25 the lysozyme of bacteriophage

7/30/2019 25 the Lysozyme of Bacteriophage

1/8

THE JOURNAL OF BIOLOGICAL CHEMISTR YVol. 244, No. 8, Issue of April 25, pp. 1968-1975, 1963PTiba in NS.A

The Lysozyme of Bacteriophage xI. PURIFICATION AND MOLECULAR WEIGHT*

(Received for publication, September 36, 1965)LINDSAY W. BLACKS AND DAVID S. HOGNESSFrom the Department of Biochemistry, Stanford University School of Medicine, Stanford, California 9.@05

SUMMARYThe lysozyme synthetized in Escherichia coti during thedevelopment of bacteriophage X (X-lysozyme) has been puri-

fied to homogeneity from either the whole cell just prior tolysis or from the lysate. The molecular weight of this pro-tein was determined from its ~20,~ 2.06 S) and its Dzo,~(10.5 X lo+ cm2.sec-l) as well as from sedimentation equi-librium data to be 17.9 X 103. Differences between thecatalytic activities of X-lysozyme and egg white lysozyme ona limited group of substrates have been observed.

It has been known for some time that a lysozyme, or endolysin,appears during the development of bacteriophage X in Escherichiacoli K12 (1). This lysozyme is not detected in sensitive E. coliprevious to infection by X or in E. coli lysogenic for X previous totheir induction. Presumably the synthesis of the lysozyme isnecessary for the breakdown of the E. coli cell wall that allows therelease of the newly synthesized X phages from the infected bac-terium.

Our initial interest in this enzyme was based on two supposi-tions. By analogy to other lysozymes, particularly that of thecoliphage T4 (2), the supposition was made that X-lysozyme is asimple protein consisting of only one small polypeptide chain.This fi rst supposition is established in this and the two succeedingarticles (3,4). The second supposition was that R,the structuralgene for X-lysozyme, is located near one end of the linear duplexDNA isolated from mature X. This supposition was induced bythe location of R as the most terminal gene on the right side ofthe genetic map of vegetative X (5), and has been established bydetermining the gene content of fragments of X DNA which v aryin size but have in common the right end of the whole X DNA (6).

Smallness of the polypeptide is a virtue for the determination ofthe position and nature of changes in its structure caused by mu-tation in its gene. Insofar as smallness indicates simplicity, itshould also be a useful characteristic in developing a system for

* This work was supported by grants from the National Insti-tutes of Health and the National Science Foundation.1 Present address, Laboratoire de Biophysique, Universit6 deGenBve, Geneva, Switzerland.

the synthesis of a polypeptide starting from DNA. In this case,smallness of the DNA molecules containing the required genemay also be a virtue. The location of R near the end of X DNAhas allowed the isolation of small DNA fragments which containR and the right-hand terminus of X DNA, but none of the otherX genes.l

With these possibilities and associated advantages in mind, wereport on the characterization of X-lysozyme in this set of threearticles. In this first article we determine the molecular weightof X-lysozyme by sedimentation analysis and describe the purifi-cation procedures which yield the necessary homogeneous prepa-rations. The experiments which relate to the primary structureof X-lysozyme are described in the succeeding two articles (3, 4)of this set.

EXPERIMENTAL PROCEDURE

MaterialsBacteria and Phages-The source of X-lysozyme was E. coli

K12 (strain W3104 of E. Lederberg) made doubly lysogenic forX and Xdg (7) and induced by ultraviolet light. As W3104 lyso-genie for X alone yields lysates with the same specif ic activ ity forX-lysozyme as does the above double lysogen, it would presum-ably represent as good a source.

Reagents-Hydroxylapatite was Hypatite C from the ClarksonChemical Company. Carboxymethyl Sephadex C-50 was ob-tained from Pharmacia. Amberlite XE-64 was a product ofRohm and Haas and was treated by the method of Hirs, Moore,and Stein (8). Streptomycin sulfate was a gif t from Merck,Sharp and Dohme. Hydrolyzed starch used for gel electro-phoresis was purchased f rom Connaught Laboratories, Toronto.The reagents used for preparation of polyacrylamide gels wereobtained from Eastman Kodak.

Egg white lysozyme (twice crystallized, Batch LY616) wasobtained from Worthington and E. coli DNA labeled with a2Pwas a gif t of A. V. Paul.

MethodsAssayof X-Lysozyme-The act ivi ty of the enzyme was assayed

by observing its ability to decrease the absorbance at 600 rnpcaused by the turbidity of chloroform-treated E. coli K12 (strain

1 J. B. Egan and D. S. Hogness, unpublished experiments.

1968

atStanfordUniversity,onMarch13,2012

www.jbc.org

Dow

nloadedfrom
http://www.jbc.org/http://www.jbc.org/http://www.jbc.org/http://www.jbc.org/http://www.jbc.org/http://www.jbc.org/http://www.jbc.org/http://www.jbc.org/http://www.jbc.org/http://www.jbc.org/http://www.jbc.org/http://www.jbc.org/http://www.jbc.org/


2/8

Issue of April 25, 1969 L. W. Black and D. 8. Hogness 1969W3104). The bacterial substrate was prepared from cells grow-ing exponentially at, 37 in 3 liters of M-9 medium (9) supple-mented with 3 mg of glucose per ml. The cells were harvestedby centrifugation (O-2, 15 min, 13,200 X 9) when their concen-tration was 4 x lo* cells per ml and then resuspended at 25 in400 ml of dilution solvent (0.005 M EDTA, 0.06 M potassiumphosphate buffe r, pH 7.0) saturated with chloroform. Im-mediately upon completing the suspension, it was chilled to 0,and the cells were washed at O-2 by three cycles of centrifugationand resuspension in dilution solvent, with 200 ml in the firs t twocycles and 50 ml in the last. Samples were frozen in liquidnitrogen and stored at -20. No loss of sensitivity in the assaywas detected after storing for 6 months, but repeated thawingand freezing of the same sample should be avoided as this signifi-cantly decreases the sensitivity .

