Resolution of the DNA strands of the specialized transducing bacteriophage lambda-h80C 1-857 dargF.

JOIURNALOFVIROLOGY,JUlY1975,p.

Copyrigit01975 American Society for Microbiology Printed in U.S.A.Vol. 16, No. 1

Resolution of the DNA

Strands of the

Specialized

Transducing

Bacteriophage Xh80C1857dargF

DON SENS, DON ESHENBAUGH, AND ERIC JAMES*

DepartmentofChemistry, University of South Carolina, Columbia, South Carolina 29208 Received forpublication 14February 1975

The

DNA strands of lambdoid

phages

with deletions or

substitutions of

the

guanine plus

cytosine-rich region

in theleftarmarenotresolvable

by complexing

with

poly UG

followed

by

centrifugation

in CsCl. This work describes a

completely general procedure

for the

strand

resolution

of these

phages by

hybridization

with

fragments

of

separated

strands of the parent

phage.

In

particular,

resolution of the DNA strands of the

specialized transducing phage

XhBOC,857dargF

is described, and

evidence is presented

which indicates that

argF

istranscribed from the rstrand.

Genetic information

is,

with

few

exceptions,

encoded

by double-stranded

DNA with

one

strand

complementary to the other as first

pro-posed

by

Watson

and Crick

(21).

The

specific

information for any given gene is defined

by

one

of

the

two

strands. All

sense

information may be

encoded

by

one

strand

as

in

the

case

of

bacterio-phage T3 and T7 (17, 18); however, in

general

some

genes

are

transcribed

from

one

strand and

other genes

are

read

in

the

opposite direction

from

the other

strand.

This has

been

clearly

demonstrated for the

coliphages

X

and

T4

(5,

19), and in

some cases

particular

regions

are

transcribed from both DNA strands

(1). The

direction of transcription of

a

number

of

genes

in

Escherichia coli has been determined (20),

and of particular

interest

is

the

divergent

tran-scription

of

the

argECBH

genes of the

arginine

regulon.

It

has been shown that argE is

tran-scribed

anticlockwise from

a

control region

between

argE

and

argC

and that the

argCBH

genes

are

transcribed

in

the

opposite direction

from a

closely linked control region (7, 11, 16).

The determination of the orientation of

tran-scription

of

various

genes and the detailed study

of

their

regulation in vitro have been made

possible by

the

availability of

techniques for the

resolution of the DNA strands of various

bacte-riophages.

The method of choice for resolving the DNA

strands of

X,

k80,

and T7 is

to

complex the DNA

with a

ribopolymer, poly UG, followed by

cen-trifugation to equilibrium in a CsCl gradient as

described by

Hradecna and Szybalski (10). In

the case of

X

this

procedure takes

advantage of a

guanine plus cytosine-rich region on the left

arm; however,

X

mutants with

deletions in this

region and

specialized transducing

bacterio-phages

(Xh8OC1857dara

[9],

Xh8OC1857dhisgnd

8g

[22

],

Ah80C 1857dargF1

[E.

James and L. Gorini,

Fed. Proc.

31:3710, 1972 ]) which have this

region

replaced by

bacterial genes have

DNA

which

is

not

amenable

to

strand

separation

by

established

procedures.

We

require

resolved strands of Xh8OC

i-857dargF

for

studying

the

regulation

of

gene

expression

in

the

arginine

biosynthetic

path-way. A

facile technique for

resolving the

DNA

strands of any X80

or X

phage is

described,

and

in

particular

we

report that the

r

strand

of

Xh80C1857dargF

carries the sense

information

for

argF.

MATERIALS AND METHODS

Materials. Ready-Solv IV and VI scintillation fluid and scintillation vials were obtained from Beckman Instruments, Atlanta, Ga. All radioisotopes were purchasedfromNewEnglandNuclearCorp., Boston, Mass.ElectrophoreticallypurifiedDNase and RNase wereobtainedfromWorthingtonBiochemicalsCorp., Freehold, N.J. Selectron B6 filters were obtained from Arthur Thomas. Media was purchased from Difco Laboratories, Detroit, Mich. Cesium chloride was obtained from Columbia Organic Chemicals, Columbia, S.C. Poly UG was purchased for Bio-polymers,Inc. andhad a U-G ratio of 1.9:1.

Media.

Bacteria

used tor preparing phage stocks weregrown in L liquid medium (13) or L agar plates except for the preparation of

32P-labeled

phage when bacteriaweregrowninS medium (Table 1). Bacterio-phageweretitered on H plates in H top agar (9). F top agar wasusedforplating cells on minimal medium (14). Selection plates contained medium A (6) and supplemental growth factors as required, 2% agar, and 0.5% glucose as carbon source. Supplements were usedatthe concentrations previously described (8).

Bacteria and bacteriophage. Bacterial strains and bacteriophage used in this work are listed in Table2.

Propagation of bacteriophage. XCb2was

on November 10, 2019 by guest

http://jvi.asm.org/

86 SENS, ESHENBAUGH, AND JAMES TABLE 1. Composition of Sl mediuma

Contents Concn

Tris-hydrochloride (pH 7.4) 0.1M

NaCl 7mM

NH4Cl 10mM

KH2PO4 3mM

MgSO4 2mM

CaCl2 0.2mM

VitaminB1 4mg/liter

20%Glucose 20mi/liter

a

Si

andS2 medium also contained all amino acids

[image:2.498.72.262.84.185.2]

and nucleic acidbasesatthe concentrationsdescribed (8). S2 medium has thesamecomposition except for the omission ofphosphate.

