Phylogeny of Ips DeGeer Species (Coleoptera: Scolytidae) Inferred from Mitochondrial Cytochrome Oxidase I DNA Sequence

Molecular Phylogenetics and Evolution Vol. 14, No. 3, March, pp. 445–460, 2000 doi:10.1006/mpev.1999.0705, available online at http://www.idealibrary.com on

Phylogeny of Ips DeGeer Species (Coleoptera: Scolytidae) Inferred from Mitochondrial Cytochrome Oxidase I DNA Sequence Anthony I. Cognato1 and Felix A. H. Sperling Department of Environmental Science, Policy, and Management–Division of Insect Biology, University of California, Berkeley, California 94720 Received January 14, 1999; revised July 1, 1999

We used 766 bp of DNA sequence data from the mitochondrial cytochrome oxidase I gene to reconstruct a phylogeny for 39 of 43 Ips species, many of which are economically important bark beetles. The phylogeny was reconstructed using equally weighted and weighted parsimony. In both analyses, peripheral clades were well supported while internal clades were poorly supported. Phylogenetic analysis of translated amino acids produced a poorly resolved tree that was discordant with trees reconstructed with nucleotide sequence data. Two main conclusions are drawn about the monophyly of Ips and traditional systematic groups within Ips. First, Ips is monophyletic only when I. mannsfeldi, I. nobilis, and the concinnus and latidens species groups are excluded. The latidens group, I. mannsfeldi, and I. nobilis form a monophyletic group with 3 Orthotomicus species, while the concinnus group has a more basal position. Second, the majority of the species groups in the current classification for Ips are not monophyletic. European Ips species do not form a monophyletic group, contrary to common usage, and are dispersed on the phylogenetic tree among North American species. These results indicate that a formal systematic revision of Ips is needed. r 2000 Academic Press Key Words: molecular systematics; beetle.

The bark beetle genus Ips DeGeer (Scolytidae) is composed of 43 species that are distributed throughout conifer forests of the northern hemisphere (S. L. Wood, 1982; Lanier, 1987; Wood and Bright, 1992; Bright and Skidmore, 1997). Twenty-seven Ips species occur only in North and Central America while the remaining 16 species occur in Eurasia (Wood and Bright, 1992). Eight Ips species are restricted to Asia (Wood and Bright, 1992) and none are restricted to Europe. The beetles use the phloem–cambium inner bark layer of conifers

1 To whom correspondence should be addressed at current address: The Natural History Museum, Department of Entomology, Cromwell Road, London SW7 5BD, United Kingdom. E-mail: a.cognato @nhm. ac.uk.

as food and substrate for rearing their young. Their status as forest pests has led to considerable research on their biology, ecology, systematics, and control (Mitton and Sturgeon, 1982; D. L. Wood, 1982; S. L. Wood, 1982; Bright, 1993). Several authors (Hopping, 1963a; Lanier, 1970a,b, 1972; S. L. Wood, 1982; Pfeffer, 1995) have revised North American and Eurasian Ips systematics; however, a unified classification of Ips species groups has not been attained. Also, the relationships of the endemic Asian Ips species have not been investigated in relation to the other Ips species. Hopping (1963a) provided the first extensive systematic treatment of the genus. He placed all North American species and three Eurasian species, I. duplicatus, I. typographus, and I. sexdentatus, into 10 ‘‘natural’’ groups (species groups) (Table 1) based on morphological characters, including the number of elytral declivity spines, antennal sutures, and male genitalia. The species in each group were revised in subsequent publications (Hopping, 1963b,c, 1964, 1965a,b,c,d,e). Lanier (1966, 1970a,b, 1972, 1987, 1991) tested the validity of Hopping’s groups III, IV, IX, and X through breeding experiments and karyology. Cryptic species were discovered among populations of I. plastographus, I. confusus, and I. calligraphus (Lanier, 1970b; Lanier et al., 1991). I. duplicatus and I. sexdentatus were removed from Groups IV and X, respectively, based on previously known morphological differences (Lanier, 1972). S. L. Wood (1982) suggested an alternative division of North American Ips species for practical purposes (Table 1). I. latidens and I. spinifer were moved to Ips from Orthotomicus. All European Ips were placed in North American groups, including I. sexdentatus, which was again grouped with I. calligraphus. However, the European Ips species have frequently been considered separately from North American species (Table 1) (Postner, 1974; Pfeffer, 1995). Recently Stauffer et al. (1997) presented a phylogeny of the seven European Ips species based on mitochondrial cytochrome oxidase I (mt DNA COI) sequences,

445

1055-7903/00 $35.00 Copyright r 2000 by Academic Press All rights of reproduction in any form reserved.

446

COGNATO AND SPERLING

TABLE 1 Systematic Groupings of North American and European* Ips Hoppins (1963,a,b,c, 1964, 1965a,b,c,d,e)

Lanier (1970a,b, 1972, 1987, 1991)

Group I—I. concinnus (Mannerheim), I. mexicanus (Hopkins). Group II—I. emarginatus (LeConte), I. knausi Swaine.

Not addressed

Group III—I. plastographus (LeConte).

Group III—I. plastographus, I. integer (Eichhoff).

Group IV—I. pini (Say), I. avulsus (LeConte), I. bonanseai (Hopkins), I. duplicatus* Sahlberg. Group V—I. perroti Swaine, I. amitinus* (Eichhoff). Group VI—I. perturbatus (Eichhoff), I. hunteri Swaine, I. utahensis Wood, I. woodi Thatcher, I. typographus* (Linneaus), I. cembrae* (Heer). Group VII—I. borealis Swaine, I. swainei R. Hopping, I. thomasi G. Hopping. Group VIII—I. tridens (Mannerheim), I. yohoensis Swaine, I. pilifrons Swaine, I. interruptus (Eichhoff). I. sulcifrons Wood, I. semirostris G. Hopping, I. amiskwiensis G. Hopping, I. engelmanni Swaine. Group IX—I. confusus (LeConte), I. lecontei Swaine, I. montanus (Eichhoff), I. cribricollis (Eichhoff), I. grandicollis (Eichhoff). Group X—I. calligraphus (Germar), I. interstitialis (Eichhoff), I. ponderosae Swaine, I. sexdentatus* Boerner. latidens group in Orthotomicus

Group IV—I. pini, I. avulsus, I. bonanseai.

Not addressed

Not addressed Not addressed

Not addressed

Not addressed

S. L. Wood (1982) concinnus group—I. concinnus, I. mexicanus. emarginatus group—I. emarginatus, I. knausi, I. acuminatus* Gyllenhal. plastographus group—I. plastographus, I. integer. I. typographus, I. longifolia (Stebbing), I. stebbingi (Strohmeyer).* pini group—I. pini, I. avulsus, I. bonanseai, I. mannsfeldi* Wachl. Combined with the perturbatus group perturbatus group—I. perturbatus, I. hunteri, I. woodi, I. perroti, I. cembrae,* I. amitinus* (Eichhoff), I. duplicatus* Sahlberg Combined with the tridens group

tridens group—I. tridens, I. pilifrons, I. borealis.

Postner (1974) Not addressed acuminatus group—I. accuminatus,* I. duplicatus,* I. mannsfeldi.* typographus group—I. typographus,* I. sexdentatus,* I. amitinus,* I. cembrae.* Ips mannsfeldi* in acuminatus group, other species not addressed. Not addressed Ips cembrae* and I. amitinus* in typographus group, and I. duplicatus* in acuminatus group, other species not addressed Not addressed

Not addressed

Group IX—I. paraconfusus grandicollis group—I. paraconNot addressed Lanier, I. confusus, I. hoppingi fusus, I. confusus, I. hoppingi, I. Lanier, I. lecontei, I. montanus, lecontei, I. montanus, I. grandiI. cribricollis, I. grandicollis. collis. Group X—I. calligraphus, I. calligraphus group—I. calligraIps sexdentatus* in typographus apache Lanier. phus, I. sexdentatus.* group, other species not addressed. latidens group in Orthotomicus

latidens group—I. latidens (Leconte), I. spinifer (Eichhoff).

Not addressed.

