damage to aspergillus fumigatus and rhizopus hyphaeby ...iai.asm.org/content/38/2/487.full.pdf ·...

INFECTION AND IMMUNITY, Nov. 1982, P. 487-495 Vol. 38, No. 20019-9567/82/110487-O9S02.00/0Copyright 0 1982, American Society for Microbiology

Damage to Aspergillus fumigatus and Rhizopus oryzaeHyphae by Oxidative and Nonoxidative Microbicidal Products

of Human Neutrophils In VitroRICHARD D. DIAMOND* AND ROBERT A. CLARK

Evans Memorial Department of Clinical Research and Department of Medicine, University Hospital, BostonUniversity Medical Center, Boston, Massachusetts 02118

Received 21 May 1982/Accepted 14 July 1982

Our previous studies established that human neutrophils could damage andprobably kill hyphae of Aspergillus fumigatus and Rhizopus oryzae in vitro,primarily by oxygen-dependent mechanisms active at the cell surface. Thesestudies were extended, again quantitating hyphal damage by reduction in uptakeof "4C-labeled uracil or glutamine. Neither A. fumigatus nor R. oryzae hyphaewere damaged by neutrophils from patients with chronic granulomatous disease,confirming the importance of oxidative mechanisms in damage to hyphae. Incontrast, neutrophils from one patient with hereditary myeloperoxidase deficien-cy damaged R. oryzae but not A. fumigatus hyphae. Cell-free, in vitro systemswere then used to help determine the relative importance of several potentiallyfungicidal products of neutrophils. Both A. fumigatus and R. oryzae hyphae weredamaged by the myeloperoxidase-hydrogen peroxide-halide system either withreagent hydrogen peroxide or enzymatic systems for generating hydrogen perox-ide (glucose oxidase with glucose, or xanthine oxidase with either hypoxanthineor acetaldehyde). Iodide with or without chloride supported the reaction, butdamage was less with chloride alone as the halide cofactor. Hydrogen peroxidealone damaged hyphae only in concentrations 21 mM, but 0.01 mM hypochlorousacid, a potential product of the myeloperoxidase system, significantly damaged R.oryzae hyphae (a 1 mM concentration was required for significant damage to A.fumigatus hyphae). Damage to hyphae by the myeloperoxidase system wasinhibited by azide, cyanide, catalase, histidine, and tryptophan, but not bysuperoxide dismutase, dimethyl sulfoxide, or mannitol. Photoactivation of the dyerose bengal resulted in hyphal damage which was inhibited by histidine, trypto-phan, and 1,4-diazobicyclo(2,2,2)octane. Lysates of neutrophils or separatedneutrophil granules did not affect A. fumigatus hyphae, but did damage R. oryzaehyphae. Similarly, three preparations of cationic proteins purified from humanneutrophil granules were more active in damaging R. oryzae than A. fumigatushyphae. This damage, as with the separated granules and whole cell lysates, wasinhibited by the polyanion heparin. Damage to R. oryzae hyphae by neutrophilcationic proteins was enhanced by activity of the complete myeloperoxidasesystem or by hydrogen peroxide alone in subinhibitory concentrations. These datasupport the importance of oxidative products in general and the myeloperoxidasesystem in particular in damage to hyphae by neutrophils. Cationic proteins mayalso contribute significantly to neutrophil-mediated damage to R. oryzae hyphae.

Our previous studies established that human increased frequency of mycotic infections, espe-blood neutrophils could damage and probably cially invasive aspergillosis, in patients withkill hyphae ofAspergillusfum;igatus or Rhizopus chronic granulomatous disease (8). In that he-oryzae in vitro (10). Indirect evidence derived reditary disorder of leukocyte function, microbi-from use of inhibitory substances suggested the cidal activity is impaired by virtue of an inabilityimportance of oxidative mechanisms in neutro- to generate normal amounts of hydrogen perox-phil-mediated hyphal damage, although polyan- ide and other oxidative intermediates (3).ions also inhibited damage to R. oryzae. The Although our previous studies were consistentimportance of oxidative mechanisms in host with a critical role for oxidative metabolism indefense against fungi has been supported by the neutrophil-mediated damage to A. fumigatus

487

on April 22, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

488 DIAMOND AND CLARK

and R. oryzae hyphae (10), they did not clarifywhich of several possible oxidative intermedi-ates or pathways (3) might act directly againstthese fungi. Effects of inhibitors of oxidativemetabolism are not necessarily completely spe-cific (9, 12, 38). For example, sodium azide andsodium cyanide may have general inhibitoryeffects on cell respiration in addition to theirinhibition of myeloperoxidase (18), and sodiumazide in high concentrations acts as a singletoxygen scavenger as well (14). Therefore, theeffects of various inhibitors on whole cells maynot reflect specific effects of these substances onan isolated system. Oxidative and nonoxidativemechanisms present in whole neutrophils maybe simulated in vitro by using purified microbi-cidal substances directly or by using cell-freecomponents of known systems to generate mi-crobicidal intermediates (6, 9). Thus, hydrogepperoxide may be used in reagent form or may begenerated by the reactions of glucose with glu-cose oxidase or by xanthine oxidase with eitherhypoxanthine or acetaldehyde (5, 14, 36). Thelatter reactions generate superoxide radical,which may then dismutate to form hydrogenperoxide or may, by itself, be microbicidal forsome organisms (36). Like whole neutrophils,these reactions, with or without interaction withmyeloperoxidase and halide, may produce otheroxidative products such as hydroxyl radical (30,36, 39, 43), hypochlorous acid (12, 13, 15, 38), orsinglet oxygen (14, 34, 36). Singlet oxygen alsomay be produced in vitro as the major oxidativeproduct of photooxidation of the dye rose bengal(14). Our previous studies using such cell-freesystems supported the critical involvement ofoxidative mechanisms in general and the myelo-peroxidase-H202-halide system in particular inneutrophil-mediated damage to Candida albi-cans hyphae and pseudohyphae (9). We nowextend these studies to mechanisms of damageof A. fumigatus and R. oryzae hyphae.

