Identification and analysis of Toll-related genes in the domesticated silkworm, Bombyx mori

ARTICLE IN PRESS Developmental and Comparative Immunology (2008) 32, 464–475

Available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/devcompimm

Identification and analysis of Toll-related genes in the domesticated silkworm, Bombyx mori Ting-Cai Cheng, Yu-Li Zhang, Chun Liu, Ping-Zhen Xu, Zhi-hong Gao, Qing-You Xia, Zhong-Huai Xiang The Key Sericultural Laboratory of Agricultural Ministry, College of Life Science, Southwest University, Chongqing 400716, China Received 7 February 2007; received in revised form 15 March 2007; accepted 19 March 2007 Available online 20 April 2007

KEYWORDS Toll-related genes; Expression; Microarray; Innate immunity; Bombyx mori

Abstract Silkworm (Bombyx mori), a model system for Lepidoptera, has contributed enormously to the study of insect immunology especially in humoral immunity. But little is known about the molecular mechanism of immune response in the silkworm. Toll receptors are a group of evolutionarily ancient proteins, which play a crucial role in the innate immunity of both insects and vertebrates. In human, Toll-like receptors (TLRs) are the typical pattern recognition receptors for different kinds of pathogen molecules. Toll-related receptors in Drosophila, however, were thought to function as cytokine receptors in immune response and embryogenesis. We have identified 11 putative Toll-related receptors and two Toll analogs in the silkworm genome. Phylogenetic analysis of insect Toll family and human TLRs showed that BmTolls is grouped with Drosophila Tolls and Anopheles Tolls. These putative proteins are typical transmembrane receptors flanked by the extracellular leucine-rich repeat (LRR) domain and the cytoplasmic TIR domain. Structural prediction of the TIR domain alignment found five stranded sheets and five helices, which are alternatingly joined. Microarray data indicated that BmToll and BmToll-2 were expressed with remarkable enrichment in the ovary, suggesting that they might play a role in the embryogenesis. However, the enriched expression of BmToll-2 and -4 in the midgut suggested that the proteins they encode may be involved in immune defense. Testisspecific expression of BmToll-10 and -11 and BmToLK-2 implies that these may be involved in sex-specific biological functions. The RT-PCR results indicated that 10 genes were induced or suppressed with different degrees after their immune system was challenged by different invaders. Expression profiles of BmTolls and BmToLKs reported here provide insight into their role in innate immunity and development. & 2007 Elsevier Ltd. All rights reserved.

Corresponding author. Tel.: +86 23 68250099;

fax: +86 23 68251128. E-mail address: [email protected] (Q.-Y. Xia). 0145-305X/$ - see front matter & 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.dci.2007.03.010

ARTICLE IN PRESS Identification and analysis of Toll-related genes in the domesticated silkworm, Bombyx mori

1. Introduction Innate immunity is the first line of defense by animals against invading microorganisms such as bacteria, fungi and viruses. Despite the lack of an adaptive immune system, invertebrates are able to rapidly recognize non-self antigens and amplify the infected signal. In the process, signal transduction pathways such as the Toll pathway and the Imd pathway are triggered and the transcription of downstream genes are activated afterwards. Effector molecules, especially antimicrobial peptides, will be highly expressed in fat body and secreted into the hemolymph. In this process, the pattern recognition receptors known as Toll-like receptors (TLRs) play a key role in the induction of innate immunity and in the inflammatory responses. Since Medzhitov and his colleagues successfully cloned a human homolog of Drosophila Toll [1], at least 10 TLRs (TLR1-10) have been found in the human genome. In contrast to Drosophila Toll proteins, the mammalian TLRs as pattern-recognition receptors (PRRs) are able to recognize directly their specific pathogen-associated molecular patterns (PAMPs) in the innate and adaptive immune responses [2]. So far, we know that: (i) TLR-2 is primarily involved in recognition of peptidoglycan and bacterial lipopeptides [3], (ii) TLR-3 is a cell-surface receptor for double-stranded RNA [4], (iii) the recognition of lipopolysaccharide (LPS) by TLR-4 is mediated by a complex (CD14, MD2, and TLR-4) [5], (iv) TLR-5 recognizes the flagellin that forms bacterial flagella [6], (v) TLR-6 cooperates with TLR2 for the recognition of mycoplasmal lipopeptides [7], (vi) TLR-9 functions as a receptor for unmethylated CpG motifs that are abundant in bacterial genome [8]. Indeed, the Drosophila genome encodes a family of nine Toll-related receptors. Most Tolls are highly expressed in normal embryos and pupae and probably have important developmental functions [9]. The Drosophila Toll functions not only in the establishment of dorsal–ventral polarity in the early embryo, but in the innate humoral and cellular immune response of larvae and adults [10]. Drosophila Toll-6–8 are expressed at high levels in embryos and pupae, suggesting that the proteins they encode function in embryogenesis and molting stages [9]. Drosophila Toll-9 similar to gain-of-function Drosophila Toll-1 activates strongly the expression of Drosomycin and antifungal genes by similar signaling components to Drosophila Toll-1. Drosophila Toll-1 has been implicated in the regulation of immune responses [11]. During Drosophila’s immune response, the broad-spectrum microbial recognition pattern is reflected by PRRs such as peptidoglycan recognition proteins (PGRPs) and Gram-negative bacteria-binding proteins (GNBPs). Toll, a cytokine receptor, is activated by an endogenous ligand, Spa ¨tzle [12]. In human, however, TLRs directly interact with PAMPs acted as typical PRRs. Toll receptors are part of an ancient receptor family involved in immune defense, conserved from lower metazoans to higher vertebrates. A typical TLR contains generally extracellular leucine-rich repeats (LRRs) connected to a cysteine-rich domain and an intracytoplasmic Toll-interleukin homolog domain (TIR) [13,14]. Toll-related proteins have been predicted and identified from several genomes of model species in addition to human, including Drosophila melanogaster [9], Anopheles gambiae [15], and Danio rerio [16], etc. In Bombyx mori, a BmToll has also been cloned and

465

characterized [17]. It is interesting to analyze phylogenetic relationships of Toll-related receptors in insects. Recent completion of the genome sequencing of B. mori provides a unique opportunity to analyze Toll-related genes [18,19]. Due to its broad experimental potential as molecular and genetic tools, B. mori is an important model system for studying insect innate immunity and development biology. The silkworm has contributed enormously to the study of insect pathology and insect immunology, particularly humoral immunity. In this paper, we have (1) identified B. mori Toll-related genes and their TIR domains from the silkworm genome using bioinformatics, (2) compared the evolutionary relationship of TIR domains with those found in dipteran insects, and (3) analyzed their temporal and spatial expression patterns. The study on Tollrelated genes may help us better understand the immuneresponse mechanism in silkworm.

