Morphological and molecular characterization of the rice root-knot nematode, Meloidogyne graminicola, Golden and Birchfeild, 1965 occurring in Zhejiang, China

TIAN Zhong-ling, Munawar Maria, Eda Marie Barsalote, Pablo Castillo, ZHENG Jing-wu,

1 Laboratory of Plant Nematology, Institute of Biotechnology, College of Agriculture and Biotechnology, Zhejiang University,Hangzhou 310058, P.R.China

2 Institute for Sustainable Agriculture (IAS), Spanish National Research Council (CSIC), Campus de Excelencia Internacional Agroalimentario, Córdoba 14004, Spain

3 Key Lab of Molecular Biology of Crop Pathogens and Insects, Ministry of Agriculture, Hangzhou 310058, P.R.China

Abstract The rice root-knot nematode Meloidogyne graminicola is a severe pest of rice. In China, it was flrst reported from Hainan Province, and later from several other provinces. In the present study, a rice root-knot nematode population found from the rice cultivation areas of Zhejiang Province, China is characterized via molecular analysis using internal transcribed spacer(ITS) and cytochrome c oxidase subunit II (coxII)-16S rRNA genes and scanning electron microscopy (SEM) observations of males and the second-stage juveniles. Morphometric data and molecular sequence comparisons for all M. graminicola populations occurring in China are also provided. The overall morphology of M. graminicola found in Zhejiang match well with the original description, though males have a slightly longer body and stylet, and a shorter tail, while the second-stage juvenile is also slightly longer than in the original description. This is the flrst report of M. graminicola from Zhejiang. Phylogenetic studies based on coxII suggest that all the Chinese populations belong to Type B. This study expands knowledge of the increasing distribution and phylogenetic relationships of M. graminicola that occur in China.

Keywords: Meloidogyne graminicola, morphology, morphometric, molecular, rice, rDNA, cytochrome c oxidase subunit II(coxII), China

1. Introduction

The rice root-knot nematode,Meloidogyne graminicola,Golden and Birchfleld, 1965, is a major pathogen of rice(Oryza sativaL.) and causes signiflcant yield losses in both upland and flooded flelds (Bridgeet al. 1990). Due to its potential damage to crops, this nematode is considered a quarantine pest in many countries (Bridge and Page 1982;Pokharelet al. 2010).

M.graminicolais an obligate sedentary endoparasite(Abadet al. 2003). Root pathology after infection includes the formation of galls localized at the root tips, and females laying eggs within the root cortex. After hatching, the second-stage juveniles (J2s) begin migrating within the root,allowing them to multiply even under flooded conditions(Vanholmeet al. 2004; Dubreuilet al. 2007). The nematode has a short life cycle that can be completed in only 15 days at 27–37°C (Jaiswalet al. 2011). Consequently, the presence of even a small number ofM.graminicolaat planting can lead to an increase to a high population density during a single crop cycle (Shresthaet al.2007).

M.graminicolahas been reported to primarily parasitize irrigated and rainfed rice (Richet al.2009; Pokharelet al.2010) in South and Southeast Asian countries such as China,India, the Philippines, Burma, Bangladesh, Laos, Thailand,Vietnam, and Nepal (Upadhyayet al. 2014). Damage due toM.graminicolawas reported to cause 28–87% of grain yield loss in upland rice in Indonesia (Netscher and Erlan 1993), 16–20% loss in lowland rainfed rice in Bangladesh(Padghamet al. 2004), 16–32% loss in irrigated and 11–73%loss in flooded rice in India (Sorianoet al. 2000).

In China, Zhaoet al. (2001) reported the flrst discovery ofM.graminicolafrom Hainan Province, where it was found associated with the roots ofAllium fistulosum. More than a decade later, Liuet al. (2011) found this nematode associated with rice and weeds growing near major rice cultivation areas of Fujian Province. By 2017,M.graminicolahad been reported from different provinces of China, and was described in association not only with rice but with other hosts as well. Longet al. (2017) reported a population ofM.graminicolafrom soybean (Glycine max) in Hainan,while the Guangdong (Htayet al. 2016), Hubei (Wanget al.2017), Hunan (Songet al. 2017), and Jiangsu (Fenget al.2017) populations were reported from samples of rice root and soil. In the present study, this nematode was found in Zhejiang Province where it parasitizes several rice flelds.

