Identification, distribution, and hosts of Meloidogyne spp. infecting horticultural crops in Florida, USA with focus on Meloidogyne enterolobii

Sampling and processing: A total of 304 root and soil samples from 56 different mostly fruit and vegetable crops were collected in 12 counties located in Central and South Florida from September 2019 to January 2023. The Global Positioning System (GPS) coordinates for each site were recorded using the GPS Tracks version 4.0.6 smartphone app (DM Software Solutions, LLC). Samples were collected from different cropping systems, including commercial vegetable farms; fruit and Asian vegetable farms; research farms; horticultural gardens; and natural landscapes. Samples were collected throughout the year and typically towards the end of the crop season, when symptoms, such as stunting, chlorosis and wilting, were most visible. A conical core sampler and shovel were used to take soil and root samples up to a depth of about 20 cm; the soil was collected with roots, with an effort to maintain their quality). The number of subsamples taken per sample depended on the location and size of the field and varied from a single sample in horticultural gardens and natural landscapes to three to ten plants per sample in commercial and research fields. Together with the roots, at least 500 cm³ of soil were collected and stored in a cooler (about 4 °C) for further nematode extraction, using a modified Baermann method (Regmi and Desaeger, 2020) when needed to confirm the absence of RKN in certain sample as stated below. Roots were carefully washed, and the presence of females and egg masses were checked under a dissecting microscope (Leica KL300 LED, Danaher Corporation, Washington, DC, USA). Three individual females per root system were hand-picked and used for RKN species identification. If no root galling, females, or egg masses were found, J2 were extracted from the soil sample, and in their absence, the sample was recorded as free from RKN. Plants were considered hosts of a given RKN species when root galls, females, and/or egg masses of that RKN were detected in their roots.

DNA extraction, PCR amplification and sequencing: DNA from each hand-picked female (N = 3) was extracted using the NaOH digestion method (Hübschen et al., 2004). Nematode species identification was conducted using nonspecific primer sets followed by species-specific primers (Table 2) and DNA sequencing of selected populations. DNA sequencing was only performed for populations that were previously identified as M. enterolobii and for a few other RKN populations where PCR identification gave inconsistent results. At least one nonspecific and species-specific primer set was used to identify the RKN present in each sample. PCR was performed using an Eppendorf Mastercycler Pro Thermal Cycler (Enfield, CT) in a 25 μl reaction volume consisting of 12.5 μl of 2 × Apex™ Taq Blue DNA Polymerase Master Mix (Genesee Scientific, San Diego, CA), 1 μl of DNA extract, 0.25 μl of each 10 μM primer stock, and 11 μl of sterile water. The PCR amplification conditions are reported in Table 3. The amplicons were separated by gel electrophoresis using a 1.2% Apex general purpose agarose gel (Genesee Scientific, El Cajon, CA), at 110 V for 60 min. Gels were stained with GelRed Nucleic Acid Stain (Biotium Inc., Hayward, CA), and DNA fragments were visualized under UV light using the Gel Doc EZ Imager (Bio-Rad Laboratories, Hercules, CA). DNA samples of M. arenaria, M. enterolobii, M. incognita, M. javanica (RKN collection, Division of Plant Industry, Florida Department of Agriculture and Consumer Services, Gainesville, FL) and M. hapla (RKN collection, Gulf Coast and Education Center University of Florida, Wimauma, FL) were used as positive controls for PCR in this study. As mentioned above, PCR products identified as positive for M. enterolobii were subject to DNA sequencing. For each M. enterolobii-positive sample, the best PCR product from at least one primer set was sent to Genewiz Company (South Plainfield, NJ, USA) for DNA purification and Sanger DNA sequencing. Newly obtained sequences were edited using Geneious Prime (version 2023.0.3, Biomatters, Ltd.) and deposited in the National Center for Biotechnology Information (NCBI) with the accession numbers indicated (Table 4).