2. Materials and methods 2.1. Identification of Toll-related genes in the silkworm genome Complete Toll-related protein sequences of Drosophila retrieved from GenBank (http://www.ncbi.nih.gov/Genbank/) were used as queries to search for Toll-related genes in the 9x coverage silkworm genome map assembled by the China and Japan silkworm genome projects. The following protein sequences and accession numbers were used as queries: D. melenogaster: DmToll (AAA28941), Dm18w (AAF57509), DmToll-3 (AAF54021), DmToll-4 (AAF52747), DmToll-5 (AAF53306), DmToll-6 (AAF49645), DmToll-7 (AAF57514), DmTollo (AAF49650), DmToll-9 (AAF51581); A. gambiae: AgToll (EAA45376), AgToll-1 (EAA07066), AgToll-5 (EAA04891), AgToll-6 (EAA00379), AgToll-7 (EAA00348), AgTrex (AAL37904), AgToll-9 (EAA04650), AgToll-10 (EAA05150), AgToll-11 (EAA05071). A PSI-BLAST search of the silkworm genome database was performed using the TIR conserved domain from BmToll (BAB85498) as a query. To predict the exon/ intron boundaries, the assembled genomic sequences were then analyzed using FGENESH program (http://sun1.softberry.com/berry.phtml). The results were further manually annotated by comparing with insect Toll-related receptors using BLASTX (http://www.ncbi.nlm.nih.gov/BLAST). In addition, to test whether or not these BmTolls are conceptual translations of pseudogenes, a BLAST search of the all-available ESTs was performed and the matched sequences can be retrieved from NCBI or the web (http://papilio.ab.a.u-tokyo.ac.jp/Bombyx_EST/). The extracellular, transmembrane, and cytoplasmic domains of protein sequences were predicted using SMART (http://smart.embl.de/) [20]. The secondary and tertiary structures were predicted by the neural network program PHD [21] and by comparative modeling with SWISS-MODEL [22], respectively.

2.2. Alignment and phylogenetic analysis of BmTolls Multiple sequence alignments of TIR domains were performed using Clustal X [23]. In order to compare equivalent regions, the TIR domains were retrieved using their

ARTICLE IN PRESS 466 secondary structure prediction from SMART. We used additional sequences for phylogenetic tree from GenBank with the following accession numbers: Homo sapiens: HsTLR-1 (AAC34137), HsTLR-2 (AAC34133), HsTLR-3 (AAC34134), HsTLR-4 (AAC80227), HsTLR-5 (AAC34136), HsTLR-6 (BAA78631), HsTLR-7 (AAF60188), HsTLR-8 (NP_057694), HsTLR-9 (Q9NR96), HsTLR-10 (XP_223422); Caenorhabditis elegans: CeTLR-1 (AAK37544). Phylogenetic trees were reconstructed using MEGA v3.0 with 1000-times bootstrap sampling [24].

2.3. Insect maintenance and immune challenge For infection experiments, we used the silkworm larvae of Dazao strain, day 3 of fifth instar. The silkworms were reared on the artificial diet (Nihonnosanko) at 25 1C. The larvae were injected with 10 ul/head of Escherichia coli (optical density at 600 nm or OD of 0.8) and Staphylococcus aureus (OD ¼ 0.8), respectively. Fat bodies were collected from fifth instar larvae (day 3) at different times after injection (1, 2, 4, 10, and 24 h). The larvae injected with distilled water were used as a negative control. Natural infection was initiated by rearing the silkworm larvae on the artificial diet containing Beauveria bassiana and incubated at 25 1C for specific times (2, 10, 24, and 48 h).

2.4. RNA extraction Each tissue sample from four larvae was stored in liquid nitrogen before pulverizing. Total RNA was isolated using Trizol Reagent (Invitrogen) according to the manufacturer’s instructions. The contaminating genomic DNA was digested with Rnase-free Dnase I (Promega) for 15 min at 37 1C. The RNA samples were eluted with Rnase-free water and stored at 80 1C.

2.5. Expression-profile analysis of BmTolls by oligonucleotide microarray The genome-wide 69-mer oligonucleotide microarray for the silkworm, constructed in the CapitalBio Corporation (Beijing, China), contains approximately 23,000 known and predicted genes based on the silkworm Whole Genome Shotgun and other silkworm sequences from Genbank. The probes for BmTolls are shown in Table 1. Double-stranded cDNAs, synthesized from 5 mg total RNA isolated from different tissues and developmental stages of B. mori, were labeled with a fluorescent dye (Cy5 or Cy3-dCTP; Amersham Pharmacia Biotech) and used to hybridize with the array at 42 1C overnight after being denatured at 95 1C for 3 min. For each sample, two hybridizations were performed by using a reverse fluorescence strategy. All microarrays were scanned with a ScanArray Express scanner using ScanArray 2.0 software (Packard Bioscience). Images were analyzed with GenePix Pro 4.0 (Axon Instruments). The global intensitybased LOWESS program was applied to normalize data [25]. Hierarchical clustering of the data was performed using the program Cluster (http://rana.stanford.edu), and the cluster data were visualized using the program TreeView (http:// rana.stanfordedu/software/).

T.-C. Cheng et al.

