Artificial induction of third-stage dispersal juveniles of Bursaphelenchus xylophilus using newly established inbred lines

The pine wood nematode, Bursaphelenchus xylophilus, is the causal agent of pine wilt disease. This nematode has two developmental forms in its life cycle; i.e., the propagative and dispersal forms. The former is the form that builds up its population inside the host pine. The latter is specialized for transport by the vector. This form is separated into two dispersal stages (third and fourth); the third-stage dispersal juvenile (JIII) is specialized for survival under unfavorable conditions, whereas the fourth-stage juvenile (JIV), which is induced by a chemical signal from the carrier Monochamus beetle, is transported to new host pines and invades them. Because of its importance in the disease cycle, molecular and chemical aspects of the JIV have been investigated, while the mechanism of JIII induction has not been sufficiently investigated. In an effort to clarify the JIII induction process, we established inbred lines of B. xylophilus and compared their biological features. We found that the total number of nematodes (propagation proportion) was negatively correlated with the JIII emergence proportion, likely because nematode development was arrested at JIII; i.e., they could not develop to adults via the reproductive stage. In addition, JIII induction seemed to be regulated by a small number of genes because the JIII induction proportion varied among inbred lines despite the high homozygosity of the parental line. We also demonstrated that JIII can be artificially induced by the nematode’s secreted substances. This is the first report of artificial induction of JIII in B. xylophilus. The dauer (dispersal) juvenile of the model organism Caenorhabditis elegans corresponds functionally to JIII of B. xylophilus, and this stage is known to be induced by a chemical signal referred to as daumone, derived from the nematodes’ secretion. The artificial induction of JIII suggests the presence of daumone-like material in B. xylophilus.

Introduction [26][27][28][29][30][31]. However, compared with the J IV , the J III stage has not been sufficiently investigated for its biological characteristics and induction, despite its importance for nematode dispersal. J III is the only stage that molts to J IV (i.e., J III formation is a prerequisite for J IV formation) and it can survive for a long period, which allows it to hide inside trees and thus become a source of the pathogen. Therefore, the mechanisms and detailed conditions for J III induction must be clarified to further our understanding of vector association and transportation of PWNs.
The model organism C. elegans has a specific stage, called the dauer stage [32], which is morphologically and functionally similar to the dispersal stages of B. xylophilus [24,33]. Dauer-stage worms can survive for about 3 months, which is much longer than the propagative life span of the worm (ca. two weeks), and this stage is induced by the chemical signals that indicate lack of food, bacteria, and high population density [34][35][36]. Dauer-inducing pheromone (daumone) is the main substance signaling nematode population density [37]. Small molecule pheromones, called ascarosides, which induce daumone, are associated with various behaviors such as gender-specific attraction, repulsion, aggregation, olfactory plasticity, and dauer induction [36,[38][39][40][41][42][43][44]. Several studies have suggested that J III of B. xylophilus is induced by a mechanism similar to the dauer induction of C. elegans, i.e., B. xylophilus is assumed to produce daumone-like materials [45].
In this study, as a first step in the investigation of J III induction, we established several inbred lines of B. xylophilus with various J III formation proportions, and developed a method for artificial induction of J III using crude extracts of cultured nematodes, which is expected to include a dauer-inducing pheromone-like substance, to understand J III induction in this nematode.

Nematode
In this experiment, four Japanese field isolates of B. xylophilus, namely, virulent S10, T4, and Ka4, and avirulent C14-5, which is physiologically and biologically different from other strains and generally used as a comparative strain for other highly virulent strains [46,47], were used as the candidate nematode materials for the following experiments. These isolates were considered as candidates for the parent strains of inbred lines with high J III -emerging proportions. S10, T4, and Ka4 were originally collected from dead trees of the Japanese red pine, Pinus densiflora, in the Shimane Prefecture in 1982, the Iwate Prefecture in 1992, and the Ibaraki Prefecture in 1994, respectively. C14-5 was obtained from an adult M. alternatus beetle in Chiba Prefecture in 1975 [48]. These isolates have been maintained for more than 20 years as laboratory cultures in the Forestry and Forest Products Research Institute, and thus, no special permit was required to use them. For S10, T4, and Ka4, nematodes were reared on a fungal lawn of Botrytis cinerea on potato dextrose agar (PDA) media, and kept at 25˚C for 7 days, while C14-5 was kept at 25˚C for 16 days because the propagation of C14-5 is slower.
Hereafter, we define isolate as a field-collected population without genetic purification, and inbred line as the genetically purified population derived from the field population (= isolate).

