From two to three dimensions: The importance of the third dimension for evaluating the limits to neuronal miniaturization in insects
Although various parameters can accommodate the scaling of body size that accompanies miniaturization (e.g., musculature, circulatory system, and fat body: Polilov & Makarova, 2017), others are less flexible. This holds true especially for the nervous system. A relative increase in brain size with respect to body size has long been reported for small invertebrates (Rensch, 1948; Beutel, Pohl, & Hünefeld, 2005; Eberhard & Wcislo, 2011; Quesada et al., 2011; Seid, Castillo, & Wcislo, 2011; Polilov & Makarova, 2017), suggesting that there may be a limit to minimal brain size, one that is probably imposed by the smallest size of neuron cell bodies (Beutel et al., 2005), and the limits these impose in turn on the number of neurons that contribute neurites to a neuropil of given volume. Soma size is indeed reduced, among identified amine‐expressing neurons in isogenic sister Trichogramma wasps, and accompanying the 5‐fold reduction that can occur in some brain volumes (van der Woude & Smid, 2017). A reduction in cell size rather than a loss of entire neurons is however in this case an example of phenotypic plasticity, and not true miniaturization. In general, small cell size tests the functional limits of organs and, at a cellular level, the size of a neuron limits its ability to accommodate organelles and sustain neuron function. For the latter, in particular, miniaturized nervous systems face two major challenges for information processing: signal noise and energy consumption (Niven & Farris, 2012; Niven, 2016). Distinct behavioral limitations have hitherto never been ascribed to miniaturized species (Eberhard, 2007), although in the specific case of sensory organs adaptations are reported for the compound eyes of miniature Hymenoptera that prevent the loss of function by maintaining sensitivity and a specific resolution for tiny photoreceptors (Fischer, Müller, & Meyer‐Rochow, 2011; Makarova, Polilov, & Fischer, 2015).
Functional limits for the diameters of neurites have previously been discussed (Llinás, 2003; Niven, 2016) and modeled (Faisal, White, & Laughlin, 2005; Sengupta, B., Faisal, A. A., Laughlin, S. B., & Niven, J. E. (2013); Neishabouri & Faisal, 2014), that limit the reduction of axon diameters in miniature species to about 0.1 µm. For neurites of smaller diameter the increasing noise from spontaneously opening sodium channels is predicted to generate spontaneous action potentials. Even so, smaller axon diameters down to 45 nm have been reported for tiny Hymenoptera (Hustert, 2012), although these are reported never to contain mitochondria, presumably because their lumen is too small to allow the passage of these organelles. The tiny neurons of small brains with short conduction distances also more frequently use graded potentials (Roberts & Bush, 1981).
Dimensional constraints have also been described for the cell bodies of neurons, with a proposed minimum diameter of 2 µm (Beutel et al., 2005) that is limited by the dimensions of the nucleus.