Biological Control 181 (2023) 105216 Contents lists available at ScienceDirect Biological Control journal homepage: www.elsevier.com/locate/ybcon Effectiveness of the parasitoid Anisopteromalus calandrae (Hymenoptera: Pteromalidae) in the control of Sitophilus zeamais and Rhyzopertha dominica in paddy rice Lidia del Arco a, b, *, Jordi Riudavets b, José Miguel Campos-Rivela c, María Teresa Martínez- Ferrer c, Nuria Agustí b, Cristina Castañé b, * a Department of Animal Biology, Plant Biology and Ecology, Faculty of Biosciences, Autonomous University of Barcelona, 08193 Bellaterra (Cerdanyola del Vallès), Spain b IRTA, Sustainable Plant Protection, Ctra. Cabrils Km 2, 08348 Cabrils (Barcelona), Spain c IRTA, Sustainable Plant Protection, Ctra. de Balada Km 1, 43870 Amposta (Tarragona), Spain H I G H L I G H T S G R A P H I C A L A B S T R A C T • A. calandrae preferred S. zeamais to R. dominica larvae when both pests were mixed. • R. dominica did not interfere in A. calandrae control efficacy of S. zeamais. • A. calandrae could suppress S. zeamais in the bottom of 500 kg bags of paddy rice. A R T I C L E I N F O A B S T R A C T Keywords: Rice is an important cereal crop in Spain that is frequently attacked by pests that cause significant quantitative Biological control and qualitative losses. Among them, the maize weevil, Sitophilus zeamais, is the key pest, and the lesser grain Ectoparasitoid borer, Rhyzopertha dominica, is a serious pest found later in the season. The cosmopolitan ectoparasitoid, Ani­ Grain weevil sopteromalus calandrae, which attacks late-instar larvae of coleopterans that develop inside grain kernels, has Grain borer Stored grain been reported as an efficient natural enemy. In this study, we evaluated whether the control efficacy of this parasitoid on S. zeamais was affected by the presence of the alternative host, R. dominica. Our laboratory experiment demonstrated the parasitoid preference for S. zeamais larvae when mixed with R. dominica larvae at different proportions. This preference for the maize weevil larvae allowed us to maintain a similar control ef­ ficacy of S. zeamais when it was offered alone than when it was offered together with R. dominica larvae. In our second experiment, done in 500 kg bags of paddy rice, similar results were obtained, since the presence of R. dominica larvae did not alter the control efficacy of A. calandrae on S. zeamais larvae. Therefore, our results * Corresponding authors at: IRTA, Sustainable Plant Protection, Ctra, Cabrils Km 2, 08348 Cabrils (Barcelona), Spain. E-mail addresses: Lidia.delarco@irta.cat (L. del Arco), cristina.castane@irta.cat (C. Castañé). https://doi.org/10.1016/j.biocontrol.2023.105216 Received 14 December 2022; Received in revised form 10 March 2023; Accepted 14 March 2023 Available online 20 March 2023 1049-9644/© 2023 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). L. del Arco et al. Biological Control 181 (2023) 105216 confirmed a good perspective for implementing the use of this natural enemy as a control tool in big bags of paddy rice. 1. Introduction Sitophilus spp. and R. dominica coexist. The aim of this study was to assess whether the control efficacy of Pest management to control stored-product insects currently relies A. calandrae on S. zeamais is affected by the presence of the alternative on the widespread use of conventional insecticides, particularly in host, R. dominica. We hypothesized that the parasitoid’s effectiveness in warmer climates. The few authorized synthetic insecticides and their suppressing S. zeamais would not be affected, and it would also be able excessive use have led to an increase in pesticide resistance, resulting in to effectively control minor contaminations of R. dominica. To test this a lack of efficacy (Corrêa et al., 2011; Kavallieratos et al., 2015). In hypothesis, we first conducted a laboratory trial with different S. zeamais addition, there is growing social concern about environmental pollution / R. dominica ratios and two parasitoid densities to determine parasitoid and human health problems from insecticides on the grains. Therefore, it host selection on mixed populations of both pest species. Then, a sub­ is necessary to develop adequate alternative control measures, among sequent experiment was done in large commercial bags, a common which biological control is a suitable option. The use of natural enemies method of grain storage, filled with paddy rice and offering to the has been reported as an effective method to control and prevent insect parasitoid only S. zeamais, or a combination of S. zeamais and a small populations from reaching pest status, especially in closed environ­ proportion of R. dominica as hosts. ments, such as storage facilities (Riudavets, 2018; Schöller and Flinn, 2000). 2. Materials and methods Rice is an important cereal crop in Spain, with around 800,000 tons produced in recent years (FAOSTATS, 2020). When stored, it is 2.1. Insect colonies frequently attacked by a range of insect pests that cause important quantitative and qualitative losses. Among them, the maize weevil, Insect colonies were maintained under controlled conditions (28 ± 2 Sitophilus zeamais (Motschulsky) (Coleoptera: Curculionidae), is the ◦ C, 70 ± 2 % relative humidity, 16:8 light:dark photoperiod). The lab­ most concerning species present in rice facilities in the northeast of oratory experiment of host selection was performed under the same Spain, as it happens in the whole Mediterranean region (Adler et al. conditions. Stock colonies of the two coleopteran species and of the 2022; Carvalho et al., 2013; Pascual-Villalobos et al., 2006; Riudavets parasitoid were started with samples collected from commercial store­ et al., 2002; Trematerra et al., 2004). The damage produced in kernels