INFECTION AND IMMUNITY, Apr. 1983, p. 97-1020019-9567/83/040097-06$02.00/0Copyright C 1983, American Society for Microbiology

Vol. 40, No. 1

Generation of Hydrogen Peroxide by Candida albicans andInfluence on Murine Polymorphonuclear Leukocyte Activity

DAVID L. DANLEY,1* ANTHONY E. HILGER,2 AND CRAIG A. WINKEL'Department of Clinical Investigation, Letterman Army Medical Center, San Francisco, California 94129,1and Division ofMedical Technology, Medical School, The University of North Carolina, Chapel Hill, North

Carolina 275142

Received 15 October 1982/Accepted 18 January 1983

Iodide fixation by murine polymorphonuclear leukocytes (PMN) incubated withviable Candida albicans blastoconidia increases directly with yeast cell concen-tration up to about 3 x 106 cells per ml, but above this concentration boundactivity declines dramatically. To understand the basis for this decline, weexamined the oxidative metabolism offungi and stimulated PMN and found someremarkable similarities between these cell types. Both produced 14C02 whenincubated with [1-14C]glucose, both reduced cytochrome c, and both fixedradiolabeled iodide, although the fungi required exogenous lactoperoxidase. Indose-response experiments, iodination by fungi with lactoperoxidase was identi-cal to that with PMN, i.e., the maximum bound activity occurred in cultures with106 to 3 x 106 blastoconidia per ml. lodination by fungi with lactoperoxidase wasreduced when blastoconidia were incubated at 25°C or in the presence of catalaseand the metabolic inhibitors rotenone, antimycin A, and 2-deoxyglucose. Resultsfrom assays for oxidation of scopoletin and o-dianisidine showed that 106blastoconidia in 1.0 ml of medium released 0.5 to 0.7 nmol of I 202 after 1 h, but 3x 106 and 107 cells released significantly less H202. These results suggest thatiodide fixation by PMN and low numbers of fungal cells may reflect a cooperativeeffort, with fungi generating some H202 that reacts with the myeloperoxidasereleased from the PMN. With high concentrations of blastoconidia, H202 activityappeared to be specifically inhibited, possibly to protect fungal cells from damage.

Candida albicans and some related speciesare part of the normal flora of the upper respira-tory, alimentary, and urogenital tracts of mosthumans. However, in a small percentage of thepopulation, Candida spp. may colonize epi- orendothelial tissues and produce primary infec-tions, such as thrush or vaginitis, and in immu-nosuppressed patients these fungi may behavelike pathogens and become invasive. Althoughconsiderable research has been conducted onthe physiology of Candida spp. and their mecha-nisms of pathogenesis, no single factor has beenidentified that consistently correlates with inva-sion by this fungus.

In previous studies on the interaction of mu-rine polymorphonuclear leukocytes (PMN) withC. albicans cells in vitro, we reported that iodidefixation by PMN is considerably different whenstimulation is by viable or killed fungi; i.e.,iodination increases when PMN are incubatedwith increased numbers of killed blastoconidia,but decreases when PMN are incubated withincreased numbers of viable blastoconidia (5).Since phagocytic cells may damage and kill fungithrough iodination (10, 13), these results suggestthat viable blastoconidia may augment their sur-

vival by modulating PMN activity. To test thishypothesis, we examined the oxidative metabo-lism of blastoconidia and stimulated PMN. Inthis communication, we report the similaritiesbetween the oxidative metabolisms of these twocell types. These similarities make it difficult toevaluate the individual aspects of metabolicfunction when PMN and fungi are incubatedtogether; however, in the iodination assay ourdata suggested that both cell types may cooper-ate to fix iodide.

(This paper was presented in part at the 82ndAnnual Meeting of the American Society forMicrobiology, 7 to 12 March 1982, Atlanta, Ga.)

MATERIALS AND METHODSFungal cells. A strain of C. albicans originally isolat-

ed from a patient (9) was grown as blastoconidia inSabouraud dextrose broth at either 25 or 37°C andsubcultured every 24 to 48 h. The yeast cells used inthe following assays were inoculated at 5 x 106 cellsper ml into 50 ml of fresh medium and incubated at37°C for 2 to 3 h unless otherwise noted. The cellswere washed twice in Krebs-Ringer phosphate-buff-ered saline (pH 7.2) with 3.0 mg of glucose per ml(KRPG) and suspended in an appropriate medium forassay.

