A SPECIFIC REVERSIBLE INHIBITION OF BACTERIOPHAGE T2'

BERNARD P. SAGIK'Department of Bacteriology, University of Illinois, Urbana, Illinois

Received for publication March 30, 1954

With most phages, the titer of a lysate meas-ured by plaque count remains stable from thetime of lysis or shows a progressive decrease asphage becomes inactivated by a variety of en-vironmental agents. Freshly prepared lysates ofbacteriophage T2 of Escherichia coli rise in titerspontaneously with time, the maxmum titerbeing reached in several weeks under refrigera-tion. Bertani (unpublished data, 1950) notedthat this rise could be accelerated greatly bydiluting a lysate in distilled water or by addingZnSO4 to a final concentration of one per cent.Watson (personal communication, 1951) ob-served that lysates of T2 to which antibacterialserum was added before lysis did not show anyrise in titer after dilution in distilled water. Thissuggested that normal lysates contained a cer-tain proportion of particles that are preventedfrom forming plaques because of combinationwith specific host materials. The present paperprovides evidence of the occurrence of suchcombination and of the properties of the com-bined phage.

MATERIALS AND METHODS

Cultures. The bacteriophages studied were T2,T4, T6, and their mutants, whose common hostis E. coli, strain B (Delbrilck, 1946).

Methods. Phage activity was determined byplaque'- counts using the agar layer method(Gratia, 1936). Stocks of phages T2, T4, and T6were lysates from "lysis inhibited" cultures(Doermann, 1948) clarified by filtration or lowspeed centrifugation. The organisms were grownin Difco nutrient broth (containing 0.5 per cent

1 Submitted in partial fulfillment of the require-ments for the Ph.D. degree, University of Illinois.This work was supported in part by a grant fromthe American Cancer Society (upon recommenda-tion of the Committee on Growth) to Dr. S. E.Luria, to whom the author is indebted for adviceand encouragement.

' Present address: Department of Biophysics,Florence R. Sabin Laboratories, University ofColorado Medical Center, Denver, Colorado.

NaCi) or in M9 medium3 to a density of about1 X 101 cells per ml. The cultures were infectedwith about 10 phage particles per cell, and asecond equal phage input was made 7 to 10minutes later. The cultures were aerated at 37 Cuntil clear. Clearing generally occurred about 9hours after infection; it was rapid, being com-pleted within 15 minutes after the beginning ofvisible lysis.

RESULTS

Spontaneous and induced rise in the titer of T2lysates. Increase in phage titer. Freshly madelysates of phage show a spontaneous rise in titerwith time (see figure 1). The amount of rise isvariable, but all T2 stocks show an increase. Theextent and rate of the increase can be reducedby suspending the phage in m/15 phosphatebuffer plus 0.85 per cent NaCl (Watson, personalcommiunication, 1951).

Diluting fresh lysates of T2 in distilled watercauses a rapid increase in titer. Maxmum titerswere reached in a 1:50 dilution in distilled waterat 37 C. The distilled water effect is observedwith T2 lysates, T2r lysates, and T2h lysatesprepared either in nutrient broth or in syntheticmedium. It is observed with T2 lysates preparedon E. coli, strain B/4, or strain B/6, as hostcells and in fresh lysates of T2h prepared on E.coli, strain B/2.

Fresh lysates of T4 and T6 phages preparedon E. coli, strain B, do not show any rise indistilled water. Thus, the increase in titer withdilution in distilled water appears to be peculiarto phage T2 and its mutants.

Filtration of inhibited lysates. Filtration offresh T2 lysates through Mandler filters (testedat 9 lb pressure) did not eliminate nor reduce

3M9 medium. Solution A: KH2PO4, 3 g; MgSO4,0.2 g; NH4Cl, 1 g; Na2HP04, 6 g; distilled water,900 ml. Solution B: Glucose, 4 g; H20, 100 ml.Solutions A and B are sterilized separately andmixed in a ratio of 9:1 as needed. The final pH is7.0.