MATERIALS AND METHODSFungi. Previously described isolates ofA. fumigatus

and R. oryzae were maintained and harvested as in ourprior studies (10). After removal of clumps and hyphalfragments by filtration through cheesecloth, sporeswere washed, dispensed into individual 15-ml centri-fuge tubes (Coming Glass Works, Corning, N.Y.) in 1-ml working samples, and incubated to induce germina-tion at 37C. For optimum, consistent germination to.30 ,um in length, R. oryzae spores were incubated at10- spores per ml for 5 to 6 h in Emmons' modificationof Sabouraud broth and 106 A.fumigatus spores per mlwere incubated for 17 to 20 h in Eagle minimalessential medium supplemented with nonessentialamino acids (M. A. Bioproducts, Walkersville, Md.)as in our prior studies (9).

Neutrophils. Heparinized human peripheral venousblood was obtained from normal volunteer controlsubjects and patients, four with chronic granuloma-

tous disease (three typical male, sex-linked recessiveand one female autosomal, identified and kindly sup-plied by Harvey Cohen, Children's Hospital, Boston,Mass.) and one with hereditary complete myeloperoxi-dase deficiency (identified and kindly supplied byRichard K. Root, Yale-New Haven Hospital, NewHaven, Conn.). With the latter patient, blood fromcontrol subjects was obtained and processed simulta-neously. Both the normal and myeloperoxidase-defi-cient leukocytes were as active as freshly preparedleukocytes in causing cytotoxic damage to 1Cr-la-beled liposomes, although the patient's cells requiredadded purified myeloperoxidase to accomplish this.As in our previous studies (9), neutrophils were sepa-rated by dextran sedimentation after centrifugation ona Hypaque (sodium and meglumine diatrizoates; Win-throp, New York, N.Y.)-Ficoll (Pharmacia FineChemicals, Piscataway, N.J.) gradient as described byBoyum (4). Contaminating erythrocytes were lysed,and neutrophils were washed as in our prior studies(10). Subcellular fractions of neutrophils, includingthose enriched with cytoplasmic granules, were sepa-rated by centrifugation on sucrose gradients by themethod of Kimball et al. (17).

Preparations of cationic proteins previously de-scribed in studies of tumor cell cytotoxicity (7) wereused, having been prepared by Inge Olsson (Universi-ty of Lund, Lund, Sweden). These consisted of twoseparate lots designated cationic protein A, consistingof a mixture of cationic proteins 1 and 2, and cationicprotein B, consisting of a mixture of cationic proteins 3and 4 (7, 27). These proteins were dissolved in distilledwater at 500 ,ug/ml and stored at -70C until justbefore use.

Ingredients of in vitro systems for generation andinhibition of microbicidal activity. As in our priorstudies (9), myeloperoxidase was prepared from ca-nine pus neutrophils by the method of Agner throughthe end of the sixth step (1) and was assayed by the o-dianisidine method. One unit of myeloperoxidase uses1 ,umol of substrate per min at 25C (19). Glucoseoxidase (type V, 200 U/mg) was obtained from SigmaChemical Co. (St. Louis, Mo.), as were hypoxanthineand xanthine oxidase (bovine buttermilk suspended in2.3 M ammonium sulfate 10 mM sodium phosphatebuffer [pH 7.8] and containing 1 mM EDTA and 1 mMsodium salicylate). Catalase (bovine liver, 6.1 mg/ml,50,000 U/mg) was obtained from Worthington Diag-nostics (Freehold, N.J.) and dialyzed against waterbefore use. Superoxide dismutase (bovine erythro-cyte, lyophilized powder, 3,000 U/mg) obtained fromMiles Laboratories, Inc. (Elkhart, Ind.) was dissolvedin water at a concentration of 5 mg/ml and stored at-20C until used. Whenever active enzymes weretested, simultaneous control tubes contained heat-inactivated enzymes (100C for 15 min for myeloper-oxidase, xanthine oxidase, glucose oxidase, and cata-lase and autoclaving at 121C for 30 min for superoxidedismutase). Acetaldehyde (Fisher Scientific Co., Fair-lawn, N.J.) was distilled and stored in working sam-ples at -20C. Rose bengal was also obtained fromFisher. Hypochlorite was obtained from Sigma, andthe amount of active substances was quantitated bymeasuring absorbance at 350 nm, resulting from dis-charge of 12 and 13 from 10 to 100 mM KI as describedby Awtrey and Connick (2) and modified by Harrisonand Schultz (12). Inhibitors tested included sodium

INFECT. IMMUN.


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

DAMAGE TO ASPERGILLUS AND RHIZOPUS HYPHAE 489

azide, sodium cyanide, histidine, tryptophan, dimethylsulfoxide, supplied by Sigma, and 1,4-diazobicy-clo(2,2,2)octane, supplied as triethylenediamine by theEastman Organic Chemicals Division of the EastmanKodak Co., Rochester, N.Y.Damage to hyphae by neutrophils and cell-free sys-

tems. Triplicate tubes contained either 101 R. oryzae or106 A. fumigatus hyphae together with a 10-fold highernumber of neutrophils or various concentrations ofcell-free substances for generating microbicidal activi-ty. As in our previous studies (9), neutrophils weresuspended in Hanks balanced salt solution, whereaschlori&e-free 0.1 M phosphate buffer (sodium salts),pH 7.4, was used with cell-free systems, with no addedserum or albumin in reaction mixtures. Hyphae andneutrophils or cell-free systems were rotated togetherfor 1 h at 37C. Control tubes for cell-free systemscontained hyphae with buffer only or with incompletesystems (e.g., heat-inactivated enzyme or one compo-nent of the system lacking). Control tubes for intactneutrophils contained buffer only, with neutrophilsadded at the end of incubations. Neutrophils were thenlysed by the addition of 2.5% sodium deoxycholate(for A. fumigatus) or repeated washes in distilledwater (R. oryzae) as in our prior studies (10). Hyphaldamage was then assessed as previously described byquantitating reduction in uptake of radioisotopes([14C]glutamine or [14C]uracil for A. fumigatus,[14C]uracil for R. oryzae) induced by neutrophils orcell-free systems, calculated as follows: [(mean countsper minute in control tubes - mean counts per minutein experimental tubes)/mean counts per minute incontrol tubes] xlOO (10).

