Caddo agilis and C. pepperella (Opiliones, Caddidae) diverged phylogenetically before acquiring their disjunct, sympatric distributions in Japan and North America.
The harvestmen Caddo agilis Banks 1892 and C. pepperella Shear 1975
(Caddidae, Caddinae) share a disjunct distribution in eastern Asia and
eastern North America that has been attributed to either recent
(Pleistocene) evolution of a C. pepperella morph from C. agilis in each
region or to a pre-glacial separation within each of two established
species. The present study used 2,130-base sequences from two nuclear
protein-coding genes (EF1a, Pol II) to test the phylogenetic predictions
of both hypotheses using representatives from the two Caddo species from
both regions and two acropsopilionine outgroup species. The results
supported the hypothesis that the two Caddo species were distinct prior
to their respective biogeographic disjunctions; C. agilis and C.
pepperella were each recovered as monophyletic and each appears to have
undergone separation into Asian and North American groups.
Keywords: Harvestmen, phylogeny, elongation factor-1a, RNA polymerase II
Harvestmen (Genetic aspects)
Harvestmen (Natural history)
Shultz, Jeffrey W.
Regier, Jerome C.
|Publication:||Name: Journal of Arachnology Publisher: American Arachnological Society Audience: Academic Format: Magazine/Journal Subject: Biological sciences; Zoology and wildlife conservation Copyright: COPYRIGHT 2009 American Arachnological Society ISSN: 0161-8202|
|Issue:||Date: May, 2009 Source Volume: 37 Source Issue: 2|
|Topic:||Event Code: 310 Science & research; 690 Goods & services distribution Advertising Code: 59 Channels of Distribution Computer Subject: Company distribution practices|
|Geographic:||Geographic Scope: Japan Geographic Code: 9JAPA Japan; 100NA North America|
This study focuses on the phylogeny and biogeography of the two
extant species of Caddo, C. agilis Banks 1892 and C. pepperella Shear
1975 (Caddidae, Caddinae), both of which occur in eastern North America
and Japan. Two scenarios have been offered in the arachnological
literature to explain the disjunct sympatric distribution of the two
species: the parallel-evolution hypothesis (Shear 1975, 1996, 2004) and
the habitat-fragmentation hypothesis (Suzuki 1976). Shear (1975)
originally suggested that the smaller C. pepperella evolved in North
America as a paedomorphic (progenetic) variant of C. agilis, perhaps as
an adaptation to shorter growing seasons associated with glacial
conditions during the Pleistocene. The subsequent discovery of C.
pepperella in Japan (Suzuki 1976) was inconsistent with Shear's
hypothesis, as it seemed to require an improbable recent dispersal
between eastern North America and Japan. Shear (1996, 2004) countered
that a C. pepperella morph may have evolved independently in North
America and Japan during the Pleistocene. In contrast, Suzuki (1976)
proposed that both species originally inhabited an ancient circumboreal
ecosystem that now consists of isolated postglacial fragments in eastern
Asia and eastern North America.
The parallel-evolution and habitat-fragmentation hypotheses can be tested by means of molecule-based phylogenetic analysis, with the former predicting diphyly within C. pepperella and the latter predicting monophyly of both C. agilis and C. pepperella across their ranges. Here we test these hypotheses using 2,130 base pairs of the nuclear protein-coding genes elongation factor-1a) (EF-1a) and RNA polymerase II (Pol II) from representative C. agilis and C. pepperella from North America and Japan as well as two acropsopilionine outgroup species, Austropsopilio sudamericanus Shultz & Cekalovic 2003 and Acropsopilio chilensis Silvestri 1904. Our results corroborate Suzuki's habitat-fragmentation hypothesis.
Terminal taxa and sequences.--Specimens were collected alive and preserved in > 95% ethanol. They were stored in > 95% ethanol at -20[degrees] C up to 2 yr prior to RNA extraction. The analysis was based on sequences from six specimens, with collection data and GenBank accession numbers as follows:
1. Caddo agilis Banks 1892. USA: New Hampshire: Cheshire County, Pisgah State Park, 42.868[degrees]N, 72.448[degrees]W, 7-11 July 2001, J.W. Shultz (EF-1a: FJ361272; Pol II: FJ476262-FJ476264).
