PHYLOGENY OF ASTYLOPSIS CASEY (COLEOPTERA: CERAMBYCIDAE) SPECIES AND PATTERNS OF HOST PLANT USE By Ellen M. Camerato A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Entomology – Master of Science 2021 ABSTRACT PHYLOGENY OF ASTYLOPSIS CASEY (COLEOPTERA: CERAMBYCIDAE) SPECIES AND PATTERNS OF HOST PLANT USE By Ellen M. Camerato Cerambycid (longhorn) beetles are diverse in their morphology and life history traits, but all share the common trait of being larval plant-borers. The larvae bore into and develop inside various plant tissues thus they can potentially cause significant economic and ecological damage, especially when transported to non-native localities. There is little empirical data on cerambycid life history traits that are essential in understanding their ecological and economic effects. Astylopsis Casey (Lamiinae: Acanthocinini) is an eastern North American genus of six species. Host preference varies greatly among the species, including both angiosperms and gymnosperms. I used morphological characters and molecular data to reconstruct phylogenies of Astylopsis to test the hypothesis that host plant use among Astylopsis species is conserved. I constructed phylogenies using partial COI and CAD DNA sequences from Astylopsis species and outgroups using parsimony methods. Astylopsis collaris, A. macula, A. sexguttata, and A. arcuata were monophyletic in both COI and combined gene phylogenies, with the genus also exhibiting monophyly in the combined gene tree. Evidence of host shift from angiosperms to gymnosperms in some species was also observed. These results confirm current taxonomic separations among the four species and their outgroups and provide important host use information. No conclusions could be drawn regarding DNA variation in association with geographic locality. These findings will inform future studies expanding the molecular dataset for Astylopsis with additional genes (arginine kinase, 28S, and EF1-α) and species (Astylopsis perplexa and A. fascipennis). ACKNOWLEDGMENTS I would like to acknowledge the United States Department of Agriculture Animal and Plant Health Inspection Service for funding this project. I would like to thank my supervisor, Dr. Anthony Cognato, for his guidance in each stage of the process, as well as my committee members Dr. Henry Chung and Dr. Alan Prather for their valuable feedback and advice. I would also like to thank Dr. Patrick Gorring and Dr. Sarah Smith for assistance and direction in my project methodology. Lastly, I would like to express my appreciation for my family and friends for their encouragement and endless support during my degree, and my cats for their comfort and companionship while working from home. iii TABLE OF CONTENTS LIST OF TABLES .....................................................................................................................v LIST OF FIGURES ................................................................................................................. vi INTRODUCTION .....................................................................................................................1 MATERIALS AND METHODS...............................................................................................4 Specimens ......................................................................................................................4 Morphology....................................................................................................................4 DNA Sequencing ...........................................................................................................5 Phylogenetic Analyses .................................................................................................. 6 RESULTS ..................................................................................................................................7 Preliminary Phylogeny...................................................................................................7 Morphological Phylogeny..............................................................................................7 Molecular Phylogenies...................................................................................................7 DISCUSSION ............................................................................................................................9 APPENDIX...............................................................................................................................11 REFERENCES ........................................................................................................................22 iv LIST OF TABLES Table 1. Institutions from which Astylopsis specimens were borrowed...................................12 Table 2. Intraspecific character matrix for Astylopsis species A. collaris, A. macula, A. sexguttata, A. arcuata, A. perplexa, and A. fascipennis. Characters are ordered as mentioned in the