Investigation of the emerging proportion of J III
Propagated nematodes were collected using the Baermann funnel technique, and surface-sterilized using lactic acid following the method described in Mamiya et al. [49]. In short, nematodes were collected in a 10-mL glass tube, washed three times with distilled water, and the same volume of ca. 6% lactic acid solution was added to the nematode suspension. After 30 sec of soaking the nematodes in the lactic acid solution, the suspension was centrifuged at 1,500 rpm for 30 sec, and the supernatant was removed. Nematodes were then rinsed three times with sterilized water. Fifty surface-sterilized nematodes of each isolate were reared on the hyphae of Cosmospora viridescens, which is frequently used in experiments involving dispersal stage induction [50,51], growing on malt extract agar (2% malt extract, 1.5% agarose) in 4-cm-diameter Petri dishes, and kept at 25˚C for 10, 20, and 30 days. Each treatment was repeated eight times.
Then, nematodes were collected using the Baermann funnel technique, and killed by heat (ca. 60˚C for 1 min), and fixed in TAF (2% triethanolamine, 2.775% formaldehyde) [52]. Fixed nematodes were kept at room temperature until microscopic observation. The numbers of total nematodes and J III were counted for each sample using a light microscope (Eclipse 80i, Nikon), and the proportion of J III was calculated.

Establishment of inbred lines and evaluation of their J III emerging proportion
Field-collected isolates of B. xylophilus usually have high genomic diversity [53]; therefore, it is necessary to establish inbred lines for molecular and biochemical analyses. We selected and used the isolate T4, which showed the highest J III formation proportion in the above-mentioned test, as the parental isolate.
Botrytis cinerea was inoculated on 4% plain (water) agar in 48-well microplates, and kept at 25˚C for 1 day. Distilled water (400 μL) was added on a fungal mat in the microplate, and one T4 J III nematode was inoculated into the water in each well, and reared for 2 days at 25˚C to obtain unmated adult females and males. Botrytis cinerea was cultured on 4% plain agar in 4-cm-diameter Petri dishes for 1 day. An unmated female and unmated male were then transferred to fungal hyphae on the dish. The culture plates were kept at room temperature and observed every day, and a young gravid female of the next generation was transferred to a new dish. Offspring were isolated using a microplate and used for the next mating. This procedure was repeated for 12 generations, and eight inbred lines were thus established from T4, and coded as ST1 to ST8.
The J III formation proportions of the eight newly established inbred lines were examined with the method described above.

Extraction of J III -inducing pheromone
Based on studies of C. elegans and B. xylophilus, we hypothesized that B. xylophilus has a substance like daumone, which is a heat-tolerant and water-soluble material [18,35,37,45], and tentatively named it DBX for daumone of B. xylophilus. Therefore, we prepared crude DBX for artificial induction of J III . In this experiment, we used the ST2 line, which had the highest J III production proportion in the J III induction test. ST2 nematodes were reared for four days on B. cinerea grown on 10-mL barley grains in a 50-mL Erlenmeyer flask. Nematodes were collected using the Baermann funnel technique for three hours, rinsed with distilled water three times, and a sterilized water suspension of the nematodes (10 worms/μL) was prepared. The suspension was shaken using a Triple Shaker NR-80 (Taitec, Japan) at 95 rpm for 48 hours at 25˚C, and the suspension was filtrated using filter paper (Kiriyama Filter paper No. 5A, 7 μm; Kiriyama Glass, Tokyo) twice to remove worms. The filtered solution supposedly containing DBX was concentrated by freezedrying, and the dried materials were dissolved in sterilized water, adjusting to 2,000,000 nematodes' secretions per 3 mL. The solution was tentatively called CDBX for crude DBX.

Artificial induction of J III
CDBX (200 μL) was inoculated on the surface of 4% plain agar in 4-cm Petri dishes, and 20 μL of yeast (Saccharomyces cerevisiae; 0.25 mg/mL or 1 mg/mL) suspension was added as food for the nematodes. In the control group, the same volume of sterilized distilled water was added instead of CDBX, along with the same amount of yeast. The plates were kept for two days at 25˚C. Thereafter, ten females and three males of ST2 were reared on each plate, and cultured for 5 days. Each treatment was repeated three times. After confirming the propagation (presence of offspring), we extracted the nematodes using the Baermann funnel technique overnight and counted the number of total and J III nematodes to calculate the J III proportion.
In addition to the above experiment, we examined J III formation under different conditions: without higher food concentrations (4 mg/mL suspension of S. cerevisiae), and with a shorter incubation (4 days) for all treatments. After four or five days of incubation, all juvenile nematodes were first generation progenies because the life cycle of B. xylophilus on yeast is longer than five days at 25˚C [22].

