Identification of midgut-specific promoters in FAW
In model insects, including D. melanogaster and Bombyx mori, a number of midgut-specific promoters have been identified and used for tissue-specific expression of transgenes (Fischer et al. 1988; Hu et al. 2015; Jiang et al. 2013; Park and Kwon 2011; Zeng et al. 2010). Midgut-specific promoters were also identified in several mosquito species (Moreira et al. 2000; Nolan et al. 2011; Skavdis et al. 1996; Zhao et al. 2014). Although some midgut-specific promoters showed cross-species activity (Skavdis et al. 1996), it is still not feasible to use midgut-specific promoters in non-host insects. Information on midgut-specific promoters in non-model pest insects is not available.
Several highly active promoters from FAW have been identified, and their activity in vivo was determined (Chen et al. 2020a). In this study, a group of eight highly expressed genes in the midgut and other tissues and a group of 16 genes that are expressed only in the midgut were identified in RNA-Seq data. The mRNA levels of these genes were investigated by RT-qPCR. All eight highly expressed genes showed significantly higher expression levels in the midgut than in the other tissues. The expression levels of SfCPH38 in the midgut is 2,048,465-fold higher than in other tissues (Figure 1A). In the group of midgut-specific genes, two genes, SfZCPase2 and SfmALP, did not express in other tissues. However, their expression levels in the midgut are much lower than the other 14 genes. Among these 14 genes, SfTrypsinC showed the highest specificity in the midgut, 1,842,596-fold higher than in other tissues (Figure 1B). The overall relative expression levels of 16 midgut-specific genes are much lower than those of eight highly expressed genes. NCBI accession numbers of these genes are provided in Table S2.
Following the method described in our previous paper (Chen et al. 2020a), the putative promoter regions of 11 midgut specific genes, including SfTrypsinC/P844, SfSP/P415, SfLipase3/P1558, SfCBP-I/P360, SfCalphotin/P799, SfTrypsin/P1705, SfCptlike/P857, SfMG17/P2000, SfSP11/P2000, SfSP38/P2000, and SfCPH38/P2000 were amplified from the genomic DNA and cloned.
Determination of the specificity of candidate FAW midgut promoters in vivo using baculovirus
Two FAW ovarian cell lines, Sf9 and Sf21, and one FAW midgut cell line Sf17 were used to investigate the FAW promoter activity. The FAW promoters from highly expressed genes displayed weak or no activity in these cell lines (Bleckmann et al. 2015; Chen et al. 2020a). The cell lines are not ideal for investigating the activity of tissue-specific promoters, which is likely due to the lack of essential transcription factors required for the activation of tissue-specific genes. Also, the promoter activity in the cell line may not always match with that in the insect in vivo. Recently, the baculoviruses were used to evaluate the performance of B. mori promoters in vivo (Tian et al. 2018; Zhang et al. 2015; Zhao et al. 2014).
To test whether the baculovirus could be used to measure the promoter activity in FAW, a reporter baculovirus expressing the luciferase gene under the control of the promoter region of an early baculovirus gene, pag1, was constructed. This reporter baculovirus, pag1:Luc, was then added to the Sf9 cells and Sf17 cells. Both cells showed luciferase activity at 24 hr post-infection, and the luciferase activity in virus-infected Sf9 cells is much higher than in virus-infected Sf17 cells (Figure S1). The virus was then injected into the newly molted 6th instar larvae. Midgut and other tissues were dissected on 1, 2, 3, 4, and 5 days post-injection. In the midgut, the maximum luciferase activity was detected on day 3. In samples containing all other tissues, the maximum luciferase activity was detected on day 4 (Figure S2). The luciferase activity was not detected in uninfected wild-type larvae or those infected with another reporter baculovirus expressing EGFP (data not shown). These results indicated that reporter baculovirus could be employed to evaluate FAW promoter performance in vivo. Since the baculoviruses could quickly spread to different tissues of FAW, we hypothesized that the reporter baculovirus carrying tissue-specific promoter may support tissue-specific transgene expression in infected FAW larvae. The reporter baculoviruses expressing the luciferase controlled by 11 candidate midgut-specific promoters were generated. The plaque-forming units (PFU) of each reporter baculovirus was determined by RT-qPCR (Figure S3). Unlike the pag1:Luc baculovirus, most of these 11 reporter baculoviruses showed higher luciferase activity in the Sf17 cells than in Sf9 cells (Figure 2A). These reporter baculoviruses were injected into the hemocoel of FAW larvae, and the luciferase activity in the midgut and other tissues was measured on day 3 after injection. The result showed that weak or no luciferase activity was detected in the midgut while the other tissues showed some luciferase activity (Figure 2B). The reporter baculoviruses failed to show the expected midgut-specific expression of the reporter gene, which is likely due to the injection of the virus into the hemocoel. Also, baculovirus infection may have altered cellular gene expression patterns (Blissard 1996; Clem and Passarelli 2013), making endogenous promoters inactive and/or changes in their tissue specificity.
