggVE

T V -T — -T ....TlDRSK

ggQg

N-S--N-S--N-Q--EVgNS

RY G SS gVTAARY G SS QVTAARV G SS HIGATRY A SS QVGKT

AKDAKDAKDAKD

HIHIVVAl

TDTDTDTD

AS-AA-AA-TAG

ssss

GLA-G-ASG-VVI-V-

DDDD

W ASQQWDS-W VSQQWNS-IFHGID-KAVADTWNN-VFRGADQK-VADVYNNY

KDGKDGKDGKDG

GIGIGIGI

-PI.-SLGS-ES-

NNNN

T-T-LSH-

Fl-RR - KR - K

Fig. 3. Comparisons of the #K virional proteins J, F, G, and Hwith those of a3, ^X174, and G4. Amino acid sequences of <#K, a3,<^X174, and G4 are from this study, Kodaira et aL (6), Sanger et aL(4), and Godson et aL (oh respectively: A, core (J) protein; B, capsid(F) protein; C, major spike (G) protein; D, minor spike (H) protein.

Dashes represent gaps inserted to optimize the protein alignment.Conserved aa residues through the four phage species are boxed. Theaa alignment of synthetic gene J protein (J-chem, see text) is shownin (A). Amino acid changes found in the G and H proteins of <£KhT areindicated by I .

J. Biochem.

Structure and Function of the <j>K Viral Gene Products 1067

( A )

65 70 80 85

S F I V A H K V D T S P A D P ,V C V I S A C V V L STCTTTTATTGTTGCTATG^AGGTTGACACTTCTCCTGCTGACCCCAATTGCGTTATCTCTGCTGGTGTTAACCTTAGT

2 1 0 TS

ofKhT

240

( B )

60 67 70 80

D A G I Q S A I Q G S N V P P A G Q L P A S X T S GGATGCTGGTATTCAGTCTGCTATTCAGGGTTCTAATGTTCCTCCTGCTGGTCAATTGCCCGCTTCTAATACTTCTGGT

190 220

Fig. 4. Nucleotide sequences of the genes G and H of ^K and (JKhT. Relevant parts of nucleotide and amino acid sequences are presented.The nucleotide numbers of the genes G (A) and H (B) are from their starting codons, ATG. The sites of the changed nucleotides and amino acidsin ^KhT are indicated by the arrows ( I ).

mutation. After Ca2+-dependent transfection and singleplaque isolation, the properties of the missense G (msG)strain and the missense H (msH) were tested.

In E. coli C and B, plating efficiency of the msG strainwas, like that of the wild-type <f>K, negligibly low, whereasthe msH mutant infected E. coli C at the relative efficiencyof 10"3 of K12 W3110, and easily reached the level of<f>KhT by further propagation on C cells. As to heatstability at 60"C, the msG strain was indistinguishable fromwild-type <f>K. The heat inactivation profile of the msHmutant was complex: an initial slow decline was followedby a rapid loss of infectivity, suggesting that the msH phagepreparation is somewhat heterogenous. When the msHmutant was repeatedly propagated on E. coli C, the progenyexhibited a heat inactivation profile similar to that of<£KhT. [As can be seen in Fig. lb, the <£KhT preparationcontained a minor fraction relatively resistant to the heattreatment. The ratio of this fraction increased duringstorage at 4'C, accompanied with gradual loss of bulkinfectivity. The nature of this change (confomational?) ispresently unknown.]

These results suggest that plaques of <f> K initially formedon E. coli C were relatively heat-stable msH type withlower plating efficiency on this host, and <t> KhT, which is themsH msG double mutant thermosensitive and highlyinfective to C, was derived from the msH strain duringpropagation on C. At any rate, cells of E. coli C and B wererather insensitive to the single mutants of msH or msGtype.

