RESPONSES OF FLEA BEETLE Phyllotreta ...

42 downloads 0 Views 782KB Size Report
Louda and Mole, 1991). The mustard oil allyl isothiocyanate (AITC), a glu- .... on top of the pipette. A 30 x 20 rom diameter foam plug was placed in the hole ...
#9186

Journal of Chemical Ecology, Vol. 31, No.8, August 2005 (©2005) DOl: 10.1007/sl 0886-005-5929-2

RESPONSES OF FLEA BEETLE Phyllotreta cruciferae TO SYNTHETIC AGGREGATION PHEROMONE COMPONENTS AND HOST PLANT VOLATILES IN FIELD TRIALS

JULIANA 1. SOROKA,I,* ROBERT 1. BARTELT,2 BRUCE W. ZILKOWSKI,2 and ALLARD A. COSSE2 lAgriculture and Agri-Food Canada, Saskatoon Research Centre, 107 Science Place, Saskatoon, Saskatchewan, Canada, S7N OX2 2USDA Agricultural Research Service, National Center for Agricultural Utilization Research, Crop Bioprotection Research Unit, 1815 N. University Street, Peoria, IL 61604, USA

(Received December 23, 2004; revised April 6, 2005; accepted April 9, 2005)

Abstract-Male-specific compounds, previously identified from Phyllotreta cruciferae and synthesized or isolated from natural sources, attracted both sexes of the beetle in field trials and therefore function as components of a male-produced aggregation pheromone. Six field experiments of 7 to 10 d duration each were conducted over 2 yr using modified boll weevil traps and two doses of pheromone. Treatments containing two doses of allyl isothiocyanate (AITC), a breakdown product of glucosinolates in Brassica napus L., a host plant of the beetles, were included in the study. A dose response was observed for both the pheromone components and AlTC, and combinations of the pheromone and AlTC generally attracted greater numbers of flea beetles than did either component itself. This increased attraction to a combination of beetle-produced compounds and host odors has not been previously demonstrated in halticine beetles and could help explain patterns of movement by P. cruciferae into field crops. Key Words-Phyllotreta cruciferae, crucifer-feeding flea beetle, aggregation pheromone, Chrysomelidae: Alticinae, field trials, kairomone.

* To whom correspondence should be addressed.

E-mail: [email protected]

1829 0098-0331/05/0800-1829/0 © 2005 Springer Science + Business Media, Inc.

1830

SOROKA ET AL. INTRODUCTION

The crucifer flea beetle Phyllotreta cruciferae (Goeze) (Coleoptera: Chrysomelidae) is a major pest of canola and crucifer crops in North America and is consistently found feeding in oilseed rape or canola fields (Brassica rapa L. or B. napus L.) on the Canadian prairies (Burgess, 1977; Lamb and Turnock, 1982; Lamb, 1989). Damage and control costs for this and a related species, P. striolata (F.), exceed $300 million annually (Knodel and Olson, 2002). Insecticides are the main means of control of flea beetles in canola crops, with more than 90% of the 5 million ha seeded to canola in North America treated with insecticides (Waite et aI., 2001). Integrated methods of management of this chronic agricultural pest are urgently needed, including tools based on pheromones and host plant volatiles. Males of P. cruciferae were previously reported to emit an aggregation pheromone that attracts both sexes based on field and laboratory experiments with live beetles (Peng and Weiss, 1992; Peng et aL, 1999). Subsequently, the volatiles emitted by the beetles were analyzed, and six sesquiterpenes were identified (Figure 1). These were produced only by males, but at least one was readily sensed by the antennae of both sexes (Bartelt et aL, 2001). Thus, these compounds or a subset of them was thought likely to constitute the pheromone. All six compounds are chiral, and only the enantiomers shown in Figure 1 are emitted by the beetles. Racemic forms (50:50 mixtures of the two enantiomers) of 1, 3, 5, and 6 were synthesized (Bartelt et aL, 2003). Subsequently, Muto et aI. (2004) synthesized the individual enantiomers of compounds 1, 3, 5, and 6, using citronellal of known configuration as the chiral starting materiaL These studies confirmed the basic structures determined by Bartelt et aI. (2001) and resolved uncertainty about the absolute configurations of these compounds. The host range of P. cruciferae is confined to the order Capparales, principally to the family Brassicaceae (Feeny et aI., 1970). All members of this family contain one or more anionic glucosinolate compounds. Glucosinolates and their metabolites are thought to act not only as deterrents for generalist insect feeders, but also as attractants and stimulants for specialist crucifer feeders such as crucifer-feeding flea beetles (Phyllotreta spp.) (Chew, 1988; Louda and Mole, 1991). The mustard oil allyl isothiocyanate (AITC), a glucosinolate breakdown product, is attractive to P. cruciferae in the field (Vincent and Stewart, 1984; Pivnick et aL, 1992). Host plants also can affect the production of, and responses to, insect pheromones. In particular, maleproduced pheromones of various bark beetles, weevils, nitidulid beetles, and other insects act synergistically with volatiles from the host plant (Landolt and Phillips, 1997). However, there is no specific information about such interactions in halticine flea beetles.

