201 Entomologia Experimentalis et Applicata 82: 201–211, 1997. c 1997 Kluwer Academic Publishers. Printed in Belgium. Effects of lectins, CRY1A/CRY1B Bt  -endotoxin, PAPA, protease and -amylase inhibitors, on the development of the rice weevil, Sitophilus oryzae, using an artificial seed bioassay B.R. Pittendrigh, J.E. Huesing, R.E. Shade & L.L. Murdock Department of Entomology, 1158 Entomology Hall, Purdue University, West Lafayette, IN 47907-1158, USA Accepted: September 27, 1996 Key words: biomonitor, Coleoptera, Curculionidae, lectin, PAPA, E-64, artificial seeds, protease inhibitor, weevil, Sitophilus Abstract An artificial maize seed bioassay was developed to evaluate potential resistance factors against the rice weevil, Sitophilus oryzae. Weevils reared in artificial seeds compared to those reared in whole maize seeds: (i) developed faster, (ii) had similar within-seed developmental mortalities, (iii) were lighter in weight upon emergence and (iv) oviposited the same number of eggs. Using this bioassay we found that E-64, a cysteine protease inhibitor, decreased the number of emerged adults per seed and delayed within-seed developmental time, suggesting that the rice weevil utilizes a cysteine protease to digest its dietary protein. Weevils fed inhibitors of trypsin and chymotrypsin, Bowman-Birk and Kunitz inhibitors respectively, developed normally. Para-amino-L-phenylalanine (PAPA), a non-protein amino acid implicated as an insect resistance factor in Vigna vexillata, was lethal at dietary levels of 0.2% (w/w) and higher. An extract from Amaranthus caudatus seeds delayed the developmental time of the rice weevil at dietary levels of 0.2% (w/w) and increased mortality at dietary levels of 1.0% (w/w). Several proteins tested, including Griffonia simplicifolia agglutinin II, phytohemagglutinin extract containing common bean -amylase inhibitor, pokeweed agglutinin, Bacillus thuringiensis CRY1A/CRY1B endotoxin, and an -amylase inhibitor from wheat, had no effect on the rice weevil. The artificial maize seed bioassay was adapted by pelleting the seed for use with an ultrasonic insect feeding monitor to determine the finding activity of rice weevils as they developed from egg hatch to pupation. Introduction The availability of maize as food and feed is limited in part by pests that destroy grain post harvest. Members of the Sitophilus complex, Sitophilus oryzae (L.), Sitophilus zeamais Motschulsky, and Sitophilus granarius (L.), attack maize in storage. Synthetic fumigants, such as phosphine, have been used to control these storage pests (Herron, 1990), but will likely have a reduced role in the future due to environmental concerns and increasing levels of insecticide resistance. One attractive alternative control strategy is host plant resistance, where the resistance factor in the plant is not also toxic to the intended mammalian consumer. Since transgenic maize can now be produced through recombinant DNA techniques (Rhodes et al., 1988; Gordon-Kamm et al., 1990), this approach may have a role in producing resistant seed. At present, the bottleneck to producing insect-resistant seeds is the scarcity of resistance genes. One of the keys to discovering insect resistance genes is a bioassay system that: i) is rapid and easy to use, ii) is as similar as possible to the insect’s natural food source, iii) uses only small quantities of test chemical, and iv) will allow researchers to assess the impact of candidate resistance proteins on developmental time and mortality rate of the insect. None of the systems developed (Baker, 1973a; Gomez et al., 1982; Morgan et al. 1993), to date, fulfill all the aforementioned criteria. The objectives of the present research were to develop an artificial seed system in which rice weevils can complete normal development and to use this sys- 202 tem to evaluate selected bioactive molecules such as lectins, protease-, and -amylase inhibitors ( -AI) on rice weevil biology. Materials and methods Insects and general procedures. The Sitophilus oryzae used in this research were reared from a colony maintained on wheat at Manhattan, Kansas. At Purdue, the stock colony was maintained for more than fifteen generations on a local variety of maize seed prior to the beginning of this research (maize seeds, Heartland Co-op, Crawfordsville, IN). The colony was kept at 27  2  C, 70  5% r.h. The variety of maize used in the experiments was Pioneer IN 4-34-358. The seeds were equilibrated at  70% r.h. for two months prior to each experiment. Rice weevil egg plugs were detected using the acid fuchsin technique (Pfadt & Brown, 1985). Artificial seeds were equilibrated for three