INTRODUCTION

The agamid genus Gonocephalus Kaup, 1825, originally considered as consisting of 29 species distributed in Southeast Asia, Australia, New Guinea and the Solomons (Wermuth, 1967), was highlighted as a zoogeographical exception because of its occurrence on both sides of the Wallacia, a border of the Oriental and Australian faunal realms (e.g., Darlington, 1957). In his unpublished dissertation, Moody (1980), on the basis of cladistic analyses using morphological characters, argued that species of the genus Gonocephalus from Australia, New Guinea and the Solomons were not actually closely related to the Oriental congeners. He thus re-assigned the former to the resurrected genus Hypsilurus Peters, 1867. Such an arrangement, published by Welch et al. (1990), was favored by recent immunological (Baverstock and Donnellan, 1990; King, 1990), karyological (Ota et al., 1992), electron-microscopic (Ananjeva and Matveyeva-Dujsebayava, 1996), and molecular studies (Honda et al., 2000; Macey et al., 2000).

Recent karyological studies further revealed prominent chromosomal variation within Gonocephalus sensu stricto: four of the five species examined shared a 2n = 42 karyo-type including 22 biarmed macrochromosomes, whereas the remainder, G. robinsonii, had a 2n = 32 karyotype including no more than 12 macrochromosomes (Ota et al., 1992; Diong et al., 2000). Based on the fact that the former arrangement is exceptional as of the agamid karyotype, Diong et al. (2000) interpreted it as a synapomorph of Gonocephalus exclusive of G. robinsonii, and went so far as to argue for the possible paraphyletic nature of the genus. They, on the basis of similarities in the number of microchromosomes between G. robinsonii and a few Australian agamids, also noted that G. robinsonii may be an Asiatic representative of the Australian agamid radiation, like Physignathus cocincinus (Honda et al., 2000; Macey et al., 2000). However, these hypotheses seriously suffer due to the lack of appropriate phylogenetic analysis of chromo-some characters and of comparable karyological data for many other related genera.

In the present study, we examined the phylogenetic relationships of the two karyomorphs of Gonocephalus and other agamid genera by analyzing mitochondrial DNA sequence data. Our purposes are to examine consistency between patterns in chromosomal and phyletic diversification within the Gonocephalus, and to test the closer affinity of G. robinsonii to the Australian agamids as hypothesized by Diong et al. (2000).

MATERIALS AND METHODS

Three species of the Gonocephalus sensu stricto, G. chamaeleontinus from West Malaysia and Indonesia, G. miotympanum from Borneo and G. robinsonii from West Malaysia (Welch et al., 1990; Manthey and Grossmann, 1997), were newly examined in the present study (see Appendix for locality data of specimens examined). We incorporated published data representing the six major agamid groups of Moody (1980) (Groups I–VI), the Chamaeleonidae (Bradypodion fischeri), the Polychridae (Anolis carolinensis), and the Iguanidae (Iguana iguana) (see Appendix for accession numbers). Of these, the last three species were regarded as out-groups, because these families are considered to be basal to the Agamidae (Frost and Etheridge, 1989; Macey et al., 1997; Honda et al., 2000)

Extraction, amplification and sequencing of DNA are described in detail elsewhere (Honda et al., 1999a, b). A part of mitochondrial 12S and 16S rRNA genes consisting of approximately 860 base pairs (bp) were amplified using the polymerase chain reaction (PCR) with primers L1091, H1478, L2606, and H3056 (Kocher et al., 1989; Hedges et al., 1993). Alignments for DNA sequences were unambiguously determined based on maximum nucleotide similarity following Honda et al. (2000).

The neighbor-joining (NJ) method (Saitou and Nei, 1987) was applied to infer relationships among taxa on the basis of a pairwise matrix of the distance from Kimura's (1980) two-parameter model, using CLUSTAL X 1.8 (Thompson et al., 1994). The maximum-likelihood (ML, empirical base frequencies and equal rate substitution model) and maximum-parsimony (MP, no bias between transition and transversion) analyses were also conducted using heuristic search option of PAUP* 4.0b (Swofford, 1998). In these three analyses, gap sites were excluded, and confidences of branched were assessed by bootstrap resamplings (Felsenstein, 1985).