houses. S. zeamais was reared on brown rice, R. dominica on wheat, and enables other species to infest the grain and cause additional damage. In A. calandrae on Callosobruchus chinensis (L.) (Coleoptera: Chrys­ the northeast of Spain, the lesser grain borer, Rhyzopertha dominica (F.) omelidae) larvae that had developed on dried beans. The maize weevil (Coleoptera: Bostrychidae), has been reported as a serious pest of stored identity was confirmed through genitalia dissection and molecular grain (Castañé and Riudavets, 2015; Trematerra et al., 2000). Since this methods. species is more thermophilic than Sitophilus spp. (Hagstrum and Milli­ ken, 1988), it is frequently found later in the season coexisting with 2.2. Host selection laboratory experiment S. zeamais but in lower abundance (personal observations). However, the presence of R. dominica in stored grain facilities is increasing due to To determine the preference of the parasitoid for each pest species, the rising temperatures caused by climate change (Adler et al., 2022). we offered them a mix of second-third instar of S. zeamais larvae and Anisopteromalus calandrae (Howard) (Hymenoptera: Pteromalidae) is a fourth instar of R. dominica larvae, since it has been previously described cosmopolitan and solitary ectoparasitoid that attacks late-instar larvae as the preferred stage for the parasitoid (Wen et al., 1994; Solá et al., of coleopterans that develop concealed inside grain kernels, such as 2020). First, we planned to provide females with the following host those of S. zeamais and R. dominica (Smith, 1993; Smith and Press, ratios: 100/0, 75/25, 50/50, 25/75 and 0/100. However, as the larvae 1992). The female maize weevil feeds on the kernel, lays an egg inside it, of both species develop inside the rice, and there is no external sign and then seals the hole. The larva feeds inside the seed, pupates, and indicating that the grains contain a larva, it is not possible to know the emerges as an adult (Longstaff, 1981). In contrast, R. dominica females exact number of hosts offered to the female parasitoids. Since this lay their eggs loosely in the grain, and the larvae bore into a kernel near number can only be estimated from the number of adults that emerged the germ to feed, pupate, and emerge as an adult (Edde, 2012). Para­ in a control treatment, a preliminary test was done to determine the sitoid searching behavior depends on visual, tactile, and chemical reproduction rates of each pest species when paddy rice was infested stimuli received from certain factors, such as environmental conditions with 175 adults/kg of S. zeamais or R. dominica for one week. The (Hong and Ryoo, 1991), host size, and host density (Choi et al., 2001; emergence rate of S. zeamais was 0.69 adults/gr paddy rice, and that of Steidle and Schöller, 2002). R. dominica was 1.45 adults/gr paddy rice. According to these results, Several studies have reported the significant potential of A. calandrae the amount of S. zeamais and R. dominica infested rice needed to offer a to reproduce and successfully suppress beetle populations in stored 100/0, 75/25, 50/50, 25/75 and 0/100 ratio of S. zeamais/R. dominica wheat (Mahal et al., 2005; Menon et al., 2002), maize (Wen and Brower, larvae was calculated (Table 1). To obtain the second and third-instar 1994), and rice (Belda and Riudavets, 2012; Chaisaeng et al., 2010; Nam larvae of S. zeamais, the paddy rice was incubated for two weeks, and et al., 2011). It was equally attracted to arenas of rice infested with in the case of R. dominica for three weeks after being infested with adults R. dominica and S. oryzae larvae when presented in a Y-tube olfactometer as explained before for the preliminary test. test (Belda and Riudavets, 2010). In another study of infested rice The amount of infested rice corresponding with each different ratio samples, the parasitoid produced a higher progeny reduction of S. oryzae was placed in 710 mL plastic containers to which extra clean paddy rice than of R. dominica (Belda and Riudavets, 2012). Although the para­ was added up to 300 g. Containers had a meshed lid for ventilation, two sitoid has been reported to have more progeny on S. oryzae than on small tubes containing sugary water, and a cotton plug for feeding the R. dominica, parasitoid-induced mortality (PIM), which includes host adult parasitoids. We released two different densities, one with 12 pairs feeding, unsuccessful parasitism and/or superparasitism, was higher in (high) and one with three pairs (low) of A. calandrae (0–7 days old). R. dominica, which means that its population may also be considerably After one week, the parasitoids were removed, and the containers were reduced (Ghimire and Phillips, 2007; Solá et al., 2020). However, no incubated until adult emergence. A control treatment without parasit­ information is available regarding parasitoid performance when both oids was also included. Ten replicates per host ratio and parasitoid 2 L. del Arco et al. Biological Control 181 (2023) 105216 density were performed. compared using GLM (quasi-binomial error distribution). Significant deviation from 1:1 female-to-male in emerged A. calandrae was deter­ 2.3. Big bags experiment mined using the chi-squared test. In the big bags experiment, only the data from the cylindrical cages where parasitoids emerged were This experiment was conducted in eight large woven polypropylene included in the analysis. The percentage of emergency reduction of bags that were maintained under ambient conditions in a warehouse S. zeamais, of parasitism, and of PIM were evaluated using a GLM (quasi- during spring and summer. The bags measured 90 cm × 90 cm × 110 cm binomial error distribution), with % S. zeamais and treatment as factors. (height) and were filled with 500 kg of paddy rice. Two