Introduction
Chitin is a characteristic constituent of the cuticles of insects and other arthropods (Kramer et al., Reference Kramer, Hopkins and Schaefer1995; Merzendorfer, Reference Merzendorfer2006). In addition to cuticles, chitin is also found in the peritrophic matrix, tracheae, and at muscle attachment points (Kramer & Koga, Reference Kramer and Koga1986; Hegedus et al., Reference Hegedus, Erlandson, Gillott and Toprak2009). The chitin-containing tissues must be remodeled during insect growth and development, and the chitinases (EC 3.2.1.14) are the enzymes responsible for the degradation of the linear polysaccharides in chitin (Kramer & Muthukrishnan, Reference Kramer, Muthukrishnan, Lawrence, Kostas and Sarjeet2005). Insect chitinases belong to glycosyl hydrolase family 18 (GH18) and are responsible for the endo-degradation of chitin; thus, chitinases are a target in pest management (Merzendorfer & Zimoch, Reference Merzendorfer and Zimoch2003). Insect chitinases and chitinase-like genes were recently identified from the completed genome sequences in Drosophila melanogaster, Tribolium castaneum and Anopheles gambiae (Zhu et al., Reference Zhu, Deng, Vanka, Brown, Muthukrishnan and Kramer2004, Reference Zhu, Arakane, Banerjee, Beeman, Kramer and Muthukrishnan2008c ; Zhang et al., Reference Zhang, Zhang, Arakane, Muthukrishnan, Kramer, Ma and Zhu2011a ). Moreover, the deduced proteins of these genes have been classified into eight groups based on the phylogenetic analysis of catalytic domain (CD) sequences (Kramer & Muthukrishnan, Reference Kramer and Muthukrishnan1997; Arakane & Muthukrishnan, Reference Arakane and Muthukrishnan2010).
Although several chitinase genes from lepidopteran species have been characterized, including those of Manduca sexta (Kramer et al., Reference Kramer, Corpuz, Choi and Muthukrishnan1993), Bombyx mori (Kim et al., Reference Kim, Shin, Bae, Kim and Park1998; Mikitani et al., Reference Mikitani, Sugasaki, Shimada, Kobayashi and Gustafsson2000; Abdel-Banat & Koga, Reference Abdel-Banat and Koga2001), Hyphantria cunea (Kim et al., Reference Kim, Shin, Bae, Kim and Park1998), Spodoptera litura (Shinoda et al., Reference Shinoda, Kobayashi, Matsui and Chinzei2001), Choristoneura fumiferana (Zheng et al., Reference Zheng, Zheng, Cheng, Ladd, Lingohr, Krell, Arif, Retnakaran and Feng2002), Helicoverpa armigera (Ahmad et al., Reference Ahmad, Rajagopal and Bhatnagar2003), Lacanobia oleracea (Fitches et al., Reference Fitches, Wilkinson, Bell, Bown, Gatehouse and Edwards2004), Spodoptera frugiperda (Bolognesi et al., Reference Bolognesi, Arakane, Muthukrishnan, Kramer, Terra and Ferreira2005) and Ostrinia nubilalis (Khajuria et al., Reference Khajuria, Buschman, Chen, Muthukrishnan and Zhu2010), only one chitinase gene or cDNA was identified in most of these species. Furthermore, most of the identified lepidopteran chitinases are group I chitinases, which are the enzymatically well characterized chitinases that are isolated from molting fluid or integument; there are some exceptions, such as a few bacterial-type chitinase genes (Cht-h) (Daimon et al., Reference Daimon, Hamada, Mita, Okano, Suzuki, Kobayashi and Shimada2003, Reference Daimon, Katsuma, Iwanaga, Kang and Shimada2005), a gut-specific group IV chitinase gene in O. nubilalis (Khajuria et al., Reference Khajuria, Buschman, Chen, Muthukrishnan and Zhu2010) and certain chitinase-like genes (Imaginal disk growth factor, IDGF) in group V (Zhang et al., Reference Zhang, Iwai, Tsugehara and Takeda2006; Wang et al., Reference Wang, Sakudoh, Kawasaki, Iwanaga, Araki, Fujimoto, Takada, Iwano and Tsuchida2009). However, as whole-genome sequences are continually reported, remarkable advances have occurred in the understanding of the entire family of lepidopteran chitinases. Based on the screening of the genome of B. mori, lepidopteran species have multiple genes that encode chitinase proteins (Nakabachi et al., Reference Nakabachi, Shigenobu and Miyagishima2010; Pan et al., Reference Pan, Lu, Wang, Yin, Ma, Ma, Chen and He2012). In addition to the eight groups that were described previously, two new groups of chitinases (groups IX and X) were created in a genome-wide analysis of the tobacco hornworm M. sexta (Tetreau et al., Reference Tetreau, Cao, Chen, Muthukrishnan, Jiang, Blissard, Kanost and Wang2015).
