INVASIVE ASPERGILLOSIS (IA) is caused predom-inately by Aspergillus fumigatus, a common

mold distributed freely in the environment[1–3]. Invasive aspergillosis is a significantlysecondary infection detected in patients withcancer, immunosuppressive disease, solid or-gan and bone marrow transplants, and ac-quired immunodeficiency syndrome (AIDS)[4–9]. The infection results in high mortality.The roles played by various host defense fac-tors in suppressing or promoting this invasive

infection are not understood completely. Clin-ical studies have emphasized repeatedly therole of neutropenia and corticosteroids in po-tentiating this disease [6–9].

Generally, impairment of both macrophageand neutrophil functions is a prerequisite to in-vasive infection with Aspergillus [10]. In the faceof infection, pulmonary alveolar macrophagesare rendered less efficient in handling As-pergillus spores, resulting in germination of theimplanted spores [11–15]. Impairment of neu-

SURGICAL INFECTIONSVolume 6, Number 4, 2005© Mary Ann Liebert, Inc.

A Murine Model of Invasive Aspergillosis: VariableBenefit of Interferon-Gamma Administration

under In Vitro and In Vivo Conditions

CHRISTOPHER P. JOHNSON,1 CHARLES E. EDMISTON, JR.,2 YONG-RAN ZHU,1MARK B. ADAMS,1 ALLAN M. ROZA,1 and VISWANATH KURUP3

ABSTRACT

Background: Interferon-� modulates host defense in a number of infectious diseases. Previ-ous studies have shown that systemic administration of interferon-gamma (IFN-�) can en-hance survival in experimental invasive aspergillosis (IA).

Methods: Using a novel model of murine IA that is characterized by primary pulmonaryinfection, we investigated the role of IFN-� in the phagocytosis and killing of Aspergillus fu-migatus by murine neutrophils and pulmonary alveolar macrophages in vitro and the impactof systemic and regional administration of IFN-� on the course of IA in glucocorticoid-treatedmice.

Results: In vitro, IFN-� significantly enhanced phagocytosis and killing function of bothneutrophils and alveolar macrophages from normal animals, but not cortisone-treated ani-mals. In vivo, intravenous administration of IFN-� did not improve phagocyte recruitment,in vivo killing, or mortality from IA. Regional (intranasal) administration of IFN-� to thelungs enhanced recruitment of phagocytic cells to the lungs and improved in vivo killing, butdid not alter (and actually worsened) mortality from IA.

Conclusions: The in vitro and in vivo effects of IFN-� in IA are contingent on many vari-ables, including the route of administration and the specific pathogenesis of infection.

Divisions of 1Transplant and 2Vascular Surgery, Department of Surgery, and the 3Department of Medicine, Med-ical College of Wisconsin and Zablocki Veterans Medical Center, Milwaukee, Wisconsin.

397

398 JOHNSON ET AL.

trophil recruitment and phagocytic functionpermits tissue invasion and eventful dissemi-nation to distant organs [16].

Cytokines that can augment macrophage orneutrophil function may have a role in treatmentor prevention of IA. Interferon-gamma (IFN-�),a cytokine secreted primarily by T lymphocytes,increases oxidative metabolism and the an-timicrobial killing function of macrophages[17–19]. Interferon-gamma enhances additionalcytokine release from lymphocytes andmacrophages, upregulates the expression ofcell adhesion molecules, and increases the effi-ciency of antigen presentation by macrophages[20]. A number of studies have demonstratedincreased killing (in vitro) of Candida species,Toxoplasma gondii, Leishmania, and mycobacte-ria by IFN-�-treated macrophages [21]. Oxida-tive metabolism and phagocytic function ofneutrophils is also increased by IFN-� [22–24].

The role of IFN-� as an immunomodulatorycytokine in IA is yet unclear. We have shownpreviously that depletion of IFN-� and tumornecrosis factor (TNF)–� resulted in enhancedmortality in a murine model of IA [25]. This wasfurther supported by the protection renderedwith exogenous administration of IFN-� andTNF-� in mice exposed to A. fumigatus [25]. Pre-vious studies in our laboratory have alsoshown that administration of interleukin (IL)–12 results in reduced mortality in mice exposedto A. fumigatus spores. This enhanced survivalwas also shown to be associated with increasedproduction of IFN-� [26].

