Moth caterpillar embryos and parasitoid egg infection as revealed in vivo and visualized by micro-CT scanning

Abstract

The Lepidopteran pest Trichoplusia ni and the parasitoid wasp Trichogramma brassicae represent a fascinating biological system, important for sustainable agricultural practices but challenging to observe. We present a nondestructive method based on micro-CT scanning technology (CT: computed tomography) for visualizing the internal parts of caterpillar embryos and of emerging parasitoids from infected eggs. Traditional methods of microscopic observation of the opaque egg contents require staining or dissection. To explore the biological system nondestructively, we optimized the application of micro-CT scanning to construct 3-D images of insects in vivo.

Introduction

One fascinating tritrophic interaction is between plants, caterpillar pests, and beneficial parasitic wasps (Turlings and Fritzsche 2007). In response to herbivory, plants synthesize and release volatile bouquets that can be highly attractive to the natural enemies of the caterpillar pest, such as parasitic wasps. Female parasitic wasps, also known as parasitoids, lay their eggs in or on the pest host. As the wasp eggs hatch and develop, they eventually kill the pest insect. This biological control is used extensively as part of sustainable agricultural practices of integrated pest management.

Researchers traditionally use destructive means to study parasitoid development inside the Lepidopteran egg, as the opacity of the chorion (shell of the insect egg) does not allow direct visualization. Using a compound microscope, the chorion must be removed before imaging the host or parasitoid development. Even when imaging the “live” embryo, the chorion removal inevitably results in the death of the insect (Jarjees et al. 1998). The same is true for fluorescence and confocal microscopy techniques, with the added requirement of staining the tissues to be observed (Masci and Monteiro 2005, Zucker 2006, Peterson 2010). Thus, potential artifacts may arise from using these techniques and the development of a single caterpillar embryo or parasitoid cannot be followed temporally. In contrast, X-ray computed tomography (CT) scanning technology would allow for nondestructive visualization.

In CT scanning, X-rays are projected from a source through a subject and the unabsorbed X-rays are captured by detectors on the opposite side of the source (Kalender 2000). The absorption rate is a function of the initial intensity of the X-ray beam and the material density and thickness of the subject. Higher density components will absorb more X-rays and will appear in lighter tones of gray classically in the reconstructed images, compared to the less dense material in darker gray. The goal of this project was to develop a protocol for obtaining X-ray micro-CT scans of lepidopteran development and that of their parasitoid in the egg, allowing for nondestructive visualization. The use of X-ray CT scanning should allow the visualization of the internal structures of an insect egg in vivo.

The technique was optimized for the visualization of Trichoplusia ni eggs and first-instar caterpillars. The cabbage looper is a prolific pest of Brassicaceous crops, particularly canola, broccoli, and cabbage (Sarfraz et al. 2011, Farias et al. 2022). The eggs of this insect can be parasitized by wasps of the genus Trichogramma (Family Trichogrammitidae) (Razinger et al. 2016, Khan et al. 2020), such as T. brassicae. As egg parasitoids, females in nature lay their eggs within the eggs of host species, which kill the host embryo as the parasitoid develops before the host can mature and damage crops. Trichogramma are pro-ovigenic parasitoid wasps; the entire life cycle of the wasp, from egg to adult, occurs within the hemispherical host egg (~0.6 × 0.4 mm in size) and oogenesis is completed prior to emergence, as females lay most of their eggs shortly after emergence (Pak and Oatman 1982, Fleury and Bouletreau 1993, Smith 1996).

Methods

Rearing Trichoplusia ni

From eggs and first-instar larvae originally obtained from Insect Production Services (Sault-Ste-Marie, ON), a T. ni colony was established in growth chambers maintained at 27 ± 3°C, and 16:8 light:dark. Colony larvae were reared on modified McMorran diet as advised. Pupae were moved to a 10-gallon aquarium to emerge as adult moths and mate, and adults were provided a sugar water/honey mixture (Great Lakes Forestry Center 2015). Eggs were collected on 100% polyester mosquito netting, where the insect “egg glue” sticks the eggs to the mesh.

