The cys-loop ligand-gated ion channel gene superfamily of the Colorado potato beetle, Leptinotarsa decemlineata

Abstract

Background

The insect cys-loop ligand-gated ion channel (cysLGIC) superfamily includes nicotinic acetylcholine receptors (nAChRs), γ-aminobutyric acid (GABA) receptors, glutamate- or histamine-gated chloride channels (GluCls and HisCls), pH-sensitive chloride channels (pHCls) and several other functionally uncharacterized receptors. Several of these receptors are target sites of neonicotinoids and other insecticides. Characterizing sequences of cysLGIC genes can facilitate the study of functional expression of subunits allowing insecticide/receptor interaction research, provide candidate targets for RNAi interventions, and can promote molecular resistance monitoring tools development. The Colorado potato beetle (CPB) is an agricultural pest that threatens the production of solanaceous crops. Although this insect shows frequent evolution of insecticide resistance, its cysLGIC superfamily is not well characterized.

Results

Twenty-two candidate CPB cysLGIC subunit genes were identified, and the functional regions of their protein sequences were annotated. CPB possesses twenty-two candidate cysLGIC subunit genes such as nAChR α4, nAChR α6, Rdl, and GluCl subunits, with similar sequence, structure, and alternative exon use as that of other insects. RNA A-to-I editing was observed of nAChR α6. Two copies of the pHCl subunit gene were identified, the first duplication of this gene observed in insects.

Conclusion

The number of cysLGIC superfamily genes is similar to that of other insect species. Alternative splicing and RNA editing conserved in insect species were also identified in expected subunits, potentially contributing to structural and functional diversity of the receptor. Evidence of naturally truncated nAChR α4 and duplicated pHCl was observed, which invites future validation.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12864-025-11867-5.

Background

Members of the cys-loop ligand-gated ion channel (cysLGIC) superfamily mediate both fast excitatory and inhibitory synaptic transmission in the nervous system of vertebrates and invertebrates. The cysLGIC are formed by the assembly of five subunits that share similar structural features: a large extracellular N-terminal region containing the signature cys-loop (two cysteines separated by 13 amino acids and bonded by disulfide bonds), ligand-binding motifs (loops A–F), and four transmembrane (TM) domains that form the ion-selective pore [1] (Fig. 1). Insect cysLGICs include the cation-permeable nicotinic acetylcholine receptors (nAChRs) [2], the γ-aminobutyric acid (GABA) receptor with both cationic and anionic (chloride) channel properties [3], the glutamate- or histamine-gated chloride channel (GluCl and HisCl) [4, 5], the pH-sensitive chloride channel (pHCl) [6], and the insect group1 subunits and several other functionally uncharacterized receptors [7] (Table 1, See Supplementary Table S1, Additional file 1). In addition to their roles in the insect nervous systems, these receptors are of particular interest as target sites of widely used insecticides. For example, nAChRs are target sites of the neonicotinoid and spinosyn insecticides, and GABA receptors are target sites of cyclodienes and the phenyl pyrazoles. According to the insecticide resistance action committee (IRAC) mode of action classification, members of the cysLGIC superfamily are target sites of seven insecticide groups (Table 1).

Fig. 1.

Structural features of a typical cys-loop receptor. a The pentameric receptor with ion permeability; b Topology of a cys-loop receptor subunit. The extracellular domain contains the signature cys-loops and Loops A-F (Gray shaded region) for ligand binding. Sequences connecting transmembrane domains 1 to 3 are usually short and can be longer between TM3 and TM4. Figure created with BioRender.com

Table 1.

Sub-category of insect cys-loop ligand-gated ion channels and their role as insecticide target sites

Receptor Name/Description	Endogenous ligand	Nomenclature of subunits	Targeted Insecticide Group
nicotinic acetylcholine receptor	Acetylcholine (ACh)	α (possessing two consecutive cysteines in loop C) and β	Group 4: neonicotinoid, Sulfoximines, etc Group 5: Spinosyns Group 14: Nereistoxin analogues Group 32: GS-omega/kappa HXTX-Hv1a peptide
γ-amino butyric acid-gated anion channels	γ-amino butyric acid (GABA)	Rdl, LCCH3 and GRD	Group 2: Cyclodiene Organochlorines & Fiproles Group 30: Meta-diamides & Isoxazolines
Glutamate-gated chloride channel	Glutamate	GluCl	Group 6: Avermectins & Milbemycins)
Histamine-gated chloride channel	Histamine	HisCl
pH-sensitive chloride channel	unclear	pHCl
Insect group 1 cysLGIC. Initially annotated in D. melanogaster as CG6927, CG7589 and CG11340	unclear	CLGC
Ungrouped subunits. Initially annotated in D. melanogaster as CG8916 and CG12344	unclear	8916 or 12,344