stainless-steel Post hoc comparisons were conducted using the Tukey correction for screened cylindrical cages (7 cm long, 5 cm internal diameter, and 1 multiple comparisons. Also, the percentage of R. dominica reduction × 2.5 mm screen) containing either 54 g of rice infested with second- between the treatments was analyzed through GLM with quasibinomial and third-instar S. zeamais larvae + 6 g of non-infested rice (S. zeamais error distribution. For all statistical analyses, a nominal significance of 5 treatment [Sz]), or 38 g of S. zeamais infested rice + 4.5 g of R. dominica % (P < 0.05) was applied. All statistical analyses were conducted with infested rice + 17.5 g of non-infested rice (S. zeamais:R. dominica the R Studio software (RStudio Team, 2022). treatment [Sz/Rd]) were placed at the bottom of each bag. Six pairs of A. calandrae (0–7 days old) were released on the surface of the grain, 3. Results together with a tube containing sugary water, to provide food for the females. Next, all the bags were closed with a rope and covered with 3.1. Host selection laboratory experiment polyester mesh. After two weeks, the screened cages were removed from the bags, and the rice inside the cages was incubated at 25 ◦ C until pest The emergence of each pest species in paddy rice differed from that emergence. Two control treatments were placed inside each bag. The estimated in the preliminary experiment, as expected (Table 1). There­ first, with pests only, consisted of two stainless-steel cylindrical cages fore, the parasitoid-to-host ratio tested ranged from the highest 1:53 in filled with the same amount of infested rice as in the treatment but with the 100 % S. zeamais to the lower 1:88 in the 100 % R. dominica when the perforated laterals covered with a sticky tape to prevent the entry of three A. calandrae females were released. When 12 A. calandrae females the parasitoids released in the big bag. The second, with parasitoid + were released, the parasitoid-to-host ratios tested ranged from 1:09 in pests, consisted of a plastic container with 120 g of rice (twice the same the 100 % S. zeamais to 1:19 in the 100 % R. dominica (Table 1). rice infestation than in the stainless-steel cylindrical cage) but included Significant differences in total host emergence (S. zeamais + six pairs of A. calandrae (0–7 days old). Four rounds of experiments were R. dominica), in comparison with the control treatment without para­ carried out in mid-May, mid-June, mid-July, and mid-August. In all sitoids, were detected among each host ratio offered to the parasitoids rounds, four replicates were conducted for each pest combination and and for both parasitoid densities tested: a higher host reduction was control treatment. observed in ratios with a predominance of S. zeamais than in those with a predominance of R. dominica (Table 2). When examining each host 2.4. Data analysis separately, almost two times higher host reduction was observed for S. zeamais (ranging from 82 % to 95 %) than for R. dominica (ranging In the host selection experiment, the percentage of emergence from 2 % to 45 %). When three pairs of parasitoids were released, there reduction vs. the control treatment of the total host offered, of were significant differences in the emergence of S. zeamais among the S. zeamais, of R. dominica, the percentage of effective parasitism, and the ratios, with a higher reduction as the proportion of the weevil in the percentage of PIM between the different R. dominica:S. zeamais ratios mixed treatments decreased. When 12 pairs of parasitoids were released, offered for each parasitoid density tested were analyzed using a gener­ no significant differences in the emergence of S. zeamais among the ra­ alized linear model (GLM) with a quasi-binomial error distribution, to tios were detected, although a similar trend was observed. When three account for overdispersion. Post hoc comparisons were conducted using or 12 parasitoids were released, significant differences were found in the the Tukey correction for multiple comparisons. PIM was computed as emergence of R. dominica among the ratios. In contrast to S. zeamais, the the difference between the total percentage of host mortality and the reduction was smaller, as the proportion of R. dominica in the mixed host mortality that produced adult parasitoids (effective parasitism). treatments was lower (Table 2). It is worth noting that the parasitoid Differences in the percentage of emergence of each species in the densities tested did not differentially affect the emergence of each pest parasitoid and in the control treatment for each host-to-host ratio and species when offered alone (F1,18 = 1.75, P = 0.203) for S. zeamais and parasitoid density tested were compared with a GLM assuming a qua­ (F1,18 = 2.83, P = 0.110) for R. dominica). sibinomial error distribution. Differences in the percentage of reduction The preference of A. calandrae could also be observed by comparing in emergence compared with the control treatment, percentage of the percentage of hosts that emerged from each host species with those effective parasitism, and percentage of PIM between both parasitoid that emerged in the control treatment for each host ratio offered (Fig. 1). densities when S. zeamais and R. dominica were offered alone were In the case of no preference, the proportion of emergence of each host Table 1 Amount (g) of infested rice with S. zeamais or R. dominica included in each ratio tested, number (mean ± SEM) of emerged pest adults (S. zeamais + R. dominica), of emerged S. zeamais adults, and of emerged R. dominica adults in the treatment without parasitoids (Control) and the resulting host-to-host and parasitoid-to-host ratios tested in the two parasitoid densities. A. Calandrae S. zeamais/ R. dominica Total No. of