Integrative and Comparative Biology Advance Access originally published online on October 30, 2007
Integrative and Comparative Biology 2007 47(6):815-830; doi:10.1093/icb/icm093
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
Phylogenetic analysis of lineage relationships among hyperiid amphipods as revealed by examination of the mitochondrial gene, cytochrome oxidase I (COI)
*Kewalo Marine Lab, Pacific Biosciences Research Center, University of Hawaii, 41 Ahui St Honolulu, HI 96813, USA;
Monterey Bay Aquarium Research Institute, 7700 Sandholdt Road, Moss Landing, CA 95039, USA
Correspondence: 1E-mail: wbrowne{at}hawaii.edu
 |
Synopsis |
|---|
Top
 Synopsis IntroductionMaterials and methodsResultsDiscussionSupplementary dataAcknowledgmentsReferences |
|---|
Among metazoans, crustaceans display the greatest disparity between body plans and are second only to the insects in overall species diversity. Within the crustaceans, the Amphipoda rank as one of the most speciose extant orders. Amphipods have successfully invaded a variety of ecosystems, including the pelagic midwater environment. Despite their abundance in varied and dissimilar habitats, and the use of traditional morphological and systematic comparative analyses, phylogenetic relationships among amphipods remain uncertain. The pelagic amphipods, hyperiids, have highly divergent life histories and morphological attributes in comparison to more familiar benthic, nearshore, intertidal, and terrestrial amphipods. Some of these adaptations are likely correlated with their pelagic life history and include features such as hypertrophied olfactory and visual systems, duplications of the eyes, and an array of modifications to the appendages. Many of these morphological features may represent homoplasies, thus masking the true phylogenetic relationships among extant hyperiid amphipods. Here, we sample a wide range of amphipod taxa for the COI gene and present the first preliminary molecular phylogeny among the hyperiids.
|
Introduction |
|---|
|
Top
SynopsisIntroductionMaterials and methodsResultsDiscussionSupplementary dataAcknowledgmentsReferences |
|---|
Amphipoda [Crustacea; Malacostraca; Peracarida] is a monophyletic, species-rich, assemblage (Schram 1986
; Schmitz 1992). They occur in nearly all known marine, freshwater, and brackish water environments as well as in highly humid terrestrial ecosystems (Barnard and Karaman 1991; Van Dover 1992; Vinogradov et al. 1996; Poltermann et al. 2000; Takhteev 2000; Serejo 2004). This ecological diversity is reflected in similarly high levels of morphological variation. The unique structure and arrangement of the posterior-most three pairs of appendages (uropods) represent the exclusive, synapomorphic, characters that unite the amphipods as a natural, monophyletic group. They are further distinguished by a combination of additional features including lateral compression of the body, presence of sessile compound eyes, the general orientation of the thoracomere appendages (pereopods) to the body axis, and the close arrangement of the anterior gnathal appendages, including the maxillipeds, into a basket-like shape around the mouth to form a compact buccal mass. The amphipods have traditionally been organized into four groups, the largely benthic taxa Gammaridea, Caprellidea, Ingolfiellidea, and the exclusively pelagic midwater taxon Hyperiidea (Martin and Davis 2001). The pelagic hyperiid amphipods are a major constituent of crustacean zooplankton (Bowman and Gruner 1973) and in some regions their swarming behavior leads to their being a primary food source for large planktivores (Vinogradov et al. 1996).
The phylogenetic relationships among the hyperiids remain a mystery. In fact, the relationships among, and within, all four of the major amphipod taxonomic groups remain poorly resolved due to conflicting suites of morphological characters currently in use by systematists (Martin and Davis 2001). This presents a conundrum in which we have no large-scale phylogenetic hypotheses for the Amphipoda, a highly successful monophyletic group that has high biological diversity coupled with evolutionary radiations in dissimilar environments. Although there have been recent advances in morphological and molecular analyses among some of the gammaridean groups of amphipods (Meyran et al. 1997; Englisch et al. 2003; Myers and Lowry 2003; Lörz and Held 2004; Serejo 2004; Davolos and Maclean 2005; Macdonald et al. 2005), there have been no molecular phylogenetic studies and few comparative morphological analyses among the hyperiids (Pirlot 1932; Coleman 1994; Zeidler 1999, 2003a, 2003b, 2004). This is surprising given that many descriptions of hyperiid species date from the 19th century and nearly all midwater plankton tows, at any depth, recover representatives from the group. Some of the major biological and evolutionary questions that rely on a hypothesis of lineage relationships, and thus remain unaddressed in this group include how patterns of biological diversity arise, how biological form and function are linked to evolutionary radiations, and in the particular case of the pelagic hyperiids, what aspects of their body plan are required for successful colonization and radiation within midwater niches. Phylogenetic analysis can ultimately address whether an ancestral benthic amphipod stem group invaded the pelagos, or conversely, whether an ancestral pelagic amphipod stem group invaded the benthos, or whether there has been a more complex interplay between hyperiid lineages with regard to colonization of oceanic midwater habitats.
An accurate phylogenetic assessment using purely morphological features is problematic. Thus far, there is no single synapomorphy known for the hyperiids. As a consequence the notion of hyperiid polyphyly has been suggested in some of the major taxonomic works (reviewed by Vinogradov et al. 1996). This suggests extensive convergence, or homoplasy, of groups of morphological attributes required for a pelagic existence and for their frequently observed associations with gelatinous zooplankton.
