Oak-hickory forests in northwestern Arkansas, eastern Oklahoma and southern Missouri have recently experienced an oak decline event with widespread oak mortality. The oak mortality is associated with an outbreak of a native wood-boring cerambycid, Enaphalodes rufulus (Haldeman), the red oak borer. Taxonomic identification, below the family level, of larval Cerambycidae through traditional morphological methods is not usually possible. We employed molecular diagnostics, with polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP), to distinguish E. rufulus from other closely related species of cerambycids. A portion of the mitochondrial DNA 16S rRNA gene, isolated from legs or thoraxes of adult museum specimens, was amplified and digested with Alu I and Hind III restriction enzymes. Both restriction enzymes independently produced fragments for E. rufulus that were significantly different from any other cerambycid tested. Alu I had one restriction site for E. rufulus and two restriction sites for all other cerambycids tested, while Hind III did not cut for E. rufulus but did cut at one restriction site for all other cerambycids. Eggs, larvae, and pupae of E. rufulus along with an unknown cerambycid larva and pupa were successfully amplified and digested by this method to verify validity of this technique for multiple life stages.
The red oak borer, Enaphalodes rufulus (Haldeman) (Coleoptera: Cerambycidae), is an important wood-boring species native to eastern hardwood forests of the United States (Donley & Acciavatti 1980). A variety of oak species are attacked by E. rufulus, but trees in the red oak group Erythrobalanus are preferred, especially black oak, Quercus velutina Lam., scarlet oak, Q. coccinea Muenchh., and northern red oak, Q. rubra L. (Hay 1974). Since E. rufulus attacks and reproduces in living trees, significant degrade in lumber quality is an important issue in commercial stands (Hay 1964). Damage caused by borers often goes unnoticed until trees are felled and sawn for timber, and by this time as much as 40% of the 120-year value of the tree may be lost (Donley & Worley 1976). In comparison, damage caused by defoliators is much more noticeable, but defoliation typically causes only a 15-20% reduction in value (Donley & Worley 1976). Donley and Acciavatti (1980) estimated that 38% of oak wood used for lumber, cooperage and veneer in the Eastern United States is affected by E. rufulus.
Enaphalodes rufulus population densities historically have been documented at low levels. Hay (1969) found an average of 3.7, 2.8, and 2.5 attacks on the bottom 1.8 m of black oak, northern red oak, and scarlet oak, respectively in Ohio, and Donley and Rast (1984) found an average of 2.0 attack sites per red oak in Pennsylvania and 3.6 in Indiana. Recently, however, an unprecedented outbreak with significant economic and ecological impacts has occurred in the Ozark oak/hickory forests of northern Arkansas and southern Missouri (U.S.A.) (Stephen et al. 2001; Starkey et al. 2000). Analysis of recent data from the Ozark National Forest reveal an average of 599 active attacks (or current generation galleries) and 77 live larvae per tree in northern red oak (Fierke et al. 2005). USDA Forest Service estimates 450,000 ha of forest in the Ozark Mountains will be impacted by E. rufulus with an estimated 68,000 m2 of timber loss or degradation (Guldin et al. 2005).
Removal of infested trees is a recommended control method for E. rufulus (Donley 1981, 1983), but diagnosis of infestation may be delayed as easily identifiable adults emerge only every two years and other identification methods are difficult and unreliable. Historically, attack sites were identified by observing frass (Hay 1969), but this method is limited by observer ability, weather conditions, and multiple insects with similar life histories. Larval keys for North American cerambycid species exist (Craighead 1923) but are difficult to follow and outdated. In addition, morphological differences among closely related cerambycid larvae are often minute. Larval identification is also important for detecting E. rufulus in tree hosts related to red oaks, which may harbor several cerambycid species, such as Elaphidion spp., Goes spp., or Noeclytus spp. (Yanega 1996). Molecular genetic techniques are an alternative to traditional methods for distinguishing E. rufulus larvae from other cerambycids.
