Pratylenchus pseudocoffeae Mizukubo, 1992 infecting dahlia (Dahlia hybrida) in the United States expands its known host and geographic ranges

1.1. Nematode samples

Tuber-derived populations of Pratylenchus spp. were recovered from symptomatic dahlia tubers submitted by commercial producers and home gardeners to The Penn State Plant Disease Clinic and Nematode Diagnostics Laboratory and to The Ohio State University Plant Diagnostic Network. Infected tubers exhibited extensive lesions and necrosis; symptoms consistent with damage caused by migratory endoparasitic nematodes. Infected samples originated from five counties in Pennsylvania (Delaware, Erie, Lancaster, Franklin, and Centre) and four counties in Ohio (Darke, Franklin, Greene, and Cuyahoga). Herein, a nematode population refers to nematodes recovered from each of these counties. Fig. 1 shows the counties of origin of Pratylenchus spp. populations analyzed in this study and their distribution relative to previously reported occurrences of this nematode in the United States.

Figure 1

1.2. Nematode isolation and morphological characterization

Nematodes were extracted from symptomatic dahlia tubers using a modified Baermann tray method from tuber peels or macerated pieces of tubers (Viaene et al., 2021). Individual nematodes were hand-picked under a dissecting microscope and transferred to embryo dishes containing distilled water. Representatives from each population were used for molecular analyses, while one population from Pennsylvania (Franklin County population) and one from Ohio (Cuyahoga County population) were randomly selected for detailed morphological characterization. For morphological observations and morphometric analyses, nematodes were heat-killed and mounted on temporary slides for imaging and measurements. Micrographs were captured using a Zeiss AXIO Imager M2 compound microscope (Carl Zeiss Microscopy GmbH, Jena, Germany) equipped with a Zeiss AxioCam 712 camera. Morphometric measurements were obtained using ZEN Pro software (Carl Zeiss Microscopy GmbH, Jena, Germany).

1.3. Molecular characterization

Genomic DNA was extracted from single specimens from each county population using a modified Proteinase K protocol (Kawasaki 1990; Frey et al.,2022). Briefly, a single nematode was cut into 12 µl of Milli-Q water and transferred to a 0.2 ml PCR tube containing 12 µl of Kawa buffer (10 mM Tris HCl, 1 mM EDTA, 0.5 % Tween 20, 50 μg/ml Proteinase K, pH 8.0). Samples were centrifugated for 1 min at 1,400 rpm, incubated at 37°C for 30 min and 94°C for 2 min, followed by a centrifugation at 1,400 rpm for 1 min. DNA samples from Greene, Franklin, and Cuyahoga counties (Ohio), were obtained using Worm Lysis protocol (Williams et al., 1992).

Two loci of the ribosomal DNA (rDNA) were amplified and Sanger sequenced. The D2-D3 expansion of the large rDNA subunit (LSU; 28S) was amplified using the D2A (5′-ACA AGT ACC GTG AGG GAA AGT TG-3′) and D3B (5′ – TCG GAA GGA ACC AGC TAC TA – 3′) primers as described by De Ley et al. (2005). The internal transcribed spacer (ITS) region was amplified using primers TW81 [5′ – GTT TCC GTA GGT GAA CCT GC – 3′; Joyce et al. (1994)] and AB28 [5′ – ATA TGC TTA AGT TCA GCG GGT – 3′; Howlett et al. (1992)] as described in Subbotin et al. (2000). The PCR products were purified using QIAquick PCR purification kit (Qiagen, Germantown, MD, USA) and sequenced directly using the respective forward and reverse PCR primers, or amplicons were first cloned using either the NEB (New England Biolabs, Ipswich, MA, USA) or TOPO (Thermo Fisher Scientific, Carlsbad, CA, USA) PCR cloning kits. Inserts were amplified and validated by colony PCR using the vector-specific primers supplied with the kits following the manufacturers’ protocols. The resulting amplicons were cleaned and subsequently sequenced with the vector-specific (M13 universal) primers at The Pennsylvania State University Genomics Core Facility (State College, Pennsylvania, USA) or Eurofins Genomics (Louisville, KY, USA).