1831

FLEA BEETLE PHEROMON"E

5

FIG. 1. Male-specific compounds previously identified from P. cruciferae. The compounds are numbered in the order of elution from a nonpolar GC column.

The primary objective of this research was to test whether the compounds identified from male P. cruciferae and subsequently synthesized were attractive to beetles in the field, relative to controls and to the host-derived attractant, AITC. A second objective was to explore beetle responses to the combination of male-specific compounds and the host plant volatile AITC.

METHODS AND MATERlALS

Male-Specific Compounds. Racemic compounds 1, 3, 5, and 6 were synthesized as described previously (Bartelt et aI., 2003). Compound 2 was not available for field tests. Compound 4, y-cadinene, was isolated from citronella essential oil· (Vig et aI., 1970) in a procedure described in the Appendix (available online at www.springerlink.com; search for DOl: 10.1007/s10886005-5929-2; Electronic Supplementary Material can be found at the end of the article).

1832

SOROKA ET AL.

Pheromone Lures. Rubber septa (McDonough, 1991) were used as dispensers for the beetle-produced compounds in field experiments. Before loading the chemicals, all septa (20 x 11 mm diameter, Aldrich Chemical Co., Milwaukee, WI, USA) were cleaned by Soxhlet extraction for 8 hr with methylene chloride. Septa with the high dose were prepared such that the total weight of the natural enantiomers of 1,3,4,5, and 6 was 500 )..Lg and that the proportions of these enantiomers in the initial emissions from the septa were the same as the emissions from male P. cruciferae (Bartelt et aI., 2001). The correct blend was found by an iterative approach because the proportions of the applied compounds differed from those emitted due to widely differing volatilities. Details are given in the Results section. Low-dose septa were prepared with 50 )..Lg of the same blend used in the higher dose septa. After loading the chemicals, all septa were aired in a hood until the solvent had evaporated and then stored in tightly closed vials at - 20 c C until used for field tests. The emissions from four high-dose septa were monitored in the laboratory for 2 wk to evaluate changes over time. Each septum was placed into a 50-ml flask through which air was passed at a rate of 100 ml/min. Volatile emissions were collected onto Super Q filters (Bartelt et aI., 2001). The collection apparatus was kept in an incubator in the dark at 27 c C. Volatiles were recovered every 1-3 d by rinsing the filters with hexane and quantified, relative to the internal standard nonadecane, by gas chromatography (GC) with a flame ionization detector. Emission per day was modeled for each component over time by linear regression. Host Volatile Lures. Technical grade (95%) allyl isothiocyanate (AITC) (Sigma-Aldrich Chemicals, Oakville, ON, Canada) was placed in the traps in one of two configurations. In high-dose AITC lures, disposable glass culture vials (50 x 6 mm, 3.8 mm inside diameter, 1 m1 capacity, Kimble Co., Nepean, ON, Canada) were filled with 0.2 m1 AITC so that approximately 40 mm of the vial remained empty. At the end of the first trial, some of the vials contained one to several flea beetles in or near the surface of the AITC. Thus, in all subsequent trials the top of each vial was plugged with a small piece of foam rubber to prevent beetles from entering the vial. Evaporation rates of AITC were tested from capillary tubes with and without plugs by weighing before and after several days in the field. Plugs did not appear to hamper evaporation of the AITC. For low-dose AITC lures, a 200-)..L1 capillary tube (125 x 2.2 mm, 1.6 mm inside diameter, Go1dseal Glassware, Becton Dickinson and Co, Parsippany, NJ, USA), was filled with approximately 10 )..L1 AITC so that 70 rom of the inside length of the tube was not filled. All vials and microcapillary tubes containing the AITC were weighed immediately before being placed in and after removal from the field traps to determine the average rate of AITC emission per day.