The above-cited studies were conducted us-ing a murine model of IA that is established byintravenous injection of A. fumigatus spores incortisone-treated mice. Pathogenesis of clinicalinfection with A. fumigatus requires inhalationof spores and establishment of pulmonary in-fection prior to distant dissemination. In thecurrent study, we have investigated the poten-tial utility of IFN-� as an immunomodulator of experimental pulmonary aspergillosis. Theresults demonstrate that, whereas IFN-� en-hances the phagocytic function of normal pul-monary alveolar macrophages and peripheralblood neutrophils in vitro, it does not improvephagocytic function or outcome (as measuredby mortality) when administered to cortisone-treated mice rendered susceptible to pul-monary infection by A. fumigatus.

METHODS

Animals

Specific-pathogen-free, 10–12-week-old fe-male BALB/c mice (Sasco Inc., Omaha, NE)were used in this study. Four mice were keptin each cage. Cages, material, water, and foodwere autoclaved before use, and cages werecovered with sterile filter caps. The experi-mental protocol was reviewed and approvedby the animal research review committee at theZablocki VA Medical Center.

Generation of Aspergillus spores anddetermination of spore viability

One strain of A. fumigatus (Af-102, ATCC-42202 American Type Culture Collection,Rockville, MD) was used in the study. The or-ganism was grown on Sabouraud’s dextroseagar (SDA) tubes (Difco Laboratories, Detroit,MI) for 72 h at 37°C. Conidia from uniformlysporulating cultures were obtained by washingin sterile physiological saline and separatedfrom larger clumps by allowing the suspensionto stand for 10 min. A suspension of mostly sin-gle conidia was thus obtained, which was freeof hyphae and conidial heads. The spores werewashed three times with sterile phosphate-buffered saline. Total spores were counted in ahemacytometer, and viability was determinedby the dilution method [27]. The preparationsshowing �95% viability were selected for thisstudy and did not change significantly afterstorage at �70°C for four weeks. Fresh sporeswere prepared every four weeks.

Immunomodulating agents

For in vitro experiments, a soluble glucocorti-coid corticosteroid, methylprednisone sodiumsuccinate (Upjohn Co., Kalamazoo, MI) wasused. For in vivo experiments, cortisone ace-tate (Sigma Chemical Co., St. Louis, MO) was in-jected intraperitoneally (i.p., 125 mg/kg in 0.5mL of 0.15 M NaCl) to facilitate induction of invasive infection. Murine recombinant IFN-�(MuIFN-�) (Biosource-International, Camarillo,CA) was used for in vitro and in vivo procedures.

Intranasal inoculation of A. fumigatus spores

Mice were anesthetized lightly with methoxy-flurane (Pitman-Moore, Inc., Mundelein, IL),

IFN-� AND ASPERGILLOSIS 399

and spores of A. fumigatus were administeredas described previously [28]. A total of 5 �106 conidia in a volume of 50 �L of 0.15 Msaline per mouse were instilled into the nostrilsusing a pipette tip. Animals were held in anupright position for 1–2 min to resume normalbreathing.

Quantitative lung cultures

Selected animals were euthanized immedi-ately and at 4, 24, and 96 h following inocula-tion of spores. Lungs were removed asepticallyand homogenized separately in sterile physio-logical saline by maceration on a wire mesh us-ing the sterile piston of a 10-mL plastic syringe[29]. Spore viability in the homogenate was de-termined quantitatively by colony count fromdilution plate assay in Sabouraud-dextrosesugar (SDA) after 48–96 h of incubation at 37°C.

In vitro measurements of pulmonary alveolarmacrophage function

Pulmonary alveolar macrophages were har-vested by bronchoalveolar lavage, which wasperformed by cannulation of the trachea with an18-gauge catheter (B-D Vascular Access, Becton-Dickinson, Sandy, UT) [30]. Each mouse under-went five 1.0-mL lavages with RPMI 1640medium (Cell Culture Facility, Medical Collegeof Wisconsin, Milwaukee, WI) supplementedwith streptomycin 100 mcg/mL and penicillin100 U/mL (Sigma). Cells in the lavage fluidwere collected by centrifugation (800g) for 15min. Lavage cells were washed three times andfinally suspended in RPMI 1640 medium sup-plemented with antibiotics as described aboveand 1% fetal bovine serum (Biocell Laboratories,Rancho Dominquez, CA). The lavage fluid con-tained �95% viable cells as measured by trypanblue staining. Cells consisted of 90–98% macro-phages by Wright-Giemsa stain.