Convergent evolution and associated character homoplasy are one of the central themes in evolutionary biology and arguably one of the most difficult problems facing phylogenetic reconstruction of relationships within and among lineages. Character homoplasy can easily mask relationships between related lineages (examples among amphipods include structural simplifications such as reductions of the maxilliped, the pleopods, and the uropods). Environmental factors play a strong role in the evolutionary shaping of morphologies through time. Since all hyperiid taxa are pelagic, it is instructive to consider some of the major features affecting organisms in the marine midwater environment. Volumetrically, the oceanic midwater is massive, accounting for 
90–99.8% of the habitable space on the planet (Cohen 1994). Although vast, the oceanic midwater environment is far from being featureless and many biologically relevant partitions can, and should be, taken into account when considering organismal diversity in the midwater.
One of the most critical parameters affecting midwater organisms is light (reviewed by Warrant and Locket 2004). The average depth of the open ocean is 3800 m. Down-welling ambient light (sun, moon, and starlight) penetrates as deep as 1000 m. Long wavelength light is rapidly absorbed within the first 150 m of the water column, the epipelagic zone, leaving only shorter wavelength light to illuminate the mesopelagic zone (150–1000 m depth). The remaining 75% of midwater habitat, the bathypelagic (1000–2500 m depth) and abyssopelagic (2500 m depth and beyond) zones, is characterized by an absence of down-welling surface light. In the bathypelagic and abyssopelagic zones down-welling light is completely replaced by omnidirectional bioluminescent point light sources. Many hyperiids participate in well-characterized cyclical, vertical, diel migrations driven by light, in which zooplankton move upward in the water column at dusk and downward at dawn. This includes many hyperiid species described as being associated with the mesopelagic and bathypelagic midwater zones (Vinogradov et al. 1996). Two additional biologically important parameters of consequence at mesopelagic, bathypelagic, and abyssopelagic depths are dissolved oxygen content and temperature. Between 400 and 1000 m is an oxygen poor water layer of variable thickness, the hypoxic oxygen minimum layer (OML), a niche in which there are particular adaptations such as reduced metabolic rates, improved oxygen uptake, and increased dependence on anaerobic metabolism (Childress and Seibel 1998). Additionally, these depths are characterized by constant temperatures near 4°C.
Another significant physical attribute of the midwater environment is its inherent patchiness. This patchiness, or granularity, creates substantial 3D surfaces with which small clinging planktonic organisms, in particular, can interact. This patchiness is a combination of both the planktonic organisms themselves, as well as a host of other organic aggregates collectively known as "marine snow" (Silver et al. 1978; Robison et al. 2005). Among the Amphipoda, modifications of morphologies associated with a clinging benthic life style could play a role in exploiting the variety of 3D surfaces present in midwater habitats (Laval 1980).
The exclusively pelagic life style of the hyperiids has historically made detailed studies of these amphipods difficult. As in situ observations of hyperiids continue to emerge, it is clear that most species have complex life histories involving obligate commensalism or parasitic relationships with a wide variety of midwater gelatinous zooplankton such as siphonophores, salps, ctenophores, and cnidarians (Harbison et al. 1977; Madin and Harbison 1977; Harbison et al. 1978; Laval 1980; Gasca and Haddock 2004; Gasca et al. 2007). Thus, the availability of surfaces in midwater appears to play an important ecological role among hyperiid amphipods and morphological specializations in different hyperiid lineages may derive from interactions with preferred surfaces.
We have used recent advances in SCUBA technology and modern techniques for collecting in midwater in combination with traditional netting to sample a diversity of hyperiid amphipods for the mitochondrial gene, cytochrome oxidase I (COI), for a preliminary molecular assessment of hyperiid phylogenetic relationships. COI encodes for a subunit of cytochrome oxidase, an enzymatic protein complex absolutely required for aerobic metabolism (Castresana et al. 1994). COI was chosen for preliminary analysis for several reasons including maternal inheritance of the mitochondrial genome, reduced occurrence of mitochondrial gene recombination, extensive use of COI in phylogenetic reconstructions, and the availability of related sequences from publicly curated databases (Avise 1986; Avise et al. 1987; Simon et al. 1994; Simon et al. 2006).
|
Materials and methods |
|---|
|
Top
SynopsisIntroductionMaterials and methodsResultsDiscussionSupplementary dataAcknowledgmentsReferences |
|---|
Specimen collection
Amphipod specimens were collected using a variety of techniques. Intertidal amphipods were collected by hand. Sub-surface benthic amphipods were collected via snorkeling, open-circuit (OC), and closed-circuit rebreather (CCR) SCUBA (Ambient Pressure Diving Ltd., UK). Pelagic amphipods were collected using a combination of snorkeling, OC, and CCR SCUBA blue-water diving (Hamner 1975
; Hamner et al. 1975; Haddock and Heine 2005), remotely operated underwater vehicles (MBARI), ring nets, and opening-closing trawling nets (Childress et al. 1978). Physical vouchers exist for all specimens and are housed at the Kewalo Marine Lab (Honolulu, HI, USA). Specimens of Parhyale hawaiensis came from a breeding colony maintained at the Kewalo Marine Laboratory (Browne et al. 2006) and those of Jassa slatteryi came from a breeding colony maintained at the University of California at Berkeley (Patel). Specimens of hyperiid amphipods were identified using the taxonomic keys of Bowman and Gruner (1973), Vinogradov et al. (1996), and Zeidler (1999, 2003a, 2003b, 2004). Thirteen samples included in this analysis were not identified to species level. They have been assigned temporary names indicating their affinity to described species.