text................................................................................................................13 Table 3. Forward and reverse primers used for DNA sequencing of Astylopsis and outgroup beetles.......................................................................................................................14 Table 4. Thermal cycler parameters for PCR of each gene. All temperatures are Celsius and all time is displayed as minutes : seconds. PCR was performed using tissue samples from specimens preserved in ethanol or dried specimens collected within the previous 20 years. The amplified PCR products were obtained in order to be sequenced and compared among species......................................................................................................................................15 Table 5. Sequence identifiers for sequences sourced from Barcode of Life Data System (BOLD). Sequence label corresponds with labels in Figure 2................................................15 Table 6. Interspecific and intraspecific p-distances between species in the preliminary phylogeny. P-distance measures represent percent difference between nucleotide sites. P-distance measures between sets of two individuals were calculated in PAUP* and then averaged to obtain the overall percent distance between nucleotide sites between the two species being compared...........................................................................................................16 v LIST OF FIGURES Figure 1. Map of the eastern United States depicting known geographic ranges of each Astylopsis species (Dillon 1956a, Linsley and Chemsak 1995, and Monné and Nearns 2020).........................................................................................................................................17 Figure 2. Preliminary phylogenetic tree using barcoded COI sequences for Astylopsis sexguttata, A. collaris, and A. macula, constructed in PAUP* using parsimony methods. Individuals are labeled with their species, serial numbers to differentiate individuals, and locality. Locality abbreviations: AR Arkansas, USA; QC Quebec, Canada; PEI Prince Edward Island, Canada; ON Ontario, Canada ..............................................................18 Figure 3. Phylogeny constructed from morphological data from 58 individuals (ten individuals each of Astylopsis collaris, A. macula, A. sexguttata, A. arcuata, and A. fascipennis, and eight Astylopsis perplexa individuals). This tree was constructed in PAUP* using a heuristic search with parsimony methods .....................................................19 Figure 4. Phylogeny of A. collaris, A. macula, A. sexguttata, and A. arcuata individuals using barcoded COI sequences, indicated by an asterisk, and experimental COI sequences for the species. This tree was constructed in PAUP* using a heuristic search with parsimony methods. Eupromus ruber, Acanthocinus obsoletus, Leptostylopsis planidorsus, Pseudastylopsis pini, Astylidius parvus, and Astyleiopus variegatus were included as outgroups, rooted with E. ruber. Darker values above branch lines indicate bootstrap support for those relationships; lighter grey values below lines indicate Bayesian posterior probabilities. Tree icons indicate whether the species uses broad-leaved or coniferous hosts. Clades without bootstrap values were not found in the strict consensus of all parsimonious trees. Locality abbreviations: AR Arkansas, USA; QC Quebec, Canada; MA Massachusetts, USA; MI Michigan, USA; FL Florida, USA; LA Louisiana, USA; PEI Prince Edward Island, Canada; TN Tennessee, USA; ON Ontario, Canada; NY New York, USA; PA Pennsylvania, USA; AZ Arizona, USA; JAP Japan .............................20 Figure 5. Phylogenetic tree visualization of A. collaris, A. macula, A. sexguttata, and A. arcuata individuals using experimental sequences of genes COI and CAD, represented by one of three parsimonious trees found in a heuristic search. Eupromus ruber, Astyleiopus variegatus, Acanthocinus obsoletus, Leptostylopsis planidorsus, Astylidius Parvus, and Pseudastylopsis pini were included as outgroups, rooted with E. ruber. Darker values above branch lines indicate bootstrap support for those relationships; lighter grey values below lines indicate Bayesian posterior probabilities. Tree icons indicate whether the species uses broad-leaved or coniferous hosts. Clades without bootstrap values were not found in the strict consensus of the three trees. Locality abbreviations: FL Florida, USA; MA Massachusetts, USA; MI Michigan, USA; LA Louisiana, USA; NY New York, USA; PA Pennsylvania, USA; TN Tennessee, USA; AZ Arizona, USA; JAP Japan .................................................................................................21 vi INTRODUCTION Cerambycid (longhorn) beetles are diverse in their morphology and distribution, host plant selection, feeding, mating behavior, and life cycle ecology, but they all share