In this study, we characterized the entire chitinase gene family from Plutella xylostella, which is a major pest of several agricultural crops worldwide. We first cloned four chitinase genes using degenerated polymerase chain reaction (PCR) and rapid amplification of cDNA ends (RACE-PCR) strategies before the complete genome sequence and the transcriptomic database of P. xylostella were released (Jouraku et al., Reference Jouraku, Yamamoto, Kuwazaki, Urio, Suetsugu, Narukawa, Miyamoto, Kurita, Kanamori, Katayose, Matsumoto and Noda2013; You et al., Reference You, Yue, He, Yang, Yang, Xie, Zhan, Baxter, Vasseur, Gurr, Douglas, Bai, Wang, Cui, Huang, Li, Zhou, Wu, Chen, Liu, Wang, Xu, Lu, Hu, Davey, Smith, Chen, Xia, Tang, Ke, Zheng, Hu, Song, You, Ma, Peng, Zheng, Liang, Chen, Yu, Zhang, Liu, Li, Fang, Li, Zhou, Luo, Gou, Wang and Yang2013). BLAST searches of these databases identified an additional 11 genes that encoded chitinase proteins. The characteristics of the chitinase genes in the diamondback moth P. xylostella, 15 in total, were determined in this study, including the domain structures, expression patterns, and phylogenetic relationships among the chitinase genes from different orders, particularly Lepidoptera.
Materials and methods
Insect culture
The P. xylostella were maintained in a growth chamber under a 16:8 h light:dark photoperiod at 25 ± 2°C and 60–70% relative humidity (RH). The larvae were reared on 10–15-day-old cabbage (Brassica rapa L.), and the adults were fed honey as a dietary supplement.
Total RNA extraction and synthesis of first-strand cDNA
The total RNA was extracted from the larvae of all instars of P. xylostella using a TRIzol total RNA isolation kit (Invitrogen) according to the manufacturer's instructions. One microgram of total RNA was used as a template for first-strand cDNA synthesis using the ImProm-II™ Reverse Transcription system (Promega) with an Oligo-(dT)20 primer. The reverse transcription was performed at 42°C for 60 min and was terminated at 72°C for 10 min.
Degenerate PCR amplification of putative chitinase gene fragments
The partial chitinase fragments were amplified with a pair of degenerate primers. The forward and reverse primers that corresponded to two conserved regions of the Family 18 insect chitinases were synthesized according to the amino acid sequences DLDWEYP and WAIDMDDF, respectively. The primers and the conditions for PCR are shown in tables S1 and S2, respectively. The PCR products were separated on 1% agarose gel, and a product mixture of approximately 700 bp was excised from the gel. After purification using a DNA gel extraction spin column (Bioman), the 700 bp fragments were subcloned into the T and A cloning vectors (RBC Bioscience). The positive clones were selected, and the plasmids were prepared for DNA sequencing. The sequencing results showed that five different chitinase fragments were amplified by degenerated PCR after a BLAST search of the NCBI database.
Rapid amplification of cDNA ends
The SMART™ RACE cDNA Amplification kit (Clontech) was used to clone the full length of the chitinase cDNAs. The 3’ RACE and 5’ RACE cDNAs were synthesized from the total RNA isolated as previously described using SMARTScrible™ Reverse Transcriptase (Clontech), according to the manufacturer's instructions. Three forward and three reverse gene-specific primers were designed based on the known cDNA sequences amplified by the degenerate PCR. The amplification of the cDNA to end was achieved using the Universal Primer A Mix (UPM) supplied in the kit and pairing with one forward gene-specific primer in the 3’ RACE PCR reaction and with one reverse gene-specific primer in the 5’ RACE PCR reaction. The primers, conditions of the PCR, and amplified range are shown in tables S1 and S2. The PCR products were recovered and were cloned into the RBC vector as previously described. The positive clones were selected for insert sequencing.
Identification of chitinase-like genes from P. xylostella genome and transcriptomic databases
The genome of P. xylostella (GenBank accession AHIO00000000) (http://www.ncbi.nlm.nih.gov/assembly/GCF_000330985.1) was screened for genes that encoded chitinase-like proteins by using BLAST searches. The BLASTP searches were performed at the website of the NCBI (http://www.ncbi.nlm.nih.gov/) using the amino acid sequences of the chitinase-like proteins of D. melanogaster, T. castaneum and A. gambiae obtained from the NCBI and of M. sexta from the Manduca base (http://agripestbase.org/manduca) as queries. KONAGAbase (http://dbm.dna.affrc.go.jp/px/), a DBM comprehensive transcriptomic and draft genomic sequences database, was also searched using the identical BLASTP method.
Default parameters were used and then the candidate chitinase-like genes were confirmed by searching the BLASTX algorithm against the nonredundant (nr) NCBI nucleotide database.
Domain architecture and phylogenetic analysis
The domain structures of the putative chitinase genes were analyzed using the SMART online program (http://smart.embl-heidelberg.de/), and the signal peptides were predicted through the SignalP 4.0 program (http://www.cbs.dtu.dk/services/SignalP). The transmembrane regions were analyzed using TMHMM Server version 2.0 (http://www.cbs.dtu.dk/services/TMHMM). The CD search at the NCBI (http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi) was used to confirm the predicted domains. The phylogenetic tree was generated based on the amino acid sequences of the CD. First, the amino acid sequences were aligned using ClustalX version 2.0 (Larkin et al., Reference Larkin, Blackshields, Brown, Chenna, McGettigan, McWilliam, Valentin, Wallace, Wilm, Lopez, Thompson, Gibson and Higgins2007), and the phylogenetic trees were inferred with the neighbor-joining (NJ) method (p-distance model, uniform rates and complete deletion of gaps or missing data). The NJ tree was constructed using the program MEGA version 6.0 (Tamura et al., Reference Tamura, Stecher, Peterson, Filipski and Kumar2013). The bootstrap probability for each node was calculated by generating 5000 bootstrap replicates. The letters A–E denote multiple CDs from the N- to the C-terminuses in the same protein.
