Evaluation of performance of four genotypic methods for studying the genetic epidemiology of Aspergillus fumigatus isolates

Abstract

In the present investigation, 49 Aspergillus fumigatus isolates obtained from four nosocomial outbreaks were typed by Afut1 restriction fragment length polymorphism (RFLP) analysis and three PCR-based molecular typing methods: random amplified polymorphic DNA (RAPD) analysis, sequence-specific DNA primer (SSDP) analysis, and polymorphic microsatellite markers (PMM) analysis. The typing methods were evaluated with respect to discriminatory power (D), reproducibility, typeability, ease of use, and ease of interpretation to determine their performance and utility for outbreak and surveillance investigations. Afut1 RFLP analysis detected 40 types. Thirty types were observed by RAPD analysis. PMM analysis detected 39 allelic types, but SSDP analysis detected only 14 types. All four methods demonstrated 100% typeability. PMM and RFLP analyses had comparable high degrees of discriminatory power (D = 0.989 and 0.988, respectively). The discriminatory power of RAPD analysis was slightly lower (D = 0.971), whereas SSDP analysis had the lowest discriminatory power (D = 0.889). Overall, SSDP analysis was the easiest method to interpret and perform. The profiles obtained by PMM analysis were easier to interpret than those obtained by RFLP or RAPD analysis. Bands that differed in staining intensity or that were of low intensity were observed by RAPD analysis, making interpretation more difficult. The reproducibilities with repeated runs of the same DNA preparation or with different DNA preparations of the same strain were high for all the methods. A high degree of genetic variation was observed in the test population, but isolates were not always similarly divided by each method. Interpretation of band profiles requires understanding of the molecular mechanisms responsible for genetic alternations. PMM analysis and Afut1 RFLP analysis, or their combination, appear to provide the best overall discriminatory power, reproducibility, ease of interpretation, and ease of use. This investigation will aid in planning epidemiologic and surveillance studies of A. fumigatus.

FIG. 1.

Southern blotting profiles of EcoRI-digested genomic DNAs of 17 A. fumigatus isolates (lanes 1 to 17, respectively) hybridized with digoxigenin-labeled Afut1 DNA. Lane M, molecular size standards, with the sizes indicated on the left.

FIG. 2.

RAPD profiles produced by amplification with primer NS3 electrophoresed through a 1.5% agarose gel. The numbers above the lanes denote the NS3 genotype. Lane M, molecular size standards, with the sizes indicated on the right; lane C, control consisting of the PCR mixture without DNA template.

FIG. 3.

SSDP analysis with SSDP type 3-specific primer pair Afd1 and Afd2. Isolates were scored for the presence (+) or the absence (−) of a 550-bp band. Lane M, molecular size standards, with the sizes indicated on the right; lane C, control consisting of the PCR mixture without DNA template.

FIG. 4.

PMM analysis of PCR profiles for isolate 19723. (A) Analysis of PCR products was performed with 6-carboxyfluorescein-labeled primer B (←B) and 4,7,2′,4′,5′,7′-hexachloro-6-carboxy-fluorescein-labeled primer D (D→). (B) The molecular sizes of the bands were automatically calculated by using 6-carboxytetramethylrhodamine-labeled internal size standards loaded with each sample. Dye-sample peaks were 1B for primer B and 1G for primer D. The sizes of the PCR fragments was calculated by using the maximum height of the fluorescent peak (∗).