Note. Some species were synonymized by subsequent authors. The systematics of I. nobilis, and I. schmutzenhoferi were not addressed by these authors.

which suggested an alternative relationship to Postner’s (1974) classification. Two monophyletic groups, I. typographus ⫹ I. cembrae and I. duplicatus ⫹ I. amitinus ⫹ I. acuminatus, were resolved. I. mannsfeldi grouped with Orthotomicus erosus while I. sexdentatus constituted a separate lineage from the rest of the species. The inconsistencies among these various Ips classifications can potentially be resolved by including all Ips species in a single phylogenetic analysis. There is now agreement that taxa should be recognized on the basis of monophyly (Hennig, 1966; deQueiroz and Gauthier, 1992), because monophyly represents unique evolutionary events shared by the included species. A phylogenetic classification using monophyly as the basis for

systematics is clearly needed for Ips taxa. Species groups under the current classification (S. L. Wood, 1982) are assumed to be evolutionary units in ecological and behavioral studies, most of which pertain to pheromone production and response (Vite et al., 1972; Lanier and Wood, 1975; Lewis and Cane, 1990; Seybold, 1992; Seybold et al., 1995). However, Ips species group generalizations based on current classification (S. L. Wood, 1982) are open to reinterpretation due to the purely practical and nonphylogenetic basis on which some groupings were conceived. The paucity of phylogenetically informative morphological characters in Ips has hampered attempts at phylogenetic reconstruction (Cane et al., 1990). Cuticular hydrocarbons and molecular characters, such as

447

Ips mtDNA COI PHYLOGENY

allozymes and random amplified polymorphic DNAs (RAPDs), have been phylogenetically informative for closely related Ips species (Cane et al., 1990; Cognato et al., 1995; Page et al., 1997). The mitochondrial genome has been successfully used at the intrageneric level for beetle systematics (Simon et al., 1994; Funk et al., 1995; Normark, 1996; Langor and Sperling, 1997; Vogler and Welsh, 1997). The mitochondrial cytochrome oxidase I gene is a good candidate for resolving relationships among Ips species, given the copious amount of nucleotide substitution observed among the European Ips species (Stauffer et al., 1997). Using the 58 half of the mtDNA COI gene, we reconstructed a phylogeny of 39 of 43 Ips species, including individuals from more than one population for 21 species. We then used this phylogeny to reassess the current classification of Ips species (S. L. Wood, 1982). MATERIALS AND METHODS Specimens and DNA Extraction Locality data and tissue preservation methods for samples of Ips species are listed in Appendix 1. Specimens of four endemic Asian species, I. chinensis Kurenzov & Kononov, I. orientalis Wood & Yin, I. ussuriensis Reitter, and I. nitidus Eggers, were unavailable for analysis. I. cribricollis and I. schmutzenhoferi have been synonymized with I. grandicollis (Wood, 1977) and I. stebbingi (Wood, 1992), respectively. However, for this analysis, these species are considered separately because of distinct morphological and biological differences (Lanier, 1987; Holzschuh, 1988; Schmutzenhofer, 1988). All other species identifications were determined by morphology, host, and range (S. L. Wood, 1982; Lanier, 1987; Holzschuh, 1988; Pfeffer, 1995) and nomenclature follows Bright and Skidmore (1997) and Wood and Bright (1992). Mitochondrial DNA was extracted using one of two procedures. In the first method, frozen thoraces were pulverized separately in 1.5-ml microfuge tubes with a stainless steel pestle. Then, 50 µl of LTE (10 mM Tris, pH 8.0, 1 mM EDTA) was pipeted into each microfuge tube and the tubes were boiled for 20 min. Without being removed from the tubes, homogenized samples were frozen for later use as template in the polymerase chain reaction (PCR). Alternatively, a QiaAmp tissue kit (QIAGEN Inc., Santa Clara, CA) was used for DNA extraction. Thoraces from the beetles were removed for extraction and the manufacturer’s tissue protocol was followed, except for the DNA elution, which consisted of one elution of 200 µl of AE buffer. The remaining body parts were placed into gelatin capsules or glued onto a mounting board and pinned. These specimens were deposited in the Essig Museum of Entomology, University of California, Berkeley.

DNA Amplification and Sequencing A region of approximately 800 bp of the mtDNA COI gene was amplified using a combination of primers (Table 2). Primers were reproduced from Simon et al. (1994) and Juan et al. (1995) or customized for optimal polymerase chain reaction on Ips species after a few species had been sequenced using universal primers. Each PCR cocktail contained 35 µl ddH2O, 5 µl 10⫻ Promega buffer, 4 µl 25 mM Promega MgCl2, 1 µl 40 mM dNTPs, 2 µl of each 5 mM primer, 1 µl of DNA template, and an overlay of mineral oil (1 drop). The following steps were performed on a programmable thermal cycler (Ericomp, San Diego, CA): cycle 1, 3 min at 95°C, 1 min at 45°C, 1.5 min at 72°C; cycles 2–36, 1 min at 94°C, 1 min at 45°C, 1.5 min at 72°C; cycle 37, 1 min at 94°C, 1 min at 45°C, 5 min at 72°C. The thermal cycler was put on pause at 45°C in the first cycle and 0.2 µl of Promega Taq was added to each tube. PCR products were cleaned of primers and unincorporated nucleotides either with a Millipore ultra free-MC filter (Millipore Corp., Bedford, MA) or with a QIAquick PCR Purification Kit (QIAGEN Inc.) following the manufacture’s instructions and were directly sequenced using ABI 377 automated sequencing with fluorescent dye terminators (Applied Biosystems, Inc., Foster City, CA). The DNA sequence for each species was confirmed with both sense and anti-sense strands. All specimen sequences were submitted to GenBank (U82236, AF113325–AF113396) (Appendix). Sequence Analysis Calculations for phylogenetic analyses, nucleotide ratios, transition/transversion (ts/tv) ratios, and Kishino–Hasegawa tests were performed with PAUP* 4.0b.1 (Swofford, 1998). Structural regions of mtDNA COI were determined according to Lunt et al. (1996). In all tree searches, Orthotomicus caelatus (Eichhoff), Orthotomicus erosus (Wollaston), Orthotomicus laricis (Fabricius), Pityogenes hopkinsi Swaine, Pityogenes calcara-

TABLE 2 PCR Primer Pairs (A and B) Used to Amplify Two Overlapping Fragments of the mtDNA COI Gene Sense primers A C1-J-2183 B C1-J-2410

58-CAA CAT TTA TTT TGA TTT TTT GG 58-CCT ACA GGA ATT AAA ATT TTT AGT TGA TTA GC 58-CCC TCA AGA CTT TGA TCT TTA GG 58-CCA TCC TCA ATC TGA TCG TTA GG 58-CCA TCA AGG CTT TGA TCT CTA GG 58-CCC TCT TCA ATT TGA TCC CTG GG 58-CCT TCA TCT CTT TGA TCA CTA GG

B C1-J-2495a B C1-J-2495b B C1-J-2495c B C1-J-2495d B C1-J-2495e Anti-sense primers A C1-N-2611 58-GCA AAA ACT GCA CCT ATT GA B TL2-N-3014 58-TCC AAT GCA CTA ATC TGC CAT ATT A Note. Primer labels follow Simon et al. (1994).

448

COGNATO AND SPERLING

tus (Eichhoff), Pityogenes carinulatus (LeConte), and Pityokteines sparsus (LeConte) were used as outgroups. These species were chosen as outgroups on the basis of their previously hypothesized close relationship to Ips (S. L. Wood, 1982). A heuristic search with 50 random stepwise addition replicates was performed with PAUP*. Branch swapping was performed with tree bisection– reconnection. Branch support was represented as decay index values (Bremer, 1994) and bootstrap values calculated with 500 replicates. Two additional analyses were performed with character matrices weighted to take into account potential heterogeneity in nucleotide substitution rates observed in this data set. Both analyses involve a posteriori weighting, as opposed to a priori weighting (Brower and Desalle, 1994). A priori character-state weighting is advantageous because it does not derive weights from the data in question, which in turn avoids circular reasoning (Neff, 1986; Albert et al., 1993). Although, intuitively, a posteriori weighting may seem circular (Neff, 1986; Albert et al., 1993), this method is not circular because it can reconstruct a tree not found among the trees of the equally weighted data set (Fitch and Ye, 1991). However, a priori weighting requires an initial hypothesis of gene evolution or a robust phylogeny for the taxa in question (Simon et al., 1994; Albert et al., 1993). Neither a model of mtDNA COI gene evolution nor a reliable phylogeny of Ips exists. Therefore, we weighted this data set using a posteriori methods. First, the mtDNA COI data set was weighted by successive approximation (Farris, 1969), based on the rescaled consistency index values for each character derived from the trees obtained from the unweighted parsimony analysis. Analysis of this weighting scheme used the same heuristic search conditions as for the unweighted data. In the second analysis, nucleotide sequences were translated into amino acid sequences. This effectively weights the data set by increasing weight to 1st and 2nd codon positions and downweighting 3rd positions. Nucleotide sequences were translated using GeneJockey II (Biosoft, Ferguson, MO) according to the Drosophila yakuba mtDNA COI

