https://orcid.org/0000-0003-4371-3388
Abstract
Agricultural crops around the world are attacked by approximately 3,000–10,000 species of pest insect. There is increasing interest in resolving this problem using environmentally friendly approaches. Wolbachia (Hertig), an insect endosymbiont, can modulate host reproduction and offspring sex through cytoplasmic incompatibility (CI). The incompatible insect technique (IIT) based on CI-Wolbachia is a promising biological control method. Previous studies have reported an association between CI and Wolbachia density, which may involve a quorum sensing (QS) mechanism. In this study, we investigated the effect of manipulating QS in Wolbachia using several chemicals including 3O-C12-HSL; C2HSL; spermidine (QS inducers), 4-phenylbutanoyl; and 4-NPO (QS inhibitors) on American serpentine leafminer (Liriomyza trifolii [Burgess]), an agricultural pest. The results showed that inducing QS with 3O-C12-HSL decreased the proportion of hatched eggs and increased Wolbachia density, whereas QS inhibition with 4-phenylbutanoyl had the opposite effects. Thus, manipulating QS in Wolbachia can alter cell density and the proportion of hatched eggs in the host L. trifolii, thereby reducing the number of insect progeny. These findings provide evidence supporting the potential efficacy of the IIT based on CI-Wolbachia for the environmentally friendly control of insect pest populations.
Around 3,000–10,000 species of insects are agricultural pests (Pimentel 2009, Dhaliwal et al. 2015, Lara and Saldivar 2015). As this is a major problem in agriculture, there is increasing interest in control methods using insect endosymbionts including Wolbachia for pests control as an alternative environmentally and economically sustainable solution (Zabalou et al. 2004, Kandel et al. 2021). Wolbachia is a ubiquitous endosymbiotic bacterium found in Nematoda and Arthropoda (Tagami et al. 2006a, Werren et al. 2008). Approximately 60% of species in both taxa are infected with Wolbachia (Hilgenboecker et al. 2008). Wolbachia has several adaptive strategies to manipulate reproduction in the host including cytoplasmic incompatibility (CI), male-killing, parthenogenesis, and feminization (Werren 1997). CI, which is among the most widespread strategies, involves modifying host male sperm, which yields fertilized eggs that do not hatch due to embryo mortality (Zabalou et al. 2004, Dong et al. 2006). CI occurs when a male insect infected with Wolbachia mates with a female that is uninfected or infected with a different Wolbachia strain (Hoffmann et al. 1986, Engelstädter and Telschow 2009).
The induction of CI with Wolbachia (CI-Wolbachia) can be constituted as an incompatible insect technique (IIT). The principle of this technique is similar to the sterile insect technique (SIT). The SIT method reduces insect pest populations through sterilization of adult males via irradiation (Knipling 1955, Nikolouli et al. 2018), while the IIT produces embryonic mortality in offspring, when infected males mate with uninfected females or males and females with different Wolbachia strain copulate. (Zabalou et al. 2004, Nikolouli et al. 2018). The IIT using Wolbachia has been applied to several insect pests including, Culex pipiens (Linnaeus) (Diptera: Culicidae) (Atyame et al. 2015), Aedes polynesiensis (Marks) (Diptera: Culicidae) (O’Connor et al. 2012), and Aedes albopictus (Skuse) (Diptera: Culicidae) (Zhang et al. 2015, Moretti et al. 2018), and could potentially be used in other insect pests.
Liriomyza trifolii is a major insect pest around the world. While the first cases were recorded in the Nearctic and Neotropical regions, an outbreak was later reported in the U.S. state of Florida (Parrella and Keil 1984, Minkenberg 1988). More recently, the geographic distribution of L. trifolii has expanded to Asia (Tagami et al. 2006b, Abe 2017), causing failure of crops such as tomato, eggplant (Solanaceae), melon, cucumber (Cucurbitaceae), kidney bean (Fabaceae), celery (Apiaceae), onion (Amaryllidaceae), and pak choi (Brassicaceae) as well as ornamental flowers in the Asteraceae family (Abe 2017). Insecticides have been used to reduce Liriomyza populations, which have responded by developing resistance (Ferguson 2004, Weintraub et al. 2017).
Liriomyza trifolii is infected only with Wolbachia and not with Cardinium (Zchori-Fein) (Cytophagales : Amoebophilaceae) (Tagami et al. 2006b), proving that the CI effect in L. trifolii is only because of Wolbachia. Wolbachia colonizing L. trifolii has an excellent vertical transmission and induces CI, making it suitable as a possible insect pest control agent (Tagami et al. 2006a). CI-Wolbachia levels vary across insect species. For instance, CI can range from strong (low proportion of hatched eggs with nearly all offspring dying) as in Rhagoletis cerasi (Linnaeus) (Diptera: Tephritidae), Glossina morsitans (Westwood) (Diptera: Glosinidae), and Aedes aegypti (Linnaeus) (Diptera: Culidae) (Riegler and Stauffer 2002, Alam et al. 2011, Hoffmann et al. 2015, Sicard et al. 2019), to weak (high proportion of hatched eggs with most offspring surviving) as in Drosophila simulans (Sturtevant) (Diptera: Drosophilidae), Drosophila mauritiana (Tsacas & David) (Diptera: Drosophilidae), and Anopheles sp. (Meigen) (Diptera: Culicidae) (Hoffmann et al. 1996, Meany et al. 2019, Sicard et al. 2019). CI-Wolbachia level is influenced by several factors including Wolbachia strain, Wolbachia diversity, host species, age, temperature, superinfection, food availability, antibiotics, and Wolbachia density (Breeuwer and Werren 1993, Clancy and Hoffmann 1998, Noda et al. 2001, Ikeda et al. 2003, Thomas and Blanford 2003, Mouton et al. 2006, Hoffmann et al. 2015).