1833

FLEA BEETLE PHEROM01\TE

1 - - - - - - plastic collecting vial "'--"::Jlj;;;;;=::S!~----- Eppendorf tube, tip removed - - - - - - - - # 1 0 cork

~------- 2 litre plastic soda bottle top c."1--~------

rubber septum with pheromone

-\+--~-----foam

~----

111111111111---

plug

trap entrance

nylon net

1 1 - - - - - 4 - - - - capillary tube 1 - - - - - yellow boll weevil trap base i - - - - - I - - - - - allyl isothiocyanate, low cone.

l(f-~--

wire anchor

FIG. 2. Schematic diagram of trap used to test attractiveness of synthetic flea beetle compounds and/or AITC to crucifer-feeding flea beetles, Phyllotreta spp., in the field near Saskatoon, SK., 2001-2002. Trap shows Treatment 9, low concentration of AITC in a capillary tube, and low concentration of pheromone on a rubber septum. Enlargement shows Treatment 6, high concentrations of both allyl isothiocyanate and pheromone.

Traps. Yellow plastic boll weevil traps were used (Story Chemical Co., Starkville, MS, USA) modified as follows (Figure 2): the funnel-shaped plastic top of each trap was replaced by the upper third portion of a clear, 2-1 plastic soft drink bottle and glued to the base of the trap. A No. 10 cork (30 x 20 rom diameter) was drilled with a No.2 cork borer, and a Gilson Diamond DlOOO plastic micropipette tip (Mandel Co., Guelph, ON, Canada) was placed through the cork with the tip up. Approximately 1 cm of the micropipette tip end was removed to form a hollow tube of the remainder, and the cork and modified pipette were inserted into the neck of the soft drink bottle. A #2 sized hole was bored into the side of a 100-ml plastic collecting vial, which was placed firmly on top of the pipette. A 30 x 20 rom diameter foam plug was placed in the hole

1834

SOROKA ET AL.

at the top of the yellow base through or onto which was inserted the appropriate treatment. The three lure elements, that is, glass vials with AITC, capillary tubes with AITC, and rubber septa with beetle-produced compounds, were inserted onto the foam plug so that the top of each was at the same level above the top of the foam plug. In the first two trials, larger insects congested the traps. In Trials 3 to 6, a strip of nylon netting wide enough to hang one third of the way down the yellow base was taped to the soda bottle top to prevent access by large insects to the trap. Entry by flea beetles and other small insects into the traps was not impeded by the netting. Field SUes and Experimental Design. Field trials were conducted at the Agriculture and Agri-Food Canada, Saskatoon Research Centre farm and at the Crop Development Centre of the University of Saskatchewan, near Saskatoon, SK., Canada, latitude 52°09'N, longitude I06°34'W, for 7 to 10 d duration on six occasions: early June, late June, and early September, 2001, and mid-June, late June-early July, and early September, 2002. The trials were set up in a randomized complete block design of six replicates and nine treatments per replicate, with randomization only at the beginning of each trial. For five of the trials, traps were set in a line along the edge of canola fields, with a minimum of 10 m between traps and with traps within 0.3 m of canola (Brassica napus) plants. In the trial conducted in September 2001, the traps were placed 5 m apart within replicates, with traps separated from canola by approximately 1.5 m. The first trials in both years were arranged in an east-west orientation, whereas all other trials were arranged in a north-south orientation. Prevailing winds in the area are from the northwest. The nine experimental treatments in each trial were the high and low pheromone doses; the high and low AITC doses, the four possible combinations of high and low doses of pheromone and AITC, and the control (foam plug alone). The contents of the collecting vials were emptied daily (every second day in Trial 3) into plastic bags, returned to the laboratory, and the number of flea beetles counted. Flea beetles were identified to species level according to the characters described by Burgess (1977). In a subsample, the sex of 20 randomly selected flea beetles in each sample was determined by examining the ventral surface of the abdominal tip or by gently squashing it and examining the extruded genitalia. The tip of the male terminal abdominal segment is cup shaped; the male copulatory organ is a sclerotized, tobacco-colored, spoon-shaped aedeagus. Female abdomens are usually larger, with a smooth terminal abdominal segment and a clear spermatheca with a red line through the center, ending in cilia externally. The number of newly emerged or teneral adults from a IOO-beetle subsample from each replicate on September 10, 2001, and from a 20-beetle subsample from each treatment and replicate of each day of Trial 6 in September 2002 also was determined. Teneral flea beetle adults, those that had emerged from

