Isolation of nematodes

We collected different beetles at the adult stage using pheromones, sweeping nets, blacklight traps and pitfall traps baited with dung. The beetles were transferred to the lab alive, sacrificed by cutting them in half transversely, and put on NGM agar plates (6 cm diameter) seeded with 300 μl of the slowly growing E. coli strain OP50. The plates were checked daily using a Zeiss Stemi 2000 dissecting scope over a period of one to three weeks for emerging and reproducing nematodes. From the emerging nematodes we produced isogenic lines by transferring single gravid females or hermaphrodites to new plates. To determine whether the isolated nematodes belonged to gonochoristic or hermaphroditic species, virgin larvae were singled out onto plates. The presence of offspring indicated that they represented a hermaphroditic species.

Emerging nematodes were determined to family level with a Zeiss Stemi 2000 dissecting scope and to genus level with a Zeiss Axioplan 2 microscope using the key by Sudhaus and Fürst von Lieven (2003). For determination several worms were transferred onto microscopic slides covered with a 0.5 mm thick layer of 5% agar and either immobilized by heating the slide over an open flame to about 60°C for a few seconds or anesthesized with sodium azide. Within the genus Pristionchus many species can not be identified by morphological methods. We therefore chose to use molecular tools and mating experiments with reference strains to distinguish the different species (Herrmann et al., 2006a, b).

Collection by pheromone traps

Pheromone baits (Fuji Flavour Co., Tokyo, Japan) for the Oriental beetle were provided by Dr. Paul Robbins, Geneva NY. These baits were either put into funnel traps made out of empty 1.5-L green tea PET bottles or directly put on the ground and observed for attracted beetles which were then caught manually.

Molecular species identification

Molecular species identification was done using the small subunit rRNA gene (SSU). In brief, genomic DNA from single nematodes was prepared using the NaOH lysis procedure described by Floyd et al. (2002) and as in Herrmann et al. (2006a). A 1-kb fragment of the SSU was amplified by PCR using the primers SSU18A (5′-AAAGATTAAGCCATGCATG-3′) and SSU26R (5′-CATTCTTGGCAAATGCTTTCG-3′) (Blaxter, 1998). The reactions were performed in 25 μl of 1× PCR buffer (Amersham Biosciences, Freiburg, Germany) containing 2.5 mM MgCl2, 0.16 mM each deoxynucleo-side triphosphate, 0.5 μM each primer, 1 μl of lysate, and 1 unit of Taq DNA polymerase (Amersham Biosciences). After an initial denaturation step at 95°C for 2 min in a T gradient thermocyler (Biometra, Göttingen, Germany), the amplification was performed by 35 to 40 cycles of 95°C for 15 sec, 50°C for 15 sec, and 72°C for 2 min, followed by 72°C for 7 min. The PCR mix was diluted 20x and 1 μl was used for sequencing of approximately 500 bp of the 5′-terminal end using the primer SSU9R (5′-AGCTGGAATTAC-CGCGGCTG-3′) and the Big Dye terminator protocol (Applied Biosystems, Darmstadt, Germany). Sequences were aligned manually using Seqpup 0.6f software for Macintosh (Gilbert, 1996).

Mating experiments

To confirm the species identification by the molecular sequence of a new strain, we performed mating experiments with the reference strain of the respective species (see below for definitions). Five virgin females were put on a plate with a small spot of E. coli OP50 together with five males of the reference strain of a certain species. On a second plate we picked the opposite sexes of the two strains to test for reciprocity. If there were no offspring after one week, the experiments were repeated two more times. If fertile offspring occurred, we considered the two strains to belong to the same species.

Isolate and strain definitions

We use the following definitions to distinguish “isolates” and “strains”. An isolate is an isogenic female line derived from a beetle sample and subjected to molecular and experimental analysis. After species identification, we established one distinct isolate per species and location as a strain. The strains permanently cultured in the lab have strain numbers and are also kept as frozen stocks. For each new species designated by molecular sequence analysis and mating experiments, one strain was defined as a reference strain. The number of strains can be higher than the number of beetles sampled, i.e., multiple strains from a given beetle.

Olfaction assay

Population chemotaxis assays were performed on 8.5 cm wide NGM agar plates, as previously described for Pristionchus species (Hong and Sommer, 2006b). Briefly, mixed stage nematodes containing mostly adults were washed three times in M9 buffer and then loaded onto the agar plates prepared with two point sources of odors (attractant and solvent control) and sodium azide on opposite sides of each plate. The chemotaxis index is defined as [the number of worms at the attractant site – worms at control site]/total number of worms scored. At least two separate experiments were conducted for each strain and pheromone concentration to control for worm-batch variation, and each experiment contained 3–5 replicates. On average, each replicate represented the outcome for 10–50 worms. Only the highest chemotaxis index was recorded, which peaked between 9–15 hours at 23°C. Pristionchus pacificus PS1843 is our standard strain for chemotaxis assays. (Z)-7-tetrade-cen-2-one (97% pure) and (E)-2-nonen-1-ol were purchased from Bedoukian Research, Inc. (Danbury, CT, USA). (E)-2-nonenal was obtained from Sigma-Aldrich (MO, USA).

