Journal of Economic Entomology Advance Access published January 10, 2017

Journal of Economic Entomology, 2017, 1–8 doi: 10.1093/jee/tow292 in Relation to Plant Disease Research article

Presence and Prevalence of Raffaelea lauricola, Cause of Laurel Wilt, in Different Species of Ambrosia in Florida, USA

Randy C. Ploetz,1,2 Joshua L. Konkol,1 Teresa Narvaez,1,3 Rita E. Duncan,1Ramon J. Saucedo,1 Alina Campbell,3 Julio Mantilla,1 Daniel Carrillo,1 and Paul E. Kendra3

1University of Florida, IFAS, 18905 SW 280th St., Homestead, FL 33031 (kelly12@ufl.edu; jkonkol@ufl.edu; tnarvaez1@ufl.edu; ritad@ufl.edu; che447@ufl.edu; juliomanti@ufl.edu; dancar@ufl.edu), 2Corresponding author, e-mail: kelly12@ufl.edu, and 3USDA-ARS Subtropical Horticulture Research Station, Miami, FL 33158 ([email protected]; [email protected]) Subject Editor: John Trumble Downloaded from Received 19 October 2016; Editorial decision 16 November 2016

Abstract We summarize the information available on ambrosia beetle species that have been associated in Florida with

Raffaelea lauricola T.C. Harr., Fraedrich & Aghayeva, the primary symbiont of glabratus Eichhoff and http://jee.oxfordjournals.org/ cause of laurel wilt. In total, 14 species in Ambrosiodmus, Euwallacea, Premnobius, Xyleborus, Xyleborinus, and Xylosandrus were either reared from laurel wilt-affected host trees or trapped in laurel wilt-affected stands of the same, and assayed for R. lauricola. In six collections from native species in the southeastern United States [Persea borbonia (L.), Persea palustris (Raf.) Sarg., and Persea humilis Nash] and four from avocado (Persea americana Mill.), extracted mycangia or heads (taxa with mandibular mycangia) or intact bodies (taxa with mycangia in other locations) were surface-disinfested before assays on a semi-selective medium for the isolation of Raffaelea (CSMAþ). Raffaelea lauricola was identified based on its characteristic phenotype on CSMAþ, and the identity of a random subset of isolates was confirmed with taxon-specific microsatellite by guest on January 11, 2017 markers. The pathogen was recovered from 34% (246 of 726) of the individuals that were associated with the na- tive Persea spp., but only 6% (58 of 931) of those that were associated with avocado. Over all studies, R. lauri- cola was recovered from 10 of the ambrosia beetle species, but it was most prevalent in Xyleborus congeners. This is the first record of R. lauricola in Ambrosiodmus lecontei Hopkins, Xyleborinus andrewesi (Blandford), and Xyleborus bispinatus Eichhoff. The potential effects of R. lauricola’s promiscuity are discussed.

Key words: Lateral transfer, vascular wilt, Lauraceae, , Scolytinae

Laurel wilt is caused by Raffaelea lauricola T.C. Harr., Fraedrich & The rapid spread of laurel wilt in the southeastern United States Aghayeva (Ophiostomatales), a fungal symbiont of an invasive has been due to the efficiency of X. glabratus as a vector, the an- Asian ambrosia beetle, Xyleborus glabratus Eichhoff (Scolytinae: thropogenic dissemination of wood that is infested with the ) (Fraedrich et al. 2008, Harrington et al. 2008). and pathogen, and the presence of native and non-native plants that Xyleborus glabratus was detected for the first time outside Asia in are susceptible to the disease and in which X. glabratus reproduces Port Wentworth, Georgia (Haack 2006, Rabaglia et al. 2006), and (Hughes et al. 2015b, Ploetz et al. 2016). Soon after the loss of the the pathogen was associated with all assays of the beetle that have first avocado tree was reported in Jacksonville, Florida (Mayfield been made to date in the United States (Fraedrich et al. 2008, et al. 2008), the disease moved southward along the state’s east Harrington et al. 2008). Harrington et al. (2011) recovered R. lauri- coast. In February 2011, the disease was confirmed adjacent to the cola from specimens of X. glabratus from its native range (Japan Everglades in stands of swamp bay (Ploetz et al. 2011); by and Taiwan). Because there are no reports of laurel wilt before November of that year, it had spread to Florida’s primary commer- 2003, it is probable that the beetle carried the pathogen when it was cial avocado production area (CAPA) in southeastern Miami-Dade first detected in Port Wentworth in 2002, and that that introduction County; and within 2 yr, it had spread throughout the CAPA. As of established the beetle and pathogen in the United States (Harrington August 2016, laurel wilt was present in all but six western counties et al. 2011, Ploetz et al. 2013). Clonal populations of both R. lauri- of Florida (Barton et al. 2016). cola and X. glabratus are associated with the laurel wilt epidemic in Unexpected insights have emerged as this disease has been re- the United States (Hughes 2013, Hughes et al. 2016). searched. Laurel wilt is unique, in that symbionts of ambrosia

VC The Authors 2017. Published by Oxford University Press on behalf of Entomological Society of America. All rights reserved. For Permissions, please email: [email protected] 1 2 Journal of Economic Entomology, 2017, Vol. 0, No. 0

Table 1. Occurrence of R.lauricola in ambrosia associated with laurel wilt-affected native Persea spp. in Florida

Species/Genera (total Number of individuals assayed Number individuals with R. lauricola % Assayed individuals with R. lauricola Mean number of individuals % sampled) ABCDEABCDEABCDE