97

on July 28, 2018 by guesthttp://iai.asm

.org/D

ownloaded from

98 DANLEY, HILGER, AND WINKEL

PMN-enriched spleen cells. Murine PMN were ob-tained from DBA/2 female mice as previously reported(5). Briefly, mice were injected intraperitoneally withcomplete Freund adjuvant H37RA (Difco Labora-tories, Detroit, Mich.), and 3 to 5 days later a spleencell suspension was prepared from these animals.Cells were treated with iron carbonyl to remove mac-rophages and put on a two-step Ficoll-Hypaque gradi-ent. Cells at the first interface from the bottom wererecovered and washed twice in KRPG before beingused. Cell viability, determined by trypan blue dyeexclusion, was greater than 99%, and the PMN con-centration ranged from 35 to 45%, with less than 5%contamination by macrophages.

Assays for oxidative metabolism. The oxidative me-tabolism of PMN and fungi was measured with stan-dard assays. Hexose monophosphate shunt (HMPS)activity in the PMN was determined by measuring14CO2 production by cells incubated with D-[1-'4C]glu-cose (16). Fungal cells or PMN were incubated in 15-by 85-mm culture tubes containing 1.0 ml of KRPGwith 1.0 p.Ci of D-[1-14C]glucose (specific activity, 7.7mCi/mmol; New England Nuclear Corp., Boston,Mass.). Each tube was sealed with a serum stopperholding a center well, and the cultures were incubatedat 37°C for 1 h. Afterwards, 0.2 ml of 2 N H2SO4 wasadded, and 14CO2 was collected over 60 min on filterpaper saturated with 1.0 N NaOH. The contents of thecenter well were mixed with 10 ml of Aquasol (NewEngland Nuclear), the vials were refrigerated over-night to reduce chemiluminescence, and the radioac-tivity was measured in a Packard Tricarb liquid scintil-lation counter.

Superoxide anion (O2-) production was determinedby measuring the reduction of cytochrome c (1) in thepresence and absence of superoxide dismutase (SOD).Fungal cells or PMN were incubated in 15- by 85-mmsiliconized glass tubes containing 0.8 ml of KRPG for15 min at 37°C. After that time, 500 p.g of cytochromec (type III; Sigma Chemical Co., St. Louis, Mo.) in 0.1ml of KRPG and either 0.1 ml of KRPG or 300 U ofSOD (Sigma) in 0.1 ml of medium were added to eachtube. Cultures were incubated for an additional 45 minat 37°C, and the optical density at 550 nm of eachculture supernatant was determined with a BeckmanDU8 spectrophotometer. Concentrations of O2 werecalculated by subtracting the mean absorbance ofculture supernatant with SOD from those values with-out SOD and dividing the difference by the millimolarextinction coefficient for cytochrome c (23 mM cm- ').To detect H202 release, we employed three assays:

iodide fixation (11), scopoletin oxidation (14), and o-

dianisidine oxidation (15). In assays of iodide fixation,cells were incubated in siliconized 15- by 85-mm glasstubes with 1.0 ml of KRPG, 2.0 U of lactoperoxidase(LP) (Sigma), and 1.0 R.Ci of Na'251 (specific activity,13 to 17 mCi/,g; Amersham Corp., Arlington Heights,Ill.). Control tubes lacked cells or LP. The cultureswere incubated at 370C for 1 h, and reactions were

terminated by adding 0.2 ml of 0.01 M sodium thiosul-fate and 3.0 ml of 10% trichloroacetic acid. Precipi-tates were sedimentated by centrifugation and washedthree times with 3 ml of l1o trichloroacetic acid.Radioactivity was measured in a Packard autogammacounter. In assays of scopoletin oxidation, cells wereincubated in 15- by 85-mm culture tubes with 1.0 ml ofKRPG, 2.0 U of LP, and 2.0 ,ug of scopoletin (Sigma).