TABLE 2. Listofstrains

Strain Source

CA 8000 Hfr thi J. Beckwith

EJ 81thi thr leu argFargl Our collection gal lac

EJ 83.1thi thr leu argFargl Our collection gal lac strr

(Gh8OC1857)

(Xh8OC1857dargF)

EJ107thi argI

argR,5

spcr Our collection

ACb2 W.Szybalski

Xh80C1857 Our collection

Xh80C1857dargF Our collection

XC1857S7 N. Kelker

gated by the plate procedure using the sensitive bacterial strain CA8000; Xh8OC1857dargF and XC1857S7were prepared from appropriate lysogenic strains as described by E. James and L. Gorini

(manuscript inpreparation).

Bacteriophage labeled with 32Pwerepreparedfrom lysogens grown in S1 (Table 1) medium with 2 mM KH2PO4.Cellsweregrowntoanopticaldensityat575

nm

(OD.7.)

of 0.40, at which time they were

sedi-mentedby centrifugationat8,000rpmfor4min;the resulting pelletwasrapidlyresuspended in S2(Table

1) medium, grownat42C for15min, transferred to

37C, and grown until confluent lysishad occurred. PhageDNA withaspecific activityof 1.0 x105to7.0

x 106counts permin/gg wasprepared asdescribed

except the Casamino Acids solution used inboth Si and S2 media was treated with 1.90 ml of 1.0 M

calcium chloride per25 ml of 20% Casamino Acids, and precipitated calciumphosphate wasremovedby

centrifugationat6,000xgfor 10min.

Purification ofbacteriophage.Phage lysateswere

clarified by centrifugation at 8,000rpm for 15 min, and phage was sedimented for 12 to 16 h in the

presence of 10% polyethylene glycol 6000 (Arthur Thomas)and 0.5 MNaClasdescribedbyYamamoto

etal. (24).The sedimentwascollected by

centrifuga-tion at 5,000 rpm for 5 min and the pellet was

carefully resuspended in l/5o of the original lysate

volume of T1buffer (6 x 10-4MMgSO4,5 x 10-4M CaCla, 6 x 10- M Tris-hydrochloride, p-H 7.3, 0.1% [wt/vol] gelatin). The concentrated phage was sedi-mentedfor2h at 30,000 rpm into acushion compris-ing 2ml of1.4-density CsCl above 2 ml of1.6-density CsClusinga 50.2Tirotorin the Beckman ultracentri-fuge.Phagewerefurtherpurified byrepeated centrif-ugation in a cesium chloridedensity gradient using 1.4, 1.5, 1.6block gradients for a period of4to 6h, 1.4, 1.6 block gradients for 12 to 16 h, or running to equilibrium in 1.5-density CsCl using either the 50.2 Ti, 50Ti, or 75 Ti angle rotors. Aftercentrifugation, gradients were harvested by upward displacementby fluorientFC 40(Minnesota Mining and Manufactur-ing Co., St. Paul, Minn.) with an ISCO gradient fractionator coupled to a Chromatronix dual-wave-length absorbance monitor with an 0.5-mm path length flow cell.

Resolutionof phage DNA strands by complexing with poly UG. Strand resolution of phage DNAby complexing with poly UGwasperformedasdescribed by Hradecna and Szybalski (10) and Szybalski (per-sonalcommunication).Cesiumchloride was added to high puritywatertogiveasaturated solution at 60C, and thewarmsolutionwasrapidly filtered through a

0.45-jum

membrane filter (Millipore).Saturated CsCl solutions hadan

OD2,0

ofless than 0.010.Polyallomer centrifuge tubes (% by3 inch; ca. 1.7by7.6cm)were placedin 1literof0.1MEDTA, pH10.0, and heated to 100C for15min; thetubeswerewashed withglass distilled water and stored at room temperature. A quantity of phage containing 350 to 450 jtg of DNA, freshly purified in a CsCl gradient, with an

OD2,0

greaterthan60,wasplacedintoadialysis bag,

both endswereclosed withhemostats,and thephage wasdialyzed for2 hagainst 1 liter ofphage dialysis bufferat 4C. Thephageweretransferredto a testtube (16by125mm), and the volumewasadjustedto1.83 ml withdialysis buffer;0.50ml ofanaqueous solution ofpoly UG (1 mg/ml)and0.015 ml of 10%Sarkosyl NL97wereadded,and the mixturewasheated for2.5 minat95Cin ahotwaterbath withgentleswirlingfor the first 30 s.The mixturewasrapidlycooledto 5C followed by the addition of 0.45 ml of 0.5 M Tris-hydrochloride (pH 7.6) and 10.9 ml of saturated CsCl. The density was adjusted to 1.725 (refractive index 1.4022), and the mixture was centrifugedina 75 Ti or 50Tirotor at35,000 rpmfor67 to 72hat 17C. At the termination of the run each preparation was frac-tionatedasdescribed for the isolation of phage.

The separated strands were dialyzed overnight against140mMphosphate buffer (pH 7.6),adjustedto 0.3 NKOH, and incubatedfor 12hat 37C toremove polyUG. Thereactionmixture was neutralized with 3 NHCl, annealed at 68Cfor 3 handpassed througha 5-mlhydroxyapatite column at 60C to remove dou-ble-stranded DNA. Single-stranded DNA was dialyzed against 1mM EDTA(pH 8.0) and storedat 4C over adropofchloroform.