the common trait of being larval plant-borers. Comprising over 36,000 species (Haack et al. 2017), members of Cerambycidae have radiated to all continents except for Antarctica. They thrive in habitats from sea level to an elevation of 4,200 meters in the mountains of China and Bolivia (Haack et al. 2017, Kariyanna et al. 2017, and Linsley 1959). Nearly all adults and larvae alike are phytophagous, selecting primarily woody plants and trees as their hosts, including both angiosperms and gymnosperms. The larvae of different lineages bore into and develop in different plant parts such as roots, stems, and branches; bark, sapwood, heartwood, and pith. Additionally, different genera attack trees at various growth stages (live, weakened, dying, dead), and can live in wet, damp, or dry wood (Haack et al. 2017, Kariyanna et al. 2017, and Linsley 1959). Mating usually occurs on the host plant and cerambycids generally do not participate in any courtship behavior: the male will simply approach the female and attempt to mate (Kariyanna et al. 2017). Pheromones play a role in mate selection and host plant location for oviposition. The few cerambycid species that rely on long-range pheromones tend to have shorter antennae with increased surface area with serrate, lamellate, or pectinate forms, observed in some males in subfamily Prioninae. A few other species sensitive to long-range pheromones also have shorter antennae but exhibit less sexual dimorphism (Kariyanna et al. 2017). Female cerambycids will oviposit in, on, or near the host plant. Members of Lamiinae are unique in that they use their mouthparts and/or ovipositors to create an opening in the outermost tissues of the host tree and lay their eggs inside. Larvae later tunnel into the host’s tissues, or the surrounding 1 soil if this was the oviposition site, and feed and develop for months to years. In many cases, pupation occurs in the same location as larval development and can last from days to months, depending on species and ambient temperature (Haack et al. 2017 and Kariyanna et al. 2017). Given their habit of larval development inside plant tissues, cerambycids can potentially cause significant economic and ecological damage, especially when transported to non-native localities (Cocquempot and Lindelöw 2010 and Wu et al. 2017). There are cerambycid taxa with little empirical data on the life history traits that are essential in understanding their ecological and economic effects. Astylopsis Casey (Lamiinae: Acanthocinini) is a genus native to eastern North America with six species (A. macula, A. collaris, A. sexguttata, A. arcuata, A. perplexa, and A. fascipennis) about which little information is known regarding their host use. Astylopsis adults vary between 6-15 mm in length and generally bear reddish to brown integument pubescent with varying brown, black, grey, and white mottling. These elytral patterns are often used to diagnose the species. Their known forest habitats range from eastern coastal United States, north into Canada, south to Florida, and west to Kansas and Texas (Fig. 1). Host preference varies greatly among the species, which is unusual within cerambycid genera. Astylopsis perplexa, for example, bores into the stems of Baccharis halimifolia, the sea-myrtle shrub. In contrast, A. sexguttata and A. arcuata both prefer coniferous trees such as pine, spruce, and larch, while A. collaris and A. macula primarily select hardwood trees (Dillon 1956a, Linsley and Chemsak 1995, and Schiefer 2000). Astylopsis fascipennis is the most recently described species and its larval host is unknown, however, it was collected in hardwood- dominated forests and specifically collected from elm and dead sweetgum branches using beating sheets (Shiefer 2000). 2 Little is known of the evolutionary relationships among Astylopsis species and their associations with host plants. To address this knowledge gap, I reconstructed phylogenies for some Astylopsis species using DNA sequences of partial mitochondrial gene cytochrome oxidase subunit I (COI) and partial nuclear protein carbamoyl-phosphate synthetase 2, aspartate transcarbamylase, and dihydroorotase (CAD) and morphological characters. Specimens were sourced from the Albert J. Cook Arthropod Research Collection or borrowed from other institutional collections in the states where these beetles are native. I used the most parsimonious trees to map host plant use, scored as broad categories such as conifers and hardwoods, on the resulting phylogenies to test the hypothesis that host plant use among Astylopsis species is conserved. 