This mini-review focuses on recent research with avian TLRs (Toll-like receptors), highlighting shared and distinct features compared with the more intensively studied mammalian TLR. These include the avian TLR repertoire and the response to various agonists. Studies with avian TLR can be applied to development of new approaches to control diseases of birds and is especially relevant to bird-borne zoonoses including avian influenza, Salmonella and Campylobacter.

Introduction

The TLR (Toll-like receptor) family is a highly conserved group of molecules that plays a major role in pathogen detection and initiation/regulation of immune responses. These receptors have been described in many animal phyla and the genomes of most vertebrate species encode between 10 and 13 different TLRs [1]. Comparative studies will help define the evolution of vertebrate TLR in the context of pathogen challenge and provide the potential to develop host-species-tailored control strategies. We will describe the nature of the avian TLR repertoire and function that may be relevant to the transmission of infection to humans.

The avian TLR repertoire

In evolutionary terms, avian and mammalian lineages diverged from a common ancestor approx. 300 million years ago, and the avian TLR repertoire comprises both orthologous and distinct TLR genes. To date, ten avian TLRs have been described and there are clear orthologues of mammalian TLR3, 4, 5 and 7, encoded within genomic regions with conserved synteny [15].

The chicken genome encodes at least one (probably two) members of the TLR1/6/10 family [58] and two tandemly arranged genes encoding TLR2 receptors [9,10]. In humans, TLR1 and 6 have been shown to form heterodimers with TLR2 which influence agonist reactivity; the configuration of these receptors in chickens may also serve a similar purpose (see below). Interestingly, there are remnants of a second disrupted TLR2-like gene in tandem with the functional TLR2 gene in mice and humans. Hence, the duplication of TLR2 occurred prior to the divergence of mammals and birds and was lost as a functional gene in the mammalian lineage.

The TLR8 gene in chickens is fragmented and disrupted by a CR1 (chicken repeat-1) retrovirus-like element; this feature was detected in galliform bird species but not in non-galliform bird species [4]. The galliforms diverged from other birds approx. 90 million years ago and a lineage-specific deletion of TLR8 may contribute to differences in viral susceptibility. Although an attractive and topical hypothesis (especially in the context of avian influenza), our analysis did not determine the status of TLR8 in non-galliform birds and this is currently under scrutiny (M. MacDonald, K. Magor and A.L. Smith, unpublished work). There is no evidence for a chTLR9 (chicken TLR9) orthologue and, if present, it is not located in an area of conserved synteny with mammals. However, chicken cells respond to unmethylated CpG motifs that stimulate mammalian TLR9 (e.g. [11]) and TLR9 can be identified in Xenopus [1]. Hence, either an avian TLR9 remains to be defined or a different receptor is involved in CpG recognition. Orthologues of murine TLR11, 12 or 13 cannot be clearly identified in the chicken genome.

Genes encoding two distinct chTLR have also been identified within the chicken genome ([1,12], and V.J. Philbin, A. Boyd and A.L. Smith, unpublished work). One of these, chTLR15 is not closely related to any known vertebrate TLR [1,12]. The other distinct chTLR shares homology with the fish/amphibian TLR21–TLR23 family [1] although the preliminary identification as chTLR21 (implying orthologous status with fish/amphibian TLR21) requires experimental confirmation.

The level of amino-acid-sequence identity between specific chTLRs and their mammalian counterparts is high (Table 1) compared with many other immune-related genes shared by these species. The sequence-based analyses can be supplemented by prediction of structural organization, especially LRR (leucine-rich repeat) distribution (Figure 1).

Table 1
Percentage of amino acid identity between vertebrate TLRs

Comparisons made between the entire TLR amino acid sequence (A) or between the TIR (Toll/IL-1 receptor) homologous domains (B). Fu indicates Takifugu rubipres. Ch indicates chicken (Gallusgallus). Shading indicates orthologue or family member. (A)