Statstical methods. Means and standard errors ofmeans were compared by two-tailed, two-sample t-tests.

RESULTS

Damage to hyphae by intact neutrophils withdeficient oxidative metabolism. Compared withneutrophils from normal subjects, cells frompatients with chronic granulomatous diseasewere defective in damaging A. fumigatus or R.oryzae hyphae. Likewise, neutrophils from asingle patient with myeloperoxidase deficiencydid not damage A. fumigatus hyphae, althoughdamage to R. oryzae hyphae by these cells wassignificant (Table 1).

Effects of microbicidal products of oxidativemetabolism on damage to hyphae. Lack of dam-age to hyphae by chronic granulomatous diseaseneutrophils indicated the importance of oxida-tive metabolism in damage to hyphae by neutro-phils, but did not specify which of the manypossible oxidative products of neutrophils (3)were capable of damaging hyphae. Preliminaryexperiments were then done by using cell-freesystems which generated oxidative products todefine concentrations of reagents that did notdamage hyphae by themselves, but whichcaused maximum hyphal damage when presentwith all components of the systems.The myeloperoxidase-H202--halide system

TABLE 1. Effect of defects in oxidative metabolismon ability of intact neutrophils to damage hyphae

Damage to hyphaea (%)Source of neutrophils Aspergillus Rhizopus

Chronic granuloma- 8.9 7.7 (4) 0.0 (2)tous disease

Myeloperoxidase defi- 0.0 (1) 32.9 (1)ciency

Normal control sub- 42.4 4.3 (7) 40.8 5.9 (5)jectsa Mean standard error of the mean of damage to

fungi by neutrophils (number of different subjectstested), determined by reduction of uptake of 14Cglutamine or uracil by A. fumigatus or R. oryzaehyphae after 1 h of incubation. Hyphal damage wascalculated as follows: [(mean counts per minute incontrol tubes - mean counts per minute in experimen-tal tubes)/mean counts per minute in control tubes] x100.

was effective in damaging both types of hyphae,whether reagent hydrogen peroxide was useddirectly or was replaced by systems for generat-ing H202 (Table 2). These systems includedglucose oxidase with glucose (5) and xanthineoxidase together with hypoxanthine or acetalde-hyde (14, 36). Without myeloperoxidase andhalide, higher concentrations ofhypoxanthine oracetaldehyde caused significant damage to hy-phae, but this was completely eliminated by theaddition of active, unheated catalase, but notsuperoxide dismutase. Heat-inactivated xan-thine oxidase or glucose oxidase did not inducedamage to hyphae, even when other compo-nents of the myeloperoxidase system were pres-ent. Damage to hyphae by the myeloperoxidasesystem was greater when iodide (with or withoutchloride) was present as a cofactor than withchloride alone (Table 2). Active myeloperoxi-dase was essential, as there was no hyphaldamage when heat-inactivated myeloperoxidasewas used or when myeloperoxidase was omittedfrom incubations. Without myeloperoxidase,higher concentrations of iodide (.1 mM) with orwithout H202 damaged hyphae. Hydrogen per-oxide in the concentration used in Table 2 (0.1mM) did not damage hyphae by itself, but diddamage hyphae directly in higher (1 to 10 mM)concentrations (Table 3). Moreover, hypochlo-rous acid, a product of the myeloperoxidasesystem (12, 13, 38), significantly damaged R.oryzae hyphae in concentrations as low as 0.01mM, although damage to A. fumigatus hyphaerequired a 100-fold higher concentration.

Putative inhibitors of oxidative microbicidalmechanisms were then used to further definewhich substances might have antifungal activity(Table 4). Damage to hyphae by the myeloper-oxidase system was inhibited by sodium azide

VOL. 38, 1982


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

TABLE 2. Damage to hyphae by the myeloperoxidase (MPO)-peroxide-halide system with reagent H202 orgenerators of H202: glucose oxidase (GO) plus glucose, xanthine oxidase (XO) plus hypoxanthine (HX), or

acetaldehyde (CH3CHO)

MPO Nal NaCi H202 GOa (1.4 mU) XO (4 mU) XO (4 mU) % Damage to hyphaeb(4 mU)a (0.1 mM) (0.1 M) (0.1 mM) + glucose + HX + CH3CHO*4 * (1.0 mM)a (0.1 mM)a (0.1 mM)a A. fumigatus R. oryzae

+ + + + - - - 83.5 5.0(8) 59.4 6.8 (3)+ + - + - - - 78.8 6.0(8) 57.4 9.4(5)+ - + + - - - 65.1 4.1 (4) 25.9 5.0(3)+ + + - + - - 85.8 3.4(6) 46.2 7.4(4)+ + + - - + - 49.7 7.8(3) 51.0 2.0(4)+ + - - - + - 45.5 4.8(9) 43.3 6.9 (4)+ - + - - + - 39.6 4.5(4) 12.1 4.7(3)+ + + - - - + 50.3 10.4(7) 38.7 4.2 (5)+ + - - - - + 54.3 8.8(4) 36.9 8.4(6)+ - + - - - + 41.4 12.2(4) 8.8 5.9(4)

a Controls which included heat-inactivated enzymes did not damage hyphae.b Mean standard error of the mean (number of experiments) determined by reduction of uptake of