430

on February 12, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

REVERSIBLE INHIBITION OF BACTERIOPHAGE T2

o broth assay* buffer assay

* water assay

doys

Figure 1. Spontaneous and induced rise in thetiter of T2 lysates. Freshly made T2 lysates show aspontaneous rise in titer with time (broth assay).The extent and rate of this increase in titer arereduced by suspending the phage in m/15 phos-phate buffer plus 0.85 per cent NaCl (bufferassay). Dilution 1:50 in distilled water induces arapid increase in titer, generally to a higher levelthan is achieved spontaneously with time (waterassay).

the distilled water effect. This shows that theeffect is due neither to disruption of unlysedcells by distilled water treatment nor to releaseof phage combined with unfiltrable cell frag-ments.

Comparison of mass lysates with "one-step"lysates. The large majority of the potentiallyactive phage in fresh T2 lysates is unable toform plaques without pretreatment. If thisinability originated within the infected cell, itmight be present also in the phage of "one-step" lysates, in which the infected bacteria arediluted to a concentration of a few thousand perml before lysis (Ellis and Delbrtick, 1939).Several experiments were performed in whichcells infected with T2 were allowed to lyseeither at a concentration of 2 X 108 per ml orat a concentration of 2 X 101 per ml. The titerof both lysates was measured with and withoutdilution in distilled water. The results of atypical experiment, given in table 1, show thatthere is no distilled water effect in "one-step"

TABLE 1Plaque titer in parallel one-sep and ma8s lysatesA culture of Escherichia coli, strain B, in

nutrient broth (2.1 X 108 cells per ml) was in-fected with 4 particles of T2 per cell. Seventeenminutes after infection, an aliquot was diluted1:10,000 (dilute lysate). When the mas lysatecleared, both lysates were assayed before andafter treatment with distilled water. Note thatthese lysis inhibited cultures yield about 500particles per cell.

TITER AlTERn= M

LYSIS, EXIlE- TITERNAlTERLYSATI DISTCLUT 1:50

CONCENTRA- DISTIRlLEDUTION IN TE WATER 17 CUMSS LYSATE AT3

Mass lysate.............. 2.6 X 109 1.4 X 1011Dilute lysate ............ 1.2 X 1011 1.3 X 1011

lysates because all the potentially active phageis already active.

The rise in titer as unmasking of inhibitedphage partiles. The rise in titer of concentratedT2 lysates might be due either to a dispersal ofphage clumps or to a release of phage fromcombination with some inhibitor. The followingexperiment was designed to distinguish betweenthe two alternatives. A filtered, freshly pre-pared lysate of T2 was diluted so that one mlcontained about 5 plaque forming units beforewater treatment. One drop (equals 0.05 ml) ofthe dilution was added to each of 50 tubes con-taining 0.5 ml of distilled water. The tubes wereincubated at 37 C for one hour. The entire con-tents of each tube were plated then by the agarlayer method. If the low counts before watertreatment were due to the presence of phageclumps, each clump being counted as one plaque,and if the rise were due to disaggregation indistilled water, there should be similar numbersof blank plates in both series, If, on the otherhand, the rise is due to an activation of in-hibited particles, which are presumably dis-tributed at random, the distribution of plaqueswill, in both series, be random (Poisson); but theaverage will be higher by the rise factor in thewater treated series since the inhibited particleswill manifest themselves in this series. The re-sults, illustrated by the experiment in table 2,are in complete agreement with the expectationsfrom the inhibitor hypothesis. The release of

1954] 431

._0-

on February 12, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

BERNARD P. SAGII

TABLE 2Test of the clump hypothesis versus the inhibitor

hypothesisA filtered, fresh broth lysate of T2 was diluted

in broth to contain about 3 particles per drop afterwater treatment. One drop of the broth dilutionwas added to each of 50 tubes containing 0.5 mlnutrient broth and to each of 50 tubes containing0.5 ml distilled water. After 1 hr at 37 C the con-tents of each tube were plated. From the meannumber of plaques per plate in the distilled waterplates (= 3.4), the expected number m of platescontaining r plaques was calculated from theequation P(r) - nren/rl where P(r) is the frac-tion of plates with r plaques and n is the meannumber of particles per drop.