or more days at 27  2  C and 70  5% r.h. The number of eggs laid in the artificial seeds was determined by dissection. Artificial seed bioassay system. The artificial seed system described here was adapted from that developed for the cowpea weevil, Callosobruchus maculatus (F.) (Shade et al., 1986). Maize flour was prepared from Pioneer IN 4-34-358 maize seeds milled in a Tecator Cyclotecr sample mill (Fisher Scientific, Chicago, IL) using a 0.5 mm screen. A paste of maize flour and doubly distilled (DD) H2 O (3:2 ratio) was injected into holes (6 mm dia., 7 mm deep) in a Teflon mold. The filled mold was then frozen in liquid N2 and lyophilized as described by Shade et al. (1986). The resultant pellets were removed from the mold and each maize pellet was dipped into liquid nitrogen for ca. 5 sec, next into a gelatin solution (maintained at 58  2  C for ca. 1 sec, then immediately returned to the liquid nitrogen for ca. 5 sec, and subsequently left to dry on a plastic surface. The gelatin solution contained 50 ml DDH2 O, 2.0 g gelatin (Difco Laboratory, Detroit, MI), and 20 drops of blue food coloring (Durkee, Wayne, New Jersey). The food coloring treatment aided in detection of rice weevil eggs when the seeds were dissected. To determine if the artificial seeds permitted normal growth and development of the rice weevil, comparisons were made between insects reared in artificial and whole seeds. Each experiment was blocked (repeated) three times. Within-seed developmental time (WSDT). Seven vials were used for each of the two treatments: (i) whole maize seeds and (ii) artificial maize seeds. Ten adult rice weevils were placed in each glass vial (20 mm diam  70 mm) with five seeds and the insects were allowed to oviposit in the seeds. Three days after the beginning of the experiment the adults were removed from the vials. The five seeds in each vial were then separated into individual vials. Vials were held at 27  2  C, 70  5% r.h. and observed daily for emerged adults. Insect weight and longevity of starved adults. The experimental design was the same as the previous experiment. When adults emerged, their sex was determined and they were either immediately frozen or allowed to starve to death. The number of days from emergence to death was recorded. Adults were frozen shortly after emergence from the seeds by placing them in a ,20  C freezer for ca. 30 min and then dried at 70  C for 24 h. Dry weights were determined using an analytical balance (accurate to 0.1 mg). Adults allowed to starve to death were also frozen, dried and weighed in the same manner. Oviposition. The experimental design was the same as that for the within-seed developmental time experiment, except seeds were not separated into individual vials. Vials containing five whole maize seeds (treatment one) were paired with vials containing five artificial maize seeds (treatment two). Emerged adults were removed daily from the vials. When at least two males and at least two females emerged from each treatment on the same day, a block was established. Each block consisted of two mating pairs from each treatment for a total of four mating pairs. One mating pair from each treatment was allowed to oviposit in three artificial seeds and the other was allowed to oviposit in three whole seeds. The three seeds in each vial were replaced with the same seed type every two days for a total of 20 days. Seeds were dissected or stained to determine the number of eggs. Within-seed developmental mortality (WSDM). Sixteen vials, with one seed per vial, were prepared for each of the three treatments: (i) whole Pioneer IN 4-34358 maize seeds, (ii) whole maize seeds of a local variety (Heartland Co-op), and (iii) artificial maize seeds made from the flour of Pioneer IN 4-34-358 maize seeds. The local variety of maize seed, which appeared to be very susceptible to weevil attack, was used as a control to determine if the Pioneer IN 4-34-358 vari- 203 ety was resistant to rice weevils. Each vial contained one seed and two mated adult female rice weevils aged 10–17 days post-emergence. Insects were allowed to oviposit on the seeds for 24 h and were then removed. Eight of the seeds in each treatment were dissected to determine the number of eggs per seed. The remaining seeds were observed to determine the number of emerged adults. Impact of selected biologically-active molecules on rice weevil development. Based on the results of the previous experiments, Stein’s two-stage sample (Steel & Torrie, 1980) was used to determine the n-size for the bioassay experiments. Ten artificial seeds were used per treatment with each seed in a separate vial. Two 