3. Caddo pepperella Shear 1975. USA: New Hampshire: Cheshire County, Pisgah State Park, 42.868[degrees]N, 72.448[degrees]W, 7-11 July 2001, J.W. Shultz (EF-1a: FJ361272; Pol II: FJ476265-FJ476267).
4. Caddo pepperella. JAPAN: Tottori Prefecture: Mt. Nagi, 630 m elev., 20 June 1998, N. Tsurusaki (EF-1[alpha]: AF240863; Pol II: AH010457).
5. Acropsopilio chilensis. CHILE: Provinica de Concepcion: Cerro Caracol, 5 October 2003, T. Cekalovic (EF-1a: FJ361275; Pol II: FJ476256 - FJ476258).
6. Austropsopilio sudamericanus. CHILE: Provincia de Valdivia: Cerro Oncol, April 2001, T. Cekalovic (EF-1a: FJ361274; Pol II: FJ476259-FJ476261).
A voucher specimen of each species is deposited in the National Museum of Natural History (Smithsonian Institution) except for C. pepperella from Japan, because the only specimen available was consumed in genomic extraction.
Molecular methods.--Detailed procedures for generating sequence data, including primer sequences, have been published elsewhere (Regier & Shultz 1997). In brief, total nucleic acids were isolated; complementary DNA of EF-1[alpha] and Pol II mRNA was made by reverse transcription; ds-DNA copies were amplified by PCR and subsequently gel isolated; the resulting PCR fragments were used as templates for another round of PCR amplification with nested primers; and the resulting fragments were gel isolated and sequenced. When the resulting fragment concentration was too low to sequence directly, it was either concentrated or reamplified using the M13 sequences present at the 5' ends of all primers. The same M13 sequences were also used as primers for thermal cycle/dideoxy sequencing. Sequencing reactions were fractionated and preliminary analyses were performed with Perkin-Elmer/ABI automated DNA sequencers. Automated DNA sequencer chromatograms were edited and contigs were assembled using the pregap and gap4 programs within the Staden software package (Staden et al. 1999). Sequences were aligned and Nexus-formatted nucleotide data sets were constructed using the Genetic Data Environment, version 2.2 (Smith et al. 1994). All sequences lacked indels. Amino acid data were inferred from nucleotide sequences using the universal nuclear genetic code option in MacClade, ver. 3.08 (Maddison & Maddison 1992).
Phylogenetic analysis.--Parsimony analyses of three data sets [all nucleotides (nt1-3), third codon positions (nt3) and inferred amino acids (aa)] were performed in PAUP4.0 (Swofford 1998) using unordered, equally weighted characters. Analyses consisted of exhaustive searches followed by bootstrap analyses (Felsenstein 1985) based on branch-and-bound searches of 1,000 pseudoreplicates. In conducting maximum-likelihood (ML) analysis, the program Modeltest, ver. 3.7 (Posada & Crandall 1998) was used to choose a model for the nt1-3 and nt3 data sets using AIC (Posada & Buckley 2004), with specific parameter values being estimated during subsequent phylogenetic analysis. The ML analyses were conducted in PAUP* using exhaustive searches and nonparametric bootstrap analyses were performed using branch-and-bound searches of 1,000 pseudoreplicates.
Parsimony analyses of all nucleotides (nt1-3), 3rd codon positions (nt3) and inferred amino acids (aa) produced identical, fully resolved topologies with bootstrap percentages (BP) of 98-100 for all internal nodes (Fig. 1) (nt1-3: 2,130 characters, 398 informative, length = 897, CI = 0.8305; nt3: 710 characters, 317 informative; length = 756, CI = 0.816; aa: 710 characters, 55 informative; length = 94, CI = 0.9492). Caddo agilis and C. pepperella were each recovered as monophyletic and reconstructed as sister groups with respect to the acropsopilionines, Acropsopilio chilensis and Austropsopilio sudamericanus.
