Introduction

Root-knot nematodes (RKN) are microscopic worms that live in soil and feed on the roots of many crops and weeds. The nematode gets its name because its feeding causes galls to form on the roots of infected plants. They are sedentary endoparasitic nematodes that depend on the induction of a permanent feeding site in living roots to complete their life cycle. RKN are the most widespread and serious plant-parasitic nematode pests, damaging a very wide range of crops throughout the world1. They are scientifically classified in the genus Meloidogyne (Tylenchida: Meloidogynidae) with over 100 species described2.

The southern RKN, M. incognita (Kofoid & White) Chitwood, 1949, is the most important nematode parasite of cotton in Arkansas3 and it has replaced the soybean cyst nematode as the premier nematode pest of soybean4. RKN are also commonly found in corn and grain sorghum fields and are associated with various horticultural and ornamental crops and turf grasses in the state. Because of the significance of the agricultural production to Arkansas’ economy5, understanding the Meloidogyne species associated with crops in the state is vital to formulation of effective and sustainable management strategies.

Previous surveys of RKN in Arkansas were conducted by using classical morphological methods. In a few surveys from soybean6, cotton7, wheat8 and blueberry9, RKN were found but species identification was not attempted. Meloidogyne graminis (Sledge & Golden, 1964) Whitehead, 1968 was first found in 1967 by R. D. Riggs on Zoysia spp. in Arkansas10. Meloidogyne hapla Chitwood, 1949 was reported on black locust (Robinia pseudoacacia) near the Mississippi River in Arkansas11. Norton et al.12 documented the occurrence of M. arenaria (Neal, 1889) Chitwood, 1949, M. hapla, and M. incognita in Arkansas. Wehunt et al.13 reported M. incognita, M. hapla, M. arenaria, M. graminis, and M. javanica from soybean fields near the Mississippi river. Elmi et al.14,15 recorded M. marylandi Jepson & Golden in Jepson, 1987 from tall fescue. Walters and Barker16 reported M. hapla, M. incognita, M. arenaria, and M. javanica (Treub, 1885) Chitwood, 1949 in Arkansas. In a recent survey from 106 soil and root samples, M. incognita, M. marylandi, M. haplanaria, M. hapla, M. arenaria and M. partityla Kleynhans, 1986 were identified through molecular diagnosis and M. incognita was the most abundant species17.

Development of resistant varieties that suppress nematode growth and reproductions is the most desirable, cost-effective and environmentally sustainable strategy for managing plant-parasitic nematodes18. Host plant resistance is effective against certain species or races; thus, accurate identification of RKN species is critical to the success of the use of host resistance or rotation. Species of RKN has been traditionally identified based on female perineal pattern, second-stage juvenile and male morphology and morphometrics, isozyme analysis, and host differential test. The traditional methods are always challenging due to highly conserved and similar morphology across species, lack of certain life stages, high intraspecies variability, potential hybrid origin and polyploidy19. In the past 20 years, molecular tools have been progressively developed to identify RKN species using polymerase chain reaction (PCR), Restriction Fragment Length Polymorphism (RFLP), and DNA sequencing, because they are usually fast, sensitive, less subjective and applicable to any life stages of a population19,20,21,22,23,24,25. The objective of this study was to collect RKN samples from field crops and natural sites in the state of Arkansas and to characterize the DNA sequences of RKN on the ribosomal DNA 18S-ITS-5.8S, 28S D2/D3 and mitochondrial DNA cytochrome oxidase gene subunit II-the intergenic spacer (CoxII-IGS) to determine the species and their distribution.

Results

RKN problem in Arkansas

RKN are common in field samples submitted to the Arkansas Nematode Diagnostic Laboratory. Infected roots have typical gall formation and RKN females, juveniles and egg masses could be recovered from the galled tissues (Figs 1 and 2). Meloidogyne marylandi does not produce galls on turfgrasses, and only semi-penetrates the roots (Fig. 3A). The female is lemon-shaped, with a much harder cuticle and a slightly protruding vulva-anus region (Fig. 3B), that is different from the pear-shaped female and rounded vulva-anus region in other common RKN living inside the galls (Fig. 1).