2.6. Reverse transcriptase polymerase chain reaction (RT-PCR) The predicted genes were used to design primers by Primer Premier 5.0 (Table 1). Primers (forward: AACACCCCGTCCTGCTCACTG; reverse: GGGCGAGACGTGTGATTTCCT) were used for amplifying the silkworm actin-3 as an internal control. The cDNA fragments were amplified in 25 ml reaction volume containing normalized cDNA, 1.5 mM MgCl2, 50 mM KCl, 10 mM Tris-HCl (pH 8.3), 0.25 mM each of dNTPs, and 2.5 units of Taq DNA polymerase (Promega). The cycle numbers were empirically chosen to produce comparable band intensities while avoiding saturation. The RT-PCR experiment was repeated three times. PCR products were separated by gel electrophoresis, recovered from the gel, and directly sequenced using an ABI3100 automated sequencer.

3. Results 3.1. A family of Toll-related genes in the silkworm We have searched the 9x coverage silkworm genome for homologs of Toll-related genes. The protein sequences for fruitfly and mosquito Toll-related genes were used as queries with the TBLASTN program. Because the TIR domain is the most conserved region between Toll-related receptors in insects and TLRs in mammalians, we have also searched for the gene sequences encoding the TIR domain in the silkworm genome by using the PSI-BLAST program. Both searches revealed the existence of 14 genes containing the conserved TIR domain, of which one gene named BmToll has also been cloned and characterized [17]. The exon prediction was performed using FGENESH program and further manually annotated. Like reported Toll receptors, each of the deduced protein sequences of eleven genes contains a transmembrane region flanked by LRR repeats in the N-terminal region and the TIR domain at the C-terminal end, which strongly suggests that these genes are the Toll-related genes of B. mori. These genes were designated as BmToll-1 to -9 because of their similarity to Drosophila counterparts and BmToll-10, -11 are homologous to Anopheles Toll-10. Another two genes with very low similarity to the key TIR domain were labeled as Toll analogs, designated as BmToLK-1 and -2. A gene containing a conversed TIR domain is homologous to Drosophila MyD88; it also contains a death domain in the N-terminal end. The Tollrelated proteins of the silkworm share higher similarity with other insect Tolls than with any mammalian TLRs, suggesting that these two groups of proteins have evolved independently [26]. Based on the 9x genome map, eight genes were found to cluster with very wide gaps for each other in a big scaffold (Fig. 1), which was located on chromosome 23 by means of the markers of simple sequence repeat and single nucleotide polymorphism. Similarly, BmToll-3 and -4 were located on chromosome 14. [27,28]. Compared with fruitfly and mosquito Tolls in amino acid sequences, BmTolls show higher similarities with Drosophila Tolls, except for BmToll-10 and -11 with Anopheles Toll-10 (Table 2). The sequence similarity in the TIR domain ranges from 59% to 80%, whereas the similarity in the extracellular domain ranged from 43% to 76%. Interestingly, the

The PCR primer sequences used in the tissue expression profiling Primer

Sequences

Oligo probes

BmToll

Forward Reverse

TTCCCATGATGGTTTCAACTC TATATACCTGAGACGCTCCCAG

TCAGATCACATATATTCCTCGATAGATTCTGACTATTCATCCGTGGAACATGGTATGGCCCCGGGCAGG

BmToll-2

Forward Reverse

TGCCTATGATGGTAACGACTC TTTCAAGTACGGCCTTAAATC

TTCAGGCATTGGCCTCCGCCGCCACCTTTAATTGACACTCAAAGCTCAGGTCAAGCTTATCTTGTCTAG

BmToll-3

Forward Reverse

CGTGGCTGCATCACTCTGTTT GTCGATCCTGGCGTCGTTAGT

GCGGCCTCGGTGCAAGCGGCGCGCCGCACCCTCATCGTAGTCTCGCGGCACTTCCTGCGCTCCAAGTGG

BmToll-4

Forward Reverse

TGGCTTTGTTGGCAGTGATG CTTTAGTTTGTGCCAGAACCAGG

TGGCGCTTAATGCTCGCCCACAACCACATCCAGACCGTTCGGCTTCAGGATCTGCCCGACACGATATTG

BmToll-5

Forward Reverse

GCAGTGATGGTCGGAGTTAC GTTCCAGAGCAGGTAGGTGTT

GAAGACGATCCACCGATTGGACTGCGGCAAGATATGCCACTAGAGTTCTTTAGTTTGCCTGATACACCA

BmToll-6

Forward Reverse

CGTTACTGTTGATAGCCCTGTG AACTGTCTATCACCCATTCG

ATGCCGGACTTGCGGAAGTGTCAGTATCACCGTTCCACAGTGAATATTTACGCTTCGGTGTCACCAGTG

BmToll-7

Forward Reverse

ACAACCTTCCGTTATTGGCGT CCGTTTCTCCTTCCATGTTATCTC

GAAAGGTTACGTTATGCGATGCCGTCTTCAAAGCGGCACGGGCATAAGTTAAAAATACTTAATTACGGA

BmToll-8

Forward Reverse

CATACCTCTTCTTATCGCAACAC GTGCAAACTTCAACTTCTCCC

CCGAATTCTGGTGGCGGCATGCCCATGCACCGACATCATCATCCGAGGAACCACTTGGGCATGTTGCCA

BmToll-9

Forward Reverse

CGTTGCGATGCCTGATG CACCATTGGGATTTAGCA

AAAAACCCACCGCGCGACTTGATGGAAATAATAGCTGAAACCAGTGTGAGAAGATTAATTTTATCACAT

BmToll-10

Forward Reverse

CTGTAGATGGAAACAACTGG GCACTCACTAAACTATACCC

GACCTGCAGCTTATAGAGAGACGTGCGGGTGATAGTTTAGTGAGTGCCGCGGAAAGTTCTAAGAGACTG

BmToll-11

Forward

ATCAGTACGATGAAGTAAGCA

CACTTTTGCTGGCCTTATTCGCTTAGTCATTTTAAATCTTTCGTATAACTTGCTTGCACGAATCGATCC

ARTICLE IN PRESS

BmTolls

Identification and analysis of Toll-related genes in the domesticated silkworm, Bombyx mori