Determination of the specificity of candidate FAW midgut promoters in vivo using transgenic insects
In recent years, the transgenic insects have been employed to evaluate the B. mori and A. aegypti promoters in vivo (Anderson et al. 2010; Deng et al. 2013; Jiang et al. 2013; Moreira et al. 2000; Totten et al. 2013; Xu et al. 2019; Xu et al. 2014). However, germline transformation technologies are not well established in non-model insects, likely due to the low transformation efficiency and difficulty of delivering transformation components into the fresh embryos. We recently established piggyBac-based transformation system in FAW (Chen et al. 2020b), which was used to test FAW promoters in vivo.
A piggyBac vector, expressing a marker protein EGFP under the control of ie1 promoter, was used to make the reporter vectors. The promoters from the genes that are highly expressed in the midgut were cloned into this piggyBac vector to drive the luciferase expression. The hr5 enhancer, which is capable of boosting promoter activity, was placed between ie1 promoter and chosen midgut-specific promoter to enhance the expression of both the EGFP and the luciferase (Figure 3A). Four reporter vectors, hr5-SfSP38/P2000:Luc, hr5-SfMG17/P2000:Luc, hr5-SfCalphotin/P799:Luc, and hr5-SfCPH38/P2000:Luc were produced. A mixture of each reporter vector and the mRNA of hyperactive transposase, which was reported to increase the transformation efficiency in several insect species (Eckermann et al. 2018; Otte et al. 2018), was injected into 2000 fresh embryos collected within 6 hr after oviposition. Forty-six transgenic hr5-SfSP38/P2000: Luc neonate larvae, 31 transgenic hr5-SfMG17/P2000:Luc neonate larvae, 56 transgenic hr5-SfCalphotin/P799: Luc neonate larvae and 41 transgenic hr5-SfCPH38/P2000:Luc neonate larvae were identified based on the EGFP marker gene expression in the G1 generation (Table 1). The adults developed from G1 positive larvae were crossed with wild-type adults to produce G2 generation. The positive transgenic larvae from all four lines showed strong GFP signals in G2 (Figure 3B), indicating that the transgenic insertions are inheritable. We then tested the luciferase activity in four tissues, including the head, epidermis, fat body, and midgut dissected from each transgenic line. The results showed that the midgut has significantly higher luciferase activity than in other tissues. The luciferase activity in the midgut is 1,731-, 1,558-, 2,337-, and 14,124-fold higher than in fat body of hr5-SfSP38/P2000:Luc, hr5-SfMG17/P2000:Luc, hr5-SfCalphotin/P799:Luc, and hr5-SfCPH38/P2000:Luc transgenic larvae, respectively (Figure 3C). SfCalphotin/P799 promoter displayed higher activity in the midgut and fat body than the other three promoters tested. SfCPH38/P2000 promoter showed the best specificity in the midgut. Low luciferase activity was also detected in other tissues, indicating the weak activity of these promoters in other tissues (Figure 3C).
Table 1
Germline transformation of transgenic constructs.