Prior to phylogeny inference, sequences were aligned and trimmed to remove ambiguously aligned residues using Geneious Prime. Phylogenetic assumptions, stationarity (constant nucleotide frequencies over time), and homogeneity (constant substitution rates over time) were assessed using matched-pairs tests of symmetry implemented in IQ-TREE (version 2.1.2; Minh et al., 2020; Naser-Khdour et al., 2019). The best-fit substitution model was then determined by ModelFinder for each partition according to Bayesian Information Criterion (BIC) (Kalyaanamoorthy et al., 2017), resulting in HKY + F + G4 for both COX2-l-rRNA and l-rRNA genes. Phylogenetic tree construction, based on the two nonspecific primer sets mentioned above, was performed using MrBayes (version 3.2.7; Ronquist et al., 2012) with the following settings: two independent runs, one million generations, sampling every 1,000 generations, and discarding the first 25% of trees as burn-in. The remaining trees were used for generating a 50% majority-rule consensus tree with posterior probabilities for each node, and nodal support was estimated based on 1,000 standard bootstrap replicates. The phylogenetic trees were rooted to the outgroup species of each gene, visualized, and edited using FigTree (version 1.4.4; http://tree.bio.ed.ac.uk/software/figtree/) and Inkscape (version 1.0.1; https://inkscape.org/), respectively. The sequence alignments used in this study to build the phylogenetic trees can be found in the GenBank accession numbers listed on each phylogenetic tree.

For the phylogenetic tree based on DNA fragments of the COX2-l-rRNA gene from C2F3/1108 primer, a total of 44 sequences –36 from GenBank and 8 of M. enterolobii – were selected for this study. These sequences were from 13 countries (Australia, Brazil, Burkina Faso, China, Kenya, Mexico, Portugal, Italy, Ivory Coast, Japan, Slovenia, South Africa, and USA). The USA sequences were from six states (Arkansas, Florida, Georgia, Nebraska, North Carolina, and South Carolina). For the phylogenetic tree based on the tRNA-His and l-rRNA gene from TRNAH/MRH106 primers, a total of 36 sequences – 26 from GenBank and 10 of M. enterolobii – were selected for this study. These sequences were from five countries (Brazil, China, Kenya, Mexico, and USA). The USA sequences were from four states (Georgia, Georgia, North Carolina, and South Carolina).

Survey data collection and display: Data concerning the RKN species identified were listed according to their geographical location, plant host, and the management practices of the field operations sampled.

The findings of our survey have agronomic significance for agricultural specialists at local and national level. They enrich their knowledge about the host preference of the RKN identified in the vegetable crop systems adopted in Florida operations. The survey data are therefore presented in different categories to facilitate their interpretation by growers, agricultural specialists and extension agents in Florida and in the USA.

By geography: A total of 304 root and soil samples were collected from 12 counties (Alachua [2], Charlotte [20], Collier [10], DeSoto [26], Hardee [1], Hendry [3], Highlands [8], Hillsborough [176], Manatee [11], Miami Dade [14], Palm Beach [13] and, Sarasota [20]) in Central and South Florida from September of 2019 to January of 2023 (Fig. 1). Samples were collected from different locations, including commercial vegetable farms, fruit and Asian vegetable farms, research farms, horticultural gardens, and natural landscapes (Fig. 2). At least one RKN species was detected in every county sampled, except Charlotte (Fig. 1). RKN species were detected in 247 (81.25%) of the 304 samples collected. At least 741 females were subject to molecular analysis (247 × 3).

Figure 1:

Florida's map showing the 12 counties sampled, distribution, and prevalence of each Meloidogyne species identified in this study. Out of 304 samples collected, 247 were positive for root-knot nematodes. Counties sampled and number of samples positive for Meloidogyne spp. were Alachua (2), Collier (1), DeSoto (20), Hardee (1), Hendry (1), Highlands (8), Hillsborough (171), Manatee (10), Miami-Dade (5), Palm Beach (8), and Sarasota (20). No root-knot nematodes were found in the samples collected in Charlotte County. Map was made using ArcGIS Prop software (Esri, Redlands, CA).

Figure 2:

Distribution of Meloidogyne species across different cropping systems found in this study.