Lavage cells were incubated for 30 min with250 U rMuIFN-� before addition of 2.5 � 106 A.fumigatus spores and 0.5 � 106 macrophages(5:1 ratio of spores/macrophages) and incu-bated at 37°C for 10 min, and 4 and 8 h. Themacrophages were then lysed by adding 1 mLof distilled water to a 0.1-mL mixture ofmacrophages and spores. The Aspergillus sporeviability was determined by colony count from

dilution plate assay in SDA after 48–72 h of in-cubation at 37°C.

In a separate series of experiments, thelavage cells were obtained as described above,and 0.5 � 106 macrophages were mixed with2.5 � 106 spores. Cultures were set up in com-plete RPMI 1640 medium in 1-mL aliquots us-ing sterile tubes incubated at 37°C in a 5% CO2and 95% air humidified incubator. One hun-dred microliters was removed after incubationat 0.2, 4, and 8 h. Cytospin preparations weremade and fixed on glass slides, and stained by Wright-Giemsa stain. Two hundred mac-rophages were counted microscopically andevaluated for spore content. Phagocytosis wascalculated as percentage of macrophages in-gesting A. fumigatus spores. The phagocytic index was calculated as follows:

Total number ofspores internalized

Phagocytic index �Total numbers of

macrophagescontaining spores (1)

In vitro measurements of peripheral bloodneutrophil function

Mice were euthanized by CO2 asphyxia;blood was obtained by cardiac puncture andcollected in heparinized tubes. Erythrocyteswere removed by incubation of 1.0 mL of thediluted blood with 1 mL of 3% Dextran for 20min at room temperature. Neutrophils wereisolated by layering 0.6 mL of the supernatantover 2 mL of Ficoll-Isopaque (Sigma) in sterileplastic 15-mL centrifugation tubes (Fisher Sci-entific International Inc., Hampton, NH), fol-lowed by centrifugation for 20 min at 1000 rpm.Differential counting revealed a neutrophil pu-rity of �90%. Neutrophils were incubated withspores and rMuIFN-� as described above.

In vivo interaction of A. fumigatus spores withpulmonary phagocytic cells

To assess in vivo spore killing by pulmonaryphagocytic cells, mice challenged with 5 � 106

spores intranasally were sacrificed at 10 min,and 4 and 8 h. The viable spores in broncho-alveolar lavage (BAL) fluid were determinedby colony count as described above. Lavagecells were counted by a hemacytometer. Per-

400 JOHNSON ET AL.

cent neutrophils were determined by cytocen-trifugation (Cytospin 2; Shadon Inc., Pitts-burgh, PA). Cells were fixed in methanol andstained by Wright-Giemsa stain, and countedvisually under a microscope. A minimum of200 cells was counted.

Administration of recombinant murine interferon-gamma (rMuIFN-�) in vivo

Animals received either systemic or regional(intranasal) IFN-�. Animals that received sys-temic rMuIFN-� were dosed according to twodifferent schedules: One group was given 5,000units of rMuIFN-� i.p. 24 h prior to exposure. Asecond group was given 1000 units of rMuIFN-� intranasally 24 h before intranasal challengewith A. fumigatus spores and continued for fivedays (total dose 5,000 units). Animals given re-gional IFN-� were dosed as follows: The firstgroup received 5,000 units intranasally 24 hprior to infection. The second group received1000 units of intranasal IFN-� daily for five daysbeginning 14 h prior to inoculation with spores.The third group received 100 units of intranasalIFN-� according to the same schedule as the sec-ond group. Mortality after challenge with A. fu-migatus spores was observed for two weeks asreported previously [10,11].

Statistical analysis

Analysis of variance with repeated measuresdesign was used to assess in vitro and in vivointeractions between host phagocytic cells andA. fumigatus spores [31]. Mortality data was an-alyzed using Kaplan Meier life table analysis[31]. Data are expressed as mean � SEM. A pvalue of �0.05 was considered significant.

RESULTS

Effects of IFN-� on ingestion and killing of A. fumigatus spores by pulmonary alveolarmacrophages in vitro

Pulmonary alveolar macrophages, harvestedby BAL were incubated with A. fumigatusspores in a 1:5 ratio (cells/spores), with andwithout 250 IU of rMuIFN-�. The macrophagesfrom different mice groups ingesting spores are shown in Figure 1, Panels A and B. The per-

centage of macrophages containing spores andthe mean number of spores/cells were signifi-cantly increased in the IFN-�-treated groups(p � 0.0005 and p � 0.006, respectively) (Fig. 2).