sequence (Clary and Wolstenholme, 1985), and a phylogeny was reconstructed with a parsimony analysis of the amino acid sequences. The Kishino–Hasegawa (1989) test was performed to examine which trees from the equally weighted and weighted data analyses provide a better phylogeny. Ln-likelihood values were calculated with PAUP* for the trees from analyses using the Hasegawa–Kishino– Yano model. The following assumptions were used for this model: no molecular clock, estimated ts/tv, empirical base frequencies, among-site rate variation approximated to a gamma distribution with four rate categories, and shape parameter ⫽ 0.5. These assumptions were chosen because they allow a robust model of nucleotide evolution (Swofford et al., 1996). RESULTS Nucleotide Patterns A 766-bp region of the mtDNA COI gene, corresponding to bases 2211–2976 of Drosophila yakuba mtDNA COI sequence (Clary and Wolstenholme, 1985), was used to reconstruct the phylogeny of Ips. This region contained 326 phylogenetically informative characters. High sequence divergence and the majority of phylogenetic information occurred at the 3rd codon position, followed by the 1st and 2nd codon positions (Table 3, Fig. 1). Minimal nucleotide saturation is observed at the 1st and 2nd codon positions, as indicated when the number of observed substitutions is similar to the number of unobserved substitutions (Fig. 1: x ⫽ y). Saturation at 3rd codon positions is high, as indicated by the divergence from the x ⫽ y line. Overall ts/tv ratio was approximately 1:1; however, a high ts/tv ratio occurs at the 1st codon position (Table 3). The 2nd codon position is most conserved and exhibits a low ts/tv ratio, as observed in other studies on beetles (Funk et al., 1995; Sperling and Langor, 1997; Stauffer et al., 1997). These data also show a pattern of high ts/tv ratio corresponding to low nucleotide divergence and vice versa (Fig. 2). The observed

TABLE 3 Nucleotide and Amino Acid Divergence Patterns Observed for the 58 Half of the mtDNA COI Gene for Ips and Outgroup Species

Total sequence 1st Codon position 2nd Codon position 3rd Codon position Amino acids

Mean sequence dissimilarity* (uncorrected ‘‘p’’)

Transition/* transversion

Variable sites

Informative characters

Number of characters

0.16 0.07 0.02 0.39 0.05

1.22 4.47 0.89 1.23 —

354 71 32 251 65

326 62 20 244 45

766 255 255 256 255

Note. An asterisk (*) indicates values calculated from average pairwise species comparisons.

Ips mtDNA COI PHYLOGENY

449

Lepidoptera (Sperling et al., 1996). Most Ips intraspecific sequence divergence was ⬍3.0% and I. cembrae from China exhibited the highest divergence (4.9%), compared to two European I. cembrae populations. Among species within well-supported clades (⬎90% bootstrap), sequence divergences range between 3.0 and 10.0% (Table 4). However, sequence divergences between the species pairs I. hunteri/I. perturbatus and I. borealis/I. pilifrons are 0.6 and 1.0%. Sequence divergence between ingroup and outgroup species ranges between 15.0 and 23.0%. Sequence divergence similar to that found between ingroups and outgroups is observed among the outgroup species (Table 4). Sixty-five of 255 amino acids were variable, including 45 phylogenetically informative sites. Mean amino acid variability is highest in the membrane-spanning helix M12 (1.53) and the carboxy terminal (2.0) protein regions and lowest (1.0) in the E4, M8, and M10 regions (Fig. 3). Nucleotide variation was high in all functional regions (Fig. 3) and most of this was due largely to 3rd position synonymous substitutions. These results are generally concordant with those of Lunt et al. (1996), Langor and Sperling (1997), and Caterino and Sperling (1999). Phylogenetic Analyses

FIG. 1. Relationship between uncorrected nucleotide divergences and corrected nucleotide divergences (Tamura and Nei, 1993), partitioned by codon position among pairwise OTU (species and populations) comparisons. Deviation of data points from the x ⫽ y line suggests degree of saturation.

ts/tv ratio is high between recently diverged species and approaches 0.5 with increased divergence, due to saturation of transitions. These patterns of nucleotide turnover for Ips mtDNA COI have been observed for other beetle species and insects in general (Simon et al., 1995; Funk et al., 1995; Stauffer et al., 1997). Intraspecific pairwise sequence divergence ranged between 0.3 and 4.9% (Table 4). The mean of these values (0.0158) is within the range of mtDNA COI and COII intraspecific sequence divergences observed for

Six most-parsimonious trees were found for Ips species and outgroups with the equally weighted analysis. I. tridens, I. pilifrons, I. borealis, and branches with 0 decay index (Fig. 4) are unresolved in a consensus of the six most-parsimonious trees (not shown). The internal nodes of these six most-parsimonious trees were not well supported by bootstrap values; however, good support was observed for peripheral clades. Clades of closely similar species (⬍10% sequence divergence) have high bootstrap values (⬎90%), while clades with ⬎10% sequence divergence generally have lower bootstrap values (⬍90%) (Fig. 4). One iteration of successive approximation analysis resulted in a single most-parsimonious tree (Fig. 5). This tree was not represented in the six mostparsimonious trees found in the equally weighted analysis. This result supports the hypothesis that successive approximation analysis is recursive rather than circular (Wenzel, 1997). The topologies of the equally weighted and successive approximation trees differ in the arrangement of O. laricis, I. mannsfeldi, I. woodi, and clades A, B, and C (Fig. 4). The positions of these species and clades have low support values and their rearrangement is therefore not surprising. Also, the rearrangement of these species and clades in the successive approximation tree does not disrupt any well-supported clades found in the equally weighted analysis. Bootstrap values and decay indices of the other clades were similar for both of these trees. Phylogenetic analysis of translated amino acid sequences resulted in 18,000 trees from an incomplete

450

COGNATO AND SPERLING

FIG. 2. Relationship between uncorrected nucleotide divergences and transition/transversion ratios observed among pairwise OTU comparisons.

heuristic search. Exhaustion of computer memory precluded a complete heuristic search. Low phylogenetic resolution was observed due to the lack of variability among amino acid sequences (Table 3). The consensus tree from analysis of amino acids was less resolved than the trees reconstructed with nucleotide data. Ten of 16 monophyletic groups (Fig. 6) were concordant with the trees reconstructed with the nucleotide data (Figs. 4

Mean sequence dissimilarity

Range

and 5); however, a few clades are morphologically improbable (Fig. 6). The trees reconstructed with the amino acid data set were not included in the Kishino–Hawasaga test, because of the large number of trees. Also, the tree search was incomplete; thus, the most likely tree is not guaranteed for comparison with the other analyses. The ln likelihoods for the trees resulting from equal weighting (most likely tree out of six) and successive approximation were 13018.69 and 12997.23, respectively. The ln-likelihood of the equally weighted tree is less likely (P ⫽ 0.1) compared to the successive approximation tree (Kishino and Hawasaga, 1989), although this value is not statistically significant. We use the successive approximation tree (Fig. 5) to facilitate discussion of Ips systematics because it provides the most resolved Ips phylogeny.