Table 1

467

ARTICLE IN PRESS 468

T.-C. Cheng et al. BmToll-2 LG23

445kb

BmToll

352kb

BmToll-7

168kb

BmToll-10

430kb

BmToLK-1 BmToll-8 BmToll-11 BmToll-6 260kb 232kb 388kb T013C09

T093F02

S2319 T012H19

T032I05

Fig. 1 The genome structure of the Toll multigene family. The black boxes indicate the members of the multigene family; the broken lines with numbers indicate the numbers of necleotides between the members of the multigene family; the black arrows indicate the direction of transcription. The markers of simple sequence repeat and single-nucleotide polymorphism are indicated by the blank triangle and the blank arrows, respectively.

Table 2 Genomic DNA scaffolds containing BmTolls, predicted protein size and position of TIR domain, and protein homologs of BmTolls compared to DmTolls Predicted protein

TIR

Extracellular

BmToll

Scaf

Exon

Size

TIR position

E-value

Homolog

Ide, Sim (%)

Ide, Sim (%)

BmToll BmToll-2 BmToll-3 BmToll-4 BmToll-5 BmToll-6 BmToll-7 BmToll-8 BmToll-9 BmToll-10 BmToll-11 BmToLK-1 BmToLK-2

nscaf3015 nscaf3015 scaffold416 scaffold416 nscaf2987 nscaf3015 nscaf3015 nscaf3015 nscaf2927 nscaf3015 nscaf3015 nscaf3015 nscaf2851

1 1 1 7 8 3 1 4 2 4 1 3 5

1295 1283 299 922 510 1291 1171 1249 784 1201 1251 1150 796

1024–1166 1023–1165 153–288 750–885 322–460 1062–1199 1007–1145 1042–1177 640–783 1070–1198 967–1109 948–1067 611–745

7.00E-18 6.43E-17 1.95E-25 1.08E-23 3.86E-29 2.95E-34 1.11E-20 5.63E-31 8.36E-16 5.65E-02 7.72E-05 1.28E+02 1.29E+00

Dm18w Dm18w DmToll-5 DmToll-5 DmToll-5 DmToll-6 DmToll-7 DmTollo DmToll-9 AgToll-10 AgToll-10 DmToll AgToll-1

47, 49, 47, 48, 49, 58, 43, 65, 37, 42, 40, 32, 32,

55, 54, 33, 33, 39, 60, 52, 51, 27, 54, 55, 40, 31,

similarities of TIR domains of BmToll and BmToll-2, -7, -10, and -11 are obviously lower than those of extracellular LRR domains to their homologs. On the contrary, the similarities of TIR domains of BmToll-3 to -6, -8, and -9 are higher than those of extracellular LRR domains to their homologs. Searching the ESTs database, we found several EST sequences corresponding to BmToll and BmToll-4, and -10 in the wing disc of larvae. Each of BmToll-2 and -6 only has one tag in the dispause eggs. BmToll-11 has one EST in the ovary and two ESTs in dispause eggs, respectively. In the holometabolous insects, the wing disc is a typical developmental tissue like the embryo. This result suggested that these BmToll proteins, like Drosophila 18w [29], might have a function in the development of silkworm. Interestingly, BmToll-5 has an EST sequence in BmN culture cell, indicating that it is not a pseudogene. BmToLK-1 has one EST in testis, whereas BmToLK-2 has one EST in brain tissue. No corresponding tag sequence for the rest of the BmTolls was found in the EST database.

3.2. Analysis of the conserved domain of BmTolls The Toll family is a typical transmembrane receptor with extracellular LRR domain and a cytoplasmic TIR domain. The cytoplasmic domain of 13 BmTolls contains a TIR domain, a motif of about 140 residues, which is essential for Toll signaling (Fig. 2(A)). With respect to the TIR domains, BmToll-3 and

65 63 70 70 69 79 61 80 59 68 67 51 52

72 70 54 48 62 76 70 66 43 73 72 61 47

BmToll-4 have 96% similarity; BmToll and BmToll-2 have 82% similarity, as is also the case for BmToll and BmToll-7. BmToll-10 is closely related to BmToll-11. On the other hand, the extracellular LRR domains, containing the characteristic consensus sequence L(X2)LXL(X2)NXL(X2)L(X7)L(X2), are responsible for the high-affinity binding of peptides [30]. The BmTolls also share high similarity in the extracellular region. To determine the conserved regions of the TIR domain of the silkworm Tolls, we aligned these domains using Clustal X (Fig. 2(B)). Three primary domains are conserved in the silkworm as described previously [17]. The core TIR domain starts from the conserved F/YDA motif, which is very similar to that in the mammalian. The alanine residue of this motif is also the start point of a b-sheet as described in human TLRs [14]. The second domain is the conserved CLHYRD motif, LH of which probably is a border between a b-sheet and a loop as human TLR2. At the C-terminal of the core TIR domain, the last domain is the conserved FW motif. But the functions of this domain remain to be clarified. Another three hydrophobic regions are defined. On the basis of the neural network program PHD, the consensus secondary structure of BmTolls encompasses five stranded sheets (labeled 1-5) and five helices (labeled 1-5), which are alternatingly joined similar to that of human TIRs [13]. The predicted tertiary structure is similar to the crystal structure of human TIR domains, which contains a large conserved surface patch crucial for receptor signaling.

ARTICLE IN PRESS Identification and analysis of Toll-related genes in the domesticated silkworm, Bombyx mori

469

Fig. 2 The silkworm Toll family. (A) Schematic representation of the domain organization of the BmToll receptors. All domains were predicted using the SMART program. In the intracytoplasmic region, the TIR domain is indicated with a long hexagon, whereas the LRR are shown as small rectangles and characteristic cysteine-rich carboxy-flanking and amino-flanking motifs by triangles. (B) Multiple-sequence alignment of TIR domains from BmToll receptors. TIR domains of deduced protein sequences were defined by SMART program (http://smart.embl.de/) and aligned by Clustal X.

