Journal of Chemical Ecology, Vol. 15, No. 3, 1989
P R O D U C T I O N OF ONION FLY 1 A T T R A C T A N T S A N D OVIPOSITIONAL STIMULANTS BY BACTERIAL ISOLATES C U L T U R E D ON ONION 2
S.M. H A U S M A N N 3 and J.R. MILLER Department of Entomology and Pesticide Research Center Michigan State University East Lansing, Michigan 48824 (Received August 25, 1986; accepted March 25, 1988)
Abstract--Decomposing onions at certain microbial successional stages produce potent volatile attractants and ovipositional stimulants of the onion fly, Delia antiqua (Diptera: Anthomyiidae). A reproducible source of these compounds was obtained by culturing Erwinia carotovora var. carotovora (EC) on sterile onion tissue. In laboratory choice tests, EC-inoculated onion was more attractive than Klebsiella pneumoniae (KP) cultured on onion, EC cultured on potato (a nonhost of onion fly), or the chemical synthetic baits dipropyl disulfide and an aqueous solution of 2-phenylethanol and pentanoic acid. Onion flies were mildly attracted to potato after inoculation with EC, but females did not accept EC-inoculated potato for oviposition. This work emphasizes that sources of semiochemicals may need to be defined microbiologically as well as physically and chemically. Key Words--Onion fly, Delia antiqua, Diptera, Anthomyiidae, Erwinia carotovora var. carotovora, Klebsiella pneumoniae, food attractant, ovipositional stimulant, dipropyl disulfide, 2-phenylethanol, pentanoic acid.
INTRODUCTION
Onion fly, Delia antiqua (Meigen) (Diptera: Anthomyiidae), is a major pest of commercial onions in the temperate northern hemisphere. Decomposing onions produce more potent onion fly attractants and ovipositional stimulants than do
1Diptera: Anthomyiidae. 2paper No. 12106 of the Michigan State University Agricultural Experiment Station. 3Current address: Building E402, DuPont Experiment Station, Wilmington, Delaware 19898.
905 0098-0331/89/0300-0905506.00/0 9 1989PlenumPublishingCorporation
906
HAUSMANN AND MILLER
healthy or freshly cut onions (Dindonis and Miller, 1981a; Ishikawa et al., 1981), but only at certain successional stages of microbial and physical damage (Vernon et al., 1981; Miller et al., 1984). Using gas-liquid chromatography (GLC), Vernon et al. (1981) demonstrated that changes in onion fly attractancy were associated with qualitative and quantitative changes in the headspace chemical profile of onion as decomposition progressed. Many factors influence volatile metabolite production in decomposing plant material. A complex interaction of environmental factors such as temperature, oxygen concentration, characteristics of the host plant, and presence of competing microorganisms and their metabolites may influence which organisms and volatile metabolites predominate. Under controlled growth conditions, microorganisms give reproducible headspace profiles that are often unique to a given genus, and, sometimes, species (Larsson et al., 1984). Indeed, using GLC for the qualitative and quantitative determination of volatile compounds has become an accepted technique in identifying anaerobic bacteria (Zechman and Labows, 1984) and some groups of aerobic bacteria (e.g., Lee et al, 1979). Bacteria infecting urine, meat, and other substrates are analyzed by direct headspace analysis of the sample or by subculturing the bacteria onto media that are selective or enriched in precursors to targeted metabolites (Larsson et al., 1984). Pure cultures may produce a distinctive series of volatile organics or a distinctive odor such as the grapelike 2-aminoacetophenone, a diagnostic marker for Pseudomonas aeruginosa (Cox and Parker, 1979). The variable attractancy of decomposing onion has hindered work aimed at identifying potent onion fly attractants and ovipositional stimulants. This paper describes how a reproducible source of these compounds has been obtained by culturing bacterial isolates associated with onion decomposition on sterilized onion tissue.
METHODS AND MATERIALS
Insects. Parental D. antiqua stock was collected from commercial onion fields in Eaton Rapids, Michigan. Insects used were 5 to 7 generations removed from the field. Adults were fed water and a dry artificial diet (Ticheler, 1971) and housed as described by Schneider et al. (1983), except onion foliage in ovipositional dishes was replaced by green surrogate stems (Harris and Miller, 1983). Larvae were reared in sand-filled boxes provisioned with an excess of bisectioned onion bulbs. All behavioral bioassays were conducted at 35 _+ 5 % relative humidity, 23 • I~ and in a 16 : 8 light-dark regime; alternating cool 55-W and warm 85-W white fluorescent bulbs were placed 10 cm above the cages. Three days before assaying, flies from the laboratory culture were placed in cylindrical,
B A C T E R I A L I S O L A T E S ON ONI ON
907
metal-screened cages (42 cm diam. x 46 cm). Between assays, flies were given food, water, and access to ovipositional dishes of freshly cut onion covered with moist sand. In the center of each dish we placed a 4-mm-diam. x 9-cmlong surrogate onion stem. Before each assay, the water dish was placed in the center of the cage, but food and ovipositional dishes were removed. Bacterial Cultures. Erwinia carotovora var. carotovora (EC) causes soft rot on a wide range of vegetables. Onion maggots moving through infested soil may spread the bacterium (Agrios, 1978). EC was isolated from damaged stored onions (Spartan Banner hybrid) potted in moist muck soil taken from a commercial onion field in East Lansing, Michigan. Damage was inflicted by 20 second-instar D. antiqua larvae taken from our laboratory culture and added to each potted bulb. After five days of larval feeding, 10 g of the decomposing onion tissue was homogenized with a sterile mortar and pestle in 25 ml sterile buffer (0.2 M KH2PO4 and 0.2 M K2HPO4, pH 7.0). For the isolation of softrot bacteria, dilutions of the homogenate were made in buffer and plated on a selective medium containing crystal violet and sodium polypectate (CVP) (Cuppels and Kelman, 1974). Plates were incubated overnight at 25 + 2~ Colonies of Erwinia spp. formed deep, cuplike depressions in the selective medium. They could be distinguished from pectolytic Pseudomonas spp., which form shallow, wider depressions (Cuppels and Kelman, 1974). A bacterial isolate of Erwinia sp. was obtained and designated as strain EC6. Isolate EC6 was identified using current taxonomic keys and texts (Lelliott, 1974; Kelman and Dickey, 1980). Klebsiella pneurnoniae (KP) had been stored at - 6 5 ~ in 4 % (v/v) DMSO in nutrient broth (0.3 % beef extract and 0.5 % peptone; Difco). The original strain, JM 1, was isolated and identified as a predominant microorganism from decomposing onions that were five times more attractive in the field than fresh cut onions (Miller et al., 1984). For use in experiments, colonies of both EC and KP were grown on Bacto nutrient agar (Difco) at 25 _+ 2~ Stock cultures were maintained on nutrient agar in the refrigerator at 4-6~ Preparation of Axenic Onion and Potato. Axenic plant tissues were prepared from Russet Burbank potatoes and an undetermined cultivar of yellow onions from commercial storage. The outer, papery scales and upper and lower 2 cm of bulb were removed; potatoes were peeled. Plant tissues were soaked in 75% ethanol for 20 min and flame sterilized 1-2 min. The flamed outer 0.5 cm of potato and the flamed outer two onion scales were removed with sterilized instruments and discarded. Two sizes of potato and onion tissue were cut. Squares of tissue measuring 3.5 • 3.5 x 0.5 cm and 2 x 2 • 0.5 cm were placed individually in polystyrene Petri dishes (diameters of 5.5 cm and 3.5 cm, respectively). Onion squares were cut only from the 2 to 3 outer scales. After sterile distilled water was
908
HAUSMANN AND MILLER
added to cover the dish bottom, tissues were incubated at 25 _+ 2~ Because the concentration of sulfur-and nitrogen-containing compounds in onion are known to differ in the outer and inner scales (Freeman, 1975), only tissue taken from the same onion scale (or the same potato) was used within the same cage for bioassays. Experiment 1: Attractancy of Axenic Onion. Approximately 100-125 flies of mixed sexes and ages were placed in each cylindrical cage. Wire screen baskets were hung in the tube traps of Weston and Miller (1985); all parts were autoclaved. Between 4 and 15 hr into photophase, treatments were enclosed individually in the wire baskets and assayed in choice tests by spacing traps equally along the periphery of cylindrical cages. After the bioassay, flies in each trap were sexed and counted. In each of four cages, traps were baited with either a 3.5 x 3.5 x 0.5-cm axenic onion square cut immediately before assaying, cut axenic onion held for one, two, and three days before assaying, or a petri dish filled with sterilized, distilled water. At the time of cutting, two parallel grooves ca. 1 cm apart and 2.5 cm deep were cut into each square. Onion tissues were assayed in their original water-filled polystyrene petri dishes with the covers removed.