Interferon-gamma also significantly enhancedthe killing function of normal macrophages invitro (Fig. 3). Pulmonary alveolar macrophageswere incubated with A. fumigatus spores andtwo different concentrations of methylpredni-solone sodium succinate (1 and 100 mcg/mL).The higher concentration of methylpredniso-lone inhibited killing to a greater extent, butboth concentrations inhibited killing when com-pared to controls. Interferon gamma partiallyrestored killing function of macrophages incu-bated with the higher concentration of methyl-prednisolone (p � 0.008), but did not restorekilling function significantly in the presence ofthe lower concentration (p � 0.07).

In vitro killing of A. fumigatus spores byneutrophils incubated with IFN-�

Peripheral blood neutrophils were incubatedwith A. fumigatus spores (1:5 ratio) with andwithout 250 IU rMuIFN-� and in the presenceof 1 mcg/mL methylprednisolone. The in vitroingestion of spores by peripheral blood neu-trophils are shown in Figure 1, Panels C and D.Again, the addition of IFN-� improved thephagocytosis and killing function of normalneutrophils (Fig. 4; low-dose, p � 0.007; high-dose, p � 0.007).

In vivo effects of IFN-� treatment on miceexposed to intranasal A. fumigatus spores

Administration of systemic rMuIFN-� (5,000IU i.p. 24 h prior to exposure) did not improvein vivo killing of A. fumigatus spores in normalor cortisone-treated mice. At four h after expo-sure to 5 � 106 spores, 18.3 � 4.5% of sporeswere viable in the lungs of control animals ver-sus 16.7 � 0.5% for IFN-�–treated controls (Fig.5). Pretreatment with cortisone acetate (125mg/kg i.p. 24 h prior to spore exposure) signif-icantly impaired in vivo killing of spores (p �0.008), and treatment with IFN-� in the presenceof cortisone acetate did not substantially im-prove in vivo killing (29.3 � 1.5% viable sporesremaining in cortisone IFN-�–treated mice atfour h versus 18.3 � 4.5% for normal controls).

IFN-� AND ASPERGILLOSIS 401

Failure of systemic rMuIFN-� to improve invivo killing was associated with impaired re-cruitment of neutrophils to the lungs in thepresence of corticosteroids (Fig. 6). Adminis-tration of cortisone acetate 125 mg/kg i.p. 24 hprior to spore exposure reduced the number ofneutrophils recruited to the lungs immediatelyafter exposure to A. fumigatus spores. Normal

mice recruited 21.8 � 5.5 � 104 neutrophils atfour hours post-exposure versus 3.8 � 104 neu-trophils for cortisone-treated mice (p � 0.05).Treatment with systemic IFN-� did not en-hance neutrophil recruitment for normal orcortisone-treated mice. To the contrary, de-creased neutrophil recruitment was seen whensystemic IFN-� was given.

FIG. 1. Examples of pulmonary alveolar macrophages and peripheral blood neutrophils incubated with 5 � 106 A.fumigatus spores in a 1:5 ratio (with and without 250 IU of fMuIFN-�). (A) Normal macrophages. (B) Macrophagesplus IFN-�. (C) Normal neutrophils. (D) Neutrophils plus IFN-�. At least 200 phagocytic cells were counted to de-termine phagocytic index.

FIG. 2. Incubation of pulmonary alveolar macrophages with 5 � 106 A. fumigatus spores in a 1:5 ratio, with and with-out 250 IU of rMuIFN-�. Percentage macrophages containing at least one spore is shown in (A) the left panel, whereasphagocytic index (mean number of spores ingested per cell) is shown in (B) the right panel (n � 3 experiments).

402 JOHNSON ET AL.

Additional experiments were performed todetermine whether systemic IFN-� either alone,or in combination with corticosteroids might im-pair neutrophil recruitment to inflammatorysites by causing neutropenia. The number of cir-culating neutrophils was not significantly al-tered by IFN-� therapy (data not shown).