Statistical analysis
The statistical analysis was performed using R ver. 3.2.4. We calculated the emergence proportion of J III using the following formula: Proportion ð%Þ ¼ ½ðnumber of J III Þ = ðtotal number of wormsÞ Â 100 Data represent the mean ± standard error (SE). For the statistical analysis, we performed the normality test to reveal the normality of the data to decide whether to use parametric or non-parametric tests. For the J III emergence proportion of the isolates, Kruskal-Wallis oneway analysis was performed to test among isolates, and subsequent Steel-Dwass multiple comparison tests were conducted. To analyze J III emergence proportions of inbred lines, one-way analysis of variance (ANOVA) and post hoc Tukey-Kramer multiple comparison tests were conducted. When analyzing artificial induction of J III tests, one-way ANOVA and post hoc Tukey-Kramer multiple comparison tests were performed. To analyze the correlation of all numbers and J III proportions, we conducted Spearman's rank correlation test.

Proportion of J III formation in the conventional isolates
The results are summarized in Fig 3 and Supplementary S1 Table. Two virulent isolates, S10 and Ka4, propagated rapidly, and maintained high population densities over the 30 days of the experimental period. Conversely, virulent T4 and avirulent C14-5 propagated slowly (P < 0.01). However, the numbers and proportions of J III were significantly higher in T4 compared with the other three isolates (P < 0.01 for both).

Proportion of J III formation in inbred lines
The results are summarized in Fig 4, and Supplementary S2 Table. The total number of nematodes in lines ST4 and ST5 was significantly higher than in the other six lines, while the J III formation proportion of ST1 and ST2 was significantly higher than in the other lines, and that of ST4 and ST5 was significantly lower than the others. This tendency was supported by the significant negative correlation between the total number of nematodes and J III formation proportion (r 2 = 0.50, P < 0.01; Fig 5).

Artificial induction of J III
The results are summarized in Fig 6 and Supplementary S4 Table. The total number of nematodes was higher in the treatment group than the control group (P = 0.00001809), and the proportion of J III was also higher in the treatment group than in the control group (P = 0.0000002041). We also checked the 4-day incubation experiment. The results showed a similar tendency, i.e., the total number of nematode and the proportion of J III were higher in the treatment group than the control group (S1 Fig and S3 Table). When observing the plates, the nematodes that remained were mostly J 2 (data not shown).