The relationship between Wolbachia density and CI level may involve a quorum sensing (QS) mechanism. QS is a bacterial cell-to-cell signal communication according to population density (Kalia 2014, Tommonaro 2019), when a quorum of cells reaches a threshold density, a gene regulatory system is activated that results in phenotypic changes (Kalia 2014, Tommonaro 2019, Coquant et al. 2020). Bacterial cells communicate with each other by detecting or producing signaling molecules known as autoinducers such as N-acylated homoserine lactone (AHL) in Gram-negative bacteria (Papenfort and Bassler 2016). AHL is produced by the highly conserved LuxI gene or its homologs, resulting in the expression of LuxR transcriptional regulator (Whitehead et al. 2001, Papenfort and Bassler 2016). Several commercial AHL can induce QS of bacteria, including 3O-C12-HSL that promote QS of Pseudomonas aeruginosa (Schroeter) (Pseudomonadales: Pseudomonadaceae), mice pathogen (Smith et al. 2002), 3O-C6-HSL which induce QS of Sodalis glossinidius (Dale & Maudlin) (Enterobacterales: Bruguierivoracaceae), an endosymbiont of Glossina spp. (Wiedemann) (Diptera: Glossinidae) (Pontes et al. 2008). QS is a critical mechanism for controlling activities associated with a symbiosis that influences host physiologic functions (Borlee et al. 2010, Pietschke et al. 2017).
In this study, we investigated the effect of manipulating QS in Wolbachia within L. trifolii, using several chemicals including 3O-C12-HSL; C2HSL; spermidine (QS inducers), 4-phenylbutanoyl; and 4-NPO (QS inhibitors). We show for the first time that Wolbachia density and the proportion of hatched eggs of L. trifolii can be controlled by inducing or inhibiting QS in Wolbachia, which is potent to apply as an IIT.
Materials and Methods
Insect Rearing
Wolbachia-infected L. trifolii isolated in Hamamatsu, Shizuoka, Japan in 1991 (Tagami et al. 2006a) was reared on the leaves of kidney bean plants in a 34 cm (width) × 35 cm (length) × 34 cm (height) cage at 23°C under a 16:8-h light/dark photoperiod.
DNA Extraction and Wolbachia Detection
Wolbachia infection status was confirmed by PCR using forward and reverse primers for Wolbachia surface protein (wsp81F and wsp691R, respectively) (Braig et al. 1998). For PCR analysis, DNA was extracted from 10–15 adults L. trifolii by sodium dodecyl sulfate (SDS)–ethanol precipitation according to a previously described method (Haukedal 2013), with slight modifications. Insect specimens in 50 µl SDS lysis buffer were crushed with a motorized pestle (Power Masher II; Nippi, Tokyo, Japan); 150 µl of SDS buffer and 20 µl proteinase-K (20 mg/ml) were added to the homogenate, which was incubated at 56°C for at least 6 h. The proteinase-K was deactivated at 99°C for 3 min and the sample was centrifuged (CT 15RE; Hitachi, Tokyo, Japan) at 7,000 rpm for 4 min. The supernatant was collected and a one-eighth volume of 8 M ammonium acetate and double the volume of 99%–100% ethanol was added, followed by incubation at −20°C for one h or 4°C overnight. To purify the DNA, the sample was centrifuged at 12,000 rpm for 30 min. The supernatant was removed, and the pellet was washed with 500 µl of 70% ethanol and then centrifuged at 12,000 rpm for 10 min. The supernatant was discarded and after drying, the pellet was resuspended in 100 µl Tris-EDTA buffers. DNA purity and yield (Bustin et al. 2009), were verified with a Nanodrop 1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA).
To verify Wolbachia infection status, DNA was extracted from a single adult specimen crushed in 30 µl sodium chloride-Tris-EDTA buffer containing 2 µl proteinase-K (20 mg/ml) and incubated at 56°C for two h. PCR was performed using GoTaq (Promega, Madison, WI) according to standard procedures on a PCR Thermal Cycler (Takara, Otsu, Japan). The cycling conditions were 98°C for 10 s; 35 cycles of 98°C for 10 s, 55°C for 30 s, and 72°C for 1 min; and 72°C for 4 min.
Antibiotic Treatment
To obtain Wolbachia-uninfected L. trifolii, infected pupae were collected in a new cage with dimensions of 6.5 cm (width) × 6.5 cm (length) × 9.5 cm (height), emerged flies were treated with 50 mg/g tetracycline hydrochloride in honey for a single day. Newly emerged adults were transferred after one day to a new cage with dimensions of 34 cm (width) × 35 cm (length) × 34 cm (height); the leaves were sprayed with 20 mg/ml tetracycline hydrochloride (Tagami et al. 2006b). This process was repeated for two to three generations. The infection status was confirmed by PCR. The treated Wolbachia-uninfected L. trifolii were used for crossing experiments.
