TLR15 Is Unique to Avian and Reptilian Lineages and Recognizes a Yeast-Derived Agonist

This information is current as Amy C. Boyd, Marylene Y. Peroval, John A. Hammond, of September 27, 2021. Michael D. Prickett, John R. Young and Adrian L. Smith J Immunol published online 12 October 2012 http://www.jimmunol.org/content/early/2012/10/12/jimmun ol.1101790 Downloaded from

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision http://www.jimmunol.org/

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication


Subscription Information about subscribing to The Journal of Immunology is online at: by guest on September 27, 2021 http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2012 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published October 12, 2012, doi:10.4049/jimmunol.1101790 The Journal of Immunology

TLR15 Is Unique to Avian and Reptilian Lineages and Recognizes a Yeast-Derived Agonist

Amy C. Boyd,* Marylene Y. Peroval,† John A. Hammond,† Michael D. Prickett,‡ John R. Young,† and Adrian L. Smithx

The TLRs represent a family of pattern recognition receptors critical in the induction of immune responses. Between 10 and 13 different TLR genes can be identified in each vertebrate species, with many represented as orthologous genes in different species. The agonist specificity of orthologous TLR is also highly conserved. In contrast, TLR15 can only be identified in avian and reptilian genomes, suggesting that this receptor arose ∼320 million ago after divergence of the /reptile and mammalian lineages. Transfection of a constitutively active form of chicken TLR15 led to NF-kB activation in HEK293 cells and induced cytokine mRNA upregulation in chicken cell lines. Full-length TLR15 mediated NF-kB induction in response to lysates from yeast,

but not those derived from viral or bacterial pathogens, or a panel of well-characterized TLR agonists. TLR15 responses were Downloaded from induced by whole-cell lysates derived from Candida albicans, Saccharomyces cerevisiae, and Schizosaccharomyces pombe, but not zymosan preparations from S. cerevisiae. The ability of yeast lysate to activate TLR15-dependent NF-kB pathways (in transfection assays) or stimulate IL-1b mRNA upregulation in chicken macrophages was abrogated by heat inactivation or pre-exposure of the lysate to PMSF. Identification of yeast as an agonist source for TLR15 provides a functional framework for consideration of this TLR within the context of pattern recognition receptor and may impact on the development of novel adjuvants. The

Journal of Immunology, 2012, 189: 000–000. http://www.jimmunol.org/

attern recognition receptors (PRRs) are germline-encoded TLR families respond to distinct groups of agonists: the TLR1/2/6/ receptors that respond to a variety of conserved pathogen- 10 family recognizes lipopeptides (2, 3), the TLR3 family responds P associated molecular patterns (PAMPs) or microbial- to double-stranded RNA (4), the TLR4 family recognizes LPS (5), associated molecular patterns to initiate the inflammatory response. TLR5 family members respond to flagellin (6), and the TLR7–9 The TLRs are type I transmembrane glycoproteins characterized family recognizes nucleic acid motifs (7–10). by an agonist recognition domain comprising variable numbers of The sixth family, the TLR11 family (including TLR11–13 and leucine-rich repeats (LRRs) and the intracellular Toll/IL-1R ho- TLR20–23), is less well characterized and contains members that are mologous (TIR) domain, which mediates signaling. TLR genes and less widely distributed in (1). Murine TLR11 (mTLR11) by guest on September 27, 2021 gene families are conserved in divergent species suggesting strong has been reported to respond to agonists derived from uropathogenic selective pressure for maintenance of function (1). The selective bacteria (11) and the apicomplexan protozoan PAMP, profilin (12). pressures to maintain TLR sequence are likely defined by the con- Pufferfish (Takifugu rubripes) TLR22 recognizes long double- servation of PAMPs (and damage-associated molecular patterns) and stranded RNA (13), and chicken TLR21 (cTLR21) has been re- the requirement to associate with accessory molecules (1). Between ported to respond to DNA rich in unmethylated CpG motifs (14, 15). 10 and 13 TLR genes are present in most vertebrates, and these can The diversity of agonist specificities in this family that includes be phylogenetically grouped into 6 major families (1). Five of the six nucleic acid and nonnucleic acid structures suggests that this family should be functionally subdivided. The cTLR repertoire comprises 10 genes including members of *The Jenner Institute, Nuffield Department of Clinical Medicine, University of Ox- each of the six families. Many cTLRs have been characterized ford, Oxford OX3 7DQ, United Kingdom; †Institute for Health, Compton, Berkshire RG20 7NN, United Kingdom; ‡Cluster in Biomedicine, 34149 Basovizza, including TLR3, TLR4, TLR5, and TLR7, which are orthologous Trieste, Italy; and xDepartment of , University of Oxford, Oxford OX1 3PS, to their mammalian counterparts and share agonist specificity United Kingdom (16–21). Unlike , chickens possess two TLR1 genes Received for publication July 5, 2011. Accepted for publication September 7, 2012. (TLR1.1 and 1.2) and two TLR2 genes (TLR2.1 and TLR2.2) that This work was supported by student support from the Biotechnology and Biological can form homodimers or heterodimers and respond to peptido- Sciences Research Council and the Institute for Animal Health, with additional support from Department for Environment, and Rural Affairs Project OD0547. A.L.S. is glycan, diacylated lipopeptides, and triacylated lipopeptides (22– a fellow of The Jenner Institute. 24). In mammalian species, these agonists are recognized by Address correspondence and reprint requests to Dr. Adrian L. Smith, Department of various heterodimers composed of TLR1 or TLR6 with the single Zoology, University of Oxford, Tinbergen Building, South Parks Road, Oxford OX1 TLR2. The duplication of TLR2 in the chicken and zebra finch 3PS, U.K. E-mail address: [email protected].ox.ac.uk (22, 25) appears to have occurred before the divergence of Abbreviations used in this article: ALV, avian leukosis virus; ATCC, American Type and mammals, as TLR2 pseudogenes can be found upstream of Culture Collection; CA, constitutively active; CAcTIR4, constitutively active chicken TLR4; CAcTIR15, constitutively active chicken TLR15; CAmTIR4, constitutively the functional TLR2 sequences in opossum, dog, and (1). active murine TLR4; CATLR, constitutively active TLR; Ct, threshold cycle; cTLR, Additional differences between mammalian and avian TLR rep- chicken TLR; GOI, gene of interest; LRR, leucine-rich repeat; mTIRcon, murine TIR deficient control construct; mTLR, murine TLR; mya, million years ago; PAMP, ertoires include the disruption of TLR8 in galliform and anseri- pathogen-associated molecular pattern; PRR, pattern recognition receptor; qRT-PCR, form birds by a retroviral insertion (17, 26) and the lack of a TLR9 quantitative RT-PCR; SEAP, secreted human embryonic alkaline phosphatase; TIR, homolog in the chicken. Before this study, TLR15 was identified Toll/IL-1R homologous. in the chicken and zebra finch genomes (1, 25). We identified Copyright Ó 2012 by The American Association of Immunologists, Inc. 0022-1767/12/$16.00 TLR15 in the genomic assemblies of two further avian species,