Pristionchus pacificus was the most common Pristionchus species found on scarab beetles in Japan

To determine whether P. pacificus and other species of the genus Pristionchus are associated with scarab beetles in Japan, we analyzed beetles from more than 60 sampling sites on the main island of Honshu (Fig. 1). The scarab beetle fauna of Japan differs from those in Europe and North America. In total, more than 450 species of scarab beetles are known from Japan (Kawai et al., 2005). We investigated 1,650 beetles of more than 10 genera and obtained 40 individual beetles infested with Pristionchus (Fig. 1, Table 1), which represents an infestation rate of 2.4%. This number is substantially lower than infestation rates of scarab beetles in western Europe and the eastern United States, where 10.2% and 23.4% of the beetles were infested, respectively (Herrmann et al., 2006a, b). Interestingly, another nematode that was commonly found on Japanese Melolonthine was a member of the genus Diplogasteroides related to Pristionchus. This observation is similar to findings in Europe, where D. magnus and D. nasuensis can be found on scarabs of the genus Melolontha (Manegold and Kiontke, 2001; Herrmann et al., 2006a).

Fig. 1.

Map of Japan showing the distribution of sampled Pristionchus species. Locations; 1: Hiroshima, Kawajiri-cho, Norosan, 2: Hiroshima, Kumano-cho Hagiwara, 3: Hiroshima, Miharashi, Kui, 4: Hyogo, Himeji-shi, Shiratori, 5: Hyogo, Miki-shi, Kasa, 6: Hyogo, Miki-shi, Yoshikawacho, Kamiarakawa, 7: Hyogo, Sasayama-shi, Jyonan, 8: Hyogo, Amagasaki-shi, Mokawa, 9: Osaka, Takatsuki-shi, Karasaki, 10: Kyoto, Hachiman-shi, Nishiyama, 11: Kyoto, Jyoyo-shi, 12: Shiga, Otsu-shi, Hieidaira, 13: Shiga, Otsu-shi, Hieidaira, 14: Hukui, Obama-shi, Anou, 15: Hukui, Minami-echizencho, Nanjyo, 16: Gifu, Gunjyou-shi, Takawashi-cho, 17: Toyama, Oyabe-shi, 18: Gunma, Shibukawa-shi, Ishihara, 19: Osaka, Takatsuki-shi, Tsutsumicho, 20: Osaka, Takatsuki-shi, Tsutsumicho, 21: Nara, Kasuganocho, Kasugataisha.

Table 1.

Frequencies of Pristionchus isolates on beetles.

From the 40 individual beetles infested with Pristionchus, we generated a total of 43 distinct isolates. Of these 43 Pristionchus isolates only three represent gonochoristic strains, whereas the majority of the 40 isolates is hermaphroditic. SSU sequence analysis and mating experiments of the 43 Pristionchus isolates revealed that they fall into three distinct species. All hermaphroditic strains had SSU sequences that were identical to P. pacificus. Mating experiments confirmed that all new isolates from Japan belong to this species. Of the 40 P. pacificus isolates, 37 were obtained from the oriental beetle Exomala orientalis (Table 1). Taken together, P. pacificus represents the most common Pristionchus species of our scarab beetle survey and the oriental beetle E. orientalis represents an important insect host species in Japan.

In addition to the beetle samples, we also took soil samples and looked for the occurrence of Pristionchus nematodes. We obtained one additional gonochoristic isolate from soil samples, which was analyzed together with the beetle-derived gonochoristic isolates (see below).

E. orientalis in the United States of America is also a host to P. pacificus

The oriental beetle was introduced to the East Coast of the United States around 1920 (Friend, 1929; Hallock, 1933) and is a known pest in particular in the northeastern part of the US (Vittum et al., 1999). To determine if P. pacificus is also associated with E. orientalis in the US, we collected oriental beetles at Cold Spring Harbor Laboratory (Long Island, New York). From 20 oriental beetles, four were infested with P. pacificus (data not shown). Also, one of the three P. pacificus strains obtained in a field study in the United States in 2005 has been isolated from E. orientalis (Herrmann et al., 2006b). Therefore, we conclude that E. orientalis represents an important host species for P. pacificus both in Japan and in the US.

