Impact of the Parasitoids Anisopteromalus calandrae (Howard) and Lariophagus distinguendus (Förster) on Three Pests of Stored Rice

Simple Summary Various pest species cause significant damage to stored rice, and polyphagous parasitoids could offer an appropriate biological control strategy. Among the beneficial species colonising silos and warehouses of rice, the generalist pteromalid parasitoids Anisopteromalus calandrae and Lariophagus distinguendus have been found to be prevalent in samples collected in the northeast of Spain. In this study, the Spanish strains of both parasitoids have shown that they may play important roles in controlling three coleopteran pests of stored rice: Sitophilus oryzae, Rhyzopertha dominica and Lasioderma serricorne. They reduced the number of pests that emerged due to effective parasitism (parasitoids reaching the adult stage and emerging from the grain), combined with other sources of host mortality, such as parasitoid immature mortality and parasitoid host feeding. Both parasitoids preferentially attack S. oryzae, which is a key pest of stored rice, but they also play an important role in the control of R. dominica, a less important pest. Although both parasitoids could consume L. serricorne larvae, their efficacy is low, and more specific natural enemies should be evaluated for the control of this pest. Abstract This study evaluated the ability of pteromalid parasitoids Anisopteromalus calandrae and Lariophagus distinguendus reared on Sitophilus zeamais to control stored product coleopteran pests Sitophilus oryzae, Rhyzopertha dominica and Lasioderma serricorne. In trials of parasitoid treatment with A. calandrae, fewer pests (S. oryzae and R. dominica) emerged than in the control. Parasitoid reproduction was highest with S. oryzae as a host, followed by R. dominica and L. serricorne. In trials of parasitoid treatment with L. distinguendus, fewer pests (S. oryzae, R. dominica and L. serricorne) emerged than in the control treatment. Sitophilus oryzae was the host with the highest rate of parasitoid reproduction, although the greatest level of reduction was seen in R. dominica (i.e., host feeding levels were higher for this host species). For L. serricorne, no L. distinguendus progeny was produced. For both species, parasitoids with significantly longer bodies and tibiae emerged from S. oryzae. These results suggest that both parasitoids have potential for use as biocontrol agents for different coleopteran species that attack stored rice.


Introduction
Stored products are attacked by many species of pests. Today, they are controlled primarily using residual insecticides and fumigants. However, the development of pest resistance to many pesticides and the reduced availability of permitted active ingredients for chemical pest control has motivated the search for effective biological alternatives [1].
In Spain and other countries in the Mediterranean Basin, the most abundant and widely distributed primary pest species in stored rice are Sitophilus zeamais Motschulsky and Sitophilus oryzae (L.) (Coleoptera: Curculionidae) [2][3][4][5][6]. Rhyzopertha dominica (Fabricius) (Coleoptera: Bostrychidae) has also been identified as a serious pest of stored rice [7,8], the Insect colonies were reared in the laboratory under controlled conditions (25 ± 2 • C; 70 ± 10% RH). The colonies were started with adults collected from several rice silos in the Barcelona region of northeastern Spain and periodically refreshed. Laboratory colonies of S. oryzae, R. dominica and L. serricorne were reared on standard media (S. oryzae and R. dominica on brown rice and L. serricorne on coarsely ground brown rice) [33]. The parasitoids A. calandrae and L. distinguendus were reared on S. zeamais larvae that developed in brown rice.

Control of Three Coleopteran Species by A. calandrae and L. distinguendus
To obtain separated populations containing all larval stages of each of the three hosts, the rice was infested with 600 adults/800 gr and kept in 2 L ventilated plastic containers under controlled conditions (25 ± 2 • C and 70 ± 10% RH). Three hundred S. oryzae or R. dominica adults were added fortnightly. For L. serricorne, 150 adults were added weekly because this species has a shorter lifespan than the other two hosts.
