Literature studies in North America (US and Canada), Europe, and Asia (particularly Russia, China, Japan, and the Korean peninsula) were reviewed to identify parasitoid guilds associated with Agrilus woodborers. There are at least 12 species of hymenopteran parasitoids attacking eggs of Agrilus beetles and 56 species (36 genera), attacking Agrilus larvae infesting various host plants in North America, Asia, and Europe. While most of the egg parasitoids (9 species) belong to the family Encyrtidae, a majority of the larval parasitoids are members of five families: Braconidae (24 species/11 genera), Eulophidae (8 species/4 genera), Ichneumonidae (10 species/9 genera), and Eupelmidae (6 species/5 genera). The highest rate of Agrilus egg parasitism (>50%) was exerted by encyrtid wasps (4 species) in North America, Asia, and Europe. In contrast, the highest rate of Agrilus larval parasitism (>50%) was caused by species in two genera of braconids: Atanycolus (North America) and Spathius (Asia), and one eulophid genus, Tetrastichus (Asia and Europe). Reported rate of Agrilus larval parasitism ichneumonids was frequent in North America, but generally low (<1%). Potential for success in biological control of emerald ash borer (Agrilus planipennis Fairmaire) in the USA with North American native parasitoids and old-association Asian parasitoids is discussed.

1. Introduction

Agrilus is the largest genus within the family Buprestidae (Coleoptera), with nearly 3,000 described species worldwide [1]. Generally, Agrilus spp. only attack angiosperms and do not develop in conifers [2]. Moreover, they tend to be specialists, most species being confined to a single genus or species of host plant. While most Agrilus species are not considered to be serious pests of agriculture or forests, at least two species have recently become seriously damaging in forests in their newly invaded areas in North America: the emerald ash borer (EAB), A. planipennis Fairmaire, and the gold spotted oak borer (GSOB), A. auroguttatus Shaefer. EAB was accidentally introduced to Michigan in late 1990s from its native range (northeast Asia, in parts of China, Russia, and Korea) possibly via wooden crates or pallets for cargo shipment [3]; it has since spread to 14 additional US states and two Canadian provinces and killed millions of North American ash (Fraxinus spp.) since its detection in 2002 [4, 5]. By contrast, GSOB is native to the oak forests of southwestern Arizona, and while its damage to oak trees in its invaded range has been on a smaller scale, it has killed more than 25,000 oaks in the oak savannahs of California since first discovered there in 2002 [68]. A few other exotic Agrilus species have also been recently detected in the United States (e.g., soapberry borer—A. prionurus Chevrolat in Texas [9]) and Canada (e.g., European oak borer—A. sulcicollis Lacordaire in Ontario [10]). Although some of the recently detected, exotic Agrilus species have not become as widespread or damaging as EAB and GSOB, the pest status of Agrilus borers as a whole along with other woodborers appears to have increased in recent years [11].

Management for the invasive (exotic) Agrilus woodborers (EAB and GSOB) in the United States initially focused on attempted eradication but changed to integration of several approaches when eradication failed to reduce the pests’ populations in infested areas and slow spread of the pests to the noninfested areas [12, 13]. In some cases, control methods being used include delimitation of infested areas, regulatory restriction of movement of pest-infested wood or plant materials, insecticide treatment or physical destruction of infested trees [1214], and biological control via introduction and release of natural enemies collected from pests’ native ranges [7, 1518]. Although none of these approaches individually is adequate, biological control, which relies on self-propagating and dispersing natural enemies, has potential to reduce invasive pest populations, particularly in forests [1921].

Agrilus adults normally lay their eggs under loose bark or in crevices of host plant tissues and rarely cause significant damage; in contrast, Agrilus larvae typically bore into the living tissue (stems, trunks, branches, or roots) of their host plants, interrupt the translocation of water and nutrients as they feed, and can kill plants within one or a few years of infestation (e.g., EAB [22]; GSOB [6]; A. prionurus [9]). In their native habitats, Agrilus populations are generally suppressed by a diverse group of natural enemies and/or host tree resistance and only occasionally become serious pests. However, when introduced into ecosystems where host plants lack coevolutionary resistance, or where appropriate specialized natural enemies are absent, they can become severe pests. The recent invasions of North America by EAB from northeast Asia and GSOB from southwestern Arizona are excellent examples of this. For example, EAB is considered a sporadic pest of ash stands in its native range in Asia [2326] but has become a serious pest threatening the existence of North American ash trees since it was accidentally introduced there [22]. Similar observations have been made for GSOB in its home range. Field studies in Asia found that a complex of natural enemies (primarily parasitoids) and host plant resistance by Asian ash trees appear to be the factors responsible for suppressing EAB populations and preventing them from frequently causing ash mortalities [15, 19].