Plasmids
|
Injected eggs
(n)
|
G0 Larvae
(n)
|
G0 Pupae
(n)
|
G1 Positive
(n)
|
hr5-SfSP38/P2000: Luc
|
~2200
|
~800
|
~600
|
46
|
hr5-SfMG17/P2000: Luc
|
~2400
|
~900
|
~700
|
31
|
hr5-SfCalphotin/P799: Luc
|
~2400
|
~1000
|
~700
|
56
|
hr5-SfCPH38/P2000: Luc
|
~2000
|
~800
|
~600
|
41
|
SfCPH38/P2000: Luc
|
~2000
|
~700
|
~600
|
29
|
hr5-SfCPH38/P2000: CYP321A8
|
~2400
|
~600
|
~500
|
38
|
SfCPH38/P2000: CYP321A8
|
~2400
|
~800
|
~600
|
79
|
As promoter fused with hr5 enhancer showed elevated activity (Chen et al. 2020a), we hypothesized that removing the hr5 enhancer might decrease the promoter activity in other tissues and increase its specificity in the midgut. To test this hypothesis, a new reporter vector, SfCPH38/P2000:Luc, was constructed (Figure 3A) injected into 2000 eggs. Twenty-nine EGFP-positive neonate larvae were obtained (Table 1). The EGFP signals in transgenic SfCPH38/P2000:Luc G2 positive larvae were not bright as transgenic hr5-SfCPH38/P2000:Luc G2 positive larvae (Figure 3B), as hr5 enhancer also boosts ie1 promoter activity resulting in the production of more EGFP protein. The luciferase activity in the midgut is 13,152-fold higher than in the fat body of SfCPH38/P2000:Luc larvae. Also, the luciferase activity in SfCPH38/P2000:Luc larval tissues is lower than in the tissues of hr5-SfCPH38/P2000 Luc larvae (Figure 3C). These results suggest that hr5 enhancer could significantly enhance the activity, but not the midgut-specificity, of CPH38 promoter.
Midgut-specific promoter mediated over-expression of P450 confers deltamethrin tolerance to S. frugiperda
P450 enzymes metabolize insecticides and plant toxins (Dermauw et al. 2020; Feyereisen 2012), and their overexpression is involved in insecticide resistance (Feyereisen 2012; Jiang et al. 2015). Although many P450 genes were found in the genome of FAW (Gui et al. 2020), little is known about their function in insecticide resistance. A recent study revealed that overexpression of CYP321A8 in Spodoptera exigua, a close relative of S. frugiperda, conferred resistance to deltamethrin (Hu et al. 2021). The insect midgut is the major organ encountering plant toxins and insecticides (Hakim et al. 2010; Smagghe and Tirry 2001). Transgenic overexpression of P450 genes in the midgut likely confers tolerance to insecticides. To test whether the identified midgut-specific promoters could be used in transgenic FAW to study insecticide resistance, SfCPH38/P2000 promoter was used to drive expression of SfCYP321A8 gene in the midgut.
The SfCYP321A8 gene was cloned. Two vectors, SfCPH38/P2000:CYP321A8 and hr5-SfCPH38/P2000:CYP321A8 were produced and injected into embryos (Figure 4A). Seventy-nine transgenic SfCPH38/P2000:CYP321A8 neonate larvae and 38 transgenic hr5-SfCPH38/P2000:CYP321A8 neonate larvae were identified based on the EGFP marker gene expression in the G1 generation (Table 1). The EGFP signals in transgenic SfCPH38/P2000:CYP321A8 G2 positive larvae are less bright than the transgenic hr5-SfCPH38/P2000:CYP321A8 G2 positive larvae (Figure 4B).
The relative mRNA levels of SfCYP321A8 were determined in the head, midgut, fat body, and epidermis dissected from the 6th instar larvae of wild-type and two G2 SfCYP321A8 transgenic insects. The highest expression levels of SfCYP321A8 mRNA were detected in the midgut of both transgenic lines. An increase by 155.5- and 287.6-fold in the SfCPH38/P2000:CYP321A8 and hr5-SfCPH38/P2000:CYP321A8 mRNA levels transgenic larvae, respectively, when compared to their levels in the wild-type larvae were detected (Figure 4C). The expression also increased by 6.3- and 10.6-fold in the fat body of SfCPH38/P2000:CYP321A8 and hr5-SfCPH38/P2000:CYP321A8 transgenic larvae, respectively. In the epidermis of transgenic hr5-SfCPH38/P2000:CYP321A8 larvae, the mRNA level of SfCYP321A8 was slightly higher than in wild-type and control transgenic larvae. No significant changes in SfCYP321A8 expression were detected in the epidermis of both transgenic larvae (Figure 4C). To test whether overexpressing of SfCYP321A8 in the midgut leads to an overall increase in total P450 activity in the midgut of the transgenic animals; the midgut and fat body were tested for P450 activity. The total P450 activity in the midgut dissected from SfCPH38/P2000:CYP321A8 and hr5-SfCPH38/P2000: CYP321A8 transgenic larvae increased by 1.41- and 1.48-fold, respectively, when compared to the P450 activity in the wild-type larvae. However, no increase in P450 activity was detected in the fat body of both transgenic larvae (Figure 4D). These results suggest that SfCPH38/P2000 promoter is capable of driving midgut-specific expression of SfCYP321A8.