Trichogramma brassicae

Pupae of the parasitoid Trichogramma brassicae were obtained from Anatis Bioprotection (Saint-Jacques-le-Mineur, QC) on cards with ~4,000 individual pupae glued to the inside of a folded card. T. brassicae were kept in a growth chamber at 24 ± 3°C and a photoperiod of 16:8 (light:dark).

Trichoplusia ni Preparation

Eggs of T. ni were prepared for micro-CT scanning by cutting clusters of 10 or more individuals out of the netting. Eggs were kept in a Petri dish with diet at 27°C in an incubator in between scanning runs. Before the first scan and after the last scan, when caterpillars were seen in the Petri dishes, eggs were taken for observation under the microscope to identify healthy eggs that had hatched.

Trichogramma brassicae Preparation

To observe parasitoids inside of host eggs, a random selection of clusters of 10 or more fresh T. ni eggs (<12 h after oviposition) were cut from the netting and exposed to adult T. brassicae for ~12 h and then allowed to incubate under rearing conditions. T. brassicae developmental time in T. ni eggs varies from 1 to 2 wk in the literature; therefore, micro-CT scanning was done on days 7–10 each with 2 sessions per day (one in the morning and one in the afternoon). In between the CT scanning sessions, the mesh with eggs with parasitoids were kept in a sealed Petri dish in an incubator at 27°C. As with the caterpillar eggs, parasitized eggs were observed under the microscope before and after micro-CT scanning.

Micro-CT Scanning Protocol

Trichoplusia ni eggs either infected with the parasitoid or noninfected were micro-CT scanned using a SKYSCAN 1174 X-ray CT scanner (Bruker, Kontich, Belgium) at a resolution of 6.6 μm under the following specifications: Object to Source distance (mm) = 221, Camera to Source distance (mm) = 266.5, Source voltage (kV) = 32, and Source current (μA) = 566. Image rotation was set to 0.4 degree, with an exposure time of 1,600 ms and a frame averaging of 3 (i.e., 3 shots are taken per angle and the average is used to generate the projection image).

The source voltage (32 kV) and source current (566 μA) values were determined as follows. After the mesh with the eggs or the wasps (see below) was installed on the plate inside the micro-CT scanner and before the micro-CT scanning session actually started, a shot was taken and the micro-CT scanner system, via the control main program, produced and displayed a profile line in red color on the screen of the companion computer. The position of this profile line advises on the need, or not, to increase or decrease the energy requested from the Source through the voltage and current settings. Because of its material density, the mesh (on which the female moths laid eggs or on which the wasps that just emerged stayed; see below) absorbed much more energy than the insect materials, but increasing the energy excessively would lower the contrast within the eggs or the wasps.

Fresh T. ni eggs (<12 h after oviposition) were collected from the colony, where they were oviposited on 100% polyester mosquito netting. A net with eggs was installed at the center of the stage of the micro-CT scanner. Two CT scanning sessions a day (once in the morning and once in the afternoon) were conducted for 4 d, which covered the span of T. ni egg development (Capinera 2002). In between the CT scanning sessions, eggs on netting were sealed in a Petri dish and kept in an incubator at 27°C. The mesh was carefully handled to not displace any of the eggs from the mesh. After the final CT scanning session, eggs were observed under a light microscope for signs of hatching (eclosion).

Micro-CT images were rendered in 3-D space with program CTVox, and the internal structures were investigated using a heatmap of estimated density values. From lowest to highest, the estimated density values were distributed in 5 classes. Accordingly, the voxels (3-D extension of pixels) were colored in black, blue, green, yellow, and red, instead of gray tones.

Results and Discussion

Trichoplusia ni eggs scanned about 12 h after oviposition show a small area of higher density material condensed in one area (Fig. 1a). At this time (day 1-time 1), the small red portion surrounded by yellow, visible at the top of the egg hemisphere in the cross-section, corresponds to the thickening of the blastoderm and the formation of the germ band in Lepidopteran embryogenesis (Chun-Li and Lu 2006). By day 2-time 1 (~36 h after oviposition), the area of green density values spreads out from the high-density point of day 1-time 1 (Fig. 1b). Finally, by day 3-time 1 (~60 h after oviposition), the high-density region has both increased and become more condensed at the top of the egg hemisphere (Fig. 1c, d). At this point, the caterpillar is in the final stages of development, the dorsal closure of the midgut is complete, and the body is positioned ventral concave (Chun-Li and Lu 2006).