10-17 day-old adult females were placed in each vial for 24 h, providing for a similar number of eggs in each seed. Three artificial seeds were dissected to determine the number of eggs. The remaining seven artificial seeds were observed daily for emerged adults. Treatments consisted of microgranular cellulose, bovine serum albumin (BSA), trans-epoxysuccinylL-leucylamido-(4-guanidino)butane (E-64), Paraamino-L-phenylalanine (PAPA), -AI from wheat, phytohemagglutinin containing common bean -AI (PHA-P), Kunitz inhibitor, and Bowman-Birk inhibitor (BBI) which were obtained from Sigma (St. Louis, MO). Griffonia simplicifolia agglutinin II (GSA II) and pokeweed agglutinin were obtained from E-Y Laboratories (San Mateo, CA). Amaranth seed extracts were from a dialyzed ammonium sulfate (30–70% saturation) precipitate. Bt CRY1A/CRY1B  -endotoxin was generously provided by Prof. William Moar from Auburn University, Auburn, AL. Behavioral assay. Artificial seeds dense enough to permit the detection of larval feeding activity were prepared by breaking apart uncoated artificial maize seeds and compressing 50 mg of resultant flour into a 1 mm deep, 6 mm diameter pellet, using a pellet press (Parr Instrument Company, Moline, IL). The compressed pellets were coated with a gelatin layer, as previously described. In the behavioral assay it was essential to have only one insect developing in each seed. In a preliminary study, female rice weevils left on an artificial seed for exactly 2 h were found to lay no more than one egg in the seed, although many of the seeds had no eggs present. Accordingly, for infestation, one female rice weevil was placed with each artificial seed for 2 h. Females were allowed to oviposit in whole seeds for 24 h. Infested seeds were placed on the transducers of an eight-channel Purdue insect feeding monitor (Shade et al., 1990). If no feeding events were observed for 1 h, then the seed was replaced with another. Feeding events were recorded on the feeding monitor in 10 min intervals. Feeding was monitored continuously over a period of 4 weeks. Data analysis. Data analysis for experiments comparing whole and artificial seeds was as follows. Within-seed developmental time (WSDT) comparisons were made on a per seed basis. The insect weight experiment was analyzed as a 2  2  2 factorial experiment. Student’s t-test was used to compare the starved longevity of the insects reared on whole and artificial seeds. Student’s t-test was also used to compare developmental times of insects reared in real and artificial seeds in the behavioral assay. Analysis of variance was performed using SuperAnova (Abacus concepts, 1991), unless otherwise noted, and Fisher’s protected LSD was used for means separation (P < 0:05). Within-seed developmental mortality (WSDM) was analyzed using the 2 test. The overall mortality rate in each experiment was first calculated using 2 (Steele & Torrie, 1980), where the total number of adults per treatment was the ‘observed’ and the ‘expected’ was calculated using the formula E  S=R (E = number of eggs per treatment; S = total number of adults per experiment; R = total number of eggs per experiment). If there was a significant difference in the ratio of eggs to adults within the experiment, then a two-celled table (2 ) was used to compare each treatment to the control (Steele & Torrie, 1980). The seeds in which the insects developed and the seeds they subsequently oviposited in were arranged in a split plot. The seed type in which the insects developed was a whole unit arranged in randomized complete block design with the type of seeds they oviposit in as sub-units Time was a split block factor over the split plot. For the oviposition experiment four blocks were established and an ANOVA was used (SAS, 1989). For the bioassay experiment, all parameters (eggs laid, WDST, and numbers of emerged adults) were evaluated using an analysis of variance for a completely randomized design. For the number of eggs laid an the number of emerged adults there was equal replication. For WSDT unequal replication occurred. Linear regression analyses were performed to describe the response of the rice weevil (1) to a compounds tested at three or more doses, where X = dose 204 and Y = mean developmental time in days or number of emerged adults per seed, (2) mortality rate to number of eggs laid per seed, where X = number of eggs laid and Y = mortality rate, and (3) developmental time in response to the WSDM rate, where X = mortality rate and Y = developmental time. Results Within-seed developmental time (WSDT). The mean ( SE) WSDT for insects reared in artificial seeds was slightly but