Xyleborus 33 13 50 25 20 5 12 43a 23 20b 15 92 86 92 100 73 glabratus (141) Xyleborus nr np nr 16 np nr nr nr 10 np nr nr nr 63 np 63 bispinatus (16) Xyleborus 9 np 118 nr np 2 nr 70a nr np 22 nr 59 nr np 57 ferrugineus (127) Xyleborus affinis (61) 5 np 41 15 np 2 nr 5a 1 np 40 nr 12 7 np 13 Xyleborus 24 np 39 nr np 0 nr 20a nr np 0 nr 51 nr np 32 volvulus (63) Xyleborus (408) 71 13 248 56 20 9 12 138 34 20 13 92 56 61 100 52 Ambrosiodmus nr np 25 nr np nr nr 0 nr np nr nr 0 nr np 0 devexulus (25) Ambrosiodmus 3 np41nrnp1c nr 0 nr np 33 nr 0 nr np 2 lecontei (44) Ambrosiodmus (69) 3 0 66 nr np 1 nr 0 nr np 33 nr 0 nr np 1 Xyleborinus nr np nr 15 np nr nr nr 2 np nr nr nr 13 np 13 Downloaded from andrewesi (15) Xyleborinus nr np 52 nr np nr nr 26a nr np nr nr 50 nr np 50 gracilis (52) Xyleborinus 22 np 68 15 np 0 nr 2a 0np0nr30np2 saxesenii (107) Xyleborinus (174) 22 0 120 30 np 0 0 28 2 np 0 0 23 7 np 17 http://jee.oxfordjournals.org/ Xylosandrus nr np nr 15 np nr nr nr 0 np nr nr nr 0 np 0 compactus (15) Xylosandrus 1 np3915npnr0 1a 1np0nr37np4 crassiusculus (55) Xylosandrus (70) 1 0 39 30 np nr 0 1 1 np 0 nr 3 3 np 3 Totals (721) 97 13 473 116 20 10 12 167 37 20 10 92 35 32 100 34

Beetle taxa are listed in the descending order of their relatedness to X. glabratus (Cognato et al. 2011). Beetles that were assayed for R. lauricola were: A) reared or trapped from a laurel wilt-affected stand of redbay (P. borbonia) in Lake Kissimmee in 2009 and 2010 (two experiments), and assayed dead 5 to 305 days after collection; B) reared in 2013 from a laurel wilt-affected redbay originating in Alachua County, and assayed within a few days after collection; C) reared by guest on January 11, 2017 from laurel wilt-affected swamp bay (P. palustris) in 2012, and assayed alive and soon after emergence; D) captured in flight in Highlands County in 2013, as described by Kendra et al. (2012), and processed alive and within days of collection; and E) reared from laurel wilt-affected silk bay (P. humilis) in Highland County in 2015, and assayed alive and within days of capture. Xyleborus bispinatus was not distinguished from X. ferrugineus when beetles were assayed in A and C (i.e., some of those listed as X. ferrugineus may have been X. bispinatus). Raffaela lauricola was not recovered from three individuals of Euwallacea nr. for- nicatus that were assayed. nr ¼ not recovered; np ¼ not processed. a Individuals confirmed as R. lauricola with microsatellite data (Carrillo et al. 2014). b Individual confirmed as R. lauricola (see Fig. 2A). c Individual confirmed as R. lauricola (see Fig. 2C).

beetles are rarely plant pathogens and were not known previously to Brevard County; Carrillo et al. 2012, unpublished data). Previously, move systemically in the host (Fraedrich et al. 2008, Ploetz et al. Brar et al. (2013) determined that X. glabratus developed at com- 2013); however, one inoculation with R. lauricola is sufficient to parable rates in logs of avocado, redbay, and swamp bay, but that colonize and kill an entire tree (Hughes et al. 2015a, Ploetz et al. fewer progeny were produced in avocado. Thus, low numbers of X. 2012). In little more than a decade, the pathogen was laterally trans- glabratus that have been associated with avocado may be the result ferred from X. glabratus to several other ambrosia beetles of it being a poor reproductive host for the beetle. (Harrington and Fraedrich 2010, Ploetz et al. 2013, Carrillo et al. In no-choice studies, Carrillo et al. (2014) showed that in addition 2014). All other species were present in the United States before the to X. glabratus, Xyleborus affinis (Eichhoff), Xyleborus ferrugineus introduction of X. glabratus and, presumably, obtained the fungus (Fabricius), Xyleborinus gracilis (Eichhoff), Xyleborus volvulus in laurel wilt-affected trees after that introduction (Ploetz et al. (Fabricius), Xyleborinus saxesenii Ratzeburg, and Xylosandrus cras- 2013). siusculus (Motschulsky) could transmit R. lauricola to swamp bay During studies of laurel wilt-affected trees, X. glabratus has pre- and X. ferrugineus and X. volvulus to avocado. This experimental evi- dominated in Persea spp. that are native to the southeastern United dence and the general absence of X. glabratus in avocado orchards in States (e.g., 48% of the individuals from a study of swamp bay), but South Florida has led to the hypothesis that other species of ambrosia it has been rarely associated with laurel wilt-affected avocado (0 of beetles naturally transmit R. lauricola to avocado. Although their 79,025 ambrosia beetles from recent trapping or rearing efforts in roles in the laurel wilt epidemic seem probable, little is known about Miami-Dade County and 11 of 4,181 reared individuals from their acquisition and spread of the pathogen. Journal of Economic Entomology, 2017, Vol. 0, No. 0 3

Table 2. Propagule densities of R.lauricola in ambrosia beetles associated with laurel wilt-affected native Persea spp. in Florida

Species/Genera (total Mean CFUs per individual Ranges of CFUs for assayed individuals Range for number of individuals species/genus sampled) AB C D E A B C D E

Xyleborus glabratus (141) 2 1,557 2,783 2,868 6,970 0–34 0–3,375 0–7,800 0–8,533 3,600–11,600 0–11,600 Xyleborus bispinatus (16) nr np nr 7 np nr nr nr 0–53 np 0–53 Xyleborus 18 np 33 nr np 0–99 nr 0–560 nr np 0–560 ferrugineus (127) Xyleborus affinis (61) 26 np 1 2 np 0–60 nr 0–20 0–30 np 0–60 Xyleborus volvulus (63) nr np 28 nr np nr nr 0–300 nr np 0–300 Xyleborus (408) 4 1,557 581 1,283 6,970 0–99 0–3,375 0–7,800 0–8,533 3,600–11,600 0–11,600 Ambrosiodmus nr np 0 nr np nr nr 0 nr np 0 devexulus (25) Ambrosiodmus 600 np 0 nr np 0–1,800 nr 0 nr np 0–1,800 lecontei (44) Ambrosiodmus (69) 600 np 0 nr np 0–1,800 nr 0 nr np 0–1,800 Xyleborinus nr np nr 1 np nr nr nr 0–17 np 0–17 andrewesi (15) Xyleborinus nr np 101 nr np nr nr 0–1,240 nr np 0–1,240 gracilis (52) Downloaded from Xyleborinus 0 np 1 0 np 0 nr 0–60 0 np 0–60 saxesenii (107) Xyleborinus (174) 0 np 44 1 np 0 nr 0–1,240 0–17 np 0–1,240 Xylosandrus nr np nr 0 np nr nr nr 0 np 0 compactus (15)