Based on the molecular analyses, including species-specific primers (Fig. 3) and DNA sequencing a total of five RKN species were identified in this study: M. arenaria, M. enterolobii, M. hapla, M. incognita, and M. javanica. M. incognita and M. enterolobii were the two most prevalent species, with each present in 76 (25.00%) samples. M. javanica was identified in 50 samples (16.44%), followed by M. arenaria with 25 samples (8.22%) and M. hapla with 16 samples (5.27%). Mixed populations of M. enterolobii and M. incognita were identified in four samples (1.31%). Hillsborough County was the most intensively sampled, with 171 samples positive for RKN. Other counties sampled where RKN was detected on at least one sample were Alachua (2), Collier (1), DeSoto (20), Hardee (1), Hendry (1), Highlands (8), Manatee (10), Miami-Dade (5), Palm Beach (8), and Sarasota (20) (Fig. 1). M. enterolobii was found in seven counties: Alachua, Hendry, Highlands, Hillsborough, Manatee, Palm Beach, and Sarasota. This represents 58% of the sampled counties and 63% of the counties found to be positive for RKN (Fig. 1).

Figure 3:

PCR amplification product generated using species-specific primer sets. A) Me = Meloidogyne enterolobii genomic DNA amplified using Me-F/Me-R primer set; B) Ma = Meloidogyne arenaria genomic DNA amplified using Far/Rar primer set; C) Mi = Meloidogyne incognita genomic DNA amplified using MI-F/MI-R primer set; D) Mj = Meloidogyne javanica genomic DNA amplified using Fjav/Rjav primer set; N = Water; M = 100-bp DNA ladder (New England Biolabs, MA), with the arrow indicating the position of the 500/517 bp band.

By growing conditions and farm type: Most samples were collected from commercial vegetable, fruit, and ornamental farms, representing a total of 145 samples (47.70%) with 106 positive samples for RKN. The crops sampled were tomato (46), pepper (23), strawberry (11), squash (7), cantaloupe (6), caladium (6), okra (5), cucumber (4), eggplant (3), guava (3), cherry tomato (3), boniato (3), parsley (3), sugar cane (3), sugarbeet (2), watermelon (2), bean (2), Caesar weed (1), corn (1), elephant ear (1), indigo (1), jackfruit (1), lily (1), luffa (1), Italian parsley (1), and peach (1) (Table 5). Five species of RKN nematodes were identified from these farms: M. incognita, M. enterolobii, M. javanica, M. hapla, M. arenaria and mixed populations of M. enterolobii and M. incognita. These represented 40 (27.58%), 20 (13.80%), 19 (13.10%), 15 (10.34%), 10 (6.90%) and 2 (1.38%) samples, respectively (Fig. 2).

Asian vegetable farms represented 58 samples (19.07%), with 50 of them positive for RKN. Three RKN species were detected with M. enterolobii as the predominant species, accounting for 42 samples (72.41%), followed by M. incognita with 6 samples (10.34%), and M. arenaria, with two samples (3.44%) (Fig. 2). The crops sampled were sweet potato (14), luffa (8), basil (7), pumpkin (7), perilla (2), Malabar spinach (2), amaranth (2), ginger (2), golden egg (2), jute (2), Thai basil (1), water spinach (1), radish (1), Napa cabbage (1), lettuce (1), leek (1), eggplant (1), chrysanthemum (1), cauliflower (1), and bok choy (1) (Table 5).

A total of 68 samples (22.37%) represented research institutions, including private companies and University of Florida Research Centers, with 59 positive samples for RKN. Five species of RKN were detected, with M. javanica representing 31 positive samples (45.59%), followed by M. arenaria, M. incognita, M. enterolobii and M. hapla, representing 11 (16.18%), 9 (13.23%), 7 (10.30%) of samples, and one (1.48%) positive sample, respectively (Fig. 2). The crops sampled were tomato (15), cowpea (10), caladium (9), cucumber (8), strawberry (3), squash (2), okra (2), pepper (2), squash (2), artichoke (2), corn (2), zucchini (1), watermelon (1), Thai basil (1), sunflower (1), sugar cane (1), radish (1), lettuce (1), lavender (1), indigo (1), eggplant (1), and cantaloupe (1) (Table 5).

