Distinct functional properties of sodium channel variants are associated with usage of alternative exons in Nilaparvata Lugens

Abstract

Voltage-gated sodium channels (Na_v) are essential for electrical signaling in the nervous system. They are also the primary targets of several classes of insecticides including pyrethroids. There is only one sodium channel gene in most insect species, whereas mammals possess at least nine sodium channel genes. Extensive alternative splicing and RNA editing of sodium channel transcripts have been documented in many insect species. However, the functional consequences of these post-transcriptional events have been evaluated only in DmNa_v and BgNa_v from Drosophila melanogaster and Blattella germanica, respectively. In this study, we isolated 41 full-length cDNA clones encoding 34 sodium channel (NlNa_v) variants from a major rice pest, the brown planthopper (Nilaparvata lugens Stål). The 34 NlNa_v variants represent 24 distinct splicing types based on the usage of nine alternative exons, six of which, including exon b, have been previously reported in other insect species. When expressed in Xenopus oocytes, NlNa_v variants lacking exon b generated significantly larger sodium currents than variants possessing exon b, suggesting an inhibitory effect of exon b on sodium current expression. A similar effect has been reported for exon b from BgNa_v. Mutational analysis showed that three conserved amino acid residues encoded by exon b are critical for its inhibitory effect. In addition, mutually exclusive exons k/l contribute to distinct functional properties and channel sensitivity to pyrethroids. Altogether, these results show that alternative splicing generates functional diversity of sodium channels in this insect species and that the role of exon b in regulating neuronal excitability is likely conserved among insect species.

Graphical Abstract

1. Introduction

Voltage-gated sodium channels (VGSC) are responsible for the generation and propagation of action potentials in most excitable cells (Catterall, 2012). Sodium channels are also the targets of various natural and synthetic neurotoxins including DDT, pyrethroid insecticides and sodium channel blocker insecticides (Dong et al., 2014; Silver et al., 2014). Mammalian sodium channel isoforms are encoded by at least nine genes, which exhibit distinct expression patterns in various nerve cells, tissues and developmental stages (Frank and Catterall, 2003; Goldin et al., 2000). However, in insects there is only one sodium channel gene in the genome of most species (Davies et al., 2007; Severson et al., 2004; Shao et al., 2009; Xue et al., 2014) with a few exceptions (Amey et al., 2015; Jiang et al., 2017; Moignot et al., 2009). The first insect sodium channel gene, para (also known as DmNa_v), was isolated from Drosophila melanogaster (Feng et al., 1995; Loughney et al., 1989). Another putative sodium channel gene (called Drosophila Sodium Channel 1, DSC1) was identified in D. melanogaster (Littleton and Ganetzky, 2000; Salkoff et al., 1987), but was later functionally confirmed to encode a Ca²⁺-selective cation channel functionally distinct from sodium channels (Dong et al., 2015; Zhang et al., 2011). Because one major mechanism of pyrethroid resistance, knockdown resistance (kdr), is caused by mutations in sodium channels, there have been many reports of isolation and characterization of orthologs of the para gene from a wide range of arthropod pests and disease vectors for investigation of the mechanisms of insecticide resistance (Du et al., 2016; Field et al., 2017; Rinkevich et al., 2013; Silva et al., 2014).

The Drosophila sodium channel gene, para, undergoes extensive alternative splicing (Lin et al., 2009; Loughney et al., 1989; Olson et al., 2008; Thackeray and Ganetzky, 1994). A total of 15 alternatively spliced exons have been reported, including 11 optional exons (a, b, e, f, h, i, j, 7, 8, 22, 23) in intracellular linkers and four mutually exclusive exons (c/d and k/l) in transmembrane domains (Lin et al., 2009; Olson et al., 2008). Most of these alternative splicing sites are also conserved in D. virillis (Thackeray and Ganetzky, 1995). Some of the alternative exons identified in Drosophila species are also highly conserved in other insect species, such as mosquito, house fly, cockroach and bee, especially optional exons a, b and j, and mutually exclusive exons c/d and k/l (Chang et al., 2009; Davies et al., 2007; Field et al., 1999; Jiang et al., 2013; Kwon et al., 2010; Lee et al., 2002; Lin et al., 2009; Olson et al., 2008; Silva and Scott, 2019; Song et al., 2004; Sonoda et al., 2008; Tan et al., 2002a; Thackeray and Ganetzky, 1994, 1995; Tsagkarakou et al., 2009; Wang et al., 2003; Zhen and Gao, 2016). Particularly, a recent study (Silva and Scott, 2019) conducted a comprehensive comparative analysis of sodium channel genes from 68 insect species covering 10 orders and revealed significant conservation of alternative exons that were first identified in D. melanogaster (Loughney et al., 1989). The high conservation of these alternative splicing events suggests possible biological importance of alternative splicing of sodium channel genes in insects. In addition to alternative splicing, RNA editing is another post-transcriptional modification that increases the molecular and functional diversity of sodium channels in insect species (Hanrahan et al., 2000; Liu et al., 2004; Palladino et al., 2000; Reenan et al., 2000; Song et al., 2004). Particularly, extensive RNA editing has been reported in the DmNa_v transcripts (Hanrahan et al., 2000; Palladino et al., 2000; Reenan et al., 2000). So far, functional characterization of alternative spliced/RNA edited sodium channel variants in Xenopus oocytes has been carried out only in D. melanogaster (DmNa_v) and the German cockroach Blattella germanica (BgNa_v) (Hanrahan et al., 2000; Lee et al., 2002; Lin et al., 2009; Olson et al., 2008; Palladino et al., 2000; Reenan et al., 2000; Song et al., 2004). Collectively, these studies showed that alternative splicing and RNA editing generate functionally and pharmacologically distinct DmNa_v and BgNa_v channels (Baines and Lin, 2017; Lin et al., 2009; Olson et al., 2008; Song et al., 2004). For example, exclusion or inclusion of the optional exon b modulates the sodium current expression of BgNa_v channels in Xenopus oocytes (Song et al., 2004). Both A-to-I and U-to-C RNA editing events contribute to the functional diversity of BgNa_v channels (Song et al., 2004; Liu et al., 2004). However, it remains to be determined whether the functional consequences of alternative splicing and RNA editing events revealed from the analyses of DmNa_v and BgNa_v variants are conserved in other insect species.