Xylosandrus 0 np 3 113 np 0 nr 0–100 0–1,690 np 0–1,690 http://jee.oxfordjournals.org/ crassiusculus (55) Xylosandrus (70) 0 np 3 56 np 0 nr 0–100 0–1,690 np 0–1,690 Total (721) 21 1,557 316 634 6,970 0–1,800 0–3,375 0–7,800 0–8,533 3,600–11,600 0–11,600

Beetle taxa and collections are described in Table 1. nr ¼ not recovered; np ¼ not processed.

We summarize results from 10 experiments in which 1,655 am- microsatellite DNA markers as described by Dreaden et al. (2014; by guest on January 11, 2017 brosia beetles were reared from laurel wilt-affected trees or trapped Fig. 2, Table 5). in laurel wilt-affected areas, and subsequently assayed for R. lauri- In total, 724 individuals were collected during six experiments cola. Non-casual, putatively symbiotic associations were determined (A–E) in Central and South Florida in natural areas that were popu- by assaying surface-disinfested mycangia or mycangium-containing lated with native species of Persea (Tables 1 and 2). Beetles were: A) body parts of individuals for the pathogen’s presence and reared or trapped from a stand of redbay in Lake Kissimmee in prevalence. 2009 and 2010 (two experiments that were combined in A because of their similarities), and assayed dead 5 to 305 days after collection; B) reared in 2013 from laurel wilt-affected redbay from Alachua Materials and Methods County, and assayed within a few days of collection; C) reared from swamp bay collected in Miami-Dade County in 2012, and assayed Adult females of 14 species in Ambrosiodmus, Euwallacea, alive and within days of emergence (Carrillo et al. 2014); D) cap- Premnobius, Xyleborus, Xyleborinus, and Xylosandrus were either tured in flight in Highlands County in 2013, as described by Kendra reared from laurel wilt-affected host trees or trapped in laurel wilt- et al. (2012), and processed alive and within days of collection; and affected stands of the same (Tables 1–4). Extracted mycangia or E) reared from laurel wilt-affected silk bay (P. humilis) in Highlands heads of species that have mandibular mycangia, or intact bodies of County in 2015, and assayed alive and within days of capture. those with other types of mycangia, were surface-disinfested for 15 s Because Xyleborus bispinatus Eichhoff was not distinguished from in 70% ethanol, rinsed three times in sterile water, and macerated in X. ferrugineus when beetles were assayed in A, B, and C, some of sterile water as previously described (Carrillo et al. 2014). Serial di- those listed as X. ferrugineus in those studies may have been X. bis- lutions (1:1, 1:10, and 1:100) of the macerates were streaked on a pinatus (see Atkinson et al. 2013). In total, 25 individuals of semi-selective medium for Raffaelea, Harrington’s (1981) CSMA Ambrosiodmus devexulus (Wood), 44 of Ambrosiodmus lecontei amended with 0.25-g ampicillin and 0.005-g rifampicin L1 Hopkins, three of Euwallacea nr. fornicatus (Eichhoff), 15 of (CSMAþ; Ploetz et al. 2012), and incubated under diffuse light on a Xyleborinus andrewesi (Blandford), 52 of X. gracilis, 107 of X. sax- laboratory bench. After 7–10 d, colony-forming units (CFUs) of R. esenii,61ofX. affinis,16ofX. bispinatus, 127 of X. ferrugineus, lauricola were enumerated based on the species’ characteristic 141 of X. glabratus,63ofX. volvulus,15ofXylosandrus compac- phenotype on CSMAþ (Fig. 1) and the numbers of CFUs per individ- tus, and 55 of X. crassiusculus were assayed (Tables 1 and 2). ual were calculated. The identity of a subset of isolates with this A total of 931 individuals were collected during four experi- phenotype was confirmed as R. lauricola with taxon-specific ments (F–I) with avocado trees (Tables 3 and 4). Live beetles were 4 Journal of Economic Entomology, 2017, Vol. 0, No. 0

Table 3. Occurrence of R. lauricola in ambrosia beetles associated with laurel wilt-affected avocado (Persea americana) in Florida

Species/Genera (total Number of individuals assayed Number of individuals with R. lauricola % Assayed individuals with R. lauricola Mean number of individuals % assayed) FGHIF G H I F G H I

Xyleborus nr nr 5 nr nr nr 3 nr nr nr 60 nr 60 glabratus (5) Xyleborus 46 5 8 20 7 5 2 14a 15 100 25 70 35 bispinatus (79) Xyleborus 2 2 23 nr 1 2 0 nr 50 100 0 nr 11 ferrugineus (27) Xyleborus 79 16 7 20 1 0 0 3b 10 0153 affinis (122) Xyleborus 134 53 10 20 3 10 0 0 2 19 0 0 6 volvulus (217) Xyleborus (450) 261 76 53 60 12 17 5 17 5 22 9 28 11 Ambrosiodmus nr nr 16 nr nr nr 0 nr nr nr 0 nr 0 lecontei (16) Ambrosiodmus (16) nr nr 16 nr nr nr 0 nr nr nr 0 nr 0 Xyleborinus 310nrnr00nrnr00 nrnr0 gracilis (13) Downloaded from Xyleborinus 238 51 nr 20 4 0 nr 0 2 0 nr 0 1 saxesenii (309) Xyleborinus (322) 241 61 nr 20 4 0 nr 0 2 0 nr 0 1 Xylosandrus 92 24 4 20 1c 110 14 250 2 crassiusculus (140) Xylosandrus (140) 92 24 4 20 1 1 1 0 1 4 25 0 2 http://jee.oxfordjournals.org/ Total (928) 594 161 73 100 17 18 6 17 3 11 8 17 6

