Evaluation for the Clinical Diagnosis of Pythium insidiosum Using a Single-Tube Nested PCR

Pythiosis is a rare infectious disease caused by Pythium insidiosum, which typically occurs in tropical and subtropical regions. The high mortality rate may be in consequence of the lack of diagnosis. The objective of this study was to evaluate reliability of a new single-tube nested PCR for detection of P. insidiosum DNA. A total of 78 clinical isolates of various fungi and bacteria, 106 clinical specimens and 80 simulated positive blood samples were tested. The developed primer pairs CPL6–CPR8 and YTL1–YTR1 are located on 18S subunit of the rRNA gene of P. insidiosum. The specificity, negative and positive predictive values were 100, 100 and 87.5 %, respectively, as compared with direct microscopy and cultivation. The detection limit of the single-tube nested PCR was 21 zoospores corresponding to 2.7 pg of the DNA. The results demonstrate that the new single-tube nested PCR offers a highly sensitive, specific and rapid genetic method for detecting P. insidiosum.


Introduction
Pythiosis is an emerging and life-threatening infectious disease caused by the oomycete Pythium insidiosum, which can develop in humans and animals [1][2][3][4]. The infective stage of P. insidiosum is a biflagellate zoospore [1]. Human pythiosis occurs predominantly in tropical areas of the world and particularly in Thailand [1,5,6]. There, the first documented case of human pythiosis was reported in 1985 [7]. According to clinical signs, human pythiosis can be divided into four types: cutaneous/subcutaneous, vascular, ocular and disseminated. The first type is characterized by chronic swelling, painful subcutaneous granulomatous infiltration and ulceration, usually located in the face or legs [8]. Chronic arthritis in the lower extremities resulting in arterial occlusion and gangrenous ulceration of feet or legs is typical for vascular pythiosis [9]. The ocular form is usually manifested as corneal ulcers or keratitis. As a result of all these forms of infection, P. insidiosum can spread via the bloodstream to various internal organs or organ systems such as the gastrointestinal tract, brain, liver, kidney or rhinosinus [10]. Currently, the diagnosis of pythiosis is based on microscopy, culture, detection of antibodies and molecular genetic techniques [9,[11][12][13]. However, microscopy cannot distinguish zygomycetes because of the coenocytic form of the mycelium [7]. Culture is time-consuming, and obtaining infected tissue samples may be difficult [1]. Because of low antibody response, false-negative results frequently occur in serological tests, particularly in ocular pythiosis [11]. Nested PCR has been developed for the diagnosis of pythiosis using the internal transcribed spacer 1 (ITS1) of the gene for rRNA [14]. Although it is highly sensitive, the main problem of this PCR is a high risk of contamination as the product of the first reaction needs to be transferred into another tube for the second reaction. The purpose of this study was to solve this problem by developing a nested PCR for detecting P. insidiosum in a single tube and to evaluate its reliability using various clinical specimens, including simulated positive blood samples and clinical isolates of bacteria and fungi.

Evaluation by Phenotypic Methods
All clinical specimens were evaluated microscopically in 20 % potassium hydroxide preparation for the presence of P. insidiosum hyphae. Culture was performed, with each specimen being inoculated on two Sabouraud dextrose agars (SDA; Oxoid, UK), two Mycosel agars (MCA; BD Diagnostics) and one blood agar (Oxoid, UK) for detection of P. insidiosum growth. One SDA and MCA each were incubated at 25°C and the other media at 37°C. All agars were evaluated for the P. insidiosum growth until 30 days. The suspected colonies were identified as P. insidiosum by induction of zoospores [15]. Results of these phenotypic methods were then compared with a single-tube nested PCR for sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV).
DNA Extraction for the PCR DNA from all clinical and simulated positive specimens was extracted with the NucleoSpin Tissue kit  [16]. Purity of isolated DNA was calculated using a spectrophotometer at OD 260 and 280.

Single-Tube Nested PCR
The single-tube nested PCR was performed using outer primers CPL6 ( Amplification was performed in a thermal cycler (Mastercycler Personal, Eppendorf International) using the following temperature cycles: initial denaturation at 95°C for 5 min, followed by 30 cycles (denaturation at 95°C for 1 min, annealing at 68°C for 1 min and extension at 72°C for 1 min), another 30 cycles (denaturation at 95°C for 1 min, annealing at 57°C for 1 min and extension at 72°C for 1 min) and final extension for 10 min at 72°C. Amplicons were visualized under UV light using a manual documentation system (InGenius3, Syngene, USA) after electrophoresis of 10 ll of the reaction solution in 2 % UltraClean agarose gel (Mo Bio Laboratories) containing the SYBR Gold nucleic acid gel stain (Invitrogen).

Evaluation of PCR Sensitivity
To determine the lowest detection limit of our nested PCR, seven 10-fold dilutions were prepared from simulated positive blood specimen. Then, 180 ll of each dilution containing 2.07 9 10 5 -2.07 9 10 -1 zoospores was used for nested PCR. As a control, DNA of P. insidiosum hyphae was also prepared as 10-fold serial dilution from 1 ng to 1 fg, and then, it was also amplified by a single-tube nested PCR. This study was approved by the Ethical Committee on Human Experimentation in Khon Kaen University in accordance with the Declaration of Helsinki (HE 541111).

