|Home | About | Journals | Submit | Contact Us | Français|
A standardized protocol for extracting DNA from Aspergillus fumigatus has been proposed by the European Aspergillus PCR Initiative (EAPCRI). Using meta-regression analysis, the EAPCRI showed certain stages of the process to be critical to providing a satisfactory analytical sensitivity. The study investigated each step of the EAPCRI protocol by elimination and monitored the influence on Aspergillus PCR performance.
The European Aspergillus PCR Initiative (EAPCRI) recently published their research evaluating fungal DNA extraction and Aspergillus PCR amplification techniques (5). The paper focused on DNA extraction from whole blood with the aim of providing a standard that could be used to assess clinical validity and utility to allow its inclusion in future disease-defining criteria (1). The results showed that the DNA extraction process was critical to the success of most PCR amplification systems that performed adequately with similar analytical sensitivities.
The standardized whole-blood fungal DNA extraction protocol involved lysis of human blood cells before lysis of the fungal cell by mechanical disruption and DNA purification, precipitation, and elution using commercial kits or instruments. Individual steps and parameters in the DNA extraction process, namely, white cell lysis, fungal lysis by bead beating, and determination of the elution volume, were shown statistically to be crucial to its success, whereas other steps, such as using the entire specimen and red cell lysis, were not critical but were associated with better assay performance. The focus of this report is the evaluation of the performance of the recommended EAPCRI whole-blood DNA extraction protocol with stepwise exclusion/modification of individual steps with the aim of corroborating the critical stages of fungal DNA extraction and the rationale behind the EAPCRI recommendations.
EDTA-whole blood was spiked with A. fumigatus (ATCC strain 1022) conidia at two clinically representative fungal burdens (33 conidia/ml and 10 conidia/ml) and then frozen at −80°C. Five replicates of 3 ml EDTA-blood with each fungal concentration were tested for each extraction variant, and controls were included to monitor interexperimental variation and possible contamination. The full DNA extraction protocol was that recommended by the EAPCRI, and on this occasion, the Roche High Pure template DNA kit was used for final DNA purification (5). For most methods, DNA was eluted in a final volume of 60 μl, providing DNA eluate concentrations of 8.8 to 88.3 rRNA copies per μl, dependent on the fungal burden and methodology (Table (Table1).1). DNA extraction was then performed, with individual steps of the protocol omitted or modified as described in Table Table1.1. To monitor PCR inhibition, an internal control (IC), a plasmid containing the capsular transport (CTRA) gene of Neisseria meningitidis, was introduced at the start of the process that used the High Pure kit. As the IC was not included from the start, it could not be used to determine extraction efficiency. The EAPCRI protocol was also modified to include the additional step of a hot NaOH incubation to help permeabilize the fungal cell wall as previously described (3). Aspergillus PCR was performed in duplicate by using the method of White et al., albeit using volumes that allowed PCR to be performed using 15 μl of DNA template in a final volume of 50 μl amplified using the Corbett Rotorgene 3000 system (6). PCR standards were included to monitor amplification efficiency and maintain interexperimental threshold consistency. The IC PCR used 10 μl of DNA template in a final volume of 20 μl, and PCR inhibition was determined by comparison of the individual threshold cycle (CT) for each specimen to that of a positive plasmid control.
At the higher fungal burden (33 conidia/ml), only the omission of bead beating significantly affected Aspergillus PCR detection rates compared to those attained using the EAPCRI protocol (Table (Table2).2). Four modifications affected PCR amplification, indicating suboptimal quality and/or quantity of DNA as represented by later Aspergillus PCR CT values (Table (Table2).2). In using 1 ml of the 3-ml specimen, the fungal burden potentially available was reduced from 33 conidia/ml to 11 conidia/ml, a load almost identical to that of the 3-ml specimen spiked with 10 conidia/ml extracted using the EAPCRI method (Table (Table1).1). CT values for these specimens were also the same (Table (Table2).2). Removal of a bead-beating step reduced the quantity of fungal DNA extracted, not only affecting reproducibility but also the Aspergillus PCR CT values. Omission of white cell lysis affected the IC, indicating the presence of an inhibitor, and without prior lysis, human DNA survived the extraction, possibly in DNA concentrations potentially detrimental to PCR amplification, although the higher burden was still reproducibly detected. It is postulated that the high fungal DNA concentration combined with a large Aspergillus PCR volume compared to those of the IC PCR limited the inhibitory effect. Elution in a larger volume, 200 μl, resulted in a DNA concentration that was lower than those attained using the EAPCRI method and is reflected by later CT values in both the Aspergillus PCR and IC and a slight reduction in Aspergillus PCR reproducibility (Tables (Tables11 and and22).
At the lower fungal burden (10 conidia/ml), seven modifications affected Aspergillus PCR performance (Table (Table2).2). As bead beating was so detrimental for the detection of the higher fungal burden, it was not evaluated at this lower load. Using less than the entire 3-ml specimen, excluding white cell lysis, and eluting in volumes of ≥100 μl greatly reduced detection rates compared to those obtained using the EAPCRI protocol. Assuming 100% extraction efficiency and elution in 60 μl, using a 15-μl template input for PCR incorporates 7.5 (95% confidence interval [CI], 2.7 to 12.3) A. fumigatus genomes per reaction, certifying the presence of DNA within each reaction, whereas elutions in 100-μl and 200-μl volumes incorporate 4.5 (95% CI, 0.5 to 8.5) and 2.25 (95% CI, 0 to 5.0) genomes per reaction, respectively, reducing replicate reproducibility.
A reduced detection rate was coupled with later CT values compared to those obtained with the EAPCRI method, and four modifications resulted in mean CT values greater than 40 cycles, a threshold that reduces real-time PCR reproducibility below 100% (4). Later CT values were also seen if red cell lysis was not done, and this was coupled with a slight reduction in the detection rate. Again, the IC was affected by the absence of the white cell lysis step and by the elution of DNA in 200-μl volumes (Table (Table22).
The removal of one red cell lysis step had no affect on PCR performance, but in this investigation, the blood was frozen prior to testing, resulting in the lysis of the red cells. It is likely that when using fresh blood, two red cell lysis steps are required. Incorporating incubation with hot NaOH after white cell lysis did not improve PCR reproducibility or CT values for either fungal burden. The primary function of a NaOH incubation is to permeabilize the fungal cell wall and, in so doing, aid downstream lyticase digestion. The use of bead beating obviates this additional step, though hot alkali treatment may still be beneficial when testing blood from nonneutropenic patients, as a greater abundance of white cells may persist despite attempts at conventional lysis (personal practical experience).
In conclusion, to improve analytical sensitivity, the EAPCRI recommends the use of ≥3 ml EDTA blood specimens, a red and white cell lysis step, bead beating to lyse the fungal element, and elution volumes of less than 100 μl. This research investigating the influence of the individual steps of the EAPCRI protocol supports the original findings and emphasizes their importance at low fungal burdens.
Published ahead of print on 18 August 2010.
†The authors have paid a fee to allow immediate free access to this article.