Grower-Based In Vivo Propagation of Entomopathogenic Nematodes

Entomopathogenic nematodes (EPNs), commonly represented by species of Steinernema and Heterorhabditis, have been relied upon as effective biocontrol agents of soil-borne insect pests for many years¹. EPNs forage in the soil for insect hosts, and when a nematode finds a suitable host, it enters the insect's body and deploys symbiotic bacteria to kill the insect². These symbiotic bacteria consume the host, multiply within the cadaver, and become the primary food source for the nematodes³. Ultimately, after multiple generations of nematodes within the host cadaver, the infective juvenile nematodes (IJs) emerge to search for new insect hosts. It is at this stage that the nematodes can be harvested and deployed as a bio-insecticide for crop protection. Farmers (and other agricultural stakeholders) can purchase formulations of such beneficial nematodes, but rarely can farmers grow their own nematodes to use as part of pest management programs.

Culturing and harvesting nematodes via typical production methods can be quite difficult. Small-scale production of EPNs is often based on in vivo methods, while at larger scales, in vitro production is used⁴. Most in vitro methods use homogenates of animal byproducts as the diet for the nematodes, thereby side-stepping the handling of insects. Not surprisingly, in vivo production can be challenging and expensive because it involves inoculation of live insect hosts, often using the White trap method⁵ with either Galleria mellonella L. ('waxworms') or Tenebrio molitor L. ('mealworms'). Culturing nematodes within insect hosts adds not only the high cost of live insects but also the labor associated with handling them.

As EPNs develop within their insect hosts, the cadavers show characteristic signs of infection corresponding to the EPN genus infecting them⁶. The cadavers can then be placed onto a moistened surface within a holding dish that has been placed inside a larger, water-filled dish⁷. When nematodes have completed their life cycle inside the host cadaver, the IJs exit the cadaver and are collected, typically using gentle irrigation with water, and then stored in a variety of media, including inert substrates like gels (Figure 1), sawdust, or vermiculite⁸. In vivo production generally reflects a low economy of scale due to labor and insect costs^7,9.

In vitro production is based on the rearing of EPNs on artificial media containing monoxenic cultures of EPNs and their symbiotic bacteria, Photorhabdus (for Heterorhabditis) and Xenorhabdus (for Steinernema)^4,10. This method necessitates advanced expertise and significant capital outlay for the materials and instrumentation attendant to high-throughput fermentation¹¹. As a result, reliable in vitro EPN production is likely to be cost-prohibitive for many farming operations.

For farmers and other agricultural stakeholders, the technical challenges associated with in vitro techniques can be circumvented by improving in vivo methodologies, particularly enhanced insect rearing¹² combined with automated nematode inoculation and harvest^7,13. Such in vivo methods can become substantially more efficient, given that they optimize EPN inoculation, production, and storage procedures^7,14,15,16. Further, in vivo production can be tailored to the needs of the end users by facilitating EPN production by the farmers themselves¹⁷. One basic disadvantage of this method is that the nematode inoculum must be resupplied to the growers from a commercial or publicly funded source for each round of production⁹, rendering the grower perpetually dependent on the externally provided inoculum source. Therefore, a sustainable, long-term EPN production method was needed for growers. The resulting sustainable methods are presented in this paper.

To explore these approaches, we set up two different in vivo production methods. In the first method, we compared in vivo production from the standard White trap approach to the novel grower-oriented method. The result is a relatively simple, low-cost approach for EPN production that is sustainable and easily adaptable to various EPN species⁹. One of the key advantages of this method is that a small subset of the EPNs produced in any 'batch' can be used to inoculate insect hosts for subsequent batches. Thus, once a grower has been provided the initial EPN inoculum, the external source material is no longer needed. This process is responsive to the continual use by the grower.