In contrast to systemic treatment with IFN-�,treatment with regional IFN-� (5,000 units intra-nasally) at 24 h before spore exposure did sig-

nificantly augment neutrophil recruitment tothe lungs and increased spore killing in vivo(Fig. 7A,B).

Effects of IFN-� on mortality of mice exposed toA. fumigatus spores

The survival data for mice treated with var-ious combinations of systemic and regionalIFN-� and exposed to A. fumigatus spores are

FIG. 3. In vitro killing of 5 � 106 A. fumigatus spores by pulmonary alveolar macrophages incubated with and with-out 250 IU rMuIFN-�, in the presence of low dose (1 mcg/mL) and high dose (100 mcg/mL) methylprednisolone. (A) Low dose. (B) High dose. M, untreated macrophages from mice; C, macrophages treated with methylprednisolone;IFN, interferon-� (n � 3 experiments).

FIG. 4. In vitro killing of 5 � 106 A. fumigatus spores by peripheral blood neutrophils incubated with and without250 IU rMuIFN-�, in the presence of low-dose (1 mcg/mL) and high-dose (100 mcg/mL) methylprednisolone. (A)Low dose. (B) High dose. N, untreated neutrophils from mice; C, neutrophils treated with methylprednisolone IFN-interferon-� (n � 3 experiments).

IFN-� AND ASPERGILLOSIS 403

shown in Figure 8, Panels A and B. Both sys-temic and intranasal IFN-� failed to prolong sur-vival of cortisone-treated mice exposed to 5 �106 spores. To the contrary, survival appeareddecreased in both groups and was significantlyworse in one of the intranasal IFN-� groups.

DISCUSSION

The present murine model of IA utilizing in-tranasal instillation of A. fumigatus spores es-

tablishes a pulmonary infection that is similarin pathogenesis to human infection. This maybe a more relevant model (compared to intra-venous injection of spores) for examining theeffects of systemic and regionally administeredcytokines on various parameters of host de-fense and survival from invasive aspergillosis.In humans, Aspergillus infections occur pre-dominantly in corticosteroid-treated patients[32–34], and infection is acquired through theinhalation of A. fumigatus spores. The spores(1.5–3.5 �m diameter) settle in small airways

FIG. 5. In vivo killing of 5 � 106 A. fumigatus spores at 4 and 8 h in normal, cortisone-treated, and cortisone plusIFN-�–treated mice. Cortisone-treated animals received 125 mg/kg; i.p. 24 h prior to spore exposure. IFN-� was givenas single dose, 5,000 U i.p., 24 h prior to exposure (n � 3 experiments).

FIG. 6. Effect of cortisone and systemic IFN-� treatment on neutrophil recruitment to the lungs of the mice chal-lenged with 5 � 106 A. fumigatus spores (n � 3 experiments).

404 JOHNSON ET AL.

where defective pulmonary alveolar macro-phage function permits germination and hy-phal growth [35,36]. Local tissue invasion withsubsequent dissemination to distant organs oc-curs in the presence of impaired neutrophil re-cruitment as occurs in the setting of corticos-teroid therapy or neutropenia.

Our results demonstrate that IFN-� im-proves the phagocytic and killing functions of normal macrophages and neutrophils for A. fumigatus, but corticosteroid-treated macro-phages and neutrophils remain impaired de-spite treatment with IFN-�. Other investigatorshave shown that IFN-� can restore certain func-tions of corticosteroid-treated monocytes invitro, but that the effects are pathogen-depen-dent. For example, Schaffner et al. reportedthat, in the presence of corticosteroid, IFN-�improved macrophage antimicrobial activityfor Listeria and Salmonella, but not Aspergillusor Nocardia [37]. This suggested that effectivekilling of A. fumigatus spores requires protein

synthesis that may be inhibited by steroids(through their inhibitory effects on gene tran-scription), whereas killing of other pathogenssuch as Listeria may occur through oxidativemechanisms that are not steroid-dependent[38]. Gaviria et al. reported that proinflamma-tory cytokines such as IFN-� may play an ad-junctive role in selected fungal infection byaugmenting the antifungal activity of polymor-phonuclear leukocytes/peripheral blood mono-nuclear cells (PMNL/PBMC) by mediating hy-phal damage in Aspergillus fumigatus, Fusariumsolani, and Candida albicans [39].

