Characterization of Pterotylenchus cecidogenus in Desmodium ovalifolium cover crop from oil palm plantations in central Colombia

Abstract Until recently, the stem gall nematode Pterotylenchus cecidogenus was only registered in eastern Colombia. However, the disease has recently been observed in central Colombian oil palm plantations that use Desmodium ovalifolium as a cover crop. Soil, root, stem, and leaf samples were collected from D. ovalifolium. Plants showed foliar yellowing, leaf drying, and galls within stem nodes. Nematodes were identified, and the distribution, population density, and relative importance of different genera were determined. We performed morphometric and molecular identification of nematodes associated with gall symptoms. The D2-D3 segment of the large subunit-28S of ribosomal ribonucleic acid (RNA) and internal transcribed spacer (ITS) was sequenced, and phylogenetic analysis was performed. P. cecidogenus mainly occurred in the galls and to a lesser extent in the roots and soil. Nematodes were not found in leaf or inflorescence tissue. Morphological and morphometric data confirm the presence of P. cecidogenus in the stems of D. ovalifolium with gall symptoms. This study is the first to report deoxyribonucleic acid (DNA) sequences of P. cecidogenus. Based on D2-D3 and ITS partial sequences, P. cecidogenus is a sister species of the leaf-galling nematode Ditylenchus phyllobius (Sinm. Orrina phyllobia).

Colombia is the fourth largest palm oil producer in the world and the first in Latin America. The country has more than 535,000 hectares planted in 112 towns across 20 states, making palm oil one of the main national agricultural sectors (Fedepalma, 2018;SISPA 2019). However, Colombian oil palm production is affected by two lethal diseases (sudden wilt and lethal wilt). In both cases, the primary management strategy is to eliminate grasses and establish legume cover crops (Arango et al., 2011;Sierra et al., 2014).
Legume cover crops are widespread and are considered essential components of productive systems, including oil palm, rubber, coffee, and bananas (Baligar et al., 2006). For oil palm, the cover is established in the immature stage of the crop, during which palm foliage cannot protect the soil from solar radiation, wind, and erosion (Ruíz and Molina, 2014). D. heterocarpon (L.) DC. (Fabaceae) (cv. Maquenque) is commonly used in palm oil plantations; alternatively, Desmodium ovalifolium is sometimes favored owing to of its covering capacity and resistance to shading. However, on the eastern Colombian plains, D. ovalifolium is affected by the stem gall nematode Pterotylenchus cecidogenus (Varón et al., 2017).
The first symptoms associated with P. cecidogenus in Desmodium are shoot leaves with chlorosis, which later wilt and finally causes necrosis and defoliation. The stems and shoots of plants with these symptoms host galls located at the nodes. Young galls are small and light brown; however, they form a subberous that cracks and dries out, turning the gall a dark, almost black color. The galls cannot detach without damaging the stem. The nematode destroys the stem's cortical and vascular tissues, causing the death of the plant (Lenné, 1983;Lehman, 1991;Varón et al., 2017).
The nematode does not inhibit Desmodium seed germination but reduces plant survival and root and stem growth (Stanton, 1986). Gall formation is related to the nematode population and age of the plant. In Stanton (1986), the nematode population increased 100 times 52 days after plants were inoculated. Although the nematode does not need wounds to penetrate the stem, injuries caused by mechanical damage are quickly colonized by the parasite (Lehman, 1991). This nematode's life cycle takes approximately 2 weeks and presents four juvenile stages (the first occurring inside the egg) for female members; thus far, male members have not been identified (Stanton, 1990). In addition, Stanton (1990) found that P. cecidogenus moves faster from dead tissue than within tissues of D. ovalifolium plants, and that nematodes probably move from one site to another by a film of water outside the stem. Nematodes move very little in the soil, and so it is difficult to find them in soil samples.
The stem gall nematode P. cecidogenus was first identified as a new genus and species belonging to the Anguinidae family in 1981, affecting Desmodium from Carimagua in the Llanos Orientales of Colombia. In the original description, morphological and morphometric data were reported for P. cecidogenus (Siddiqi and Lenné, 1984). However, thus far, no molecular data are reported for this nematode in worldwide databases; therefore, its evolutionary relationships based on deoxyribonucleic acid (DNA) sequences remain unknown.
The cultivation of oil palm is a growing industry in the Colombian agricultural sector. Until recently, P. cecidogenus was only registered in eastern Colombia. However, the disease has recently been observed in central Colombian oil palm plantations that use Des modium ovalifolium as a cover crop. The D. ovalifolium is affected by yellowing and subsequent drying of the plants, causing losses for palm growers. As such, timely management strategies to prevent the dissemination and/or establishment of the disease to other plantations is needed. Thus, the present study had the following objectives: (i) to verify the presence of P. cecidogenus in D. ovalifolium from the central region of Colombia, (ii) confirm the taxonomic identity of P. cecidogenus through morphological, morphometric, and molecular analyses, and (iii) determine the phylogenetic relationships of P. cecidogenus through molecular data.

