Abstract
Molecular diagnosis based on genomic amplification methods such as real-time PCR assay has been reported as an alternative to conventional culture for early detection of invasive candidiasis. However, a major limitation of the molecular method is the difficulty associated with breaking fungal cell walls since the DNA extraction step still requires more than half of a working day. It has been suggested that PCR detection of free template DNA in serum is preferred over the use of whole blood for the diagnosis of systemic candidiasis. In this study, two conventional procedures (the first [the HLGT method] consists of boiling sera in an alkaline guanidine-phenol-Tris reagent, and the second [the PKPC method] uses proteinase K digestion, followed by organic extraction) and three commercially available kits for DNA isolation were evaluated for sensitivity, purity, cost, and use of template for most clinically important Candida species in a TaqMan-based PCR assay. To optimize these procedures, we evaluated the effect of adding 0.5% bovine serum albumin to DNA extracts and found that it decreased the effects of inhibitors. The QIAamp DNA blood kit did significantly shorten the duration of the DNA isolation but was among the most expensive procedures. Furthermore, the QIAamp DNA blood kit proved to be as sensitive as the HLGT DNA isolation method for PCR amplification from 52 serum samples from hematology or oncology patients with clinically proven or suspected systemic Candida infections. All PCR-positive samples showed approximately the same Candida species load by both procedures (100% correspondence), whereas one discordant result was obtained between PCR and blood culture.
The management of invasive fungal infections has been hampered by the inability to diagnose the infection at an early stage of disease. However, diagnosis remains difficult, since the only sign of infection may be a prolonged fever that is refractory to antibacterial treatment. In recent years, efforts have been made to develop molecular-biology-based methods for rapid diagnosis, which is crucial to the treatment and recovery of patients suffering from systemic candidiasis. In a comparison of the molecular diagnoses obtained by a real-time PCR-based method to the results of blood culture, the sensitivity and specificity of the molecular method with a C. albicans-specific probe observed with 122 clinical blood samples were 100 and 97%, respectively (12). However, a major limitation of the molecular method in comparison to blood culture was the difficulty associated with problems in breaking fungal cell walls since the DNA extraction step is still a limiting factor, requiring more than half of a working day.
Actually, there is no consensus concerning the best blood fractions to be tested for diagnosis of systemic candidiasis. Several PCR methods have been developed for use either on whole-blood samples (7, 11, 12) or on serum samples (1, 3-5). However, in addition to being too time-consuming and labor-intensive, protocols for extraction of cellular candidal DNA from blood samples may also (i) show significant amounts of PCR inhibitor compounds, (ii) introduce fungal DNA from most of the commercialized enzymatic preparations used in various phases of sample processing, (iii) amplify human white blood cells DNA when PCR primers based on the rRNA gene are used, and (iv) show different efficiencies of the release of genomic DNA from different Candida species, which may exhibit changes in physical composition of the cell wall and therefore susceptibility to digestion. In contrast to blood samples, DNA in serum is not present in a cellular component, and its isolation can therefore be easily achieved. The procedures to isolate solubilized DNA from serum rely solely on its separation from serum proteins, including nucleases, and on its purification. It has previously been suggested that PCR detection of free template DNA in serum is preferred over whole blood for diagnosis of systemic candidiasis (3) and, in the present study, a comparison of five DNA isolation procedures from serum of most clinically important Candida species was performed. These included two conventional methods (5, 13) and three commercially available kits (a QIAamp DNA blood kit, a HighPure PCR template preparation kit, and DNAzol). These conventional procedures used two different strategies for nuclease digestion: boiling lysis in alkaline guanidine thiocyanate reagent or proteinase K. Based on the sensitivity of TaqMan-based PCR assay detection and the purity of DNA isolated and their ease of integration in the routine work flow, we chose two of these methods for DNA isolation from 52 serum specimens from hematology or oncology patients with suspected or proven disseminated candidiasis. The results obtained by TaqMan-based PCR assay were compared to those of immunoenzymatic detection of circulating Candida mannan antigen by using blood cultures as the reference assay.
MATERIALS AND METHODS
Yeast isolates.