Insect nAChRs respond to acetylcholine (ACh) and mediate fast excitatory transmission at cholinergic synapses in the central nervous system (CNS). They involve processing sensory signals, locomotion, and cognitive processes [8]. Insect nAChR subunits are of two types, α (characterized by the two consecutive cysteines in Loop C) and β [2]. Most insect nervous nAChRs are hetero-pentameric, containing both α and β subunits, while in some rare cases, particular α subunits can form homomers [9]. Imidacloprid acts at the ACh binding site (the orthosteric site) on insect nAChRs, thereby activating the integral cation-permeable ion channel of the receptor [10] (Fig. 1). GABA receptors are widespread in insect CNS, mediating inhibitory and excitatory neurotransmission. The first insect GABA subunit gene was identified from a Drosophila melanogaster strain resistant to dieldrin (a cyclodiene insecticide), named Rdl [11]. GluCls have a wide range of functions in invertebrates’ nervous systems associated with locomotion, feeding, and sensory inputs [4]. The first insect GluCl subunit cDNA was cloned from D. melanogaster [12]. Most insect species, including Apis mellifera, Tribolium castaneum, and Nasonia vitripennis, have a single GluCl gene. Histamine is the primary neurotransmitter of arthropod photoreceptors, and these histamine-gated chloride channels likely play a role in vision, although their function remains unclear [12, 13]. The pH-sensitive chloride channel was initially identified in D. melanogaster. While its native ligand has not yet been identified, its chloride channel is opened in response to shifts in extracellular pH [6]. Insects possess 1—3 group1 subunits. In early studies, these subunits were labeled by numbers, consistent with the first discovered D. melanogaster orthologs (CG6927, CG7589, and CG11340); later, subunits belonging to this group, were labeled as CLGC, standing for Cys-loop Ligand Gated ion Channel [14]. Finally, there are two functionally uncharacterized subunits, initially annotated as CG8916 and CG12344, in the D. melanogaster genome assembly.

Insect species typically possess 21–26 cysLGIC subunit genes [7, 14–20], except for two cockroach species showing over 30 members, mainly because of expansion of nAChR β subunits [21]. These numbers are relatively small compared with other arthropod and nematode cysLGIC superfamilies; for example, the spider Pardosa pseudoannulata, the spider mite Tetranychus urticae, and the nematode Caenorhabditis elegans have 34, 29, and 102 cysLGIC genes respectively [22–24].

Genomic analysis of these proteins in pest insects has several benefits. Variation in cysLGIC receptor genes conferring target site insensitivity have been identified in insects (Table 2). Characterizing this variation provides the basis for molecular resistance monitoring. Similarly, sequence knowledge allows the development of nucleic acid-based control methods such as RNA interference which can be designed based on variant sites between beneficial species and pests to achieve effective pest management while minimizing risks to non-target species.

Table 2.

Target site insensitivity mutations on cys-loop receptors

Subunits	Mutations	Species	Resistance to	Reference
nAChR β1	R81T	Myzus persicae	Neonicotinoid	[25, 26]
		Aphis gossypii	Neonicotinoid	[27, 28]
	V101I	Myzus persicae	Neonicotinoid	[25]
	L80S	Aphis gossypii	Imidacloprid	[29]
	A58T&R79E	Bemisia tabaci	neonicotinoid	[30]
nAChR α6	G275E	Frankliniella occidentalis	Spinosad	[31]
Rdl	A301S/G	Aedes aegypti	Fiproles	[32]
		Nilaparvata lugens	Fipronil	[33, 34]
		Blattella germanica	Fipronil and dieldrin	[35, 36]
	T350M T350M + A301 G	Drosophila simulans	fipronil	[37]
GluCl	L315F	Musca domestica	ivermectin	[38]
	L319F	Bombyx mori	ivermectin	[39]

Characterizing cysLGIC subunit sequences also aides the study of subunit function, including the expression of subunits in well-established expression vehicles such as Xenopus laevis oocytes and gene editing in multiple insect species [8, 40–42].

The Colorado potato beetle, Leptinotarsa decemlineata (CPB), is a damaging pest, threatening the production of solanaceous crops, including potatoes, tomatoes, and eggplants. Intensive insecticide use has led to resistance to cysLGIC targeting insecticides, such as dieldrin [43], organochlorines [44], neonicotinoids [45, 46] and spinosyns [47] in CPB populations. Except for a few cysLGIC genes [48–50], this gene superfamily has not been studied in the CPB despite the importance of insecticides targeting cysLGICs for its control. A recently published chromosome-level CPB genome assembly [51] provides an opportunity to annotate and characterize the cysLGIC superfamily in this species. Here, we report the first systematic study of the cysLGIC gene superfamily in Colorado potato beetles. The protein coding regions of all subunits were successfully identified in the CPB genome. Characterizing the full complement of CPB cys-loop LGIC subunits represents a critical step in identifying key components of the CPB nervous system and pinpointing particular molecular targets underlying responses to insecticides.

Materials and methods

Bioinformatic analysis

To identify potential CPB cysLGIC genes, a tBLASTn [52] search was executed against the chromosome-level CPB genome (Genbank accession number JANJPO000000000.1) [51], with all T. castaneum and Periplaneta americana cysLGIC subunit protein sequences as queries (See Supplementary Table S2, Additional file 1 for NCBI accession number of query sequences). T. castaneum was chosen because, among all species with well-characterized cysLGIC, it is most closely related to CPB. P. americana was chosen because it has a large number of insect cysLGIC subunits. Candidate CPB cysLGIC genes were determined based on their identity with queries, particularly in the N-terminal ligand-binding region and the four transmembrane domains. In a second round, T. castaneum cysLGIC subunits were mapped to the CPB genome using the protein2genome model in Exonerate [53] to predict the intron and exon boundaries. To find the complete protein coding sequence of each CPB cysLGIC subunit, as well as to validate its expression in CPB, chromosomal hits identified by tBLASTn and Exonerate were extracted and manually assembled to obtain the genomic DNA (gDNA) sequence of each subunit. Then, the assembled gDNA sequences were used as queries to Blastn against a CPB transcriptome (NCBI accession GGNV00000000.1) [54]. Transcript hits with bit scores higher than 900 were collected.

