Parasitology Research

, 106:499

Evaluation of a molecular beacon real-time PCR assay for detection of Baylisascaris procyonis in different soil types and water samples

Authors

  • Rachel R. Gatcombe
    • National Center for Zoonotic, Vector-Borne and Enteric DiseasesCenters for Disease Control and Prevention
    • Atlanta Research and Education Foundation
  • Narayanan Jothikumar
    • National Center for Zoonotic, Vector-Borne and Enteric DiseasesCenters for Disease Control and Prevention
  • Sriveny Dangoudoubiyam
    • Department of Comparative PathobiologyPurdue University
  • Kevin R. Kazacos
    • Department of Comparative PathobiologyPurdue University
    • National Center for Zoonotic, Vector-Borne and Enteric DiseasesCenters for Disease Control and Prevention
Original Paper

DOI: 10.1007/s00436-009-1692-6

Cite this article as:
Gatcombe, R.R., Jothikumar, N., Dangoudoubiyam, S. et al. Parasitol Res (2010) 106: 499. doi:10.1007/s00436-009-1692-6

Abstract

Baylisascaris procyonis is a helminth parasite commonly found in North American raccoons (Procyon lotor) that is a cause of clinical neural, ocular, and visceral larva migrans in humans when infective eggs are ingested. Rapid detection of B. procyonis eggs in contaminated soil and water would assist public health analysts in evaluating risks associated with public exposure to areas of known raccoon activity. In this study, a molecular beacon probe-based real-time polymerase chain reaction (PCR) assay was developed to enable rapid and specific detection of eggs of Baylisascaris spp. The molecular beacon assay targeted the cytochrome oxidase subunit 2 (cox-2) gene of B. procyonis. To determine method sensitivity, experiments testing various egg levels (250, 25, and five eggs) were performed by seeding into 0.5-g soil samples or 0.5-mL water samples. Different soil sample types were extracted using a commercial nucleic acid extraction kit. Specificity testing using previously characterized helminth tissue specimens indicated that the assay was specific to Baylisascaris spp. Little real-time PCR inhibition was observed for most of the soil and water samples. A seed level of 250 eggs was detected for all soil types, and two seed levels (25 and five eggs) were detected for surface water samples. These results demonstrate that the reported real-time PCR assay was effective for the sensitive detection of B. procyonis in a wide range of soil types, and should be a useful tool for investigations of soil or water potentially contaminated with eggs of this parasite.

Introduction

Baylisascaris procyonis is an increasingly recognized cause of neural larva migrans in humans and animals, causing severe neurological deficits or fatalities in most domestic cases (Gavin et al. 2005). The parasite is also well recognized as a cause of ocular larva migrans in humans (Goldberg et al. 1993; Kazacos 2001). Raccoons (Procyon lotor) are the definitive host for B. procyonis, a roundworm that infects raccoons through ingestion of infective eggs, or larvae in the tissues of other animals such as rodents and birds (Gavin et al. 2005). The documented prevalence rates of raccoons infected with B. procyonis vary from 22% or lower in the southeast to as high as 82–100% in the midwest, northeast, and west coast of the USA (Eberhard et al. 2003; Evans 2002; Kazacos 2001; Park et al. 2000). In North America, raccoons thrive in close proximity to human residential areas, particularly near wooded recreational parks and in urban and suburban settings where food and water sources are close by (Page et al. 1998; Roussere et al. 2003). Raccoon desensitization to human residence and interaction poses a public health threat, since raccoon feces are commonly found close to homes, making accidental ingestion of B. procyonis eggs a real possibility.