0.016 (n ⫽ 30)

0.003–0.05

DISCUSSION

0.066 (n ⫽ 17)

0.03–0.10

0.186 (n ⫽ 397)

0.15–0.23

0.182 (n ⫽ 17)

0.15–0.20

TABLE 4 Mean Sequence Dissimilarity (Uncorrected ‘‘p’’) of the 58 Half of the mtDNA COI Gene for Pairwise Comparisons among Ips Species and Outgroups

Intraspecific (comparing ingroup taxa only) Interspecific (comparing taxa within well-supported clades, ⬎90% bootstrap) Interspecific (comparing ingroups vs outgroups) Interspecific (comparing outgroup taxa only)

Nucleotide and Amino Acid Divergence Patterns mtDNA COI nucleotide sequence divergence is high for both Ips intra- and interspecific comparisons. High intraspecific divergence for mtDNA COI (5.0%) has also been observed for other beetles (Funk et al., 1995). The

Ips mtDNA COI PHYLOGENY

451

FIG. 3. Mean variability in nucleotides and amino acids of the structural regions for half of the mtDNA COI gene. Variability was calculated as the average number of nucleotides or amino acids per site observed in each region for 38 Ips species and 7 outgroup species.

high level of nucleotide substitution observed among Ips species may reflect either high nucleotide substitution rates or a long period of time since their divergence from a common ancestor. For Pissodes weevils the former hypothesis is favored because low enzyme and morphological differences are associated with high mtDNA nucleotide substitutions (Langor and Sperling, 1997). Comparison of the pairwise allozyme differences (Cane et al., 1991) to the corresponding uncorrected sequence dissimilarity for Ips species (I. confusus, I. hoppingi, I. paraconfusus, I. grandicollis, I. lecontei, I. pini, and I. latidens) (Fig. 5) shows relatively fast mtDNA COI sequence evolution (Fig. 7). The mean [range] sequence (0.063, [0.04–0.074]) and allozyme (0.085, [0.073–0.091]) divergence for three closely related species (I. confusus, I. hoppingi, and I. paraconfusus) are similar to values calculated for closely related Pissodes species (Langor and Sperling, 1997). Comparison of the above Ips species to the distantly related species I. latidens indicates that the sequence dissimilarity has stabilized at approximately 0.18, while the allozyme distance ranges from 0.688 to 1.092 (Fig. 7). The amount of mtDNA COI sequence divergence affects the amount of homoplasy in the Ips phylogeny. mtDNA COI sequences are probably effectively saturated at internal nodes relative to the peripheral clades, as was found for Papilio species (Reed and Sperling, 1999). For Ips, saturation is highest at 3rd codon positions, followed by 1st and 2nd codon positions for the internal nodes (Fig. 1). Third positions are not

excluded from these analyses; they provide the majority of phylogenetically informative sites (Table 3). However, saturation of nucleotide substitutions at the internal nodes decreases the potential for characters to be phylogenetically informative and consequently decreases phylogenetic support. Mean amino acid variation within structural regions of mtDNA COI is concordant with that of other taxa (Lunt et al., 1996; Langor and Sperling, 1997; Caterino and Sperling, 1999). However, amino acid variability is higher and occurs within more protein regions for Ips as compared to Pissodes (Langor and Sperling, 1997). This is most likely an effect of taxon sampling and phylogenetic divergence, because amino acid variability for this study was calculated from 46 taxa ranging among 4 genera, while amino acid variability for Pissodes was calculated from only 5 closely related species. The Ips phylogeny reconstructed with amino acids produced a poorly resolved strict consensus tree (Fig. 6). The tree topology is discordant with the trees reconstructed with the nucleotide sequence (Figs. 4–6). Stauffer et al. (1997) used amino acids (translated from mtDNA COI sequence between primers C1-J-2410 and TL2-N-3014) to reconstruct a single most-parsimonious tree for the seven European Ips. Taxon sampling is a potential factor causing the different results obtained by Stauffer et al. (1997) compared to this study. The addition of more species confounds the phylogenetic signal, resulting in the thousands of trees obtained in this study (Fig. 6).

452

COGNATO AND SPERLING

FIG. 4. Ips species phylogram of one of six most-parsimonious trees from the unweighted analysis. Letters indicate clades discussed in text. Bootstrap values ⬎50% are given above line and decay index is given below line. Branches are unresolved in consensus tree with decay index ⫽ 0. Bootstrap value of 100 ⫽ decay index ⱖ6. CI ⫽ 0.207, RI ⫽ 0.568, RC ⫽ 0.122.

Ips mtDNA COI PHYLOGENY

453

FIG. 5. Most-parsimonious tree of Ips species calculated with characters weighted by successive approximation using rescaled consistency index. *Eurasian species. Letters indicate clades discussed in text. Bootstrap values ⬎50% are given above line and decay index is given below line. Bootstrap value of 100 ⫽ decay index ⱖ3.

454

COGNATO AND SPERLING

FIG. 6. Ips species strict consensus tree of 18,000 most-parsimonious trees (incomplete heuristic search) from translated amino acid sequence analysis. Only one specimen of each species was used because of the lack of intraspecific amino acid difference. Asterisks indicate clades that are morphological improbable.

Ips Phylogeny and Classification Two main conclusions resulted from the phylogenetic analyses (Figs. 4 and 5). First, Ips is polyphyletic when all 39 Ips species and outgroups are analyzed together. The concinnus and latidens groups, I. mannsfeldi, and I. nobilis showed relationship to the outgroup species. Second, most previously recognized species groups (S. L. Wood, 1982) (Table 1) are not monophyletic, with the exception of the concinnus and tridens groups (Fig. 5). While most of the intraspecific clades and a few interspecific clades have high bootstrap values, the majority of

internal nodes exhibit less then 50% bootstrap values (Fig. 5). This phylogeny is thus a preliminary hypothesis of relationships of closely related Ips species, and more distantly related clades with low bootstrap values are open to reinterpretation with future analyses. Species groups defined by S. L. Wood (1982) are discussed below with reference to the phylogeny (Table 5) and the discussion of these results follows from the bottom of the phylogeny to the top (Fig. 5). Monophyly of Ips, with the exclusion of the concinnus and latidens species groups, I. mannsfeldi, and

Ips mtDNA COI PHYLOGENY

455

FIG. 7. Scatter plot showing the association between allozyme distance (Nei, uncorrected) and sequence dissimilarity (uncorrected ‘‘p’’) for seven pairwise Ips species comparisons. R2 ⫽ 0.84.

I. nobilis (Fig. 5), is supported by the morphological synapomorphy of a gradual elytral declivity (S. L. Wood, 1982). The concinnus group is monophyletic, including I. mexicanus and I. concinnus. The basal placement agrees with the observation of morphological characteristics shared by these Ips species and other Ipini genera, such as Orthotomicus and Acanthotomicus (S. L. Wood, 1982). The latidens group, I. mannsfeldi, and I. nobilis are monophyletic with three outgroup species, O. caelatus, O. erosus, and O. laricis (Fig. 5, clade D). As with the concinnus group, the morphological boundaries between Ips and Orthotomicus are obscure for the species mentioned above and the generic placement of these species has been debated (S. L. Wood, 1982; Pfeffer, 1995). The mtDNA phylogeny (Fig. 5) suggests that inclusion of these species with Orthotomicus is warranted. However, systematic revision of Ipini is necessary to redefine the morphological boundaries between Ips and Orthotomicus. The placement of the other outgroup species is basal to Ips species, excluding the concinnus and latidens groups (Fig. 5). The majority of species groups within clade E (Fig. 5) are not monophyletic. However, I. knausi and I. emarginatus are monophyletic, as was hypothesized by Hopping (Table 1). This clade was reconstructed for all analyses (Figs. 4 and 5). I. sexdentatus is basal to the emarginatus group and these species together are basal to all Ips in clade F. Monophyly is supported for the plastographus group (Figs. 4 and 5, excluding I. typographus). This group is

embedded in the larger clade G (Fig. 5), rendering the I. pini group paraphyletic. S. L. Wood’s (1982) placement of I. mannsfeldi in the pini group is not supported. As shown above, it is closely related to Orthotomicus spp., and this relationship is also supported by the presence of a steep elytral declivity (S. L. Wood, 1982; Pfeffer, 1995). The perturbatus group is a conglomerate of related and unrelated species comprising two monophyletic groups and I. woodi (Fig. 5, clades H and K). Clade H comprises the North American I. perroti and the European I. amitinus, I. typographus, and I. cembrae. However, bootstrap values are less than 50% for this clade and the placement of I. perroti is in question. The remaining species of the perturbatus group, I. perturbatus and I. hunteri, are monophyletic and form the sister to the tridens group (Fig. 5, clades K and J). These two groups form a well-supported clade (Fig. 5, clade I) found in all analyses (Figs. 4 and 5). The species of both groups comprise all North American spruce-feeding Ips species. These species are closely related and the relationship among I. tridens, I. pilifrons, and I. borealis is not completely resolved (Fig. 4). All species (Fig. 5, clade I) have been shown to interbreed (Lanier and Burkholder, 1974) and the above observation is not surprising. The population of I. pilifrons 1 is disjunct from the populations of I. tridens, I. pilifrons 2, and I. borealis (Appendix 1) and its phylogenetic position is plausible on the basis of the disjunct range. In contrast, I. perturbatus 3 is not monophyletic with I. perturbatus 2, even though it was a sample from the same popula-

456

COGNATO AND SPERLING

TABLE 5 Comparison of Ips Species Groups (defined by S. L. Wood, 1982) to Phylogeny Reconstructed with Successive Approximation Weighting (Fig. 5) % Bootstrap

S. L. Wood, 1982

Phylogeny (Fig. 5)

concinnus Group—I. concinnus, I. mexicanus.

concinnus Group—I. concinnus, I. mexicanus. Not monophyletic with all Ips species. emarginatus Group—I. emarginatus, I. knausi. Monophyletic. plastographus Group—I. plastographus, I. integer. Monophyletic with pini group. pini Group—I. pini, I. avulsus, I. bonanseai. Paraphyletic. perturbatus Group—I. perturbatus, I. hunteri. Monophyletic with tridens group.