Aside from effects on phagocytic functionsof neutrophils and macrophages, IFN-� modu-lates other aspects of host immunity that couldbe important in host defense from Aspergillusinfection. Interferon-gamma increases T-cell re-sponsiveness by increasing MHC class I and

FIG. 7. Effect of intranasal treatment with IFN-� on neutrophil recruitment (A) and spore killing (B) in thelungs of mice challenged with 5 � 106 A. fumigatus spores (n � 3 experiments).

FIG. 8. Mortality data for cortisone-treated mice chal-lenged with 5 � 106 A. fumigatus spores. (A) Upper panelrepresents systemically administered IFN-� 5,000 U � 1or 1000 U daily for five days. (B) Lower panel representsregional (intranasal) IFN-�: 5,000 U � 1, 1000 U daily forfive days or 100 U daily for five days.

IFN-� AND ASPERGILLOSIS 405

class II expression on antigen presenting cells(APC), thereby increasing the effectiveness ofthese cells. Interferon-gamma also increases ac-tivation of T-cells directly by APC-independentmechanisms, and augments the function of by-stander cells such as B-cells and NK cells. Theseadditional actions of IFN-� are, in fact, the ba-sis for its use in the treatment of hepatitis C andin patients undergoing bone marrow and stemcell transplants. Although these effects on lym-phocyte function are important, we were notable to evaluate them specifically in this model.

Cortisone-treated mice exhibited impairedrecruitment of neutrophils into lungs infectedwith A. fumigatus spores. This defect did notimprove after treatment with systemic IFN-�.In fact, systemic IFN-�, given to normal mice,appeared to impair neutrophil recruitment tothe lungs (Fig. 6). This may be possible sinceIFN-� is normally elaborated as a local cytokineand has been shown to decrease neutrophilchemotaxis, thereby functioning to sequesterneutrophils at the site of inflammation [40]. Al-though systemic administration did not causeneutropenia, it could interfere with the cellularlocalization to inflammatory sites.

An alternative explanation for a lack of ben-efit associated with systemic IFN-� could bethat the preparation lacked activity or that aninsufficient amount was given. These possibil-ities are unlikely, as an identical preparation,concentration, and route was used in a previ-ous study that demonstrated efficacy. In thatstudy, we found that IFN-� was beneficialwhen administered systemically to cortisone-treated mice exposed intravenously to A. fumi-gatus spores [25]. Other examples of protectionwith systemic IFN-� include infections withMycobacterium avium-intracellulare, Toxoplasmagondii, and experimental leishmaniasis [41–43].The current study was designed to examine thebenefit of IFN-� in a more clinically pertinentmodel of pulmonary infection.

Delivery of intranasal IFN-� was designed toachieve high concentrations of IFN-� in thelung. Following the administration of in-tranasal IFN-�, neutrophil recruitment to thelungs was enhanced and killing function im-proved, even in the presence of corticosteroids.However, outcomes were worse for each IFN-� regimen utilized. Increased leukocyte migra-

tion into the lungs during the acute phase ofpneumonia may have contributed to increasedtissue damage and increased mortality. Similaradverse effects from augmented neutrophil re-cruitment have been seen in experimentalmodels of acute respiratory distress syndrome(ARDS) [44].

Although we were unable to show a benefi-cial effect for either systemic or regional IFN-�in these experiments, it should be acknowl-edged that the infection model is still some-what artificial and only a limited number oftherapeutic combinations were investigated. Itis conceivable that a different inoculating dose(or doses) of A. fumigatus spores or varying thecombinations of intranasal and systemic IFN-�might eventually show a protective effect forcytokine administration.

Our findings demonstrate that the overall ef-fects of IFN-� therapy, when given to augmenthost defense for invasive aspergillosis, mayvary under different in vitro and in vivo con-ditions. Furthermore, the in vivo responses tocytokine therapy may be influenced by theroute of infection (and route of administrationfor the cytokine). Accordingly, experimentalmodels which are used to investigate the po-tentially salutory effects of adjuvant cytokinetherapy for invasive mycoses should be basedon a pathogenesis of infection that is clinicallyrelevant.

REFERENCES

1. Kurup VP, Kumar A. Immunodiagnosis of asper-gillosis. Clin Microbiol Rev 1991;4:439–456.

2. Lentino JR, Rosenkranz MA, Michaels JA, et al. Noso-comial aspergillosis: A retrospective review of air-borne disease secondary to road construction andcontaminated air conditioners. Am J Epidemiol 1982;116:430–437.