Deliberate efforts have been recently undertaken in the United States to achieve biological control of EAB and GSOB through introduction of natural enemies (parasitoids) from the native ranges of these pests [7, 17]. These classical biological control efforts for EAB have led to the discovery and introduction of several egg and larval parasitoids that have the potential to establish and suppress the pests’ populations in the newly introduced regions [1921, 27]. Similar programs for GSOB commenced in 2010 and are too immature to reach tentative conclusions about natural enemy diversity and impacts. In reviewing the literature, we found that many groups of parasitoids and other natural enemies have reported attacking Agrilus beetles. An overview of the composition of the parasitoid guild attacking this group of woodborers will contribute to the current and future development of biological control programs to manage these pests, particularly those Agrilus that have invaded new regions or environments. In the present study, we first review the diversity of natural enemy complexes in particular, hymenopteran parasitoid guilds associated with egg and larval stages of Agrilus species, and then discuss the potential of those parasitoids for use as agents of classical biological control against this group of pests.

2. Literature Reviewed

We searched seven major online data bases using the key word “Agrilus” either alone or in combination with any of the key words “parasitoid,” “parasite”, “natural enemy”, or “biological control” to locate relevant literature. Databases examined were (1) Agricola, (2) BioAbstracts (BIOS), (3) Biological Sciences Set, (4) Biological and Agricultural Index Plus, (5) CSA Illumina, (6) CAB Abstract, and (7) ISI Web of Sciences set. The key word “Agrilus” alone resulted in 1942 articles (Table 1), of which 142 articles remained when combined with “parasitoid or parasite, natural enemies, or biological control.” It must be noted that database searches concluded in March 2011. For this paper, we included only those original research articles that provide information on parasitoid identity at the family, genus, or species levels (Table 2).

In addition to searching databases, we contacted colleagues who work on invasive EAB and GSOB beetles and their biological control in the United States and Canada for information on parasitoid guilds of these species. All relevant studies were read and analyzed for mention of Agrilus species, associated parasitoids, known host associations, host plants, and geographic distributions. If available, parasitism rates by each group, guild, or species of parasitoids were noted.

3. Results and Discussion

At the genus level, the guilds of egg and larval parasitoids of Agrilus species were similar in North America, Europe, and Asia. While several families of North American parasitoids (including Braconidae, Chalcididae, Ichneumonidae, and Eupelmidae) are capable of utilizing larvae of the newly introduced emerald ash borer (A. planipennis) as a novel host, some Asiatic species of parasitoids appear to be more specific and only utilize Asian Agrilus species as hosts. From the geographic distribution point of view, it appears that there is more diversity in the parasitoid complex associated with Agrilus beetles in North American than in Asia and Europe. However, this geographic difference in parasitoid diversity may actually reflect different levels of research activities on the subject. For example, the invasion of North America by EAB has certainly resulted in much more research activities on the parasitoid complex of this group of woodborers in North America.

There are at least 12 species of hymenopteran parasitoids that attack eggs of Agrilus beetles and 56 parasitoid species that attack Agrilus larvae in various plants in North America, Asia, or Europe (Table 2). While most of these egg parasitoids (9 species) belong to the family Encyrtidae, a majority of the larval parasitoids are members of five families: Braconidae (24 species/11 genera), Eulophidae (8 species/4 genera), Ichneumonidae (10 species/9 genera), and Eupelmidae (6 species/5 genera). One species of larval parasitoid (Phasgnophora sulcata Westwood) (Chalcididae) is frequently associated with native Agrilus woodborers in North America [2830]. In addition, there is one larval parasitoid (Foenatopus sp.) in the family Stephanidae that was reported attacking A. sexsignatus (Fisher) infesting Eucalyptus trees in southeast Asia [31, 32].