100

tridens Group—I. tridens, I. pilifrons, I. borealis. Monophyletic with perturbatus group. grandicollis Group—I. paraconfusus, I. confusus, I. hoppingi, I. lecontei, I. montanus, I. grandicollis, I. cribricollis. Paraphyletic. calligraphus Group—I. calligraphus, I. apache. Monophyletic with grandicollis group. latidens Group—I. latidens, I. spinifer. Monophyletic with Orthotomicus species.

100

emarginatus Group—I. emarginatus, I. knausi, I. acuminatus.* plastographus Group—I. plastographus, I. integer, I. typographus.* pini Group—I. pini, I. avulsus, I. bonanseai, I. mannsfeldi.* perturbatus Group—I. perturbatus, I. hunteri. I. woodi, I. perroti, I. cembrae,* I. amitinus,* I. duplicatus.* tridens Group—I. tridens, I. pilifrons, I. borealis.

grandicollis Group—I. paraconfusus, I. confusus, I. hoppingi, I. lecontei, I. montanus, I. grandicollis. calligraphus Group—I. calligraphus, I. sexdentatus.* latidens Group—I. latidens, I. spinifer, I. erosus.*

100

100

0

100

0

100

100

*European species.

tion (Appendix 1). I. perturbatus 3 may represent a gynogenetic lineage (Lanier and Oliver, 1966) or may indicate incomplete lineage sorting. Monophyly of the grandicollis group (Fig. 5, clade L) is not supported in all analyses (Figs. 4 and 5). The populations of I. cribricollis constitute a lineage distinct from the populations of I. grandicollis, thus supporting the validity of the former species (Lanier, 1987). I. montanus, I. grandicollis, I. cribricollis, and I. lecontei are monophyletic with, and basal to, the calligraphus group (Fig. 5, clade M). I. apache is distinct from I. calligraphus (Lanier et al., 1991) and the exclusion of I. sexdentatus from the calligraphus group is confirmed (Lanier et al., 1972). I. paraconfusus, I. confusus, and I. hoppingi are monophyletic (Fig.

5) and the monophyly of I. confusus and I. hoppingi is concordant with their pinyon pine hosts (Pinus spp.). The grandicollis group has received phylogenetic attention from separate analyses of three character sets; allozymes (Cane et al., 1990), RAPDs (Cognato et al., 1995), and cuticular hydrocarbons (Page et al., 1997). The phylogenies resulting from these studies are generally concordant with those from the mtDNA COI data (Fig. 5). Hypotheses proposed by the previous studies (Cane et al., 1990; Cognato et al., 1995; Page et al., 1997), monophyly of I. paraconfusus, I. confusus, and I. hoppingi, and the ambiguous placement of I. lecontei within the grandicollis group are corroborated with the mtDNA COI data. Unknown to all authors was the close relationship of the grandicollis and calligraphus groups (Fig. 5). The inclusion of the calligraphus groups in this analysis reveals the paraphyly of the grandicollis group. The mtDNA COI phylogenies suggests that I. grandicollis, I. cribricollis, I. lecontei, and the calligraphus group are monophyletic (Fig. 5). This hypothesis is supported by morphology; the 1st elytral declivity spine on the second interstrial space is synapomorphic for these species (S. L. Wood, 1982). The multitude of data sets offers an opportunity for a single robust phylogenetic analysis of the grandicollis group. However, it is now necessary to include the species of the calligraphus group, given their close relationship to the grandicollis group. Unfortunately, allozyme, RAPD, and cuticular hydrocarbon data were not collected for the species of the calligraphus group. A more rigorous combined analysis awaits these data. The European Ips species are not a monophyletic group. They are dispersed in the phylogeny among the North American species, although the exact arranagement is uncertain because of low bootstrap values (Fig. 5). The close relationships of I. typographus and I. cembrae, and of I. mannsfeldi and O. erosus hypothesized by Stauffer et al. (1997) are corroborated. However, the current analysis does not support the hypothesized monophyly of I. duplicatus, I. amitinus, and I. acuminatus (Stauffer et al., 1997). I. acuminatus is basal to most other Ips species (Fig. 5) and has an emarginate 3rd elytral declivity spine, as observed with I. knausi and I. emarginatus. The phylogenetic position of I. amitinus is also difficult to ascertain; this species may be related to the North American I. perroti but bootstrap values are below 50% (Fig. 5). I. duplicatus is most closely related to the morphologically similar I. hauseri; however, their phylogenetic placement is not well supported (Fig. 5). A phylogeny reconstructed with few taxa and characters can be misleading (Lecointre et al., 1993; Cummings et al., 1995). Based on a phylogeny of 7 European Ips species, Stauffer et al. (1997) hypothesized that speciation of I. acuminatus occurred with a host shift between tree genera (Picea abies to Pinus silvestris).

457

Ips mtDNA COI PHYLOGENY

However, the Ips phylogeny reconstructed with 39 of 43 species (Fig. 5) indicates that I. sexdentatus, I. emarginatus, and I. knausi are basal to I. acuminatus and all use Pinus spp. as hosts. Therefore, it is probable that the ancestor of I. acuminatus did not feed on spruce. This suggests that speciation of I. acuminatus did not involve a shift between two host tree genera but possibly occurred as a host shift among Pinus species. The Himalayan Ips species I. longifolia, I. stebbingi, and I. schmutzenhoferi (Fig. 5, clade N) are monophyletic and within this group I. longifolia and I. stebbingi are also monophyletic (Fig. 5). Both these groups are associated with high bootstrap values and are also present in the amino acid consensus tree (Fig. 6). The placement of the Himalayan species among the other Ips species is tentative; however, they may be related to the North American spruce Ips (Fig. 5, clade I). The removal of I. schmutzenhoferi from synonymy with

I. stebbingi is suggested, given its basal position to I. longifolia and I. stebbingi. The large amount of nucleotide difference observed for I. schmutzenhoferi indicates that it has evolved separately for a long period from both I. longifolia and I. stebbingi (Fig. 4). Many Asian Ips species have been synonymized because of morphological similarity (for example, I. subelongatus Motschulsky, I. fallax Eggers, and I. shinanoensis Yono were synonymized with I. cembrae) (Wood and Bright, 1992). However, the high amount of genetic divergence among the Himalayan Ips species suggests that Ips species diversity in Asia may be underestimated. These phylogenetic differences compared to traditional Ips systematics (Table 5) indicate that a revision of Ips is necessary. Formal systematic revision of Ips, including additional data from elongation factor-1␣ and mtDNA 16S genes, is in progress (A.I.C., unpublished data).

APPENDIX Collection Data for Ips and Outgroup Species (All Specimens Were Collected by A. I. Cognato Unless Noted)

Species Ips acuminatus (Gyllenhal) Ips amitinus 1 (Eichhoff) Ips amitinus 2 Ips apache 1 (Lanier) Ips apache 2 Ips avulsus 1 (Eichhoff) Ips avulsus 2 Ips bonanseai 1 (Hopkins) Ips bonanseai 2 Ips borealis Swaine Ips calligraphus 1 (Germar) Ips calligraphus 2 Ips cembrae 1 (Heer) Ips cembrae 2 Ips cembrae 3 Ips concinnus (Mannerheim) Ips confusus 1 (LeConte)

Ips confusus 2 Ips cribricollis 1 (Eichhoff) Ips cribricollis 2 Ips duplicatus 1 (Sahlberg) Ips duplicatus 2 Ips emarginatus 1 (LeConte) Ips emarginatus 2 Ips grandicollis 1 (Eichhoff)

Preservation method

Genbank Accession No.