A similarly low-cost/low-input approach to nematode propagation has been developed in-parallel, focused largely on biocontrol of cranberry pests in Wisconsin, USA. Recently, two nematode strains of the species, Oscheius onirici (Nematoda: Rhabditidae) and Heterorhabditis georgiana (Nematoda: Rhabditidae), were discovered in the acidic peatlands of central Wisconsin¹⁸ and tested repeatedly in field studies^18,19. These species are native and have been shown to be highly virulent against historically problematic insect pests^18,19,20. As such, many farmers have been interested in using these nematodes as biocontrol agents for crop pests (sparganothis fruitworm and redheaded flea beetle). The nematodes can be mass-propagated in vivo using beetle larvae (Tenebrio molitor) that have been embedded within wet, well-aerated layers of a polymer-coated cotton wool (i.e., 'horse gauze'). Importantly, the fibers are wettable yet stiff enough to maintain the thickness of the gauze after water has been added. As a result, the bilayers of gauze do not compress when wet, facilitating better air circulation and nematode survivorship. The bilayer gauze system better simulates the pore spaces within moist soil and is suitable for nematode storage once the IJs have emerged. Harvesting of IJs from the gauze bilayer is readily accomplished via systematic irrigation of the gauze and collection of the nematode-rich slurry.

1. Obtaining the initial inoculation and basic procedures

NOTE: Mass production for the grower-based methods is based on infected insect hosts. In the first method, once the initial inoculum is produced, aqueous suspensions of nematodes are no longer needed to infect new hosts.

1. To start the initial inoculum, obtain the desired EPN species. Use the infective juvenile (IJ) stage of Steinernema carpocapsae (All strain) and Heterorhabditis bacteriophora (VS strain).
2. Using IJs, infect last-instar G. mellonella larvae using the White trap method⁷.
3. Inoculate live 50 G. mellonella larvae with IJs (100 IJs/insect) of S. carpocapsae or H. bacteriophora in Petri plates (90 mm) lined with filter paper. Keep the inoculated insects at 25 °C until IJs are near emergence (5-7 days for S. carpocapsae, 7-10 days for H. bacteriophora).
4. Proceed with the Continuous grower-based production process⁹.
   1. For mass inoculation, place one infected cadaver (e.g., infected with S. carpocapsae or H. bacteriophora) with 50 live G. mellonella larvae in a plastic container (5.7-14.2 L containers) lined with a moist paper towel and keep at 25 °C.
   2. After 72 h, remove cadavers not showing appropriate color changes (tan/brown for Steinernema; red/orange for Heterorhabditis) from the container to avoid contamination.
   3. Move the cadavers to containers with polyacrylamide gel powder in clear plastic containers and moisten with water (20 mL/dry gram of gel). Place the cadavers on top of a nylon screen mesh (2 mm × 2 mm pore size) to separate them from the gel⁷.
   4. Harvest the IJs from the infected hosts (which stay on the screen on top of the gel). Continue the harvest into the gel until ~5-7 days after IJs begin to emerge; remove and discard the depleted cadavers (IJs harvest should be allowed to continue until IJs diminish substantially, usually about 5-7 days).
      NOTE: IJs harvested into the gel can be used to start inoculum for a subsequent round of EPN production, used directly for application, or stored under refrigeration for later use for several weeks to months with no loss in viability or even 3-6 months to use for re-inoculation.
5. To develop inoculum for subsequent production rounds, expose healthy G. mellonella larvae to IJs within the gel in Petri dishes (90 mm). Then, use these newly infected hosts to inoculate the next group of G. mellonella larvae (as mentioned above in steps 1.2 and 1.3).
6. For EPN applications, rinse IJs that have entered the gel out using tap water or screens (pore size < 0.3 cm). Alternatively, apply EPNs in the gel directly to small arenas without rinsing.
   NOTE: EPNs can be used in aqueous suspensions and applied using standard grower practice (sprayers, irrigation, etc.)⁷. The gel can be stored for several weeks under refrigeration with slight reductions in EPN viability, while at room temperature it can be stored for no more than 2 weeks. Under refrigeration, the gel can be stored for months (exact times will vary by EPN species).