significantly shorter (36:2  0:4days, n = 19, range 31.5-39.2) than for insects reared in whole maize seeds (39:5  0:6 days, n = 16, range 35.0–42.8; d:f : = 33 t = 4:9, P  0:0001). An analysis of variance was also performed to determine if crowding of insects within the seed would result in delays of WSDT. Within-seed developmental time was independent of the number of emerging adults per seed (artificial seeds: F = 0:01, d:f : = 3, 28, P > 0:25, range of 1–4 adults emerged per seed; whole seeds: F = 0:146, d:f : = 2, 74, P > 0:25, range of 1–3 adults emerged per seed). Insect weight. The mean ( SE) weight of adults reared in whole seeds (1:01  0:05 mg, n = 12, range 0.68–1.28) was significantly greater than for those reared in artificial seeds (0:80  0:03 mg, n = 12, range 0.67–0.98; d:f : = 1, 14, F = 60:69, P < 0:001). Starved longevity. The time (days) required for adult insects to starve to death did not differ between those reared in whole and artificial seeds (whole seeds 5:79  0:38 days, n = 14, range 4–9; artificial seeds 5:59  0:274 days, n = 19, range 3–8; d:f : = 31, t = 31, t = 0:43, P > 0:25). Within-seed developmental mortality (WSDM). The mortality rate of insects reared in whole seeds (Pioneer IN 4-34-358 seed: 35 eggs laid, 5 emerged adults; and Heartland Co-op seeds: 37 eggs laid, 4 emerged adults) did not differ from insects reared in artificial seeds (55 eggs laid, 9 emerged adults; 2 = 0:48, d:f : = 2, P > 0:25). Since there was a high mortality rate and a high egg/seed ratio for each treatment, we tested the hypothesis that the number of eggs per seed affected mortality rate. A positive linear correlation (Figure 1) between number of eggs laid per seed and the percentage mortality supported this hypothesis (whole seeds: y = 28:24 + 9:12 x, r2 = 0:87, d:f : = 1; 4, Figure 1. Effect of number of Sitophilus oryzae eggs laid per whole seed on percent mortality of the immatures. Each point has an n 6. Error bars indicate the standard error of the mean.  F = 26:83, P < 0:01). In whole seeds there was no increase in the developmental time as mortality rates increased (y = 33:51 + 0:03 x, r2 = 0:61, d:f : = 1; 4, F = 6:327, 0:25 > P > 0:05). Oviposition. Fecundity of adults reared in whole seeds (20:06  2:23 SE eggs per 2 days) did not differ from those reared in artificial seeds (21:32  1:96 SE eggs per 2 days; F = 0:20, d:f : = 1; 6, P > 0:25). Impact of selected biologically-active compounds on rice weevil development. To evaluate whether added protein on inert bulk per se would affect rice weevil development, seeds were prepared containing BSA or microgranular cellulose. BSA at 1% or 5% (w/w) had no effect on the rice weevils (Table 1). Similarly, there was no effect of cellulose at concentrations up to 15% (w/w) on any of the parameters tested except mortality rate (Table 1). A regression analysis showed no effect of cellulose on the number of emerged adults (F = 3:66, d:f : = 1; 8, 0:25 > P > 0:05) or on WSDT (F = 0:92, d:f : = 1; 8, P > 0:25). The significant Pvalue for the mortality rate for the overall experiment (Table 1) was due to the 8% cellulose treatment in which survivorship was extremely high. Despite this result, presumably a statistical anomaly, it appears that bulk has little effect on the rice weevil. E-64 caused a significant decrease in the number of emerged adults per seed as well as delaying WSDT (Table 2). The number of emerged adults per seed significantly decreased at 0.10, 0.15, and 0.20% (w/w) levels (r2 = 0:80, F = 11:90, d:f : = 1; 3, P < 0:05). E-64 increased WSDM at 0.10 and 0.15% (w/w) levels, but not at 0.20%. E-64 prolonged WSDT of the rice 205 Table 1. Effects of bovine serum albumin (BSA) (1A) and microgranular cellulose (1B) on the number of emerged adults, mortality (WSDM) and within-seed developmental time (WSDT) of Sitophilus oryzae reared in artificial maize seeds (AMS)ab Treatment Number of adults emerged per seed from seven artificial seeds Mean SEe % w/w of AMS A AMS BSA B AMS cellulose  Ratio of eggs to Adultscd WSDT Mean SEe  n 0.0 1.0 5.0 3:7 3:0 3:3  0:9  0:6  0:3 11:12 9:10 10:8 34:6 34:8 35:3  0:8  0:8  1:6 6 6 5 0.0 1.0 2.0 3.0 4.0 5.0 6.0 8.0 10.0 15.0 2:3 2:0 2:7 2:1 2:0 1:6 1:5 1:6 1:3 1:9  0:4  0:2  0:3  0:4  0:4  0:3  0:2  0:4  0:2  0:3 40:14 47:14 37:16 32:15 35:14 23:11 38.9 19:18  f 35:8 38:13 37:6 35:8 35:1 36:8 36:9 38:2 39:3 36:0 38:5 37:3  1:4  0:9  0:5  1:1  0:8  0:7  1:2  1:7  1:4  0:9 6 7 6 7 7 7 6 7 6 7 a Table 1A–B: represents two experiments performed on different dates. There was no significant difference in the numbers of eggs laid between treatments within 0:192, d:f : 8, 18, P > 0:25; and for 1B: F 