Comparison of alternative likelihood models in Modeltest indicated that the nt1-3 data should be analyzed using a GTR+[[GAMMA].sub.4]+I model and that the nt3 matrix should be analyzed under the TVM+[[GAMMA].sub.4] model. Exhaustive likelihood searches using these models recovered topologies identical to those derived from parsimony-based analyses (nt1-3: -ln likelihood = 6649.2923; nt3, -ln likelihood = 3391.9731). Specifically, the clades C. agilis, C. pepperella, and Caddo were each recovered as monophyletic with strong support (BP 99-100%) (Figs. 2, 3), a result predicted by Suzuki's hypothesis. The two data sets were reanalyzed under their respective models with the analyses constrained to yield Shear's hypothesis of independent evolution of C. pepperella from C. agilis in Asia and North America and Suzuki's hypothesis of monophyly of both C. agilis and C. pepperella throughout their ranges. Kishino-Hasegawa tests (Kishino & Hasegawa 1989) showed the trees constrained to the prediction of Shear's hypothesis to be significantly less likely than those constrained to Suzuki's hypothesis (P < 0.001).
Our results indicate that Caddo agilis and Caddo pepperella are monophyletic species that diverged phylogenetically before each acquired a disjunct geographic distribution in Japan and eastern North America. This supports Suzuki's (1976) habitat-fragmentation hypothesis and is inconsistent with Shear's (1975, 1996, 2004) paralle-levolution hypothesis. These findings are consistent with current understanding of climatic and biogeographic events during the late Tertiary (Sanmartm et al. 2001). The eastern Asia-eastern North America disjunction exemplified by Caddo parallels a long-known biogeographic pattern among flowering plants (Wen 1999; Xiang et al. 2000). During the mid-Cenozoic, eastern Asia and eastern North America were spanned by mesophytic forests that were eventually separated into Asian and American components during the early Pliocene, a culmination of long-term trends in the cooling and drying of central and northern North America. As a consequence, many plant genera have representative species in both eastern Asia and eastern North America, and relative rates tests conducted on 12 species pairs using the rbcL gene indicated a divergence time of 5.4 6 2.6 million years ago (Xiang et al. 2000). Zoologists have not explored the Asian-North American disjunction to the same extent as botanists, but several examples are known among animals (Sanmartin et al. 2001), including the non-caddine harvestmen Acropsopilio boopis (Crosby 1904) and Crosbycus dasycnemus (Crosby 1911), Okeantobates millipedes (Enghoff 1993), plethodontid salamanders (Min et al. 2005) and, among late Tertiary fossils, lesser pandas and meline badgers (Tedford & Harington 2003; Wallace & Wang 2004).
[FIGURES 1-3 OMITTED]
In contrast, because the two Caddo species are roughly sympatric and often syntopic in both North America and Japan, vicariant or climatic events cannot readily explain their phylogenetic divergence or morphological differences. It is possible that the two extant Caddo species diverged due to resource or habitat partitioning, as C. agilis tends to occupy exposed surfaces (e.g., tree trunks, logs, stones) and C. pepperella occurs on the ground in the leaf litter and under fallen objects (Suzuki 1976; Shultz, unpubl. obs.). Given the similarity between C. pepperella and juvenile C. agilis,itispossiblethatsuch habitat specialization produced morphological differences between the two species via heterochrony, in a manner similar to that proposed by Shear (1975, 1996, 2004). Still, there is no clear evidence as to whether C. agilis is peramorphic with respect to its ancestor, whether C. pepperella is paedomorphic with respect to its ancestor, both, or neither. Outgroup comparison with the small soil- or litter-dwelling acropsopilionines (Caddidae) would seem to favor C. pepperella as the better model for the common ancestor of extant Caddo and thus evolution of C. agilis via hypermorphosis. However, without relevant information about the morphology and development of ancestral and extant Caddo,this matter will remain an exercise in speculation.
We thank Tomas Cekalovic and Nobuo Tsurusaki for specimens and two anonymous reviewers for comments. This research was supported by NSF grants 9981970 and 0640179. JWS was supported by the Maryland Agricultural Experiment Station.
Manuscript received 1 August 2008, revised 5 December 2008.
