CHEMICAL ECOLOGY Electroantennogram and Behavioral Responses of suspensa (Diptera: ) to Putrescine and Ammonium Bicarbonate Lures

1 PAUL E. KENDRA, WAYNE S. MONTGOMERY, NANCY D. EPSKY, AND ROBERT R. HEATH

USDAÐARS, Subtropical Horticulture Research Station, 13601 Old Cutler Rd., Miami, FL 33158

Environ. Entomol. 38(4): 1259Ð1266 (2009) ABSTRACT At present, the most effective synthetic lures for pest Anastrepha ßies are mul- ticomponent blends that include and the diamine synergist putrescine (1,4-diaminobutane). Both chemicals generally have been regarded as protein cues that result in female-biased attraction. Using electroantennography (EAG) and ßight tunnel bioassays, this study evaluated response of the Caribbean fruit ßy, Anastrepha suspensa (Loew) to vapors released from commercial lure formulations of ammonium bicarbonate and putrescine. Over a range of doses tested, EAG response to ammonium bicarbonate was equivalent for both , but female response was signiÞcantly greater than male response to putrescine and to a 1:1 mixture of ammonium bicarbonate and putrescine. Amplitude of EAG response to the mixture was approximately equal to the summation of responses to the individual substrates. Using a Þxed dose of substrate, EAG measurements from females 1Ð14 d old indicated that antennal sensitivity to both lures varied according to physiological state of the ßy. Peak response to ammonium bicarbonate was recorded from immature females, peak response to putrescine from sexually mature females. In bioassays, more females were captured with ammonium bicarbonate plus putrescine than with ammonium bicarbonate alone. This difference was not observed in males, resulting in a higher female to male ratio in captures with ammonium bicarbonate plus putreseine (3:1) versus ammonium bicarbonate alone (1:1). Results suggest that separate olfactory receptors are involved in detection of the two semiochemicals, and that the putrescine component is primarily responsible for the female-biased attraction.

KEY WORDS Caribbean fruit ßy, synthetic attractants, 1,4-diaminobutane, olfaction, semiochemicals

The New World genus Anastrepha includes a number cent years (Thomas 2004) and its preferred hosts in- of invasive fruit ßies that are serious pests of tropical clude (Citrus x paradisi Macfad), one of and subtropical fruit crops (Aluja 1994). Of the 198 FloridaÕs major agricultural exports (Weems et al. described species of Anastrepha (Norrbom 2000), only 2004). Other species of concern include A. obliqua 1 species of economic importance has become estab- (Macquart), A. fraterculus (Wiedemann), A. striata lished in FloridaÑthe Caribbean fruit ßy, A. suspensa (Schiner), and A. grandis (Macquart) (White and (Loew). Indigenous to the West Indies, A. suspensa is Elson-Harris 1992). now common in south , where it is a quarantine Because of the economic impact of Anastrepha pest of citrus and a production pest of other , pests, much attention has been focused on develop- most notably common (Psidium guajava L.) ment of trapping systems for monitoring their popu- (Greany and Riherd 1993). Several other Anastrepha lations. Surveillance programs traditionally have used species pose a threat to Florida because of McPhail traps baited with liquid protein, speciÞcally close proximity of populations in Mexico and the Ca- hydrolyzed buffered with (Lopez et ribbean basin. In addition, the increasing volume of al. 1971, Steyskal 1977, Heath et al. 1993a). IdentiÞ- foreign produce shipments entering the stateÕs ports cation of ammonia vapor as the key fruit ßy attractant creates a potential pathway for pest entry and spread emanating from proteinaceous baits (Bateman and (Kendra et al. 2007 and references therein). The Mex- Morton 1981, Mazor et al. 1987) has led to develop- ican fruit ßy, A. ludens (Loew), represents a particular ment of ammonia-based synthetic lures. Subsequent threat because it has been detected in Florida in re- research found that amines such as putrescine, meth- ylamine, and trimethylamine enhance the efÞcacy of traps baited with synthetic ammonia, and current This article reports the results of research only. Mention of a Anastrepha monitoring programs use multicomponent proprietary product does not constitute an endorsement or recom- mendation by the USDA. lures, including combinations of ammonium acetate 1 Corresponding author, e-mail: [email protected]. and putrescine (Heath et al. 1995, Thomas et al. 2001); 1260 ENVIRONMENTAL ENTOMOLOGY Vol. 38, no. 4 ammonium bicarbonate, methylamine, and putrescine females were dissected to assess maturity status ac- (Robacker and WarÞeld 1993); and ammonium carbon- cording to a six-stage classiÞcation system developed ate, methylamine, and putrescine (Robacker 1999). The previously (Kendra et al. 2006). With this system, commercial two-component lure containing ammonium stages 1Ð4 comprise sequential steps in the ovarian acetate and putrescine (BioLure; Suterra, Bend, OR) has maturation process, stage 5 denotes presence of ma- been shown to be equal to or better than aqueous ture oocytes, and stage 6 indicates the ovipositional torula yeast/borax for capture of A. suspensa in Florida phase. For sexually mature females (stages 5 and 6), (Thomas et al. 2001, Hall et al. 2005, Holler et al. 2006) ovaries were teased apart to count the number of but less effective in tests conducted in mature oocytes per ovary ( load). (Pingel et al. 2006). Capture of other Anastrepha spe- Sample Preparation. Two commercially available cies with synthetic lures versus liquid protein baits has lures were used as test substrates, ammonium bicar- been variable as well (Epsky et al. 2004, Thomas et al. bonate, NH4HCO3 (AgriSense-BCS, Mid Glamorgan, 2008). United Kingdom), and putrescine, NH2(CH2)4NH2, As part of an ongoing effort to develop improved (Suterra). Lures were placed individually and in com- synthetic lures for pest Anastrepha fruit ßies, we bination (one of each type) into 250-ml gas-tight glass initiated a research program to address tephritid bottles equipped with lids that had been Þtted with olfactory ecology, integrating electroantennogra- short thru-hull ports (Swagelok, Solon, OH) and sil- phy (EAG) and developmental with be- icone septa (Alltech, DeerÞeld, IL). Sample bottles havioral response to known olfactory attractants. Our were sealed and equilibrated for 24 h at room tem- goal is to identify the principal factors inßuencing perature (Ϸ24ЊC) to allow for headspace saturation attraction to chemical cues and develop improved with volatiles from the lure(s). 