In this study, we conducted molecular and functional characterization of sodium channel variants from the brown planthopper (Nilaparvata lugens Stål). Specifically, we isolated 41 full-length cDNA clones representing 24 distinct splicing types. Our functional analysis of 11 variants expressed in Xenopus oocytes revealed significant differences in channel functional properties and channel sensitivity to pyrethroids among the variants. Moreover, we identified three residues encoded by one conserved alternative exon, exon b, are crucial for exon b-mediated regulation of sodium current expression in Xenopus oocytes.

2. Materials and methods

2.1. The brown planthopper

The insecticide-susceptible strain of the brown planthopper was obtained from the China National Rice Research Institute in September 2001 and reared on rice crop without any contact of insecticides as a laboratory strain in an incubator at 25 ± 1 °C, humidity 70% – 80% and 16 h light/8 h dark photoperiod.

2.2. Isolation of full-length NlNa_v cDNA clones by reverse transcription- polymerase chain reaction (RT-PCR)

Total RNA was isolated from two adults and two 5^th instar nymphs of the insecticide-susceptible strain using TRIzol Reagent (Invitrogen, CA, USA) according to the manufacturer’s instructions. Developmental regulation of alternative splicing of insect sodium channels has been documented (Lee et al., 2002). In this study we used both adults and nymphs which was intended to identify maximal numbers of splice types. RNA integrity was checked by agarose gel electrophoresis (1%) and RNA concentration was estimated by measuring OD₂₆₀ with NanoDrop2000 (Thermo Scientific, MA, USA). First-strand cDNA was synthesized from 1 μg of total RNA using PrimeScript II RTase (Takara, Kyoto, Japan) and random 6mers, using the following cycling parameters: 30 °C for 10 min, and then 95 °C for 5 min. The full-length NlNa_v gene and NlTipE gene were amplified by RT-PCR with a sense primer (5’-CTTTCGGAGACAGAGAGAGCGTATCA-3’ for NlNa_v and 5’-CTTGAGGAGACCCGGATCATAATG-3’ for NlTipE) and an anti-sense primer (5’-TACAACAACGTCTCAGGTGCTAGGTG-3’ for NlNa_v and 5’-CGTAGTCTACGTTGTCTAATAATCG-3’ for NlTipE), which were designed according to the genome sequence of the brown planthopper (GenBank accession number AOSB00000000, BioProject PRJNA177647). The PCR products were purified and cloned into the pGH19 expression vector using ClonExpress II One Step Cloning Kit (Vazyme, Jiangsu, China). The ligated DNA was transformed into E. coli DH5α competent cells (TransGen Biotech, Beijing, China) and 41 full-length cDNA clones were isolated. The inserts were sequenced at GenScript Co. Ltd. (Nanjing, Jiangsu, China).

2.3. Site-Directed Mutagenesis

S772, Y774 and Y775 were highly conserved among 20 arthropod species (see below) and they could regulate sodium current expression by phosphorylation (Song et al., 2004). In this case, we made alanine substitutions of S772, Y774 or Y775 in an exon b-containing variant, NlNa_v2–1 (GenBank accession number: MN480445), which is one of the most common splice type. Site-directed mutagenesis was performed by Polymerase Chain Reaction (PCR) using mutant primers and Pfu Turbo DNA polymerase (Agilent, CA, USA) with the following cycling parameters: 95 °C for 2 min followed by 18 cycles of 95 °C for 30 s, 50 °C for 30 s and 72 °C for 20 min, then followed by 72 °C for 10 min, 4 °C forever. The forward and reverse primer of each mutation were complementary and carried with mutant sites. The region where mutations were introduced was sequenced.

