Molecular Cloning, Characterization and Expression Profiling of a Ryanodine Receptor Gene in Asian Corn Borer, Ostrinia furnacalis (Guenée)

Ryanodine receptor (RyR) Ca2+ release channel is the target of diamide insecticides, which show selective insecticidal activity against lepidopterous insects. To study the molecular mechanisms underlying the species-specific action of diamide insecticides, we have cloned and characterized the entire cDNA sequence of RyR from Ostrinia furnacalis (named as OfRyR). The OfRyR mRNA has an Open Reading Frame of 15324 bp nucleotides and encodes a 5108 amino acid polypeptide that displays 79–97% identity with other insects RyR proteins and shows the greatest identity with Cnaphalocrocis medinalis RyR (97%). Quantitative real-time PCR showed that the OfRyR was expressed at the lowest level in egg and the highest level in adult. The relative expression level of OfRyR in first, third and fifth-instar larva were 1.28, 1.19 and 1.99 times of that in egg. Moreover, two alternative splicing sites were identified in the OfRyR gene. One pair of mutually exclusive exons (a/b) were present in the central part of the predicted SPRY domain, and an optional exon (c) was located between the third and fourth RyR domains. Diagnostic PCR demonstrated that exons a and b existed in all developmental stages of OfRyR cDNA, but exon c was not detected in the egg cDNA. And the usage frequencies of these exons showed a significant difference between different developmental stages. These results provided the crucial basis for the functional expression of OfRyR and for the discovery of compound with potentially selective insect activtity.


Introduction
Ryanodine receptors (RyRs) are members of a superfamily of intracellular Ca 2+ channels. These channels regulate the release of calcium from the lumen of sarcoplasmic/endoplasmic reticulum to the cytosol of muscle and non-muscle cells [1][2]. The RyRs are homomeric tetramers with a total molecular mass of 2 to 2.5 MDa, and each contains about 5000 amino acids with a large hydrophilic N-terminal domain and a small membrane-spanning C-terminal domain. The large N-terminal cytoplasmic domain termed ''foot structure'' spans the junction gap between the transverse tubules and sarcoplasmic reticulum membranes, while the hydrophobic C-terminal domain, which was predicted to contain 4-12 transmembrane segments, forms the putative pore of the Ca 2+ release channel [3][4]. It has been proposed that the specific interaction between the cytoplasmic and transmembrane domains is an important mechanism in the intrinsic modulation of the RyR [5].
To date, three types of ryanodine receptor (RyR1, RyR2 and RyR3) have been identified in mammals. RyR1 and RyR2 are abundant in skeletal and cardiac muscles. RyR3 is ubiquitously expressed throughout different tissues but is relatively abundant in brain and certain skeletal tissues [6][7][8]. There is approximately 66% amino acid sequence homology among the three mammalian RyR isoforms, except for three regions with high degrees of variability [9]. Two isoforms (RyRA and RyRB) were found in fish, amphibians and birds [10][11]. These two isoforms showed homology with mammals RyR1 and RyR3. However, only one RyR gene was found in Drosophila melanogaster (DmRyR) [12], and DmRyR shared 45-47% amino acid identity with three mammalian RyR isoforms [13]. The regions with a high degree of structural divergence between the mammalian and insect isoforms of the RyR may serve as potential targets for insecticides that interact specifically with the insect but not mammalian isoforms [9].
The Ca 2+ -release channel properties of RyR have been studied with increasing interest in the last 30 years in the medical field. Recently, insect RyRs attracted researchers' great attention in the pesticide field. Because, insect RyRs were found to be the targets of two classes of diamides chemicals: the phthalic diamides such as flubendiamide [14][15][16][17], and the anthranilic diamides such as chloranthraniliprole [18][19][20]. The diamides insecticides activate the ryanodine-sensitive intracellular calcium release channels in insects and show good selectivity for insect RyRs compared to mammalian RyRs by evoking typical symptoms including poisoning, body contraction, feeding cessation, paralysis and subsequent mortality [21]. It was reported that flubendiamide and chlorantraniliprole caused release of intracellular Ca 2+ in isolated Heliothis virescens and Periplaneta americana neurons [15,16,20]. Meanwhile, binding studies on microsomal membranes from Heliothis virescens and Periplaneta americana muscles showed that diamides interact with a site different from the ryanodine binding site on the insect RyR complex. The binding site for ryanodine is located on the C-terminal portion of RyR, within or close to the pore region of the Ca 2+ release channel [16,20]. However, the exact binding sites for flubendiamide or chlorantraniliprole have not yet been identified.