3. Warris A, Voss A, Verweij PE. Hospital sources of As-pergillus: New routes of transmission? Rev IberoamMicol 2001;18:156–162.

4. Beyer J, Schwartz S, Heineman V, Siegert W. Strate-gies in prevention of invasive pulmonary aspergillo-sis in immunosuppressed or neutropenic patients.Antimicrob Agents Chemother 1994;38:911–917.

5. Park SB, Kang MJ, Whang EA, et al. A case of pri-mary cutaneous Aspergillus in a renal transplant re-cipient. Transplant Proc 2004;36:2156–2157.

6. Kusne S, Torre-Cisneros J, Manez R, et al. Factors as-sociated with invasive lung aspergillosis and the sig-

406 JOHNSON ET AL.

nificance of positive Aspergillus culture after livertransplantation. J Infect Dis 1992;166:1379–1383.

7. Ohmagari N, Raad II, Hachem R, Kontoyiannis DP.Invasive aspergillosis in patients with solid tumors.Cancer 2004;101:2300–2302.

8. Denning DW, Follansbee SE, Scolaro M, et al. Pul-monary aspergillosis in the acquired immunodefi-ciency syndrome. N Engl J Med 1991;324:654–662.

9. Altes A, Remacha AF, Sarda P, et al. Frequent severeliver iron overload after stem cell transplantation andits possible association with invasive aspergillosis.Bone Marrow Transplant 2004;34:505–509.

10. Miyamoto GY, Barlam TF, Vander Els NJ. Invasive as-pergillosis in patients with AIDS. Clin Infect Dis1992;14:66–74.

11. Schaffner A, Douglas H, Braude A. Selective protec-tion against conidia by mononuclear and againstmycelia polymorphonuclear phagocytes in resistanceto Aspergillus. J Clin Invest 1982;69:617–631.

12. Rinehart JJ, Balcerzak SP, Sagone AL, LoBuglio AF.Effects of corticosteroids on human monocyte func-tion. J Clin Invest 1974;54:1337–1343.

13. Schaffner A. Therapeutic concentrations of glucocor-ticoids suppress the antimicrobial activity of humanmacrophages without impairing their responsivenessto gamma interferon. J Clin Invest 1985;76:1755–1764.

14. Wasylnka JA, Moore MM. Aspergillus fumigatusconidia survive and germinate in acidic organelles ofA549 epithelial cells. J Cell Sci 2003;116:1579–1587.

15. Rhame F. Prevention of nosocomial aspergillosis. JHosp Infect 1991;18:466–472.

16. Gerson SL, Talbot GH, Hurwitz S, et al. Prolongedgranulocytopenia: The major risk factor for invasivepulmonary aspergillosis in patients with acuteleukemia. Ann Intern Med 1984;100:345–351.

17. Beaman L. Fungicidal activation of murine macro-phages by recombinant gamma interferon. Infect Im-mun 1987;55:2951–2955.

18. Brummer E, Morrison CJH, Stevens DA. Recombinantand natural gamma-interferon activation of micro-phages in vitro: Different dose requirements for induc-tion for killing activity against phagocytizable and non-phagocytizable fungi. Infect Immun 1985;49:725–730.

19. Farmaki E, Roilides E. Immunotherapy in patientswith systemic mycoses: A promising adjunct. Bio-Drugs 2001;15:207–214.

20. Williams JG, Jurkovich GH, Maier RV. Interferon-�:A key immunoregulatory lymphokine. J Surg Res1993;54:79–93.

21. Murray HW. Interferon-gamma, the activated mi-crophage and host defense against microbial chal-lenge. Ann Intern Med 1988;108:595–608.

22. Berton G, Zeni L, Cassatella MA, Rossi F. Gamma in-terferon is able to enhance the oxidative metabolismof human neutrophils. Biochem Biophys Res Com-mun 1986;138:1276–1282.

23. Morrison CJ, Brummer E, Isenberg RA, Stevens DA.Activation of murine polymorphonuclear neutrophilsfor fungicidal activity by recombinant gamma inter-feron. J Leukocyte Biol 1987;41:434–440.

24. Abe S, Hu W, Ishibashi H, Hasumi K, Yamaguchi H.Augmented inhibition of Candida albicans growth bymurine neutrophils in the presence of a tryptophanmetabolite, picolinic acid. J Infect Chemother 2004;10:181–184.