adult No. S. zeamais No. R. dominica S. zeamais: R. Parasitoid/ host released (pairs) infested rice (g) hosts emerged adult emerged adults emerged dominica ratio tested ratio tested 3 261/0 158.4 ± 3.9 158.4 ± 3.9 0 100/0 1:53 196/31 190.2 ± 10.9 128.5 ± 3.9 61.7 ± 12.7 70/30 1:63 130/62 188.4 ± 8.5 82.2 ± 3.3 106.2 ± 6.9 44/56 1:63 65/93 220.4 ± 5.7 42.9 ± 2.4 177.5 ± 5.0 19/81 1:73 0/124 264.1 ± 5.5 0 264.1 ± 5.5 0/100 1:88 12 261/0 102.9 ± 3.8 102.9 ± 3.8 0 100/0 1:09 196/31 130.5 ± 2.5 77.0 ± 3.3 53.5 ± 2.9 59/41 1:11 130/62 160.7 ± 5.1 52.8 ± 3.0 107.9 ± 2.9 33/67 1:13 65/93 190.2 ± 4.8 27.6 ± 1.9 164.6 ± 4.3 15/85 1:16 0/124 225.6 ± 10.6 0 225.6 ± 10.6 0/100 1:19 3 L. del Arco et al. Biological Control 181 (2023) 105216 Table 2 Number (mean ± SEM) of emerged S. zeamais and R. dominica adults, and percentage (mean ± SEM) of emergence reduction in comparison with the control treatment without parasitoids; of total hosts (S. zeamais + R. dominica); of S. zeamais or of R. dominica, at each host-to-host ratio offered and for the two parasitoid densities tested. A. calandrae released S. zeamais: R. No. S. zeamais adult No. R. dominica adult % Total host % S. zeamais % R. dominica (pairs) dominica ratio emerged emerged reduction reduction reduction 3 100/0 19.8 0.0 87.5 ± 1.7 a 87.5 ± 1.7b 0 70/30 11.3 54.5 65.4 ± 4.4b 91.2 ± 1.8 ab 19.1 ± 4.7 bc 44/56 4.3 90.9 49.5 ± 2.0c 94.8 ± 0.6 a 15.4 ± 2.8c 19/81 2.8 127.7 40.8 ± 2.3c 93.5 ± 1.6 ab 28.1 ± 2.8b 0/100 0.0 145.6 44.9 ± 2.6c 0 44.9 ± 2.6 a F4,45 = 41.51 F3,36 = 4.71 F3,36 = 18.59 P < 0.001 P < 0.01 P < 0.001 12 100/0 17.6 0.0 82.9 ± 3.2 a 82.9 ± 3.2 0 59/41 8.5 60.3 47.3 ± 3.1b 89.0 ± 1.9 2.0 ± 1.4c 33/67 5.6 76.1 49.2 ± 2.6b 89.4 ± 1.5 29.5 ± 3.4b 15/85 4.1 97.2 46.7 ± 2.2b 85.1 ± 2.3 41.0 ± 2.3 a 0/100 0.0 130.0 42.4 ± 2.8b 0 42.4 ± 2.8 a F4,45 = 22.15 F3,36 = 1.84 F3,36 = 154.40 P < 0.001 P = 0.157 P < 0.001 Values in the same column for each parasitoid density followed by the same letter are not significantly different (Tukey test, P < 0.05). A. calandrae A. calandrae Fig. 1. Percentage (mean ± SEM) of S. zeamais and R. dominica adults emerged in the parasitoid (solid line) versus control treatment (dotted lines) at the different host-to-host ratios offered when: A) three female’s and B) 12 female’s parasitoids were released. * Denotes significant differences (P < 0.001) in pest emergence rates between parasitoid and control treatments. species at each tested ratio would be similar in both the parasitoid and Table 3 the control treatments. However, the emergence rates of S. zeamais and Number (mean ± SEM) and sex ratio of the A. calandrae emerged at each host-to- R. dominica were significantly different from those of the control treat­ host (S. zeamais: R. dominica) ratio offered and for both parasitoid densities ment in both parasitoid density releases. In all cases, the proportion of (three or 12 pairs) released. S. zeamais that emerged was lower than expected. For three pairs of A. calandrae S. zeamais: No. A. calandrae Sex A. calandrae released, F1,18 = 184.55, P < 0.001 (70Sz/30Rd), F1,18 = released (pairs) R. dominica ratio adults emerged ratio1 404.98, P < 0.001 (44Sz /56Rd) and F1,18 = 51.22, P < 0.001 (19Sz 3 100/0 149.3 ± 3.9 0.64 ± /81Rd), and for 12 pairs of A. calandrae, F1,18 = 83.64, P < 0.001 (59Sz 0.07* /41Rd), F1,18 = 320.70, P < 0.001 (33Sz /67Rd) and F1,18 = 304.46, P < 70/30 98.1 ± 3.8 0.76 ± 0.001 (15Sz /85Rd). Conversely, the proportion of R. dominica that 0.03* emerged was higher than expected. 44/56 77.5 ± 3.0 0.77 ± When S. zeamais was offered alone, almost three times more para­ 0.02* 19/81 45.0 ± 3.9 0.74 ± sitoids emerged from three pairs than from 12 pairs of parasitoids 0.02* released (Table 3) (F1,18 = 72.46, P < 0.001) and, consequently, a 0/100 24.2 ± 3.6 0.61 ± significantly lower percentage of parasitism was obtained in the high 0.05 parasitoid density. On the contrary, when R. dominica was offered alone, 12 100/0 52.2 ± 4.5 0.43 ± no differences in parasitism were observed between parasitoid-released 0.60 59/41 41.7 ± 2.9 0.51 ± densities (F1,18 = 1.06, P = 0.316). Also, when three female parasitoids 0.06 were released, a bias toward female production in most of the ratios 33/67 34.4 ± 2.7 0.53 ± tested was observed, while no bias was obtained in any ratio when 12 0.04 female parasitoids were released (Table 3). When parasitoid preferences 15/85 24.9 ± 1.9 0.40 ± 0.04 were evaluated as the percentage of effective parasitism (adult parasit­ 0/100 16.9 ± 2.1 0.42 ± oids emerged) at the different pest species ratios, it significantly 0.04 increased with the proportion of S. zeamais at both parasitoid-released 1 Ratio of females to the total number of emerged parasitoids. densities (F4,45 = 181.61, P < 0.001 and F4,45 = 63.60, P < 0.001 for * Denotes a significant deviation from 1:1 female-to-male (chi-squared test). three and 12 pairs of A. calandrae, respectively) (Fig. 2, darker bars). 