Five reference yeast strains of Candida albicans ATCC 24433, C. tropicalis ATCC 66029, C. glabrata ATCC 90030, C. parapsilosis ATCC 22019, and C. krusei ATCC 6258 were cultivated on Sabouraud glucose-agar plates (BBL/Becton Dickinson, Sparks, Md.) for 72 h at 30°C. Species identification was established by using the API 20C kit (bioMérieux, Marcy l'Étoile, France). Cultures were routinely inoculated from single colonies. Serial dilutions of candidal cells were prepared with sterile saline suspensions that were adjusted to a 0.5 McFarland standard (which is ∼106 CFU/ml).
Clinical samples.
During the prospective study period, a total of 52 clinical blood specimens from 39 patients were analyzed. Samples were obtained from patients hospitalized in our institution with clinically proven or suspected systemic Candida infection according to the 2002 Consensus Conference definitions of invasive fungal infections in patients with cancer and recipients of hematopoietic stem cell transplants (2). In brief, the patient symptoms and characteristics included persistent fever, unresponsiveness to broad-spectrum antibacterial therapy, specimen positivity by histology, the isolation of Candida from blood or other sterile sites, colonization of multiple sites with Candida, and repeated computed tomography and ultrasound scans, suggesting a mycotic lesion (e.g., in the liver, spleen, or lung). Blood cultures were evaluated in the BACTEC 9600 blood culture system (Becton Dickinson Diagnostic Instrument Systems), which detected microbial growth by continuous monitoring. Blood (8 to 10 ml) was inoculated into two culture vial types, BACTEC Plus Aerobic/F and BACTEC Plus Anaerobic/F for aerobic and anaerobic cultures, respectively. Whole blood (5 ml) collected in plain tubes for PCR assay and Candida antigen detection were taken in parallel from each patient. Serum samples were obtained by centrifugation of clotted blood and were stored at −70°C until use. Blood cultures were considered negative if they remained negative after 14 days of incubation in the automated BACTEC system. In addition, 15 control blood samples from healthy volunteers were also tested.
Sensitivity of the real-time PCR assay.
For the testing of sensitivity, whole blood from healthy volunteers was artificially spiked in a titration experiment with genomic DNA of Candida spp. (final concentrations of 1 μg to 100 fg of DNA, corresponding to 108 to 101 CFU) in serial dilutions (11). Sera samples obtained from infected blood were isolated and analyzed.
DNA isolation.
In order to compare the purity of DNA prepared to the listed protocols, sera from patients and from spiked blood were subjected to the isolation procedures according to each protocol, with the exception that the enzyme, the lysis buffer and reagents were increased proportional to the serum volume. Altogether, five methods for DNA isolation were compared. We took 400 μl of sera as the uniform volume for all protocols, and 200 μl as the final volume of DNA preparation. Two conventional DNA isolation procedures published in the literature (5, 13) and three commercially available methods were compared. Extraction with the HighPure PCR template preparation kit (Roche Diagnostics, Mannheim, Germany) and DNAzol (ICN Biomedical, Aurora, Ohio) was performed according to the manufacturer's instructions. In the QIAamp DNA blood kit (Qiagen, Hilden, Germany), serum samples were treated with proteinase K and mixed. Subsequently, 400-μl of lysis buffer was added, mixed again, and incubated for 10 min at 56°C. After the addition of 400-μl of absolute ethanol, the mixture was loaded on the QIAamp spin column and washed twice with the AW1 and AW2 washing buffers for 1 min at 6,000 × g and for 3 min at 20,000 × g, respectively. Then, 200 μl of TE buffer (Tris-HCl [pH 8.0], 1 mM EDTA) at 70°C was applied to the spin column, followed by centrifugation for 1 min at 6,000 × g. Finally, the initial filtrate