2.4. Expression of NlNa_v sodium channel in Xenopus oocytes

Ovaries from oocyte-positive female Xenopus laevis were purchased from Xenopus 1 (Dexter, MI, USA) and cRNAs were produced by in vitro transcription with T7 polymerase using the mMESSAGE mMACHINE Kit (Ambion, TX, USA). The TipE gene, first found in D. melanogaster (Feng et al., 1995; Warmke et al., 1997), and its orthologs in other insects (Bourdin et al., 2013; Du et al., 2013) are known to facilitate robust expression of insect sodium channels in Xenopus oocytes. Therefore, NlNa_v cRNA was co-injected into oocytes with the brown planthopper NlTipE cRNA at a 1:1 molar ratio. The procedures for oocytes preparation and cRNA injection were identical to those described previously (Tan et al., 2002a; Tan et al., 2002b). Injected oocytes were incubated in ND96 at 15°C for 1–4 days.

2.5. Electrophysiological recording and statistical analysis

All oocyte recordings were performed at room temperature (20–24 °C) in ND96 bath solution by two-electrode voltage clamp using an OC725C oocyte clamp (Warner Instruments, CT, USA) and a Digidata 1440A interface (Axon Instruments, CA, USA) as those previously described (Tan et al., 2005). pCLAMP 10.2 software (Axon Instruments, CA, USA) was used for data acquisition and analysis. The voltage-dependence of activation, fast inactivation, slow inactivation and recovery from inactivation were determined following the protocols previously described and fitted with a Boltzmann equation to generate V_1/2, the midpoint of the activation or inactivation, and k, the slope factor (Tan et al., 2002a; Tan et al., 2002b).

The peak current of NlNa_v sodium channel variants was measured by a 20-ms depolarization to −15 mV from the −120 mV holding potential at 17 or 70 hours after injection of 0.8 ng of cRNA for NlNa_v sodium channel variants or 1 ng of cRNA for NlNa_v2–1 and its recombinants. Results were reported as mean ± SEM. Statistical significance was determined by using one-way analysis of variance with Scheffe’s post hoc analysis, Student’s t-test analysis and significant values were set at P < 0.05.

2.6. Insecticides

Etofenprox, permethrin (a mixture of 1R-cis, 1R-trans, 1S-cis and 1S-trans isomers,) and deltamethrin were purchased from Sigma-Aldrich (Sigma-Aldrich, MO, USA). The purities of all insecticides were greater than 95%.

2.7. Measurement of tail currents induced by pyrethroids

Pyrethroids were dissolved in dimethyl sulfoxide (DMSO) in a stock concentration of 100 mM. The working solution (at the concentration of 1 μM) was prepared in ND96 recording solution just prior to experiments. The concentration of DMSO in the final solution was < 0.5%, which had no effect on the function of sodium channel. The methods of applying pyrethroids for measurement of pyrethroid-induced tail currents were identical to those described previously (Tatebayashi and Narahashi, 1994). The pyrethroid-induced tail currents were recorded during a 100-pulse train of 5-ms step depolarizations from −120 to 0 mV with 5-ms interpulse intervals 10 min after pyrethroids application. The percentage of channels modified by pyrethroids was calculated using the formula (Tatebayashi and Narahashi, 1994).

$$\text{M}=\left\{\left[{\text{I}}_{\text{tail}}/\left({\text{E}}_{\text{h}}-{\text{E}}_{\text{Na}}\right)\right]/\left[{\text{I}}_{\text{Na}}/\left({\text{E}}_{\text{t}}-{\text{E}}_{\text{Na}}\right)\right]\right\}\times 100$$

Where I_tail is the maximal tail current amplitude, E_h is the potential to which the membrane is repolarized, E_Na is the reversal potential for sodium currents determined from the current-voltage curve, I_Na is the amplitude of the peak current during depolarization before pyrethroids exposure, and E_t is the potential of the step depolarization.

3. Results and discussion

3.1. Identification of nine alternative exons in NlNa_v

We isolated 41 full-length cDNA NlNa_v clones using RT-PCR. Comparison between the genomic sequence of the NlNa_v gene (GenBank accession number: XM_022340088.1) and NlNa_v cDNA variants revealed a total of nine alternative exons, which are named based on the nomenclature of the DmNa_v exons (Olson et al., 2008). The positions of these alternative exons are indicated in Fig. 1A. Based on the usage of the alternative exons, we grouped the 41 clones into 24 unique splice types (Fig. 1B). The amino acid sequences encoded by these alternative exons are presented in Fig. 1C and are compared to those of DmNa_v and BgNa_v.

Fig. 1.

Identification of 24 alternative splice types of NlNa_v. A: Schematic drawing of the NlNa_v protein topology indicating the locations of alternative exons using patterned boxes. Exons j, 13, E⁶⁰⁶, b, e, f and 36 are optional, whereas exon k/l is mutually exclusive. B: Usage of the identified alternative exons in a total of 41 full-length cDNA clones that were isolated in this study. The variants were named according to the splice types. The four clones in the first splice type were identical in sequence and therefore were named as NlNa_v1. Two clones in the second splice type were identical. Therefore, the variants in splice type 2 were named as NlNa_v2–1, NlNa_v2–2 and NlNa_v2–3. All three clones in the third splice type were identical and were named as NlNa_v3. The two or three clones in each of splice types 4–11 were different except for the two clones in splice type 11 which were identical. C: Alignments of the deduced amino acid sequences of alternative exons among para, BgNa_v and NlNa_v. Gaps introduced to preserve optimal alignments are indicated as dashes; the sequence identity are indicated by the asterisks. D: The exon-intron structure of the region where the optional E⁶⁰⁶ is located. Exclusion of E⁶⁰⁶ (boxed) is the result of alternative splicing using 3’ alternative acceptor site and the intron upstream is indicated in shade and lowercase letters. The splicing consensus sites (gt/ag) are indicated in bold. The 3’ alternative acceptor site “AG” is indicated in bold and underlined.