3 MATERIALS AND METHODS In order to understand the evolutionary history of the genus Astylopsis and test the hypothesis that host plant use is conserved, I studied individuals of each species for variation in morphology, sequenced their DNA, and constructed phylogenetic trees with these data. Specimens Due to the COVID-19 pandemic in 2020-2021, I was unable to travel out of state to collect specimens for my project. Instead, Astylopsis specimens were sourced from the MSU A.J. Cook Arthropod Research Collection and borrowed from the institutes listed in Table 1. Additional specimens for molecular analysis were collected by hand and preserved directly into >90% ethanol or using flight interception traps through the US Forest Service Early Detection Rapid Response program (Rabaglia et al. 2019). Specimens of Pseudastylopsis pini Schaeffer, Astylidius parvus LeConte, Astyleiopus variegatus Haldeman, Leptostylopsis planidorsus LeConte, Acanthocinus obsoletus Olivier, and Eupromus ruber Dalman were included as outgroups. Morphology Astylopsis specimens were compared for interspecific and intraspecific variation in morphological characters. Ten specimens from each species (except eight specimens for A. perplexa) were examined with a Leica (Wetzlar, Germany) MZ6 compound microscope illuminated with two Ikea (Delft, Netherlands) JANSJÖ LED lamps. The individuals were scored for variation in external morphological characters and placed into a character matrix (Table 2). The matrix was scored according to the following list of characters and states. 4 1. Cuticle color (dorsal, anterior views, 18x): (0) Cinnamon Buff; (1) Cinnamon; (2) Clay Color; (3) Mikado Brown; (4) Sayal Brown; (5) Tawny Olive; (6) Verona Brown; (7) Snuff Brown 2. Pubescence color (dorsal view, 18x): (0) Pale Pinkish Buff; (1) Light Vinaceous-Cinnamon; (2) Light Pinkish Cinnamon 3. Pubescence density, % (dorsal, anterior views, 18x): (0) 70-79; (1) 80-89; (2) 90-100 4. Ventral pubescence density, % (ventral view, 18x): (0) 70-79; (1) 80-89; (2) 90-100 5. Dorsoventral flatness, mm (left lateral view, 7x): (0) 1.0-1.9; (1) 2.0-2.9; (2) 3.0-3.9 6. Diameter of median callus, mm (dorsal view, 7x): (0) 0.2-0.3; (1) 0.4-0.5; (2) 0.6 Characters 1 and 2 were scored corresponding to colors on plate 29 of a PDF file of Ridgway (1912) displayed on a 2014 Apple (Cupertino, California, United States) MacBook Pro Retina with Intel Iris 1536 MB graphics at 75% brightness with the Color LCD option selected. Characters 3 and 4 were scored as a percentage of the whole visible area that is covered by pubescence, and characters 5 and 6 were measured in millimeters with a micrometer in the stereoscope. Data was analyzed and used to construct a tree in PAUP* 4.0 b10 PPC (Swofford 2001) and the tree was edited for clarity in Inkscape v1.0.2. DNA Sequencing DNA was extracted from one leg (trochanter, femur, tibia, and tarsi) of each beetle using a Qiagen DNEasy blood and tissue kit (Hilden, Germany) with the protocol from the manufacturer. The purified DNA was used in a PCR of the partial mitochondrial gene cytochrome C oxidase subunit I (COI) and partial nuclear protein carbamoyl-phosphate synthetase 2, aspartate transcarbamylase, and dihydroorotase (CAD) using various primer pairs 5 (Table 3). I obtained PCR products with 12.5-µl ddH2O, 2-µl 5x PCR Buffer (Qiagen), 2-µl MgCl2 (Qiagen), 0.4-µl dNTP mix (Qiagen), 1-µl template DNA, 0.1-µl HotStar Taq DNA polymerase (Qiagen), and 1-µl each of forward and reverse primers. PCR was performed on a thermal cycler (PTC-2000, MJ Research, Waltham, Massachusetts, USA), or MyCycler Thermocycler (BioRad, Hercules, California, USA) under various conditions (Table 4). Products were then treated with Exo-SAP (USB Corp., Cleveland, Ohio, USA) and sequenced at the Michigan State University Research Technology Support Facility using BigDye Terminator v1.1 (Applied Biosystems, Foster City, California, USA). I visualized forward and reverse DNA sequences in Sequencher (Ann Arbor, Michigan, USA) to trim primer sequences, edit ambiguities, and create consensus sequences. Phylogenetic Analyses I constructed phylogenetic trees from the morphology matrix and DNA sequences as well as the publicly available barcode sequences archived in the Barcode of Life Data System (BOLD) (Table 5) (Ratnasingham and Hebert 2007). PAUP* with parsimony methods was used to generate the trees. Heuristic searches with 100 stepwise random additions with tree bisection- reconnection were performed and characters were unordered and equally weighted for the parsimony analysis. Bootstrap values under parsimony for the same parameters as above were calculated by performing 500 pseudoreplicates with simple additions in PAUP*. Nucleotide difference among sequences was measured as uncorrected p-distance. Bayesian confidence values for phylogenetic relationships were obtained with MRBAYES software (Huelsenbeck and Ronquist 2001). Tree files were edited for better visualization and clarity in FigTree v1.4.4 and Inkscape v1.0.2. 