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1101790 2 EVOLUTION AND FUNCTION OF TLR15 the duck (Anas platyrhynchos) and (Meleagris gal- CCACCCAATCCAGGAAATGTTAACCCA-39. Primers were synthesized lopavo). A partial TLR15 sequence was also identified in the by Sigma. Probes labeled with the fluorescent reporter dye FAM at the 59 9 genome of the Carolina anole ( carolinensis). These end and the quencher TAMRA at the 3 end were obtained from Euro- gentec (Southampton, U.K.). Primer and probe sequences for 28S RNA sequences afforded an opportunity to resolve the phylogenetic and cytokine mRNA assays were as previously published (32, 33). relationship between TLR15 and the other TLR families. Fur- Quantitative RT-PCR (qRT-PCR) was performed using the Applied thermore, we demonstrated the functional capability of TLR15: Biosystems TaqMan FAST Universal PCR Master Mix. Amplification and it activates innate immunity and specifically recognizes novel detection of target sequences were performed using the Applied Biosystems 7500 FAST Real-Time PCR System with the following cycle conditions: agonist structures present in whole-cell lysates derived from one cycle of 48˚C for 30 min and 94˚C for 20 s, and 40 cycles of 94˚C for yeast/fungal pathogens. 3 s, 60˚C for 30 s. For analysis of cTLR mRNA expression in tissues, 28S values were used Materials and Methods to adjust for differences in input RNA. Samples were diluted 1:5 for gene of interest (GOI) and 1:500 for 28S reactions. For analysis of cTLR mRNA Phylogenetic analysis expression in cell subsets and analysis of cytokine mRNA expression To identify TLR15 sequences in other genomes, we used the cTLR15 in transfected cells, the concentration of RNA within each sample was mRNA sequence (NM001037835) in BLAST searches of the following determined using a NanoDrop 1000 spectrophotometer (Thermo Fisher genomes: the duck (A. platyrhynchos), wild turkey (M. gallopavo), zebra Scientific, Loughborough, U.K.). These samples were adjusted to the same finch (Taeniopygia guttata), and Carolina anole lizard (Anolis caro- concentration by dilution in nuclease-free water (Qiagen). These normal- linensis). TLR15 has not been found in any mammalian genome, or in the ized RNA samples were used directly in cytokine or TLR qRT-PCR assays Xenopus tropicalis, Salamander, or piscine genomes (1) (A.C. Boyd, M.Y. and diluted 1:100 in nuclease-free water for the reference gene (28S) Peroval, J.A. Hammond, M.D. Prickett, J.R. Young, and A.L. Smith, un- assays as RNA amounts were lower than those obtained from tissue. For published observations). Putative TLR15 sequences were aligned with TLR and 28S standard curves, serial 10-fold dilutions of a single prepa- Downloaded from cTLR15 sequences and avian, human, murine, Xenopus tropicalis,and ration of RNA from a Rhode Island Red chicken were used. For TLR1 and TLR2 family sequences using ClustalX cytokine standard curves, RNA from COS-7 (African Green monkey (27). TLR15 has been described as being most closely related to the TLR1 cell line transformed by replication defective Simian vacuolating and TLR2 subfamily (1). The resulting alignment was evaluated manually, virus 40; American Type Culture Collection [ATCC] no. CRL-1651) cells and where necessary adjusted on the grounds of structural constraints using transfected with plasmids expressing cytokine genes was used. Bioedit (28). Phylogenetic analysis was performed in MEGA4.1 to pro- Quantification was based on the increased fluorescence detected be- cause of hydrolysis of the target-specific probes by the 59 to 39 exonuclease duce bootstrapped neighbor joining trees using the Tamura-Nei (if se- http://www.jimmunol.org/ quence distances allowed) and Tajima-Nei methods (29, 30). activity of the hot-start DNA polymerase during amplification. Normali- zation of the reporter signal was achieved through use of the passive ref- Tissue samples and ex vivo cell populations erence dye 6-carboxy-c-rhodamine, which is not involved in amplification. Results are expressed as the threshold cycle value (Ct) at which the change Specific pathogen-free Rhode Island Red chickens were supplied by the in the reporter dye passes a significance threshold. The threshold level was Poultry Production Unit of the Institute for Animal Health, Compton set to the same value on each plate where a given gene was assayed. Ct Laboratory (all work involving was performed in accordance with values for TLR or cytokine mRNA product for each sample were adjusted local and national regulations). All tissue and cell samples were obtained using the 28S rRNA Ct value for the same sample, to account for variation from birds at 6 wk posthatch. Tissue samples were collected and stored in in sampling and RNA preparation. The slopes of a plot of Ct against log10 RNAlater until processing. Tissues were homogenized using a mini-bead of the standard dilution series present on each plate where a given gene beater (Biospec Products, Newmarket, U.K.), and RNA was extracted was assayed were calculated. The slopes of the respective GOI and 28S by guest on September 27, 2021 using an RNeasy mini kit (Qiagen, Crawley, U.K.). dilution series were then used to calculate GOI Ct values and adjust for Blood-derived adherent cell populations (predominantly composed of differences in input total RNA as follows: corrected Ct value = Ct + macrophages) were isolated as previously described (31). In brief, hepa- (Nt – Ct9) 3 S/S9, where Ct is the mean of triplicate GOI Ct values, Nt is rinized blood was mixed with an equal volume of PBS, overlaid onto an the median 28S Ct for all samples within an experiment, Ct9 is the mean of equal volume of Histopaque 1083 (Sigma, Gillingham, U.K.), and centri- triplicate 28S values of individual sample, S is the GOI slope, and S9 is the 3 fuged at 1200 g for 40 min without braking. Cells at the interface were 28S slope. Results were expressed as 40 2 Ct values or in fold change (34). washed three times in ice-cold PBS supplemented with 5% FBS (Autogen Bioclear, Caine, U.K.) by centrifugation at 400 3 g, then pelleted by cen- Cloning of TLR15 and creation of constitutively active TLR trifugation at 400 3 g for 5 min at 4˚C. Cells were then resuspended in RPMI-1640 (Invitrogen, Paisley, U.K.) with 5% chicken serum (Invi- Total RNA was isolated from the spleen of a 6-wk-old Rhode Island Red 21 trogen), 10 U ml penicillin and streptomycin, and 100 mM L-glutamine. chicken using an RNeasy mini kit (Qiagen), and cDNA was prepared using Cells were adjusted to 5 3 106 cells/ml in medium, and 1-ml aliquots of the an iScript Select cDNA synthesis kit (Bio-Rad Laboratories). All PCRs cell preparation were dispensed into 24-well plates and incubated for 48 h at were performed with the proofreading BIO-X-ACT (Bioline, Little 37˚C, 5% CO2, with a change of medium after 24 h. All other cell subsets Clacton, U.K.). Sequencing was performed using the Applied Biosystems were isolated from the spleen using FACS. were removed from BigDye Terminator v3.1 sequencing kit (Applied Biosystems). Primers used birds and transferred immediately to PBS/FBS. Spleens were disrupted by for cloning of TLR15 and constitutively active (CA) constructs are given in fragmentation and passed through a 40-mm cell strainer into a petri dish. Table I. The full-length coding sequence of TLR15 mRNA was amplified The cell suspension was layered onto Ficoll density gradient media (1.077 and cloned into the TA vector pTargeT (Promega, Southampton, U.K.). g/ml; GE Healthcare, Hatfield, U.K.) and centrifugated at 450 3 g for Constitutively active TLRs (CATLRs) were created as previously described 30 min (no brake). Cells were removed from the interface, washed twice (35). In brief, a PCR and cloning approach was used to link the extra- in PBS/FBS, counted, and resuspended to a concentration of 1 3 107/ml. cellular domain of murine CD4 to the TIR domains of mTLR4 (designated Primary biotinylated Abs and Abs directly conjugated to FITC or R-PE CA mTLR4 [constitutively active murine TLR4 (CAmTIR4]), cTLR4 were added at 1 mg/106 cells, and reactions were incubated on ice for 20 (constitutively active chicken TLR4 [CAcTIR4]), and cTLR15 (constitutively min (all Abs from Cambridge Bioscience, Cambridge, U.K.). Cells were active chicken TLR15 [CAcTIR15]). A control construct was generated washed three times by addition of PBS/FBS and centrifugation at 400 3 g comprising the extracellular domain of murine CD4 and the transmembrane for 4 min at 4˚C. Cells were resuspended in PBS/FBS containing strepta- domain of mTLR4, but lacking the TIR domain and thus unable to signal vidin conjugated with allophycocyanin (Cambridge Biosciences) to a final (designated mTIRcon). The pTargeT vector was used as the backbone for concentration of 1 mg/ml and were incubated on ice for 20 min. Cells were all constructs. Plasmids encoding the CA and control constructs were then then washed twice and resuspended in PBS supplemented with 5% FBS and transfected into cells to assess induction of innate immune responses. 2 mM EDTA solution (Sigma). FACS was carried out using a FACSAria (Becton Dickinson, Oxford, U.K.). Cell culture Quantitative RT-PCR The human embryonic kidney (HEK293; ATCC no. CRL-1573) cell line was cultured in Eagle’s MEM (Sigma), with 2 g/l sodium bicarbonate, TLR15 primer and probe sets were designed using the Primer Express 2nML-glutamine (Invitrogen), 10% FBS (Autogen Bioclear), and 13 software program (Applied Biosystems, Warrington, U.K.). Sequences were nonessential amino acids (Invitrogen) at 37˚C, 5% CO2. HEK293 were as follows: forward primer, 59-AGCTGAACTGCTGCCACATTT-39; re- plated at 1.5 3 105 cells/well on a 24-well plate (Thermo Fisher Scientific) verse primer, 59-TTTCCTCTGTTCTTCTTTGTCTGAATC-39; probe, 59- for transfection 24 h later. The Journal of Immunology 3