Sequence cloning
Genomic DNA was isolated with DNeasy Tissue Kit (Qiagen, Inc.) from isolated pleopod and/or dissected and isolated trunk muscle tissue. COI PCR was completed using conserved primers LCO1490 5-GGT CAA CAA ATC ATA AAG ATA TTG G-3 and HCOoutout 5-GTA AAT ATA TGR TGD GCT C-3 (Folmer et al. 1994; Schwendinger and Giribet 2005). PCR products of the appropriate size were direct sequenced by Macrogen, Inc (South Korea).
Phylogenetic analysis
Relevant COI sequences from 94 gammarids and two isopod outgroups were selected and added to the analysis from Genbank. Sequences were initially aligned and edited using Sequencher (Gene Codes Corporation). Multiple sequence alignments were generated using T-Coffee (Notredame et al. 2000). Alignments were then manually viewed and adjusted in MacClade v4.08 (Maddison and Maddison 2005).
Bayesian phylogenetic inference analysis was executed with MrBayes 3.1.2 (Ronquist and Huelsenbeck 2003). The molecular evolution model, GTR+I+
, was selected using the Akaike Information Criterion (AIC) as implemented in Model-Test 3.7 (Posada and Crandall 1998). Four independent searches were run from 30 million generations to 40 million generations, and trees were sampled every 100 generations. Likelihoods were visualized with Tracer v1.3 (Rambaut and Drummond 2003) to determine the number of generations to burn-in and to assess convergence of data sets. Consensus trees from the independent runs were compared to assess convergence and topology congruence of data sets. A consensus tree from the combined Bayesian runs was constructed and represents a total of 130 million generations from the four independent runs with trees sampled every 1000 generations and an initial burn-in of 5 million generations. Consensus trees were rooted with the isopod Armadillidium vulgare COI sequence. Resulting trees were visualized with TreeView X (Page 1996) and FigTree (http://evolve.zoo.ox.ac.uk/software.html?id=figtree).
Maximum likelihood (ML) analysis was executed with RAxML-VI-HPC (Stamatakis 2006). The model GTRMIX was selected for executing four initial runs, each with 250 searches. Searches from the ML runs were combined and the consensus from the top 500 trees was visualized to assess convergence of topologies. Bootstrapping of ML analysis was executed with Seqboot (Felsenstein 1989) followed by 20 independent searches for the most likely tree per bootstrap iteration with RAxML-VI-HPC. The best tree was retained for each of the 129 iterations that were executed. Resulting trees were visualized with FigTree (http://evolve.zoo.ox.ac.uk/software.html?id=figtree).
A mutational saturation plot (Philippe et al. 1994; Philippe and Forterre 1999) was generated using a maximum parsimony (MP) analysis with 10 random sequence addition MP heuristic searches. The best tree was used to generate a patristic distance matrix. Individual pair-wise values were compared on X–Y scatter plots. The X-axis was used for the inferred substitution values and the Y-axis was used for the observed substitution values.
|
Results |
|---|
|
Top
SynopsisIntroductionMaterials and methodsResultsDiscussionSupplementary dataAcknowledgmentsReferences |
|---|
Phylogenetic analysis
A diversity of amphipods from the Northeast, Northwest, and Central Pacific, the western edge of the Atlantic Gulf Stream, the Caribbean Sea, Germany, and the Weddell Sea were collected and sampled for
815 bp of the COI gene (Fig. 1, Table 1). The 72 new amphipod COI sequences obtained for this study have been deposited with Genbank (Table 1). In an effort to test the hyperiids as a monophyletic group within Amphipoda, additional nonhyperiid amphipod COI sequences available from Genbank were incorporated into our phylogenetic analyses (Table 2). Our COI analysis included 168 taxa of which 52 represent a broad sampling among known hyperiidean forms.
|
|
|
Bayesian phylogenetic inference was used to infer COI gene lineage relationships. The number of generations required to stability (burn-in) was as high as 4.2 million, with an average burn-in of 3.5 million generations across four independent runs. This was due, in part, to the large number of taxa analyzed. Our combined bayesian COI analysis recovered three clades of hyperiid amphipods (Figs. 2 and 4, Supplementary data). The hyperiids comprising clade 1 appear to branch early among the taxa included in this analysis. However, the relationship between hyperiids comprising clades 2 and 3, as defined in this analysis, remain unclear. Although independent runs always recovered the same topologies among hyperiid taxa within each of the three clades, the relationship between clade 2 and clade 3 varied. The clade 2 hyperiids alternate between sister to clade 3 hyperiids (Supplementary data) and sister to the iphimediid gammarid clade (Supplementary data); however, in both cases the resolution at these deep nodes is low. ML was also used to infer COI gene lineage relationships and bootstrap values were mapped to the most likely ML tree. Our combined ML COI analysis recovers three clades of hyperiids nearly identical to those from our Bayesian results (Figs. 2, 3, and 4, Supplementary data). The most likely ML tree unites the three hyperiid clades. However, there is no bootstrap support for this result (Fig. 3, Supplementary data). With a high degree of support, we also recovered monophyletic caprellids (from the small number included in this study) as branching from within the gammariids (Figs. 2 and 3, Supplementary data). An internal assessment of our results was by recapitulation of the two Lake Bailkal gammariid radiations (Macdonald et al. 2005
) and recapitulation of the Antarctic radiations of the Epimeria and Iphimediid gammariids (Lorz and Held 2004) (Supplementary data).