Control tubes lacked LP. Cultures were incubated at37°C for 1 h and centrifuged, and 0.1 ml of supernatantwas added to 1.9 ml of water with 10' M cyanide.Fluorescence was measured with an FOCI A-4 fluo-rometer with a 7-39 primary filter and a 3-72 secondaryfilter. To determine the amount of H202 produced, wecompared these results with a standard curve obtainedwith predetermined concentrations of H202 in lieu ofviable cells. In assays of o-dianisidine oxidation, cellswere incubated in 15- by 85-mm culture tubes with 1.0ml of KRPG for 1 h at 37°C. The tubes were centri-fuged, and 0.8 ml of supernatant was recovered andmixed with 1.0 U of LP and 100 jig of o-diansidine in0.2 ml of KRPG. The optical density at 460 nm wasmeasured spectrophotometrically, and the amount ofH202 in each culture was determined from a standardcurve.

RESULTS

Measurement of HMPS activity. During anoxidative metabolic burst, PMN metabolize glu-cose through the HMPS, which can be assessedby measuring their production by 14CO2 from [1-14C]glucose. For instance, in our assays PMNreleased (mean ± standard error of the mean)1,109 ± 122 cpm when incubated alone, but inthe presence of 100 ,ug of zymosan they released9,670 ± 953 cpm. The 14CO2 released by PMNincubated with viable blastoconidia increasedfrom 23,198 ± 293 to 104,097 ± 7,607 cpm as thenumber of fungi increased from 105 to 107 cells(Fig. 1). However, when comparable numbers ofviable blastoconidia were incubated alone,14CO2 release increased from 31,453 ± 1,288 to156,908 ± 14,477 cpm. Because blastoconidiaalone released 2 to 10 times more 14CO2 than didPMN activated with zymosan, we could notdetermine the contribution made by each celltype when they were incubated together. More-over, 14co2 release was lower in cultures con-taining both PMN and fungi than in culturescontaining fungi alone.02 release by PMN and blastoconidia. As an

alternate means of detecting PMN activation byblastoconidia, we measured 02 release byquantifying the reduction of cytochrome c in thepresence and absence of SOD. Spleen cellsenriched for PMN released 0.2 nmol of O2-when incubated alone and 4.7 ± 0.4 nmol of O2-when activated with 100 ,ug of zymosan; thesevalues were in close agreement with earlierobservations (5). PMN stimulated with 3 x 105viable blastoconidia generated 1.7 nmol of O2-,but only 0.9 nmol when stimulated with 107blastoconidia. However, blastoconidia alonehad significant activity in the assay, since 3 x105 yeast cells released 1.1 nmol and 3 x 10' and107 cells released about 1.7 nmol of O2- (Fig. 2).Once again, the oxidative metabolism of thefungus obscured determinations of PMN metab-olism; in fact, these data suggest that little

INFECT. IMMUN.

on July 28, 2018 by guesthttp://iai.asm

.org/D

ownloaded from

HYDROGEN PEROXIDE RELEASE BY C. ALBICANS 99

1831-

140

o 120

0~x2100

LU0 400-

.0

3 x 10' 1 0. 3 x 10oNUMBER OF BLASTOSPORES

FIG. 1. Release of "CO2 by increasing numbeiC. albicans blastoconidia (0) or blastoconidia wix 106 PMN-enriched spleen cells (0) incubated1.0 ILCi of [1-_4C]glucose for 1 h at 37°C. Each prepresents the geometric mean of six values fduplicate experiments. Bars represent the stanerror of the mean.

difference exists in the amounts of O2- geneed whether blastoconidia are incubated alonwith PMN.

Iodination by viable blastoconidia. Thethat viable blastoconidia released O2- gavcreason to hypothesize that they would alsolease H202 and fix iodide in the presencesuitable peroxidase. 1251 fixation by PMN itbated with various numbers offungi and by fiincubated with LP were compared (Fig.lodination by PMN (mean ± standard errothe mean) increased from 14,329 ± 1,805with 3 x 105 blastoconidia to 46,728 ± 6,352with 3 x 106 yeast cells, but declined dramat'ly to 8,408 ± 1,491 cpm with 107 fungi. Ycells incubated with LP had a similar patteriodination, i.e., 3 x 10 cells fixed 43,027,485 cpm, and 106 and 3 x 106 cells fixed al60,000 cpm, but 107 cells fixed only 20,663,495 cpm.To determine whether iodination in the F

ence of LP was the result of H202 releas4blastoconidia, we measured the cell-boundtivity when yeast cells were incubated uivarious conditions (Table 1). Iodination didoccur in cultures incubated at 25°C or in tilacking glucose or LP. The addition of cataat 3,000 or 10,000 U/ml inhibited iodinationdose-dependent fashion, and the metabolikhibitors antimycin A and rotenone also inhitiodination. Antimycin A and rotenone, ho.er, did not inhibit iodination of blastocoi