After five weeks, the adult beetles were sieved out, and 100 g of rice from each host population was placed in 0.5 L ventilated glass jars. Seven female and three male A. calandrae (less than 7 days old) were introduced to each jar. In the case of L. distinguendus, Insects 2023, 14, 355 3 of 9 three females and one male (less than 7 days old) were introduced to each jar due to the low availability of individuals. Two Eppendorf tubes (Eppendorf Ibérica S.L.U., San Sebastián de los Reyes, Madrid, Spain) with sucrose solution and a cotton plug were added to all jars to feed the wasps. The parasitoids were kept in the jars for one week and then removed. During the following seven weeks, the numbers of host adults and parasitoids that emerged from the hosts were recorded weekly. The parasitoids were collected, sexed and stored in 70% alcohol. The length of the right hind tibia and the total body length from the top of the head to the end of the abdomen, excluding the ovipositor protrusion, of 20 females and 20 males from each species that emerged from each of the different hosts were measured using a stereomicroscope (Carl Zeiss Meditec Iberia S.A.U., Tres Cantos, Madrid, Spain) at 40× [34]. In the control treatments, the three hosts were placed in jars, but no parasitoids were released. Six replicate trials of each parasitoid and host combination were performed.
Host mortality measured as the difference between adult emergence in each parasitoid treatment compared to emergence in their respective control treatment, the number of parasitoid adult progeny (effective parasitism) and parasitoid size and sex ratio were used as response variables to assess host suitability [19]. The ratio of parasitoid-induced mortality (PIM) in the hosts (i.e., host mortality caused by the presence of the parasitoid that did not result in adult parasitoid emergence) [35] to the total number of hosts that emerged from the control treatment was also calculated.

Statistical Analyses
Before statistical analyses, the homogeneity of variances was tested using Bartlett's test, and data sets were transformed when necessary. The number of S. oryzae, R. dominica and L. serricorne that emerged in the trials with each parasitoid species was compared to the number that emerged in the control treatments by a two-way ANOVA. The size (tibia and body length) of adult parasitoids of each sex that emerged from the different hosts was also compared by a two-way ANOVA. The number of A. calandrae and L. distinguendus adults that developed in each host species, the percentage of reduction in host emergence, the percentage of effective parasitism and the PIM caused by the two parasitoids were analysed using a one-way analysis of variance (ANOVA). Post hoc comparisons were conducted using the Tukey correction for multiple comparisons. All statistical analyses were conducted using JMP 16.2.0 (SAS Institute Inc., Cary, NC, USA, 2020-2021).

Host Suitability Trials with A. calandrae
Although we infested brown rice with the same number of adults of each host, the number of progeny produced by each species was significantly different (F = 702.17; df = 5.35; p < 0.001): More S. oryzae adults emerged than R. dominica or L. serricorne adults ( Figure 1A). In the presence of parasitoids, significantly fewer S. oryzae and R. dominica adults emerged compared to the control treatment without parasitoids (S. oryzae: t = 721.76, df = 1, p < 0.001; R. dominica: t = 93.07, df = 1, p < 0.001). However, the numbers of L. serricorne that emerged from the control and parasitoid treatments did not differ significantly (t = 0.645; df = 1; p = 0.4281).
The number of A. calandrae that emerged from each host species tested differed significantly (F = 61.10; df = 2, 17; p < 0.001). The largest number of descendants emerged from S. oryzae; fewer than half as many emerged from R. dominica as from S. oryzae. The number of parasitoid offspring collected from L. serricorne was very low ( Figure 1B). The sex ratio of parasitoid offsprings was 52% females for S. oryzae, 45% females for R. dominica and 54% females for L. serricorne.
The parasitoid treatment reduced the number of pests that emerged due to effective parasitism (parasitoids reaching the adult stage and emerging from the grain) combined with other host mortality causes, such as parasitoid immature mortality and parasitoid host feeding (PIM) ( Table 1). Higher percentages of total mortality were observed for R. dominica and S. oryzae (62% and 47%, respectively); L. serricorne had the lowest total Insects 2023, 14, 355 4 of 9 mortality (10%). In R. dominica and S. oryzae, host mortality was due primarily to effective parasitism (57% and 36%, respectively), while in L. serricorne, mortality was due mainly to PIM (7.8%). There were significant differences in the percentage of total mortality and in the effective parasitism that the parasitoid produced in the three host species tested but not in the percentage of PIM.