2-Butanone (99% pure; female-targeted trapping systems compatible with Sigma-Aldrich, St. Louis, MO) was used as a reference sterile male release programs. In a previous report standard in all tests. This compound, a naturally oc- (Kendra et al. 2005b), we quantiÞed antennal re- curring fruit volatile, has been shown to elicit consis- sponse of male and female A. suspensa to ammonia and tently strong EAG responses in both sexes of A. sus-

CO2, the volatile compounds released from ammo- pensa, and its purity has been conÞrmed by gas nium bicarbonate lures. For females, we further doc- chromatography (Kendra et al. 2005b). umented relationships between stages of ovarian de- Electrophysiology. Olfactory responses were re- velopment, EAG responses, and attraction to ammonia corded with Syntech equipment and EAG 2000 soft- and CO2 (Kendra et al. 2005a, 2006). In this study, we ware (Hilversum, The Netherlands), using published examine response of A. suspensa to putrescine, the methods (Kendra et al. 2005a, b). In brief, freshly second attractant of the two-component lure. Recent dissected antennal preparations were mounted be- Þeld trials that deployed ammonium acetate lures in tween micropipette electrodes with salt-free conduc- combination with a series of Þve homologous terminal tive gel (Spectra 360; Parker Laboratories, FairÞeld, diamines (including putrescine) showed that captures NJ) and placed under a stream of puriÞed air. Using of A. suspensa were heavily biased toward females gas-tight syringes (VICI Precision Sampling, Baton (Kendra et al. 2008). Initial EAG analyses with these Rouge, LA), saturated vapor samples were withdrawn substrates suggested that the putrescine (or compa- from the test bottles, injected into the airstream, and rable diamine) component was responsible for the presented to the antennae. There was a 2-min interval female-biased attraction. To test that hypothesis, this (clean air ßush) between sample injections to prevent study further evaluates antennal response of male and antennal adaptation (diminished EAG response as a female A. suspensa to putrescine, presented alone and result of repeated exposure to speciÞc olfactory stim- in combination with an ammonium bicarbonate lure. uli). EAG responses to test substrates were measured We also determine the effect of female reproductive initially in millivolts (peak height of depolarization) state on antennal sensitivity to the two lures and con- and converted to a percentage relative to the standard duct ßight tunnel bioassays to evaluate behavioral response (20 ␮l 2-butanone saturated vapor). Nor- response of both sexes to the lures. malization with this standard chemical corrected for time-dependent variability in antennal performance, and all statistical analyses were performed using nor- Materials and Methods malized EAG results. . Anastrepha suspensa were obtained from a For mature male and female ßies (9Ð12 d after laboratory colony maintained at the USDAÐARS, Sub- eclosion), volumetric doses ranging from 0.25 to 4.0 ml tropical Horticulture Research Station, Miami, FL. (a series of Þve two-fold increases in headspace vol- Rearing conditions consisted of a 12:12-h (L:D) pho- ume) were used to measure response to putrescine, toperiod, 25 Ϯ 2ЊC, and 75 Ϯ 5% RH. All test insects ammonium bicarbonate, and an equal concentration were of known age, staged at 1-d intervals using es- of the two. For each and substrate, EAG responses tablished protocols (Kendra et al. 2006). Adult ßies were recorded from 10 insects, and responses from were housed in mixed-sex cages (Ϸ1:1 sex ratio) and each were the average of three replicate mea- provisioned with water (from agar blocks) and food surements. To assess effect of female age (sexual ma- (reÞned cane sugar and yeast hydrolysate, 4:1 mix- turity status) on EAG response to putrescine and am- ture) ad libitum before collection for EAG analysis. monium bicarbonate, Þxed 2-ml doses of each After completion of electrophysiological recordings, substrate were presented to ßies 1Ð14 d after eclosion. August 2009 KENDRA ET AL.: A. suspensa RESPONSE TO SYNTHETIC LURES 1261

