Insect Science (2014) 21, 301–312, DOI 10.1111/1744-7917.12113

ORIGINAL ARTICLE

Evolutionary relationships of Pemphigus and allied genera (Hemiptera: Aphididae: Eriosomatinae) and their primary endosymbiont, Buchnera aphidicola Lin Liu1,2 , Xing-Yi Li1,2 , Xiao-Lei Huang1 and Ge-Xia Qiao1 1 Key

Laboratory of Zoological Systematics and Evolution, Institute of Zoology, Chinese Academy of Sciences, and 2 University of Chinese

Academy of Sciences, Beijing, China

Abstract Aphids harbor primary endosymbionts, Buchnera aphidicola, in specialized cells within their body cavities. Aphids and Buchnera have strict mutualistic relationships in nutrition exchange. This ancient association has received much attention from researchers who are interested in endosymbiotic evolution. Previous studies have found parallel phylogenetic relationships between non-galling aphids and Buchnera at lower taxonomic levels (genus, species). To understand whether relatively isolated habitats such as galls have effect on the parallel relationships between aphids and Buchnera, the present paper investigated the phylogenetic relationships of gall aphids from Pemphigus and allied genera, which induce pseudo-galls or galls on Populus spp. (poplar) and Buchnera. The molecular phylogenies inferred from three aphid genes (COI, COII and EF-1α) and two Buchnera genes (gnd, 16S rRNA gene) indicated significant congruence between aphids and Buchnera at generic as well as interspecific levels. Interestingly, both aphid and Buchnera phylogenies supported three main clades corresponding to the galling locations of aphids, namely leaf, the joint of leaf blade and petiole, and branch of the host plant. The results suggest phylogenetic conservatism of gall characters, which indicates gall characters are more strongly affected by aphid phylogeny, rather than host plants. Key words codivergence, gall, parallel evolution, phylogenetic conservatism

Introduction Bacterial symbionts play important roles on the diversification and evolution of many insect groups. Some endosymbionts provide their hosts with essential nutrients, especially insect hosts that feed on plant sap (Moran & Telang, 1998; Baumann et al., 2013). About 10%–20 % of insect species depend on intracellular bacterial mutualists for their viability and reproduction (Douglas, 1989).

Correspondence: Ge-Xia Qiao and Xiao-Lei Huang, Key Laboratory of Zoological Systematics and Evolution, Institute of Zoology, Chinese Academy of Sciences, 1 Beichen West Road, Beijing 100101, China. Tel: +86 10 64807133; fax: +86 10 64807099; email: [email protected]; [email protected]

Such endosymbionts (also called P-endosymbionts) live inside insect specialized cells and undergo maternal transmission to developing eggs or embryos (Buchner, 1965; Moran et al., 2008). With the establishing long-term, mutualistic processes, consistent with this stable transmission, the phylogenies of P-endosymbionts may match those of their insect hosts (Munson et al., 1991a; Moran et al., 1993; Thao et al., 2000; Spaulding & von Dohlen, 2001; Baumann & Baumann, 2005; Rosenblueth et al., 2012). To date, the cospeciation of hosts and symbionts from various environments has been analyzed to disclose their symbiotic history (Aanen et al., 2002; Noda et al., 2007; Liu et al., 2013). Aphids are plant sap-feeding insects that harbor primary endosymbionts, Buchnera aphidicola, in specialized cells within their body cavity. Buchnera can provide their aphid 301

 C 2014

Institute of Zoology, Chinese Academy of Sciences

302

L. Liu et al.

hosts with essential amino acids poorly represented in the plant phloem sap (Moran et al., 2003; Baumann et al., 2013). The mutualistic relationship between aphid and Buchnera is a well-known example of symbiosis. This ancient association, which evolved from about 80–150 million years ago (Moran et al., 1993), has received much attention from researchers who are interested in symbiotic evolution. Considerable phylogenetic congruence observed at deeper evolutionary levels (e.g., distantly related taxa) may not adequately expose the biological processes underlying them (Moran et al., 1993; Baumann et al., 1999; Clark et al., 2000; Funk et al., 2000), because the horizontal transfer of symbionts occurs more easily among closely related hosts (Jousselin et al., 2009). In studies of distantly related taxa, evolutionary congruence may be overestimated when the horizontal transfer could not be detected (Chen et al., 1999; Wernegreen & Riley, 1999). Thus, most studies on aphid–Buchnera relationships have been focused on lower taxonomic levels (Clark et al., 2000; Funk et al., 2000; Wernegreen et al., 2001; Jousselin et al., 2009; Peccoud et al., 2009a, b; Liu et al., 2013). Previous studies usually support parallel phylogenies and cospeciation between aphids and Buchnera. However, previous studies were performed only for several aphid groups, including Uroleucon spp., Brachycaudus spp., Mollitrichosiphum spp. and Acyrthosiphon pisum (Clark et al., 2000; Funk et al., 2000; Jousselin et al., 2009; Peccoud et al., 2009a, b; Liu et al., 2013). Pemphigini belongs to the aphid subfamily Eriosomatinae, together with Eriosomatini and Fordini (Blackman & Eastop, 1994; Remaudi`ere & Remaudi`ere, 1997; Nieto Nafria et al., 1998). The life cycle of Pemphigini is mostly of the holocyclic-heteroecious type, that is, they typically infest Populus as woody primary hosts on which sexual reproduction occurs, as well as herbaceous plants such as Graminaceae as secondary hosts on which asexual parthenogenesis occurs (Ghosh, 1984; Zhang et al., 1999). Most Pemphigini species are gall makers that induce pseudo-galls or galls only on their woody primary hosts (Ghosh, 1984; Zhang et al., 1999; Zhang & Qiao, 2007a). The gall traits are highly species specific and show great variation in shape, size, structure and galling-site. Galls form relatively isolated habitats, provide improved nutrition and better protection against natural enemies to the inducer and its offspring (Cornell, 1983; Price et al., 1986; Price et al., 1987; Inbar et al., 2004; Koyama et al., 2004; Zhang & Qiao, 2007a, b; Miller et al., 2009). Considering that previous studies on aphid–Buchnera relationships mainly sampled non-galling aphid groups, Pemphigini species provide an opportunity to understand whether relatively isolated habitats (i.e., galls) have effect on the  C 2014