The parasitoid treatment reduced the number of pests that emerged due to effective parasitism (parasitoids reaching the adult stage and emerging from the grain) combined with other host mortality causes, such as parasitoid immature mortality and parasitoid host feeding (PIM) ( Table 1). Higher percentages of total mortality were observed for R. dominica and S. oryzae (62% and 47%, respectively); L. serricorne had the lowest total mortality (10%). In R. dominica and S. oryzae, host mortality was due primarily to effective parasitism (57% and 36%, respectively), while in L. serricorne, mortality was due mainly to PIM (7.8%). There were significant differences in the percentage of total mortality and in the effective parasitism that the parasitoid produced in the three host species tested but not in the percentage of PIM. Male and female parasitoids developed in S. oryzae and in L. serricorne larvae were larger than those developed in R. dominica; that is, they had longer tibiae and longer bodies. All females had longer tibiae and bodies than the males (Table 1). Table 1. Percentage (mean ± SEM) of host mortality, parasitism and PIM (host feeding, unsuccessful parasitism) caused by A. calandrae when the larvae of S. oryzae, R. dominica or L. serricorne were offered. Mean (±SEM) hind tibia and body length of A. calandrae (males and females) emerging from each host species. N = 40 (20 males and 20 females). In each row, means followed by a different letter are significantly different (p < 0.05, Tukey). Body size measurements are given in mm.

Host Suitability Trials with L. distinguendus
Again, although we infested brown rice with the same number of adults of each host, the different species produced significantly different numbers of progeny (F = 167.39; df = 5, 35; p < 0.001). More S. oryzae than L. serricorne adults emerged, and even fewer R. dominica adults emerged ( Figure 2A). Nevertheless, in the treatment with L. distinguendus, significantly fewer adult S. oryzae, R. dominica or L. serricorne emerged than in the control treatment (S. oryzae: t = 46.97, df = 1, p < 0.001; R. dominica: t = 139.76, df = 1, p < 0.001; L. serricorne: t = 3.90, df = 1, p < 0.05).  parasitism (parasitoids reaching the adult stage and emerging from the grain) combined with other host mortality causes, such as parasitoid immature mortality and parasitoid host feeding (PIM) ( Table 1). Higher percentages of total mortality were observed for R. dominica and S. oryzae (62% and 47%, respectively); L. serricorne had the lowest total mortality (10%). In R. dominica and S. oryzae, host mortality was due primarily to effective parasitism (57% and 36%, respectively), while in L. serricorne, mortality was due mainly to PIM (7.8%). There were significant differences in the percentage of total mortality and in the effective parasitism that the parasitoid produced in the three host species tested but not in the percentage of PIM. Male and female parasitoids developed in S. oryzae and in L. serricorne larvae were larger than those developed in R. dominica; that is, they had longer tibiae and longer bodies. All females had longer tibiae and bodies than the males (Table 1). Table 1. Percentage (mean ± SEM) of host mortality, parasitism and PIM (host feeding, unsuccessful parasitism) caused by A. calandrae when the larvae of S. oryzae, R. dominica or L. serricorne were offered. Mean (±SEM) hind tibia and body length of A. calandrae (males and females) emerging from each host species. N = 40 (20 males and 20 females). In each row, means followed by a different letter are significantly different (p < 0.05, Tukey). Body size measurements are given in mm.
0.42 ± 0.003 a 0.34 ± 0.013 b 0.40 ± 0.013 a 47.12 5.119 <0.001 Hind tibia length parasitism (parasitoids reaching the adult stage and emerging from the grain) combined with other host mortality causes, such as parasitoid immature mortality and parasitoid host feeding (PIM) ( Table 1). Higher percentages of total mortality were observed for R. dominica and S. oryzae (62% and 47%, respectively); L. serricorne had the lowest total mortality (10%). In R. dominica and S. oryzae, host mortality was due primarily to effective parasitism (57% and 36%, respectively), while in L. serricorne, mortality was due mainly to PIM (7.8%). There were significant differences in the percentage of total mortality and in the effective parasitism that the parasitoid produced in the three host species tested but not in the percentage of PIM. Male and female parasitoids developed in S. oryzae and in L. serricorne larvae were larger than those developed in R. dominica; that is, they had longer tibiae and longer bodies. All females had longer tibiae and bodies than the males (Table 1).