Reverse transcription PCR (RT-PCR) analysis
To investigate the stage-specific expressions of the chitinase genes, the total RNA was isolated from each of eight developmental stages, including the egg, first to fourth instar larvae, prepupa, pupa and adult, using the TRIzol Total RNA Isolation kit (Invitrogen). The total RNA was also isolated from fourth instar larval tissue samples, including the midgut, hemolymph, Malpighian tube and carcass (whole larva after the gut was removed), to study the tissue-specific expression. The cDNA synthesis was previously described, and the information for the primer sequences is shown in table S3. The PCR products were resolved on 1.5% agarose gel and were visualized by staining with ethidium bromide. The β-actin gene from P. xylostella was used as the loading reference for RT-PCR analyses. The RT-PCR was repeated a minimum of three times for each gene at each developmental stage and for each tissue type of the fourth instar larva.
Results
Identification and classification of chitinase genes in P. xylostella
Four RACE-PCR cloned chitinase genes from P. xylostella were denoted as PxCht1, PxCht2, PxCht3 and PxCht3-like. The complete cDNA sequences and the corresponding amino acid sequences were submitted to the GenBank database with the nucleotide accession numbers of FJ613480, JQ417265, JQ417267 and JQ417266, respectively. The phylogenetic analysis of the four chitinase proteins (accession: ACU42267, AFI55112, AFI55114 and AFI55113) with proteins from A. gambiae (AgCht), Aedes aegypti (AaCht), T. castaneum (TcCht) and D. melanogaster (DmCht) was generated using the method proposed by Zhang et al. (Reference Zhang, Zhang, Arakane, Muthukrishnan, Kramer, Ma and Zhu2011a ). In fig. 1, PxCht1 and PxCht2 are grouped into groups I and III, respectively, with bootstrap values greater than 90%. However, PxCht3 and PxCht3-like were tentatively placed in the group IV with low bootstrap values.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary-alt:20161111063157-88646-mediumThumb-S0007485316000511_fig1g.jpg?pub-status=live)
Fig. 1. Phylogenetic analysis of chitinase proteins based on catalytic domain sequences. The RACE-PCR cloned chitinase (accession: ACU42267, AFI55112, AFI55114 and AFI55113) and chitinases listed in table S4 were used. Phylogenetic trees of insect chitinases were generated with the MEGA6 software after alignment using ClustalX2 software. Bootstrap values were obtained with the neighbor-joining method using 5000 replications. The letters A–E denote multiple CDs from N- to C-terminuses in the same protein.
The BLAST searches of the P. xylostella genome (assembly DBM_FJ_V1.1) and transcriptomic databases (KONAGAbase) identified 15 genes that encoded chitinase proteins: thirteen Cht genes, one IDGF gene and one Cht-h gene respectively. In fig. 2, the molecular phylogenetic analysis of the 15 P. xylostella chitinases (table 1) with other four lepidopteran chitinases is shown. The 13 Danaus plexippus, 13 Papilio xuthus, 11 Bombyx mori and 11 M. sexta chitinase proteins obtained from the NCBI, NCBI, both NCBI and silkDB (http://silkworm.genomics.org.cn/) and the Manduca base, respectively (table S5), were included in the analysis. The chitinase genes detected in P. xylostella were named with a number that corresponded to the phylogenetic group in which they were grouped: PxCht (chitinases, all groups except group V) or PxIDGF (chitinase-like proteins, the Imaginal Disk Growth Factors, group V). The four RACE-PCR cloned Cht genes that were described previously were found again in the screening of the P. xylostella genome. Thus, hereafter in this study, PxCht1, PxCht2, PxCht3 and PxCht3-like were renamed PxCht5, PxCht7, PxCht25-1 and PxCht25-2, respectively. The detailed information is listed in table 1 for the genes that encoded the chitinase proteins searched from P. xylostella.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary-alt:20161111063157-84726-mediumThumb-S0007485316000511_fig2g.jpg?pub-status=live)
Fig. 2. Phylogenetic analysis of lepidopteran chitinase proteins based on catalytic domain sequences. The chitinase proteins listed in table 1 and table S5 were used. Phylogenetic trees of lepidopteran chitinases were generated with the MEGA6 software after alignment of first catalytic domain using ClustalX2 software. Bootstrap values were obtained with the neighbor-joining method using 5000 replications.