was reapplied to the column, followed by incubation for 5 min at 70°C prior to centrifugation. The first conventional DNA isolation procedure (the heat lysis-guanidine thiocyanate [HLGT] method) was performed as outlined by Sandhu et al. (13), except that the volumes were adjusted proportional to the serum volume. Serum samples were mixed with 1-ml of GPT reagent (6 M guanidine thiocyanate dissolved in 50 mM Tris-HCl [pH 8.0] and mixed with an equal volume of phenol buffered in Tris-HCl [pH 8.0]), followed by incubation in a boiling water bath for 15 min. Then, 500-μl of chloroform-isoamyl alcohol (24:1) was added, followed by further mixing and centrifugation for 10 min at 20,000 × g. The aqueous (upper) phase was transferred to a clean tube and mixed with 500 μl of 100% isopropanol. After 1 h of incubation at −20°C, the mixture was centrifuged (15 min at 20,000 × g), and the supernatant was discarded. Finally, the DNA pellet was washed with 70% ethanol, dried, and dissolved in 200 μl of TE buffer. The second conventional DNA isolation procedure (the proteinase K-phenol-chloroform [PKPC] method) was performed as described by Chryssanthou et al. (5). In brief, DNA was extracted from serum by using proteinase K digestion, followed by phenol-chloroform extraction and ethanol precipitation. The DNA recovered in all procedures was immediately analyzed or stored at −20°C until testing. The purity of DNA was measured by UV spectrophotometry (DU 640 spectrophotometer; Beckman Instruments, Fullerton, Calif.) by calculating the ratio of absorbance at 260 nm to absorbance at 280 nm.
PCR amplification.
The presence of DNA in isolated samples was determined by TaqMan-based PCR assay (GeneAmp 5700 Sequence Detection System; Applied Biosystems, Foster City, Calif.). The level of fluorescence emitted in the reaction is directly proportional to the quantity of PCR product. Each reaction was repeated three times and was performed as previously described (12) with minor modifications. Briefly, each 25-μl of TaqMan PCR mixture consisted of a 5-μl aliquot of sample DNA, 1× PCR buffer, 3.5 mM MgCl2, a 0.2 μM concentration of each primer, each deoxynucleoside triphosphate at a concentration of 0.2 mM, a 0.2 μM concentration of a Candida species-specific probe, 2 U of Taq DNA polymerase (Roche Diagnostics), 0.5 μl of ROX passive reference dye (Invitrogen, Groningen, The Netherlands), and 2.5 μl of 5% bovine serum albumin (BSA; Sigma, St. Louis, Mo.). Thermal cycling conditions consisted of heating at 94°C for 10 min, which preceded a two-stage temperature profile of 30 s at 95°C and 1.5 min at 60°C for 40 cycles. Reactions are characterized by the time during cycling when a threshold of baseline fluorescence (Ct) is exceeded.
Primers ITS86 and ITS4 (15), used for DNA amplification, are complementary to conserved sequences in 5.8S ribosomal DNA and 28S ribosomal DNA, respectively. The fluorogenic probes were designed by Shin et al. (14) for the detection of C. albicans, C. tropicalis, C. glabrata, C. parapsilosis, and C. krusei DNAs, respectively. In each PCR run, a serially diluted standard of genomic Candida species DNA (1 μg to 100 fg of DNA, corresponding to 108 to 101 CFU) (11) was used for the generation of a standard curve. The latter is generated by plotting the Ct values versus log10(N), where N is the concentration of the standard. The Candida species DNA level in each serum sample is determined by locating its Ct on the standard curve. Negative controls were tested by using the same PCR mixture under the amplification conditions described above but without template DNA.
Antigen detection.
Platelia Candida Ag (Bio-Rad, Marnes la Coquette, France), an immunoenzymatic assay based on a rat monoclonal antibody, was used for the detection of Candida antigen (β1-5-oligomannosides) in the sera. The assay was performed according to the manufacturer's instructions.
RESULTS
Comparison of DNA isolation procedures.