0.52 ± 0.006 a 0.41 ± 0.011 b 0.50 ± 0.010 a Body length 54% females for L. serricorne. The parasitoid treatment reduced the number of pests that emerged due to effective parasitism (parasitoids reaching the adult stage and emerging from the grain) combined with other host mortality causes, such as parasitoid immature mortality and parasitoid host feeding (PIM) ( Table 1). Higher percentages of total mortality were observed for R. dominica and S. oryzae (62% and 47%, respectively); L. serricorne had the lowest total mortality (10%). In R. dominica and S. oryzae, host mortality was due primarily to effective parasitism (57% and 36%, respectively), while in L. serricorne, mortality was due mainly to PIM (7.8%). There were significant differences in the percentage of total mortality and in the effective parasitism that the parasitoid produced in the three host species tested but not in the percentage of PIM. Male and female parasitoids developed in S. oryzae and in L. serricorne larvae were larger than those developed in R. dominica; that is, they had longer tibiae and longer bodies. All females had longer tibiae and bodies than the males (Table 1).
The parasitoid treatment reduced the number of pests that emerged due to effective parasitism (parasitoids reaching the adult stage and emerging from the grain) combined with other host mortality causes, such as parasitoid immature mortality and parasitoid host feeding (PIM) ( Table 1). Higher percentages of total mortality were observed for R. dominica and S. oryzae (62% and 47%, respectively); L. serricorne had the lowest total mortality (10%). In R. dominica and S. oryzae, host mortality was due primarily to effective parasitism (57% and 36%, respectively), while in L. serricorne, mortality was due mainly to PIM (7.8%). There were significant differences in the percentage of total mortality and in the effective parasitism that the parasitoid produced in the three host species tested but not in the percentage of PIM. Male and female parasitoids developed in S. oryzae and in L. serricorne larvae were larger than those developed in R. dominica; that is, they had longer tibiae and longer bodies. All females had longer tibiae and bodies than the males (Table 1). Table 1. Percentage (mean ± SEM) of host mortality, parasitism and PIM (host feeding, unsuccessful parasitism) caused by A. calandrae when the larvae of S. oryzae, R. dominica or L. serricorne were offered. Mean (±SEM) hind tibia and body length of A. calandrae (males and females) emerging from each host species. N = 40 (20 males and 20 females). In each row, means followed by a different letter are significantly different (p < 0.05, Tukey). Body size measurements are given in mm.
2.12 ± 0.034 a 1.58 ± 0.032 b 2.08 ± 0.053 a Male and female parasitoids developed in S. oryzae and in L. serricorne larvae were larger than those developed in R. dominica; that is, they had longer tibiae and longer bodies. All females had longer tibiae and bodies than the males (Table 1).
This parasitoid was able to reproduce only in S. oryzae and R. dominica. No adult parasitoids emerged from the L. serricorne larvae ( Figure 2B). Significantly different total numbers of L. distinguendus emerged from these two pest species (F = 228.54; df = 2.17; p < 0.001); S. oryzae produced the most adult parasitoids. The sex ratio in both host species was 64% females for S. oryzae and 53% females for R. dominica.
There were significant differences in the total percentage of host mortality and in the percentage of mortality that was due to effective parasitism or PIM in each host species. The highest total mortality was observed in R. dominica (73%), which experienced approximately twice as high mortality as L. serricorne or S. oryzae (37% and 30%, respectively). Nearly half of the mortality in R. dominica was due to effective parasitism and half to PIM; similarly, in S. oryzae, approximately half of the total mortality was due to parasitism and half to PIM. However, in L. serricorne, mortality was due entirely to PIM since no adult parasitoids emerged ( Table 2). This parasitoid was able to reproduce only in S. oryzae and R. dominica. No adult parasitoids emerged from the L. serricorne larvae ( Figure 2B). Significantly different total numbers of L. distinguendus emerged from these two pest species (F = 228.54; df = 2, 17; p <0.001); S. oryzae produced the most adult parasitoids. The sex ratio in both host species was 64% females for S. oryzae and 53% females for R. dominica. There were significant differences in the total percentage of host mortality and in the percentage of mortality that was due to effective parasitism or PIM in each host species. The highest total mortality was observed in R. dominica (73%), which experienced approximately twice as high mortality as L. serricorne or S. oryzae (37% and 30%, respectively). Nearly half of the mortality in R. dominica was due to effective parasitism and half to PIM; similarly, in S. oryzae, approximately half of the total mortality was due to parasitism and half to PIM. However, in L. serricorne, mortality was due entirely to PIM since no adult parasitoids emerged (Table 2).