4 L. del Arco et al. Biological Control 181 (2023) 105216 Fig. 2. Percentage (mean ± SEM) of host reduction in comparison with the control treatment and its components: effective parasitism and PIM at the different host- to-host ratios offered when three females and 12 females were released. S. zeamais: R. dominica ratios with the same lowercase letter for parasitism (darker bars) or for PIM (lighter bars) in each parasitoid density are not significantly different (Tukey test, P < 0.05). PIM significantly varied among the different host-to-host ratios in both replicates of the parasitoid + pests control treatment, we recovered parasitoid release rates (F4,45 = 25.32, P < 0.001 and F4,45 = 4.84, P < adults of S. zeamais and of R. dominica when offered both species and of 0.001 for three and 12 parasitoid females, respectively), increasing as the parasitoid. However, in the parasitoid + pests treatment, pest in­ the proportion of R. dominica increased (Fig. 2, lighter bars). When dividuals were recovered in all replicates, but parasitoids were not S. zeamais was offered alone, a significant increase in PIM was observed present in six of the thirty-two replicates conducted in the four rounds, when 12 pairs of parasitoids were released (F1,18 = 49.13, P < 0.001). three in the S. zeamais treatment and three in the S. zeamais + However, when R. dominica was offered alone, no significant differences R. dominica treatment. The resulting parasitoid-to-host ratio was 1:14 were detected between parasitoid-released rates (F1,18 = 0.04, P = and 1:12 for the S. zeamais and S. zeamais + R. dominica, respectively, 0.851). considering the four rounds together. In each treatment, S. zeamais reduction by the parasitoid was similar 3.2. Big bags experiment when offering only S. zeamais than when offering S. zeamais + R. dominica. This reduction was more than 90 % in the parasitoid + pests Temperature in the warehouse during the trial ranged from 17.8◦C to control treatment, being higher than in the parasitoid + pests treatment 30.36◦C, with a daily mean of 25.0 ± 0.5◦C; and humidity ranged from (43–53 % reduction) (treatment F1,62 = 33.81, P < 0.001; % S. zeamais 77.8 % to 52.3 %, with a daily mean of 68.6 ± 0.5 %. Temperatures F1,61 = 0.07, P = 0.790; treatment × % S. zeamais F1,60 = 3.67, P = inside the bags were very similar to those of the warehouse, ranging 0.060) (Fig. 4). The reduction of R. dominica emergence was similar in from 19.1◦C to 27.9 ◦C, with a mean of 24.5 ± 0.5 ◦C, and the relative the parasitoid + pests control than in the parasitoid + pests treatment humidity ranged from 74.6 % to 72.0 %, with a mean of 73.7 ± 0.1 %. (25.0 % and 25.9 %, respectively) (F1,22 = 0.02, P = 0.885), but lower The mean infestation of rice offered to the parasitoid (data in the than the reduction of S. zeamais emergence (Fig. 4). control treatment estimated as adult emergence) when only S. zeamais When examining host reduction in relation to the number of para­ was offered was 68.3 ± 3.4 larvae. In contrast, when S. zeamais + sitoids emerged and the hosts killed by other causes (PIM), parasitism R. dominica were offered, the mean infestation obtained was 83.4 ± 5.2 was similar in the control (parasitoid + pests) than in the treatment, and of S. zeamais larvae and 2.8 ± 2.2 of R. dominica larvae, corresponding to it was not affected by the presence of R. dominica (treatment F1,62 = a host-to-host ratio of 97 S. zeamais: 3 R. dominica (Fig. 3). In all 1.91, P = 0.167; % S. zeamais F1,62 = 0.07, P = 0.798; treatment × % Fig. 3. Number (mean ± SEM) of S. zeamais, of R. dominica, and of A. calandrae adults emerged in the pests control, in the parasitoid + pests control, and in the parasitoid + pests treatment, when rice was infested only with S. zeamais or with both pest species. 5 L. del Arco et al. Biological Control 181 (2023) 105216 Fig. 4. Percentage (mean ± SEM) of S. zeamais and R. dominica reduction in emergence in the control with parasitoid + pests and in the parasitoid + pests treatment, when rice was infested only with S. zeamais or with S. zeamais + R. dominica. Comparison of host ratios between parasitoid + pests control and parasitoid + pests treatment with different letters are significantly different (Tukey’s test, P < 0.05). S. zeamais F1,62 = 0.06, P = 0.801) (Fig. 5). However, PIM was signifi­ previous test, with a higher reproduction of R. dominica compared to cantly higher in the control group than in the treatment group (treat­ S. zeamais, as can be observed by the total number of emerged adults ment F1,62 = 42.27, P < 0.001; % S. zeamais F1,62 = 1.32, P = 0.250; from each species when they were not mixed (100 % R. dominica or 100 treatment × % S. zeamais F1,62 = 0.43, P = 0.513) (Fig. 5). Therefore, the % S. zeamais). This large variability in hosts’ emergence in both tests is fact that there was more host reduction in the parasitoid + pests control inherent to the methodology, since neither the sex nor the age of the than in the parasitoids + pests treatment was due to the difference in individuals used in the rice infestation were determined. Sex determi­ PIM values. nation of adults of both pest species is not an easy task, and if done, individuals are injured. Although the siliceous hull of paddy rice pro­ 4. Discussion tects it from attack by many storage pests (Nadeem et al., 2011), the mouthparts of R. dominica compared to those of S. zeamais may facilitate Results obtained in this study showed that the presence of access to the rice grain. It is also worth noting that the parasitoids were R. dominica at low levels did not diminish the efficacy of the parasitoid able to disperse and parasitize the infested grains despite the presence of A. calandrae in reducing S. zeamais emergence at either a small (labo­ the hull, as already reported in Riudavets et al. (2021). ratory) or large (big bags) scale. In the host selection laboratory To a large extent, the effectiveness of parasitoids in controlling a experiment, we offered different proportions of rice infested by each specific insect pest infestation depends on its host preference. Differ­ pest species. As mentioned, we conducted a previous test to obtain the ences in parasitism rates among host species are used as an estimate of approximate emergence ratios of both pest species in paddy rice and host preference and host suitability for the parasitoid (Timokhov and used these results for calculating the proportions of infested rice to use in Gokhman, 2003). The host selection laboratory experiment showed that the experiment. It turns out that in the laboratory experiment, the A. calandrae preferred to parasitize S. zeamais larvae