Table I. Primers used to create CATLR and to clone TLR15


DF1 cells (spontaneously transformed chicken embryonic fibroblast cell Oxford). S. cerevisiae was cultured from a dried commercial preparation line; ATCC no. CRL-12203) were cultured in DMEM (Sigma) with 3.7 g/l (Sainsbury’s, U.K.). To investigate the of the TLR15-agonist activity sodium bicarbonate, 13 L-glutamine, 10% FBS, and 1% sodium pyruvate in the yeast lysates, these were incubated with 1 mM PMSF or heat treated at 41˚C, 5% CO2. DF1 were plated and cultured at the same densities and (100˚C for 10 min). under the same conditions as HEK293 cells.

HD11 cells (chicken macrophage-like cell line) (36) were cultured in Downloaded from RPMI-1640 medium (Invitrogen) containing 20 mM L-glutamine, 2.5% Results FBS, 2.5% chicken serum, and 10% tryptose phosphate broth at 41˚C, 5% TLR15 is conserved between avians and reptiles, and is distinct 3 5 CO2 in air. HD11 were plated at 3 10 cells/well, in 24-well plates, for from the TLR1/2 subfamily transfection 2 d later. TLR15 orthologs were identified in the four available avian Transfection genomes (chicken, duck, turkey, and zebra finch). We also iden-

The pNiFty-SEAP (secreted human embryonic alkaline phosphatase) re- tified a 1.7-kb TLR15 gene fragment in the Carolina anole lizard http://www.jimmunol.org/ porter plasmid (Invivogen, Autogen Bioclear) was cotransfected into cells genome assembly. Reptiles shared a common ancestor with birds with either empty vector or CA constructs plus empty vector to normalize 259.7–299.8 million years ago (mya), and both diverged from the transfected DNA amounts. HEK293 cells were transfected with either 0.125 ∼ mg recombinant TLR plasmid or the control plasmid, plus 0.4 mg of the mammalian lineage 312.3–330.4 mya (37). TLR15 was not reporter plasmid, made up to a total of 0.8 mg plasmid DNA per well with found in any other genome assembly. empty vector. The reporter plasmid was not functional in chicken cells, Alignment of predicted TLR15 amino acid sequences revealed which were transfected with 0.125 mg recombinant TLR plasmid or the that TLR15 is highly conserved between birds and the anole, the m control plasmid, plus 0.675 g empty vector. To transfect cells grown in only reptilian species with an available genome assembly (Fig. 1). 24-well plates, 1.5 ml Lipofectamine 2000 (Invitrogen) was made up to 50 ml with OptiMEM I reduced serum medium and incubated for 5 min at The TIR domain is particularly conserved at 88–97% amino acid room temperature. Plasmid DNA was diluted in 50 ml OptiMEM I, mixed identity between chicken and other birds, and 80% compared with by guest on September 27, 2021 with the diluted Lipofectamine 2000, and incubated for 20 min at room the anole TLR15. Overall, there is also high sequence similarity temperature. Each well received 100 ml of the appropriate plasmid/ within the LRR regions, and the distribution of predicted LRR is Lipofectamine 2000 mixture with 400 ml fresh media and was incubated generally conserved between avian and reptilian TLR15 mole- for 6 h (at 37˚C or 41˚C, 5% CO2). Culture wells were washed and replenished with 500 ml of the appropriate media before further incubation cules. Within the external domain there are four regions where the or manipulation. Transfected cells were incubated with or without agonist TLR15 genes contain indels. With reference to the cTLR15, these or microbial lysates for appropriate periods (see Results). The supernatant lie at positions 31 (2 aa shorter in zebra finch and 1 aa shorter in was collected and the catalytic activity of SEAP for p-Nitrophenyl phos- the duck); at position 136, a 9-aa deletion in duck; at position 180, phate determined using the SEAP reporter assay kit (Autogen Bioclear) following the manufacturer’s instructions. an insertion of 5 (duck) or 6 aa (zebra finch); and at position 370, a 2-aa insertion in zebra finch and a 5-aa deletion in anole (Fig. 1). TLR agonists and microbial lysates The consequences of these polymorphisms for TLR15 function All purified TLR agonists were obtained from Invivogen and used at the are unknown. following concentrations unless otherwise indicated; Pam3CSK4 (1 mg/ml), Initial analysis indicated that TLR15 was most closely related to FSL-1 (1 mg/ml), Staphylococcus aureus peptidoglycan (100 mg/ml), the TLR1 subfamily as previously reported (1). To investigate the m polyinosinic-polycytidylic acid (50 g/ml), Escherichia coli LPS (10 phylogeny of TLR15, TLR1/2 subfamily members were identified mg/ml), enterica serovar Typhimurium flagellin (10 mg/ml), R848 (1 mg/ml), and E. coli endotoxin-free DNA (5 mg/ml). Cells were in the four available avian genomes and in the human, murine, stimulated with agonist diluted in complete media and cells cultured under anole, and Xenopus genomes. Although homologs of the four appropriate conditions before analysis. cTLR1/2 family members were identified in the zebra finch ge- The following microbial lysates were used: sonicated S. enterica serovar nome, TLR1.2 was not located in the duck genome (because of an Typhimurium lysate (1 mg/ml), UV-inactivated avian leukosis virus (ALV; used at 1 3 105 tissue culture infectious units); yeast lysates; Candida identifiable gap in the genome sequence), and only a partial se- albicans (1 mg/ml); Schizosaccharomyces pombe (1 mg/ml); Saccharo- quence for TLR2.2 (lacking the TIR domain) was identified in the myces cerevisiae (1 mg/ml) were prepared by bead-milling 2 3 2 min turkey genome. Sequences encoding two homologs of TLR1 and Mini-Beadbeater (Biospec Products) followed by 103 freeze–thaw cycles. three homologs of TLR2 were identified in the anole lizard ge- Lysate sample concentrations were standardized by protein concentration nome compared with two of each in the chicken. using A280 absorbance using a NanoDrop 1000 spectrophotometer (Thermo Fisher Scientific). ALV (subgroup A, strain RAV-1) was a gift from the Alignment of TLR sequences from these diverse species reveals laboratory of Prof. Venugopal Nair (Institute for Animal Health, Compton, conservation of the TIR domains in contrast with the extensive U.K.) and was inactivated using a Stratalinker Crosslinker (Stra- diversification of the extracellular domains. The regional pattern of tagene) set at 120 mJ for 10 min. S. enterica serovar Typhimurium F98 was sequence conservation between the TIR and extracellular domains originally provided by Prof. Paul Barrow (Nottingham, UK). C. albicans was a gift from Dr. Nigel Saunders and Dr. David Greaves (Sir William Dunn of TLR15 in different avian species suggests that these domains School of Pathology, Oxford, U.K.). S. pombe was a gift from Dr. Stephen have been subject to distinct selection pressures. Phylogenetic Kearsey and Marianne Shepherd (Department of Zoology, University of analysis of the TIR domains from mammalian, reptilian, and 4 EVOLUTION AND FUNCTION OF TLR15 Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021