[14C]glutamine or uracil by A. fumigatus or R. oryzae hyphae after 1 h of incubation.

and sodium cyanide in concentrations known toinhibit myeloperoxidase (18, 22). Moreover, cat-alase blocked hyphal damage, emphasizing theimportance of H202 in the reaction, althoughsuperoxide dismutase did not. Histidine andtryptophan, antagonists of hypochlorous acid(13, 15, 38) and also putative quenchers ofsinglet oxygen, also inhibited the myeloperoxi-dase system. However, potential scavengers ofhydroxyl radical, dimethyl sulfoxide and manni-tol (33, 36, 39), did not alter hyphal damage.Thus, these inhibitor studies did not support arole for hydroxyl radical in hyphal damage anddid not distinguish between the possibilities ofdamage by hypochlorous acid or singlet oxygen.Since the above studies showed that the formerwas, by itself, sufficient to damage hyphae (Ta-ble 3), we then studied the antifungal potential ofthe latter. Singlet oxygen is the major oxidativeproduct formed by photoactivation of the dyerose bengal (14). This reaction induced hyphaldamage which was inhibited by the singlet oxy-gen quenchers (16), tryptophan (37) and 1,4-diazobicyclo(2,2,2)octane (21, 30), so that sin-

TABLE 3. Damage to hyphae by reagent hydrogenperoxide or hypochlorous acid

Reagent added % Damage to hyphaeato incubations Concn (mM)A. fumigatus R. oryzae

H202 10 68.4 74.7H202 1 24.8 36.2H202 0.1 0.0 7.3HOCI 1 52.7 95.1HOCI 0.1 12.2 68.2HOCI 0.01 7.6 47.4HOCI 0.001 0.0 10.9

a Mean damage to hyphae after 1 h of incubation atpH 7.0 in two separate experiments.

glet oxygen appeared to be capable of damaginghyphae if produced (Table 5).Damage to hyphae by neutrophil cationic pro-

teins. Our previous studies suggested that non-oxidative mechanisms, especially cationic pro-teins, might contribute to damage to R. oryzaehyphae, as polyanions inhibited damage by in-tact, normal neutrophils (10). This hypothesiswas supported by experiments using fractionsprepared from lysed, normal neutrophils (Table6). Lysates of whole neutrophils and fractionsenriched with neutrophil granules damaged R.oryzae (but not A. fumigatus) hyphae, and thisdamage was reversed by the polyanion heparin.Citric acid extracts (0.01 M, pH 2.7) of wholecell lysates or granular pellets (31) gave compa-rable results (data not shown). Fractions rich incomponents from cell nuclei also damaged R.oryzae hyphae. Since these latter fractions con-tain significant amounts of potentially fungicidalcationic histones (11), heparin was added tothese fractions as well and similarly neutralizedantifungal activity.

Purified preparations of neutrophil cationicproteins damaged R. oryzae hyphae more thanA. fumigatus hyphae (Table 7). The former weresignificantly damaged by protein concentrationsas low as 10 ,ug/ml, whereas the latter required50 ,ug/ml. Heating at 90C for 10 min eliminatedenzymatic activity of the proteins, but consis-tently increased antifungal activity. In fact,these proteins in their natural, unheated stateshowed no significant activity against A.fumiga-tus hyphae, in contrast to results with R. oryzaehyphae. In all cases, antifungal activity of thecationic proteins was neutralized by the polyan-ion heparin.

It appeared equally plausible that these neu-trophil cationic proteins might act with differingspecificity toward different organisms as occurs

490 DIAMOND AND CLARK INFECT. IMMUN.


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


TABLE 4. Inhibition of damage to hyphae by the myeloperoxidase (MPO) system with H202, glucoseoxidase (GO) plus glucose, xanthine oxidase (XO) plus hypoxanthine (HX), or acetaldehyde

% Inhibition of damage to hyphaea

Inhibito added concnA. fumigatus with supplements to MPO + I- R. oryzae with supplements to MPO + I-

Inhibitor added (concn)Reagent GO + XO + HX xo + Reagent GO + HX +H202 glucose CH3CHO H202 glucose XO + CH3CHO

NaN3 (0.1 mM) 85.8 63.4 100.0 54.0 64.1 67.3 84.0 50.7NaCN (1.0 mM) 58.3 31.5 43.6 41.2 50.1 38.5 100.0 32.1Catalase (750 U)b 66.3 82.5 40.1 56.5 95.2 76.5 42.0 74.8SODC (,Lg/ml)b 1.8 0.0 0.0 0.0 0.0 4.2 0.7 0.0Histidine (1.0 mM) 47.4 54.5 96.0 67.9 22.3 49.6 72.9 38.3Tryptophan (1.0 mM) 49.1 12.2 88.2 92.8 63.2 60.2 76.5 46.9DMSOd (14 mM) 0.7 0.0 16.0 0.0 2.1 0.0 0.0 0.0Mannitol (400 mM) 0.7 0.0 0.0 0.0 0.0 0.0 0.4 0.0

a Mean inhibition of hyphal damage in two or more separate experiments, calculated as follows: (damagewithout inhibitor - damage with inhibitor)/damage without inhibitor x 100.

b Heat-inactivated enzyme was used in each experiment as a control.c SOD, Superoxide dismutase.d DMSO, Dimethyl sulfoxide.

in interactions with some bacteria (27, 28) ormight act primarily due to the charge density ofthe polycation. To investigate the latter possibil-ity, hyphae were incubated with a variety ofpolycations and polyanions, including polyami-no acids, protamine, and heparin (Table 8).Hyphae were consistently damaged by polycat-ions, but not by polyanions, and the latter neu-tralized the antifungal effects of the former. A.fumigatus hyphae were, if anything, more sensi-tive than R. oryzae hyphae to damage by thesepolycations.