Male and female parasitoids developed in S. oryzae larvae were larger than those that developed in R. dominica; also, in both host species, female parasitoids had longer tibiae and body lengths than males ( Table 2). Table 2. Percentage (mean ± SEM) of host mortality, effective parasitism and PIM (host feeding, unsuccessful parasitism) caused by L. distinguendus when the larvae of S. oryzae, R. dominica or L. serricorne were offered. Mean (±SEM) hind tibia and body length of L. distinguendus (males and females) emerging from each host species. N = 40 (20 males and 20 females). In each row, means followed by a different letter are significantly different (p < 0.05, Tukey). Body size measurements are given in mm.

Discussion
In the experiment conducted with each parasitoid, the host S. oryzae produced many more progenies in the control treatment than did R. dominica (3.8 times as many with A. calandrae or 19 times as many with L. distinguendus) or L. serricorne (4.6 times as many with A. calandrae or 4.7 times as many with L. distinguendus) (Figures 1A and 2A). This occurred even though the rice was infested with a similar number of adults for each host tested,  Table 2. Percentage (mean ± SEM) of host mortality, effective parasitism and PIM (host feeding, unsuccessful parasitism) caused by L. distinguendus when the larvae of S. oryzae, R. dominica or L. serricorne were offered. Mean (±SEM) hind tibia and body length of L. distinguendus (males and females) emerging from each host species. N = 40 (20 males and 20 females). In each row, means followed by a different letter are significantly different (p < 0.05, Tukey). Body size measurements are given in mm. The parasitoid treatment reduced the number of pests that emerged due to effective parasitism (parasitoids reaching the adult stage and emerging from the grain) combined with other host mortality causes, such as parasitoid immature mortality and parasitoid host feeding (PIM) ( Table 1). Higher percentages of total mortality were observed for R. dominica and S. oryzae (62% and 47%, respectively); L. serricorne had the lowest total mortality (10%). In R. dominica and S. oryzae, host mortality was due primarily to effective parasitism (57% and 36%, respectively), while in L. serricorne, mortality was due mainly to PIM (7.8%). There were significant differences in the percentage of total mortality and in the effective parasitism that the parasitoid produced in the three host species tested but not in the percentage of PIM. Male and female parasitoids developed in S. oryzae and in L. serricorne larvae were larger than those developed in R. dominica; that is, they had longer tibiae and longer bodies. All females had longer tibiae and bodies than the males (Table 1). Table 1. Percentage (mean ± SEM) of host mortality, parasitism and PIM (host feeding, unsuccessful parasitism) caused by A. calandrae when the larvae of S. oryzae, R. dominica or L. serricorne were offered. Mean (±SEM) hind tibia and body length of A. calandrae (males and females) emerging from each host species. N = 40 (20 males and 20 females). In each row, means followed by a different letter are significantly different (p < 0.05, Tukey). Body size measurements are given in mm.

Host Suitability Trials with L. distinguendus
Again, although we infested brown rice with the same number of adults of each host, the different species produced significantly different numbers of progeny (F = 167.39; df = 5, 35; p < 0.001). More S. oryzae than L. serricorne adults emerged, and even fewer R. dominica adults emerged (Figure 2A). Nevertheless, in the treatment with L. distinguendus, significantly fewer adult S. oryzae, R. dominica or L. serricorne emerged than in the control treatment (S. oryzae: t = 46.97, df = 1, p < 0.001; R. dominica: t = 139.76, df = 1, p < 0.001; L. serricorne: t = 3.90, df = 1, p < 0.05). The parasitoid treatment reduced the number of pests that emerged due to effective parasitism (parasitoids reaching the adult stage and emerging from the grain) combined with other host mortality causes, such as parasitoid immature mortality and parasitoid host feeding (PIM) ( Table 1). Higher percentages of total mortality were observed for R. dominica and S. oryzae (62% and 47%, respectively); L. serricorne had the lowest total mortality (10%). In R. dominica and S. oryzae, host mortality was due primarily to effective parasitism (57% and 36%, respectively), while in L. serricorne, mortality was due mainly to PIM (7.8%). There were significant differences in the percentage of total mortality and in the effective parasitism that the parasitoid produced in the three host species tested but not in the percentage of PIM. Male and female parasitoids developed in S. oryzae and in L. serricorne larvae were larger than those developed in R. dominica; that is, they had longer tibiae and longer bodies. All females had longer tibiae and bodies than the males (Table 1).