25. Nagi H, Guo J, Choi H, Kurup VP. Interferon-� andtumor necrosis factor-� protect mice from invasive as-pergillosis. J Infect Dis 1995;172:1554–1560.

26. Kurup VP, Guo J, Choi H, Fink JN. The protective effect of exogenous interleukin-12 in murine asper-gillosis. Immunol Infect Dis 1996;6:175–182.

27. Kurup VP. Interaction of Aspergillus fumigatus sporesand pulmonary alveolar macrophages of rabbits. Im-munobiology 1984;166:53–61.

28. Kurup VP, Mauze S, Choi H, et al. A murine modelof allergic bronchopulmonary aspergillosis with ele-vated eosinophils and IgE. J Immunol 1992;148:3783–3788.

29. Kurup VP, Seymour BWP, Choi H, Coffman RL. Par-ticulate Aspergillus fumigatus antigens elicit a Th2 re-sponse in BALB/c mice. J Allergy Clin Immunol1994;93:1013–1020.

30. Grunig G, Corry DB, Leach MW, et al. Interleukin-10is a natural suppressor of cytokine production and in-flammation in a mouse model of allergic broncho-pulmonary aspergillosis. J Exp Med 1997;185:1089–1099.

31. BMDP Statistical Software. Berkeley, CA: Universityof California Press, 1990.

32. Gustafson TL, Schaffner W, Lavely GB, et al. Invasiveaspergillosis in renal transplant recipients: Correla-tion with corticosteroid therapy. J Infect Dis 1983;148:230–238.

33. Sohail MR, Smilack JD. Aspergillus fumigatus septicarthritis complicating intra-articular corticosteroid in-jection. Mayo Clin Proc 2004;79:578–579.

34. Wang H, Ding Y, Li X, et al. Fatal aspergillosis in apatient with SARS who was treated with cortico-steroids. N Engl J Med 2003;349:507–508.

35. Walsh TJ, Dixon DM. Nosocomial aspergillosis: Envi-ronmental microbiology, hospital epidemiology, diag-nosis and treatment. Eur J Epidemiol 1989;5:131–142.

36. Shibuya K, Ando T, Hasegawa C, et al. Pathophysi-ology of pulmonary aspergillosis. J Infect Chemother2004;10:138–145.

37. Schaffner A, Douglas H, Braude AI, Davis CE. Killingof Aspergillus spores depends on the anatomicalsource of the macrophage. Infect Immun 1983;2:1109–1115.

38. Nathan CF, Murry HW, Weibe ME, Rubin BY. Iden-tification of interferon-gamma as the lymphokine thatactivates human macrophage oxidative metabolismand antimicrobial activity. J Exp Med 1983;158:670–689.

39. Gaviria JM, van Burik JA, Dale DC, et al. Compari-son of interferon-gamma, granulocyte colony-stimu-lating factor, and granulocyte-macrophage colony-stimulating factor for priming leukocyte-mediatedhyphal damage of opportunistic fungal pathogens. JInfect Dis 1999;179:1038–1041.

IFN-� AND ASPERGILLOSIS 407

40. Beilfeldt HO, Babiuk LA. Alteration of some leuko-cyte functions following in vivo and in vitro exposureto recombinant bovine alpha and gamma interferon.J Interferon Res 1986;6:123.

41. Squires KE, Murphy WF, Madoff LC, Murry HW. In-terferon-� and Mycobacterium avium intracellulare in-fection. J Infect Dis 1989;159:599–601.

42. Murray HW. Gamma interferon, cytokine-inducedmacrophage activation, and antimicrobial host de-fense. In vitro, in animal models, and in humans. Di-agn Microbiol Infect Dis 1990;13:411–421.

43. Vanloubbeeck Y, Jones DE. Protection of c3heb/fejmice against Leishmania amazonensis challenge afterprevious leishmania major infection. Am J Trop MedHyg 2004;71:407–411.

44. Baron RM, Carvajal IM, Fredenburgh LE, et al. Nitricoxide synthase-2 down-regulates surfactant protein-B expression and enhances endotoxin-induced lunginjury in mice. FASEB J 2004;18:1276–1278.

Address reprint requests to:Dr. Charles E. Edmiston, Jr.

Department of SurgeryMedical College of Wisconsin

9200 West Wisconsin Ave.Milwaukee, WI 53226

E-mail: edmiston@mcw.edu