NO. OPPLAQUESPER PLATE

(r)

01234567891011

NO. OrPLATESwIT rPLQUEsIN THE

NUTRIENTBROTHSERIES

37121000000000

X2 = Z(x -

P = 0.85.

NO. OPPLATESwiT rPLAQUEZSIN TH

DISTILLEWAT1SERIES

(x)

4491196400101

NO. OFPLATES

IN TEWATER

ASSURING3.4 PAiT-CLES PERDlROP ONTEE

AVERAGE(m)

1.655.609.5510.809.256.253.551.760.750.30,0.1010.05J

x -m

2.351.60

0.550.20

-0.25-0.250.45

m)"/m = 4.182,

30 60 120

(E -m)'/m

inhibition hereafter will be called activation ofinhibited particles.

Rate of activation of inhibited ly8ates. Freshlyprepared phage lysates either in broth or in M9medium, diluted 1:50 in distilled water at 37 C,reach maximum titer in 15 seconds. At 5 C themaximum titer is reached in 90 to 120 seconds.Because of these rapid rises, a "dump experi-ment" (Stent and Wollman, 1950) was devisedin which phage was diluted with distilled waterinto several tubes kept at the desired tempera-ture. The reaction was stopped by "dumping"into one tube a large volume of broth. Thismethod is justified by the fact that upon dilution

300seconds

Figure S. The activation of inhibited T2 lysatesin distilled water and in broth as a function of thetemperature and the dilution. The 1:5 dilution inwater reaches a maximum titer after two hours.

of lysates in broth there is no rise in titer at 5 Cnor at 37 C over a three hour period, even whenthe titer had risen partially in distilled water.The results are shown in figure 2. The rate ofrise in titer depends both on the extent of dilu-tion in water and on the temperature. A rapidrise is observed even in broth if the samples arekept at 65 C. The maxmum titer is essentiallythe same after treatment in distilled water at5 C and at 65 C (figure 2). Rapid activation at65 C occurs also in the undiluted lysates. Thephage in heat activated lysates does not becomereinhibited upon cooling.

The properties of inhibited lymates. Heat treat-ment. The rapid rise in titer of phage T2 whenexposed to high temperatures suggested thatactivation might complicate heat inactivationstudies on T2 lysates. In fact, unlike the other Tphages, which show exponential rates of heatinactivation, T2 has been reported as exhibitinga multiple-hit type of inactivation curve (Adams,1949).

Fresh phage lysates were diluted 1:50 indistilled water or in broth. After one hour at37 C further 1:100 dilutions in nutrient brothwere made, and aliquots were placed in separatetubes in a 65 C water bath. The reaction wasstopped at various intervals by dumping 9

432 [VOL. 68

on February 12, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

4REVERSIBLE INHIBITION OF BACTERIOPHAGE T2

minutes, 65 a

Figure 3. Treatment of lysates at 65 C.

volumes of chilled nutrient broth into a reactiontube. Each sample was asyed both with andwithout further treatment in distilled water at37 C; the "treated" assay gives the full titer ofresidual phage, while the "untreated" asay

gives the titer of residual noninhibited phageplus residual heat activated phage. The resultsof such an experiment are shown in figure 3.The activated phage decays exponentially. Thenonactivated phage does not; in platings with-out water treatment there is a rise in titer to a

peak at 2 to 3 minutes and then an exponentialdecay. Nonactivated phage plated in watershows no rise but a plateau with subsequent ex-

ponential decay.Treatment with antiphage serum. Most phages

are iMeativated by specific antiphage sera at an

exponential rate. With phage T2 exponentialinactivation is observed with one-step lysatesand with activated mass lysates. Nonactivatedlysates treated with dilute antiphage serum

exhibit a rise in titer followed by a decline atthe same rate as for activated phage (figure 4).