Unrecorded Picea abies Spruce

Frozen Ethanol Ethanol

AF113325 AF113326 AF113327

Pinus engelmanni

Frozen

AF113328

Unrecorded

Ethanol

AF113329

Unrecorded Pinus Pinus

Pinned Frozen Frozen

AF113330 AF113331 AF113332

Unrecorded Picea glauca

Frozen Frozen

AF113333 AF113334

Pinus strobus Pinus ponderosa Larix decidua Larix decidua Larix gmelini Picea sitchensis Pinus edulis

Frozen Pinned Ethanol Ethanol Ethanol Frozen Frozen

AF113335 AF113336 AF113337 AF113338 AF113339 AF113340 AF113341

Pinus monophylla Pinus ponderosa

Frozen Frozen

AF113342 AF113343

Pinus leiophylla

Ethanol

AF113344

Picea abies

Ethanol

AF113345

Spruce

Ethanol

AF113346

Pinus ponderosa

Frozen

AF113347

Pinus ponderosa Pinus rigida

Frozen Frozen

AF113348 AF113349

Collection locality, date and collector Czech Republic: Moravia, Tulasice V-1995. M. Knizek Czech Republic: Moravia, Chuchelna V-1995. M. Knizek Russia: Lenbolovo, St. Petersburg Reg. 8-10-VIII-1998. M. Mandelstan AZ: Cochise Co. Coronado National Forest. Rd 42, 10 mi NW of Portal. 6-IX-1996. Mexico: Michoacan: Uruapan. Barranca del Cupatitzio 1996. A. Storer Louisiana. 1996. G. Lenhard GA: Cobb Co. Marietta. 19-IX-1997. M. Caterino AZ: Graham Co. Coronado National Forest Rd. 366. 5-IX-1996. Mexico: Nuevo Leon. XII-1993. S. Seybold Canada: Quebec, Gatineau Park nr. Tayor Lake. 20-V-1995. NY: Suffolk Co. Smithtown 11-IX-1994. AZ: Gila Co. 14 mi. S. Young 26-XI-1985. G. N. Lanier Czech Republic: Moravia, Resice. 15-V-1997. M. Knizek France: Brianc¸on. 16-VI-1996. A. Roques China: Heilongjiang, Lo-Shan. 1-VI-1996. A. Roques WA: Clallam Co. Hoh River Rd. 28-VI-1997. AZ: Gila Co. San Carlos Indian Reservation. 11 mi. NE of Point of Pines 26-V-1993. S. Seybold, A. I. Cognato & D. L. Wood NV: Douglas Co. Gardenerville. II-1994. S. Seybold. NM: Otero Co. Lincoln National Forest Rt. 82 8.05 km E. Cloudcroft, 11-V-1994. Mexico: Michoacan: San Juan Nuevo, LaPila Forest, 31-V-1998. A. I. Cognato & A. Del Rio-Morea Czech Republic: Moravia, Chuchelna. V-1995. M. Knizek Russia: Lenbolovo, St. Petersburg Reg. 8-10-VIII-1998. M. Mandelstan CA: Lassen Co. Black’s Mountain Experimental Forest. VII-1994. S. Seybold, A. I. Cognato & D. L. Wood WA: Kittitas Co. Roslyn. 30-VI-1997. NY: Suffolk Co. Brookhaven. 28-VIII-1992

Host

458

COGNATO AND SPERLING

APPENDIX—Continued

Species Ips grandicollis 2 Ips grandicollis 3 Ips hauseri Reitter Ips hoppingi 1 Lanier Ips hoppingi 2 Ips hunteri Swaine Ips integer (Eichhoff) Ips knausi Swaine Ips latidens 1 (Leconte) Ips latidens 2 Ips lecontei 1 Swaine Ips lecontei 2 Ips longifolia (Stebbing) Ips mannsfeldi (Wachtl) Ips mexicanus (Hopkins) Ips montanus (Eichhoff) Ips nobilis (Wollaston) Ips paraconfusus 1 Lanier Ips paraconfusus 2 Lanier Ips perroti Swaine Ips perturbatus 1 (Eichhoff) Ips perturbatus 2, 3 Ips pilifrons 1 Swaine Ips pilifrons 2 Ips pini 1 (Say) Ips pini 2 Ips pini 3 Ips plastographus (LeConte) Ips schmutzenhoferi Holzschuh Ips sexdentatus (Boerner) Ips spinifer (Eichhoff) Ips stebbingi 1 Strohmeyer Ips stebbingi 2 Ips tridens (Mannerheim) Ips typographus 1 (Linnaeus) Ips typographus 2 Ips typographus 3

Preservation method

Genbank Accession No.

Pinus banksiana

Pinned

AF113350

Unrecorded

Pinned

AF113351

Unrecorded

Pinned

AF113352

Pinus cembroides

Frozen

AF113353

Pinus cembroides

Frozen

AF113354

Picea pugens

Frozen

AF113355

Pinus ponderosa

Frozen

AF113356

Pinus ponderosa

Ethanol

AF113357

Pinus lambertinan Pinus ponderosa Pinus ponderosa

Frozen Frozen Frozen

AF113358 AF113359 AF113360

Pinus ponderosa

Frozen

AF113361

Chir pine

Pinned

AF113362

Unrecorded Pinus muricata Pinus monticolla

Ethanol Frozen Frozen

AF113363 AF113364 AF113365

Unrecorded Pinus radiata Pinus lambertinana Pinus bankisana

Pinned Frozen Frozen Frozen

AF113366 AF113367 AF113368 AF113369

Picea glauca

Frozen

AF113370

Picea glauca

Ethanol

Picea engelmannii

Frozen

2: AF113371 3: AF113372 AF113373

Collection locality, date and collector Canada: Ontario: Constance Bay, Turbolton Forest. 13-V-1995. FL: Pinellas Co., Harpon Spring, Howard Park. 12-V-1995. D. Czokajlo USSR: Kazachstan. Isyk 1767m Zailij. Ala Tau. 29-V-1974. A. Pfeffer Igt. Mexico: Nuevo Leon Mpio. Galeana Rd. Galeana-Dr. Arroyo, Las Crucites 2,470m. X-1993. S. Seybold AZ: Cochise Co. Coronado N.F. Rd. 42 10mi NW. Portal. 7-IX-1996. AZ: Greenlee Co. Rd. 191, 3 mi. N. Hannangan’s Meadow. 3-IX-1996. AZ: Greenlee Co. Rd. 191, 7 mi. S. Hannangan’s Meadow. 1-IX-1996. CO: Fairfeild nr. Payosa Springs N. Jose. VII-1994. S. Kelley CA: Mariposa Co. Fish Camp 15-IV-1996. WA: Roslyn. 30-VI-1997. AZ: Gila Co. San Carlos Indian Reservation. Point of Pines 27-V-1993. S. Seybold, A. I. Cognato & D. L. Wood AZ: Greenlee Co. Apache National Forest Rd. 191, 20 mi N. Morceani. 4-IX-1996. Bhutan: Autsu Lhuntshi. 1100 m. 9-X-1985. H. Schmutzenhofer Austria: Vienna. VI-1989. C. Stauffer CA: Marin Co. Olema. 8-III.1994. CA: Modoc Co. Warner Mountains, T44N, R15E, N1/2, S17. 25-VII-1994. S. Seybold, A. I. Cognato, D. L. Wood. Canary Islands: Tenerife. Villaflor. 12-III-1993. M. Knizek. CA: Alameda Co. Berkeley. Wildcat Cayon Rd. 14-X-1995. CA: Mariposa Co. Fish Camp 15-IV-1996. Canada: Ontario: Constance Bay, Torbolton Forest. 13-V1995. Canada: Ontario: Muni. of Ottawa-Carton, Marlborough Forest, 19-V-1995. D. E. Bright & A. I. Cognato Canada: Alberta. nr Manning. VIII-1996. D. Langor

Host

AZ: Greenlee Co. Rd. 191, 7 mi. S. Hannangan’s Meadow. 1-IX-1996 CO: Larmie Co. Roosevelt N.F., Hwy 14 10,000 ft. 18-VI1997. NM: Otero Co. Lincoln N.F. 11-V-1994. AZ: Apache Co. Apache N.F. Rd. 56. 3-IX-1996. Canada: Calgary 9-X-1990. J. Bordan CA: Marin Co. Inverness 14-VIII-1995. Bhutan: Chelaila, 3100 m. 5-VI-1986. H. Schmutzenhofer

Picea engelmannii

Frozen

AF113374

Pinus ponderosa Pinus ponderosa Pinus contorta Pinus muricata Larix griffithiana

Frozen Frozen Ethanol Frozen Pinned

AF113375 AF113376 AF113377 AF113378 AF113379

Czech Republic: Moravia, Tulasice V-1995. M. Knizek CA: Contra Costa Co. Diablo 3-IX-1995. Pakistan: Swat, Utror. 3-VIII-1987. C. Holzschuh

Unrecorded Pinus sabiniana Pinus gerardiana

Ethanol Frozen Pinned

AF113380 AF113381 AF113382

Nepal: Dhawalagiri, Kall-Gandaki-Khola, Mustang, D. Kalopani 2500-2800 m 3-VIII-1987. C. Holzschuh WA: Clallam Co. Hoh River Rd. 28-VI-1997.