Like the three phages (a3, <£X174, and G4), the <£K geneJ codes for a small and highly basic core protein of 24-aaidentical to that of a 3 (11), which is shorter than that of<£X174 (4) composed of 38 aa and G4 of 25 aa (5). The fourgene J proteins are structurally similar (Fig. 3D): (i) a basicN-terminal half rich in Lys and Arg (albeit diverged in aasequence), (ii) a conspicuously conserved C-terminal half,whose identical decapeptide KGARLWYVGG is thought to

be a DNA-binding domain (R. McKenna et al., personalcommunication), and (iii) a C-terminal aromatic residue(Phe in <£K, aZ, and #X174; Tyr in G4). As reportedpreviously (6), a3 can use the <£K gene J product in vivo,but not that of G4 or <£X174, and vice versa.

These results on the four virional proteins (F to J)indicate that <£K and a 3 are functionally related, andremote from (£X174 and G4, whereas <£X174 and G4 areevolutionally closer.

DNA-Binding Properties of the Gene J Protein—Thedetailed structure and function of the gene J protein arestill unknown, because of difficulty in preparation of thegene J protein from phage particles in sufficient amountsfor analysis in vitro.

To investigate DNA-binding properties, the gene Jprotein of <f>K was chemically synthesized as described in"MATERIALS AND METHODS." The synthetic protein (term-ed J-chem) is 23 aa long, with the sequence shown in Fig.3A; in the present synthesis, the N-terminal residue Metwas omitted, according to a previous report (4) that thegene J protein isolated from <£X174 particle was devoid ofthe N-terminal residue. The J-chem protein with a calcu-lated MW of 2.7 kDa is soluble in H2O as well in Tris-basedbuffers (e.g., 50 mM Tris-HCl, pH7.3), and shows arandom structural circular dichroism profile with nosignificant bands. Its electrophoretic mobility on a 17%SDS-polyacrylamide gel is the same as that of the naturalgene J protein isolated from 4>K particle (data not shown).

Binding of the J-chem protein to <f>K circular SS DNAwas examined by gel retardation analysis. SS DNA (0.2 //g)was mixed with various amounts of J-chem ranging from 0to 0.2 //g in a reaction buffer (50 mM Tris-HCl, pH 7.3),incubated at 25° C for 30 min, and then electrophoresed ona 1% agarose gel. As shown in Fig. 5 (lanes 1-4), SS DNAincubated with J-chem moved in front of the naked SSDNA, faster than expected from the mass of the complex,and its electrophoretic mobility was accelerated with

Vol. 119, No. 6, 1996

1068 K.-I. Kodaira et al.

7 8 9 10 11 12 13 U

TABLE HI Infectivity of ^K DNA treated with J-chem or SSB.Single- or double-stranded DNA of <f>K was mixed with syntheticgene J protein (J-chem) or SSB in 50 mM Tns-HCl (pH7.3),incubated at 25'C for 30 tnin, and transfected to Ca2+-treated E cohK12 W3110 (7).

Fig 5 Characterization of in vitro interaction between chemi-cally synthesized gene J protein (J-chem) and ^K DNA. a' 4>KSS DNA (0 2 fig) was mixed with J-chem (lane 1, 0 fig, lane 2, 0.05fig; lane 3, 0 1 fig; lane 4, 0.2 fig), incubated at 25" C for 30 mm, andeach mixture was subsequently electrophoresed on a 1% agarose gelb: <j>K SS DNA (0.2 fig) was mixed with SSB (lane 6,0 fig; lane 6,0 1fig; lane 7, 0 2 fig, lane 8, 0 3 fig) and analyzed as in (a) c: <£K SSDNA (0 2 fig) was incubated with SSB (0 1 fig) for 30 mm at 25"C,and then J-chem (lane 9, 0 fig; lane 10, 0.05 fig; lane 11, 0 1 fig; lane12, 0 2 fig) was added After further incubation for 30 nun, eachmixture was analyzed as in (a), d: i^KSSDNA(0 2^g) was incubatedwith J-chem proteins (lane 13,0.05 fig, lane 14, 0.1 //g) for 30 min at25'C, and then SSB (0 I fig) was added. After further incubation for30 min, each mixture was analyzed as in (a).

increasing J-chem concentration. Preliminary studies withfluorescence spectroscopy have shown that an SS DNAmolecule of <f>K seems to be saturated by the J-chemproteins at a weight ratio (J-chem/SS DNA) of 1.0, that is,approximately 600 molecules of J-chem cover one SSDNA. In addition, J-chem bound to <f>K RF DNA tightly,but the complex moved more slowly than the naked RFDNA, as in E. coli SSB (data not shown). These resultsclearly indicate that the gene J proteins bind to both of SSand RF DNAs, and selectively condense SS DNA, but notRF DNA, into a compact form. Hamatake et al. {12)reported that gene J proteins purified from <£X174 parti-cles have an ability to bind to RF as well as to SS DNA.