6 RESULTS Preliminary Phylogeny A preliminary phylogeny was constructed using barcoded COI sequences of Astylopsis specimens through BOLD to determine whether there was intraspecific variation based on locality differences as well as the evolutionary placement of each species in the overall genus clade. Sequences from the three most commonly encountered species, Astylopsis sexguttata, A. collaris, and A. macula, were available for this preliminary analysis (Fig. 2). This phylogeny exhibited monophyly for A. collaris and A. macula, and A. sexguttata. Intraspecific p-distance values (Table 6) were 4.3% or less. Interspecific p-distance was 14% between A. macula and A. collaris, and there was 17% nucleotide difference between A. macula and A. sexguttata, as well as between A. collaris and A. sexguttata. Morphological Phylogeny The phylogeny constructed in PAUP* of the morphology data (Fig. 3) showed no resolution among species, having had very little variation among individuals in the characters that I chose. Bootstrap values were not greater than 50%. Molecular Phylogenies Another tree was constructed with the COI sequences from the preliminary phylogeny and a set of newly generated COI sequences from A. collaris, A. macula, A. sexguttata, and A. arcuata (Fig. 4). Outgroups Astyleiopis variegatus, Astylidius parvus, Pseudastylopsis pini, Leptostylopsis planidorsus, Acanthocinus obsoletus, and Eupromus ruber were included in this tree. Each Astylopsis species was monophyletic. Astylopsis collaris, A. sexguttata, and A. macula 7 all exhibited bootstrap support of 100% and Bayesian posterior probabilities of 1.00. The A. sexguttata clade exhibits a sister group relationship with the one A. arcuata individual, having bootstrap support of 96% and Bayesian confidence of 1.00. Two barcoded sequences labeled A. sexguttata appeared in a separate clade with the outgroup Astyleiopus variegatus. In a more comprehensive dataset including both mitochondrial and nuclear genes, the genus and all species were also monophyletic (Fig. 5). Astylopsis exhibited 99% bootstrap support and Bayesian confidence of 1.00. Astylopsis sexguttata, A. collaris, and A. macula each exhibited 100% bootstrap support and Bayesian confidence of 1.00. The Astylopsis arcuata individual formed a clade with A. sexguttata in this tree, with 98% bootstrap support and Bayesian confidence of 0.99. 8 DISCUSSION Despite the diversity and pest importance of cerambycid beetles, data is still lacking on their evolutionary relationships and host use. This study further delineated this information in the lamiine genus Astylopsis, providing a view of morphological and molecular phylogenetic relationships. The morphological data examined showed no pattern among the six Astylopsis species, indicating that morphology is very conserved among individuals in each species, however it is not phylogenetically informative. Examination of partial COI and CAD sequence data was more informative. There was no resolution at the intraspecific level for Astylopsis sexguttata in either of the strict consensus trees, which is expected as there are few character differences at this level. This was also supported with p-distance values found in a preliminary tree in that intraspecific nucleotide difference for COI were 2.5% for A. macula, 0.5% for A. collaris, and 1.4% for A. sexguttata. Interspecific variation for these three species was 14-17%, which is consistent with uncorrected pairwise distances measured among cerambycid species in previous studies (Farrell 2001 and Nearns 2013). In the COI tree, two A. sexguttata individuals from BOLD formed a clade with Astyleiopus variegatus, separate from the rest of the Astylopsis members. This is likely due to these individuals having been misidentified and correctly belonging to the outgroup, considering their bootstrap support and Bayesian confidence values of 100 and 1.00, respectively. COI phylogenetic data supported my hypothesis of monophyly for each Astylopsis species included (A. collaris, A. macula, A. sexguttata, and A. arcuata), confirming their current separations. The hypothesis of monophyly for each species, as well as the genus as a whole, was also supported in the combined tree. Monophyletic clades have been resolved in similar studies of cerambycid taxa at the subfamily, tribe, and genus level that are consistent with current separations (Farrell 2001, 9 Gutiérrez et al. 2020, Lee and Lee 2020, Li et al. 2016, Nakamine and Takeda 2008, Ohbayashi et al. 2009, Souza et al. 2020, and Toki and Kubota 2010). Despite varying localities among individuals in the COI phylogeny, no real conclusions could be formed on variation in DNA based on locality. Both trees also exhibited informative relationships based on host plant use. Astylopsis sexguttata and A. arcuata are the conifer-feeding members of this genus, which is reflected in the clade that they formed with high bootstrap support and Bayesian posterior probability in both the COI and combined gene trees. Hardwood feeding species A. collaris and A. macula show earlier divergence than the conifer feeders, indicating hardwood trees represent the more ancestral host selected by Astylopsis beetles. Although conifers predate broad-leaved trees evolutionarily, a shift from angiosperms to