Utility of blood as the clinical specimen for the diagnosis of ocular toxoplasmosis using uracil DNA glycosylase-supplemented loop-mediated isothermal amplification and real-time polymerase chain reaction assays based on REP-529 sequence and B1 gene

This cross-sectional study was designed using 117 immunocompetent patients with clinically diagnosed OT in Farabi Eye Hospital, Tehran, Iran, from October 2015 to August 2017. The OT was clinically diagnosed by a retina specialist (MZ). Chorioretinal involvements of T. gondii were classified into three major groups based on fundoscopic examination, including (1) Group A (active lesion): an area of white edematous retina with ill-defined borders and overlying vitritis, usually accompanied by nearby vasculitis (Fig. 1A). This appearance represents the active form of disease and when seen with no scars is considered as a consequence of the primary acquired infection [20]; (2) Group B (old scar): a chorioretinal scar with sharp borders and variable degrees of dark pigmentation (Fig. 1B). This appearance represents the inactive form of disease and is considered as a reminiscence of the active prior chorioretinitis [21, 22] and (3) Group C (reactivated disease): a combination of the two forms, including an area of active chorioretinitis in eyes with toxoplasmic chorioretinal scars, usually close to an active chorioretinitis focus (Fig. 1C). This form may be resulted from the failure of immune system to continually limit the infected foci and subsequent release of the tissue cyst content into the close tissues [23]. Moreover, 200 patients with other eye diseases other than toxoplasmosis, including viral keratitis (39 samples), amebic keratitis (38 samples), fungal keratitis (42 samples), bacterial endophthalmitis (64 samples) and cataract (17 samples) were participated as controls. Blood samples (up to 3.5 mL) were collected from the participants using EDTA-containing tubes. These blood samples have been verified using various biological techniques by the authors in previous studies [24, 25].

DNA was extracted from the whole blood using QIAamp Genomic DNA Blood Mini Extraction Kit (Qiagen, Hilden, Germany) based on the manufacturer instructions and kept frozen at − 20 °C for further use in qPCR and UDG-LAMP reactions [25]. The reference DNA was extracted from a virulent RH strain of T. gondii (Type I), which was previously collected from peritoneal cavity of the infected mice [26].

A TaqMan-probe qPCR targeting REP-529 and B1 was developed using StepOne Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) with 45 cycles of amplification. REP-529 is a highly repetitive sequence with 200–300 copies in T. gondii genome [27]. The B1 gene has 35 copies in the genome and is conserved indifferent parasite strains [28]. The PCR reaction included RealQ Plus Master Mix for probes labeled with reporter fluorophor 6-carboxyfluorescein (Ampliqon, Odense, Denmark) (10 µL), TagMan probe (FAM-CCCTCGCCCTCTTCTCCACTCTTCAA-3-BHQ1) (0.2 µM), extracted DNA (5 µL), forward (5ʹ-CTTCGTCCAAGCCTCCGA-3ʹ) and reverse (5ʹ-GACGCTTTCCTCGTG GTGAT-3ʹ) primers (0.4 µL) and distilled water (4 µL). The PCR thermal cycling was carried out at 95 °C for 5 min and continued for 40 cycles of 95 °C for 1 min, annealing at 55 °C for 1 min and extension at 72 °C for 1 min. The gapdh gene was used as housekeeping gene with primers common to all mammalian species to check the quality of DNA [18]. An internal control (TaqMan exogenous positive control) was used, and each run was considered valid if the positive, negative, and internal controls were acceptable. The result was considered negative if repeated atypical amplification curves with proper amplification of the internal control was present. A sample was considered inhibited if amplification of the internal control failed and there was no amplification or atypical amplification curve for the target of interest.