parallel relationships between Buchnera and closely related aphid species with the same host plant group (i.e., Populus spp.) and overlapping geographical distributions. In this study, we chose gall-forming aphid species from Pemphigus and its allied genera (Pemphigini) that have different galling locations on poplars and investigated their phylogenetic relationships with Buchnera. The molecular phylogenies were inferred based on three aphid genes (COI, COII and EF-1α) and two Buchnera genes (gnd, 16S rRNA gene), and the congruence between these phylogenies was then tested. The evolutionary relationships among Pemphigus and its allied genera as well as the gall evolution are also discussed. Materials and methods Taxon sampling Forty-one samples representing 15 gall-forming species from Pemphigus and allied genera, which were collected from galls located on leaf, joint of leaf blade and petiole, as well as branches of poplars, were used in this study. We chose two aphid species, namely Phloeomyzus passerinii (Phloeomyzinae) and Chaitophorus populeti (Chaitophorinae), which also feed on poplar, as outgroups for the aphid phylogeny. Escherichia coli, Salmonella enterica and Klebsiella pneumoniae, which are closely related to Buchnera aphidicola (Munson et al., 1991b), were selected as outgroups for the bacteria phylogeny. The aphid samples used in this study are listed in Table S1. All the voucher specimens (individuals from which DNA were extracted, as well as their gall mates) preserved in 95 % or 100 % ethanol were deposited in the National Zoological Museum of China, Institute of Zoology, Chinese Academy of Sciences, Beijing, China. Data collection Total DNA, including Buchnera genome, was extracted from single aphids using the DNeasy Tissue Kit (Qiagen, Germantown, MD, USA). Five gene fragments were amplified, including the Buchnera genes 16S rRNA gene and gluconate-6-phosphate dehydrogenase (gnd) gene, and the aphid mitochondrial cytochrome c oxidase subunits I and II (COI/COII) genes, and the nuclear elongation factor 1α (EF-1α) gene. All polymerase chain reaction (PCR) primers used in this study are listed in Table 1. PCRs contained 3 μL total DNA, 3 μL 10× EasyTaq DNA Polymerase Buffer (+Mg2+ ) (TransGen Biotech, Beijing, China), 2.0 U EasyTaq DNA polymerase (TransGen Biotech), 6.0 mmol/L each deoxynucelotide Institute of Zoology, Chinese Academy of Sciences, 21, 301–312

Phylogenetic congruence between gall aphids and Buchnera

303

Table 1 Primers used for aphid and Buchnera gene amplification and sequencing. Primers (5 –3 )

Gene 16S rRNA gene gnd COI COII EF1α

8–30: AGAGTTTGATCATGGCTCAGATTG 1507–1484: TACCTTGTTACGACTTCACCCCAG BamHI: CGCGGATCCGGWCCWWSWATWATGCCWGGWGG ApaI: CGCGGGCCCGTATGWGCWCCAAAATAATCWCKTTGWGCTTG LepF: ATTCAACCAATCATAAAGATATTGG LepR: TAAACTTCTGGATGTCCAAAAAATCA 2993+: CATTCATATTCAGAATTACC A3772: GAGACCATTACTTGCTTTCAGTCATCT EF3: GAACGTGAACGTGGTATCAC EF2: ATGTGAGCAGTGTGGCAATCCAA

triphosphate (dNTP: TransGen Biotech), 6 pmole each primer, and were brought to a final volume of 30 μL with double-distilled water. PCR conditions were as follows: an initial denaturation at 95 °C for 4–5 min, followed by 35 cycles of 95 °C for 30–60 s; annealing temperatures (depending on the primer sets) for 1 min; extension at 72 °C for 1–2 min; final extension at 72 °C for 7–10 min. The annealing temperatures of each specific primer set were used as follows: 52 °C for COI, 42 °C for COII, 51 °C for EF-1α, 65 °C for 16S rRNA gene, and 55 °C for gnd. PCRs were visualized using 1 % agarose gel electrophoresis and purified using a DNA Fragment Purification kit (Labest, Beijing, China). Some products were cloned into plasmid vectors to obtain sufficient template for DNA sequencing. PCR products and clones were sequenced on an ABI 3730 automated sequencer (Applied Biosystems, Foster City, CA, USA). All sequences acquired in this study have been deposited in GenBank under the accession numbers given in Table S1. Raw sequences were assembled and corrected by SeqmanII (version 5.01, 2001; DNAstar Inc., Madison, WI, USA). All DNA sequences for each fragment were aligned in MEGA 5.05 (Tamura et al., 2011) using default parameters and then adjusted by eye to correct for spurious insertions/deletions. For EF-1α, only coding regions were used in phylogenetic analysis. The introns were removed by using GT-AG rule and by comparing with the complementary DNA (cDNA) sequence of Epipemphigus niisimae (GenBank accession no. DQ499607). Some ambiguous sites from COII, EF-1α, gnd, 16S rRNA gene, which contained many gaps, were removed. The sequences of Buchnera outgroups, Escherichia coli, Salmonella enterica and Klebsiella pneumoniae, were obtained from GenBank (16S rRNA gene, accession no. J01859, DQ153191, X87276; gnd, accession no. U00096, U14495, CP000964).

 C 2014

References van Ham et al. (1997) Clark et al. (1999) Foottit et al. (2008) Stern (1994) Normark (1996) von Dohlen et al. (2002) Palumbi (1996)

Phylogenetic analysis Before combining gene fragments, we used 100 replicates of the partition homogeneity test (PHT) (Farris et al., 1994) as contained in PAUP* 4.0 (Swofford, 2002), to estimate congruence between datasets. The results of PHT showed that the sequence data sets for COI+COII (P = 0.6), COI+COII+EF-1α (P = 0.05) and gnd+16S rRNA gene (P = 0.3) were congruent and could therefore be combined in the following analyses. To assess the relative stability of trees, we used two tree construction methods: maximum likelihoods (ML) and Bayesian. These analyses were conducted for aphid (COI, COII, EF-1α and the combined datasets of COI+COII, COI+COII+EF-1α) and Buchnera (gnd, 16S rRNA gene and gnd+16S rRNA gene) datasets independently. The most appropriate models of evolution were selected independently for each dataset using the AIC (Posada & Buckley, 2004) in jModeltest (Posada, 2008). ML analyses were implemented in PAUP*4.0, and compared with results of RaxML (Stamatakis et al., 2005) and PhyML (Guindon & Gascuel, 2003). ML conducted in PAUP*4.0 under heuristic searches used tree bisection reconnection (TBR) branch-swapping and associated parameters estimated by jModelTest. ML bootstrap values were computed by repeating the same ML heuristic search on 100 pseudo-replicates. PhyML and RAxML were run by applying the optimal substitution model obtained from jModelTest and estimating all model parameters. Branch support was assessed by bootstrap analysis with 1 000 replicates. Bayesian analyses were implemented in MrBayes 3.0 (Larget & Simon, 1999) for the single and combined datasets. In the concatenated analysis the data were partitioned by gene, where each partition had its corresponding model. Substitution model parameters were unlinked among partitions, and rate priors were set to variable to