After diluting the enzyme in the dilution solvent, 0.05 ml isadded to one of two tubes, each of which contains 1.0 ml ofthawed bacteria diluted in the same solvent to an absorbance at600 rnp of 0.5. Immediately thereafter 0.05 ml of dilution solventis added to the other tube and the two tubes are placed in a waterbath at 37 for 4.0 min. The contents are then transferred to acuvet te, and the absorbance is determined at 4.5 min. The dif-ference in absorbance between the two suspensions (A absorbance)is proportional to enzyme concentration as long as the A ab-sorbance does not, exceed 0.3. One unit, of X-lysozyme is definedas that amount which will cause a A absorbance of 0.1.The following compounds used in the purification of the enzymeinhibit its act ivi ty and should be removed prior to assay:(NH&Sob, streptomycin sulfate, and protamine sulfate.

The assay was standardized with respect to different batches ofbacterial substrate by testing their sens itiv ity to egg white lyso-zyme which had a specif ic activity within 10% of 5000 X-lyso-eyme units per mg for all batches of substrate.

Other Assays and Measurements-The procedure for the phageassay has been described (7), as has that for determining the ac-tiv ity of X-exonuclease (10).

Protein was measured by the method of Lowry et al. (II), withcrystalline bovine serum albumin as a standard. The proteinconcentration of purified X-lysozyme calculated with this stand-ard is evidently accurate since the refractive index increment de-termined according to Van Holde and Baldwin (12) using pro-tein concentrations determined in this manner is 1.86 x 10e4 mlper mg, in agreement with the values found for other simple pro-teins (1.86 f 0.02 X 10e4 ml per mg (13)).

In some cases, the protein concentration of purified X-lysozymewas determined from the absorbance at 280 rnp and the value forits extinction coeff icient which we observed to be 1.07 cm2 per mg.Absorbance was measured in a Zeiss PM&II spectrophotometerwith a l-cm light path.

Analytical sedimentations were performed in a Spinco model Eultracentrifuge with schlieren optics and Eastman SpectroscopicII G plates to photograph the patterns, which were measuredwith a Gaertner M2001RS two-dimensional microcomparator.

Purification of X-LysozymeInduction of Lysogenic Cells-The kinetics of appearance of

X-lysozyme, X-exonuclease, and newly formed X phages followinginduction of lysogenic bacteria with ultraviolet light is given inFig. 1. They indicate the late nature of R, the structuralgene for X-lysoeyme, as compared to the early characteristicof the gene for X-exonuclease, this same temporal sequence hav-

0 30 60 90Minutes after itiductiok

FIG . 1. Kinetics of appearance of X-lysozyme after induc tionwith ultraviolet light. W3104 lysogenic for X were grown expo-nentially in 200 ml of H medium (14) containing 4 mg of glucoseper ml to a bacterial density of 7 X lo* bacteria per ml. Afterchilling to O, the suspension was irradiated for 70 set with ultra-violet light at a surface intensity of 19 ergs mm-2 set-1 in a tray,22 X 33 cm, under mild agitation. Twenty millilite rs of 10%Difco Bactotryptone were added, the culture was returned to 37at zero time on the above abscissa , and the growth was followedby measuring the absorbance at 600 rnp. Samples were removedat various times, chilled to O, and assayed for the indicatedcharacteristics. Total infective phage were assayed after treatinga 2-ml sample with 2 drops of chloroform plus 0.02 ml of a 5 mg perml solution of egg white lysozyme in 0.1 M potassium EDTA, pH7.0, shaking vigorously, and incubating at 37 for 20 min. Ex-tracts for enzyme assays were made by sonic oscillation (MullardSonicator, O, 6Q set) of baterial suspension s prepared by centri-fuging the sample and resuspending the pellet in an equal volumeof 0.05 M potassium glycyl-glycine buffer, pH 7.0. Note that thephage yield is plotted loga rithmically and consequently gives theimpression of premature appearance in comparison with the otherquantities; e.g. the time at which the phage yield attains 5% of itsmaximum value is later than the comparable time for A-lysozyme,although the logarithmic phage curve appears to precede theX-lysozyme curve.ing been observed previously when lysogenic cells are inducedwith mitom ycin (15) or by thymine starvation (16) and whensensitive cells are infected with X (16). Clearly, the best sourceof X-lysozyme is either the cells just prior to lysi s or the lysateitself. Both sources have been used. As the lysate is the pre-ferred source for the isolation of large amounts of the wild typeenzyme, we have described the purification from this source insome detail. The purification of the enzyme from cells priorto lysis differs appreciably only in the initial steps. These stepsare described subsequently since such cells may provide the bestsource for less stable forms of X-lysozyme, e.g. those derived fromtemperature-sensitive mutants. The growth and induction ofthe lysogenic bacteria given below are the same for each proce-dure.

