JOURNAL OF BACTERIOLOGY, Feb., 1966 Vol. 91, No. 2Copyright © 1966 American Society for Microbiology Printed in U.S.A.

Cationic Proteins of PolymorphonuclearLeukocyte Lysosomes

II. Composition, Properties, and Mechanism of Antibacterial ActionH. I. ZEYA AND J. K. SPITZNAGEL'

Department of Bacteriology and Immunology, University ofNorth Carolina School of Medicine,Chapel Hill, North Carolina

Received for publication 11 October 1965

ABSTRACT

ZEYA, H. I. (University of North Carolina, Chapel Hill), AND J. K. SPITZNAGEL.Cationic proteins of polymorphonuclear leukocyte lysosomes. II. Composition,properties, and mechanism of antibacterial action. J. Bacteriol. 91:755-762. 1966.-A basic proteins fraction from guinea pig polymorphonuclear (PMN) granules wasobtained by acid extraction and precipitation with 20% (v/v) ethyl alcohol. Thefraction accounted for most of the antibacterial activity of the PMN granules andcorresponded to the antibacterial cationic components of intact granules (bands I,II, and III) resolved by zone electrophoresis. Absence from the fraction of com-ponents identical to the enzymatic components of intact lysosomes showed that thefraction was essentially free from lysosomal enzymes. The amino acid analysis ofproteins in the fraction gave a preponderance of basic amino acids (25 %), especiallyof arginine (16%). The comparative amino acid analysis showed that the lysosomalcationic proteins (LCP) fraction was markedly different from nuclear histones. TheLCP fraction manifested antibacterial activity against certain gram-negative andgram-positive microorganisms, including Candida albicans, and exhibited stoichio-metric relationship in its activity. Microorganisms treated with LCP fraction wereagglutinated. Anionic substances such as nucleic acids, heparin, and endotoxin effec-tively blocked the antibacterial activity of the fraction. The LCP fraction causedsuppression of oxygen uptake by bacterial cells and damaged the permeability bar-riers of cells as manifested by rapid release of P32 as well as ultraviolet-absorbing ma-terial at 260 m,u, in the supernatant fluid.

The preceding paper dealt with the electropho-retic resolution of the antibacterial activity ofpolymorphonuclear (PMN) lysosomes, distinctfrom known lysosomal enzymes. The activity wasassociated with cationic protein componentswhich migrated fastest towards the cathode aheadof lysozyme and other enzymes.

This paper describes some of the biological andbiochemical properties of the lysosomal cationicproteins in relation to the mechanism of theirantibacterial activity.

MATERIALS AND METHODS

Preparation of whole granules. The PMN granulefraction was prepared from guinea pig peritonealexudate cells by homogenization and differentialcentrifugation as described in the preceding paper (33).

1 Public Health Service Research Career Develop-ment Awardee.

Preparation of ethyl alcohol fraction. Fractionalprecipitation of acid extracts of granules was carriedout according to the method of Ui (30) with modifica-tions. Granule suspension (0.2 ml) in 0.25 M sucrosewas extracted with 1.8 ml of 0.2 N H2S04 at 4 C for30 min. The mixture was then centrituged at 10,000 Xg for 10 min to remove the debris. The clear super-natant fluid was transferred to an ice-chilled glasscentrifuge tube, and the temperature was brought to-4 C. Absolute ethyl alcohol at the same tempera-ture was added very slowly with constant gentle stir-ring to a final concentration of 20% (v/v). The mix-ture was allowed to stand for 3 to 4 hr at -4 C forprecipitate formation. The precipitate was then col-lected by centrifugation at 250 X g for 15 min at -4C. The precipitate was immediately redissolved inice-cold distilled water and was dialyzed overnightagainst changes of distilled water. The dialysate waslyophilized and stored in tightly stoppered bottles at-10 C.

Antibacterial assay. Assay of the 20% ethyl alcohol755

on July 14, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

ZEYA AND SPITZNAGEL

fraction was carried out according to the method ofHirsch (14).

Paper electrophoresis. Electrophoresis was doneaccording to the method described earlier (32).Amino acid analysis. Analysis of 20% ethyl alcohol

fraction was carried out according to the procedure ofSpackman et al. (26) in an automatic amino acidanalyzer. For hydrolysis, 1.5 mg of sample was treatedat 110 C with 6 N HCI in evacuated sealed tubes for22 hr.The ultraviolet-absorption curve of the fraction was

prepared in citrate phosphate buffer (pH 5.6) with aBeckman DB spectrophotometer.