The major sources of B. procyonis contamination in such areas are raccoon “latrines”, which are accumulations of feces at the base of trees or on logs, woodpiles or other areas, resulting from habitual defecation at these specific sites. On average, an infected raccoon sheds about 20,000 to 26,000 B. procyonis eggs per gram of feces and some can shed in excess of 250,000 eggs per gram of feces (Kazacos 2001). A single latrine may contain large quantities of feces and thus millions of infective eggs (Page et al. 1999; Roussere et al. 2003). B. procyonis eggs are ellipsoidal and dark brown and measure from 63 to 88 µm by 50 to 70 µm in size. Eggs develop into infective second-stage larvae in 2 to 4 weeks, are resistant to decontamination and degradation, and can remain viable in the environment for several years (Kazacos and Boyce 1989). Because raccoons have become peridomestic, there is increased risk for human exposure to infective eggs from raccoon latrines. In clinical cases of baylisascariasis, raccoon inhabitance near the patient’s home is frequently described (Gavin et al. 2005). Young children are particularly at risk of infection due to raccoon inhabitance near and around recreational areas such as playgrounds. Infants who may explore their environment by oral means are at great threat of ingestion of contaminated soil. The majority of published clinical cases of B. procyonis have been found in children under the age of three (Murray and Kazacos 2004). Studies also indicate that playground soil can be highly contaminated with ascarid eggs (Chorazy and Richardson 2005).

Considering the potential for contact with infective eggs, the development of rapid diagnostic assays is important for proper assessment of infection with eggs from contaminated soil. In addition to being an emerging pathogen that is generally underdiagnosed, B. procyonis has also been suggested as a potential biothreat agent (Sorvillo et al. 2002). Rapid and sensitive molecular assays (polymerase chain reaction, PCR) have recently been published for detection of B. procyonis eggs and larvae in fecal, environmental, and tissue samples (Dangoudoubiyam et al. 2009). These researchers identified the parasite’s mitochondrial cytochrome oxidase 2 (cox-2) gene as the target for amplification, and examined both conventional and SYBR Green dye-based real-time PCR. The SYBR Green dye-based real-time PCR assay had 100-fold greater sensitivity for detecting B. procyonis genomic DNA than conventional PCR and successfully detected the parasite’s eggs or larvae in spiked canine fecal samples, naturally contaminated soil, feces from a clinically patent dog, and brain tissue from rodents and birds that died from clinical neural larva migrans due to the parasite (Dangoudoubiyam et al. 2009). Concerning environmental sampling for eggs, it was not known if different soil types would have an effect on assay sensitivity, or if the assays could also be applied to surface water contaminated with eggs. In the present study, the cytochrome oxidase subunit 2 (cox-2) gene was targeted to develop a probe-based (using a molecular beacon) real-time PCR assay to detect B. procyonis eggs. The assay was evaluated using an array of previously characterized soil types, field soil samples, and surface water to represent the various environmental matrices that might be encountered during B. procyonis investigations.

Materials and methods

Specimens

Embryonated and unembryonated B. procyonis egg stocks were provided by the authors at Purdue University. Unembryonated egg stocks were quantified by microscopy and determined to contain 200 eggs/20 µL. All B. procyonis egg stock specimens were stored at −20°C. Specificity testing was performed using a stock of Ascaris suum eggs (Excelsior Sentinel, Inc., Trumansburg, NY, USA) and adult nematode tissues (Ancylostoma caninum, Ascaris lumbricoides, Baylisascaris columnaris, Baylisascaris transfuga, Toxocara cati, Toxocara canis, and Toxascaris leonina) provided by the authors at Purdue University.

Soil samples and seeding

A panel of seven soil samples was obtained from the United States Environmental Protection Agency’s (USEPA) National Exposure Research Laboratory, Characterization and Monitoring Branch, Las Vegas, NV. The panel included sand (two types), loam, loamy sand, sandy loam, clay loam, and silt loam having soil composition, pH values, and total organic carbon (TOC) concentrations shown in Table 1. Non-seeded USEPA soil samples were tested for background Baylisascaris and were found to be negative. All soils were stored at 4°C in separate capped-glass test tubes. Three seed levels of B. procyonis eggs (250, 25, and five eggs) were added to 0.5-g aliquots of each USEPA soil sample to investigate molecular beacon real-time PCR assay detection performance. All USEPA soil testing was done in triplicate, with one negative nuclease-free water PCR control per PCR run.
Table 1