Figure 1
figure 1

Photographs of root galls and females of southern root-knot nematode (Meloidogyne incognita) from tomato in Pulaski, Arkansas (RT122).

Full size image
Figure 2
figure 2

Photographs of the infested potato and the juveniles of southern root-knot nematode (Meloidogyne incognita) from potato in Van Buren County, Arkansas (RT139).

Full size image
Figure 3
figure 3

Photographs of females of Maryland root-knot nematode (Meloidogyne marylandi) from Sedge like grass in Washington County, Arkansas (RT106). (A) Female on the root. (B) Female.

Full size image

RKN identification

Five RKN species were identified including M. incognita, M. hapla, M. haplanaria, M. marylandi and M. partityla; the results are presented in Table 1. Species identification in this study was based on the combined analysis of DNA sequencing on the rDNA 18S-ITS-5.8S, 28S D2/D3 and CoxII-IGS (Table 1) and PCR by species-specific primers (Table 2). Meloidogyne incognita, the most prevalent species, was found in 232 samples (95%) from soybean, cotton, corn and various vegetables in 36 of the 39 counties from which samples were collected (Ashley, Bradley, Chicot, Clay, Cleburne, Columbia, Conway, Craighead, Crittenden, Cross, Desha, Drew, Garland, Greene, Jackson, Jefferson, Johnson, Lafayette, Lawrence, Lincoln, Logan, Lonoke, Miller, Mississippi, Montgomery, Phillips, Pope, Prairie, Pulaski, Randolph, Saline, Sebastian, Van Buren, Washington, Woodruff, and Yell) (Fig. 4). Meloidogyne hapla was found in only one sample from rose in Craighead County (Fig. 5). Meloidogyne haplanaria was found in seven samples from okra, tomato, peanut, Indian hawthorn, ash, willow and elm trees in Baxter, Faulkner, Logan, Saline, Van Buren, and Washington counties (Fig. 6). Meloidogyne marylandi was found in three samples from grasses in Hempstead, Logan, and Washington counties (Fig. 7). Meloidogyne partityla was found in only one sample from pecan in Logan County (Fig. 8). There were no samples with mixtures of species found.

Table 1 Species and isolates of root-knot nematodes (Meloidogyne spp.) sequenced in the present study.
Full size table
Table 2 Primers used for polymerase chain reaction and DNA sequencing.
Full size table
Figure 4
figure 4

Distribution of southern root-knot nematode (Meloidogyne incognita) in Arkansas.

Full size image
Figure 5
figure 5

Distribution of Texas peanut root-knot nematode (Meloidogyne haplanaria) in Arkansas.

Full size image
Figure 6
figure 6

Distribution of northern root-knot nematode (Meloidogyne hapla) in Arkansas. Logan and Washington counties were from results by Khanal et al.17.

Full size image
Figure 7
figure 7

Distribution of Maryland root-knot nematode (Meloidogyne marylandi) from Arkansas. Drew, Craighead and Perry counties were from results by Khanal et al.17.

Full size image
Figure 8
figure 8

Distribution of pecan root-knot nematode (Meloidogyne partityla) from Arkansas.

Full size image

DNA sequencing

The rDNA 18S-ITS-5.8S (182 sequences), 28S D2/D3 (226 sequences) and CoxII-IGS (197 sequences) were deposited in GenBank and their GenBank accession numbers are presented in Table 1. Although attempts were made to perform DNA sequencing on all three genes for each sample, not all PCR or DNA sequencing was successful. However, at least one gene was sequenced from all RKN populations except for one population (TK42). One hundred forty-two samples (58.2%) have all three genes sequenced. Many of the sequences from different populations are identical, thus their sequences were assigned the same accession number. Minor DNA sequence variations within the same species were observed in each gene among some populations.

