Buthus, Leach, 1815

Fet, Victor, Gantenbein, Benjamin, Gromov, Alexander V., Lowe, Graeme & Lourenço, Wilson R., 2003, The first molecular phylogeny of Buthidae (Scorpiones), Euscorpius 4 (4), pp. 1-10 : 4-7

Buthus , Androctonus , Leiurus , Vachoniolus and Apistobuthus

This group, only weakly supported in our analysis, includes widespread Buthus , Androctonus , and Leiurus , all very typical for the western Palearctic deserts (North Africa, Middle East) ( Vachon, 1952; Levy & Amitai, 1980; Sissom, 1994), and two psammophile genera from the Arabian Peninsula, Vachoniolus and Apistobuthus (two species each) It is worth noting that this loose group includes the most toxic of Old World scorpion genera, Androctonus and Leiurus . This result suggests that Androctonus and Leiurus could be more closely related to each other than to any other genus represented in our study, though their coupling is weakly supported in this analysis. This is consistent with the data on their venom (Loret & Hammock, 2001; see also below). The weak support for Vachoniolus and Apistobuthus could indicate their common origin as psammophile species in the Arabian deserts; Apistobuthus possesses some highly derived features (“flared” metasomal segments; Vachon, 1960). Placement of Vachoniolus within this clade, apart from Buthacus , was surprising because these two genera are closely similar in their external morphology, and indeed female Vachoniolus was originally confused with Buthacus (Levy et. al., 1973). Both genera express relatively abbreviated pedipalps and the loss of carapacial and mesosomal carination. The data here implies these similarities may be a result of convergence, possibly associated with psammophilic lifestyles. However, the grouping of Buthacus yotvatensis (the example sequenced here) with Hottentotta jayakari is also correlated with distinctive characters separating this species from other members of Buthacus , i.e., large size and development of dense pilosity on the metasomal carinae, a feature typical of the “hirsute” group of Hottentotta sp. , as exemplified by H. jayakari . The divergence of other diagnostic characters between Hottentotta and Buthacus , related to tarsal setation, are likely due to ecological radiation into different substrata (i.e., rock and gravel vs. sand). Molecular data from the smaller species of Buthacus may yet reveal a relationship to Vachoniolus .

Other genera

Outgroups used in our analysis originate from the New World (genera Centruroides and Rhopalurus ), Madagascar (endemic genus Grosphus ) and Southeast Asia ( Lychas ). Not surprisingly, we see the expected, strongly supported close relationship of the genus Centruroides (North and Central America, Caribbean, northern South America) and Rhopalurus (South America and Caribbean). Absence from our analysis of additional genera from tropical Africa and Asia did not allow us to achieve phylogenetic resolution outside of the “13 genera clade”; it is most important, however, that both tropical Old World genera ( Grosphus and Lychas ) group outside of the “13 genera clade”, although their common clade with New World species is not supported. A possible evolutionary scenario for Buthidae could include separate evolution in Laurasia and Gondwana, after the split of Pangaea, as it was for many other groups of organisms. In this case the “13 genera clade” could represent lineages of northern (Laurasian) evolution. If support is obtained in the future for the clade of modern Neotropical and Afrotropical scorpions (represented by Centruroides , Rhopalurus , and Grosphus in our analysis), it would strongly suggest a typical Gondwanan “signature”, with New World buthids originating from South America as tropical forms. The latter hypothesis would correlate with patterns of modern diversity of arid scorpions in the New World, where scorpion radiation in the deserts of both Americas is largely observed in Bothriuridae , Vaejovidae and Iuridae . The few Buthidae which inhabit deserts of North America (species of Centruroides ) could have dispersed to the arid landscapes from the New World tropics. It is worth noting that one of the major characters subdividing Buthidae at this moment, so-called alpha versus beta orientation of dorsal trichobothrial on pedipalp femur ( Vachon, 1975; Sissom, 1990) also is consistent with our phylogeny, the “13 genera clade” all being beta pattern.