A total of 30 (9.87%) samples were collected from horticultural gardens, with 29 positive samples for RKNs. Only two RKN species were found, with M. incognita the most common, being found in 21 (70.00%) of the samples; M. enterolobii in six of the samples (20.00%); and mixed populations of M. incognita and M. enterolobii in two of the samples (6.67%) (Fig. 2). The crops sampled were tomato (6), lettuce (4), okra (3), pepper (3), sunflower (1), squash (1), pumpkin (1), mustard (1), Malabar spinach (1), eggplant (1), cucumber (1), cowpea (1), coriander (1), cherry tomato (1), cactus (1), bok choy (1), and amaranth (1) (Table 5). Three samples were collected from natural areas; two samples were from hemp (Cannabis sativa) at a private home and were positive for M. arenaria; and one sample was collected from a wild tomato (Solanum capsicoides) in a natural preserve and was positive for M. enterolobii (Fig. 2).

By crop and plant species: A total of 56, crops belonging to 22 different families, were sampled (Table 5). Tomato was the most-sampled crop, followed by pepper, caladium, sweet potato, and strawberry. Among the plant families, Solanaceae had 108 samples, followed by Cucurbitaceae (51), Fabaceae (21), Convolvulaceae (18), and Araceae (16) (Fig. 4, Table 5). M. enterolobii was found infecting 20 different plant species, with pepper (16), sweet potato (14), basil (7), luffa (7), pumpkin (6), and tomato (6) being the most common hosts. For comparison, M. incognita was found on 21 different plant species, M. javanica on 12, M. arenaria on six and M. hapla on three (Table 5).

Figure 4:

Distribution of Meloidogyne species by crop found in this study.

Several new host records for M. enterolobii were found in this study: new worldwide host records of M. enterolobii infecting wild tomato (Solanum capsicoides); new continental USA host records of M. enterolobii infecting cowpea (Vigna unguiculata) and cactus (Nopalea cochenillifera); and new Florida host records of M. enterolobii infecting bean (Phaseolus vulgaris L.). There were also new state host records for Florida of M. arenaria infecting hemp (Cannabis sativa); elephant ear (Colocasia esculenta); and lily (Lilium sp.), M. hapla infecting cantaloupe (Cucumis melo), and M. incognita infecting lavender (Lavandula angustifolia) and sunflower (Helianthus annuus).

Phylogenetic study: Meloidogyne enterolobii sequences obtained from this study formed a monophyletic clade with other M. enterolobii populations with 100% support. All selected sequences from this study and other locations collected in the US and other parts of the world were used for comparisons grouped together, presenting genetic similarities among them.

The phylogenetic tree, based on DNA fragments of the COX2-l-rRNA gene from C2F3/1108 primer set, placed RKN populations into five distinct groups (Fig. 5). The RKN populations clustered in five distinct clades containing sequences of 11 RKN species. In this tree, M. enterolobii, M. hapla, M. haplanaria and M. partityla showed genetic variation and clustered in four different clades. M. arenaria, M. ethiopica, M. floridensis, M. hispanica, M. incognita, M. javanica and M. luci clustered together in another separate clade, presenting genetic similarities among those species.

Figure 5:

Bayesian 50% majority-rule consensus tree inferred from mitochondrial DNA DNA fragment of cytochrome c oxidase subunit II to large subunit 16S rRNA genes from C2F3/1108 primer set. Populations of Meloidogyne enterolobii obtained in this study are represented in bold. GenBank accession number proceed species and origin. ^*Recently reclassified as Meloidogyne luci (Stare et al., 2017).

