Fumigation Activity against Phosphine-Resistant Tribolium castaneum (Coleoptera: Tenebrionidae) Using Carbonyl Sulfide

Simple Summary Phosphine is one of the most widely used fumigants for the control of stored grain pests in quarantine. However, PH3 resistance to many stored pests has been reported. In this study, the fumigation activity of carbonyl sulfide was researched as an alternative fumigant to control PH3-resistant pests. In PH3 treatment, there was a clear difference in the fumigation activity of domestic strain Tribolium castaneum and resistance strain T. castaneum, but both the d- and r-strains of T. castaneum showed similar results in the carbonyl sulfide (COS) treatment. Furthermore, both PH3 and COS showed no significant difference in fumigation activity in Oryzaephilus surinamensis. Abstract Phosphine resistance is occurring among stored-grain pests worldwide. This study investigated the fumigation activity of phosphine (PH3) and carbonyl sulfide (COS) against domestic strain (d-strain) Tribolium castaneum, resistance strain (r-strain) T. castaneum and Oryzaephilus surinamensis. All developmental stages of the pests were exposed to two fumigants (PH3 and COS), and the fumigation activity according to the dose and exposure time was evaluated in a 12-L desiccator and 0.5 m3 fumigation chamber. The rice sorption rate and quality following exposure to thetwofumigants were evaluated. The mortality was 2.9% in r-strain T. castaneum, 49.5% in d-strain T. castaneum and 99.2% in O. surinamensis when 2 mg/L PH3 was used in a 12-L desiccator for 4 h. However, all pest developmental stages showed 100% mortality after 24 h of exposure in the 0.5 m3 fumigation chamber, except for the r-strain T. castaneum. A mortalityof 100% was observed in all of the r-strain T. castaneum developmental stages at an exposure time of 192 h. For COS applied at 40.23 mg/L and 50 g/m3 in the 12-L desiccator and the 0.5 m3 fumigation chamber, respectively, 100% mortality was observed across all developmental stages regardless of species and strain. The sorption of COS was 10% higher than that of PH3, but there was no significant difference in rice quality compared to that in the control. Therefore, this study suggests that COS can be used for controlling T. castaneum resistant to PH3.


Fumigation Experiments
The fumigation activities against all developmental stages of T. castaneum (d-and r-strain) and O. surinamensis were investigated using PH 3 and COS in a 12-L desiccator and 0.5 m 3 fumigation chamber modified according to Kyung et al. [36].
Twenty individuals of each developmental stage of the 3 kinds of stored-grain pests were placed in a Petri dish (5 cm i.d. × 1 cm) containing 0.05 g of food in a 12-L desiccator. PH 3 and COS were injected using 100 mL and 500 mL gastight syringes (Hamilton, NV, USA), respectively. The d-, and r-strain T. castaneum and O. surinamensis were exposed to PH 3 at various doses for 4 h, while the r-strain T. castaneum were exposed to PH 3 at 2 mg/L for 24 h to 250 h of exposure. The COS treatment was conducted with 24 h of exposure to all stored grain pests. The fumigation treatment was applied at 20 ± 1 • C, and the dishes were incubated at 26 ± 1 • C and 60-70% relative humidity with a 16:8 h (L:D) photoperiod after treatment. The controls were not treated with any fumigants.
Each Petri dish (5 cm) containing each developmental stage of the stored-grain pests was placed in a 0.5 m 3 fumigation chamber (125 × 50 × 80 cm) at a loading ratio of 50% with rice. The Petri dishes were placed in the top, middle and bottom parts of the chamber, respectively. The PH 3 and COS treatments were administered at 2 g/m 3 and 50 g/m 3 , respectively, with 24 h of exposure, except in the case of r-strain T. castaneum, which was exposed to PH 3 for 192 h.
All experiments were repeated 3 times, and mortality was evaluated 13 days after treatment for larvae, pupae and adults and 20 days after treatment for the egg stage.