Experiment 2: Time course of Attractancy of Bacteria Cultured on Onion. To determine the minimal time needed to incubate soft rot bacteria on onion to obtain significant differences in onion fly attractancy, traps were baited with axenic onion inoculated with a bacterial isolate and incubated one, two, or three days before assaying. These treatments were compared to freshly cut axenic onion and a water control. To inoculate onion with bacteria, two parallel grooves ca. 1 cm apart and 2.5 mm deep were cut into each square. A loopful of 24-hr inoculum was placed in each groove. Tissues were incubated at 25 _+ 2~ Experiments using EC and using KP were replicated seven times as described in experiment 1. Experiment 3: Relative Attractancy of EC- and KP- inoculated Onion. To determine the relative attractancy of the two isolates cultured on onion, traps were baited with onion inoculated with EC, KP, or EC plus KP and incubated for three days at 25 +_ 2~ before assaying along with freshly cut axenic onion. Onion was inoculated as described in experiment 2 with EC or KP inoculum placed in both grooves of a square or with EC inoculum in one groove and KP inoculum in the other~ Eight replications were conducted as described in experiment 1. Experiment 4: Attractancy of EC-Inoculated Host and Nonhost Plant. The volatile onion fly attractants produced by EC-inoculated onion may have been due to EC, regardless of the nutritive medium used, or to the interaction of EC with onion, the primary host plant of the onion fly. To determine the importance of the host plant in the production of attractants, EC was cultured on axenic onion and axenic potato, a nonhost plant of onion fly, as described in experi-
B A C T E R I A L I S O L A T E S ON ONI ON
909
ment 2. Traps were baited with EC-inoculated onion, inoculated potato, axenic onion, axenic potato, and water control. Axenic onion was freshly cut; other treatments were incubated three days before assaying. This experiment was replicated seven times as described in experiment 1. Experiment 5: Attractancy of EC-Inoculated Onion vs. Chemical Synthetics. The attractancy of onion inoculated with EC and incubated three days before assaying was compared to synthetic chemical attractants. One chemical attractant was a solution of 100 txl 2-phenylethanol (2-PhEt) and 25/zl pentanoic acid mixed in 50 ml distilled water. These chemicals were reported by Ishikawa et al. (1984) as D. antiqua attractants extracted from decomposing onions. This formulation was reported to give an optimal release rate for onion fly attractancy in the field (Ishikawa et al., 1984). Ten-milliliter glass beakers containing 10 ml of this aqueous solution were placed in traps immediately before testing. A second chemical attractant was n-dipropyl disulfide (Pr2S2), a predominant volatile secondary metabolite from fresh onions (Block, 1985). For an optimal release rate (ca. 100/~g/hr), 100/~1 Pr2S2 (purity 99%) was placed into size 3 BEEM (Ted Pella Co., Box 510, Tustin, California 92680) polyethylene enclosures (Dindonis and Miller, 1981b). One capsule per trap was used. To attain a stable release rate, filled capsules were held for 6 hr prior to assaying. The experiment was replicated four times as described in experiment 1. Experiment 6: Oviposition on EC-Inoculated Host and Nonhost Plant. Ovipositional dishes (35 ml) were filled with sterilized sand moistened with sterilized, distilled water. A 4-mm-diam x 9-cm-long surrogate onion stem stood vertically in the center of each dish. By spacing ovipositional dishes equally along the cage periphery, five different test materials were assayed simultaneously in cages containing six to eight gravid D. antiqua females. Treatments were assayed 8-14 hr into photophase, thus spanning the peak diurnal ovipositional period of D. antiqua (Havukkala and Miller, 1987). In one ovipositional experiment, the 2.0 x 2.0 x 0.5-cm squares of plant tissue, along with associated water, were placed ca. 0.5 cm beneath the sand surface immediately before assaying. No plant material was placed in the cup designated as a water control. In a second experiment, to determine whether volatiles alone from test materials stimulated oviposition, Petri dishes containing the 3.5 • 3.5 x 0.5-cm squares of tissue and water were placed in sterilized glass quart canning jars. A GLC septum was inserted into each of two holes drilled into the jar lid. Teflon tubing (1 mm ID) was inserted into each septum. One tube passed through the screened cage wall and was threaded through a septum inserted into a hole in the side of the ovipositional cup. The tubing opened ca. 5 mm from the surrogate stem and ca. 1.25 cm below the sand surface. The second tube led from the jar lid to a small plastic manifold and then to the filter of a compressed air tank. Air carrying volatiles from the jars to the ovipositional cups flowed at 10 ml/min.
910
HAUSMANN AND MILLER
In both ovipositional experiments, the treatments used were onion and potato inoculated with EC and incubated for three days before assaying, axenic potato held three days before assaying, freshly cut axenic onion, and a water control. Preparation of treatments followed the procedure outlined in experiments 1 and 2. Both experiments were replicated four times. RESULTS
Identification oflsolate EC6. Bacterial cells of EC6 were motile, straight rods occurring primarily singly, but sometimes in pairs or short chains. This isolate was gramnegative, facultatively anaerobic, oxidase negative, phosphatase negative, reduced nitrate, did not produce reducing substances from sucrose, and utilized Simmon's sodium citrate. Growth occurred at 36~ and in 5% NaC1. No special growth factors were required. Colonies on yeast extract-dextrose-CaCO2 agar (YDC) were creamy white with no diffusible pink or blue pigments; colonies on nutrient agar were transluscent white with no yellow pigments. Inoculation on sterile potato disks caused soft rot. Based on these results and the characteristic growth on CVP medium, EC6 was identified as a strain of Erwinia carotovora var. carotovora. Experiment 1: Attractancy of Axenic Onion. Sterilized onion tissues incubated one day or more before the assay lost activity (Figure 1); therefore, sterilized onions used as controls in the remaining bioassays were cut immediately before assaying.