UDG-LAMP assay was successfully performed in our laboratory as described previously [29]. Specific oligonucleotide primers for T. gondii used for the LAMP assay were designed based on REP-529 and B1 regions of the parasite genome (Table 1). The total reaction master mix volume was 25  μL consisting of 1 μL of template DNA, 40 picomol of each of FIP and BIP primers, 20 pmol of each LF and LB primers (used only in REP-LAMP), 5 pmol of each of F3 and B3 primers, 8 U of Bst2 DNA polymerase (New England Biolabs, USA), 1.4 mmol/L of deoxyuridine triphosphates (dUTP) instead of dTTP and 2X reaction buffer, containing 1.6 mol/L betaine (Sigma Aldrich, USA), 40 mmol/L Tris–HCl (pH 8.8), 20 mmol/L of KCL, 20 mmol/L of (NH4)2SO4, 16 mmol/L of MgSO4 and 0.2% tween 20). The LAMP assay were carried out at 60–67 °C for 30, 45, 60 and 75 min to find the optimum time and temperature conditions and finally; 63 °C and 60 min were the best temperature and time for both target genes, respectively. Addition of 3 µL of the fluorescent detection reagent, diluted SYBR Green I (Invitrogen, Carlsbad, California, USA), to the reaction tube after LAMP reaction enables visible analysis of the results under daylight and/or UV light. The positive amplification was read through observation of change in color of the reaction mixture following addition of dye to the tube and the color changes from orange (negative reaction) to green (positive reaction) (Fig. 2). The results were further verified by electrophoresis of LAMP products in 1.5% agarose gel electrophoresis stained with DNA Safe Stain (SinaClon, Iran), and visualized under UV light. To prevent self-amplification of the primers, reactions were carried out in the absence of DNA templates. The T. gondii RH-strain DNA and double distilled water were used as positive and no template controls, respectively. To assess reproducibility of the assay, all experiments were carried out in duplicate. The protocol is fully described in a previous study by the authors [29, 30].

To assess analytical sensitivity of the qPCR and UDG-LAMP techniques for the detection of T. gondii (RH strain) DNA, tenfold serial dilutions of T. gondii DNA with 1 pg to 0.01 fg concentrations were prepared and used in qPCR and LAMP assays targeting REP-529 and B1. The construction is performed with various concentrations of DNA. Next, the cycle threshold (CT) values were plotted as mean (triplicate) against the standard curve values to determine the detection limit of both primer sets. Parasite concentrations are determined after the calculation of the linear regression equation (y = ax + b), where y = CT; a = curve slope (slope); x = parasite number; and b = where the curve intersects y-axis (y intercept) [31, 32].

To assess the analytical specificity of the qPCR and UDG-LAMP techniques, extracted DNAs from other microorganisms, including Leishmania tropica, L. major, Giardia lamblia, Cryptosporidium parvum, Entamoeba dispar, Trichostrongylus colubriformis, Corynosoma capsicum, Candida albicans and Acanthamoeba spp. as well as human chromosomal DNA, were used as templates in qPCR and LAMP assays using specific primers of REP-529 and B1.

Data were analyzed using SPSS Software v.21 (SPSS Inc. Chicago, IL, USA). The χ² test was used to calculate diagnostic sensitivity and specificity of the qPCR and UDG-LAMP techniques. Agreements between the clinical and molecular assays were demonstrated using Cohen’s kappa coefficient [33]. Linear regressions were constructed from the standard curves for REP-529 and B1 and CT mean comparison of both primer pairs. Both curves were statistically analyzed using an unequal-variance t-test based on a critical value of p ≤ 0.05.

Of the total samples, 55.55% (65/117) belonged to males and 44.44% (52/117) to females. Thirty six (30.76%) patients were up to 10 years old, 50 (42.73%) patients were 11–20 years old, 23 (19.65%) patients were 21–30 years old and eight (6.83%) patients were 31–40 years old. Of 117 patients, 37 (31.62%) patients included active lesions (Group A), 13 (11.11%) patients included pigmented scars (Group B) and 67 (57.26%) patients included pigmented scars with active lesions (Group C).

Out of 37 patients in Group A (patients with active lesions), 18 and 16 patients were positive by qPCR using REP-529 and B1 and 15 and 14 patients were positive by UDG-LAMP using REP-529 and B1, respectively. All positive results using UDG-LAMP had 100% overlaps with positive results of qPCR using REP-529 and B1. None of the patients in other clinical groups (Groups B and C) were positive using molecular assays (Table 2). All patients in control group were negative for REP-529 and B1 using qPCR and UDG-LAMP. Amplification plot and derivative melt curve analysis of the T. gondii based on REP-529 and B1 targets are shown in Fig. 3. The results of chi-square test indicated that qPCR method based on REP-529 was the most sensitive than UDG-LAMP using REP-529 and B1 (P = 0.003).

The first step was to assess the reportable range of the reaction that was determined utilizing T. gondii DNA isolate from tachyzoites (RH strain) using the REP-529 (Fig. 4A) and B1 (Fig. 4B). The current results showed that the detection limits for UDG-LAMP using REP-529 and B1 were 1 and 100 fg, respectively. Detection limits of qPCR using REP-529 and B1 were estimated as 0.1 and 1 fg of T. gondii genomic DNA, respectively (Figs. 4, 5).

Using qPCR and UDG-LAMP, no amplifications were detected in reactions with DNA of G. lamblia, C. parvum, E. dispar, L. major, Acanthamoeba spp., T. colubriformis, C. capsicum and C. albicans as well as human genomic DNA (Fig. 6).