Fig. 1.

Cross-sections of a Trichoplusia ni egg micro-CT scanned at (a) 12 h (day 1-time 1), (b) 36 h (day 2-time 1), (c) 60 h (day 3-time 1), and (d) 84 h (day 4-time 1) after oviposition. The density heatmaps show the development of the caterpillar embryo from the blastoderm at 12 h (red+yellow) in (a), with the complete formation of the organ tissue (green) at 60 h with the body positioned ventral concave in (c) and by 84 h (d) the caterpillar is nearing eclosion.

A first-instar T. ni caterpillar was captured mid-hatch in the last of one micro-CT scanning session (day 3-time 2) (Fig. 2; see also Supplementary Animation 1). The red points at the most dorsal position of the body may represent the dorsal artery and the points of highest density in the hindgut region may be the Malpighian ampulla (Fig. 2, arrows a). The green-highlighted structure that makes up most of the body (Fig. 2, arrows b) corresponds to the digestive tract of the caterpillar, which is surrounded by lower density air filled trachea which feed oxygen to the digestive system (Sutton 2010, Wagner and Hoyt 2022). In the head of the caterpillar (Fig. 2, arrows c), clusters of green-yellow-red density materials highlight the brain of the insect at the back of the head and the sub-esophageal ganglia near the mouthparts.

Fig. 2.

Trichoplusia ni 1st instar caterpillar emerging from the egg during micro-CT scanning on day 3-time 2. Fixed central axial cross-section showing: a) Dorsal artery and malpighian ampulla (red+yellow) and hindgut (green), b) convoluted midgut (zigzag green interspersed with blue), and c) brain and suboesophageal ganglion (red+yellow).

The use of micro-CT imaging to capture the caterpillar anatomy demonstrates the power of this nondestructive technology. The insect’s brain and the sub-esophageal ganglion are well defined in these images and appear to be surrounded by material of lower density in the head (Fig. 2, arrows c). Contrary to the general belief that the caterpillar gut is unconvoluted (Wagner and Hoyt 2022), the midgut of the newly hatched first-instar caterpillar here is shown to have pronounced folds, visible as the zigzag green density values in the micro-CT images (Fig. 2, arrows a). This could be hinting at a more complex internal anatomy of the midgut than is widely known, as dissection of a caterpillar’s hydrostatic system causes the loss of its turgidity (Lin et al. 2011).

To observe T. ni egg parasitism by T. brassicae, fresh T. ni eggs (<12 h after oviposition) oviposited on netting, were exposed to T. brassicae adults for ~12 h and then incubated under rearing conditions. As T. brassicae developmental time in T. ni eggs varies from 1 to 2 wk (Ozder and Saglam 2004), micro-CT scanning was performed 7–10 d after parasitism 2 sessions per day (one in the morning and one in the afternoon). A newly emerged T. brassicae wasp was thus captured (Fig. 3). Like in the T. ni first-instar caterpillar (Fig. 2), the midgut (green) is the most prevalent organ shown in the T. brassicae wasp (Fig. 3, arrows a). High-density material (red) highlights the wasp abdomen, around or within the gut. This may corroborate the finding (Jarjees and Merritt 2002) that concentrated waste material is stored in crystalline urate form within cells that line the midgut epithelium near the hemocoelic side in Trichogramma australicum larvae. This waste material is then released as uric acid into the gut of the adult wasp through the malpighian tubules. A newly emerged wasp is likely to still possess a high concentration of waste material that has yet to be excreted, and uric acid would show up as higher density (1.87 g/ml) compared to surrounding cells which are mostly water (1 g/ml). The thoracic ganglion and ventral nerve chord (Fig. 3, arrows b) also show up as high density in micro-CT images, compared to surrounding tissue.