Table 1. Information on genes encoding chitinase and chitinase-like proteins searched from P. xylostella genome and transcriptomic databases. Genome assembly gap found in scaffold is denoted with an asterisk, the presence of glutamic acid residues in the CD is shown in boldface, and NS represents not shown in the prediction.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary-alt:20161111063157-23289-mediumThumb-S0007485316000511_tab1.jpg?pub-status=live)
The chitinase genes from five lepidopterans were clustered into 12 groups (fig. 2): 11 of the groups (I−X and h) were previously reported in M. sexta (Tetreau et al., Reference Tetreau, Cao, Chen, Muthukrishnan, Jiang, Blissard, Kanost and Wang2015) and one new group was created. The 15 P. xylostella chitinases were clustered into 10 of these 11 groups, not including group VIII, and two chitinases might be in a new group of the chitinase family. Eight of the 11 groups in P. xylostella (groups I-III, V, VI, IX, X and h) contained a single chitinase gene: PxCht5 in group I, PxCht10 in group II, PxCht7 in group III, PxIDGF in group V, PxCht6 in group VI, PxCht1 in group IX, PxCht3 in group X and PxCht-h in group h. The other three groups (IV, VII and the new group) contained 2, 3 and 2 chitinase genes, respectively: PxCht8-1 and PxCht8-2 in group IV, PxCht2-1, PxCht2-2 and PxCht2-3 in group VII, and PxCht25-1 and PxCht25-1 in the new group. Moreover, both PxCht2-3 and PxCht6 were predicted to have two transcript variants, and the letters a and b denote each of the isoforms.
Structure of chitinase genes from P. xylostella
Fig. 3 shows the domain architecture of the deduced amino acid sequences from the P. xylostella chitinase genes. Most of the predicted amino acid sequences contained single copy of putative CD (GH18 domain: pfam00704) (see: http://www.ncbi.nlm.nih.gov/cdd, for more information on the conserved domains), whereas PxCht7 had two copies of this domain and PxCht10 had five copies. One or more chitin-binding domains (CBD) (CBM-14: pfam01607) were detected in PxCht3, PxCht5, PxCht6, PxCht7, PxCht8-1, PxCht8-2 and PxCht10. Additionally, PxCht-h contained a polycystic kidney disease 1 domain (PKD1: smart00089).
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary-alt:20161111063157-72970-mediumThumb-S0007485316000511_fig3g.jpg?pub-status=live)
Fig. 3. Domain architecture of the chitinase proteins of Plutella xylostella. The domain structure was analyzed with the SMART tool (http://smart.embl-heidelberg.de/). The accession numbers of all the proteins used are listed in table 1.
Fig. 4 shows the alignment with the putative CD of the 15 chitinase proteins from P. xylostella. Most sequences contained four highly conserved regions (CR I-CR IV), a characteristic feature of all insect chitinases (Kramer & Muthukrishnan, Reference Kramer and Muthukrishnan1997; De la Vega et al., Reference De la Vega, Specht, Liu and Robbins1998). The signature sequences were KXXXXXGGW, FDGXDLDWEYP, MXYDXXG and GXXXWXXDXD in which X was a non-specified amino acid. In PxCht1 and PxCht3, all four regions were poorly conserved and devoid of the residue E, which is the putative proton donor in the catalytic mechanism in the CR II (Watanabe et al., Reference Watanabe, Kobori, Miyashita, Fujii, Sakai, Uchida and Tanaka1993; Lu et al., Reference Lu, Zen, Muthukrishnan and Kramer2002). The CD range label in boldface indicates the presence of E residue in CR II (table 1). Moreover, genome assembly gaps were found in several PxCht sequences (denoted with asterisks in table 1).
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary-alt:20161111063157-68876-mediumThumb-S0007485316000511_fig4g.jpg?pub-status=live)
Fig. 4. Conserved regions in catalytic domains of chitinase proteins in Plutella xylostella. Multiple alignments of the CD sequences were generated with the GeneDoc tool. Four conserved regions (CR I-IV) are labeled. Their amino acid sequences are shown above the boxes. The accession numbers of all the proteins used are listed in table 1.
Expression of PxCht genes in different developmental stages and tissues
The stage- and tissue-specific expression patterns of the PxCht genes were analyzed with RT-PCR. Fig. 5 shows that PxCht1, PxCht2-1, PxCht2-2, PxCht2-3, PxCht5, PxCht7, PxCht8-1, PxCht8-2 and PxIDGF were expressed at all developmental stages, PxCht6 and PxCht-h were expressed in most stages except for the adult stage, and PxCht10 was not detected in pupae and adults. PxCht3 was detected only at the prepupal and pupal stages, and PxCht25-1 and PxCht25-2 were larval stage specific. Based on the tissue-specific expression patterns, PxCht5, PxCht7 and PxIDGF were expressed in all the tissues that were examined, PxCht1, PxCht6, PxCht10 and PxCht-h were expressed in most of the tissues, PxCht1, PxCht10 and PxCht-h were not detected in the hemolymph, and PxCht6 was not expressed in the gut tissue. Additionally, PxCht8-1, PxCht8-2, PxCht25-1 and PxCht25-2 were gut-specific, and PxCht2-1, PxCht2-2 and PxCht2-3 were carcass-specific. Furthermore, the expression of PxCht3 was not found in all the tissues dissected from fourth instar larvae.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary-alt:20161111063157-41379-mediumThumb-S0007485316000511_fig5g.jpg?pub-status=live)
Fig. 5. Expression of P. xylostella chitinase genes in different developmental stages and different tissues as evaluated by RT-PCR. RNA samples were isolated from eight developmental stages and five tissues in the fourth instar larvae. The primers used for the RT-PCR are shown in table S3. The actin (PxActin) gene was used as the reference gene.