In intact neutrophils from patients with chron-ic granulomatous disease, it appeared that cat-ionic proteins by themselves were insufficient tocause hyphal damage. We then investigated thepotential for interactions of cationic proteinswith the myeloperoxidase system. Subinhibitorylevels of hydrogen peroxide, together with othercomponents of the myeloperoxidase system or

separately (produced by the reaction of glucosewith glucose oxidase), damaged hyphae whencombined with subinhibitory levels of cationicproteins (Table 9). The effect was particularlystriking with R. oryzae hyphae, where concen-trations of cationic proteins as low as 1 ,ug/mlcontributed to hyphal damage.

DISCUSSIONOur previous studies established that neutro-

phils could damage and probably kill hyphae ofA. fumigatus and R. oryzae. These findings havenow been confirmed and extended, emphasizingthe critical importance of oxidative mechanismsof neutrophils in hyphal damage. Neutrophilsfrom patients with chronic granulomatous dis-ease did not damage hyphae, presumably be-cause oxidative intermediates, including super-oxide, hydrogen peroxide, hypochlorous acid,hydroxyl radical, and perhaps singlet oxygen are

TABLE 5. Damage to hyphae by rose bengal and inhibition of damageComponents of system

Photoactivated % Damage % Inhibitionrhoseactiengad Histidine Tryptophan DABCOc A. fumigatus R. oryzae to a nofbamag(1 ,uM) (1 mM) (1 mM) (0.1 mM) hyphae hyphae yphae

+ - - - + - 38.4+ + - - + - 10.9 71.5+ - + - + - 15.8 58.8+ - - + + - 7.1 81.4+ - - - - + 83.1+ + - - - + 0.0 100.0+ - + - - + 0.0 100.0+ - - + - + 58.5 29.6

a Mean of five experiments, calculated as in Table 1.b Calculated as in Table 4.c DABCO, 1,4-Diazobicyclo(2,2,2)octane.

VOL. 38, 1982


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


TABLE 6. Damage to hyphae by lysates andfractions of normal neutrophils

% Damage to hyphaeLeukocyte fraction

A. fumigatus R. oryzae

Whole cell lysate 2.1 31.2Nuclear pellet 14.9 29.8Postgranular supernatant 0.0 0.0Granular pellet 1.0 37.6Granular pellet + 25 ,ug 0.0

of heparin per mlNuclear pellet + 25 FLg of 0.0 0.3

heparin per ml

not produced in normal amounts (3, 20). Thesepatients have a propensity to recurrent mycoses,which developed in 20.4% of 245 cases; most ofthese were aspergillosis (8). In contrast, al-though potentially microbicidal oxidative prod-ucts are produced by myeloperoxidase-deficientneutrophils (3, 24), fungicidal activity, especiallyfor C. albicans (3, 9, 18, 23, 25), is impaired. Inour studies, myeloperoxidase-deficient neutro-phils failed to damage A. fumigatus hyphae, butdid damage R. oryzae hyphae. However, dam-age to hyphae might be attributable to a numberof different oxidative or nonoxidative mecha-nisms independent of myeloperoxidase (9, 24,26, 31, 32).Using cell-free systems (6, 7, 9, 21, 36), we

found that the myeloperoxidase system withchloride and iodide or iodide alone as cofactorefficiently damaged A. fumigatus and R. oryzaehyphae. Concentrations of sodium azide or sodi-um cyanide which primarily affect myeloperoxi-dase (18) inhibited antifungal effects, but both ofthese inhibitors may affect cell respiration, andsodium azide in high concentrations is also a

quencher of singlet oxygen (14, 36). As empha-sized by the pronounced inhibitory effect ofcatalase, hydrogen peroxide was required forhyphal damage by the myeloperoxidase system.Superoxide likely functions in these antifungalsystems primarily to supply hydrogen peroxide,rather than providing any direct antimicrobialeffects, since catalase, but not superoxide dis-mutase, inhibited hyphal damage mediated bythe xanthine oxidase system in the absence ofmyeloperoxidase. Hydrogen peroxide by itselfdamaged hyphae, but only in concentrations 10-to 100-fold higher than those active in the myelo-peroxidase system.A variety of specific antimicrobial substances

have been postulated as being produced byneutrophils either directly or as byproducts ofthe myeloperoxidase system (3, 20). Hypochlo-rous acid, a product of the myeloperoxidasesystem (12, 13, 38), damaged both A. fumigatusand R. oryzae hyphae, although the latter weresusceptible to 100-fold lower concentrationsthan the former. The potential effectiveness ofhypochlorous acid as an antimicrobial agent maybe increased by acidification compared withactivity at pH 7.4. Recent studies have demon-strated proton release at the surfaces of stimulat-ed neutrophils (40), so phagocytic vacuoles neednot be formed for acidification to occur. Inhibi-tion of the myeloperoxidase system by histidineand tryptophan is also consistent with inhibitionof the antifungal activity of hypochlorous acid,although these agents are not completely specif-ic as they also are singlet oxygen quenchers (13,15, 34). We did find that singlet oxygen itselfappeared to be able to damage hyphae, as evi-denced by the antihyphal activity of the pho-toactivated dye rose bengal (14). However, sin-glet oxygen may or may not be definitely proved

TABLE 7. Damage to hyphae by neutrophil granular cationic proteaseCationic Concn Heated (90'C, Heparin Damage to hyphaeaprotein (ILg/ml) 10 min) (25 ~Lg/ml) A uiau .oyaadded A uiau .oyaA, lot 2 50 Yes No 51.1 62.0A, lot 2 10 Yes No 22.8 34.9A, lot 2 50 No No 0.0 21.7A, lot 2 10 No No 0.0 26.0A, lot2 10 No Yes 0.0A, lot 1 50 Yes No 26.0 64.1A, lot 1 10 Yes No 0.0 24.8A, lot 1 50 No No 4.8 32.5A, lot 1 50 No Yes 0.0A, lot 1 10 No No 0.4 12.2B 50 Yes No 13.5 71.7B 10 Yes No 0.0 30.9B 50 No No 0.2 36.2B 50 No Yes 0.9B 10 No No 0.0 19.2

a Mean of two separate experiments.

