An improved polymerase chain reaction assay for the detection of Tritrichomonas foetus in cattle

Abstract

An improved polymerase chain reaction test has been developed to detect Tritrichomonas foetus, the causative agent of trichomoniasis in cattle. The test amplifies a region of the 5.8S ribosomal RNA gene of T. foetus, and it is simple, sensitive, and specific when compared with traditional methods to examine field samples.

Tritrichomonas foetus is the causative agent of bovine trichomoniasis, a sexually transmitted disease of economic importance to livestock producers because it causes early abortions, uterine infections, and sterility (1). It has a worldwide distribution. It is transmitted from infected bulls to cows at the time of coitus, wherein the parasite invades the vagina, uterus, and oviducts and causes embryonic death and infertility. Although testing of breeding animals and widespread use of artificial insemination has helped to control its distribution in North America, Europe, and Australia, it is still present in range herds and natural breeding situations, with reports of the prevalence of infection in bulls ranging from relatively low percentages (2% to 5%) in some herds to significant levels (15% to 40%) in others (2,3,4).

The diagnosis of bovine trichomoniasis has been based mainly upon microscopic examination of cultures in Diamond's medium (5) of organisms from preputial washings from bulls and cervicovaginal secretions from cows. This procedure is labor intensive and expensive in both the field and laboratory, and it can be insensitive if only a small number of organisms are present in preputial washings and vaginal swabs, or if viability of the organism in inoculated medium is compromised by field conditions, such as cold temperature. Therefore, efforts have been made to establish DNA-based diagnostic methods for the detection of T. foetus.

Initially, attempts were based on restriction digestion of DNA from T. foetus and other organisms and the identification of an 0.85 kb band from HindIII digestion of genomic DNA that was present in all T. foetus isolates that were tested (6). A set of polymerase chain reaction (PCR) primers labeled TF1 and TF2 were developed to amplify a 162 bp amplicon within this fragment. These primers also yielded a 400 bp product from bovine genomic DNA and multiple amplification products from DNA from a related organism, T. vaginalis, although these could be distinguished from the 162 bp T. foetus amplicon following Southern blot of the agarose gel and hybridization with the 0.85 kb T. foetus DNA probe (6,7). Recently, comparative sequence analysis of the 5.8S rRNA gene and internal transcribed spacer regions (ITS1 and ITS2) from several different T. foetus isolates demonstrated strict intraspecies conservation of these sequences and distinctiveness of these sequences from those in 5.8S rRNA genes in other trichomonad protozoa (8). This sequence information was applied to design primers labeled TFR3 and TFR4 for the PCR amplification of a 347 bp fragment from T. foetus DNA that spanned the entire 5.8S rRNA gene and the flanking ITS1 and ITS2 regions (9). No amplification products were observed with DNAs from other trichomonads, bacterial DNA, or purified bovine genomic DNA, and the sensitivity, as in the previous PCR test, was found to be sufficient to detect DNA from just one or a few organisms in an assay (9). However, some additional PCR products not related to the 5.8S rRNA T. foetus sequences were still generated in addition to the 347 bp fragment when the TFR3 and TFR4 PCR primers were used with Tritrichomonas DNA or DNA from clinical diagnostic samples (9). For this reason, we have designed another set of primers from the 5.8S rRNA gene that do not yield multiple PCR products with T. foetus DNA and, besides generating only a single amplicon of 211 bp in length, share the same level of sensitivity and specificity described for the TFR3 and TFR4 primer pair.

