Introduction

Phosphine (PH3) has been widely used as an insecticide for grain reserves since the 1930s. Due to various advantages, including its low cost, high toxicity, minimal residue, and rapid action, PH3 has replaced the ozone-harming chemical methyl bromide as the most popular fumigant worldwide1. However, with prolonged use of PH3, pests with high levels of resistance have been reported in many countries, such as the United States, Australia, and Brazil2,3,4. As there is no effective chemical replacement for PH3, much effort has been devoted to studying the mechanisms of PH3 toxicity and resistance5.

To date, the precise mode of PH3 toxicity is still unclear. One of the mechanisms involved is the generation of reactive oxygen species (ROS)6. ROS can lead to oxidative stress, lipid peroxidation, and toxic changes to the redox state of the cell, leading to a failure of oxidative respiration7,8,9,10. In support of this, antioxidants such as glutathione and melatonin can prevent most of the oxidative damage induced by PH3 11,12.

PH3 can also disrupt the antioxidant defence system13. Typically, high levels of ROS will activate the antioxidant defence system of the cell14, inducing the up-regulation of various antioxidant enzymes such as superoxide dismutase (SOD) and catalase (CAT). However, while SOD is up-regulated after PH3 treatment, expression and activity of CAT, which can decompose hydrogen peroxide to water and oxygen, are inhibited by PH3 15,16. It was proposed that PH3 may inhibit CAT by reducing the metal ion cofactor in the active site17. However, the inhibitory effect could only be observed in vivo but not in vitro, suggesting that there may be other regulatory mechanisms16,18. Previously, we found that PH3 fumigation could reduce mRNA production of the DmCAT gene in Drosophila melanogaster, which was the first evidence that PH3 may disrupt the antioxidant defence system at transcriptional level, adding another layer of complexity to the function of PH3 19. However, the mechanism by which PH3 inhibits DmCAT transcription is unknown.

In the current study, we investigated whether PH3 could directly regulate the transcription of DmCAT by dissecting the role of the DmCAT promoter in PH3-mediated DmCAT inhibition in D. melanogaster S2 cells. We first established a PH3 treatment system using S2 cells and confirmed that it led to down-regulation of DmCAT. We then developed a luciferase reporter assay and identified an essential fragment containing a DNA replication-related element (DRE) within the DmCAT promoter. This element was sufficient and necessary for PH3-mediated DmCAT repression. We then showed that levels of DREF, the DRE-binding transcription factor, were reduced upon PH3 treatment and that this phenomenon was an essential prerequisite for DmCAT repression. These data suggest that PH3 inhibits DmCAT promoter activity by reducing levels of the transcription factor DREF, providing a new mechanism for the mode of action of PH3.

Results

PH3 treatment reduces the expression of DmCAT in D. melanogaster S2 cells

To gain insight into the molecular mechanism of PH3-mediated CAT gene repression, we established a PH3 treatment system in D. melanogaster S2 cells. We observed that PH3 treatment inhibited the growth of S2 cells, with clear dose and time dependencies (Fig. 1A). At a concentration of 14 µg/L and a treatment time of 48 h, PH3 led to the death of 40% of cells.

Figure 1
figure 1

Establishment of phosphine (PH3) treatment system in Drosophila melanogaster S2 cells. (A) Cell viability assays for S2 cells treated with different concentrations of PH3 for various lengths of time. (B) Real-time PCR analysis of the DmCAT gene in S2 cells treated with 14 µg/L PH3 for 0, 1, 2, or 4 h. CK: control. Lowercase letters indicate significant differences at p < 0.05. (C) Western blot against DmCAT in S2 cells treated with 14 µg/L PH3 for 0, 1, 2, or 4 h (Full-length blots are presented in Supplementary Figs S1 and S2).

Full size image

To examine the expression of the DmCAT gene, we performed reverse transcription real-time PCR and western blotting after treating S2 cells with 14 µg/L PH3 for 1, 2, and 4 h. We observed a considerable decrease in DmCAT at both mRNA and protein level (Fig. 1B,C). These data indicated that this system could faithfully recapitulate PH3-mediated growth inhibition and CAT gene repression19.