INFECT. IMMUN.


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


TABLE 8. Damage to hyphae by polyanions andpolycations

% Damage to hyphae'Substances added (concn) Aspergilus Rhizopus

Polyaspartic acid (10 nM) 2.4 0.0Polyglutamic acid (10 nM) 2.3 0.0Polyarginine (10 nM) 42.0 22.0Polylysine (10 nM) 43.4 28.3Polylysine (10 nM) + 6.9 7.4

polyglutamic acid (10 nM)Polyarginine (10 nM) + 8.2 8.9

polyaspartic acid (10 nM)Protamine (2.5 p.g/ml) 60.1 37.7Heparin (2.5 ,ug/ml) 0.0 3.4Protamine (2.5 izg/ml) + 2.8 0.0

heparin (2.5 .ag/ml)a Mean of two or more separate experiments.

to be produced by neutrophils (12, 34, 38), andour data with inhibitors cannot separate singletoxygen-induced antihyphal effects from activityattributable to hypochlorous acid (13, 15, 38).Hydroxyl radical, another potentially microbi-cidal product of neutrophil oxidative metabolism(33, 35, 39, 43), damaged C. albicans hyphae inour previous studies (9). Whereas scavengers ofhydroxyl radical failed to inhibit damage to A.fumigatus or R. oryzae hyphae, the use ofoptimum concentrations of some inhibitors wasprecluded by their effects on hyphal growth inthe absence of neutrophils. Nevertheless, di-methyl sulfoxide was used in the same concen-trations which previously inhibited damage to C.albicans hyphae (9) and successfully scavengedhydroxyl radical in other systems (33). Inhibi-tory substances also may not penetrate to sitesof microbicidal activity between closely apposedcell surfaces, but our studies at least do notsupport a role for hydroxyl radical in damage toA. fumigatus and R. oryzae hyphae.

Potential nonoxidative antifungal mechanismsof leukocytes (24, 26, 31, 32) include granule-associated cationic proteins of neutrophils,which can kill Candida parapsilosis yeasts (24,26) . We previously have noted that neutrophil-mediated damage to R. oryzae hyphae was in-hibited by polyanions (10), and we have nowshown that whole lysates and isolated granulesfrom human neutrophils can damage R. oryzae(but not A. fumigatus) hyphae. This antifungaleffect was neutralized by polyanions. Moreover,cationic proteins purified from granules of hu-man neutrophils (7) and known to be active indestroying bacteria (27) and tumor cells (7) alsodamaged hyphae. Elimination of the chymotryp-sin-like activity of these cationic proteins, ifanything, enhanced rather than impaired antihy-phal effects. In their natural, undenatured state,these proteins primarily damaged R. oryzae hy-

phae, having little effect on A. fumigatus hy-phae. However, specificity of the cationic pro-teins for organisms was not attributable merelyto the effects of charge density of these arginine-rich proteins, since A. fumigatus and R. oryzaehyphae were equally susceptible to damage bythree different polyanions. Thus, differentialsusceptibility of the fungi to neutrophil cationicproteins would appear to depend upon someother aspect of protein structure.Although these as well as our previous (10)

studies are consistent with a role for cationicproteins in damage to R. oryzae hyphae by intactneutrophils, our studies with chronic granuloma-tous disease neutrophils indicated the necessityfor intact oxidative metabolism for the hyphaldamage to occur. However, this need not pre-clude a contributory role for cationic proteins inthe process, as various mechanisms may inter-act within phagocytic vacuoles or at cell sur-faces to contribute to microbicidal activity (3,20, 28, 41). For example, elastase from neutro-phil granules may enhance the bactericidal activ-ity of either the myeloperoxidase system orcationic proteins (28). Cationic proteins mayenhance somewhat the rate of killing of somebacteria by the myeloperoxidase system (28).Like myeloperoxidase, the cationic proteinsused in our studies have been localized to theazurophil granules of neutrophils (29) and somight interact. Concentrations of these particu-lar cationic proteins present in normal neutro-phils may well be insufficient to lyse target cellswithout interaction with other mechanisms, as107 lysed neutrophils yield only 12 ,ug of these

TABLE 9. Interaction of subinhibitory levels ofcationic proteins with the myeloperoxidase (MPO)

system% Damage to hyphae

Components of systemA. fumigatus R. oryzae

MPO + 1- + Cl- + H202 0.0 14.9(0.01 mM)

Unheated cationic protein 0.0 26.0A, lot 2 (10 ,ug/ml)

MPO + I- + Cl + H202 + 24.1 58.9unheated cationic proteinA, lot 2 (10 ,ug/ml)

Unheated cationic protein 0.0 2.2A, lot 2 (1 Fg/ml)

MPO + I- + Cl- + H202 + 0.3 30.2unheated cationic proteinA, lot 2 (1 ,ug/ml)

Glucose oxidase (1.4 mU) + 0.0 0.0glucose (1.0 mM)

Glucose oxidase (1.4 mU) + 0.7 21.8glucose (1.0 mM) +unheated cationic proteinA, lot 2 (1 ,ig/ml)

VOL. 38, 1982


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


cationic proteins (41). Damage to R. oryzaehyphae required 10 to 50,ug of these proteins perml in our studies. However, we noted thatsubinhibitory levels of these cationic proteins (1,ug/ml) damaged hyphae when combined withsubinhibitory levels of hydrogen peroxide, withor without other constituents of the myeloperox-idase system. Moreover, granule-associated cat-ionic proteins other than those which we testedinhibit some bacteria (42) and so may prove tocontribute to antifungal activity. In myeloperox-idase deficiency, where hydrogen peroxide andother oxidants may be produced in increasedamounts (3, 24), cationic proteins also mighthave contributed to the observed damage to R.oryzae hyphae.