Fig. 3.

Parasitoid wasp, Trichogramma brassicae, newly emerged from a Trichoplusia ni egg during a micro-CT scanning session. a) Wasp midgut (peripheral green) shows large portions of high-density material (red+yellow) inside. This is evidence of the presence of concentrated host material within the gut of the insect, as host material is stored throughout larval development as urate crystals and released into the gut as uric acid in the early adult stage (Jarjees and Merritt 2002). b) Other structures that show up as high-density (red+yellow) are the thoracic ganglion and ventral nerve chord.

Within the host egg, it was sometimes observed that more than one wasp occupied a single egg, apparent by the 2 abdomens that show up in scans of parasitized eggs at late stages of development, giving a coffee bean appearance to the eggs in the micro-CT images (see Supplementary Animation 2). Superparasitism, when an individual parasitoid or a conspecific oviposits within an already parasitized host, is known to occur in several Trichogramma species (Suzuki et al. 1984, van Dijken and Waage 1987). However, to our knowledge, it has not been reported for T. brassicae in T. ni eggs. Superparasitism often has negative consequences for the parasitoid development due to limited nutritional and spatial resources. However, it has been proposed that under stress conditions, superparasitism may be an adaptive strategy (van Alphen and Visser 1990). Here, we present images of T. brassicae superparasitism with 2 adult wasps in one T. ni egg.

On a technological or more technical note, we used a resolution (voxel side length) of 6.6 μm, which was the smallest possible with the micro-CT scanner available to us. There exist micro-CT scanners offering a spatial resolution of 5 μm or less; the smaller the better as long as the scale of observation is large enough to include the whole volume of interest. There is also the concept of CT number (CTN), which is derived from the X-ray attenuation coefficient for the CT scanned material relative to air, water, and skull (Kalender 2000). After due calibration of a CT scanner, CTN_air is expected to be equal to −1,000 HU, CTN_water to 0 HU, and CTN_skull to +1,000 HU. We used CT numbers indirectly, in the coloring of different parts of an egg or a wasp depending on the corresponding estimated material density (Figs. 1–3). Last but not least, mosquito netting was the most appropriate method that we found for egg yield. We tested a number of other materials to collect moth eggs, including paper and cardboard (with different material densities and thus, different X-ray attenuation coefficients than polyester or nylon), but the female moths would only lay their eggs on mesh and Insect Production Services (where we originally received our colony from) uses a white mesh to collect eggs for their caterpillar insect colonies.

In closing, our application of X-ray micro-CT scanning technology provides 3-D image data for a parasitoid–host system nondestructively. We have shown the ability to monitor caterpillar embryogenesis and observe internal structures of both lepidopteran caterpillars and egg parasitic wasps. As well, the method provides an opportunity of observing superparasitism in vivo. Two videos can be downloaded from the Supplementary Materials and watched as complements to Figs 2 and 3; 8 microscopy images are also provided for comparison purposes.

Funding

The Authors are grateful to Centre SÈVE (FRQNT Strategic Clusters grant #264708) for funding the micro-CT scanning work (equipment operation and image analysis) and the Canada Foundation for Innovation (CFI) for the micro-CT scanning equipment (grant #36368).

Author contributions

Ryan Smith (Conceptualization [equal], Formal analysis [equal], Investigation [equal], Visualization [equal], Writing—original draft [equal], Writing—review & editing [equal]), Liwen Han (Formal analysis [equal], Investigation [equal], Methodology [equal], Software [equal], Visualization [equal]), Jacquie Bede (Conceptualization [equal], Funding acquisition [equal], Investigation [equal], Methodology [equal], Project administration [equal], Resources [equal], Supervision [equal], Writing—original draft [equal], Writing—review & editing [equal]), and Pierre Dutilleul (Conceptualization [equal], Funding acquisition [equal], Investigation [equal], Methodology [equal], Project administration [equal], Resources [equal], Supervision [equal], Validation [equal], Writing—original draft [equal], Writing—review & editing [equal])