The gonochoristic isolates belong to three novel species

In addition to the 40 isolates of P. pacificus, we obtained four isolates that belong to gonochoristic species, three from scarab beetles and one from a soil sample (Table 1). SSU sequence analysis and mating experiments of these isolates revealed that they fall into three species, all of which were not observed in our surveys in Europe and the US (Herrmann et al., 2006a, b). Interestingly, the strain obtained from the soil sample is identical in sequence to a strain provided by Dr. Walter Sudhaus (FU Berlin, Germany) from another soil sample in Japan. Mating experiments confirmed that both soil sample-derived strains belong to the same species.

In their catalogue of the Diplogastridae, Sudhaus and Fürst von Lieven (2003) reported 27 valid species descriptions in the genus Pristionchus. However, none of these species was reported from Japan. The three species discovered in our survey in Japan have distinct SSU sequences, do not mate succesfully with any of the established European and American species, but are morphologically similar to other Pristionchus species that we culture in the laboratory. We consider them to be novel species based on the molecular SSU sequence data and the results of mating experiments (Fig. 2, Tables 1 and 2). We designate them as Pristionchus sp. 11, Pristionchus sp. 14 and Pristionchus sp. 15. We will provide more detailed morphological descriptions elsewhere. The SSU sequence data show complete identity of the 471-bp fragment analyzed within all members of a given species (Fig. 2). The SSU of P. sp. 11 differs from that of P. pacificus by three diagnostic substitutions (54G, 213C, 243C). Pristionchus sp. 14 is characterized by the combination of the unique substitution 78C, substitution 128T shared with P. sp. 6, and the substitutions 211T and 226A shared with P. pacificus and P. sp. 11. The diagnostic substitutions for P. sp. 15 are 47G, 67A, 82T, 118T, 120T, 122G, 123T, 148T, 165A, 198C, 201A, 233T, 238C, 268T, 276A, 282A, 419T, 459A, and 469G. The reference strains of all novel species are available as living cultures and frozen stocks in our laboratory and can be provided to other researchers upon request.

Fig. 2.

SSU sequence comparison of Pristionchus species. The alignment shows the part of the SSU sequences used for species identification. The sequences of the species found in Japan are compared to those from Western Europe and the Eastern United States (Herrmann et al., 2006a, b). Dashes (−) indicate identity to the majority consensus sequence at the top; asterisks (*) indicate alignment gaps. Numbering refers to the SSU segment obtained in this study. The sequences have been submitted to the GenBank database and are available under accession codes DQ270018–DQ270025, DQ419900–DQ419904, and EF216858–EF216860.

Table 2.

Description of three novel Pristionchus species

The E. orientalis sex pheromone (Z)-7-tetradecen-2-one is specifically attractive to P. pacificus

Previous studies indicated that four closely related Pristionchus species associated with beetles have distinct chemotaxis profiles toward plant and insect compounds (Hong and Sommer, 2006b). To test if compounds from E. orientalis could be used by P. pacificus nematodes to locate its natural host, we performed chemotaxis assays using (Z)-7-tetradecen-2-one, the known E. orientalis sex pheromone, as an attractant (Leal et al., 1994; Zhang et al., 1994) (Fig. 3). We found that (Z)-7-tetradecen-2-one is attractive to only P. pacificus at 10- to100-fold dilutions. Closely related P. maupasi, P. entomophagus and P. uniformis were either not affected or were repulsed by this pheromone (Fig. 3). Furthermore, the model nematode C. elegans, which has not been found to be associated with beetles, was also repulsed by (Z)-7-tetradecen-2-one (Fig. 3). Finally, we tested two other known pheromones from closely related Anomala species, (E)-2-nonenal (A. schoenfeldti) and (E)-2-nonen-1-ol (A. vitis), but found no attraction of P. pacificus (Leal et al., 1992; Hasegawa et al., 1993; Toth et al., 1994). Therefore, consistent with our field findings, P. pacificus is specifically attracted to the sex pheromone of its preferred natural host, and may possibly be able to distinguish closely related Exomala/Anomala species using beetle species-specific pheromones.

Fig. 3.

Chemoattraction assay of closely related Pristionchus species and the distantly related model nematode C. elegans using the Exomala orientalis female sex pheromone (Z)7-tetradecen-2-one (Leal et al., 1994; Zhang et al., 1994). (Z)-7-tetradecen-2-one is highly attractive only to Pristionchus pacificus, whereas it is highly repulsive to P. uniformis and C. elegans, and neutral for P. maupasi and P. entomophagus. Error bars denote 95% confidence intervals. Pristionchus pacificus attraction at 10% and 1% dilutions of (Z)-7-tetradecen-2-one is significantly different to all other species except for P. uniformis at 1% attractant concentration (Student's T-test).