Institute of Zoology, Chinese Academy of Sciences, 21, 301–312

304

L. Liu et al.

allow differing substitution rates among partitions. For each dataset, Bayesian analyses consisted of two independent runs, each with four chains, run for one million generations sampling every 100 generations. The first 25 % of trees generated were then discarded. After the summarizing consensus tree was assigned, the posterior probabilities were accordingly obtained. Finally, Shimodaira–Hasegawa (SH) tests were used to compare alternative topologies obtained during the various analyses (Shimodaira & Hasegawa, 1999), as implemented in PAUP.

Test of phylogenetic congruence Several methods for testing codivergence hypotheses are available, most of which are reviewed in Paterson and Banks (2001), Stevens (2004) and De Vienne et al. (2013). Two representative methods were used to analyze the hostsymbiont interactions in the present study: TreeMap 1.0 (Page, 1994) and Jane 4.01(Conow et al., 2010) programs. TreeMap 1.0 (Page, 1994) is a tree-based program that reconciles two trees using four types of events (i.e., cospeciation, host-switching, duplication and sorting events) to test whether more cospeciation events are present than would be expected by chance. According to the requirement of TreeMap that each parasite should have a corresponding host but not every host has to have a corresponding parasite, we deleted outgroups from the Buchnera and aphid trees. We input the best pruned ML trees of aphid (COI+COII+EF-1α) and Buchnera (gnd+16S rRNA gene) combined analyses. Both exact search and heuristic search algorithms were implemented to find optimal reconstructions. Randomization tests with the proportional-to-distinguishable search were conducted with 1 000 random Buchnera trees to test whether the two phylogenies contain more cospeciation events than expected by chance. Jane runs faster than TreeMap and is more precise in estimating the cophylogeny events (Keller-Schmidt et al., 2011). As these two programs are using different heuristics models, we complemented the analysis with Jane 4.01 to compare the solution results. According to Charleston’s cost scheme (Charleston, 1998), the event cost used here were zero for a codivergence event, one for duplication and host switching, and two for sorting events. We performed analyses with 500 generations and population size of 100. Significant matching of host and bacterial phylogenies was tested by computing the costs of 100 replicates with random tip mapping and random parasite methods and comparing the resulting costs to the cost of the original associations.  C 2014

Results Phylogenetic relationships among Pemphigus and allied genera The detail attributes of nucleotide sequences are summarized in Table 2. In the analyses of the three single genes, COI, COII and EF-1α, all the obtained topologies grouped aphid samples into clades corresponding to species of the current classification. For each single gene, ML and Bayesian analyses produced similar topologies that were not significantly different using the likelihoodbased SH test. However, in both analyses, branch support was relatively weaker for the relationships higher than generic level. Consequently, the analyses of the individual genes are not presented here. The monophyly of Epipemphigus as well as Thecabius plus Epipemphigus was supported in all analyses (ML/BI>95). However, the monophyly of Pemphigus was not recovered in all the analyses. Pemphigus matsumurai and P. sinobursarius occupied the basal position. The rest of Pemphigus species clustered together and then clustered with Thecabius and Epipemphigus. In the COI, COII trees, P. bursarius was closely related to P. tibetensis, and then clustered with P. borealis. However, the lack of resolution for the relationship of Pemphigus species in EF-1α trees was especially noticeable. The only difference among COI, COII and EF-1α trees was the position of Thecabius, which grouped with Epipemphigus in COI and COII analyses, but formed a weaker clade with some Pemphigus species in the EF-1α tree (ML/BI: 67/0.55). ML and Bayesian analyses for the two combined datasets (COI+COII, COI+COII +EF-1α) showed almost identical topologies for each dataset. Also, the topologies of each combined dataset analyses were highly congruent, and enhanced the phylogenetic resolution when compared to those of COI, COII and EF-1α. As shown by the tree of the COI+COII combined analysis (Fig. S1), Epipemphigus and Thecabius clustered together with relatively high statistical support values. The genus Pemphigus was always polyphyletic. P. matsumurai and P. sinobursarius grouped together and occupied a basal position. The rest of Pemphigus species formed a clade and then grouped with the genera Epipemphigus + Thecabius. The only difference between the two combined data analyses was that P. bursarius clustered with P. tibetensis and then grouped with P. borealis in COI+COII analyses (Fig. S1), but positioned as sister to P. tibetensis + P. borealis in COI+COII+EF-1α analyses (Fig. S2). Finally, the SH test determined that there was no significant difference between the topology of the COI+COII and COI+COII+EF-1α trees (P = 0.215) based on the Institute of Zoology, Chinese Academy of Sciences, 21, 301–312

Phylogenetic congruence between gall aphids and Buchnera

305

Table 2 Characters and evolutionary models of DNA sequences and the datasets used in phylogenetic analyses.