Discussion
The first insect chitinase gene was cloned from M. sexta (Kramer et al., Reference Kramer, Corpuz, Choi and Muthukrishnan1993). In the early stages of research, the cDNA cloned from several insects indicated that only a single chitinase gene was found in each species. However, in later studies, large and diverse groups of the chitinase genes were found. The phylogenetic analysis based on the protein sequences of the CDs first assigned these chitinase proteins into five separate groups (I–V) (Zhu et al., Reference Zhu, Deng, Vanka, Brown, Muthukrishnan and Kramer2004, Reference Zhu, Arakane, Banerjee, Beeman, Kramer and Muthukrishnan2008c ), and then the family of chitinase proteins was expanded to eight groups (I–VIII) (Zhang et al., Reference Zhang, Zhang, Arakane, Muthukrishnan, Kramer, Ma and Zhu2011a ). Although B. mori was the first species from which multiple chitinase genes were identified (Kim et al., Reference Kim, Shin, Bae, Kim and Park1998; Abdel-Banat & Koga, Reference Abdel-Banat and Koga2001; Daimon et al., Reference Daimon, Hamada, Mita, Okano, Suzuki, Kobayashi and Shimada2003), the progression of investigation on the entire family of lepidopteran chitinase genes is much slower than that of dipteran and coleopteran species. A better understanding of the lepidopteran chitinase genes was achieved as genome sequences became available (Pan et al., Reference Pan, Lu, Wang, Yin, Ma, Ma, Chen and He2012; Tetreau et al., Reference Tetreau, Cao, Chen, Muthukrishnan, Jiang, Blissard, Kanost and Wang2015). Currently, the chitinase family of genes is classified into 11 groups (groups I–X and h, a lepidopteran-specific chitinase) (Tetreau et al., Reference Tetreau, Cao, Chen, Muthukrishnan, Jiang, Blissard, Kanost and Wang2015).
In this study, we successfully identified 15 individual chitinase genes from P. xylostella, which is a recalcitrant pest of several agricultural crops. Based on the phylogenetic and structural analyses, these chitinase genes were named in accordance with the nomenclature reported in recent studies (Zhang et al., Reference Zhang, Zhang, Arakane, Muthukrishnan, Kramer, Ma and Zhu2011a ; Tetreau et al., Reference Tetreau, Cao, Chen, Muthukrishnan, Jiang, Blissard, Kanost and Wang2015). PxCht5, PxCht7, PxCht25-1 and PxCht25-2 were first cloned by RACE-PCR and were also detected in the later genome searching method later. Our cloned cDNA added 153 and 27 transcript bases to patch the genome assembly gaps in PxCht5 and PxCht7, respectively. According to the results of the domain and phylogenetic analyses (figs 1, 2 and 3), PxCht5 was classified in group I and was typically multi-domain, including a signal peptide, a CD, an S/T-rich linker that was heavily glycosylated, and a CBD with six consensus cysteines (belonging to chitin-binding module 14). Based on the results of RNA interference to silence the TcCht5 gene, the group I chitinase might only be required for pupal-adult molting in T. castaneum (Zhu et al., Reference Zhu, Arakane, Beeman, Kramer and Muthukrishnan2008a ). However, the results were different in S. exigua and suggested that the group I chitinase had an important role during the larval–pupal and pupal–adult stages, and down-regulated expression caused abnormal and lethal effects (Zhang et al., Reference Zhang, Chen, Yao, Pan and Zhang2012). The current results suggested that PxCht5 might be involved in multiple functions associated with chitin turnover because the transcripts were detected in all stages and tissues, similar to other lepidopteran group I chitinases (fig. 5).
PxCht7 contained 2 CDs and 1 CBD, which was a characteristic of group III chitinases (fig. 3). The Cht7 proteins typically possess a transmembrane region at the N-terminus and are predicted to be membrane-anchored proteins. Additionally, a signal peptide existed in front of the transmembrane segment of the Cht7 proteins in D. melanogaster and A. gambiae but not in T. castaneum (Zhu et al., Reference Zhu, Arakane, Banerjee, Beeman, Kramer and Muthukrishnan2008c ) and M. sexta (Tetreau et al., Reference Tetreau, Cao, Chen, Muthukrishnan, Jiang, Blissard, Kanost and Wang2015). In the current study, PxCht7 had only a signal peptide and lacked a transmembrane region. The lack of a transmembrane region in PxCht7 suggested that the protein was not membrane-anchored, similar to the other putative group III chitinases. For example, a chitinase from the hard tick Haemaphysalis longicornis (You et al., Reference You, Xuan, Tsuji, Kamio, Taylor, Suzuki and Fujisaki2003) is the only biochemically well characterized group III chitinase that contains only a signal peptide in front of 2 CDs. The chitinase of hard ticks is located between the old and the new cuticle in molting nymphs, suggesting a role in molting. Another example is Cht4 from a pea aphid (Acyrthosiphon pisum) (Nakabachi et al., Reference Nakabachi, Shigenobu and Miyagishima2010), whose expression level is dominant in embryos, indicating that this chitinase was essential for embryonic development in aphids. In T. castaneum, Cht7 functions in tissue differentiation rather than in molting because when the TcCht7 was silenced, there was defective abdominal contraction, elytral expansion and hindwing folding (Zhu et al., Reference Zhu, Arakane, Beeman, Kramer and Muthukrishnan2008a ). PxCht7 was expressed in all stages and tissues (fig. 5); however, the function of group III chitinases in Lepidoptera remains unclear.
