R

ice (Oryza sativa L.) is a principal staple food crop for nearly half of the population in the world. The rice seed is an environment for rich diversity of microorganisms to thrive, hence its potential as a vehicle for transmission of beneficial or harmful bacteria as a planting material (Cottyn et al., 2009).

Global food demand is rising with increasing population (Win, 2008) but rice production has stagnated in recent years in some major rice-producing regions of Asia (Cassman et al., 2003). One of the principal reasons for the decline in rice yield is constituted by the diseases that affect rice production. Some of the major diseases that greatly affect rice are bacterial blight and bacterial leaf streak, caused by pathovars of Xanthomonas oryzae (Xo) which are Xanthomonas oryzae pv. oryzae (Xoo) and Xanthomonas oryzae pv. oryzicola (Xoc).

Through commercial seed trading and germplasm exchange activities of public and private institutions, seeds are exchanged throughout the world (Cottyn, 2002) and the movement of plant pathogens, through international borders has been made possible through these activities. Thus, rapid and accurate identification of pathogens are critical for the containment or elimination of pathogens and for regulatory reasons since incorrect diagnoses can result in large financial losses (Lang et al., 2010).

The causal agent of bacterial leaf blight and bacterial leaf streak in rice, Xoo and Xoc respectively, are quarantine yellow-pigmented bacteria regulations in numerous countries. However, due to the presence and similarity in terms of colony morphology of other seed-associated, non-pathogenic, yellow-pigmented bacteria to Xoo and Xoc, has been one of the main reasons of misdiagnosis by quarantine laboratories as well as a challenge for accurate detection (Sakthivel et al., 2001).

Reliance on one or a few unique features of diagnostic tools is risky due to the high degree of genotypic diversity and plasticity among many microbial pathogens. Specific fragments unique for specific pathogens will allow accurate detection and identification of pathogens. Lang identified specific diagnostic markers with the use of a comparative genomic pipeline. From a list containing numerous Xanthomonas oryzae species and pathovar-specific primers, four primers were selected and used in a multiplex PCR and optimized by IRRI, to detect Xanthomonas oryzae from rice seeds and differentiate the two pathovars present in this species, Xanthomonas oryzae pv. oryzae (Xoo) and Xanthomonas oryzae pv. oryzicola (Xoc) in a single multiplex PCR reaction.

The objective of this study was to determine the specificity of the IRRI-optimized multiplex PCR developed by Lang for the detection and differentiation of Xanthomonas oryzae pv. oryzae and Xanthomonas oryzae pv. oryzicola.

MATERIALS AND METHODS

Source of the Rice Seed Bacterial Isolates

One hundred nine (109) unidentified bacteria isolated from rice seeds accessed from the Seed Health Unit (SHU) were provided by the Host Plant Disease-Resistance Unit of IRRI. These isolates produced yellow colonies, similar to the colony morphology of Xanthomonas oryzae pv. oryzae and Xanthomonas oryzae pv. oryzicola on modified Wakimoto’s Medium Without Potato (WF-P) (Karganilla and Buddenhagen, 1972; Karganilla et al, 1973; Paulsen and Karganilla, 1973). These yellow colony-forming strains were grown on Suwa’s medium or modified Wakimoto’s Medium without Potato (WF-P) and were previously confirmed by IRRI to be non-pathogenic. Pure isolates were maintained and grown in 30℃.

Isolation and Standardization of Genomic DNA

Bacterial genomic DNA was extracted using the GES (Guanidium thiocyanate-EDTA-sarcosyl) method by Pitcher et al (1989) but modified by the Plant Pathology Division of IRRI. The quality and quantity of the isolated DNAs were checked by running them on a 1% (w/v) agarose gel with GelRed (10 μL/100 mL), along with several dilutions of λ DNA of known concentrations (e.g. 10, 20, 40, 80 ng/ μL) for comparison. The DNA bands were viewed under UV light using Alpha Imager Imaging System (Alpha Innotech Corp., USA).

BOX-PCR Amplification and Gel Electrophoresis

BOX-PCR DNA fingerprinting was done using the BOX A1R primer (5’-CTACGG CAAGGCGACGCTGACG- 3’) and PCR amplifications were done using the Programmable Thermal Controller TM (PTC-100 model, MJ Research Inc., USA). The BOX-PCR reaction mixture contained 1.25 μL of 5μM primer 1, 11.85 μL of sterile distilled water, 5.0 μL of 5x Gitschier buffer (83 mM (NH₄)₃SO₄; 335 mM, pH 8.8, Tris-HCl; 33.5 mM MgCl₂; 33.5 μM EDTA and 150 mM ß–mercaptoethanol), 0.2 μL Bovine Serum Albumin (20 mg/μL), 2.5 μL of Dimethylsufoxide (DMSO), 2.0 μL of 25mM dNTPs and 2.0 μL of genomic DNA. The PCR reaction profile consisted of: initial denaturation at 95℃ for 7 minutes, 30 cycles of denaturation at 94℃ for 1 minute, annealing at 53℃ for 1 minute and extension at 65℃ for 8 minutes and a cycle of final extension at 65℃ for 15 minutes.