Insects and RNA preparation
Asian corn borer larvae were obtained from the stalk of field corn at the Hengshui experimental field of Chinese Academy of Agricultural Sciences, Hebei Province, China. Colonies have been maintained on an agar-free semi-artificial diet in the laboratory for several generations. Total RNA was isolated from Asian corn borer using an RNeasyH Mini Kit (Qiagen, Germany) following the manufacturer's instructions.

OfRyR cDNA cloning
Complementary DNA was transcribed from total RNA (1 mL) of third instar larvae following the manufacturer's protocol (Primescript TM First-Strand cDNA Synthesis kit; Takara, Dalian, China). The strategy used for PCR cloning is shown in Figure 1. Several primer sets based collectively on conserved sequences from bombyx mori, Spodoptera exigua, plutella xylostella, Cnaphalocrocis medinalis, Heliothis virescens, Aedes aegypti, Anopheles gambiae and Drosophila Melanogaster RyR mRNAs were synthesized (BGI, Beijing, China), and employed for PCR amplification ( Table 1). The thermal cycling for the PCR reaction was: initial denaturing at 94uC for 5 min, 30 cycles at 94uC for 30 s, 50-65uC (depending on the Tm of primers) for 30 s and 72uC for 1-3 min (determined by the length of the amplified fragments), and an additional polymerisation step at 72 uC for 10 min. The PCR products were electrophoresed on 1% agarose gels, and the expected DNA fragments were subcloned into pMD19-T Simple Vector (Takara, Dalian, China). The nucleotide sequences of the generated clones were determined at the BGI (Beijing, China), and compared with those of other RyR mRNA sequences previously reported in GenBank.

Sequence analysis
The nucleotide sequences of the full-length OfRyR cDNA was determined by all of the overlapping cDNA clones. The molecular weight, protein length and pI were predicted using DNAMAN v.7.0 (Lynnon Corporation). The sequence alignment was performed using CLUSTALW with the default settings, and the aligned sequences were used to construct the phylogenetic tree in MEGA5. The conserved sequence was analysed using Conserved Domains (http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb. cgi). The transmembrane (TM) segments were predicted from the secondary structure of the deduced amino acid sequence (www.cbs.dtu.dk/services/TMHMM). The hydropathy profile of the consensus amino acid sequence of OfRyR was analysed using the website (http://web.expasy.org/cgi-bin/protscale/ protscale.pl.).