Isolation of DNA from the phage

Xh8OC1857dargF.

Bacteriophage was purified

imme-diately priortoisolationofDNAby centrifugationto equilibrium in1.5-density cesiumchloride, and puri-fied phage was adjusted with 1.5-density cesium J. VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:2.498.69.265.246.387.2]

STRAND SEPARATION OF Xh8OC1857dargF 87 chloride toyieldaconcentration correspondingto an

OD2,,0

of 13.The phage solution was dialyzed against 1

xSSC(0.15 MNaClplus0.015M sodiumcitrate, pH 7.4) for 3 h without stirring, and the phage was extractedthree times with an equal volume of redis-tilled phenol saturated with 1 x SSC. For the first extraction 15

Al

of a 25% solution of sodium dodecyl sulfate wasadded per ml. The third phenol extraction was followed by extraction with chloroform-isoamyl alcohol (24:1, vol/vol); the aqueous phase was ad-justed to 25

Ag/ml

inPronase andincubated for2h at 37C. Two additional phenol extractions were per-formedfollowed by a final chloroform-isoamyl alcohol extraction. The aqueous layer was dialyzed exhaus-tively against 10 mMEDTA, pH 8.0, and stored at 4Coveradropofchloroform.

Preparation of 4 to 5S siingle-stranded DNA fragments. A quantity ofsingle-strandedDNA, usu-ally 800Mg, in 13ml of 1 mM EDTA was sonically treated with a Bronson Heat Systems sonifier at a power output of 85 W using four pulses of 30-s duration with 5-min cooling at 0C between each pulse. The fragments wereconcentrated to avolume of 3 ml by centrifugal lyophilization in a silanized corextube, and 0.5-mlaliquotswerelayeredon a 5 to 30%sucrosegradient (0.5 MNaCl-1mMEDTA)with a0.5-mlcushion of 50% sucrose and centrifuged in an SW50.1rotorat45,000 rpmto anw2tvalue of 55,000. Thegradientswerefractionated from thebottom;the upper 2.0 mlcontained4 to5S fragments whichwere pooled,concentratedby centrifugal lyophilizationto a

volume of 5 ml, and desalted by gel filtration on a Sephadex G15 column equilibrated with 1 mM EDTA. DNA larger than4 to5S,recovered from the bottom of the gradient, was used in subsequent solification experiments.

Hydroxyapatite chromatography.Denatured and native DNA were separated by differential ad-sorption to and elution from hydroxyapatite (Bio-Rad, Richmond, Calif., Bio Gel control no. 11704) according to Kohne and Britten (12). Columns of hydroxyapatite were poured in disposable plastic syringebarrels ina60C water bath (J. F.Sambrook, personal communication). Mixtures of native and renatured DNAwere applied in 140 mM phosphate buffer(pH 7.0).Single-strandedDNApassedthrough the column, which was washed with eight column volumes of 140 mM phosphate buffer. Duplex DNA was eluted with six column volumes of 400 mM phosphate buffer.

Determination of radioactivity. Bovine serum albumin was addedascarrier forsampleswhichwere precipitated with 10% trichloroacetic acid. The pre-cipitates were collectedon glass fiber filters (What-man), dried, andcounted inReady-SolvIV scintilla-tion fluid. Samples containing 32p were counted without fluorby Cerenkov radiationinaBeckmanLS 230scintillation spectrometer.

Hybridization conditions. Allhybridizationswere

performed at68C in 140mMphosphate buffer(pH 7.6)-0.4% sodium dodecyl sulfate. Native DNA was denatured at low salt concentration by heating to 97C for 4 min, quick cooled, adjusted to 140 mM phosphate, and passed through a hydroxyapatite

column to remove any tangled DNA. In experiments to determine the kinetics of reassociation, samples were withdrawn from the annealing mixture at appropriate times,diluted 10-fold into 140 mM phos-phate buffer, and storedat 4 C until the proportion of single- anddouble-stranded DNA was determined by chromatography on hydroxyapatite.

Preparation of argF mRNA. Strain EJ107 was grown at 37 C in S1 medium without uridine to a concentration ofapproximately 5 x 108cells/ml, at which time the culture was pulsed for 50 s by the addition of 4uCi of [3HJuridineper ml. At the end of the pulse period a freshly prepared solution of sodium azide was added to 0.02 M final concentration, and the culture was immediately poured on to the same volume of frozen crushed buffer containing 20 mM Tris-hydrochloride (pH 7.3)-5 mM MgCl2. The cells were allowed to thaw,concentrated by centrifugation for 10 min at 8,000 rpm, and resuspended in a 0.1 volume of 0.1 x SSC(pH 7.0). The cells were lysed by a single passage through a French press at 8,000 lb/in2.Bentonite was immediately added to the lysate to afinal concentration of 0.01 mg/ml, the mixture wasstirred gently at 4 C for 5 to 10 min,and bentonite was removed by centrifugation for 10min at 13,000 rpm at4C.