FIGURE 1. Sequence analysis reveals conservation between TLR15 from avian species and Anolis carolinensis. Amino acid alignment of avian and lizard TLR15 sequences. The entire chicken sequence is shown in each top row in gray text. Amino acids in other sequences are only shown where they differ compared with the chicken sequence. Gray dashes within the sequences represent regions that do not align, or missing sequence in the case of the A. carolinensis sequence (alignment position 1–299). Black text within the chicken sequence represents LRRs; thin dashed black line represents the LRR C- terminal region; thick black dashed line represents the transmembrane domain region; dotted line represents the TIR domain. Boxes indicate motifs defined for the IL-1R family (47). Ac, Anolis carolinensis (Carolina anole lizard); Ap, A. platyrhynchos (duck); Gg, Gallus gallus (chicken); Mg, M. gallopavo (wild turkey); Tg, T. guttata (zebra finch). species TLR1 and TLR2 subfamily members were Similar analysis of the extracellular domains was performed included to assess validity of, and increase support for, the TIR using the Tajima-Nei model (Fig. 2C). Distances between these domain tree (Fig. 2A). The overall architecture of the tree shows sequences were too great to use the Tamura-Nei model, but all the expected phylogeny of species (Fig. 2B), with Xenopus as the other models tested produced almost identical results. In contrast most vertebrate in this analysis (37). Trees derived using with Fig. 2A, sequences group mainly by gene family, as opposed other methods produced essentially identical results. The TIR to species for the avian TLR1 subfamily. The clustering of these domains from each TLR subfamily group together by species, agonist recognition domain sequences in a gene-specific and not with strong support at the relevant branch nodes. The identifica- species-specific manner implies a strong selective pressure to re- tion of orthologous genes from the anole genome is also well tain specific PAMP recognition capacity in theTLR1 family supported. This analysis illustrates that the TIR domains of TLR1 members. For example, the avian TLR1.1 extracellular domains and TLR2 have evolved in a species-specific manner since these group together, as do the TLR1.2 sequences. The TLR2 subfamily gene families duplicated, a feature that may be related to inter- extracellular domains do not replicate this pattern, and grouping is actions with species-specific signaling molecules. Fig. 2A also by species with the exception of chicken and turkey TLR2.1. This reveals that the TIR domain of TLR15 falls within the large TLR1 suggests that the TLR2 subfamily has evolved in a species-specific and TLR2 , as previously reported (1), and appears most manner similar to that seen with the TIR domains. The consid- closely related to TLR1. erable divergence of the extracellular domains precluded mean- The Journal of Immunology 5

FIGURE 2. Bootstrap neighbor joining tree showing phylogenetic relationships between the TIR domains and extracellular domains of the TLR1/2 subfamily and TLR15. Bootstrap neighbor joining tree showing phylogenetic relationships between (A) the TIR domains and (C) extracellular domains of the TLR1/2 sub- family and TLR15. Sequences were aligned us- ing the ClustalW program and checked manually using the Bioedit program. Trees were derived using MEGA4.1 software. Node support scores ,50% have been removed. (B) Diagrammatic representation of phylogenetic relationships be- tween vertebrate groups (not to ). The data set used in the phylogenetic analysis is available on request. Downloaded from http://www.jimmunol.org/ ingful phylogenetic comparison with TLR genes from distantly in the spleen and thymus, with slightly less signal in the bursa, related groups such as mammals. , and colon. The lowest level of TLR15 mRNA in the tissue The TIR domains of TLR1/2 and TLR15 formed a distinct clade, samples was in the . In the cell subset samples, the whereas the extracellular domains of the TLR15 sequences were highest level of TLR15 mRNA was detected in the adherent cell isolated from the TLR1 and TLR2 subfamilies. The phylogeny and fraction, a population known to contain significant numbers of the low node score (,50) suggest that although functional con- macrophages. The FACS-sorted lymphocyte subsets expressed straints have maintained a degree of similarity between the TLR1/2 similar levels of TLR15 mRNA, but this was less than in adherent and TLR15 signaling (TIR) domains, the extracellular agonist cells. TLR15 expression in the tissues may therefore reflect the recognition domain has undergone extensive diversification from relative abundance of macrophages. by guest on September 27, 2021 the ancestral sequence. This divergence of the extracellular do- CATLR15 activates NF-kB induction in HEK293 cells and main of TLR15 from TLR1 and TLR2 implied a different agonist cytokine mRNA induction in chicken cell lines specificity, necessitating investigation of the expression patterns and functional of TLR15. To determine the capacity for cTLR15 TIR domain to mediate a proinflammatory signal, we constructed a CAcTLR15 TIR domain TLR15 mRNA expression in tissues and cell subsets isolated (CAcTIR15) (Table I). Cotransfection of HEK293 cells with from 6-wk-old chickens CAcTIR15 and the NF-kB reporter plasmid induced significant TLR15 mRNA was detected in all of the tissue and cell subsets NF-kB activation (p , 0.001) at comparable levels with that in- analyzed (Fig. 3). The highest levels of TLR15 mRNA were found duced by CAmTIR4 (Fig. 4). To assess the impact of activation of the TLR15-mediated signaling pathway in chicken cells, we transfected the CAc-