Soil panel composition, pH, and TOC valuesa

Soil ID

Soil type

Soil composition

pH

TOC

A

Nevada C

Sand 98%

11

0.02%

Silt 1%

Clay 1%

B

C horizon sand

Sand 95%

5.5

4.50%

Silt 4%

Clay 1%

C

Loam

Sand 50%

8.4

1.85%

Silt 40%

Clay 10%

D

GA Bt2 clay loam

Sand 42%

5

0.20%

Silt 22%

Clay 32%

E

Bs loamy sand

Sand 81.9%

4.4

2.90%

Silt 10.9%

Clay 0.2%

F

Charleston sandy loam

Sand 61%

7.7

3.80%

Silt 31%

Clay 8%

G

Bs silt loam

Sand 26%

5.1

1.50%

Silt 67%

Clay 7%

aSoil type, composition, pH, and TOC data provided by USEPA, Las Vegas, NV, USA

In addition to the USEPA soil panel, three soil samples were collected from a local park in Decatur, GA. Sterile, 50-mL centrifuge tubes were packed with soil from three sampling sites: a children’s playground, a forestry area, and a baseball field. pH values for these soils were 5.2, 6.5, and 7.1, respectively, and TOC values were 14.6%, 2.99%, and 4.55%, respectively (Midwest Laboratories Inc., Omaha, NE, USA). Two seed levels (250 and 25 eggs) of B. procyonis eggs were added to the 0.5-g sample of soil from each sampling site for nucleic acid extraction and real-time PCR testing. Non-seeded Decatur, GA soils were tested for background B. procyonis and were found to be negative. Soils were stored at 4°C and processed within a week of sampling. All testing was done in duplicate with one negative nuclease-free water PCR control per PCR run.

Water samples and seeding

A 20-L surface water sample was obtained in a cubitainer from Murphy Candler Lake (Atlanta, GA, USA) and had pH and TOC values of 7.71 and 10.4 mg/L, respectively. The water was concentrated using ultrafiltration (UF) using Fresenius F200NR ultrafilters, which were blocked using 5% calf serum (Invitrogen cat. no. 16170-078) and allowed to rotate overnight on a rotisserie (Hill et al. 2007). Prior to beginning the UF procedure, sodium polyphosphate (NaPP; Sigma-Aldrich) was added to the water sample to achieve a final concentration of 0.01% (wt/vol) (Hill et al. 2007). After UF was completed and the retentate collected, a 500-mL elution solution consisting of 0.01% Tween 80, 0.01% NaPP, and 0.001% Antifoam Y-30 Emulsion was recirculated through the ultrafilter and allowed to reduce in volume (Hill et al. 2007). The final ultrafiltration sample (retentate + eluate = 208 mL) was centrifuged at 4,000×g at 4°C for 30 min, the supernatant discarded, and the pellet resuspended using phosphate-buffered saline to achieve a final sample volume of 4.5 mL. The water sample was stored at 4°C and used within 1 week of collection. To investigate B. procyonis real-time PCR detection limits for water concentrates, two seed levels (25 and five eggs) were added to 0.5 mL of lake water concentrates (in duplicate) and one spiked nuclease-free water matrix control (for each egg level). Non-seeded water samples were tested for background B. procyonis and were found to be negative.

Nucleic acid extraction

Egg seeding and DNA extraction were performed on the same day. DNA was extracted from 0.5-g soil samples using an UltraClean Soil DNA Isolation kit (MoBio Laboratories, Carlsbad, CA, USA) according to the manufacturer’s instructions. DNA extracts were stored at −20°C. DNA was extracted from 500-µL volumes of concentrated water samples using an UltraClean Soil DNA Isolation kit according to the manufacturer’s instructions.