2. Steinernema carpocapsae experiment

NOTE: The objective of this study was to validate IJ production in a grower-based approach relative to the standard White trap approach by varying arena size.

1. Set up the experiment to comprise five treatments with different cadaver densities varying from 5 to 50 hosts per arena.
   1. T1: Control: Place 5 cadavers in White traps using Petri dishes (10.0 cm × 1.5 cm; diameter [D] × height [H]) or 0.06 cadavers/cm².
   2. T2: Set 30 cadavers in larger White traps (15.0 cm × 1.5 cm; D × H) in 14.2 L clear-plastic containers (43.2 cm × 28.3 cm × 16.5 cm; L × W × H ) or 0.025 cadavers/cm².
   3. T3: Set 25 cadavers in polyacrylamide gel (15 g of gel powder + 300 mL of tap water) in 5.7 L clear plastic containers (36.2 cm × 21 cm × 12.4 cm; L × W × H) or 0.033 cadavers/cm².
   4. T4: Set 30 cadavers in polyacrylamide gel (35 g of dry gel + 700 mL of tap water) in 14.2 L clear-plastic containers or 0.025 cadavers/cm².
   5. T5: Set 50 cadavers in polyacrylamide gel (35 g of dry gel powder + 700 mL of tap water) in 14.2 L clear-plastic containers or 0.041 cadavers/cm².
2. Use the standard White trap method (T1) as a control (see step 2.1).
3. Ensure that each treatment has four replicates and repeat the entire experiment twice.
4. Place the cadavers with gel treatments on top of the 2 mm × 2 mm nylon screen mesh.
5. Cover all the containers with loosely tightened perforated lids and keep at 25 °C.
6. Observe the White traps daily for IJ emergence and terminate 7 days after the first IJ emergence.
7. At the end, assess the total IJ production per insect based on standard dilution procedures⁶.
8. In the gel, count the IJs after mixing the gel thoroughly, recording the total volume of gel, diluting 10 mL of it in water (1:100 ratio), and stirring it for almost 20 min.

3. Heterorhabditis bacteriophora experiment

NOTE: This experiment was designed to test a grower-based production procedure for H. bacteriophora (procedures were based on the results of the S. carpocapsae experiment).

1. Set up the treatments: Place 5 cadavers in White traps in Petri dishes (100 mm D x 15 H mm) and 50 cadavers in polyacrylamide gel (35 g of dry gel powder + 700 mL of tap water) in 14.2 L clear-plastic containers.
2. Use the White trap method (see step 3.1) to serve as the control.
3. Repeat the whole experiment twice with five replicates of each treatment.
4. Perform the experiment and assess the IJ yield in the same way as the previous experiment (see section 2).

4. Oscheius onirici and Heterorhabditis georgiana propagation

NOTE: The objective of this study was to validate IJ production using a second grower-based approach that relied upon a commercially available polymer-coated cotton wool (horse gauze) as the culturing substrate. In this procedure, the horse gauze serves as the substrate for emerging IJs because its physical properties allow for both aeration and an efficient harvest of IJs. The horse gauze approach was compared to a standard nematode substrate (simple cotton wool).