Gas Concentration and Sorption Measurements
The gas concentration in each fumigation chamber was measured with a Tedlar gas sampling bag (1 L, SKC, Dorset, UK) using a gastight syringe (100 µL, Hamilton, NV, USA). The collected gas concentration was analyzed with gas chromatography (GC) (Agilent Technology 6890N, and 7890A, Santa Clara, CA, USA). The detector used for PH 3 was a nitrogen phosphorus detector (NPD), and that used for COS was a flame photometric detector (FPD). The GC conditions were as follows: The GC NPD injector temperature was 250 • C, the oven temperature was 240 • C, the detector temperature was 320 • C, and the column was an HP-5 (0.53 mm × 15 m, Agilent Technology, Santa Clara, CA, USA) operating in splitless mode. The FPD had an injector temperature of 200 • C, detector temperature of 200 • C and oven temperature of 200 • C, and the column was a DB-Sulfur SCD (0.53 mm × 70 m, Agilent Technology, Santa Clara, CA, USA).
The concentration and time (CT) values were determined by collecting gases at 30 min and 1, 4, 8, 12, and 24 h after treatment with PH 3 and COS, and gas was collected until 48, 96, and 192 h had passed for PH 3 in 0.5 m 3 fumigation chamber (AFHB/ACIA, 1989).
The sorption rates of PH 3 and COS were determined using a 12-L desiccator with 0% and 50% loading ratios (w/v) with rice. PH 3 and COS were applied at doses of 2 mg/L and 50 mg/L, respectively, at 20 • C for 24 h. The gas concentration for sorption was determined at 1, 4, 8, 12, 18 and 24 h after treatment and analyzed following the GC conditions described above. C/C 0 (%) values were calculated as the concentration at each time after treatment (C) divided by the initial concentration 30 min after treatment (C 0 ) and multiplied by 100. A 12-L desiccator without rice was used as the control.

Rice Quality Evaluation
The rice quality was evaluated to determine the effects of the two fumigants on rice. Either 2 g/m 3 PH 3 or 50 g/m 3 COS was applied in the 0.5 m 3 fumigation chamber at a loading ratio of 50% with rice for 48 h of exposure, and then rice quality was measured 24 h after treatment. The control was not treated with any fumigants.
Rice quality was analyzed by collecting 1 kg of rice from each treatment. As a component analysis, the protein and amylose content in 300 g of rice were measured using an Infratec TM 1241 grain analyzer (Foss, Hilleroed, Denmark). Measurement of the physical change ratio (%) was carried out with Insects 2020, 11, 750 4 of 12 1000 grains of head rice, broken rice, chalky kernels, and damaged rice using a Single-Grain Rice Inspector, model RN-600 (Kett Electric Laboratory, Tokyo, Japan). All measurements were conducted at Chungcheongbuk-do Agricultural Research and Extension Services (Cheongju, Korea).

Statistical Analysis
The fumigation activities of PH 3 and COS against T. castaneum (d-strain and r-strain) and O. surinamensis were analyzed statistically according to the doses and locations, and the exposure times of the r-strain T. castaneum were also compared using Tukey's test [37]. The differences in rice quality between the fumigant-treated and untreated rice were analyzed using a t-test [37].