Alignment

aphid

COI COII EF-1α COI+COII COI+COII+EF-1α Buchnera gnd 16S rRNA gnd+16S rRNA

No. of samples

Aligned sequence length (bp)

Variable site

Parsimony informative site

p-distance

Nucleotide composition (T:C:A:G)

AIC best model

36 39 33 30 30 33 32 29

657 704 762 1361 2123 828 1339 2167

135 (20.6 %) 167 (23.7 %) 54 (7.1 %) 270 (19.8 %) 323 (15.2 %) 250 (30.2 %) 67 (5.0 %) 311 (14.4 %)

115 (17.5 %) 120 (17.1 %) 48 (6.3 %) 221 (16.2 %) 269 (12.7 %) 221 (26.7 %) 56 (4.2 %) 277 (12.8 %)

0.064 0.063 0.022 0.063 0.048 0.102 0.016 0.049

41.1:14.6:34.4:10.0 39.9:12.3:40.7:7.2 25.7:21.8:28.0:24.4 40.5:13.4:37.6:8.5 35.2:16.4:34.2:14.2 34.7:11.2:38.2:15.9 21.6:21.3:28.4:28.7 26.6:17.5:32.1:23.8

TIM2+I+G TIM2+G TIM2+G TIM2+I+G GTR+I+G TIM1+G GTR+I+G GTR+I+G

analysis of COI+COII dataset, as well as when COI+COII +EF-1α dataset was used (P = 0.118). Phylogenetic relationships among Buchnera For the phylogenetic analyses of the two individual datasets (gnd and 16S rRNA gene), the branch support values were insufficient to resolve relationships between genera, but major species were highly supported, similar to the results obtained in aphid single gene analyses. We used the ‘Buchnera-aphid species name’ to represent the Buchnera of corresponding aphid species in the present study (Liu et al., 2013). ML and Bayesian analyses for 16S rRNA gene and gnd fragments showed similar topologies, respectively. The monophyly of Buchnera-Epipemphigus and Buchnera-Thecabius was supported robustly in both 16S rRNA gene (Fig. S3) and gnd (Fig. S4) analyses. A conspicuous finding in the 16S rRNA gene and gnd trees is that the members of Buchnera–Pemphigus could not form a monophyletic clade. Two major clades were identified based on 16S rRNA gene analyses: Buchnera– Epipemphigus and Buchnera–Thecabius clustered together, while the second clade consisted of Buchnera–P. borealis, Buchnera–P. tibetensis, Buchnera–P. bursarius, Buchnera–P. sp. 1 and Buchnera–P. sp. 2. The samples of Buchnera–P. matsumurai and Buchnera–P. sinobursarius grouped together and adopted a basal position. However, in the gnd trees, Buchnera–P. matsumurai + Buchnera– P. sinobursarius grouped with Buchnera–Epipemphigus and Buchnera–Thecabius. ML and Bayesian analyses for gnd and 16S rRNA gene combined dataset (Fig. 1) showed identical topologies, and similar to those of 16S rRNA gene. The only topological difference between these trees was that the combined dataset reconstructed Buchnera–Thecabius to be sister to the remaining Buchnera–Pemphigus clade (including Buchnera–P. borealis, Buchnera–P. tibetensis,  C 2014

Buchnera–P. bursarius, Buchnera–P. sp. 1 and Buchnera– P. sp. 2), and then grouped with Buchnera–Epipemphigus. These topological differences were not significant by SH tests (ML vs. Bayesian topology, P = 0.166). Phylogenetic congruence analyses TreeMap found 642 optimal reconstructions with 18 cospeciations, 1–8 duplications, 0–7 host switches and 15–33 sorting events. The probability of obtaining the observed number of cospeciation events cannot be explained as a result of random establishments of host–parasite association (P < 0.01). Results obtained in Jane (Fig. 2) corroborated this finding, which also showed significant levels of cospeciation in the aphid–Buchnera association (P < 0.01, in both permutation tests). Fourteen cospeciations, one duplications, 13 host switches and two loss events were identified. Discussion Phylogenetic congruence between Buchnera and their aphid hosts The phylogenies of Buchnera and galling aphid species from Pemphigus and its allied genera were generally congruent. This confirms and expands previous evidence of parallel evolution between aphids and Buchnera (Clark et al., 2000; Funk et al., 2000; Wernegreen et al., 2001; Jousselin et al., 2009; Peccoud et al., 2009a, b; Liu et al., 2013). The Buchnera phylogeny reflected major features of the aphid phylogeny, including the monophyly of Epipemphigus, and P. borealis + P. tibetensis were the sister group with P. bursarius. These relationships are congruent with those presented by Zhang and Qiao (2007a), who used molecular data (COI and EF-1α) and

Institute of Zoology, Chinese Academy of Sciences, 21, 301–312

Fig. 1 Comparison of phylogenies of aphids (Pemphigus and allied genera) and Buchnera. (A) Aphid phylogeny based on maximum likelihood (ML) analysis of COI/COII/EF1α combined data set. (B) ML phylogenetic tree derived from gnd/16S rRNA gene combined dataset of Buchnera. Values above branches are bootstrap values of maximum likelihood analysis. The bootstrap values higher than 50 are shown. Solid circles indicate cospeciation events. See text for the gall types and gall locations of each of the clades A–C.

306 L. Liu et al.

 C 2014

Institute of Zoology, Chinese Academy of Sciences, 21, 301–312

Fig. 2 Cophylogeny of aphid and Buchnera from Jane 4.01 (Conow et al., 2010). The reconciled trees based on the COI/COII/EF-1α combined dataset of aphid and gnd/16S rRNA gene of Buchnera are shown here. Blue and black lines indicate the phylogenies of the Buchnera and aphids, respectively. Hollow red circles indicate cospeciation events; yellow or red solid circles indicate duplications; arrows indicate host switch events; dashed line indicates sorting events. Support values were shown near each event.

Phylogenetic congruence between gall aphids and Buchnera

 C 2014

Institute of Zoology, Chinese Academy of Sciences, 21, 301–312

307

308

L. Liu et al.

more species to infer the phylogenetic relationships within Pemphigini. However, the genus Pemphigus appeared as polyphyletic in all analyses based on Buchnera and aphid sequences. P. matsumurai + P. sinobursarius were placed at a distant position with the other Pemphigus species. The slight incongruence between analyses based on Buchnera and aphid combined datasets were concentrated at the genera Epipemphigus and Thecabius. Although the sister relationship of Epipemphigus and Thecabius indicated in the aphid-based topologies was not recovered in the Buchnera tree based on combined datasets, it was really supported by the analyses based on 16S rRNA gene. However, relationships in the Buchnera tree based on combined data are consistent with those obtained in Zhang and Qiao (2007a). In addition, both TreeMap and Jane analyses detected a significant pattern of cophylogeny between the aphid lineages and their Buchnera symbionts. These results support the view that topological congruence between aphid and Buchnera trees are not due to chance alone. The strong cospeciation signal detected between Buchnera and aphid species from Pemphigus and allied genera could be indicative of a specialized interaction. Buchnera could follow the radiation of aphids and diversified in parallel with their hosts. Although Jane analysis (Fig. 2) evidenced 13 host switches, 11 of them occurred at the intraspecies level. As the obtained aphid species reproduced asexually, and they are living in the gall which rules out the influence of parasitic wasps, we infer that the horizontal transfer cannot happen at the intraspecies level. Two other host switches were the Buchnera with ancestor nodes of P. matsumurai + P. sinobursarius, Epipemphigus + Thecabius, Epipemphigus + Thecabius and that of the rest of the Pemphiginus species. However, the support values of these two switching events are quite low (P < 50 %). In addition, if the host switches occurred at the ancestor period, the aphid–Buchnera phylogenies should show significant incongruence. Therefore, as Clark et al. (2000) and Jousselin et al. (2009) mentioned, we suggest that the slight conflict between host and symbiont trees in our study may be due to potential influence of methodological artifacts, such as the inadequacy of the models of evolution or limited taxon sampling, as well as lack of adequate signal for certain nodes.