Due to the increasing distribution ofM.graminicolain China and the associated potential for damage to rice crops, the objectives of the present study were to: 1) provide additional morpho-biological and morphometric data for the second-stage juveniles and males, as well as scanning electron microscopy (SEM) observations; 2) compare morphometric analyses of all reported Chinese populations;and 3) phylogenetic reconstruction of internal transcribed spacer (ITS) andcoxII-16S rRNA genes to examine genetic diversity.

2. Materials and methods

2.1. Nematode population sampling, extraction, and morphological assessment

Rhizosphere soil and root samples were collected from irrigated rice flelds in the Shangjing Village of Jinhua,Zhejiang Province, China. Samples were washed and nematodes were extracted following the sieving and decanting method of Brown and Boag (1988). All nematodes were stored at 4°C for morphological and molecular examination. Specimens ofM.graminicolafrom Hubei and Hunan were requested through personal communication for molecular studies. Nematodes were carefully examined under the stereomicroscope and individually chosen for further morphological assessment.For morphometric studies, nematodes were killed and flxed in hot formalin and processed in glycerin, following the method of Seinhorst (1959) as modifled by De Grisse(1969). The measurements and light micrographs of nematodes were completed with a Nikon Eclipse Ni-U 931845 compound microscope (Nikon, Japan). Perineal patterns of mature females were prepared according to Hartman and Sasser (1985). Adult females were gently removed from root tissues and teased apart with tweezers and needle. Lip and neck regions were excised, and the posterior end was cleared in a solution of 45% lactic acid.The perineal pattern was trimmed and transferred to a drop of glycerin for microscopic observations.

For SEM examination, nematodes were flxed in a mixture of 2.5% paraformaldehyde and 2.5% glutaraldehyde,washed three times in 0.1 mol L–1cacodylate buffer, postflxed in 1% osmium tetroxide, dehydrated in a series of ethanol solutions, and critical point-dried with CO2. After mounting on stubs, the samples were coated with gold.

2.2. DNA extraction, PCR and sequencing

DNA samples were prepared according to Zhenget al.(2003). Three sets of primers (synthesized by Invitrogen,Shanghai, China) were used in the PCR analyses to amplify the ITS, D2-D3 expansion segments of 28S of rRNA and thecoxIIregion. Primers for ampliflcation of ITS were TW81 (5′-GTT TCC GTA GGT GGT GAA CCT GC-3′) and AB28 (5′-ATA TGC TTA AGT TCA GCG GGT-3′) (Curranet al. 1994).Primers for ampliflcation of the D2-D3 28S rRNA gene were the forward D2A (5′-ACA AGT ACC GTG AGG GAA AGT TG-3′) and the reverse D3B (5′-TCG GAA GGA ACC AGC TAC TA-3′) (De Leyet al. 1999). The mitochondrial fragment between the partialcoxIIand the partial 16S was amplifled using the forward primer C2F3 (5′-GGT CAA TGT TCA GAA ATT TGT GG-3′) and the reverse primer 1108 (5′-TAC CTT TGA CCA ATC ACG CT-3′) (Powers and Harris 1993). PCR conditions were as described by Yeet al. (2007) and Fanelliet al. (2017). PCR products were evaluated on 1% agarose gels stained with ethidium bromide. PCR products were purifled and sent for sequencing by Invitrogen (Shanghai,China). The newly obtained sequences were submitted to GenBank under the accession numbers indicated on the phylogenetic trees.

2.3. Phylogenetic analyses

Newly obtained sequences of ITS rDNA andcoxII-16S rRNA, and available sequences from other nematodes were obtained from GenBank and used for phylogenetic reconstructions ofMeloidogynespecies. Outgroup taxa for the dataset were chosen according to previously published data (Htayet al. 2016; Fanelliet al. 2017). Sequences were compared using BLAST homology, and sequence alignments were manually edited using BioEdit (Hall 1999).

Phylogenetic trees inferred from the ITS dataset and thecoxII-16S rRNA was performed with the Maximum Likelihood(ML) method using MEGA version 6 software (Tamuraet al.2013). The phylograms were bootstrapped 1 000 times to assess the degree of support for the phylogenetic branching.Heuristic search with Nearest-Neighbor-Interchange (NNI)and the complete deletion option were used to remove positions with gaps and missing data.