Beetle taxa are listed in the descending order of their relatedness to X. glabratus (Cognato et al. 2011). Beetles assayed for R. lauricola were: F) trapped in laurel wilt-affected avocado groves; G) reared from laurel wilt-affected avocado bolts from Miami-Dade County, FL; H) reared from laurel wilt-affected avocado bolts from Fort Pierce, FL; and I) reared from bolts of laurel wilt-affected avocado bolts from Miami-Dade County, FL, assayed live, 2015. Raffaelea lauricola was not recovered from three individuals of Premnobius cavipennis that were assayed. nr ¼ not recovered, np ¼ not processed. a Individuals confirmed as R. lauricola (see Fig. 2A). b Individuals confirmed as R. lauricola; four isolates from three individuals were recovered on CSMAþ and Malt Extract Agar (see Fig. 2B). c Individuals confirmed as R. lauricola (see Fig. 2C). by guest on January 11, 2017

Table 4. Propagule densities of R.lauricola in ambrosia beetles associated with laurel wilt-affected avocado (P.americana) in Florida

Species (total number of Mean CFUs per individual Ranges of CFUs for assayed individuals Range for individuals assayed) species/genus FGHIF G H I

Xyleborus glabratus (5) nr nr 480 nr nr nr 0–1,480 nr 0–1,480 Xyleborus bispinatus (79) 5 41 5 41 0–60 4–80 0–20 0–320 0–320 Xyleborus ferrugineus (27) 50 5 0 nr 0–100 4–6 0 nr 0–100 Xyleborus affinis (122) 3 0 0 9 0–200 0 0 0–80 0–200 Xyleborus volvulus (217) 0 30 0 0 0–20 0–1,140 0 0 0–1,140 Xyleborus (450) 2 24 46 39 0–200 0–1,140 0–1,480 0–320 0–1,480 Ambrosiodmus lecontei (16) nr nr 0 nr nr nr 0 nr 0 Ambrosiodmus (16) nr nr 0 nr nr nr 0 nr 0 Xyleborinus gracilis (13) 0 0 nr nr 0–0 0 nr nr 0 Xyleborinus saxesenii (309) 2 0 nr 0 0–200 0 nr 0 0–200 Xyleborinus (322) 2 0 nr 0 0–200 0 nr 0 0–200 Xylosandrus crassiusculus (140) 589 15 3 0 0–54,200 0–360 0–10 0 0–54,200 Xylosandrus (140) 589 15 3 0 0–54,200 0–360 0–10 0 0–54,200 Total (928) 75 13 33 10 0–54,200 0–1,140 0–1,480 0–320 0–54,200

Beetle taxa and collections are described in Table 1. nr ¼ not recovered; np ¼ not processed. Journal of Economic Entomology, 2017, Vol. 0, No. 0 5

Fig. 1. Colony phenotype of R. lauricola on a cyclohexamide-containing semi-selective medium, CSMAþ. Arrows (<) indicate colonies of R. lauricola recovered from X. bispinatus after (A) 20 d and (B) 7 d, respectively, some of which are not indicated in B. Downloaded from http://jee.oxfordjournals.org/ by guest on January 11, 2017

Fig. 2. Microsatellite markers reported by Dreaden et al. (2014), CHK and IFW, were used to confirm visual identifications of a subset of isolates with the pheno- type of R. lauricola noted in Fig. 1.(A) First and last lanes are 100-bp ladders (L), lanes 1 and 10 are DNA from an isolate recovered from X. glabratus, and lanes 2– 9 and 11–18 are DNA from isolates recovered from X. bispinatus. DNA in lanes 1–9 was amplified with primers for CHK and that in lanes 10–18 was amplified with primers for IFW. (B) First and last lanes are 100-bp ladders (L), lanes 1–4 are DNA from isolates recovered from X. affinis, lanes 5 and 9 have no template (negative control), and lanes 6 and 10 are DNA from a reference isolate of R. lauricola, CBS 127349 (Centraalbureau voor Schimmelcultures, Fungal Centre, Utrecht, The Netherlands). (C) First lane is a 100-bp ladder (L), lanes 1 and 5 are DNA from an isolate recovered from X. crassiusculus (identified in Table 2), lanes 2 and 6 are DNA from an isolate recovered from A. lecontei (identified in Table 1), lanes 3 and 7 are positive-control DNA for CBS 127349, and lanes 4 and 8 have no template (negative control). Faint bands in A and B are primer-dimer bands, which result from complementarity between primers (Dieffenbach et al. 1993). processed within 2 h to 2 d of collection from: F) Lindgren funnel silkbay (E in Tables 1 and 2), swamp bay (C in Tables 1 and 2), and traps deployed in 2012 in different avocado orchards in Miami- avocado (H in Tables 3 and 4) were compared using the Welch ana- Dade County; G) rearing containers in 2012 with avocado bolts lysis of variance (ANOVA), which adjusts for unequal variances from 14 trees in 10 orchards in Miami-Dade County; H) rearing (SAS 9.3). Means were separated using Tukey–Kramer comparison containers in 2012 with avocado bolts from an experimental plant- lines for least squares means of the species. Data that were available ing in Fort Pierce, FL (USDA-ARS); and I) rearing containers in for X. glabratus reared from redbay were not included in these com- 2015 with avocado bolts from different avocado orchards in Miami- parisons, as they were obtained from bolts harvested a significant Dade County. In total, 16 individuals of A. lecontei, three of distance from the geographical origin of the silkbay, swamp bay, Premnobius cavipennis Eichhoff, 13 of X. gracilis, 309 of X. saxese- and avocado samples. Campbell et al. (2016) had previously shown nii, 122 of X. affinis,79ofX. bispinatus,27ofX. ferrugineus,5of that the geographical origin of sample trees influenced R. lauricola X. glabratus, 217 of X. volvulus, and 140 of X. crassiusculus were recovery from X. glabratus. assayed. The effect of differences in beetle storage on R. lauricola recov- To determine whether the host tree in which a beetle developed ery was examined for subsets of X. glabratus (n ¼ 23) caught in affected the amount of R. lauricola recovered from their mycangia, flight in Highlands County, which were not included in Tables 1–4. mean CFUs recovered from X. glabratus (n ¼ 75) reared from Individuals were refrigerated for 0, 10 or 53 d before assay, and 6 Journal of Economic Entomology, 2017, Vol. 0, No. 0