In the phylogenetic tree based on the tRNA-His and l-rRNA gene from TRNAH/MRH106 primer set, the RKN populations clustered in five distinct clades (Fig. 6) containing sequences of 11 RKN species. Meloidogyne enterolobii, M. hapla, M. haplanaria and M. mali clustered in four separate and distinct clades. M. arenaria, M. ethiopica, M. floridensis, M. hispanica, M. incognita, M. javanica and M. luci clustered in another separate clade, presenting genetic similarities among those species. The clade with M. enterolobii sequences contained a subclade with 66% support. In this subclade, two M. enterolobii isolates (OQ835725 and OQ835726) from this study clustered together with other isolates (KJ146864, MH800965, MN840972, MT094620, MW802190, ON320402, ON320404 and ON320406) from China, Florida, Georgia, North Carolina, and South Carolina.

Figure 6:

Bayesian 50% majority-rule consensus tree inferred from mitochondrial DNA fragment of partial l-rRNA gene from TRNAH/MRH106 primer set. Meloidogyne enterolobii from this study are represented in bold. Branch lengths represent means of posterior distribution. GenBank accession number precedes species and origin. ^*Recently reclassified as Meloidogyne luci (Stare et al., 2017).

Root-knot nematodes find perfect conditions in Florida and have been a menace to vegetable and specialty crop growers in the state since the early days of agriculture (Neal, 1889; Desaeger, 2021). Since the first report of M. enterolobii in Florida (Brito et al., 2002; 2004), two Meloidogyne surveys were carried out in the state (Brito et al. 2008, 2010). Both studies used isozyme analysis, mainly esterase (EST) and malate dehydrogenase (MDH) phenotypes, to characterize and identify the species of Meloidogyne present. In both these surveys, the most common RKN species found were M. arenaria, M. incognita and M. javanica, with M. enterolobii the most frequently found in ornamental plants. Many ornamental plants have been reported to be susceptible to M. enterolobii, and it is becoming clear that infected ornamental plants, during planting, are one of the major pathways of introducing and spreading M. enterolobii among several parts of the world (Moore et al. 2020a; 2020b; EPPO, 2023; de Heij, 2023).

In the southern US, M. incognita is the most identified RKN species in commercial fields (Schwarz et al., 2020; Marquez and Hajihassani, 2023; Faske et al., 2023). Our findings confirm the prevalence of M. incognita as the most common RKN species in Florida, but they also (and maybe surprisingly) confirm that M. enterolobii is just as prevalent. M. enterolobii was identified in 20 different crops, from large commercial fields to small community gardens. Several new host records for M. enterolobii were found in this study (as well as for other RKN species), confirming, and further expanding the already extremely broad host range that has been reported from ornamental, horticultural and agronomic crops, as well as turfgrass and weed plants, in previous studies from Florida (Brito et al., 2007, 2010; Baidoo et al., 2016; Joseph et al., 2016; Kaur et al., 2007).

In Asian vegetable farms, RKN, especially M. enterolobii, were very prevalent, often in high numbers. Most of these farms grow crops year-round, providing a continuous food supply to the nematodes, as most crops are good hosts to many RKN species (Bui et al, 2022a; 2022b). Nematode management is non-existent in these farms, and farmers generally have no knowledge about RKN or other nematodes. Due to language and cultural barriers (most farmers are of Vietnamese origin), Asian vegetable farms have had limited access to pest management information in general. It is quite likely that RKN, including M. enterolobii, has been spread across these farms with infested plant material, such as ginger rhizomes and sweetpotato tubers.

Meloidogyne arenaria was found in hemp, tomato, and cucumber, but was especially common in caladium, a very valuable ornamental crop that is grown for its tubers. Caladium has been reported to be susceptible to the three major RKN species, as well as to M. enterolobii and M. floridensis (Kokalis-Burelle et al., 2017). Additionally, caladium has a long growing period of more than six months, and RKN is the most important pathogen infecting this crop, especially when grown in sandy soil (Gu and Desaeger, 2021). Although hot water treatment, by soaking the tubers in water at 43° C for two hours, is considered an effective procedure to sanitize caladium tubers from RKN and Pythium, it has been observed that some nematodes are likely to survive (Gu and Desaeger, 2021), possibly (just like ginger in Asian vegetable farms) facilitating the spread of M. arenaria across caladium fields with tubers infested by nematodes that escaped the treatment.