NlNa_v1 (four clones in this splicing type are identical in sequence; GenBank accession number: MN480444) contains optional exons j, 13, e, f and 36, and mutually exclusive exon l and has a 6,276 bp open reading frame (ORF) which encodes a protein of 2,091 amino acid residues. All variants contain four homologous domains (I–IV) and each domain possessing six transmembrane segments (S1–6), except that NlNa_v16 lacks both exons k/l and NlNa_v24 lacks exon 36. All the structural features that are hallmarks of sodium channels are present in the deduced amino acid sequences of these full-length clones, including the DEKA motif (in the loops connecting S5 and S6 of each domains) that determines ion selectivity; a conserved MFM motif (in the linker between domains III and IV), which is critical for fast inactivation (Dong, 2007). All variants except NlNa_v24 encode voltage sensors formed by each S4 segment containing repeated motifs of a positively charged amino acid residue followed by two hydrophobic residues.

Six of the alternative exons, exons j, b, e, f, k, and l, were previously reported in D. melanogaster and other insect species (Lee et al., 2002; Olson et al., 2008; Song et al., 2004). Exon j is optional and encodes a stretch of 10 amino acid residues at the N-terminus. Other optional exons b, e and f encode sequences in the intracellular linkers connecting domains I and II, and domains II and III. Exon f is an optional exon. Exons b and e are the result of alternative splicing using a 3’ alternative acceptor site and a 5’ alternative donor site, respectively. Mutually exclusive exons k/l encode IIIS3–4, equivalent to G2/G1 in BgNa_v (Song et al., 2004; Tan et al., 2002a). Similar to exon k/l in DmNa_v and G2/G1 in BgNa_v, transcripts containing exon l were more abundant than those carrying exon k in NlNa_v gene (Tan et al., 2002a). One NlNa_v (NlNa_v16) lacked both k/l likely producing a truncated protein missing part of domain IV and the N-terminus. Because we amplified and cloned the entire coding region of sodium channel transcripts using RT-PCR, the 24 splice types including the splice type 16 (i.e., NlNa_v16) we identified based on the analysis of the full-length clones (Fig. 1B) represent real splice types of transcripts in vivo. A similar protein truncation has been observed in BgNa_v, but due to the inclusion of a third alternative exon G3 which contains a stop codon (Song et al., 2004; Tan et al., 2002a).

We designated the remaining three alternative exons that have not been reported in other insect species as exon 13, E⁶⁰⁶ and exon 36. Exon 13 is optional and encodes a stretch of five amino acid residues in the first intracellular linker connecting domains I and II. Although the amino acid sequence encoded by exon 13 is unique to N. lugens, the location of exon 13 in the genome corresponds to an optional exon, exon n, in the AaNa_v gene from Aedes aegypti (Chang et al., 2009), optional exon 12 in BmNa_v from Bombyx mori (Shao et al., 2009) and optional exon 2 in BgNa_v from Blattella germanica (Dong, 1997). E⁶⁰⁶, is the result of alternative splicing using a 3’ alternative acceptor site (ag, underlined in Fig. 1D) which induced an internally optional spliced exon encoding a glutamate (E⁶⁰⁶) in the middle of the first intracellular linker connecting domain I and II of NlNa_v channels (Fig. 1D). The same splicing mechanism was previously reported for G¹¹¹¹ in the intracellular linker connecting domains II and III in BgNa_v (Du et al., 2009). The optional exon 36 encodes IVS3–5. Exclusion of this exon is predicted to generate a truncated protein lacking most of domain IV and the C-terminus. We did not find additional alternative splice exons, such as exons i, a and c/d, in isolated NlNa_v clones.

In addition to alternative exons, scattered amino acid changes were detected in individual variants compared with NlNa_v1 (Table S1). One clone, NlNa_v6–2, contained a premature stop codon (A⁵⁷⁹⁴AG to T⁵⁷⁹⁴AG) downstream of IVS6, resulting in lack of 160 amino acids in the C-terminus. It is not clear whether these scattered amino acid changes were caused by errors introduced during RT-PCR cloning or represent potential RNA editing events. These full-length clones were not further functionally characterized in Xenopus oocytes. Because this study was not aimed at functionally characterizing the role of scattered amino acid differences in modulating NlNa_v properties, we did not further confirm whether these differences were caused by RT-PCR errors, a possibility that needs to be examined in future functional studies of NlNa_v variants.