The objective of this research was to develop a molecular diagnostic technique for all life stages of E. rufulus with polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP). Genetic research has never been conducted on E. rufulus and molecular diagnostics have been reported for only one other cerambycid, Anoplophora glabripennis (Motschulsky) (Kethidi et al. 2003). The A. glabripennis diagnostic technique utilizes sequence characterized amplified regions (SCARs) derived from randomly amplified polymorphic DNA (RAPD). While the end result of this technique is a very simple polymerase chain reaction diagnostic, development of this procedure is time-consuming and expensive. On the other hand, PCR-RFLP is a simple, inexpensive, established and reliable technique (Taylor & Szalanski 1999) that has been used to identify many economically important insects, including termites, Reticulitermes spp. (Isoptera: Reticulitermatidae) (Szalanski et al. 2003), screwworm flies, Cochliomyia spp. (Diptera: Calliphoridae) (Litjens et al. 2001), and corn rootworm, Diabrotica spp. (Coleoptera: Chrysomelidae) (Clark et al. 2001).
Materials and Methods
Specimen Collection
Enaphalodes rufulus adults were collected from two areas, Fly Gap and White Rock, in the Ozark Mountains of northwestern Arkansas with standard black-lighting techniques during the flight period between mid-June and late July of 2003 (UTM Zone 15–S NAD83: Fly Gap--0431660, 3954978, White Rock--412668, 3949429). Beetles were placed in 100% ethanol immediately upon capture.
Adult cerambycids, other than E. rufulus, that are common to the Ozark Mountain region and one closely related Enaphalodes species were collected during concurrent research at the University of Arkansas. Voucher specimens are stored in the Forest Insect Collection at the University of Arkansas Forest Entomology Lab (Table 1). Specimens were caught in clear plexiglass, passive flight intercept panel traps during the summer of 2001. Specimens were collected in 50% ethylene glycol. Upon return to the lab, specimens were transferred to 95% alcohol until pinning. Morphological identification of adult specimens was made by Dr. J. K. Barnes, Arthropod Museum Curator at the University of Arkansas.
Eggs of E. rufulus were collected from a lab colony in March 2005. Larvae and pupae were collected from red oak trees harvested between October 2002 and March 2005. Eggs and early instars were frozen, and late instars and one pupa were stored in 95% alcohol until used. An unknown larva and pupa were collected from a white oak tree and stored in 95% alcohol until used.
PCR-RFLP Protocol
DNA was extracted from one leg or thorax of E. rufulus and the other 13 adult cerambycids used in this procedure. DNA extraction was accomplished by the protocol of the Qiagen DNeasy tissue kit (Valencia, CA). Re-suspended DNA was stored at -20°C until used. An approximately 420-bp portion of the 16S rRNA gene was amplified using the primers 16S-r (5'-CGCCTGTTTATCAAAAACAT-3') (Simon et al. 1994) and 16S-f (5'-TTACGCTGTTATCCCTAA-3') (Kambhampati & Smith 1995). PCR reactions were conducted with 1 μl of extracted DNA as per Szalanski et al. (2000) with a thermocycler profile consisting of 40 cycles of 94°C for 45 s, 46°C for 45 s, and 72°C for 45 s. PCR products were purified and concentrated with the Wizard SV Gel and PCR clean-up kit (Promega, Madison, WI). One sample from each adult was sent to the University of Arkansas for Medical Sciences DNA Sequencing Core Facility (Little Rock, AR), for direct sequencing in both directions. Consensus sequences for each species were acquired by manual alignment and editing of forward and reverse sequences in BioEdit (Hall 1999). GenBank accession numbers are DQ417758 to DQ2417771.
Webcutter 2.0 (Heiman 1997) was used to predict restriction sites from DNA sequence data. Amplified DNA was digested according to manufacturer's (Promega, Madison, WI) recommendations with the enzymes Alu I or Hind III. Fragments were separated by 2% agarose gel electrophoresis. Gels were stained with ethidium bromide and photographed with the UVP BioDoc-it documentation system (Upland, CA).
Results
The rRNA 16S amplicon ranged from 411 to 418 bp in all cerambycids studied (Table 1). Webcutter 2.0 analysis of the 16S sequences revealed that either Alu I or Hind III could effectively distinguish E. rufulus from all other species (Table 2). Either enzyme was also capable of distinguishing E. atomarius (Drury), a less common but closely related species. All other species produced similar restriction profiles.