DNase(RNase free) was added to the supernatant liquid at a concentration of 20Mg/ml,and the mixture wasincubated at room temperature for 15 min. The solution of RNAwasthenextracted three times with phenol saturated with 0.1 xSSC. The first extraction was performed in the presence of 0.4% SDS. After each extraction the mixture was centrifuged at 6,000 rpm for 5 min to facilitatephase separation. After the last phenol extraction, residual phenol was removed from the aqueous phase by extraction with chloro-form-isoamyl alcohol (24:1, vol/vol). The aqueous phase was heated in aboilingwaterbath for 4 min and quick chilled, and 20 x SSC was added to give an overall concentration of 2 x SSC. The cold solution was passed through a double layer of nitrocellulose filters. To the filtered solution a0.1 volume of 20% sodium acetate (pH 5.2) and2volumes of ethanol (at -20C)wereadded, and precipitationwasallowedto occurovernight at -20 C.

TheRNA precipitate wascollected by centrifuga-tion and resuspended in 0.1 x SSC and passed throughaSephadexG-50(fine) column (0.9by30cm, equilibratedwith0.1 x SSC).RNA elutedinthe void volume and was concentrated by ethanol precipita-tion. RNA was resuspended in 2 x SSC, saturated with phenol, and stored at4C.

RESULTS

The resolution of the DNA strands of

bacte-riophage

X

and

080

is

readily affected by

com-plexing

with

poly UG

followed

by centrifugation

to

equilibrium

in a

1.725-density

cesium

chlo-ride

gradient

as shown in

Fig.

la and b for

XC1857S7 and

XCb2,

respectively. Separations

such

as those shown in

Fig.

la

and b

are efficacious due to the specific complexing of VOL.16,1975

on November 10, 2019 by guest

http://jvi.asm.org/

88 SENS,ESHENBAUGH, AND JAMES

4 3 2

9 (0

I-a 2

0

I

0L

0

0

4

3

2

0 10 20 30

FRACTION NUMBER

FIG. 1. Comparison of efficacy of complexing with poly UGfollowed by centrifugationtoequilibrium in 1.710-density cesium choride for the resolution of the DNA strands ofvarious bacteriophage. (a) XCb2; (b)

Xh8OC1857; (c) Wh8OC1857dargF.

poly UG with the guanine plus cytosine-rich region ofthe left arm ofthe

bacteriophage.

In

thosecases inwhich this guanineplus

cytosine-rich region is removed by a deletion or

substi-tution, as in the case of

080dargF

and

Xh8OC1857dargF,

the guanineplus cytosine-rich

region is no

longer

available to complex poly UG.The DNAisolated from suchphages isnot amenable to strand separation using poly UG

(Fig.

lc). The

work of

Westphal

(22)

and of Sambrook et al.

(16)

with simian virus 40 and

the

following

considerations

suggest

an

alter-nate

procedure

for

resolving

the DNA

strands

of

Ah8OC,857dargF.

(i)

DNA isolated from

Xh8OC1857

and

080

may

be

readily

resolved

using

poly

UG.

(ii)

Single-stranded Xl fragments

should

com-pete

for the

Xh8OCidargFr

strand in a

reas-sociation

experiment

with

denatured

Xh8OC1857dargF.

(iii)

The rate of

reassociation

of

high

concen-trations

of Xl

fragments with the

complemen-tary

strand

of

Xh8OC1857dargF should be

faster

than that of the

corresponding

strand of

Xh8OC1857dargF,

in

direct proportion

to the

increased concentration of

fragments relative

to

Ah8OC1857dargF.

(iv) The

unit

length-fragment DNA hybrid

duplex

may

be

separated from single-stranded

DNA

by chromatography

on

hydroxyapatite.

(v) Single-stranded fragments

of 4 to5S can

be

readily

separated

from unit

length

Xh8OC1857dargF DNA by

centrifugation in

a

sucrose

density gradient.

If

unit

length-denatured Xh8OC1857dargF

DNA

was

hybridized

to

either

Xh8OC1857

ror

I

strand

fragments

of 4 to

5S,

only

one

of the

DNA strands

of

the specialized

phage should

hybridize

to

the

4 to

5S-resolved DNA. By

performing the

reaction

with

a

large

excess

of

4

to

5S

fragments, the

rate

of

hybridization

of

fragments

to

their

complement

should

be

con-siderably

faster than

that

of

the

homologous

Xh8OC1857dargF

r or I

strand;

excess

single-stranded

fragments

together with the

same

strand

of

Xh8OC 1857dargF should

not

be

re-tained when

chromatographed

on

hydroxy-apatite at 140

mM

phosphate, whereas unit

length-fragment hybrid DNA would be retained

and

eluted

at 400mM

phosphate.

Data

are

presented

in

Fig.

2

which

summarize

the

results

of a

number

of

reassociation

experi-ments in

which

3H-labeled

unit

length

Xh8OC

,dargF

DNA

was

denatured

and

per-mitted toreassociate in thepresence of various concentrations of 4 to

5S fragmented

r

strand

DNA isolated

fromXCb2. The fraction of

radio-active

single-stranded

DNA present after

hy-bridization had been permitted

to

proceed

for

varying

times was ascertained

by

chromatog-raphy

on

hydroxyapatite. Increasing

concentra-tionsofXCb2fragments lead toadecreaseinthe percentage of

labeled

duplex DNA, but

even with a 30-fold excess offragments less than 50%

radioactive-labeled

single-stranded DNA was

present;

however,

the use of

Xh80C1857

frag-ments ora mixture offragments from

XCb2

and J.VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:4.498.75.261.49.505.2]

