Neuroanatomy of sea spiders implies an appendicular origin of the protocerebral segment

Independent specialization of arthropod body segments has led to more than a century of debate on the homology of morphologically diverse segments(1,2), each defined by a lateral appendage and a ganglion of the central nervous system. The plesiomorphic composition of the arthropod head remains enigmatic because variation in segments and corresponding appendages is extreme. Within extant arthropod classes (Chelicerata, Myriapoda, Crustacea and Hexapoda-including the insects), correspondences between the appendage-bearing second (deutocerebral) and third (tritocerebral) cephalic neuromeres have been recently resolved on the basis of immunohistochemistry(1) and Hox gene expression patterns(3,4). However, no appendage targets the first ganglion, the protocerebrum, and the corresponding segmental identity of this anterior region remains unclear(5). Reconstructions of stem-group arthropods indicate that the anteriormost region originally might have borne an ocular apparatus and a frontal appendage innervated by the protocerebrum(6). However, no study of the central nervous system in extant arthropods has been able to corroborate this idea directly, although recent analyses of cephalic gene expression patterns in insects suggest a segmental status for the protocerebral region(7-10). Here we investigate the developmental neuroanatomy of a putative basal arthropod(11), the pycnogonid sea spider, with immunohistochemical techniques. We show that the first pair of appendages, the chelifores, are innervated at an anterior position on the protocerebrum. This is the first true appendage shown to be innervated by the protocerebrum, and thus pycnogonid chelifores are not positionally homologous to appendages of extant arthropods but might, in fact, be homologous to the 'great appendages' of certain Cambrian stem-group arthropods.