Table 5. Isolates from Tables 1 and 3 for which microsatellite markers confirmed visual identifications as R.lauricola

Beetle species No. of isolates tested Number of isolates positive Source of beetle(s)

CHK IFW

Ambrosiodmus lecontei 1 1 1 Reared from laurel wilt-affected redbay Xyleborus glabratus 8 8 8 Reared from laurel wilt-affected silk bay Xyleborus bispinatus 8 8 8 Reared from laurel wilt-affected avocado Xyleborus affinis 4 4 4 Reared from laurel wilt-affected avocado Xylosandrus crassiusculus 1 1 1 Trapped in laurel wilt-affected avocado orchard

Taxon-specific microsatellite markers, CHK and IFW, developed by Dreaden et al. (2014), were used to identify putative isolates of R. lauricola. mean CFU recoveries for R. lauricola were compared with a 63%). Furthermore, considerably higher CFUs of R. lauricola were Kruskal–Wallis ANOVA (SAS 9.3); post hoc Wilcoxon–Mann– found in X. glabratus than in the other species. With the exception Whitney tests (also nonparametric) were used to determine which of a rare individual of X. crassiusculus, from which 54,200 CFUs groups differed. Bonferroni correction was applied to control the were recovered (Fig. 2C, Table 4), CFUs per individual were often familywise error rate (Bonferroni 1950). In this case, the critical one to three orders of magnitude greater in X. glabratus than in the value, P ¼ 0.05, was divided by three, which was the number of other species (Tables 2 and 4). comparisons made; thus, differences were significant at P ¼ 0.017. Over all experiments, R. lauricola was recovered from A. lecon- Downloaded from tei, X. andrewesi, X. gracilis, X. saxesenii, X. affinis, X. bispinatus, X. ferrugineus, X. glabratus, X. volvulus, and X. crassiusculus (Fig. Results 2, Tables 1–4). This is the first record of R. lauricola in A. lecontei, Raffaelea lauricola was routinely identified based on its characteris- X. andrewesi, and X. bispinatus. tic colony phenotype on CSMAþ (Fig. 1). In no case was a putative phenotype of R. lauricola not validated with the diagnostic micro- http://jee.oxfordjournals.org/ satellite markers CHK and IFW that were developed by Dreaden Discussion et al. (2014) (Fig. 2, Table 5, and data not shown). Thus, visual Raffaelea lauricola had been isolated previously from several differ- identification of R. lauricola on CSMAþ was reliable for identifying ent species of ambrosia beetle (Carrillo et al. 2014; Harrington et al. and enumerating the species. 2010; Harrington and Fraedrich 2010; Ploetz et al. 2012, 2013). We The condition of the beetles that were assayed influenced R. laur- summarize the results from published (Carrillo et al. 2014, icola recovery, as a mean of three CFUs was isolated from dead indi- Campbell et al. 2016) and several unpublished studies (Campbell viduals in two experiments with native Persea spp. (A in Table 2) 2014, Ploetz et al. and Saucedo et al., unpublished data), and report,

compared with 1,557, 2,783, 2,872, and 6,970 CFUs in four experi- for the first time, R. lauricola in three additional species. We also by guest on January 11, 2017 ments with living beetles (Table 2). Refrigeration, which had been summarize the extent to which different species and genera of am- used previously to store beetles before assay (Harrington et al. 2011, brosia beetles carry the fungus and the impact different host tree spe- Harrington and Fraedrich 2010), had a significant impact on recov- cies had on these observations. Hopefully, these results will prompt ery (v2 ¼12.92, df ¼ 2, P < 0.0016). Although more CFUs additional research on these topics. (2,143 6 664, n ¼ 7) were isolated from unrefrigerated beetles (0 d) Although the movement of symbionts among ambrosia beetle than from refrigerated ones for 10 d (1,717 6 1,429, n ¼ 6) or 53 d species was recognized previously (Batra 1966, Carrillo et al. 2014, (0 6 0, n ¼ 10), only recovery after 53 days differed significantly Gebhardt et al. 2004, Kostovcik et al. 2015), the magnitude and from the other treatments. Host tree species also affected recovery, speed with which this has occurred for R. lauricola appears to be un- as the fungus was isolated from 34% (246 of 724) of the beetles that precedented. In little more than a decade after its introduction into were associated with native Persea spp., but only 6% (58 of 931) of the United States, R. lauricola has been laterally transferred from X. those that were associated with avocado (Tables 1 and 3). Mean re- glabratus to at least nine other ambrosia beetle species. covery of R. lauricola also differed among individuals of X. glabra- It seems probable that some of the other species are involved in tus reared from silkbay, avocado, and swamp bay (F ¼ 5.35, the ongoing epidemic in avocado orchards in South Florida (Ploetz numerator df ¼ 2, denominator df ¼ 19.6, P < 0.0001); the greatest et al. 2016). In no-choice experiments, Carrillo et al. (2014) reported recoveries of CFUs were from silkbay (6,970 6 2,620; n ¼ 20), fol- that six and two species other than X. glabratus transmitted R. lauri- lowed by individuals from swamp bay (2,783 6 1,994; n ¼ 50) and cola to potted redbay and avocado trees, and that laurel wilt de- then avocado (480 6 606, n ¼ 5). veloped in six and one of the interactions, respectively. Considering The data were also influenced by the prevalence of X. glabratus the rarity of X. glabratus in laurel wilt-affected avocado orchards in the native Persea spp., compared with avocado, and the greater and the ability of other ambrosia beetle species to harbor and trans- prevalence of R. lauricola in X. glabratus compared with the other mit the pathogen, further study of these species is warranted. species of ambrosia beetle that were examined. Xyleborus glabratus Additional attention should be given to whether species other was usually the most frequent ambrosia beetle recovered from laurel than X. glabratus may expand the host range of R. lauricola and ex- wilt-affected native Persea spp., but it was hardly ever associated acerbate the spread of laurel wilt (Ploetz et al. 2016). In the present with laurel wilt-affected avocado, particularly in the CAPA (Carrillo assays, host tree species affected whether, and the extent to which et al. 2012, unpublished data). Considering only live specimens, R. the pathogen was recovered from an individual. More information lauricola was present in 89% of the individuals of X. glabratus (101 is needed on the acquisition of this fungus by different species of am- of 113 individuals in experiments B, C, D, E, and H) compared with brosia beetle and its transmission to different host tree species within lower proportions of all other species (means between 0% and and outside the Lauraceae. Journal of Economic Entomology, 2017, Vol. 0, No. 0 7