Pinus gerardiana

Pinned

AF113383

Picea sitchensis

Frozen

AF113384

Czech Republic: Moravia. V-1995. M. Knizek

Unrecorded

Frozen

AF113385

Russia: Lenbolovo, St. Petersburg Reg. 8-10-VIII-1998. M. Mandelstan Russia: St. Khekhcir evirons, Khabarousk 2-IX-1990. M. Mandelstan

Spruce

Ethanol

AF113386

Pinus koraiensis

Pinned

AF113387

459

Ips mtDNA COI PHYLOGENY

APPENDIX—Continued

Species

Collection locality, date and collector

Ips woodi 1 Thatcher Ips woodi 2 Orthotomicus caelatus 1 & 2 (Eichhoff) Orthotomicus erosus (Wollaston) Orthotomicus laricis (Fabricius) Pityogenes calcaratus (Eichhoff) Pityogenes carinulatus (LeConte) Pityogenes hopkinsi Swaine Pityokteines sparsus (LeConte)

Host

Preservation method

Genbank Accession No.

AZ: Greenlee Co. Rd. 191, 7 mi. S. Hannangan’s Meadow. 1-IX-1996. CA: Mono Co. Inyo National Forest Crooked Creek Research Station 21-VIII-1997. NY: Onondago Co. nr. Syracuse 28-IX-1995. S. Teale

Pinus strobiformis

Frozen

AF113388

Pinus flexilis

Frozen

AF113389

Pinus resinosa

Frozen

Greece: Athens. V-1994.

Pinus

Ethanol

1: AF113390 2: AF113391 U82236

Romania: Retezat. Pietrele. 29-VIII-1982. M. Knizek

Unrecorded

Pinned

AF113392

Greece: Attica. Delase 17-VI-1994.

Pinus

Pinned

AF113394

NM: Otero Co. Lincoln N.F. 11-V-1994.

Pinus ponderosa

Frozen

AF113393

RI: Providence 18-VII-1997.

Pinus strobus

Frozen

AF113395

Canada: Quebec, Gatineau Park 14-V-1995.

Abies balsamea

Pinned

AF113396

ACKNOWLEDGMENTS We thank everyone who so generously provided us with Ips specimens (Appendix 1), including Arthur Fong (State of California, Department of Parks and Recreation) and Paddy Hardy (USDA Forest Service, White Mountain Ranger District) who facilitated collection permits. B. Mishler and D. L. Wood kindly reviewed early manuscripts and the helpful comments of M. Caterino, M. Cronin, K. Galacatos, D. Kain, R. Matthews, J. Wells, and an anonymous reviewer greatly improved this study. This work was financed by Hatch funds to F.S. and grants from the Margaret C. Walker Fund and Sigma Xi to A.I.C. This study was funded partially by a grant from the Pacific Southwest Research Station, USDA Service. We thank Nancy Rappaport, Pacific Southwest Research Station, who collaborated in the study and administered the grant.

REFERENCES Albert, V. A., Chase, M. W., and Mishler, B. D. (1993). Character-state weighting for cladistic analysis of protein-coding DNA sequences. Ann. Missouri Bot. Gard. 80: 752–766. Bright, D. E. (1993). Systematics of bark beetles. In ‘‘Beetle– Pathogen Interactions in Conifer Forests’’ (T. Schowalter and G. Flip, Eds.), pp.23–36. Academic Press, New York. Bright, D. E., and Skidmore, R. E. (1997). ‘‘A Catalog of Scolytidae and Platypodidae (Coleoptera), Supplement (1990–1994),’’ NRC Research Press, Ottawa. Bremer, K. (1994). Branch support and tree stability. Cladistics 10: 295–304. Brower, A. V. Z., and DeSalle, R. (1994). Practical and theoretical considerations for choice of a DNA sequence region in insect molecular systematics, with a short review of published studies using nuclear gene regions. Ann. Entomol. Soc. Am. 87: 702–716. Cane, J. H., Stock, M. W., Wood, D. L., and Gast, S. J. (1990). Phylogenetic relationships of Ips bark beetles (Coleoptera:Scolytidae): Electrophoretic and morphometric analysis of the grandicollis group. Biochem. Syst. Ecol. 19: 359–368. Caterino, M., and Sperling, F. A. H. (1999). Papilio phylogeny based on mitochondrial cytochrome oxidase I and II genes. Mol. Phylogenet. Evol. 11: 122–137.

Clary, D. O., and Wolstenholme, D. R. (1985). The mitochondrial DNA molecule of Drosophila yakuba: Nucleotide sequence, gene organization and genetic code. J. Mol. Evol. 22: 252–271. Cognato, A. I., Rogers, S. O., and Teale, S. A. (1995). Species diagnosis and phylogeny of the Ips grandicollis group (Coleoptera: Scolytidae) using Random Amplified Polymorphic DNA. Ann. Entomol. Soc. Am. 88: 397–405. Cummings, M. P., Otto, S. P., and Wakeley, J. (1995). Sampling properties of DNA sequence data in phylogenetic analysis. Mol. Biol. Evol. 12: 814–823. de Queiroz, K., and Gauthier, J. (1992). Phylogenetic taxonomy. Annu. Rev. Ecol. Syst. 23: 449–480. Farris, J. S. (1969). A successive approximations approach to character weighting. Syst. Zool. 18: 374–385. Fitch, W., and J., Ye. (1991). Weighted parsimony: Does it work? In ‘‘Phylogenetic Analysis of DNA Sequences’’ (M. M. Miyamoto and J. Cracraft, Eds.), pp. 147–154. Oxford Univ. Press, Oxford. Funk, D. J., Futuyma, D. J., Orti, G., and Meyer, A. (1995). Mitochondrial DNA sequences and multiple data sets: A phylogenetic study of phytophagous beetles (Chrysomelidae: Ophraella). Mol. Biol. Evol. 12: 627–640. Hennig, W. (1966). ‘‘Phylogenetic Systematics,’’ Univ. of Illinois Press, Urbana, IL. Holzschuh, C. (1988). Eine neue art der gattung Ips aus Bhutan (Coleoptera: Scolytidae). Entomol. Basiliensia 12: 481–485. Hopping, G. R. (1963a). The natural groups of Ips De Geer (Coleoptera: Scolytidae). Can. Entomol. 95: 508–516. Hopping, G. R. (1963b). The North American species in group I of Ips DeGeer (Coleoptera: Scolytidae). Can. Entomol. 95: 1091–1096. Hopping, G. R. (1963c). The North American species in groups II and III of Ips DeGeer (Coleoptera: Scolytidae). Can. Entomol. 95: 1202–1210. Hopping, G. R. (1964). The North American species in group IV and V of Ips DeGeer (Coleoptera: Scolytidae). Can. Entomol. 96: 970–978. Hopping, G. R. (1965a). The North American species in group VI of Ips DeGeer (Coleoptera: Scolytidae). Can. Entomol. 97: 533–541. Hopping, G. R. (1965b). The North American species in group VII of Ips DeGeer (Coleoptera: Scolytidae). Can. Entomol. 97: 193–198.