qRT-PCR analysis of OfRyR expression
The relative expression levels of OfRyR mRNA in different developmental stages were examined by quantitative real-time PCR (qRT-PCR). Complementary DNA was synthesized with equal amount of total RNA (1 ug) from eggs, first-instar larvae, third-instar larvae, fifth-instar larvae, pupae and adults according to the instructions of the PrimeScript TM RT reagent Kit with gDNA Eraser (Perfect Real Time) (Takara, Dalian, China). The housekeeping gene b-actin from O. furnacalis was utilized as a reference to normalize the amount of transcript. Primers (Table 2) for OfRyR and Ofb-actin gene were designed to generate products located in the conserved domain without deletion sites. Each 20 mL PCR master mix contained 2 mL diluted cDNA template, 10 mL SYBR Premix Ex Taq TM , 0.4 mL forward primer (10 uM), 0.4 mL reverse primer (10 uM), 0.4 mL Rox Reference Dye II(506) and 6.8 mL RNase free H 2 O, which follows the instructions of the SYBRH Premix Ex Taq TM (Perfect Real Time) (Takara, Dalian, China). Thermal cycling was completed on a 7500 Real-time PCR System (Applied Biosystems) using the program of 95uC for 30 s, followed by 40 cycles of 95uC for 5 s and 60uC for 34 s. After the cycling protocol, a melting curve analysis from 60uC to 95uC was applied to all reactions to verify a single PCR product. The qPCR amplifications were carried out in 96-well plates. Four biological replicates were examined. The amplification efficiency of the target gene and housekeeping gene were estimated by using E = (10 21/slope )21, where the slope was derived from the plot of the cycle threshold (Ct) value versus the log of the serially diluted template concentration. Quantification of transcript level of OfRyR gene was conducted according the 2 2DDCt method [25]. The OfRyR expression data were expressed Figure 1. Cloning strategy for the isolation of OfRyR cDNA sequence. The thick black line indicates the full-length OfRyR cDNA. S1-S17 represent the nucleotide sequences of the overlapping cDNA clones. Six PCR fragments (S2, S6, S9, S13, S15 and S17) were amplified with degenerate primer sets. A degenerate primer coupled with a specific primer were used to obtain fragments S1, S3 and S7. And the rest fragments were amplified with specific primer sets. The 59-end and 39-end fragments were obtained through 59-RACE and 39-RACE. doi:10.1371/journal.pone.0075825.g001 as means 6 standard deviation (SD). Statistical analysis was performed by Duncan's Multiple Range Test for significance (P,0.05) by using SPSS 17.0 (SPSS, Inc., Chicago, IL, USA).

Diagnostic PCR analyses for the detection of alternative exons
Diagnostic PCR was conducted to detect the putative alternative exon in the individual cDNA clones. The names and nucleotide sequences of the primers used in the diagnostic PCR   reactions were listed in Table 3. Firstly, segments containing the alternative exons were amplified by primers flanking the alternative exon region. The PCR products were cloned into pMD19-T vector (Takara, Dalian, China) and cultured overnight at 4uC. Then the recombinant plasmid was transformed into the DH5a competent cells and grew in the solid LB culture medium with ampicillin, IPTG and X-Gal. The positive clones were selected and used as templates for diagnostic PCR. Mutually exclusive exons were identified using two exon-specific primer sets. Optional exons were identified using a primer spanning the exon and flanking region or a primer spanning the flanking region excluding the exon sequence. These exon-specific primers were used to generate unique PCR products representing the presence or absence of alternative exon in each clone. The PCR products were examined by agarose gel electrophoresis and visualized under the UV transilluminator. Representative clones exhibiting unique bands were sequenced to confirm the reliability of the diagnostic PCR assay. Data were collected from 20-35 clones for each segment and developmental stage and three cDNA templates of each developmental stage were tested as replications.

Cloning and analysis of OfRyR cDNA
Totally, nineteen overlapping cDNA clones of O. furnacalis were amplified, and they were spliced together to make a full-length cDNA (GenBank accession number KC355370). This cDNA contains a 15327 bp open reading frame (ORF) sequence. The ORF encodes a protein of 5108 amino acid residues with molecular weight of 577.4 kDa, and the predicted isoelectric point was 5.35.
The amino acid sequences alignment showed that the identities of OfRyR with CmRyR, CsRyR, SeRyR, PxRyR, BmRyR and DmRyR were 97, 96, 95, 92, 95 and 79%, respectively. However, identities of OfRyR with H. sapiens RyR1-3 were only 51-60% (Table 4). To investigate the evolutionary relationships among RyR sequences, a phylogenetic analysis was performed using the ClustalW and MEGA 5.0 program based on the ORF amino acid sequence of eleven insects, one nematode and three mammalian RyRs (Fig 2). The results of phylogenetic analysis revealed that OfRyR is grouped with CmRyR, CsRyR, SeRyR, BmRyR and PxRyR. The phylogenetic tree showed that the RyRs of eleven insects formed a bigger cluster, and the three mammalian RyRs formed a smaller one. This two clusters were well segregated from each other, and also from the Caenorhabditis elegans RyR.

