Background
Toxoplasmosis is an infectious disease caused by the protozoan
Toxoplasma gondii effecting individuals throughout the world. Two main subpopulations are highly susceptible to this parasite: the fetus and an immunocompromised individual. Congenital infection may induce spontaneous abortion or serious sequelae when maternal infection occurs during pregnancy. Both the damage for the fetus (the sooner, the more deleterious) and the frequency of trans-placental transmission (the later, the more frequent) is usually depending on the stage of gestation. In the immunocompromised host, the prognosis for cerebral disseminated toxoplasmosis is poor, with a mortality rate of 63% [
1]. For these two subpopulations, a rapid and accurate diagnosis is required to initiate treatment. Diagnosis with radiological findings, histology, tissue culture, or inoculation into mice is difficult, time consuming, or impractical. Since anti-toxoplasma antibody titers are often unchanged or decreased at the onset of clinical symptoms, and most cases of active toxoplasmosis are due to reactivation of latent infection, the serological status is solely useful in order to know whether the patient is at risk for reactivation. This is why the direct demonstration of the parasite in tissues or other biological fluids by PCR is a major breakthrough for the diagnosis of toxoplasmosis in these patients [
2,
3].
The reliability of the detection of
T. gondii DNA in amniotic fluids (AF) or blood is of utmost importance. Real-time PCR assays have recently emerged as a dramatic improvement in the reliability of PCR assays. Since the reaction tubes need not be opened after amplification, avoiding potential contamination of the environment with amplicons, the risk of false positive results are dramatically reduced. Quantitative PCR also provides additional data to direct the choice of specific treatments [
4].
Real-time PCR also provides the opportunity to compare the sensitivity of different DNA targets. Several studies report a relatively high level of false negative results in prenatal diagnosis [
5] or huge differences between immunocompromised patients using current PCR assays [
6,
7]. This may be due to pathophysiological reasons, explaining the absence or intermittent presence of the parasite in the specimens tested, or by poor performance of the PCR assays. The use of more repetitive but not polymorphic DNA targets may help to resolve this issue.
We have developed a real-time PCR assay targeting a recently discovered repetitive 529-bp DNA fragment in
T. gondii [
8], and compared the results to our previously developed PCR assay targeted to the multicopy gene B1 [
9]. Whereas the number of B1 gene copies has been estimated around 35, the number of repeats of the 529-bp DNA fragment is reported to be between 230 and 330 [
8]. We compared the sensitivity our two LightCycler (LC) PCR assays with serial dilutions of
T. gondii DNA and subsequently analyzed amniotic fluids to compare the efficiency of the two PCR protocols.
Methods
T. gondiiisolates, clinical samples and bacterial strains
Five reference strains of
T. gondii, including the RH strain, two of zymodeme 1, two of zymodeme 2, one of zymodeme 3 [
10], and seven strains derived from human amniotic fluids (AF), were selected from the culture collections of our institutions. DNA preparations from 51 amniotic fluid samples, which were received by the Hôpital américain de Paris between 1996 to 2001 and tested
T. gondii positive using a previously published PCR assay [
11], were used as PCR templates. In addition, 160 clinical specimens of various types originating from patients without a suspected
T. gondii infection (about 50 % of these patients were seropositive for
Toxoplasma but do not show clinical symptoms of acute disease), and 118 strains from different bacterial genera, were included in the study.
Template DNA preparation
Genomic DNA of
T. gondii strains was prepared as desribed previously [
11]. Template DNA from clinical specimens was prepared using the High Pure PCR Template Preparation kit (Roche Diagnostics, Mannheim, Germany) according to the manufacturer's instructions. Following the centrifugation and washing steps, total DNA was eluted from the spin columns with 100 μl of elution buffer, and 5 μl aliquots were directly transferred to a PCR reaction.
Nucleotide sequence analysis of the cryptic T. gondiitarget
To identify potential sequence variations within the target sequence among different T. gondii strains, or among different copies of the gene within the chromosome of one organism, total genomic DNA of 5 reference strains and 7 clinical isolates were amplified using primer Tox-8 (5'-CCC AGC TGC GTC TGT CGG GAT-3') and primer Tox-5 (5'-GAC GTC TGT GTC ACG TAG ACC TAA G-3'). Specific amplification products of 450-bp were purified using the HighPure PCR Product Purification kit (Roche Diagnostics), and cycle sequencing reactions were performed as described in the PRISM Ready Reaction Dye Deoxy Terminator cycle sequencing kit protocol (Applied Biosystems, Weiterstadt, Gemany). Both strands of amplicons originating from different amplification reactions were sequenced in duplicate to rule out the possibility of Taq DNA polymerase-induced errors. Optimal results were obtained using amplification primers Tox-8 and Tox-5 as forward and reverse sequencing primers, respectively. The fluorescent-labeled reaction products were analyzed with a ABI PRISM 310 Genetic Analyzer (Applied Biosystems).
