Molecular Diagnosis of Baylisascaris schroederi Infections in Giant Panda (Ailuropoda melanoleuca) Feces Using PCR

The giant panda (Ailuropoda melanoleuca), one of the world's most iconic and threatened species, is considered a world natural heritage. Baylisascaris schroederi is the most common intestinal nematode in wild and captive giant pandas (Zhang et al., 2008). At the adult stage, this parasite usually inhabits the small intestine of the giant panda and can cause severe baylisascariasis (Zhang et al., 2008). The pathogenesis of baylisascariasis in pandas is mainly the consequence of anorexia, anemia, abdominal pain, diarrhea, and weight loss. When heavy infections or heterotopic parasitism occurs, the roundworms can cause intestinal obstruction, inflammation, and death (Loeffler et al., 2006). Currently, B. schroederi infections are widespread in wild and captive giant pandas, with an estimated prevalence of infection of 50 to 100% (Zhang et al., 2008). Primary and secondary infections with B. schroederi are one of the leading causes of death in giant pandas. Zhang et al. (2008) demonstrated that the probability of death of wild pandas caused by B. schroederi increased significantly between 1971 and 2005 and that visceral larva migrans caused by B. schroederi represented the most important cause of death between 2001 and 2005, responsible for 12 of 24 deaths reported (Zhang et al., 2008). Recently, fecal samples from six mountain ranges of China were used to study ascarid load in wild pandas. Results indicated that B. schroederi was still the most significant threat to health and survival of wild pandas, with infection rates up to 54% across the regions sampled (Zhang et al., 2011). The current approach for detection and identification of B. schroederi infection is based on conventional flotation and morphologic examination, which requires parasitologic expertise (Loeffler et al., 2006). However, due to the low density of B. schroederi eggs in feces or possible environmental cross-contaminating eggs of other parasites, such as morphologically similar Baylisascaris spp., the detection or identification is difficult and time consuming (Loeffler et al., 2006). A sensitive and efficient PCR technique has been developed and used successfully for the diagnosis of parasitic infestations (e.g., Baylisascaris transfuga eggs) from fecal samples of bears (Ursus arctos; Ambrogi et al., 2011). We designed a set of PCR primers for specifically amplifying a 291 base pair (bp) DNA fragment of 12S rRNA of the B. schroederi mitochondrial genome (mt DNA). We used the primers to detect specifically B. schroederi eggs in giant panda fecal samples. Based on these results, we evaluated the detection threshold, the cross-reactivity with other nematodes and the overall sensitivity of the method, as compared with classical flotation.

We used 50 fresh fecal samples of different giant pandas (500–600 g) from the Ya'an Bifengxia Base of China Conservation and Research Center for Giant Panda (30°07′25.5″ N, 102°59′48.6″ E). For flotation, each sample (50 g) was centrifuged at 12,000 × G for 15 min and subjected to flotation saturated with MgSO₄ for 30 min (Markovics and Medinski, 1996). Based on the morphologic characteristics (Sprent, 1968), B. schroederi-positive samples were identified and differentiated under light microscopy. For each negative sample, the flotation was repeated eight times. Afterwards, all samples (including the negative) were stored at 4 C and further examined using our PCR protocol. To determine the PCR detection threshold or sensitivity, specific numbers of B. schroederi eggs (1, 2, 3, 4, 5, 10, 25, 50, and 100) were obtained from a positive fecal sample with a capillary tube (0.5-mm diameter) under light microscopy. Each number of countable eggs was tallied twice independently, and in case of discrepancies, a third count was performed. Subsequently, the series of countable eggs with 20 µl of ddH₂O were subjected to alternate boiling and freezing (100 C and liquid nitrogen) for 10 cycles (Leles et al., 2009) and genomic DNA extraction (proteinase K digestion/phenol-chloroform DNA extraction/DNA purification; Iñiguez et al., 2002). A similar sample treatment (alternate boiling and freezing) and total DNA isolation were also performed in the 50 tested fecal samples for PCR detection. Finally, we evaluated the potential PCR cross-reactivity with eggs of other parasites (including morphologically similar Baylisascaris spp. and coinfected hookworms) using the adult DNA of B. transfuga, Baylisascaris procyonis, and Ancylostoma caninum (provided by Sichuan Agriculture University, China) as the controls in all amplification processes. After sequence alignment of the complete 12S genes derived from mt DNAs of B. schroederi (No. HQ671081), B. transfuga (No. NC_015924), B. procyonis (No. NC_016200), and A. caninum (No. NC_012309; Jex et al., 2009; Xie et al., 2011a, b), a set of primers specifically targeting a 291-bp fragment of 12S was designed with Primer Premier version 5.0 (Premier Biosoft International, Palo Alto, California, USA) and the corresponding oligonucleotide primer sequences were BpF (forward: 5′-TTTTACCTTGGCATTTTGTC-3′) and BpR (reverse: 5′-CTCTCAATTACTACTCAACCTCC-3′).