1. Examine the two factors in a fully crossed two-way factorial design: Nematode Identity (O. onirici or H. georgiana) × Culturing Substrate (horse gauze or cotton wool). This experimental design (2 × 2) will create four unique treatment combinations, each replicated 7 times and arrayed in a completely randomized design.
2. For each replicate, place a horse gauze or cotton wool sheet in a shallow plastic tray (35.6 cm × 25.4 cm × 2 cm) to serve as the substrate for the insect hosts. Spread 200 T. molitor larvae (freeze-killed) evenly on the sheet. Place another sheet of the same dimensions over the larvae, creating three layers (a lower layer of wool, a layer of mealworms, and a top layer of wool). For both nematode species, prepare seven trays using horse gauze and seven trays using cotton wool.
3. Derive the IJ inocula for both O. onirici and H. georgiana from in vivo nematode cultures with T. molitor (mealworms) as the host on moistened cotton rounds.
   1. Place two small cotton rounds within a Petri dish and moisten them with a soak (10-15 mL) of tap water.
   2. Place 4-6 T. molitor larvae (live or freeze-killed) in a non-overlapping pattern between the two cotton rounds.
   3. Add 10 mL of a nematode slurry (400-500 IJs/mL) to the top cotton round, allowing the IJs to search for and infect the mealworms.
      NOTE: At 21-22 ˚C, the IJs begin emerging 2-4 weeks after inoculation.
   4. Harvest the IJs from the Petri dishes by pouring (twice) 20-25 mL of water over the top cotton round and collecting the slurry that has percolated through the cotton.
4. Infect the insect hosts within the large, shallow trays via a thorough drenching of the wool sheets with approximately 400 mL of a nematode suspension (80,000-90,000 nematodes/L). Allow the nematodes to infect the insect larvae and incubate for 4 weeks at 21-22 ˚C.
5. After 4 weeks, harvest the nematodes by irrigating each tray twice with drenches (500 mL each) of tap water, collecting a total of 1 L of nematode slurry.
6. Count the nematode density per mL under a dissecting microscope and tally the counts for each replicate tray.

S. carpocapsae experiment
Treatment effects were significant based on the mean numbers of IJs/cadaver (One-way ANOVA: F4,39 = 2.95, P = 0.0362). Only T2 and T4 were significantly different from each other, as T4 produced more IJs per cadaver (244,029 ± 16,241) than T2 (159,114 ± 9,669) (Figure 2).

H. bacteriophora experiment
A significant treatment effect was observed on the mean number of IJs per cadaver in this experiment (One-way ANOVA: F1,19 = 5.0, P = 0.040). The treatment based on the grower-oriented method had significantly more IJs/cadaver (186,389 ± 28,645) than the White trap treatment (124,470 ± 19,129) (Figure 3).

Oscheius onirici and Heterorhabditis georgiana propagation
The bilayer of horse gauze was equally effective as cotton wool at providing a culturing substrate for O. onirici and H. georgiana nematodes (Two-way ANOVA: main effect of Nematode Identity: F1,23 = 0.146; P = 0.706; main effect of Substrate Type: F1,23 = 0.104; P = 0.751; interaction between Nematode Identity and Substrate Type: F1,23 = 0.346; P = 0.563). Mean O. onirici per L from trays with cotton and horse gauze were, respectively (in millions), 101.11 ± 13.73 and 95.93 ± 20.19. Mean H. georgiana per L from trays with cotton and horse gauze were, respectively (in millions), 82.23 ± 18.69 and 99.93 ± 18.40.

Figure 1: Diagram of the grower-based method. Step 1: inoculation; Step 2: mass incubation; Step 3: transfer to polyacrylamide gel with screen. Abbreviation: IJs = infective juvenile nematodes. Please click here to view a larger version of this figure.

Figure 2: Mean Steinernema carpocapsae (All) infective juvenile (IJ) production per Galleria mellonella cadaver across five treatments. Abbreviation: IJs = infective juvenile nematodes. This figure was taken with permission from Oliveira-Hofman et al.9. Please click here to view a larger version of this figure.

Figure 3: Mean Heterorhabditis bacteriophora (VS strain) infective juvenile production per Galleria mellonella cadaver from a White trap and a polyacrylamide gel treatment. This figure was taken with permission from Oliveira-Hofman et al.9. Please click here to view a larger version of this figure.

This research has revealed that grower-oriented methods (with polyacrylamide gel and/or a polymer-coated gauze bilayer) can reliably produce high numbers of nematodes for pest control programs. Importantly, basic parameters of the processes mentioned above can be improved to fulfill grower needs (e.g., host density, arena sizes, and substrates used). The two approaches may be suitable for different cropping systems; a comparison and further optimization of the methods will be a major topic of future research.