Meloidogyne javanica was found on 12 different plants and was the most common RKN species on tomato. M. javanica was also detected infecting a specific strawberry cultivar, Winterstar, but it did not infect any of six other strawberry cultivars that were planted together with this cultivar (Oliveira et al., 2023). This was the first report of M. javanica infesting strawberry in Florida, where M. hapla has been the most reported RKN species on this crop. This has been confirmed by the findings in this study and previous field observations conducted by Oliveira (2022), indicating that M. hapla is being introduced into Florida fields with nematode-infested strawberry propagative material imported from northern states and Canada. Every year, more than 100 million strawberry transplants carrying the nematode are imported into Florida strawberry operations. Meloidogyne hapla can cause some late-season damage to Florida strawberry, but it can also reproduce rapidly on double- or relay-cropped vegetables grown soon after strawberry, preventing their establishment and severely stunting their growth (Desaeger, 2018; Khanal and Desaeger, 2020).

Significant regional differences in the prevalence of RKN throughout the state's vegetable area were noted. In the drier deep sand soils in west-central Florida, RKN has been long reported to be widely distributed (Neal, 1889), and visible crop damage was common in this area during our survey. In the Everglades agricultural region in south Florida (sections of Collier, Hendry, Palm Beach, and Miami-Dade counties) which has large areas of organic “muck” soils and a naturally high water table, vegetable crops rarely showed nematode damage, and RKN were also much more sporadic. The high water table and seasonal flooding of some of these fields, as well as the high organic content of the muck soils, naturally suppress RKN. Flooding has been used to control RKN since the 1930s (Brown, 1933), and organic soils are well-known to often provide excellent natural suppression of many plant-parasitic nematodes.

Meloidogyne enterolobii sequences from this study formed a monophyletic clade with other M. enterolobii populations with 100% support. The phylogenetic analysis of 12 different M. enterolobii isolates obtained from this study and 24 isolates of M. enterolobii from the USA and other countries clearly separated M. enterolobii from other RKN species and revealed low variability amongst them, clustering the isolates in one clade. This confirms the results of other studies from Brazil, South Africa and China (Tigano et al., 2010; Onkendi and Moleleki, 2013; Rashidifard et al., 2018, Shao et al., 2020). The lack of variation among M. enterolobii species can be explained by their mode of reproduction through mitotic parthenogenesis, and could also indicate possible transportation or movement of infected plant material and/or infested soil around the world. M. enterolobii has been intercepted in several European countries, including the Netherlands, Germany, United Kingdom, several times in plant material imported from Asia, South America, and Africa, including Cactaceae, Syngonium sp., Ficus sp., Ligustrum sp., Brachychiton sp., and Rosa sp. (EPPO, 2023; de Heij et al., 2023). While the M. enterolobii sequences here presented a lack of genetic diversity, it is important to point out the low number of analyses (< 100 samples), as well as the fact that phylogenetic trees based on l-rRNA indicated little variability among the isolates and placed the M. enterolobii isolates into a subclade with 66% support. Further studies analyzing more sequences are necessary to keep investigating possible genetic variation among populations found infecting different plant species in the USA and other counties.

This study confirmed the prevalence of RKN in South and Central Florida and revealed that M. incognita and M. enterolobii are the most common species in these parts of the state. On the other hand, M. enterolobii reports from North Florida are few, which may be related to the difference in crops grown there (Z. Grabau, UF, pers. comm.). Peanut (Arachis hypogeae) and corn (Zea mays), which are both non-hosts or poor hosts of M. enterolobii in the US, are mostly grown in North Florida, and much less in South and Central Florida. Knowing the prevalence, hosts, and geographic distribution of M. enterolobii is extremely important to help limit its introduction and dissemination in new areas. Because of its broad host range and the lack of RKN resistance in commercially-available vegetable and fruit cultivars, current options for managing M. enterolobii are extremely limited, which is why increased awareness is so important. This is especially important considering that M. enterolobii has been included in the list of quarantinable pathogens in some states in the USA and Europe. Knowledge from this study on the prevalence, hosts, and distribution of the nematode in Florida is critical and can be used for improving and establishing new control measures for RKN, particularly M. enterolobii.