derived gymnosperm feeding has been found in lamiine cerambycids in the tribe Lamiini (Toki and Kubota 2010), and also within subfamily Cerambycinae (Lee and Lee 2020). This pattern of host use may be more common, as it is also observed in scolytine bark beetles (Kirkandall et al. 2015). This host shift is not surprising, as the more ancestral ougroups use both angiosperm and gymnosperm hosts. Derived conifer-feeding Astylopsis species also may have acquired special adaptations to overcome conifer properties such as resin and lignin. Overall, the relationships shown in these phylogenies confirm these four Astylopsis species and outgroups as being distinct taxa and provide another example of the host shift from angiosperms to gymnosperms. These results also confirm that our methods will continue to provide meaningful phylogenetic information on the relationships between Astylopsis species when incorporating more genes (Wingless, arginine kinase, 28S, and EF1-α) and other Astylopsis species (A. perplexa and A. fascipennis). 10 APPENDIX 11 APPENDIX Table 1. Institutions from which Astylopsis specimens were borrowed. Collection, Institution Location Acronym # Loan Corresponding Specimens Curator Clemson University Arthropod Clemson, South CUAC 59 Mike Ferro Collection Carolina Florida Department of Gainesville, FDACS 8 Paul Skelley Agriculture and Consumer Florida Services, Florida State Collection of Arthropods Georgia Museum of Natural Athens, Georgia GMNH 40 E. Richard History, University of Georgia Hoebeke Kansas University Biodiversity Lawrence, KUBI 22 Zachary Falin Institute Kansas Albert J. Cook Arthropod East Lansing, MSUC 30 Gary Parsons Research Collection, Michigan Michigan State University Mississippi Entomological Mississippi MEM 92 Terence Museum, Mississippi State State, Schiefer Mississippi Texas A&M University Insect College Station, TAMU 163 Karen Wright Collection Texas University of Texas Austin, Texas UTBC 4 Alex Wild Biodiversity Center Wisconsin Insect Research Madison, WIRC 58 Craig Brabant Collection, University of Wisconsin Wisconsin-Madison 12 Table 2. Intraspecific character matrix for Astylopsis species A. collaris, A. macula, A. sexguttata, A. arcuata, A. perplexa, and A. fascipennis. Characters are ordered as mentioned in the text. ID 1 2 3 4 5 6 A. collaris 1 6 2 2 2 1 1 2 7 2 2 2 1 1 3 3 2 2 2 1 1 4 3 2 0 1 1 1 5 6 2 2 2 1 0 6 6 2 2 2 1 1 7 6 2 2 2 1 1 8 4 2 2 2 1 0 9 5 2 2 2 1 0 10 4 1 2 2 1 1 A. macula 11 4 2 2 2 2 1 12 3 2 2 2 1 1 13 3 2 2 2 1 1 14 2 2 2 1 1 1 15 3 2 1 2 1 1 16 4 2 2 1 0 0 17 3 2 2 1 1 1 18 0 2 2 1 1 0 19 4 2 1 0 0 1 20 6 2 2 1 1 1 A. sexguttata 21 6 2 2 2 0 1 22 4 1 2 2 1 1 23 4 2 2 2 1 1 24 7 2 2 2 1 1 25 6 2 2 2 0 1 26 3 2 2 2 1 1 27 1 1 2 2 1 1 28 3 2 2 1 1 2 29 6 1 2 2 1 1 30 3 2 2 2 1 1 A. arcuata 31 3 0 2 2 1 1 32 4 2 2 2 1 1 33 5 0 2 2 1 1 34 4 2 2 2 1 1 13 Table 2. (cont.) 35 2 2 2 2 1 1 36 7 2 1 1 1 1 37 3 2 2 2 1 1 38 3 2 2 2 2 1 39 5 0 2 2 1 1 40 4 2 2 2 1 1 A. perplexa 41 1 1 2 1 2 2 42 7 2 2 2 2 1 43 6 2 2 2 2 1 44 1 1 2 2 2 2 45 3 2 2 2 1 1 46 4 1 2 2 1 1 47 6 1 2 2 2 1 48 4 1 2 2 1 1 A. fascipennis 49 3 2 1 2 0 1 50 3 2 2 2 1 1 51 6 2 2 2 1 1 52 3 2 2 2 1 1 53 6 2 2 2 0 1 54 6 2 2 2 1 1 55 6 1 2 2 1 1 56 6 2 2 2 1 1 57 1 1 2 2 1 0 58 6 2 2 2 1 1 Table 3. Forward and reverse primers used for DNA sequencing of Astylopsis and outgroup beetles. Gene Primer Sequence Source COI LCO1490 GGTCAACAAATCATAAAGATATTGG Hebert et al. 2003 HCO2198 TAAACTTCAGGGTGACCAAAAATCA Hebert et al. 2003 CAD CD338F ATGAARTAYGGYAATCGTGGHCAYAA Moulton and Wiegmann 2004 CD668R ACGACTTCATAYTCNACYTCYTTCCA Wild and Maddison 2008 14 Table 4. Thermal cycler parameters for PCR of each gene. All temperatures are Celsius and all time is displayed as minutes : seconds. PCR was performed using tissue samples from specimens preserved in ethanol or dried specimens collected within the previous 20 years. The amplified PCR products were obtained in order to be sequenced and compared among species. Gene Initial Denaturation Annealing Elongation Final Hold Denaturation Elongation COI 95 94 50 72 72 12 15:00 0:30 1:00 1:30 5:00 ¥ [ 37 cycles ] CAD 94 94 50 72 72 12 2:00 0:30 1:00 1:30 5:00 ¥ [ 37 cycles ] Table 5. Sequence identifiers for sequences sourced from Barcode of Life Data System (BOLD). Sequence label corresponds with labels in Figure 2. Sequence Label (as appears in Fig. 2) BOLD Sequence ID Astylopsis sexguttata1 BBCCA3340-12 Astylopsis sexguttata2 CERLF444-08 Astylopsis collaris1 CERLF432-08 Astylopsis collaris2 CERLF433-08 Astylopsis collaris3 CERLF435-08 Astylopsis collaris4 CERLF434-08 Astylopsis macula1 CERLF445-08 Astylopsis macula2 CERLF446-08 Astylopsis macula3 CERLF447-08 Astylopsis macula4 CERLF448-08 Astylopsis macula5 CERLF623-08 Astylopsis macula6 CERLF624-08 Astylopsis macula7 CERLF625-08 Astylopsis macula8 CERLF626-08 Astylopsis macula9 CNLMF980-14 KR123196 Astylopsis macula10 CNLMP2009-14 KR127106 Astylopsis macula11 CNLMR092-14 KR126679 Astylopsis