Binding of SSB to <pK SS DNA was examined under thesame conditions as used for J-chem. SS DNA (0.2 jug) in thepresence of SSB (ranging from 0 to 0.3//g) moved moreslowly than the naked SS DNA, and its mobility decreasedwith increasing SSB (Fig. 5, lanes 5-8), although twocomplexes appeared: their conformations are presentlyunknown. These results suggest that SSB binds tightly toSS DNA, but the complex is not so compact as to movefaster than the naked SS DNA on the gel. To clarifywhether SSB is replaced by gene J protein in morphogene-sis, competition experiments between J-chem and SSBwere performed. <f>K SS DNA (0.2 fig) was incubated withSSB (0.1 fig) in the reaction buffer at 25"C for 30 min, andthen J-chem (from 0.05 to 0.2 fig) was added. After furtherincubation for 30 min, the mixtures were electrophoresedon a 1% agarose gel. As presented in Fig. 5 (lanes 9-11),

Protein

NoneJ-chem

SSB

(Mi)

0.10.20 30.40 50.10.20.30.40.5

RelativeSS-DNA (0 2 fig)

1.02.4x10"'5.0X10-'

< 1.0X10"'——

5 5X10"1

1.7x10"'7 0X10"'

<3 .0xl0- '—

titerDS-DNA(0 4/*g)

101.010101.01.01 01 01.01.01.0

SSB associated with SS DNA is substantially replaced byJ-chem.

Infectivity of SS or RF DNA complexed with J-chem wasinvestigated by Ca2+-dependent transfection assay (7). Asshown in Table in, infectivity of <£K SS DNA wasmarkedly reduced in the presence of J-chem as well as SSB,although the latter was less inhibitory than J-chem,whereas transfecting activity of <f>K RF DNA was notprevented by J-chem or SSB.

The present in vivo and in vitro observations thusstrongly indicate that <f>K nascent SS DNA preferentiallybinds to gene J (core) protein, replacing host SSB, andhence is packaged into a virional vesicle composed of threegene proteins F (capsid), G (major spike), and H (minorspike). It is also evident that two spike proteins G and Hdetermine the host-range of (j>K, through still unidentifiedprotein-protein and protein-LPS interaction(s). Takingadvantage of the c£K host-range mutant and synthetic geneJ protein, further studies on structural and functionalinteractions among the virion-forming proteins are inprogress.

We thank Drs S Aimoto and S. Ono for chemical synthesis ofJ-chem, and Dr. N. Shimamoto for a gift of E. cok SSB.

REFERENCES

1. Kornberg, A. (1980) DNA Replication, WH Freeman, SanFrancisco

2 Baas, P.D. and Jansz, H.S. (1988) Single-stranded DNA phageorigins Curr. Top. Microbiol. Immunol. 136, 31-70

3. Taketo, A. and Kodaira, K.-I (1978) Host-factor requirementsand some properties of a3 and related phages in The Single-Stranded DNA Phages (Denhardt, D.T , Dressier, D , and Ray,D S , eds.) pp. 361-367, Cold Spring Harbor Laboratory, NewYork

4. Sanger, F , Air, G.M., Barrell, B.G., Brown, N.L., Coulson, A.R.,Fiddes, J C, Hutchinson, C A , m, Slocombe, P.M., and Smith,M (1977) Nucleotide sequence of bactenophage <£X174 DNA.Nature 285, 687-695

5. Godson, G N., Fiddes, J.C., Barrell, B G., and Sanger, F (1978)Comparative DNA sequence analysis of the G4 and c£X174genomes in The Single-Stranded DNA Phages (Denhardt, D T.,Dressier, D , and Ray, D S., eds.) pp. 51-86, Cold Spring HarborLaboratory, New York

6. Kodaira, K.-I., Nakano, K., Okada, S., and Taketo, A. (1992)Nucleotide sequence of the genome of the bactenophage aZ:Interrelationship of the genome structure and the gene products

J, Biochem.