3.3. Evolutionary relationship of BmTolls The conserved cytoplasmatic domains of Toll-related genes have the most reliable determination of phylogeny than the extracellular domains exposed to high mutational pressures over long periods of time. This is the main reason we used

TIR domains to reconstruct phylogenetic trees with the neighbor-joining method (Fig. 3). The results revealed that insect Toll-related genes formed seven major clusters. Toll-1, -2, and -7 have a common origin in insect. BmToll-7 diverged before the separation of BmToll and BmToll-2. BmToll-3 and -4 diverged after the split of Lepidopteron and

ARTICLE IN PRESS 470

T.-C. Cheng et al. 99

BmToll

91

BmToll-2

99

BmToll-7 Toll-1, 2, 7

Dm18w 69

DmToll-7

100

AgToll-7

99

BmToll-10 BmToll-11 85

Toll-10, 11

AgToll-10

79

AgToll-11

99 89

BmToLK-1 ToLK

BmToLK-2 CeTLR-1 100

BmToll-3 BmToll-4 BmToll-5

92

DmToll-5 DmToll

70

AgToll

Toll-3, 4, 5

AgToll-5

99

AgToll-1

100

77

DmToll-3 DmToll-4

100 BmToll-6

98

DmToll-6

Toll-6

AgToll-6

97 98

BmToll-8 DmTollo

98

Toll-8

AgTrex

96

HsTLR-9

98

HsTLR-8 65

HsTLR-7

100

HsTLR-5 HsTLR-3

58 100

AgToll-9

86

DmToll-9

Toll-9

BmToll-9 HsTLR-4 HsTLR-2 100 97 0.1

HsTLR-6 HsTLR-1

100

HsTLR-10

Fig. 3 A phylogenetic tree of TLR and Toll protein sequences from the related species. The phylogenetic tree was reconstructed based on the conserved TIR domains found with MEGA v3.0. Alignment gaps were not excluded. Bm is an abbreviation for B. mori, Dm for D. melanogaster, Ag for A. gambiae, Hs for H. sapiens, Ce for C. elegans. Accession numbers of proteins and method used to identify TIR domains are detailed in the Methods and Materials section.

Dipteran, similar to Toll-3 and -4 in Drosophila and Toll-1 and -5 in A. gambiae. BmToll-5 clusters closely with Drosophila Toll and Toll-5. The group of Toll-6 and -8 seems

to have a common origin. BmToll-6 and -8 are homologs of Drosophila Toll-6 and Tollo, respectively. BmToll-10 and -11, which seem to have separated from a common ancestor of

ARTICLE IN PRESS Identification and analysis of Toll-related genes in the domesticated silkworm, Bombyx mori mosquito Toll-10 and -11, are located on branches more closely to the group of Toll-1, -2, and -7. The group of Toll-9 including BmToll-9 is located on branches more distinct from other groups of insect Tolls as reported [26]. It is suggested that the Toll-9 shares structural and functional similarities with mammalian TLRs. BmToLK-1 and -2 are clustered together and have probably close relationship with the TLR of C. elegans.

3.4. Spatial and temporal expression profiles of BmTolls To investigate the expression pattern of BmTolls, we picked the signals of these genes from the genome-wide microarray data from nine silkworm tissues, including the ovary, testis, head, epiderm, midgut, malpighian tubule, fat body, hemocyte, and silk gland. The results indicated that BmTolls represent tissue-specific expression (Fig. 4(A)). BmToll-2 was widely expressed, with enrichment in the ovary and the midgut and ordinary expression in the testis, epiderm, and head. In contrast to BmToll-2, BmToll and BmToll-4 were highly expressed in the ovary and the midgut, respectively. BmToll was not expressed in the midgut, whereas BmToll-4 was not expressed in sex glands. Interestingly, BmToll-10, and -11 and BmToLK-2 appeared to be testis-specific expression. BmToll-9 was specifically expressed in the malpighian tubule. In addition, no expression of these genes was detected in the main immune tissues, such as fat body and hemocyte. One possibility for this is that expression in these tissues was below the noise level. No significant levels of BmToll-3, -5, -6, and -8 and BmToLK-1 were also detected in tissue-expression experiments and in developmentalstage experiments, with the exception of BmToll-6, which was expressed in the wandering and earlier pupae. Four main stages of the silkworm life cycle, including the fifth-instar larvae, wandering, pupae, and adults, were used to explore the developmental expression profiles. Each sample was from a single individual with sex division. The stage- and sex-specific expression of BmTolls is shown in Fig. 4(B). Both BmToll and BmToll-2 were expressed ubiquitously from the fifth-instar larvae to adults, but the average expression value of the former was about twice of that of the latter. BmToll-6 was weakly expressed from the first day of wanders to the second day of pupae, and not expressed in fifth-instar larvae, latter-day pupae, and adults. BmToll-4 and -10 were scarcely expressed in wanders and pupae. Interestingly, BmToLK-2 had sex-specific expression from the beginning of wandering to adults.

3.5. Expression profile of BmTolls in immunized tissues To assess the time course of BmTolls expression after infection, we performed the semi-quantitative RT-PCR to analyze total RNA samples of fat body at different times after S. aureus (a Gram-positive bacteria), E. coli (a Gramnegative bacteria), and B. bassiana (a fungi) infected the silkworm larvae. We only tested BmTolls that have clear homologs in insects, so the BmToLKs were excluded from this experiment. The results indicated that the expression profiles of BmTolls were obviously distinct after being

471

induced by different pathogens (Fig. 5). For instance, BmToll, reported to suppress remarkably after LPS challenge, was significantly suppressed after being injected by S. aureus and E. coli. By contrast, BmToll-2 was obviously down-regulated after E. coli and B. bassiana. BmToll-3 was acutely up-regulated from 1 to 10 h after being induced by S. aureus and E. coli, whereas the peak levels of BmToll-3 appeared at 48 h after being induced by B. bassiana. BmToll-4 was strongly suppressed by S. aureus and induced by E. coli and B. bassiana. Contrary to this, BmToll-10 and -11 were induced by S. aureus and suppressed by E. coli and B. bassiana. BmToll-9 was remarkably induced by B. bassiana and S. aureus; however, it was rare after injection of E. coli even when increasing the PCR cycles. These results suggest that there might be differential immune responses to different pathogens. Almost all BmTolls show remarkable basal expression by control RT-PCR. However, no EST or microarray expression of these genes was found in fat body, suggesting that they might be expressed at low abundance.