Time sequence ofbactericidal action of ethyl alcoholfraction. An 18-hr culture of Escherichia coli0117:H27 was washed three times in buffer solutionand resuspended to give 1.3 X 109 viable units per mlof citrate phosphate buffer (pH 5.6). The suspensionwas treated with 50 ,ug of 20% ethyl alcohol fractionand incubated at 37 C. Immediately, and at 15, 30,60, and 120 min of incubation, samples were with-drawn for serial dilutions. Measured amounts ofsuitable dilutions were inoculated into melted nutrientagar, and pour plates were prepared in sterile 9-cmpetri dishes. The plates were incubated at 37 C for 24hr, and the colonies were counted. The decrease inviable count over the period of time was plottedgraphically.

Effect on oxygen consumption ofE. coli. To a seriesof Warburg flasks, 1 ml of 18-hr washed culture of E.coli suspension containing 0.25 mg (dry weight) ofbacteria, 0.5 ml of 0.06 M glucose, and 0.5 ml ofethyl alcohol fraction containing 150 pg of proteinwere added. Oxygen consumption was recorded every10 min in the water bath at 37 C. The data obtainedwere plotted as microliters of oxygen consumedagainst time.

Permeability experiments. For permeability experi-ments, 18-hr washed cultures of E. coli were used. A1-ml amount of E. coli suspension containing 0.25 mg(dry weight) of bacteria was added to 2 ml of ethylalcohol fraction containing 50 pg of protein per ml incitrate phosphate buffer (pH 5.6), with proper con-trols. The tubes were kept at ice-cold temperatureduring the period. After addition of bacteria, thetubes were immediately transferred, and incubationwas carried out for 1 hr at 37 C. At the end of theperiod, the tubes were centrifuged for 15 min at 5,000rev/min, and the clear supernatant fluid was analyzedin a Beckman DB spectrophotometer between wave-lengths of 300 and 220 m, for the presence of ultra-violet light-absorbing substances released from thebacterial cells.

Release of P". E. coli 0117:H27 cells were labeledwith P3-orthophosphate. The medium employed forgrowth of E. coli contained a complex mixture ofamino acids (21), water-soluble vitamins, and salts.For maximal labeling, the orthophosphate concentra-tion in the medium was adjusted to 0.4 mmole perliter.A twice-washed suspension of labeled E. coli cells

containing 0.18 mg (dry weight) of bacteria wastreated with 50 pug of ethyl alcohol fraction in citratephosphate buffer (pH 5.6) and was incubated for 1 hr

at 37 C. A measured quantity of clear supematantfluid obtained after centrifugation was dried in plan-chets and analyzed for radioactivity in a proportionalcounter (Nuclear Measurement Corp., Indianapolis,Ind.). The supernatant fluids at zero time and after1 hr of incubation were analyzed with proper con-trols. The released radioactivity was expressed ascounts per minute per 0.1 ml of supernatant fluid.

Stoichiometric relationship of antibacterial activity.Twofold serial dilutions of ethyl alcohol fraction(from 5 to 0.15 jug) were prepared in citrate phosphatebuffer (pH 5.6) in 2-ml cups as described earlier (33).Six rows of such cups containing the fraction were setup. A measured amount of E. coli cells containing1.5 X 107 bacteria was added to each cup in the firstrow. The next five rows received a 10-fold decreasingconcentration of E. coli suspension, so that the lastrow received 1.5 X 102 bacteria per cup. The incuba-tion was carried out in the manner described previ-ously. The colonies were examined after 18 hr, and the50% inhibition end point in each row was recorded.

Effect of anionic substances. Commercial deoxyri-bonucleic acid (DNA), ribonucleic acid (RNA), hepa-rin, E. coli cell wall, endotoxin, and mucopeptides(streptococcal) were tested for blocking effects againstthe antibacterial activity of ethyl alcohol fraction.A measured amount of 20% ethyl alcohol fraction

was mixed with varying concentrations of anionic sub-stances and was assayed for its antibacterial effect onE. coli. Controls containing buffer and anionic sub-stances but without antibacterial protein and withbuffer and antibacterial substances but without ani-onic substances were used.

RESULTSEthyl alcohol fractionation. Various methods

for the fractionation of basic proteins have em-ployed acid extraction (8) because of the acidsolubility of such proteins. Among the mineralacids, preference has been expressed for dilutesulfuric acid over HCI, because a large and vari-able excess of the latter is required for precipita-tion (30). Ethyl alcohol has been considered oneof the most effective precipitants, besides beingmild and convenient (7).