3. Results

3.1. Distribution of the rice root-knot nematode in China

M.graminicolahas previously been reported from Fujian,Guangdong, Hainan, Hubei, Hunan, and Jiangsu provinces,and in the present study, this nematode was reported for the flrst time in Zhejiang Province.

3.2. Field symptoms and morphological assessment

Aboveground symptoms ofM.graminicolaincluded patches in the fleld, and poor plant growth, with stunted appearance and chlorotic leaves; belowground symptoms appeared as swollen and hooked root tips on affected root systems (Fig. 1). Dissection of galled root revealed that they contained both females and males. Females were pear-shaped and were found embedded within the cortical layer of the root. The perineal pattern of the females appeared oval-shaped with dorsal round arches, smooth striae, sometimes with a few lines converging at either end of the vulva, and lateral flelds obscure or absent. Light and scanning electron micrographs of J2, female, and male nematodes are shown in Figs. 2–4. The comparative morphometrics of the Zhejiang population ofM.graminicolaand other Chinese populations are presented in Tables 1–3.

The overall morphology and morphometrics of the Zhejiang population ofM.graminicolacorrespond well with the original description. Of theM.graminicolapopulations reported from China, female morphometrics were provided only for the Fujian, Hainan, Hunan, and Jiangsu populations,and each of them matched the original description with the exception of slightly longer stylets.

Descriptions of males were reported in all Chinese populations except for the populations from Hainan (Longet al. 2017) and Hubei (Wanget al. 2017). The overall morphology of males from the Zhejiang population also agreed with the original description except for a slightly longer body of 1 270 (1 043.4–1 553.4) μmvs. 1 222(1 020–1 428) μm, a longer stylet 17.2 (15.2–19.0) μmvs.16.8 (16.2–17.3) μm, a slightly smaller tail 9.2 (8.1–10.3)μmvs. 11.1 (6.1–15.1) μm and a shorter spicule 21.1(20.1–22.0) μmvs. 28.1 (27.4–29.1) μm. However, this small morphometric variation is within the range of species variation. The morphometrics of the males of the Zhejiang population correspond well with those of other Chinese populations except for the Hunan population, which had slightly elongated bodies, stylets, and spicules compared to those of the other Chinese populations.

Fig. 1 Symptoms caused by Meloidogyne graminicola in severely infested rice flelds of China. A, lowland rice flooded fleld in Zhejiang Province. B, upland rice drought fleld in Hunan Province with apparent chlorosis (arrow indicating hook-shaped root galls).

Fig. 2 Light photomicrographs of the Meloidogyne graminicola perineal pattern. A, entire female body. B and C, oval shaped perineal pattern. D, anus. E, dorsal arch. F, smooth striae.

Fig. 3 Light photomicrographs of the second-stage juveniles (J2s) and adult males of Meloidogyne graminicola. A, entire body of J2. B, pharyngeal region of J2 with arrows showing position of excretory pore. C, tail region of J2 with arrow showing position of anus. D, entire body of male. E, anterior region with arrow showing position of dorsal pharyngeal gland oriflce of male. F,posterior pharyngeal region of male with arrow showing position of excretory pore. G, male testis. H, male tail. I and J, lateral lines. Scale bars: A, D=100 μm; B, C, E–J=10 μm.

The morphology and morphometrics values of J2 agreed with the original description, though they had slightly longer bodies: 456.7 (402.7–509.0) μmvs. 441 (415–484) μm. The other Chinese populations matched the original description except for the slightly longer stylet length found in the Fujian(13.7 (13.0–15.0) μm), Hainan (14.3 (13.5–15.6) μm), and Hunan (14.0 (13.2–15.5) μm) populations, respectively.Similarly, the tail was slightly longer in the Hainan (72.9 (60.0–85.0) μm) and Hunan (73.7 (68.4–85.3)μm) populations as compared to the original description. The morphometrics of J2 in the Zhejiang population correspond well with those of the other Chinese populations. Overall, the morphometrics are within the range of species variation ofM.graminicola.