Symptom observation and sampling
We visited 30 oil palm plantations (Elaeis guineensis) of between 3 and 6 years of establishment with Desmodium ovalifolium as a cover crop. To verify the health status of the cover crop and identify plants with abnormal aspects, one sample per lot was taken, each of which was composed of five subsamples (Table 1) Symptomatic tissue samples as stems (with or without galls), leaves, pods, and inflorescences were collected and kept in labeled plastic bags and refrigerated until processing. Simultaneously, at locations where plants with symptoms were found, samples of young roots and rhizospheric soil were collected from close to the area with the highest concentration of roots in order to determine if the nematode affected the root system or survived in the ground.

Sample processing
Nematodes were extracted using the oxygenationdecantation method (Ravichandra, 2014). In brief, 1 g of fresh tissue from each of the sampled organs was cut into small portions and placed in a decantation sieve without a paper towel but with enough water to cover the sample. After 24 h, the decantation plate with the nematode suspension was removed and concentrated to 20 mL with a 400-mesh sieve (Varón de Agudelo and Castillo, 2001). Soil nematodes were extracted by suspension, filtration, and decantation methods (Ravichandra, 2014). In brief, 100 cm 3 of soil was placed in a container with water. After stirring for 2 min, the suspension was passed through a series of three sieves arranged from larger to smaller mesh diameters (reference No. 20 = 840 µm, 200 = 75 µm, 400 = 37 µm). The contents of the last two sieves were collected for decantation using a sieve previously arranged with a paper towel, and rested on a decantation plate with sufficient water. After 24 hr, the nematode suspension contained in the decantation plate was removed and concentrated to 20 mL with a 400-mesh sieve (Varón de Agudelo and Castillo, 2001).

Phylogenetic analysis
The consensus sequences obtained (12 of D2-D3 and 13 of ITS) were edited using the Geneious software (Kearse et al., 2012). Once the sequence editions were carried out, their identities were confirmed using the BLASTn software (http://www.ncbi. nlm.nih.gov/BLAST). Subsequently, the sequences presented under the accession numbers in Table 3 were independently aligned and analyzed using the MUSCLE algorithm included in the MEGA6 program (Tamura et al., 2013). Based on both obtained matrices, the nucleotide substitution models were determined by taking into account the Bayesian information criterion (BIC) using the ModelGenerator v.0.851 software (Keane et al., 2006). The phylogenetic relationships based on D2-D3 and ITS were determined by the maximum likelihood (ML) method based on the Tamura-Nei model (Tamura et al., 2013), which was used to model the differences in evolutionary speed between locations. Internal support of the nodes was performed using the bootstrap method with 1000 replicates. The Cervidellus cervus sequence was used as an external group (HM452377) for D2-D3 and Radopholus similis (GQ281456) for ITS.