Analysis of OfRyR amino acid sequence
Important modulator-binding sites were identified in OfRyR amino acid sequence (Fig S1) [26], was found in three sites (residues 1093-1096, 1145-1148 and 5073-5076) of OfRyR sequence. Three possible nucleotide binding sites, identified on the basis of the consensus GXGXXG motif [27], were located at positions 2734-2739, 3976-3981 and 4687-4691 in the OfRyR sequence. A potential binding site of calmodulin (CaM) in OfRyR was recognised in residues 3722-3751, corresponding to residues 3614-3643 in rabbit RyR1 sequence [28]. Two putative Ca 2+binding EF-hand motifs (residues 4186-4213 and 4221-4248), which interact with the cytoplasmic Ca 2+ to regulate the ion channel were present in OfRyR. It has been shown that C-terminal region of RyR could form the functionally important domains. For example, the pore helix was predicted to be in the loop region between the fifth and sixth transmembrane helices (residues 4949-4963) [13]. Covering part of this sequence (residues 4960-4969), the sequence motif (GVRAGGGIGD), which has been shown to constitute pore-forming segments of the Ca 2+ release channels was present in OfRyR [22]. All of these residues are also conserved in the OfRyR sequence as well as in other RyRs sequence. Moreover, a glutamate that is proposed to be involved in the Ca 2+ sensitivity in rabbit RyR1 (E 4032 ) [29] was conserved in OfRyR (E 4145 ). Residues corresponding to I 4897 , R 4913 , and D 4917 of rabbit RyR1, which play an important role in activity and conductance of the Ca 2+ release channel [30], were also conserved in OfRyR (I 4967 , R 4983 , and D 4987 ).

Expression of OfRyR mRNA in different developmental stages
The amplification efficiency (E) values of RyR and b-actin from O. furnacalis were calculated to be 0.925 and 0.963 respectively. Moreover, the relative expression level of OfRyR showed a significant difference between different developmental stages ( Fig 5). It showed the lowest expression level in egg and the highest in adult, and the relative expression level in adult and pupae were 24.1 and 13.6 times when compared to that in egg. In addition, the expression level of OfRyR in first, third and fifthinstar larvae were 1.28, 1.19 and 1.99 times of that in egg.

Developmental regulation of alternative exon usage
Multiple sequences alignment revealed two putative alternative splicing sites (named alternative splicing 1 (AS1) and alternative splicing 2 (AS2)) existed in OfRyR. This two splicing sites form three alternative exons. Figure 6 showed the nucleotide and putative amino acid sequences of the three alternative exons. AS1 located between amino acid residues 1141-1173, which corresponding to the central part of the predicted second SPRY domain and forms one pair of mutually exclusive exons (a/b). AS2 located between the predicted third and fourth RyR domains (residues 2926-2931) and forms the optional exon c.
Diagnostic PCR was used to determine the usage of each alternative exon for OfRyR mRNA at different developmental stages including egg, first instar larva, third instar larva, fifth instar larva, pupae and adult. The usage frequencies of each putative alternative exon were summarized in Figure 7