DNA sequences of MK102775 (1,980 bp), MK102776 (2,296 bp), MK102777 (1,227 bp), MK102778 (968 bp) and MK102779 (1,815 bp) are different regions of 18S-ITS-5.8S and have more than 99% identity with many sequences of M. incognita, M. javanica and M. arenaria from GenBank. MK102771 (2,180 bp), MK102772 (2,216 bp) and MK102773 (2,180 bp) matched with two sequences of M. haplanaria (AY919178, 637 bp and AY757867, 637 bp) with 100% identity in aligned region. These three sequences are 98–99% identical to many tropical species sequences including M. incognita, M. javanica and M. arenaria from GenBank. The 2,215-bp DNA sequence of 18S-ITS-5.8S (MK102780) is 99–100% identical to DNA sequences of M. hapla from the GenBank (KP901065, KJ636268, AY268119, AY593892, EU669941, EU669942, AY942628, MH011983, EU669943 and KJ636267). DNA sequences of MK102774 (2,015 bp) and MK102781 (790 bp) are 100% identical to M. marylandi (KP901041) and 99% identical to M. marylandi (KP901049 and KP901043).

The DNA sequence of 28S D2/D3 (MK102787, 1,006 bp) of M. incognita is fairly conserved; no sequence variation was observed among most Arkansas populations. It has minor nucleotide differences with other Arkansas sequences of M. incognita (MK102786, 1,006 bp, MK102788, 643 bp, MK102789, 641 bp, MK102790, 643 bp, and MK102791, 643 bp). Blast search of these sequences revealed 97–100% identity with many tropical species sequences including M. incognita, M. javanica and M. arenaria from GenBank (KP901082, KP901083, KP901078, etc.). The DNA sequences of 28S D2/D3 (MK102784, 1,003 bp and MK102785, 935 bp) on M. haplanaria are 95–96% identical to many tropical species sequences including M. incognita, M. javanica and M. arenaria from GenBank (KP901082, KP901083, KP901078, etc.). No 28S DNA sequence of M. haplanaria from GenBank is available to compare with the study populations. The 1,042-bp DNA sequence (MK102780) of M. hapla is 99–100% identical to DNA sequences of M. hapla from GenBank (GQ130139, KU180679, KP306534, KP306532, KU587712, KP901086, DQ145641, KJ598136 and KJ755183). The 678-bp DNA sequence (MK102782) of M. marylandi is close to many sequences of M. marylandi (KP901066 etc.) with 99–100% identity. The 667-bp DNA sequence of M. partityla (MK102783) is 94% identical to M. ethiopica (KY882483), M. hispanica (EU443606) and M. luci (LN626951). No 28S DNA sequence of M. partityla from GenBank is available to compare with the study population.

The DNA sequences of mitochondrial DNA CoxII-IGS of M. incognita (MK102798, 912 bp, MK102799, 879 bp, MK102800, 909 bp, MK102801, 771 bp, MK102802, 831 bp) are comprised of 138-bp CoxII and the rest IGS. The CoxII sequences are highly conserved and identical which encode a polypeptide GQCSEICGINHSFMPILVEITLFDFFKLNLLTNWLFYFCWSKSKY. However, there are five types of IGS sequences that showed four significant gaps, six mutations and one insertion/deletion as shown in Fig. 9. Blast search of these sequences revealed 99–100% identity to many tropical species sequences including M. incognita, M. javanica and M. arenaria from GenBank (MH152335, MF043913, LN864824, etc.). The DNA sequences of CoxII-IGS (MK102793, 660 bp, MK102794, 541 bp and MK102795, 541 bp) on M. haplanaria are 99% identical to sequences of M. haplanaria (KT783539, KM881682, AY757905 and AY757906). The 470-bp DNA sequence (MK102792) of M. hapla is 99% identical to DNA sequences of M. hapla from the GenBank (KJ598134, AY757887, AY757888, AY757899, KP681265, KM881684 and KF993633). The 533-bp DNA sequence (MK102797) of M. marylandi is identical to sequence of M. marylandi (JN241918) and a few bp differences with other sequences of M. marylandi (JN241917, KM881683 and KC473862). The 511-bp DNA sequence of M. partityla (MK102796) is 99% identical to M. partityla (AY672412, AY757908, AY672413 and KM881686).