Reactions in these cell-free systems are analo-gous to those occurring within and at surfaces ofintact neutrophils. Our results indicate that anti-hyphal mechanisms of leukocytes may not nec-essarily be identical for all types of hyphae.Moreover, interactions of several potential oxi-dative and nonoxidative antihyphal mechanismsmay define the host's ability to limit fungalinfections. For example, whereas aspergillosiswas a common complication of chronic granulo-matous disease (8), mucormycosis was not ob-served despite the ubiquity of R. oryzae in theenvironment. This seeming disparity may be, atleast in part, related to the availability of alterna-tive, nonoxidative mechanisms active against R.oryzae but not A. fumigatus in host neutrophils.Granule-associated cationic proteins acting to-gether with other constituents such as elastase(28) might explain such a process. Where con-centrations of oxidative (hydrogen peroxide) ornonoxidative (cationic proteins) substances arelimiting or suboptimal, interactions of mecha-nisms may be required for antihyphal activity. Ifso, this may have important implications inactivity of defense mechanisms against opportu-nistic mycoses in the intact host.

ACKNOWLEDGMENTS

We thank Henry Rosen for valuable advice, and ElizabethHuber, Nelson Erickson III, Steven Leong, and WilliamOrmerod for technical assistance.

This work was supported by Public Health Service grantA115338 from the National Institute of Allergy and InfectiousDiseases.

LITERATURE CITED

1. Agner, K. 1958. Crystalline myeloperoxidase. Acta Chem.Scand. 12:89-94.

2. Awtrey, A. D., and R. E. Connick. 1951. The absorptionspectra of 12, I3-, I-, I03- 5406 and S2037. Heat of thereaction 13- = 12 + I. J. Am. Chem. Soc. 73:1842-1843.

3. Babior, B. 1978. Oxygen-dependent microbial killing byphagocytes. N. Engl. J. Med. 298:659-668, 721-725.

4. Boyum, A. 1968. Isolation of mononuclear cells by onecentrifugation, and of granulocytes by combining centrifu-gation and sedimentation at lg. Scand. J. Clin. Lab.Invest. Suppl. 97:77-89.

5. Clark, R. A., and S. J. Klebanoff. 1977. Myeloperoxidase-H202-halide system: cytotoxic effects on human bloodleukocytes. Blood 50:65-70.

6. Clark, R. A., and S. J. Klebanoff. 1980. Neutrophil-plate-let interaction mediated by myeloperoxidase and hydro-gen peroxide. J. Immunol. 124:399-405.

7. Clark, R. A., I. Olsson, and S. J. Klebanoff. 1976. Cyto-toxicity for tumor cells of cationic proteins from humanneutrophil granules. J. Cell. Biol. 7:719-723.

8. Cohen, M. S., R.-E. Isturiz, H. L. Malech, R. K. Root,C. M. Wilfert, L. Gutman, and R. Buckley. 1981. Fungalinfection in chronic granulomatous disease: the impor-tance of the phagocyte in defense against fungi. Am. J.Med. 71:59-66.

9. Diamond, R. D., R. A. Clark, and C. C. Haudenschild.1980. Damage to Candida albicans hyphae and pseudohy-phae by the myeloperoxidase system and oxidative prod-ucts of neutrophil metabolism in vitro. J. Clin. Invest.66:908-917.

10. Diamond, R. D., R. Krzesicki, B. Epstein, and W. Jao.1978. Damage to hyphal forms of fungi by human leuko-cytes in vitro: a possible host defense mechanism inaspergillosis and mucormycosis. Am. J. Pathol. 91:313-328.

11. Gadebusch, H. H., and A. G. Johnson. 1966. Natural hostresistance to infection with Cryptococcus neoformans.IV. The effect of some cationic proteins on the experimen-tal disease. J. Infect. Dis. 116:551-565.

12. Harrison, J. E., and J. Schultz. 1976. Studies on thechlorinating activity of myeloperoxidase. J. Biol. Chem.251:1371-1374.

13. Harrison, J. E., B. D. Watson, and J. Schultz. 1978.Myeloperoxidase and singlet oxygen: a reappraisal. FEBSLett. 92:327-332.

14. Hasty, N., P. B. Markel, P. Radlick, and D. R. Kearns.1972. Role of azide in singlet oxygen reactions: reaction ofazide with singlet oxygen. Tetrahedron Lett. 1:49-52.

15. Held, A. M., and J. K. Hurst. 1978. Ambiguity associatedwith use of singlet oxygen trapping agents in myeloperoxi-dase-catalyzed reactions. Biochem. Biophys. Res. Com-mun. 81:878-885.

16. Hodgson, E. K., and I. Fridovich. 1974. The production ofsuperoxide radical during the decomposition of potassiumdichromate. Biochemistry 13:3811-3814.

17. Kimball, H. R., G. H. Ford, and S. M. Wolff. 1975. Lyso-somal enzymes in normal and Chediak-Higashi bloodleukocytes. J. Lab. Clin. Med. 86:616-630.

18. Klebanoff, S. J. 1970. Myeloperoxidase contributions tothe microbicidal activity of intact leukocytes. Science160:1095-1097.

19. Klebanoff, S. J., and R. A. Clark. 1975. Hemolysis andiodination of erythrocyte components by a myeloperoxi-dase-mediated system. Blood 45:699-707.