A DRE consensus motif within the DmCAT promoter is required for PH3-mediated transcription inhibition

To directly investigate the role of the DmCAT promoter in PH3-mediated DmCAT repression, we cloned the full-length (−1,944 to 0 bp)20 and various truncated fragments of the DmCAT promoter into the pGL4.11 [luc2P] luciferase reporter vector. PH3 treatment reduced luciferase activity to 32.9% when the full-length DmCAT promoter was used, indicating that the promoter was directly involved in PH3-mediated DmCAT repression (Fig. 2A). However, four truncated promoter fragments showed similar efficacy in reducing luciferase activity upon PH3 treatment (Fig. 2B), suggesting that the core response unit lay within the −94 to 0 bp region.

Figure 2
figure 2

A minimal fragment containing a DNA replication-related element (DRE) motif is required for phosphine (PH3)-mediated DmCAT repression. (A) Promoter activity of wild-type (−1,944 full length and −94) and mutant (−94 M) DRE DmCAT promoters treated with or without PH3. (B) Luciferase assay comparing PH3 inhibition efficiencies of full-length (−1,944) and various truncated fragments of DmCAT promoter. Relative promoter activity was defined as the ratio of luciferase activity in PH3-treated cells to that in untreated cells. DRE-WT, wild-type DRE; DRE-M, mutant DRE. Lowercase letters indicate significant differences at p < 0.05.

Full size image

It was previously reported that a DRE element was present in the DmCAT promoter at positions −78 to −85 bp20. This element, together with its binding protein, DREF, is required for the positive regulation of DmCAT during fruit fly development20. To investigate whether this DRE element was required for PH3-mediated DmCAT repression, we mutated three critical nucleotides within the DRE consensus sequence and assessed the mutant promoter with the luciferase assay. Consistent with a positive role of DRE in DmCAT regulation, DRE mutations strongly reduced promoter activity (by 99%) in S2 cells without PH3 treatment (Fig. 2A). Interestingly, after treatment with PH3, relative luciferase activity was reduced by only 16.5% when using the DRE mutant promoter rather than by 61.7%, as observed for the full-length promoter (Fig. 2A,B and Supplementary Table S1). These data suggested that DRE was critical for PH3-mediated DmCAT repression.

PH3 reduces the activity of the DmCAT promoter through inhibition of DREF expression

Previously, we found that PH3 treatment led to reduced DREF levels in the fruit fly19. We observed similar effects in S2 cells at both protein and mRNA level (Fig. 3A,B). Additionally, a Chip-qPCR analysis was performed to compare the occupancy of Pol II on the DREF promoter with or without PH3 treatment. Our results showed that Pol II occupancy was greatly reduced upon PH3 treatment, suggesting that transcription activity could directly contribute to reduced DREF mRNA abundance (Fig. 3C). These data provided a possible explanation for PH3-mediated DmCAT repression, with PH3 down-regulating the expression of DREF at transcriptional level, and low amounts of DREF resulting in reduced DmCAT promoter activity.

Figure 3
figure 3

Repression of DREF by phosphine (PH3). (A) Western blot (Full-length blots are presented in Supplementary Figs S3 and S4) and (B) real-time PCR analysis of DREF in S2 cells treated with 14 µg/L PH3 for 0, 1, 2, or 4 h. CK: control. (C) Chromatin immunoprecipitation (ChIP)-qPCR analysis of Pol II enrichment on the DREF promoter with or without PH3 treatment. Lowercase letters indicate significant differences at p < 0.05. TSS, transcription start site.

Full size image

To test this hypothesis, we performed the luciferase assay under DREF knockdown or overexpression conditions. In both cases, we would expect luciferase activity not to respond to PH3 treatment, since there would be either no DREF or constitutively high levels of DREF, respectively. Double-stranded RNA-mediated DREF knockdown efficiently depleted DREF protein levels (Fig. 4A). Consistent with the positive role of the DRE/DREF system in DmCAT regulation, we observed reduced luciferase activity after DREF knockdown (Fig. 4B). After PH3 treatment, we observed a dramatic decrease in luciferase activity in mock knockdown cells but not in DREF knockdown cells, suggesting that DREF is essential for PH3-mediated DmCAT repression (Fig. 4B).