|
|
|
Although, we found many amphipod crown-group relationships to be highly supported, the majority of the deep nodal relationships among the Amphipoda are poorly supported and remain largely unresolved from COI sequencing alone (Supplementary data). Interestingly, most hyperiid COI sequences have long branches relative to the other amphipod sequences examined. This is indicative of a differential rate of change occurring among hyperiid COI sequence clades relative to the other amphipod COI sequences examined.
Morphological homoplasy among the Amphipoda is a well-characterized phenomenon that has historically confounded taxonomic descriptions (Barnard and Karaman 1991; Vinogradov et al. 1996), and the use of molecular data for phylogenetic methodologies is not immune from the effects of homoplasy. At the level of DNA sequences, homoplasy can manifest as site-specific variation that escapes detection due to two proximate causes (1) inadequate taxon sampling which thereby misses phylogenetically informative sequence variation, and (2) a rapid rate of sequence change among the taxa being examined which eliminates evidence of past, phylogenetically useful, sequence variation. To assess the potential influence of sequence homoplasy and to further characterize the variation in branch lengths that we observed among the taxa sampled, we generated a mutation saturation plot. We used maximum parsimony to generate a patristic distance matrix (Philippe and Forterre 1999). In a mutation saturation plot, the presence of a curve with a mutational plateau indicates that progressive saturation for mutational changes is occurring and thus there is a loss of the phylogenetic signal necessary for accurately resolving deep branching nodes (Philippe et al. 1994; Philippe and Forterre 1999).
Our scatter plot of observed versus inferred COI substitutions has an atypical, wide, cloud-like distribution (Fig. 5A). In general, there were fewer observed substitutions than predicted to have occurred between COI sequence pairs. No plateau corresponding to COI mutational saturation could be easily discerned from the full data set. Sampling subsets of COI sequence, however, revealed three distinguishable patterns of COI gene variation segregating among the taxa in this study (Fig. 5B–D). This indicates divergent, nonuniform, patterns of sequence change among COI gene lineages in the Amphipoda.
|
Pair-wise comparisons among the three hyperiid clades demonstrates that among these taxa there is a saturation curve containing many observed substitutions that are relatively close to inferred substitutions as well as an expected mutational plateau (Fig. 5B). Therefore, although the hyperiid COI branches are generally longer than the other amphipod sequences considered, they should contain sufficient sequence variation to reliably resolve many crown-group relationships (Fig. 4). Each of the three detected hyperiid COI clades independently retains the general shape of this curve (Supplementary data).
Superimposing pair-wise comparisons between the largest recovered monophyletic gammariid COI clade exclusive of hyperiid sequences (Fig. 2) reveals a cloud-like cluster (Fig. 5C). The cluster of substitution values does not display a discernable curve and plateau trend. The cluster departs strongly from the ideal observed/inferred substitution ratio, and a significant proportion of pair-wise values exhibit low inferred substitution values coupled with low observed substitution values. This supports the general observation of shorter branch lengths observed in this gammariid clade. The shape of this cluster strongly suggests that mutational saturation has removed most of the phylogenetic signal associated with comparisons between these sequences. Indeed, a majority of the nodes in this clade are characterized by low Bayesian posterior probabilities and no ML bootstrap support, and the deep nodes in particular are not well supported (Fig. 2).
Finally, superimposing pair-wise comparisons between the hyperiids and all other amphipod taxa results in an additional informative shift in the cluster of substitution values (Fig. 5D). The distribution of the values observed in this case suggests that the majority of the pair-wise comparisons between gammarid and hyperiid COI sequences have particularly low phylogenetic signal, as observations of sequence substitution fall dramatically relative to inferred sequence substitutions. Consideration of the shape of clustered values in Fig. 5C and D confirms that mutational saturation for COI has occurred among the gammarid amphipods included in this analysis. The observation of homoplasy among gammarid versus gammariid and gammariid versus hyperiid COI sequences does not prevent reconstruction of lineage relationships within hyperiid clades. However, COI sequence homoplasy does prevent the accurate recovery of relationships between the three recovered hyperiid clades that contain deeper nodes that may interdigitate with ill-defined gammariid taxa.