Molecular beacon (cox-2)-based real-time PCR assay

A molecular beacon real-time PCR assay was developed by the Centers for Disease Control (CDC) for the present study. To target the B. procyonis cox-2 gene sequence, alignments were made using four Genbank sequences: AF179908 (Baylisascaris procyonis cytochrome oxidase subunit 2 (cox-2) gene), X54253 (Ascaris suum complete mitochondrial genome), AF179922 (Toxascaris leonina cytochrome oxidase subunit 2 (cox-2) gene), AF179907 (Ascaris lumbricoides cytochrome oxidase subunit 2 (cox-2) gene), and AF179909 (Baylisascaris transfuga cytochrome oxidase subunit 2 (cox-2) gene) using Clustal X version 1.8 (Thompson et al. 1994; http://www.ebi.ac.uk/clustalw). The forward primer (SrF1; 5′-GAGTTATGAGTTTAGTGATATTCCTGGA-3′) had a Tm = 60°C. The reverse primer (JVBPR-5′-GCAAAGCCCAAGAATGAATCAC-3′) had a Tm = 66°C. The molecular beacon probe (JVBPP-5′-cgcgatcTATTAACATCACAAGGTACAACACAACGgatcgcg-3′) had a Tm = 75°C and a stem ∆G = −4.96. The molecular beacon probe was synthesized with a FAM fluorophore label at the 5′ end and a Black Hole Quencher molecule as a quencher at the 3′ end (Fig. 1). The probe was designed to form a stem–loop structure to bring into close proximity the 5′ and 3′ ends of the probe, resulting in minimal fluorescence for the non-hybridized probe. In the presence of a complementary target sequence, the probe hybridizes to the target, separating the reporter dye from the quencher and resulting in a measurable increase in fluorescence. Primer and probe positions were based on Genbank Sequence #AF179908. Primers were designed to have no secondary structures, self-annealing, or significant identity (E < 1.0) to non-Baylisascaris sequences found by BLAST searching (Altschul et al. 1997; http://www.ncbi.nlm.nih.gov/BLAST/). The primers and probe were synthesized in the CDC Biotechnology Core Facility (Atlanta, GA, USA).
https://static-content.springer.com/image/art%3A10.1007%2Fs00436-009-1692-6/MediaObjects/436_2009_1692_Fig1_HTML.gif
Fig. 1

Predicted secondary structure of B. procyonis molecular beacon probe. The folding algorithm used to generate the figure designed by D. Steward and M. Zucker (https://www.bioinfo.rpi.edu/applications/hybrid/quikfold/php)

Real-time PCR

Real-time PCR reactions were performed in a volume of 20 μL, containing 10 μL of 2× QuantiTect Probe PCR kit (Qiagen, Valencia, CA, USA), 250 nM of each specific primer, 100 nM probe, and 2 µL of the genomic DNA (from the water or soil sample extract). Amplification consisted of 15 min at 95°C followed by 45 cycles of 10 s at 95°C, 30 s at 55°C (with fluorescence data collection), and 20 s at 72°C and was performed with the ABI 7500 Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). Potential real-time PCR inhibition in the soil matrices was tested by adding B. procyonis egg DNA to nuclease-free water and nucleic acid extracts from non-seeded soil samples and comparing the crossing threshold (CT) values between the soil and control samples to determine if the soil samples exhibited inhibited PCR (Table 2). Potential real-time PCR inhibition associated with water samples was evaluated using the same approach used to evaluate potential real-time PCR inhibition associated with soil samples. All inhibition testing was done in duplicate with one negative nuclease-free water PCR control per PCR run.
Table 2

Average CT values for B. procyonis eggs in 0.5-g soil samples and associated delays in CT values from PCR inhibition testing

