Background

The beetle Dinoderus porcellus Lesne (1923) is the most abundant and economically important pest of stored yam chips in West Africa (Adedire & Gbaye, 2002; Loko et al., 2013; Osuji, 1980). It causes significant economic impact because yam chips represent the main form of long-term preservation of yam tubers (Omohimi et al., 2018), which are the staple food of million people in West Africa with an estimated food supply quantity of 93.84 kg/capita/year in 2018 (FAO, 2019). D. porcellus digs galleries by feeding on the yam chips and crushes the latter into powder within a few weeks causing significant qualitative and quantitative damage (John et al., 2020). Losses caused by D. porcellus were estimated at 27.4% of yam chips after 2 months of storage (Loko et al., 2019). The use of synthetic insecticides is the main control method used by farmers to reduce D. porcellus populations in stored yam chips (Loko et al., 2013). However, in addition to their negative impacts on the environment, treated yam chips with synthetic insecticides have led to numerous cases of food poisoning (Adedoyin et al., 2008; Adeleke, 2009). The use of biological control agent appears to be an eco-friendly alternative method to control this storage pest.

The polyphagous predator Alloeocranum biannulipes Montr. and Sign. (1861), which coexists in the stored yam chips with D. porcellus appears to be an important biological control agent against this pest (Loko et al., 2013). Indeed, studies on the functional response of A. biannulipes feeding on larvae and pupae of D. porcellus at different densities (Loko et al., 2017), and a test on the suppressive effect of this predator with respect to D. porcellus populations showed its potential as biological agent for controlling D. porcellus in stored yam chips (Loko et al., 2019). The reduviid A. biannulipes has also proved to be a good candidate for biological control of Prostephanus truncatus Horn in stored cassava chips (Loko et al., 2020) and other storage pest species such as Corcyra cephalonica Stainton, Tribolium confusum Duval, and Anagasta kuehniella Zeller (Awadallah et al., 1984). However, prior to the efficient use of A. biannulipes as a biological control agent against D. porcellus, there is a need for information regarding its reproductive numerical response. This knowledge is crucial because it not only allows to evaluated the ability of A. biannulipes to reduce the abundance of D. porcellus in stored yam chips, but also provides baseline data for a mathematical model to calculate the number of predators needed to regulate the pest population (Parween & Ahmad, 2015; Tangkawanit et al., 2018). In fact, the successful development and deployment of biological control programs against D. porcellus in stored yam chips using A. biannulipes will need to rely on an efficient mass rearing of this predator. However, for mass rearing of reduviids, pre-imaginal survival and development are major constraints and known as important factors in the life history of reduviids, which are strongly influence by the prey density (Grundy et al., 2000; Sahayaraj, 2002). Therefore, for the development of a cost-efficient rearing system for A. biannulipes, it is important to identify the most appropriate prey-predator ratio. The purpose of this study is to determine the numerical response of A. biannulipes to different levels of abundance of D. porcellus infesting yam chips, and assess the effect of prey density on the development and survival of A. biannulipes under laboratory conditions.

Methods

Pest rearing

Adults of D. porcellus were obtained from yam chips bought at the Dassa-Zoumé market (Collines department). The infested yam chips were gently broken with a hand mortar to collect D. porcellus adults. The pest was reared in cylindrical plastic boxes (15 cm diameter × 25 cm high) open at one end containing dried and sterilized yam chips. The open end of the plastic boxes was covered with a muslin cloth to allow adequate ventilation and prevent the escape of insects. The plastic boxes were stored in the laboratory conditions (T°: 25 ± 2 °C; RH: 75–80%; photoperiod 12:12 h (L/D)). Every two weeks the D. porcellus adult were removed from the rearing boxes, and the larvae were used for experiments.

Predator rearing

Adults of the predator A. biannulipes were collected in the Magoumi village (Collines department, latitude 8°10′26″ N and longitude 2°13′59″ E) in stored rice. The collected insects were reared in plastic boxes (15 cm diameter × 25 cm high) containing 500 g of yam chips infested by 100 D. porcellus adults of undetermined sex and age. The rearing boxes were maintained under laboratory conditions (T°: 25 ± 2 °C; RH: 75–80%; photoperiod 12:12 h (L/D). Two weeks later, 20 adults of undetermined sex and age of A. biannulipes were added to the plastic boxes containing the infested yam chips. Every two weeks the predators were removed from the rearing boxes, and the adult progenies were used for the experiments.