Fumigation Activities of PH 3 and COS in a 12-LDesiccator
The fumigation activities were investigated after the exposure of d-strain T. castaneum, r-strain T. castaneum and O. surinamensis to PH 3 for 4 h ( Figure 1). Exposure for 4 h was not sufficient to control 100% of d-or r-strain T. castaneum. Specifically, the r-strain T. castaneum showed very low susceptibility across all developmental stages. However, the r-strain did not show 100% mortality after 24 h of exposure (data not shown), but the d-strain T. castaneum exhibited 100% fumigant activity across all developmental stages under these conditions ( Figure 2). In particular, the egg stage showed the lowest mortality of 2.9% after 24 h of exposure to 2 mg/L, and all r-strain T. castaneum developmental stages showed 100% mortality only when exposed to this amount for 192 h. The egg and pupal stages did not show 100% mortality in the 1 mg/L PH 3 treatment, but all developmental stages of O. surinamensis showed >98% mortality (99.2% in egg, 98.3% in pupa, and 100% in larvae and adult stage, respectively) when exposed to 2 mg/L of the PH 3 treatment for 4 h. The fumigation activities of PH3 and COS against T. castaneum (d-strain and r-strain) and O. surinamensis were analyzed statistically according to the doses and locations, and the exposure times of the r-strain T. castaneum were also compared using Tukey's test [37]. The differences in rice quality between the fumigant-treated and untreated rice were analyzed using a t-test [37].

Fumigation Activities of PH3 and COS in a 12-LDesiccator
The fumigation activities were investigated after the exposure of d-strain T. castaneum, r-strain T. castaneum and O. surinamensis to PH3 for 4 h ( Figure 1). Exposure for 4 h was not sufficient to control 100% of d-or r-strain T. castaneum. Specifically, the r-strain T. castaneum showed very low susceptibility across all developmental stages. However, the r-strain did not show 100% mortality after 24 h of exposure (data not shown), but the d-strain T. castaneum exhibited 100% fumigant activity across all developmental stages under these conditions ( Figure 2). In particular, the egg stage showed the lowest mortality of 2.9% after 24 h of exposure to 2 mg/L, and all r-strain T. castaneum developmental stages showed 100% mortality only when exposed to this amount for 192 h. The egg and pupal stages did not show 100% mortality in the 1 mg/L PH3 treatment, but all developmental stages of O. surinamensis showed >98% mortality (99.2% in egg, 98.3% in pupa, and 100% in larvae and adult stage, respectively) when exposed to 2 mg/L of the PH3 treatment for 4 h.  After 24 h of exposure to various doses of fumigation with COS, mortality was evaluated in d-strain T. castaneum, r-strain T. castaneum and O. surinamensis ( Figure 3). Treatment with COS had similar results in both d-strain and r-strain T. castaneum. The d-strain T. castaneum adults and larvae showed 100% mortalities at 14 mg/L and 25 mg/L, respectively, and the same percentage of activity was observed in r-strain T. castaneum at 12 mg/L and 17 mg/L. The eggs and pupae of both strains showed 100% mortality only when 30 mg/L of COS was used. Although the fumigation activity detected in COS-treated eggs and pupae was lower than that in other developmental stages, fumigant activity was observed regardless of the strain. O. surinamensis showed 100% mortality in adult, larva and pupa stages at 20 mg/L, but 100% mortality was observed in the egg stage at 40 mg/L.  After 24 h of exposure to various doses of fumigation with COS, mortality was evaluated in dstrain T. castaneum, r-strain T. castaneum and O. surinamensis (Figure 3). Treatment with COS had similar results in both d-strain and r-strain T.castaneum. The d-strain T. castaneumadults and larvae showed 100% mortalities at 14 mg/L and 25 mg/L, respectively, and the same percentage of activity was observed in r-strain T. castaneum at 12 mg/L and 17 mg/L. The eggs and pupae of both strains showed 100% mortality only when 30 mg/L of COS was used. Although the fumigation activity detected in COS-treated eggs and pupae was lower than that in other developmental stages, fumigant activity was observed regardless of the strain. O. surinamensis showed 100% mortality in adult, larva and pupa stages at 20 mg/L, but 100% mortality was observed in the egg stage at 40 mg/L.

Scale-UpFumigation Experiment with PH3 and COS in a 0.5 m 3 Fumigation Chamber
In the 0.5 m 3 fumigation chamber experiment, when d-strain T. castaneum and O. surinamensis were treated with 2 g/m 3 PH3 for 24 h at a 50% loading ratio with rice, all developmental stages showed 100% mortality regardless of the location (Table 1). However, the r-strain T. castaneum larvae showed less than 70% mortality only following 24-h exposure to PH3, and the other developmental stages showed very low fumigation activities (<29%). When the time of exposure to PH3 was increased to 192 h, 100% mortality was observed across all developmental stages of r-strain T. castaneum at all locations.