20. Klebanoff, S. J., and R. A. Clark. 1978. The neutrophil:function and clinical disorders, p. 424-428. North HollandPublishing Co., Amsterdam.

21. Klebanoff, S. J., R. A. Clark, and H. Rosen. 1976. Myelo-peroxidase-mediated cytotoxicity, p. 267-288. In J.Schultz and F. Ahmad (ed.), Cancer enzymology. Aca-demic Press, Inc., New York.

22. Klebanoff, S. J., and H. Rosen. 1978. Ethylene formationby polymorphonuclear leukocytes. Role of myeloperoxi-dase. J. Exp. Med. 148:490-505.

23. Lehrer, R. I. 1969. Antifungal effects of peroxidase sys-tem. J. Bacteriol. 99:361-365.

24. Lehrer, R. I. 1975. The fungicidal mechanisms of humanmonocytes. I. Evidence for myeloperoxidase-linked andmyeloperoxidase-independent candidacidal mechanism.J. Clin. Invest. 55:338-346.

25. Lehrer, R. I., and M. J. Cline. 1969. Leukocyte myelo-peroxidase deficiency and disseminated candidiasis: therole of myeloperoxidase in resistance to Candida infec-tion. J. Clin. Invest. 48:1478-1488.

26. Lehrer, R. I., K. M. Ladra, and R. B. Hake. 1975. Non-oxidative fungicidal mechanisms of mammalian granulo-

INFECT. IMMUN.


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


cytes: demonstration of components with candidacidalactivity in human, rabbit, and guinea pig leukocytes.Infect. Immun. 11:1226-1234.

27. Odeberg, H., and L. Olsson. 1976. Mechanisms for themicrobicidal activity of cationic proteins of human granu-locytes. Infect. Immun. 14:1269-1275.

28. Odeberg, H., and I. Olsson. 1976. Microbicidal mecha-nisms of human granulocytes: synergistic effects of granu-locyte elastase and myeloperoxidase or chymotrypsin-likecationic proteins. Infect. Immun. 14:1276-1283.

29. Ohlsson, K., I. Olsson, and J. K. Spitznagel. 1977. Local-ization of chymotrypsin-like cationic protein, collagenaseand elastase in azurophilic granules of human neutrophilicpolymorphonuclear leukocytes. Hoppe-Seyler's Z. Physi-ol. Chem. 358:361-366.

30. Ouannes, C., and T. Wilson. 1968. Quenching of singletoxygen by tertiary aliphatic amines. Effect of DABCO. J.Am. Chem. Soc. 90:6527-6529.

31. Patterson-Delafield, J., R. J. Martinez, and R. I. Lehrer.1980. Microbicidal cationic proteins in rabbit alveolarmacrophage: a potential host defense mechanism. Infect.Immun. 30:180-182.

32. Peterson, E. M., and R. A. Calderone. 1978. Inhibition ofspecific amino acid uptake in Candida albicans by lyso-somal extracts from rabbit alveolar macrophages. Infect.Immun. 21:506-513.

33. Repine, J. E., J. W. Eaton, M. W. Anders, J. R. Hoidal,and R. B. Fox. 1979. Generation of hydroxyl radical byenzymes, chemicals, and human phagocytes in vitro.Detection with the anti-inflammatory agent, dimethyl sulf-oxide. J. Clin. Invest. 64:1642-1651.

34. Rosen, H., and S. J. Klebanoff. 1977. Formation of singlet

oxygen by the myeloperoxidase-mediated antimicrobialsystem. J. Biol. Chem. 252:4803-4810.

35. Rosen, H., and S. J. Klebanoff. 1979. Hydroxyl radicalgeneration by polymorphonuclear leukocytes measuredby electron spin resonance. J. Clin. Invest. 64:1725-1729.

36. Rosen, H., and S. J. Klebanoff. 1979. Bactericidal activityof a superoxide anion-generating system. A model for thepolymorphonuclear leukocyte. J. Exp. Med. 149:27-39.

37. Singh, H., and J. A. Valdez. 1978. Singlet oxygen a majorreactive species in the furocoumarin photosensitized inac-tivation of E. coli ribosomes. Photochem. Photobiol.28:539-545.

38. Slivka, A., A. F. LoBuglio, and S. J. Weiss. 1980. Apotential role for hypochlorous acid in granulocyte-medi-ated tumor cell cytotoxicity. Blood 55:347-350.

39. Tauber, A. I., and B. B. Babior. 1977. Evidence forhydroxyl radical production by human neutrophils. J.Clin. Invest. 60:374-379.

40. Van Zwieten, R., R. Wever, M. N. Hamers, R. S. Ween-ing, and D. Roos. 1981. Extracellular proton release bystimulated neutrophils. J. Clin. Invest. 68:310-313.

41. Venge, P., A. Stromberg, J. H. Braconler, L.-E. Roxin,and I. Olsson. 1978. Neutrophil and eosinophil granulo-cytes in bacterial infection: sequential studies of cellularand serum levels of granule proteins. Br. J. Hematol.38:475-483.

42. Weiss, J., S. Beckerdite-Quaglinta, and P. Elsbach. 1980.Resistance of gram-negative bacteria to purified bacteri-cidal leukocyte proteins. Relation to binding and bacteriallipopolysaccharide structure. J. Clin. Invest. 65:619-628.

43. Weiss, S. J., P. K. Rustagi, and A. F. LoBuglio. 1978.Human granulocyte generation of hydroxyl radical. J.Exp. Med. 147:316-323.

VOL. 38, 1982


http://iai.asm.org/

Dow

nloaded from
http://iai.asm.org/

damage to aspergillus fumigatus and rhizopus hyphaeby ...iai.asm.org/content/38/2/487.full.pdf ·...

Documents