For this test, responses were recorded from Þve fe- cator variables that were coded with a 1 for ages to males of each age, with 10 replicate measurements per be included and with a 0 for ages to be excluded in individual. All results are presented as mean (ϮSE) the analysis. For example, coding ßies 1Ð3 d old with normalized EAG responses. a 1 and all other ßies with a 0 allowed comparisons Flight Tunnel Bioassays. Bioassays were conducted between ßies 3 d old or less with ßies 4 d old or older. as two-choice tests using 30.2 by 30.2 by 122-cm Plexi- Twelve indicator variables were created, and these glas ßight tunnels (Heath et al. 1993b). were indicator variables were used sequentially as the given the choice of volatiles from an ammonium bi- classiÞcation variable in homogeneity of slopes carbonate lure alone or in combination with a pu- models in Proc GLM (SAS Institute 1985) to test for trescine lure. Each tunnel had two traps made from differences in slopes and intercepts between age clear plastic, snap-cap vials (140 ml; BioQuip, Gar- groups created. dena, CA) mounted to the inner wall of the upwind To account for differences in number of ßies re- side of the tunnel. Test substrates were placed in leased per test, total number captured per test and 500-ml ßasks Þtted with an entry port for the external number captured per treatment were converted to air supply and an efßuent port for delivery of volatiles percentage captured by dividing by number of ßies into the tunnels (Analytical Research Systems, released. Two-sample t-tests using Proc TTEST were Gainesville, FL). Volatile chemicals were introduced used to compare percentage capture of females versus into the tunnel through the back of the trap. The males per test in the bioassays, and percentage capture outside face of the trap (the snap-on cap) was covered of ßies responding to volatiles from ammonium bicar- with green tape to provide a visual cue, and it con- bonate plus putrescine versus ammonium bicarbon- tained a 1.5-cm-diameter hole in the center to provide ate alone, with separate analysis for each sex. Unless a point source release of volatiles from test substrate stated otherwise, summary statistics are presented and a point of entry for the ßies. A wax matrix bait as mean Ϯ SD. station strip (2.54 cm by 6.35 cm by 4 mm thick; Heath et al. 2009) with 1% granulated sugar as feeding stim- Results ulant, 1% methomyl (Lannate, 90% [AI]; DuPont Ag Products, Wilmington, DE) as toxicant, and green EAG Dose–Response Profiles. For all three sub- food color as visual cue was placed in each trap to strates, the relationships between dose and amplitude retain ßies that had entered. All ßies were sexually of antennal response were best Þt by regression with mature (11Ð13 d) and had been protein-starved for hyperbolic models (Fig. 1). The general equation is 24 h before the test. Tests using one sex only were expressed in the form: y ϭ ax/b ϩ x, where y repre- conducted, with the two choices in opposite positions sents the EAG response elicited with x amount of to account for position effect, and tests of the other sex odorant substrate, and the coefÞcients a and b repre- run the following day. For each test, 15Ð25 ßies, de- sent maximum EAG response and receptor binding pending on availability, were released at the down- afÞnity, respectively. Hyperbolic equations are com- wind end of the tunnel at 1400 hours each day, and the monly used for ligand-binding studies and have been number of ßies captured inside the traps was recorded shown previously to serve well for characterizing EAG after 20 h. Tests were conducted under natural doseÐresponse relationships (Kendra et al. 2008). supplemented by ßuorescent light (12:12 L:D), and With the ammonium bicarbonate lure (Fig. 1A), tests were replicated 10 times. there was no difference between EAG responses re- Statistical Analysis. Regression analysis (Systat Soft- corded from mature males and females at any of the ware 2006a) was used to describe the relationships doses tested. Regression equations were as follows: between substrate dose and EAG response, with sep- female y ϭ 119.616x/1.036 ϩ x; R2 ϭ 0.994; male y ϭ arate analyses for each test substrate and each sex. 108.071x/1.141 ϩ x; R2 ϭ 0.997. EAG response of both Several regression models were explored, including sexes increased with ammonium bicarbonate dosage linear, logarithmic, polynomial, and hyperbolic mod- up through the 2-ml dose. There was no signiÞcant els. Analysis by t-test (Systat Software 2006b) was increase in response when dose increased from 2 to 4 performed to evaluate differences (P Յ 0.05) between ml, indicating that the dose range evaluated encom- EAG responses of males and females at each substrate passed substrate sufÞcient to elicit maximal antennal dose, differences within sex between adjacent doses, response. and differences within sex between substrates at each With the putrescine lure (Fig. 1B), mean female dose. response (y ϭ 48.287x/0.858 ϩ x; R2 ϭ 0.998) was Regression analysis was also used to describe the signiÞcantly greater than male response (y ϭ 25.757x/ relationships between female chronological age, 0.634 ϩ x; R2 ϭ 0.985) throughout the upper dose stages of ovarian development, and EAG responses range (doses Ͼ0.5 ml, denoted by asterisks in Fig. 1B). to single substrates. Sigmoidal and peak regression In males, maximal EAG response to putrescine was models were evaluated in addition to the models achieved with the 0.5-ml dose, with no appreciable listed above. Indicator variables (Zar 1999) were increase in antennal response thereafter. In contrast, used to determine the line to either side of a hy- there was incremental increase in amplitude of the pothetical age break point where the regression line female EAG response as dosage increased up through changed slope (Capinera et al. 1987, Kendra et al. the 2-ml dose and response leveled off in the 2- to 4-ml 2005a). This was done by creating a series of indi- range. Comparing EAG responses to putrescine (Fig. 1262 ENVIRONMENTAL ENTOMOLOGY Vol. 38, no. 4

Fig. 1. Mean (ϮSE) EAG responses of mature Caribbean fruit ßies to vapor emitted from commercially available lures: ammonium bicarbonate (A), putrescine (B), and a 1:1 combination of the two lures (C). Responses to lures are expressed as a percentage of the standard reference response (2-butanone vapor). DoseÐresponse curves generated with hyperbolic regression models. Experimental doses with an asterisk indicate signiÞcant differences between male and female response (t-test, P Յ 0.05).