When comparing the polyacrylamide gel to the White trap-based method, the gel substrate produces greater numbers of IJs. A critical element of these findings is that IJs can move directly into the gel, whereas in White traps they must move from inside the holding dish to the harvest dish, and, therefore, some IJs may not reach the harvest dish. There are advantages to harvesting IJs directly from gel, given that IJs tend to disperse into their surroundings, effectively concentrating themselves in the gel, facilitating storage and/or harvest¹⁵.

Similarly, IJs cultured in mealworms embedded in the gauze bilayer can also leave the cadavers, and by tracking gravity, will move down and gather on/in the wetted fibers of the gauze. The structure of the polymer-coated bilayer creates a high surface area on which nematodes can be concentrated. The ease of culturing and harvesting of the IJs makes this approach amenable to growers wishing to produce nematodes in-house. The finding that horse gauze was as productive as cotton wool is quite important because it indicates that the benefits of horse gauze (i.e., improved nematode harvest efficiency) can be sustained without reductions in productivity. It is likely that horse gauze provides greater aeration, compared to cotton wool, because the gauze fibers do not collapse into a dense, water-soaked mat. Importantly, the nematodes that emerge directly from an infected host (per our in vivo methods) will have demonstrated -- as a population -- their capacity to find, subdue, and kill insect hosts, thereby indicating a high degree of infectivity (invasion capacity), better dispersal capacity, higher virulence, and environmental (abiotic) tolerance, compared with nematodes produced via in vitro methods^21,22.

The grower-oriented production system proposed herein can be easily adapted to other EPN species, as the insect host, Galleria mellonella, is very susceptible to most Steinernema and Heterorhabditis species⁶. Other insect hosts may be used, such as T. molitor^6,7,18,19. The value of gel formulation was demonstrated by Leite et al. (2018)²³; if preserved under the right temperature in the gel system, they can be stored for several months before usage. The gel-based formulation is highly suitable for most application systems; gels with polyacrylamides have been successfully used as standard EPN formulations for decades^6,7,24. Moreover, gels have high levels of storage capacity as well as harvesting from gel via screening and rinsing is very easy²⁴, can take less than 5 min (Shapiro-Ilan personal observation) and higher yields are observed in the gel system approach relative to White traps. Gel formulations retain the IJs within the formulations (while other formulations (based on peat, vermiculite, potting mix diatomaceous and mushroom compost) had a higher percentage of IJs that escaped the medium²³. Further trials on nematode storage substrates that can completely dissolve in the water of a spray tank (e.g., phenolic sponge as utilized by Leite et al. (2023)²⁵ would be of value to growers.

Our grower-oriented approach has a potential drawback -- repeated sub-culturing can cause deterioration of the nematodes' virulence, which is a potential issue for all production systems. This issue can be avoided if growers use an initial inoculum from inbred lines²⁶ and then limit nematode densities during infection events, thereby limiting the incidence of "cheaters" (non-virulent nematodes that rely on other nematodes to kill the host) within nematode cultures²⁷.

Our novel grower-based production methods represent sustainable, accessible, high-yielding, high-longevity, and low-cost (yet labor-intensive) approaches to EPN production. The key to continuous production is the use of a reliable initial inoculum, based on infected hosts or stored gel. The process is thus responsive to continual use by the growers, which is an alternative to other methods in which the grower is continually dependent on an externally provided inoculum source. Having multiple methods for grower-oriented nematode propagation systems will allow growers to tailor the system to their own farming practices.

The impetus for growers to produce their own nematodes will be as varied as the growers themselves, but considering that nematodes produced via in vivo methods often exhibit greater searching capacity, fitness, and virulence^21,22, there is intrinsic value to in vivo rearing of nematodes. Conversely, EPNs produced via in vitro approaches do not encounter live insects during their development, which effectively denies the nematodes the opportunity to maintain high per-capita lethality (an important metric of their value as a biocontrol agent)^7,21,22. The cost-effectiveness of our methods should be considered relative to other in vivo techniques, as well as to the cost-effectiveness of commercial EPNs. Field efficacy and costs are important determinants of any biocontrol program.