Primer and hybridization probe design
One LC-PCR assay, which was described in detail previously, targeted the B1 gene of
T. gondii [
4,
11]. The other LC-PCR assay targeted a recently described
T. gondii 529-bp DNA repeat element of cryptic function. Based on the newly determined consensus sequence GenBank AF487550, two primer oligonucleotides were selected flanking a conserved 162-bp region within the
T. gondii-specific repeat element. For the sequence-specific detection of the corresponding amplicons, a pair of LightCycler hybridization probes were designed against sequences internal to the amplified region. The nucleotide sequences of primers and hybridization probes and their corresponding locations within GenBank AF487550 are shown in Table
1.
Table 1
Oligonucleotide primers and LightCycler hybridization probes used in the PCR assay.
Tox-9 | AGG AGA GAT ATC AGG ACT GTA G | Cryptic | 143–164 | AF487550 | Present study |
Tox-11 | GCG TCG TCT CGT CTA GAT CG | Cryptic | 304–285 | AF487550 | Present study |
Tox-HP-1 | GAG TCG GAG AGG GAG AAG ATG TT-[FL] | Cryptic | 214–236 | AF487550 | Present study |
Tox-HP-2 | [Red 640]-CCG GCT TGG CTG CTT TTC CTG-Ph | Cryptic | 238–258 | AF487550 | Present study |
LC-PCR assays and product detection
DNA oligonucleotide primers and hybridization probes were synthesized by TIB Molbiol, Berlin, Germany. All LC-PCR assays were performed using a fluorescence detecting temperature cycler (LightCycler; Roche Diagnostics). The amplification mixture consisted of 2 μl of 10 X reaction mix (LightCycler FastStart Master Hybridization Probes, Roche Diagnostics), 4 mM MgCl2, 0.5 μM of each oligonucleotide primer, 0.25 μM of each oligonucleotide probe, and 5 μl of template DNA in a final volume of 20 μl. Carry-over was prevented by using the heat-labile uracyl-DNA-glycosylase (UNG) (Roche Diagnostics). The reaction mixture was initially incubated for 1 min at room temperature to allow the UNG to act. This incubation was followed by a 10-min incubation step at 95°C to denature the template DNA, to inactivate the UNG enzyme, and to activate the Fast Start Taq DNA polymerase. Samples were amplified as follows: 50 cycles of denaturation at 95°C for 10 sec, annealing at 56°C for 20 sec, and an extension at 72°C for 20 sec. The temperature transition rate was 20°C/sec.
The generation of target amplicons for each sample was monitored between the annealing and the elongation steps at 640 nm. Samples positive for target genes were identified by the instrument at the cycle number where the fluorescence attributable to the target sequences exceeded that measured for background. Those scored as positive by the instrument were confirmed by visual inspection of the graphical plot (cycle number versus fluorescence value) generated by the instrument.
Quantification of LightCycler products
The quantitative interpretation of LightCycler results was assisted by the "fit point method" algorithm with the "minimize error" option (LightCycler software Vers. 3.5; Roche Diagnostics). The LightCycler software performs all additional calculation steps necessary for generation of a standard curve. Briefly, a noise band (treshold) is automatically set at a fluorescence level, at which the fluorescence signal development reflects that the PCR is in the log-linear phase. The software then calculates the logarithmic values by interpolating a straight line through 2 data points above the treshold value, and the points of intersection (crossing points) with the noise band are determined. Finally, the crossing points are plotted against the logarithm of the target DNA concentration in individual samples.
Sensitivity of the LC-PCR assays for T. gondiidetection
To assess the the analytical sensitivity of the PCR assays for the two different target genes, 10-fold serial dilutions of T. gondii DNA (RH strain) were prepared, ranging from 2 ng to 2 fg input per 20 μl PCR reaction. On the assumption that one 80-Mbp genome equivalent of T. gondii equals about 80 fg, this would be 2.5 × 104 down to 2.5 × 10-2 parasite equivalents. The dilution series were tested in triplicate with the two LC-PCR assays to determine the minimum amount of template DNA per reaction that could be detected by each protocol.
Discussion
As for all parasitic diseases, the PCR diagnosis of toxoplasmosis is not standardized. It seems highly probable that there will be a consensus on real-time PCR assays in the next few years [
4,
11,
12]. However, the consensus on the best sequence to be amplified will be more difficult. The findings of our study clearly confirm that the recently described 529-bp DNA fragment is present in much higher copy number than the 35-fold repeated B1 gene [
8], and is highly conserved on a nucleotide sequence level between strains and isolates. Therefore, this DNA target offers an improvement in the sensitivity of PCR tests for the detection of
T. gondii DNA.
Three repeated DNA sequences have been studied so far for the diagnosis of toxoplasmosis. The rRNA genes have been the basis for several independent PCR assays [
6,
13,
14] but the homogeneity of the amplified fragments has never been demonstrated to a sufficient extent. That may explain the observation of Jones et al. [
15], who found the 35-fold repeated B1 target more sensitive than the ribosomal genes [
14]. The TGR1
E is a non-coding repetitive element of
T. gondii, evaluated as a diagnostic target for PCR [
16,
17], but abandoned because of sequence heterogeneity among different copies of the repeated elements and the impossibility to find suitable PCR primers for amplification [
18]. Up till now, the most popular DNA target for diagnosing toxoplasmosis was the B1 gene, with a huge diversity of published primers and PCR protocols [
3,
6,
11,
19‐
22]. Altough the heterogeneity of the B1 repeats is well known [
23], it seems to be sufficiently conserved for diagnostic purposes. However, the primers for amplification must be carefully designed to amplify all the different types of
T. gondii. We have shown that most of the
T. gondii isolates in amniotic fluids are of type 2 [
24], whereas most of the published PCR primers are designed complementary to the type 1 sequences.