We are only beginning to understand symbioses that are estab- individual of X. crassiusculus in Table 4 with 54,200 CFUs). lished between R. lauricola and different ambrosia beetle species. Despite their rarity, highly infested individuals of species with meso- Based on the present results, there may be a phylogenetic signature notal mycangia could be important vectors in the avocado for which beetle species harbor R. lauricola. Note that the beetle pathosystem. taxa in Tables 1–4 are listed in the descending order of their related- Bark beetles are closely related evolutionary predecessors of am- ness to X. glabratus, as reported by Cognato et al. (2011). Although brosia beetles that colonize the phloem, rather than the xylem, of relatively few species were assayed during the present work, R. laur- host trees (Farrell et al. 2001). They are vectors of fungal tree patho- icola was most prevalent in Xyleborus congeners (mean incidences gens (Ploetz et al. 2013), and those associated with Dutch elm dis- of 52% and 11% for the native Persea and avocado experiments, re- ease may be the most well-known examples (Webber 1990). At least spectively), and that the closest relative of X. glabratus in these ana- 10 species of bark beetles, usually Scolytus spp. (Coleoptera: lyses, X. bispinatus, exhibited the highest frequencies of association Curculionidae: Scolytinae), are known to be phoretic vectors of the with R. lauricola after X. glabratus (considering only live specimens, Ophiostoma spp. that cause the Dutch elm disease (i.e., carry the mean incidences of recovery of 91% vs. 63% and 60% vs. 35% in pathogens as external, casual contaminants; Webber 1990). native Persea and avocado, respectively). Previous reports had sug- Bark beetles in the genus Hypothenemus (Coleoptera: gested that Raffaelea spp. were the primary symbionts of Xyleborus Curculionidae: Scolytinae) are very common in tropical and sub- spp. (Harrington et al. 2010, Kostovcik et al. 2015). In the present tropical regions, including major avocado-production areas. They study, Xyleborinus spp. exhibited a less clear tendency toward a re- are abundant in South Florida (Johnson et al. 2016), and are found lationship with R. lauricola (corresponding means of 17% and 1% in commercial avocado orchards that are affected by laurel wilt for the native Persea and avocado experiments). (Kendra and Carrillo, unpublished data). Hypothenemus is the most Downloaded from In summarizing results from a study of microbial communities in commonly intercepted genus of scolytine beetle at U.S. ports of entry mycangia of X. affinis, X. ferrugineus, and X. crassiusculus, (Haack 2001), and many species that are found in South Florida are Kostovcik et al. 2015 suggested that different types of mycangia also found in Mexico (Kirkendall 1984), Central America (Wood may “support functionally and taxonomically distinct” symbioses. 1982), and South America (Wood 2007). However, with the excep- They concluded that the mandibular mycangium found in tion of Hypothenemus hampei (Ferrari) (coffee berry borer), they Xyleborus enabled the establishment of a broader array of sym- have been rarely studied because of their minute size and difficulties http://jee.oxfordjournals.org/ bionts than the larger and more exposed mesonotal mycangium pos- with species identification (Johnson et al. 2016). sessed by X. crassiusculus (which is also found in Xyleborinus). Previously, Morales-Ramos et al. (2000) reported that a fungus, Hulcr et al. (2011) studied how different ambrosia beetle species Fusarium solani, was commonly associated with H. hampei; they responded to cultures (volatiles) of R. lauricola. In olfactometer indicated that asperities and flattened setae on the beetle’s cuticle assays, adult females of X. saxesenii and X. crassiusculus were probably assisted the phoretic movement of that fungus to coffee repelled by R. lauricola, which corresponds with the uncommon re- beans. Because of their abundance in avocado-production areas and covery of the pathogen from these species (Tables 1 and 3). ability to phoretically transmit fungi, we suggest that Interestingly, another beetle that carried R. lauricola more frequently

Hypothenemus spp. should be examined as potential vectors of R. by guest on January 11, 2017 in the present work, X. ferrugineus, had a net nonresponse in the ol- lauricola. factometer work (was repelled about as often as it was attracted (156 vs. 132, P ¼ 0.16), whereas X. glabratus was significantly attracted to R. lauricola (in 54 of 84 assays, P ¼ 0.004; Hulcr et al. 2011). Acknowledgements Whether a beetle avoids, ignores, or is attracted to R. lauricola We thank Juliette Hubbard and Wayne Montgomery for technical assistance. may affect whether it is a factor in the epidemiology of this disease This work was supported, in part, by USDA grants from the Florida on avocado (Ploetz et al. 2016). How do commonly encountered Department of Agriculture and Consumer Services, FDACS Sponsor #019730 species that are repelled by R. lauricola in olfactometer assays, such and #021757, the USDA-ARS National Plant Disease Recovery System, the as X. saxesenii and X. crassiusculus, avoid it in trees that are af- Florida Avocado Administrative Committee, and NIFA grant 201551181- fected by laurel wilt, and might they still be occasional vectors of the 24257. pathogen despite its repellency in olfactometers? What do different beetle species farm in their natal galleries in laurel wilt-affected avo- cado trees, and do repelled species weed R. lauricola out of their gar- References Cited dens or do their symbionts inhibit its growth? Are beetles that are Atkinson, T. H., D. Carrillo, R. E. Duncan, and J. E. Pena.~ 2013. Occurrence not repelled by the fungus able to cultivate it and use it as a food of Xyleborus bispinatus (Coleoptera: Curculionidae: Scolytinae) Eichhoff in source, and Does R. lauricola indirectly (or directly) manipulate the southern Florida. Zootaxa 3669: 96–100. behavior of X. glabratus or other Xyleborus congeners Barton, C., C. Bates, B. Cutrer, J. Eickwort, S. Harrington, D. Jenkins, D. interspecifically? McReynolds Stones, L. Reid, J. J. Riggins, and R. Trickel.2016. Although the extent to which the other ambrosia beetle species Distribution of counties with laurel wilt disease by year of initial detection. are factors in the laurel wilt epidemic on avocado requires further (http://www.fs.usda.gov/Internet/FSE_DOCUMENTS/fseprd513913.pdf) study, we are beginning to understand the vector portion of this puz- Batra, L. R. 1966. Ambrosia fungi: Extent of specificity to ambrosia beetles. zle. By virtue of their affinity for Raffaelea spp. and mandibular Science 173: 193–195. Bonferroni, C. 1950. Sulle medie multiple di potenze. Boll. Dell’Unione mycangia (Harrington et al. 2010, Kostovcik et al. 2015), species of Matematica Ital. 5: 267–270. Xyleborus may be expected to foster symbioses with R. lauricola Campbell, A. S. 2014. Characterizing pathogen-vector-host interactions in more readily than species in other genera. In contrast, species with laurel wilt, a disease of Avocado. PhD thesis, University of Florida, mesonotal mycangia should be able to carry greater quantities of the Gainesville, FL. pathogen (based on the larger size of this organ, relative to mandibu- Campbell, A. S., R. C. Ploetz, T. Dreaden, P. Kendra, and W. Montgomery. lar mycangia), even if they are infested with R. lauricola infrequently 2016. Geographic variation in mycangial communities of Xyleborus glabra- and not attracted to or repelled by the fungus in olfactometers (note tus. Mycologia 108: 657–667. 8 Journal of Economic Entomology, 2017, Vol. 0, No. 0