1B) versus ammonium bicarbonate (Fig. 1A), both gradually to intermediate levels (58.5 Ϯ 1.8) by 14 d. sexes showed a much higher response to the ammo- The maximal responses were recorded from immature nium bicarbonate lure at all experimental doses. females in stage 3Ð4 ovarian development (Fig. 2A). When vapors from the ammonium bicarbonate and Analysis with the indicator variables identiÞed a break putrescine lures were presented concurrently (Fig. point in the regression curve at5d(F ϭ 33.62; df ϭ 2 1C), female response (y ϭ 158.199x/1.044 ϩ x; R ϭ 3,66; P Ͻ 0.0001), and there were signiÞcant differ- 0.991) was signiÞcantly higher than male response ences in both intercept (F ϭ 87.03; df ϭ 1,66; P Ͻ ϭ ϩ 2 ϭ (y 135.807x/1.329 x; R 0.988) throughout the 0.0001) and slope (F ϭ 93.37; df ϭ 1,66; P Ͻ 0.0001) for entire range of doses. In both sexes, EAG response to the regression lines for ßies 1Ð5 d old versus ßies 6Ð14 the two-component mixture increased with dosage up d old. through the 2-ml dose, and response to the mixture EAG response to putrescine vapor (Þxed 2-ml dose) was approximately equal to the sum of responses to the as a function of female age was also best Þt by regres- individual substrates. Comparing EAG responses to ϭ ϩ the mixture (Fig. 1C) versus the individual substrates sion with a log-normal equation (y 16.57 Ϫ 2 2 ϭ (Fig. 1A and B), signiÞcantly greater antennal re- 14.25e{ 0.5[ln(x/9.55)/0.86] }; R 0.829; Fig. 2C). sponses were elicited with the mixture than with pu- As with the ammonium bicarbonate substrate, the Ϯ trescine alone, in both sexes and at all doses tested. lowest antennal response (17.4 4.5) to putrescine Female response was greater to the mixture than to was recorded from teneral adults. However, pu- ammonium bicarbonate alone for all but one of the test trescine response increased steadily throughout the doses (0.5 ml). However, male response to the mixture Þrst week (ovary stages 1Ð4) to reach a plateau level was greater than response to ammonium bicarbonate (33.2 Ϯ 5.1) by 8 d (stage 5). Although not as dramatic only at the two highest doses. a change as observed with ammonium bicarbonate, EAG Temporal Profiles. The relationship between amplitude of EAG response to putrescine doubled stages of ovarian development and female chronolog- over this period. Tests with the indicator variables ical age was best Þt by regression with a sigmoidal found that separation of ßies that were younger than model (y ϭ 6.29/{1 ϩ e[Ϫ(x Ϫ5.19)/2.10]}; R2 ϭ 0.983, 8 d from ßies that were8dorolder gave the best Þt where x is ßy age in days, and y is mean ovarian stage; (F ϭ 9.18; df ϭ 3,66; P Ͻ 0.0001), and there were Fig. 2A). Adult females were sexually immature differences in both intercept (F ϭ 8.10; df ϭ 1,66; P ϭ (stages 1Ð4) for the Þrst week after emergence, 0.006) and slope (F ϭ 8.08; df ϭ 1,66; P ϭ 0.006) for reached maturity at 8 d (stage 5) when peak egg load these two cohorts. was observed, and entered the ovipositional phase Flight Tunnel Bioassays. Overall, a higher percent- (stage 6) at9dasreßected by a signiÞcant drop in age of females were captured than males in the bio- mean egg load (Fig. 2A, bars). This temporal proÞle assays (48.1 Ϯ 9.5 and 27.8 Ϯ 9.8%, respectively, t ϭ for female reproductive maturation was consistent 4.72, df ϭ 18, P ϭ 0.0002). In the comparison between with previous results obtained with laboratory strain A. suspensa (Kendra et al. 2006). treatments (Fig. 3), more females were captured in A log-normal peak regression model best described response to ammonium bicarbonate plus putrescine ϭ ϭ the changes in EAG response to ammonium bicarbon- than to ammonium bicarbonate alone (t 2.99, df ϭ ate vapor (Þxed 2-ml dose) that occurred with in- 18, P 0.0078). There was no difference, however, in ϭ ϭ ϭ creasing age of female (y ϭ 32.19 ϩ 59.55e{Ϫ0.5[ln(x/ capture of males by these lures (t 0.33, df 18, P 5.50)/0.88]2}; R2 ϭ 0.888; where x is ßy age in days, and 0.7416). Based on the average number of ßies captured y is normalized EAG response; Fig. 2B). Lowest EAG per treatment, the female to male ratio was higher for response (42.4 Ϯ 3.2) was recorded from females 1 d ßies captured in response to ammonium bicarbonate old, it increased quickly to reach peak response plus putrescine (3:1) versus ammonium bicarbonate (92.3 Ϯ 6.9) by 5Ð6 d after eclosion, and declined alone (1:1). August 2009 KENDRA ET AL.: A. suspensa RESPONSE TO SYNTHETIC LURES 1263

Fig. 3. Mean (ϮSE) percentage of Caribbean fruit ßies captured in two-choice bioassay presenting ammonium bi- carbonate (AB) or ammonium bicarbonate plus putrescine (ABϩPu). All ßies were sexually mature, n ϭ 10 replicate trials per sex, 15Ð25 ßies per trial.