We identified P. pacificus as the most frequent Pristionchus nematode in this survey and consider E. orientalis as an important host species of P. pacificus. Previous sampling throughout the world revealed that P. pacificus is a widespread but rare species (Srinivasan et al., 2001; Zauner et al., 2007). Pristionchus pacificus was found in lower frequencies than other Pristionchus species in previous samplings, either in soil or in scarab beetle surveys in Europe, the US and South Africa. By contrast, our field studies in Japan yielded 40 new P. pacificus isolates and represent 66% of all available P. pacificus strains. Also, for the first time, our collection came from several independent isolates from the same type locality and/or beetle. Therefore, this collection will be very helpful for future studies of the population structure of P. pacificus.

The second major result of this study is the identification of a beetle host for P. pacificus. The oriental beetle E. orientalis was the origin of 93% of all Japanese P. pacificus isolates described here. These findings from Japan are consistent with the observation that one of three P. pacificus strains obtained from similar samplings trips in the US in 2005 was also from E. orientalis (Herrmann et al., 2006b). Furthermore, a small survey at Cold Spring Harbor Laboratory on Long Island (New York) identified four additional P. pacificus strains among 20 oriental beetle specimens sampled. However, it should be noted that the very high number of isolates from Japan might, in part, be the result of a sampling bias: after the first identifications of P. pacificus on E. orientalis, the use of sex pheromone traps significantly increased the numbers of E. orientalis beetles in this study. The use of pheromone traps also results in a strong sex bias, as only male beetles can be collected. However, in previous surveys in Europe (Melolontha and Geotrupes) (Herrmann et al., 2006a) we never observed any sex bias in the infestation rate. Given this reservation, our findings can identify E. orientalis as a major host of P. pacificus. The observed infestation rates, however, cannot be directly compared to infestation rates observed in our previous samplings because of different climate and soil ecology in the sampled areas. Nonetheless, we feel that the use of pheromone traps also provides a unique advantage for future studies because it provides access to beetle material around the world.

The identification of a beetle host for P. pacificus encouraged us to investigate for possible host cues involved in P. pacificus attraction toward E. orientalis. We found that only P. pacificus is attracted to the E. orientalis sex pheromone (Z)-7-tetradecen-2-one, whereas its close relatives were not attracted. Although the 10% concentration needed for robust P. pacificus attraction is high compared to the 1–10% effective range for other P. pacificus attractants with similar molecular masses, we speculate that unknown minor beetle attractants may synergize with (Z)-7-tetradecen-2-one to become a more powerful bouquet for P. pacificus in nature. Alternatively, conventional population chemotaxis assays may not be sensitive enough to determine the necessary pheromone levels for successful P. pacificus infestations, since the developmental stage for locating the host and the ability to colonize as single hermaphrodites have not yet been investigated. We hope to address these open questions in more controlled settings in the future. Nevertheless, P. pacificus' unique attraction towards the sex pheromone of its natural host underscores an emerging theme for nematode-beetle interactions showing that insect pheromones play a crucial role in host selectivity (Hong and Sommer, 2006b).

Exomala orientalis has a very broad distribution in the Japanese region and it has been suggested that the species was introduced to Hawaii before 1908 and to New Haven, Connecticut in the 1920s (Vittum et al., 1999). It has not been isolated in Europe, but several closely related species of the genus Anomala are known from various locations around the world. For example, A. solida is a vineyard and orchard pest in southeastern Europe (Toth et al., 1994). This observation is of particular importance, as one of the two strains of P. pacificus from Europe (RS5171) is from a vineyard in Montenegro. Future studies will investigate if there is any association between other Anomala species and P. pacificus and if the molecularly divergent P. pacificus laboratory strain PS312 from California (which is untypical and does not correlate to most other American isolates of P. pacificus) might represent a vine-related introduction that is independent of the other American strains.

The identification of a beetle host for P. pacificus is most useful for studies in evolutionary developmental biology and allows a progression of developmental studies from laboratory conditions towards the natural environment. Evolutionary developmental biology ultimately requires the incorporation of ecological factors (Zauner et al., 2007). This study presents the basis for the incorporation of ecological and behavioral aspects into nematode evolutionary developmental biology. Following the identification of E. orientalis as a host of P. pacificus, we will set out to establish laboratory cultures of E. orientalis with the hope to better understand the nematode life cycle in this more complex ecological context. Ultimately, such studies might provide evidence for how developmental processes that are usually studied under extremely artifical laboratory conditions are influenced by environmental factors. They might indicate how selection can act on these processes to result in well-adapted developmental patterns. Given the surprising amount of developmental flexibility seen in homologous developmental processes, such as vulva formation in P. pacificus and C. elegans, the incorporation of environmental factors might be very useful.