Introduction

Solenopsis invicta Buren (Hymenoptera: Formicidae; red imported fire ant) is an ant species native to South America that has spread around the world causing extensive ecological and economic damage in areas where it is considered an invasive species (Wetterer 2013). A major contributor to the spread of S. invicta is infested shipping container traffic (Wylie et al. 2020), and with the ever-increasing speed and volume of global containerized trade, further spread of this invasive species is expected. In Japan, S. invicta is yet to become an established species; however, instances of infested containerized cargo originating from China have been intercepted, with the first report being in 2017 (Ministry of the Environment, Japan 2021). Increases in the volume of containerized cargo arriving from China have increased the invasion pressure. Thus, improved measures to prevent S. invicta infestation of containerized cargo are urgently needed.

Synthetic repellents are currently the major means of preventing S. invicta infestation of containerized cargo; however, these repellents are associated with environmental and health hazards. In addition, shipping containers are used to transport a variety of goods, including foodstuffs. Therefore, a repellent that uses a naturally occurring compound would be useful. Allyl isothiocyanate (AITC) is a naturally occurring compound that is part of the defense against herbivores in plants such as wasabi (Eutrema japonicum [Miq.] Kiudz). AITC may pose a lower threat to the environment or to human health than the synthetic pesticides (Dhingra et al. 2004), yet strong inhibitor against various arthropods (Mansour et al. 2012; Worfel et al. 1997; Wu et al. 2014; Zanada and Ferris 2003). In addition, Du et al. (2020) have reported that S. invicta is particularly sensitive to AITC vapor, with a high lethal effect being observed even at low vapor concentrations (LC 50: 32.5 μg/L). Thus, AITC is a promising naturally occurring compound for development as a repellent against S. invicta.

A major challenge to using AITC as a repellent to prevent S. invicta infestation of containerized cargo is that AITC is highly volatile (Nerio et al. 2010) and the most widely used packaging material in shipping containers is corrugated cardboard; indeed, more than 80% of all products in the United States are shipped in corrugated cardboard boxes (Twede 2007). Cardboard is a porous cellulose-fiber-based material, which means it is a poor gas barrier (Aslannejad and Hassanizadeh 2017; Hochegger et al. 2021). Thus, a specialized delivery technology is needed for the practical application of AITC as a repellant against S. invicta infestation of containerized cargo.

In our previous field-based study, we examined the repellency of AITC using baited polypropylene centrifuge tubes lined with sheets of polyethylene (PE) resin containing microencapsulated AITC (hereafter microencapsulated AITC sheet, mAITC sheet) and found that the mAITC sheet completely prevented S. invicta from accessing the bait (Hashimoto et al. 2019). In a later study, we found that an important characteristic of mAITC sheet is that it releases AITC vapor gradually over an extended period of time (Hashimoto et al. 2020), such that it is able to repel S. invicta even under open conditions. However, polypropylene is a much better gas barrier compared with corrugated cardboard (Sekiyama et al. 1995); therefore, further studies are needed to examine the efficacy of mAITC sheet in cardboard boxes.

Here, building on our previous findings, we conducted another field-based study, this time to examine whether mAITC is able to prevent S. invicta from entering bait-loaded cardboard boxes. Under real-world shipping conditions, cardboard boxes are often covered with PE film to protect them from water damage, dust, or collapse (Macnish et al. 2012; Singh et al. 2014). Since PE film is a better gas barrier than corrugated cardboard (Sekiyama et al. 1995), the AITC vapor concentration could be expected to be higher in a covered cardboard box than in an uncovered carboard box. Therefore, we also examined the repellent effect of mAITC using boxes covered with PE film. Our data show that mAITC has high efficacy as a repellent against the entry of S. invicta into cardboard boxes, indicating that mAITC may be a safe, organic solution for protecting containerized cargo and preventing the further spread of S. invicta. To our knowledge, this is the first attempt to develop a naturally occurring repellent to prevent S. invicta infestation of containerized cardboard-box cargo.

Materials and methods

Study Site

Field experiments were conducted on October 30, 2019, at a construction site in Tucheng District, New Taipei City, Taiwan (23624°58′41.1″N, 121°26′49.6″E). The construction site was seriously infested with S. invicta. During the study, the field temperature and relative humidity remained around 27 °C and 71%, which are ideal conditions for active foraging by S. invicta workers (Lu 2014).