Figure 9
figure 9

Multiple alignment of CoxII-IGS gene in Meloidogyne incognita collected from Arkansas.

Full size image

Molecular phylogenetic relationships

A phylogenetic tree based on the rDNA 18S-ITS-5.8S is presented in Fig. 10 with two Pratylenchus species as outgroup taxa. This tree placed the study populations in three distinct groups. Meloidogyne incognita populations are in a clade with other tropical RKN species including M. incognita, M. arenaria, M. javanica, M. floridensis and M. morocciensis with 100% support. Meloidogyne haplanaria is sister to this clade with 100% support. Meloidogyne enterolobii is basal to this clade with 93% support. Meloidogyne marylandi and M. graminis are very closely related and are in a clade with M. spartinae with 100% support. Meloidogyne hapla is sister to M. microtyla with 100% support. Meloidogyne hapla and M. marylandi are in a monophyletic group with 100% support. Unfortunately, M. partityla from this study was not sequenced successfully.

Figure 10
figure 10

Bayesian consensus tree inferred from rDNA 18S-ITS-5.8S under GTR + I + G model (-lnL = 13647.8496; AIC = 27315.6992; freqA = 0.2616; freqC = 0.2077; freqG = 0.2494; freqT = 0.2813; R(a) = 1.2697; R(b) = 2.0864; R(c) = 1.6566; R(d) = 0.6843; R(e) = 3.1581; R(f) = 1; Pinva = 0.3599; Shape = 0.3398). Posterior probability values exceeding 50% are given on appropriate clades.

Full size image

A phylogenetic tree based on the rDNA 28S D2/D3 sequences is presented in Fig. 11 with two Pratylenchus species as outgroup taxa. This tree placed Arkansas RKN in four distinct groups. Meloidogyne hapla population RT83 (MN475814) is in a clade with M. hapla (KP901086). This clade is in a monophyletic clade with M. dunensis (EF612712) with 84% support. Meloidogyne incognita (MK102786-MK102791) and M. haplanaria (MK102784 and MK102785) are in a monophyletic clade with M. arenaria, M. javanica, M. incognita, M. konaensis, M. paranaensis, M. thailandica, M. enteroloii, M. hispanica, M. ethiopica and M. inornata with 100% support. Meloidogyne partityla is sister to this clade with 82% support. Meloidogyne marylandi (MK102782) is in a clade with M. marylandi (JN157852 and KP901066) and M. graminis (JN019331, KP901076 and KP901077) with 99% support.

Figure 11
figure 11

Bayesian consensus tree inferred from rDNA 28S D2/D3 under TVM + I + G model (-lnL = 5664.7959; AIC = 11347.5918; freqA = 0.2548; freqC = 0.1889; freqG = 0.2676; freqT = 0.2888; R(a) = 0.6653; R(b) = 3.0047; R(c) = 1.7303; R(d) = 0.3041; R(e) = 3.0047; R(f) = 1; Pinva = 0.2636; Shape = 0.6053). Posterior probability values exceeding 50% are given on appropriate clades.