1:05, each of the experiments, 1A: F d:f : 6, 14, P > 0:25. c Number of eggs counted from three seeds and number of adults from seven seeds. d WSDM calculated using 2 (Steele and Torrie 1980). Ratios significantly different from the control at P < 0:05 and P < 0:01 were followed by  and  , respectively. For experiments, 1A: 2 0:68, d:f : 2, P > 0:25; and for 1B: 2 22:65, d:f : 9, P < 0:01. e There was no significant difference in the number of emerged adults between treatments within each of the experiments, 1A: F 1:21, d:f : 2, 17, P > 0:25; and for 1B: F 1:54, 9, 60, 0:25 > P > 0:05 or in the developmental time between treatments within two d:f : of the experiments, 1A: F 0:11, d:f : 2, 14, P > 0:25; and for 1B: F 1:3, d:f : 9, 56, P > 0:05. f Mortality was significantly lower than the control. b = = = = = = = = = weevil (Table 2) with a delay of 9.4 days occurring for every 0.1% (w/w) increase in dose (y = 36:9 + 90:2 x, r2 = 0:90, F = 25:9, d:f : = 1; 3, P < 0:05). The following compounds, when incorporated into artificial seeds, had no effect on the rice weevils: Bowman-Birk inhibitor (number of emerged adults: r2 = 0:39, F = 1:28, d:f : = 1; 2, P > 0:25; WSDT: r2 = 0:57, F = 2:66, d:f : = 1; 2, > 0:25 > P > 0:05), GSA II (number of emerged adults: r2 = 0:06, F = 0:18, d:f : = 1; 3, P > 0:25; WSDT: r2 = 0:03, F = 0:08, d:f : = 1; 3, P > 0:25), PHA-P, and pokeweed agglutinin (Tables 1 and 2). Wheat -AI had no effect on rice weevil WSDT (r2 = 0:05, F = 0:17, d:f : = 1; 3, P > 0:25; Table 2) but regression analysis showed that the numbers of emerged adults decreased as the dose increased (r2 = 0:91, F = 28:54, d:f : = = = = = = = = 1; 3, P < 0:05). At the highest dose of -AI there was no significant reduction in the numbers of emerged adults as compared to the control (Table 2), when the data was analyzed with an ANOVA. Kunitz inhibitor did not reduce the number of emerged adults or delay WSDT of the rice weevil (number of emerged adults: r2 = 0:06, F = 0:12, d:f : = 1; 2, P > 0:25; WSDT: r2 = 0:76, F = 6:20, d:f : = 1; 2, 0:25 > P > 0:05; Table 3). Bt CRY1A/CRY1B  -endotoxin incorporated into the artificial seeds at levels up to 0.0100% (w/w) had no effect on the rice weevil when analyzed with an ANOVA (number of eggs laid: F = 8:4, d:f : = 5; 12, 0:25 > P > 0:05; WSDT: F = 1:7, d:f : = 5, 25, 0:25 > P > 0:05; and number of emerged adults: F = 0:246, d:f : = 5; 29, P > 0:25; and mortality rate: 2 = 10:91, d:f : = 5, P < 0:05) or with a regres- 206 Table 2. Effects of E-64, Bowman-Birk inhibitor (BBI), PHA-P, GSA II (seed) and -AI on the number of emerged adults, mortality (WSDM) and within-seed developmental time (WSDT) of Sitophilus oryzae reared in artificial maize seeds (AMS)a Treatment AMS E-64 BBI PHA-P GSA II -AI a % w/w of AMS Number of adults emerged per seed from seven artificial seeds Mean SEd 0.00 0.05 0.10 0.15 0.20 0.20 1.00 2.00 2.00 0.05 0.10 0.25 0.50 0.05 0.10 0.50 1.00 2:0 1:0 0:6 0:4 0:4 1:7 2:06 2:3 1:3 1:9 1:3 2:0 1:6 2:1 2:1 1:9 1:4   0:3  0:2  0:3  0:2 + 0:2  0:4  0:4 + 0:5  0:3  0:3  0:3  0:3  0:2  0:4  0:4  0:3  0:4 Ratio of eggs to Adultsbc 13:14 9:7 16:4 10:3 9:3 15:12 14:18 18:16 13:14 12:13 17:9 9:14 18:11 18:15 14:15 16:13 13:10 WSDT (days) Mean SEe  34:4  1:6 44:1  1:2 46:0  2:1 52:7  0:7 52:7  1:3 34:1  2:1 31:3  0:4 32:2  0:9 32:7  1:0 33:5  1:5 34:6  1:0 32:0  0:7 35:1  1:6 33:2  1:4 33:5  0:8 31:4  0:5 33:7  0:8 n 7 6 3 3 3 7 7 7 6 7 7 7 7 7 7 7 6 There was no significant difference in the numbers of eggs laid between treatments within the experiment: F 0:64, d:f : 16, 34, P > 0:25. b Number of eggs counted from three seeds and number of adults from seven seeds. c WSDM calculated using 2 (Steele and Torrie 1980). Ratios significantly different from the 2 control at P < 0:05 represented as  . For this experiment;  34:22, d:f : 16, P < 0:01. d Means followed an  differ significantly from the AMS control mean (LSD) at p 0:01. 2:99, d:f : 16, There was a significant difference in the numbers of emerged adults (F 102, P < 0:001) and in the developmental time between treatments within the experiment (F 20:35, d:f : 16, 87, p < 0:001). = = = = = =  = = sion (number of emerged adults: r2 = 0:59, F = 5:85, d:f : = 1; 4, 0:25 > P > 0:05; WSDT: r2 = 0:0:2, F = 0:07, d:f : = 1; 4, P > 0:25). The selected compounds that had the greatest impact on rice weevil development were PAPA and amaranth. At 0.1% (w/w) PAPA had no effect on the rice weevil, but at 0.2% (w/w) and higher doses, no adults emerged (Table 3). Extract of amaranth incorporated into the artificial seeds at 0.20% (w/w) delayed rice weevil WSDT but did not affect WSDM (Table 3). At  1% (w/w) no adults emerged (Table 3). Behavioral assay. Feeding activities of rice weevil immatures developing in intact and artificial maize seeds were observed using the biomonitor. The feeding patterns were similar in the two treatments (Figure 2a,b). Insects observed in both intact and artificial maize seeds displayed four instars. From first to fourth stadium there was an increase in the number of feeding events detected for insects reared in both treatments. Each stadium was interrupted by a period of ca. 8– 15 h when no feeding activity was observed. During this time the insects were molting as was confirmed by dissection of seeds. Mean ( SE) developmental time spent in the larval stages was not significantly different between rice weevils reared in whole (439:39  29:39 h, n = 2, range 410.00–468.78) and artificial maize seeds (489:83  8:76 h, n = 2, range 481.16–497.90; t = 1:64, d:f : = 2, P > 0:05). 207 Figure 2. Feeding events of an immature Sitophilus oryzae from the beginning of the first larval instar to pupation. Both individuals exhibited four larval instars. (A) The whole seed was placed on the biomonitor at the end of a 24 h period of adult oviposition. (B) The artificial seed was placed on the biomonitor after a 2 h period of adult oviposition. 208 Table 3. Effects of Kunitz inhibitor, amaranth, pokeweed lectin and Para-amino-Lphenylalanine (PAPA) on the number of emerged adults, mortality (WSDM) and within-seed developmental time (WSDT) of Sitophilus oryzae reared in artificial maize seeds (AMS)ab Treatment A AMS Kunitz inhibitor amaranth B AMS pokeweed lectin PAPA % w/w of AMS Number of adults emerged per seed from seven artificial seeds Mean SEe 0.0 0.2 1.0 2.0 0.2 1.0 2.0 1:3 1:9 2:9 1:7 0:6 0:0 0:0  0:3  0:3  0:5 f  0:4  0:2  0:0  0:0 29.9 21:13 17:20 g 28.12 26.4 18:0 13:0 37:5 36:5 35:4 35:3 43:3 ——– ——– 0.0 0.1 1.0 0.1 0.2 0.4 0.6 0.8 1.0 1:7 1:7 1:6 0:2 0:0 0:0 0:0 0:0 0:0  0:3  0:3  0:4  0:2  0:0  0:0  0:0  0:0  0:0 33:12 32:12 32:11 26:14 28:0 33:0 32:0 31:0 27:0 36:5 36:7 39:1 37:0 ——– ——– ——– ——– ——–  Ratio of eggs to Adultscd WSDT Mean SEe   1:2  0:8  0:6  1:0  4:4  1:0  0:6  0:7  0:5 n 7 6 7 7 3 0 0 7 7 7 7 0 0 0 0 0 a Table 3A–B: represents experiments performed on two different dates. There was no significant difference in the numbers of eggs laid between treatments within each of the experiments, 3A: F 1:05, d:f : 6, 14, P > 0:25; and for 3B: F 0:192, d:f : 8, 18, P > 0:25. c Number of eggs counted from three seeds and number of adults from seven seeds. d WSDM calculated using 2 (Steele and Torrie 1980). Ratios significantly different from the control at P < 0:05 and P < 0:01 followed by  and  , respectively. For experiments, 3A: 2 49:08, d:f : 6, P < 0:01; and for 3B: 2 46:83, d:f : 8, P < 0:01. e Means followed by an  and  respectively differ from each other (LSD) at p < 0:05 and p < 0:01. There was a significant difference in the number of emerged adults (3A: F 11:6, d:f : 6, 42, P < 0:001; and for 3B: F 21:2, d:f : 8, 54, P < 0:01) and in the 4:0, d:f : 4, 23, P < 0:05; and for 3B: F 2:83, d:f : 3, developmental time (3A: F 24, 0:25 > P > 0:05) between treatments within both experiments. f Significantly more adults emerged from the treatment than the control. g Mortality was significantly lower than the control. b = = = = = = = = = = Discussion The artificial maize seed bioassay described here can be used to assess the effects of selected proteins, peptides and other chemicals on rice weevil WSDT and WSDM. While several life-history parameters differed from the intact seed, the artificial seed makes it possible to assess the impact of specific test chemicals on weevil development. Should it be necessary, the effects of chemicals on feeding behavior can be conducted using pelleted artificial maize seeds in conjunc- = = = = = = tion with the Purdue ultrasonic feeding monitor. The latter technique can give insight into larval response to various chemicals as revealed by changes in feeding pattern, lengthening of instars, and time of death, information not available hitherto. Rice weevils reared in whole maize seeds weighed more and took longer to develop than insects reared in artificial maize seeds. According to Roff (1981) an increase in body size of an organism is often associated with longer egg-to-adult developmental time. In other parameters tested, rates of oviposition, WSDM and 209 starved longevity, there were no differences between insects reared in the two treatments. Rearing the weevils in artificial seeds did not affect their reproductive capacity. Since the longevity of starved adults was not significantly different between insects reared in whole and artificial