Test materials

Microencapsulated AITC sheets (size, 5 × 10 cm; thickness, 1 mm) containing 1 mg of AITC per 1 cm2 were obtained from PRD Co., Ltd. (Osaka, Japan) (PATENT No. JP5033232 B2 2012.9.26, WasaP”™). Square corrugated cardboard boxes (size, 6 × 6 × 6 cm; thickness, 5 mm; C5 rainier B-flute, no special surface treatment) and PE terephthalate covers (thickness, 50 μm) that protected the top and sides of the cardboard boxes leaving the bottom open were obtained from Moriya Industry Co., Ltd. (Osaka, Japan). In each cardboard box and PE cover, a hole (diameter, 5 mm) was made in one of the bottom corners to allow entry of S. invicta (Fig. 1, see Supplementary). Oily corn grits is an excellent bait for S. invicta (Lofgren et al. 1975; Williams et al. 2001); therefore, a piece of oil-fried snack made from corn grits (Housefoods’s “Tongari corn Yaki toumorokoshi” 1 g) was used as bait to attract S. invicta into the boxes.

Fig. 1
figure 1

a Corrugated cardboard boxes used to evaluate the effectiveness of allyl isothiocyanate microencapsulated in polyethylene (PE) resin (mAITC sheet) as a repellent against Solenopsis invicta. Each cardboard box (size, 6 × 6 × 6 cm; thickness, 5 mm) contained a piece of oil-fried corn grit bait and had a hole (diameter, 5 mm) cut at one of the bottom corners to allow the ants access to the bait inside the box. Four conditions were examined: with or without mAITC sheet (size, 5 × 10 cm; thickness, 1 mm) + with or without a PE cover (thickness, 50 μm). b Arrangement of the cardboard boxes around a S. invicta nest mound. One box for each test condition was placed around each nest mound

Full size image

Experimental procedure

Three experimental conditions and one control condition were examined. To examine the efficacy of mAITC sheet to repel S. invicta, box containing a single mAITC sheet was used (Condition 1). To examine the effect of covering the box with PE, box containing a single mAITC sheet and covered with PE film was used (Condition 2). To examine whether the PE film cover alone inhibited the entry of S. invicta, box without the mAITC sheet but covered with PE film was used (Condition 3). As control condition, box without the mAITC sheet and not covered with PE film was used. Ten boxes were used for each condition and each box contained a single piece of bait. The experiment was conducted using a total of ten individual S. invicta nest mounds. One box for each condition (i.e., four boxes) was placed close to the nest mound and the combination of the placement of the boxes around the nest was the same each time (Fig. 1).The number of S. invicta workers inside the box at 40 min after box placement was counted. The setting period of baited cardboard boxes was determined based on the results of an experiment conducted in Taiwan to define a monitoring method for S. invicta (Okinawa Prefecture 2020).

Data analysis

The nonparametric Kruskal–Wallis test followed by the Dwass–Steel–Critchlow–Fligner pairwise comparison were used to compare the number of S. invicta individuals in the boxes for each treatment. The statistical analysis was performed in JAMOVI 1.8.4.0 (The Jamovi Project 2021). For all comparisons, a p value less than 0.05 were considered to be statistically significant.

AITC release kinetics inside a cardboard box

Although AITC concentrations inside the cardboard boxes were not measured in this filed study, we measured the concentration of AITC in the boxes in the laboratory. AITC vapor concentrations were measured in cardboard boxes using a colorimetric gas detector tube (No. 149, Gastec Corp., Ayase, Japan; measurement range 5–200 ppm). The detector tube is a graduated glass tube filled with a chemical reagent that produces a color change when exposed to the target gas. The length of the color change along the graduated tube is proportional to the concentration of the gas being detected. The gas in the cardboard box was sampled via a gas-sampling pump (Ap-1, Komyo Rikagaku Kogyo KK, Japan) inserted in the hole made for the S. invicta to enter the box. The concentration of AITC was converted based on the conversion table in the instruction manual of the detector tube (https://nextteq.com/pdfs/Gastec_Tube_Instructions_149.pdf). Four replicates each were used for cardboard boxes with or without the PE cover. Gas samples were collected at 60 min after placing the AITC sheet in the box. This experiment was conducted at around 25 °C and 70% relative humidity in a laboratory owned by PRD Co., Ltd. and located in Osaka City, Japan.