Sequence selection for protein analysis

Transcripts from each cysLGIC subunit, were translated to their corresponding protein sequences, aligned, and manually screened for completeness of “cysLGIC subunit signature regions” (including the N-terminal ligand-binding motifs and transmembrane domains). Protein sequences proceeding to alignment were selected according to the following rule: for a subunit of which only one of the transcript-translated proteins shows all “signature regions”, the transcript-translated sequence will be selected; if multiple transcript-translated proteins can display all “signature regions”, the longest one was used; For a subunit of which none of the transcript-translated proteins can show complete “signature regions” or no transcript hit has a score higher than 900, a manual search was executed in the NCBI database to find any of its transcripts, cDNA or protein coding sequence predicted by automated computational analysis.

In order to further investigate the truncated sequence of nAChR α4, we compared the aligned nAChR α4 N-terminal extracellular region protein sequences of five insect species, including the red flour beetle, small hive beetle, western honeybee, American cockroach, and fruit fly (Additional file 3, Figure S1). Analysis of these aligned sequences revealed a high degree of structural conservation in this region, with no observed insertions or deletions, also suggesting that use of a closely related species’ nAChR α4 protein sequence as a query to map the CPB genome could be an effective strategy for locating exon regions encoding the N-terminal region.

To investigate the possible cause of highly fragmented nAChR α4 transcripts, we mapped RNA-seq data, obtained from previous studies (See Supplemental Table S3, Additional file 1) to the CPB genome assembly and calculated the mean depth of coverage at each expected nAChR α4 exon region. We also collected the exon specific mean depth of coverage of nAChR α1, which is located on the same chromosome, as a reference.

Protein sequence alignment and annotation

CPB nAChR subunits and non-nAChR subunits were aligned separately. Protein alignment was conducted by Clustal X2 [55]. The pairwise alignment was set with a gap-opening penalty of 10 and a gap-extension penalty of 0.1, applying the Gonnet 250 protein weight matrix. Sequence alignment was visualized and annotated using JalView version 2.1.1 [56]. Transmembrane domains of protein sequences were predicted by TMHMM—2.0 (https://services.healthtech.dtu.dk/service.php?TMHMM-2.0). Protein signal cleavage sites were predicted by SignalP 5.0. Genedoc calculated identities/similarities between protein sequences.

Phylogenetic analysis

Phylogenetic analyses were run for nAChR and non-nAChR subunits separately. In MEGAX [57], the MUSCLE algorithm aligned cysLGIC subunits of CPB, D. melanogaster, A. mellifera, P. americana, and T. castaneum (Additional file 2: Table S2 includes NCBI accession numbers of all sequences, used for phylogenetic analysis), and the phylogenetic tree was then reconstructed using the maximum-likelihood method and Dayhoff matrix-based model with 1000 bootstraps replication. The tree file was visualized and manipulated by TreeViewer [58].

Alternative splicing and post-transcriptional modification identification

Alternative splicing of a subunit gene was identified by comparing transcript isoforms. In addition, alternative exons can also be evidenced by tBlastn results, as a region of a protein query can hit multiple genomic regions with a reasonably high identity of each hit, and these alternative regions are tandemly arranged in the genome. Pre-mRNA editing was analyzed by comparing gDNA and transcripts of each subunit and their translated protein sequences.

Results

Overview of Leptinotarsa cysLGICs

CPB possesses 22 candidate cysLGIC subunit genes, including 11 nAChR subunit genes and 11 non-nAChR subunit genes. Sixteen subunit genes are found on autosomes and six are located on the X chromosome and a total of 21 subunit sequences were aligned and subsequently used in phylogenetic analysis (See Supplementary Tables S2 and S4, Additional file 1) [48, 49]. Subunits are labelled with a one-letter genus code plus a three-letter species code and the subunit types. For example, the nAChR α6 subunit in Tribolium castaneum was denoted as Tcas_ α6 and the Rdl ortholog of CPB is Ldec_Rdl.

CPB nAChR subunits

Eleven nAChR subunits were identified in CPB. Based on characteristic features such as the presence or absence of two consecutive cysteines in loop C (Fig. 2), ten subunits were categorized as α type and one as β type (Table S1). Most CPB nAChR subunits had amino acid sequence identities over 60% with their T. castaneum counterparts (Table 3) except for Ldec_α9 with 43% identity with Tcas_α9. This relatively low identity between nAChR α9 orthologs is not surprising because previous studies found this subunit to be highly variable across species, such as 28% and 20% identity between the T. castaneum nAChR α9 and its counterparts in A. mellifera and P. americana, respectively [14, 21]. In addition, a GEK motif preceding TM2, which is conserved among insect nAChR subunits, was present in all CPB nAChR subunits except Ldec_ α9 (Fig. 2). Similarly, replacements of the GEK motif were observed in multiple nAChR subunits of two cockroach species [21], silkworms [59], N. vitripennis [16], as well as the nAChR α9 in T. castaneum. Ldec_α1—Ldec_α4 and Ldec_α8 have insertions in Loop F (Fig. 2), also seen in other insect species, such as cockroaches [21], parasitoid wasp N. vitripennis [16] and the red floor beetle T. castaneum [14], which may contribute to interactions with the neonicotinoid, imidacloprid [60].