Extraction of granules for 30 to 40 min with0.2 N sulfuric acid was sufficient to insure com-plete lysis of granules and liberation of their con-tent into the suspending fluid. The fraction pre-cipitable from the acid extract with 2D% (*/v)ethyl alcohol was found to be soluble in istlledwater and in solutions of acid pH. The fhactionaccounted for most of the antibacterial activityand on electrophoresis was resolved into bandscorresponding to the antibacterial bands It, II,and M of intact lysosomes (Fig. 1).Afrom ethyl alcohol fraction of other bAdre-sponding to the enzymatic bands of i4tact IysoPsomes showed that a fraction practically free fromlysozyme and other known enzymes was obtaied.Fractions obtained from rabbit PMN gtantes b

756 'DJ. AcnnuoL.

on July 14, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

LYSOSOMAL CATIONIC PROTEINS. II

20% EthanolFraction (LCP)

GRN...........

OR mNE III I lb la

FIG. 1. Comparative pherogram of 20% ethyl al-cohol-precipitated fraction (LCP) from guinea pigPMN Iysosomes and intact PMN lysosomes (GRN).Antibacterial bands: I-III. Lysozyme: IV. Ribonu-clease: V. Deoxyribonuclease: VI.

precipitation with 20% ethyl alcohol have beenshown by other workers to be free from knownacid hydrolases (17) including cathepsins (Goluband Spitznagel, Federation Proc. 23:509, 1964).The fraction, precipitated as the sulfate, was

found to be stable in acid solution for a longperiod of time. However, in distilled water itsantibacterial activity deteriorated during a periodof 48 hr, possibly because of aggregation and de-naturation of the protein molecule. The fractionwas heat-stable (100 C for 10 min at pH 3), agglu-tinated bacterial suspensions, and was precipitatedby acidic polymers.Amino acid analysis. The amino acid analysis of

the ethyl alcohol-precipitable fraction is given inTable 1. The basic amino acids, arginine and ly-

sine, predominated, constituting 25% of the totalamino acids. Arginine alone contributed almost16% of amino acids. The acidic amino acids, suchas glutamic acid and aspartic acid, comprised 17%of the total amino acids of the proteins. The pre-ponderance of basic amino acids in the 20%ethyl alcohol fraction puts the fraction proteins inthe category of cationic proteins. Hence, the pro-teins in the fraction will be referred to as lyso-somal cationic proteins (LCP).

Because of the high content of basic aminoacids, the LCP fraction was compared with vari-ous histones fractions (Table 1). A striking differ-ence was revealed in the content of arginine, ala-nine, and cystine. The LCP fraction containedtwice as much arginine as compared with most ofthe histone fractions. The histone fractions wererich in alanine and lysine which formed a moder-ate percentage of the amino acids in LCP frac-tion. Significant quantities of cystine were presentin LCP fractions, whereas histones contained lit-tle or no cystine. The comparison of amino acidcontent as well as electrophoretic patterns ofhistones (not presented here) showed that thecationic proteins of PMN lysosomes were differ-ent from nucleoproteins.The ultraviolet-absorption curve of the LCP

fraction is shown in Fig. 2. The curve showed apeak at 280 m,u, which is normal for purified pro-tein preparations with minimal nucleotide con-tamination.

TABLE 1. Comparison of amino acid composition ofPMN lysosomalwith that of nuclear histone fractions*

cationic proteins

Davison and Butler Daly and Mirsky Crampton et al.Lysosomal (12) (11) (10)