WI3
I
I
I
I
+1 11 0
9
0
7 5
e- 3 i
0 Days
i
1 sterile
i
,
2 onion
i
3 held
FIG. 1. Onion fly attractancy as influenced by time sterilized onion tissue was held before assaying. Orthogonal contrasts after ANOVA: freshly cut onion vs. onion incubated 1, 2, and 3 days (P < 0.05); freshly cut onion vs. water control (P < 0.05); and water control vs. onion incubated 1, 2, and 3 days (P > 0.20).
911
B A C T E R I A L I S O L A T E S ON ONI ON
Experiment 2: Time Course of Attractancy of Bacteria Cultured on Onion. Attractancy of EC-inoculated onion increased with incubation times from one to three days (Figure 2A). Onion inoculated and incubated two or three days caught significantly more flies than either noninoculated onion or the water control. A similiar pattern of increasing attractancy with incubation times from one to three days was seen for KP-inoculated onion tissue (Figure 2B). A three-day incubation period was judged sufficient to obtain significant differences in attractancy between bacterially inoculated onion and sterile onion.
Experiment 3: Relative Attractancy of EC- and KP-Inoculated Onion. Traps baited with onion inoculated with EC caught significantly more flies than those baited with KP-inoculated onion (Figure 3). Onion tissue inoculated with both bacteria did not catch significantly more flies than onion inoculated with
W
A
~2o~
w a t e r @ @ @ @ 1 day 2 day 3 day IaJ
CO +1
25
.5 5
ware,@ @ @ @ 1 day 2day 3day F]o. 2. Influence of incubation time on onion fly attractancy of onion inoculated with bacterial isolates. Sterilized onion tissue ( 9 inoculated with EC (A) or with KP (B) and incubated 1, 2, or 3, days before the bioassay. Treatments marked by different letters are significantly different (Tukey's HSD all-pairwise comparisons after ANOVA; P < 0.05 forA; P < 0.10 for B).
912
HAUSMANN AND MILLER
•W)40 +1
e- 8o o t~ o e'~20 t~ .i,..a
O@ 9174 FIG. 3. Relative onion fly attractancy of EC and/or KP inoculated on sterilized onion tissue (O) and incubated 3 days before the bioassay. Treatments marked by different letters are significantly different (Tukey's HSD all-pairwise comparisons after ANOVA; P < 0.05).
either bacterial isolate alone (P > 0.20). Regardless of the bacterial isolate used, all inoculated onion baits caught significantly more flies than traps baited with noninoculated onion. The remaining experiments focused on the bacterial isolate EC.
Experiment 4: Attractancy of EC-InocuIated Host and Nonhost Plant. A higher mean trap catch was attained when onion rather than potato was inoculated with EC (Figure 4). Traps baited with EC-inoculated potato did catch significantly more flies than those baited with noninoculated potato.
Experiment 5: Attractancy of EC-Inocalated Onion vs. Chemical Synthetics. Traps baited with EC-inoculated onion caught significantly more flies than any other bait used (Figure 5). The mean fly catches obtained with traps baited with either chemical attractant or with noninoculated onion were all low and did not differ significantly from each other (P > 0.20). All baits tested caught significantly more flies than the water control, except Pr2S 2 (P > 0.20). Experiment 6: Oviposition on EC-Inoculated Host and Nonhost Plant. When test materials were placed directly in the ovipositional dishes, EC-inoculated onion received significantly more eggs than any other treatment (Table 1, experiment A). Freshly cut, sterilized onion received fewer eggs than ECinoculated onion but more eggs than all other treatments. No other significant differences were found (P > 0.20).
913
BACTERIAL ISOLATES ON ONION
1.1.145 ~-(2
09
cr
+1 t'-35 .i..+ O
r
t~
e-15 t~ I1) 5
water r----~ 0
[~-] @
FIG. 4. Production of onion fly attractants by the interaction of EC on onion fly host and nonhost tissue. Sterilized onion tissue (O) and potato tissue (D) inoculated with EC and incubated 3 days before the bioassay. Statistically significant interaction between plant tissue used and presence or absence of EC [two-way ANOVA of data transformed to (X +_ 0.5)1/2; p < 0.01)]. Orthogonal contrasts: potato vs. EC-inoculated potato (P < 0.01); onion vs. EC-inoeulated onion (P < 0.01); EC-inoculated potato vs. ECinoculated onion (P < 0.01); potato vs. onion (P > 0.20). When volatiles from these treatments were pumped into the dishes, ECinoculated onion received significantly more eggs than any other treatment group (Table 1 experiment B). No other differences were found among the remaining treatments ( P > 0.20), all of which received very few eggs. l.U 09 +1 o
25 20
b c
water
b
bc
.,.Am Pr2S 2
2-PhET
,
pentanoic acid
O@
Fro. 5. Comparison of chemical synthetics and EC-inoculated onion as onion fly attractants. Symbols and statistics as in Figure 3.
914
HAUSMANN AND MILLER
TABLE l. MEAN NUMBER OF EGGS LAIDa BY D.
antiqua IN RESPONSE TO E C
CULTURED
ON HOST AND NONHOST PLANT TISSUE
Experimentb Treatment
A
B
Ec-inoeulated onion Noninoculated onion EC-inoculated potato Noninoculated potato Water control
73.5a + 23.87 12.5b + 2.96 1.0c + 1.00 0.0c 0.0c
47.5a + 10.54 1.0b + 0.56 1.0b + 0.41 0.0b 0.0b
a Means + SE followed by the same letter within a column are not significantly different [one-way ANOVA of data transformed to (X + 0.5) x/2followed by Tukey's all-pairwise comparisons (Gill, 1978); P < 0.05]. /'Treatments presented in ovipositional dishes (experiment A) or as volatiles pumped into ovipositional dishes (experiment B).
DISCUSSION
The unstable sulfenic acids released from cut onion degrade to form various sulfur-containing volatiles that, in large part, give onion its characteristic odor and taste (Block, 1985). A number of these volatiles, (e.g., dipropyl disulfide) are known to attract and/or stimulate oviposition by the onion fly (Matsumoto and Thorsteinson, 1968; Vernon et al., 1978). As the volatiles from damaged onion cells are exhausted, the attractancy of freshly cut sterilized onion would be expected to decrease; hence, freshly cut sterilized onion was chosen as an appropriate control for assays testing the effects of culturing bacteria on sterilized onion tissue. The interaction of EC and of KP with onion tissue increased D. antiqua trap catches (Figure 2) and oviposition (Table 1). Behavioral activity in response to EC- or KP-inoculated onion increased with time (Figure 2), probably due to a rise in volatile release. As the bacterial population increases, a concommitant increase in amounts of pectolytic and cellulolytic enzymes would be expected (Start and Chatterjee, 1972). Qualitative changes in volatiles also would be expected as the chemical composition of the host changes and bacteria exhaust available nutrients. Human olfaction detects little difference between sterilized and EC-inoculated potato tissue. Indeed, many vegetables infected with E. carotovora vat. carotovora alone do not release the unpleasant odors commonly associated with decomposing plants. Saprophytic bacteria that colonize plant tissue after pathogenic infection generally are responsible for the foul odor. Two notable exceptions are crucifers and alliums, both of which release unpleasant sulfurous odors when inoculated with the soft-rot pathogen (Agrios, 1978). EC-inoculated potato
BACTERIAL ISOLATES ON ONION
915
may release simple volatile metabolites not particularly odoriferot~s but nevertheless mildly attractive to onion fly, e.g., short-chain alcohols and hydrocarbons (Stotzky and Schenck, 1976). The aqueous solution o f 2-PhEt and pentanoic acid reported by Ishikawa et al. (1984) to be a potent attractant o f D. antiqua was much less attractive than EC-inoculated onion tissue (Figure 5), suggesting that attractants associated with decomposing onions remain to be identified. Indeed, the low fly catches obtained with 2-PhEt/pentanoic acid as well as with PrzS2 demonstrate that currently known synthetic attractants do not closely simulate fresh or maximally attractive decomposing onion. Perhaps decomposing onions used as positive controls in the studies emphasizing the potency of 2-PhEt/pentanoic acid as attractants for D. antiqua were inadequate (Ishikawa et al., 1984). Even though they were significantly attracted, female onion flies did not accept EC-inoculated potato as an ovipositional site (Table 1). F o r D. antiqua, oviposition is a specific behavioral response restricted to stimuli from AlIium cepa and its closest relatives. On the other hand, attractancy can include both host-finding and food-finding. Thus, onion flies may be attracted to and feed on decomposing nonhost plants (Miller et al., 1984) but may not accept them for oviposition. The most notable points made here are: (1) that a source o f consistent highly active D. antiqua attractants has been found (this advance should greatly facilitate chemical identification of these attractants), and (2) sources o f behavior-modifying stimuli may need to be defined microbiologically as well as physically and chemically. Acknowledgments--We thank Joan Harlin for care of onion fly cultures. This work was supported by USDA Competitive Research Grant 83-CRCR-l-1204 to James R. Miller and by the Michigan Onion Committee.