A when exogenous H202 was added to the cul-tures. We also examined the effects of hexose oniodination, since we had previously reportedthat mannose inhibits iodination by PMN incu-bated with viable blastoconidia (5). Adding 3.0mg of glucose or mannose to KRPG decreasediodination by blastoconidia 30 to 40%, whereas2-deoxyglucose at a similar concentration re-duced iodination by 78%.

In the studies described above, blastoconidiawere recovered from cultures grown at 25°C andincubated at 37°C for 2 h. To determine whetherH202 release occurred with blastoconidia at anyage, we measured iodide fixation by cells grownin Sabouraud broth for various times at 37°C.Because fungi transferred to 37°C tended toproduce filaments, iodination was quantified as

~ counts per minute per microgram of fungal cell1O" protein, and serial threefold dilutions of cells

were assayed to compensate for the inhibitionrs of we observed in assay tubes with more than 3 xith 2 106 cells per ml. Blastoconidia removed fromwith overnight cultures at 25°C and put in KRPG at3oint 37°C failed to show iodination. However, after adard 2-h incubation at 37°C in fresh medium, cells at

low concentrations (less than 1 ,ug of protein perml) fixed 55,000 to 60,000 cpm/l,g of protein, andat higher concentrations (greater than 1 F.g of

Xrat- protein per ml) iodination declined in a dose-e or dependent fashion. A similar pattern of iodin-

ation by cells incubated in fresh medium for 4 hfact was observed. However, peak iodination wase us only 20,000 cpm/,Lg, and after a 6-h incubation at) re- 37°C the cells failed to fix iodide regardless ofof a the cell concentration (Fig. 4).ncu- Because we had transferred the cells afteruingi overnight culture at 25°C to fresh medium at3). 37°C, we hypothesized that metabolic activity in

r ofcpmcpmical-east-n of1 +bouti2 +

)res-e byac-

nderI nothosedaseiin ac in-bitedwev-.idia

-2.01

C

a

c1.0

LU00.5

0Uo-

3x 10 log 3x10NUMBER OF BLASTOSPORES

10s

FIG. 2. Reduction of cytochrome c by increasingnumbers of C. albicans blastoconidia (0) or blastocon-idia with 2 x 106 PMN-enriched spleen cells (-)incubated for 1 h at 37°C. Each point represents thegeometric mean of nine values from triplicate experi-ments. Bars represent the standard error of the mean.

VOL. 40, 1983

on July 28, 2018 by guesthttp://iai.asm

.org/D

ownloaded from

100 DANLEY, HILGER, AND WINKEL

70

6O

60

50x

C.40-

z

030-

z0-0

10-

0

3x 10' 106 3 x 106 10' 3 x 10lNUMBER OF BLASTOSPORES

FIG. 3. Iodide fixation by increasing numbers of C.albicans blastoconidia incubated alone (a), with 2 x106 PMN-enriched spleen cells (A), or with 2.0 U ofLP (0). Each point represents the geometric mean of12 values from quadruplicate experiments. Bars repre-sent the standard error of the mean.

the fungi might have increased dramatically,accounting for the transient release of H202.Therefore, we cultured blastoconidia at 37°C,transferred them to fresh medium at 37°C, andmeasured iodination at various times after thetransfer. The pattern of iodination observed inthese blastoconidia was identical to that ob-served in cells from cultures grown at 25°C, i.e.,they failed to fix iodide; after 2 h, cultures with 1,ug of fungal cell protein fixed 50,000 to 60,000cpm/,ug, and older cells fixed progressively lessiodide (data not shown). These findings indicatethat H202 release did not occur throughout thecell cycle but appeared to be limited to early log-phase growth.Examination of H202 release by other assays.