All reactions (25 µl), respectively, containing 1.5 µl of template genomic DNA (e.g., the countable feces-isolated eggs DNA, 50 fecal samples DNA, the adult of B. schroederi, B. transfuga, B. procyonis, and A. caninum DNA), 12.5 µl of PCR mixture, 8 µl of ddH₂O, 0.4% w/v bovine serum albumin (BSA; not used in DNA from the adults), and 1.5 µl of each primer (10 pmol each) were subjected to 95 C for 5 min, followed by 35 cycles at 95 C for 30 s, 50 C for 30 s, 72 C for 30 s, and a final extension at 72 C for 10 min. After amplification, 10 µl of PCR product from each sample was separated on a 1% agarose gel, and the amplicons were stained with ethidium bromide and visualized under ultraviolet light.

The results of detections on 50 giant panda fecal samples by MgSO₄-saturated flotation and by PCR assays are shown in Table 1. The initial flotation positive rate was 46% (23/50). However, after repeated flotation of the remaining 27 initially negative samples, another six samples were found positive, for a positive rate of B. schroederi of 58% (29/50). Among the repeated flotation for six positive samples, two samples were positive only in one of eight replicates, whereas the other four were positive in four of eight replicates (data not shown). Compared with the 88% (44/50) for PCR, the success of the PCR was 42% higher than the initial flotation and 30% higher than initial and repeated flotation combined. A high detection rate for the PCR examination and poor reliability of the flotation examination suggest that our PCR method is a more suitable alternative for species-specific assessment of subclinical B. schroederi infestations in giant pandas, even with low-level infestations. Flotation could remain a sufficient tool for the diagnosis of baylisascariasis in clinically affected panda, given that they shed large quantities of eggs during the course of the disease. Further field studies comparing the PCR targeting 291-bp 12S fragment and flotation methods could be carried out, with quantification of eggs, to more precisely evaluate the test.

To extract and purify DNA from ascarid eggs is particularly difficult. Generally, fertilized eggs possess a thick, mammillated proteinaceous coat and an impermeable, desiccation-resistant lipid layer (Ambrogi et al., 2011), which inhibit DNA extraction. We applied physical treatment (precipitation, flotation, boiling, freezing) to fecal eggs to break their complex shell and release the DNA. Furthermore, for DNA extraction, the phenol-chloroform DNA extraction protocol can obtain more egg DNA, than the commercial kit QIAamp DNA Stool Mini Kit (Qiagen, Venlo, Netherlands; Leles et al., 2009). Thus, we performed a similar DNA isolation with minor modifications. In a previous study, adding Taq DNA polymerase containing BSA to the reaction mixture reduced the inhibitory effect of feces, improving PCR quality (Abu Al-Soud and Rådström, 2000). Therefore, we added 0.4% BSA to our PCR reactions. Encouragingly, those preparations allowed subsequent PCR amplification.

Reports have indicated that the main intestinal nematodes in giant pandas are A. caninum and B. schroederi (Zhang et al., 2008). However, several other Baylisascaris-infected host animals occur in the area. To evaluate the cross-reactivity with possible cross-contaminating eggs of other Baylisascaris species (e.g., B. transfuga from bears [Ursidae]) and B. procyonis from raccoons [Procyon lotor]), we also used genomic DNA of B. transfuga and B. procyonis as templates. We detected no amplification of 12S-specific 291-bp product in these nematode species (Fig. 1), suggesting that our PCR allows discrimination between B. schroederi and other ascarid species, including morphologically similar Baylisascaris spp. Additionally, our extraction method combined with PCR amplification containing BSA, made the detection threshold a single feces-isolated egg (Fig. 2). This detection threshold was lower than that reported by Ambrogi et al. (2011) in B. transfuga (two eggs; Ambrogi et al., 2011). Thus, we believe that the PCR we describe will provide better performance in evaluating the prevalence of B. schroederi in fecal samples from captive or wild populations of giant pandas.

This work was supported by Research Project (200910188) from the Science and Technology Ministry, China, and Project (2010IK004) from the General Bureau of Entry-Exit Inspection and Quarantine, China. We thank Yun Sun and Jia-hai Wang (College of Veterinary Medicine, Sichuan Agricultural University, China) for technical assistance and Cai-wu Li and his staff (Ya'an Bifengxia Research Base of Giant Panda Breeding) for collection of material.