Environmental Monitoring for Aspergillus fumigatus in Association with an Immunosuppressed Rabbit Model of Pulmonary Aspergillosis

Abstract

Aspergillus fumigatus causes life-threatening pneumonia in immunocompromised patients. Conidia, the infectious form of the organism, are handled in a biologic safety cabinet under BSL2 conditions. However because germinated conidia form noninfectious hyphae in tissue, we hypothesized that rabbits inoculated intratracheally would grow A. fumigatus in their lungs but that the environment would remain free of this fungus, potentially permitting maintenance of infected animals under ABSL1 conditions. We performed a surveillance study for the presence of A. fumigatus in the environment before proceeding with antifungal therapy studies of experimental pulmonary aspergillosis. The expected outcome included absence of A. fumigatus in the environment, stool, and blood and presence in rabbit lungs. Female SPF New Zealand white rabbits were immunosuppressed and inoculated intratracheally (n = 4) or intraesophageally (n = 2) with 1.25 × 10⁸ conidia of A. fumigatus. Feces, pan liners, and walls were sampled daily during the 11-d experiment, and blood was sampled on days 2, 6, and 8 after inoculation. Samples were cultured on 5% Sabouraud glucose agar plates. Lungs were weighed and scored for hemorrhagic infarcts and homogenized for culture on 5% Sabouraud glucose agar and trypticase soy agar plates. Blood cultures, rabbit stool, and environmental swabs were all negative for A. fumigatus whereas the lungs inoculated intratracheally demonstrated 4.5 × 10² ± 0.8 × 10² CFU/g of A. fumigatus. Therefore, neutropenic rabbits with experimental invasive pulmonary aspergillosis do not shed conidia of A. fumigatus and can be safely housed under ABSL1 conditions after inoculation.

Aspergillus fumigatus is an opportunistic fungal pathogen that can cause life-threatening invasive pulmonary aspergillosis, the features of which include filamentous growth within the parenchyma of the lung, intravascular thrombosis, angioinvasion, tissue infarction, and occasionally hematogenous dissemination (Figure 1).² Invasive pulmonary aspergillosis is of particular concern for severely immunocompromised patients, such as cancer patients receiving cytotoxic chemotherapy, recipients of hematopoietic stem cells or solid-organ transplants, and patients with chronic granulomatous disease.^1,4,24 Despite advances in treatment with antifungal agents, mortality rates with this infection remain high (greater than 50%), and those patients with severe immunosuppression are at highest risk.¹⁵

Figure 1.

Invasive pulmonary aspergillosis in a neutropenic rabbit lung. A, alveoli; BV, blood vessel wall; arrow, invasion of blood vessels by Aspergillus hyphae; *, intravascular thrombosis. Hematoxylin and eosin stain; bar, 100 µm.

A. fumigatus is ubiquitous in the soil, where it thrives on organic debris. It is a widely distributed organism that sporulates abundantly, with each conidial head producing thousands of conidia.¹¹ Inhalation of conidia by immunocompetent persons rarely causes clinical disease, because the innate immune system readily eliminates the conidia.²² However, with the increase in numbers of immunosuppressed patients, this fungus has become the most prevalent airborne fungal pathogen in developed countries.¹²

A persistently neutropenic rabbit model of invasive pulmonary aspergillosis has been developed to study potential advances in diagnosis, treatment, and prevention of A. fumigatus.^3,8 For example, this model has been used to demonstrate the benefits of using ultrafast computerized tomography to improve timely identification and monitoring of infection with A. fumigatus.^20,26 A novel PCR assay for invasive pulmonary aspergillosis that was developed by using this model is now a clinical molecular diagnostic test.^13,27 In addition, this model has been used to validate an enzyme immunoassay for galactomannan, a component of the fungal cell well that is now used as a biomarker for infection.⁷ Numerous studies have been performed investigating a variety of treatments for A. fumigatus, including liposomal amphotericin B, voriconazole, caspofungin, and various combination protocols involving echinocandins, polyenes, and triazoles.^8,9,14,17-19

On our submission of an animal use protocol, we were requested to perform an environmental surveillance study to ascertain whether persistently neutropenic rabbits shed infectious fungal elements after the animals are infected intratracheally with A. fumigatus conidia. Despite its environmental ubiquity, A. fumigatus is classified as a BSL2 pathogen.⁵ Although our laboratory handles this organism on culture medium within a biologic safety cabinet to prevent propagation of conidia, our initial studies indicated that rabbits intratracheally inoculated with A. fumigatus for the establishment of experimental invasive aspergillosis did not shed conidia and could be handled under ABSL1 conditions.

Materials and Methods

Animals.