0.47 ± 0.01 a 0.32 ± 0.01 b -Body length of parasitoid offsprings was 52% females for S. oryzae, 45% females for R. dominica and 54% females for L. serricorne. The parasitoid treatment reduced the number of pests that emerged due to effective parasitism (parasitoids reaching the adult stage and emerging from the grain) combined with other host mortality causes, such as parasitoid immature mortality and parasitoid host feeding (PIM) ( Table 1). Higher percentages of total mortality were observed for R. dominica and S. oryzae (62% and 47%, respectively); L. serricorne had the lowest total mortality (10%). In R. dominica and S. oryzae, host mortality was due primarily to effective parasitism (57% and 36%, respectively), while in L. serricorne, mortality was due mainly to PIM (7.8%). There were significant differences in the percentage of total mortality and in the effective parasitism that the parasitoid produced in the three host species tested but not in the percentage of PIM. Male and female parasitoids developed in S. oryzae and in L. serricorne larvae were larger than those developed in R. dominica; that is, they had longer tibiae and longer bodies. All females had longer tibiae and bodies than the males (Table 1).
The parasitoid treatment reduced the number of pests that emerged due to effective parasitism (parasitoids reaching the adult stage and emerging from the grain) combined with other host mortality causes, such as parasitoid immature mortality and parasitoid host feeding (PIM) ( Table 1). Higher percentages of total mortality were observed for R. dominica and S. oryzae (62% and 47%, respectively); L. serricorne had the lowest total mortality (10%). In R. dominica and S. oryzae, host mortality was due primarily to effective parasitism (57% and 36%, respectively), while in L. serricorne, mortality was due mainly to PIM (7.8%). There were significant differences in the percentage of total mortality and in the effective parasitism that the parasitoid produced in the three host species tested but not in the percentage of PIM. Male and female parasitoids developed in S. oryzae and in L. serricorne larvae were larger than those developed in R. dominica; that is, they had longer tibiae and longer bodies. All females had longer tibiae and bodies than the males (Table 1). Table 1. Percentage (mean ± SEM) of host mortality, parasitism and PIM (host feeding, unsuccessful parasitism) caused by A. calandrae when the larvae of S. oryzae, R. dominica or L. serricorne were offered. Mean (±SEM) hind tibia and body length of A. calandrae (males and females) emerging from each host species. N = 40 (20 males and 20 females). In each row, means followed by a different letter are significantly different (p < 0.05, Tukey). Body size measurements are given in mm.
1.81 ± 0.04 a 1.2 ± 0.05 b -Male and female parasitoids developed in S. oryzae larvae were larger than those that developed in R. dominica; also, in both host species, female parasitoids had longer tibiae and body lengths than males ( Table 2).

Discussion
In the experiment conducted with each parasitoid, the host S. oryzae produced many more progenies in the control treatment than did R. dominica (3.8 times as many with A. calandrae or 19 times as many with L. distinguendus) or L. serricorne (4.6 times as many with A. calandrae or 4.7 times as many with L. distinguendus) (Figures 1A and 2A). This occurred even though the rice was infested with a similar number of adults for each host tested, and we used the same amount of infested rice for each host. Therefore, S. oryzae reproduced most successfully under the conditions of our experiments: with brown rice and at 25 ± 2 • C and 70 ± 10% RH. The opposite results were observed when S. zeamais and R. dominica were reared in paddy rice under the same environmental conditions as those in the present experiments: R. dominica reproduced twice as much in paddy rice as did S. zeamais (1.45 vs. 0.69 adults/gr of rice, respectively) [36]. Since R. dominica is a grain borer, it can likely penetrate the rough skin of the paddy rice better than the smooth surface of the brown rice; the opposite is likely true of the Sitophilus species tested. These results indicate