4. Discussion In this study, we described eleven Tolls and two ToLKs from the silkworm B. mori genome. This number is different from that in D. melanogaster genome (nine Toll-related genes) and that in A. gambiae genome (10 Toll-related genes) [31]. Like all reported Toll-related proteins, they contain the characteristic LRR motifs, a transmembrane region, and the TIR domain. We designated these genes homologs to D. melanogaster and A. gambiae as BmToll-2 to -11. The data from the genome map and the phylogenetic tree analysis indicate that this gene family was brought by a small number of gene duplication events. It is likely that the BmToLK-1 has a common ancestor with seven other genes on its chromosome. This was inferred from the close phylogenetic relationship with CeTLR-1. The group of BmToll-1, -2, and -7 was formed by two duplications after the separation of Lepidopteron and Dipteran, whereas the duplication event of BmToll-6 and -8 occurred before the separation of Lepidopteron and Dipteran. Additionally, BmToll-3 and -4, located on chromosome 14, were produced by an event of gene duplication after the separation of Lepidopteron and Dipteran. Expression information of BmTolls was obtained from genome-wide oligonucleotide microarray, ESTs, and semiquantitative RT-PCR after immune challenge, offering us a chance to learn more about their functions in innate immunity and in development. Based on the microarray data, the high level of expression of BmToll-2 and -4 in the midgut suggested that the proteins they encode may be involved in immune defense, similar to Anopheles Toll-9 highly expressed in the adult gut [15]. BmToll and BmToll-2 were abundant in the ovary. BmToll-6 and -7 were expressed in the pupal stage, implying that they may play a role in the embryogenesis as adhesion molecules like Drosophila Toll and 18w [29]. Sex-specific expression of BmToLK-2 in the tissue- and developmental-expression profiles and testisspecific expression of BmToll-10 and -11 in tissue-expression profile implied that they may be involved in sex-specific biological function. In addition, no expressions of these

ARTICLE IN PRESS 472

T.-C. Cheng et al.

F3d

Ovary

W0dF

Testis

W0dM

Malpighian

W0.5dF

Epiderm

W0.5dM

Head

W1dF

Hemocyte

W1dM

Fatbody

W1.5dF

Midgut

W1.5dM

PSG

W2dF

MSG BmToll-2

BmToll-4

BmToll

BmToll-7

BmToLK-2

BmToll-11

W2.5dM

BmToll-10

W2.5dF

BmToll-9

W2dM

W3dF W3dM

-2

0

2

P0dF P0dM P1dF P1dM P2dF P2dM P3dF P3dM P4dF P4dM P5dF P5dM A0dF A0dM BmToll-4

BmToll-2

BmToll

BmToll-7

BmToll-6

BmToLK-2

BmToll-10

Fig. 4 Temporal and spatial expression profiles of BmTolls and BmToLK-2. (A). The developmental expression profiles of the genes were analyzed in four main stages including the fifth-instar larvae (F), wandering (W), pupae (P), and adults (A; F, female; M, male). For each sample, two hybridizations were performed using a reverse fluorescence strategy. (B). The tissue-specific expression profiles of the genes were detected in silkworm tissues. Four hybridizations were performed for each tissue. Hierarchical clustering was performed using the average signal intensity in Supplementary Table 2. Red represents highly expressed genes whereas green shows those with low or scarce expression levels. Gray indicates the genes with missing expression data.

genes were detected by genome-wide microarray in the immune tissues of naive silkworm, such as fat body and hemocyte. The reason may be that these genes have very weak basal-expression levels in fat body and hemocyte and was considered to be below the noise level.

In Drosophila, Toll protein activates the Dif factor during immune response. Toll and the genes encoding components of Toll pathway are obviously up-regulated after bacterial infection [32]. In A. gambiae, Toll-1 and Toll-9 were weakly induced when infected by E. coli [15]. In B. mori, however,

ARTICLE IN PRESS Identification and analysis of Toll-related genes in the domesticated silkworm, Bombyx mori B.bassiana (30 cycle) C

2

10

24 48h

S.aureus (30 cycle) C

1

2

4

10 24h

473

E.coli (35 cycle) C

1

2

4

10 24h

BmToll BmToll-2 BmToll-3 BmToll-4 BmToll-6 BmToll-7 BmToll-8 BmToll-9 BmToll-10 BmToll-11 Actin-3

Fig. 5 Time courses of BmTolls expression after infection. Total RNA samples were isolated from fat bodies at indicated times (hr) after immune challenge. The control (C) was fat body RNA isolated from the silkworm injected with distilled water. RT-PCR products separated by 1.2% agarose gel electrophoresis were cloned and verified by sequencing. Actin-3 was amplified for 25 cycles as internal control.