Carrillo, D., R. E. Duncan, and J. E. Pena.~ 2012. Ambrosia beetles Hulcr, J., R. Mann, and L. L. Stelinski. 2011. The scent of a partner: Ambrosia (Coleoptera: Curculionidae: Scolytinae) that breed in avocado wood in beetles are attracted to volatiles from their fungal symbionts. J. Chem. Ecol. Florida. Fla. Entomol. 95: 573–579. 37: 1374–1377. Carrillo, D., R. E. Duncan, J. N. Ploetz, A. Campbell, R. C. Ploetz, and J. E. Johnson, A. J., P. E. Kendra, J. Skelton, and J. Hulcr. 2016. Species diversity, Pena.2014~ . Lateral transfer of a phytopathogenic symbiont among native phenology, and temporal flight patterns of Hypothenemus pygmy borers and exotic ambrosia beetles. Plant Pathol. 63: 54–62. (Coleoptera: Curculionidae: Scolytinae) in South Florida. Environ. Cognato, A. I., J. Hulcr, S. A. Dole, and B. H. Jordal. 2011. Phylogeny of Entomol. 45: 627–632. haplo–diploid, fungus-growing ambrosia beetles (Curculionidae: Kendra, P. E., W. S. Montgomery, J. S. Sanchez, M. A. Deyrup, J. Niogret, and Scolytinae: Xyleborini) inferred from molecular and morphological data. N. D. Epsky. 2012. Method for collection of live redbay ambrosia beetles, Zool. Scr. 40: 174–186. Xyleborus glabratus (Coleoptera: Curculionidae: Scolytinae). Fla. Entomol. Dieffenbach, C. W., T. M. Lowe, and G. S. Dveksler.1993. General concepts 95: 513–516. for PCR primer design. PCR Methods Appl. 3: S30–S37. Kendra, P. E., W. S. Montgomery, J. Niogret, and N. D. Epsky. 2013. An un- Dreaden, T. J., J. M. Davis, C. L. Harmon, R. C. Ploetz, A. J. Palmateer, P. S. certain future for American Lauraceae: a lethal threat from redbay ambrosia Soltis, and J. A. Smith. 2014. Development of multilocus PCR assays for beetle and laurel wilt disease. Am J. Plant Sci. 4: 727–738. Raffaelea lauricola, causal agent of laurel wilt disease. Plant Dis. 98: Kendra, P. E., W. S. Montgomery, J. Niogret, G. E. Pruett, A. E. Mayfield, III, 379–383. M. MacKenzie, M. A. Deyrup, G. R. Bauchan, R. C. Ploetz, and N. D. Farrell, B. D., A.S.O. Sequeira, B. C. Meara, B. B. Normark, J. H. Chung, and Epsky. 2014. North American Lauraceae: Terpenoid emissions, relative at- B. H. Jordal. 2001. The evolution of agriculture in beetles (Curculionidae: traction and boring preferences of redbay ambrosia beetle, Xyleborus glab- Scolytinae and Platypodinae). Evolution 55: 2011–2027. ratus (Coleoptera: Curculionidae: Scolytinae). PLoS ONE 9: e102086. Fraedrich, S. W., T. C. Harrington, R. J. Rabaglia, M. D. Ulyshen, A. E. Kirkendall, L. R. 1984. Notes on the breeding biology of some bigynous and Mayfield, III, J. L. Hanula, J. M. Eickwort, and D. R. Miller. 2008. A fungal monogynous species of Mexican bark beetles (Scolytidae: Scolytus, Downloaded from symbiont of the redbay ambrosia beetle causes a lethal wilt in redbay and Thysanoes, Phloeotribus) and records for associated Scolytidae (Hylocurus, other Lauraceae in the southeastern USA. Plant Dis. 92: 215–224. Hypothenemus, Araptus) and Platypodidae (Platypus). Z. Ang. Entomol. 97: 234–244. Gebhardt, H., D. Begerow, and F. Oberwinkler. 2004. Identification of the Kostovcik, M., C. C. Bateman, M. Kolarik, L. L. Stelinski, B. H. Jordal, and J. ambrosia fungus of Xyleborus monographus and X. dryographus Hulcr. 2015. The ambrosia symbiosis is specific in some species and promis- (Coleoptera: Curculionidae, Scolytinae). Mycol. Prog. 3: 95–102. cuous in others: evidence from community pyrosequencing. ISME J. 9: Haack, R. A. 2001. Intercepted Scolytidae (Coleoptera) at US ports of entry: 126–138. http://jee.oxfordjournals.org/ 1985-2000. Integr. Pest Manag. Rev. 6: 253–282. Kuhnholz, S., J. H. Borden, and A. Uzunovic. 2003. Secondary ambrosia bee- Haack, R. A. 2006. Exotic bark- and wood-boring Coleoptera in the United tles in apparently healthy trees; Adaptions, potential causes and suggested States: Recent establishments and interceptions. Can. J. For. Res. 36: research. Inter. Pest Manag. Rev. 6: 209–219. 269–288. Mayfield, A. E., III, J. A. Smith, M. Hughes, and T. J. Dreaden. 2008. First re- Harrington, T. C. 1981. Cycloheximide sensitivity as a taxonomic character port of laurel wilt disease caused by a Raffaelea sp. on avocado in Florida. in Ceratocystis. Mycologia 73: 1123–1129. Plant Dis. 92: 976. Harrington, T. C., D. N., Aghayeva, and S. W., Fraedrich 2010. New combi- Morales-Ramos, J. A., M. G. Rojas, H. Sittertz-Bhatkar, and G. Saldana.~ nations in Raffaelea, Ambrosiella, and Hyalorhinocladiela, and four new 2000. Symbiotic relationship between Hypothenemus hampei (Coleoptera: species from the rebay ambrosia beetle, Xyleborus glabratus. Mycotaxon