Sera obtained from whole blood artificially infected with increasing amounts of genomic DNA of Candida spp. were isolated by one of the following procedures: the PKPC method, the HLGT method, a QIAamp DNA blood kit, a HighPure PCR template preparation kit, or DNAzol. In our preliminary experiments, we found that DNA extracts from simulated serum samples demonstrated significant inhibition of the TaqMan-based PCR assay. Depending on the DNA isolation procedure, the average rate of positivity was ranged from 25 (DNAzol) to 58% (HLGT) (Table 1). Upon addition of 0.5% BSA to PCR mixtures, the HLGT DNA isolation method and QIAamp DNA blood kit yielded more positive PCR results (100%) than the PKPC method and HighPure kit (75%), or DNAzol (63%). Of note, all tested procedures showed almost the same efficiency in isolating DNA, irrespective of the tested Candida species. No signal was obtained in the PCR assay when genomic DNA isolated from serum samples prepared from no seeded blood was used as a template, regardless of whether or not BSA was included in the reaction. Hence, in ulterior experiments, 0.5% BSA was added to all PCR mixtures.
TABLE 1.
Effect of 0.5% BSA on PCR amplification of DNA extracts from sera evaluated by five DNA isolation procedures
| Candida sp. | Presence (+) or absence (−) of BSA | No. of tested samples | No. of positive samples (%) as determined with:
|
||||
|---|---|---|---|---|---|---|---|
| PKPC | HLGT | QIAamp | HighPure | DNAzol | |||
| C. albicans | − | 16 | 8 (50) | 8 (50) | 4 (25) | 6 (38) | 4 (25) |
| + | 16 | 12 (75) | 16 (100) | 16 (100) | 12 (75) | 10 (63) | |
| C. tropicalis | − | 16 | 8 (50) | 10 (63) | 6 (38) | 6 (38) | 4 (25) |
| + | 16 | 12 (75) | 16 (100) | 16 (100) | 12 (75) | 10 (63) | |
| C. glabrata | − | 16 | 8 (50) | 10 (63) | 6 (38) | 6 (38) | 4 (25) |
| + | 16 | 12 (75) | 16 (100) | 16 (100) | 12 (75) | 10 (63) | |
| C. parapsilosis | + | 16 | 12 (75) | 16 (100) | 16 (100) | 12 (75) | 10 (63) |
| C. krusei | + | 16 | 12 (75) | 16 (100) | 14 (88) | 12 (75) | 10 (63) |
| Totala | − | 48 | 24 (50) | 28 (58) | 16 (33) | 18 (38) | 12 (25) |
| + | 48 | 36 (75) | 48 (100) | 48 (100) | 36 (75) | 30 (63) | |
That is, C. albicans, C. tropicalis, and C. glabrata.
Table 2 shows the DNA purity, the detection limit, the duration of the DNA isolation, and the cost per sample obtained for testing Candida species by various DNA isolation procedures. Several observations can be made. (i) The QIAamp DNA blood kit proved to be as sensitive as the HLGT DNA isolation method. All other procedures yielded DNA isolation results that were 1 to 3 log10 units less sensitive. (ii) The more sensitive methods yielded more pure, high-quality DNA. (iii) The use of commercialized kits did significantly shorten duration of the DNA isolation procedure.
TABLE 2.
Comparison of different DNA isolation methods from sera prepared from simulated blood with Candida DNA
| Method | Reference or source | DNA puritya (range) | Detection limit (CFU/ml)b | Timec (h) | Cost per sample (U.S. dollars) |
|---|---|---|---|---|---|
| PKPC | 5 | 1.13 (1.12-1.60) | 1,000 | 2.75 | 0.61 |
| HLGT | 13 | 1.75 (1.30-1.80) | 10 | 2.50 | 0.52 |
| QIAamp DNA blood kit | Qiagen | 1.66 (1.02-1.70) | 10 | 0.5 | 3.91 |
| HighPure PCR template kit | Roche | 1.02 (1.00-1.12) | 100 | 0.75 | 5.67 |
| DNAzol | Sigma | 1.02 (0.96-1.09) | 10,000 | 1 | 1.72 |
Average DNA purity as determined by UV spectrophotometry.
The number of CFU in the most simulated samples producing a PCR-positive signal in five independent DNA isolations was considered the detection limit.
That is, the mean extraction time for DNA isolation from six samples.
Performance of selected procedures with clinical samples.