Scale-Up Fumigation Experiment with PH 3 and COS in a 0.5 m 3 Fumigation Chamber
In the 0.5 m 3 fumigation chamber experiment, when d-strain T. castaneum and O. surinamensis were treated with 2 g/m 3 PH 3 for 24 h at a 50% loading ratio with rice, all developmental stages showed 100% mortality regardless of the location (Table 1). However, the r-strain T. castaneum larvae showed less than 70% mortality only following 24-h exposure to PH 3 , and the other developmental stages showed very low fumigation activities (<29%). When the time of exposure to PH 3 was increased to 192 h, 100% mortality was observed across all developmental stages of r-strain T. castaneum at all locations.
The fumigation activity of COS was investigated after 24 h of exposure in the 0.5 m 3 fumigation chamber for d-strain T. castaneum, r-strain T. castaneum and O. surinamensis ( Table 2). The d-strain T. castaneum, r-strain T. castaneum and O. surinamensis exposed to 50 g/m 3 of COS for 24 h showed 100% mortality regardless of location.

Rice Quality and Sorption of the Two Fumigants
Six rice quality experiments under exposure to PH 3 and COS were conducted in a 12-L desiccator ( Table 3). The changes in protein content and amylose, which are components of rice, did not show Insects 2020, 11, 750 8 of 12 any statistically significant differences compared to those in the control for both fumigants. The two fumigants did not affect the rice in terms of the four kinds of physical change rates. Table 3. Quality of rice exposed to 2 g/m 3 PH 3 and 50 g/m 3 COS at a 50% loading ratio following 48 h of exposure in a 0.5 m 3 fumigation chamber. The sorption rates of rice following exposure to PH 3 and COS for up to 24 h were analyzed ( Figure 4). After 5 h of treatment of a 50% loading ratio of rice with 2 mg/L PH 3 , the concentration began to decrease to 96%, and the concentration was 91% after 24 h of treatment. When the 50% loading ratio of rice was treated with 50 mg/L COS, the concentration began to decrease to 97% after 1 h and decreased to 82% after 24 h. However, PH 3 and COS did not show a significant reduction in concentration (97% and 98%, respectively), even after 24 h of treatment, in the control without rice.

Quality Criteria
Insects 2020, 11, x FOR PEER REVIEW 8 of 12

Rice Quality and Sorption of the Two Fumigants
Six rice quality experiments under exposure to PH3 and COS were conducted in a 12-L desiccator ( Table 3). The changes in protein content and amylose, which are components of rice, did not show any statistically significant differences compared to those in the control for both fumigants. The two fumigants did not affect the rice in terms of the four kinds of physical change rates. Table 3. Quality of rice exposed to 2 g/m 3 PH3 and 50 g/m 3 COS at a 50% loading ratio following 48 h of exposure in a 0.5 m 3 fumigation chamber. The sorption rates of rice following exposure to PH3 and COS for up to 24 h were analyzed ( Figure 4). After 5 h of treatment of a 50% loading ratio of rice with 2 mg/L PH3, the concentration began to decrease to 96%, and the concentration was 91% after 24 h of treatment. When the 50% loading ratio of rice was treated with 50 mg/L COS, the concentration began to decrease to 97% after 1 h and decreased to 82% after 24 h. However, PH3 and COS did not show a significant reduction in concentration (97% and 98%, respectively), even after 24 h of treatment, in the control without rice.