RESULTS

The amplified fragment of the 12S and 16S rRNA genes consisted of 862 total sites. Of these, however, 201 sites involved in the insertion or deletion were excluded from phylogenetic analyses. The NJ dendrogram derived from aligned sequences is shown in Fig. 1A. The monophyly of the Agamidae was supported in all bootstrap iterations (node 1: 100%). The ingroup portion of this dendrogram was divided into two major lineages. One of the major clusters (node 2: 83%) further split into two subclusters (nodes 4, 5), of which node 4 (80%) consisted of West Asian and North African primitive agamids (Leiolepis and Uromastyx: Group I sensu Moody, 1980), whereas node 5 (84%) accommodated Australian genera including Hypsilurus that had been previously assigned to Gonocephalus (Groups II–IV complex). The other major cluster (node 3: 71%) split into two subclusters (nodes 6, 7), of which node 6 (100%) consisted of West Asian and African derived agamids (Agama and Phrynocephalus: Group VI). Node 7 (100%), on the other hand, contained Southeast Asian derived agamids (Acanthosaura, Aphaniotis, Calotes, Draco, Gonocephalus, Japalura, Phoxophrys and Ptyctolaemus: Group V). Within this node, G. chamaeleontinus, G. grandis and G. miotympanum (node 8: 100%), and G. robinsonii and J. polygonata (node 9: 70%) constituted exclusive clusters, respectively.

Fig. 1

(A) Neighbor-joining (NJ) dendrogram derived from distance matrix from 12S and 16S rRNA sequence data. Numbers beneath branches are BPs at least 50% of the 1,000 bootstrap replications. Nodes with bold numbers indicate relationships referred to in the text. Bar equals 0.1 unit of Kimura's two-parameter distance. Published data for the diploid chromosome numbers (2n) are given in parentheses (DeSmet, 1981; Witten, 1983; Solleder and Schmid, 1988; Ota et al., 1992; Ota and Hikida, 1989; Diong et al., 2000; Ota, unpubl data). Although Groups II–IV complex is mainly distributed in Australasia, Physignathus cocincinus is occurs in the continental part of Southeast Asia. (B) Maximum-likelihood (ML) dendrogram (ln likelihood=−11727.6). Branches without BP values were not supported in≥50% of the 100 bootstrap replicates. Bar equals 0.1 unit. Bold numbers above branches are identical with those in NJ and MP dendrograms. (C) Maximum parsimony (MP) dendrogram using heuristic option (one parsimonious tree, 2,418 steps, 448 bp informative under the condition of parsimony, consistency index=0.45, homoplasy index=0.55, retention index=0.48). Branches without BP values were not supported in≥50% of the 1,000 bootstrap replicates.

Relationships depicted as a result of ML (Fig. 1B) and MP analyses (Fig. 1C) showed no inconsistency with those expressed in the NJ dendrogram in terms of topology of nodes 1–9, except for the absence of node 3 in MP. However, when the Kishino and Hasegawa (1989) test and Templeton (1983) test were applied to ML and MP dendrograms, respectively, topologies in Fig. 1B and Fig. 1C were not significantly different from those in alternative hypotheses in which Gonocephalus was constrained monophyletic (P=0.130 and 0.167, respectively).

DISCUSSION

Several authors pointed out that there are two typical karyomorphs in the family Agamidae, of which one consists of 2n=34 or 36 chromosomes including 12 biarmed macro-chromosomes and 22 or 24 microchromosomes, whereas the other of 2n=46 or 48, all telocentric chromosomes without a distinct size break (Bickham, 1984; King, 1981; Moody and Hutterer, 1978; Olmo, 1986; Ota and Hikida, 1989; Solleder and Schmid, 1988; Witten, 1983). One of these karyomorphs is considered to be derived from the other through a series of Robertsonian rearrangements of macro-chromosomes, sometimes accompanied by addition or deletion of a pair of microchromosomes (Bickham, 1984; King, 1981). Based on these empirical assumptions, Ota et al. (1992) and Diong et al. (2000) hypothesized that the 2n=42 karyomorph with 22 biarmed macrochromosomes and 20 microchromosome of Gonocephalus bellii, G. chamaeleontinus, G. grandis, G. liogaster and G. miotympanum represents a highly specialized state. The latter authors also considered such an arrangement to be a strong evidence for the monophyly of those species against G. robinsonii, a species having 2n=32 karyotype including only 12 biarmed macro-chromosomes, and further assumed the latter species to be distant from other congeners phylogenetically.

Present results, consistently indicating the monophyly of the “2n=42 species” of Gonocephalus against G. robinsonii and other related genera with high bootstrap values, offer a substantial support to these hypotheses, although the number of species examined in the present study is too small to draw any definite conclusion. The states of morphological characters shared between G. robinsonii and other congeners (i.e., diagnostic characters of Gonocephalus), such as the presence of tubercular scales among otherwise homologous granular scales (Manthey and Grossmann, 1997), thus appear to be derived from convergence or represent symplesiomorphy.

As was pointed out by Diong et al. (2000), chromosomal arrangement of G. robinsonii most resembles that of a few Australian agamids in that it has only 20 microchromosomes besides 12 biarmed macrochromosomes. Present results, however, negate their close affinity, and indicate the origin of G. robinsonii in an Asiatic agamid radiation. Simple deletion of a pair of microchromosomes seems to be responsible for the emergence of karyotype of G. robinsonii.