macula12 CNPEP1441-14 KR130747 Astylopsis macula13 CNPER012-14 KR124596 Astylopsis CNROS762-13 KR129993 Astylopsis macula14 CNSLI559-12 KM844100 Astylopsis macula15 CNTIE051-15 MF638667 Astylopsis macula16 OPPQE395-17 Astylopsis macula17 OPPQE399-17 15 Table 5. (cont.) Astylopsis macula18 OPPQE400-17 Astylopsis macula19 OPPQI188-17 Astylopsis macula20 CERGL009-08 Astylopsis macula21 CERGL010-08 Astylopsis macula22 CNSLP1104-13 KM850268 Table 6. Interspecific and intraspecific p-distances between species in the preliminary phylogeny. P-distance measures represent percent difference between nucleotide sites. P-distance measures between sets of two individuals were calculated in PAUP* and then averaged to obtain the overall percent distance between nucleotide sites between the two species being compared. Species 1 Species 2 P-Distance (%) A. macula A. collaris 14 A. macula A. sexguttata 17 A. collaris A. sexguttata 17 A. macula (clade 1) A. macula (clade 2) 4.3 A. macula A. macula 2.5 A. collaris A. collaris 0.5 A. sexguttata A. sexguttata 1.4 16 Figure 1. Map of the eastern United States depicting known geographic ranges of each Astylopsis species (Dillon 1956a, Linsley and Chemsak 1995, and Monné and Nearns 2020). 17 1 Astylopsis sexguttata1 AR Astylopsis sexguttata2 QC Astylopsis collaris1 QC Astylopsis collaris3 QC Astylopsis collaris4 QC Astylopsis collaris2 QC Astylopsis macula1 QC Astylopsis macula9 QC Astylopsis macula10 QC Astylopsis macula12 PEI Astylopsis macula13 PEI Astylopsis macula11 QC Astylopsis macula2 QC Astylopsis macula5 QC Astylopsis macula6 QC Astylopsis macula7 QC Astylopsis macula8 QC Astylopsis macula3 QC Astylopsis macula22 ON Astylopsis macula14 ON Astylopsis macula4 QC Astylopsis macula15 ON Astylopsis macula17 ON Astylopsis macula18 ON Astylopsis macula19 ON Astylopsis ON Astylopsis macula20 ON Astylopsis macula21 ON Astylopsis macula16 ON 5 changes Figure 2. Preliminary phylogenetic tree using barcoded COI sequences for Astylopsis sexguttata, A. collaris, and A. macula, constructed in PAUP* using parsimony methods. Individuals are labeled with their species, serial numbers to differentiate individuals, and locality. Locality abbreviations: AR Arkansas, USA; QC Quebec, Canada; PEI Prince Edward Island, Canada; ON Ontario, Canada. 18 Figure 3. Phylogeny constructed from morphological data from 58 individuals (ten individuals each of Astylopsis collaris, A. macula, A. sexguttata, A. arcuata, and A. fascipennis, and eight Astylopsis perplexa individuals). This tree was constructed in PAUP* using a heuristic search with parsimony methods. 19 Figure 4. Phylogeny of A. collaris, A. macula, A. sexguttata, and A. arcuata individuals using barcoded COI sequences, indicated by an asterisk, and experimental COI sequences for the species. This tree was constructed in PAUP* using a heuristic search with parsimony methods. Eupromus ruber, Acanthocinus obsoletus, Leptostylopsis planidorsus, Pseudastylopsis pini, Astylidius parvus, and Astyleiopus variegatus were included as outgroups, rooted with E. ruber. Darker values above branch lines indicate bootstrap support for those relationships; lighter grey values below lines indicate Bayesian posterior probabilities. Tree icons indicate whether the species uses broad-leaved or coniferous hosts. Clades without bootstrap values were not found in the strict consensus of all parsimonious trees. Locality abbreviations: AR Arkansas, USA; QC Quebec, Canada; MA Massachusetts, USA; MI Michigan, USA; FL Florida, USA; LA Louisiana, USA; PEI Prince Edward Island, Canada; TN Tennessee, USA; ON Ontario, Canada; NY New York, USA; PA Pennsylvania, USA; AZ Arizona, USA; JAP Japan. 20 Figure 5. Phylogenetic tree visualization of A. collaris, A. macula, A. sexguttata, and A. arcuata individuals using experimental sequences of genes COI and CAD, represented by one of three parsimonious trees found in a heuristic search. Eupromus ruber, Astyleiopus variegatus, Acanthocinus obsoletus, Leptostylopsis planidorsus, Astylidius parvus, and Pseudastylopsis pini were included as outgroups, rooted with E. ruber Darker values above branch lines indicate bootstrap support for those relationships; lighter grey values below lines indicate Bayesian posterior probabilities. Tree icons indicate whether the species uses broad-leaved or coniferous hosts. Clades without bootstrap values were not found in the strict consensus of the three trees. Locality abbreviations: FL Florida, USA; MA Massachusetts, USA; MI Michigan, USA; LA Louisiana, USA; NY New York, USA; PA Pennsylvania, USA; TN Tennessee, USA; AZ Arizona, USA; JAP Japan. 21 REFERENCES 22 REFERENCES Cocquempot, C. and A. Lindelöw. 2010. Longhorn beetles (Coleoptera, Cerambycidae) Chapter 8.1. In: Roques A et al. (Eds) Alien terrestrial arthropods of Europe. BioRisk, 4(1): 193– 218. Dillon, L. S. 1956a. The Nearctic components of the tribe Acanthocinini (Coleoptera: Cerambycidae). Part I. Annals of the Entomological Society of America 49:134–167. Farrell, B. D. 2001. Evolutionary assembly of the milkweed fauna: Cytochrome oxidase I and the age of Tetraopes beetles. Molecular Phylogenetics and Evolution 18(3):467-578. Gutiérrez, N., F. A. Noguera, S. Zaragoza-Caballero, and J. J. Morrone. 