QS Inducer and Inhibitor Treatments
The QS inducers N-(3-oxododecanoyl) homoserine lactone (3O-C12-HSL) and N-acetyl-l-homoserine lactone (C2HSL) (both from Tocris, Minneapolis, MN) and spermidine (Cayman Chemical, Ann Arbor, MI) and QS inhibitors 4-phenylbutanoyl (4-phenyl) (Combi-Blocks, Sandiego, CA) and 4-nitro-pyridine-N-oxide (4-NPO) (Tokyo chemical industry, Tokyo, Japan) were used at the following concentrations: 3O-C12-HSL: 1, 10, and 100 µM (Smith et al. 2002, Josephson et al. 2020), C2HSL: 0.1, 1, and 10 µM (Smith et al. 2002; Liu et al. 2019); spermidine, 1, 10, and 100 µM (Nasrallah et al. 2011); 4-phenyl, 50, 100, and 150 µM (Schaefer et al. 1996); and 4-NPO: 10 and 100 µM, and 1 mM (Rasmussen et al. 2005). Unmated L. trifolii adults were collected within 24 h after emergence and were fed the chemical in honey (honey put in 1x1 cm cotton) in a 6.5 cm (width) × 6.5 cm (length) × 9.5 cm (height) cage, for one day. Treated L. trifolii were reared in a 34 cm (width) × 35 cm (length) × 34 cm (height) cage; for indirect uptake, the chemicals were sprayed daily onto the leaves up to the larva stage of the next generation.
Crossing Experiments
Crossing experiments were performed to observe and evaluate the effect of QS inducer or inhibitor in the number of oviposited eggs, larvae, and adults. We observed four crosses combinations (w+♂ and w-♀, w-♂ and w+♀, w+♂ and w+♀, w-♂ and w-♀) in every different QS inducer and inhibitor. At least six pair crosses were carried out for each cross combination in different QS inducer and inhibitor treatments. A pair of virgin individuals were reared on a kidney bean leaf in an 11.5 cm (diameter) × 8.5 cm (height) plastic cup. The number of oviposited eggs, larvae, and adult longevity (until death) were recorded.
Cytoplasmic Incompatibility Assays
The CI expression levels were checked in crosses between male infected with female uninfected. To correct the nuisance from normal mortality caused by the crossing that uncorrelated with CI, we used a corrected index of CI (CIcorr) which is described as the percentage of egg that do not hatch (from the CI crosses between infected males and uninfected females) among the crosses which survive without CI effect. Then CIcorr(%) = [(CIobs − CCM)/(100 − CCM)] × 100, where CIobs is the percentage of unhatched eggs detected in the CI cross (infected males and uninfected females) and CCM is the percentage of unhatched eggs observed in the compatible cross from reciprocal crosses (uninfected males and infected females) (Poinsot et al. 1998). For CI experiments, the status of Wolbachia infection in female and male parents and their offspring was confirmed by PCR.
qPCR
Wolbachia density under different chemical treatments was evaluated by qPCR on a Thermal Cycler Dice Real-Time System II (TP 970; Takara). The total reaction volume of 25 µl contained 12.5 µl TB Green Premix Ex Taq II (Takara), 8.5 µl sterile water, 1 µl each of forward and reverse primers, and 2 µl template DNA. The cycling conditions were 98°C for 10 s, followed by 40 cycles of 98°C for 10 s and 55°C for 30 s. Relative Wolbachia density was calculated with the 2−∆∆Ct method (Livak and Schmittgen 2001) by comparing the copy number of the wsp gene to that of the L. trifolii cytochrome oxidase I (COI) gene (Watanabe et al. 2011). The following primers were designed using NCBI Primer 3 Plus (Ye et al. 2012), wsp, L.tri wsp-F (5′-ACTCCATACGTTGGTGTTGGT-3′) and L.tri wsp-R (5′-CCAAAATAACGACTCCAGCA-3′); and COI, L.tri COI-F (5′-GGGAAAGCTATATCTGGGGCT-3′) and L.tri COI-R (5′-CCGGTGCCCTAATTGGTGAT-3′). There were six replicates for each treatment (3 biological replicates, each with at least two technical replicates for each gene) (Taylor et al. 2010).
Statistical Analysis
Effect of QS inducer and QS inhibitor on Wolbachia density in L. trifolii was estimated with ∆Ct which refer to 2−∆∆Ct (Yuan et al. 2006), followed by one-way analysis of variance (ANOVA), and the Tukey’s honestly significant difference (HSD) test for comparison of means (Tukey 1949).
The normal distribution of data was tested with the Shapiro–Wilk test (Shapiro and Wilk 1965). Data on the QS inducer and QS inhibitor effect on oviposited eggs were normally distributed and evaluated by one-way ANOVA. Data on the effect of QS inducer and QS inhibitor on adult longevity were non-normally distributed and were analyzed with the Kruskal–Wallis test (Kruskal and Wallis 1952).
CI levels were analyzed using a generalized linear model (GLM) with binomial distribution and logit link function. Those analyses are appropriate to analyze the proportion data model (Crawley 2012, Demétrio et al. 2014, McCullagh and Nelder 2019).
Statistical analyses were performed using SPSS v21.0 software (Nie et al. 1975) and R v4.0.4 (R 2013). P < 0.05 was considered statistically significant in all statistical tests.
