Exploring miniature insect brains using micro-CT scanning techniques

Abstract

The capacity to explore soft tissue structures in detail is important in understanding animal physiology and how this determines features such as movement, behaviour and the impact of trauma on regular function. Here we use advances in micro-computed tomography (micro-CT) technology to explore the brain of an important insect pollinator and model organism, the bumblebee (Bombus terrestris). Here we present a method for accurate imaging and exploration of insect brains that keeps brain tissue free from trauma and in its natural stereo-geometry, and showcase our 3D reconstructions and analyses of 19 individual brains at high resolution. Development of this protocol allows relatively rapid and cost effective brain reconstructions, making it an accessible methodology to the wider scientific community. The protocol describes the necessary steps for sample preparation, tissue staining, micro-CT scanning and 3D reconstruction, followed by a method for image analysis using the freeware SPIERS. These image analysis methods describe how to virtually extract key composite structures from the insect brain, and we demonstrate the application and precision of this method by calculating structural volumes and investigating the allometric relationships between bumblebee brain structures.

X‐ray computed tomography and its potential in ecological research: A review of studies and optimization of specimen preparation

“…This technique allows spatial resolution in the 1–10 μm range (i.e., spanning the range from whole cells down to the level of single organelles) and a temporal resolution of less than 100 ms. Furthermore, due to recent improvements, it provides enough detail to successfully distinguish either cuticular structures or soft tissues as muscles and nervous system (Sena et al., 2015; Smith et al., 2016; Westneat et al., 2008). …”

“…Images obtained from X‐ray CT possess a homogeneous illumination with isotropic resolution at each slice, which allows consistent and precise volumetric estimates and some degree of automation of the process (Sickert et al., 2015; Smith et al., 2016). As reported by Seo, Lim, Seo, and Lee (2015), X‐ray CT is sensitive enough for internal modifications that in some cases cannot be traced through resin‐sectioned images.…”

“…Metscher (2009a) suggested that stains are usually not tissue‐specific and the final quality depended mainly on the fixative. However, comparisons between different stains have shown that indeed some compounds can stain lipidic tissue more intensely (e.g., Lugol's solution) (Degenhardt et al., 2010); thus, recognition of different tissues can be greatly influenced by the chosen stain (Smith et al., 2016). …”

“…PTA has shown to yield sufficient results to recognize of brain and muscle structures in arthropods (Smith et al., 2016; Swart, Wicklein, Sykes, Ahmed, & Krapp, 2016). However, because of its slow penetration rate—as evidenced by Smith et al.…”

“…(2016) and also this study—it might be considered when the sample tissue is thin or removal of parts of the exoskeleton is possible (to facilitate stain perfusion). While osmium tetroxide (OsO 4 ) has been a popular staining agent in previous X‐ray CT studies (Jahn et al., 2018; Kim et al., 2012; Metscher, 2009a,b; Ribi, Senden, Sakellariou, Limaye, & Zhang, 2008), we did not consider it because of its undesirable toxicity, high costs, limited tissue penetration, and failure to work properly in tissues preserved in alcohol (Hayat, 1970; Metscher, 2009a; Smith, Carson, & Ferguson, 1974; Smith et al., 2016). …”