Discussion
This study evaluated the activity of PH 3 and COS against two stored-product pests, T. castaneum (d-strain and r-strain) and O. surinamensis. PH 3 shows fumigation activity against various grain pests, T. castaneum, Rhyzopertha dominica, and O. surinamensis, but an increase in the resistant strain of these insects has been reported [11,[38][39][40]. This experiment was performed to investigate an alternative fumigant that is effective against PH 3 -resistant pests, especially T. castaneum. This is because T. castaneum has a very high frequency of strong resistance to PH 3 , especially in the United States, Southern India and Turkey [11,29,30]. Although it has been reported that sulfuryl fluoride (SF) has the effect of controlling PH 3 -resistant pests, it is vulnerable to low temperatures and has a disadvantage in that it is not easy to use because there is a change in activity according to sealing techniques during treatment [41]. COS is also toxic to stored-grain pests such as T. castaneum, R. dominica, O. surinamensis, Callosobruchus chinensis, Sitophilus zeamais, Lasioderma serricorne, Cryptolestes pusillus, Callosobruchus maculatus, Trogoderma variable, and Tribolium confusum [14,42]. This experiment also showed fumigation activity against T. castaneum and O. surinamensis. Ther-strain T. castaneum treated with PH 3 showed an increasing fumigation effect only when the exposure time was increased because the fumigation effect was much lower than that observed ford-strain T. castaneum, but both the d-and r-strains of T. castaneum showed similar results in the COS treatment. It was also observed that the COS showed higher tolerance to O. surinamensis than in d-and r-strains of T. castaneum at the egg stage. This is because of the difference in the target site where PH 3 and COS exhibit toxicity; PH 3 is toxic because of the formation of reactive oxygen radicals involved in the electron transport chain, and COS is toxic because of the metabolite action of hydrogen sulfide [43][44][45]. The fumigation activity of both fumigants differed according to the developmental stage exposed. A similar effect has also been observed in association with other pests and fumigants. Under 4 h of exposure, Frankliniella occidentalis eggs showed higher tolerance to the fumigation activity of >16 mg·h/L PH 3 at 5 • C and >84.4 mg·h/L ethyl formate (EF) at 5 • C than other developmental stages [36]. In addition, Phthorimaea operculella larvae and eggs showed the highest tolerance during PH 3 treatment, and Carposina niponensis and Rhynchophorus ferrugineus showed the highest tolerance in the egg stage [46][47][48]. In this study, a difference in fumigation activity to each developmental stage was also observed, but the fumigation effect decreased in the egg and pupa stage in d-and r-strain T. castaneum and O. surinamensis for both fumigants. Even in the scale-up experiment (0.5 m 3 ) in which the dose was set based on the results of the 12-L desiccator experiment, the PH 3 r-strains of T. castaneum showed 100% mortality for 192-h exposure, but all three kinds of stored pests could be controlled 100% at the same dose in the COS treatment. Thus, even when scaling-up at the field level, it is thought that fumigation activity against pests will be observed at the COS dose (50 g/m 3 ) used inthis study.
In terms of the adsorption rate, that of COS was higher than that of PH 3 , but it did not affect the fumigation activity in the 24-h exposure treatment. The use of COS did not have obvious side effects, such as water adsorption by rice, the expansion of cooking rice or dry substances in the rice cooking water, but the flavor changed when a COS concentration of more than 100 g/m 3 was used [42]. Furthermore, in the rice quality evaluation in this study, both fumigants did not result in any significant difference in quality compared to that ofthe control, even in the treatment of 50 g/m 3 COS.
These results indicate that PH 3 is a very effective fumigant for the control of stored-grain pests and that COS could be used an alternative for the control of the T. castaneum strain resistant to PH 3 and O. surinamensis.

Conclusions
In our evaluation of the fumigation activity of PH 3 against d-and r-strain T. castaneum and O. surinamensis, it was found that r-strain T. castaneum has very high resistance to PH 3 . However, in the COS treatment, r-strain T. castaneum showed similar mortality to that of d-strain T. castaneum. Ifr-strain T. castaneum appears during the control of stored-grain pests, it is worth considering the use of COS as an alternative to PH 3 .