The highest rates of Agrilus egg parasitism (>50%) occurred with four species of encyrtid wasps reported in North America, Asia, and Europe (Table 2). In contrast, the highest rates of Agrilus larval parasitism (>50%) were caused by two groups of braconid wasps: Atanycolus spp. (in North America) and Spathius spp. (in Asia), and three species of eulophid wasps (in Asia and Europe). Although ichneumonid wasps were frequently reported attacking Agrilus woodborers in North America, the reported rate of parasitism was very low (<1%) for all the ichneumonid species.

It is interesting to note that several species of North American native parasitoids, Atanycolus spp., Spathius floridanus Ashmead, S. laflammei Provancher, S. simillimus Ashmead, Phasgonophora sulcata Westwood, and one accidentally introduced Asiatic wasp Balcha indica (Mani and Kaul), have been recently reported attacking the invasive emerald ash borer. One group of native parasitoids, Atanycolus spp., has recently become the dominant mortality factor associated with emerald ash borer, attacking >50% of A. planipennis larvae at some forest sites in Michigan (USA) [19, 33]. The potential of both the native (new-association) parasitoids and the introduced (old-association) parasitoids (e.g., Oobius agrili Zhang and Huang, Tetrastichus planipennisi Yang, and Spathius agrili Yang) for biological control of EAB, in the USA, needs further investigation.

A diverse group of hymenopteran parasitoids attacks eggs and larvae of Agrilus woodborers in North America, Asia, and Europe. Literature review of this genus, in regards to its parasitoid guild, has interest due to the introduction of two species in North America (GSOB and EAB). In biological control, parasitoid species of invasive pests are often introduced from the land of origin, if proved to be safe (not become a pest themselves). In addition, new-association parasitoids that inhabit the region prior to the pest introduction sometimes exert pressure on this newly arrived pest and offer opportunity for research and augmentation of indigenous parasitoid populations. Our literature review has provided documentation of research activities for 12 egg parasitoid species and 56 larval parasitoid species. These parasitoids are identified from 19 species of Agrilus, a small representation of almost 3000 described, that attack 18 recorded plant types (13 hardwoods, 5 shrubs). Being a diverse genus, these results show a wealth of research opportunities for further work on Agrilus parasitoids worldwide. Nearly two thirds (64.3%) of the literature found was published after year 2000. Twenty-seven of 83 entries (32.5%), in Table 2, reference A. planipennis. These findings are results of EAB postdetection in 2002.

Although Agrilus species are relatively host specific, because of larvae’s concealed nature, early stages and damage are difficult to assess and take much effort to obtain. This has implications on finding and identifying parasitoid complexes for biocontrol and may be a reason for so little literature. Of those parasitoid species found in association with EAB, some are ectoparasitoids and known to attack woodborers in different families (e.g., Cerambycidae). While data from the current literature do not show any particular relationship between host specificity and mode of parasitization (endo- versus ectoparasitoids), further research is needed to investigate such relationship.

Some species occur on multiple Agrilus spp., such as egg parasitoid Oobius zahaikevitshi. Atanycolus cappaerti is known to attack A. planipennis, A. liragus, and A. bilineatus, while Leluthia astigma attacks A. planipennis, A. difficilis, and other Agrilus spp. These may provide better access of parasitoids where poplar, chestnut, honey locust, and ash occur together.

The parasitoid guild of Agrilus in China, Russia, and North America and EAB distribution may provide species for introduction or augmentation. Though most of the parasitism rates are low (<10%), a few worthy candidates not yet used for introduction or augmentation include egg parasitoid, O. zahaikevitshi from Russia, and larval parasitoids Atanycolus cappaerti, Spathius agrilovorus, and Spathius floridanus. These species in the USA have not yet been reared in large numbers, and further studies on rearing methods need pursuing. It appears also that braconids and eulophids have provided the best potential for biological control, and the number of studies the last five years bear this out. It also indicates that species size and morphology (ovipositor length) for accessing the host from outside the host plant are important for success.

Finally, parasitoid work in biological control efforts often lack taxonomic expertise to provide accurate identifications. Some of these newly known parasitoid species are not well understood. Egg parasitoids are often disregarded due to size and inaccessibility of host eggs. These hamper ongoing biological control of invasive or cyclic native pest populations. A concluding question is should work be done now on conspecifics that have the potential to be invasive (e.g., A. coxalis attacks oaks in Mexico—California has a history of acquiring pests from MX, could A. coxalis be another threat to CA’s besieged oaks forests?).