Prey consumption and oviposition of A. biannulipes in function of D. porcellus larvae density

24-hours-old adults of A. biannulipes were placed in pairs for mating in plastic boxes (6 cm diameter × 4 cm high), each containing 50 g of yam chips infested with D. porcellus larvae. Males and females of A. biannulipes were sexed based on their external genitalia (Loko et al., 2019). Ejection of spermatophore capsules by mated females confirmed successful copulation (Ambrose et al., 2009). After mating, the females were isolated individually in petri dishes (100 × 15 mm) and starved for 24 h. Subsequently, they were exposed to different densities (1, 2, 4, 6, and 8) of D. porcellus larvae (Loko et al., 2017) supplied with yam chips. After 24 h, the females were removed from the petri dishes and the number of eggs laid and the number of prey consumed were recorded. Observations were made for 30 consecutive days (using fresh prey larvae changed each day (Loko et al., 2019). Six replicates were done with five A. biannulipes females per replicate for each prey density. The duration of pre-oviposition and oviposition was also recorded.

The efficiency of conversion of ingested food (ECI) into egg biomass at different prey density treatments was estimated using equation:

$${\text{ECI}} = \frac{{{\text{Number }}\,{\text{of}}\,{\text{eggs}}\,{\text{laid }}}}{{{\text{Number}}\,{\text{of}}\,{\text{prey}}\,{\text{consumed }}}} \times 100$$

Development and survival of A. biannulipes in relation to prey density

Freshly laid A. biannulipes eggs were removed from the substrate with a fine camel hairbrush (Srikumar et al., 2014). Each egg was placed on dry cotton wool at the bottom of plastic boxes (6 cm diameter × 4 cm high) and kept under ambient laboratory conditions (T°: 25 ± 2 °C; RH: 75–80%; photoperiod 12:12 h (L/D). A cohort of 50 eggs of different predator was established for each prey density. After egg hatching, each A. biannulipes nymph was fed individually D. porcellus larvae freshly supplied with yam chips every day at densities of 1, 2, 4, 6 and 8. The post-embryonic development time of the five nymphal stages and the survival of each A. biannulipes developmental stages at various prey densities were recorded.

Data analysis

The relationship between A. biannulipes female fecundity and the efficiency of conversion of ingested food versus prey density was determined using regression analyses. Analysis of variance (ANOVA) was used to determine the difference between parameters evaluated at different prey densities, whereby the normality of the data was tested with the Shapiro–Wilk test (Shapiro & Wilk, 1965) using IBM SPSS statistical analysis software, version 25. The means and percentages that did not show homogeneity of variances were log-transformed and arcsine-transformed, respectively, before ANOVA. The Student Newman Keuls test was used to separate means significantly different (p ≤ 0.05).

Results

Oviposition of A. biannulipes females preying on different larval densities of D. porcellus

The number of D. porcellus larvae consumed influenced the duration of pre-oviposition and oviposition of A. biannulipes females (Table 1). Although the duration of preoviposition of A. biannulipes females fed with the highest prey density was shorter, it did not differ significantly (p ˃ 0.05) from that of females fed with the lowest prey density. However, the length of oviposition in A. biannulipes females fed with the highest prey density was significantly (p ˂ 0.05) higher from that of females fed with the lowest prey density. The increased density of D. porcellus larvae had a significant impact on the number of eggs laid by females of A. biannulipes (Table 1). At the highest larval densities of D. porcellus, the number of prey eaten and eggs laid by A. biannulipes females were significantly (p ≤ 0.05) higher than at the lowest densities (Fig. 1). The total average fecundity of A. biannulipes females peaked at 68.80 ± 5.75 eggs when offered the highest prey density (8). Daily oviposition of A. biannulipes females increased significantly (p ≤ 0.000) as a function of D. porcellus larvae consumed attaining a maximum daily oviposition rate of 5.40 ± 0.32 eggs for the highest prey density. There was a significant correlation between the daily number of eggs laid by an A. biannulipes female (y) and the number of D. porcellus larvae consumed (x), which can be expressed by the equation y = 0.158x2 – 0.4073x + 3.8151 (R2 = 0.571) (Fig. 2). The efficiency of food conversion into eggs (ECI) by A. biannulipes female decreased significantly (p < 0.001) with increasing prey density. However, the curve of ECI slightly increased at 8 densities of D. porcellus larvae (Fig. 3).

Table 1 Mean number of D. porcellus larvae consumed and reproduction of A. biannulipes females at increasing prey density (N = 30 replicates for each prey density)
Full size table
Fig. 1
figure 1

Numerical response in terms of mean number of eggs laid by A. biannulipes females at increasing D. porcellus density

Full size image
Fig. 2
figure 2

Daily oviposition of A. biannulipes females as a function of the number of D. porcellus larvae consumed per day

Full size image
Fig. 3
figure 3

Relationship between the efficiency of ingested feed conversion to egg biomass of A. biannulipes at increasing D. porcellus density

Full size image

Development and survival of nymph instars of A. biannulipes preying on different larval densities of D. porcellus

The percentage of A. biannulipes eggs hatched was significantly high and did not vary significantly (p ≥ 0.05) with the prey density offered (Table 2). All first instar nymph of A. biannulipes fed with a density of four or more larvae survived. This survival rate was significantly (p ≤ 0.05) higher than that of A. biannulipes first instar nymph fed with a single larva. With the exception of the third and fifth nymphal instars of A. biannulipes, survival rates of second and fourth nymphal instars fed with a single D. porcellus larva were significantly (p ≤ 0.05) lower than those of nymphs fed with higher densities of larvae (Table 2). Egg-to-adult survival rates were significantly (p ≤ 0.05) higher for nymphs fed with the higher densities of D. porcellus larvae compared to A. biannulipes nymphs fed with 1 or 2 larvae.