FIGURE 4. NF-kB activation in HEK293 cells transfected with CATLR. FIGURE 3. TLR15 mRNA levels in chicken tissues and cell subsets. HEK293 cells transfected with control plasmid (mTIRcon) or CAmTIR4, TLR15 mRNA levels measured using qRT-PCR in tissues (left panel) and cTLR4 (CAcTIR4), or cTLR15 (CAcTIR15) were cultured for 40 h before cell subsets (right panel) from 6-wk-old chickens. The mRNA level is SEAP activity was measured. Data represent the mean of three separate expressed as 40 2 Ct and is adjusted for 28S. Tissue or cell subset is wells; error bars indicate SEM. Data are representative of three indepen- indicated on the x-axis. Cell subset abbreviations are for adherent cells dent experiments. Significant differences in NF-kB activation relative to (Ad) and then surface proteins used to isolate the remaining subsets by supernatant from cells transfected with the control plasmid were calculated FACS. Data comprise results for three birds per tissue and four birds per using Dunnett’s multiple-comparison test with a Bonferroni correction. cell subset. Error bars indicate SEM. ***p , 0.001. 6 EVOLUTION AND FUNCTION OF TLR15

TIR15 into a chicken fibroblast cell line (DF1) and a macrophage- lian cells that express the appropriate TLR. Exposure of cTLR15- like cell line (HD11), and subjected it to qRT-PCR analysis of transfected HEK293 cells to the defined agonist preparations cytokine mRNA. CAcTIR15, CAcTIR4, and CAmTIR4 induced failed to stimulate a cTLR15-dependent increase in SEAP activity significant increases in cytokine mRNA levels in cells transfected in the supernatant (Fig. 6A). A small increase in SEAP was relative to the control plasmid (Fig. 5). In both cell types, all CA detected in HEK293 cells exposed to flagellin, but this was highest TIR induced increased levels of IL-1b and CXCLi2 mRNA, al- in cells transfected with empty plasmid and is most likely due to though only the CAcTIR induced upregulation of IL-6 mRNA. the expression of TLR5 by HEK293 cells (38, 39). Because TLR15 did not mediate NF-kB induction in response to TLR15 activates NF-kB induction in HEK293 cells in response stimulation with any of the commonly used agonists, we widened to yeast lysates, but not in response to a panel of TLR agonists, our search to include lysates derived from different groups of viral lysates, or bacterial lysates pathogen (viral, bacterial, and yeast). No significant difference in Because CAcTIR15 was functional in HEK293 cells, we were able reporter induction was observed in response to ALV in TLR15- to construct a screen for sources of TLR15-stimulating agonists transfected cells compared with cells transfected with empty based on transfection with a plasmid containing full-length wild vector (Fig. 6B). Reporter induction was observed in cells stim- type cTLR15 and the NF-kB reporter plasmid. Cells transfected ulated with S. enterica serovar Typhimurium lysate, but this was with CAmTIR4 were included as a positive control. In the first not dependent on TLR15 expression and may reflect detection instance, cells were exposed to a panel of defined PRR agonists of flagellin by TLR5 expressed by HEK293 cells. In contrast, cells (Pam3Cys, FSL1, PGN, polyinosinic-polycytidylic acid, LPS, stimulated with S. cerevisiae lysate resulted in a substantial

flagellin, R848, or unmethylated CpG DNA) all known to stimu- cTLR15-dependent increase in SEAP. The levels of SEAP induced Downloaded from late proinflammatory cytokine expression in avian and mamma- with S. cerevisiae lysate were comparable with that detected after induction with CAmTLR4. Lysates prepared from C. albicans and S. pombe also induced significant TLR15-dependent reporter ac- tivation, although the latter was less potent. Collectively, these data indicate that TLR15 mediates NF-kB induction in response to

a yeast-derived molecular pattern. Mammalian TLR2 was reported http://www.jimmunol.org/ to recognize the lipopeptide component of zymosan (40, 41). However, exposure of cTLR15-transfected HEK293 cells to a range of concentrations of a zymosan preparation (from S. cer- evisiae) failed to induce reporter activation (Fig. 6B). During the preparation of this article, de Zoete et al. (42) proposed that TLR15 was activated by secreted microbial proteases. To inves- tigate the nature of the TLR agonist activity in whole-cell yeast lysates, we examined the effect of heat inactivation or pretreat- by guest on September 27, 2021

FIGURE 5. Cytokine induction in chicken cells transfected with CATLR. DF1 and HD11 cells transfected with CAmTIR4, CAcTIR4, or CAcTIR15 were cultured for 40 h, RNA extracted, and qRT-PCR per- FIGURE 6. NF-kB induction in response to TLR agonists and microbial formed for IL-1B (A, B), CXCLi2 (C, D), and IL-6 (E, F) mRNA. Data lysates in HEK293 cells transfected with TLR15. HEK293 cells trans- represent fold changes relative to cytokine induction in cells transfected fected with TLR15 (gray bars) or empty vector (white bars) were treated with control plasmid in three separate wells. Data are representative of with TLR agonists (A), microbial lysates, and a range of concentrations of three independent experiments. Significant differences in cytokine induc- zymosan (B). CAmTIR4 was used as a positive control (A). Each bar com- tion relative to supernatant from cells transfected with the control plasmid prises data from three separate wells, and data are representative of three were calculated from raw data using Dunnett’s multiple-comparison test independent experiments. Agonist and lysate concentrations used are given with a Bonferroni correction. *p , 0.05, ***p , 0.001. in Materials and Methods. ***p , 0.001 calculated using two-way ANOVA. The Journal of Immunology 7 ment of the lysate with PMSF. Both of these treatments abrogated the TLR15 stimulatory activity of yeast lysates (Fig. 7) as mea- sured in an NF-kB induction assay. Yeast lysates stimulate the inflammatory response in HD11 cells Because TLR15 mRNA was detected at high levels in macro- phages, we examined the consequences of exposure of HD11 cells to S. cerevisiae lysate. The level of IL-1b mRNA was increased by treatment with S. cerevisiae lysate relative to that detected with untreated cells (Fig. 8). In contrast, the zymosan preparation (at 1 mg/ml) failed to stimulate any changes in the level of IL-1b mRNA (data not shown). The stimulatory activity of S. cerevisiae lysate was completely inhibited by exposure to PMSF or heat inactivation (Fig. 8).