The effect of isolated habitats on aphid–Buchnera relationship Ecological factors are important to the associated organisms (Guay et al., 2009; Anbutsu & Fukatsu, 2011). Ecological factors such as feeding sites and living habitats of aphid hosts may have effect on the aphid–Buchnera  C 2014

relationship. For instance, if the aphid hosts have frequent ecological and behavioral contact and overall biological similarity, the potential for horizontal transmission could be maximized (Funk et al., 2000). Moreover, different parts of the plant have different nutrition-distribution patterns (Pich et al., 1994; Ma et al., 2002). Some previous studies suggested that the galls located more closely to the plant vascular system may absorb more nutrition (Nyman et al., 1998, 2000; Zhang & Qiao, 2007a, b). Some researchers also found that photosynthesis rate of galls were higher than those of non-gall tissues of the same organ (Fay et al., 1993; Huang et al., 2011), which indicates that gall tissue is more nutritious than nongall tissue (Koyama et al., 2004). Therefore, non-galling aphids and galling aphids with relatively isolated habitats may have experienced different selection pressures and have different phylogenetic relationships with their nutritional partner Buchnera. However, the present study indicates obvious phylogenetic congruence between gallforming aphids and Buchnera. This rules out the regular occurrence of horizontal transfer of Buchnera between closely related aphid species with the same host plants and overlapping geographical distributions. In addition, it appears that different ecological niches and relatively isolated habitats may have no obvious effect on the evolutionary relationships of aphids and Buchnera.

The phylogenetic conservatism of gall characters Three major clades (Fig. S2) were identified, which correspond to gall types and gall locations. The clade C, including P. matsumurai and P. sinobursarius, was placed at the basal position. These two species induce true galls on the joint of leaf blade and petiole with tissue lignification and smaller cubage inside (Zhang & Zhong, 1979; Zhang & Qiao, 2007a). Clade B included five species of Pemphigus. Most of them can induce true-galls with thick walls on the twig (Zhang & Zhong, 1979; Zhang & Qiao, 2007a). Clade A includes Thecabius and Epipemphigus, which often form pseudo-galls on leaves of Populus which shapes as a rolling leaf or appears as slight tissue lignification and has a primary exit (Zhang & Zhong, 1979; Zhang et al., 1995; Zhang & Qiao, 2007a). The present phylogeny showed the conservatism of gall characters on aphid phylogeny, which supported the hypothesis that gall characters are more affected by aphids rather than their host plants (Stern, 1995; Inbar et al., 2004; Sano & Akimoto, 2011). The association between gall morphology and gall inducers’ phylogeny is similar to that described in thrips (Crespi & Worobey, 1998), wasps (Stone & Cook, 1998) and sawflies (Nyman et al., 2000). Institute of Zoology, Chinese Academy of Sciences, 21, 301–312

Phylogenetic congruence between gall aphids and Buchnera

Some previous studies on the potential mechanisms of gall formation demonstrated that insect genotype, rather than plant genotype, determine the variation in gall formation (Crego et al., 1990; Dodson, 1991; H¨oglund et al., 2012). In addition, the selection of galling-site, which is determined by the fundatrix of aphid species, was confirmed by previous researches (Inbar & Wool, 1995; Inbar, 1998; Wool, 2004). These evidences indicate that insect gall form is determined primarily by the insects, and gall characters represent extended insect phenotypes (Stern, 1995; Stone & Cook, 1998; Stone & Sch¨onrogge, 2003; Chen & Qiao, 2012). However, to clarify the evolutionary trend of gall characters in Pemphigini on Populus, more taxa need to be investigated in the future. Acknowledgments We are grateful to David Voegtlin, Dong Zhang, Jian-Feng Wang, Guo-Dong Ren, Kun Guo, Li-Yun Jiang, XiaoMei Su and Tong Lei for collecting specimens; to Fen-Di Yang for making aphid slides. Dr. Masakazu Sano and two anonymous reviewers provided valuable suggestions on previous versions of the manuscript. The work was supported by National Natural Sciences Foundation of China (31272348), National Science Funds for Distinguished Young Scientists (No. 31025024), National Science Fund for Fostering Talents in Basic Research (No. J1210002), a grant from the Ministry of Science and Technology of the People’s Republic of China (No. 2011FY120200) and grants from the Key Laboratory of the Zoological Systematics and Evolution of the Chinese Academy of Sciences (No. Y229YX5105, O529YX5105). Disclosure Each author of this study declares that there is no conflict of interests in this work. References Aanen, D.K., Eggleton, P., Rouland-Lef`evre, C., GuldbergFrøslevl, T., Rosendahll, S. and Boomsma, J.J. (2002) The evolution of fungus-growing termites and their mutualistic fungal symbionts. Proceedings of the National Academy of Sciences of the United States of America, 99, 14887–14892. Anbutsu, H. and Fukatsu, T. (2011) Spiroplasma as a model insect endosymbiont. Environmental Microbiology Reports, 3, 144–153. Baumann, L. and Baumann, P. (2005) Cospeciation between the primary endosymbionts of mealybugs and their hosts. Current Microbiology, 50, 84–87.  C 2014