Fig. 2.

Protein sequence alignment of CPB nAChR subunits. The T. castaneum nAChR α1 is included for reference (NCBI accession NP_001103245.1). N-terminal signal leader peptides are indicated by lowercase letters. Loops A-F are highlighted in blue; the cys-loops are highlighted in gray, and transmembrane domains in orange. The two consecutive cysteines in Loop C, characteristic of α subunits, are enclosed in a box. Three amino acids (GEK) preceding TM2 are indicated by solid triangles underneath. NCBI accession numbers are listed in Supplementary Table S2, Additional file 1

Table 3.

CPB nAChR subunits (identity/similarity) to T. castaneum. NCBI accession numbers listed in Supplementary Table 2, Additional file 1

	Ldec_ a1	Ldec_ a2	Ldec_ a3	Ldec_ a4	Ldec_ a5	Ldec_ a6	Ldec_ a7	Ldec_ a8	Ldec_ a9	Ldec_ a10	Ldec_ b1
Tcas_ a1	90/94	53/69	55/66	52/65	29/48	33/51	34/52	53/69	15/32	18/37	38/56
Tcas_ a2	51/68	90/94	50/65	46/62	26/48	33/52	33/51	50/66	16/33	19/34	38/56
Tcas_ a3	55/67	50/65	87/92	67/76	28/48	32/49	34/51	55/67	15/32	19/34	41/59
Tcas_ a4	52/65	46/62	66/76	91/95	27/45	32/48	33/51	55/70	16/32	18/35	40/56
Tcas_ a5	23/40	21/40	23/40	22/38	63/67	26/41	25/41	21/38	13/27	15/28	24/40
Tcas_ a6	34/51	33/51	32/49	33/49	34/53	89/94	64/75	33/52	18/36	21/37	33/53
Tcas_ a7	33/51	34/52	33/49	34/50	32/51	62/74	89/94	33/51	16/33	19/37	32/51
Tcas_ a8	52/67	50/66	55/67	51/67	29/50	33/53	34/53	85/90	15/34	21/39	40/58
Tcas_ a9	15/30	15/32	15/32	14/31	16/36	15/34	14/33	14/32	43/66	23/47	14/33
Tcas_ a10	19/36	19/35	18/35	18/35	19/40	20/38	20/39	20/37	26/47	63/80	19/38
Tcas_ a11	52/66	50/66	55/66	52/66	28/48	33/51	33/50	84/90	15/34	21/38	39/57
Tcas_b1	38/55	38/55	40/57	38/54	29/51	34/52	32/51	39/57	16/31	20/39	94/96

^*Potential orthologs are highlighted in bold

In the nAChR amino acid sequence-based phylogeny (Fig. 3), α1—α8 and β1 subunits are relatively conserved across insect species [61]. The “nAChR α5 group” only includes non-dipteran orthologs, since dipteran α5 subunits were named following different nomenclatures [59]. For example, the Bombyx mori α5 shared 17% identity with D. melanogaster α5 [59]. Instead, D. melanogaster α5 subunits are clustered with non-dipteran α7 subunits, named here as “α7-like group”. In addition, two CPB nAChR, Ldec_α9, and Ldec_α10, clustered in the large divergent group.

Fig. 3.

Phylogeny of nAChR subunits in five species. Species name annotations: Ldec-L. decemlineata; Dmel-D. melanogaster; Amel-A. mellifera; Tcas-T. castaneum; Pame-Periplaneta americana. Gvio_2XQ3_A, a bacterial ancestor of cysLGICs from Gloebacter violaceus was an outgroup, to which the tree was rooted. Bootstrapping values higher than 0.5 are displayed. NCBI accession numbers are listed in Supplementary Table S2, Additional file 1

The Ldec_α7 and Ldec_β1 genes are located within 474 kb of each other on the X chromosome NCBI Accession number CM045700.1 (www.ncbi.nlm.nih.gov). Clustering of nAChR α7 and β1 subunits have similarly been observed in the genomes of B. germanica, P. americana, A. mellifera, A. gambiae, and T. castaneum [14, 19, 21, 62], suggesting conserved syntony and the possibility that their expression may be coordinated.

Preliminary evidence of alternative splicing in several subunits was supported by tBlastn results. Two alternatives were identified for Ldec_α4 exon4, labelled exon4a and exon4b respectively (Fig. 4a). Alternative splicing leads to amino acid polymorphism in the agonist binding pocket forming region, Loop E, and Loop B, which may contribute to the variable function of protein isoforms. The Ldec_α6 has two potential alternatives in exon 8 (Fig. 4b), encoding TM2, which forms the pore of the channel, possibly contributing to functional diversity [1]. Alternative splicing of exon 4 in the nAChR α4 and exon 8 in nAChR α6 has also been observed in other insect species [63].

Fig. 4.