Results
If P. insidiosum DNA was presented in clinical material, four amplicons sized 512, 452, 340 and 240 bp were detected by our nested PCR.

Evaluation of PCR Specificity Within Clinical Isolates
Four amplicons were detected in all 34 P. insidiosum isolates. Only one genus-specific product (512 bp) was found in five non-insidiosum Pythium spp. as demonstrated in Fig. 2. No product was found after amplification of DNA from all the remaining 29 fungal and 10 bacterial isolates. Therefore, our single-tube nested PCR had 100 % specificity within the group of clinical microorganisms tested in this study.

Evaluation of PCR Specificity in Clinical Specimens
Based on positivity of phenotypic methods, seven pus specimens from corneal ulcers were evaluated as positive within all tested clinical specimens. From that, three specimens were microscopically negative; culture and zoospore induction were positive in all seven specimens. Moreover, as demonstrated in Table 2, one more pus specimen, also from a corneal ulcer, which was negative in both phenotypic methods, was positive in our single-tube nested PCR.

Evaluation of PCR Sensitivity
As shown in Fig. 3, the lowest detection limit in diluted simulated positive blood specimen was 21 zoospores, which corresponded to 2.7 pg of DNA. The lowest detection limit of serially diluted DNA isolated from P. insidiosum hyphae was 1 pg.

Evaluation of PPV and NPV of PCR in Clinical Specimens
These results are summarized in Table 3. Based on results of phenotypic methods, 7 specimens were evaluated as positive for P. insidiosum. Using our single-tube nested PCR, DNA of P. insidiosum was detected in 8 specimens, whereas 7 of them were positive by phenotypic methods. Therefore, NPV and PPV were 100 and 87.5 %, respectively.

Discussion
In our set of clinical specimens, one patient was found to be positive in single-tube nested PCR, but negative in phenotypic tests. In our study, the criterion for detecting pythiosis was positivity in one of two phenotypic methods. However, Krajaejun et al. [9] defined the criteria for pythiosis as culture positivity, seropositivity or clinical manifestation. The patient who had negative results of phenotypic tests and was  positive in nested PCR had symptoms of ocular pythiosis evaluated by a clinician. The reason why phenotypic methods could not confirm pythiosis in this case was probably their low sensitivity [17]. So, this finding demonstrates the superiority of PCR over phenotypic methods from the sensitivity point of view. The detection limit of our nested PCR in this study (2.7 pg of DNA in relation to 21 counted zoospores) allows us to suppose that one zoospore contains 128.57 fg of DNA. However, our experiments showed that the lowest detected amount of DNA from P. insidiosum isolated from its mycelium was 1 pg of DNA, corresponding to about eight zoospores. According to this detection limit, one zoospore could contain 48-129 fg of DNA. This is the first report which tried to estimate the amount of DNA in P. insidiosum zoospores. Hussain et al. [18] reported a detection limit of PCR for Phytophthora infestans of 0.5 pg, which corresponded to four zoospores; in that case, one zoospore should contain 125 fg of DNA. The DNA content is similar to data about plant pathogenic organisms from the order Peronosporales (Oomycota) ranging from 46 to 163 fg [19]. The newly designed, genus-specific outer set of primers CPL6 and CPR8 amplified a 512 bp fragment from a conserved part of the 18S subunit of the rRNA gene and species-specific inner primers YTL1 and YTR1 amplified a 240 bp fragment from variable region of this subunit were firstly evaluated in this study. Although finally four PCR products were seen in a single-tube nested PCR, the test showed excellent sensitivity and specificity. The fragment of 452 bp was the product of CPL6 and YTR1 primers, and the 300 bp fragment was amplified by the YTL1 and CPR8 primers. Four PCR products were detected in all 34 strains of P. insidiosum and in clinical samples from patients with pythiosis. A sensitivity of 100 % and a low detection limit allow us to recommend our single-tube nested PCR as a   useful tool for laboratories, which need to detect P. insidiosum in clinical material. Pythiosis is a rare but fatal infection in humans. Krajaejun et al. [8] reported 102 cases of human pythiosis in Thailand from 1985 to 2003. The zoospore is the infective stage of P. insidiosum; therefore, the selected spiking of blood with zoospores simulates the state in really infected blood better than spiking it with pure DNA. The sensitivity of our nested PCR tested on 78 various clinical isolates and 186 clinical specimens including simulated positive blood samples was higher than with other tests [20][21][22]. In addition, nested PCR solves the main drawbacks of other methods for detecting pythiosis. It overcomes problems with culture and allows identification in a substantially shorter time directly in clinical specimens. Thus, the prognosis of patients can be improved because treatment can be started earlier. Moreover, performing nested PCR in a single tube markedly reduces the risk of crosscontamination. Results of this study demonstrate that our single-tube nested PCR is suitable for detecting P. insidiosum directly from clinical specimens and identification of clinical isolates as well. As this method is rapid, highly sensitive and has minimal risk of cross-contamination, it has a great potential in the diagnosis of suspected cases of pythiosis.