309

Baumann, L., Baumann, P., Moran, N.A., Sandstrom, J. and Thao, M.L. (1999) Genetic characterization of plasmids containing genes encoding enzymes of leucine biosynthesis in endosymbionts (Buchnera) of aphids. Journal of Molecular Evolution, 48, 77–85. Baumann, P., Moran, N. and Baumann, L. (2013) Bacteriocyteassociated endosymbionts of insects. The Prokaryotes (ed. M. Dworkin), pp. 465–496. Springer-Verlag, New York. Blackman, R.L. and Eastop, V.F. (1994) Aphids on the World’s Trees: An Identification and Information Guide. CAB International, Wallingford. 1024 pp. Buchner, P. (1965) Endosymbiosis of Animals with Plant Microorganisms. Interscience Publishers, Inc., New York. 909 pp. Charleston, M.A. (1998) Jungles: A new solution to the host/parasite phylogeny reconciliation problem. Mathematical Biosciences, 149, 191–223. Crego, C.L., Weis, A.E., Polans, O. and Bretz, C.K. (1990) Sympatric sibling species from three phenotypically distinct Asteromyia (Diptera: Cecidomyiidae) galls on the same host plant species. Environmental Entomology, 83, 149–154. Chen, X., Li, S. and Aksoy, S. (1999) Concordant evolution of a symbiont with its host insect species: Molecular phylogeny of genus Glossina and its bacteriome-associated endosymbiont, Wigglesworthia glossinidia. Journal of Molecular Evolution, 48, 49–58. Chen, J. and Qiao, G.X. (2012) Galling aphids (Hemiptera: Aphidoidea) in China: Diversity and host specificity. Psyche, 2012: 621934. Clark, M.A., Moran, N.A. and Baumann, P. (1999) Sequence evolution in bacterial endosymbionts having extreme base compositions. Molecular Biology and Evolution, 16, 1486– 1598. Clark, M.A., Moran, N.A., Baumann, P. and Wernegreen, J.J. (2000) Cospeciation between bacterial endosymbionts (Buchnera) and a recent radiation of aphids (Uroleucon) and pitfalls of testing for phylogenetic congruence. Evolution, 54, 517– 525. Cornell, H.V. (1983) The secondary chemistry and complex morphology of galls formed by the Cynipinae (Hymenoptera): Why and how? American Midland Naturalist, 110, 225– 234. Conow, C., Fielder, D., Ovadia, Y. and Libeskind-Hadas, R. (2010) Jane: A new tool for the cophylogeny reconstruction problem. Algorithms for Molecular Biology, 5, 16–25. Crespi, B. and Worobey, M. (1998) Comparative analysis of gall morphology in Australian gall thrips: The evolution of extended phenotypes. Evolution, 52, 1686–1696. De Vienne, D.M., Refr´egier, G., L´opez-Villavicencio, M., Tellier, A., Hood, M.E. and Giraud, T. (2013) Cospeciation vs host-shift speciation: Methods for testing, evidence from natural associations and relation to coevolution. New Phytologist, 198, 347–385.

Institute of Zoology, Chinese Academy of Sciences, 21, 301–312

310

L. Liu et al.

Dodson, G.N. (1991) Control of gall morphology: Tephritid gallformers (Aciurina spp.) on rabbitbrush (Chrysothamnus). Ecological Entomology, 16, 177–182. Douglas, A.E. (1989) Mycetocyte symbiosis in insects. Biological Reviews, 64, 409–434. Farris, J.S., Kallersjo, M., Kluge, A.G. and Bult, C. (1994) Testing significance of congruence. Cladistics, 10, 315–319. Fay, P.A., Hartnett, D.C. and Knapp, A.K. (1993) Increased photosynthesis and water potentials in Silphium intrgrifolium galled by cynipid wasps. Oecologia, 93, 114–120. Foottit, R.G., Maw, H.E.L., von Dohlen, C.D. and Herbert, P.D.N. (2008) Species identification of aphids (Insecta: Hemiptera: Aphididae) through DNA barcodes. Molecular Ecology Resources, 8, 1189–1201. Funk, D.J., Helbling, L., Wernegreen, J.J. and Moran, N.A. (2000) Intraspecific phylogenetic congruence among multiple symbiont genomes. Proceedings of the Royal Society B-Biological Science, 267, 2517–2521. Ghosh, A.K. (1984) Fauna of India, Homoptera, Aphidoidea, Subfamily Pemphiginae. The Technical and General Press, Calcutta. Guay, J.F., Boudreault, S., Michaud, D. and Cloutier, C. (2009) Impact of environmental stress on aphid clonal resistance to parasitoids: Role of Hamiltonella defensa bacterial symbiosis in association with a new facultative symbiont of the pea aphid. Journal of Insect Physiology, 55, 919–926. Guindon, S. and Gascuel, O. (2003) A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology, 52, 696–704. H¨oglund, S., R¨onnberg-W¨astljung, A.C., Lagercrantz, U. and Larsson, S. (2012) A rare major plant QTL determines nonresponsiveness to gall-forming insect in willow. Tree Genetic and Genomes, 8, 1051–1060. Huang, M.Y., Lin, K.H., Yang, M.M., Chou, H.M., Yang, C.M. and Chang, Y.T. (2011) Chlorophyll fluorescence, spectral properties, and pigment composition of galls on leaves of Machilus thunbergii. International Journal of Plant Sciences, 172, 323–329. Inbar, M. (1998) Competition, territoriality and maternal defense in a gall-forming aphid. Ethology Ecology and Evolution, 10, 159–170. Inbar, M. and Wool, D. (1995) Phloem-feeding specialists sharing a host tree: Resource partitioning minimizes interference competition among galling aphid species. Oikos, 73, 109– 119. Inbar, M., Wink, M. and Wool, D. (2004) The evolution of host plant manipulation by insects: Molecular and ecological evidence from gall-forming aphids on Pistacia. Molecular Phylogenetics and Evolution, 32, 504–511. Jousselin, E., Desdevised, Y. and Coeur d’acier, A. (2009) Fine-scale cospeciation between Brachycaudus and Buchnera aphidicola: Bacterial genome helps define species and evo-