3.2. Distinct gating properties of NlNa_v variants attribute to the inclusion/exclusion of unique alternative exons

To determine gating properties of NlNa_v variants. we co-expressed each variant with NlTipE (GenBank accession number: MN480446) in Xenopus oocytes and recorded sodium currents using two-electrode voltage clamp one to four days after cRNA injection. NlTipE enhanced the amplitude of peak current of NlNa_v channels (Fig. 2A). Twenty-four NlNa_v variants representing all identified 24 splice types were functionally examined. Among them eleven (Table 1) produced sodium currents large enough (1.5–3 μA) for further functional analysis, whereas the remaining 11 variants produced currents that were too small (< 0.5 μA) for functional analysis. Two variants, NlNa_v16 and NlNa_v24 did not generate any detectable current even 4 days after cRNA injection, likely due to the lack of transmembrane regions encoded by exons k/l (NlNa_v16) and exon 36 (NlNa_v24).

Fig. 2.

Differences in voltage dependence of activation, fast inactivation or slow inactivation of NlNa_v variants. A: Sodium peak current trace of NlNa_v1 channels. Current traces with and without NlTipE were shown on the top of Panel A. Current traces from NlNa_v1 in the presence of NlTipE were shown at the bottom of Panel A, which were measured by 20-ms depolarizing potentials from −80 to −5 mV (in 5 mV steps) from a holding potential of −120 mV. B: Voltage dependence of slow-inactivation of NlNa_v15 (exon k⁺, 36⁺) and NlNa_v18 (exon E⁶⁰⁶, f⁺, l⁺, 36⁺). C: Voltage dependence of activation of NlNa_v14 (exon 13⁺, l⁺, 36⁺), NlNa_v15 (exon k⁺, 36⁺) and NlNa_v20 (exon e⁺, k⁺, 36⁺). D: Voltage dependence of activation of NlNa_v5–1 (exon E⁶⁰⁶, e⁺, k⁺, 36⁺), NlNa_v15 (exon k⁺, 36⁺) and NlNa_v20 (exon e⁺, k⁺, 36⁺). E: Voltage dependence of fast-inactivation of NlNa_v19 (exon j⁺, 13⁺, E⁶⁰⁶, l⁺, 36⁺) and NlNa_v20 (exon e⁺, k⁺, 36⁺). F: Voltage dependence of activation of NlNa_v1 (exon j⁺, 13⁺, e⁺, f⁺, l⁺, 36⁺) and NlNa_v17 (exon j⁺, e⁺, f⁺, l⁺, 36⁺). The error bars represent SEM from at least 10 oocytes.

Table 1.

Gating properties of 11 functional NlNa_v variants.

Splice type	Activation		Inactivation		Persistent current (%)	n	Slow-Inactivation		n
	V1/2 (mV)	k (mV)	V1/2 (mV)	k (mV)			V1/2 (mV)	k (mV)
NlNav1	−32.6 ± 0.6a	4.2 ± 0.2	−50.7 ± 0.6abc	5.4 ± 0.1	9.2 ± 0.2	26	−54.3 ± 2.1bc	7.1 ± 0.4	18
NlNav5–1(k)	−32.3 ± 0.4ab	3.3 ± 0.2	−48.8 ± 0.7bcd	4.5 ± 0.1	2.8 ± 0.1*	10	−50.5 ± 1.4c	6.5 ± 0.2	13
NlNav6–1	−30.6 ± 1.0abc	4.3 ± 0.4	−51.1 ± 1.1abc	4.2 ± 0.1	10.3 ± 0.2	9	−51.2 ± 1.0bc	6.2 ± 0.5	9
NlNav8–1	−33.5 ± 1.1a	4.4 ± 0.1	−53.3 ± 1.1ab	4.2 ± 0.1	8.7 ± 0.1	8	−53.1 ± 1.0bc	4.8 ± 0.5	10
NlNav11–1	−32.2 ± 0.7ab	3.7 ± 0.4	−47.9 ± 0.6cd	4.3 ± 0.2	9.5 ± 0.3	10	−57.5 ± 2.1ab	5.0 ± 0.4	14
NlNav14	−35.4 ± 1.1a	4.2 ± 0.2	−48.9 ± 0.7bcd	3.9 ± 0.2	8.9 ± 0.1	13	−56.2 ± 1.8b	5.0 ± 0.7	9
NlNav15 (k)	−23.9 ± 0.9d	4.2 ± 0.4	−46.4 ± 0.4cd	4.9 ± 0.1	3.2 ± 0.1*	8	−49.7 ± 1.2c	6.0 ± 0.5	10
NlNav17	−27.1 ± 1.3bcd	4.5 ± 0.2	−47.4 ± 0.6cd	4.7 ± 0.1	8.8 ± 0.2	11	−50.9 ± 0.7bc	6.3 ± 0.3	12
NlNav18	−31.1 ± 0.6abc	3.7 ± 0.1	−48.2 ± 0.6bcd	4.1 ± 0.2	10.0 ± 0.2	10	−64.2 ± 1.4 a	5.6 ± 0.6	12
NlNav19	−33.4 ± 1.2a	4.6 ± 0.2	−53.4 ± 1.1a	4.3 ± 0.1	9.8 ± 0.1	15	−54.5 ± 1.7bc	6.3 ± 0.5	8
NlNav20 (k)	−25.7 ± 0.8cd	3.5 ± 0.2	−44.2 ± 0.3d	4.6 ± 0.2	2.7 ± 0.1*	7	−50.2 ± 1.5c	5.2 ± 0.4	10

The voltage dependence of activation, fast inactivation and slow inactivation were fitted with two-state Boltzmann equation to determine V_1/2, the voltage for half-maximal conductance or inactivation, and k, the slope factor for conductance or inactivation. Persistent current (%) was determined as the percentage of persistent current relative to transient peak current. Values in the table represent the mean ± SEM, and n is the number of oocytes used. Values in each column with the different letters showed significantly different among NlNa_v variants using one-way ANOVA with Scheffé’s post hoc analysis (p < 0.05).