E. coli K12 doubly lysogenic for X and Xdg (7) were grownovernight at 37 in a Biogen (American Sterilizer Corporation)containing 90 liters of H medium (14) supplemented with 81 g ofglucose which limits the overnight growth to 7 x lo8 cells per ml.After adding 320 g of glucose the next, morning to resume growth,cooling of the culture to 26 was initiated when the cell densitywas 1.5 x lo9 cells per ml and completed when i t had risen to2 x 10Q cells per ml. The bacteria were then irradiated for 400set with the ultraviolet lamps of the Biogen. Nine liters of 10%Difco Bactotryptone were added along with a small amount ofantifoam A (Dow Corning Corporation) and the temperaturewas adjusted to 37.

For purification from unlysed cells, the bacteria were allowed


www.jbc.org

Dow

nloadedfrom


3/8

1970 Lyso.zyme of Bacteriophage X. I Vol. 244, No. 8to grow for 55 min after reaching 37, and then collected rapidlyin iced vats and harvested with a Spinco:model 170 continuousflow centrifuge at 29,009 x g, keeping the temperature below 15.The centr ifuged bacteria were quickly frozen in liquid nitrogenand stored at -20.

For purification from lysates the culture was maintained at 37until completion of lys is and then cooled to 0. Under these con-ditions of growth and induction, lys is normally begins at 60 minand is complete by 130 min.

The details o f the procedure for the purification of X-lysozymefrom this lysate are given below and the results are given in TableI. When the total volume listed in Table I is greater than thatgiven below, it indicates that multiple batches were used. Unlessotherwise stated, all manipulations in the following procedureswere performed at O-5 and all centr ifugations were at 13,200 Xg for 15 min.

Precipitation at pH S.!?-HCI (250 ml) 2 N was slowly stirredinto 30 liters o f lysate until the pH reached 3.2. After 10 min offurther stirring, the precipitate was centrifuged from suspensionat 29,009 x g and 10-15 with a Spinco model 170 continuousflow centrifuge, and then resuspended in 3 liters of 0.005 MEDTA, 0.08 M potassium phosphate buf fer, pH 6.4, by stirring attop speed in a Waring Blendor for 2 min. This suspension wasstirred overnight and centrifuged, the supernatant was saved,and the pellet was resuspended in 170 ml of 0.005 M EDTA, 0.08M potassium phosphate buffe r, pH 6.4. After centrifugation ofthis second suspension, the two supernatants were combined toyield Fraction II.

Streptomycin Fractionation-To 3 liters of Fraction II wasadded, with stirring, 0.03 times its volume of 25% streptomycinsulfate dissolved in 0.005 M EDTA, 0.08 M potassium phosphatebuffer, pH 6.4. The mixture was stirred slowly for 15 min and

TABLE IPurification of X-lysozymeI I Protein

Fraction and step I I otal - CI.--mm g

I. Lysate. . . 126II. Precipitationat pH 3.2.. .. 11.5 34III. Streptomycinfractionation 11.7 32

IV. (NH&S04fractionation 1.20 10.2V. Amberlite XE-64 chroma-tography.. 0.58 0.29VI. Acetone pre-cipitation.. 0.150 0.1%VII. Carboxy-methylSephadexC-50 chro-matography. 0.155 O.ll!VIII. Hydroxylapa-tite chroma-tography.. . 0.125 0.09(

1

5

I-

- IL_-

:

3.02.78.5

units( lo-205.95.65.6

,OWXl-traton

mits/mlx 10-a0.170.510.484.7

0.50 5.6 9.61.20 5.2 35

0.74

0.72

5.1

4.5

33

36-

X-Lysozyme

Total 5pecificactivitymits/mgx 10-a

0.170.180.55

1929

45

50

centrifuged, and the supernatant was collected to obtain FractionIII.

Ammonium Sulfate Fractionation-To 3 liters of Fraction IIIwere added, with stirring, 730 g of solid (NH&Sod. After 20min of further stirring, the suspension was centrifuged and 680 gof solid (NH&SO4 were added to the resulting supernatant in thesame manner. The precipitate obtained by centrifugation wasdissolved in 250 ml of 0.005 M EDTA, 0.08 M potassium phosphatebuffer, pH 6.4, to obtain Fraction IV. Four batches of FractionIV were combined and dialyzed against 6 liters o f the precedingbuffer, the dialysate being changed once during an overnightperiod.

Amberlite XE-64 Chromatography-A column of AmberliteXE-64 (3.3 cm x 26 cm) was washed with 4 column volumes of0.005 M EDTA, 0.08 M potasium phosphate buffer, pH 6.4.After centrifugation of Fraction IV to remove the fine precipitatewhich formed during dialysis , it was applied to the column at aflow rate of 2.2 ml per min. The column was washed with 1column volume of 0.005 M EDTA, 0.1 M potassium phosphatebuf fer, pH 6.4, and a constant elution gradient from zero to 0.7M KC1 in 2 liters of the preceding solvent was applied at 2.2 mlper min, lo-ml fractions being collected. The enzyme was elutedin the second quarter of the gradient and pooled to form FractionV.