Scolytidae) and Fusarium solani (Moniliales: Tuberculariaceae). Ann. by guest on January 11, 2017 111: 337–361. Entomol. Soc. America. 93: 541–547. Harrington, T. C., and S. W. Fraedrich. 2010. Quantification of propagules of Ploetz, R. C., M. A. Hughes, P. E. Kendra, S. W. Fraedrich, D. Carrillo, L. L. the laurel wilt fungus and other mycangial fungi from the redbay ambrosia Stelinski, J. Hulcr, A. E. Mayfield, III, T. L. Dreaden, J. H. Crane, et al. beetle, Xyleborus glabratus. Phytopathology 100: 1118–1123. 2016. Recovery plan for laurel wilt of Avocado, caused by Raffaelea lauri- Harrington, T. C., S. W. Fraedrich, and D. Aghayeva. 2008. Raffaelea lauri- cola. National Plant Disease Recovery System. Homeland Security cola, a new ambrosia beetle symbiont and pathogen on the Lauraceae. Presidential Directive Number 9 (HSPD-9). Plant Health Prog. Mycotaxon 102: 399–404. Ploetz, R. C., J. Hulcr, M. Wingfield, and Z. W. de Beer. 2013. Ambrosia and Harrington, T. C., H. Y. Yun, S. S. Lu, H. Goto, D. N. Aghayeva, and S. W. -associated tree diseases: Black Swan events in tree pathology? Fraedrich. 2011. Isolations from the redbay ambrosia beetle, Xyleborus Plant Dis. 95: 856–872. glabratus, confirm that the laurel wilt pathogen, Raffaelea lauricola, origi- Ploetz, R. C., J. E. Pena,~ J. A. Smith, T. L. Dreaden, J. H. Crane, T. Schubert, nated in Asia. Mycologia 103: 1028–1036. and W. Dixon. 2011. Laurel wilt is confirmed in Miami-Dade County, cen- Hughes, M. A. 2013. The evaluation of natural resistance to laurel wilt disease ter of Florida’s commercial avocado production. Plant Dis. 95: 1599. in redbay (Persea borbonia). Ph. D. Thesis. University of Florida, Ploetz, R. C., J. M. Pe´rez-Martine´z, J. A. Smith, M. Hughes, T. J. Dreaden, S. Gainesville. A. Inch, and S. Y. Fu. 2012. Responses of avocado to laurel wilt, caused by Hughes, M. A., S. A. Inch, R. C. Ploetz, H. L. Er, A.H.C. van Bruggen, and J. Raffaelea lauricola. Plant Pathol 61: 801–808. A. Smith. 2015a. Responses of swamp bay, Persea palustris, and avocado, Ploetz, R. C., B. Schaffer, A. I. Vargas, J. L. Konkol, J. Salvatierra, and R. Persea americana, to the laurel wilt pathogen, Raffaelea lauricola. For. Wideman. 2015. Impact of laurel wilt, caused by Raffaelea lauricola,on Pathol. 45: 111–119. leaf gas exchange and xylem sap flow in avocado, Persea americana. Hughes, M. A., J. J. Riggins, F. Koch, A. Cognato, C. Anderson, T. J. Phytopathology 105: 433–440. Dreaden, J. P. Formby, R. C. Ploetz, and J. A. Smith. 2016. The laurel wilt Rabaglia, R. J., S. A. Dole, and A. I. Cognato. 2006. Review of American story: introduction and impact of an exotic vector (Xyleborus glabratus) Xyleborina (Coleoptera: Curculionidae: Scolytinae) occurring north of and pathogen (Raffaelea lauricola). Phytopathology 106S: 573. Mexico, with an illustrated key. Ann. Entomol. Soc. Am. 99: 1034–1056. Hughes, M. A., J. A. Smith, R. C. Ploetz, P. E. Kendra, A. E. Mayfield, III, J. L. Webber, J. F. 1990. Relative effectiveness of Scolytus scolytus, S. multistriatus Hanula, J. Hulcr, L. L. Stelinski, S. Cameron, J. J. Riggins, et al. 2015b. and S. kirschi as vectors of Dutch elm disease. Eur. J. For. Pathol. 20: Recovery plan for laurel wilt on redbay and other forest species caused by 184–192. Raffaelea lauricola and disseminated by Xyleborus glabratus. Plant Health Wood, R. L. 1982. The bark and ambrosia beetles of North and Central Prog. 16: 173–210. America (Coleoptera: Scolytidae), a taxonomic monograph. Great Basin Hulcr, J., and R. R. Dunn. 2011. The sudden emergence of pathogenicity in Nat. Mem. 6: 1–1359. insect–fungus symbioses threatens naive forest ecosystems. Proc. R. Soc. B Wood, R. L. 2007. Bark and ambrosia beetles of South America (Coleoptera: Sci. 278: 2866–2873. Scolytidae). Brigham Young University Press, Provo, UT.