Amino acid cationicproteins

Fast Slow I II Total A B C

Asparticacid.......... 7.5 4.9 2.7 6.1 2.4 5.5 2.1 5.2 3.3Glutamic acid......... 9.1 8.5 2.3 9.9 4.0 8.2 3.6 8.2 5.8Glycine ............... 6.1 8.7 7.9 9.8 7.0 8.5 7.1 8.0 7.4Alanine .............. 6.4 11.6 27.4 12.1 19.1 12.8 24.5 11.3 19.6Valine ................ 5.8 6.9 5.0 4.3 4.5 6.4 5.3 6.4 4.7Leucine .............. 9.4 9.3 3.1 8.1 4.9 8.0 4.7 8.0 5.9Isoleucine ............ 4.9 6.5 1.3 3.4 1.5 4.2 1.1 4.5 2.4Serine ................ 3.4 2.7 5.9 5.0 6.0 4.1 7.1 7.0 6.8Threonine ...... .... 5.6 6.4 5.8 6.1 4.4 5.4 5.7 4.9 5.8Cystine ............... 3.5 0.4 0.0 0.4 0.0 0.0 0.0Methionine ........... 1.1 1.7 0.0 1.1 0.2 1.1 0.0 0.9 0.5Proline ............... 6.2 2.3 8.9 4.0 7.7 5.0 9.2 4.4 7.6Phenylalanine . . .. 4.2 2.8 0.0 2.3 0.8 1.8 0.7 1.3 1.4Tyrosine .............. 1.2 4.5 0.7 2.5 0.8 2.4 0.7 3.0 2.0Histidine.. 0.7 3.2 0.0 1.9 0.0 2.8 0.0 2.4 1.1Lysine ............... 8.5 8.4 26.2 11.5 33.7 13.7 26.0 12.6 18.8Arginine.............. 15.5 11.6 2.7 11.3 3.0 8.8 2.4 7.3 4.7

* Data, recalculated from the indicated sources, are expressed as percentages of total moles of aminoacids recovered.

757VOL. 91, 1966

on July 14, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

ZEYA AND SPITZNAGEL

TABLE 3. Effect of anionic substances onthe bactericidal activity of lysosomalcationic protein fraction (2 pg/mi)

on Escherichia coli (70 to80 bacteria/ml)

Concn (jug/ml)Anionic substance

5 3.0 2.0 1.5 0.5

Deoxyribonucleic acid... 75* 72 82 80 76Ribonucleic acid ........ 72 82 75 76 75Lipopolysaccharide...... 82 80 76 74 80Cell wall (E. coli)....... 84 76 78 81 74Mucopeptide

(Streptococcus) ........ 72 70 68 74 82Heparin................. 84 79 80 83 80

* Number of colonies which developed at 37 Cin each cup within 18 hr. No colonies developedwhen no anionic substance was present.

240 260 280 300 320 340

A - MlimicronsFIG. 2. Ultraviolet-absorption spectrum of 20%

ethyl alcohol-precipitated fraction (LCP) of PMNlysosomes, 100 ,ug/ml; 10-mm cuvette; citrate-phos-phate buffer, pH 5.6.

TABLE 2. Range ofbactericidal activityof lysosomal cationic protein fractionfrom gram-positive PMN granules

Amt of fraction re-quired for 50% inhi-

Test organism bition of organisms(ug/150 to 200organisms)

Escherichia coli ........ ........ O.15Klebsiella pneumoniae.......... 0.15Candida albicans ....... ........ 0.07Staphylococcus aureus ...... ..... 0.30Bacillus megaterium ....... ...... O.15

Range ofantibacterial action ofLCPfraction. Anumber of gram-positive and gram-negative mi-croorganisms were inhibited by small amounts ofthe LCP fraction (Table 2). Candida albicans, inview of its size and surface area, seemed excep-tionally sensitive to the microbicidal effects ofthe fraction. The antibacterial activity could beremoved from the solution by absorption with thebacteria.

Effects ofanionic polymers. The blocking effectof anionic substances on the interaction of basicproteins with bacterial cells has been reported(4, 5, 18, 28). The antibacterial activity of theLCP fraction was effectively neutralized by smallamounts of acidic polymers such as RNA, DNA,heparin, and endotoxin (Table 3). No attempt

"I 1.5x .07-a 1.5 x 106-

1.5x105-1.5 x 104-1.5 x 103 -

1.5 x102-

50 % inhibition (end pont)

0.15 031 0.62 1.25 2.5 5.0

,ug of 20% ETOH fraction (LCP)

FIG. 3. Effect of varying concentrations of bacteriaand LCP on 50% inhibition endpoint.

was made to determine the proportionality of thereacting substances.

Effect of varying bacterial concentrations. Basicpolypeptides exhibit a stoichiometric relationshipin their antibacterial activity (3, 28). The rela-tionship of the numbers of bacteria killed to theconcentration of LCP is represented graphicallyin Fig. 3. The relationship is interesting, as every10-fold increase in bacteria required a 2-fold in-crease in LCP concentration for 50% inhibition.It was not possible to count bacteria in dilutionslower than 1.5 X 102. Instead, 50% end points inthe cups were recorded by visual comparison ofthe colony density with that of controls. Underthese circumstances, the linear relationship maynot be altogether perfect; nevertheless, a typicaldose-response curve was obtained.