Acetone Precipitation-Acetone was added slowly with stirringto Fraction V (585 ml) until a final acetone concentration of70% (v /v) was achieved. After standing for 30 min, the result-ing suspension was centrifuged for 15 min at 5860 X g and theprecipitate was dissolved in 140 ml of 0.001 M EDTA, 0.05 Mpotassium phosphate buffer, pH 6.0. This solution was dialyzedagainst 6 liters o f the same solvent, the dialysate being changedonce during the overnight dialysis (Fraction VI).

Carboxymethyl Sephadex C-60 Chromatography-A column ofcarboxymethyl Sephadex C-50 (2.2 cm x 10.5 cm) was washedwith 2 column volumes of 0.001 M EDTA, 0.05 M potassium phos-phate, 0.4 M KC1 buffe r, pH 6.0, and then with 10 column volumesof 0.001 M EDTA, 0.05 M potassium phosphate buf fer, pH 6.0.Following centr ifugation, the dialyzed Fraction V (150 ml) wasapplied at 1 ml per min, the column was then washed with 4column volumes of 0.001 M EDTA, 0.1 M KCl, 0.05 M potassiumphosphate buffer, pH 6.0. A constant gradient from 0.10 MKC1 to 0.30 M KC1 in 500 ml of 0.001 M EDTA, 0.05 M potassiumphosphate buf fer, pH 6.0, was then applied at a flow rate o f 1 mlper min and 6.5~ml fractions collected. The fractions containingenzyme were pooled and dialyzed against 6 liters of 0.005 Mpotassium phosphate buffer, pH 6.0, to obtain Fraction VII.

Hydroxylapatite Chromatography-A column of hydroxylapatite(2.2 cm by 5.3 cm) was washed with 2 column volumes of 0.225 Mpotassium phosphate buffer, pH 6.8, and then with 3 columnvolumes of 0.005 M potassium phosphate buffer, pH 6.8. Frac-tion VII (155 ml) was loaded on the column at a flow rate of0.5 ml per min and eluted at the same flow rate by a constantgradient between 0.025 M potassium phosphate buffer, pH 6.8,and 0.225 M potassium phosphate buffer, pH 6.8, in a totalvolume of 400 ml. Fractions of 7.5 ml were collected and Frac-tions 21 to 37 (Fig. 2) were pooled to form Fraction VIII ofTable I. This lysozyme was concentrated by acetone precipi-tation to varying degrees depending upon its usage. Suchconcentrated solutions are also referred to as Fraction VIII inthe succeeding sections.


www.jbc.org

Dow

nloadedfrom


4/8

Issue of April 25, 1969 L. W. Black and D. 8. Hogness 1971PuriJcation from Unlused Cells-The preparation of a crude

extract from the frozen bacteria and its fractionation with strep-tomycin are given below. Final purification was achieved byprocedures ssentially the sameas those given above, startingwith the ammonium sulfate fractionation.The frozen bacteria (650g, wet weight) were partially thawedand suspended in 2666 ml of acetone in a Waring Blendor for 15min at medium speed. The acetone was removed by filtrationand the bacteria were horoughly dried. The enzyme wasstablein the acetone powder for at least 6 months at -20. Fiftygramsof acetonepowder were suspendedn 750 ml of 0.005MEDTA, 0.08 M potassium phosphate buffer, pH 6.4, by stirringovernight. The suspension as centrifuged, the pellet was re-suspendedn 750 ml of the precedingsolvent, and this secondsuspension as centrifuged. After combining he two superna-tants, the protein concentration wasadjusted o 10 mgper ml bydilution with the abovebuffer, and 1 volume of 25% streptomycinsulfate (dissolved n the same uffer) wasadded,with stirring, toevery 10 volumesof the extract. The mixture wasstirred for 15min and centrifuged, and the supernatant (2130ml) wascollectedand fractionated with ammonium sulfate as describedabove.Enzyme prepared rom eachof the two sources specificactiv-ity of 45 to 50 x 103 nits per mg of protein) has beenused orcharacterization of the sedimentationcoefficientand the molecu-lar weight. No significant differenceswere detected, indicatingthat the low pH used o precipitate protein from the lysate haslittle, if any, irreversible effect.

RESULTS

Testsor HomogeneityFig. 2 shows he chromatography of Fraction VII on a hy-

droxylapatite column, he last step n the purification procedure.The singlepeak of X-lysozyme activity coincideswith the singleprotein peak to the extent that the specificactivities are within~15% of a singlevalue (50 x lo3 units per mg), which is withinthe combinederror of the assays.The resultsof electrophoresis t pH 5 in starch gelsof both apurified preparation and one n which about one-half of the pro-tein was X-lysozyme are shown n Fig. 3. That the strongly

20 30Fraction number

FIG. 2. Hydroxylspatite chromatography of Fraction VII.The conditions are given in the text. The maximum proteinconcentration and lysosyme activ ity , found in Fraction 29, were1.5 mg per ml and 76 X 103 units per ml, respectively.