Although iodide fixation is related directly toH202 concentration, this assay is not suitable forquantitating H202 release because the number ofsites for iodide fixation is not constant. Toeliminate this variable and to quantitate H202release, we used two assays that measure H202concentration through oxidation of the exoge-nous substrates scopoletin and o-dianisidine.Data from the scopoletin assay showed that 3 x105 and 106 blastoconidia released 0.5 nmol ofH202 after 1 h at 37°C. However, we were notable to use this assay to measure H202 incultures with more than 106 blastoconidia sincethere was a significant reduction in fluorescencein cultures lacking LP (9). When o-dianisidinewas used in lieu of scopoletin, we found that itwas not oxidized by supernatant from tubescontaining 3 x 105 to 10' blastoconidia unlessLP was present. The greatest amount of oxida-tion (optical density at 460 nm, 0.0069 ± 0.0003)was observed with supernatant from tubes con-

taining 106 blastoconidia, and this activity wasreduced 90% when fungi were incubated with100 U of catalase. Using a standard curve, wedetermined the amount of H202 in cultures withincreased numbers of blastoconidia. Superna-tants from cultures with 3 x 105 and 106fungalcells contained (mean ± standard error of themean) 0.51 ± 0.03 and 0.67 ± 0.03 nmol ofH202, respectively (Fig. 5). However, the opti-cal density at 460 nm was sipnificantly lower forthe media in which 3 x 10 and 107 cells wereincubated, and we calculated that only 0.29 ±0.2 and 0.13 ± 0.03 nmol of H202, respectively,was present. Results from this assay and thescopoletin assay agreed that 3 x 105 and 106blastoconidia produced 0.5 to 0.7 nmol of H202after 1 h at 37°C. Moreover, in cultures withgreater numbers of yeast cells, results from theo-dianisidine assay showed that less H202 isavailable for reaction.

DISCUSSIONIn studying the interaction between phagocyt-

ic cells and C. albicans, we and others (5, 6)have reported changes in the oxidative metabo-lism of PMN that are consistent with activationof the oxidative metabolic burst. However, inthese studies there was little or no considerationgiven to the contribution made by the fungi. Theresults of studies reported in this communicationindicate that C. albicans cell metabolism issignificant and definitely contributes to the re-sults observed in a number of assays.

Initially, we measured HMPS activity in PMNand showed that C. albicans cells generated'4CO2 from [1-_4C]glucose as do PMN activatedwith zymosan. On a per-cell basis, blastoconidiagenerated more activity than PMN, since 3 x105 blastoconidia released (mean ± standarderror of the mean) 31,453 ± 1,288 cpm from (1-

TABLE 1. Effect of various treatments oniodination by C. albicans blastoconidia in vitro

% Reduction in iodinationTreatment from controla

(mean ± SEM)b

Cells incubated (25°C) 87 ± 7LP deleted 98 ± 0Glucose deleted 86 ± 1Catalase added (3,000 U) 59 + 3Catalase added (10,000 U) 89 ± 1Antimycin A added (18 ,uM) 79 + 3Rotenone added (10 ,uM) 83 + 22-Deoxyglucose added (3 mg) 78 + 5Glucose added (3 mg) 43 + 6

a In treated controls, 106 blastoconidia were incu-bated in 1.0 ml of KRPG with 2.0 U of LP and 1.0 uCiof Na"2'I for 1 h at 37°C.

b Data represent at least six values from duplicateexperiments.

INFECT. IMMUN.

on July 28, 2018 by guesthttp://iai.asm

.org/D

ownloaded from

HYDROGEN PEROXIDE RELEASE BY C. ALBICANS 101

0.1 1.0 10FUNGAL PROTEIN (mg)

100

FIG. 4. Iodide fixation by increasing numbers of C.albicans cells transferred from overnight cultures at25°C and incubated in Sabouraud broth at 37°C for 0(0), 2 (0), 4 (A), or 6 h (A) before being used in theiodination assay. Each point represents the geometricmean of six values from duplicate experiments.