By comparing different DNA isolation procedures based on data from simulated sera (Tables 1 and 2), both the QIAamp DNA blood kit and the HLGT method were selected to test their value in the routine clinical laboratory. Fifty-two serum specimens from 39 patients were tested by these procedures, and DNA recovery was measured by using the TaqMan-based PCR assay. Of these, 32 samples (62%) were negative by PCR and by the other tests used (data not shown). The results obtained from 20 samples (38%) found to be PCR-positive for Candida spp. are shown in Table 3: 16 C. albicans, 1 C. parapsilosis, 1 C. tropicalis, and 2 C. glabrata. Concomitant blood cultures were not obtained in nine patients. Among the eight remaining patients, 11 samples were PCR positive and 1 was blood culture negative. In 10 of 11, the same Candida species was isolated from blood cultures (Table 3). The only case of discordant results between PCR and blood culture (patient 5) was observed in a sample that proved to be Candida antigen positive; its C. albicans load was low (70 CFU/ml of blood). Overall, Candida loads ranged between 70 and 1,840 CFU/ml of blood and between 60 and 1,710 CFU/ml of blood when DNA isolation was performed by the HLGT method and the QIAamp DNA blood kit, respectively. Moreover, a 100% correspondence of PCR results between the two procedures was documented. Finally, none of the serum samples from 15 healthy volunteers showed a PCR-positive result for Candida spp., irrespective of the DNA isolation procedure applied. The sensitivity, specificity, and positive and negative predictive values of the TaqMan-based PCR assay compared to those for the blood culture system were 100, 97, 91, and 100%, respectively; compared to those for Candida antigen detection assay, these values were 100, 86, 75, and 100%, respectively.
TABLE 3.
Comparative results of positive TaqMan based PCR assay samples processed by using HLGT and the QIAamp DNA blood kit with blood culture and antigen Candida detection results
| Patient no. | DNA load (102 CFU/ml of blood)a
|
Blood culture result | Candida antigen concn (ng/ml) | Blood originb | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
|
C. albicans
|
C. parapsilosis
|
C. tropicalis
|
C. glabrata
|
||||||||
| HLGT | QIA | HLGT | QIA | HLGT | QIA | HLGT | QIA | ||||
| 1 | -d | - | - | - | - | - | 2.8 | 2.8 | C. glabrata | − | PAC |
| 1 | - | - | - | - | - | - | 10.3 | 10.1 | C. glabrata | 1.220 | PAC |
| 2 | 2.6 | 2.5 | - | - | - | - | - | - | C. albicans | − | PAC |
| 2 | 4 | 4 | - | - | - | - | - | - | C. albicans | 0.300 | PAC |
| 3 | 2.5 | 3.2 | - | - | - | - | - | - | C. albicans | >2 | CVC |
| 3 | 10 | 10 | - | - | - | - | - | - | C. albicans | − | CVC |
| 4 | 3.4 | 4.2 | - | - | - | - | - | - | NDc | >2 | |
| 5 | 0.7 | 0.7 | - | - | - | - | - | - | − | >2 | CVC |
| 6 | - | - | - | - | 13 | 12 | - | - | C. tropicalis | >2 | PAC |
| 7 | - | - | 6.3 | 7.3 | - | - | - | - | C. parapsilosis | − | PAC |
| 8 | 6 | 4 | - | - | - | - | - | - | C. albicans | − | PAC |
| 9 | 3 | 2.7 | - | - | - | - | - | - | C. albicans | 1.300 | PER |
| 10 | 18.4 | 17.1 | - | - | - | - | - | - | ND | 1.081 | |
| 11 | 13.5 | 13.1 | - | - | - | - | - | - | ND | 1.470 | |
| 12 | 13.3 | 10.5 | - | - | - | - | - | - | ND | 0.540 | |
| 13 | 7.7 | 7.6 | - | - | - | - | - | - | ND | 0.860 | |
| 14 | 3 | 4 | - | - | - | - | - | - | ND | 0.600 | |
| 15 | 1.3 | 1.8 | - | - | - | - | - | - | ND | 0.260 | |
| 16 | 1.1 | 1.9 | - | - | - | - | - | - | ND | 0.350 | |
| 17 | 0.7 | 0.6 | - | - | - | - | - | - | ND | 0.290 | |
QIA, QIAamp DNA blood kit.