 C 2014

lutionary relationships in aphids. Proceedings of the Royal Society, Series B: Biological Science, 276, 187–196. Keller-Schmidt, S., Wieseke, N., Klemm, K. and Middendorf, M. (2011) Evaluation of host parasite reconciliation methods using a new approach for cophylogeny generation. Working paper from Bioinformatics Leipzig. Available at: http://www.bioinf.uni-leipzig.de/working/11-013. Koyama Y., Yao, I. and Akimoto, S. (2004) Aphid galls accumulate high concentrations of amino acids: A support for the nutrition hypothesis for gall formation. Entomologia Experimentalis et Applicata, 113, 35–44. Larget, B. and Simon, D. (1999) Markov chain Monte Carlo algorithms for the Bayesian analysis of phylogenetic trees. Molecular Biology and Evolution, 16, 750–759. Liu, L., Huang, X.L., Zhang, R.L., Jiang, L.Y. and Qiao, G.X. (2013) Phylogenetic congruence between Mollitrichosiphum (Aphididae: Greenideinae) and Buchnera indicates insect– bacteria parallel evolution. Systematic Entomology, 38, 81–92. Ma, S.B., Yang, G.Y. and Zhao, N.X. (2002) Nutrition distribution pattern in different organs of Sinopodophyllum hexandrum (Royle) Ying and its ecological significance. Chinese Journal of Ecology, 21, 65–67. Miller, D.G., Ivey, C.T. and Shedd, J.D. (2009) Support for the microenvironment hypothesis for adaptive value of gall induction in the California gall wasp, Andricus quercuscalifornicus. Entomologia Experimentalis et Applicata, 132, 126–133. Moran, N.A. and Telang, A. (1998) Bacteriocyte-associated symbionts of insects: A variety of insect groups harbor ancient prokaryotic endosymbionts. Bioscience, 48, 295–304. Moran, N.A., McCutcheon, J.P. and Nakabachi, A. (2008) Genomics and evolution of heritable bacterial symbionts. Annual Review of Genetics, 42, 165–190. Moran, N.A., Munson, M.A., Baumann, P. and Ishikawa, H. (1993) A molecular clock in endosymbiotic bacteria is calibrated using the insect hosts. Proceedings of the Royal Society B-Biological Science, 253, 167–171. Moran, N.A., Plague, G.R., Sandstr¨om, J.P. and Wilcox, J.L. (2003) A genomic perspective on nutrients provisioning by bacterial symbionts of insects. Proceedings of the National Academy of Sciences of the United States of America, 100, 9388–9393. Munson, M.A., Baumann, P., Clark, M.A., Baumann, L., Moran, N.A., Voegtlin, D.J. and Campbell, B.C. (1991a) Evidence for the establishment of aphid–Eubacterium endosymbiosis in an ancestor of four aphid families. Journal of Bacteriology, 173, 6321–6324. Munson, M.A., Baumann, P. and Kinsey, M.G. (1991b) Buchnera gen. nov. and Buchnera aphidicola sp. nov., a taxon consisting of the mycetocyte-associated, primary endosymbionts of aphids. International Journal of Systematic and Evolutionary Microbiology, 41, 566–568.

Institute of Zoology, Chinese Academy of Sciences, 21, 301–312

Phylogenetic congruence between gall aphids and Buchnera Nieto Nafria, J.M., Mier Durante, M.P. and Remaudi`ere, G. (1998) Les noms des taxa du groupe-famille chez les Aphididae. Revue Francxaise d’Entomologie (N.S.), 19, 77– 92. Noda, S., Kitade, O., Inoue, T., Kawai, M., Kanuka, M., Hiroshima, K., Hongoh, Y., Constantino, R., Uys, V., Zhong, J., Kudo, T. and Ohkuma, M. (2007) Cospeciation in the triplex symbiosis of termite gut protists (Pseudotrichonympha spp.), their hosts, and their bacterial endosymbionts. Molecular Ecology, 16, 1257–1266. Normark, B.B. (1996) Phylogeny and evolution of parthenogenetic weevils of the Aramigus tessellatus species complex (Coleoptera: Curculionidae: Naupactini): Evidence from mitochondrial DNA sequences. Evolution, 50, 734–745. Nyman, T., Roininen, H. and Vuorinen, J.A. (1998) Evolution of different gall types in willow-feeding sawflies (Hymenoptera: Tenthredinidae). Evolution, 52, 465–474. Nyman, T., Widmer, A. and Roininen, H. (2000) Evolution of gall morphology and host plant relationships in willowfeeding sawflies (Hymenoptera: Tenthredinidae). Evolution, 54, 526–533. Page, R.D. (1994) Parallel phylogenies: Reconstructing the history of host–parasite assemblages. Cladistics, 10, 155–173. Palumbi, S.R. (1996) Nucleic acids II: The polymerase chain reaction. Molecular Systematics (ed. D.M. Hillis), pp. 205– 247. Sinauer Press, Sunderland, UK. Paterson, A.M. and Banks, J. (2001) Analytical approaches to measuring cospeciation of host and parasites: Through a glass, darkly. International Journal for Parasitology, 31, 1012–1022. Peccoud, J., Ollivier, A., Plantegenest, M. and Simon, J.-C. (2009a) A continuum of genetic divergence from sympatric host races to species in the pea aphid complex. Proceedings of the National Academy of Sciences of the United States of America, 106, 7495–7500. Peccoud, J., Simon, J.-C., McLaughlin, H.J. and Moran, N.A. (2009b) Post-Pleistocene radiation of the pea aphid complex revealed by rapidly evolving endosymbionts. Proceedings of the National Academy of Sciences of the United States of America, 106, 16315–16320. Pich, A., Scholz, G. and Stephan, U.W. (1994) Iron-dependent changes of heavy metals, nicotianamine, and citrate in different plant organs and in the xylem exudate of two tomato genotypes. Nicotianamine as possible copper translocator. Plant and Soil, 165(2), 189–196. Posada, D. (2008) jModelTest: Phylogenetic model averaging. Molecular Biology and Evolution, 25, 1253–1256. Posada, D. and Buckley, T.R. (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.