^*Significant differences of persistent current between NlNa_v1 (exon l-carrying) and other variants were determined by Student’s t-test analysis (p < 0.05).

All 11 functional variants activated rapidly in response to depolarization and then inactivated rapidly. Besides peak currents, a non-inactivating current (known as persistent current) was also detected from many NlNa_v variants (indicated with an arrow in Fig. S1). Consistent with previous work in Drosophila (Lin et al., 2009), those NlNa_v variants containing exon l possessed greater persistent currents than NlNa_v variants containing exon k (Table 1). The parameters for the voltage dependence of activation, steady-state fast inactivation and slow inactivation are summarized in Table 1. The voltage for half-maximal activation ranged from −23 mV (for NlNa_v15) to −35 mV (for NlNa_v14) (Table 1). The half-maximal inactivation voltage of the variants was in a range of −44 mV (for NlNa_v20) to −53 mV (for NlNa_v8–1 and NlNa_v19). A broader range from −49 to −64 mV was observed for the voltage for half-maximal slow inactivation. No significant difference in kinetics of recovery from inactivation was detected among 11 functional variants (data not shown).

There are no scattered amino acid differences among the functional NlNa_v variants with one exception of NlNa_v18 which has one amino acid change A469V. Therefore, the observed differences in gating properties among these functional variants can be attributed to individual alternative exons or unique exon combinations. We found that the most notable gating modifications were mediated by sequences encoded by mutually exclusive exons k/l and optional exon 13 and E⁶⁰⁶. Below, we describe the role of specific exons in modulating NlNa_v gating properties.

All three exon k-containing variants, NlNa_v5–1, NlNa_v15 and NlNa_v20, exhibited more depolarizing voltage dependence of slow inactivation than most of the exon l-carrying variants including NlNa_v18 (Fig. 2B). Furthermore, two exon k-containing variants, NlNa_v15 and NlNa_v20 exhibited a 10–12 mV depolarizing shift in the voltage dependence of activation compared with exon l-carrying variants including NlNa_v14 (Fig. 2C). However, the third exon k-carrying variant, NlNa_v5–1, exhibited a more hyperpolarizing voltage dependence of activation compared with NlNa_v15 and NlNa_v20 (Table 1). NlNa_v5–1 differs from NlNa_v20 only in the usage of E⁶⁰⁶ (Fig. 2D), indicating that the inclusion of E⁶⁰⁶, which encodes one amino acid residue, E⁶⁰⁶, contributed to the difference in activation between the three exon k-containing NlNa_vs. The impact of exon k on the voltage dependence of fast inactivation was less certain than its effect on activation and slow inactivation. Compared with exon l-carrying variants, such as NlNa_v19, NlNa_v20 has a distinct depolarizing shift in the voltage dependence of fast inactivation (Fig. 2E), but not for NlNa_v5–1 and NlNa_v15 variants (Table 1). Potentially, other optional exons contributed to different gating properties of these variants. In our previous study, BgNa_v11 containing exon G2 (equivalent to exon k) had a more hyperpolarized inactivation voltage (Song et al., 2004), suggesting that other alternative exons and/or RNA editing of BgNa_v transcripts might play an important role in determining the gating properties of BgNa_v variants.

NlNa_v1 and NlNa_v17 differ only in the usage of exon 13. Inclusion of exon 13 in NlNa_v1 caused a 5-mV shift of the voltage dependence of activation in the hyperpolarizing direction compared with NlNa_v17 which lacks exon 13 (Table 1 and Fig. 2F). In addition, the sequences encoded by exon j are quite conserved among NlNa_v, DmNa_v and BgNa_v (Fig. 1C). In DmNa_v, inclusion of exon j was predicted to shift the voltage-dependent inactivation in a more hyperpolarizing direction (Lin et al., 2009). Here we found that NlNa_v19 variant containing the exons j + E⁶⁰⁶ had a significant hyperpolarizing shift in the voltage dependence of fast inactivation compared with NlNa_v14 and NlNa_v20 variants lacking exons j and E⁶⁰⁶. These results suggest potential functional conservation of exon j between DmNa_v and NlNa_v. In addition, DmNa_v variants containing exon f seem to activate at more hyperpolarizing membrane potentials than those containing exon j and e (Lin et al., 2009). However, we did not find any significant alteration in gating properties associated with exon f in NlNa_v variants, possibly because the amino acid sequences of exon f is quite diverse among NlNa_v, DmNa_v and BgNa_v even though the genomic position of exon f is conserved. Furthermore, comparisons of NlNa_v6 vs. NlNa_v19 and NlNa_v20 vs. NlNa_v15 showed that exon e does not appear to have a significate impact on gating properties of NlNa_v variants.