PAC, Port-a-Cath implantable intramuscular device; PER, peripheral blood; CVC, central venous catheter.
ND, not done.
-, negative result.
The Candida antigen titers as determined by the enzyme immunoassay of 15 of the 20 PCR-positive serum samples ranged from intermediate (0.26 ng/ml) to high (>2 ng/ml). Three Candida antigen-negative samples were from patients 1, 2, and 3. For patient 1, both samples were collected prior to antifungal therapy. Samples from patient 2, who had a catheter infection, were collected during treatment (intravenous fluconazole). In patient 3, the first serum was withdrawn before antifungal treatment and the second was withdrawn after treatment. Using blood culture and PCR assays, the other Candida antigen-negative serum samples obtained from patients 7 and 8 were identified as containing C. parapsilosis and C. albicans, respectively.
DISCUSSION
The aim of this study was mainly to compare different procedures for candidal DNA isolation from serum in order to test their efficiency and practicability in obtaining PCR-compatible material. We addressed the reliability of our TaqMan-based PCR assay (12) for the sensitive detection and identification of most clinically important Candida species involved in systemic candidiasis by including an amplification facilitator to avoid false-negative results. To determine whether PCR amplification would agree with conventional diagnostic and identification methods, selected purification methods suitable for PCR utilization in the routine clinical laboratory were used to process serum specimens of patients with clinically proven or suspected systemic Candida infections.
Neither commercial nor conventional procedures applied for DNA isolation in the present study were sufficient for the removal of PCR-inhibitory compounds from serum prepared from spiked blood with Candida DNA. If 0.5% BSA was not added to the PCR mixture, the rate of positivity was only 25 to 58%. As a consequence of this inhibition, tedious purification steps to generate DNA suitable for PCR amplification analysis can add to the expense of DNA extraction, as well as to the loss of target DNA, and increase sample processing time. In order to overcome the inhibitory effect of clinical samples for PCR testing without introducing practical modifications to DNA isolation procedures, a simple absorption of these inhibitors to BSA offers a potentially valuable means to relieve interference. Our data indicate that, depending on the DNA isolation procedure applied, inhibitory substances whose interference is relieved by BSA will be present at various amounts in DNA extracts. By supplementing PCR mixtures with 0.5% BSA, the inhibitory effects on amplification were either significantly relieved (the PKPC method, HighPure PCR template preparation kit, and DNAzol) or completely eliminated (the HLGT method and QIAamp DNA blood kit).
To date, different procedures have been used to generate candidal template DNA from serum samples for PCR assay (1, 3-5). However, to our knowledge the efficiencies of these procedures applied simultaneously to the same sera sets have not been compared to same study. For instance, approaches with the classic organic extraction have been applied successfully for DNA isolation from fungal pathogens. Chryssanthou et al. (5) utilized proteinase K and sodium dodecyl sulfate for protein digestion and phenol-chloroform for DNA purification. Sandhu et al. (13) used chaotropic agents, such as guanidine thiocyanate, followed by DNA purification with chloroform-isopropanol. Although these conventional procedures can be applied to other types of clinical specimens and the reagents are inexpensive in comparison to the kits, these methods are impractical for the processing of large numbers of samples that arrive in the laboratory at different times. In our comparison study, even when awkward chemicals were used, these conventional procedures did not prove to be superior to the QIAamp DNA blood kit when they were applied to clinical specimens. All other kit-form DNA isolation systems tested failed to detect relatively low amounts of candidal DNA in our experimental settings. The observed sensitivity of QIAamp DNA blood kit (101 CFU/ml of blood), which is a prerequisite for its application in the routine clinical laboratory is very encouraging. This kit offers also other needed characteristics: it is simple, reliable, and reproducible. Moreover, the QIAamp DNA blood kit is suitable for TaqMan-based PCR assay utilization in the clinical laboratory because it provides a fast pretreatment procedure for isolating candidal DNA from serum samples. The complete time required for Candida species detection and identification is less than 3 h: 0.5 h for DNA isolation and ca. 2.0 h for PCR amplification and analysis of data. All of these characteristics make the QIAamp DNA blood kit an ideal candidate for use as a routine diagnostic test.
