Development of a Rapid, Sensitive, and Field-Deployable Razor Ex BioDetection System and Quantitative PCR Assay for Detection of Phymatotrichopsis omnivora Using Multiple Gene Targets

Phymatotrichopsis root rot, also known as cotton root rot, Texas root rot, or Ozonium or Phymatotrichum root rot, is an important soilborne disease of over 2,000 dicotyledonous species, including cotton, alfalfa, vegetable crops, and fruit and nut trees (1). The causal fungus, Phymatotrichopsis omnivora, is a member of the family Rhizinaceae (Ascomycota: Pezizomycetes) and allied to the genera Psilopezia and Rhizina (2). The disease occurs in most of Texas (excluding the panhandle), southern Oklahoma, and New Mexico, southern and western Arizona, and northern Mexico (1, 2). Because of its extensive host range, P. omnivora represents a significant threat to agricultural productivity throughout its geographic range. Phymatotrichopsis root rot is a considerable economic concern, causing up to $100 million in annual losses to the U.S. cotton crop alone (2). Due to its high economic impact and broad host range, P. omnivora has been included in lists of regulated organisms by the European and Mediterranean Plant Protection Organization (EPPO A1 list no. 21 [http://www.eppo.org/QUARANTINE/listA1.htm]), the California Department of Food & Agriculture (no. 3261 in the CDFA Plant Quarantine Manual, 1989 [http://pi.cdfa.ca.gov/pqm/manual/pdf/309.pdf]), and the United Nations Security Council's monitoring program of Iraq (S/1995/208 [http://www.fas.org/news/un/iraq/s/s1995-0208.htm]). Isolation of P. omnivora from infected plants and infested soils is difficult and often not possible. Hence, culture-independent (molecular) approaches to pathogen identification are desirable. Accurate and sensitive detection and discrimination assays for early diagnosis would facilitate disease management. Methods developed for plant pathogen detection include enzyme-linked immunosorbent assay (ELISA) (3), loop-mediated isothermal amplification (LAMP) (4), oligonucleotide array (5), endpoint PCR (6), and real-time quantitative PCR (qPCR) (7). PCR-based techniques are generally more sensitive than immunological methods and have high specificity and discriminatory capabilities. Real-time qPCR offers greater sensitivity and speed than endpoint PCR in the detection of target DNA (7, 8). TaqMan and SYBR green qPCR are the two most popular qPCR formats and chemistries (9). SYBR green qPCR detects all amplified double-stranded DNA, including nonspecific reaction products, while TaqMan qPCR detects only specific amplification products because its amplification-dependent cleavage of probes, incorporating reporter and quencher dyes, results in increased fluorescence (10).

Currently, methods for on-site, rapid, reliable, and sensitive detection of P. omnivora are not available. On-site accurate and sensitive detection of fungi is a challenge. Available PCR assays require PCR inhibitor-free DNA purification from fungus-infected plants and access to laboratory facilities equipped with PCR and qPCR machines. Simple on-site DNA purification and PCR protocols for accurate pathogen identification would facilitate diagnosis and regulation, as well as disease monitoring and management. Ideally, these protocols would use a portable, battery-operated real-time qPCR platform designed for on-site molecular testing, which allows plant pathogen detection by minimally trained operators in the absence of laboratory facilities and conditions, including electricity, centrifuges, liquid nitrogen, water baths, incubators, and hazardous chemicals. Portable instruments developed previously for on-site pathogen detection include the SmartCycler (Cepheid, Sunnyvale, CA), the R.A.P.I.D. system (Idaho Technologies, Salt Lake City, UT), the LightCycler (Roche Applied Science, Indianapolis, IN), and the Bio-Seeq instrument (Smiths Detection, Edgewood, MD). Recently, Tomlinson and coworkers (10, 11) demonstrated the on-site detection of Phytophthora ramorum using qPCR (Cepheid SmartCycler II) and loop-mediated isothermal amplification. Use of the SmartCycler has been successful for on-site detection of the bacterium Xylella fastidiosa in grapevines (12) and for diagnosis of foot-and-mouth disease (13). These methods take about 2 h and detect a single target gene. The Razor Ex BioDetection system (Idaho Technology, Inc., Salt Lake City, UT) was designed originally for military use to identify biological-threat organisms on-site. The Razor Ex BioDetection system offers ready-to-use, freeze-dried reagent pouches and bar code-based PCR cycling program upload. These features make the Razor Ex BioDetection system user-friendly. Matero and coworkers (14) developed 40-min on-site detection methods based on the Razor Ex for anthrax, brucellosis, tularemia, and plague.

Nucleic acid-based detection of fungi in plant material requires the purification of genomic DNA (15), which can often be contaminated with phenolics and other PCR inhibitors. Therefore, on-site molecular detection requires both a battery-operated real-time qPCR thermocycler and a rapid and PCR inhibitor-free genomic DNA extraction method.

In this paper, we describe the development of TaqMan qPCR assays that may be performed rapidly, on-site, in about 30 min using the field-deployable Razor Ex BioDetection system protocols for reliable, sensitive, and accurate detection of P. omnivora using three target fungal genes. We also describe a modified magnetic-bead-based method for on-site DNA extraction (∼10 min) from fungus-infected plant roots. These tools provide enhanced investigative capability for applications in plant disease diagnostics and management, pathogen population monitoring, agricultural biosecurity, and microbial forensics.

We report the development of a reliable, accurate, and sensitive qPCR assay for detection of the plant-pathogenic fungus P. omnivora, using three pairs of primers and probes targeting three different genes, ITS-rDNA, beta-tubulin, and RPB2 genes. These primer and probe sets appear to be ready for use in the field-deployable Razor Ex BioDetection system. ZEN double-quenched probes having the internal ZEN quencher with the traditional 3′ end quencher shortened the distance between the reporter and quencher, leading to highly efficient quenching with significantly less background signal and greater precision (GEN, New Rochelle, NY).