|
Discussion |
|---|
|
Top
SynopsisIntroductionMaterials and methodsResultsDiscussionSupplementary dataAcknowledgmentsReferences |
|---|
The phylogeny of pelagic midwater hyperiid amphipods
The histories of patterns of descent by modification are difficult to discern when shared attributes are the result of convergence. Among the Amphipoda, morphological convergence (homoplasy) is well documented. To help improve our understanding of uncertain relationships and the relative importance of morphological characters, we examined the COI gene among 168 amphipodan taxa including 52 hyperiids. Our results indicate that there may have been as many as three independent radiations among the hyperiids. In addition, our COI data suggest taxonomic affinities for several previously enigmatic hyperiid taxa. Bayesian results suggest that at least one of the hyperiid radiations (hyperiid clade 1 in Fig. 2) may be an early diverging lineage within Amphipoda. Our Bayesian results also suggest that the hyperiids are likely to be polyphyletic (Fig. 2). ML results unite the three hyperiid clades but with no bootstrap support (Fig. 3). We also find a highly supported monophyletic clade of caprellids from the COI sequence data included in this analysis. In our analyses, the caprellid clade is always sister to the lineage that includes the ischyrocerid J. slatteryi (Supplementary data). This result is in agreement with recent morphological analyses of corophiid and caprellid groups (Myers and Lowry 2003
). COI sequence variation among the amphipod taxa sampled does not, however, resolve either the relationships between separate hyperiid radiation events or the relationships of lineages among the majority of the gammarids due to inadequate resolution at deeper nodes.
Several interesting observations from our preliminary molecular analysis of the hyperiids can be made. First, there is a fair amount of agreement between the clades recovered in our molecular results (Fig. 4) and traditional taxonomic groupings of hyperiids based on morphology alone (Vinogradov et al. 1996). For example, several monophyletic lineages composed of classical taxonomic groupings are recovered in our analyses including the infraorder Physosomata within clade 2, and the superfamilies Platysceloidea, Vibilioidea, and Phronimoidea within clade 1 and clade 3, respectively (Fig. 4, Supplementary data).
However, a number of interesting questions regarding hyperiid relationships arise from our analyses. For example, traditional morphological examinations have been unable to confidently place the genus Cystisoma relative to other hyperiid taxa (Vinogradov et al. 1996; Zeidler 2003a). In our analyses, Cystisoma appears to diverge at the base of clade 2 (Fig. 4, Supplementary data) that includes representatives of the classically defined infraorder Physosomata (Vinogradov et al. 1996). Additional sampling of Cystisoma and other members of clade 2 are needed to further explore this inferred relationship. Among the representatives of clade 1, we consistently recovered, albeit with low support, Lycaea nasuta splitting the classically defined family Oxycephalidae (Fig. 4, Supplementary data) suggesting that Oxycephalidae, as currently recognized (Vinogradov et al. 1996), maybe polyphyletic. Further sampling between Lycaea and other members of the Oxycephalidae would be informative with regard to this tentative result. We also consistently recovered Tryphana malmi as a well-supported member of clade 3, whereas morphological analysis alone suggests Tryphana would be affiliated with the Platysceloidea (and thus clade 1 in our analyses). This is a rather surprising result that suggests further analyses of Tryphana relative to members of both clade 1 and clade 3 are warranted.
The presence of significant intraspecific variation in COI has been used among some amphipod groups to discriminate between populations of closely related species (Meyran et al. 1997; Witt et al. 2003; Kaliszewska et al. 2005; Lefebure et al. 2006; Witt et al. 2006). In our analysis we sampled two monotypic genera, Phronimella elongata and Phrosina semilunata, from both the Central Pacific and the Atlantic Gulf Stream. Both are members of hyperiid clade 3 (Fig. 4). We recovered surprisingly long branch-lengths between Pacific and Atlantic isolates. Two alternative scenarios can be considered. The COI sequence divergence could indicate cryptic speciation within these two genera that have historically been considered monotypic. Alternatively, it could represent substantial isolation and divergence of the panoceanic populations of these two species. Further comparisons between populations of these taxa from both the Atlantic and the Pacific oceans, are warranted.
One of the main mechanisms contributing to sequence homoplasy and the loss of phylogenetic signal is site-specific mutational saturation (Philippe et al. 1994; Philippe and Forterre 1999). Our test for mutational saturation among the COI sequences used in this analysis indicates saturation (loss of phylogenetic signal) among the majority of gammarid COI sequences (Fig. 5C). Comparisons between hyperiid COI sequences demonstrate differential evolution of this gene and the retention of phylogeneticaly significant information among these sequences (Fig. 5B, Supplementary data). However, the low signal present among the majority of the gammarid sequences included prevent the recovery of lineage relationships outside of the three well-supported hyperiid clades.
Morphological and molecular homoplasy play a central role in evolutionary biology and represent major impediments to understanding links between organismal genotype and phenotype. Improving support for deep branching nodes within Amphipoda is crucial to hypothesizing credible evolutionary relationships of the hyperiid radiations reported here to other amphipodan faunas. Future analyses should include multiple genetic loci from both mitochondrial and nuclear markers to moderate the effects of differential rates of gene evolution. Increased taxon sampling is also urgently needed to better represent the enormous diversity of extant Amphipoda (currently more than 8000 described species). Hypotheses generated from these additional molecular analyses can, and should be, used to independently evaluate morphological attributes that appear to play important roles in patterns of amphipod diversity but that have proven recalcitrant to traditional morphological based evolutionary analyses due to homoplasy. Given the high branch support observed within hyperiid clades in our analysis, independent contrast methodologies could prove particularly useful in exploring morphological character convergence within and between these lineages (Felsenstein 1985).