Six SPF female New Zealand white rabbits (Covance Research Products, Denver, PA) weighing 2.6 to 3.5 kg at the time of intratracheal inoculation were used. Animals were maintained in accordance with the Guide for the Care and Use of Laboratory Animals¹⁰ in an AAALAC-accredited facility. All procedures in the study were approved by Weill Cornell Medical College's IACUC. Rabbits were individually housed in stainless steel, mobile 6-cage units (4.17 ft² in each cage; Lock Solutions, Laurence Harbor, NJ) containing removable plastic (GE Noryl, Selkirk, NY) sleeves with perforated floors and removable excreta pans. Pans were lined with cage board (Techboard; Shepherd Specialty Products, Somerville, NJ) and were changed 4 times each week, whereas cage sleeves and the rack were changed every 2 wk. The rabbits received a commercial high-fiber pelleted laboratory diet (Hi-Fiber Rabbit Diet, Purina Mills, St Louis, MO) ad libitum and were maintained on municipal water in glass bottles with rubber stoppers and stainless steel sippers. The rabbits were housed in a self-contained isolation cubicle (Full Isolation Cubicle, Britz and Company, Wheatland, WY) with vertically telescoping doors. The cubicle was ventilated with HEPA-filtered room air supplied through a ceiling diffuser in the front of the cubicle. Air was exhausted through registers on both back cubicle corners and was HEPA-filtered before being returned to the holding room. Rabbits were housed at 68 ± 2 °F (20 ± 1 °C), with 30% to 70% relative humidity and a 12:12-h light:dark cycle (0600 to 1800). Cubicle ventilation provided 25 air changes hourly. The cubicle was maintained at negative pressure with regard to the room. The room was maintained at a negative pressure (0.03 in. of water difference) relative to the corridor that served it. The room contained 3 isolation cubicles along one wall. Three times each week, the rack was removed from the cubicle, and the floor was mopped with Quatricide (Pharmacal Research Laboratories, Naugatuck, CT); the walls were wiped with a damp cloth as needed.

Cage components were sanitized in a mechanical washer (Better Built R600, Northwest Systems, Delta, British Columbia) that provided a 180 °F final rinse. An inorganic acid detergent (Acid Power, Pharmacal Research Laboratories) was used during the wash cycle to remove urine scale.

Vascular access in each rabbit was established by the vendor prior to the animals’ arrival at the facility by the surgical placement of a silastic tunneled central venous catheter, as previously described.²⁵ The silastic catheter permitted nontraumatic venous access for the administration of parenteral agents and repeated blood sampling.

Isolate.

NIH Aspergillus fumigatus isolate 4215 (ATCC no. MYA1163), obtained from a patient with fatal pulmonary aspergillosis, was used in the experiment.¹⁹ A frozen isolate (stored at −70 °C) was subcultured onto Sabouraud glucose slants and incubated for 24 h at 37 °C. The slants were then left at room temperature for an additional 5 d before conidia were harvested.

Immunosuppression and maintenance of neutropenia.

Profound and persistent neutropenia (neutrophil concentration of fewer than 100 neutrophils per µL) was established and maintained by using cytarabine (Hospira, Lake Forest, IL) and methylprednisolone (Solu-Medrol, Pfizer for Pharmacia and Upjohn, New York, NY), as described previously.¹⁹ This condition was achieved by using 525 mg/m² cytarabine administered daily for 5 consecutive days beginning on experimental day −1 (day 0 was the day of A. fumigatus inoculation). A maintenance dose of 484 mg/m² cytarabine was administered for 2 additional days, on experimental days 6 and 7, to maintain persistent neutropenia. Methylprednisolone at a dose of 5 mg/kg IV once daily was administered on days −1 and 0 of the study. Ceftazidime (75 mg/kg twice daily), gentamicin (5 mg/kg every other day), and vancomycin (15 mg/kg daily) were administered intravenously 4 d after beginning immunosuppression (day 4) to prevent opportunistic bacterial infections. Vancomycin also was administered in the drinking water (50 mg/L) beginning on day −1 to prevent antibiotic-associated diarrhea due to Clostridium spp.¹⁹ Water bottles were changed every 2 d or more often when water consumption warranted more frequent changes.

Inoculation.