1835

FLEA BEETLE PHEROMOl\!E

pupation less than 2 d from collection, had discernibly lighter elytra and other body parts than older flea beetles. Statistical Analyses. For each of the six trials, the daily trap catches were summed over the days of the experiment for each of the 54 individual trap sites (9 treatments x 6 blocks). These sums were then transformed to IOglO(X + 1) to stabilize variance, and analysis of variance was conducted using the mixed model procedure of SAS ("Proc Mixed") (SAS Institute, Inc., 2001) or Statistix software (Analytical Software, 2003). Treatments were considered as fixed effects and blocks were considered as random effects. If a significant F statistic resulted, comparisons of means were conducted. Initially, comparisons were made between the various treatments and the control using Dunnett's test (Steel and Torrie, 1980). Linear contrasts were then constructed, using means in the IOglO(X + 1) scale, to compare the overall effects of pheromone or AITC at differing doses with the four pheromone/AITC interactions. For example, the contrast for effect of the high-pheromone dose was the mean over the three treatments that contained the high-pheromone dose (high pheromone, high pheromone plus high AITC, and high pheromone plus low AITC) minus the mean over three corresponding treatments that did not contain pheromone (control, high AITC, and low AITC). The other contrasts involved analogous sets of means. Probability values for the t statistics for these contrasts are presented with the results. The antilog of each contrast value represents the factor by which the subject effect increased the trap catch above the number found in the unbaited control. In analyses of beetle sex ratios, data were transformed by arcsine square root to stabilize variances. Analyses were conducted on data from individual trials, on data combined over June and September trials, and on data combined over all trials. When a significant F statistic resulted, Tukey's Studentized comparisons of means were determined.

RESULTS

Lure Emission Data. The preparation and emission properties for the high dose of male-specific compounds are summarized in Table 1. The proportions of emitted compounds differed slightly from those that were applied because of different volatilities. Compound 1 was the most volatile and became depleted most rapidly from the lures. However, the composition of the emitted blend was relatively stable over a 1-wk period. The high-dose dispensers released 4300, 5500, 5200, 7200, 8200, and 4800 flg per day of AITC for Trials 1-6, respectively (means over 18 dispensers per experiment). The low-dose AITC dispensers had mean (N = 18) release rates of 620, 220, 340, 310, 590, and 280 flg per day for experiments

I-'

00

U.)

0\

TABLE 1. EMISSION RATES OF SYNTHETIC Phyllotreta cruciferae PHEROMONE COMPONENTS FROM HIGH-DOSE RUBBER SEPTA

Weight of natural enantiomer per septum Compound

(~lgt

Relative amounts applied (1 = 100)"

1

250 17 123 28 82

100 6.7 49.0 11.1 32.6

3 4 5

6

Natural relative emission rates (1 = 100r 100 4.4 26.6 6.0 3.9

Measured relative emission rates (1 = 100; mean ± SE for others)" Initial 100 4.4 25.2 5.7 2.8

After 7 d 100

(:1:0.3) (:!:2.8) (:1:0.8) (±0.9)

504 (±0.2)

39.0 (± 1.6) 9.2 (±0.5) 7.7 (±0.5)

Measured emission rates (flg/day) (mean ± SEt Initial 11.8 (±OA) 0.54 (±0.03) 3.5 (±0.3) 0.80 (:to.08) 0.56 (:1.:0.06)

After 7 d 8.2 (±0.3) 0041 (±0.02)

2.9 (:!0.2) 0.67 (:to.04) 0.54 (:to.03)

Synthetic compounds 1, 3, 5, and 6 were racemic, whereas compound 4 was the pure beetie-produced enantiomer. Thus, the actual gravimetric and achiral GC measurements for the amounts of 1, 3, 5, and 6 were double the values shown. SE = standard error of fitted value from regression (see text). " Ratios of compounds applied or measured are based on the amounts of natural enantiomers present. SE = standard error of fitted value from regression. C Relative emission rates based on Bartelt et aI. (2001). (l