tabolite (Tabor and Tabor 1976) and as a volatile organic compound, liberated through bacterial de- composition of protein. In this latter form, putrescine can serve as a semiochemicalÑan olfactory cue for protein-rich food sources (Wakabayashi and Cun- ningham 1991)Ñfunctionally analogous to ammonia vapor (Bateman and Morton 1981, Drew et al. 1983). In addition, putrescine can undergo spontaneous ox- idation to generate 1-pyrroline, another odoriferous compound known to be a protein cue for A. ludens (Robacker 2001). Female Anastrepha are sexually im- mature at eclosion and dependent on consumption of protein for completion of ovarian maturation (Kendra et al. 2006). This critical requirement for dietary pro- tein is the biological factor underlying fruit ßy attrac- Fig. 2. Female physiological state (A) as indicated by tion to liquid protein baits, ammonia-based lures, pu- Ϯ mean ( SE) stage of ovarian development (left axis, regres- trescine, and other alkyl amines generated from sion curve) and number of mature oocytes per ovary, or egg microbial decay. load (right axis, bar graph), and corresponding EAG re- sponses to Þxed doses (2 ml saturated vapor from commercial Results from both the EAG analyses and the ßight lures) of ammonium bicarbonate (B) and putrescine (C) tunnel bioassays support the hypothesis that pu- measured from Caribbean fruit ßies 1Ð14 d after emergence. trescine is the chemical component generating the EAG responses expressed as a percentage of the standard strong female bias in trap captures with the two-com- reference response (2-butanone vapor). Temporal proÞles ponent lure. EAG response to pure ammonia (Kendra for ovarian development and EAG responses generated with et al. 2005b) and to the ammonium bicarbonate lure sigmoidal and log-normal peak regression models, respec- (presented herein) was equivalent in males and fe- tively. Dotted lines indicate break points in EAG regression males. EAG response to the commercial putrescine curves as determined by indicator variable analysis. lure, as well as to reagent-grade putrescine and ca- daverine (1,5-diaminopentane, a terminal diamine ho- mologous to putrescine; Kendra et al. 2008), was sig- Discussion niÞcantly higher in females, given sufÞcient dosage. The most effective synthetic lures currently avail- When the putrescine lure was combined with ammo- able for pest Anastrepha are multicomponent blends nium bicarbonate, female EAG response was signiÞ- that include an ammonia-emitting substrate and the cantly higher over all doses tested. Differences were terminal diamine putrescine (1,4-diaminobutane). Al- also observed between the sexes with respect to the though not attractive when deployed alone, pu- putrescine dose at which maximal EAG response was trescine has been shown to be a strong synergist to achieved. Male response leveled off at the 0.5-ml dose, ammonium acetate for capture of both A. ludens whereas female response continued to increase with (Heath et al. 1995, 2004) and A. suspensa (Kendra et the addition of substrate up through the 2.0-ml dose. al. 2008). In nature, putrescine is derived from the This four-fold difference in saturation levels suggests amino acids or ornithine (Morris and Pardee that the female antenna may contain more olfactory 1966) and can be found both as an intracellular me- receptors for putrescine as compared with the male 1264 ENVIRONMENTAL ENTOMOLOGY Vol. 38, no. 4 antenna. In Drosophila, females typically have a larger sory system can be very different in larval and adult number of olfactory sensilla than males, with the third stages of the same insect (Couto et al. 2005). Plasticity antennal segment (the primary olfactory organ in within the adult stage has also been reported; for most higher dipterans) bearing Ϸ419 in males and 457 instance, in Aedes aegypti L., the antennal receptor in females (Shanbhag et al. 1999). Sexual differences sensitivity to lactic acid (a vertebrate cue for in peripheral olfactory systems have been docu- mosquitoes) is much reduced after a bloodmeal, when mented in tephritid species (e.g., Toxotrypana curvi- host-seeking behaviors are replaced by oviposition- cauda; ArzufÞ et al. 2008). Likewise, sexual dimor- seeking behaviors (Davis 1984). Although the vari- phism in Anastrepha antennal morphology may be a ability in olfactory response of Anastrepha can be contributing factor in the female-biased attraction ob- correlated with major developmental events in the served with the two-component blend. Additional ßyÕs life, the underlying cellular processes have not yet studies, including electron microscopy and single-cell been studied in the