Molecular detection and species-specific identification of medically important Aspergillus species by real-time PCR in experimental invasive pulmonary aspergillosis

Abstract

Diagnosis of invasive pulmonary aspergillosis (IPA) remains a major challenge to clinical microbiology laboratories. We developed rapid and sensitive quantitative PCR (qPCR) assays for genus- and species-specific identification of Aspergillus infections by use of TaqMan technology. In order to validate these assays and understand their potential diagnostic utility, we then performed a blinded study of bronchoalveolar lavage (BAL) fluid specimens from well-characterized models of IPA with the four medically important species. A set of real-time qPCR primers and probes was developed by utilizing unique ITS1 regions for genus- and species-specific detection of the four most common medically important Aspergillus species (Aspergillus fumigatus, A. flavus, A. niger, and A. terreus). Pan-Aspergillus and species-specific qPCRs with BAL fluid were more sensitive than culture for detection of IPA caused by A. fumigatus in untreated (P < 0.0007) and treated (P ≤ 0.008) animals, respectively. For infections caused by A. terreus and A. niger, culture and PCR amplification from BAL fluid yielded similar sensitivities for untreated and treated animals. Pan-Aspergillus PCR was more sensitive than culture for detection of A. flavus in treated animals (P = 0.002). BAL fluid pan-Aspergillus and species-specific PCRs were comparable in sensitivity to BAL fluid galactomannan (GM) assay. The copy numbers from the qPCR assays correlated with quantitative cultures to determine the pulmonary residual fungal burdens in lung tissue. Pan-Aspergillus and species-specific qPCR assays may improve the rapid and accurate identification of IPA in immunocompromised patients.

Fig. 1

Locations of real-time PCR assay targets. An rDNA single repeat is shown. The pan-Aspergillus assay targets the 18S rRNA gene region. The species-specific assays target the ITS1 region.

Fig. 2

Real-time PCR linearity and dynamic range. Five to 5 × 10⁷ copies of each assay's target region were introduced into real-time PCR wells and analyzed to determine crossing thresholds. The log copy number detected per reaction was plotted versus the crossing threshold for pan-Aspergillus (squares), A. fumigatus (triangles), and A. terreus (circles).

Fig. 3

Limit of detection. A mixture of A. fumigatus and A. terreus germlings was introduced at low concentrations (7 to 50 germlings) into BAL fluid, followed by nucleic acid extraction. The eluates were tested by each of the Aspergillus PCR assays. The number of copies detected per reaction was plotted against the number of organisms per 0.5 ml BAL fluid for pan-Aspergillus (squares), A. fumigatus (triangles), and A. terreus (circles).

Fig. 4

Detection of rDNA copies in BAL fluid, showing efficacy of amphotericin B (AmB) treatment in an experimental model of invasive pulmonary aspergillosis. Copies/ml and CFU/ml of BAL fluid are shown for real-time PCR assays (pan-Aspergillus [black bars] and A. fumigatus or A. terreus species-specific [white bars] assays) and culture (gray bars), respectively. Utilizing the A. fumigatus and pan-Aspergillus assays, there was a statistically significant difference in numbers of copies/ml of BAL fluid between the untreated and treated groups for A. fumigatus (P < 0.0001) but not for A. terreus (P = 0.813) (two-tailed Mann-Whitney t test).

Fig. 5

Detection of rDNA copies in tissue homogenates, showing efficacy of amphotericin B (AmB) treatment (A. fumigatus) and ravuconazole (Ravu) treatment (A. terreus) in an experimental model of invasive pulmonary aspergillosis. Log copies per tissue sample and log CFU/g tissue are shown for real-time PCR assays (pan-Aspergillus [black bars] and A. fumigatus or A. terreus species-specific [white bars] assays) and culture (gray bars), respectively. Utilizing the two-tailed Mann-Whitney t test, there was a statistically significant difference in numbers of copies per tissue extraction between the untreated and treated groups for A. fumigatus (P < 0.0001 for amphotericin B at 2.5 mg/kg of body weight/day and P < 0.0003 for amphotericin B at 5.0 mg/kg/day) but not for A. terreus (P = 0.8988).