STRAND Xh8QC,57dargF 89

and

the solution was rapidly cooled

to 0

C and

applied to a 3-ml hydroxyapatite column

main-tained at 60

C, with occasional

stirring of

40

p:)

the

hydroxyapatite

bed. The fraction

passing

40

through the column contained

more

than 90%

of the single-stranded DNA;

the

remaining

49

/single-stranded

DNA

eluted with

10

ml of

140

cz

30

X 30

-0~~~~~~~~~~~~~~~~~10

*C 20 -

75-.0

0 . * a 2

0 5 10 IS 20

ACb2

(pg/mi)

FIG. 2. Relationship between

'H-labeled

single- 01 10 10 100

stranded

Ah8OC,857dargF

present and concentration Tira. (hr)

of added single-stranded XCb2 fragments. Xh8OC1- FIG. 3. Rate of conversion of single-stranded

857dargF

DNA (1.0

ug)

wasdenatured in 0.5 ml of 1

Xh8OC1857dargF

DNA into duplex DNA in the pres-mM EDTA by heating at 97 C for 5

min

and adjusted ence of a

30-fold

excess ofsingle-strandedXh8OC,8574 to a total volume of 2

ml

in 140 mMphosphate buffer to 5Sfragments.

Xh8OC1857dargF

DNA (2.0

Mg)

was containing various concentrations of 4 to 5S XCb2 denatured in I ml of 1 mMEDTAbyheatingat97C single-strandedDNA. The mixture was incubated at for 5min and adjusted to a totalvolumeof 4 mlin 140 68 C for 60 h, at which time 1 ml was removed, diluted mM phosphate buffercontaininga30-fold excessof 4 with 9 ml of 140 mM phosphate buffer-0.4% sodium to 5S

Xh8OC1857,

strand DNA. The mixture was dodecyt sulfate, and stored at 4 C until assayed for incubated at 68 C. Samples(0.5 ml) wereremovedat single-stranded

'H-labeled

DNA by chromatography intervals, diluted with 3.5 mlof 140 mMphosphate on hydroxyapaptite. buffer-0.4%sodiumdodecylsulfate, andstoredat 4 C until assayed forsingle-stranded'H-labeled DNA by

480 led to the rapid reassociation of 50% of the

chromatography

on

hydroxyapatite.

labeled DNA to the duplex form (Fig. 3 and 4)

with 50% of the radioactive-labeled single-

100

stranded DNA present. These data clearly

indi-cate that annealing denatured DNA, isolated

from a specialized transducing bacteriophage,

It

75

-with a 30-fold excess of single-stranded frag-

a

ments from an appropriate phage, should per-

C

mit separation of the strands of the specialized

50

-phage.

Preparation

of separated

strands

of

,,

Xh8OC1857dargF.

The strands of

Xh80C1857-

2-dargF

are

separated routinely by using the

following protocol: 5

ug

of

Xh8OC1857dargF

DNA was diluted to 4.0 ml with sterile

water,

heated at 97C for 5 min, and immediately

°01

lo

10 100

cooled to

0 C.

Purifieca

4 to

5S fragmented

Time

(hr)

DNA(150

M&g)

isolated from

Ah80C1857

1strand FIG. 4. Rate of conversion of single-stranded

was

added,

and the mixture was adjusted

h0C0857dargF

DNA

into duplex DNA

in

the

pres-to 140 mM phosphate (pH 7.6)-0.4% sodium ence of a

20-fold

excessof

single-stranded

Xh8OC,8574

dodecyl sulfate in a total volume of 10 ml. to 5S fragmentsanda10-fold

excess

of

XCb2.

Experi-The mixture was incubated at 68 C for 1.5 h, mentalconditionsas

described in

the

legend

to

Fig. 3.

VOL.16,1975

on November 10, 2019 by guest

http://jvi.asm.org/

[image:5.498.39.222.63.307.2] [image:5.498.237.431.167.323.2] [image:5.498.240.429.463.615.2]

90

SENS,ESHENBAUGH, ANDJAMES mM phosphate buffer. Double-stranded DNA waseluted from the hydroxyapatite column with 6

ml

of 400

mM phosphate buffer

(pH 7.6) with 0.4% sodium dodecyl sulfate; an addi-tional 8 ml of the same buffer removed more

than

95% of

the double-stranded

DNA.

Radioactive DNA was equally* divided be-tween

the

fractions

eluting

at 140 mM

and

400

mM

phosphate, and the

total

yield

wasgreater

than

95%.

The fraction eluting

at 140 mM

phosphate was concentrated to dryness

by

cen-trifugal lyophilization,

resuspended

in 3 ml of sterile glass distilled water,

and

desalted on a

column

(1.5

by

20cm) of

Sephadex G15

equili-brated

with 1 mM EDTA (pH 8.0). DNA was

eluted

in

the void volume and concentrated

to a

volume of 2 ml

by

centrifugal lyophilization.

Aliquots

of 0.5 ml were

layered

on 5 to 30%

sucrose

gradients with

a

0.25-ml

cushion

of 50% sucrose and

centrifuged

at42,000 rpm in an

SW50.1

rotor to an w2t

value

of 55,000.

The

sucrose

gradients

were

fractionated

from

the

bottom

into

three

fractions;

the lower

2.5 ml

contained Xh8OC 1857dargF light strand, the

next 0.5 ml

contained

no DNA, and the upper

2.5

ml

of

gradient

contained

Xh8OC,857

I

strand

4 to

5S fragments.

Fractions

containing

Xh80C1857dargF light strand

were

pooled,

250

,ug

of

poly(A) carrier

was

added,

and the solution

was

concentrated

to a

volume

of 7

ml

by

centrifugal lyophilization, desalted

as

de-scribed, and again concentrated

to a

volume

of 3

ml.