seeds, the two groups likely accumulate similar levels of nutritional reserves. The apparently high mortality rate observed in the WSDM experiment was due to the high number of eggs laid per seed. The WSDM was based on the number of eggs laid per seed and the number of emerged adults. The only way to determine the number of eggs per artificial seed was by breaking open several seeds for each treatment (sample without replacement). As the variation in the number of eggs per seed decreased so to did the number of artificial seeds needed to sample for egg numbers. In preliminary experiments we observed that when two females were allowed to oviposite for 24 h, the number of eggs oviposited per seed was uniform enough that as few as three seeds per treatment could be sacrificed. To estimate the number of eggs laid per seed, more eggs needed to be laid per seed than the seed was capable of supporting to adulthood; resulting in an apparently high mortality rate. As shown in figure one, mortality increases as more eggs were oviposited per seed. When one egg was ovisposited per seed the mortality rate was approximately 30%, a reasonable level of mortality at the temperature and humidity levels maintained in this study (Longstaff, 1981). Bovine serum albumin (BSA), an inert protein, incorporated at a 5% (w/w) level into the diet did not affect the number of emerged adults per seed or WSDT of the rice weevil. Thus, any delays in WSDT or change in number of emerged adults observed in the rice weevil, when proteins are incorporated into the diet at the 5% (w/w) level, are likely due to the qualitative properties of the test protein and not to the presence of bulk protein. Nutritional bulk, i.e., microgranular cellulose, in the diet did not cause delays in developmental time or increases in WSDM at levels up to 15% (w/w). By contrast, increases in WSDM of the cowpea weevil were observed when microgranular cellulose was present in the diet at 8% (w/w) (R.E. Shade, unpubl.). The cowpea weevil may be more dependent on efficiently extracting nutrients from its diet than is the rice weevil. WSDT is a good measure of the effects of test compounds on the rice weevil, since this parameter is independent of the effects of crowding or increased levels of mortality. As was shown in the WSDT experiment, developmental time of the rice weevil was independ- ent of the number of emerged adults per seed. In the WSDM experiment we observed that developmental time was not affected by the mortality rate. Therefore, developmental time is a direct measure of the effects of potential resistance factors on the rice weevil. WSDM is a less useful measure of the impact of selected compounds on rice weevil development. In the presence of E-64, rice weevil WSDM significantly decreased at 0.10, 0.15% (w/w) levels, but not at 0.2%. The lack of significant mortality at a higher dose does not fit established toxicological models. A better measure of the efficacy of selected compounds on rice weevil survivorship in the artificial seed is the comparison between the number of emerged adults per seed. Since all three doses of E-64 [0.10, 0.15, and 0.20% (w/w)] significantly reduced the number of emerged adults, this parameter is concurrent with conventional toxicological models. For most of the compounds tested, analysis by regression or ANOVA resulted in similar conclusions. There was one exception; a regression analysis suggested that wheat -AI had a significant effect on the number of emerged adults. We observed no significant decrease in the number of emerged adults per artificial seed even at the highest doses of -AI (Table 2). Regression analysis should be performed only in combination with an ANOVA to ensure a reasonable assessment of a compounds activity against the rice weevil. This bioassay will allow researchers to assay for compounds that increase rice weevil mortality rates, or delay developmental time or both. Compounds that delay insect developmental time will likely have an important role in controlling multivoltine insects that attack stored grains. The results presented in Table 2 document the effects of dietary E-64 on one generation of the rice weevil. The full impact of a dietary cysteine protease inhibitor is obscured when a single generation is considered. Rice weevil infestations in stored grain can begin at low levels and build up to high levels in several generations until the entire store is destroyed. Murdock et al. (1990) suggested that delays in WSDT, as well as increased mortality of the cowpea weevil, would reduce the buildup of the insect’s population. Based on a mathematical model, it was estimated that