Results and discussion

Microencapsulated AITC significantly reduced the number of S. invicta inside the cardboard boxes (Kruskal–Wallis test, chi-square = 28.7, df = 3; p < 0.001) (Fig. 2). At 40 min after the start of the experiment, in the boxes not containing an mAITC sheet, an average (±standard deviation) of 69.4 ± 25.9 individuals were counted in the boxes without the PE cover and 61.3 ± 25.2 individuals were counted in the boxes with the PE cover. No significant difference was detected between these two conditions (Dwass–Steel–Critchlow–Fligner pairwise comparison, p = 0.942), indicating that the PE cover did nothing to inhibit the entry of S. invicta to the boxes. In contrast, in the boxes containing mAITC sheets, an average of 26.5 ± 14.3 individuals were counted in the uncovered boxes and 2.1 ± 2.0 individuals were counted in the covered boxes. Statistical analysis revealed that the numbers of ants in the boxes with the mAITC sheets were significantly smaller than that in boxes without mAITC sheets, irrespective of whether a PE cover was used (Dwass–Steel–Critchlow–Fligner pairwise comparison: with vs. without mAITC in uncovered box, p = 0.01; with vs. without mAITC in covered box, p < 0.001). In addition, the number of ants in the boxes with the mAITC sheet and a PE cover was significantly smaller than that with the mAITC sheet but without a PE cover (Dwass–Steel–Critchlow–Fligner pairwise comparison: uncovered vs. covered box with mAITC, p < 0.001); indeed, the boxes with a mAITC sheet and a PE cover contained hardly any ants.

Fig. 2
figure 2

Number of Solenopsis invicta individuals per cardboard box at 40 min after the start of the experiment. Four conditions were examined: with or without allyl isothiocyanate microencapsulated in polyethylene (PE) resin (mAITC sheet) + with or without a PE cover (n = 10 boxes per treatment). In the boxplots, the boxes represent the interquartile range, the horizontal lines within the boxes represent medians, and the upper and lower whiskers represent the maximum and minimum values. Values shown above the boxplots are presented as mean ± standard error of the mean. Different letters above the boxplots indicate significantly different means (Kruskal–Wallis test followed by the Dwass–Steel–Critchlow–Fligner pairwise comparison; p value < 0.05)

Full size image

In the present experiment, we used a bait—oily corn grits— that is known to be effective at attracting S. invicta (Lofgren et al. 1975; Williams et al. 2001), and we cut a hole in the box and PE cover to facilitate entry of ants into the box. Despite this setup, which was expected to attract a large number of S. invicta, we found that the mAITC sheet significantly reduced the number of S. invicta in the box irrespective of whether a PE cover was used, although the number was very low when a PE cover was used. Cardboard is a porous material and poor gas barrier. In particular, low-molecular-weight compounds with high volatility can easily escape through cardboard (Sekiyama et al. 1995). AITC is one such highly volatile compound; however, the present data suggested that mAITC sheet was able to slow the release of AITC to the atmosphere outside of the box. When a PE cover was used, a higher concentration of AITC vapor may be retained in the box. These results indicate that mAITC sheet, especially in combination with a PE cover, is a promising repellent for protecting containerized cardboard-box cargo against S. invicta infestation. Although AITC concentrations inside the cardboard boxes were not measured in this filed study, we measured the concentration of AITC in the boxes in the laboratory at around 25 °C and 70% relative humidity using detector tubes (No. 149, Gastec Corp., Ayase, Japan). The average of four repetitions, each with and without cover, showed that the AITC concentration in the box after 60 min was 41.8 ± 6.9 ppm without cover and 85.3 ± 11.3 ppm with cover. The data may support that PE covers prevented AITC from leaving the carboard boxes.

Our present findings demonstrate that the sustained release property of mAITC is key to its practical use as a repellent against S. invicta infection of containerized cargo, but there are also several other advantages associated with using the microencapsulation approach. For example, the technology allows for the incorporation of AITC into PE materials, which can take various forms. This allows AITC to be easily incorporated as part of a cargo’s packaging material, and the dose of AITC can be easily controlled by adjusting the size or number of sheets used.

A limitation of the present study is that the cardboard boxes that were used were smaller than those generally used for containerized shipping. Therefore, further research is needed to verify the repellent efficacy of mAITC sheet in the types of boxes used in real-world shipping containers. Our group plans to evaluate the repellent efficacy of mAITC sheet on S. invicta using actual cardboard cargo and an actual shipping container in a field infested with S. invicta in Taiwan. However, since AITC has strong antifungal activity, mAITC is already in use as a fungicide in cardboard box cargo for container transport, demonstrating its effectiveness. In case mAITC is used as a fungicide in containerized cargo, approximately 25 cm2 of mAITC sheet is packed per 1 L of space volume. The minimum growth inhibition concentration of fungus by AITC gas has been shown to be about 40 μg/L (Iida et al. 2010). Therefore, it is presumable that the cardboard box packing the mAITC sheet can hold more than 40 μg/L of AITC gas during transportation in a marine container. Since the median lethal concentration of AITC gas to S. invicta has been shown to be 32.5 μg/L (Du et al. 2020), we consider that the same usage of mAITC sheet as the fungicide in containerized cargo may have potential for use as a repellent against S. invicta infestation, unless its preferred bait is in the cargo.