Potential alternative splicing. Variable sites are highlighted. a Ldec_α4 exon four splicing variants. Both exon 4a and 4b were translated from genomic DNA extracted from the chromosome CM045695.1. b Ldec_α6 exon eight splicing variants. Exon 8a was translated from genomic DNA extracted from the chromosome CM045700.1; exon 8b was translated from the transcript GGNV01207620.1. c Ldec_Rdl exon three splicing variants. Both exon 3a and 3b were translated from genomic DNA extracted from the chromosome CM045709.1; d Ldec_Rdl exon six splicing variants. Both exon 6a and 6b were translated from genomic DNA extracted from the chromosomeCM045709.1. e Ldec_GluCl exon three splicing variants. All three alternative exons were translated from genomic DNA extracted from the chromosome CM045700.1

A comparison of the CPB nAChR α6 gDNA and a transcript (GGNV01207620.1) sequence indicate three potential cases of adenosine (A) to guanosine (G) editing in gDNA sequence positions 27, 415, and 468 (Fig. 5a). This conversion results in an S-to-L, N-to-D, and I-to-M amino acid substitution on protein positions 10, 139, and 156, respectively (Fig. 5b). The S-to-L substitution may not affect the protein product since it is in the signal peptide region (refer to Fig. 2), which will be trimmed in the final protein. The N-to-D substitution removed a potential N-glycosylation site, which may affect receptor maturation, channel desensitization, and conductance [64, 65]. It is worth noting that nucleotide differences between reference genomes may be due to genomic DNA variation rather than RNA A-to-I editing since these two sequences came from different CPB samples.

Fig. 5.

Potential RNA A-to-I editing on CPB nAChR α6. a A-to-G conversion between the gDNA and the transcript at site 27, 415, and 468 respectively (indicated by solid triangles) b Amino acid replacement due to RNA editing, indicated by solid triangles

GABA receptor subunits

In CPB, one ortholog of each GABA receptor subunit was identified and named Ldec_Rdl, Ldec_GRD, and Ldec_LCCH3, respectively. The Ldec_Rdl shows a high identity with its T. castaneum counterpart (Table 4) and an anion channel signature PAR motif preceding TM2 which is conserved across insect species (Fig. 6). Initially, we found no evidence of an alternative Ldec_Rdl exon 3 based on transcript data or tBLASTn results. Subsequently, because RDL is known to have splice variants of exon 3 in other species [14, 17, 19, 66], we mapped all three Rdl protein isoforms of T. castaneum (NCBI accession numbers NP_001107808.1, NP_001107809.1, and NP_001107764.1) to the CPB genome using Exonerate and successfully identified alternative exon 3 variants (Fig. 4c). Exon 3 of the CPB Rdl gene encodes 23 amino acids preceding Loop D. Additionally, we identified two alternative exon 6 variants, which introduce 12 amino acid polymorphisms (Fig. 4d). Some of these variations occur between Loops F and C, within the ligand-binding pocket, potentially affecting receptor function. By combining exon 3 and exon 6 alternatives, CPB could theoretically possess three Rdl isoforms, potentially contributing to the functional diversity of GABA receptors. Notably, Ldec_Rdl has a Serine (S) amino acid residue at position 280, which is equivalent to the 301S allele conferring resistance to fiproles (Table 2).

Table 4.

CPB non-nAChR cysLGIC subunits homology statistics (identity/similarity) to Tribolium. NCBI accession numbers listed in Supplementary Table 2, Additional file 1

	Ldec_12344	Ldec_8916	Ldec_CLGC	Ldec_GluCl	Ldec_GRD	Ldec_HisCl1	Ldec_HisCl2	Ldec_LCCH3	Ldec_pHCl	Ldec_Rdl
Tcas_12344	70/81	17/31	16/30	21/41	23/39	26/45	28/48	21/40	17/36	23/40
Tcas_8916	19/33	77/85	14/27	23/40	37/53	20/36	21/36	27/43	17/30	26/43
Tcas_CLGC1	17/34	16/30	52/67	20/37	18/32	18/34	19/36	19/38	21/35	18/34
Tcas_CLGC2	18/34	15/28	51/66	20/35	17/30	17/33	17/35	17/35	18/33	17/33
Tcas_CLGC3	21/41	15/30	35/53	21/39	17/37	19/39	19/38	17/37	21/37	18/35
Tcas_GluCl	22/41	22/38	20/33	95/97	25/44	28/44	27/45	27/46	24/43	30/44
Tcas_GRD	20/36	39/53	16/28	24/41	72/81	21/38	21/38	30/46	17/31	30/45
Tcas_HisCl1	27/47	20/34	15/31	28/45	22/39	82/89	54/67	24/42	19/37	24/41
Tcas_HisCl2	27/50	21/36	16/33	28/45	24/42	54/67	84/89	26/42	20/39	26/42
Tcas_LCCH3	23/40	26/42	17/34	27/46	29/47	25/42	26/43	85/92	18/35	32/48
Tcas_pHCl	18/36	16/30	19/32	25/44	17/33	18/37	18/38	18/35	86/93	20/36
Tcas_Rdl	23/40	26/43	17/31	30/45	30/46	24/40	26/41	32/50	20/36	94/95

^*Potential orthologs are highlighted in bold

Fig. 6.