To validate the diagnostic, this procedure was tested on 29 adult E. rufulus and as many replicates as possible for the additional adult species (Table 1). All species were replicated except Distenia undata (Fabricius), Bellamira scalaris (Say), Dorcaschema wildii Uhler, Goes tigrinus (DeGeer), and Aegomorphus morrisii (Uhler). There was only one individual available from the collection for these five species. Samples digested with Alu I revealed clearly defined results (Fig. 1) as did samples digested with Hind III (Fig. 2).
All life stages of E. rufulus also were tested with this procedure. Five E. rufulus eggs, four E. rufulus early larval head capsules, six E. rufulus late larval incised head capsules and one E. rufulus incised pupa head were used for confirmation of the use of this diagnostic procedure for multiple life stages. All samples produced positive results when either enzyme was used. One unknown cerambycid larva, and one unknown cerambycid pupa collected from a white oak tree also were tested. Results gave fragment lengths similar to those of all other cerambycids except E. rufulus and E. atomarius.
Discussion
Results from this research show that Enaphalodes rufulus readily can be distinguished from all other cerambycids tested by PCR-RFLP and either of the two restriction enzymes, Alu I or Hind III. The ability to digest the PCR amplicon by either enzyme could be important for future studies where use of only one enzyme is possible.
PCR-RFLP is a well established means for species diagnostics of many organisms (Taylor & Szalanski 1999; Slade et al. 1993; Sperling et al. 1994; Roehrdanz 1997; Szalanski et al. 1997; Harrington & Wingfield 1995; Taylor et al. 1996). This method of identification requires less equipment and is less expensive than other practiced methods. It is also easy to repeat and can be used not only for diagnostics but also for phylogenetic analyses (Taylor & Szalanski 1999).
DNA extraction from dried adult legs or thoraxes worked well with Qiagen DNeasy extraction kit and PCR-RFLP. Extracting DNA from pinned, dried adult cerambycids could prove useful in future studies where genetic information may need to be extracted from stored museum specimens. Adults were used to create this procedure as they easily are identified by morphological characteristics, and positive identification was necessary for sequence comparisons.
Morphological identification of early instar cerambycid larvae is difficult if not impossible. PCR-RFLP is an established diagnostic tool for larval identification and easily could be used to distinguish morphologically similar, yet genetically distinct species of cerambycids that have similar life history strategies. This should prove useful in detecting E. rufulus as a contributing factor in other oak mortality events especially during the larval stage in which they spend about 90% of their two-year life cycle. Early detection of E. rufulus in other oak decline events should help foresters or landowners make informed management decisions in regard to harvest options and/or silvicultural remediation.
White oak mortality is prevalent throughout the Ozark Mountains in Arkansas. White oaks can harbor several species of cerambycids including E. rufulus and white oak borer, G. tigrinus. An unknown larva and pupa from a white oak tree, which may have been G. tigrinus, were tested to confirm the validity of this diagnostic technique and to show that this procedure works with larvae and pupae of species other than E. rufulus. The resulting larval and pupal fragments were similar to those of all other cerambycids tested, except E. rufulus and E. atomarius. A molecular diagnostic procedure possibly could be created for other cerambycids from the Ozarks. This would clarify to what extent G. tigrinus or other cerambycids are contributing to white oak mortality, and may offer insight into the prevalence of other immature cerambycids in economically important oak trees.
It was difficult to obtain samples from areas other than northern Arkansas as E. rufulus is normally at low population levels. However, this study provides a good foundation for E. rufulus identification with PCR-RFLP and our results can be expanded as specimens are collected from other regions of the eastern U.S.
Acknowledgments
The authors thank Dana Kinney, Vaughn Salisbury, Damon Crook, Leah Chapman, Jarrett Bates, and Ben Rowe for help in collecting specimens; Dr. J. K. Barnes for specimen identification; and Melissa Fierke and Larry Galligan for reviews and suggestions. Financial support for this research was provided in part through the Arkansas Agricultural Experiment Station, the Arkansas Forestry Research Center, Special Technology Development and Forest Health Monitoring Programs of the USDA Forest Service, Atlanta, Georgia.