Full size image

A phylogenetic tree based on the mitochondrial DNA CoxII-IGS sequences is presented in Fig. 12 rooted with M. partityla (MK102796) based on the multiple sequence alignment whose sequence is most distinct from the other sequences. No outgroup species was included in the analysis because of the large sequence divergency. This tree placed Arkansas RKN in five distinct groups. Meloidogyne partityla (MK102796) is at the basal position. Meloidogyne hapla population RT83 (MK102792) is in a clade with other M. hapla (AY757887, AY757888, KP681265, KM881684, KF993633 and AY757899). Meloidogyne haplanaria (MK102793-MK102795) is in a clade with other M. haplanaria (KT783539, KM881682, AY757905 and AY757906). This clade is sister to M. enterolobii with 100% support. Meloidogyne marylandi (MK102797) is in a clade with two other M. marylandi (JN241917 and JN241918). Meloidogyne incognita (MK102798- MK102802) is in a monophyletic clade with M. incognita, M. arenaria, M. javanica, M. luci, M. ethiopica, M. arabicida, M. lopezi, M. paranaensis, and M. izalcoensis with 98% support. This clade is sister to M. arenaria, M. morocciensis, M. thailandica and M. incognita with 100% support.

Figure 12
figure 12

Bayesian consensus tree inferred from mitochondrial DNA CoxII-IGS under TVM + G model (-lnL = 4936.4829; AIC = 9888.9658; freqA = 0.3513; freqC = 0.0315; freqG = 0.1032; freqT = 0.5139; R(a) = 2.3466; R(b) = 4.1635; R(c) = 1.2778; R(d) = 4.0003; R(e) = 4.1635; R(f) = 1; Pinva = 0; Shape = 0.7173). Posterior probability values exceeding 50% are given on appropriate clades.

Full size image

PCR by species-specific primers

The species identification of M. incognita was confirmed using PCR by M. incognita-specific SCAR primers Inc-K14-F/Inc-K14-R which produced a 399-bp DNA fragment (Fig. 13a) or Finc/Rinc which produced a 1200-bp PCR fragment (Fig. 13b). Only one population (RT83) is positive to primers MH0F/MH1R which were M. hapla-specific with 960-bp amplicon (Fig. 13b). None of these study samples were positive to primers Fjav/Rjav and Far/Rar which are species-specific to M. javanica and M. arenaria respectively. One population TK42 failed to get any good DNA sequencing results on three genes, but it is positive for M. incognita when using PCR by M. incognita-specific SCAR primers (Fig. 13).

Figure 13
figure 13

Photographs of an example of agarose gel electrophoresis of root-knot nematode (Meloidogyne spp.) from Arkansas by species-specific primers. (a) Primers Inc-K14-F/Inc-K14-R, M. incognita-specific. Lane A: TK3; B: TK42; C: TK156; D: TK196; E: TK206; F: RT131; G: RT128; H: Water negative control; 100 bp low scale DNA ladder. (b) A–D: primers Finc/Rinc, M. incognita-specific; E–H: primers MH0F/MH1R, M. hapla-specific. Lane A: TK3; B: TK42; C: TK190; D: RT137; E: RT83-female 1; F: RT83-female 2; G: VW9, M. hapla-positive control; H: Water negative control; 1 kb DNA ladder.

Full size image

Discussion

This study characterized DNA sequences on ribosomal DNA 18S-ITS-5.8S, 28S D2/D3 and a mitochondrial DNA CoxII-IGS on 244 RKN populations from various hosts, collected from 39 counties in Arkansas. Five species were identified, including M. incognita, M. hapla, M. haplanaria, M. marylandi and M. partityla through a combined analysis of DNA sequencing and PCR by species-specific primers. The phylogenetic relationships agreed broadly, i.e. sequences analysed were grouped into clades as reasonably expected with no contradictions irrespective of the three loci sequenced. Although DNA sequencing can determine M. hapla, M. haplanaria, M. marylandi and M. partityla by any of the three genes, it is impossible to determine M. incognita because these genes are too conserved among other closely related RKN as shown in blast search and phylogenetic trees. PCR by species-specific primers is needed for the identification of M. incognita. Unlike earlier surveys of the state, M. arenaria, M. javanica and M. graminis were not detected from any of the samples. One RKN population with the second-stage juveniles having very short tails was found in a sample collected at the Lon Mann Cotton Research Station near Brinkley, Arkansas. This sample was found below an oak tree in a mixture of grasses and dicot weeds. Several attempts to find females failed and no DNA study was ever performed. There were some RKN samples forwarded to the second author by the Arkansas Nematode Assay Service and by the Arkansas Plant Health Clinic that contained soil with little or no roots, thus only the second-stage juveniles were available. These second-stage juveniles were reared in a greenhouse using tomato and bermudagrass as possible hosts. While some success in producing a population of RKN resulted, most testing resulted in failure. This failure was disappointing in that two samples identified with the second-stage juveniles appeared to be M. arenaria17. Another failure was not establishing a RKN population when finding males along with the second-stage juveniles in grass samples in experimental plots from the main University Experiment Station in Fayetteville.