With respect to the recently described 529-bp repeat element, the sequencing data indicate a highly conserved nucleotide sequence among various strains and isolates of T. gondii and, equally important for hybridization probe-based detection of amplicons, among different intragenomic copies of this particular target. Alignment of our sequencing results with 62 different T. gondii isolates (5 reference strains, 7 clinical isolates and 51 T. gondii-positive amniotic fluids) revealed 3 nucleotide mutations, one deletion, and one defined nucleotide ambiguity compared to GenBank AF146527. The newly determined 404-bp partial sequence within the T. gondii 529-bp repeat region was deposited as GenBank AF487550.
Since the sequence of the original GenBank entry AF146527 was obtained with cloned amplicons, only single molecules of the 529-bp amplicons have been investigated by Homan et al. [
8]. Without doubt, this project was an important step towards the molecular characterization of the
T. gondii-specific 529-bp repeat element. Knowing about, and, as a consequence, omitting the extremely "T"-rich region at the 3' end of the target gene, we used the primer pair Tox-5 and Tox-8 for amplification and direct sequencing of a 404-bp internal segment. This integrated sequencing approach, in connection with the ABI four-dye one-lane sequencing technology, led to a more detailed insight into the nucleotide variations present at individual positions within different intragenomic copies of the
T. gondii 529-bp repeat element.
Based on the newly determined consensus sequence GenBank AF487550, we designed a set of LightCycler PCR primers and hybridization probes specific for the T. gondii 529-bp repeat element. Compared to a well-established LC protocol targeting the B1 gene, the analytical sensitivity of the novel LC protocol was evaluated with 10-fold dilution series of T. gondii genomic DNA. Here a detection limit of 20 fg of genomic DNA was consistently observed, which was found to be ten times more sensitive than the results obtained with the B1-specific LC assay.
Real-time PCR provides quantitative results, since the cycle number in which amplicons become detectable is proportional to the logarithm of initial number of templates. Identical amplification efficiencies and the fact that a gain in crossing point values of about 3.5 cycles was observed with the 529-bp repeat element LC assay for a given starting concentration of template DNA, the actual copy number of 529-bp repeat elements within the genome of T. gondii must be at least tenfold higher than the copy number of the 35-fold repeated B1 gene. Statistical exploitation of quantitative LightCycler results obtained with 51 T. gondii-positive human amniotic fluids showed a gain in crossing point values of 4.6 in average, thereby supporting the practical advantage of the 529-bp repeat element over the B1 gene as diagnostic target with respect to assay sensitivity.
With respect to the quantitative PCR protocol of Homan et al. [
8], who amplified the complete 529-bp repeat region and observed a detection limit of four tachyocytes, we intentionally amplified only an internal 162-bp segment of the target gene, thereby omitting the extremely "T"-rich region at the 3'-end. It is well known that long stretches of identical nucleotides in the template strand may cause a "stuttering" behaviour of the
Taq DNA polymerase reducing the speed of the elongation process or causing the enzme to fall off the template strand. Leaving this problematic "T"-rich segment of the target gene improved the efficiency of the amplification process and a detection limit of less than one tachyzoite equivalent was observed in the present study.
The specificity of the newly developed LC assay targeting the the T. gondii 529-bp repeat element was evaluated with DNA preparations of 62 T. gondii isolates or clinical specimens positive for T. gondii, 160 clinical specimens from patients without a suspected T. gondii infection, and 118 strains of different bacterial species. Positive PCR results were observed with only the 62 T. gondii-positive DNA preparations whereas all of the other DNA preparations remained negative. No cross-reactions or borderline PCR results in the presence of high amounts of human chromosomal DNA were observed. Consistently negative PCR results were observed with specimens of various types originating from individuals who had anti-toxoplasma antibodies but no clinical symptoms of acute toxoplasmosis.
Although we evaluated the novel LC assay with a collection of
T. gondii-positive amniotic fluid samples, other biological fluids like blood, aqueous humor, vitreous, or tissues may also serve as starting material for the sensitive detection of the parasites in clinical samples [
25,
26]. It should be considered that the repartition of the toxoplasmosis cysts in organs, such as heart or brain, is not homogenous and the size of the sample subjected to DNA extraction is a key issue. Consequently, a negative PCR result with a small biopsy section does not definitively exclude the presence of toxoplasmosis cysts in a given organ or clinical sample.
Authors' contributions
U. Reischl, S. Bretagne and J.-M. Costa were responsible for the development, probe design, and evaluation of the novel PCR assay. P. Ernault was involved in testing DNA preparations of human amniotic fluids. Last but not least, D. Krüger made his colleagues aware of this novel T. gondii-specific target sequence.