850 Nature Vol. 276 21/28 December 1978
Myriapoda-Hexapoda assemblage' and to the Uniramia or against a polyphyletic Second, an increase in amount (and in 7 give a name, the Uniramia, to this group • concept of arthropodan evolution. radioactive labelling) of two polypeptide The component taxa are united on their S. M. MANTON bands in polyacrylamide gels was found in 4 basic arthropodan features; their parti Zoology Department, CEV-infected plants • MWs assigned to cular type of jaws, and their limbs lacking British Museum (Natural History), the two polypeptides were 18,000 and any trace of true outer ramus. The jointed London SW7, UK 15,000. I have recalculated these MWs as limbs of myriapods and hexapods could, 13,200 and 10,400, using as internal arguably, have been derived from lobo I. Locke, M. & Huie. P. Narure 270,341- 343 (1977). markers the large and small subunits podia! limbs essentially resembling those 2. Manton. S.M. J. Linn. Soc .• Zoo/. 56,58-72 (1958). of ribulose-1 ',5' -diphosphatecarboxylase 3. Manton, S.M. 1. nat. Hisr. 1, 1-22 (1967). of the extant Onychophora, and these 4. Manton. S.M. J. Linn. Soc .• Zoo/. 53,257- 375 (1973). (MWs 54,000 and 12,000). The larger of limbs could have been derived from limbs 5. Manton, S.M. J. zoot. R-..171, 111-130 (1974). the CEV -induced polypeptides was of metamerically segmented worms with a 6. Manton , S. M. The Arthropoda : Habits, Functional located preferentially in the histone frac Morphology und Evolurion 1- 527 (Odord University haemocoel which, along with the muscles, Press 1977). tion, the smaller was in the non-histone 4 worked the uniramous limbs. As long ago 7. Manton. S.M. J. Linn. Soc .. Zoot. 51,203-400 (1972). fraction . This partitioning fits with the 8 9 8. Ticg>. 0 . W. Q. 11. microsc. Sci. 82, 1-225 (1940); 85, as 1940, Tiegs · emphasised the unity of 191-328 (1945). amino acid composition of the two poly these three taxa and Anderson 10 has 9. Tiegs. 0. W. 0. 11. Microsc. Sci. 88, 165-336 (1947). peptide components that could be derived 10. Anderson. D. T. The Embryology und Phylogeny of from the negative strand. shown in detail the common ontogenetic Annelids and ArthFopods 1-495 (Pergamon, Oxford, theme which exists throughout the Uni 1973) ramia and contrasts absolutely with that II. Robson. E. A. Q. 11. microsc. Sci. 105,281-299 (1964). 120-127 12. Hackman. R. H. & Goldberg, M. Science 190, 582-583 /AAAGGGGG pervading the entire crustacean group. (1975). Further, crustecean ontogeny could not 13. Manton. S.M. Phil. Trans. R. Soc. B. 2Z7, 411-464. have been derived from that of any known annelidan group. In contrast, the uni Are viroids ramian ontogeny might have been derived negative-strand viruses? from that of some yolky-egged annelid or annelid-like worm which possessed a IN their paper detailing the complete haemocoel and lobopodia, but not sequence of 359 bases in potato spindle coelomate parapodia. tuber viroid (PSTV) RNA, Gross et a/. 1 2 There is no justification on the basis of considered, as have other workers , that Golgi phenomena for the restriction of the the circular single-stranded molecule is term Uniramia to the Myriapoda and unlikely to function as a messenger RNA. Hexapoda alone, nor for the exclusion The possibility that PSTV RNA might be from the arthropods of any animal on one a non-coding strand rather than a positive character alone. That the Onychophora (coding) strand may have been over are thoroughly arthropodan is shown by looked. Fig. 1 Model of a PSTV RNA strand 11 12 their cuticle · , which contains protein I have examined the base sequence complementary to the viroid strand 1 and chitin, but not collagen, in contrast to complementary to that established by sequenced by Gross el a/. • The heavy line 1 represents the circular RNA. The annelids; by the presence of an ecdysial Grosset al. • Four possible initiation trip 1 numbers 1 and 359 indicate the same start cycle as in other arthropods \ the cuticle lets (all GUG) and six possible strong being shed in one piece, as in no other termination triplets could result in four position as that used by Gross eta/., but because the strand represented here is invertebrates on Locke and Huie's list\ polypeptides containing 108, 79, 43 or 28 antiparallel to the viroid strand, the posi and by the general anatomy of the Ony amino acids, of which the two larger (with tion of base 1 in this diagram is equivalent chophora, which tallies with that of other molecular weights (MWs) 11,300 and to that of base ]59 in the viroid sequence. uniramian taxa. We do not know the 8,500) are the more interesting. The Dashed line, polypeptide of MW 11,300; significance of onychophoran Golgi longest of these sequences begins with a dotted-dashed line, polypeptide of MW phenomena nor of its association with GUG at positions 134-136 and teminates 8,500. unstriated muscle, but this one feature is at a double U AG at positions 99-104 (Fig. no acceptable reason to disrupt the Uni 1). The GUG is preceded by a well-placed In the light of the above evidence I ramia or exclude the Onychophora from potential ribosome recognition sequence suggest that viroids may be very small the arthropods. as indicated in Fig. 1. The second potential negative-strand viruses with the following Thus, without further consideration of polypeptide is in a different reading frame properties. (1) They use secondary struc the several other arthropodan taxa, it is and begins with a GUG at residues 177- ture rather than a protein coat for protec clear that the Uniramia and Crustacea 179. It is preceded by an unbroken tion of the genome; (2) they use a host each represent well defined arthropodan sequence of 14 purines, and teminates in polymerase to make a complementary taxa, which could not have originated an UAG at residues 55-57. The amino RNA strand; and (3) this strand serves as a from common annelidan ancestry, and acid composition of the small polypeptide template for progeny viroid particles and that their cuticles must have evolved in is unremarkable except for a high content as message for one or two viroid proteins. parallel. of phenylalanine (1 0% ), but the larger This hypothesis can be tested by a search The exoskeleton of all arthropods has polypeptide has a content of arginine in PSTV -infected tissues for one or two the same mechanical functions in provi (11 %) and lysine (5%) that would give it viroid-induced polypeptides of the pre ding unstretchable cuticles, sclerites and histone-like properties. dicted size and composition. arthrodial membranes concerned with Two experimental observations would R. E. F. MATTHEWS trunk movements and muscle insertions. fit with the view that one or both of these Department of Cell Biology, The requirements of other invertebrates polypeptides are produced by viroids. University of Auckland, are different. The cuticles of the various First, there is evidence for an RNA strand Private Bag, Auckland, arthropodan taxa are not identical, complementary to citrus exocortis viroid 3 New Zealand although protein and fine-fibre systems (CEV) RNA in infected tissues ; CEV is 2 form the basis of most sclerites. There is fairly closely related to PSTV . Some of I. Gross. H. J.
0028-0836/78/02 76--V850$0 1.00 © Macmillan Journals Lid Wl\\