Tephritidae. Possible mechanisms recordings are needed to address this area. to account for this variability include hormone-medi- The temporal variation in female EAG response to ated temporal regulation of the protein components the ammonium bicarbonate lure was comparable to (and encoding genes) that comprise the peripheral that obtained with pure ammonia vapor in a prior olfactory system. These include the transmembrane study with A. suspensa. The ammonium bicarbonate olfactory receptors themselves, as well as a variety of break point occurred at 5 d and the ammonia break soluble protein constituents of the sensillum lymph, point occurred at 4 d (Kendra et al. 2005a). With both including odorant-binding proteins, chemosensory substrates, peak antennal response was measured from proteins, and enzymes that remove active odorants immature females (stage 3Ð4) with ovaries actively from the dendritic membrane (de Bruyne and Baker undergoing vitellogenesis, the deposition of yolk pro- 2008). teins (Kendra et al. 2006). This time frame coincides The long-term goal of this research is development with the age of peak consumption of protein (Landolt of improved synthetic lures for early detection and and Davis-Hernandez 1993) and with development of monitoring of economically important Anastrepha the tephritid gut, when the peritrophic membrane and fruit ßies. This objective may be achieved through a normal bioÞlm of symbionts are present (Lauzon et al. better understanding of tephritid olfactory ecology, 2009) and active, at which point the ßy likely digests including the cellular and physiological mechanisms protein meals more effectively (C. R. Lauzon, per- underlying odor-mediated behaviors. Our strategy sonal communication). In contrast, the temporal re- consists of integrating comparative EAG analyses with sponse to putrescine was not as expected. Our hy- aspects of development, sexual maturation, and nutri- pothesis was that both ammonia and putrescine were tional requirements to determine how these variables equivalent protein cues and that the two substrates inßuence initial chemoreception of, and ultimate be- would elicit similar patterns of antennal sensitivity havioral response to, known volatile attractants. Al- correlated with female reproductive physiology. though the relationship between behavioral signiÞ- However, the EAG evidence indicates otherwise, with cance and amplitude of EAG response is not clearly maximal response to putrescine recorded from sexu- understood in many insect systems (Cha et al. 2008), ally mature females. Ammonia and putrescine appar- we saw good correlation between EAG response and ently serve slightly different functional roles as pro- tephritid behavior with the food-based attractants tein-feeding cues for the ßies. The differences in EAG evaluated thus far (Kendra et al. 2005a, 2008). Devel- response proÞles strongly suggest that there are sep- opment of improved female-targeted lures may be arate olfactory receptors, with separate temporal reg- realized by combining semiochemicals that (1) con- ulatory mechanisms, involved in processing informa- vey the same message (e.g., multiple feeding cues), tion from these two semiochemicals. Further support (2) elicit higher EAG responses in females, (3) have is provided by the additive EAG responses obtained additive EAG responses (recruit additional recep- when the two odorants were combined (i.e., increase tors), and (4) elicit peak EAG responses at different in receptor potential above the plateau level observed stages (physiological ages) of the adult femaleÕs life, with individual substrates) and by the presence of thereby achieving broad attraction of ßies regardless separate olfactory neuron types for detection of am- of age structure of the population. Results of this study monia and putrescine documented from Drosophila suggest that enhanced female-biased attraction may (Yao et al. 2005). be realized through improvements in the diamine Through EAG analysis of known-aged A. suspensa, component, such as optimization of dose, formulation, we showed that antennal response to a speciÞc olfac- or synergistic combinations of homologous diamines. tory stimulus is not constant throughout the life of an adult ßy but varies according to the physiological state and nutritional needs of the insect. Thus far, we eval- uated ammonia, CO2 (Kendra et al. 2005a), ammo- Acknowledgments nium bicarbonate, and putrescine and saw temporal The authors thank C. Allen, I. Filpo, E. Schnell, A. changes in EAG response to each of these odorants. Va´zquez, M. Gill, D. Long, L. Mazuera, J. Sanchez, and J. Plasticity in the olfactory system based on changing Tefel (USDAÐARS, Miami, FL) for technical assistance and ecological needs is adaptive for insects undergoing C. Lauzon ( State University, East Bay), V. Soroker metamorphic changes, and the peripheral chemosen- (Agricultural Research Organization of Israel, Beit-Dagan), August 2009 KENDRA ET AL.: A. suspensa RESPONSE TO SYNTHETIC LURES 1265