Ul

~

...,tTJ ~

FLEA BEETLE PHEROMOl\'E

1837

1-6, respectively. Typical standard errors were in the range of 5-20% for both dosages. Release rates per day did not appear to vary appreciably over the short length of the experiments (7-10 d). For example, in the third trial, high-rate vials were removed after the second day and were found to have lost an average of 7700 Ilg AITC per trap per day, whereas over the entire 10 d of the test the average release rate was 5180 Ilg per trap per day. Flea beetle elytra were noted in some of the microcapillary tubes and may have contributed to the variability in the amount of AITC dispersed in this treatment by partially blocking air flow. Field Experiments: General Observations. Large numbers of flea beetles were captured during the study, and over 99.5% of them were P. cruciferae. The totals for P. cruciferae captured in Trials 1 to 6, respectively, were 8,610; 10,151; 13,908; 12,940; 1,460; and 73,321. Other flea beetles captured included a total of 5 P. striolata and 94 Psylliodes punctulata, numbers too low to make inferences about treatment effects. Hymenoptera of various species also were captured in the experiments. Traps in Trials 1 and 2 frequently contained up to four Banchus flavescens Cresson, a large ichneumonid parasitoid of the bertha armyworm (Mamestra configurata Walker). Flea beetles may have been hindered in entering traps already containing Banchus because numbers of beetles tended to be lower in traps with the large parasitoids.After the netting was put on in Trials 3-6, larger insects were excluded from the traps. Of the micro-Hymenoptera collected, most were Microctonus vittatus 1., a native parasitoid of P. cruciferae. Details of parasitoid responses to the various treatments will be the subject of a subsequent paper. Weather conditions had a critical effect on the number of flea beetles captured, with highest numbers found after warm, sunny, and calm days, and low numbers found after cold, rainy, or windy days. Total numbers of flea beetles caught in the 54 traps in a 24-hr period varied from 1 flea beetle on June 6, 2001 (average daily temperature 12.9°C, 13 mm precipitation, north wind 10-20 kph), to 20,061 flea beetles on September 15, 2002 (average daily temperature 19.0°C, sunny, light winds). Flea beetles displayed little activity and were not attracted to any traps in very hot weather. For example, relatively few beetles were collected in Trial 5, during which maximum temperatures exceeded 35°C for several days. Treatment Effects. Differences among treatment means were found in all six experiments (Table 2). The treatment containing high levels of both pheromone and AITC collected the greatest number of flea beetles, being more attractive than the control at P::S 0.001 in all trials except the second one. All combined pheromone-AITC treatments had greater numbers of flea beetles than did the control. Furthermore, all later treatments with combinations of attractants had numerically greater numbers of flea beetles than single attractants (Table 2). In later trials, when modifications excluded large para-

1838

SOROKA ET AL.

TABLE 2. MEAN DAILY NUMBER OF Phyllotreta cruciferae PER TRAP IN FIELD TRIALS TESTING ATTRACTION OF SYNTHETIC MALE PHEROMONE ORlAND AITC

Trial

Treatment Control Pheromone (high) Pheromone (low) AlTC (high) AITC (low) Pheromone (high) + AlTC (high) Pheromone (high) + AlTC (low) Pheromone (low) + AlTC (high) Pheromone (low) + AlTC (low) F statistic (8, 40 df) P value Pooled % SE

(Jun 5Jun 12)

2

3

4

5

6

(Jun 22Jun 29)

(Aug 31Sep 10)

(Jun 13Jun 21)

(Jun 25Ju12)

(Sep 11Sep 18)

6.99 17.1 9.23 28.9** 8.07 46.3***

17.3 22.0 13.6 28.0 13.4 27.6

16.3 123*** 58.1 *** 103*** 50.2** 209***

0.13 1.90*** 0.66 16.0*** 5.21 *** 88.1 ***

0.16 0.31 0.85 1.08 0.60 8.26***

75.0 156 105 129 171 420***

11.5

37.9

149***

33.4***

3.47***

277**

24.7*

37.9

143***

47.1***

5.00***

287**

27.3**

19.6

103***

29.9***

5.96***

223*

6.39