Figure 4
figure 4

Effect of DREF RNAi and overexpression on phosphine (PH3)-mediated DmCAT repression. (A) Representative western blot validating the efficiency of DREF RNAi. CK: control (Full-length blots are presented in Supplementary Figs S4 and S5). (B) Luciferase assay comparing relative luciferase activities of control and DREF RNAi cells with or without PH3 treatment. *p < 0.05 vs. untreated cells. (C) Representative western blot validating the overexpression of DREF (Full-length blots are presented in Supplementary Figs S4 and S6). (D) Luciferase assay comparing relative luciferase activities of control or DREF overexpressing cells with or without PH3 treatment. *p < 0.05 vs. untreated cells.

Full size image

For DREF overexpression, we cloned the full-length DREF open reading frame into the pAc5.1/V5-His expression vector and transfected this vector along with the pGL4.11-pCAT [luc2P] luciferase reporter. Western blot confirmed the overexpression of DREF protein (Fig. 4C). We observed higher luciferase activity in DREF overexpressing cells, again supporting its positive role in DmCAT expression. After PH3 treatment, we found that luciferase activity was reduced only in control transfected cells and not in DREF overexpressing cells, further supporting our hypothesis (Fig. 4D).

Effect of PH3 on the expression of DREF target genes

As recently reviewed by Nguyen et al., DREF is a multifunctional factor, and could positively regulate the expression of genes other than CAT 21. To determine whether these DREF target genes were also targets of PH3, we investigated the expression of seven DREF target genes including WARTS 22, TTF 23, TFB2 24, HIPPO 25, OSA 26, MOIRA 26, and P53 27 in S2 cells after PH3 treatment. Our results showed that the expression of TTF and TFB2, which are related to mitochondrial function, were partly reduced after PH3 treatment, indicating that the DRE/DREF system might play a common role in relaying the PH3 signal to its target genes (Fig. 5).

Figure 5
figure 5

Effect of phosphine (PH3) on the expression of seven DREF target genes. *p < 0.05 vs. untreated cells.

Full size image

Discussion and Conclusion

This and our most recent studies demonstrate that PH3 inhibits catalase activity via transcriptional regulation19. Previous studies mostly focused on the direct inhibition of catalase enzymatic activity by PH3 15,16. While these two mechanisms are not necessarily mutually exclusive, it should be noted that the enzymatic inhibition of catalase could only be observed in vivo but not in vitro, suggesting that enzymatic regulation might not be direct16,18. In addition, reduction of the metal ion cofactor of catalase by PH3 could not explain why PH3 has the opposite effect on superoxide dismutase, which also has metal cofactors.

The observation that PH3 treatment results in down-regulation of DmCAT mRNA in D. melanogaster led us to investigate how PH3 affected DmCAT transcription at the molecular level. By devising a luciferase reporter system in D. melanogaster S2 cells, we demonstrated that DmCAT promoter activity was strongly repressed upon PH3 treatment. This response was quick (within 1 h), indicating that it may not be a downstream effect of antioxidant system failure. Further dissection of the DmCAT promoter showed that a small fragment (−94 to 0 bp) was necessary and sufficient for PH3 responsiveness (Fig. 2). Interestingly, we found that a DRE motif (−78 to −85 bp) was involved in this process. The DmCAT promoter with a mutated DRE sequence no longer responded to PH3 treatment, although basal promoter activity was greatly reduced. It has been reported that the CAT gene is positively regulated by the DRE/DREF system in D. melanogaster 20, which raises an interesting question as to why a positive regulatory element would be involved in the negative regulation of the CAT gene.

The finding that levels of the positive regulatory factor DREF decrease after PH3 treatment provides a reasonable explanation for this discrepancy. PH3 treatment does not directly inhibit the DmCAT promoter, but rather it reduces levels of the positive regulatory factor DREF. By manipulating levels of DREF via RNAi or overexpression experiments, we report that (1) the DREF level positively correlates with DmCAT promoter activity, consistent with a previous study20; and (2) either depleting or overexpressing DREF renders the DmCAT promoter nonresponsive to PH3 treatment, suggesting that DREF is the critical factor for PH3-mediated DmCAT repression. Thus, the down-regulation of DREF by PH3 treatment is a prerequisite for repression of DmCAT.