 Human TLR entire protein 
 TLR1 TLR2 TLR3 TLR4 TLR5 TLR6 TLR7 TLR8 TLR9 TLR10 FuTLR21 
chTLR1.1 49 28 19 23 21 45 20 21 22 44 24 
chTLR1.2 47.1 29 19 21 18 41 18 19 18 42 14 
chTLR2.1 29 50 20 24 27 28 25 23 21 29 28 
chTLR2.2 29 50 20 24 21 27 24 23 22 28 30 
chTLR3 22 23 58 24 23 23 24 24 19 22 24 
chTLR4 21 26 22 44 24 23 24 23 24 20 22 
chTLR5 21 21 23 22 50 20 22 22 22 21 25 
chTLR7 23 25 24 20 22 24 62 40 35 24 25 
chTLR15 32 32 20 22 25 32 27 27 24 28 26 
chTLR21 28 28 23 25 25 27 26 23 25 25 43 
(B) 
 Human TLR TIR domain 
 TLR1 TLR2 TLR3 TLR4 TLR5 TLR6 TLR7 TLR8 TLR9 TLR10 FuTLR21 
chTLR1.1 82.4 48 21 37 24 70 27 29 27 66 42 
chTLR1.2 81.2 57 28 41 27 81 28 34 28 76 46 
chTLR2.1 52 81 27 38 29 52 38 36 34 50 43 
chTLR2.2 52 81 27 38 29 52 38 36 34 50 43 
chTLR3 28 28 67 30 28 27 31 32 29 26 28 
chTLR4 34 37 29 60 30 33 31 29 28 31 34 
chTLR5 31 31 28 32 70 28 30 32 32 28 38 
chTLR7 34 39 28 30 25 33 72 56 47 34 36 
chTLR15 45 40 26 29 35 44 34 29 25 39 39 
chTLR21 38 42 26 40 36 37 34 36 39 35 63 
 Human TLR entire protein 
 TLR1 TLR2 TLR3 TLR4 TLR5 TLR6 TLR7 TLR8 TLR9 TLR10 FuTLR21 
chTLR1.1 49 28 19 23 21 45 20 21 22 44 24 
chTLR1.2 47.1 29 19 21 18 41 18 19 18 42 14 
chTLR2.1 29 50 20 24 27 28 25 23 21 29 28 
chTLR2.2 29 50 20 24 21 27 24 23 22 28 30 
chTLR3 22 23 58 24 23 23 24 24 19 22 24 
chTLR4 21 26 22 44 24 23 24 23 24 20 22 
chTLR5 21 21 23 22 50 20 22 22 22 21 25 
chTLR7 23 25 24 20 22 24 62 40 35 24 25 
chTLR15 32 32 20 22 25 32 27 27 24 28 26 
chTLR21 28 28 23 25 25 27 26 23 25 25 43 
(B) 
 Human TLR TIR domain 
 TLR1 TLR2 TLR3 TLR4 TLR5 TLR6 TLR7 TLR8 TLR9 TLR10 FuTLR21 
chTLR1.1 82.4 48 21 37 24 70 27 29 27 66 42 
chTLR1.2 81.2 57 28 41 27 81 28 34 28 76 46 
chTLR2.1 52 81 27 38 29 52 38 36 34 50 43 
chTLR2.2 52 81 27 38 29 52 38 36 34 50 43 
chTLR3 28 28 67 30 28 27 31 32 29 26 28 
chTLR4 34 37 29 60 30 33 31 29 28 31 34 
chTLR5 31 31 28 32 70 28 30 32 32 28 38 
chTLR7 34 39 28 30 25 33 72 56 47 34 36 
chTLR15 45 40 26 29 35 44 34 29 25 39 39 
chTLR21 38 42 26 40 36 37 34 36 39 35 63 

Predicted domain organization of chicken TLR

Figure 1
Predicted domain organization of chicken TLR

Primary amino acid sequence of the chTLR was subjected to domain prediction [SMART (simple modular architecture research tool); http://smart.embl-heidelberg.de] and results were confirmed by manual analysis. For comparison, similar analyses were performed with human TLR1–TLR10 and with Takifugu rubipres (fu) TLR21–TLR23.

Figure 1
Predicted domain organization of chicken TLR

Primary amino acid sequence of the chTLR was subjected to domain prediction [SMART (simple modular architecture research tool); http://smart.embl-heidelberg.de] and results were confirmed by manual analysis. For comparison, similar analyses were performed with human TLR1–TLR10 and with Takifugu rubipres (fu) TLR21–TLR23.

Biology of chTLR

Many studies have demonstrated that chicken cells respond to exogenous addition of pathogen components with agonist activity for mammalian TLR. The distribution of chTLR mRNA in isolated cell populations and tissues (e.g. [7,10,1214]) is broadly similar to that described for mammals. Determination of the TLR repertoire expressed by different cell types has allowed more rational description of the cellular responses induced after exposure to pathogens or their components. Ectopic expression of chTLR in mammalian cells, such as HEK-293 cells (human embryonic kidney cells), has been used to confirm chTLR agonist specificities for chTLR1/6/10-like molecules, chTLR2 (2.1 and 2.2), chTLR3 and chTLR7 ([6,8,10,15], and V.J. Philbin, A. Boyd and A.L. Smith, unpublished work).