Thirteen clones of a total of 64 DmNa_v clones (20%) belong to the most common splice type (Olson et al., 2008), compared to only 4 clones of a total of 41 NlNa_v clones (9%) belong to the most common splice type. In addition, we did not find any RNA editing site in our analysis of the NlNa_v transcripts. The four clones in the most common splicing type of NlNa_v are identical. In contrast, the 10 clones in the most common splicing type of DmNa_v are all different and contain scattered amino acid changes, which appear to be caused by RNA editing (unpublished results). Interestingly, DmNa_v and BgNa_v variants appear to exhibit a broader range of voltage-dependence of activation and inactivation (Olson et al., 2008; Song et al., 2004) than NlNa_v variants (Table 1). Both A-to-I and U-to-C RNA editing events have been shown to contribute to functional diversity of BgNav channels (Liu et al., 2004; Song et al., 2004). Collectively, these results suggest that the usage of alternative exons may be different between the brown planthopper and DmNa_v and BgNa_v and that the brown planthopper may rely more on alternative splicing to increase functional diversity, whereas DmNa_v and BgNa_v use both extensive alternative splicing and RNA editing.

3.3. Genomic conservation of exon b across insect species

Previous studies have shown that deletion of exon b in exon b-carrying BgNa_v variants drastically increased the amplitude of sodium currents in Xenopus oocytes, whereas addition of exon b in an exon b-lacking BgNa_v variant reduced the amplitude of sodium currents (Song et al., 2004). Deletion of exon b in the Varroa mite sodium channel VdNa_v1 also enhanced sodium current in Xenopus oocytes (Du et al., 2009b). For the remaining 11 variants that produced currents that were too small for the analysis of gating properties, they all possess exon b. To further confirm the association of poor expression with exon b, we injected the same amount of cRNA (0.8 ng/per oocyte) into oocytes and record sodium currents at 17 hours after injection for the 11 variants lacking exon b and at 70 hours for the variants containing exon b. For NlNa_v variants lacking exon b, at 70 hours after injection, the amplitude of sodium currents was too large to record. As shown in Fig. 3, NlNa_v variants that produced robust sodium currents all lacked exon b, whereas NlNa_v variants that generated only small sodium currents all contained exon b. These results suggest a conserved role of exon b in regulating sodium current expression in insects.

Fig. 3.

Sodium peak currents of NlNa_v variants in Xenopus oocytes. Amplitude of the peak current was measured by a 20-ms depolarization to −15 mV from the holding potential of −120 mV. The error bars indicate the SEM for at least 10 oocytes. The representative current traces were shown on the left of the figure.

Examination of the genomic sequence of the NlNa_v gene revealed that exclusion or inclusion of exon b in different NlNa_v variants is the result of alternative splicing at two 3’ alternative acceptor sites (Fig. 4A). Further comparison of the genomic region where exon b resides among 20 arthropod species revealed an extremely conservative genomic structure including the two 3’ alternative acceptor sites (Fig. 4B). Deduced amino acid sequences encoded by exon b are also highly conserved among 20 arthropod species (Fig. 4B). While our manuscript was in preparation, another study (Silva and Scott, 2019) reported a more comprehensive analysis of sodium channel genes from 68 insect species and also established the conservation of exon b across insect species. Highly conserved S772, Y774 and Y775 (Fig. 5) were predicted to be potential phosphorylation sites which could be involved in regulating sodium current expression (Song et al, 2004). To examine whether mutations of these amino acids affect exon b-mediated inhibition of sodium currents, we made alanine substitutions of S772, Y774 and Y775 in one exon b-bearing variant, NlNa_v2–1. All three alanine substitutions significantly increased the amplitude of peak current (Fig. 5), suggesting that serine and tyrosine at these positions are critical for regulating the amplitude of sodium peak current possibly via phosphorylation. but the mutations did not alter voltage dependence of activation and fast inactivation (Table 2).

Fig. 4.

Exclusion of exon b is the result of alternative splicing using a 3’ alternative acceptor site. A and B: Exon-intron structure of the region where exon b is located from NlNa_v (A) and other 20 arthropod species (B). The amino acid sequence encoded by exon b is boxed and the intron upstream is indicated in shade and lowercase letters. The splicing consensus sites (gt/ag) are indicated in bold. The 3’ alternative acceptor site “AG” is indicated in bold. The highly conserved amino acids S772 and Y774 are marked in red.

Fig. 5.

Sodium peak currents of NlNa_v2–1 and its mutants in Xenopus oocytes. Amplitude of the peak current was measured by a 20-ms depolarization to −15 mV from the holding potential of −120 mV. The error bars indicate the SEM for at least 10 oocytes. *Significant differences between NlNa_v2–1 and mutant channels were determined by Student’s t-test analysis (p < 0.05). The representative current traces were shown on the left of the figure.

Table 2.

Voltage dependence of activation and fast inactivation of NlNa_v2–1 and its recombinants.