Primer sequences were designed to amplify the region of the T. foetus 5.8S rRNA gene from nucleotide 1615 in the 18S rRNA gene bordering the internal transcribed spacer, ITS1, through ITS1 into the 5.8S rRNA sequence, ending at nucleotide 1825. The upstream primer, which we have designated as TF211A, has the sequence: 5'CCTGCCGTTGGATCAGTTTCGTTA3'. The downstream primer, which we have designated as TF211B, has the sequence: 5'GCGCAATGTGCATTCAAAGATT- CG3'. A BLASTn (basic local alignment search tool nucleotide) search (10) indicated that these primers have only limited homology to the respective sequence of T. vaginalis, a related human parasite that is not found in cattle. By using DNA derived from a T. foetus reference sample obtained from the American Tissue Culture Collection (ATCC #30233; Rockville, Maryland, USA), conditions for optimal PCR amplification were established. Parasite DNA was prepared from pelletted, cultured organisms, after a previously described method (11), wherein the organisms were lysed with buffered sodium dodecyl sulfate (SDS), digested with proteinase K (Gibco BRL, Grand Island, New York, USA), the proteinaceous material precipitated by high salt, the purified DNA recovered by alcoholic precipitation and redissolved in TE buffer (10 mM Tris HCl, 1 mM Na₂EDTA. pH 8.0). The PCR reactions containing T. foetus DNA were carried out in a 50 μL volume in PCR buffer (final concentration of 50 mM KCl, 10 mM Tris HCl, pH 8.8 [25°C] and 0.05% [v/v] of each of Tween-20 and Nonidet-P40), deoxynucleoside triphosphates (200 μM final concentration of each of deoxyadenosine, -cytosine, -guanine, and -thymine triphosphates), oligodeoxynucleotide primers (200 nM final concentration of each), 1.5 mM MgCl₂, and 1 unit (0.2 μL) of Taq DNA polymerase in a thermal cycler (Model 480 Thermal Cycler; Perkin Elmer, Norwalk, Connecticut, USA) for 35 cycles, where each cycle consisted of annealing at 60°C for 1 min, extension at 72°C for 1 min and denaturation at 94°C for 1 min. A variety of sources of Taq polymerase from a number of different suppliers were used successfully. These included AmpliTaq Gold (Perkin Elmer, Applied Biosystems Division, Foster City, California, USA), Platinum Taq DNA Polymerase (Life Technologies, Rockville, Maryland, USA), and Tsg DNA Polymerase (Bio Basic, Scarborough, Ontario). The PCR reactions were also carried out successfully in the same reaction mix as above, using other thermal cyclers (Model 2400 or Model 9600 Gene Amp System; Perkin Elmer, Applied Biosystems Division), except that 25 μL volumes were used and each step of the cycle program was reduced from 1 min to 30 s duration. In the presence of T. foetus DNA, these conditions yielded a strong signal for a single amplicon of the predicted size of 211 bp (Figures 1, 1b). This PCR product was sequenced by using the Prism Dye Terminator Cycle Sequencing protocol (Perkin Elmer, Applied Biosystems Division) and the sequence obtained was identical to the published sequence, with the exception that the nucleotide at position 1664 was a guanine in our PCR product and a thymine in the published sequence (8, GenBank M81842). We have verified this observed difference through sequencing of several PCR amplicons in both strands. Thus, this is likely a true strain polymorphism reflected in the particular reference sample obtained from American Type Culture Collection in comparison with the isolate used for the sequence published (8).

Figure 1. Specificity and sensitivity of the PCR assay to detect Tritrichomonas foetus DNA. The same PCR conditions were used throughout. A: DNA samples (20 ng) were subjected to 35 cycles of PCR and 15 μL of the PCR reaction was resolved electrophoretically on a 2% agarose gel containing 0.5 μg/mL ethidium bromide and photographed under UV light. Lanes 1, 7: 100 bp DNA ladder; lanes 2, 6: T. foetus DNA; lane 3: bovine genomic DNA; lane 4: Mycoplasma bovigenitalium DNA; lane 5: Ureaplasma diversum DNA. B: The gel on the left shows PCR products resolved electrophoretically from reactions containing decreasing quantities of T. foetus DNA. Lanes 1,12: 100 bp DNA ladder; lanes 2–9: T. foetus DNA at concentrations of 5 ng, 1 ng, 500 pg, 100 pg, 50 pg, 10 pg, 5 pg, and 1 pg, respectively; lane 10: 100 ng of bovine genomic DNA; lane 11: DNA blank (negative control). The gel on the right shows PCR products resolved from DNA from 5 (lane 2) and 1 (lane 3) T. foetus organisms, respectively; lane 1: 100 bp ladder. C: Detection of T. foetus DNA in samples of smegma cultures of bulls. PCR products from assay of 2 negative (lanes 1 and 2) and 4 positive (lanes 3 to 6) clinical samples were resolved electrophoretically on a 2% agarose gel; lane 7: 100 bp ladder (Gibco BRL).