Structure and Function of the <j>K Viral Gene Products 1069

with those of the pages, i£X174, G4, and <j>K. Biochim. Biophys.Acta 1130, 277-288

7. Taketo, A. and Kodaira, K.-I. (1978) Growth of single-stranded 15.DNA phages in replication mutants of Escherichia coli. Mol. Gen.Genet. 162, 151-155

8. Sims, J., Capon, D., and Dressier, D. (1979) dnaG (primase)- 16.dependent origins of DNA replication: Nucleotide sequences ofthe negative strand initiation sites of bacteriophages St-1, <f>K,and aZ. J. Biol. Chem. 254, 12615-12628 17.

9. Kodaira, K.-I., Godson, G.N., and Taketo, A. (1995) Compara-tive studies of the minus origin mutants of Escherichia colispherical single-stranded DNA phages. Biochim. Biophys. Acta1260, 191-199 18.

10. Meyer, R. and Laine, P. (1990) The single-stranded DNA-binding protein of Escherichia coli. Microbiol. Rev. 54, 342-380

11. Kodaira, K.-I. and Taketo, A. (1984) Isolation and some prop-erties of bacteriophage aZ gene J mutant. Mol. Gen. Genet. 196, 19.541-543

12. Hamatake, R.K., Aoyama, A., and Hayashi, M. (1985) The Jgene of bacteriophage <f>X174: In vitro analysis of J proteinfunction. J. Virol. 54, 345-350 20.

13. Kodaira, K.-I., Nakano, K., and Taketo, A. (1985) Function andstructure of microvirid phage aZ genome. DNA sequence of Hgene and properties of missense H mutant. Biochim. Biophys. 21.Acta 825, 255-260

14. Kodaira, K.-I. and Taketo, A. (1984) Function and structure of 22.microvirid phage aZ genome I. Electrophoretic characterization

of proteins encoded by wild-type phage. Biochim. Biophys. Acta788, 333-338Sambrook, J., Fritech, E.F., and Maniatis, T. (1989) MolecularCloning: A Laboratory Manual, Cold Spring Harbor Laboratory,Cold Spring Harbor, NYSanger, F., Nicklen, S., and Coulson, A.R. (1977) DNA sequenc-ing with chain-terminating inhibitors. Proc. Natl. Acad. Sci.USA 74, 5463-5467Shine, J. and Dalgarno, L. (1974) The 3'-terminal sequence ofEscherichia coli 16S ribosomal RNA: Complementary to non-sense triplets and ribosome binding sites. Proc. Natl. Acad. Sci.USA 71, 1342-1346Tessman, E.S. and Tessman, I. (1978) The genes of the isometricphages and their functions in 77ie Single-Stranded DNA Phages(Denhardt, D.T., Dressier, D., and Ray, D.S., eds.) pp. 9-29, ColdSpring Harbor Laboratory, New YorkBullock, W.O., Fernandez, J.M., and Short, J.M. (1987) XL1-blue: A high efficiency plasmid transforming recA Escherichiacoli strain with beta-galactosidase selection. Biotechnology 5,376-378Yanish-Perron, C, Vieira, J., and Messing, J. (1985) ImprovedM13 phage cloning vectors and host strains: Nucleotide sequencesof the M13mpl8 and pUC19 vectors. Gene 33, 103-119Messing, J. (1983) New M13 vectors for cloning. MethodsEnzymol. 101, 20-78Vieira, J. and Messing, J. (1987) Production of single-strandedplasmid DNA. Methods Enzymol. 153, 3-11

Vol. 119, No. 6, 1996