14C]glucose, whereas 2 x 106 activated spleencells containing about 106 PMN released only9,670 ± 953 cpm. When PMN were incubatedwith various numbers of yeast cells, the 14CO2release was about 30% less than in cultures withfungi alone, suggesting that PMN may inhibitblastospore metabolism. However, the extent ofthis inhibition could not be determined from ourtest results.Chattaway et al. (3) reported significant

HMPS activity in blastoconidia grown at 30°C,and noted that the HMPS is a major source ofNADPH that could be used by protein disulfidereductase to promote budding. Our data suggestthat this metabolic pathway may be very activein blastoconidia grown at 37°C, and the reducingpower generated by HMPS activity also may bespent in generating 2-. The amount of O2released by yeast cells alone was comparable tothat released by PMN plus blastoconidia, so thisassay was not suitable for measuring PMN acti-vation by viable fungi. However, these resultssuggest that H202 was also being generatedextracellularly by fungi, since O2 spontaneous-ly dismutates to H202. Using an iodination assayto measure H202 release, we found that blasto-conidia incubated with Na125I and LP fixediodide in a manner similar to PMN incubatedwith viable blastoconidia, i.e., bound radioactiv-ity was highest in cultures with 106 to 3 x 106yeast cells and declined dramatically in cultureswith 107 cells.That iodination did not occur in the absence of

LP and was significantly reduced by catalaseindicate that iodide fixation was mediated by theH202 released by the blastoconidia. Generationof H202 within the mitochondria of baker'syeast (Saccharomyces cerevisiae) has been re-

ported by Boveris (2). He postulated that O2-generated from ubisemiquinone is dismutated toH202 by mitochondrial SOD in the intermem-brane space. He noted, however, that H202 isprobably neutralized by cytochrome c peroxi-dase, since H202 diffusion from the mitochon-dria to the yeast cytosol could result in cellinjury or death. Our results showed that H202 orO2 was generated in the mitochondria of theblastoconidia, since iodide fixation by blasto-conidia was inhibited by antimycin A and rote-none, compounds that block the mitochondrialrespiratory chain. However, it is not clear whythese reactive oxygen species did not cause cellinjury. We have observed that the mitochondriain yeast cells are localized next to the cellmembrane (unpublished results), which wouldreduce the intracellular distance through whichthese potentially cytotoxic oxygen species mi-grate.Eremina and Lozinov (7) used an o-dianisi-

dine assay to measure H202 in filtrates fromcultures of C. albicans. They reported that theblastoconidia released H202 throughout log-phase growth and that resting cells inoculatedinto Tris-phosphate buffer produced a maximumconcentration of H202 of 20 nmol/ml after 30min. These results differed from our findingsthat H202 release by blastoconidia was greatlyreduced in the absence of glucose and that 106cells released 0.5 to 0.7 nmol of H202 after 1.0 hat 37°C. Of interest, however, is our observationthat H202 was released by cells during early log-phase growth, a time during which many blasto-conidia are forming germ tubes. Evans et al. (8)

0.7r

0.6j

0.5-

00.4E0.3

I

0.1 -

0

3 x 105 1r 3 x 106 107NUMBER OF BLASTOSPORES

FIG. 5. Measurement of H202 in supernatantsfrom cultures containing various numbers of viableblastoconidia as determined by the oxidation of o-dianisidine in the presence (0) or absence (0) of 1.0 Uof LP. Each point represents the geometric mean ofsix values from duplicate experiments. Bars representthe standard error of the mean.

VOL. 40, 1983

on July 28, 2018 by guesthttp://iai.asm

.org/D

ownloaded from

102 DANLEY, HILGER, AND WINKEL

reported that blastoconidia transferred to freshSabouraud medium at 37°C initiated filamenta-tion and that filamentation peaked within 1.5 to2.5 h. After this time, reproduction occurred bylateral budding, and the blastoconidia formeddid not produce germ tubes. There appears to bea correlation between filamentation and maxi-mal H202 release per microgram of cell protein;however, we do not know whether H202 releaseoccurs in all cells or exclusively in cells forminggerm tubes.The process of filamentation has been associ-