Imamura and Yamakawa reported one Toll gene of the silkworm, which was expressed in the fat body and suppressed strongly after LPS challenge. By RT-PCR analysis, BmToll was also strongly suppressed after the E. coli and S. aureus challenges, similar to results reported by Imamura and Yamakawa [17]. BmTolls were induced or suppressed strongly, suggesting that they may be involved in immune response to different pathogens. In addition, genome-wide expression showed that none of the silkworm Toll-related genes were significantly expressed in response to fungi attack. This was probably because of their low amounts of transcripts (data not shown). The silkworm is one of the most important models that contribute to the research on insect immunity. With the exception of inducible antimicrobial peptides [33], many immune-related proteins involved in pattern recognition, signal transduction, and transcription regulation have been characterized by immune challenge. The PGRP-S was discovered in the silkworm as the protein that recognizes PGN and Gram-positive bacteria and initiates activation of the prophenoloxidase cascade [34]. The GNBP of B. mori is constitutively expressed in the fat body and has strong affinity for the Gram-negative bacteria [35], whereas the b-1,3-glucan recognition protein (bGRP) has strong specific affinity to b-1,3-glucan, a component of the fungal cell wall [36]. One transcription factor, belonging to the Rel family, was identified to be involved in the activation of antimicrobial peptide genes in B. mori [37]. Interestingly, the BmRelA and BmRelB strongly activated the Lebocin 4 and Attacin in vitro, respectively. The two proteins are structurally identical, with the exception of the N-terminal 52 amino acid region. This probably has a dual role as an activator and a repressor for regulating antimicrobial peptide genes; it may enhance or suppress the function of the proline-rich domain. Innate immunity is an evolutionarily conserved system and is present in all organisms [38]. In Lepidoptera, immunity-related proteins in hemocytes, such as antibacterial peptides and pattern recognition proteins, have been

a focus in research to understand immune mechanism. Immune signaling pathways connecting recognition with expressing effectors have been primarily outlined in Drosophila [39]. The two signaling pathways, Toll and Imd, activate responses to infection to various classes of microorganisms and induce the production of effector molecules, among which antimicrobial peptides are prominent [40]. An endogenous protein, Spa ¨tzle, activates Toll protein. Toll binding to a ligand activates a cytoplasmic cascade involving Drosophila MyD88 and the adaptor protein Tube. By comparative analysis, the homologs of Drosophila Spa ¨tzle and MyD88 were found in the silkworm genome. The cDNA of B. mori Spa ¨tzle has been cloned and overexpressed in E. coli. After the silkworm larvae were injected with renatured BmSpz1, mRNA levels of antimicrobial peptide genes increased remarkably [41]. A MyD88 of B. mori similar to Drosophila MyD88 contains conserved death domain and TIR domain. Therefore, it is speculated that the Toll-related proteins of the silkworm function probably as cytokine receptors activated by certain endogenous ligands in immune response rather than as pattern recognition proteins. Our comparative genome analysis also indicates that the genes involved in the Imd pathway, such as the homologs of Drosophila IMD, FADD, and DREDD, did exist in the silkworm genome (data not shown). On the basis of the above observations, it is speculated that the two pathways are present in the silkworm and probably control the expression of antimicrobial peptide genes in the innate immune response.

Acknowledgments This work was supported by grants from the National Basic Research Program of China (No. 2005CB121000), the National Key Scientific and Technological Project (No.2005BA711A07), and the National Natural Science Foundation (No. 30470991). The microarray hybridizations were done in

ARTICLE IN PRESS 474 the CapitalBio Corporation (Beijing, China). We are grateful to Jesse Frumkin (Keck Graduate Institute) for his critical reading of the manuscript.

Appendix A. Supplementary materials The online version of this article contains additional supplementary data. Please visit doi:10.1016/j.dci.2007.03.010.

References [1] Medzhitov R, Preston-Hurlburt P, Janeway Jr. CA. A human homologue of the Drosophila Toll protein signals activation of adaptive immunity. Nature 1997;388:394–7. [2] Takeda K, Akira S. Toll-like receptors in innate immunity. Int Immunol 2005;17:1–14. [3] Schwandner R, Dziarski R, Wesche H, Rothe M, Kirschning CJ. Peptidoglycan- and lipoteichoic acid induced cell activation is mediated by toll-like receptor 2. J Biol Chem 1999;274: 17406–9. [4] Alexopoulou L, Holt AC, Medzhitov R, Flavell RA. Recognition of double-stranded RNA and activation of NF-kappaB by Toll-like receptor 3. Nature 2001;413:732–8. [5] Hoshino K, Takeuchi O, Kawai T, Sanjo H, Ogawa T, Takeda Y, et al. Cutting edge: Toll-like receptor 4 (TLR4)deficient mice are hyporesponsive to lipopolysaccharide: evidence for TLR4 as the Lps gene product. J Immunol 1999; 162:3749–52. [6] Hayashi F, Smith KD, Ozinsky A, Hawn TR, Yi EC, Goodlett DR, et al. The innate immune response to bacterial flagellin is mediated by Toll-like receptor 5. Nature 2001;410: 1099–103. [7] Hajjar AM, O’Mahony DS, Ozinsky A, Underhill DM, Aderem A, Klebanoff SJ, et al. Cutting edge: functional interactions between toll-like receptor (TLR) 2 and TLR1 or TLR6 in response to phenol-soluble modulin. J Immunol 2001;166: 15–9. [8] Hemmi H, Takeuchi O, Kawai T, Kaisho T, Sato S, Sanjo H, et al. A Toll-like receptor recognizes bacterial DNA. Nature 2000; 408:740–5. [9] Tauszig S, Jouanguy E, Hoffmann JA, Imler JL. Toll-related receptors and the control of antimicrobial peptide expression in Drosophila. Proc Natl Acad Sci USA 2000;97:10520–5. [10] Hashimoto C, Hudson KL, Anderson KV. The Toll gene of Drosophila, required for dorsal–ventral embryonic polarity, appears to encode a transmembrane protein. Cell 1988;52:269–79. [11] Ooi JY, Yagi Y, Hu X, Ip YT. The Drosophila Toll-9 activates a constitutive antimicrobial defense. EMBO Rep 2002;3:82–7. [12] Weber AN, Tauszig-Delamasure S, Hoffmann JA, Lelievre E, Gascan H, Ray KP, et al. Binding of the Drosophila cytokine Spatzle to Toll is direct and establishes signaling. Nat Immunol 2003;4:794–800. [13] Rock FL, Hardiman G, Timans JC, Kastelein RA, Bazan JF. A family of human receptors structurally related to Drosophila Toll. Proc Natl Acad Sci USA 1998;95:588–93. [14] Xu Y, Tao X, Shen B, Horng T, Medzhitov R, Manley JL, et al. Structural basis for signal transduction by the Toll/interleukin1 receptor domains. Nature 2000;408:111–5. [15] Luna C, Wang X, Huang Y, Zhang J, Zheng L. Characterization of four Toll related genes during development and immune responses in Anopheles gambiae. Insect Biochem Mol Biol 2002; 32:1171–9. [16] Jault C, Pichon L, Chluba J. Toll-like receptor gene family and TIR-domain adapters in Danio rerio. Mol Immunol 2004;40: 759–71.