Time course ofbactericidal action ofLCPfrac-tion. The results (Fig. 4) showed that thee bacteri-cidal action of the LCP fraction expressed itselfas a rapid decrease in viable counts over a periodof 30 min. After 30 min, no more changes in theviable counts were observed, except for a slightrise at the end of 2 hr. This rise could be due to

I.00-

.9-

.8 -

.7-~U1)

CZ)

'031

.6

.5 "

.4

.3

.2-

.I

0r220

758 J. BACMOL.

.61

on July 14, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

LYSOSOMAL CATIONIC PROTEINS. II

I Xl

x1080 15 50 60 90 120

Time - Minutes

FIG. 4. Changes in the viability of Escherichia colitreated with LCP fraction (50 ,ug) in citrate-phosphatebuffer (pH 5.6) at 37 C. Symbols: A, untreated cells;0, cells treated with LCP.

TABLE 4 Effect of time on the antibacterialtiter of lysosomal cationic proteins

Protein concn (ug per ml per 150 bacteria)Time(min) 2.5 1.25 0.63 0.31 0.15 0.07 0.0

0 150* 148 152 155 153 155 15415 0 0 0 72 140 154 15730 0 0 0 0 74 152 15460 0 0 0 0 75 148 155120 0 0 0 0 72 150 152

* Number of colonies which developed at 37 Cin each cup within 18 hr.

reutilization by remaining viable cells of cellularconstitutents released from nonviable cells.

Similar results are shown in Table 4 where aseries of cups containing various concentrationsof LCP and a small number of bacteria were in-cubated for different lengths of time. It was ob-served that at 15 min only half of the bacterialnumber was viable in the cup containing 0.31 m,uof LCP per ml, and at 30 min there was completeinhibition in these cups. In cups containing 0.15m,i of LCP per ml, little inhibition occurred at 15min, whereas 50% inhibition of bacterial colonieswas seen at 30 min, and this inhibition continuedunchanged even on prolonged incubation.

Effect ofLCP on bacterial respiratory activity.Tissue basic proteins and synthetic basic polypep-tides have been reported to inhibit bacterial oxy-gen consumption (3, 18, 22, 28). In our study,immediate inhibition of oxygen consumption bybacterial cells treated with LCP fraction was ob-served starting within 10 to 15 min of incubation(Fig. 5). The bacterial suspension without LCPcontinued to consume oxygen in a linear fashionwith respect to time. The respiratory activity oftreated bacteria continued at reduced rate for 2 hrand then stopped.

Effect of LCP fraction on the permeability ofbacteria. In view of the reports that cationic de-tergents (24) and basic polypeptides such as tyro-cidin (16), polymyxin (23), and plakin (1) damagethe bacterial cellular membrane, resulting in therelease of cell constituents, the LCP fraction wastested for injurious action on bacterial permeabil-ity. Examination of the ultraviolet-absorptionspectra of the supernatant fluids from buffered,LCP-containing suspension of E. coli revealed anabsorption maximum at 260 m,t (Fig. 6). Com-pared with controls, an increase of over twofoldoccurred in 260-m,u absorbing material in themedium containing bacterial cells treated withLCP fraction. A dose relationship was found be-tween the concentration ofLCP in the suspendingmedia and the release of 260-m,u absorbing ma-terial from the bacterial cells.

Effect of LCP fraction on release of P32 from

360-340-320-300

280-

} 260-240-220-200-80-

8 160-

r 140-

120-oAoO

10 30 60 90 120 150

Time - Minutes

FIG. 5. Respiratory activity of Escherichia coli (1.2mg, dry weight) at 37 C in citrate-phosphate buffer(pH 5.6) with 0.06 M glucose. Symbols: 0, untreatedcells; X, cells treated with 150 ug/lml of LCP fraction.

759VOL.91, 1966

3

IA

e55si

on July 14, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

ZEYA AND SPITZNAGEL

1.00-

0t

.9

.8

.7

.6-

.5

.4-

.3

.2-

. 1

0O l

220 240 260 280 300 320

A - MilimicronsFIG. 6. Ultraviolet spectra ofsupernatantfluids from

Escherichia coli suspension (0.25 mg, dry weight, ofbacteria/ml) incubated at 37 C for I hr in 0.05 m

citrate-phosphate buffer, pH 5.6. Symbols: *, un-

treated cells; A, cells treated with 100 ug of LCPfraction.