 C 2014

311

Price, P.W., Fernandes, G.W. and Waring, G.L. (1987) Adaptive nature of insect galls. Environmental Entomology, 16, 15–24. Price, P.W., Waring, G.L. and Fernandes, G.W. (1986) Hypotheses on the adaptive nature of galls. Proceedings of the Entomological Society of Washington, 88, 361–363. Rosenblueth, M., Sayavedra, L., S´amano-S´anchez, H., Roth, A. and Mart´ınez-Romero, E. (2012) Evolutionary relationships of flavobacterial and enterobacterial endosymbionts with their scale insect hosts (Hemiptera: Coccoidea). Journal of Evolutionary Biology, 25, 2357–2368. Remaudi`ere, G. and Remaudi`ere, M. (1997) Catalogue des Aphididae du Monde: Homoptera Aphidoidea. Institut National de la Recherche Agronomique, Paris. Sano, M. and Akimoto, S. (2011) Morphological phylogeny of gall-forming aphids of the tribe Eriosomatini (Aphididae: Eriosomatinae). Systematic Entomology, 36, 607–627. Shimodaira, H. and Hasegawa, M. (1999) Multiple comparisons of log-likelihoods with applications to phylogenetic inference. Molecular Biological and Evolution, 16, 1114–1116. Spaulding, A.W. and von Dohlen, C.D. (2001) Psyllid endosymbionts exhibit patterns of co-speciation with hosts and destabilizing substitutions in ribosomal RNA. Insect Molecular Biology, 10, 57–67. Stamatakis, A., Ludwig, T. and Meier, H. (2005) RAxML-III: A fast program for maximum likelihood-based inference of large phylogenetic trees. Bioinformatics, 21, 456–463. Stern, D.L. (1994) A phylogenetic analysis of soldier evolution in the aphid family Hormaphididae. Proceedings of the Royal Society B-Biological Science, 256, 203–209. Stern, D.L. (1995) Phylogenetic evidence that aphids, rather than plants, determine gall morphology. Proceedings of the Royal Society B-Biological Science, 260, 85–89. Stevens, J. (2004) Computational aspects of host–parasite phylogenies. Briefings in Bioinformatics, 54, 339–349. Stone, G.N. and Cook, J.M. (1998) The structure of cynipid oak galls: Patterns in the evolution of an extended phenotype. Proceedings of the Royal Society B-Biological Science, 265, 979–988. Stone, G.N. and Sch¨onrogge, K. (2003) The adaptive significance of insect gall morphology. Trends in Ecology and Evolution, 18, 512–522. Swofford, D. (2002) PAUP*: Phylogenetic analysis using parsimony (*and other methods), version 4.0b.10. Sinauer Associates, Sunderland, Massachusetts. Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M. and Kumar, S. (2011) MEGA5: Molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Molecular Biology and Evolution, 28, 2731–2739. Thao, M.L., Moran, N.A., Abbot, P., Brennan, E.B., Burckhardt, D.H. and Baumann, P. (2000) Cospeciation of psyllids and

Institute of Zoology, Chinese Academy of Sciences, 21, 301–312

312

L. Liu et al.

their primary prokaryotic endosymbionts. Applied and Environmental Microbiology, 66, 2898–2905. van Ham, R.J., Moya, A. and Latorre, A. (1997) Putative evolutionary origin of plasmids carrying the genes involved in leucine biosynthesis in Buchnera aphidicola (endosymbiont of aphids). Journal of Bacteriology, 179, 4768– 4777. von Dohlen, C.D., Kurosu, U. and Aoki, S. (2002) Phylogenetics and evolution of the eastern Asian–eastern North American disjunct aphid tribe, Hormaphidini (Hemiptera: Aphididae). Molecular Phylogenetics and Evolution, 23, 257–267. Wernegreen, J.J. and Riley, M.A. (1999) Comparison of the evolutionary dynamics of symbiotic and housekeeping loci: A case for the genetic coherence of rhizobial lineages. Molecular Phylogenetics and Evolution, 16, 98–113. Wernegreen, J.J., Richardson, A.O. and Moran, N.A. (2001) Parallel acceleration of evolutionary rates in symbiont genes underlying host nutrition. Molecular Phylogenetics and Evolution, 19, 479–485. Wool, D. (2004) Galling aphids: Specialization, biological complexity, and variation. Annual Review of Entomology, 49, 175– 192. Zhang, G.X., Qiao, G.X., Zhong, T.S. and Zhang, W.Y. (1999) Fauna Sinica Insecta, Vol. 14. Homoptera: Mindaridae and Pemphigidae. Science Press, Beijing. Zhang, H.C. and Qiao, G.X. (2007a) Systematic status of the genus Formosaphis Takahashi and the evolution of galls based on the molecular phylogeny of Pemphigini (Hemiptera: Aphididae: Eriosomatinae). Systematic Entomology, 32, 690– 699. Zhang, H.C. and Qiao, G.X. (2007b) Molecular phylogeny of Fordini (Hemiptera: Aphididae: Pemphiginae) inferred from nuclear gene EF-1α and mitochondrial gene COI. Bulletin of Entomological Research, 97, 379–386. Zhang, G.X. and Zhong, T.S. (1979) Five new species of Pemphigus Hartig and a new species of Sanpupemphigus gen. nov. from China (Aphidoidea: Pemphigidae). Acta Entomologica Sinica, 22, 324–332. Zhang, G.X., Zhong, T.S. and Qiao, G.X. (1995) A study on Chinese Thecabius Koch 1857, with description of new sub-

 C 2014

genus, new species and new subspecies. Entomologia Sinica, 2, 206–224.

Accepted January 26, 2014

Supporting Information Additional Supporting Information may be found in the online version of this article at the publisher’s web-site: Table S1 Collection information for aphid samples used in this study and sequence GenBank accession numbers. Fig. S1 ML tree obtained from the analysis of the mitochondrial combined dataset (COI/COII). Bootstrap values from maximum likelihood analysis (50 % and greater) are shown above the branches, and the posterior probabilities of Bayesian analysis (BI) (50 % and greater) are shown below. Fig. S2 Aphid phylogenetic tree based on maximum likelihood analysis of the combined dataset (COI/COII/EF-1α). Bootstrap values from maximum likelihood analysis (50 % and greater) are shown above the branches, and the posterior probabilities of Bayesian analysis (BI) (50 % and greater) are shown below. See text for the gall types and gall locations of each of the clade A–C. Fig. S3 Bayesian phylogenetic tree of Buchnera reconstructed from 16S rRNA gene sequences. Buchnera samples are indicated as names of their aphid hosts. Values above and below branches are bootstrap values of maximum likelihood analysis and posterior probabilities of Bayesian inference. Only nodes supported by 50 % or greater are shown. A dash (-) marks the bootstrap values of those nodes that were lower than 50 %. Fig. S4 Bayesian phylogenetic tree of Buchnera reconstructed from gnd sequences. Buchnera samples are indicated as names of their aphid hosts. Values above and below branches are bootstrap values of maximum likelihood analysis and posterior probablities of Bayesian inference. Only nodes supported by 50 % or greater are shown.

Institute of Zoology, Chinese Academy of Sciences, 21, 301–312