SEM observation had not been previously provided for any Chinese population. SEM observation of males and juveniles of the Zhejiang population (Fig. 4) showed that the oral opening is slit-like and is surrounded by six inner labial sensilla. The labial disc is ovoid shaped and is slightly elevated above the medial lips, which are roughly bean shaped. The labial disc and medial lips are of similar shape in males and juveniles. The amphidial apertures appear as elongate ovals in juveniles but are slit-like in adult males.The lateral fleld is smooth in juveniles but forms ridges in males. The body annulations of juveniles are interrupted or deviate slightly from their course as they pass through the anal opening, which is a small and rounded hole in the cuticle. The cloacal opening in males is transversely elongate, and the annulations on the anterior and posterior side of the opening are indistinct or absent.

Fig. 4 Scanning electron micrographs (SEM) of the second stage juveniles (J2s) and adult males of Meloidogyne graminicola. A and B, lip region of J2. C, lip region of male.D, lateral fleld at midbody of J2. E, anus region of J2. F, lateral fleld at midbody of male. G and H, male tail. I, tail of J2. Arrows show the position of anus. Scale bars: 2 μm.

3.3. Molecular characterization and phylogenetic relationship of the M. graminicola populations

The sequence size of ITS,coxII-16S rRNA, and D2-D3 region of 28S are 579, 531, and 766 bp, respectively.All the amplifled sequences were identical and showed 99–100% similarity withM.graminicolasequences present in GenBank. Five new sequences were obtained for theM.graminicolaZhejiang population and deposited in GenBank under accession numbers KY660542–KY660543 for ITS,KY660544–KY660545 for D2-D3 of 28S, and MG356945 forcoxII-16S rRNA genes. Additionally, twocoxII-16S rRNA sequences were obtained for the Hubei and Hunan populations and were deposited in GenBank under accession numbers MH033621 and MG356944, respectively.

The ITS ML tree (Fig. 5) was constructed using 30 sequences ofM.graminicolapopulations occurring in China and other countries, along with 22 sequences from otherMeloidogynespecies andPratylenchus penetrans(FJ99117) as the outgroup. All the populations ofM.graminicolaformed a well-supported group (94%) within other species ofMeloidogyne. However, the sequences obtained from different countries does not group together according to geographic location, suggesting intraspecies population variation. Three Chinese populations, Hunan(KX461934), Fujian (HQ420902), and Taiwan (KJ57283), grouped with Indian populations (HM623443, HM581973 and JF949754) while another population from Fujian(KR604730) and Guandong (KR604731-32) grouped with the Myanmar (KR604740, KR604734, KR604734,KR604736, and KR604739) populations. Only the Hainan(KU646999) population appeared between two USA(JN241866-67) populations. Four Chinese populations,Zhejiang (KY660543-42), Jiangsu (KY250090), Hubei(KY346887), and Fujian (HQ420903), appeared between the Myanmar populations. The twoM.trifoliophilapopulations from Australia (AF077091) and New Zealand(JX465593) clustered withM.graminicolapopulations from Italy (LT669809–LT669810) and the USA (EF432571).Similarly, the fourM.graminicolapopulations from the USA (DQ909025, DQ909031, DQ909047, and EF432570)and one Indian population (KM921775) appeared as a basal subclade in theM.graminicolaclade and showed comparatively long branches, suggesting a sequence variation amongM.graminicolapopulations. Moreover,three Fujian populations (KR234082, KM236560, and KM111531) do not appear with other Chinese populations.

Table 1 Comparative morphometric analysis of adult female Meloidogyne graminicola from China

Table 2 Comparative morphometric analysis of adult male of Meloidogyne graminicola in China

Table 3 Comparative morphometric analysis of the second-stage juvenile (J2) of Meloidogyne graminicola in China

Fig. 5 Maximum likelihood tree showing the phylogenetic relationships of Meloidogyne graminicola based on internal transcribed spacer (ITS) rDNA sequences. Numbers indicated on nodes are bootstrap values for each cluster based on 1 000 permutations.Bold means the sequences produced in this study.

The ITS sequence identities of the Zhejiang population showed 99% similarity (1 nucleotide difference) with all reported populations from China except Jiangsu(2 bp difference), Taiwan (3 bp), and those three Fujian populations (5–6 bp) which did not cluster with the other Chinese populations.