T.-C. Cheng et al. [17] Imamura M, Yamakawa M. Molecular cloning and expression of a Toll receptor gene homologue from the silkworm, Bombyx mori. Biochim Biophys Acta 2002;1576:246–54. [18] Xia Q, Zhou Z, Lu C, Cheng D, Dai F, Li B, et al. A draft sequence for the genome of the domesticated silkworm (Bombyx mori). Science 2004;306:1937–40. [19] Mita K, Kasahara M, Sasaki S, Nagayasu Y, Yamada T, Kanamori H, et al. The genome sequence of silkworm, Bombyx mori. DNA Res 2004;11:27–35. [20] Letunic I, Copley RR, Schmidt S, Ciccarelli FD, Doerks T, Schultz J, et al. SMART 4.0: towards genomic data integration. Nucleic Acids Res 2004;32:D142–4. [21] Rost B, Yachdav G, Liu J. The PredictProtein server. Nucleic Acids Res 2004;32:W321–6. [22] Schwede T, Kopp J, Guex N, Peitsch MC. SWISS-MODEL: an automated protein homology-modeling server. Nucleic Acids Res 2003;31:3381–5. [23] Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG. The Clustal X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 1997;25:4876–82. [24] Kumar S, Tamura K, Nei M. MEGA3: Integrated software for molecular evolutionary genetics analysis and sequence alignment. Brief Bioinform 2004;5:150–63. [25] Yang YH, Dudoit S, Luu P, Lin DM, Pengn V, Ngai J, et al. Normalization for cDNA microarray data: a robust composite method addressing single and multiple slide systematic variation. Nucleic Acids Res 2002;30:e15. [26] Kimbrell DA, Beutler B. The evolution and genetics of innate immunity. Nat Rev Genet 2001;2:256–67. [27] Yamamoto K, Narukawa J, Kadono-Okuda K, Nohata J, Sasanuma M, Suetsugu Y, et al. Construction of a single nucleotide polymorphism linkage map for the silkworm, Bombyx mori, based on bacterial artificial chromosome end sequences. Genetics 2006;173:151–61. [28] Miao XX, Xub SJ, Li MH, Li MW, Huang JH, Dai FY, et al. Simple sequence repeat-based consensus linkage map of Bombyx mori. Proc Natl Acad Sci USA 2005;102:16303–8. [29] Eldon E, Kooyer S, D’Evelyn D, Duman M, Lawinger P, Botas J, et al. The Drosophila 18 wheeler is required for morphogenesis and has striking similarities to Toll. Development 1994;120: 885–99. [30] Kobe B, Deisenhofer J. Proteins with leucine-rich repeats. Curr Opin Struct Biol 1995;5:409–16. [31] Christophides GK, Zdobnov E, Barillas-Mury C, Birney E, Blandin S, Blass C, et al. Immunity-related genes and gene families in Anopheles gambiae. Science 2002;298:159–65. [32] De Gregorio E, Spellman PT, Rubin GM, Lemaitre B. Genomewide analysis of the Drosophila immune response by using oligonucleotide microarrays. Proc Natl Acad Sci USA 2001;98: 12590–5. [33] Cheng T, Zhao P, Liu C, Xu P, Gao Z, Xia Q, et al. Structures, regulatory regions, and inductive expression patterns of antimicrobial peptide genes in the silkworm Bombyx mori. Genomics 2006;87:356–65. [34] Yoshida H, Kinoshita K, Ashida M. Purification of a peptidoglycan recognition protein from hemolymph of the silkworm, Bombyx mori. J Biol Chem 1996;271:13854–60. [35] Lee WJ, Lee JD, Kravchenko VV, Ulevitch RJ, Brey PT. Purification and molecular cloning of an inducible Gramnegative bacteria-binding protein from the silkworm, Bombyx mori. Proc Natl Acad Sci USA 1996;93:7888–93. [36] Ochiai M, Ashida M. A pattern-recognition protein for beta-1, 3-glucan. The binding domain and the cDNA cloning of beta-1, 3-glucan recognition protein from the silkworm, Bombyx mori. J Biol Chem 2000;275:4995–5002. [37] Tanaka H, Yamamoto M, Moriyama Y, Yamao M, Furukawa S, Sagisaka A, et al. A novel Rel protein and shortened isoform that

ARTICLE IN PRESS Identification and analysis of Toll-related genes in the domesticated silkworm, Bombyx mori differentially regulate antibacterial peptide genes in the silkworm Bombyx mori. Biochim Biophys Acta 2005;1730:10–21. [38] Brennan CA, Anderson KV. Drosophila: the genetics of innate immune recognition and response. Annu Rev Immunol 2004; 22:457–83. [39] Hoffmann JA. The immune response of Drosophila. Nature 2003;426:33–8.

475

[40] Lemaitre B. The road to Toll. Nat Rev Immunol 2004;4: 521–7. [41] Wang Y, Cheng TC, Subrahmanyam R, Zou Z, Xia QY, Xiang ZH, Jiang H. Proteolytic activation of pro-spa ¨tzle is required for the induced transcription of antimicrobial peptide genes in lepidopteran insects. Dev Comp Immunol 2007, in press., doi:10.1016/j.dci.2007.01.001.