(01 puriFied A-lysoz yme+ -

- -I

DistanceCram Origin, cm.FIG. 3. Starch gel electrophoresis of X-lysozyme. The purifiedenzyme was used in a while b epresents the result with a partiallypurified preparation whose specif ic act ivi ty is one-half the maxi-mum value. Gels were made in 0.05 M sodium acetate buffer , pH5.0, according to the method of Smithies (17). The slot was loadedwith 0.05 ml of enzyme solution (3 mg per ml for Fraction VI and5 mg per ml for Fraction VIII), and 175 volt s were applied for 14hours across electrodes connected to reservoirs containing 0.05 M

sodium citrate buffer, pH 5.0. The gels were sectioned andstained according to Wilson and Hogness (18). The lower curverepresents the X-lysozyme act ivi ty eluted from 0.5-cm sections ofthe unstained half of the gel into 1 ml of 0.06 M potassium phos-phate buffe r, pH 7.0, after a 14-hour exposure at 2-4.

FIG. 4. Polyacrylamide gel electrophoresis of X-lysozyme.The electrophoresis of Fraction VIII was performed according tothe method of Ornstein and Davis (19). A 15yo small pore gelstacked at pH 5 and run at pH 4.3 was used (see Reference 20 forthe composition of gels and buffers). Following electrophoreeisof 75 rg o f X-lysozyme for 1 hour (voltage drop, 150 volt s; current,2.5 ma; 24), the gel was stained with Amido black and washedwith 7% acetic acid.staining band is the X-lysozyme is made clear by the measure-ment of enzyme activities of the material in segments f the gel(Fig. 3b). Starch gel electrophoresis as also performedat pH7.0, but resolutionof impurities wasgreatly reducedat this pH;


www.jbc.org

Dow

nloadedfrom


5/8

Lysozyme of Bacteriophage A. I Vol. 244, No. 8only one band, moving as posit ively charged protein, was de-tected in the partially purified preparation.

The result of electrophoresis of the purified X-lysozyme in apolyacrylamide gel at pH 4.3 is illustrated in Fig. 4. Again onlyone component is detectable.

The schlieren pattern obtained during the centri fugation ofFraction VIII consisted of a single symmetrical peak indicatingthe existence o f only one sedimenting component in the purifiedpreparation. The conditions of centrifugation were thosedescribed below for the determination of the sedimentationcoefficient.Sedimentation Behavior and Determination of Molecular Weight

The previous data indicate that the purified X-lysozyme issuff iciently homogeneous to allow a determination of its molec-ular weight by sedimentation analysis. In the following sections,the molecular weight has been determined from the sedimentationand diffusion coefficients and from the radial distribution of theprotein after centrifugation to equilibrium.

Sedimentation and Diffusion Coeficients-The sedimentationcoefficient (~20,~) of X-lysozyme, measured at a protein concen-tration of 10.0 mg per ml, is 2.02 5, and at 4.2 mg per ml it is2.10 S. For these determinations the enzyme, dissolved in0.10 M KCl, 0.01 M potassium phosphate buffer, pH 7.0, wascentrifuged at 59,780 rpm and 23.5 in a 12-mm, 2, doublesector synthetic boundary-forming cell. Although there appears

-2.5

d3.5fG+ !.Jf-4.5

-5.5

Time in minuksFIG. 5. The determination of the diffus ion coeff icient by themethod of Sophianopoulos et al. (21). The conditions of centrifu-gation are those given for the determination of the sedimentation

coefficient except that a 2.04-mm column of solution in a 12-mm,2 double sector Epon c ell was centrifuged at 15,220 rpm, for 25hours at 18.3. The schlieren patterns during the approach toequilibrium as well as at equilibrium were photographed and thevertical displacement from the base line at (b + m)/2 was meas-ured, where b and m refer to the radial distances to the bottomand menisc us of thesolution column, respectively. At equilibriumthis vertical displacement is referred to as ye, while at othertimes, t, it is referred to as yt. The diffusion coefficient can thenbe calculated from the equation (21):D = cb - ml2 d n (Y, - yt). f czt 1 + (b - rn)79

since the slope of the above curve is equal to d In (ye - yt)/dt.The terms in brackets are small correction terms having a value of1.052 under the conditions used here.

10 20 30Time in minutes

FIG. 6. Determination of the diffusion coefficient of X-lysozymefrom boundary spreading. The conditions of centrifugation werethose given for determination of the sedimentation coefficientexcept that the speed was 8000 rpm and the temperature was 18.3.The heigh t (h) and area (a) of the sing le boundary represented inthe schlieren photographs were determined as a function of timeand the ratio of the area (A) to height (H) of the true boundarycurve (&/& against T) calculated from A/H = a/(h . F) where Fis the magnification factor for T. D is computed from the slopeof the above curve and the equatio n (22):

(A/H)2 = 477Dtto be some dependence of the ~20,~ on concentration (the abovevalues extrapolated to zero concentration by plotting either sor s-l versus c give a value of 2.16 at c = 0) the data are insufli-cient for an accurate determination of this dependence. Conse-quently, we use the average value of 2.06 S for the concentrationrange 4 to 10 mg per ml.The diffus ion coeff icient (020,~) of X-lysozyme was measuredby two different methods at a protein concentration of 10 mg perml. The method of Sophianopoulos et al. (21) was used inanalyzing the transient states during sedimentation to equi-librium. The method and data are summarized in Fig. 5, fromwhich a value of 10.7 X lo+ cm2.se0 for the Dz,,,, was calcu-lated.