PxCht25-1 and PxCht25-2 were cloned by using the same primers because their open reading frame shared a high level of sequence similarity, and the major difference between the two mRNA sequences was found in the 3'UTR. Table 2 shows the percentage of protein sequence identity. Similar to multiple chitinases from Dipteran and T. castaneum that could not be clustered in other groups, PxCht25-1 and PxCht25-2 were tentatively placed in the most divergent group IV (fig. 1). In the phylogenetic relationship among Lepidopteran chitinases, they fell in group I with NJ tree (fig. 2) but not with Maximum likelihood tree (fig. S1). Based on the sequence similarity and the phylogenetic analysis, we proposed that PxCht25-1 and PxCht25-2 were paralogous genes that could be in a new group of the chitinase family because they could not form monophyletic groups with any known chitinase consistently. Both PxCht25-1 and PxCht25-2 proteins possessed a signal peptide and a CD (fig. 3) and showed similar patterns of expression, exclusively expressed in the larval stages and gut tissue, whereas PxCht25-2 was highly expressed in the first and second larval instars (fig. 5). The domain architectures and expression patterns of PxCht25-1 and PxCht25-2 were similar to one O. nubilalis chitinase, which was the first identified group IV chitinase in Lepidopteran (Khajuria et al., Reference Khajuria, Buschman, Chen, Muthukrishnan and Zhu2010). The larval and gut-specific chitinases are important in regulating the chitin content of the peritrophic matrix (PM) and are essential for larval growth and development; PxCht25-1 and PxCht25-2 were likely to have a similar function.
Table 2. Protein sequence identity profiles of PxCht2s (A), PxCht8s (B) and PxCht25s (C). PxChts belong to the same group and were aligned and statistically analyzed using the GeneDoc tool. Statistical reports show the calculations for exact matches, Juxtaposition greater than zero and aligned with gaps from top to bottom; absolute values are on the left and percent values are on the right.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary-alt:20161111063157-97512-mediumThumb-S0007485316000511_tab2.jpg?pub-status=live)
Sequencing gaps were found in PxCht8-2, which lead to incorrect genome assembly. A genomic PCR was performed to obtain the missing sequences. The complete PxCht8-2 genome sequence was reassembled and the deduced protein sequence length was updated from 370 to 567 bp (fig. S2). Most group IV chitinases are characterized by the absence of a CBD and an S/T-rich region and are expressed only in the gut or the fat body (Yan et al., Reference Yan, Cheng, Narashimhan, Li and Aksoy2002; Arakane & Muthukrishnan, Reference Arakane and Muthukrishnan2010). PxCht8-1 and PxCht8-2 were orthologous to Cht8s (fig. 2), members of group IV chitinases that contain a CBD (fig. 3) (Zhu et al., Reference Zhu, Arakane, Banerjee, Beeman, Kramer and Muthukrishnan2008c ; Tetreau et al., Reference Tetreau, Cao, Chen, Muthukrishnan, Jiang, Blissard, Kanost and Wang2015). Additionally, PxCht8-1 and PxCht8-2 had similar patterns of expression and were expressed exclusively in gut tissue and throughout all stages of development (fig. 5). No developmental defects were observed when the expression of TcCht8 was down-regulated by RNAi (Zhu et al., Reference Zhu, Arakane, Beeman, Kramer and Muthukrishnan2008a ). Further experiments are required to determine whether the Cht8s have a role in the digestion of chitin-containing material or in the immunity against pathogens containing chitin.
Similar to the new group and group IV chitinases, group VII in P. xylostella consisted of multiple genes (PxCht2-1, PxCht2-2 and PxCht2-3). Multiple group VII chitinase genes were also observed in D. plexippus, B. mori and P. xuthus (table S5). Multiple group I chitinases have been identified in three mosquito species, including five in A. gambiae, four in Ae. aegypti and three in C. quinquefasciatus (Zhang et al., Reference Zhang, Zhang, Arakane, Muthukrishnan, Kramer, Ma and Zhu2011b ). A gene cluster consisting of multiple Cht5 genes may have resulted from gene tandem duplications. A similar phenomenon of gene duplication for two LmCht5 genes in Locusta migratoria suggested that the gene duplication of Cht5 might be not unique to the mosquito lineage (Li et al., Reference Li, Zhang, Wang, Liu, Ma, Sun, Li and Zhu2015). However, the gene duplication of group I chitinase was not observed in P. xylostella, but was observed in the new group and groups IV and VII. The locations of PxCht25s, PxCht8s and PxCht2s were on the different scaffold (table 1). The duplication of these chitinases in the new group and groups IV and VII have not been previously reported and the mechanisms are required to further determine. The similar patterns of expression among PxCht2s suggested that they might be involved in the chitin turnover associated with molting because the transcripts were detected primarily in the carcass (fig. 5).