TABLE 5. Effect of lysosomal cationic proteins(LCP) on release of p32 from labeledEscherichia coli* in citrate-phos-

phate buffer (pH 5.6)

Time (min)LCP (pg/ml)

0 60

0.0 337 t 6,72950.0 992 12,500

* The amount of cells was 0.18 mg (dry weight)/ml.

t Counts per minute per 0.1 ml of supernatantfluid.

labeled E. coli. As shown in Table 5, the releaseof P32 from bacteria treated with LCP fractionwas rapid, and was twofold greater than that ofuntreated bacterial cells. The release of P3 im-mediately after mixing of proteins with bacterialsuspension suggested rapid interaction at the levelof bacterial membrane.

DIscussIoNGuinea pig PMN lysosomes contain bacteri-

cidal proteins that possess isoelectric points higherthan those of a well-known basic protein lysozyme

(isoelectric point, 10.5 to 11.0). Although hetero-geneous electrophoretically, the proteins can beprecipitated, practically free from lysosomal en-zymes, from the acid extracts of PMN granules asone fraction in 20% (v/v) ethyl alcohol. The pro-teins contain a high concentration of basic aminoacids, particularly arginine, and resemble tissuebasic proteins in physicochemical and certainbiological properties. Basic proteins like histonesand protamine are essentially nuclear in origin,but the evidence for the existence of similar sub-stances outside nuclei is not completely lacking.Butler, Cohn, and Simson (6) isolated basic pro-teins from a microsomal fraction of cytoplasm.That histonelike proteins in ribosomes may belatent ribonucleases was suggested by Leslie (19).In our experiments, the lysosomal cationic pro-teins did not seem to possess any enzyme activityunder the limited number of enzyme asys car-ried out on the proteins.

Despite some degree of correspondence in theamino acid composition of LCP fraction and var-ious histone preparations (Table 1), the LCP frac-tion nevertheless differed significantly in posses-sing a very high concentration of arginine. Theheterogeneity of histones is well known, and theLCP also seemed to be heterogeneous electropho-retically; however, a significantly high content ofcystine in the LCP and its absence in histoneswould suggest that some components of the LCPfraction are structurally more complex than arenuclear histones.

Basic protein components thought to have orig-inated from the nuclei of the PMN (leukdn) wereisolated by Skarnes and Watson (25). The frac-tion contained (on a moles per cent basis) smalleramounts of arginine and lysine, and lacked anycystine. However, the method of extraction (ex-traction of whole cells with lactic acid) would notpreclude contamination with lysosomal cationicproteins. On the other hand, the fraction mani-fested selective bactericidal activity against gram-positive organisms, whereas LCP was equallyeffective against gram-positive and gram-negativeorganisms as well as C. albicans. The bactericidalactivity in acid extracts ofPMN granules ha beendefined operationally and named "phagocnyti"(9). In our experience, "phagocytin" turned outto be a heterogeneous mixture of enzymes andbasic proteins. It is likely that the antibterialactivity ascribed to phagocytin resides in thelysosomal cationic polypeptides.One of the important features of tissue and

synthetic basic polypeptides is their inactivationby anionic polymers (3, 5, 18, 28). Nucleic acids,heparin, endotoxin, and other microbial cell waRcomponents effectively neutralized the antibac-terial activity of the LCP fraction. The results

J. BA4bnRIOL.760

II

on July 14, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

LYSOSOMAL CATIONIC PROTEINS. II

lend credence to the concept that acidic polymersof bacterial and tissue origin may neutralize basicproteins of PMN granules, thus augmenting theinvasive properties of certain pathogenic orga-nisms. The resistance-lowering activity of mucin(20) may be partly attributable to neutralizationof such proteins from PMN leukocytes.The strongly basic nature of the LCP fraction,

the rapidity with which it reduced bacterial viabil-ity, and its high reactivity with anionic substancesindicate that the interaction with bacterial cellsinvolves the formation of strongly electrostaticbonds between the acidic groups in cell compo-nents and amino and guanidinium groups of thebasic amino acids in the proteins. Rapid clumpingof the bacterial cells is yet another evidence ofsuch interaction.The rapid release from the bacterial cells ex-