In the second method, the diffusion coeff icient was calculatedby measuring the spreading of the boundary formed at 8000rpm in a synthetic boundary cell as a function of time (22). Thedata and method are summarized in Fig. 6, from which a valuefor the Dzo,, of 10.2 X lo+ cm2 se0 was calculated. Thesmall sedimentation coeff icient of X-lysozyme, the low speed ofcentrifugation, and the time interval used (36 min) are suchthat the radial position of the maximum in the schlieren pattern(rmBx) increases only slightly during measurement of boundaryspreading (AT,, at 36 min is about 1% the increment in thewidth of the schlieren peak at half of the maximum height).Consequently, corrections for sharpening of the boundary dueto a dependence of s on concentration can be neglected.

The average value of the diffusion coeff icient calculated by thetwo methods in 10.5 X lop3 cm2.sec-l and, when this is combinedwith the previous ly determined sedimentation coeff icien t, amolecular weight for X-lysozyme of 1.78 x lo4 was calculatedfrom the Svedberg equation,

M=sR T

D(1 - fiP)


www.jbc.org

Dow

nloadedfrom


6/8

Issue of Apri l 25, 1969 L. W. Black and D. X. Hogness 1973The partial specif ic volume, 0, was computed from the amino acidcomposit ion (3) to be 0.732 cm3 per g (23).

Xedimentation Equilibrium-Centrifugation of the enzyme tosedimentation equilibrium in a short column of solution (3.52mm) at 33,450 rpm yields a distribution (Fig. 7) in which theprotein concentration at the meniscus is zero. This distributionallows the calculation of the molecular weight from a singleschlieren pattern by two methods: (a) from a determination ofd(lnc,) /dW, where T is the distance from the axis o f rotationand cr is the protein concentration at r (Fig. 7u), and (b) froma determination of d((l/r) .(dc,/dr))/dc, (Fig. 7b). The valuescalculated by these two methods are in good agreement, being1.75 x 10 and 1.79 X 104, respectively (see Table II). A dis-advantage of the distribution obtained at this relatively highspeed is that only a fraction of the total protein added to thecentr ifuge cell is analyzed (the remainder being in a region o f too

47.6 48.1 48.6r; crn2b I

I I I2 4C,, mq/ml.

1

-1

FIG. 7. Determination of the molecular weight of X-lysozymeby sedimentation to equilibrium. Except for the speed (33,450rpm) and temperature (16) the conditions of centrifugation wereidentical with those given in Fig. 5. The schlieren pattern of theequilibrium distribution was photographed and analyzed to givethe plots a and b above. a, the molecular weight, M, is calculatedfrom the equation (24) :M= 2R T d(ln CJ.-(1 - &J)w2 d(7.2)

b, the molecular weight is calculated from the equation (12):.1! 24

RTM = (1 - Bp)w2

\r -ar /dc,

TABLE IIMolecular UI(

Method

Equilibrium centrifugationMethod a (see Fig. 7a)Method b (see Fig. 7b)

Sedimentation-diffusioncAverage

yht of X-lysozymeAngularvelocity

33,45035,60033,45035,60016,40015,220

PH

7.04.97.04.97.09.37.0

4.2 1.755.0 1.824.2 1.795.0 1.774.2 1.825.0 1.831.78

1.790 The solvent at pH 7.0 was that given in Fig. 5. At pH 4.9,

the solvent was 0.10 M KCI, 0.025 M potassium acetate buf fer.At pH 9.3, the solvent was 0.20 M KCl, 0.020 M potassium car-bonate buffer.b COstands for the initial protein concentration in the equilib-rium determinations.c ~20,~ = 2.06 S; Dzo,~ = 10.5 X 10-T cm2.sec-l.

high a dc/dr value to be registered in the schlieren pattern).Consequently, the values of molecular weight obtained byMethods a and b do not necessarily represent the weight average(M,) and z-average (1M,) molecular weight, respectively , thatwould otherwise prevail.

At the lower speed of 16,200 rpm, the equilibrium concentra-tion at the meniscus is not zero and must be evaluated for thefirst of the above methods, although not for the second. Sincethis evaluation introduces an error in the first method not in-herent in the second, we present here only that molecular weightdetermined by the second method at this lower speed. Thisvalue is 1.82 x lo4 and compares favorably with the value of1.79 x lo4 calculated by the same method at the higher speed(Table II). Since essentially all of the column of solution isanalyzed at the lower speed and consequently most of the proteinmolecules contribute to the distribution, the agreement betweenthe values calculated at the two speeds indicates that the dis-advantage associated with the higher speed measurements is noteffect ive for these preparations. The agreement between thethree values is also indicative of the homogeneity of the material.

E$ect of pH on Molecular Weight-Sophianopoulos and VanHolde (25) have shown that egg white lysozyme undergoes areversible dimerization between pH 5 and pH 9, the extent ofdimerization increasing with increasing pH. As a consequence,the molecular weight calculated from sedimentat ion equilibriummeasurements is a function of pH and represents the monomeronly below pH 5. This type o f aggregation does not occur withX-lysozyme. The molecular weight of X-lysozyme was deter-mined at pH 4.9 and pH 9.3 by sedimentation to equilibrium.Both Methods a and b were used at the lower pH, while Method balone was used at the higher pH. It is clear from an examinationof the values for molecular weight given in Table II that thereis no significant change as the pH is raised from 4.9 to 9.3. Allvalues fall into the range (1.79 f 0.04) x 104.