The generation of multiple isoforms from a single group I chitinase gene through alternative splicing was first reported in B. mori (Abdel-Banat & Koga, Reference Abdel-Banat and Koga2002). Both PxCht2-3 and PxCht6 were predicted to have two transcript variants, and PxCht6 is a member of the group VI chitinases (fig. 2). The alternative splicing event in PxCht6 led to the generation of two isoforms that differed in protein length and structure, and PxCht6b had an additional CBD at the C-terminal region (fig. 3). A similar phenomenon was also found for MsCht6 during the search for the transcripts in Manduca Base (ftp://ftp.bioinformatics.ksu.edu/pub/Manduca/OGS2/OGS2_20140407_transcripts.fa). In PxCht2-3, the alternative splicing event led to equal lengths and only three differences in their deduced amino acid sequences (table 2).
The group II chitinase (PxCht10) was large chitinase that had multiple CDs and CBDs (fig. 3), whose number and location showed conserved arrangements in the same order. The arrangement of CD and CBD units in lepidopterans is in which
represents a CD and
represents a CBD, according to three group II chitinases from B. mori (Pan et al., Reference Pan, Lu, Wang, Yin, Ma, Ma, Chen and He2012), M. sexta (Tetreau et al., Reference Tetreau, Cao, Chen, Muthukrishnan, Jiang, Blissard, Kanost and Wang2015) and Danaus plexippus (GeneBank: EHJ65741). Dipterans have 4 CDs and 4 CBDs (Zhu et al., Reference Zhu, Arakane, Banerjee, Beeman, Kramer and Muthukrishnan2008c
; Zhang et al., Reference Zhang, Zhang, Arakane, Muthukrishnan, Kramer, Ma and Zhu2011a
), with the arrangement
, and Coleopterans have 5 CDs and 5 CBDs (Royer et al., Reference Royer, Fraichard and Bouhin2002; Arakane & Muthukrishnan, Reference Arakane and Muthukrishnan2010), with the arrangement
. A comparison of the CRIIs in each of the CDs from the three lepidopteran group II chitinases showed that the glutamate residue was uniformly detected in the third and the fourth CRIIs. TcCht10 was expressed in all stages and played a vital role in the embryo hatch, larval molt, pupation and adult metamorphosis (Zhu et al., Reference Zhu, Arakane, Beeman, Kramer and Muthukrishnan2008a
); PxCht10 might be not involved in pupation and adult metamorphosis because the transcripts were not detected in the pupal and adult stages (fig. 5). However, further investigations are needed to verify this assumption.
PxCht1 was in group IX (fig. 2) and was previously identified as Stabilin-1 interacting chitinase-like protein (SI-CLP). The SI-CLP proteins are secreted by lysosomes to interact with the transmembrane receptor stabilin-1 and are involved in protein sorting during endocytosis (Kzhyshkowska et al., Reference Kzhyshkowska, Gratchev and Goerdt2006). Similar to MsCht1 (Tetreau et al., Reference Tetreau, Cao, Chen, Muthukrishnan, Jiang, Blissard, Kanost and Wang2015), PxCht1 had a different set of conserved residues in CRII (fig. 4). In group I to IV chitinases, the consensus sequence for CRIIs consisted of FDG(L/F)DLDWE(Y/F)P, whereas the CRII sequence in PxCht1 was FDGVVLEMLSQ, which was devoid of the residue E, only had the first aspartic acid and appeared to lack the other acidic groups shown to influence catalytic activity (Lu et al., Reference Lu, Zen, Muthukrishnan and Kramer2002; Zhang et al., Reference Zhang, Huang, Fukamizo, Muthukrishnan and Kramer2002). PxCht3 was in group X (fig. 2), which contained one CD followed by two very closely spaced tandem CBDs and a very long C-terminal stretch ending with a third CBD (fig. 3). PxCht3 appeared to be missing the CRI and have suffered an alteration of the typical CRII to VQGLE (fig. 4), which was similar toMsCht3 (Tetreau et al., Reference Tetreau, Cao, Chen, Muthukrishnan, Jiang, Blissard, Kanost and Wang2015). Notably, different patterns of expression were found in group IX and X chitinases between P. xylostella and M. sexta (fig. 5) (Tetreau et al., Reference Tetreau, Cao, Chen, Muthukrishnan, Jiang, Blissard, Kanost and Wang2015). MsCht1 was expressed exclusively in adult testes and ovaries, whereas PxCht1 was expressed in all stages, and MsCht3 was expressed in the fourth instar larvae, whereas PxCht3 was expressed exclusively in the pupae.