The expression levels of both SfCPH38 and SfCYP321A8 were investigated in the eggs and 1st instar larvae of wild-type and transgenic lines. Expression of SfCYP321A8 in 1st instar larvae of both transgenic lines gradually increased similar to the SfCPH38 gene expression (Figure S4). Leaf-disc assays were performed to evaluate the efficacy of deltamethrin in wild-type and transgenic 1st instar larvae. The results showed that the LC50 of deltamethrin in SfCPH38/P2000:CYP321A8 and hr5-SfCPH38/P2000: CYP321A8 larvae increased by 4.2- and 4.5-fold, respectively, when compared to the wild-type larvae (Figure 4E). These data suggest that transgenic overexpression of SfCYP321A8 in the midgut increased deltamethrin tolerance in FAW larvae.
Identification and characterization of midgut-specific promoters from a non-model insect, FAW are included in this paper. The de novo mining of tissue-specific promoters needs gene expression data, which is not available for many insects. Since the information on many tissue-specific promoters from the model insects is available, it may be possible to identify the homologous promoters in the target non-model insects. In the lepidopteran model insect, B. mori, promoter of APN gene, is only active in the midgut (Jiang et al. 2015). We found six APN genes in S. frugiperda. RT-qPCR analysis revealed midgut-specific expression of all APN genes (Figure S5). The candidate promoter region of the SfAPN6 gene was cloned and used to drive the expression of SfCYP321A8. Transgenic SfAPN6/P1949:CYP321A8 animals were obtained (Table S3, Figure S6A). However, the mRNA levels of SfCYP321A8 only increased by 2.3-fold in the midgut of this transgenic line than in the wild-type (Figure S6B). Also, this transgenic line showed similar levels of deltamethrin susceptibility as the wild-type (Figure S6C). The candidate promoter region of SfSP11, homolog of a midgut-specific serine protease in B. mori (Liu et al. 2017), was also tested to drive expression of SfCYP321A8 in animals (Table S3, Figure S6A). A 16.7-fold increase of SfCYP321A8 mRNA levels was detected in the midgut of transgenic animals than in the wild-type (Figure S6B). However, no increase in deltamethrin tolerance was observed (Figure S6C). Since these two midgut-specific promoters of FAW were identified from information from B. mori did not show good activity as SfCPH38/P2000 promoter, our hypothesis on using information from model insects to identify promoters in non-model insects is not supported. We used ~2 kp region upstream to the ATG of each gene as candidate promoter region. The low activity of SfAPN6 and SfSP11 promoters might be because this region did not include complete promoter.
Transgenic reporter insects have been used to investigate the promoter performance in vivo. However, germline transformation technologies are not available in most non-model insects. We tried to use baculoviruses to deliver the reporter gene expression cassettes to different tissues of FAW larvae but failed to determine the activity of midgut-specific promoters in vivo. A recent report revealed one baculovirus species, Autographa californica multiple nucleopolyhedrovirus (AcMNPV), is an efficient vector for gene delivery into several mosquito species; it can transduce both larvae and adults with little or no tissue barriers and without obvious negative effects (Naik et al. 2018). It seems that baculovirus could be used to investigate the promoter performance in non-permissive hosts in vivo.
In conclusion, transgenic reporter insects were successfully used to determine the in vivo performance of midgut-specific promoters in an important agricultural pest, FAW. Midgut-specific promoter was used to investigate the role of P450 in insecticide resistance. This work could serve as a model for exploring other tissue-specific promoters, which will benefit functional genomic studies in FAW and other non-model insects.