REFERENCES AGRIOS,G.N. 1978. Bacterial soft rots, pp. 477-483, in Plant Pathology. Academic Press, New York. BLOCK,E. 1985. The chemistry of garlic and onion. Sci. Am. 252:114-119. Cox, C.D., and PARKER,J. 1979. Use of 2-aminoacetophenone production in identification of Pseudomonas aeruginosa. J. Clin. Microbiol. 9:479-484. CUt'PELS,D., and KELMAN,A. 1974. Evaluation of selective media for isolation of soft-rot bacteria from soil and plant tissue. Phytopathology 64:468-475. DINDONIS, L.L., and MILLER,J.R. 1981a Onion fly and little house fly host finding selectively mediated by decomposing onion and microbial volatiles. J. Chem. Ecol. 7:419-426. DINDON~S, L.L., and MILLER,J.R. 1981b. Onion fly trap catch as affected by release rates of n-dipropyl disulfide from polyethylene enclosures. J. Chem. Ecol. 7:411-418. FREEMAN,G.G. 1975. Distribution of flavour components in onion (Allium cepa L.), leek (Allium porrum ) and garlic ( Allium sativum ). J. Sci. Food Agric. 26:471-481.
9t6
HAUSMANN AND MILLER
GILL, J.L. 1978. Design and Analysis of Experiments in the Animal and Medical Sciences, Vol. 1. Iowa State University Press, Ames, Iowa. HARRIS, M.O., and MILLER, J.R, 1983. Color stimuli and oviposition behavior of the onion fly, Delia antiqua (Meigen) (Diptera: Anthomyiidae). Ann. Entomol, Soc. 76:766-771. HAVUKKALA,lJ., and MILLER, J.R. 1987. Daily periodicity in the ovipositional behavior of the onion fly, Della antiqua (Diptera: Anthomyiidae). Environ. Entomol. 16:41-44. ISHIKAWA, Y., IKESHOJ1,T., and MATSUMOTO,Y. 1981. Field trapping of the onion and seed-corn flies with baits of fresh and aged onion pulp. Appl. Entomol. Zool. 16:490-493. ISHIKAWA,Y., MATSUMOTO,Y., TSUTSUMI,M., and MITSUI, Y. 1984. Mixture of 2-phenylethanol and n-valerie acid, a new attractant for the onion and seedcom flies. Hylemya antiqua and H. platura (Diptera: Anthomyiidae). Appl. Entomol Zool. 19:448-455. KELMAN, A., and DICKEY, R,S. 1980 Erwinia. 2. Soft rot or carotovora group, pp. 31-35, in N.W. Schaad (ed.). Laboratory Guide for Identification of Plant Pathogenic Bacteria. American Phytopathology Society, St. Paul, Minnesota. LARSSON, L., MARDH, P.A., and ODHAM,G. 1984. Analysis of volatile metabolites in identification of microbes and diagnosis of infectious diseases, pp. 207-235, in G. Odham, L. Larsson, and P.-A. Mardh (eds.). Gas Chromatography/Mass Spectrometry, Applications in Microbioiogy. Plenum Press, New York. LEE, M:L., SMITH, D.L,, and FREEMAN, L.R. 1979. High resolution gas chromatographic profiles of volatile organic compounds produced by microorganisms at refrigerated temperatures. Appl. Environ. Microbiol. 37:86-90. LELLIOT% R.A. 1974. Description of the species of the genus Ervvinia, pp. 469-476, in R.F. Buchanan and N.E. Gibbons (eds.). Bergey's Manual of Determinative Bacteriology, 8th Ed. Witliams & Wilkins, Baltimore, Maryland. MATSUMOTO,Y., and THORSTEINSON,A.J. 1968. Effect of organic sulfur compounds on oviposition in onion maggot, Hylemya antiqua Meigen (Diptera: Anthomyiidae). Appl. Entomol. Zool. 3:5-12. MILLER, J.R., HARRIS, M.O., and BREZNAK,J.A. 1984. Search for potent attractants of onion flies. J. Chem. Ecol. 10:1477-1488. SCtfNEIDER, W.D., MILLER, J.R., BREZNAK, J.A., and FOBES, J.F. 1983. Onion maggot, Delia antiqua, survival and development on onions in the presence and absence of microorganisms. EntomoL Exp. AppL 33:50-56. STARR, M.P., and CHATTERJEE,A.K. 1972. The genus Erwinia: Enterobacteria pathogenic to plants and animals. Annu. Rev. Microbiol. 26:389-426. STOTZKY, G., and SCHENCK, S. 1976. Volatile organic compounds and microorganisms. C.R.C. Crit. Rev. Microbiol. 333. TICHELER, J. 1971. Rearing of the onion fly, Hylemya antiqua Meigen), with view to release sterilized insects, pp. 341-346. in Sterility Principle for Insect Control or Eradication. Proceedings of a Symposium, Athens, 1970. IAEA, Vienna. VERNON, R.S., PmRCE, H.D., JR., BORDEN, J.H., and OEHLSCHLAGER,A.C. 1978. Host selection by Hylemya antiqua: Identification of oviposition stimulants based on proposed active thioalkane moieties. Environ. Entomol. 7:728-731. VERNON, R.S., JUDD, G.J.R., BORDEN, J.H., PIERCE, H.D., Jr., and OEHLSCHLAGF~, A.C. 1981. Attraction of Hylemya antiqua (Meigen) (Diptera: Anthomyiidae) in the field to host-produced oviposition stimulants and their nonhost analogues. Can. J. Zool. 59:872-881. WESTON, P.A., and MILLER, J.R. 1985. Influence of cage design on precision of tube-trap bioassay for attractants of the onion fly, Delia antiqua. J. Chem. Ecol. 11:435-440. ZECHMAN, J.M., and LABOWS, J.N., JR. 1984. Volatiles of Pseudomonas aeruginosa and related species by automated headspaee concentration-gas chromatography. Can. J. Microbiol. 31:232-237.