Results
Wolbachia Density After Treatment With QS Inducer
Induction of QS using 3O-C12-HSL increased the density of Wolbachia (one-way ANOVA, F = 5.62, df = 3, 8, P = 0.023) (Fig. 1). The density of Wolbachia was increased 3.3-fold with respect to the control (Tukey’s HSD, df = 2, 2, P = 0.081) due to treatment with 1 µM of 3O-C12-HSL. Furthermore, the density was increased 5.35 fold (Tukey’s HSD, df = 2, 2, P = 0.017) due to treatment with 10 µM of 3O-C12-HSL. The density of Wolbachia was not increased anymore when 100 µM of 3O-C12-HSL were applied (2.94 fold of control) (Tukey’s HSD, df = 2, 2, P = 0.12). Therefore, 10 µM of 3O-C12-HSL was considered the most effective concentration for increasing the density of Wolbachia in infected L. trifolii.
Relative density of Wolbachia in Lirimyoza trifolii following treatment with the QS inducer 3O-C12-HSL. Values are shown as means with standard errors of six replicates (Three biological replicates, each consisting of two technical replicates). Different letters above the bars indicate statistically significant differences (Tukey‘s honestly significant difference test, P < 0.05). 3O-C12-HSL 10 μM showed Wolbachia density significant difference with control (3O-C12-HSL 0 μM).
Application of C2HSL (Fig. 2) and spermidine (Fig. 3) increased the density of Wolbachia in a dose-dependent manner, but the results were not statistically significant (C2HSL: one-way ANOVA, F = 1.61, df = 3, 9, P = 0.254; spermidine: one-way ANOVA, F = 0.31, df = 3, 8, P = 0.814).
Relative density of Wolbachia in Lirimyoza trifolii following treatment with the QS inducer C2HSL. Values are shown as means with standard errors of six replicates (Three biological replicates, each consisting of two technical replicates). Data were analyzed by one-way ANOVA (P > 0.05).
Relative density of Wolbachia in Lirimyoza trifolii following treatment with the QS inducer spermidine. Values are shown as means with standard errors of six replicates (Three biological replicates, each consisting of two technical replicates). Data were analyzed by one-way ANOVA (P > 0.05).
Wolbachia Density After Treatment With QS Inhibitor
Inhibition of QS using 4-phenyl decreased the density of Wolbachia in a significant way (one-way, ANOVA, F = 4.37, df = 3,8, P = 0.042). Significant difference compared to the control group was observed at the highest concentration (150 µM) of 4-phenyl (Tukey’s HSD, df = 2, 2; P = 0.038) (Fig. 4.). There was a similar dose-dependent effect noted for 4-NPO (Fig. 5.), but the results were not statistically significant (one-way, ANOVA, F = 1.17, df = 3, 8, P = 0.377).
Relative density of Wolbachia in Lirimyoza trifolii following treatment with the QS inhibitor 4-phenyl. Values are shown as means with standard errors of six replicates (Three biological replicates, each consisting of two technical replicates). Different letters above the bars indicate statistically significant differences (Tukey’s honestly significant difference test, P < 0.05). 4-Phenyl 150 μM showed Wolbachia density significant difference with other concentration including control (4-phenyl 0 μM).
Relative density of Wolbachia in Lirimyoza trifolii following treatment with the QS inhibitor 4-NPO. Values are shown as means with standard errors of six replicates (Three biological replicates, each consisting of two technical replicates). Data were analyzed by one-way ANOVA (P > 0.05).
Crossing Experiments and Cytoplasmic Incompatibility Assays
Crossing experiments were performed to investigate the effect of QS manipulation on the CI expression of L. trifolii. Treatment with the QS inducer 3O-C12 HSL reduced egg hatching rates/ induced strong CI (GLM, z = −2.40, P < 0.05). The CIcorr values after 3O-C12 HSL treatment was (99.4% ± 0.59). Whereas treatment with the QS inhibitor 4-phenyl had the opposite effect (GLM, z = 4.94, P < 0.001), reduced CIcorr values (68.38% ± 18.8) (Fig. 6).
Cytoplasmic incompatibility levels resulting from a cross between a Lirimyoza trifolii male infected with Wolbachia and an uninfected female under QS inducer or inhibitor treatment. The corrected index of CI (CIcorr) was performed to correct the nuisance from normal mortality caused by the crossing that uncorrelated with CI. A generalized linear model (GLM) binomial family analysis was used to determine the differences between QS treatment (*P < 0.05, ***P < 0.001).
Number of Ovipositing Eggs
There was no difference in the number of oviposited eggs across groups compared to the control under the different chemical treatments (one-way ANOVA, F = 0.97, df = 23, 126, P = 0.49) (Table 1).