Enghoff, H. 1993. Phylogenetic biogeography of a Holarctic group: the julidan millipedes. Cladistic subordinateness as an indicator of dispersal. Journal of Biogeography 20:525-536.
Felsenstein, J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39:783-791.
Kishino, H. & M. Hasegawa. 1989. Evaluation of the maximum likelihood estimate of the evolutionary tree topologies from DNA sequence data, and the branching order in Hominoidea. Journal of Molecular Evolution 29:170-179.
Maddison, W.P. & D.R. Maddison. 1992. MacClade: Analysis of Phylogeny and Character Evolution. Version 3. Sinauer Associates, Sunderland, Massachusetts. 398 pp.
Min, M.S., S.Y. Yang, R.M. Bonett, D.R. Vieites, R.A. Brandon & D.B. Wake. 2005. Discovery of the first Asian plethodontid salamander. Nature 435:87-90.
Posada, D. & T.R. Buckley. 2004. Model selection and model averaging in phylogenetics: advantages of Akaike Information Criterion and Bayesian approaches over likelihood ratio tests. Systematic Biology 53:793-808.
Posada, D. & K.A. Crandall. 1998. Modeltest: testing the model of DNA substitution. Bioinformatics 14:917-818.
Regier, J.C. & J.W. Shultz. 1997. Molecular phylogeny of the major arthropod groups indicates polyphyly of crustaceans and a new hypothesis for the origin of hexapods. Molecular Biology and Evolution 14:902-913.
Sanmartin, I., H. Enghoff & F. Ronquist. 2001. Patterns of animal dispersal, vicariance and diversification in the Holarctic. Biological Journal of the Linnean Society 73:345-390.
Shear, W.A. 1975. The opilionid family Caddidae in North America, with notes on species from other regions (Opiliones, Palpatores, Caddoidea). Journal of Arachnology 2:65-88.
Shear, W.A. 1996. Hesperopilio mainae, a new genus and species of harvestman from Western Australia (Opiliones: Caddidae: Acropsopilioninae). Records of the Western Australian Museum 17:455-460.
Shear, W.A. 2004. Description of the female of Acropsopilio chomulae (Goodnight & Goodnight 1948) from Chiapas, Mexico (Opiliones, Caddidae, Acropsopilioninae). Journal of Arachnology 32:432-435.
Smith, S.W., R. Overbeek, C.R. Woese, W. Gilbert & P.M. Gillevet. 1994. The genetic data environment and expandable GUI for multiple sequence analysis. Computer Applications in the Biosciences 10:671-675.
Staden, R., K.F. Beal & J.K. Bonfield. 1999. The Staden Package, 1998. Pp. 115-130. In Bioinformatics Methods and Protocols. (S. Misener & S. Krawetz, eds.). The Humana Press, Totowa, New Jersey.
Suzuki, S. 1976. The harvestmen of the family Caddidae in Japan (Opiliones, Palpatores, Caddoidea). Journal of Science, Hiroshima University, Series B, Division 1 26:261-273.
Swofford, D.L. 1998. PAUP* 4.0 Beta Version. Sinauer Associates: Sunderland, Massachusetts.
Tedford, R.H. & C.R. Harington. 2003. An arctic mammal fauna from the early Pliocene of North America. Nature 425:388-390.
Wallace, S.C. & X. Wang. 2004. Two new carnivores from an unusual late Tertiary forest biota in eastern North America. Nature 431:556-559.
Wen, J. 1999. Evolution of eastern Asian and eastern North American disjunct distributions in flowering plants. Annual Review of Ecology and Systematics 30:421-455.
Xiang, Q-Y., D.E. Soltis, P.S. Soltis, S.R. Manchester & D.J. Crawford. 2000. Timing of the eastern Asian-eastern North American floristic disjunction: molecular clock corroborates paleontological estimates. Molecular Phylogenetics and Evolution 15:462-472.
Jeffrey W. Shultz: Department of Entomology, University of Maryland, College Park, Maryland 20742, USA. E-mail: email@example.com
Jerome C. Regier: Center for Biosystems Research, University of Maryland Biotechnology Institute, College Park,
|Gale Copyright:||Copyright 2009 Gale, Cengage Learning. All rights reserved.|