ated with the pathogenicity of C. albicans sincethe physical extrusion of a germ tube permits theextension of the fungus into tissue and hindersits containment by phagocytic cells. The releaseof reactive oxygen species by C. albicans cellsalso could serve to augment invasion, since O2and H202 are toxic to mammalian cells as well asto microorganisms (12). However, the release ofH202 by fungi also could enhance their killing byphagocytic cells. We observed that PMN incu-bated with 106 viable blastoconidia fixed fourtimes more radioiodide than did PMN incubatedwith comparable numbers of killed furngi (5).This difference could indicate that PMN arestimulated more by viable organisms than bykilled organisms (6). However, we postulate thatviable blastoconidia release H202, which reactswith PMN myeloperoxidase to fix halide. Wedid not determine whether iodination by blasto-conidia affects cell function, but it may be signif-icant that iodination is suppressed when thefungal cell concentration is greater than 3 x 106blastoconidia per ml, whereas glucose metabo-lism and O2 release by the fungi are not. Wesuggest that blastoconidia may tolerate low con-centrations of H202 but will inhibit levels thatwould be detrimental to cell function. If and howthese functions are related to invasion by C.albicans remains to be determined; but we be-lieve that reexamination of the oxidative metab-olism in both fungi and phagocytes may lead to abetter understanding of Candida pathogenicity.

ACKNOWLEDGMENTSThis investigation was supported in part by a research grant

from the National Foundation for Infectious Disease.

We thank Linda Ickes for technical assistance and MarinaVillaflor for editorial assistance.

LITERATURE CITED

1. Babior, B. M., R. S. Kipnes, and J. T. Curnutte. 1973.Biological defense mechanisms: the production by leuko-cytes of superoxide, a potential bactericidal agent. J. Clin.Invest. 52:741-744.

2. Boveris, A. 1978. Production of superoxide anion andhydrogen peroxide in yeast mitochondria, p. 65-79. In M.Bacila, B. L. Horecker, and A. 0. M. Stoppani (ed.),Biochemistry and genetics of yeast. Academic Press-, Inc.,New York.

3. Chattaway, F. W., R. Bishop, M. R. Holmes, and F. C.Odds. 1973. Enzyme activities associated with carbohy-drate synthesis and breakdown in the yeast and mycelialforms of Candida albicans. J. Gen. Microbiol. 75:97-109.

4. Crabtree, H. G. 1929. Observations on carbohydrate me-tabolism of tumors. Biochem. J. 23:536-545.

5. Danley, D. L., and A. E. Hilger. 1981. Stimulation ofoxidative metabolism in murine polymorphonuclear leu-kocytes by unopsonized fungal cells: evidence for amannose-specific mechanism. J. Immunol. 127:551-556.

6. Diamond, R. D., R. Krzesicki, and W. Jao. 1978. Damageto pseudohyphal forms of Candida albicans by neutro-phils in the absence of serum in vitro. J. Clin. Invest.61:349-359.

7. Eremina, S. S., and A. B. Lozinov. 1981. Release ofhydrogen peroxide into medium by growing and restingyeast cells. Mikrobiologiya 50:414-417.

8. Evans, E., F. Odds, M. Richardson, and K. Hoiland. 1974.The effect of growth medium on filament production inCandida albic ans. Sabouraudia 12:112-119.

9. Hilger, A. E., and D. L. Danley. 1980. Alteration of poly-morphonuclear leukocyte activity by viable Candida albi-cans. Infect. Immun. 27:714-720.

10. Howard, D. H. 1977. Fungicidal systems derived fromphagocytic cells, p. 198-206. In K. Iwata (ed.), Recentadvances in medical and veterinary mycology. UniversityPark Press, Baltimore.

11. Klebanoff, S. J., and R. A. Clark. 1977. lodination byhuman polymorphonuclear leukocytes: a re-evaluation. J.Lab. Clin. Med. 89:675-687.

12. Klebanoff, S. J., and R. A. Clark. 1978. The neutrophil:function and clinical disorders, p. 414-416. North-Hol-land Publishing Co., New York.

13. Lehrer, R. I. 1969. Antifungal effects of peroxidiase sys-tems. J. Bacteriol. 99:361-365.

14. Root, R. K., J. Metcalf, N. Oshino, and B. Chance. 1975.H202 release from human granulocytes during phagocyto-sis. J. Clin. Invest. 55:945-955.

15. Steinman, R. M., and Z. A. Cohn. 1972. The interactionof soluble horseradish peroxidase with mouse peritonealmacrophages in vitro. J. Cell. Biol. 55:186-204.

16. Strossel, T. P., and Z. A. Cohn. 1976. Phagocytosis, p.289. In C. A. Williams and M. W. Chase (ed.), Methodsin immunology and immunochemistry, vol. 5. AcademicPress, Inc., New York.

INFECT. IMMUN.

on July 28, 2018 by guesthttp://iai.asm

.org/D

ownloaded from