Two groups have also recommended the QIAamp DNA blood kit for DNA isolation from serum samples (6, 16). Dixon et al. (6), in investigating a variety of methods for the extraction of genomic DNA from serum samples from prostate cancer patients, found that the QIAamp DNA blood kit reliably yielded pure, high-quality DNA for PCR. Wahyuningsih et al. (16) used this kit to purify C. albicans DNA from serum specimens. QIAamp DNA blood kit proved to be a fast and simple method for isolating minute amounts of DNA for the diagnosis of invasive candidiasis, when followed by PCR amplification and a microtitration plate enzyme immunosorbent assay. Regarding the cost per assay, QIAamp DNA blood kit proved to be among the most expensive procedures. However, the cost significantly varies from study to study. For Dixon et al. (6), this kit is twice as expensive as other DNA isolation procedures. Löeffler et al. (10), in a comparative study for extraction of DNA of C. albicans and Aspergillus niger cells from blood specimens, have found that QIAamp tissue kit is about 23 times more expensive than an in-house DNA isolation method. In our experimental settings, we have used large volume of serum (400 μl) in order to increase the sensitivity of the TaqMan-based PCR assay. Therefore, a higher cost representing 7.5 times the cost of the conventional procedure was calculated. Nevertheless, if technician remunerations can be also calculated to include extract sample processing cost, the total cost of DNA isolation by using QIAamp DNA blood kit will be more advantageous than conventional procedures.
The QIAamp DNA blood kit and the conventional HLGT method were subsequently applied to 52 serum samples from hematology or oncology patients. A 100% correspondence of PCR results could be demonstrated. Interestingly, all PCR-positive samples showed approximately the same Candida species load by both procedures. No Candida DNA could be detected by TaqMan-based PCR assay in serum samples from healthy volunteers, irrespective of the DNA isolation procedure applied. On the other hand, compared to blood culture results, the sensitivity and specificity of the TaqMan-based PCR assay in serum were 100 and 97%, respectively. However, the sole serum specimen, which was blood culture negative and Candida antigen positive, showed identical C. albicans loads by TaqMan-based PCR assay when DNA isolation was performed either by the QIAamp DNA blood kit or the HLGT method (70 CFU/ml of blood). The reason for the negative result of this blood culture from a suspected patient of candidemia is not understood, however, probably because of the lack of terminal subcultures of blood culture bottles without detected growth, the BACTEC system may lack detection of an important percentage of culture bottles even when the yeasts remain viable in the culture medium (8). This is also consistent with previous studies where PCR has been reported to be more sensitive than blood culture in the diagnosis of candidemia (1, 5, 9). The specificity and positive predictive values were better when the TaqMan-based PCR assay was compared to the blood culture than with Candida antigen detection, but it should be considered the lower number of comparable data (nine patients without blood cultures available). In summary, our results demonstrated that the QIAamp DNA blood kit was the sole commercially available assay tested that yielded the same sensitivity and purity of fungal DNA isolation from serum as the HLTG method. Moreover, besides being rapid, the QIAamp DNA blood kit, combined with the real-time quantitative TaqMan-based PCR assay, provides better reliability and safety than the conventional DNA isolation procedures previously used.