For a number of reasons, the expansion of sampled genetic loci and the inclusion of additional taxa should be undertaken to further improve lineage relationships within the hyperiid clades identified here. Pelagic hyperiid amphipods have very sharp differences in the organization of their heads and anterior nervous systems relative to most other benthic amphipods, likely due to constraints imposed by disparate life histories (W.E.B., personal observation). Clearly, associations of hyperiids with gelatinous zooplankton are sufficiently old to have played an important role in the evolution of various hyperiid morphologies, as evidenced by the modifications of appendages used for feeding on/with and for attachment to their preferred hosts (Bowman and Gruner 1973; Vinogradov et al. 1996). Many ecdysozoans (e.g., crustaceans, insects, chelicerates, nematodes, and nematomorphs) have representatives that possess a parasitic life history (Price 1977; Poulin and Morand 2000). Coevolution between hosts and their parasites is well characterized as leading both to strong stabilizing and strong destabilizing selection (Frank 1993). As details of associations between hyperiids and gelatinous zooplankton continue to emerge, it will be important to consider the role that hosts have played during the evolution of various hyperiid lineages. For example, in contrast to the majority of benthic amphipods, the hyperiids exhibit a range of posthatching larval stages that in most cases reflect specific interactions with preferred hosts (Laval 1980). We suggest that the hyperiid amphipods offer a unique intersection of attributes that further our understanding of biological evolution; they are members of an extremely successful clade of metazoans, display a range of parasitic associations, and are well adapted to the largest habitat on the planet.
|
Supplementary data |
|---|
|
Top
SynopsisIntroductionMaterials and methodsResultsDiscussionSupplementary dataAcknowledgmentsReferences |
|---|
Supplementary data are available at ICB online.
|
Acknowledgments |
|---|
|
Top
SynopsisIntroductionMaterials and methodsResultsDiscussionSupplementary dataAcknowledgmentsReferences |
|---|
This work has been supported by the NSF (W.E.B., DBI-0310269; M.Q.M., EF-0334871) and the NAS (W.E.B.). This work has also benefited substantially from access to Smithsonian Institution field stations: the Smithsonian Institution Caribbean Coral Reef Ecosystems (CCRE) Program at Carrie Bowe Caye as contribution #806 (partially supported by the Hunterdon Oceanographic Research Fund), the Smithsonian Tropical Research Institution (STRI) at Bocas del Toro (partially supported by the Hunterdon Oceanographic Research Fund), and the Smithsonian Marine Station at Fort Pierce. We thank the UH Diving Safety Program for providing OC and CCR SCUBA training and support. Specimens critical to this study were contributed by Robert L. Humphreys, Karen Osborn, Nipam H. Patel, Rebecca Scheinberg Hoover, Carsten Wolff, Wolfgang Zeidler, and the pilots of MBARI's; ROVs, Tiburon and Ventana. We thank Casey W. Dunn for computation advice. We also thank the following for discussion and providing critical comments that have significantly improved this communication: Rebeca Gasca, Wolfgang Zeidler, Casey W. Dunn, Andreas Hejnol, David Q. Matus, and Amy Maxmen.
|
Footnotes |
|---|
From the symposium "Integrative Biology of Pelagic Invertebrates" presented at the annual meeting of the Society for Integrative and Comparative Biology, January 3–7, 2007, at Phoenix, Arizona.
|
References |
|---|
|
Top
SynopsisIntroductionMaterials and methodsResultsDiscussionSupplementary dataAcknowledgmentsReferences |
|---|
Avise JC. Mitochondrial DNA and the evolutionary genetics of higher animals. Philos Trans R Soc Lond B (1986) 312::325–42.
Avise JC, Arnold J, Ball RM, Bermingham E, Lamb T, Neigel JE, Reeb CA, Saunders NC. Intraspecific phylogeography: the mitochondrial DNA bridge between population genetics and systematics. Annu Rev Ecol Syst (1987) 18::489–522.[Web of Science]
Barnard JL, Karaman GS. The families and genera of marine Gammaridean Amphipoda (except marine Gammaroids). Rec Aus Mus Suppl (1991) 13::1–866.
Bowman TE, Gruner H-E. The families and genera of Hyperiidea (Crustacea: Amphipoda). Smithsonian Contributions to Zool (1973) 146::1–64.
Browne WE, Schmid BGM, Wimmer EA, Martindale MQ. Expression of otd orthologs in the amphipod crustacean, Parhyale hawaiensis. Dev Gene Evol (2006) 216::581–95.[CrossRef][Web of Science][Medline]
Castresana J, Lübben M, Saraste M, Higgins DG. Evolution of cytochrome oxidase, an enzyme older than atmospheric oxygen. EMBO (1994) 13::2516–25.[Web of Science][Medline]
Childress JJ, Barnes AT, Quetin LB, Robison BH. Thermally protecting cod ends for the recovery of living deep-sea animals. Deep-Sea Res (1978) 25::419–22.[Web of Science]
Childress JJ, Seibel BA. Life at stable low oxygen levels: adaptations of animals to oceanic oxygen minimum layers. J Exp Biol (1998) 201::1223–32.[Abstract]
Cohen JE. Marine and continental food webs: three paradoxes? Philos Trans R Soc Lond B (1994) 343::57–69.