Sample name

Average CT values for 250 egg seed

Average CT values for 25 egg seed

Average CT values for five egg seed

CT value difference between soil sample and positive control

Soil A

35.5 ± 0.66

37.8 ± 1.43

41.3b

5–8

Soil B

27.5 ± 0.87

31.6 ± 0.23

34.0 ± 0.42

3–4

Soil C

34.7a

Neg

ND

3–4

Soil D

36.1 a

Neg

ND

4–5

Soil E

27.8 ± 0.88

33.2  ± 0.61

39.0b

NAc

Soil F

28.9 ± 4.95

Neg

ND

5–8

Soil G

31.4 ± 0.85

35.8b

36.3 ± 3.14

4–5

aResult from one positive assay (two of three replicates were negative)

bAverage of two positive assays (one of three replicates were negative)

cNA not applicable, average CT of soil was negative; Neg all three replicates were negative; ND not done

Sensitivity and specificity testing

A standard curve for the real-time PCR assay was prepared using duplicate dilution series of DNA extracted from a stock of 250 B. procyonis eggs in nuclease-free water. PCR efficiency (%) based on the standard curve was calculated using the formula: \( \left( {{\text{1}}{0^{\left( { - {\text{1}}/{\text{slope}}} \right)}} - {\text{1}}} \right) \times {\text{1}}00 \). The egg equivalents corresponded to the following values: 25, 10, 5, 2.5, 1, 0.5, 0.1, 0.05, 0.01, 0.001, and 0.0001 eggs. Specificity testing was performed using previously identified helminth tissue and eggs to examine potential cross-reactivity among similar nematodes. The following tissues were extracted and tested: A. lumbricoides, A. suum, An. caninum, B. columnaris, B. transfuga, T. cati, T. canis, and T. leonina.

Results

Assay specificity and sensitivity

No cross-reactivity was observed when the molecular beacon real-time PCR assay was tested using a DNA panel prepared from the following specimens: A. lumbricoides, A. suum, An. caninum, T. cati, T. canis, and T. leonina. However, the assay did react with the three Baylisascaris species tested (B. procyonis, B. columnaris, and B. transfuga).

A standard curve was prepared using nine dilutions of DNA extracted from a stock having a known quantity of B. procyonis eggs. The standard curve for the assay had an R2 value of 0.98 and a slope of −3.87 (Fig. 2), which corresponds to a calculated PCR efficiency of 81%. A theoretical egg equivalent of 0.001 eggs was determined to be the detection limit of the assay.
https://static-content.springer.com/image/art%3A10.1007%2Fs00436-009-1692-6/MediaObjects/436_2009_1692_Fig2_HTML.gif
Fig. 2

Linear standard curves for detection of Baylisascaris procyonis using molecular beacon real-time PCR assay

B. procyonis detection for USEPA soil panel

Based on CT results, a seed level of 250 eggs was associated with an average CT value of 28–36 for all of the USEPA soils (Table 2). Soils C, D, and F were associated with significant real-time PCR inhibition and rendered negative results for the 25 egg levels. Soils C, D, and F all share a loamy consistency and relatively high clay content (Table 1), which may correspond to elevated humic acid concentrations and associated PCR inhibitory effects. PCR inhibition testing indicated that soil samples A and F exhibited substantial inhibition, with CT value delays ranging from 5 to 8. At the 25 egg level, B. procyonis that could be repeatedly detected in soils A, B, E, and G had good CT values of 31 to 38. At the five egg seeding level, soils A, B, E, and G had CT values ranging from 34 to 41.

B. procyonis detection for field soil and water samples

The molecular beacon probe-based real-time PCR assay was found to be effective in amplifying B. procyonis in various types of environmental soil and water (Table 3). The qPCR results for soil show that the assay was effective at detecting 250 and 25 eggs of B. procyonis in field soil samples and 25 and five eggs in water samples. The soil samples had an average CT value of 30 at the 250 egg level and 34 for the water sample at the 25 egg level. Positive CT values at even lower egg levels (i.e., 25 for the soil sample and five for the water sample) demonstrate that this assay can consistently render positive results in spiked environment field samples.
Table 3