Quantification and analysis of nematode populations
To quantify the nematode populations present in 1 gram of fresh tissue in each sample, three aliquots of 1 mL were taken and counted in a chamber under a light microscope (Olympus PX40, Allentown, PA) and an adapted camera (Olympus DP 73, Allentown, PA). The same methodology was used to quantify the nematode population in 100 cm 3 of soil from each sample. For the analysis of the populations of nematodes present in the soil and tissue of Desmodium plants, parasitic and ecological parameters such as frequencies and prominence values were taken into consideration, including (Norton, 1978;Volcy, 1998): Absolute frequency = (number of samples in which a genus appears/total samples evaluated) * 100; Relative frequency = (absolute frequency of the nematode/sum of absolute frequencies) * 100; Absolute density = number average of individuals per 100 g of soil or number of individuals per 1 g of tissue; Relative density = (absolute density of the nematode genus/sum of the absolute densities of all genera) × 100; Prominence value = absolute density * √ absolute frequency; Relative prominence value = (genus prominence value/sum of the prominence value of all genera) × 100.

Description of symptoms
We collected 30 samples from 30 plots located on 28 palm oil plantations in the departments of Cesar (26), Santander (3), and Norte de Santander (1), all of which utilized Desmodium ovalifolium as the cover crop (Table 1).
The symptoms observed in diseased plants of D. ovalifolium were characterized by yellowing, chlorosis, drying of leaves, wilting of shoots, and death of branches and plants (Fig. 1). In the stem, galls were observed in the nodes, more frequently in the basal part of the stems, and in some cases in the upper nodes. In the initial stages, galls were light brown in color; as they became older, galls became dark brown. The oldest galls presented a cracked suberous, which, when dried, acquired a dark, almost black color. The galls were an integral part of
-GQ281456 Múnera et al. (2010) the stem, and it was not possible to separate them without damaging the tissue (Fig. 1E).

Morphological and morphometric identification of the nematode
The nematodes extracted from D. ovalifolium stems with gall symptoms presented morphological and morphometric characteristics similar to those reported for the nematode P. cecidogenus. Females were morphologically distinguished by the following features: a post-mortem habitus that was straight or slightly ventrally arcuate, a vulva covered by large flaps, and a that was tail elongate-conoid to a sharply pointed tip; no males were identified (Table 2; Fig. 2).

Molecular characterization and phylogenetic analysis
The amplification of the segment D2-D3 and ITS from the ribosomal DNA region yielded amplicons of 728 and 1000 bp, respectively. The comparison of the sequences against the GenBank database did not present percentages of similarity equal to or greater than 99% with other reference sequences previously reported. However, the sequences of segment D2-D3 were very similar to those of KT192617 and KT192618 (identity levels of 81.06% and 81.14%, respectively; E-value: 0.0) of the species Ditylenchus phyllobius (Sinm. Orrina phyllobia). Similar results were obtained with the partial sequences of ITS, with 88% similarity to D. phyllobius (KT192616.1; E-value of 0.0). The partial sequences obtained in this study are the first reported for P. cecidogenus and were deposited for consultation in the NCBI database (Table 3). Phylogenetic analysis based on the use of segment D2-D3 comprised a total of 36 taxa and 824 characters, including gaps, of which 166 were conserved, 636 were variable, and 363 were informative parsimonious sites. Phylogenetic analysis based on the ITS region included 35 taxa and 1499 characters, including gaps, of which 717 were conserved, 736 were variable, and 645 were informative parsimonious sites. In both analyses, the maximum likelihood algorithm grouped P. cecidogenus consensus sequences of this study with high bootstrap values of 100%, in a clade separated from other species of the Anguinidae family but close to the clade of Ditylenchus phyllobius (Sinm. Orrina phyllobia) with 100% bootstrap support (Figs. 3, 4).