DRE and the DRE-binding factor DREF are recognised as a multifunctional system with important roles in D. melanogaster, including tumour suppression, cell development, tissue growth, chromatin organisation, and mitochondrial biogenesis21. Therefore, it is interesting to know whether other DRE/DREF target genes are also down-regulated by PH3 and, if so, what are their roles in PH3 toxicity. Our results showed that two DREF target genes related to mitochondrial function were down-regulated after PH3 treatment (Fig. 5). Considering that PH3 could also disrupt mitochondrial function and inhibit respiration in invertebrate organisms6, it is possible that the DRE/DREF system is also involved in this process and is essential for relaying the PH3 signal to its targets genes. This will require comprehensive analysis of transcriptional changes upon PH3 treatment. Additional questions, which will be explored in future studies, include the exact down-regulation of DREF by PH3.

Investigations into how PH3 disrupts the antioxidant defence system and leads to the generation of ROS have produced important results in the field, especially in the context of emerging PH3 resistance. First reported more than two decades ago, PH3 resistance has now been observed in many countries. It may become a worldwide problem in the future, considering its predominant use and the consequent likelihood of imposing strong selection for PH3 resistance in pests. Studies on the molecular mechanism of PH3 toxicity will provide information to help address PH3 resistance, ultimately benefitting industries that rely on pest control.

Materials and Methods

Preparation of the PH3 solution

Gaseous PH3 was generated by adding aluminium phosphide tablets to acidified water28. To make the PH3 solution, PH3 gas (1 mL) was injected into a sealed, sterile Wheaton narrow-mouth bottle (Z250007; Sigma, St. Louis, MO, USA) containing 1 × PBS, pH 7.4 (15 mL), through a Mininert Valve (33304; Sigma). The bottle was inverted overnight at 28 °C to allow it to equilibrate. The concentration of the PH3 solution was determined to be 14 µg/mL using a QuantiChrom Phosphate Assay Kit (BioAssay Systems, Hayward, CA, USA) according to the manufacturer’s protocol.

S2 cell culture and PH3 treatment procedures

D. melanogaster S2 cells were cultured in TC100 medium supplemented with heat-inactivated FBS (10%) and antibiotics at 28 °C. For cell viability assays, the cells were seeded into a 96-well plate at a density of 0.5 × 105 per well. After 24 h, the medium was changed to fresh medium, and cells were treated with PH3 at a concentration of 1.4, 7, or 14 µg/L for 0, 12, 24, or 48 h. At each time point and for each treatment concentration, cell viability was evaluated using the CellTiter 96® AQueous One Solution Cell Proliferation Assay System (Promega, Madison, WI, USA)29.

For real-time PCR and western blot assays, S2 cells were seeded into a 6-well plate at a density of 5 × 105 per well. Cells were treated with PH3 at a concentration of 14 µg/L for 1, 2, or 4 h, and then were harvested for RNA and protein extraction.

Luciferase reporter construction and luciferase assay

There is only one identified gene encoding for CAT in D. melanogaster. The full-length promoter of DmCAT (−1,944 to 0 bp, relative to the transcriptional start site of DmCAT) was amplified with the following two primers: Full-Pcat-S-XhoI, 5′-CTCGAGACCTGGGTTTATGGGCTAA-3′; Full-Pcat-A-BglII, 5′-AGATCTGTAGTCAATCAACTGATTGGA-3′20. Differently truncated fragments of the DmCAT promoter were amplified with the forward primers shown in Table 1, and Full-Pcat-A-BglII was used as the reverse primer. To mutate the DRE sequence, which corresponds to positions −85 to −78 bp, the forward primer −94M-Pcat-S-XhoI, carrying three nucleotide mutations, was used.