REFERENCES
- 1.Ahmad, S., Z. Khan, S. Mustafa, and U. Khan. 2002. Seminested PCR for diagnosis of candidemia: comparison with culture, antigen detection, and biochemical methods for species identification. J. Clin. Microbiol. 40:2483-2489. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Ascioglu, S., J. H. Rex, B de Pauw, J. E. Bennett, J. Bille, F. Crokaert, D. W. Denning, J. P. Donnelly, J. E. Edwards, Z. Erjavec, D. Fiere, O. Lortholary, J. Maertens, J. F. Meis, T. F. Patterson, J. Ritter, D. Selleslag, P. M. Shah, D. A. Stevens, and T. J. Walsh. 2002. Defining opportunistic invasive fungal infections in immunocompromised patients with cancer and hematopoietic stem cell transplants: an international consensus. Clin. Infect. Dis. 34:7-14. [DOI] [PubMed] [Google Scholar]
- 3.Bougnoux, M.-E., C. Dupont, J. Mateo, P. Saulnier, V. Faivre, D. Payen, and M. H. Nicolas-Chanoine. 1999. Serum is more suitable than whole blood for diagnosis of systemic candidiasis by nested PCR. J. Clin. Microbiol. 37:925-930. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Burnie, J. P., N. Golbang, and R. C. Matthews. 1997. Semiquantitative polymerase chain reaction enzyme immunoassay for diagnosis of disseminated candidiasis. Eur. J. Clin. Microbiol. Infect. Dis. 16:346-350. [DOI] [PubMed] [Google Scholar]
- 5.Chryssanthou, E., B. Anderson, B. Petrini, S. Löfdahl, and J. Tollemar. 1994. Detection of Candida albicans DNA in serum by polymerase chain reaction. Scand. J. Infect. Dis. 26:479-485. [DOI] [PubMed] [Google Scholar]
- 6.Dixon, S. C., J. Horti, Y. Guo, E. Reed, and W. D. Figg. 1998. Methods for extracting and amplifying genomic DNA isolated from frozen serum. Nat. Biotechnol. 16:91-94. [DOI] [PubMed] [Google Scholar]
- 7.Flahaux, M., D. Sanglard, M. Monod, J. Bille, and M. Rossier. 1998. Rapid detection of Candida albicans in clinical samples by DNA amplification of common regions from C. albicans-secreted aspartic proteinase genes. J. Clin. Microbiol. 36:395-401. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Horvath, L. L., D. R. Hospenthal, C. K. Murray, and D. P. Dooley. 2003. Detection of simulated candidemia by the BACTEC 9240 system with Plus aerobic/F and anaerobic/F blood culture bottles. J. Clin. Microbiol. 41:4714-4717. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Khan, Z. U., and A. S. Mustafa. 2001. Detection of Candida species by polymerase chain reaction (PCR) in blood samples of experimentally infected mice and patients with suspected candidemia. Microbiol. Res. 156:95-102. [DOI] [PubMed] [Google Scholar]
- 10.Löeffler, J., H. Hebart, U. Schumacher, H. Reitze, and H. Einsele. 1997. Comparison of different methods for extraction of DNA of fungal pathogens from cultures and blood. J. Clin. Microbiol. 35:3311-3312. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Löeffler, J., N. Henke, H. Hebart, D. Schmidt, L. Hagmeyer, U. Schumacher, and H. Einsele. 2000. Quantification of fungal DNA by using fluorescence resonance energy transfer and the LightCycler system. J. Clin. Microbiol. 38:586-590. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Maaroufi, Y., C. Heymans, J.-M. De Bruyne, V. Duchateau, H. Rodriguez-Villalobos, M. Aoun, and F. Crokaert. 2003. Rapid detection of Candida albicans in clinical blood samples by using a TaqMan-based PCR assay. J. Clin. Microbiol. 41:3293-3298. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Sandhu, G. S., B. C. Kline, L. Stockman, and G. D. Roberts. 1995. Molecular probes for diagnosis of fungal infections. J. Clin. Microbiol. 33:2913-2919. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Shin, J. H., F. S. Nolte, B. P. Holloway, and C. J. Morrison. 1999. Rapid identification of up to three Candida species in a single reaction tube by a 5′ exonuclease assay using fluorescent DNA probes. J. Clin. Microbiol. 37:165-170. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Turenne, C. Y., S. E. Sanche, D. J. Hoban, J. A. Karlowsky, and A. M. Kabani. 1999. Rapid identification of fungi by using the ITS2 genetic region and an automated fluorescent capillary electrophoresis system. J. Clin. Microbiol. 37:1846-1851. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Wahyuningsih, R., H.-J. Freisleben, H.-G. Sonntag, and P. Schnitzler. 2000. Simple and rapid detection of Candida albicans DNA in serum by PCR for diagnosis of invasive candidiasis. J. Clin. Microbiol. 38:3016-3021. [DOI] [PMC free article] [PubMed] [Google Scholar]