460

COGNATO AND SPERLING

Hopping, G. R. (1965c). The North American species in group VIII of Ips DeGeer (Coleoptera: Scolytidae). Can. Entomol. 97: 159–172. Hopping, G. R. (1965d). The North American species in group IX of Ips DeGeer (Coleoptera: Scolytidae). Can. Entomol. 97: 422–434. Hopping, G. R. (1965e). The North American species in group X of Ips DeGeer (Coleoptera: Scolytidae). Can. Entomol. 97: 803–809. Juan, C., Oromi, P., and Hewitt, G. M. (1995). Mitochondrial DNA phylogeny and sequential colonization of Canary Islands by darkling beetles of the genus Pimelia (Tenebrionidae). Proc. R. Soc. Lond. B. 261: 173–180. Kishino, H., and Hasegawa, M. (1989). Evaluation of the maximum likelihood estimate of the evolutionary tree topologies from DNA sequence data and branching order in Hominoidea. J. Mol. Biol. 157: 105–132. Langor, D. W., and Sperling, F. A. H. (1997). Mitochondrial DNA sequence divergence in weevils of the Pissodes strobi species complex (Coleoptera: Curculionidae). Insect Mol. Biol. 6: 255–265. Lanier, G. N. (1966). Interspecific matings and cytological studies of closely related species of Ips and Orthotomicus Ferrari (Coleoptera: Scolytidae). Can. Entomol. 98: 175–188. Lanier, G. N. (1970a). Biosystematics of the genus Ips (Coleoptera: Scolytidae) in North America: Hopping’s group III. Can. Entomol. 102: 1404–1423. Lanier, G. N. (1970b). Biosystematics of North American Ips (Coleoptera: Scolytidae): Hopping’s group IX. Can. Entomol. 102: 1139–1163. Lanier, G. N. (1972). Biosystematics of the genus Ips (Coleoptera: Scolytidae) in North American: Hopping’s groups IV and X. Can. Entomol. 104: 361–388. Lanier, G. N. (1987). The validity of Ips cribricollis (Eichhoff) as distinct from I. grandicollis (Eichhoff) and occurrence of both species in Central America. Can. Entomol. 119: 179–187. Lanier, G. N., and Oliver, J. H. (1966). ‘‘Sex-Ratio’’ condition: Unusual mechanisms in bark beetles. Science 153: 208–209. Lanier, G. N., and Burkholder, W. E. (1974). Pheromones in speciation of Coleoptera. In ‘‘Pheromones’’ (M. C. Birch, Ed.), pp. 161–189. North-Holland, London. Lanier, G. N., and Wood, D. L. (1975). Specificity of response to pheromones in the genus Ips (Coleoptera: Scolytidae). J. Chem. Ecol. 1: 9–23. Lanier, G. N., Teale, S. A., and Parjares, J. A. (1991). Biosystematics of the genus Ips (Coleoptera: Scolytidae) in North American: Review of Ips calligraphus group. Can. Entomol. 23: 1103–1124. Lecointre, G., Philippe, H., Le, H. L. V., and Guyader, H. L. (1993). Species sampling has a major impact on phylogenetic inference. Mol. Phylogenet. Evol. 2: 205–224. Lewis, E. E., and Cane, J. H. (1990). Pheromonal specificity of southeastern Ips pine bark beetles reflects phylogenetic divergence (Coleoptera: Scolytidae). Can. Entomol. 122: 1235–1238. Lunt, D. H., Zhang, D.-X., Szymura, J. M., and Hewitt, G. M. (1996). The insect cytochrome oxidase I gene: Evolutionary patterns and conserved primers for phylogenetic studies. Insect Mol. Biol. 5: 153–165. Mitton, J. B., and Sturgeon, K. B. (1982). ‘‘Bark Beetles in North American Conifers: A System for the Study of Evolutionary Biology,’’ Univ. of Texas Press, Austin. Neff, N. A. (1986). A rational basis for a priori character weighting. Syst. Zool. 35: 110–123. Normark, 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. Page, M., Nelson, L. J., Blomquist, G. J., and Seybold, S. J. (1997). Cuticular hydrocarbons as chemotaxonomic characters of pine engraver beetles (Ips spp.) in the grandicollis subgeneric group. J. Chem. Ecol. 23: 1053–1099.

Pfeffer, A. (1995). ‘‘Zentral- und Westpala¨arktische Borken- und Kernka¨fer,’’ Pro Entomologia, c/o Naturhistorisches Museum, Basel. Postner, M. (1974). Scolytidae (⫽Ipidae), Borkenkafer. In ‘‘Die Forstschadlinge Europas’’ (W. Schwenke, Ed.), Vol. 2, pp. 334–487. Hamburgund, Berlin. Reed, R. D., and Sperling, F. A. H. (1999). Interaction of process partitions in phylogenetic analysis: An example from the swallowtail butterfly genus Papilio. Mol. Biol. Evol. 16: 286–297. Schmutzenhofer, H. (1988). Mass outbreaks of Ips bark beetles in Bhutan and the revision of the genus Ips deGeer for the Himalayan region. In ‘‘Integrated Control of Scolytid Bark Beetles’’ (T. L. Payne and H. Saarenmaa, Eds.), pp. 345–355. Virginia Polytechnic Inst. and State Univ. Seybold, S. J. (1992). ‘‘The role in the Olfactory-Directed Aggregation Behavior of Pine Engraver Beetles in the Genus Ips (Coleoptera: Scolytidae),’’ Ph.D. dissertation, Univ. of California, Berkeley. Seybold, S. J., Ohtsuka, T., Wood, D. L., and Kubo, I. (1995). The enantiomeric composition of ipsdienol: A chemotaxonomic character for North American populations of Ips spp. in the pini subgeneric group (Coleoptera: Scolytidae). J. Chem. Ecol. 21: 995–1016. Simon, C., Frati, F., Beckenbach, A., Crespi, B., Liu, H., and Flook, P. (1994). Evolution, weighting, and phylogenetic utility of mitochondrial gene sequences and a compilation of conserved polymerase chain reaction primers. Ann. Entomol. Soc. Am. 87: 651–701. Sperling, F. S., Byers, J. R., and Hickey, D. A. (1996). Mitochondrial DNA sequence variation among pheromotypes of the dingy cutworm, Feltia jaculifera (Gn.) (Lepidoptera: Noctuidae). Can. J. Zool. 74: 2109–2117. Stauffer, C., Lakatos, F., and Hewitt, G. M. (1997). The phylogenetic relationships of seven European Ips (Scolytidae, Ipinae) species. Insect Mol. Biol. 6: 1–8. Swofford, D. L. (1998). PAUP*: Phylogenetic Analysis Using Parsimony, 4.0b.1. Smithsonian Institute, Washington, DC. Swofford, D. L., Olsen, G. J., Waddell, P. J., and Hillis, D. M. (1996). Phylogenetic inference. In ‘‘Molecular Systematics’’ (D. M. Hillis, C. Moritz, and B. K. Mable, Eds.), pp. 407–425. Sinauer, Sunderland, MA. Tamura, K., and Nei, M. (1993). Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Mol. Biol. Evol. 10: 512–526. Vite´, J. P., Bakke, A., and Renwick, J. A. A. (1972). Pheromones in Ips (Coleoptera: Scolytidae): Occurrence and production. Can. Entomol. 104: 1967–1975. Vogler, A. P., and Welsh, A. (1997). Phylogeny of North American Cicindela tiger beetles inferred from multiple mitochondrial DNA sequences. Mol. Phylogenet. Evol. 8: 225–235. Wenzel, J. W. (1997). When is a phylogenetic test good enough? In ‘‘The Origin of Biodiversity in Insects: Phylogenetic Tests of Evolutionary Scenarios’’ (P. Grandcolas, Ed.), pp. 31–45. Mem. Mus. Natn. Hist. Nat. 173. Wood, D. L. (1982). The role of pheromones, kairomones, and allomones in the host selection and colonization behavior of bark beetles. Annu. Rev. Entomol. 27: 411–446. Wood, S. L. (1977). New synonymy and new species of American bark beetles (Coleoptera: Scolytidae), part V. Great Basin Nat. 37: 78–88. Wood, S. L. (1982). ‘‘The Bark and Ambrosia Beetles of North America (Coleoptera: Scolytidae), A Taxonomic Monograph,’’ Great Basin Nat. Mem. 6. Wood, S. L. (1992). Nomenclatural changes and new species of Platypodidae and Scolytidae (Coleoptera), part II. Great Basin Nat. 52: 78–88. Wood, S. L., and Bright, D. E. (1992). ‘‘A Catalog of Scolytidae and Platypodidae (Coleoptera), Part 2: Taxonomic Index, Volumes A and B,’’ Great Basin Nat. Mem. 13.