Downloaded from learnmem.cshlp.org on September 28, 2021 - Published by Cold Spring Harbor Laboratory Press Evolution, Discovery, and Interpretations of Arthropod Mushroom Bodies Nicholas J. Strausfeld,1,2,5 Lars Hansen,1 Yongsheng Li,1 Robert S. Gomez,1 and Kei Ito3,4 1Arizona Research Laboratories Division of Neurobiology University of Arizona Tucson, Arizona 85721 USA 2Department of Ecology and Evolutionary Biology University of Arizona Tucson, Arizona 85721 USA 3Yamamoto Behavior Genes Project ERATO (Exploratory Research for Advanced Technology) Japan Science and Technology Corporation (JST) at Mitsubishi Kasei Institute of Life Sciences 194 Machida-shi, Tokyo, Japan Abstract insect orders is an acquired character. An overview of the history of research on the Mushroom bodies are prominent mushroom bodies, as well as comparative neuropils found in annelids and in all and evolutionary considerations, provides a arthropod groups except crustaceans. First conceptual framework for discussing the explicitly identified in 1850, the mushroom roles of these neuropils. bodies differ in size and complexity between taxa, as well as between different castes of a Introduction single species of social insect. These Mushroom bodies are lobed neuropils that differences led some early biologists to comprise long and approximately parallel axons suggest that the mushroom bodies endow an originating from clusters of minute basophilic cells arthropod with intelligence or the ability to located dorsally in the most anterior neuromere of execute voluntary actions, as opposed to the central nervous system. Structures with these innate behaviors. Recent physiological morphological properties are found in many ma- studies and mutant analyses have led to rine annelids (e.g., scale worms, sabellid worms, divergent interpretations. One interpretation nereid worms) and almost all the arthropod is that the mushroom bodies conditionally groups, except crustaceans. The most primitive lo- relay to higher protocerebral centers bopods, the Onychophora (velvet worms), possess information about sensory stimuli and the these structures as well as the most advanced so- context in which they occur. Another cial insects. interpretation is that they play a central role Like all other parts of an organism, mushroom in learning and memory. Anatomical studies bodies are the products of evolution. Their struc- suggest that arthropod mushroom bodies ture and functions reflect their evolutionary history are predominately associated with olfactory and the specific sensory and behavioral adapta- pathways except in phylogenetically basal tions that characterize a particular taxon. Present insects. The prominent olfactory input to understanding and interpretations of mushroom the mushroom body calyces in more recent body function also reflect the evolution of research on insect brain and behavior, which has resulted in disparate views of mushroom body function. One 4Present address: National Institute for Basic Biology, Myodaiji-cho, Japan. view holds that mushroom bodies are the site of 5Corresponding author. olfactory learning and memory. Another view is LEARNING & MEMORY 5:11–37 © 1998 by Cold Spring Harbor Laboratory Press ISSN1072-0502/98 $5.00 LEARNING& MEMORY 11 Downloaded from learnmem.cshlp.org on September 28, 2021 - Published by Cold Spring Harbor Laboratory Press Strausfeld et al. that they support a variety of other functions, not Flo¨gel (1876) was the first to define criteria for all necessarily represented in the same species. identifying mushroom bodies across insect species: Such functions include sensory discrimination and the presence in the supraoesophageal mass of integration with other modalities, the control of paired groups of several hundred to several hun- complex behavioral repertoires, and spatial orien- dred thousand minute cells (termed globuli cells tation. but now known as Kenyon cells; Strausfeld 1976) that surmount lobed neuropils that Flo¨gel pro- posed were composed of parallel fibers. Kenyon EARLY HISTORY OF MUSHROOM BODY RESEARCH (1896a,b), who was the first to use Golgi methods Mushroom bodies were discovered in 1850 by on insect brains, confirmed Flo¨gel’s ideas about the the French biologist Fe´lix Dujardin, who called fibrous nature of globuli cell morphology. Kenyon these structures corps pe´doncule´s, likening their showed that the dendritic branches of globuli cells appearance to the fruiting bodies of lichens. Many invade the head of the mushroom body to form a recent accounts have co-opted Dujardin’s paper in structure called the calyx. He described the parallel support of the widely held belief that mushroom axons of globuli cells forming a pedunculus that bodies are learning and memory centers. However, extends