wheat germ agglutinin (WGA) present in the diet of the cowpea weevil at 1% (w/w) would reduce the population after 180 days from 256 023 insects (non-resistant seeds) to 3,901 insects (resistant seeds). Using a similar approach, the populations of the rice weevils reared in control and 0.1% (w/w) E-64 artificial maize seeds were modeled. The model was based on the formula 210 Figure 3. Simulation model of the effect of dietary E-64 at 0.1% (w/w) on the Sitophilus oryzae population over time. The X-axis represents time (days) from the introduction of the first mating pair. The Y-axis represents the total number of insects developed (x 1000). r = loge Ro =T (r = instantaneous rate of increase, Ro = the net replacement rate per generation and T = mean generation time). The results presented here assumed (i) an initial infestation of 1 female/kg, (ii) a 50:50 sex ratio, (iii) 30% WSDM, (iv) that each female laid 45 eggs over the first 20 days of her life (v) and a 30% WSDM. The WSDM is directly correlated with the number of eggs/seed. In a sparse population, such as 1 female/kg of seed, one egg should occur per seed, resulting in a WSDM of ca. 30% (Figure 1). As the population density increases so should the number of eggs per seed and the WSDM. For the sake of simplicity, the WSDM was held at 30% throughout the model. For E-64 and the control, a WSDT of 46.0 days and 34.4 days, respectively, were assumed. Using these parameters, the population growth of weevils in a model grain bin was estimated. At the end of 300 days, the total number of adults per kg was approximately ten fold larger for the control seeds than those reared in seed containing E-64 (Figure 3). Acknowledgements The authors would like to thank J. B. Santini and Dr. W. E. Nyquist for their assistance with the statistics. This research was supported in part by funds from the Indiana Agricultural Experiment Station Crossroads ‘90’ program. This is paper has number 14702 of the Purdue Agricultural Experiment Station, W. Lafayette, IN 47907-1158. References Abacus Concepts, SuperAnova, 1991. Abacus Concepts Inc, CA, USA. Baker, J. E. & J. M. Mabie, 1973a. Growth and development of larvae of the granary weevil, Sitophilus granarius (Coleoptera: Curculionidae), on natural and meridic diets. Canadian Entomologist 105: 249–256. Baker, J. E. & J. M. Mabie, 1973b. Growth responses of larvae of the rice weevil, maize weevil and granary weevil on a meridic diet. Journal of Economic Entomology 66: 681–683. Gomez, L. A., J. G. Rodriguez, C. G. Poneleit & D. F. Blake, 1982. Preference and utilization of maize endosperm variants by rice weevil. Journal of Economic Entomology 75: 363–367. Gordon-Kamm, W. J., T. M. Spencer, M. L. Mangano, T. R. Adams, R. J. Daines, W. G. Start, J. V. O’Brien, S. A. Chambers, W. R. Jr. Adams, N. G. Willetts, T. B. Rice, C. J. Mackey, R. W. Krueger, A. P. Kausch & P. G. Lemaux, 1990. Transformation of maize cells and regeneration of fertile transgenic plants. Plant Cell 2: 603–618. Herron, G. A., 1990. Resistance to grain protectants and phosphine in Coleopterous pests of grain stored on farms in New South Wales. Journal of the Australian Entomology Society 29: 183– 189. Longstaff, B. C., 1981. Biology of the grain pest species of the genus Sitophilus (Coleoptera: Curculionidae): A critical review. Protection Ecology 2: 83–130. Morgan, T. D., B. Oppert, T. H. Czapla & K. J. Kramer, 1993. Avidin and streptavidin as insecticidal and growth inhibiting dietary proteins. Entomology Experimentalis et Applicata 69: 97–108. Murdock, L. L., J. E. Huesing, S. S. Nielsen, R. C. Pratt & R. E. Shade, 1990. Biological effects of plant lectins on the cowpea weevil. Phytochemistry 29: 85–89. Pfadt, R. E. & L. R. Brown, 1985. Fundamentals of Applied Entomology. MacMillan Publishing Co., New York. Rhodes, C. A., D. A. Pierce, I. S. Mettler, D. Mascarenhas & J. J. Detmer., 1988. Genetically transformed maize plants from protoplasts. Science 240: 204–207. 211 Roff, D., 1981. On being the right size. American Naturalist 118: 405–422. SAS Users Guide, 1989. SAS Institute Inc. Box 8000, Cary, NC 27511-8000. Shade, R. E., L. L. Murdock, D. E. Foard & M. A. Pomeroy, 1986. Artificial seed system for bioassay of cowpea weevil (Coleoptera: Bruchidae) growth and development. Environmental Entomology 15: 1286–1291. Shade, R. E., E. S. Furgason & L. L. Murdock, 1990. Detection of hidden insect infestations by feeding-generated ultrasonic signals. American Entomologist 36:3 231–234. Steele, G. D. & J. H. Torrie, 1980. Principles and Procedures of Statistics. A Biometrical Approach, Second edition. McGrawHill Publishing Company, New York, 633 p.