2020. Phylogenetic revision of the floricolous longhorn beetle genus Strangalidium Giesbert (Coleoptera: Cerambycidae). Systematics and Biodiversity 18(1):79-101. Haack, R. A., M. A. Keena, and D. Eyre. 2017. Life history and population dynamics of cerambycids. In: Wang Q (ed) Cerambycidae of the world—biology and pest management. CRC Press, Boca Raton, 71–103. Hebert, P. D. N., A. Cywinska, S. L. Ball, and J. R. deWaard. 2003. Biological identifications through DNA barcodes. Proceedings of the Royal Society of London B 270:313–321 Huelsenbeck, J. P. and F. Ronquist. 2001. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17(8), 754–755. Kariyanna, B., M. Mohan, and R. Gupta. 2017. Biology, ecology, and significance of longhorn beetles (Coleoptera: Cerambycidae). Journal of Entomology and Zoology Studies 5(4):1207-1212. Kirkendall, L. R., P. H. W. Biedermann, and B. H. Jordal. 2015. Chapter 3. Evolution and diversity of bark and ambrosia beetles [pp. 85-156]. In: Bark Beetles (F. E. Vega and R. W. Hofstetter, editors). Academic Press. Lee, S. and S. Lee. 2020. Multigene phylogeny uncovers oviposition-related evolutionary history of Cerambycinae (Coleoptera: Cerambycidae). Molecular Phylogenetics and Evolution 145 (2020) 106707. Li, W., X. Yang, L. Quian, Y. An, and J. Fang. 2016. The complete mitochondrial genome of the citrus long-horned beetle, Anoplophora chinensis (Coleoptera: Cerambycidae). Mitochondrial DNA Part A 27(6):4665-4667. Linsley, E. G. 1959. Ecology of Cerambycidae. Annual Review of Entomology 4:99-138. 23 Linsley, E. G., and J. A. Chemsak. 1995. The Cerambycidae of North America, Part VII, No. 2: Taxonomy and classification of the subfamily Lamiinae, tribes Acanthocinini through Hemilophini. University of California Publications in Entomology, Vol. 114. University of California Press, Berkeley, xii-292. Monné, M. A. and E. H. Nearns. 2020. Catalogue of the Cerambycidae (Coleoptera) of Canada and United States of America. Part IV. Subfamily Lamiinae. Available from https://cerambycids.com/catalog/. Moulton J. K. and B.M. Wiegmann. 2004. Evolution and phylogenetic utility of CAD (rudimentary) among Mesozoic-aged Eremoneuran Diptera (Insecta). Molecular Phylogenetics and Evolution 31(1):363-78. Nakamine, H. and M. Takeda. 2008. Molecular phylogenetic relationships of flightless beetles belonging to the genus Mesechthistatus Breuning, (Coleoptera: Cerambycidae) inferred from mitochondrial COI gene sequences. Journal of Insect Science 8(70). Nearns, E. H. 2013. Systematics of longhorned beetles (Insecta: Coleoptera: Cerambycidae). [Doctoral dissertation, University of New Mexico]. UNM Digital Repository. Ohbayashi, N., J. Ogawa, and Z. Su. 2009. Phylogenetic analysis of the lamiine genus Anoplophora and its relatives (Coleoptera: Cerambycidae) based on the mitochondrial COI gene. Special Bulletin of the Japanese Society of Coleopterology 7:309-324. Rabaglia, R. J., A. I. Cognato, E. R. Hoebeke, C. Johnson, J. R. Labonte, M. E. Carter, and J. Vlach. 2019. Early detection and rapid response: A 10-year summary of the USDA Forest Service program of surveillance for non-native bark and ambrosia beetles. American Entomologist 65(1), 29–42. Ratnasingham, S. and P. D. N Hebert. 2007. BARCODING: Bold: The Barcode of Life Data System (http://www.barcodinglife.org): BARCODING. Molecular Ecology Notes 7(3), 355–364. Ridgway, R. 1912. Color standards and color nomenclature. Washington, D. C: The author. Shiefer, T. L. 2000. A new species of ASTYLOPSIS Casey (Coleoptera: Cerambycidae: Acanthocinini) from the Southeastern United States. The Coleopterists Bulletin 54(4):533-539. Souza, D. d. S., L. Marinoni, M. L. Monné, and J. Gómez-Zurita. 2020. Molecular phylogenetic assessment of the tribal classification of Lamiinae (Coleoptera: Cerambycidae). Molecular Phylogenetics and Evolution 145 (2020) 106736. Swofford, D. L. 2001. PAUP*. Phylogenetic Analysis Using Parsimony (*and Other Methods). Version 4. Sinauer Associates, Sunderland, MA. 24 Toki, W. and K. Kubota. 2010. Molecular phylogeny based on mitochondrial genes and evolution of host plant use in the long-horned beetle tribe Lamiini (Coleoptera: Cerambycidae) in Japan. Molecular Ecology and Evolution 39(4). Wild, A. L. and D. R. Maddison. 2008. Evaluating nuclear protein-coding genes for phylogenetic utility in beetles. Molecular Phylogenetics and Evolution 48:877–891. Wu, Y., N. F. Trepanowski, J. J. Molongowski, P. F. Reagel, S. W. Lingafelter, H. Nadel, S. W. Myers, and A. M. Ray. 2017. Identification of wood-boring beetles (Cerambycidae and Buprestidae) intercepted in trade-associated solid wood packaging material using DNA barcoding and morphology. Scientific Reports 7:40316. 25