Conidia were harvested under a class 2 type A/B3 biologic safety cabinet by using a solution of 10 mL 0.025% Tween 20 (Fisher Scientific, Pittsburgh, PA) in sterile normal saline, transferred to a conical tube, washed, and counted on a hemacytometer. An inoculum of 1.25 × 10⁸ conidia A. fumigatus in a volume of 250 to 350 µL was prepared for each rabbit. Four rabbits were inoculated intratracheally and the remaining 2 received intraesophageal inoculation. Rabbits were anesthetized intravenously with a solution of ketamine (13 to 14 mg/kg) and xylazine (1.3 to 1.4 mg/kg) for inoculation on the second day of immunosuppression (day 0), as described previously.¹⁹ The A. fumigatus inoculum then was administered intratracheally or intraesophageally via a Flagg 0 straight-blade laryngoscope (Welch Allyn, Skaneateles Falls, NY), with a tuberculin syringe attached to a 5.25-in. 16-gauge Teflon catheter (Becton Dickinson Infusion Therapy Systems, Franklin Lakes, NJ). All rabbits were euthanized on day 9 after inoculation with A. fumigatus.

Outcome variables.

The following panel of outcome variables was used to assess presence of A. fumigatus: pulmonary infarct score, lung weight, pulmonary residual fungal burden, and blood cultures. For environmental surveillance, samples of stool and swabs of the pan liners, walls, and floor of the cubicle were cultured on SGA plates. The outcome variable panel was applied to all experimental rabbits.

Pulmonary lesion scores, lung weights, and residual fungal burden.

The lung and heart block were resected at necropsy. The lungs were weighed and inspected for hemorrhagic infarct lesions in each individual lobe. The number of infarcted lobes was calculated and the mean number of positive lobes per rabbit (the infarct score) was calculated for each group. Bronchoalveolar lavage was performed on each lung preparation by instilling and subsequently withdrawing 20 mL normal saline into the clamped trachea by using 2 sterile 12-mL syringes each containing 10 mL sterile normal saline. The lavage fluid was centrifuged for 10 min at 1500 × g. Most of the supernatant was discarded, and the pellet was resuspended in the remaining 2 mL of fluid. A 0.1 mL sample of this fluid and 0.1 mL of a 1:10 dilution of this fluid were cultured on SGA plates. Lung tissue from each rabbit was sampled and cultured by a standard excision of tissue from each lobe. Each fragment was weighed individually, homogenized with sterile normal saline, and dilutions (10⁻¹ and 10⁻²) were prepared in sterile normal saline. Aliquots (0.1 mL) were plated onto SGA plates and incubated at 37 °C for 24 h and then at room temperature for an additional 24 h. The number of A. fumigatus colony-forming units was counted and recorded for each lobe, and the number of colonies per gram of lung tissue (cfu/g) was calculated.¹⁹

Environmental surveillance.

At least 3 pellets of stool were selected from each rabbit's cage on 2 separate days before inoculation and every day after inoculation until the end of the experiment. The number of samples taken on each day depended on the survival of the rabbits as 2 were euthanized on day 5 and one on day 7. All animals that were euthanized before the end of the study (day 9) were intratracheally inoculated. Pellets were placed in a sterile bag and homogenized with 2 mL of sterile normal saline. A 0.1-mL sample of this fluid was cultured on SGA and trypticase soy agar plates.

Pan liners as well as cubicle walls and floor were swabbed 2 d before inoculation, on the morning of inoculation and every day thereafter until the end of the experiment using sterile, normal saline dampened, cotton-tipped applicators (6-in. wood shaft Cotton Tipped Applicator Sterile 2 Pack, Solon Manufacturing Company, Skowhegan, ME) and roll-cultured on SGA plates (HealthLink, Jacksonville, FL). For all swabs, the sample surface was swabbed in 2 directions at right angles to each other in a close zig-zag pattern while the swab was rotated. A single swab was used to sample the entire floor of the cubicle, another applicator for each pan liner, and one additional swab was used to sample all 3 cubicle walls. One colony per plate for fungus or bacteria was considered positive.

Samples (0.1 mL each) of whole blood collected from the vascular catheter from each rabbit were cultured on SGA plates on days 2, 6, and 8 after inoculation with A. fumigatus. For blood collection, the catheter was cleaned with 70% ethanol, approximately 0.5 mL of blood and lock solution was removed, and 0.1 mL was aspirated for blood culture. The catheter then was flushed with 1 mL sterile normal saline and 1 mL heparinized saline.

Statistical analysis.

Statistical analysis was performed by using Prism 4 (GraphPad Software, San Diego, CA). Groups were compared by using the Mann–Whitney U test. A 2-tailed P value of 0.05 or less was considered to be statistically significant. Values are expressed as mean ± SEM.

Results

A. fumigatus -mediated pulmonary injury.

Rabbits with intratracheal inoculation demonstrated well-established pulmonary aspergillosis caused by A. fumigatus-mediated pulmonary injury, as measured by lung weight and infarct score. Rabbits with intratracheal inoculation demonstrated significantly (P < 0.05) greater lung weight than did rabbits with intraesophageal inoculation (46.4 ± 8.7 g compared with 10.3 ± 0.2 g, respectively). Rabbits with intratracheal inoculation demonstrated significantly (P < 0.01) higher infarct scores (5.8 ± 0.2) compared with no infarcts (0 ± 0) in rabbits with intraesophageal inoculation.