Neither normal rabbit serum nor serum

against the host bacteria raised the titer of thecrude lysates. Antiphage serum subjected totwo absorption cycles with young E. coli, strainB (2 X 108 cells per ml; mixture kept at 48 Cfor one hour and at 9 C for twenty-three hours),could still raise the titer of crude T2 lysates.Absorption of the antiviral antibody with ultra-violet inactivated phage T2 (1 X lO1 phage perml) completely removed the ability of the serum

to raise the T2 titer as well as its ability to in-activate T2. The results suggest that specificantibody can remove inhibitor from the phageparticles without necessarily inactivating them.

Ultraviolet inactivation. Nonexponential inac-tivation of T2 (and T4, T5, T6) is found withboth inhibited and noninhibited T2. Both lysatesgive curves that extrapolate to an initial valueabove the point of origin (Benzer et al., 1950;Latarjet and Morenne, 1951).

Prevention of inhtion. The results above sug-gest that inhibition is the result of phage par-ticles combining with srface elements of bac-teria. In fact, inhibition is prevented bytreatments that prevent attachment. The rise oftiter is absent in lysates produced by allowinglysis to occur in salt-free broth or in 10 per cent(NH4)2S04, where T2 attachment to host bac-teria does not occur. Antiserum against thebacterial host partially prevents the formationof inhibited phage.

The properties of inhibited phage particles.Plaque morphology. The most striking charac-

ouwO-T& s WOdWtd *s

*eTaSdse) sehisikm*eA-TvtooiniitwpWh*ftheb wi-

pse-Tmaponibitd lsait afte

400serum&diue1250 Theconros inlued

(o"-)ttwtaigsr subdimentmas hwith both

250

PHASE 20

0 2 4 6 a 105?NUTES

Figure 4. Activation of inhibited phage by anti-phage serum. Samples of an inhibited lysate, afterdilution in nutrient broth, were treated with anti-T2 serum diluted -1:2,500. The controls included:(a) treatment with antiphage sierum subjected totwo absorption cycles; with young cells, (b) treat-ment with antiphage serum absorbed with bothyoung cells; and ultraviolet inactivated phageT2, (c) treatment with normal rabbit serum, (d)treatment with antibacterial serum.

1954] 433

on February 12, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

BERNARD P. SAGIK

teristic of inhibited T2 lysates is their poorplaque forming ability. The inhibited lysatesproduce a majority of small, irregular plaques.The more inhibited the lysate, the greater is theproportion of these poor plaques. Distilled wateractivation, in addition to activating phage,changes all plaques to their normal morphology.This finding suggests that inhibition is not an all

or none phenomenon, but that there are transi-tion forms between noninhibited phage and fullyinhibited phage. Presumably, the poor plaquesstem from partly inhibited phage particles thatare still adsorbed by bacteria, although more

slowly.Adsorption of inhibited and activated T2. Ad-

sorption was studied by means of streptomycintreated bacteria, which permitted observationover a long period of time. Streptomycin (50 or

100 micrograms per ml) was added to activelygrowing cultures of E. coli, strain B, in nutrientbroth, and the cultures were aerated vigorouslyfor one hour; the proportion of survivors was

lower than 10-7. These streptomycin killed cells

can adsorb phage specifically but do not releasenew phage.

Aliquots of the cell suspension were mixedwith samples of inhibited lysates or of wateractivated lysates. Assays were made at intervalsto determine residual phage. Activated phage isadsorbed rapidly at an approximately exponen-tial rate. The noninhibited fraction of phagepresent in mas lysates is adsorbed less rapidly.The total phage, noninhibited as well as in-

hibited, is adsorbed very slowly; even after 24hours, 43 per cent of the total phage in freshlysates remains unadsorbed.