Discussion TLR15 could be described as an orphan TLR, because phylogenetic b analysis reveals it does not firmly group with any of the TLR FIGURE 8. Upregulation of IL-1 mRNA levels in HD11 cells treated Downloaded from families identified to date. We report the identification of TLR15 in with yeast lysate. HD11 cells were cultured with S. cerevisiae lysate at m all four currently available avian genomes (representing divergent a concentration of 1 g/ml or with PMSF-treated lysate (1 mM) or heat- inactivated (HI) lysate (1 mg/ml) for 1 h. Levels of IL-1b mRNA were bird groups) and identification of a partial sequence in the genome determined by qRT-PCR. Data represent fold changes relative to mRNA of a reptile, the lizard Anolis carolinensis. Birds, reptiles, and levels in the untreated cells (3 wells/treatment). *p , 0.05 calculated using mammals belong to the clade Reptiliomorpha, which diverged from one-way ANOVA. ∼ its sister clade, Amphibia, 330–350 mya (37) (Fig. 2B). The http://www.jimmunol.org/ Reptiliomorpha subgroups, the Synapsida (mammals and extinct relatives) and Diapsida (reptiles, birds, and extinct relatives), di- exception in this data set was TLR2.1 in the chicken and turkey, verged 312–330 mya (37). The phylogenetic distribution of TLR15 although this may reflect the fact that TLR2.1 is significantly suggests that this TLR arose by gene duplication from a TLR1/2- different from TLR2.2, and the divergence of chickens and turkeys like ancestor between 330 and 260 mya, after the divergence of is a recent evolutionary event. Inclusion of turkey TLR2.1 as the the Reptiliomorpha from Amphibia and Mammalia, but before the genome is completed may alter this distribution. The avian TLR1 divergence of birds from (37). This suggests that TLR15 extracellular domain sequences grouped by gene family, suggest- represents an expansion of the TLR repertoire restricted to birds ing that selection has maintained patterns of agonist recognition and reptiles. with each TLR1 family member. The mammalian TLR1 gene by guest on September 27, 2021 The TIR domain of the different TLR families has remained family includes TLR1, TLR6, and TLR10, reflecting the extent of highly conserved in phylogenetically diverse vertebrate groups, the diversification of this gene family from an ancestral TLR. The suggesting a strong influence of purifying selection (1). In contrast, TLR1 subfamily (the heterodimeric partners of TLR2) appears to the extracellular domains of TLR exhibit considerably greater have evolved under lesser constraints than the TLR2 subfamily (1). levels of divergence between species (43). In this study, TLR2 TIR Why the TLR1 family should diversify so freely in comparison domains and extracellular domains grouped by species, suggesting with TLR2 is intriguing. TLR15 was tentatively assigned to the species-specific constraints dominated the evolution of TLR2. The TLR1 subfamily, and our data support this scenario with the TIR domain, although the extracellular domain is more divergent. The presence of the TLR1.1/1.2 and TLR2.1/2.2 pairs in several avian species (and in the reptile Anolis carolinensis) suggests that gene duplication in each family occurred before the divergence of these avian and reptilian lineages. However, this might be ex- pected to produce a phylogenetic tree where, for example, the TIR domains of TLR1.1 and TLR1.2 were located in separate containing all species. This implies some constraint that has maintained similarity of the TIR domain within a gene family, whereas allowing it to diverge between species, which may relate to the need to interact with signaling molecules recruited by the TIR domains. Purifying selection also appeared to act on the ag- onist recognition domains of the TLR2 subfamilies. Conversely, diversification of the TLR1 extracellular domain appears to have been driven by the requirement to interact with ligand. The phy- logenetic association of TLR15 with the TLR1/2 subfamily FIGURE 7. TLR15 activation after heat or PMSF inactivation of lysates. implicates TLR15 as the most extreme example of divergence HEK293 cells transfected with TLR15 (gray bars) or empty vector (white from the ancestral TLR1 sequence, possibly driven by a require- bars) were exposed to yeast whole-cell lysates (1 mg/ml) or lysates treated with heat (100˚C, 10 min, heat inactivation [HI]) or 1 mM PMSF. Each bar ment to recognize a particular agonist type. represents mean 6 SEM activation of the SEAP NF-kB reporter construct To define the functional capability of the TIR domain, we from three separate wells, and all data are representative of two indepen- constructed CA forms of the TLR15 TIR domain. CAcTIR15 (and dent experiments. ****p , 0.0001 calculated using one-way ANOVA of CAmTIR4 positive control) induced upregulation of NF-kB acti- TLR15-transfected cell data. vation in HEK293 cells and increased expression of cytokine 8 EVOLUTION AND FUNCTION OF TLR15 mRNA in chicken cells. These data clearly demonstrated that the sponse to yeast lysate suggests that TLR15 is the primary PRR TLR15 TIR domain is functionally intact and can mediate involved in the avian macrophage response to yeast. proinflammatory signals. There were some differences in the ca- The discovery of a specific source of agonist for TLR15 rep- pacity of CAmTIR4 and CAcTIR4 to drive responses in human or resents a major step forward and has the potential to impact chicken cell lines, with the strongest signals being achieved by practically on adjuvant development for use in the poultry industry. matched species transfections. CAcTIR15 induced similar levels Identification of a specific agonist for this TLR may reveal an avian of NF-kB induction to CAmTIR4 in HEK293 cells, but lower (and possibly reptilian)-specific agonist or a universal agonist for levels of cytokine induction in DF1 cells. In HD11 macrophage- which a mammalian TLR is yet to be described. It is possible that like cells, CAcTIR15 induced cytokine mRNA upregulation to another mammalian TLR or a completely different type of innate levels comparable with CAcTIR4 and then CAmTIR4. Members immune receptor may fulfill the role of TLR15 in mammals. The of the TLR1 family form heterodimers with TLR2 to recognize diversification of TLR gene families, exemplified by the restricted different microbial lipopeptides (3). The phylogenetic isolation of lineage distribution of TLR15, illustrates pathogen-driven evolu- TLR15 raised the possibility that this TLR was capable of agonist tionary pressure at the basic level of pattern recognition. TLR15 recognition without a heterodimeric partner. The data presented in appears to be another instance where these forces have resulted in this report indicate that this is indeed the case. a unique molecular solution in the avian/reptilian lineage. Transfection of full-length TLR15 into HEK293 cells was used to screen defined agonists and microbial preparation as potential Acknowledgments sources of TLR15 agonists. Exposure of transfected cells to a wide We gratefully acknowledge Paul Sopp (Institute for Animal Health) for per- range of defined agonists and microbial preparations of viral or forming FACS for this project and Media Services for provision of cell lines. Downloaded from bacterial origin did not induce NF-kB induction. The only mi- We also thank Prof. Venugopal Nair (Institute for Animal Health), Prof. Paul crobial lysates capable of inducing significant reporter activation Barrow (University of Nottingham), Dr. Nigel Saunders, Dr. David Greaves, in a TLR15-dependent manner were those derived from yeasts. Dr. Stephen Kearsey, and Marianne Shepherd (University of Oxford) for The fact that all the yeast whole-cell lysates induced NF-kB in- providing microbial strains. duction in a TLR15-dependent manner suggests that the specific Disclosures agonist for TLR15 is a conserved PAMP. TLR15-dependent NF- http://www.jimmunol.org/ kB induction was more pronounced with lysates derived from The authors have no financial conflicts of interest. C. albicans or S. cerevisiae than with S. pombe, which may indi- cate differences in agonist level or structure between the different References groups of yeast. Species-specific differences in cell wall compo- 1. Roach, J. C., G. Glusman, L. Rowen, A. Kaur, M. K. Purcell, K. D. Smith, sition are well documented in yeast (44), and these may contribute L. E. Hood, and A. Aderem. 2005. The evolution of vertebrate Toll-like receptors. Proc. Natl. Acad. Sci. USA 102: 9577–9582. to the different magnitude of the response. Nonetheless, the finding 2. Chuang, T., and R. J. Ulevitch. 2001. Identification of hTLR10: a novel human that TLR15 responds to yeast-derived agonists is functionally sig- Toll-like receptor preferentially expressed in immune cells. Biochim. Biophys. nificant. The TLR15 stimulatory activity in the yeast lysates was Acta 1518: 157–161. 3. Ozinsky, A., D. M. Underhill, J. D. Fontenot, A. M. Hajjar, K. D. Smith, by guest on September 27, 2021 heat labile and inhibited by PMSF, supporting the recent findings of C. B. Wilson, L. Schroeder, and A. Aderem. 2000. The repertoire for pattern de Zoete et al. (42) with culture supernatants. recognition of pathogens by the innate immune system is defined by cooperation Mammalian TLR2 and Dectin-1 are responsible for the cellular between toll-like receptors. Proc. Natl. Acad. Sci. USA 97: 13766–13771. 4. Alexopoulou, L., A. C. Holt, R. Medzhitov, and R. A. Flavell. 2001. Recognition response to zymosan, a complex cell-wall preparation from yeast of double-stranded RNA and activation of NF-kappaB by Toll-like receptor 3. that contains lipopeptides, b-glucans, mannans, mannoproteins, Nature 413: 732–738. 5. Poltorak, A., X. He, I. Smirnova, M. Y. Liu, C. Van Huffel, X. Du, D. Birdwell, and chitin (40, 41). The TLR2 simulating properties of zymosan E. Alejos, M. Silva, C. Galanos, et al. 1998. Defective LPS signaling in C3H/HeJ were attributable to cell-wall lipopeptides, as treatment of zy- and C57BL/10ScCr mice: mutations in Tlr4 gene. Science 282: 2085–2088. mosan with organic solvents abrogated TLR2-mediated signaling 6. Hayashi, F., K. D. Smith, A. Ozinsky, T. R. Hawn, E. C. Yi, D. R. Goodlett, J. K. Eng, S. Akira, D. M. Underhill, and A. Aderem. 2001. The innate immune but did not affect signaling via Dectin-1 (45). TLR15 is most response to bacterial flagellin is mediated by Toll-like receptor 5. Nature 410: closely related to the TLR1 family, comprising TLR1, TLR2, 1099–1103. TLR6, and TLR10 (1), and TLR15 was activated by yeast-derived 7. Diebold, S. S., T. Kaisho, H. Hemmi, S. Akira, and C. Reis e Sousa. 2004. Innate antiviral responses by means of TLR7-mediated recognition of single-stranded agonists. This prompted consideration of zymosan as an agonist RNA. Science 303: 1529–1531. source for TLR15. A range of concentrations of zymosan were 8. Hemmi, H., O. Takeuchi, T. Kawai, T. Kaisho, S. Sato, H. Sanjo, M. Matsumoto, k K. Hoshino, H. Wagner, K. Takeda, and S. Akira. 2000. A Toll-like receptor not sufficient to elicit NF- B activation in TLR15-transfected recognizes bacterial DNA. Nature 408: 740–745. HEK293 cells. These data suggest that the specific agonist for 9. Heil, F., H. Hemmi, H. Hochrein, F. Ampenberger, C. Kirschning, S. Akira, TLR15 may be a conserved yeast molecule that is either not G. Lipford, H. Wagner, and S. Bauer. 2004. Species-specific recognition of single-stranded RNA via toll-like receptor 7 and 8. Science 303: 1526–1529. present in the cell fraction from which zymosan is derived or lost 10. Lund, J. M., L. Alexopoulou, A. Sato, M. Karow, N. C. Adams, N. W. Gale, during the purification process. Yeast cell-wall–associated chitin A. Iwasaki, and R. A. Flavell. 2004. Recognition of single-stranded RNA viruses has also been shown to activate murine macrophages in a TLR2- by Toll-like receptor 7. Proc. Natl. Acad. Sci. USA 101: 5598–5603. 11. Zhang, D., G. Zhang, M. S. Hayden, M. B. Greenblatt, C. Bussey, R. A. Flavell, and MyD88-dependent manner (46). Because chitin is present in and S. Ghosh. 2004. A toll-like receptor that prevents infection by uropathogenic zymosan preparations, it is unlikely that chitin is an agonist for bacteria. Science 303: 1522–1526. 12. Yarovinsky, F., D. Zhang, J. F. Andersen, G. L. Bannenberg, C. N. Serhan, TLR15. Intriguingly, a recent report implicates proteolytic cleav- M. S. Hayden, S. Hieny, F. S. Sutterwala, R. A. Flavell, S. Ghosh, and A. Sher. age by bacterial and yeast-secreted proteases as a mechanism of 2005. TLR11 activation of dendritic cells by a protozoan profilin-like protein. activation for TLR15 (42). Our data support the hypothesis that Science 308: 1626–1629. 13. Matsuo, A., H. Oshiumi, T. Tsujita, H. Mitani, H. Kasai, M. Yoshimizu, whole-cell lysates from yeast contain a potent enzymatically active M. Matsumoto, and T. Seya. 2008. TLR22 recognizes RNA duplex to TLR15 agonist because the effect is abrogated by PMSF or heat induce IFN and protect cells from birnaviruses. J. Immunol. 181: 3474–3485. treatment. Importantly, we also demonstrate that yeast lysates ac- 14. Brownlie, R., J. Zhu, B. Allan, G. K. Mutwiri, L. A. Babiuk, A. Potter, and + P. Griebel. 2009. Chicken TLR21 acts as a functional homologue to mammalian tivate proinflammatory cytokine mRNA production in the TLR15 TLR9 in the recognition of CpG oligodeoxynucleotides. Mol. Immunol. 46: chicken macrophage cell line, HD11. Although pathogen prepa- 3163–3170. 15. Keestra, A. M., M. R. de Zoete, L. I. Bouwman, and J. P. van Putten. 2010. rations would normally include a variety of PAMPs (stimulating a Chicken TLR21 is an innate CpG DNA receptor distinct from mammalian range of TLRs), the heat and PMSF sensitivity of the HD11 re- TLR9. J. Immunol. 185: 460–467. The Journal of Immunology 9