D. Jenkins (USDAÐARS, Mayagu¨ ez, P.R.), and two anony- Holler, T., J. Sivinski, C. Jenkins, and S. Fraser. 2006. A mous reviewers for helpful suggestions with the manuscript. comparison of yeast hydrolysate and synthetic food at- tractants for capture of Anastrepha suspensa (Diptera: Tephritidae). Fla. Entomol. 89: 419Ð420. References Cited Kendra, P. E., W. S. Montgomery, D. M. Mateo, H. Puche, N. D. Epsky, and R. R. Heath. 2005a. Effect of age on Aluja, M. 1994. Bionomics and management of Anastrepha. EAG response and attraction of female Anastrepha sus- Annu. Rev. Entomol. 39:155Ð178. pensa (Diptera: Tephritidae) to ammonia and carbon Arzuffi, R., N. Robledo, and J. Valdez. 2008. Antennal sen- dioxide. Environ. Entomol. 34: 584Ð590. silla of Toxotrypana curvicauda (Diptera: Tephritidae). Kendra, P. E., A. Va´zquez, N. D. Epsky, and R. R. Heath. Fla. Entomol. 91: 669Ð673. 2005b. Ammonia and carbon dioxide: quantitation and Bateman, M. A., and T. C. Morton. 1981. The importance of electroantennogram responses of Caribbean fruit ßy, ammonia in proteinaceous attractants for fruit ßies (Fam- ily: Tephritidae). Austr. J. Agric. Res. 32: 883Ð903. Anastrepha suspensa (Diptera: Tephritidae). Environ. En- Capinera, J. L., D. R. Horton, N. D. Epsky, and P. L. Chap- tomol. 34: 569Ð575. man. 1987. Effects of plant density and late-season de- Kendra, P. E., W. S. Montgomery, N. D. Epsky, and R. R. foliation on yield of Þeld beans. Environ. Entomol. 16: Heath. 2006. Assessment of female reproductive status 274Ð280. in Anastrepha suspensa (Diptera: Tephritidae). Fla. En- Cha, D., S. Nojima, S. Hesler, A. Zhang, C. Linn, W. Roelofs, tomol. 89: 144Ð151. and G. Loeb. 2008. IdentiÞcation and Þeld evaluation of Kendra, P. E., M. K. Hennessey, W. S. Montgomery, E. M. grape shoot volatiles attractive to female grape berry Jones, and N. D. Epsky. 2007. Residential composting of moth (Paralobesia viteana). J. Chem. Ecol. 34: 1180Ð1189. infested fruit: a potential pathway for spread of Anas- Couto, A., M. Alenius, and B. J. Dickson. 2005. Molecular, trepha fruit ßies (Diptera: Tephritidae). Fla. Entomol. 90: anatomical, and functional organization of the Drosophila 314Ð320. olfactory system. Curr. Biol. 15: 1535Ð1547. Kendra, P. E., N. D. Epsky, W. S. Montgomery, and R. R. Davis, E. E. 1984. Development of lactic acid-receptor sensi- Heath. 2008. Response of Anastrepha suspensa (Diptera: tivity and host-seeking behavior in newly emerged female Tephritidae) to terminal diamines in a food-based syn- Aedes aegypti mosquitos. J. Insect Physiol. 30: 211Ð215. thetic attractant. Environ. Entomol. 37: 1119Ð1125. de Bruyne, M., and T. C. Baker. 2008. Odor detection in Landolt, P. J., and K. M. Davis-Hernandez. 1993. Temporal insects: volatile codes. J. Chem. Ecol. 34: 882Ð897. patterns of feeding by Caribbean fruit ßies (Diptera: Drew, R.A.I., A. C. Courtice, and D. S. Teakle. 1983. Bac- Tephritidae) on and hydrolyzed yeast. Ann. En- teria as a natural source of food for adult fruit ßies tomol. Soc. Am. 86: 749Ð755. (Diptera: Tephritidae). Oecologia (Berl.) 60: 279Ð284. Lauzon, C. R., S. D. McCombs, S. E. Potter, and N. C. Pea- Epsky, N. D., P. E. Kendra, and R. R. Heath. 2004. Devel- body. 2009. Establishment and vertical passage of En- opment of lures for detection and delimitation of invasive terobacter (Pantoea) and Klebsiella Anastrepha fruit ßies, pp. 84Ð89. In W. Klassen, W. Colo´n, pneumoniae through all life stages of the Mediterranean and W. I. Lugo (eds.), Proceedings of the 39th Annual fruit ßy (Diptera: Tephritidae). Ann. Entomol. Soc. Am. Meeting of the Caribbean Food Crops Society, 13Ð18 July 102: 85Ð95. 2003, Grenada, West Indies. C.F.C.S., St. Croix, USVI. Lopez, F. F., L. F. Steiner, and F. R. Holbrook. 1971. A new Greany, P. D., and C. Riherd. 1993. Preface: Caribbean fruit yeast hydrolysate-borax bait for trapping the Caribbean ßy status, economic importance, and control (Diptera: fruit ßy. J. Econ. Entomol. 64: 1541Ð1543. Tephritidae). Fla. Entomol. 76: 209Ð211. Mazor, M., S. Gothilf, and R. Galun. 1987. The role of am- Hall, D. G., R. E. Burns, C. C. Jenkins, K. L. Hibbard, D. L. monia in the attraction of females of the Mediterranean Harris, J. M. Sivinski, and H. N. Nigg. 2005. Field com- fruit ßy to protein hydrolyzate baits. Entomol. Exp. Appl. parison of chemical attractants and traps for Caribbean 43: 25Ð29. fruit ßy (Diptera: Tephritidae) in Florida citrus. J. Econ. Morris, D. R., and A. B. Pardee. 