Discussion
Many isolates of B. xylophilus have been isolated and maintained in laboratories, and are quite variable in some biological characteristics, e.g., virulence, propagation, and efficiency of phoresy [46]. However, regardless of its importance, J III formation proportion has not been closely compared among isolates. In this study, a T4 isolate showed a significantly higher J III formation proportion than the other three isolates (Fig 3, S1 Table). Interestingly, although the T4 isolate is known to propagate quickly [54], its population size was significantly smaller than the other two pathogenic isolates throughout the experiment. In previous studies, J III began to appear after the nematode population reached the maximum level, and the proportion and number of J III increased as the population decreased [23,24,55]. Considering the relationship between population size and J III formation proportion, nematodes of T4 are suspected to be more sensitive to J III induction signals, such as unfavorable conditions and the concentration of DBX, than other isolates, and they shifted to the dispersal life cycle in the early phase of incubation. In comparison, most individuals of the S10 and Ka4 isolates were J 2 after 30 days of the experimental period (S5 Table). This suggests that the nematodes of these two isolates, S10 and Ka4, were starved and suspended their growth before turning to J III . In the model model species, Pristionchus pacificus, has natural variation in pheromone production and sensitivity derived from genetic differentiation among several strains collected from different localities all over the world [63,64]. The variation of J III formation among PWN isolates seems similar to that of dauer formation among C. elegans and P. pacificus strains, and therefore, nematodes of the isolate T4 may have some mutations associated with J III formation so they more readily form J III compared with other isolates. In this study, we could not reveal which Artificial induction of dispersal juveniles of Bursaphelenchus xylophilus production or sensitivity of pheromone resulted in the high J III proportion in T4; further physiological and chemical analyses are required to reveal the answer. It is interesting that T4 nematodes have maintained a high J III production ability for a long sub-culturing history, despite Artificial induction of dispersal juveniles of Bursaphelenchus xylophilus extended sub-culturing of PWN isolates sometimes causing reductions in the pathogenicity against host pine trees and decreases in the proportion of J III formation [24,65]. In a previous study, T4 showed a high degree of homozygosity (only 4.26% of variants in the T4 isolate genome) [66], suggesting that the genetic diversity is rather low in this isolate. However, in the present study the proportion of J III formation varied among inbred lines derived from the isolate; i.e., T4 retains genetic diversity associated with J III formation within the highly homozygous isolate. The number of dauer-related (daf) genes varies among the nematode groups. For example, C. elegans has 1,259 daf genes in the latest version of gene association analysis (WS258) (http://www.wormbase.org), and only 52 daf-gene groups were identified in Pristionchus pacificus (http://www.wormbase.org). In the case of T4, although the isolate has low genetic diversity, the J III -related genes seem variable among inbred lines because of their discontinuous J III formation proportions, i.e., the J III emerging proportions of inbred lines were separated statistically into high (ST1, ST2, and ST3), medium (ST6, ST7, and ST8), and low (ST4 and ST5) proportion groups (Fig 4). Therefore, J III formation of T4 is likely regulated by a relatively small number of key genes as in P. pacificus.
The total number of nematodes was negatively correlated with the J III proportion; i.e., the lines showing a high J III formation proportion did not propagate well during the experimental period. This negative correlation is probably because of the developmental arrest in J III . J III is a kind of dauer (dormant) stage, and thus the individual that molted to J III does not develop to the reproductive stage (= adult) under certain conditions. Therefore, the next generation of nematodes was produced only by the adults that had emerged at an earlier stage of culture (= early generation of adults). However, in the lines in which the J III formation proportion was lower, nematodes developed to adults and reproduced more effectively than the lines with higher J III formation, and the total nematode population became larger in such lines.
After the first-generation adults were exposed to CDBX, most offspring of the next generation (J 2 ) molted to J III , suggesting the presence of water-soluble material(s) that induce J III formation. In C. elegans, daumone was first identified as ascaroside (-)-6-(3,5-dihydroxy-6-methyltetrahydropyran-2-yloxy) heptanoic acid [37]; i.e., an ascarylose component with a fatty acid side chain that can be dissolved in water. Ascarosides are often used as signaling molecules, such as dauer-inducing pheromone, in various nematode species including free-living and parasitic species (C. elegans, C. afra, Heterorhabditis bacteriophora, Allodiplogaster seani, Nippostrongylus brasiliensis, Oscheius carolinensis, O. tipulae, Parastrongyloides trichosuri, Pelodera strongyloides, P. pacificus, Rhabditis sp. AF5, Steinernema carpocapsae, S. glaseri, and S. riobrave) [67][68][69][70][71]. Therefore, CDBX as a J III inducer can be hypothesized to include ascaroside-like substances, which can induce the J III formation of B. xylophilus. It was reported that the dauer-inducing pheromones of C. elegans and P. pacificus did not induce dauer formation of P. pacificus and C. elegans, respectively, although the physical properties of those two pheromones were similar to one another [72]. This suggests that dauer-inducing pheromones are species-specific or affect only closely related (narrow-ranged) species; that of B. xylophilus should be no different. Further physiological analyses on the J III induction of B. xylophilus will be necessary to reveal this species specificity.

Conclusion
Bursaphelenchus xylophilus, the causal agent of pine wilt, is an important forest pathogen. However, the mechanism of induction of dispersal stages has not been studied in detail despite its importance for expansion of the disease. In the present study, we clarified that 1) the manner of J III formation is different among field isolates of B. xylophilus, 2) a set of newly established inbred lines varied in J III production proportion, and 3) water soluble substances of nematode secretion induced J III production. The inbred lines derived from the T4 isolate may be useful for investigating J III formation in future studies, e.g., RNA-sequencing analyses of nematodes after J III induction treatment to identify the genetic pathways of J III formation. Furthermore, the J III formation assay system established here will be applicable to chemical analyses of DBX.
J III of B. xylophilus (and its close relatives) is an interesting stage because it is induced in a similar manner to dauers of other nematode species, and has biological characteristics similar to dauers, e.g., long-term survival [45,73,74]. However, in B. xylophilus, it is not J III , but rather J IV , that is the true dauer stage [18,75]. By conducting further analyses of the molecular biology and chemistry associated with J III formation, information that is important not only for disease control, but also for developmental biology, will be revealed.