Discussion
Although insect RyR has been used as target of insecticide for more than six decades [31], little research was conducted upon it until the emergency of diamides insecticides. These insecticides show highly insecticidal activity against lepidopteran insect pests but exhibit low mammalian toxicity.  [23]. In addition, the CmRyR cDNA encoding 5087 amino acid residues was cloned by JJ Wang et al. [13]. In our study, we cloned and sequenced the full-length cDNA encoding OfRyR from third instar larva of O. furnacalis. The deduced 5108 amino acid residues of OfRyR shares a high degree of identity (92%-97%) with reported lepidopteran insect RyRs. But OfRyR shares only 48-60% identity with mammalian isoforms. The amino acid divergence between insect and mammalian RyR isoforms might be a reason why diamide insecticides show a high selectivity for insect RyRs compared to mammalian RyRs. Moreover, seven residues (N 4922 , N 4924 , N 4935 , L 4950 , L 4981 , N 5013 and T 5064 ), which are unique to lepidopteran RyR homologues have been found in CmRyR sequence [13]. The identical residues (N 4943 , N 4945 , N 4956 , L 4971 , L 5002 , N 5034 and T 5085 ) were also identified in OfRyR sequence. In contrast, different residues at the corresponding positions are shared by non-lepidopteran insect RyRs. And these residues are highly conserved in the non-lepidopteran insects, other invertebrate and vertebrate RyRs [13]. Interestingly, the residue N 4956 in OfRyR just locate in the central part of the putative pore-helix motif, and the residue L 5002 is present in the sixth transmembrance domain.
These results indicated that these residues might be involved in the differences in Ca 2+ release channel properties between lepidopteran and non-lepidopteran insects RyRs.
The diamide insecticides are potent activators of insect RyRs. It has been proved that the action of flubendiamide is highly specific, with selective toxicity in restricted insect taxa, including Lepidoptera [15]. For example, flubendiamide could induce release of Ca 2+ through the BmRyR, but not through the rabbit RyR isoforms [9]. However, the binding sites of diamide insecticides are not clear. Binding studies on microsomal membranes from insect muscles suggest that flubendiamide and chlorantraniliprole interact with a site distinct from the ryanodine binding site on the insect RyR complex [16,20]. And flubendiamide is mainly incorporated into the transmembrane domain (amino acids 4111-5084) of the sRyR, while the N-terminal sequence (residues 183-290) is a structural requirement for flubendiamide-induced activation of the sRyR [9]. This segment of sRyR (residues 183-290) shares 94-99% identity with the equivalent region of other insects RyRs (OfRyR, CmRyR, SeRyR and PxRyR), but only 49% identity between sRyR and rabbit RyR1. This is possibly an important reason why diamide insecticides show high selectivity between pest insects and mammals.
Adenine nucleotides (including ATP, ADP, AMP, cAMP, adenosine, and adenine) are also activators of RyRs. Studies using photoreactive ATP analog Bz2ATP indicated the regulation of Ca 2+ release by ATP involves an ATP binding site(s) located on the 27-kDa and 13-kDa fragments of the ryanodine receptor protein [22,32]. The consensus sequence for an ATP-binding site, GXGXXG, occurs between three and six times in the RyR sequence. And there were three possible nucleotide binding sites, located at positions 2734-2739, 3976-3981 and 4687-4691 in the OfRyR sequence. In addition, OfRyR might be regulated by cytosolic Ca 2+ , similar to mammalian RyRs [33]. Because two conserved Ca 2+ -binding EF-hands were present in the C-terminus of the OfRyR. Ca 2+ is the principal regulation mechanism for RyR in excitation-contraction coupling. And RyR was activated by micromolar concentrations of cytoplasmic Ca 2+ [3], but millimolar Ca 2+ could inhibit RyR. This biphasic behavior may result from two classes of Ca 2+ binding sites, a high-affinity activation site and a low affinity inactivation site. Another important regulator of OfRyR is the bifunctional protein calmodulin (CaM). The   [34]. When free Ca 2+ is between 10 29 and 10 27 mol/L, apo-CaM activates RyR1 and RyR3 (but not RyR2), whereas the concentration of Ca 2+ .10 25 mol/L, Ca 2+ -CaM inhibits all three RyR isoforms [35]. A potential binding site for both apo-CaM and Ca 2+ -CaM was identified in the OfRyR sequence (residues 3722-3751) [28]. So this region in OfRyR sequence might be the candidates for CaM binding and play a crucial role in the physiology process. Real-time quantitative PCR indicated the expression level of OfRyR gene was quite different in the development stages. It showed high expression level in adult and pupae and the lowest in egg. However, Guo et al have reported that the expression levels of the PxRyR in different-instar larvae and adults were much higher than those of the prepupae and pupae in P. xylostella [31]. Therefore, we could deduce that the expression level of RyRs might vary in different insects.