Independent origin of psammophily

An important issue in scorpion evolution, discussed in detail by Fet et al. (1998) is parallelism in the highly adaptive features allowing psammophily (life in sand). For the first time, within the “13 genera clade”, we document an independent origin of buthid psammophily in at least four clades: (a) Liobuthus (Central Asia); (b) Buthacus (North Africa, Middle East); (c) Anomalobuthus (Central Asia); and (d) Vachoniolus and Apistobuthus (Middle East) (the latter two could have shared a psammophilic ancestor). Especially impressive is strongly supported independent evolution of psammophily in Liobuthus and Anomalobuthus , two virtually sympatric monotypic genera inhabiting the vast deserts of Central Asia from the Caspian Sea in Turkmenistan to Syr-Darya River basin ( Fet, 1989, 1994). A further study of molecular phylogeny of these genera is under work (Fet et al., in progress).

Origin of mammal-specific toxins

Another important issue in scorpion evolution is, of course, evolution of toxins. Our phylogeny supports a common origin of Old World lineages of Buthidae with the most potent neurotoxic venom: the genera Androctonus and Leiurus . These and other Palearctic species in which toxins have been studied ( Buthus , Mesobuthus , Hottentotta , Orthochirus ) are all present in our “13 genera clade” All these forms share separate mammal- and insect-specific neurotoxins specific for Na + channels (Loret & Hammock, 2001). At the same time, New World genera have potent toxins ( Centruroides , Tityus ) acting on both mammals and insects. We suggest that the origin of this feature could lie in the Laurasia versus Gondwana split as discussed above. Thus, the separate mammal-specific Na + toxins could have evolved during aridification of the Palearctic in the Tertiary period, when one of the most important selective factors was rapid radiation of small burrowing mammals (mostly rodents) in arid landscapes. Naturally, such newcomers to the scorpion environment as rodents would be a direct competitor for space (burrows) and in addition important nocturnal predators, as many of them are today (McCormick & Polis, 1990). Such pressure explains emergence of specific mammal-targeting toxins (used for defense, not for foraging) in predominantly burrowliving Palearctic buthids – as opposed to largely vegetation-inhabiting New World buthids.

Loret et al. (1991) discovered a toxin (AahIT4) with “ancestral” properties in a North African population of Androctonus australis . It is able to bind to both alpha and beta sites in both vertebrates and insects, as opposed to other, more specialized toxins. This led to a biogeographic hypothesis that buthid toxins acting on mammals “originated from North Africa” ( Loret et al., 1991; Loret & Hammock, 2001). However, the currently accepted A. australis ranges from Mauritania to India (Fet & Lowe, 2000: 69), and the center of origin of this species is unknown; further, none of the congeneric Androctonus species or non-African Old World buthids were tested for the existence of the AahIT4 or related toxins. The constant reference to A. australis as the “Sahara scorpion” in toxicological literature leads to unjustified and superficial conclusions about evolution of both species and toxins, further confusing this complicated issue. According to our analysis, Androctonus does not have a basal position in buthid phylogeny, thus “ancestral” properties of Androctonus toxin should be a plesiomorphic character. Only an independently assessed evolutionary history of Buthidae will allow us to “plot” toxin features on the supported phylogeny. In fact, some studies demonstrated that toxicity to both mammals and insects, as opposed to either mammal- or insect-specificity, could depend on minor amino acid changes ( Kopeyan et al., 1993), thus a long evolutionary history is not required for explanation of toxin evolution.

Further Directions of Research

Although our preliminary study covered only a selection of all existing buthid genera, we detected several important evolutionary trends in the radiation of Palearctic buthids, which include some of the most toxic scorpion species (genera Androctonus and Leiurus ). In future, a combined morphological and molecular (DNA) phylogenetic analysis including more taxa will further clarify relationships within Buthidae . The major issues remaining to be addressed are the relationship between Palearctic, Afrotropical, and Oriental buthids in the Old World, and the origin of New World buthids (according to our hypothesis, from Gondwana). Many speciose buthid genera, first of all Tityus and Centruroides in the New World, have to be subjected to phylogenetic analysis, and could be eventually split into several genera. A large number of branches should be added to our preliminary tree in the future, many from exotic taxa in remote and difficult localities and hardly accessible. We are convinced that only the collaborative efforts of many scorpion taxonomists from many parts of the world will bring sufficient knowledge on evolution of Buthidae .