The sample

was

stored

at 4

C

over a

drop

of

chloroform with

a

final

yield

of77%.

Fractions

from

the

upper 40% of

the gradients

were

pooled, concentrated by centrifugal

lyophiliza-tion,

desalted, and reconcentrated

as

described

to

yield

88% of

the

fragments

originally used

in

the

experiment.

The

fraction

eluting

in400

mM

phosphate

was

concentrated, desalted, and

con-centrated

as

described. Aliquots

of 0.5 ml were

layered

on a 5 to 30%

alkaline

sucrose

gradient

(0.5 M

NaCl,

0.2M

NaOH,

1

mM

EDTA) with

a

0.25-ml cushion of

50%

sucrose

and

cen-trifuged at 42,000 rpm at 4 C in an SW50.1 rotor to an

w2t value

of55,000. The lower 2.5 ml of the gradients was pooled and neutralized with 1.0 N

HCl,

and 250 ug of

poly

A carrier was

added,

and the mixture was concentrated,

desalted,

and concentrated once more as described to give a final recovery of 82% of

Xh8OC,857dargF

r strand. The individual resolved strands were

unit

length

as

demonstrated

in

Fig.

5 inwhich

the

sedimentation

profile

of Xh8OC

,857dargF

freshly prepared

by phenol

extraction

(Fig. 5a),

after resolution of the I strand from the 140mM fraction

(Fig.

5b),

and after resolution of the r strand from the 400 mM fraction

(Fig. 5c)

are

ro

0

E

0. 3o

3

2

0 5 10 15 ZO 0 5

FRACTION NUMBER

FIG. 5. Sedimentation of XhOC,857dargF single-stranded DNA through alkaline sucrose gradients.

Samples of 32P-labeledIandrstrandXh8OC,857dargF

DNA were sedimented through a5 to 30% alkaline

sucrosegradient at42,000rpm at4CinanSW50.1

rotor toanW2t valueof55,000.Fractions(0.2ml)were

collected, and the 32Pwas determined by measuring

Cerenkov radiation. (a) Xh80C,857dargF freshly

pre-pared by phenol extraction; (b)Istrand obtainedfrom 140 mM phosphate fraction; (c) r strand obtained

from 400 mM phosphate fraction; (d) I strand

Xh8OC1857dargF after self-annealing for 16hat68C in 0.2 MNaCl.

compared by sedimentation through a 5to30% alkaline sucrose gradient (0.5 M NaCl-0.2 M

NaOH).

Each of the resolved fractions was

self-annealed for 16 h at 68 C in 0.2 M NaCl to permit any contaminating complementary strandtoreanneal; the phosphate concentration

wasadjusted to 140mM, and each fractionwas

chromatographed on hydroxyapatite at 60C. More than 98% ofthe DNApassed through the column, indicating that therewas insignificant

cross-contamination with the complementary strand. A sample of the single-stranded DNA self-annealed as described was sedimented

through an alkaline sucrose gradient as

de-scribed (Fig. 5d). In view of the considerable thermal shearing obtained after self-annealing at68C andbecause ofnegligible cross-contami-nation of one resolved strand with the other,

this step of the preparation procedure was

usually omitted.

We have performed twoexperiments to

con-firm that the 140and400mMfractions indeed constitute resolved strands of

Xh8OC1857dargF.

(i) Self- and cross-hybridization. When

d

c

J. VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:6.498.281.467.76.282.2]

STRAND SEPARATION OF

Xh8OC,857dargF

91

Xh8OC1857dargF

I

and

rstrand DNAwere

incu-bated separately at 68 C in 140 mM phosphate bufferat aconcentration of 0.5 Mg ofDNA/ml,

there was no conversion to a double-stranded

form detectable bychromatographyon

hydroxy-apatite; however, a mixture containing equal

proportions of both strands renatured with second-order kinetics (Fig. 6).

(ii) Hybridization of argF mRNA to

Xh-80C1857dargFrand

I

strands. Purified mRNA wasprepared fromastrain making constitutive

levels of argF message and ornithine transcar-bamylase asdescribed. Six aliquots, each

com-prising 3.05 x 106 counts/min of [3H

]uridine-labeled RNA, were

hybridized

to 5 ,gof

dena-tured Xh8OC1857dargF DNA immobilized

on a

nitrocellulose filter. One filterwaswashed, and

boundradioactivity wascounted in Ready-Solv

IV; RNA was eluted from the other filters as

describedby B0vre and Szybalski (2). Approxi-mately 0.2% of the RNA input was recovered

from specific hybridization todenatured DNA

isolated from

Xh8OC1857dargF (Table 3). In

100

75

z

C

50

25

0o1 10 10 100

Time(hr)

FIG. 6. Reassociation kinetics of I and r strand

Xh8OC,dargF DNA. Hybridization solutions

con-tainedatotalof2.5 Mgof strand DNA(U),rstrand DNA (A),or amixtureof equal proportion of andr

strand DNA (0) in 2.5 ml of 140 mMphosphate buffer-0.4% soldium dodecyl sulfate. After various timesof incubationat68C, samples (0.250 ml)were

diluted into2.5 mlof140mMphosphate buffer-0.4% sodium dodecyl sulfate, and thequantity of single-stranded and nativeDNAwasdeterminedby chroma-tographyonhydroxyapaptite.

constrast to

these

data, when mRNA

was

pre-pared under conditions of physiological

repres-sion

from

CA8000,

a

strain

carrying

the argR+

allele, 0.06% of the RNA input was found to

hybridize

specifically to DNA isolated from

Xh8OC1857dargF (Table

4). The purity of argF

mRNA isolated under

conditons of constitutive

synthesis

was

demonstrated

by rehybridization

to

denatured

Xh8OC,857dargF

DNA and to

'80

DNA

immobilized

on nitrocellulose filters. It

was

found that

93% of

this

input radioactivity

bound specifically

to

argF-containing DNA,

whereas

only

2%

bound

to

080

DNA.