Na+ channel type	Activation		Inactivation		n
	V1/2(mV)	k(mV)	V1/2(mV)	k(mV)
NlNav2–1	27.8 ± 0.9	5.6 ± 0.2	50.6 ± 1.0	5.3 ± 0.1	22
S772A	26.2 ± 0.7	6.0 ± 0.3	48.6 ± 0.5	5.3 ± 0.1	17
Y774A	28.1 ± 0.6	5.6 ± 0.2	49.1 ± 0.3	5.1 ± 0.1	30
Y775A	27.3 ± 0.9	5.4 ± 0.4	47.8 ± 0.8	5.1 ± 0.1	9

The voltage dependence of activation and fast inactivation were fitted with two-state Boltzmann equation to determine V_1/2, the voltage for half-maximal conductance or inactivation, and k, the slope factor for conductance or inactivation. Values in the table represent the mean ± SEM, and n is the number of oocytes used. Voltage dependence of activation and fast inactivation have no significant different between NlNa_v2–1 and its recombinants, which determined by Student’s t-test analysis (p>0.05).

3.4. Different sensitivities of NlNa_v variants to pyrethroids

We also evaluated the sensitivities of the 11 functional NlNa_v variants to etofenprox, an ether pyrethroid used in the control of the brown planthopper (Hemingway, 1995; Wu et al., 2017; Yaméogo et al., 2001), permethrin and deltamethrin, two widely used pyrethroids in controlling many insect pests. The amplitude of pyrethroid-induced tail currents was used to quantify the potency of pyrethroid modification of NlNa_v variants. As shown in Fig. 6A, no tail currents were detected in the absence of pyrethroids in all 11 NlNa_v variants. As expected, tail currents induced by etofenprox and permethrin (Type I pyrethroids) decayed rapidly, which returned to the baseline within 2 s, whereas tail currents induced by deltamethrin (a Type II pyrethroid) decayed slowly and a portion of the tail current remained at the end of the 8 s repolarization (Fig. 6A). We found three variants, NlNa_v5–1, NlNa_v15–1 and NlNa_v20, were more resistant to all three pyrethroids. Interestingly, all three variants possess alternative exon k instead of l (Fig. 6B). These results are consistent with an earlier report that BgNa_v variants containing exon k (i.e., G2 in BgNa_v) were less sensitive to pyrethroids than those containing exon l (i.e., G1 in BgNa_v) (Tan et al., 2002a). One major mechanism of pyrethroid resistance, known as knockdown resistance (kdr), is caused by point mutations in sodium channels, and many kdr related mutations have been reported across insect species (Rinkevich et al., 2013; Dong et al., 2014; Field et al., 2017). However, so far there has been no report of an association between pyrethroid resistance in field populations with changes in the frequency of alternative exons including exon k/l.

Fig. 6.

Sensitivity of NlNa_v splice variants to etofenprox (ETO), permethrin (PMT) and deltamethrin (DMT). A: Tail currents induced by ETO, PMT and DMT from NlNa_v1 at the concentration of 0.1 and 1 μM. B: Sensitivity of 11 functional NlNa_v splice variants to ETO, PMT and DMT at the concentration of 1 μM. The error bars indicate the SEM for at least 10 oocytes. The pyrethroid-induced tail currents were recorded during a 100-pulse train of 5-ms step depolarizations from −120 to 0 mV with 5-ms interpulse intervals 10 min after pyrethroid application. The percentage of channels modified by pyrethroids was calculated using the equation described in Materials and methods (Tatebayashi and Narahashi, 1994).

In conclusion, our study showed that sodium channel splice variants from the brown planthopper exhibit different functional properties when expressed in Xenopus oocytes. In particular, our results showed that alternative exons b and k/l are highly conserved not only in sequence and genomic structure among insect species, but also in modulating sodium channel function. Furthermore, we found that exon k-containing variants exhibit a higher level of resistance to pyrethroid insecticides. Sodium channels are critical for initiation and propagation of action potentials in neurons. Changes in sodium channel activity due to alternative splicing events likely modify activities of neurons to alter the neuronal excitability to meet the unique function of the neurons in vivo. Collectively, our results complement earlier studies of Drosophila and cockroach sodium channel variants on the role of alternative splicing in regulating sodium channel function and further implicate the functional importance of these alternative splicing events in fine-tuning the function of the nervous system in insects.

Supplementary Material

1

Fig. S1. The representative voltage dependence of activation current traces of exon k-carrying variants NlNa_v5–1 (exon E⁶⁰⁶, e⁺, k⁺, 36⁺), NlNa_v15 (exon k⁺, 36⁺) and NlNa_v20 (exon e⁺, k⁺, 36⁺), and exon l-carrying variants NlNa_v14 (exon 13⁺, l⁺, 36⁺), NlNa_v18 (exon E₆₀₆, f⁺, l⁺, 36⁺) and NlNa_v19 (exon j⁺, 13⁺, E⁶⁰⁶, l⁺, 36⁺), which were recorded by 20-ms depolarizing potentials from −80 to −5 mV (in 5 mV steps) from a holding potential of −120 mV. Persistent current (P), indicated with a double-headed arrow, was recorded at the end of a 20-ms depolarizing potential.

Highlights:

1. NlNa_v splicing variants exhibited distinct functional properties.

2. The optional exon b inhibited the amplitude of sodium currents.

3. Three amino acid residues in exon b are critical for its inhibitory effect.