in comparison with emergence of R. dominica was higher than the one observed in the R. dominica larvae. This result was supported by: a) a higher reduction in host emergence, almost double, occurred when S. zeamais larvae were offered alone to the parasitoid than when R. dominica larvae were offered alone; b) a reduction in S. zeamais emergence in the mixed host ratios was always above 80 %, while that of R. dominica was at the most around 41 %; c) when host emergence in the parasitoid treatment was compared to the control treatment, significantly fewer S. zeamais adults and more R. dominica adults emerged than expected; and d) A. calandrae preferentially reproduced on S. zeamais larvae than on R. dominica larvae. These results are in line with earlier studies of host suitability, which found higher parasitism rates of A. calandrae in S. zeamais and S. oryzae than in R. dominica (Solá et al., 2020; Belda and Riudavets, 2012; Ghimire and Phillips, 2007). Ghimire and Phillips (2008) also indicated that A. calandrae has an innate preference for S. zeamais over R. dominica. Nevertheless, it is worth noting that Timokhov and Gokh­ man (2003) showed intraspecific variation of different strains of A. calandrae in host preference between Curculionidae or Anobiidae families. The higher effective parasitism observed in maize weevil larvae at the lowest parasitoid density suggests that A. calandrae females tend to Fig. 5. Percentage (mean ± SEM) of host reduction in emergence and its components (effective parasitism in dark green bars and PIM in light green use hosts of larger sizes for oviposition and offspring development, bars) in the parasitoid + pests control and in the parasitoid + pests treatment, whereas the smallest ones are used for their nutrition (Choi et al., 2001). when rice was infested only with S. zeamais or with S. zeamais + R. dominica. This is because the host size on which the parasitoids develop affects the Comparison of host ratios between control and treatment with different letters size of the emergent parasitoid and thereby, also its reproductive are significantly different (Tukey’s test, P < 0.05). 6 L. del Arco et al. Biological Control 181 (2023) 105216 potential (Kim et al., 1995). Bruins et al. (1994) proposed that the S. zeamais would be reduced, and this reduction would increase as the relative host size may influence host selection because larger hosts are proportion of S. zeamais increased. Nevertheless, S. zeamais is the key perceived earlier than smaller ones. We deliberately offered S. zeamais pest in paddy rice, and the use of A. calandrae seems to be a feasible larvae of a younger instar than that preferred by A. calandrae to avoid alternative to the use of fumigants or contact insecticides when this size-host selection between the two pest species. However, because the commodity is stored in big bags. rice infestation period was one week and our pests are internally developing species, the assessment of the larval size offered was just an CRediT authorship contribution statement estimate, and we may have offered smaller R. dominica and larger S. zeamais than expected. Lidia del Arco: Methodology, Investigation, Data curation, Writing – In contrast with what could be expected, parasitoid offspring were review & editing. Jordi Riudavets: Conceptualization, Methodology, almost double in the low parasitoid density than in the high density Formal analysis, Writing – review & editing, Funding acquisition. José when S. zeamais was offered alone. This was probably caused by su­ Miguel Campos-Rivela: Methodology, Investigation, Data curation, perparasitism due to intraspecific competition between female parasit­ Writing – review & editing. María Teresa Martínez-Ferrer: Method­ oids, as documented in other studies (Choi et al., 2001; Lebreton et al., ology, Investigation, Data curation, Writing – review & editing. Nuria 2009; Solá et al., 2020). Anisopteromalus calandrae females are unable to Agustí: Conceptualization, Writing – review & editing. Cristina identify hosts parasitized by sibling females (Benkhellat et al., 2015). Castañé: Conceptualization, Methodology, Formal analysis, Writing – Therefore, increasing the parasitoid density increases the probability of review & editing, Funding acquisition. females laying eggs on already parasitized larvae, which could end up killing the host since only one egg could develop in one host larva (Lebreton et al., 2009). Therefore, quadrupling the number of parasit­ Declaration of Competing Interest oids introduced did not result in further host reduction in either of the two pest species. In the case of S. zeamais alone, it increased the PIM. In The authors declare that they have no known competing financial the case of R. dominica, as supported by other authors who studied host interests or personal relationships that could have appeared to influence suitability (Ghimire and Phillips, 2007; Solá et al., 2020), higher PIM the work reported in this paper. values were observed in both parasitoid densities because of host feeding or high mortality during parasitoid development. Acknowledgements The female-biased sex ratio observed when three female parasitoids were released could be explained by the regulation of the sex ratio of This research was funded by the Instituto Nacional de Investigación y pteromalid offspring, which preferentially deposit female eggs on large Tecnología Agraria y Alimentaria (INIA) from the Spanish Ministry of hosts and male eggs on small hosts due to its lower nutritional value Economy and Competitiveness (grant RTI2018- 095526-R-100) and by (Choi et al., 2001; Lebreton et al., 2009). Therefore, the fact that the sex ratio was balanced with R. dominica suggests that it is a less suitable host the CERCA Program (Centres de Recerca de