ThecoxII-16S rRNA ML tree (Fig. 6) was constructed from 17 sequences ofM.graminicolapopulations occurring in China and other countries, along with 23 sequences ofMeloidogynespecies. All theM.graminicolapopulations formed a well-supported group (99%), with the two populations from USA (AY757885-84) and from Italy(LT669811–LT669812) appearing as a basal clade. The Fujian (KR234081) and USA (JN241926) populations showed slightly longer branches, suggesting a sequence variation amongM.graminicolapopulations. To determine if there was any genetic variability among the Chinese populations, the newly described Zhejiang population was compared with the other populations. All the Chinese populations ofM.graminicolashowed only one nucleotide difference with the Zhejiang population except for the Fujian(KR234081), Hubei (MG356943), and Hunan (MG356944)populations which showed 2, 4, and 2 nucleotide differences,respectively.

No intra-speciflc variation was found between D2-D3 regions of 28S sequences of all the Chinese populations.Thus, this marker was not used for phylogenetic analysis or for comparison of sequence similarities.

4. Discussion

Fig. 6 Maximum likelihood tree showing the phylogenetic relationships of Meloidogyne graminicola based on coxII-16S rRNA gene sequences. Numbers indicated on nodes are bootstrap values for each cluster based on 1 000 permutations. Bold means the sequences produced in this study.

In the present study, molecular identiflcation of Chinese populations ofM.graminicolawas made based on ITS-rDNA, D2-D3 of 28S, andcoxII-16S rRNA. In the ITS tree,M.trifoliophilapopulations from Australia and New Zealand clustered withM.graminicolapopulations, suggesting this might be the same species. Different populations from Myanmar, the USA, and India did not group together,indicating a signiflcant level of intra-population variability.This was also observed among the Chinese populations such asM. graminicolaFujian populations, three of them occupied basal positions to other Chinese populations.The ITS tree showed different geographical groupings ofM.graminicolawhich may represent different evolutionary origins. In thecoxIItree, theM.graminicolaclade divided into two subclades based on the type of mitochondrial DNA.McClureet al. (2012) designed two types of mitochondrial DNA for the presence ofDraIandSspI restriction sites.Use of the different types of mitochondrial DNA aids in quick identiflcation, and works well to identify different species. All the Chinese and several of the USA populations(JN241939, JN241927, JN241929, and JN241926) formed one clade and belonged to type B, while the other clade containing the Italian (LT669811–LT669812) and remaining USA (AY757884–AY757885) populations belonged to type A (McClureet al. 2012; Fanelliet al. 2017).

A low level of nucleotide difference indicated a lack of genetic diversity between ChineseM.graminicolapopulations. Similar results were reported by Htayet al.(2016), where a low degree of sequence dissimilarity among Chinese populations was also observed. There has no clear evidence to indicate whether this species is native or introduced in China. Phylogenetic studies (Fig. 6) revealed that the most of the Chinese populations grouped together,and this nematode may have continued to spread from one initial introduction location to different localities by means of either infected planting material or by unhygienic agronomic practices.

In addition to the use of molecular identiflcation, variation in morphological characters such as the length of stylet,spicule, and vulva are considered to have diagnostic importance for identiflcation ofM.graminicola(Salaliaet al.2017). However, previous studies have shown that morphometrics among and within populations did not correlate with the geographic source of the population(Pokharelet al. 2010). Our results agree with those of Salaliaet al. (2017) who demonstrated that for identiflcation ofM.graminicola, molecular analyses are more reliable,because observed differences in morphology and slight morphometrical variations could be due to geographic distribution. Hence, we consider that the small morphometric variation observed among the Chinese populations could be species variation or phenotypic plasticity commonly exhibited by nematodes (De Oliveiraet al. 2017).

5. Conclusion

The present study provides a detailed morphological and molecular characterization of the population ofM.graminicolafound in Zhejiang Province, China, together with SEM observations of adult males and the secondstage juveniles. Furthermore, this study compiles all the host records and provides comprehensive morphometric and molecular analysis of all populations of ChineseM.graminicola. Our research expands available information on the increased distribution and phylogenetic relationships of different populations ofM.graminicolaoccurring in China. Additionally, the molecular sequencing data and comparative morphometry obtained in this study will help future identiflcation ofM.graminicola.

Acknowledgements

This research was supported by the Special Fund for Agro-scientiflc Research in the Public Interest in China(201503114). The authors thank Dr. Rong Nianhang, from the Center of Electron Microscopy, Zhejiang University,China for providing assistance in preparation of SEM.