Protein sequence alignment of CPB cysLGIC subunits other than nAChR. T. castaneum Rdl subunit (NCBI accession NP_001107809.1) is included as a reference. N-terminal signal leader peptides are indicated by lowercase letters. Loops A-F are highlighted in blue, the cys-loops are highlighted in gray, and transmembrane domains in orange. Three amino acids preceding TM2 are indicated by solid triangles underneath. NCBI accession numbers listed in Supplementary Table S2, Additional file 1. The Ldec_Rdl protein sequence includes an amino acid substitution at TM2 (boxed), which is equivalent to the 301S mutations (Table 2) conferring resistance to fiproles

The Ldec_GRD and Ldec_LCCH3 lack the PAR motif (Fig. 6), indicating their potential role as a cation channel. Two alternative exons 6, leading to 12 amino acid polymorphisms of the Ldec_Rdl were detected by the tBlastn result (Fig. 4c). Some of the amino acid variation occurs between loops F and C, within the ligand binding pocket, which may cause functional effects on receptors.

CPB glutamate and histamine-gated chloride ion channels

One glutamate ion channel gene (Ldec_GluCl) and two histamine-gated chloride ion channel genes (Ldec_HisCl1, and Ldec_HisCl2) were identified in the CPB genome. T. castaneum and CPB have high sequence similarity (over 82%) of these orthologs (Table 4). These three subunits display a PAR motif before TM2, consistent with their role as anion channels. Comparison among transcript-translated protein sequences indicates three alternative splicing products of exon 3 of the Ldec_GluCl gene (Fig. 4e). Numbers of alternative GluCl exons 3 vary among species, with honeybees, fruit flies and two cockroach species having two [7, 19, 21] and the red flour beetles having three [14]. Ldec_GluCl exon 3 encodes the region proximate to loop D, which participates in the formation of ligand-binding pockets; therefore, amino acid polymorphisms due to alternative splicing may lead to variable ligand-binding properties among isoforms.

CPB pH-sensitive chloride channel subunits

The Tcas_pHCl query matched two similar, tandemly arrayed CPB genomic regions on chromosome CM045712.1. Protein sequences deduced from these two genes are equal in length (426AA), having only one amino acid variant at position 356 (S/A), resulting in a 99.8% identity. This suggests that these two genes may be recently duplicated homologues of CPB pHCl. Therefore, they were named Ldec_pHCl copy1 and Ldec_pHCl_copy2 respectively (See Supplementary Table S4, Additional file 1). However, only one CPB pHCl protein isoform was validated via transcript sequence and this isoform was used for phylogenetic analysis (Fig. 7) and protein annotation (Fig. 6).

Fig. 7.

Phylogeny of non-nAChR cysLGIC subunits in five insect species. Gvio_2XQ3_A was an outgroup, with which the tree was rooted. Ldec_α1 is included as a reference. Bootstrapping values higher than 0.5 are displayed. See Supplementary Table S2, Additional file 1, for accession numbers of sequences used in phylogeny

Other CPB cysLGIC subunits

In CPB, only one gene belonging to Insect Group 1 cysLGICs was identified, based on its relative similarity with three T. castaneum CIGCs and position in the phylogeny (Fig. 7). Therefore, it was named Ldec_CLGC. Insect Group 1 subunit homologs form an independent group in the phylogeny (Fig. 7) [14, 18].

Orthologs of Drosophila CG8916 and CG12344 were identified in the CPB genome and named Ldec_8916 and Ldec_12344, respectively. The Ldec_12344 displays a PAR motif before TM2 (Fig. 6), indicating its potential role as an anion channel. The phylogeny shows that the Ldec_12344 and its orthologs grouped with HisCl subunits, while the 8916-group clustered with GABA-gated ion channel subunits (Fig. 7).

Discussion

We characterized the CPB cys-loop LGIC gene superfamily, which encodes receptors that play major roles in the nervous system and are also targets of widely used insecticides. CPB has a similar number (22) of receptor subunit genes as other sequenced insects with greatest similarity to another beetle, Tribolium. Mechanisms increasing subunit diversity, such as alternative splicing and RNA A-to-I editing, are observed in nAChR α4, nAChR α6, Rdl, and GluCl subunits.

Ldec_α4, Ldec_α9 subunits of nAChR and Ldec_pHCl have unique features

Compared to other subunits, the Ldec_α9 subunit has low similarity with its T. castaneum counterpart, though T. castaneum remains the most similar of those included in the phylogram (Table 3). Ldec_α9 also lacks the GEK motif preceding TM2, which is usually conserved in nAChR subunits. To evaluate whether the Ldec_α9 is properly categorized as a nAChR subunit we calculated the protein sequence identity between Ldec_α9 and all other CPB cysLGIC subunits. We observed that Ldec_α9 shares 9% to 14% identity with other CPB non-nAChR subunits, while its identity with other CPB nAChR can reach 25%. Further, the relative similarity to the Tribolium α9 in Table 3, indicates that Ldec_α9 is more likely a nAChR subunit than another non-nAChR cysLGIC subunit. Whether Ldec_α9 possesses the cation selectivity of other GEK-motif-containing nAChR subunits remains to be determined.