Meloidogyne incognita (Southern RKN) is the most abundant species and was identified in 95% samples. It was the only species found in field crops including soybean and cotton, except for one population of M. haplanaria from soybean in Logan County (TK201). This species has worldwide distribution and numerous hosts and is the most damaging species throughout the tropics and warmer regions of the world. Meloidogyne incognita is predominantly found in warmer climates, at latitudes between 35°S and 35°N26. This study revealed M. incognita is the most common and widespread species in field crops in Arkansas.

Meloidogyne hapla (Northern RKN) is widely distributed, particularly in temperate regions and the cooler, higher altitude areas of the tropics. Taylor & Buhrer27 reported that in the USA, M. hapla was most common north of 39°N. It is polyphagous and affects over 550 crops and weeds28 including many agricultural and horticultural plants (vegetables, fruits, ornamentals), but few grasses or cereals28. From the current and previous study17, this species was found from knockout rose, oak, elm and poke weed (Phytolacca americana) from three northern counties including Craighead, Logan, and Washington (Fig. 5), but not from any field crops.

Meloidogyne haplanaria (Texas peanut RKN) was originally found attacking peanut in Texas29 and was also reported from Arkansas17 and Mi-resistant tomato in Florida30. Host range studies revealed that it can parasitize several legumes and crucifer crops29 and infect M. arenaria-susceptible cultivars of peanut, garden pea and radish31. Although watermelon, cotton, corn, tobacco and wheat are nonhosts for M. haplanaria, peper, eggplant, soybean and common bean are moderate hosts for this nematode29,31. In our study, this species was found on ash, tomato, peanut, willow, elm, Indian hawthorn and soybean from six counties including Baxter, Faulkner, Logan, Saline, Van Buren and Washington (Fig. 6). It’s worthy to note that only one soybean field (TK201) had M. haplanaria. This species is distinct by mitochondrial DNA CoxII-IGS, but similar to M. incognita, M. arenaria and M. javanica in ribosomal DNA 18S-ITS and 28S D2/D3.

Meloidogyne marylandi (Maryland RKN) was first described by Jepson & Golden32 on bermudagrass (Cynodon dactylon) in College Park, Maryland, USA. It has been reported from Arkansas17, Texas33, Florida34, Oklahoma35, North Carolina, South Carolina36, Arizona, California, Nevada, Utah and Hawaii37. Outside USA, M. marylandi has been found in Japan38, Israel39, and Costa Rica40. From current and previous study17, this species was found from grasses from six counties including Craighead, Drew, Hempstead, Logan, Perry and Washington (Fig. 7). Another closely related species, M. graminis, is native to USA. It was first described infecting St. Augustine grass (Stenotaphrum secundatum) in Winter Haven, Florida, in 196441. This species has been reported on cultivated grasses from Florida to California and Hawaii, as far north as New England, on native grasses in the Konza Prairie in Kansas42,43, North Carolina, and South Carolina36. The M. graminis from grass reported in 1974 by Grisham et al.10 and in 1982 by Robbins6 was believed to be M. marylandi which was described much later in 198726. Before M. marylandi was described in 1987, no DNA analysis was available and species found from grass in Arkansas was assigned as M. graminis. Thus, no M. graminis is really confirmed in Arkansas.