Catalunya, Spain) of the Generalitat de Catalunya. L. del Arco is grateful for her scholarship, for parasitoid development, as previously reported by Solá et al. (2020), as well as when compared with S. oryze (Belda and Riudavets, 2012; funded by the grant PRE2019-087497 from the Spanish Ministry of Science, Innovation and Universities. Ghimire and Phillips 2008, 2007). No bias resulted when offered 12 female parasitoids; in the case of superparasitism, Choi et al. (2001) estimated a balanced progeny sex ratio. References In the big bags, the parasitoid was able to successfully disperse Adler, C., Athanassiou, C., Carvalho, M.O., Emekci, M., Gvozdenac, S., Hamel, D., through the paddy rice kernels and locate host larvae at 1 m depth in the Riudavets, J., Stejskal, V., Trdan, S., Trematerra, P., 2022. Changes in the bottom of the bags, as already shown in Riudavets et al. (2021). Pref­ distribution and pest risk of stored product insects in Europe due to global warming: erence of A. calandrae for S. zeamais larvae was also confirmed in this Need for pan-European pest monitoring and improved food-safety. J. Stored Prod. Res. 97, 101977 https://doi.org/10.1016/J.JSPR.2022.101977. experiment because a similar reduction of S. zeamais adult emergence Belda, C., Riudavets, J., 2012. Reproduction of the parasitoids Anisopteromalus calandrae was observed when it was offered alone than when it was offered (Howard) and Lariophagus distinguendus (Förster) on arenas containing a mixed together with a small proportion of R. dominica. Likewise, the reduction population of the coleopteran pests Sitophilus oryzae and Rhyzopertha dominica. in R. dominica adult emergence was lower, as occurred in those ratios J. Pest Sci. 85, 381–385. https://doi.org/10.1007/s10340-011-0401-2. Belda, C., Riudavets, J., 2010. Attraction of the parasitoid Anisopteromalus calandrae with less R. dominica in the host selection trial. The differences in (Howard) (Hymenoptera: Pteromalidae) to odors from grain and stored product S. zeamais reduction between the parasitoid + pests control and the pests in a Y-tube olfactometer. Biol. Control 54, 29–34. https://doi.org/10.1016/j. parasitoid + pests treatment were probably due to the larger dimensions biocontrol.2010.02.005. Benkhellat, O., Jaloux, B., Moali, A., Chevrier, C., Monge, J.P., 2015. Host discrimination of the container and the amount of rice to be explored. In the big bags, and egg laying in Anisopteromalus calandrae (Hymenoptera: Pteromalidae) the odors used by the parasitoids to find the host larvae (Belda and ectoparasitoid of Callosobruchus maculatus (Coleoptera: Chrysomelidae). J. Stored Riudavets, 2010; Ghimire and Phillips, 2008) were more diluted than in Prod. Res. 61, 48–53. https://doi.org/10.1016/J.JSPR.2015.02.005. Bruins, E.B.A.W., Wajnberg, E., Pak, G.A., 1994. Genetic variability in the reactive the parasitoid + pest control plastic container. Similar results in the distance in Trichogramma brassicae after automatic tracking of the walking path. decrease of parasitoid efficacy against S. zeamais were obtained in Entomol. Exp. Appl. 72, 297–303. https://doi.org/10.1111/j.1570-7458.1994. Riudavets et al. (2021) when comparing results obtained in PVC pipes tb01830.x. Carvalho, M.O., Faro, A., Subramanyam, B., 2013. Insect population distribution and (filled with 30 kg of paddy rice) and in big bags (500 kg of paddy rice). In density estimates in a large rice mill in Portugal – A pilot study. J. Stored Prod. Res. this experiment, effective parasitism was similar in the parasitoid + pest 52, 48–56. https://doi.org/10.1016/J.JSPR.2012.07.002. control than in the treatment, indicating that the priority of parasitoid Castañé, C., Riudavets, J., 2015. Sampling arthropod pests and natural enemies in stored barley. J. Stored Prod. Res. 64, 54–61. https://doi.org/10.1016/j.jspr.2015.08.005. females would be first to lay eggs and second to prey on the hosts. Chaisaeng, P., Chongrattanameteekul, W., Visarathanonth, P., Vajarasathiara, B., 2010. However, we would expect parasitism to be higher in the parasitoid + Laboratory studies on control of the maize weevil Sitophilus zeamais by the parasitoid pest control, as the larvae were more accessible. However, perhaps due Anisopteromalus calandrae. Sci. Asia 36, 6. https://doi.org/10.2306/scienceasia1513- to the high values of PIM obtained, superparasitism could also have 1874.2010.36.006. Choi, W.I., Yoon, T.J., Ryoo, M.I., 2001. Host-size-dependent feeding behaviour and occurred. progeny sex ratio of Anisopteromalus calandrae (Hym., Pteromalidae). J. Appl. In conclusion, the effectiveness of A. calandrae in controlling Entomol. 125, 71–77. https://doi.org/10.1111/j.1439-0418.2001.00496.x. S. zeamais in big bags would not be hampered by the presence of Corrêa, A.S., Pereira, E.J.G., Cordeiro, E.M.G., Braga, L.S., Guedes, R.N.C., 2011. Insecticide resistance, mixture potentiation and fitness in populations of the maize R. dominica at any studied host proportion. However, the effectiveness of weevil (Sitophilus zeamais). Crop Prot. 30, 1655–1666. https://doi.org/10.1016/j. the parasitoid in reducing R. dominica populations in the presence of cropro.2011.08.022. 7 L. del Arco et al. Biological Control 181 (2023) 105216 Edde, P.A., 2012. A review of the biology and control of Rhyzopertha dominica (F.) the Pascual-Villalobos, M.J., Carreres, R., Riudavets, J., Aguilar, M., Bozal, J.M., García, M. lesser grain borer. J. Stored Prod. Res. 48, 1–18. https://doi.org/10.1016/J. C., del Estal, P., 2006. Plagas del arroz almacenado y sus enemigos naturales en JSPR.2011.08.007. España. Bol. San. Veg. Plagas 32, 225–231. FAOSTAT, 2020. Food and Agriculture Organization Statistics. Available online: https:// Riudavets, J., 2018. Biological control of stored products pests. IOBC/WPRS Bull. 130, www.fao.org/faostat/en/#data/QCL (accessed on Sep 2022). 