P R O D U C T I O N OF ONION FLY 1 A T T R A C T A N T S A N D OVIPOSITIONAL STIMULANTS BY BACTERIAL ISOLATES C U L T U R E D ON ONION 2
S.M. H A U S M A N N 3 and J.R. MILLER Department of Entomology and Pesticide Research Center Michigan State University East Lansing, Michigan 48824 (Received August 25, 1986; accepted March 25, 1988)
Abstract--Decomposing onions at certain microbial successional stages produce potent volatile attractants and ovipositional stimulants of the onion fly, Delia antiqua (Diptera: Anthomyiidae). A reproducible source of these compounds was obtained by culturing Erwinia carotovora var. carotovora (EC) on sterile onion tissue. In laboratory choice tests, EC-inoculated onion was more attractive than Klebsiella pneumoniae (KP) cultured on onion, EC cultured on potato (a nonhost of onion fly), or the chemical synthetic baits dipropyl disulfide and an aqueous solution of 2-phenylethanol and pentanoic acid. Onion flies were mildly attracted to potato after inoculation with EC, but females did not accept EC-inoculated potato for oviposition. This work emphasizes that sources of semiochemicals may need to be defined microbiologically as well as physically and chemically. Key Words--Onion fly, Delia antiqua, Diptera, Anthomyiidae, Erwinia carotovora var. carotovora, Klebsiella pneumoniae, food attractant, ovipositional stimulant, dipropyl disulfide, 2-phenylethanol, pentanoic acid.
INTRODUCTION
Onion fly, Delia antiqua (Meigen) (Diptera: Anthomyiidae), is a major pest of commercial onions in the temperate northern hemisphere. Decomposing onions produce more potent onion fly attractants and ovipositional stimulants than do
1Diptera: Anthomyiidae. 2paper No. 12106 of the Michigan State University Agricultural Experiment Station. 3Current address: Building E402, DuPont Experiment Station, Wilmington, Delaware 19898.
905 0098-0331/89/0300-0905506.00/0 9 1989PlenumPublishingCorporation
906
HAUSMANN AND MILLER
healthy or freshly cut onions (Dindonis and Miller, 1981a; Ishikawa et al., 1981), but only at certain successional stages of microbial and physical damage (Vernon et al., 1981; Miller et al., 1984). Using gas-liquid chromatography (GLC), Vernon et al. (1981) demonstrated that changes in onion fly attractancy were associated with qualitative and quantitative changes in the headspace chemical profile of onion as decomposition progressed. Many factors influence volatile metabolite production in decomposing plant material. A complex interaction of environmental factors such as temperature, oxygen concentration, characteristics of the host plant, and presence of competing microorganisms and their metabolites may influence which organisms and volatile metabolites predominate. Under controlled growth conditions, microorganisms give reproducible headspace profiles that are often unique to a given genus, and, sometimes, species (Larsson et al., 1984). Indeed, using GLC for the qualitative and quantitative determination of volatile compounds has become an accepted technique in identifying anaerobic bacteria (Zechman and Labows, 1984) and some groups of aerobic bacteria (e.g., Lee et al, 1979). Bacteria infecting urine, meat, and other substrates are analyzed by direct headspace analysis of the sample or by subculturing the bacteria onto media that are selective or enriched in precursors to targeted metabolites (Larsson et al., 1984). Pure cultures may produce a distinctive series of volatile organics or a distinctive odor such as the grapelike 2-aminoacetophenone, a diagnostic marker for Pseudomonas aeruginosa (Cox and Parker, 1979). The variable attractancy of decomposing onion has hindered work aimed at identifying potent onion fly attractants and ovipositional stimulants. This paper describes how a reproducible source of these compounds has been obtained by culturing bacterial isolates associated with onion decomposition on sterilized onion tissue.
METHODS AND MATERIALS
Insects. Parental D. antiqua stock was collected from commercial onion fields in Eaton Rapids, Michigan. Insects used were 5 to 7 generations removed from the field. Adults were fed water and a dry artificial diet (Ticheler, 1971) and housed as described by Schneider et al. (1983), except onion foliage in ovipositional dishes was replaced by green surrogate stems (Harris and Miller, 1983). Larvae were reared in sand-filled boxes provisioned with an excess of bisectioned onion bulbs. All behavioral bioassays were conducted at 35 _+ 5 % relative humidity, 23 • I~ and in a 16 : 8 light-dark regime; alternating cool 55-W and warm 85-W white fluorescent bulbs were placed 10 cm above the cages. Three days before assaying, flies from the laboratory culture were placed in cylindrical,
B A C T E R I A L I S O L A T E S ON ONI ON
907
metal-screened cages (42 cm diam. x 46 cm). Between assays, flies were given food, water, and access to ovipositional dishes of freshly cut onion covered with moist sand. In the center of each dish we placed a 4-mm-diam. x 9-cmlong surrogate onion stem. Before each assay, the water dish was placed in the center of the cage, but food and ovipositional dishes were removed. Bacterial Cultures. Erwinia carotovora var. carotovora (EC) causes soft rot on a wide range of vegetables. Onion maggots moving through infested soil may spread the bacterium (Agrios, 1978). EC was isolated from damaged stored onions (Spartan Banner hybrid) potted in moist muck soil taken from a commercial onion field in East Lansing, Michigan. Damage was inflicted by 20 second-instar D. antiqua larvae taken from our laboratory culture and added to each potted bulb. After five days of larval feeding, 10 g of the decomposing onion tissue was homogenized with a sterile mortar and pestle in 25 ml sterile buffer (0.2 M KH2PO4 and 0.2 M K2HPO4, pH 7.0). For the isolation of softrot bacteria, dilutions of the homogenate were made in buffer and plated on a selective medium containing crystal violet and sodium polypectate (CVP) (Cuppels and Kelman, 1974). Plates were incubated overnight at 25 + 2~ Colonies of Erwinia spp. formed deep, cuplike depressions in the selective medium. They could be distinguished from pectolytic Pseudomonas spp., which form shallow, wider depressions (Cuppels and Kelman, 1974). A bacterial isolate of Erwinia sp. was obtained and designated as strain EC6. Isolate EC6 was identified using current taxonomic keys and texts (Lelliott, 1974; Kelman and Dickey, 1980). Klebsiella pneurnoniae (KP) had been stored at - 6 5 ~ in 4 % (v/v) DMSO in nutrient broth (0.3 % beef extract and 0.5 % peptone; Difco). The original strain, JM 1, was isolated and identified as a predominant microorganism from decomposing onions that were five times more attractive in the field than fresh cut onions (Miller et al., 1984). For use in experiments, colonies of both EC and KP were grown on Bacto nutrient agar (Difco) at 25 _+ 2~ Stock cultures were maintained on nutrient agar in the refrigerator at 4-6~ Preparation of Axenic Onion and Potato. Axenic plant tissues were prepared from Russet Burbank potatoes and an undetermined cultivar of yellow onions from commercial storage. The outer, papery scales and upper and lower 2 cm of bulb were removed; potatoes were peeled. Plant tissues were soaked in 75% ethanol for 20 min and flame sterilized 1-2 min. The flamed outer 0.5 cm of potato and the flamed outer two onion scales were removed with sterilized instruments and discarded. Two sizes of potato and onion tissue were cut. Squares of tissue measuring 3.5 • 3.5 x 0.5 cm and 2 x 2 • 0.5 cm were placed individually in polystyrene Petri dishes (diameters of 5.5 cm and 3.5 cm, respectively). Onion squares were cut only from the 2 to 3 outer scales. After sterile distilled water was
908
HAUSMANN AND MILLER
added to cover the dish bottom, tissues were incubated at 25 _+ 2~ Because the concentration of sulfur-and nitrogen-containing compounds in onion are known to differ in the outer and inner scales (Freeman, 1975), only tissue taken from the same onion scale (or the same potato) was used within the same cage for bioassays. Experiment 1: Attractancy of Axenic Onion. Approximately 100-125 flies of mixed sexes and ages were placed in each cylindrical cage. Wire screen baskets were hung in the tube traps of Weston and Miller (1985); all parts were autoclaved. Between 4 and 15 hr into photophase, treatments were enclosed individually in the wire baskets and assayed in choice tests by spacing traps equally along the periphery of cylindrical cages. After the bioassay, flies in each trap were sexed and counted. In each of four cages, traps were baited with either a 3.5 x 3.5 x 0.5-cm axenic onion square cut immediately before assaying, cut axenic onion held for one, two, and three days before assaying, or a petri dish filled with sterilized, distilled water. At the time of cutting, two parallel grooves ca. 1 cm apart and 2.5 cm deep were cut into each square. Onion tissues were assayed in their original water-filled polystyrene petri dishes with the covers removed.