Meloidogyne partityla (pecan RKN) is a plant pathogenic nematode infecting pecan. It was first described in pecan trees in South Africa by Kleynhans (1986)44. It is thought to have been introduced into South Africa by pecan seedlings that came from USA in 1912, 1939 and 194044. Today, this nematode is seen infecting pecan trees in Arizona45, Arkansas46, Florida47,48, Georgia49, New Mexico50, Oklahoma45, South Carolina51 and Texas52. In addition to pecans, they also infect the California black walnut (Juglans hindsii), English walnut (J. regia), shagbark hickory (Carya ovate), post oak (Quercus stellate), water oak (Quercus nigra) and laurel oak (Q. laurifolia). The health of infested trees continues to decline every year50. In this study, only one sample from pecan in Logan County was identified as M. partityla (Fig. 8).

Meloidogyne enterolobii (Guava RKN) is a recent emerging and highly pathogenic RKN species in the USA. It was originally described from China in 198353 and later reported in Florida in 200454, North Carolina in 201355, Louisiana in 201956 and South Carolina in 201957 attacking field crops, vegetables, ornamental plants, guava tree and weeds. Meloidogyne enterolobii is considered as a tropical species; due to its limited distribution and high damage impact, it was added to the European and Mediterranean Plant Protection Organization A2 Alert list58 and became a regulated nematode in South Korea, Costa Rica and USA (Florida, Louisiana, Mississippi, North Carolina)54,55,56,57,58,59,60. Fortunately, M. enterolobii was never detected in our survey and thus it is listed as a regulated species to prevent its disperse61.

In this study, DNA sequencing and PCR by species-specific primers were employed successfully to characterize and identify RKN from a wide range of plants from 39 counties in Arkansas. The results revealed the presence of five RKN species with M. incognita being the most predominant. Their hosts, distribution, DNA sequences of three genes and phylogenetic relationships were investigated. This study provides basic information for future management of these economically important species in Arkansas.

Methods

Nematode sample collection

A total of 244 RKN populations from various hosts from 39 counties in Arkansas were sampled in this study from 2014 to 2018 (Table 1) (Fig. 14). These samples were collected during the growing season. No specific permissions were required in sampling for plant-parasitic nematodes and no endangered or protected species were involved. Two hundred and six RKN samples (TK1-TK206) were initially collected from soil samples that were taken by Arkansas Cooperative Extension Service agents as a part of a statewide nematode survey sponsored in part by the Arkansas Soybean Promotion Board. Samples were collected during the period from September 1 – November 1 in 2014–2016 and were from fields that were either in soybean in the year they were sampled, or they were cropped to corn, grain sorghum, or cotton as a rotation crop with soybean. Samples were stored and transported to the Arkansas Nematode Diagnostic Laboratory in Hope, Arkansas in plastic bags inside insulated coolers. Samples were stored no longer than two weeks prior to assay. When RKN was extracted through routine elutriation62 and sugar flotation63 of a sub-sample, the remaining soil was placed into a 15-cm-diameter clay pot filled with 50:50 mixture of fine builders’ sand and sandy loam topsoil. A single tomato seedling (Solanum lycopersicon L var. lycopersicum, cv. ‘Rutgers’) at the age of three to four week old from gemination was grown in the soil in a greenhouse. Tomato plants were then removed from the soil and the root systems were washed to remove excess soil at harvest. Root galls on tomato were collected after 60–70 days of inoculation and shipped to Nematode Lab at Agronomic Division in North Carolina Department of Agriculture. Thirty-eight other populations were collected by the second author. Galls or dissected females were shipped to NCDA without rearing nematodes on tomato.

Figure 14
figure 14

Sampled counties and sample numbers for root-knot nematode survey from Arkansas.