Twenty μL of the amplified PCR product was analyzed using gel electrophoresis. The gel was composed of 0.75% Synergel^TM and 0.5% agarose and was electrophoresed for 15 hours at 50-55 volts. Sterile distilled water, DNAs of Xanthomonas campestris pv. dieffenbachia, Xanthomonas oryzae pv. oryzae and Xanthomonas oryzae pv. oryzicola were loaded in the 2nd, 3rd, 4th and 5th wells, respectively. The DNA fragments that were separated after the gel electrophoresis were stained for 2-3 hours with Sybersafe (10 μL/100 mL) diluted in 300 mL nanopure water and were viewed under the UV light and photographed using the Alpha Imager 4.1.0 System (Alpha Innotech Corporation, USA) connected to a computer.

Normalization of BOX-PCR Patterns and Cluster Analysis

The GelCompar version 3.5 software was used for normalization and cluster analysis of the unidentified isolates with similar morphological characteristics with Xoo and Xoc, Xoc and X. campestris pv. dieffenbachia. Normalization was done by aligning each fingerprint produced to a common reference template. Comparative analysis of BOX-PCR genomic fingerprints was performed by the Unweighted Pair Group Method with Arithmetic averages (UPGMA) using the Jaccard’s correlation coefficient to calculate the level of similarity between BOX-PCR patterns. Correlation levels were expressed in percentages of similarity in the dendrogram scale and upon normalization, the software produced a dendrogram from which cluster candidate isolates to be used for multiplex PCR amplification.

Multiplex PCR Amplification and Gel Electrophoresis

The multiplex PCR reaction mixture was composed of contained 1.6 μL 10 mM dNTPs, 2.0 μL 10x PCR buffer, 2.0 μL 25 mM MgCl₂, 0.5 μL Taq polymerase and 0.5 μL of each 10μM primer. Four primer pairs were used: 1 primer pair for Xanthomonas oryzae, that generated a 331 bp-fragment; 1 primer pair for Xanthomonas oryzae pv. oryzae that generated a 162 bp-fragment; 2 primer pairs for Xanthomonas oryzae pv. oryzicola that generated two fragments, 691 bp and 945 bp, respectively (Table 1). The PCR reactions involved an initial denaturing step at 94℃ for 3 minutes, denaturation for 30 seconds, annealing for 30 seconds at 64℃, extension for 2 minutes at 68℃ and 31 cycles of 94℃ for 30 seconds, 64℃ for 30 seconds and 68℃ for 2 minutes. Lastly, a final extension was done at 68℃ for 10 minutes. The PCR products gel electrophoresed at 110V for 2 hours. Alpha Imager System connected to a computer was used to view and photograph the bands on the gel.

RESULTS AND DISCUSSION

Colony Morphology of the Yellow Colony-Forming Rice Seed-Borne Bacteria

Pure isolates of 109 unidentified bacteria isolated from rice seeds accessed from the Seed Health Unit (SHU) produced yellow-pigmented colonies on Suwa’s medium or modified Wakimoto’s Medium without Potato (WF-P), similar to the colony morphology of Xoo and Xoc (Fig. 1).

DNA Fingerprinting of the Rice Seed-Borne Isolates using BOX-PCR

In order to determine genotypic diversity among the unidentified bacteria isolated from rice seeds, BOX-PCR analysis (de Brujin et al, 1996) was performed (Supplementary data 1). The dendrogram generated by the Gel- Compar ver. 3.5 software using the BOX-PCR fingerprints, revealed the genotypic diversity of the 109 different SHU isolates (Supplementary data 2). The majority of the isolates were widely distributed under different clusters, signifying that these isolates were different from each other and from Xoo and Xoc.

Forty one clusters were identified using the 49% similarity coefficient. This similarity coefficient was used for the selection of isolates to be subjected to multiplex PCR analysis since it was at this percent similarity level where Xoo and Xoc separated into two different clusters. The dendrogram data, thus, indicated that rice seed-associated/inhabiting bacteria isolated from rice seeds constituted a relatively vast, diverse, genotypic range of yellow-colony forming bacteria.