High assay specificity, accuracy, and reliability are critical in biosecurity, quarantine, and microbial forensics applications. The use of a multigene format maximizes reliability, specificity, and broad-range detection within P. omnivora variants by minimizing the risk of false positives and false negatives because each gene acts as an internal control for the two other genes, which enhances the assay's applicability in microbial forensics. No nonspecific cross-reactivity (false positives) was observed for the three primer and probe sets when tested against plant and microbial exclusivity panels (Table 2). Positive PCR results with genomic DNA of 13 diverse P. omnivora isolates collected from different locations confirmed broad-range detection of the target pathogen (Table 3). All 14 symptomatic, asymptomatic, and dead plant roots were PCR positive when tested by all three primer and probe sets. Thirty-one samples of 52 symptomatic and asymptomatic plant and soil samples from alfalfa and cotton fields were qPCR positive for P. omnivora using the PO4 primers and probe. A few of the positive samples included asymptomatic or infected/dead weed plants collected from the alfalfa field.

The developed assays are sensitive, detecting as little as 1 fg of plasmid DNA (positive control) and 10 fg (PO4) to 100 fg (Pobt1 and PoRPB2-2) of P. omnivora genomic DNA. The higher sensitivity achieved with primer set PO4 may be due to the greater number of rDNA copies per genome (50 to 200 copies per fungal genome [23]) relative to the single or low-copy-number protein-encoding beta-tubulin and RPB2 genes. The high copy number of rDNA allowed detection even from fragmented P. omnivora DNA. The genome size of P. omnivora is estimated to be ∼115 Mb (24). Using a previously published formula (25), 1 pg = 978 Mb (or 1 Mb = 1.022 � 10⁻³ pg), the detection limit of a single genome of P. omnivora should be around 117.5 fg. However, the PO4 primers and probe detect down to 10 fg. Using SYBR green qPCR, Arif and coworkers (26) detected purified P. omnivora genomic DNA at amounts as low as 1 fg. However, SYBR green qPCR can result in false positives due to the ability of SYBR green dye to bind to any double-stranded DNA, including primer dimers, limiting the application of this protocol for forensic applications and quarantine inspections.

Previous work had shown that the addition of an AT-rich sequence at the 5′ terminus of primer sequences could lead to increases in PCR yields, sensitivity, and fluorescence (21, 22, 27). But in this study, the addition of 5′ AT-rich flaps made no difference in C_T values, indicating that the designed primers and probes already had optimal PCR thermodynamics.

An important capability for biosecurity, quarantine, and microbial forensics applications is the ability to use the procedure in the field. Tomlinson and coworkers (4, 10) developed on-site, loop-mediated isothermal amplification for phytoplasmas and SmartCycler protocols for Phytophthora ramorum. The SmartCycler has been used also for Xylella fastidiosa detection (12). Razor Ex BioDetection system-based methods are more rapid than those employing the SmartCycler (10, 12, 14). Matero and coworkers (14) found no difference in sensitivity or accuracy between the Razor Ex and Applied Biosystem 7300/7500 formats. Recently, Idaho Technology, Inc., announced the development of the Razor CRP BioThreat-X kit, which represents the first U.S. Department of Defense system capable of testing for 10 biothreat agents at one time in the field within 30 min. In our investigation, the Razor Ex BioDetection system protocol took about 30 min from sample preparation (DNA extraction, ∼10 min) to final detection (∼20 to 25 min). The rapid DNA isolation method, unlike many laboratory DNA extraction methods, does not require centrifugation, incubation, organic solvents, or the use of liquid nitrogen for sample homogenization. Commercially available Razor Ex BioDetection system pouches generally contain lyophilized PCR reagents to minimize contamination and eliminate the need for cold storage, but in our experiment we injected regular TaqMan qPCR components into the pouch using a disposable syringe. The primers for the Razor Ex BioDetection system were biotinylated, so as to capture the amplified product using streptavidin magnetic beads for further confirmation, if required. No assay inhibition was observed with biotinylated primers.

Multigene TaqMan qPCR and field-deployable Razor Ex BioDetection system-based detection assays for P. omnivora are rapid, reliable, sensitive, and efficient, and their use can speed phytosanitary diagnostics and pathogen detection. Early detection can help to prevent pathogen dissemination through plant materials during interstate or international commerce. The assays also have potential applications for farm management and monitoring of resistance in plant breeding programs. The modified Dynabeads-based rapid DNA isolation method, validated in this work for P. omnivora, can be applied to other plant pathogens, as well as for field and routine diagnostics.

We thank George Opit and Trenna Blagden (Oklahoma State University) for reviewing the manuscript. We extend our appreciation to Carolyn Young and Eddie Funderburg (Noble Foundation) for providing P. omnivora-infected samples used for validation of the techniques developed in this study. Thanks to Rimvydas Vasaitis (Swedish University of Agricultural Sciences [SLU]) for kindly providing dried R. undulata apothecia for primer validation. We are grateful to Carla Garzon (Oklahoma State University) for providing the genomic DNA of P. omnivora, R. undulate, and P. aphanidermatum isolates and reviewing the manuscript. We thank Ouyang Ping (Oklahoma State University) for technical assistance. We also appreciate the help provided by Astri Wayadande, Mindy James, Shefali Dobhal, and Donna Ria Cassi (Oklahoma State University).

This work was supported by the Oklahoma Agricultural Experiment Station, Oklahoma State University (project number OKL 02773).

The mention of trade names or commercial products in this publication does not imply recommendation or endorsement by Oklahoma State University.