Nucleotide sequences were assembled using Geneious version 11.1.5 (Biomatters Ltd, Auckland, New Zealand), and aligned using ClustalX 1.83 (Chenna et al., 2003). Alignments were manually edited, when necessary, in GenDoc 2.5 (Nicholas et al., 1997) and subsequently analyzed using Bayesian inference in MrBayes 3.1.2 (Ronquist and Huelsenbeck, 2003). The best-fit models of nucleotide substitution were determined using jModelTest 0.1.1 (Posada, 2008) under the Akaike Information Criterion. Pairwise sequence divergence was calculated using PAUP*4a (Swofford, 2003). Bayesian analysis for each DNA region was initiated from random starting trees and run with four Markov chains for 1.0 × 10⁶ generations (Subbotin, 2021). Posterior probabilities (PP) above 0.5 are indicated on the corresponding nodes.

1.4. Nematode multiplication and greenhouse pathogenicity assays

To confirm pathogenicity, the response of commercial dahlia cuttings (cv. Mystic Spirit) obtained from ColorLink, Ball Horticultural Company (West Chicago, IL), to infestation by a P. pseudocoffeae population originating from Cuyahoga County, Ohio, was evaluated in a greenhouse experiment conducted at The Ohio State University, Columbus, Ohio, from August 22 to October 31, 2025. The inoculum used in the experiment consisted of P. pseudocoffeae specimens isolated from infected dahlia tubers and subsequently multiplied on surface-sterilized carrot discs as described by Coyne et al. (2014).

Rooted cuttings were transplanted individually into a commercial soilless potting mix in 8.9 cm × 8.9 cm pots and arranged in a randomized complete block design on a 25.4 cm × 50.8 cm tray. Four inoculation treatments were applied: 0 (control), 4,000 (low), 8,000 (medium), and 16,000 (high) nematodes per plant, corresponding to, approximately, 0, 7, 14, and 28 nematodes cm⁻³ of soil. Inoculum density was estimated from the total number of nematodes present in a stock suspension by averaging counts from three aliquots examined under a microscope at 40× magnification. The volume of suspension required to obtain the target inoculum density for each treatment was subsequently calculated. Each treatment was replicated four times. Nematode suspensions containing a mixture of vermiform life stages (juveniles and adults) were applied as a soil drench around the base of each stem and covered with potting mix. Plants were watered daily and maintained for 10 weeks at 24 ± 5°C and 16 h light and 8 h dark cycles.

At harvest, plants were carefully removed from the pots, and substrates were collected for nematode extraction. Roots were gently washed in beakers to retain adhering substrate, blotted dry with paper towels, and weighed. Nematodes were extracted from both substrate and macerated roots using Baermann funnels and are reported as the total mean number recovered per plant.

1.5. Statistical analysis

Plant growth parameters were analyzed using linear models with treatment (inoculum density) as a fixed effect. Pairwise comparisons were conducted using Tukey-adjusted tests, and compact letter displays were used to summarize differences among treatments. Dunnett-adjusted contrasts were applied to compare each inoculum density with the noninoculated control.

Nematode counts were analyzed after ln (x + 1) transformation. Percent reduction in root weight relative to controls was calculated from treatment mean values, and 95% confidence intervals were estimated using bootstrap resampling (5,000 iterations). All analyses were performed in R version 4.3.2 (R Core Team, 2023) using the packages tidyverse (Wickham et al., 2019), multcomp (Hothorn et al., 2008), and car (Fox and Weisberg, 2019).

2.1. Host symptoms

Dahlia tubers submitted for diagnosis exhibited bark-like texture with longitudinal grooves, scarring of the tuber skin, and necrotic lesions; additional necrotic lesions were observed after peeling the tubers (Fig. 2). Population densities of dahlia root-lesion nematodes obtained from these tubers ranged from 5 nematodes g⁻¹ tissue in asymptomatic tubers to as high as 544 nematodes g⁻¹ tissue in symptomatic tubers.