Table 2 Survival (percentage ± standard error) of A. biannulipes feeding on different densities of D. porcellus larvae
Full size table

Except for the fourth nymphal instar of A. bannulipes where the development time of nymphs fed with one larva was significantly different from nymphs fed with higher larval densities, no significant differences were observed in all other nymphal instars (Table 3). Furthermore, the development time from egg to adult showed no significant difference (p ≥ 0.05) according to the density of D. porcellus larvae.

Table 3 Development times (mean ± standard error; days) of immature stages of A. biannulipes feeding on different densities of D. porcellus larvae
Full size table

Discussion

The predator A. biannulipes, which showed Holling’s functional response type II in previous studies (Loko et al., 2017), presented a strong reproductive numerical response to increasing D. porcellus larvae density. As opposed to the case where A. biannulipes females were fed with larvae of P. truncatus (Loko et al., 2020), the density of D. porcellus larvae significantly influenced the fecundity of the predator. This difference could be explained by the quality of the prey, which is the most important factor affecting the prey-predator relationship (Aragón-Sánchez et al., 2018). Females of A. biannulipes preying on the highest D. porcellus larval densities presented the shortest preoviposition period and the highest oviposition period as observed in other reduviid predators, namely Rhynocoris marginatus (Fabricius) (Rajan & Sreelatha, 2019) and Rhynocoris fuscipes F. (Ambrose & Claver, 1997). The observed density-dependent numerical responses of A. biannulipes females suggest this predator could be an efficient biological control agent of D. porcellus. Predators displaying strong density-dependent reproduction rates are considered as promising biocontrol agents (Milonas et al., 2015), through their capacity to suppress pest populations before they reach damaging levels (Amiri-Jami & Sadeghi-Namaghi, 2014).

The high proportion of prey eaten and number of eggs laid by A. biannulipes females at the higher prey density suggest that this predator mainly allocates ingested food for egg production. Thereby, the decreasing efficiency of prey conversion into eggs with increasing prey density is consistent with a type II functional response of A. biannulipes. This is explained by the fact that female predators invest available resources in their own maintenance only after reaching their maximum egg-laying rate (Omkar, 2004). However, at the higher D. porcellus larvae density, the efficient food conversion value of A. biannulipes females slightly increased, indicating that A. biannulipes females may invest part of their energy in both oviposition and predation activities. Therefore, this implies that for the mass rearing of A. biannulipes the optimal prey density seems to be six D. porcellus larvae per day to maximize offspring production.

The number of prey items offered did not influence the hatching rate of A. biannulipes eggs, in contrast with previous observations where A. biannulipes adults were fed with P. truncatus larvae (Loko et al., 2020). As reported by Sahayaraj et al. (2016), offering prey species of diverse nutritional qualities to the same predator species has resulted in different egg hatching rates. However, A. biannulipes nymphs fed with high D. porcellus larval densities (2–8) displayed a significantly higher survival rate than those fed with a single larva, suggesting that the nutrients provided by one prey per day are not enough for optimal survival of A. biannulipes nymphs. Reduced survival at low prey density was also observed in other reduviids such as R. fuscipes (Ambrose & Claver, 1997), R. marginatus (Rajan & Sreelatha, 2019) and Cosmoclopius nigroannulatus Stal (Jahnke et al., 2002).

Our results showed that A. biannulipes was able to complete its development at each of the five prey densities and the prey density did not affect developmental time of this predator. This suggest that one D. porcellus larva per day is sufficient to ensure the requirements for the pre-imaginal development of A. biannulipes, albeit at lower survival rates. Knowing that prey consumption by a predator depends on several biotic and abiotic factors, further studies are required to evaluate the effect of temperature on the predatory potential of A. biannulipes and the biocontrol efficiencies of the mass-produced predator against D. porcellus in stored yam chips under farm conditions.

Conclusions

This study revealed that A. biannulipes females preying on D. porcellus exhibited a positive, density-dependent numerical response. Prey density impacted nymph survival but not their development. We can hence conclude that for the mass rearing of A. biannulipes the optimal prey density seems to be six D. porcellus larvae per day to maximize offspring production.