to the front of the brain where axons then Dujardin did not suggest this, but proposed that branch to provide a vertical and a medial lobe these centers endowed an insect with a degree of (Kenyon 1896a,b). Kenyon identified afferents to free will or intelligent control over instinctive ac- the calyces and suggested that these carried olfac- tions. He supported this idea from comparative tory, visual, and tactile information. He proposed studies of the brains of ichneumons, solitary bees, that mushroom bodies provided a center for sen- and honeybees, showing that advancing sociality sory-motor integration, quite separate from direct was correlated with the possession of enlarged sensory-motor relays that characterize other brain mushroom bodies (Dujardin 1850). He observed areas or thoracic or abdominal ganglia (Kenyon that insects with small mushroom bodies showed 1896a). Kenyon did not suggest that mushroom greater coordination of thoracic motor actions af- bodies are involved in learning and memory. ter decapitation than did insects with large mush- Golgi studies on the mushroom bodies of Blat- room bodies and proposed that the smaller the todea (cockroaches: Sanchez 1933), Hemiptera mushroom body, the more automatic or instinctive (true bugs: Pflugfelder 1937), and Hymenoptera that insect’s behavior. Dujardin also carried out ex- (ants, wasps, bees: Goll 1967) all confirmed periments on homing abilities in ants (Dujardin Kenyon’s (1896a,b) findings, as have descriptions 1853) to underpin his ideas of insect intelligence. of these neuropils in the cricket Acheta domesti- Two other French biologists, Faivre (1857) and Bi- cus (Schu¨rmann 1973, 1974), the sphingid moths net (1894), furthered Dujardin’s ideas, performing Sphinx ligustri (Pearson 1971) and Manduca sophisticated ablation experiments to demonstrate sexta (Homberg et al. 1989), the house fly Musca that although the suboesophageal ganglion is nec- domestica (Strausfeld 1976), and the honeybee essary for maintaining synchronized movements of Apis mellifera (Mobbs 1982, 1984). the limbs (Faivre showed that, if fed, a dytiscid beetle can survive for months without its supra- FIRST EVOLUTIONARY STUDIES esophageal ganglion), it was insufficient for provid- ing complex and varying patterns of motor activity. Even before the advent in the early 1900s of Faivre, in particular, demonstrated these complex methods that selectively reveal neural architecture movements to be under the control of the supra- (Cajal and de Castro 1933), early anatomists pro- esophageal ganglion (the brain proper). Several vided reasonably accurate descriptions of the other 19th century studies supported the idea that mushroom bodies. Viallanes (1893), for example, mushroom bodies of insects mediate intelligent was the first to recognize the enormous number of versus innate behavior on the basis of comparative mushroom body globuli cells in the horseshoe crab anatomy, with particular emphasis on the social Limulus, a feature that led the Swedish neurologist Hymenoptera and the differences between the Holmgren (1916) to wonder why this animal re- brains of different castes (Leydig 1864; Forel 1874; quired such a huge center when, in his view, it so Flo¨gel 1876, 1878). Like Dujardin’s, none of these obviously lacked behavioral sophistication. Vi- studies explicitly suggested that the mushroom allanes (1887a,b) also identified mushroom bodies bodies underlie learning and memory. in dragonflies, wasps, and crickets, thereby provid- LEARNING& MEMORY 12 Downloaded from learnmem.cshlp.org on September 28, 2021 - Published by Cold Spring Harbor Laboratory Press MUSHROOM BODY EVOLUTION ing early comparative descriptions of this neuropil bodies in termites (Holmgren 1909) reinforced from sectioned material. Bretschneider (1913, Holmgren’s opposition to the idea that mushroom 1914, 1918, 1924) published a series of studies on bodies endowed an arthropod with intelligence. insect brains, including that of the cockroach Peri- He found that cockroaches, near relatives of ter- planeta orientalis (Bretschneider 1914) in which mites, had mushroom bodies almost as advanced he identified the characteristic striations in the (in cell number and gross morphology) as those of mushroom body lobes, now known to be caused Hymenoptera. But not a keen observer of animal by the layered arrangements of Kenyon cell axons behavior, he wrote that ‘‘the psychic ability of the (Li and Strausfeld 1997; Mizunami et al. 1997; roach is hardly worth comparing with that of the Strausfeld 1998a,c). Like Flo¨gel (1876), Bret- termite.’’ schneider (1918) also attempted to use brain fea- The second serious attempt to reconstruct tures for inferring insect relationships. However, arthropod phylogeny from brain anatomy was the