Coleman CO. Comparative anatomy of the alimentary canal of hyperiid amphipods. J Crustacean Biol (1994) 14::346–70.[CrossRef][Web of Science]
Davolos D, Maclean N. Mitochondrial COI-NC-COII sequences in talitrid amphipods (Crustacea). Heredity (2005) 94::81–6.[CrossRef][Web of Science][Medline]
Englisch U, Coleman CO, Wagele JW. First observations on the phylogeny of the families Gammaridae, Crangonyctidae, Melitidae, Niphargidae, Megaluropidae, and Oedicerotidae (Amphipoda, Crustacea), using small subunit rDNA gene sequences. J Nat Hist (2003) 37::2461–86.[CrossRef]
Felsenstein J. Phylogenies and the comparative method. Am Nat (1985) 125::1–15.[CrossRef][Web of Science]
Felsenstein J. PHYLIP – phylogeny inference package (Version 3.2). Cladistics (1989) 5::164–6.
Folmer O, Black M, Hoeh W, Lutz R, Vrijenhoek R. DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Mol Mar Biol Biotech (1994) 3::294–9.[Medline]
Frank SA. Evolution of host-parasite diversity. Evolution (1993) 47::1721–32.[CrossRef][Web of Science]
Gasca R, Haddock SHD. Associations between gelatinous zooplankton and hyperiid amphipods (Crustacea: Peracarida) in the Gulf of California. Hydrobiologia (2004) 530/531::529–35.[CrossRef]
Gasca R, Suarez-Morales E, Haddock SHD. Symbiotic associations between crustaceans and gelatinous zooplankton in deep and surface waters off California. Mar Biol (2007) 151::233–42.[CrossRef]
Haddock SHD, Heine JN. Scientific blue-water diving. (2005) La Jolla: California Sea Grant College Program.
Hamner WM. Underwater observations of blue-water plankton: logistics, techniques, and safety procedures for divers at sea. Limnol Oceanogr (1975) 20::1045–51.
Hamner WM, Madin LP, Alldredge AL, Gilmer RW, Hamner PP. Underwater observations of gelatinous zooplankton: sampling problems, feeding biology, and behavior. Limnol Oceanogr (1975) 20::907–917.
Harbison GR, Biggs DC, Madin LP. The associations of Amphipoda Hyperiidea with gelatinous zooplankton - II. Associations with Cnidaria, Ctenophora, and Radiolaria. Deep-Sea Res (1977) 24::465–88.[Web of Science]
Harbison GR, Madin LP, Swanberg N. On the natural history and distribution of oceanic ctenophores. Deep-Sea Res (1978) 25::233–56.[Web of Science]
Kaliszewska ZA, et al. Population histories of right whales (Cetacea: Eubalaena) inferred from mitochondrial sequence diversities and divergences of their whale lice (Amphipoda: Cyamus). Mol Ecol (2005) 14::3439–56.[CrossRef][Medline]
Laval P. Hyperiid amphipods as crustacean parasitoids associated with gelatinous zooplankton. In: Oceanography and marine biology: an annual review—Barnes M, ed. (1980) Aberdeen: Aberdeen University Press. 11–56.
Lefebure T, Douady CJ, Gouy M, Trontelj P, Briolay J, Gibert J. Phylogeography of a subterranean amphipod reveals cryptic diversity and dynamic evolution in extreme environments. Mol Ecol (2006) 15::1797–806.[CrossRef][Medline]
Lorz A-N, Held C. A preliminary molecular and morphological phylogeny of the Antarctic Epimeriidae and Iphimediidae (Crustacea, Amphipoda). Mol Phylogenet Evol (2004) 31::4–15.[CrossRef][Web of Science][Medline]
Macdonald KS III, Yampolsky L, Duffy JE. Molecular and morphological evolution of the amphipod radiation of Lake Baikal. Mol Phylogenet Evol (2005) 35::323–43.[CrossRef][Web of Science][Medline]
Maddison W, Maddison D. (2005) MacClade. v4.08. Sunderland: Sinauer Associates.
Madin LP, Harbison GR. The associations of Amphipoda Hyperiidea with gelatinous zooplankton - I. Associations with Salpidae. Deep-Sea Res (1977) 24::449–63.[Web of Science]
Martin JW, Davis GE. An updated classification of the recent Crustacea (2001) Los Angeles: Natural History Museum of Los Angeles County. 124.
Meyran J-C, Monnerot M, Taberlet P. Taxonomic status and phylogenetic relationship of some species of the genus Gammarus (Crustacea, Amphipoda) deduced from mitochondrial DNA sequences. Mol Phylogenet Evol (1997) 8::1–10.[Medline]
Myers AA, Lowry JK. A phylogeny and a new classification of the Corophiidea Leach, (1814) (Amphipoda). J Crustacean Biol (2003) 23::443–85.[CrossRef][Web of Science]
Notredame C, Higgins DG, Heringa J. T-Coffee: a novel method for fast and accurate multiple sequence alignment. J Mol Biol (2000) 302::205–17.[CrossRef][Web of Science][Medline]
Page RDM. TREEVIEW: an application to display phylogenetic trees on personal computers. Comput Appl Biosci (1996) 12::357–8.
