Detection of Aspergillus fumigatus in a rat model of invasive pulmonary aspergillosis by real-time nucleic acid sequence-based amplification

Abstract

Rapid and sensitive detection of Aspergillus from clinical samples may facilitate the early diagnosis of invasive pulmonary aspergillosis (IPA). A real-time nucleic acid sequence-based amplification (NASBA) method was investigated by use of an inhalational rat model of IPA. Immunosuppressed male Sprague-Dawley rats were exposed to Aspergillus fumigatus spores for an hour in an aerosol chamber. Bronchoalveolar lavage (BAL) fluid, lung tissues, and whole blood were collected from five infected rats at 1, 24, 48, 72, and 96 h postinfection and five uninfected rats at the end of the experiment. Total nucleic acid (TNA) was extracted on an easyMAG instrument. A primer-molecular beacon set targeting 28S rRNA was designed to detect Aspergillus spp. The results were compared to those of quantitative PCR (qPCR) (18S rDNA) and quantitative culture. The analytical sensitivity of the real-time NASBA assay was <1 CFU/assay. A linear range of detection was demonstrated over 5 log units of conidia (10 to 10(5) spores). Both NASBA and qPCR showed a progressive increase in lung tissue burdens, while the CFU counts were stable over time. The fungal burdens in BAL fluid were more variable and not indicative of a progressive infection. The results of both real-time assays correlated well for both sample types (r = 0.869 and P < 0.0001 for lung tissue, r = 0.887 and P < 0.0001 for BAL fluid). For all whole-blood specimens, NASBA identified Aspergillus-positive samples in the group from which samples were collected at 72 h postinfection (three of five samples) and the group from which samples were collected at 96 h postinfection (five of five samples), but no positive results were obtained by culture or PCR. Real-time NASBA is highly sensitive and useful for the detection of Aspergillus in an experimental model of IPA.

FIG. 1.

(A) Linearity of the pan-Aspergillus NASBA assay determined by triplicate amplification of total nucleic acids extracted from serially diluted A. fumigatus conidia with 10⁵ to 10¹ spores. (B) Linearity of Aspergillus qPCR assay. The same TNA series used for the experiment whose results are presented in panel A were used to generate the standard curve.

FIG. 2.

Aspergillus burdens in lung tissues (A) and BAL fluids (B) in terms of CFU counting, qPCR, and real-time NASBA measurements. UI, uninfected group. A linear increase in the Aspergillus burden in lung tissues was observed by PCR (F = 40.66, P < 0.0001) and NASBA (F = 141.10, P < 0.0001).

FIG. 3.

The correlation between the results of the real-time NASBA and qPCR assays for the detection of the Aspergillus burden in lung tissue (A) and BAL fluids (B) from rats with IPA was statistically significant (P < 0.0001). Spearman's correlation coefficients (r) were 0.869 (A) and 0.887 (B). The black lines represent the linear regression, and the dotted lines represent the 95% confidence intervals.