Hybridiza-tion of

1,400

counts/min

of this

purified argF

mRNA to 0.33

,ug

of

Xh8OC1857dargF

r strand

and

l

strand, isolated

as

described, resulted

in

76%

of the

input

radioactivity

hybridizing

spe-cifically

to

the

r

strand and

9% to

the

l

strand

(Table

5).

DISCUSSION

The

results

presented

in this work show

that

it is

possible

to separate the strands of the

specialized

transducing

bacteriophage

Ah8OC 1857dargF utilizing

a procedure

which

takes

no cognizance ofthe specific parts of

the

molecules which

are

either

present or

deleted. It

is,

therefore, possible

toresolve the strands of

[image:7.498.43.224.327.482.2]

any

deletion

or

substitution

mutant

of

X or

080

TABLE 3. Purification ofargFmRNAisolatedfrom EJ107

counts/min counts/min

TotarRNA

bound to: % bound to:

input('H

counts/min) argF 080DNA

argF

480DNA

DNA DNA

3,054,800 6,210

1227

10.2 10.007

TABLE 4. Isolationof argFmRNAfrom CA8000 grown in the presenceof 100

Mg

ofarginineper ml

'Hcounts/min 'Hcounts/min

bound to: %bound to: Total RNA

input

argF

4)80DNA

argF

O8DNA

DNA DNA

901,000 564 49 0.06 0.005

TABLE 5. Hybridization of purified argFmRNAtoargFand

480

DNA

PurifiedargF 'Hcounts/minbound to: %'Hcounts/minbound to: mRNA'H

counts/min

input argFDNA 480 DNA argFr argF I argF DNA O80DNA argFr argF I

1,400 1,290 330 | 1,060 | 130 93 | 2 | 76 99

VOL.16, 1975

on November 10, 2019 by guest

http://jvi.asm.org/

[image:7.498.243.434.390.495.2]

92 SENS,ESHENBAUGH, AND JAMES

by

use ofthe technique

described

using

single-stranded

fragments

with

an

S value

of

between

4

and

5

derived

from one or more

lambdoid

phages to

provide

the maximum

homology

be-tween the fragments

and

the genome to

be

resolved. The results

obtained

using

XCb2

frag-ments to

resolve

a

Xh8O

hybrid phage indicated

that the choice

of fragments

isolated

from

another source

should be

more efficacious. The

use of

resolved

fragments of

Xh8OC,857

or a

mixture of

XCb2 and

080

fragments was

found

to provide a statisfactory solution to the

prob-lem. An important aspect of this work is the

applicability

of

the

procedure

we have

de-scribed

for the

resolution

of the DNA

strands

of

specific DNA species

obtained

after

digestion

with the

endonuclease

Eco

RI

of DNA

isolated

from a specialized transducing

bacteriophage.

The onlymodification

required

for

this

purpose isthe sucrose gradient centrifugation step

used

for

the

separation of

the

resolved

restricted

DNA

strand

from excess

single-stranded

frag-ments.

It is

possible

that

duplex

DNA

comprising

unit

length Xh8OC 1857dargF

r

strand and

Xh8OC,857

I

strand

fragments could be

sepa-rated

from excess

single-stranded

fragments

and

Xh8OC1857dargF

I

strand

by centrifugation

to

equilibrium

in a

1.7250-density cesium

chlo-ride density

gradient; however,

the

use of

chro-matography

on

hydroxyapatite

is

much

more

rapid,

considerably

less expensive,

and

un-doubtedly leads

to a

cleaner

separation.

The

data

presented

in

Table

4

clearly

indi-cate that

control

of

the

argF

operon in

the

arginine

biosynthetic regulon

is

effected,

at

least in part, at

the

transcriptional level. This

is

consistent with

data presented by

Pouwels

etal.

(15)

for

the

argECBH

genes of

the arginine

regulon.

These workers

demonstrated

that

argECBH

mRNA

levels

were

between

0.02 tc

0.036%

under conditions

of

physiological

re-pression

compared

toa

constitutive

level of

0.23

to 0.42% of

the

total RNA

synthesized

in a 50-s pulse of

[3Hluridine

asjudged by studies

of mRNA

hybridized

to DNA isolated from a

080

transducing

phage

carrying the genes ppc

and

argECBH.

It is

clear

that synthesis of

mRNA from

bacterial operons neighboring argF

is

observed

under conditions of

physiological

derepression of the arginine biosynthetic

regu-lon;

however,

the data shown in

Table

4indicate

that at least 70% of the mRNA,

synthesized

inastraincarrying the argR- allele, which binds

specifically

to DNA isolated from the

phage

Xh8OC

,857dargF

is indeed argF mRNA. This

is

entirely

consistent with the observation

(Table

5) that

76%

of

purified

argF

mRNA

binds specifically with

XhSOC1857dargF

r

strand

DNA

and

clearly

indicates

that

the

Xh8OC

,-857dargF

r

strand

carries sense information

for the

argF

gene.