Philippe H, Forterre P. The rooting of the universal tree of life is not reliable. J Mol Evol (1999) 49::509–23.[CrossRef][Web of Science][Medline]
Philippe H, Sorhannus U, Baroin A, Perasso R, Gasse F, Adoutte A. Comparison of molecular and paleontological data in diatoms suggest a major gap in the fossil record. J Evol Biol (1994) 7::247–65.[CrossRef][Web of Science]
Pirlot JM. Introduction a l’Etude des Amphipodes Hyperides. Annales de L’Institut Oceanographique (1932) 12::1–36.
Poltermann M, Hop H, Falk-Peterson S. Life under Arctic sea ice - reproduction strategies of two sympagic (ice-associated) amphipod species, Gammarus wilkitzkii and Apherusa glacialis. Marine Biol (2000) 136::913–20.[CrossRef]
Posada D, Crandall K. ModelTest: testing the model of DNA substitution. Bioinformatics (1998) 14::817–8.
Poulin R, Morand S. The diversity of parasites. Quart Rev Biol (2000) 75::277–93.[CrossRef][Medline]
Price PW. General concepts on the evolutionary biology of parasites. Evolution (1977) 31::405–20.[CrossRef][Web of Science]
Rambaut A, Drummond A. (2003) Tracer v1.3: www.evolve.zoo.ox.ac.uk.
Robison BH, Reisenbichler KR, Sherlock RE. Giant Larvacean houses: rapid carbon transport to the Deep Sea floor. Science (2005) 308::1609–11.
Ronquist F, Huelsenbeck JP. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics (2003) 19::1572–4.
Schmitz EH. Amphipoda. In: Microscopic anatomy of invertebrates. Chapter 10 (1992) New York: Wiley-Liss, Inc. 443–528.
Schram FR. Crustacea (1986) New York: Oxford University Press.
Schwendinger PJ, Giribet G. The systematics of the south-east Asian genus Fangensis Rambla (Opiliones: Cyphophthalmi: Stylocellidae). Invertebr Syst (2005) 19::297–323.[CrossRef]
Serejo CS. Cladistic revision of the talitroidean amphipods (Crustacea, Gammaridea), with a proposal of a new classification. Zoologica Scripta (2004) 33::551–86.[CrossRef][Web of Science]
Silver MW, Shanks AL, Trent JD. Marine snow: microplankton habitat and source of small-scale patness in pelagic populations. Science (1978) 201::371–3.
Simon C, Frati F, Beckenbach A, Crespi B, Liu H, Flook P. Evolution, weighting, and phylogenetic utility of mitochondrial gene sequences and a compilation of conserved polymerase chain reaction primers. Ann Entomol Soc Am (1994) 87::651–701.[Web of Science]
Simon C, Buckley TR, Frati F, Stewart JB, Beckenbach AT. Incorporating molecular evolution into phylogenetic analysis, and a new compilation of conserved polymerase chain reaction primers for animal mitochondrial DNA. Annu Rev Ecol Syst (2006) 37::545–79.[CrossRef][Web of Science]
Stamatakis A. RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics (2006) 22::2688–90.
Takhteev VV. Trends in the evolution of Baikal amphipods and evolutionary parallels with some marine Malacostracan faunas. Adv Ecol Res (2000) 31::197–219.
Van Dover CL. Deep-sea amphipod swarms. Nature (1992) 358::25–6.[CrossRef]
Vinogradov ME, Volkov AF, Semenova TN. Hyperiid Amphipods (Amphipoda, Hyperiidea) of the world oceans (1996) Washington, DC: Smithsonian Institution Libraries. 1–632.
Warrant EJ, Locket NA. Vision in the deep sea. Biol Rev (2004) 79::671–712.[Medline]
Witt JDS, Blinn DW, Hebert PDN. The recent evolutionary origin of the phenotypically novel amphipod Hyalella montezuma offers an ecological explanation for morphological stasis in a closely allied species complex. Mol Ecol (2003) 12::405–13.[CrossRef][Medline]
Witt JDS, Threloff DL, Hebert PDN. DNA barcoding reveals extraordinary cryptic diversity in an amphipod genus: implications for desert spring conservation. Mol Ecol (2006) 15::3073–82.[Medline]
Zeidler W. Review of the hyperiidean amphipod genus Oxycephalus Milne-Edwards (Crustacea: Amphipoda: Hyperiidea: Oxycephalidae). Invertebr Taxon (1999) 13::391–424.[CrossRef]
Zeidler W. A review of the hyperiidean amphipod family Cystisomatidae Willemoes-Suhm, 1875 (Crustacea: Amphipoda: Hyperiidea). Zootaxa (2003a) 141::1–43.
Zeidler W. A review of the hyperiidean amphipod superfamily Vibilioidea Bowman and Gruner, 1973 (Crustacea: Amphipoda: Hyperiidea). Zootaxa (2003b) 280::1–104.
Zeidler W. A review of the families and genera of the hyperiidean amphipod superfamily Phronimoidea Bowman & Gruner, 1973 (Crustacea: Amphipoda: Hyperiidea). Zootaxa (2004) 567::1–66.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||