Experiment 2: Time course of Attractancy of Bacteria Cultured on Onion. To determine the minimal time needed to incubate soft rot bacteria on onion to obtain significant differences in onion fly attractancy, traps were baited with axenic onion inoculated with a bacterial isolate and incubated one, two, or three days before assaying. These treatments were compared to freshly cut axenic onion and a water control. To inoculate onion with bacteria, two parallel grooves ca. 1 cm apart and 2.5 mm deep were cut into each square. A loopful of 24-hr inoculum was placed in each groove. Tissues were incubated at 25 _+ 2~ Experiments using EC and using KP were replicated seven times as described in experiment 1. Experiment 3: Relative Attractancy of EC- and KP- inoculated Onion. To determine the relative attractancy of the two isolates cultured on onion, traps were baited with onion inoculated with EC, KP, or EC plus KP and incubated for three days at 25 +_ 2~ before assaying along with freshly cut axenic onion. Onion was inoculated as described in experiment 2 with EC or KP inoculum placed in both grooves of a square or with EC inoculum in one groove and KP inoculum in the other~ Eight replications were conducted as described in experiment 1. Experiment 4: Attractancy of EC-Inoculated Host and Nonhost Plant. The volatile onion fly attractants produced by EC-inoculated onion may have been due to EC, regardless of the nutritive medium used, or to the interaction of EC with onion, the primary host plant of the onion fly. To determine the importance of the host plant in the production of attractants, EC was cultured on axenic onion and axenic potato, a nonhost plant of onion fly, as described in experi-
B A C T E R I A L I S O L A T E S ON ONI ON
909
ment 2. Traps were baited with EC-inoculated onion, inoculated potato, axenic onion, axenic potato, and water control. Axenic onion was freshly cut; other treatments were incubated three days before assaying. This experiment was replicated seven times as described in experiment 1. Experiment 5: Attractancy of EC-Inoculated Onion vs. Chemical Synthetics. The attractancy of onion inoculated with EC and incubated three days before assaying was compared to synthetic chemical attractants. One chemical attractant was a solution of 100 txl 2-phenylethanol (2-PhEt) and 25/zl pentanoic acid mixed in 50 ml distilled water. These chemicals were reported by Ishikawa et al. (1984) as D. antiqua attractants extracted from decomposing onions. This formulation was reported to give an optimal release rate for onion fly attractancy in the field (Ishikawa et al., 1984). Ten-milliliter glass beakers containing 10 ml of this aqueous solution were placed in traps immediately before testing. A second chemical attractant was n-dipropyl disulfide (Pr2S2), a predominant volatile secondary metabolite from fresh onions (Block, 1985). For an optimal release rate (ca. 100/~g/hr), 100/~1 Pr2S2 (purity 99%) was placed into size 3 BEEM (Ted Pella Co., Box 510, Tustin, California 92680) polyethylene enclosures (Dindonis and Miller, 1981b). One capsule per trap was used. To attain a stable release rate, filled capsules were held for 6 hr prior to assaying. The experiment was replicated four times as described in experiment 1. Experiment 6: Oviposition on EC-Inoculated Host and Nonhost Plant. Ovipositional dishes (35 ml) were filled with sterilized sand moistened with sterilized, distilled water. A 4-mm-diam x 9-cm-long surrogate onion stem stood vertically in the center of each dish. By spacing ovipositional dishes equally along the cage periphery, five different test materials were assayed simultaneously in cages containing six to eight gravid D. antiqua females. Treatments were assayed 8-14 hr into photophase, thus spanning the peak diurnal ovipositional period of D. antiqua (Havukkala and Miller, 1987). In one ovipositional experiment, the 2.0 x 2.0 x 0.5-cm squares of plant tissue, along with associated water, were placed ca. 0.5 cm beneath the sand surface immediately before assaying. No plant material was placed in the cup designated as a water control. In a second experiment, to determine whether volatiles alone from test materials stimulated oviposition, Petri dishes containing the 3.5 • 3.5 x 0.5-cm squares of tissue and water were placed in sterilized glass quart canning jars. A GLC septum was inserted into each of two holes drilled into the jar lid. Teflon tubing (1 mm ID) was inserted into each septum. One tube passed through the screened cage wall and was threaded through a septum inserted into a hole in the side of the ovipositional cup. The tubing opened ca. 5 mm from the surrogate stem and ca. 1.25 cm below the sand surface. The second tube led from the jar lid to a small plastic manifold and then to the filter of a compressed air tank. Air carrying volatiles from the jars to the ovipositional cups flowed at 10 ml/min.
910
HAUSMANN AND MILLER
In both ovipositional experiments, the treatments used were onion and potato inoculated with EC and incubated for three days before assaying, axenic potato held three days before assaying, freshly cut axenic onion, and a water control. Preparation of treatments followed the procedure outlined in experiments 1 and 2. Both experiments were replicated four times. RESULTS
Identification oflsolate EC6. Bacterial cells of EC6 were motile, straight rods occurring primarily singly, but sometimes in pairs or short chains. This isolate was gramnegative, facultatively anaerobic, oxidase negative, phosphatase negative, reduced nitrate, did not produce reducing substances from sucrose, and utilized Simmon's sodium citrate. Growth occurred at 36~ and in 5% NaC1. No special growth factors were required. Colonies on yeast extract-dextrose-CaCO2 agar (YDC) were creamy white with no diffusible pink or blue pigments; colonies on nutrient agar were transluscent white with no yellow pigments. Inoculation on sterile potato disks caused soft rot. Based on these results and the characteristic growth on CVP medium, EC6 was identified as a strain of Erwinia carotovora var. carotovora. Experiment 1: Attractancy of Axenic Onion. Sterilized onion tissues incubated one day or more before the assay lost activity (Figure 1); therefore, sterilized onions used as controls in the remaining bioassays were cut immediately before assaying.