1966. Multiple pathways of Entomol. 98: 1641Ð1647. putrescine biosynthesis in Escherichia coli. J. Biol. Chem. Heath, R. R., N. D. Epsky, P. J. Landolt, and J. Sivinski. 1993a. 241: 3129Ð3135. Development of attractants for monitoring Caribbean fruit Norrbom, A. L. 2000. The Diptera site, Tephritidae main ßies (Diptera: Tephritidae). Fla. Entomol. 76: 233Ð244. page, Anastrepha. (http://www.sel.barc.usda.gov/Diptera/ Heath, R. R., A. Manukian, N. D. Epsky, J. Sivinski, C. O. tephriti/Anastrep/Anastrep.htm). Calkins, and P. J. Landolt. 1993b. A bioassay system for collecting volatiles while simultaneously attracting te- Pingel, R. L., N. D. Epsky, and R. Goenaga. 2006. Field trials phritid fruit ßies. J. Chem. Ecol. 19: 2395Ð2410. to attract fruit ßies (Diptera: Tephritidae) in commercial Heath, R. R., N. D. Epsky, A. Guzman, B. D. Deuben, A. sapodilla, mamey sapote, and carambola orchards in Manukian, and W. L. Meyer. 1995. Development of a Puerto Rico. J. Agric. Univ. Puerto Rico 90: 109Ð113. dry plastic with food-based synthetic attract- Robacker, D. C. 1999. Attraction of wild and laboratory- ant for the Mediterranean and the Mexican fruit ßy strain Mexican fruit ßies (Diptera: Tephritidae) to two (Diptera: Tephritidae). J. Econ. Entomol. 88: 1307Ð1315. synthetic lures in a wind tunnel. Fla. Entomol. 82: 87Ð96. Heath, R. R., N. D. Epsky, D. Midgarden, and B. I. Katsoy- Robacker, D. C. 2001. Roles of putrescine and 1-pyrroline in annos. 2004. EfÞcacy of 1,4-diaminobutane (pu- attractiveness of technical-grade putrescine to the Mex- trescine) in a food-based synthetic attractant for capture ican fruit ßy (Diptera: Tephritidae). Fla. Entomol. 84: of Mediterranean and Mexican fruit ßies (Diptera: Te- 679Ð685. phritidae). J. Econ. Entomol. 97: 1126Ð1131. Robacker, D. C., and W. C. Warfield. 1993. Attraction of Heath, R. R., S. G. Lavallee, E. Schnell, D. G. Midgarden, and both sexes of the Mexican fruit ßy, , to N. D. Epsky. 2009. Laboratory and Þeld cage studies on a mixture of ammonia, methylamine, and putrescine. female-targeted attract-and-kill bait stations for Anas- J. Chem. Ecol. 19: 2999Ð3016. trepha suspensa (Diptera: Tephritidae). Pest Manag. Sci. Shanbhag, S., B. Muller, and A. Steinbrecht. 1999. Atlas of 65: 672Ð677. olfactory organs of Drosophila melanogaster. 1. Types, ex- 1266 ENVIRONMENTAL ENTOMOLOGY Vol. 38, no. 4

ternal organization, innervation and distribution of olfactory and capture of Anastrepha fruit ßies (Diptera: Tephriti- sensilla. Int. J. Insect Morphol. Embryol. 28: 377Ð397. dae). J. Entomol. Sci. 43: 76Ð85. SAS Institute. 1985. SAS/STAT guide for personal comput- Wakabayashi, N., and R. T. Cunningham. 1991. Four-compo- ers, version 6. SAS Institute, Cary, NC. nent synthetic food bait for attracting both sexes of the Steyskal, G. C. 1977. History and use of the McPhail trap. melon ßy (Diptera: Tephritidae). J. Econ. Entomol. 84: Fla. Entomol. 60: 11Ð16. 1672Ð1676. Systat Software. 2006a. SigmaPlot 10 userÕs manual. Systat Weems, H. V., Jr., J. B. Heppner, G. J. Steck, T. R. Fasulo, and Software, Point Richmond, CA. J. L. Nation. 2004. Mexican fruit ßy, Anastrepha ludens Systat Software. 2006b. SigmaStat 3.5 userÕs manual. Systat (Loew) (Insecta: Diptera: Tephritidae). (http://creatures. Software, Point Richmond, CA. ifas.uß.edu/fruit/tropical/mexican_fruit_ßy.htm). Tabor, C. W., and H. Tabor. 1976. 1,4-diaminobutane (pu- White, I. M., and M. M. Elson-Harris. 1992. Fruit ßies of trescine), spermidine, and spermine. Annu. Rev. Bio- economic signiÞcance: their identiÞcation and bionom- chem. 45: 285Ð306. ics. CAB International, Wallingford, United Kingdom, in Thomas, D. B. 2004. Hot peppers as a host for the Mexican association with The Australian Centre for International fruit ßy, Anastrepha ludens (Diptera: Tephritidae). Fla. Agriculture Research, Canberra, Australia. Entomol. 87: 603Ð608. Yao, C. A., R. Ignell, and J. R. Carlson. 2005. Chemosensory Thomas, D. B., T. C. Holler, R. R. Heath, E. J. Salinas, and coding by neurons in the coeloconic sensilla of the Dro- A. L. Moses. 2001. Trap-lure combinations for surveil- sophila antenna. J. Neurosci. 25: 8359Ð8367. lance of Anastrepha fruit ßies (Diptera: Tephritidae). Fla. Zar, J. H. 1999. Biostatistical analysis, 4th ed. Prentice Hall, Entomol. 84: 344Ð351. Upper Saddle River, NJ. Thomas, D. B., N. D. Epsky, C. A. Serra, D. G. Hall, P. E. Kendra, and R. R. Heath. 2008. Ammonia formulations Received 26 November 2008; accepted 5 May 2009.