Effect of QS inducer or QS inhibitor on number of oviposited eggs in Lirimyoza trifolii crossing experiments
| QS treatment | Male | Female | No. of oviposited eggs* |
|---|---|---|---|
| Control | + | − | 20.60 ± 3.10 |
| − | + | 24.00 ± 1.93 | |
| + | + | 25.57 ± 3.42 | |
| − | − | 34.00 ± 2.20 | |
| QS inducer | |||
| 3O-C12HSL | + | − | 27.91 ± 0.96 |
| − | + | 26.66 ± 2.53 | |
| + | + | 33.20 ± 5.69 | |
| − | − | 32.50 ± 3.49 | |
| Spermidine | + | − | 31.50 ± 3.40 |
| − | + | 32.10 ± 2.90 | |
| + | + | 28.66 ± 3.14 | |
| − | − | 25.83 ± 4.36 | |
| C2HSL | + | − | 30.83 ± 5.10 |
| − | + | 32.00 ± 5.32 | |
| + | + | 26.00 ± 3.45 | |
| − | − | 25.16 ± 2.72 | |
| QS inhibitor | |||
| 4-Phenyl | + | − | 24.83 ± 2.95 |
| − | + | 34.10 ± 5.14 | |
| + | + | 31.00 ± 3.40 | |
| − | − | 31.66 ± 5.90 | |
| 4-NPO | + | − | 30.14 ± 4.32 |
| − | + | 31.40 ± 1.16 | |
| + | + | 21.60 ± 1.32 | |
| − | − | 28.83 ± 5.60 |
| QS treatment | Male | Female | No. of oviposited eggs* |
|---|---|---|---|
| Control | + | − | 20.60 ± 3.10 |
| − | + | 24.00 ± 1.93 | |
| + | + | 25.57 ± 3.42 | |
| − | − | 34.00 ± 2.20 | |
| QS inducer | |||
| 3O-C12HSL | + | − | 27.91 ± 0.96 |
| − | + | 26.66 ± 2.53 | |
| + | + | 33.20 ± 5.69 | |
| − | − | 32.50 ± 3.49 | |
| Spermidine | + | − | 31.50 ± 3.40 |
| − | + | 32.10 ± 2.90 | |
| + | + | 28.66 ± 3.14 | |
| − | − | 25.83 ± 4.36 | |
| C2HSL | + | − | 30.83 ± 5.10 |
| − | + | 32.00 ± 5.32 | |
| + | + | 26.00 ± 3.45 | |
| − | − | 25.16 ± 2.72 | |
| QS inhibitor | |||
| 4-Phenyl | + | − | 24.83 ± 2.95 |
| − | + | 34.10 ± 5.14 | |
| + | + | 31.00 ± 3.40 | |
| − | − | 31.66 ± 5.90 | |
| 4-NPO | + | − | 30.14 ± 4.32 |
| − | + | 31.40 ± 1.16 | |
| + | + | 21.60 ± 1.32 | |
| − | − | 28.83 ± 5.60 |
Data represent mean ± standard error of 6 replicates. +, Infected with Wolbachia; −, uninfected.
*No significant difference (P > 0.05, analysis of variance).
Effect of QS inducer or QS inhibitor on number of oviposited eggs in Lirimyoza trifolii crossing experiments
| QS treatment | Male | Female | No. of oviposited eggs* |
|---|---|---|---|
| Control | + | − | 20.60 ± 3.10 |
| − | + | 24.00 ± 1.93 | |
| + | + | 25.57 ± 3.42 | |
| − | − | 34.00 ± 2.20 | |
| QS inducer | |||
| 3O-C12HSL | + | − | 27.91 ± 0.96 |
| − | + | 26.66 ± 2.53 | |
| + | + | 33.20 ± 5.69 | |
| − | − | 32.50 ± 3.49 | |
| Spermidine | + | − | 31.50 ± 3.40 |
| − | + | 32.10 ± 2.90 | |
| + | + | 28.66 ± 3.14 | |
| − | − | 25.83 ± 4.36 | |
| C2HSL | + | − | 30.83 ± 5.10 |
| − | + | 32.00 ± 5.32 | |
| + | + | 26.00 ± 3.45 | |
| − | − | 25.16 ± 2.72 | |
| QS inhibitor | |||
| 4-Phenyl | + | − | 24.83 ± 2.95 |
| − | + | 34.10 ± 5.14 | |
| + | + | 31.00 ± 3.40 | |
| − | − | 31.66 ± 5.90 | |
| 4-NPO | + | − | 30.14 ± 4.32 |
| − | + | 31.40 ± 1.16 | |
| + | + | 21.60 ± 1.32 | |
| − | − | 28.83 ± 5.60 |
| QS treatment | Male | Female | No. of oviposited eggs* |
|---|---|---|---|
| Control | + | − | 20.60 ± 3.10 |
| − | + | 24.00 ± 1.93 | |
| + | + | 25.57 ± 3.42 | |
| − | − | 34.00 ± 2.20 | |
| QS inducer | |||
| 3O-C12HSL | + | − | 27.91 ± 0.96 |
| − | + | 26.66 ± 2.53 | |
| + | + | 33.20 ± 5.69 | |
| − | − | 32.50 ± 3.49 | |
| Spermidine | + | − | 31.50 ± 3.40 |
| − | + | 32.10 ± 2.90 | |
| + | + | 28.66 ± 3.14 | |
| − | − | 25.83 ± 4.36 | |
| C2HSL | + | − | 30.83 ± 5.10 |
| − | + | 32.00 ± 5.32 | |
| + | + | 26.00 ± 3.45 | |
| − | − | 25.16 ± 2.72 | |
| QS inhibitor | |||
| 4-Phenyl | + | − | 24.83 ± 2.95 |
| − | + | 34.10 ± 5.14 | |
| + | + | 31.00 ± 3.40 | |
| − | − | 31.66 ± 5.90 | |
| 4-NPO | + | − | 30.14 ± 4.32 |
| − | + | 31.40 ± 1.16 | |
| + | + | 21.60 ± 1.32 | |
| − | − | 28.83 ± 5.60 |
Data represent mean ± standard error of 6 replicates. +, Infected with Wolbachia; −, uninfected.
*No significant difference (P > 0.05, analysis of variance).
Adult Longevity
QS inducer and inhibitor treatment did not affect the longevity of emerged adult compared to the control group, and there were no differences between males (H(11) = 14.71, P = 0.196) and females (H(11)=15.623, P = 0.156) for any treatment including the control (Table 2, Supp Figs. S1 and S2 [online only]).