Alternative splicing of pre-mRNA transcripts is a prevalent feature of gene processing and powerful mechanism to generate remarkable protein diversity from single gene [36]. And it plays an important role in physiology, disease and developmental stagespecific processes [37,38]. Recently, many alternative splice variants of RyR isoforms have been identified in human, rabbit and mouse. For instance, alternative splicing of ASI residues (A 3481 -Q 3485 ) contribute to an inhibitory module in RyR1 that influences excitation-contraction coupling [39]. And ASI residues is developmentally regulated: the residues are present in adult ASI(+)RyR1, but absent in the juvenile ASI(2)RyR1 which is over-expressed in adult myotonic dystrophy type 1 [1]. The 24-bp  splice insertion of human RyR2, which is present at low level in embryonic and adult hearts can suppress intracellular Ca 2+ fluxes that protects cells from caffeine-evoked apoptosis [36]. And a short isoform of spliced RyR3, which lacks a 29 amino acid fragment (H 4406 -K 4434 ) containing a predicted transmembrane helix, is not a functional Ca 2+ channel but inhibits the efficiency of the Ca 2+induced Ca 2+ release mechanism by a negative modulation of RyR2 in native smooth muscle cells [36].
Similarly, alternative splicing also exists in insects RyRs. It's reported that the multiple mRNAs of D. melanogaster RyR can be generated by two different splicing patterns: (i) the mutually exclusive incorporation of exon into mRNA, and (ii) the choice of one out of two splice-acceptor sites [20]. The same splicing patterns were also identified in the C. medinalis and O. furnacalis RyR genes. In our study, two alternative splice variants (AS1 and AS2) showing developmental regulation were found in the OfRyR gene. AS1 (residues 1141-1173) located in the central of the second SPRY domain (SPRY2). The SPRY domain is a proteinprotein interaction module and is implicated in important biological pathways, including regulate innate and adaptive immunity [40]. SPRY2 has been identified as an in vitro binding partner for N-terminal residues in the II-III loop of the skeletal muscle DHPR, as well as for the scorpion toxins imperatoxin A and maurocalcine. In addition, SPRY2 domain and its F loop (P 1107 -A 1121 ) bind to the alternatively spliced residues and neighboring basic residues (ASI/basic) of RyR1 via an electrostatic interaction to regulate excitation contraction coupling. Mutation of the negatively charged amino acid of the SPRY2 F loop reduce SRRY2 domain binding to the ASI/basic region [41]. Since the mutually exclusive exons a and b are highly different in OfRyR, and exon b contains less negatively charged amino acids than exon a. Therefore, the OfRyR splice variants generated by mutually exclusive exons a and b might probably possess different function and binding effect. In our study, the splice variants showed significantly developmental regulation in OfRyR. This result implied that alternative splicing might play an important part in the temporal encoding of the Ca 2+ release channel, and might involve in the differences in channel properties and function. Figure S1 Analysis of the deduced OfRyR amino acid sequences. Predicted transmembrance domains were overlined and labeled TM1-TM6. Conserved residues for the putative EFhand motifs, pore-forming segments, pore-helix, adenine ring binding sites (Y[GAST][VG][KTQSN]), nucleotide binding sites (GXGXXG) and potential CaM binding site were also overlined and labeled. The residues in OfRyR (I4967, R4983 and D4987), which play an important role in activity and conductance of the Ca2+ release channel were marked with triangles. Pentastar indicated glutamate (E4145) that is proposed to be involved in the Ca2+ sensitivity in OfRyR.

Acknowledgments
We would like to thank the editor Subba Reddy Palli and two anonymous reviewers for their helpful comments on an earlier version of this paper.