Residual fungal burden.

Rabbits with intratracheal inoculation yielded 4.5 × 10² ± 0.8× 10² cfu/g from lung tissue in comparison to 0 cfu/g for rabbits with intraesophageal inoculation (P < 0.01).

Environmental surveillance.

Stool samples collected from all 6 rabbits, beginning 2 d before inoculation and continuing throughout the 11-d experiment, were negative for A. fumigatus. During the surveillance period, no A. fumigatus was recovered on culture plates inoculated with samples from pad liners, walls, and floors. In addition, the blood samples taken on experimental days 2, 6, and 8 were negative for A. fumigatus.

Discussion

Experimental pulmonary aspergillosis is an invasive disease characterized by invasion of blood vessels in the lungs by hyphae of A. fumigatus, closely recapitulating human infection.^3,8,23 Intratracheal inoculation of rabbits uses conidia to mimic natural infection. The inoculum colonizes the tracheobronchial mucosa with conidia. As the rabbit becomes neutropenic, the conidia germinate to form hyphae, which invade bronchial mucosa, alveolar septa, and blood vessels (Figure 1).

Although the inoculum is administered as a suspension of conidia, A. fumigatus in lung tissue develops into hyphae and invades pulmonary parenchyma (Figure 1). In the current study, all 4 rabbits with intratracheal inoculation demonstrated heavy infection in the lungs, which was expressed by increased lung weights, infarct scores, and numbers of fungal colonies per gram of tissue. Despite the marked infection demonstrated, we were unable to recover any A. fumigatus colonies from cultures of stool samples and swabs of pan liners, walls, and floor throughout all 11 d of the experiment.

The primary route of infection with A. fumigatus is inhalation of airborne conidia (3 to 5 μm), which then are deposited in bronchioles and alveoli. The hyphal form that develops in tissue ranges from 50 to 100 μm in length and is therefore not easily aerosolized. In immunocompetent persons, pulmonary alveolar macrophages clear conidia that reach alveoli and initiate a proinflammatory response. Conidia that evade macrophages germinate into hyphae and become targets for neutrophils that are recruited by the proinflammatory response.⁶ Patients with neutropenia and corticosteroid-induced immunosuppression, such as the rabbits in the current study, are at increased risk of invasive pulmonary aspergillosis.

Animal-to-animal transmission of A. fumigatus infection has not been reported. One reported cluster of cases involved person-to-person transmission, but was related to a wound infection with Aspergillus fumigatus.¹⁶ In contrast to the tissue invasive hyphal form present in intratracheally infected animals, Aspergillus infection within a wound can sporulate in vivo and is potentially transmissible.¹⁶ Infection-control standards in human hospitals focus on controlling the conidial form of Aspergillus spp. Therefore, patients with tissue-invasive Aspergillus infections are not sequestered from other at-risk populations. Infection control in at-risk populations is aimed at avoiding areas and activities where increased levels of conidia are likely, such as construction areas, gardening, and areas with large amounts of dust.²¹

A. fumigatus is an organism that is infectious only in the conidial form. Once the fungus is within the body, it germinates into hyphae and is, therefore, no longer infectious. Figure 1 demonstrates that the organism within tissue that is handled during necropsy is only in the hyphal form and, therefore, unable to aerosolize and cause infection. The conidial form used in these experiments is handled only within a biosafety cabinet (inoculum preparation and plate counting), and it is administered to rabbits intratracheally only. When administered intratracheally, the conidia are suspended in a viscous solution to prevent aerosolization. In addition, the blood samples cultured in the current study were negative for the fungus, indicating that handling it does not present an infectious risk.

Our study also confirms that A. fumigatus does not survive in the rabbit gastrointestinal tract. Acknowledging that rare accidental aspiration might occur during inoculation, we inoculated 2 rabbits intraesophageally. These rabbits did not demonstrate any pulmonary disease, as measured by lung weights, infarct scores, and number of colonies per gram of tissue. There was no evidence of shedding of this organism from the lung or gastrointestinal tract into the stool or immediate environment, given that no A. fumigatus was recovered from any surveillance cultures of stool, pan liners, walls, or floor throughout the experiment. The conidia and any hyphal forms that developed were likely damaged or destroyed by endogenous intestinal flora or the acidic gastric environment.

In conclusion, for studies involving experimental invasive pulmonary aspergillosis in rabbits, an intratracheally established infection does not appear to result in environmental contamination, and ABSL1 practices likely provide sufficient personnel and environmental protection, as is now permitted at our institution.