Effect of QS inducer or QS inhibitor on the longevity of adult Lirimyoza trifolii
| Male | Female | |||
|---|---|---|---|---|
| QS treatment | Infection status | Adult longevity, days* | Infection status | Adult longevity, days* |
| Control | + | 3.30 ± 0.33 | + | 8.00 ± 0.40 |
| − | 4.00 ± 0.26 | − | 9.60 ± 0.57 | |
| QS inducer | ||||
| 3O-C12HSL | + | 3.57 ± 0.29 | + | 7.66 ± 0.66 |
| − | 4.00 ± 0.70 | − | 7.60 ± 0.50 | |
| C2HSL | + | 3.09 ± 0.21 | + | 8.30 ± 0.95 |
| − | 3.75 ± 0.47 | − | 9.00 ± 1.18 | |
| Spermidine | + | 5.60 ± 0.87 | + | 7.77 ± 0.49 |
| − | 3.60 ± 0.33 | − | 9.60 ± 0.33 | |
| QS inhibitor | ||||
| 4-Phenyl | + | 3.50 ± 0.29 | + | 7.00 ± 0.31 |
| − | 4.00 ± 0.31 | − | 6.80 ± 1.10 | |
| 4-NPO | + | 3.10 ± 0.40 | + | 8.30 ± 1.00 |
| − | 3.75 ± 1.10 | − | 9.60 ± 1.02 |
| Male | Female | |||
|---|---|---|---|---|
| QS treatment | Infection status | Adult longevity, days* | Infection status | Adult longevity, days* |
| Control | + | 3.30 ± 0.33 | + | 8.00 ± 0.40 |
| − | 4.00 ± 0.26 | − | 9.60 ± 0.57 | |
| QS inducer | ||||
| 3O-C12HSL | + | 3.57 ± 0.29 | + | 7.66 ± 0.66 |
| − | 4.00 ± 0.70 | − | 7.60 ± 0.50 | |
| C2HSL | + | 3.09 ± 0.21 | + | 8.30 ± 0.95 |
| − | 3.75 ± 0.47 | − | 9.00 ± 1.18 | |
| Spermidine | + | 5.60 ± 0.87 | + | 7.77 ± 0.49 |
| − | 3.60 ± 0.33 | − | 9.60 ± 0.33 | |
| QS inhibitor | ||||
| 4-Phenyl | + | 3.50 ± 0.29 | + | 7.00 ± 0.31 |
| − | 4.00 ± 0.31 | − | 6.80 ± 1.10 | |
| 4-NPO | + | 3.10 ± 0.40 | + | 8.30 ± 1.00 |
| − | 3.75 ± 1.10 | − | 9.60 ± 1.02 |
Data represent mean ± standard error. +, Infected with Wolbachia; −, uninfected.
*No significant difference (P > 0.05, Kruskal–Wallis test).
Effect of QS inducer or QS inhibitor on the longevity of adult Lirimyoza trifolii
| Male | Female | |||
|---|---|---|---|---|
| QS treatment | Infection status | Adult longevity, days* | Infection status | Adult longevity, days* |
| Control | + | 3.30 ± 0.33 | + | 8.00 ± 0.40 |
| − | 4.00 ± 0.26 | − | 9.60 ± 0.57 | |
| QS inducer | ||||
| 3O-C12HSL | + | 3.57 ± 0.29 | + | 7.66 ± 0.66 |
| − | 4.00 ± 0.70 | − | 7.60 ± 0.50 | |
| C2HSL | + | 3.09 ± 0.21 | + | 8.30 ± 0.95 |
| − | 3.75 ± 0.47 | − | 9.00 ± 1.18 | |
| Spermidine | + | 5.60 ± 0.87 | + | 7.77 ± 0.49 |
| − | 3.60 ± 0.33 | − | 9.60 ± 0.33 | |
| QS inhibitor | ||||
| 4-Phenyl | + | 3.50 ± 0.29 | + | 7.00 ± 0.31 |
| − | 4.00 ± 0.31 | − | 6.80 ± 1.10 | |
| 4-NPO | + | 3.10 ± 0.40 | + | 8.30 ± 1.00 |
| − | 3.75 ± 1.10 | − | 9.60 ± 1.02 |
| Male | Female | |||
|---|---|---|---|---|
| QS treatment | Infection status | Adult longevity, days* | Infection status | Adult longevity, days* |
| Control | + | 3.30 ± 0.33 | + | 8.00 ± 0.40 |
| − | 4.00 ± 0.26 | − | 9.60 ± 0.57 | |
| QS inducer | ||||
| 3O-C12HSL | + | 3.57 ± 0.29 | + | 7.66 ± 0.66 |
| − | 4.00 ± 0.70 | − | 7.60 ± 0.50 | |
| C2HSL | + | 3.09 ± 0.21 | + | 8.30 ± 0.95 |
| − | 3.75 ± 0.47 | − | 9.00 ± 1.18 | |
| Spermidine | + | 5.60 ± 0.87 | + | 7.77 ± 0.49 |
| − | 3.60 ± 0.33 | − | 9.60 ± 0.33 | |
| QS inhibitor | ||||
| 4-Phenyl | + | 3.50 ± 0.29 | + | 7.00 ± 0.31 |
| − | 4.00 ± 0.31 | − | 6.80 ± 1.10 | |
| 4-NPO | + | 3.10 ± 0.40 | + | 8.30 ± 1.00 |
| − | 3.75 ± 1.10 | − | 9.60 ± 1.02 |
Data represent mean ± standard error. +, Infected with Wolbachia; −, uninfected.