Of the nAChR subunits, α4 is the only one whose complete expression was not confirmed by either transcripts or cDNA. Figure 8 displays an alignment of all potential CPB nAChR α4 protein sequences deduced from different sources, including three transcripts obtained from a pest population in Wisconsin (GEEF01180990.1, GEEF01026894.1, and GEEF01038441.1), one cDNA from Xinjiang, China (KP090398.1) and the gDNA sequence (CM045695.1, also from Xinjiang) used in this study, in addition to the T. castaneum nAChR α4 for reference. According to Tcas_α4 and the Ldec_α4 gDNA deduced sequence, a typical beetle nAChR α4 protein is expected to be over 550 amino acids; however, the longest transcript-deduced CPB protein (GGNV01187321.1_deduced) possesses only 182 amino acids. The cDNA-deduced sequence is slightly longer but still missing almost all N-terminal regions. This cDNA sequence was obtained in a previous study by mRNA RT-PCR, followed by a 5’- and 3’- Rapid amplification of cDNA ends (RACE) [50]. Because of this, the short nAChR α4 cDNA may not be due to incomplete sequencing coverage or mis-amplification. Additional evidence of truncation of the nAChR α4 transcripts is offered by the depth of coverage analysis. Table S5 (Additional file 1) and Fig. 9 show the mean depth of coverage of RNA reads across the investigated exon regions. For nAChR α4 (Fig. 9a), the mean depth of coverage for exons 6–11, encoding the membrane-spanning region, is at least twice that of exons 1–4, encoding for the N-terminal extracellular region. Notably, exon 1 and exon 3 exhibit few mapped reads, with an average coverage below 1 ×, and no reads were mapped to exon 2 across all RNA-seq datasets (Table S5). It is surprising since for each analysis, we used pooled RNA-seq data (See Supplementary Table S3, Additional file 1) from multiple samples from a population (Dataset 1–3), or even pooled data from multiple populations (Dataset 4), to map the CPB genome. Excluding exon 2, the ratio of the highest depth of coverage (exon 11) to the lowest (exon 1) is 27.15. This extremely low expression of exons encoding for the N-terminal extracellular region may explain the inability to assemble a complete transcript. Furthermore, the boxplot (Fig. 10a) indicates that the interquartile range for exons 1–4 does not overlap with that of exons 6–11.

Fig. 8.

Protein sequence alignment of all potential CPB nAChR α4 isoforms. T. castaneum nAChR α4 (NCBI accession NP_001103246.1) is included for reference. Respectively, GGNV01187321.1, GEEF01180990.1, GEEF01026894.1, and GEEF01038441.1 are transcript sequences, KP090398.1 is a cDNA sequence

Fig. 9.

Exon-specific depth of coverage. a the depth of coverage of nAChR α4 exons; b the depth of coverage of nAChR α1 exons

Fig. 10.

Distribution of coverage depth across exons. a nAChR α4; b nAChR α1

For nAChR α1 (Fig. 9b), more reads mapped to exons encoding the membrane spanning region (exon 6–8) than the N-terminal coding exons (exon 1–4), the ratio of the highest (exon 8) to the lowest mapped exon (exon 1) was only 2.62. Additionally, the boxplot (Fig. 10b) did not reveal a distinct difference in depth of coverage across exons, as the interquartile ranges overlapped well.

These much shorter protein sequences deduced from transcripts and cDNA may indicate that CPB nAChR α4 are naturally truncated and since the cDNA and transcripts are from five different CPB populations, this truncated version of CPB nAChR α4 may be geographically widespread. Truncation of nAChR is also seen in other insect species, such as the truncated α4 and α5 transcripts found in D. melanogaster and B. mori [7, 59]. However, these truncated transcripts lead to a partial loss of transmembrane regions or ligand binding loops, rather than a nearly complete loss of N-terminal extracellular region seen in CPB nAChR α4.

Whether this nAChR α4 truncation alters CPB’s natural biological processes and insecticide interactions remains to be determined as does the geographic and phylogenetic range of this truncation. Importantly, knocking out the nAChR α4 in a D. melanogaster strain led to a 1.491, 1.262, and 1.343-fold resistance ratios to sulfoxaflor, imidacloprid, and nitenpyram, respectively [67]. It is possible therefore that this structural change is in response to exposure of these diverse populations of CPB to neonicotinoid insecticides. Analysis of the roles of truncated nACh4 α4 subunits in the structure and function of nACh4 receptors would clearly be of value.

The successful expression of nAChR β1 is potentially vital for insect survival, as homozygous nAChR β1 knockout D. melanogaster strains are either lethal [68] or exhibit significantly reduced viability [67]. Therefore, nAChR β1 could serve as an ideal target for designing pest-specific RNAi products. Based on this premise, we compared the nAChR β1 mRNA sequences encoding the region from Loop D to Loop C in western honeybees, Colorado potato beetles, and small hive beetles (Fig. 11). We highlighted the first ten sites that are unique to honeybees but conserved among the beetle species. An ideal nucleotide-based control method, such as RNAi, could be designed based on these signature sites to selectively manage pest species while minimizing adverse effects on honeybees.

Fig. 11.

Comparison of nAChR β1 mRNA sequences among selected insect species. The displayed sequences encode the region from Loop C to Loop D. NCBI accession numbers for the mRNA sequences shown: A. mellifera (Amel)—NM_001079560.1; L. decemlineata (Ldec)—KP339870.1; A. tumida isoform 1—XM_049961093.1, isoform 2—XM_020024745.2

For non-nAChR cysLGIC subunits, the expression of Ldec_Rdl and Ldec_8916 were not validated by transcripts or database cDNA sequence. A previous study sought to amplify the CPB Rdl mRNA using PCR [69] while only a fragmented product was obtained. The complete expression of these two subunits in the CPB population still needs to be validated.