Quantification and analysis of nematode populations
The juvenile stages and females of P. cecidogenus were recovered in 56.7% of the tissue samples with galls, with an average population of 1,768 individuals per gram of fresh tissue. In addition, the juvenile and female stages were present in 10% of the node samples, with an average population of 23.8 individuals per gram of fresh tissue. Pterotylenchus was not detected in leaf samples. Only three inflorescences were analyzed because our sampling period did not match the flowering season.
Analysis of the parasitic and ecological parameters in roots showed that the genera of parasitic nematodes of plants with the highest relative importance in Desmodium roots were Meloidogyne (second-stage juveniles) followed by Pterotylenchus, Helicotylenchus, and Pratylenchus (juveniles and females). Xiphinema is a plant-parasitic nematode that is less important (Table 4). In relation to other genera, Pterotylenchus presented a prominence value of 46.7, with a low distribution (13.3) and population level (12), and eight individuals per gram of fresh root. Differences in the genera Pterotylenchus were associated with rhizospheric soil and Desmodium roots; however, their parasitic activity was not determined.
In the rhizospheric soil of Desmodium, the genera of parasitic nematodes of plants of greater relative importance were Helicotylenchus, followed by Tylenchorhynchus (juveniles and females). Xiphine ma (juveniles), Meloidogyne (second-stage juveniles), and Pterotylenchus (juveniles and females) were of intermediate importance. Pratylenchus (juveniles and females), Criconemella, and Trichodorus (juveniles) were recorded in the soil samples, but their distribution and population levels were low (Table 5). Pterot ylenchus had a prominence value of 11.1, with a low distribution (10) and population level (3.5 individuals per 100 cc of soil).

Discussion
The symptoms observed in diseased plants of D. ovalifolium used as a cover crop in oil palm in central Colombia include yellowing, foliar drying, wilting, and plant death, and are similar to the descriptions made in previous research (Lenné, 1983;Siddiqi and Lenné, 1984;Stanton, 1986;Lehman, 1991;Varón et al., 2017).
Based on morphological and morphometric diagnosis, the presence of P. cecidogenus was confirmed in the stems of D. ovalifolium with gall symptoms. The morphometric measurements registered in this study were similar to those reported for P. cecidogenus females in the original description (Siddiqi and Lenné, 1984). According to the results obtained with the BLAST tool and phylogenetic analysis with D2-D3 and ITS sequences, P. cecidogenus is a sister species of the leaf-galling nematode D. phyllobius (Sinm. Orrina phyllobia; Medina et al., 2016). These results are consistent with those of Siddiqi and Lenné (1984), who reported that P. cecidogenus is a unique species of the Anguinidae family with vulval flaps, but morphologically similar to Orrina in lacking a muscular median esophageal bulb and females that are not obese.
The stem nematode P. cecidogenus in D. ovalifolium presented low absolute frequency and prominence values because the nematode affects the aerial part of the plant; its occurrence in soil and roots does not indicate that it feeds on the root system. However, its presence is possible because when the plants die, the galls remain in the soil, giving the nematode the possibility of feeding in nearby plants (Lenné, 1983).
The nematode Pterotylenchus was not detected in the leaf samples analyzed. It is known that this parasite is a nematode that induces galls in the nodes of diseased plants, which explains the high frequency and population found in the galls. Although there were no visible symptoms in the case of nodes, it is possible that they were initiating infection and the galls had not yet formed. Although Pterotylenchus appeared in roots, according to Lenné (1983) and  Stanton (1986), the parasitic nematode of D. ovalifolium induces galls in the nodes of the plants and is not found in the other organs of the plant.
In conclusion, in this study, we found a new distribution of the stem-gall nematode P. cecidogenus affecting D. ovalifolium plants in central Colombia. This identification was confirmed using molecular tools and constitutes the first report of this technique for this species. This study confirms the spread of the nematode to new regions of the country. This could reflect new movement of plant material or asexual propagation because, since it was first recorded in 1981, it has been restricted to eastern Colombia (Siddiqi and Lenné, 1984). We suggest avoiding the moving plant tissue of Desmodium from regions reported with stem-gall nematode P. cecidogenus infestations to other areas of Colombia.