posed to lysosomal cationic proteins of P32 andmaterials having absorption maxima at 260 m,iwould indicate that the proteins interacted withand disorganized the structures of bacterial cellsresponsible for the maintenance of normal perme-ability barriers. Inhibition of bacterial oxygenconsumption and rapid drop in viable count sug-gest that the cationic proteins damage the cellmembrane, and thereby are able to enter the celland further combine with nucleic acids and intra-cellular enzyme systems. Evidence has been pre-sented by Becker and Green (2) that histones andprotamines enter mammalian cells and combinewith nucleic acids. The reaction was followed byan abrupt cessation of incorporation of labeledamino acids into the cellular proteins.High reactivity of lysosomal cationic proteins

with substances carrying an opposite charge limitsenvironmental conditions favorable to their anti-bacterial activity. In vivo, close proximity may berequired between bacteria and the cationic pro-teins. The situation in phagocytic vacuoles evi-dently provides an environment sufficiently re-stricted, and with other conditions suitable forinteraction of LCP with bacterial cells. Lysis ofgranules during phagocytosis and transfer ofgranular components including basic proteins tobacterial cell wall is well established (13, 15, 29).The membrane-damaging effect of cationic pro-teins with consequent nonviability of bacterialcells may be a prerequisite for the degradation ofbacteria by lysosomal hydrolytic enzymes in thephagocytic vacuoles.

Currently, it is becoming clearer that PMNleukocytes are not functionally restricted to therole of phagocytosis, but rather occupy a morediverse position in the mediation of inflammatoryresponse and tissue injury (27). The capacity oflysosomal cationic proteins to invoke capillarypermeability, edema (Golub and Spitznagel, Fed-

eration Proc. 23:509, 1964), leukocytic diapedesis,and petechial hemorrhage (17), and the inabilityof lysosomal enzymes to produce inflammationin equally low concentration, is significant in viewof recent reports in the literature implicatinglysosomal enzymes in tissue injury (31).Our results show that the specific granules of

PMN leukocytes contain certain strongly cationicproteins which by virtue of their unique physico-chemical properties are bactericidal to a variety ofmicroorganisms. In conjunction with their capac-ity to produce inflammation and tissue injury,the lysosomal cationic proteins may constitute animportant biochemical determinant in host defense.

ACKNOWLEDGMENTS

This investigation was supported by Public HealthService grants GMK-315-155 from the Division of Gen-eral Medical Sciences and AI-02430 from the Na-tional Institute of Allergy and Infectious Diseases, andby General Research Support Grant 1-GS-106.We gratefully acknowledge the expert technical help

of Marjorie H. Cooney.

LITERATURE CrrED

1. AMANO, T., M. SHIMA, AND K. KATO. 1953. Thedamage of semipermeability of bacteria causedby plakin. Med. J. Osaka Univ. 4:277-284.

2. BECKER, F. F., AND H. GREEN. 1961. The effects ofprotamines and histones on the nucleic acidsof ascitic tumor cells. Exptl. Cell Res. 19:361-375.

3. BLOOM, W. L., AND F. G. BLAKE. 1948. Studies onan antibacterial polypeptide extracted fromnormal tissues. J. Infect. Diseases 83:116-123.

4. BLOOM, W. L., M. G. WINTERS, AND D. W.WATSON. 1951. The inhibition of two antibac-terial basic proteins by nucleic acids. J. Bac-teriol. 62:7-13.

5. BURGER, W. C., AND M. A. STAHMAN. 1952. Theagglutination and growth-inhibition of bac-teria by lysine polypeptides. Arch. Biochem.Biophys. 39:27-36.

6. BUTLER, J. A. V., P. COHN, AND P. SIMSON. 1960.The presence of basic proteins in microsomes.Biochim. Biophys. Acta 38:386-388.

7. BUTLER, J. A. V., P. F. DAVISON, AND D. W. F.JAMES. 1953. Isolation and properties of histonesfrom thymus nucleoproteins. Biochem. J. 54:xxi.

8. BUTLER, J. A. V., P. F. DAVISON, D. W. F. JAMES,AND K. V. SHOOTER. 1954. The histones of calfthymus deoxyribonucleoproteins. I. Prepara-tions and homogeneity. Biochim. Biophys.Acta 13:224-232.

9. COHN, Z. A., AND J. G. HIRSCH. 1960. The isola-tion and properties of the specific cytoplasmicgranules of rabbit polymorphonuclear leuko-cytes. J. Exptl. Med. 112:983-1004.