Multiplex PCR amplification of SHU isolates

Forty one clusters were identified and one isolate for each cluster of the dendrogram was selected using 49% similarity coefficient as the reference. An additional 10 SHU isolates were added, which were the isolates nearest to the position of Xoo and Xoc in the dendrogram (Supplementary data 1). A total of 51 SHU isolates were, thus, selected and used for the determination of the specificity of the PCR primers developed by Lang as well as the multiplex PCR optimized by IRRI.

As for the positive control strains, PXO 99 (Xanthomo- nas oryzae. pv. oryzae and BLS 297 (Xanthomonas oryzae pv. oryzicola) were used and G 278 (Xanthomonas campestris pv. dieffenbachia) and sterile distilled water as the negative controls. On lanes 1 and 5 are the 1kb DNA ladder, and on lanes 2, 3 and 4 are BLS 297 (Xoc), PXO 99 (Xoo) and sterile distilled water, respectively. Two Xocspecific amplicons, 691 bp and 945 bp bands, were generated for BLS 297 (lane 2), while a 162-bp fragment specific for Xoo was resolved for PXO 99 (lane 3). In addition, a 331 bp-fragment was amplified for both Xoo and Xoc, confirming that these belong to Xanthomonas oryzae (Fig. 2). For all of the 51 SHU isolates, no amplification of the Xo, Xoo or Xoc bands was observed (Fig. 3). This result implied that all of the isolates were not Xanthomonas oryzae strains and were neither Xo, Xoo nor Xoc.

Smears were, however, found at the bottom of each lane of the SHU isolates but these were present in all the other lanes in the multiplex PCR assay, including those of the controls. These could not be signs of false negatives because the Xoo and Xoc strains showed very clear and highly visible amplified bands, in contrast to the SHU isolates in which no such bands were observed.

The multiplex PCR assay validated in this study was, thus, determined to be specific and appears to be a robust method to distinguish Xo, Xoo and Xoc strains from other yellow colony-forming bacteria isolated but non-Xo isolates present on rice seeds. One of the recommendations that can be made after conducting this study is to sequence at least three of the non-pathogenic, seed-associated bacteria present in high frequencies in rice seeds, as well as those genotypically closest to the Xoo or Xoc clusters in the dendrogram, in order to design a specific primer or primers that can detect non-pathogenic, yellow colony- forming bacteria similar to Xoo and Xoc and to test more geographically diverse yellow colony-forming bacteria similar to Xoo and Xoc to further test the specificity of the multiplex PCR used and to disseminate this technology to different quarantine laboratories for the determination of the repeatability and reproducibility of the results.

적 요

종자 등 유전자원의 국가간 교류는 새로운 병원균을 유입시 켜 새로운 병을 일으키고 경제적인 손실을 극대화 시킬수 있 는 원인을 제공하기 때문에 국가별로 검역이 더욱 강화되고 있는 실정이다. 그러한 연유로 국가별로는 자국에 존재하지 않 은 유사한 병원균을 발견할 경우 종자도입을 막고 있어 불필 요한 경제적 손실원인이 되기도 한다. 따라서 이러한 손실을 막기위해 각국에 검역소에서는 병원균의 형태적인 감별보다 더 정확한 검정기술이 시급한 실정이다. 본 연구에서는 다중 PCR을 통하여 한번의 PCR 증폭으로 벼종자에 존재하는 벼흰 잎마름병을 일으키는 Xanthomonas oryzae pv. oryzae (Xoo) 과 세균성 줄무늬병을 일으키는 Xanthomonas oryzae pv. oryzicola (Xoc) 를 검출하고 이 두병원균을 판별할수 있는, 국제미작연구소가 최적화한 다중PCR 검정방법의 실질적 적용 성 여부를 검증하기 위한 연구였다. 그 결과 51개의 비병원성 이나 Xoo 또는 Xoc 와 유사하며 벼종자에서 분리된 균들은 다중PCR검정에서 어떤 밴드도 증폭하지 않은 반면, 대조구 (positive control)에서는 알맞은 band size의 amplicon들을 증 폭해냈다. 따라서 본 연구에 사용된 국제미작연구소에서 최적 화된 다중 PCR방법 (IRRI-optimized multiplex PCR)은 벼종 자에서 발견할수 있는 Xoo와 Xoc 의 존재 여부 뿐만 아니라 이들과 형태적으로 유사하나 비병원성균으로부터 구분해 내는 데 탁월한 방법이었음을 확인할 수 있었다.