We show preliminary evidence of the first Ldec_pHCl gene duplication observed in an insect species. This observation deserves additional validation because insect pHCl have potential as insecticide targets due to its exclusive occurrence in invertebrates [6].

Detection of Rdl mutation in CPB reference genome

We detected the fiprole resistance-conferring mutation in the Rdl sequence of the reference CPB genome. The Ldec_Rdl protein displayed in Fig. 6 corresponds to transcript XM_023156435.1, derived from the genomic scaffold NW_019289435.1, which was assembled from a sample collected in Long Island, New York [70]. This indicates that this resistance allele is present in field populations of CPB. Future studies involving broader population sampling are necessary to determine its regional frequency and potential contribution to phenotypic resistance.

Chromosome level genome assembly indicates more genomic features

Improved CPB genome assemblies contributed here to revealing genomic features which, because genomic architecture has implications for adaptation to insecticide exposure, may provide insight into insecticide resistance [51]. To date, all known field-evolved neonicotinoid resistance-conferring mutations (not identified in CPB yet) are found on the β1 subunit, among which the R81T mutation is the most widely reported, detected in multiple M. persicae and A. gossypii populations worldwide [25–28, 71]. However, no homozygous 81TT individuals have yet been reported in any field population [25], possibly due to fitness costs associated with the 81 T genotype [72]. Colorado potato beetle is an XX-XO sex-determination system [73], of which the male individuals have a single X chromosome. Therefore, alleles present on the X chromosome of males would be functionally dominant for both resistance and fitness traits. If there were significant fitness costs of that allele, perhaps due to pleiotropic effects on neuronal function, this resistance mutation may not persist in CPB populations because of its deleterious effects on all male carriers of the resistance alleles.

Second, the improved completeness and contiguity of this newer genome assembly revealed the close physical proximity of some nAChR subunit genes. This may facilitate their coordinated expression and co-assembly into the native channel. For example, D. melanogaster nAChR α1, α2, and β2 subunit genes are clustered within 200 kb of each other on chromosome 3; these three subunits were later shown to form robust functional nAChRs in X. laevis oocytes. Here, the protein-coding regions of Ldec_α7 and Ldec_β1 were observed to cluster within 470 kb on the chromosome CM045700.1 (See Supplementary Table S4, Additional file 1), suggesting that they are likely to co-assemble in native CPB nAChR.

Conclusion

CPB has a typical number of cysLGIC superfamily genes compared to other insect species and has members of all major groups. There is evidence that CPB has a recently duplicated pHCl gene, but further testing of this observation is needed. With the exception of one subunit with truncated transcripts (nAChR α4) CPB subunit genes were typical in structure as well. Evidence of alternative splicing and RNA editing were also identified in CPB subunit transcripts, potentially contributing to subunits’ structural and functional diversity. The characterization of CPB cysLGICs provides further insights into the evolution of this gene superfamily in insects. The sequence characterization provides an important basis for future studies of these ion channels as well as for the rational design of insecticides that control CPB, ideally with less effect on non-target organisms. Such studies can include the functional expression of certain subunits to determine their interaction features with native ligands and insecticides. For example, CPB nAChR α1 and α8 subunits have been co-expressed with rat β2 subunit (rβ2) at different capped RNA (cRNA) ratios in Xenopus oocytes, and responses of different hybrid receptors to ACh and three neonicotinoid chemicals were analyzed [74]. In addition, based on the sequence characterization, RNA interference and gene editing techniques may also be applied to CPB nAChRs or other cysLGIC subunits to assess their role in neuronal function and response to insecticides.

Abbreviations

Ach

Acetylcholinesterase

BLAST

Basic Local Alignment Search Tool

cDNA

Complementary DNA

CLGC

Cys-loop Ligand Gated ion Channel

CNS

Central nervous system

CPB

Colorado Potato Beetle

cysLGIC

Cys-loop ligand-gated ion channels

DNA

Deoxyribonucleic acid

GABA

Gamma-aminobutyric acid

gDNA

Genomic DNA

GluCl

Glutamate-gated chloride channel

GRD

Gamma-aminobutyric acid receptor alpha-like

HisCl

Histamine-gated chloride receptors

IRAC

Insecticide Resistance Action Committee

LCCH3

Gamma-aminobutyric acid receptor subunit beta-like

mRNA

Messenger ribonucleic acid

nAChR

Nicotinic acetylcholine receptors

NCBI

National Center for Biotechnology Information

pHCl

PH-sensitive chloride channels

RACE

Rapid amplification of cDNA ends

Rdl

Resistance to dieldrin

RNA

Ribonucleic acid

TM

Transmembrane

Authors’ contributions

DC conceived the project, performed the research, and created all figures. DC and DJH wrote the manuscript. Both authors read and approved the final manuscript.

Funding

This work was aided by support of the Department of Entomology, University of Maryland and the National Institute of Food and Agriculture, U. S. Department of Agriculture and SARE under award number LNE23-475R.

Data availability

The datasets analyzed during the current study are available at www.ncbi.nlm.nih.gov, accession numbers are listed in table S4 and in the text.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.