The specificity of the inhibitor for phage T2.Inactivation of phage T2 by fresh lyeates of otherphages. Among the T-even phages of E. coli,strain B, only T2 exhibits the inhibition-activa-tion phenomenon. Fresh lysates of phages T4and T6 showed no rise in titer even though thesame host cells would have produced inhibitedT2 lysates. The following experiment was per-

formed to determine whether inhibitory materialwas produced in T6 lysates or if the phage insuch lysates withstood or overcame inhibitoryaction. Known quantities of fully active T2 were

added to lysis inhibited, T6-infected brothcultures of E. coli, strain B, within 15 minutesafter visible clearing. The mixture was assayed15 and 60 minutes later for T2 and T6 activity,before and after water treatment (table 3). TheT6 titer showed no increase with water treat-ment. The added active T2 lost 75 per cent ofits titer in 15 minutes. This loss was reversedcompletely by water activation. If T2 was addedto the lysate two hours after clearing, no in-hibition of T2 occurred. Lysates of T6 preparedon E. coli, strain B/2, which is unable to adsorbT2, fail to inhibit phage T2 but inhibit phageT2h, which is adsorbed by strain B/2.The fact that not all the added T2 phage was

inhibited in the T6 lysates suggested that theinhibitors may disappear rapidly after lysis. Thefollowing experiment confirmed this suggestion.

BLE -3Inactivation of phage by fresh 1y8ates and by sonically disrupted cells

Broth cultures of Escherichia coli, strain B (1 X 108 per ml), were lysed by phage (lysis inhibitedby infection with 20 to 50 particles per cell) or sonic vibration. To aliquots of the lysates, knownquantities of fully active phage were added within 15 minutes after the lysates and cleared. After 15to 60 minutes, incubation at 37 C, the lysates were assayed before and after water treatment.

FINAL TITXR AFTERADDED PHIAGE HOST CELL LYSED LYSING AGENT INPUT TITER LOW POINT TITER DISTILLED WATER

TREATMENT

T2 B T6 1.6 X 109 6.2 X 10' 1.5 X 109T2 B T6 1.3 X 109 2.5 X 108 1.1 X 109T2 B/2 T6 1.2 X 109 1.2 X 109 1.3 X 109T2h B/2 T6 1.0 X 109 7.4 X 108 1.0 X 109T2r B T2 1.2 X 109 1.2 X 109 1.2 X 10'T2h B T2 1.4 X 10' 1.3 X 109 1.3 X 109

T2 B sonic waves 5.2 X 108 3.8 X 101 6.5 X 107T2 B (filtered) sonic waves 5.2 X 10' 5.0 X 10' 5.1 X 108

434 [VOL. 68

on February 12, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

REVERSIBLE INHIBITION OF BACTERIOPHAGE T2

Known quantities of phage T2r or T2h wereadded to T2-infected, lysis inhibited cultureswithin 15 minutes after complete lysis. Fifteenminutes later the mixtures were assayed for T2,T2r, T2h, with and without water treatment.There was an eightfold rise in T2 titer but nochange in T2 or T2h titer. Inhibitory materialhad been present as shown by the rise in T2titer, but it had become unavailable. Either thefree inhibitory material is unstable, or it is re-moved by combination with the phage.Phage inactivation by sonically disrupted cells.

The experiments of the preceding sections sub-stantiated the bacterial origin of the inhibitorspresent in fresh lysates. Bacterial extracts ob-tained by methods other than phage lysis mightalso contain such inhibitory material. Youngcultures of E. coli, strain B, in broth were dis-rupted by sonic vibration (25 minutes at 9kilocycles) leaving 5 X 103 survivors out of apopulation of 2 X 108 cells per ml. Unfilteredaliquots of this suspension were mixed withknown quantities of active T2 and incubated for30 minutes at 37 C. Unfiltered cell debris in-activated over 99 per cent of the added phageT2. Distilled water treatment reactivated 12 percent of this, a fifteenfold rise from the lowesttiter. Filtration of the sonicated cell suspensionthrough a Mandler candle removed the inac-tivating material; filtration of a mixture of phageand disrupted cells removed all the inhibitedphage. Thus, phage inhibited by sonically brokenbacteria differs from phage inhibited in naturallysates, which can pass bacterial filters.