16. Keestra, A. M., and J. P. van Putten. 2008. Unique properties of the chicken 32. Kaiser, P., L. Rothwell, E. E. Galyov, P. A. Barrow, J. Burnside, and P. Wigley. TLR4/MD-2 complex: selective activation of the MyD88- 2000. Differential cytokine expression in avian cells in response to invasion by dependent pathway. J. Immunol. 181: 4354–4362. Salmonella typhimurium, Salmonella enteritidis and Salmonella gallinarum. 17. Philbin, V. J., M. Iqbal, Y. Boyd, M. J. Goodchild, R. K. Beal, N. Bumstead, Microbiology 146: 3217–3226. J. Young, and A. L. Smith. 2005. Identification and characterization of a func- 33. Poh, T. Y., J. Pease, J. R. Young, N. Bumstead, and P. Kaiser. 2008. Re- tional, alternatively spliced Toll-like receptor 7 (TLR7) and genomic disruption evaluation of chicken CXCR1 determines the true gene structure: CXCLi1 of TLR8 in chickens. Immunology 114: 507–521. (K60) and CXCLi2 (CAF/interleukin-8) are ligands for this receptor. J. Biol. 18. Schwarz, H., K. Schneider, A. Ohnemus, M. Lavric, S. Kothlow, S. Bauer, Chem. 283: 16408–16415. B. Kaspers, and P. Staeheli. 2007. Chicken toll-like receptor 3 recognizes its 34. Wu, Z., L. Rothwell, J. R. Young, J. Kaufman, C. Butter, and P. Kaiser. 2010. cognate ligand when ectopically expressed in human cells. J. Interferon Cytokine Generation and characterization of chicken bone marrow-derived dendritic cells. Res. 27: 97–101. Immunology 129: 133–145. 19. Leveque, G., V. Forgetta, S. Morroll, A. L. Smith, N. Bumstead, P. Barrow, 35. Medzhitov, R., P. Preston-Hurlburt, and C. A. Janeway, Jr. 1997. A human ho- J. C. Loredo-Osti, K. Morgan, and D. Malo. 2003. Allelic variation in TLR4 is mologue of the Drosophila Toll protein signals activation of adaptive immunity. linked to susceptibility to Salmonella enterica serovar Typhimurium infection in Nature 388: 394–397. chickens. Infect. Immun. 71: 1116–1124. 36. Beug, H., A. von Kirchbach, G. Do¨derlein, J. F. Conscience, and T. Graf. 1979. 20. Kogut, M. H., H. He, and P. Kaiser. 2005b. Lipopolysaccharide binding protein/ Chicken hematopoietic cells transformed by seven strains of defective avian CD14/ TLR4-dependent recognition of salmonella LPS induces the functional leukemia viruses display three distinct phenotypes of differentiation. Cell 18: activation of chicken heterophils and up-regulation of pro-inflammatory cytokine 375–390. and chemokine gene expression in these cells. Anim. Biotechnol. 16: 165–181. 37. Benton, M. J., and P. C. Donoghue. 2007. Paleontological evidence to date the 21. Boyd, A., V. J. Philbin, and A. L. Smith. 2007. Conserved and distinct aspects of tree of life. Mol. Biol. Evol. 24: 26–53. the avian Toll-like receptor (TLR) system: implications for transmission and 38.Gewirtz,A.T.,T.A.Navas,S.Lyons,P.J.Godowski,andJ.L.Madara.2001. control of bird-borne zoonoses. Biochem. Soc. Trans. 35: 1504–1507. Cutting edge: bacterial flagellin activates basolaterally expressed TLR5 to 22. Fukui, A., N. Inoue, M. Matsumoto, M. Nomura, K. Yamada, Y. Matsuda, induce epithelial proinflammatory gene expression. J. Immunol. 167: 1882– K. Toyoshima, and T. Seya. 2001. Molecular cloning and functional character- 1885. ization of chicken toll-like receptors. A single chicken toll covers multiple 39. Simon, R., and C. E. Samuel. 2007. Activation of NF-kappaB-dependent gene molecular patterns. J. Biol. Chem. 276: 47143–47149. expression by Salmonella flagellins FliC and FljB. Biochem. Biophys. Res. Downloaded from 23. Higuchi, M., A. Matsuo, M. Shingai, K. Shida, A. Ishii, K. Funami, Y. Suzuki, Commun. 355: 280–285. H. Oshiumi, M. Matsumoto, and T. Seya. 2008. Combinational recognition of 40. Brown, G. D., J. Herre, D. L. Williams, J. A. Willment, A. S. Marshall, and bacterial lipoproteins and peptidoglycan by chicken Toll-like receptor 2 sub- S. Gordon. 2003. Dectin-1 mediates the biological effects of beta-glucans. J. family. Dev. Comp. Immunol. 32: 147–155. Exp. Med. 197: 1119–1124. 24. Keestra, A. M., M. R. de Zoete, R. A. van Aubel, and J. P. van Putten. 2007. The 41. Gantner, B. N., R. M. Simmons, S. J. Canavera, S. Akira, and D. M. Underhill. central leucine-rich repeat region of chicken TLR16 dictates unique ligand spec- 2003. Collaborative induction of inflammatory responses by dectin-1 and Toll- ificity and species-specific interaction with TLR2. J. Immunol. 178: 7110–7119. like receptor 2. J. Exp. Med. 197: 1107–1117.