ACKNOWLEDGMENTS

We thank JoeSambrookforintroducing usto hydroxyapa-titefor thechromatographyofDNA andMichael Vodkinfor

numerousdiscussions. We also thank Dianne Winburn and Philip James for excellenttechnical assistance. Weare most

grateful to the 3M Company for the gift of fluorinertFC40. This work was supported by the Research Corporation Brown-Hazen BH-853 grant and American Cancer Society grantVC 131 provided to Eric James.

LITERATURE CITED

1. Bovre, K., and W. Szybalski. 1969. Patterns of conver-gentandoverlapping transcription within the b2 region ofcoliphageA.Virology 38:614-626.

2. Bovre,K.,and W.Szybalski.1971.Multistep DNA-RNA hybridization techniques, p.350-383.In L. Grossman

and K. Moldave(ed.), Methodsinenzymology, vol.21.

Academic Press Inc., New York.

3. Campbell,A. 1963.Segregantsfromlysogenic

heterogen-otescarrying recombinant lambda prophages. Virology 20:344-356.

4. Campbell, A.1964. Genetic recombination betweenXdg

phage.Virology 23:234-251.

5. Cohen,S. N., and J. Hurwitz. 1968. Genetic transcription

in bacteriophage A: studies of A mRNAsynthesis in vivo. J. Mol. Biol. 37:387-406.

6. Davis, D. D. 1950.Mutants ofEscherichia colirequiring

methionineorvitamin B12.J. Bacteriol. 60:17-28.

7. Elseviers, D., R., Cunin, N. Glansdorff, S. Baumberg, and E. Ashcroft.1972.Control regions within the arg-ECBH gene cluster of Escherichia coli K12. Mol.Gen.

Genet. 117:349-366.

8. Eshenbaugh, D., D. Sens, and E.James. 1974.A solid phase radioimmune assayforornithine

transcarbamyl-ase, p. 61-73. In R. B. Dunlap (ed.), Advances in experimental medicine and biology, vol. 42. Plenum Press, New York.

9. Gottesman, S., and J. R. Beckwith. 1969. Directed transposition of the arabinose operon: atechniquefor the isolation ofspecialized transducingbacteriophages.

J. Mol. Biol. 44:117-127.

10. Hradecna, A., and W. Szybalski. 1967. Fractionation of thecomplementarystrands ofcoliphageADNAbased ontheasymmetricdistributionofthepolyIGbinding

sites.Virology32:633-643.

11. Jacoby, G. A. 1972. Control of the argECBHcluster in Escherichia coli.Mol.Gen. Genet. 117:337-348. 12. Kohne, D. E., and R. J. Britten. 1971. Hydroxyapatite

techniquesfornucleic acid reassociation. p. 500-512. In G. L. Contoni and D. R. Davies (ed.), Proceduresin

nucleic acidresearch, vol.2.Harper& Row, New York.

13. Lennox, E. S. 1955. Transduction of linked genetic charactersofthe hostby bacteriophageP 1. Virology

1:190-206.

14. Miller,J.H., K.Ippen, J. G. Scaife, and J. R. Beckwith.

1968. Thepromoter-operator regionofthe lacoperonof

Escherichia coli.J.Mol. Biol.38:413-420.

15. Pouwels, P. H., R. Cunin, and N. Glansdorff. 1974. Divergent transcription in the argECBH cluster of

genesinEscherichia coliK12. J.Mol. Biol.83:421-424.

16. Sambrook, J. F., P. A. Sharp, and W. Keller. 1972.

Separationof thestrandsofSV40DNAand

hybridiza-J.VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

tion of theseparated strandstoRNA extractedfrom lytically infected and transformed cells. J. Mol. Biol. 70:57-71.

17. Summers, W. C., and W. Szybalski. 1968. Size, number and distribution ofpolyG binding siteson the

sepa-rated DNA strands of coliphage T7. Biochim. Biophys. Acta 166:371-378.

18. Summers, W. C., and W. Szybalski. 1968. Totally

asym-metric transcription of coliphage T7invivo: correlation withpolyG binding sites. Virology 34:9-16.

19. Taylor, K., Z. Hradecna, and W. Szybalski. 1967. Asym-metric distribution of the transcribing regionsonthe

complementary strands of coliphage A DNA. Proc. Natl. Acad. Sci. U.S.A. 57:1618-1625.

20.Taylor, A. L., and C. D. Trotter. Linkage map of

Escherichiacoli strain K-12. Bacteriol. Rev. 36:504.

21. Watson, J. D., and F. H. C. Crick. 1953. Astructurefor deoxyribose nucleic acid. Nature (London) 171:737-738.

22. Westphal, H. 1970. SV40 DNA strand selection by Escherichia coli RNA polymerase. J. Mol. Biol. 50:407-420.

23. Wolf, R. E., and D. G. Fraenkel. 1974. Isolation of specialized transducing bacteriophages for gluconate 6-phosphatedehydrogenase (gnd) of Escherichia coli. J. Bacteriol. 117:468-476.

24. Yamamoto, K., B. Alberts, R. Benzinger, L. Lawhorne, and G. Preiber. 1970.Rapid bacteriophage sedimenta-tion in the presence of polyethylene glycol and its application to large-scale purification. Virology 40:734-744.

on November 10, 2019 by guest

http://jvi.asm.org/