DISCUSSION

Bacteriophage T2 is present in fresh lysates ofE. coli, strain B, in an inhibited form, fromwhich it is slowly released. Several problems areof interest in relation to this inhibition: thenature of the inhibitor, the mechanism of phage-inhibitor combination, the difference between T2and other related phages.

In favor of the bacterial origin of the inhibitorand of its combination with phage after lysis arethe observations that inhibition occurs only inmass lysates and that it can be prevented bytreatments that prevent phage-bacterium com-bination (removal of NaCl, addition of an excessof (NHI)2SO4, addition of antibacterial serum).The specificity of the inhibition, shown by ex-periments on the inhibition of one phage in

fresh lysates of another, is similar to the spe-cificity of phage adsorption.

Inhibited phage is present in lysates in theform of single particles. This indicates that theinhibitor, which is filtrable, is of such a natureas not to bind together several particles. In-stead, sonic extracts of bacteria give a partiallyreversible inhibition, but these inhibitors arenot filtrable.The inhibitor in fresh lysates acts by pre-

venting or retarding the adsorption of phage T2onto the susceptible cells. Schlesinger (1932),Delbruck (1940), and Puck et at. (1951) havecompared the adsorption rate of phage withthe maximum possible rate (100 per cent colli-sion efficiency). Under optimum conditions, thetwo rates are similar. In the inhibited lysatesthe phage appears to be in a variety of formswith collision efficiencies ranging down to zero.Puck and his co-workers (Puck et al., 1951;

Garen and Puck, 1951) have analyzed the roleof various ions in phage-host combination. In amedium that by itself does not promote phageattachment, the monovalent ions (Na+, K+) aremost effective in promoting adsorption of phageT2 at a concentration of 0.1 M. The rate of ad-sorption diminishes below this concentrationand is negligible in distilled water. Higher ionconcentrations beyond the optimum level re-tard adsorption. These ionic effects are con-cerned with the first, reversible reaction ofadsorption.

Inhibition of phage T2 in fresh lysates isprevented by ammonium sulfate in a concentra-tion equivalent to 0.76 M and can be reversedby a 1:50 dilution of broth in distilled water,lowering the sodium ion concentration to about2 X 10-3 M. Slow activation occurs upon dilutionof sodium ions to a concentration around 1.7 X10-2 M, which is below the optimum for ad-sorption.We have no explanation of the mechanism of

activation by heat treatment. Foster et al. (1949)noted a rise in titer in phage T2 after heatingand proposed the following interpretations: (a)breaking up of clumps; (b) increased enzymeactivity on the part of the phage; (c) some formof heat activation of the phage itself, as inspore germination. Our experiments appear todisprove the first and third hypothesis and donot support directly the second one. Studies onthe activation of inhibited phage T2 and on the

19541 435

on February 12, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

BERNARD P. SAGIK

disappearance of inhibitors from lysates of otherphages may indeed reveal some enzymatic ac-tions of phage.An unexpected result is the activation of in-

hibited phage T2 by antiphage serum. The in-hibited phage might be combined with inhibitorat many places or spots, each combination beingindividually reversible; the over-all combinationmight be almost irreversible because of themultiplicity of simple reversible attachments.Antibody molecules capable of combining withthe phage, without inactivating it, could reducethe number of sites available to the inhibitorand, in effect, pry the inhibitor off the phage.It would be desirable to test whether activationis due to neutralizing antibody or to nonneu-tralizing antibody which combines with phageheads only (Lanni and Lanni, 1953).