Detection of seeded B. procyonis samples in soil and water

Matrix

Egg #

Location

Average CTa

Inhibition delay from pos. control

Soil

250 egg

Baseball field

31

2

Soil

25 egg

Baseball field

41

5

Soil

250 egg

Forest

30

1

Soil

25 egg

Forest

41

5

Soil

250 egg

Playground

30

1

Soil

25 egg

Playground

41

5

Water

25 egg

Murphy Candler Lake

34

1

Water

5 egg

Murphy Candler Lake

38

1

aPerformed in duplicate

Discussion

Simple and rapid analytical techniques are needed to improve the capacity of public health scientists to test soil samples for the presence of pathogens, including Baylisascaris. Using a different set of primers than previously examined (Dangoudoubiyam et al. 2009) and a molecular beacon probe-based real-time PCR methodology, the present study confirms the use of real-time PCR to detect B. procyonis eggs in soil and extends the results to specific known soil types. In addition, the method successfully detected eggs in seeded concentrated surface water samples. While other methods such as egg flotation and microscopy have been reported, these methods are time consuming and complicated and require a baseline expertise in diagnostic parasitology. The aim of this study was to develop a simple probe-based real-time PCR assay that could rapidly and sensitively detect B. procyonis in environmental samples. The assay was found to be capable of detecting DNA from the equivalent of less than one B. procyonis egg. In addition, the calculated PCR efficiency for the molecular beacon assay (81%, based on standard curve slope of −3.867) was higher than the previously reported SYBR Green-based assay (based on the reported slope of −4.827), possibly due to higher annealing temperatures of the primers used in the present assay. When the molecular beacon real-time PCR assay was applied to seeded soil and water samples, the assay was found to consistently detect 250 B. procyonis eggs in 0.5-g soil samples. For soil samples not exhibiting PCR inhibition, detection limits were as low as five eggs in 0.5-g samples (i.e., EPA soils A, B, E, and G). When the assay was applied to soil samples collected from a Decatur, GA park, the detection limit was 25 eggs in 0.5-g samples. When the assay was applied to surface water concentrates, a seed level of five eggs was consistently detected in 0.5-mL samples. In related work demonstrating the sensitivity of these methods, Dangoudoubiyam et al. (2009) used conventional and SYBR green-based real-time PCR to detect as few as 20 unembryonated B. procyonis eggs per gram of feces and to amplify the DNA extracted from a single larva. Whereas in the present study the probes used could not differentiate B. procyonis from the related parasites B. columnaris and B. transfuga, essentially being genus-specific, the primers used by Dangoudoubiyam et al. (2009) amplified only B. procyonis and B. columnaris, which are very closely related. As a cause of larva migrans, B. columnaris is less pathogenic than B. procyonis, and although B. columnaris has been involved in a limited number of clinical cases in animals, B. procyonis accounts for an estimated 96% of recognized cases and outbreaks (Kazacos 2001).

This study reports a molecular beacon probe-based rapid real-time PCR assay that can be useful for improving the effectiveness of environmental investigations for B. procyonis, an emerging pathogen that is largely under-diagnosed and may represent a widespread threat to public health (Gavin et al. 2005; Murray and Kazacos 2004; Sorvillo et al. 2002). This real-time PCR assay was found to be specific for Baylisascaris spp. and sensitive for the detection of B. procyonis in soil and water samples, but site-specific conditions (e.g., presence of PCR inhibitors) and the choice of sample preparation technique could affect the performance of this analytical technique for specific applications.

Acknowledgments

The authors thank Brian Schumacher (USEPA/NERL) for assistance in providing soil panel specimens for this study. This publication was supported in part by funds made available through the Centers for Disease Control and Prevention, Coordinating Office for Terrorism Preparedness and Emergency Response. The use of trade names and names of commercial sources is for identification only and does not imply endorsement by the Centers for Disease Control and Prevention or the US Department of Health and Human Services. The findings and conclusions in this presentation are those of the authors and do not necessarily represent those of the Centers for Disease Control and Prevention. Experiments performed for this study comply with the current laws of the USA.

Copyright information

© US Government 2009