107–109. Ghimire, M.N., Phillips, T.W., 2008. Effects of prior experience on host selection and host Riudavets, J., Lucas, E., Pons, M.J., 2002. Insects and mites of stored products in the utilization by two populations of Anisopteromalus calandrae (Hymenoptera: northeast of Spain. IOBC/WPRS Bull. 25 (3), 41–44. Pteromalidae). Environ. Entomol. 37, 1300–1306. https://doi.org/10.1093/ee/ Riudavets, J., Martínez-Ferrer, M.T., Campos-Rivela, J.M., Agustí, N., Castañé, C., 2021. 37.5.1300. Releases of the parasitoid Anisopteromalus calandrae (Hymenoptera: Pteromalidae) Ghimire, M.N., Phillips, T.W., 2007. Suitability of five species of stored-product insects as can control Sitophilus zeamais (Coleoptera: Curculionidae) in big bags of paddy rice. hosts for development and reproduction of the parasitoid Anisopteromalus calandrae Biol. Control 163, 104752. https://doi.org/10.1016/j.biocontrol.2021.104752. (Hymenoptera: Pteromalidae). J. Econ. Entomol. 100, 1732–1739. https://doi.org/ RStudio Team (2022). RStudio: Integrated Development for R. RStudio, version 10.1093/jee/100.5.1732. 2022.07.2+576. Hagstrum, D.W., Milliken, G.A., 1988. Quantitative Analysis of Temperature, Moisture, Schöller, M., Flinn, P.W., 2000. Parasitoids and Predators, in: Alternatives to Pesticides in and Diet Factors Affecting Insect Development. Ann. Entomol. Soc. Am. 81, Stored-Product IPM. Springer US, Boston, MA, pp. 229–271. https://doi.org/ 539–546. https://doi.org/10.1093/aesa/81.4.539. 10.1007/978-1-4615-4353-4_9. Hong, Y.S., Ryoo, M.il., 1991. Effect of temperature on the functional and numerical Smith, L., 1993. Effect of Humidity on life history characteristics of Anisopteromalus responses of Lariophagus distinguendus (Hymenoptera: Pteromalidae) to various calandrae (Hymenoptera: Pteromalidae) parasitizing maize weevil (Coleoptera: densities of the host, Sitophilus oryzae (Coleoptera: Curculionidae). J. Econ. Entomol. Curculionidae) larvae in shelled corn. Environ. Entomol. 22, 618–624. https://doi. 84, 837–840. https://doi.org/10.1093/jee/84.3.837. org/10.1093/ee/22.3.618. Kavallieratos, N.G., Athanassiou, C.G., Arthur, F.H., 2015. Efficacy of deltamethrin Smith, L., Press, J.W., 1992. Functional response of Anisopteromalus calandrae against stored-product beetles at short exposure intervals or on a partially treated (Hymenoptera: Pteromalidae): influence of host numbers versus host density. rice mass. J. Econ. Entomol. 108, 1416–1421. https://doi.org/10.1093/jee/tov060. J. Entomol. Sci. 27, 375–382. https://doi.org/10.18474/0749-8004-27.4.375. Kim, J., Chun, Y.S., Ryoo, M.I., 1995. Adaptation of Anisopteromalus calandrae Solá, M., Riudavets, J., Castañé, C., 2020. Control of Rhyzopertha dominica and Sitophilus (Hymenoptera: Pteromalidae) to an alternative host, Callosobruchus chinensis zeamais in stored rice with different release rates of the larval parasitoid (Coleoptera: Bruchidae). Korean J. Appl. Entomol. 34, 278–286. Anisopteromalus calandrae. Entomol. Generalis 40 (3), 323–330. https://doi.org/ Lebreton, S., Darrouzet, E., Chevrier, C., 2009. Could hosts considered as low quality for 10.1127/entomologia/2020/0899. egg-laying be considered as high quality for host-feeding? J. Insect Physiol. 55, Steidle, J.L.M., Schöller, M., 2002. Fecundity and ability of the parasitoid Lariophagus 694–699. https://doi.org/10.1016/J.JINSPHYS.2009.04.017. distinguendus (Hymenoptera: Pteromalidae) to find larvae of the granary weevil Longstaff, B.C., 1981. Biology of the grain pest species of Sitophilus (Coleoptera: Sitophilus granarius (Coleoptera: Curculionidae) in bulk grain. J. Stored Prod. Res. 38, Curculionidae) A critical review. Protect. Ecol. 2, 83–103. 43–53. https://doi.org/10.1016/S0022-474X(00)00044-8. Mahal, N., Islam, W., Parween, S., Mondal, K.A.M.S.H., 2005. Effect of Anisopteromalus Timokhov, A.V., Gokhman, V.E., 2003. Host preferences of parasitic wasps of the calandrae (Hymenoptera: Pteromalidae) in controlling residual populations of Anisopteromalus calandrae species complex (Hymenoptera: Pteromalidae). Acta Soc. Rhyzopertha dominica (Coleoptera: Bostrichidae) in wheat stores. Int. J. Trop. Insect Zool. Bohemicae 67, 35–39. Sci. 25, 245–250. https://doi.org/10.1079/IJT200580. Trematerra, P., Paula, M.C.Z., Sciarretta, A., Lazzari, S.M.N., 2004. Spatio-temporal Menon, A., Flinn, P.W., Dover, B.A., 2002. Influence of temperature on the functional analysis of insect pests infesting a paddy rice storage facility. Neotrop. Entomol. 33, response of Anisopteromalus calandrae (Hymenoptera: Pteromalidae), a parasitoid of 469–479. https://doi.org/10.1590/S1519-566X2004000400012. Rhyzopertha dominica (Coleoptera: Bostrichidae). J. Stored Prod. Res. 38, 463–469. Trematerra, P., Sciarreta, A., Tamasi, E., 2000. Behavioural responses of Oryzaephilus https://doi.org/10.1016/S0022-474X(01)00050-9. surinamensis, Tribolium castaneum and Tribolium confusum to naturally and artificially Nam, Y., Yoon, E.Y., Kim, K.D., Ryoo, M.il, 2011. Mycoflora of stored rice in relation to damaged durum wheat kernels. Entomol. Exp. Appl. 94, 195–200. https://doi.org/ the biological control of Sitophilus oryzae by Anisopteromalus calandrae. J. Asia Pac. 10.1046/j.1570-7458.2000.00619.x. Entomol. 14, 399–403. https://doi.org/10.1016/J.ASPEN.2011.04.010. Wen, B., Brower, J.H., 1994. Suppression of maize weevil, Sitophilus zeamais (Coleoptera: Nadeem, S., Hamed, M., Shafique, M., 2011. Feeding preference and developmental Curculionidae), populations in drums of corn by single and multiple releases of period of some storage insect species in rice products. Pakistan J. Zool. 43, 79–83. Anisopteromalus calandrae (Hymenoptera: Pteromalidae). J. Kansas Entomol. Soc. 67, 331–339. https://www.jstor.org/stable/25085538. 8