Full size image

DNA extraction

RKN females were dissected in water in a 9-cm petri dish under Zeiss Stemi 2000-C microscope (Gottingen, Germany). A single female was pipetted into 10-µl 1X TE buffer (10 mM Tris-Cl, 1 mM EDTA; pH 9.0) on a glass microscope slide (7.5 cm × 2.5 cm). The nematodes were then macerated with a pipette tip into pieces, collected in 50-µl 1X TE buffer and stored at −20 °C. Three DNA replicates per sample were prepared for any samples with females. If only the second-stage juveniles were available, 1–10 juveniles were macerated with a pipette tip into pieces and put in one tube as DNA template in 50-µl 1X TE buffer.

DNA amplification, cleaning and sequencing

The primers used for ribosomal and mitochondrial DNA PCR and DNA sequencing are shown in Table 2 as previously described36. These primers were synthesized by Integrated DNA Technologies, Inc. (Coralville, Iowa, USA). The 25-µl PCR was performed using 12.5-µl 2X Apex Taq red master mix DNA polymerase (Genesee Scientific Corporation, San Diego, CA, USA), 9.5-µl water, 1-µl each of 10-µM forward and reverse primers, and 1 µl of DNA template according to the manufacturer’s protocol in a Veriti® thermocycler (Life Technologies, Carlsbad, CA, USA). The thermal cycler program for PCR was as follows: denaturation at 95 °C for 5 min, followed by 40 cycles of denaturation at 94 °C for 30 s, annealing at 55 °C for 45 s, and extension at 72 °C for 1 min. A final extension was performed at 72 °C for 10 min. PCR products were cleaned using ExoSap-IT (Affymetrix, Inc., Santa Clara, CA, USA) according to the manufacturer’s protocol. DNA sequencing was performed using PCR primers for direct sequencing by dideoxynucleotide chain termination using an ABI PRISM BigDye terminator cycle sequencing ready reaction kit (Life Technologies, Carlsbad, CA, USA) in an Applied Biosystems 3730 XL DNA Analyzer (Life Technologies) by the Genomic Sciences Laboratory (North Carolina State University, Raleigh, NC, USA). The molecular sequences were compared with other nematode species available at the GenBank sequence database using the BLASTn homology search program.

Phylogenetic analyses

DNA sequences were edited with ChromasPro1.5 2003–2009 (Technelysium Pty Ltd, Helensvale, Australia) and were aligned by Mega7.0.1464 using default settings. The model of base substitution in the DNA sequence data was evaluated using MODELTEST version 3.0665. The Akaike-supported model66, the proportion of invariable sites, and the gamma distribution shape parameters and substitution rates were used in phylogenetic analyses using DNA sequence data. Bayesian analysis was performed to confirm the tree topology for each gene separately using MrBayes 3.1.067, running the chain for 1,000,000 generations and setting the ‘burnin’ at 2,500. Markov Chain Monte Carlo (MCMC) methods were used within a Bayesian framework to estimate the posterior probabilities (pp) of the phylogenetic trees68 using the 50% majority-rule. The λ2 test for homogeneity of base frequencies and phylogenetic trees was performed using PAUP* version 4.0 (Sinauer Associates, Inc. Publishers, Sunderland, MA, USA).

Species identification using PCR by species-specific primers

The species identification of M. incognita was confirmed using PCR by species-specific SCAR primers Inc-K14-F/Inc-K14-R which produce a 399-bp DNA fragment69. Another set of M. incognita-specific SCAR primers was a 1200-bp PCR fragment amplified by Finc/Rinc21. Fjav/Rjav21, Far/Rar21 and MH0F/MH1R70 were the other species-specific primers to M. javanica, M. arenaria and M. hapla which produced 670-bp, 420-bp and 960-bp DNA fragment respectively. The 25-µl PCR was performed using 12.5-µl 2X Apex Taq red master mix DNA polymerase, 7.5-µl water, 1-µl each of 10-µM forward and reverse primers, and 1-µl of DNA template. The PCR condition is the same as described above.