Table 1 Primers for luciferase reporter construction.
Full size table

The amplified fragments were cloned into the pGL4.11 [luc2P] vector with XhoI and BglII, resulting in pGL4.11-pCAT [luc2P]. This reporter, together with the control vector pGL4.74 [hRluc/TK], was co-transfected into S2 cells for 48 h30. S2 cells were then treated with PH3 for another 4 h, followed by detection of luciferase activity with the Dual-Luciferase Reporter Assay System (Promega).

Real-time PCR

Total RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. Contaminating genomic DNA was removed using DNase I (Takara, Kusatsu, Japan). RNA (2 µg) was reverse transcribed using M-MLV Reverse Transcriptase (Promega). Real-time PCR was carried out with SYBR Green Real-time PCR master mix (Toyobo, Osaka, Japan). RP49 was used as an endogenous control. The primers used are listed in Table 2.

Table 2 Primers for real-time PCR.
Full size table

Western blot

Cells (106) were harvested, and total protein was extracted with 200 µL lysis buffer (20 mM Tris-HCl pH 8.0, 2 mM MgCl2, 0.2 M sucrose). Then, protein (20 µg) was resolved by 13% SDS-PAGE and transferred to a PVDF membrane (Millipore, Billerica, MA, USA) using wet electro-transfer. CAT, DREF, and endogenous control β-actin were blotted with corresponding antibodies26,27.

ChIP-qPCR

Chromatin immunoprecipitation (ChIP) was performed using the iDeal ChIP-seq kit (Diagenode, Denville, NJ, USA) according to the manufacturer’s instruction with minor modifications. Briefly, 2 × 106 PH3-treated or untreated cells were harvested in PBS and cross-linked with formaldehyde (1%) for 2 min. After sonication, chromatin from 1 × 106 cells was subjected to ChIP with Pol II antibody (ab5408; Abcam, Cambridge, MA, USA). For ChIP-qPCR assays, four pairs of primers were used to amplify fragments at different distances downstream from the DREF transcription start site (TSS) (Table 3). Pol II enrichment was normalised to the constitutively active glyceraldehyde-3-phosphate dehydrogenase promoter region using the Drosophila Positive Control Primer Set Gapdh1 (Active Motif, Carlsbad, CA, USA).

Table 3 Primers for chromatin immunoprecipitation (ChIP)-qPCR.
Full size table

RNAi in S2 cells

RNAi was performed according to methods described by Fernández-Moreno et al. and Sawado et al.23,31. To make double-stranded RNAs, plasmid templates containing the corresponding target sequences were first constructed. For DREF, a fragment ranging from nucleotide positions 720–1,321 flanked by T7 promoter sequences at both ends was ligated into pUC19. For mock RNAi, a LacZ sequence, a fragment ranging from nucleotide positions 2,528–348 of the pUC19 vector, flanked by T7 promoters was ligated to pUC19. These templates were in vitro-transcribed using the MEGAscript® RNAi kit (Ambion, Austin, TX, USA) according to the manufacturer’s protocol. Next, in vitro-transcribed dsRNA (100 µg) was purified. For dsRNA treatment, 1.5 µg of dsRNA per 105 S2 cells was directly added to unsupplemented TC100 medium. After 1 h of incubation, medium was changed to complete medium. After 24 h, the cells were transfected with luciferase reporters; 48 h later, cells were collected for real-time PCR, western blot, or luciferase activity assays.

Transient gene overexpression in S2 cells

The full open reading frame of the D. melanogaster DREF gene (NM_078805.4) was synthesised by Beijing Protein Innovation (Beijing, China) and ligated into the pAc5.1/V5-His vector with XhoI and XbaI. The expression vector was then transfected into S2 cells together with the luciferase reporter for 48 h. The transfected cells were then treated with PH3 for another 4 h, followed by real-time PCR, western blot, or luciferase activity assays.

Statistical analysis

All data were generated from at least three independent replicate experiments and processed with Microsoft Excel (Microsoft Corporation, Redmond, WA, USA). Final data are expressed as means ± SE. One-way ANOVA and LSD test were used to test for statistically significant differences at p < 0.05.

Data availability

All data generated or analysed during this study are included in this published article.