*No significant difference (P > 0.05, Kruskal–Wallis test).
Discussion
In this study, we investigated the effect of QS inducers 3O-C12-HSL, C2HSL, spermidine, and QS inhibitors 4-phenyl, and 4-NPO on L. trifolii and Wolbachia inside L. trifolii. 3O-C12-HSL and C2HSL are specific autoinducers in Gram-negative bacteria (Liu et al. 2019); and spermidine is a polyamine that is used to stimulate intracellular growth, such as enhance the cell growth (Nasrallah et al. 2011). We found that the density of Wolbachia was increased compared to the control due to treatment with 3O-C12-HSL (Fig. 1). Probably 3O-C12-HSL are accumulated in the cells and stimulates the proliferation of Wolbachia. In contrast, neither C2HSL nor spermidine had significant effects on Wolbachia density. In our study, spermidine did not produce a specific effect on QS, probably because spermidine has a broad effect. It not only induces QS in a specific way but can also modulate gene expression, biosynthesis, degradation, transport, and cell growth by targeting RNA polymerase activity (Chan and Chua 2010, Igarashi and Kashiwagi 2010, Martino et al. 2013).
Variable effects of AHL QS inducers on the density of Wolbachia were observed in our study. It may be attributed to the fact that the different positions or lengths of the acyl chain. Gram-negative bacteria have at least 30 AHLs composed of a homoserine lactone ring and various acyl side chains, the length of which determine the signaling specificity of the molecule (Parsek and Greenberg 2000, Liu et al. 2019, Phuong et al. 2020). 3O-C12-HSL has a more extended chain structure than C2HSL (Smith et al. 2002, González et al. 2013). AHLs with a long acyl chain including 3O-C12-HSL strongly induce QS. It was shown that they enhance adhesion of Acidithiobacillus ferrooxidans (Temple & Colmer) (Acidithiobacillales: Acidithiobacillaceae) cells to substrates via a QS mechanism (González et al. 2013). Our results also support the results of a previous study demonstrating that nonacylated HSLs, such as C2HSL and C4HSL have a modest QS-inducing effect than 3O-C12-HSL in Pseudomonas aeruginosa implying that the characteristics such as length and presence of the acyl side chain influences this function (Smith et al. 2002).
Our qPCR data showed that treatment with the QS inhibitor 4-phenyl (150 µM) reduced the density of Wolbachia by 0.15 fold with respect to the control (Fig. 4). 4-Phenyl has analogous structure with the QS inducer. Therefore 4-phenyl binds to LuxR to prevent the binding of 3-oxohexanoyl as well as the expression of QS-related genes (Schaefer et al. 1996, Kalia 2014). 4-Phenyl was shown to inhibit LuxR strongly in Vibrio fischeri (Beijerinck) (Vibrionales: Vibrionaceae) (Geske et al. 2007). In our study, 4-NPO did not suppress Wolbachia density (Fig. 5), and probably did not inhibit the QS system in Wolbachia. 4-NPO suppresses the QS activity of P. aeruginosa and formation of biofilm (Rasmussen et al. 2005), but did not affect protease activity or inhibit the QS system of Pectobacterium (Waldee) (Enterobacterales: Pectobacteriaceae) (Rasch et al. 2007). Nonetheless, results of these experiments indicate that density of Wolbachia can be manipulated by the QS system targeting chemical agents.
QS inducer and inhibitor treatment did not significantly affect the number of oviposited eggs (Table 1) or adult longevity (Table 2, Supp Figs. S1 and S2 [online only]), likely because the doses used in this study (0.1–1,000 µl) had only minor effects on L. trifolii. The oral 50% lethal dose of 4-NPO is 107 mg/kg in rats and 20 mg/kg in pigeons (Goodhoue and Baumgartner 1965).
We found that the average survived days of male and female adults were 3.73 and 8.78 d respectively (Table 2, Supp Figs. S1 and S2 [online only]), this is in line with the previously reported 3 and 8–12 d, respectively (Charlton and Allen 1981, Zoebisch and Schuster 1987, Koo et al. 2012).
Treatment with the QS inducer 3O-C12-HSL induce strong CI in L. trifolii (Fig. 6). We also found that high Wolbachia density was associated with the low proportion of hatched eggs following 3O-C12-HSL treatment (Figs. 1 and 6). Our results are consistent with an earlier finding that high Wolbachia density was positively correlated with a low proportion of hatched eggs in insect hosts including Laodelphax striatellus (Fallén) (Delphacidae) (Noda et al. 2001). On the contrary, low density of Wolbachia in Drosophila simulans and Sogatella furciffera (Horvath) (Hemiptera: Delphacidae) were correlated with a high proportion of hatched eggs in both species (Bressac and Rousset 1993, Clancy and Hoffmann 1998, Noda et al. 2001).
3O-C12-HSL was reported to induce QS during Gram-negative bacteria infection (Smith et al. 2002), thereby affecting the host (Josephson et al. 2020). Dermal injection of 1–10 µM 3O-C12-HSL in mice induced QS in P. aeruginosa, leading to lung inflammation (Smith et al. 2002); and 50 µM 3O-C12-HSL modulated QS in bacteria, which impacted host mitochondria (Josephson et al. 2020). As QS is related to bacterial density and Wolbachia density influences CI level, QS may be indirectly associated with CI level. When Wolbachia density reaches a threshold level, the expression of target genes is activated or inhibited (Smith et al. 2002). Wolbachia density probably activated CI genes. This was supported by our data, which showed associations among QS chemical treatment, Wolbachia density, and CI levels in the insect host.