25. Cormican, P., A. T. Lloyd, T. Downing, S. J. Connell, D. Bradley, and 42. de Zoete, M. R., L. I. Bouwman, A. M. Keestra, and J. P. van Putten. 2011. http://www.jimmunol.org/ C. O’Farrelly. 2009. The avian Toll-Like receptor pathway—subtle differences and activation of a Toll-like receptor by microbial proteases. Proc. amidst general conformity. Dev. Comp. Immunol. 33: 967–973. Natl. Acad. Sci. USA 108: 4968–4973. 26. MacDonald, M. R., J. Xia, A. L. Smith, and K. E. Magor. 2008. The duck toll 43. Werling, D., O. C. Jann, V. Offord, E. J. Glass, and T. J. Coffey. 2009. Variation like receptor 7: genomic organization, expression and function. Mol. Immunol. matters: TLR structure and species-specific pathogen recognition. Trends 45: 2055–2061. Immunol. 30: 124–130. 27. Larkin, M. A., G. Blackshields, N. P. Brown, R. Chenna, P. A. McGettigan, 44. Xie, X., and P. N. Lipke. 2010. On the evolution of fungal and yeast cell walls. H. McWilliam, F. Valentin, I. M. Wallace, A. Wilm, R. Lopez, et al. 2007. Yeast 27: 479–488. Clustal W and Clustal X version 2.0. Bioinformatics 23: 2947–2948. 45. Ikeda, Y., Y. Adachi, T. Ishii, N. Miura, H. Tamura, and N. Ohno. 2008. Dis- 28. Hall, T. A. 1999. BioEdit: a user-friendly biological sequence alignment editor sociation of Toll-like receptor 2-mediated innate immune response to Zymosan and analysis program for Windows 95/98/NT. Nucl. Acids Symp. Ser. 41: 95–98. by organic solvent-treatment without loss of Dectin-1 reactivity. Biol. Pharm. 29. Tamura, K., and M. Nei. 1993. Estimation of the number of nucleotide sub- Bull. 31: 13–18.

stitutions in the control region of mitochondrial DNA in and chim- 46. Da Silva, C. A., D. Hartl, W. Liu, C. G. Lee, and J. A. Elias. 2008. TLR-2 and by guest on September 27, 2021 panzees. Mol. Biol. Evol. 10: 512–526. IL-17A in chitin-induced macrophage activation and acute inflammation. J. 30. Tajima, F., and M. Nei. 1984. Estimation of evolutionary distance between nu- Immunol. 181: 4279–4286. cleotide sequences. Mol. Biol. Evol. 1: 269–285. 47. Slack, J. L., K. Schooley, T. P. Bonnert, J. L. Mitcham, E. E. Qwarnstrom, 31. Wigley, P., S. D. Hulme, N. Bumstead, and P. A. Barrow. 2002. In vivo and J. E. Sims, and S. K. Dower. 2000. Identification of two major sites in the type I in vitro studies of genetic resistance to systemic salmonellosis in the chicken interleukin-1 receptor cytoplasmic region responsible for coupling to pro- encoded by the SAL1 locus. Microbes Infect. 4: 1111–1120. inflammatory signaling pathways. J. Biol. Chem. 275: 4670–4678.