Figure 2

2.2. Nematode morphological and morphometric characterization

Root-lesion nematodes recovered from symptomatic tubers measured 472 ± 14.9 to 558 ± 60 µm in body length (Table 1), with females slightly longer than males. No sexual dimorphism was observed other than reproductive structures. In females (Fig. 3a–d), the vulva was located at approximately 80% of body length; the vagina extended more than half of the body width at the vulval level (Fig. 3h); and the spermatheca was rounded to elliptical and contained sperm (Fig. 3f). The body was slightly ventrally curved when heat-killed. The labial region exhibited two to three annuli; the lateral field consisted of four incisures (Fig. 3g). The tail terminus was smooth and rounded to sub-hemispherical (Fig. 3c and d).

Figure 3

In males, lateral fields with four incisures extended along the mid-body and terminated at the bursa. Spicules were paired and slender; gubernaculum was small and simple (Fig. 2e and i). The excretory pore was located just posterior to the pharyngo-intestinal junction, approximately at the level of the pharyngeal gland origin.

2.3. Morphometric measurements

All measurements are provided in Table 1.

2.4. Molecular characterization

Based on morphological and molecular analyses, all nine nematode populations recovered from the symptomatic dahlia samples were identified as P. pseudocoffeae. The lengths of the D2D3 region of the 28S and ITS fragments generated for all populations were 741 and 889 bp, respectively, excluding primer binding sites. Newly obtained sequences were deposited in GenBank under accession numbers PV986174 to PV986178, PZ048661, PZ048662, PX945294, and PX945295 for 28S rDNA region, and PV988438 to PV988442, PX945289, PX957255, PZ028637, and PZ028638 for ITS region. In the aligned datasets, sequences were trimmed at both ends to match the shortest reference sequences obtained from GenBank. The final aligned 28S and ITS datasets had 737 and 883 bp, respectively, including alignment gaps. Phylogenies were inferred using the GTR + G model. Sequences from the original description of P. pseudocoffeae from Japan were not included because these were not produced in the original description by Mizukubo (1992), but sequences from a Japanese population reported by Kushida and Kondo (2015) were included in the analyses. Phylogenetic relationships between the Ohio and Pennsylvania populations of P. pseudocoffeae and related Pratylenchus species are shown in Fig. 4. In both 28S- and ITS-based phylogenies (Fig. 4a and b, respectively), all of the nine populations clustered with previously reported P. pseudocoffeae populations in a strongly supported monophyletic clade (PP = 1.0). This clade was clearly separated from other Pratylenchus species, including P. scribneri, P. agilis, P. hippeastri, and P. coffeae, confirming the identity of the dahlia root-lesion nematode as P. pseudocoffeae.

Figure 4

2.5. Greenhouse pathogenicity assay

Nematode inoculation significantly reduced root and shoot growth relative to noninoculated control plants (Table 2, Fig. 5). Mean root weight in control plants was 12.96 g, compared with 3.27, 4.94, and 4.40 g in the low-, medium- and high-inoculum treatments, respectively. Dunnett-adjusted contrasts (Table 3) confirmed that each inoculum level resulted in significantly lower root weight than the noninoculated controls. Although inoculation significantly reduced root weight, no significant differences were detected among the three inoculum levels. At the end of the trial, nematodes were absent in the control plants but were detected in all inoculated treatments, with mean total counts ranging from approximately 6,708–8,204 P. pseudocoffeae per plant (Table 2). Nematode densities were similar among inoculated treatments. The significant root damage and corresponding reduction in root biomass (Fig. 5) across all inoculum levels indicate the impact of nematode interaction with the plants.

Figure 5

Although shoot weight and plant height generally declined with increasing inoculum density, treatment effects were less pronounced than those observed for root weight (Fig. 4, Table 3). Mean shoot weight declined by 12.9% (P = 0.757), 20.0% (P = 0.500), and 34.2% (P = 0.191) in the low-, medium- and high-inoculum treatments, respectively, but these reductions were not statistically significant. In contrast, plant height decreased by 9.0% (P = 0.144), 16.9% (P = 0.005), and 17.2% (P = 0.008), in the low-, medium- and high-inoculum treatments, respectively, with significant reductions observed at the medium- and high-inoculum levels. Overall, these results indicate that root growth and tuber formation were more strongly affected by nematode inoculation than shoot biomass although plant height was significantly reduced at higher inoculum densities (Tables 2 and 3).