One PxIDGF sequence was reported previously in the NCBI (accession AB282642) and was also found in KONAGAbase (PXUG_V1_002709). All IDGFs were in group V chitinases (Zhu et al., Reference Zhu, Arakane, Banerjee, Beeman, Kramer and Muthukrishnan2008c ). The crystalline structure and the homology modeling suggest that all chitinase-like proteins have a (βα)8 triose-phosphate isomerase barrel structure (Varela et al., Reference Varela, Llera, Mariuzza and Tormo2002), and the group V chitinase-like proteins have an extra loop between the β−4 strand and the α−4 helix (Zhu et al., Reference Zhu, Deng, Vanka, Brown, Muthukrishnan and Kramer2004, Reference Zhu, Arakane, Banerjee, Beeman, Kramer and Muthukrishnan2008c ). The growth-promoting function was demonstrated on Drosophila IDGFs (Kawamura et al., Reference Kawamura, Shibata, Saget, Peel and Bryant1999). The IDGF proteins carry carbohydrate-binding activity but lack enzymatic activity (Zhu et al., Reference Zhu, Arakane, Beeman, Kramer and Muthukrishnan2008b ) and most likely act as chitolectins and bind to cell surface receptors or glycoproteins. Furthermore, in the current study, PxIDGF was expressed at all developmental stages and in all tissues (fig. 5), which was consistent with previous findings in B. mori (Pan et al., Reference Pan, Chen, Xia, Yao, Gao, Lu, Huojuan, He and Wang2010) and M. sexta (Tetreau et al., Reference Tetreau, Cao, Chen, Muthukrishnan, Jiang, Blissard, Kanost and Wang2015).
PxCht-h was identified in KONAGAbase (PXPG_V2_004242). Cht-h is a chitinase-like gene specific to lepidopteran insects (Zhang et al., Reference Zhang, Zhang, Arakane, Muthukrishnan, Kramer, Ma and Zhu2011a ). Because the deduced amino acid sequences showed extensive structural similarities with bacterial and baculoviral chitinases, Cht-h might be derived from a bacterial or baculoviral chitinase gene through horizontal gene transfer. Additionally, the Cht-h proteins exhibit exo-chitinase activities that act in concert with endo-chitinases. In an immunohistochemical study, the B. mori Cht-h was localized in the chitin-containing tissues during the molting stages (Daimon et al., Reference Daimon, Katsuma, Iwanaga, Kang and Shimada2005). Moreover, a lethal effect was observed at the pupal and adult stages after injection with Cht-h dsRNA into S. exigua larvae indicating that SeCht-h played an important role in the larval-pupal and pupal-adult stages (Zhang et al., Reference Zhang, Chen, Yao, Pan and Zhang2012). However, PxCht-h might be not involved with the pupal-adult stage because of low and no expression in the pupal and adult stages, respectively (fig. 5).
An endoβ-N-acetylglucosaminidase (ENGase) (EC 3.2.1.96), which belong to the glycosyl hydrolase family 85 (GH85), was also detected in our genome-wide search (table 1). The ENGases are included in the GH18 chitinase-like superfamily because they are phylogenetically closely related to the GH18 chitinases (Funkhouser & Aronson, Reference Funkhouser and Aronson2007). Moreover, both PxIDGF and PxCht-h were only detected in KONAGAbase but not in the genome search, and group VIII (Cht11) was not identified using either search approach. Tetreau et al. (Reference Tetreau, Cao, Chen, Muthukrishnan, Jiang, Blissard, Kanost and Wang2015) described the evolution of the 11 groups of chitinases in 14 species from seven different phylogenetic groups and suggested that group VIII chitinases were found in all insect species and even in Nematoda (Tetreau et al., Reference Tetreau, Cao, Chen, Muthukrishnan, Jiang, Blissard, Kanost and Wang2015). Therefore, the identification of PxCht11 is expected upon release of a new version of the DBM genome.
The insect chitinase families most likely evolved from a common ancestor, which was evident because the chitinases shared a high degree of conservation of amino acid sequences and domain organization in each group. Lepidopteran insects have a similar assortment of genes that encode chitinase and chitinase-like proteins but differ in the total number compared with the other insect orders. In conclusion, we identified 15 chitinase genes in P. xylostella, and we showed the phylogenetic relationships with other lepidopteran chitinases. Although we did not identify a group VIII chitinase, one new phylogenetic group of chitinases was found in addition to the 11 groups described previously. The different patterns of expression suggest that the DBMs carry numerous chitinase proteins to efficiently degrade the different types of chitin (α, β and γ forms) and the modified forms (deacetylated chitin) that occur in different extracellular structures and developmental stages (Merzendorfer & Zimoch, Reference Merzendorfer and Zimoch2003). Additional experiments are required to determine which of these chitinases are enzymatically active and to characterize their functions in the chitin metabolism of insects. In our laboratory, the expressions of PxCht5, PxCht7 and PxCht25 genes were performed in Escherichia coli, yeast and a baculovirus system for enzymatic characterization (data not shown). A better understanding of the chitinases in Lepidoptera might help to develop novel chitin-targeted strategies for pest control.
Supplementary Material
The supplementary material for this article can be found at http://dx.doi.org/10.1017/S0007485316000511.
Acknowledgements
We gratefully acknowledge the receipt of financial support from the Ministry of Science and Technology, Republic of China (MOST 104-2313-B-002-018 & MOST 104-2313-B-002-071). We thank the Taiwan Agricultural Chemicals and Toxic Substances Research Institute for providing the diamondback moths. The authors thank Dr. Jianzhen Zhang for helping with the phylogenetic analyses.