The specificity of the PCR reaction was demonstrated by challenging the assay with DNA templates derived from T. foetus, Mycoplasma bovigenitalium (ATCC #19852), and Ureaplasma diversum (ATCC #49783). The latter 2 are bovine urogenital isolates that potentially could be included in preputial or vaginal samples for assay collected from animals. The assay was also challenged with bovine (Bos taurus) genomic DNA. Agarose gel electrophoresis of the PCR reaction indicated that no signal was detectable with any of the DNA samples, except for T. foetus DNA where a single band of 211 bp was clearly detectable on the ethidium bromide-stained gel (Figure 1a).

To assess the sensitivity of the PCR test, 2 types of experiment were conducted. In the first, T. foetus DNA was serially diluted over a range of 5 ng to 1 pg per reaction (Figure 1b). The 211 bp PCR product was detectable to levels as low as 1 pg, an amount in the range of the genome size of related protozoans in this class (12). In a second test, either a single organism or pools of 5 organisms were lysed to provide the DNA template for the PCR reaction and a strong signal was detectable in every case (Figure 1b). Thus, as with the 2 earlier PCR assays described in the literature (6,9), the level of sensitivity of the assay is very high and sufficient to detect the presence of DNA from just a single organism.

The reliability of this test was established with clinical samples from collections of bulls on pastures in Northern Alberta, in collaboration with veterinarians and scientists working with Alberta Agriculture Food and Rural Development (AAFRD). Preputial samples were collected by AAFRD veterinarians using a standard 1-mL plastic insemination tube (ITSI Continental Canada, Princeton, Ontario) with an attached 3-mL syringe to provide suction; the collected smegma (approximately 0.3 to 0.5 mL) was placed in 15 mL of Diamond's medium. An aliquot (400 μL) of this sample, concentrated to 200 μL by centrifugation, was sent to our laboratory for PCR assay, while the remainder was used for diagnosis of T. foetus in the provincial veterinary pathology laboratory (VPL, Edmonton, Alberta) by conventional means of culture and microscopy. Our assays were carried out blind and comparison of results between PCR tests and culture was not made until all assays had been completed. The 200-μL sample of culture was subjected to centrifugation at 12 000 × g in a microfuge for 5 min and the pellet was resuspended in 200 μL of phosphate-buffered saline with a vortex mixer. The cellular material was pelleted again by centrifugation, as above, and the wash procedure was repeated a second time. The final pellet was resuspended in 100 μL of lysis buffer, which is composed of PCR buffer plus proteinase K (125 μg/mL). The lysate was incubated at 65°C for 2 h and then at 95°C for 20 min to inactivate the proteinase K. An aliquot of this lysate (20 μL) was mixed with 5 μL of a master mix containing all the other PCR components to achieve the final concentrations noted above and the PCR was conducted, as described earlier.

Of the 847 samples analyzed, only 4 were found to be positive for T. foetus DNA by the PCR test (the 4 positive samples and 2 negative samples are shown in Figure 1c). The identity of the T. foetus DNA amplification product was verified in each case by hybridization, using the method described by Felleisen et al (9). Three of these 4 were also positive by conventional culture assay; no other samples were culture positive. The additional positive observed by our test may reflect the increased sensitivity of the PCR test over the culture test, or it may reflect the presence of DNA from T. foetus organisms that were present, but dead, in the sample and, consequently, not viable for culture and, therefore, not detectable with this procedure. To test for inhibition of the PCR reaction, a selection of apparently negative samples were retested in duplicate, with one reaction spiked with T. foetus DNA; nonspiked samples remained negative and all spiked samples were positive (data not shown). Otherwise, there was absolute concurrence in the 2 assay methods and occurrence of false positives was not a problem in this large sample screen.

These results indicate that this PCR test can be used successfully to screen for the presence of T. foetus infections in bulls. The assay is simple and rapid to perform when compared with conventional culture methods followed by microscopic identification of the protozoan parasite. In addition, because there is only one specific PCR reaction product with the primer set described herein, the test can easily be taken to the next step for automation for analysis and detection of PCR product in large sample sets.