10. CRAMPTON, C. F4, S. MOORE, AND W. H. STEIN.

VOL. 911 1966 761

on July 14, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

ZEYA AND SPITZNAGEL

1955. Chromatographic fractionation of calfthymus histones. J. Biol. Chem. 215:787-801.

11. DALY, M. M., AND A. E. MIRSKY. 1955. Histoneswith high lysine content. J. Gen. Physiol. 38:405-413.

12. DAVISON, P. F., AND J. A. V. BuTLER. 1954. Thefractionation and composition of histones fromthymus nucleoproteins. Biochim. Biophys. Acta15:439-440.

13. HIRSCH, J. G. 1962. Cinematographic observa-tions on granule lysis in polymorphonuclearleukocytes during phagocytosis. J. Exptl. Med.116:827-834.

14. HIRSCH, J. G. 1958. Bactericidal actions of his-tones. J. Exptl. Med. 108:924-944.

15. HORN, R. G., S. S. SPICER, AND B. K. WETzr.1964. Phagocytosis of bacteria by heterophileleukocytes: acid and alkaline phosphatase cyto-chemistry. Am. J. Pathol. 45:327-335.

16. HOTCHKISS, R. D. 1944. Gramicidin, tyrocidinand tyrothricin. Advan. Enzymol. 4:153-199.

17. JANoFF, A., AND B. W. ZWEIFACH. 1964. Produc-tion of infammatory changes in the microcircu-lation by cationic proteins extracted from lyso-somes. J. Exptl. Med. 120-.747-764.

18. KATCHALSKI, E., L. BICHOWSKI-SLoMNffzKI, ANDB. E. VOLCANI. 1953. The action of some watersoluble poly-amino acids on bacteria. Biochem.J. 55:671-680.

19. LEsLE, I. 1961. Biochemistry of heredity: a gen-eral hypothesis. Nature 189:260-268.

20. McLEoD, C. 1941. The mode of action of mucinin experimental meningococcus infection. Am.J. Hyg. 34.B51-B63.

21. MARcus, P. I., S. J. CICCuRA, AND T. T. PUCK.1956. Clonal growth in vitro of epithelial cellsfrom normal human tissues. J. Exptl. Med. 104:615-628.

22. MILLER, B. F., R. ABRAMS, A. DORFMAN, AND M.KLEIN. 1942. Antibacterial properties of pro-tamines and histones. Science 96:428-430.

23. NEWTON, B. A. 1956. The properties and mode of

action of the polymyxins. Bacteriol. Rev. 20:14-27.

24. SALTON, M. R. J. 1951. The adsorption of cetyl-trimethyl ammonium bromide by bacteria; itsaction in releasing cellular constituents, and itsbactericidal effects. J. Gen. Microbiol. 5:391-404.

25. SKARNES, R. C., AND D. W. WATSON. 1956. Char-acterization of leukin: an antibacterial factorfrom leukocyte active against gram-positivepathogens. J. Exptl. Med. 104:829-845.

26. SPACKMAN, D. H., W. H. STEN, AND S. MooRE.1958. Automatic recording apparatus for use inthe chromatography of amino acids. Anal.Chem. 30:1190-1206.

27. SPECTOR, W. G., AND D. A. WILLOUGHBY. 1963.The inflammatory response. Bacteriol. Rev. 27:117-154.

28. SPITZNAGEL, J. K. 1961. The effect ofmaalanand other cationic polypeptides on the cyto-chemical character of bacterial cells. J. Exptl.Med. 114:1063-1078.

29. SPITZNAGEL, J. K., AND H. Y. Cm. 1963. Cationicproteins and antibacterial properties of infectedtissues and leukocytes. Am. J. Pathol. 43:697-711.

30. UI, N. 1957. Preparation, fractionation and prop-erties of calf thymus histones. Biochim. Bio-phys. Acta 25:493-502.

31. WEISSMAN, G., B. BECHER, G. WEIDERMANN, ANDA. W. BERNHEIMER. 1965. Studies on lysosomes.VII. Acute and chronic arthritis produced byintra-articular injections of streptolysin S inrabbits. Am. J. Pathol. 46:129-147.

32. ZEYA, H. I., AND J. K. SPITZNAGEL. 1963. Antibacterial and enzymic basic proteins from leu-kocyte lysosomes: separation and identifica-tion. Science 142:1085-1087.

33. ZEYA, H. I., AND J. K. SPrrZNAGEL. 1966. Cationicproteins of polymorphonuclear leukocyte lyso-somes. I. Resolution of antibacterial and en-zymatic activities. J. Bacteriol. 91:750-754.

762 J. BACTERIOL.

on July 14, 2020 by guesthttp://jb.asm

.org/D

ownloaded from