Catalytic PropertiesWhile our present interests emphasize the structural character-

istics of X-lysozyme without special reference to its catalytic


www.jbc.org

Dow

nloadedfrom


7/8

Lysoxyme of Bacteriophage A. I Vol. 244, No. 8

4 8 12 16 20Hours at 37

FIG . 8. Action of X-lysozymeandeggwhite lysozyme on penta-N-acetylglucosamine. Two tubes containing 38 mM sodiumcitrate, 3.5 my penta-N-acetylglucosamine, H 5.1, in a finalvolume of 0.54ml were ncubated at 37 for 24 hours. One tubecontained 0.146mu egg white lysozyme, the other 0.117mM X-lysozyme. Samples,0.025 ml, were removed from the sealedtubesduring the incubation, diluted to a final volumeof 1 ml, andquickly frozen. Of this solution 0.50ml wasassayedor reducingsugargroups (29). In this assay, he optical density at 690 npincreases s the concentration of reducing groups ncreases:0,A-lysozyme; , eggwhite ysozyme.properties, we have made certain observationspertinent to itscatalytic function, particularly in comparison o that of eggwhite lysozyme.The purified &lysozyme has a specificactivity 10 times thatof egg white lysozyme in our assay system with the E. colisubstrate. Because the molecular weight of X-lysozyme isslightly larger than for egg white lysozyme (M = 14.3 x 103),the ratio is slightly greater on a units/molecule basis. Inagreementwith results obtained with impure preparationsof X-lysozyme (26, 27), we find that the purified X-lysozyme has noeffect on Micrococcus ly.sodeikticus; the rate catalyzed by X-lysozyme was less han 0.001 the rate observed for egg whitelysozyme.We have also nvestigated the effect of X-lysozyme on two ofthe simpler substrates or egg white lysozyme, the tri- andpenta-N-acetylglucosamine erived from chitin and kindly sentto us by J. A. Rupley (28). Penta-N-acetylglucosamine shydrolyzed by eggwhite lysozyme at a much greater rate thanis tri-N-acetylglucosamine (28). However, the penta-2\r-acetylglucosamine s not hydrolyzed by X-lysozyme (Fig. 8).As the X-lysozyme was fully active on the E. coli substrateafterthe 22-hour incubation shown in Fig. 8, we conclude that thepenta-N-acetylglucosamine s not a substrate for this enzyme.Similarly, tri-N-acetylglucosaminewasobserved o be nsensitiveto attack by X-lysozyme.

DISCUSSIONFrom our viewpoint, the most important conclusions f thispaperare (a) that the X-lysozyme is a smallprotein, (a) that thenumber of molecules ynthesized per induced cells s compara-tively high, and (c) that respectable mountsof the pure proteincan be isolated n a reasonably simplemanner.The low molecularweight is the result that we had anticipatedby analogy to other lysozymes. The value of 17.9 x lo3 ob-

tained for X-lysozyme s remarkably similar o the value of 18.1 x

103 ecently found for the lysozyme derived from coliphageT4(2) and about 25% greater than the molecular weight of eggwhite lysozyme. The similarity between the lysozymesof thecoliphagesand the difference between them and egg whitelysozyme is further emphasizedby the comparisonof theirammoacid compositions iven in the following paper (3).The X-lysozyme and the eggwhite lysozyme alsodiffer in theircatalytic properties. However, the failure of X-lysozyme tocatalyze the hydrolysis of penta-N-acetylglucosamine oesnotprovide a very strong argument hat with cell wall substrateshetwo enzymes do not effect the hydrolysis of the samebond,namely the p-(1,4) linkage between the N-acetylmuramic acidand N-acetylglucosamine esidues, he bond attacked by eggwhite lysozyme (30). Indeed, t isquite possiblehat X-lysozyme,but not eggwhite lysozyme, demandshe presence f a muramicacid residue or catalysis. Previous work with impure prepara-tions of X-lysozyme and cell wall substratesndicates hat simi-lar, perhaps dentical, products were obtained when X-lysozymeand egg white lysozyme attacked the cell walls of Bacillusmegaterium; on the other hand, somedissimilarproducts wereobtained from cell walls of E. coli (26). Considering hat animpure enzyme was used, these nvestigations also eave openthe question as to what bond in the cell wall is destroyed byX-lysozyme.With the molecular weight and the specific activity of thepurified enzyme, one can compute the number of X-lysozymemolecules er induced cell from the enzyme activity in lysates.Under our conditionsof induction this turns out to be about 6 xlo4 molecules. It is difficult to estimate he efficiency of the Rgene becauseone does not know the fraction of the many Xgenomes resent n an induced cell which are active. One can,however, make a crude comparison o another late protein(proteins) that should be formed in large amounts, the headprotein (proteins)of the maturephage. The combinedmolecularweight of all proteins n a mature X phagecanbe calculated o beabout 33 x 106daltons either from (a) the fact that the ratio ofprotein to DNA is mature X is close o 1 (14) and the molecularweight of X DNA is about 33 x lo6 (31) or (b) from the densityof the phageprotein and the dimensions f the phagehead andtail (32). Calculating that the head represents bout SO% ofthe total protein, one obtains26 X lo6 daltons or the protein inthe head. The 6 x 104 ysozyme molecules er cell representsome1 x log daltons, or the equivalent n mass o the protein inabout 40 phageheads.This number is to be comparedwith the 100 mature phageproduced per induced cell under these conditions. The com-parison s not strict becausewe are ignorant of the number ofphagehead subunits hat were synthesizedbut do not appear nthe mature phage and the number of X-lysozyme moleculeshidden from the assayprocedureby inactivation or adsorptionto cell walls. However, even this crude comparisonsuggeststhat such disparate late proteins as X-lysozyme and headprotein (proteins) are synthesizedat rates that differ by lessthan an order of magnitude, a suggestion hat is somewhatsurprising n view of the massiveamounts of T4 head proteinsynthesized n E. coli infected with that phage 33).Finally, it should be noted that approximately 100 mg ofX-lysozyme are obtained from the purification sequence ivenin Table I. This can be accomplished n about 2 weeks byone person. Thus, the amounts necessary or an analysis of