WI3
I
I
I
I
+1 11 0
9
0
7 5
e- 3 i
0 Days
i
1 sterile
i
,
2 onion
i
3 held
FIG. 1. Onion fly attractancy as influenced by time sterilized onion tissue was held before assaying. Orthogonal contrasts after ANOVA: freshly cut onion vs. onion incubated 1, 2, and 3 days (P < 0.05); freshly cut onion vs. water control (P < 0.05); and water control vs. onion incubated 1, 2, and 3 days (P > 0.20).
911
B A C T E R I A L I S O L A T E S ON ONI ON
Experiment 2: Time Course of Attractancy of Bacteria Cultured on Onion. Attractancy of EC-inoculated onion increased with incubation times from one to three days (Figure 2A). Onion inoculated and incubated two or three days caught significantly more flies than either noninoculated onion or the water control. A similiar pattern of increasing attractancy with incubation times from one to three days was seen for KP-inoculated onion tissue (Figure 2B). A three-day incubation period was judged sufficient to obtain significant differences in attractancy between bacterially inoculated onion and sterile onion.
Experiment 3: Relative Attractancy of EC- and KP-Inoculated Onion. Traps baited with onion inoculated with EC caught significantly more flies than those baited with KP-inoculated onion (Figure 3). Onion tissue inoculated with both bacteria did not catch significantly more flies than onion inoculated with
W
A
~2o~
w a t e r @ @ @ @ 1 day 2 day 3 day IaJ
CO +1
25
.5 5
ware,@ @ @ @ 1 day 2day 3day F]o. 2. Influence of incubation time on onion fly attractancy of onion inoculated with bacterial isolates. Sterilized onion tissue ( 9 inoculated with EC (A) or with KP (B) and incubated 1, 2, or 3, days before the bioassay. Treatments marked by different letters are significantly different (Tukey's HSD all-pairwise comparisons after ANOVA; P < 0.05 forA; P < 0.10 for B).
912
HAUSMANN AND MILLER
•W)40 +1
e- 8o o t~ o e'~20 t~ .i,..a
O@ 9174 FIG. 3. Relative onion fly attractancy of EC and/or KP inoculated on sterilized onion tissue (O) and incubated 3 days before the bioassay. Treatments marked by different letters are significantly different (Tukey's HSD all-pairwise comparisons after ANOVA; P < 0.05).
either bacterial isolate alone (P > 0.20). Regardless of the bacterial isolate used, all inoculated onion baits caught significantly more flies than traps baited with noninoculated onion. The remaining experiments focused on the bacterial isolate EC.
Experiment 4: Attractancy of EC-InocuIated Host and Nonhost Plant. A higher mean trap catch was attained when onion rather than potato was inoculated with EC (Figure 4). Traps baited with EC-inoculated potato did catch significantly more flies than those baited with noninoculated potato.
Experiment 5: Attractancy of EC-Inocalated Onion vs. Chemical Synthetics. Traps baited with EC-inoculated onion caught significantly more flies than any other bait used (Figure 5). The mean fly catches obtained with traps baited with either chemical attractant or with noninoculated onion were all low and did not differ significantly from each other (P > 0.20). All baits tested caught significantly more flies than the water control, except Pr2S 2 (P > 0.20). Experiment 6: Oviposition on EC-Inoculated Host and Nonhost Plant. When test materials were placed directly in the ovipositional dishes, EC-inoculated onion received significantly more eggs than any other treatment (Table 1, experiment A). Freshly cut, sterilized onion received fewer eggs than ECinoculated onion but more eggs than all other treatments. No other significant differences were found (P > 0.20).
913
BACTERIAL ISOLATES ON ONION
1.1.145 ~-(2
09
cr
+1 t'-35 .i..+ O
r
t~
e-15 t~ I1) 5
water r----~ 0
[~-] @
FIG. 4. Production of onion fly attractants by the interaction of EC on onion fly host and nonhost tissue. Sterilized onion tissue (O) and potato tissue (D) inoculated with EC and incubated 3 days before the bioassay. Statistically significant interaction between plant tissue used and presence or absence of EC [two-way ANOVA of data transformed to (X +_ 0.5)1/2; p < 0.01)]. Orthogonal contrasts: potato vs. EC-inoculated potato (P < 0.01); onion vs. EC-inoeulated onion (P < 0.01); EC-inoculated potato vs. ECinoculated onion (P < 0.01); potato vs. onion (P > 0.20). When volatiles from these treatments were pumped into the dishes, ECinoculated onion received significantly more eggs than any other treatment group (Table 1 experiment B). No other differences were found among the remaining treatments ( P > 0.20), all of which received very few eggs. l.U 09 +1 o
25 20
b c
water
b
bc
.,.Am Pr2S 2
2-PhET
,
pentanoic acid
O@
Fro. 5. Comparison of chemical synthetics and EC-inoculated onion as onion fly attractants. Symbols and statistics as in Figure 3.
914
HAUSMANN AND MILLER
TABLE l. MEAN NUMBER OF EGGS LAIDa BY D.
antiqua IN RESPONSE TO E C
CULTURED
ON HOST AND NONHOST PLANT TISSUE
Experimentb Treatment
A
B
Ec-inoeulated onion Noninoculated onion EC-inoculated potato Noninoculated potato Water control
73.5a + 23.87 12.5b + 2.96 1.0c + 1.00 0.0c 0.0c
47.5a + 10.54 1.0b + 0.56 1.0b + 0.41 0.0b 0.0b
a Means + SE followed by the same letter within a column are not significantly different [one-way ANOVA of data transformed to (X + 0.5) x/2followed by Tukey's all-pairwise comparisons (Gill, 1978); P < 0.05]. /'Treatments presented in ovipositional dishes (experiment A) or as volatiles pumped into ovipositional dishes (experiment B).
DISCUSSION
The unstable sulfenic acids released from cut onion degrade to form various sulfur-containing volatiles that, in large part, give onion its characteristic odor and taste (Block, 1985). A number of these volatiles, (e.g., dipropyl disulfide) are known to attract and/or stimulate oviposition by the onion fly (Matsumoto and Thorsteinson, 1968; Vernon et al., 1978). As the volatiles from damaged onion cells are exhausted, the attractancy of freshly cut sterilized onion would be expected to decrease; hence, freshly cut sterilized onion was chosen as an appropriate control for assays testing the effects of culturing bacteria on sterilized onion tissue. The interaction of EC and of KP with onion tissue increased D. antiqua trap catches (Figure 2) and oviposition (Table 1). Behavioral activity in response to EC- or KP-inoculated onion increased with time (Figure 2), probably due to a rise in volatile release. As the bacterial population increases, a concommitant increase in amounts of pectolytic and cellulolytic enzymes would be expected (Start and Chatterjee, 1972). Qualitative changes in volatiles also would be expected as the chemical composition of the host changes and bacteria exhaust available nutrients. Human olfaction detects little difference between sterilized and EC-inoculated potato tissue. Indeed, many vegetables infected with E. carotovora vat. carotovora alone do not release the unpleasant odors commonly associated with decomposing plants. Saprophytic bacteria that colonize plant tissue after pathogenic infection generally are responsible for the foul odor. Two notable exceptions are crucifers and alliums, both of which release unpleasant sulfurous odors when inoculated with the soft-rot pathogen (Agrios, 1978). EC-inoculated potato
BACTERIAL ISOLATES ON ONION
915
may release simple volatile metabolites not particularly odoriferot~s but nevertheless mildly attractive to onion fly, e.g., short-chain alcohols and hydrocarbons (Stotzky and Schenck, 1976). The aqueous solution o f 2-PhEt and pentanoic acid reported by Ishikawa et al. (1984) to be a potent attractant o f D. antiqua was much less attractive than EC-inoculated onion tissue (Figure 5), suggesting that attractants associated with decomposing onions remain to be identified. Indeed, the low fly catches obtained with 2-PhEt/pentanoic acid as well as with PrzS2 demonstrate that currently known synthetic attractants do not closely simulate fresh or maximally attractive decomposing onion. Perhaps decomposing onions used as positive controls in the studies emphasizing the potency of 2-PhEt/pentanoic acid as attractants for D. antiqua were inadequate (Ishikawa et al., 1984). Even though they were significantly attracted, female onion flies did not accept EC-inoculated potato as an ovipositional site (Table 1). F o r D. antiqua, oviposition is a specific behavioral response restricted to stimuli from AlIium cepa and its closest relatives. On the other hand, attractancy can include both host-finding and food-finding. Thus, onion flies may be attracted to and feed on decomposing nonhost plants (Miller et al., 1984) but may not accept them for oviposition. The most notable points made here are: (1) that a source o f consistent highly active D. antiqua attractants has been found (this advance should greatly facilitate chemical identification of these attractants), and (2) sources o f behavior-modifying stimuli may need to be defined microbiologically as well as physically and chemically. Acknowledgments--We thank Joan Harlin for care of onion fly cultures. This work was supported by USDA Competitive Research Grant 83-CRCR-l-1204 to James R. Miller and by the Michigan Onion Committee.