SUWMARY

Fresh lysates of bacteriophage T2 exhibit aspontaneous rise in titer (activation) which canbe accelerated by a variety of treatments: dilu-tion in distilled water, heating, treatment indilute antiphage serum.

Dilute T2 lysates, in which lysis has occurredat low bacterial concentrations, do not show therise in titer; their titer corresponds to the maxi-mum titer obtainable from mass lysates. Evi-dence is presented indicating that the activationis due to removal of an inhibitor of bacterialorigin with which the phage becomes combinedfollowing lysis. Partially inhibited particles ofphage T2 are adsorbed slowly by host bacteriaand form poor plaques. Active phage T2 can beinhibited reversibly by fresh lysates of otherphages produced on T2-sensitive bacteria. It isinhibited in a partially reversible way by soni-cally disrupted noninfected bacteria. The in-hibition of phage T2 and its reversal by heatingaccount for the apparent anomalies of thekinetics of thermal inactivation of this phage.They do not account for the nonexponential rateof ultraviolet inactivation.

REFERENCESADAms, M. H. 1949 The stability of bacterial

viruses in solutions of salts. J. Gen. Physiol.,32, 579-594.

BENZER, S., DELBRttCK, M., DULBECCO, R.,HUDSON, W., STENT, G. S., WATSON, J. D.,WEIDEL, W., WEIGLE, J. J., AND WOLLMAN,E. L. 1950 Syllabus. In Viruses, 1950.Calif. Inst. Tech. Bookstore, Pasadena, Calif.

DELBRtYCK, M. 1940 Adsorption of bacterio-phage under various physiological conditionsof the host. J. Gen. Physiol., 23, 631-642.

DELBRtCK, M. 1946 Bacterial viruses or bac-teriophages. Biol. Revs. Cambridge Phil.Soc., 21, 30-40.

DOERMANN, A. H. 1948 Lysis and lysis inhibi-tion with Escherichia coli bacteriophages.J. Bacteriol., 55, 257-276.

ELLIS, E. L., AND DELBRtYCK, M. 1939 Thegrowth of bacteriophage. J. Gen. Physiol.,22, 365-384.

FOSTER, R. A. C., JOHNSON, F. H., AND MILLER,V. K. 1949 The influence of hydrostaticpressure and urethane on the inactivation ofbacteriophage. J. Gen. Physiol., 33, 1-16.

GAREN, A., AND PUCK, T. T. 1951 The first twosteps of the invasion of host cells by bacterialviruses. II. J. Exptl. Med., 94, 177-189.

GRATIA, A. 1936 Des relations numeriquesentre bact6ries lysogenes et particules debact6riophage. Ann. inst. Pasteur, 57, 652-676.

LANNI, F., AND LANNI, Y. T. 1953 Antigenicstructure of bacteriophage. Cold SpringHarbor Symposia Quant. Biol., 18, 159-168.

LATARJET, R., AND MORENNE, P. 1951 Inacti-vation d'un bact6riophage par un rayonne-ment ultra-violet de tres faible intensite-Ann. inst. Pasteur, 80, 220-224.

PuCK, T. T., GAREN, A., AND CLINE, J. 1951The mechanism of virus attachment to hostcells. I. The role of ions in the primary at-tachment. J. Exptl. Med., 93, 65,88.

SCHLESINGER, M. 1932 tJber die Bindung desBakteriophagen an homologe Bakterien. I.Die Unterscheidung von Gruppen verschiede-ner BindungsaffinitAt innerhalb der Bakteri-ophagen desselben Lysats. Die Frage derReversibilitAt oder Irreversibilitat der Bin-dung. Z. Hyg. Infektionskrankh., 111, 136-148.

STENT, G. S., AND WOLLMAN, E. L. 1950 Studieson activation of T4 bacteriophage by co-factor. II. The mechanism of activation.Biochim. et Biophys. Acta, 6, 307-316.

436 [VOL. 68

on February 12, 2020 by guest

http://jb.asm.org/

Dow

nloaded from