For CI-Wolbachia to be effective as a control measure for L. trifolii, the proportion of hatched eggs must be drastically reduced so that offspring will not be produced. It was suggested that the strain of Wolbachia is the main determinant of CI-Wolbachia efficacy (Noda et al. 2001). However, trans infection of Tribolium confusum (Jacquelin du Val) (Coleoptera: Tenebrionidae) and Tribolium castanem (Herbst) (Coleoptera: Tenebrionidae) with a strong CI strain of Wolbachia has had limited success (Hughes and Rasgon 2014), indicating that a close phylogenetic relationship between insect and hosts does not guarantee efficient trans infection (Ikeda et al. 2003, Zabalou et al. 2004). There have been numerous attempts of microinjection of Wolbachia into Anopheles spp., but many attempts have failed (Hughes and Rasgon 2014). On the other hand, when Wolbachia is successfully transferred to a new host, it can cause physiologic dysfunction and alter lifespan (McGraw et al. 2002). Trans infection of Wolbachia strain wVulC from Armadillidium vulgare (Latreille) (Isopoda: Armadillidiidae) to Porcellio dilatatus (Brandt) (Isopoda: Porcellionidae) resulted in nervous system disorder and growth inhibition of the new host (Le Clec’h et al. 2012); and transfer of Wolbachia strain wMelPop from Drosophila melanogaster (Meigen) (Diptera: Drosophilidae) into Aedes albopictus reduced the fecundity and longevity of the mosquito (Suh et al. 2009).
QS inducer treatment, which targets Wolbachia inside the insect host, could be a promising alternative method to decrease the proportion of hatched eggs by promoting CI. In the present study, the 3O-C12-HSL treatment increased Wolbachia density in insect cells and strongly induced CI (Figs. 1 and 6). Thus, this study is a good initiation that can be adopted for IIT.
IIT based on CI-Wolbachia has several advantages, such as the fact that Wolbachia can be transferred to the next generation by vertical transmission, thereby eliminating the cost associated with radiation in SIT (Moretti and Calvitti 2021). Moreover, IIT does not require gene modification, for example, through clustered regularly interspaced short palindromic repeats (CRISPR), and release of transgenic insects carrying a dominant lethal mutation (Zhang et al. 2015, Pagendam et al. 2020, Kandel et al. 2021). Thus, the IIT based on CI-Wolbachia may be applicable to L. trifolii.
The lack of a standardized and efficient method for isolating males is an obstacle for the IIT (Lutrat et al. 2019, Baton et al. 2021). Male selection in L. trifolii using a 1.54 × 0.76-mm pore sieve is based on pupal sexual dimorphism, although the effectiveness of this method is less than 100% (Sultan et al. 2017). Field contamination with Wolbachia-infected female L. trifolii can lead to the unintended replacement of a Wolbachia-infected population due to their compatibility with released males of the same strain as well as with uninfected wild males in the field. To minimize this risk, some researchers have suggested releasing insects infected with different strains of Wolbachia that are bidirectionally incompatible with the wild insect population (Calvitti et al. 2012, Baton et al. 2021).
Combining the IIT and SIT is also a possible solution. Integrating the SIT with IIT can complement each other’s weaknesses, since several insect males, reduce their survival and competitiveness after SIT radiation. Generally, the SIT uses radiation doses between 120–200 Gy to sterilize flies (Baton et al. 2021). In a combining SIT and IIT previous study, Drosophila suzukii (Matsumura) (Diptera: Drosophilidae) and Aedes albopictus infected with Wolbachia were irradiated at lower doses (45 Gy), and the result showed that the males not lose their survival and competitiveness, while the females were sterile (Zheng et al. 2019, Nikolouli et al. 2020). That combining technique was appropriate to prevent population replacement caused by female contamination of IIT. Integrating the SIT and IIT methods also can be adopted for other insect methods. Previous research reported that sterile L. trifolii using the SIT was feasible (Kaspi and Parrella 2003), our results showed 3O-C12 HSL induce strong CI that supported the IIT method. Thus, combining SIT and IIT is a potent tool to control L. trifolii populations.
In conclusion, our findings are the first report that use of a QS inducer for strong CI level on Wolbachia-infected males of L. trifolii for application on IIT, although additional studies are needed to optimize CI, mass rearing, male sorting, the competitive advantage of laboratory males over wild males, and field management following the release of Wolbachia-infected incompatible males.
Acknowledgments
We thank Kasai Atsushi Sensei and Tomoko Okamoto Sensei for valuable comments during the study; Yuki Ishihara and Yuta Ohata for technical assistance; Dwi Harya Yudistira and Mukhlis Jamal Holee for discussion on statistical analyses; Arijit Nath for helpful English proofreading on the revised manuscript; and the anonymous reviewers for suggesting improvements to the manuscript.
Author Contributions
A.K.H.: conceptualization; data curation; formal analysis; investigation; methodology; visualization; writing-original draft. A.G.: investigation; writing-review & editing. Y.T.: conceptualization; data curation; formal analysis; funding acquisition; investigation; methodology; project administration; resources; supervision; validation; writing-review & editing.