The two chTLR2 molecules exhibit different capacities to respond to mammalian TLR2 agonists [6,8,10]. In mammals, TLR1–TLR2 or TLR6–TLR2 heterodimers enhance responses and confer agonist specificity on diacylated and triacylated lipopeptides. Keestra et al. [8] demonstrated that one chTLR1/6/10-like molecule (termed TLR16) enhanced the ability of chTLR2.2 to respond to both diacylated and triacylated agonists. Creation of chimaeric receptors with exchange of chTLR1/6/10 LRR 6–16 into human TLR6 conferred dual specificity and allowed species-specific interactions with chTLR2.2. In contrast, Higuchi et al. [6] described two chTLR1/6/10 molecules and reported that a combination of TLR1.1 or TLR1.2 with either TLR2.1 or TLR2.2 afforded specificity to the agonist response. Based on sequence similarity and response profile, the Keestra TLR16 and the Higushi TLR1.1 represent the same molecule.

The agonist activity of double-stranded RNA [poly(I:C)] for chTLR3 has recently been demonstrated by transfection into HEK-293 cells [15]. Similarly, we have demonstrated activation of chTLR7-transfected HEK-293 cells with R848 (V.J. Philbin and A.L. Smith, unpublished work). Exposure of chicken splenocytes to TLR7/8 agonists induces strong up-regulation of IL-1β (interleukin 1β), IL-6 and IL-8 but fails to induce the up-regulation of IFNα (interferon α) or IFNβ mRNA [4]. Transfection of mouse or chicken TLR7 into cultured chicken kidney cells leads to agonist-dependent up-regulation of IL-1β but not IFNα or IFNβ mRNA. In contrast, chTLR7 is capable of inducing NF-κB (nuclear factor κB) reporter systems and IFNβ mRNA in HEK-293 cells, suggesting a difference in signalling pathways between birds and mammals. However, TLR7-agonist-induced induction of type I IFN has been reported in chicken cells as measured by antiviral or Mx reporter assays (e.g. [15]). Also, heterophils (avian polymorphonuclear cells) isolated from two chicken lines exhibited differential responsiveness to R848 exposure, indicating the potential for genetic polymorphism in this pathway [16].

Transfection-based studies are yet to be reported with the other chTLRs, but there is evidence for TLR4–LPS (lipopolysaccharide) and TLR5–flagellin interactions (e.g. [2,17]). The ability of chicken cells to respond to unmethylated CpG motifs is also well established (e.g. [13]) but, as indicated above, there is no evidence for a chTLR9 orthologue.

At present, in vivo analysis of chTLR function is limited, although two reports with Salmonella enterica provide evidence that TLR interactions affect host–pathogen biology. The chTLR4 gene is polymorphic (as seen with human TLR4) and a genetic mapping approach was used to identify this locus as contributing to the ability to resist the effects of systemic salmonellosis in young chicks [3]. The studies on chTLR5 were driven by the observation that flagella status of S. enterica serovars correlated with ability to cause serious systemic disease. Flagellated serovars Typhimurium and Enteritidis are largely restricted to the gut compared with aflagellate Pullorum and Gallinarum, which cause typhoid-like disease. An aflagellate Salmonella Typhimurium mutant (administered orally) established in systemic tissues more efficiently than the parental, flagella intact strain [2]. The increased translocation was associated with reduced inflammation and cytokine production in the gut, suggesting a role for TLR5–flagellin interactions in retention of flagellated Salmonella in the gut.

Conclusions

Studies of avian TLR biology are at an early phase but it is already clear that there are significant differences from mammals, especially with the TLR repertoire and cellular response to agonist exposure. These studies will continue to have an impact on our understanding of the evolution of vertebrate TLR function and avian host–pathogen biology. There are clear opportunities to develop species-specific adjuvants and pathogen control strategies that will impact on animal and human health.

Pattern-Recognition Receptors in Human Disease: A Biochemical Society Focused Meeting held at Queens' College, University of Cambridge, Cambridge, U.K., 8–10 August 2007. Organized and Edited by C. Bryant (Cambridge, U.K.), K. Fitzgerald (University of Massachusetts Medical School, U.S.A.), N. Gay (Cambridge, U.K.), P. Morley (GlaxoSmithKline, U.K.) and L. O'Neill (Trinity College Dublin, Ireland).

Abbreviations

     
  • HEK-293

    cell, human embryonic kidney cell

  •  
  • IFNα

    interferon α

  •  
  • IL

    interleukin

  •  
  • LRR

    leucine-rich repeat

  •  
  • TLR

    Toll-like receptor

  •  
  • chTLR

    chicken TLR

We thank current and past members of the Enteric Immunology Group and our collaborators (particularly Professor P. Barrow, Dr J. Young, Dr P. Kaiser, Dr M. Iqbal, Dr M. Kogut, Dr D. Kapczynski and Professor K. Magor) for many discussions and contributions. We also acknowledge financial support from BBSRC (Biotechnology and Biological Sciences Research Council), DEFRA (Department for Environment, Food and Rural Affairs) and the Jenner Institute.