REFERENCES AGRIOS,G.N. 1978. Bacterial soft rots, pp. 477-483, in Plant Pathology. Academic Press, New York. BLOCK,E. 1985. The chemistry of garlic and onion. Sci. Am. 252:114-119. Cox, C.D., and PARKER,J. 1979. Use of 2-aminoacetophenone production in identification of Pseudomonas aeruginosa. J. Clin. Microbiol. 9:479-484. CUt'PELS,D., and KELMAN,A. 1974. Evaluation of selective media for isolation of soft-rot bacteria from soil and plant tissue. Phytopathology 64:468-475. DINDONIS, L.L., and MILLER,J.R. 1981a Onion fly and little house fly host finding selectively mediated by decomposing onion and microbial volatiles. J. Chem. Ecol. 7:419-426. DINDON~S, L.L., and MILLER,J.R. 1981b. Onion fly trap catch as affected by release rates of n-dipropyl disulfide from polyethylene enclosures. J. Chem. Ecol. 7:411-418. FREEMAN,G.G. 1975. Distribution of flavour components in onion (Allium cepa L.), leek (Allium porrum ) and garlic ( Allium sativum ). J. Sci. Food Agric. 26:471-481.
9t6
HAUSMANN AND MILLER
GILL, J.L. 1978. Design and Analysis of Experiments in the Animal and Medical Sciences, Vol. 1. Iowa State University Press, Ames, Iowa. HARRIS, M.O., and MILLER, J.R, 1983. Color stimuli and oviposition behavior of the onion fly, Delia antiqua (Meigen) (Diptera: Anthomyiidae). Ann. Entomol, Soc. 76:766-771. HAVUKKALA,lJ., and MILLER, J.R. 1987. Daily periodicity in the ovipositional behavior of the onion fly, Della antiqua (Diptera: Anthomyiidae). Environ. Entomol. 16:41-44. ISHIKAWA, Y., IKESHOJ1,T., and MATSUMOTO,Y. 1981. Field trapping of the onion and seed-corn flies with baits of fresh and aged onion pulp. Appl. Entomol. Zool. 16:490-493. ISHIKAWA,Y., MATSUMOTO,Y., TSUTSUMI,M., and MITSUI, Y. 1984. Mixture of 2-phenylethanol and n-valerie acid, a new attractant for the onion and seedcom flies. Hylemya antiqua and H. platura (Diptera: Anthomyiidae). Appl. Entomol Zool. 19:448-455. KELMAN, A., and DICKEY, R,S. 1980 Erwinia. 2. Soft rot or carotovora group, pp. 31-35, in N.W. Schaad (ed.). Laboratory Guide for Identification of Plant Pathogenic Bacteria. American Phytopathology Society, St. Paul, Minnesota. LARSSON, L., MARDH, P.A., and ODHAM,G. 1984. Analysis of volatile metabolites in identification of microbes and diagnosis of infectious diseases, pp. 207-235, in G. Odham, L. Larsson, and P.-A. Mardh (eds.). Gas Chromatography/Mass Spectrometry, Applications in Microbioiogy. Plenum Press, New York. LEE, M:L., SMITH, D.L,, and FREEMAN, L.R. 1979. High resolution gas chromatographic profiles of volatile organic compounds produced by microorganisms at refrigerated temperatures. Appl. Environ. Microbiol. 37:86-90. LELLIOT% R.A. 1974. Description of the species of the genus Ervvinia, pp. 469-476, in R.F. Buchanan and N.E. Gibbons (eds.). Bergey's Manual of Determinative Bacteriology, 8th Ed. Witliams & Wilkins, Baltimore, Maryland. MATSUMOTO,Y., and THORSTEINSON,A.J. 1968. Effect of organic sulfur compounds on oviposition in onion maggot, Hylemya antiqua Meigen (Diptera: Anthomyiidae). Appl. Entomol. Zool. 3:5-12. MILLER, J.R., HARRIS, M.O., and BREZNAK,J.A. 1984. Search for potent attractants of onion flies. J. Chem. Ecol. 10:1477-1488. SCtfNEIDER, W.D., MILLER, J.R., BREZNAK, J.A., and FOBES, J.F. 1983. Onion maggot, Delia antiqua, survival and development on onions in the presence and absence of microorganisms. EntomoL Exp. AppL 33:50-56. STARR, M.P., and CHATTERJEE,A.K. 1972. The genus Erwinia: Enterobacteria pathogenic to plants and animals. Annu. Rev. Microbiol. 26:389-426. STOTZKY, G., and SCHENCK, S. 1976. Volatile organic compounds and microorganisms. C.R.C. Crit. Rev. Microbiol. 333. TICHELER, J. 1971. Rearing of the onion fly, Hylemya antiqua Meigen), with view to release sterilized insects, pp. 341-346. in Sterility Principle for Insect Control or Eradication. Proceedings of a Symposium, Athens, 1970. IAEA, Vienna. VERNON, R.S., PmRCE, H.D., JR., BORDEN, J.H., and OEHLSCHLAGER,A.C. 1978. Host selection by Hylemya antiqua: Identification of oviposition stimulants based on proposed active thioalkane moieties. Environ. Entomol. 7:728-731. VERNON, R.S., JUDD, G.J.R., BORDEN, J.H., PIERCE, H.D., Jr., and OEHLSCHLAGF~, A.C. 1981. Attraction of Hylemya antiqua (Meigen) (Diptera: Anthomyiidae) in the field to host-produced oviposition stimulants and their nonhost analogues. Can. J. Zool. 59:872-881. WESTON, P.A., and MILLER, J.R. 1985. Influence of cage design on precision of tube-trap bioassay for attractants of the onion fly, Delia antiqua. J. Chem. Ecol. 11:435-440. ZECHMAN, J.M., and LABOWS, J.N., JR. 1984. Volatiles of Pseudomonas aeruginosa and related species by automated headspaee concentration-gas chromatography. Can. J. Microbiol. 31:232-237.