www.jbc.org

Dow

nloadedfrom


8/8

Issue of April 25, 1969 L. W. Black and D. X. Hognessprimary structure and for studies of crysta.llization propertiescan be obtained in reasonable times.

il.

1.2.3.4.5.6.7.8.9.

10.

11.12.13.14.15.16.

REFERENCES 19.JACOB. F., FUERST, C., AND WOLLMAN, E., Ann. Inst. Pasteur, 20.

93, 724 (1957). . TSUCU TA, A., AND INOUYE, M., J. Mol. Biol., 37,201 (1968). 21.BLACK. L. W.. AND HOQNESS, D. S., J. Biol. Chem., 244. 1976

(1969). BLACK, L. W., AND HOGNESS, D. S., J. Biol. Chem., 244, 1982(1969).

CAMPBELL, A., Virology, 14, 22 (1961).HOQNESS, D. S., DOERFLER, W., EGAN, J. B., AND BLACK,L. W., Cold Spring Harbor Symp. Quant. Biol., 31,129 (1966).

HOQNESS, D. S., AND SIMMONS, J. R., J. Mol. Biol., 9, 411(1964).HIRS. C. H. W.. MOORE. S.. AND STEIN, W. H., J. Biol. Chem..

200; 493 (1953). ADA&, M.H., Bacteriophages, Interscience Publishing Com-

pany, Inc., New York, 1959, p. 446.LITTLE , J. W., LEHMAN, I. R., AND KAISER, A. D., J. Biol.

Chem., 342, 672 (1967).LOWRY, 0. H., ROSEBROUOH, N. J., FARR, A. L., AND RANDALL,R. J., J. Biol. Chem., 193,265 (1951).VAN HOLDE, K. E., AND BALDWIN, R. L., J. Phys. Chem., 62,735 (1958).DOTY, P., AND EDSALL, J. T., Advan. Protein Chem., 6, 55(1951).

24.25.26.27.28.29.30.

KAISER, A. D., AND HOGNESS, D. S., J. Mol. Biol., 2,392 (1960). 31.PROTAS S. J. J.. AND KORN, D., Proc. Nat. Acad. Sci. U. S. A., 32.

22.23.

66, 1089 (1966). . 33.PROTAS S, J. J., AND KORN, D., J. Biol. Chem., 241,4175 (1966).

SMITHIES, O., Biochem. J., 71, 585 (1969).WILSON, D. B., AND HOONESS, D. S., J. BioZ. Chem., 239, 2469

(1964).ORNSTEIN, L., Ann. N. Y . Acad. Sei., 121, 321 (1964); DAVIS,B. J., Ann. N. Y. Acad. Xci., 121,404 (1964).REISFELD, R. A., LEWIS, U. J., AND WILLIAMS, D. E., Nature,196, 281 (1962).

SOPHIANOPOULOS, A. J., RHODES, C. EL., HOWOMB, D. N., ANDVAN HOLDE, K. E., J. BioZ. Chem., 237, 1107 (1962); DE AL,W. C., RUTTER , W. J., AND VAN HOLDE, K. E., Biochemistry,2, 246 (1963).

WILLIAMS, J. W., VAN HOLDE, K. E., BALDWIN, R. L., ANDFUJITA, H., Chem. Rev., 68, 790 (1958).COHN, E . J., AND EDSALL, J. T., Proteins, amino acids, andpeptides, Reinhold Publishin g Company, New York, 1943,p. 371.SCHACHMAN, H. K., Ultracentrifuge in biochemistry, AcademicPress, New York, 1959, p. 202.

SOPHIANOPOULOS, A. J., AND VAN HOLDE, K. E., J. BioZ,Chem., 239, 2516 (1964).WORK, E., J. Gen. Microbial., 26, 167 (1961); Biochem. J ., 76,

38P (1960).JACOB, F., AND FUERST, C. R., J. Gen. Microbial., 18, 518

(1968).RUPLEY, J. A., Biochim. Biophys. Acta, 83,246 (1964).PARK, J. T., AND JOHNSON, M. J., J. Biol. Chem., 181, 149

(1949).SALTON, M. R. J., AND GHUYSEN, J. M., Biochim. Biophys.Acta, 36, 552 (1969).THOMAS, C. A., J. Gen. Physiol. , 49,143 (1966).KAISER, A. D., J. Gen. Physiol. , 49,171 (1966).SARABHA I, A. S., STRETT ON, A. 0. W., BRENNER, S., AND

BOLLE, A., Nature, 201, 13 (1964).atStanfordUniversity,onMarch13,2012

www.jbc.org

Dow

nloadedfrom

25 the lysozyme of bacteriophage

Documents