References

References
1
Roach
J.C.
Glusman
G.
Rowen
L.
Kaur
A.
Purcell
M.K.
Smith
K.D.
Hood
L.E.
Aderem
A.
Proc. Natl. Acad. Sci. U.S.A.
2005
, vol. 
102
 (pg. 
9577
-
9582
)
2
Iqbal
M.
Philbin
V.J.
Withanage
G.S.
Wigley
P.
Beal
R.K.
Goodchild
M.J.
Barrow
P.
McConnell
I.
Maskell
D.J.
Young
J.
, et al. 
Infect. Immun.
2005
, vol. 
73
 (pg. 
2344
-
2350
)
3
Leveque
G.
Forgetta
V.
Morroll
S.
Smith
A.L.
Bumstead
N.
Barrow
P.
Loredo-Osti
J.C.
Morgan
K.
Malo
D.
Infect. Immun.
2003
, vol. 
71
 (pg. 
1116
-
1124
)
4
Philbin
V.J.
Iqbal
M.
Boyd
Y.
Goodchild
M.J.
Beal
R.K.
Bumstead
N.
Young
J.
Smith
A.L.
Immunology
2005
, vol. 
114
 (pg. 
507
-
521
)
5
Yilmaz
A.
Shen
S.
Adelson
D.L.
Xavier
S.
Zhu
J.J.
Immunogenetics
2005
, vol. 
56
 (pg. 
743
-
753
)
6
Higuchi
M.
Matsuo
A.
Shingai
M.
Shida
K.
Ishii
A.
Funami
K.
Suzuki
Y.
Oshiumi
H.
Matsumoto
M.
Seya
T.
Dev. Comp. Immunol.
2007
 
7
Iqbal
M.
Philbin
V.J.
Smith
A.L.
Vet. Immunol. Immunopathol.
2005
, vol. 
104
 (pg. 
117
-
127
)
8
Keestra
A.M.
de Zoete
M.R.
van Aubel
R.A.
van Putten
J.P.
J. Immunol.
2007
, vol. 
178
 (pg. 
7110
-
7119
)
9
Boyd
Y.
Goodchild
M.
Morroll
S.
Bumstead
N.
Immunogenetics
2001
, vol. 
52
 (pg. 
294
-
298
)
10
Fukui
A.
Inoue
N.
Matsumoto
M.
Nomura
M.
Yamada
K.
Matsuda
Y.
Toyoshima
K.
Seya
T.
J. Biol. Chem.
2001
, vol. 
276
 (pg. 
47143
-
47149
)
11
Vleugels
B.
Ververken
C.
Goddeeris
B.M.
Poult. Sci.
2002
, vol. 
81
 (pg. 
1317
-
1321
)
12
Higgs
R.
Cormican
P.
Cahalane
S.
Allan
B.
Lloyd
A.T.
Meade
K.
James
T.
Lynn
D.J.
Babiuk
L.A.
O'Farrelly
C.
Infect. Immun.
2006
, vol. 
74
 (pg. 
1692
-
1698
)
13
He
H.
Genovese
K.J.
Nisbet
D.J.
Kogut
M.H.
Mol. Immunol.
2006
, vol. 
43
 (pg. 
783
-
789
)
14
Kogut
M.H.
Iqbal
M.
He
H.
Philbin
V.
Kaiser
P.
Smith
A.
Dev. Comp. Immunol.
2005
, vol. 
29
 (pg. 
791
-
807
)
15
Schwarz
H.
Schneider
K.
Ohnemus
A.
Lavric
M.
Kothlow
S.
Bauer
S.
Kaspers
B.
Staeheli
P.
J. Interferon Cytokine Res.
2007
, vol. 
27
 (pg. 
97
-
101
)
16
Kogut
M.H.
Swaggerty
C.
He
H.
Pevzner
I.
Kaiser
P.
Microbes Infect.
2006
, vol. 
8
 (pg. 
1866
-
1874
)
17
Dil
N.
Qureshi
M.A.
Vet. Immunol. Immunopathol.
2002
, vol. 
84
 (pg. 
191
-
207
)

Author notes

1

Present address: Harvard Medical School and Department of Medicine, Division of Infectious Diseases, Children's Hospital, 300 Longwood Avenue, Boston, MA 02115, U.S.A.