Background
Toll-like receptors (TLRs) are a family of phylogenetically conserved transmembrane pattern recognition receptors involved in innate immunity and inflammation. To date, 10 functional TLRs have been reported in humans [
1] which recognise unique and highly conserved molecular signatures from microbes known as
pathogen associated molecular patterns (PAMPs) that include lipopolysaccharides (LPS), flagellin, lipopeptides, lipotechoic acid (LTA), microbial DNA, viral RNAs and others [
1].
TLRs have been implicated in ocular inflammation. For example, activation of TLRs by PAMPs due to an initiating mucosal infection and the subsequent immune response has been hypothesised to play a key role in the pathogenesis of anterior uveitis [
2]. In addition, expression of TLR2 in human conjunctival epithelial cells was shown to play a significant role in the chronic ocular inflammatory response to
Staphylococcus aureus[
3]. Kezic et al. [
4] suggested that both epithelial cells and immune cells play a role in ocular inflammation. Specifically, radiation-resistant, non-bone marrow derived ocular cells, such as iris endothelial cells or nonpigmented ciliary body epithelial cells, play a greater role in the development of endotoxin-induced uveitis (EIU) than bone marrow-derived macrophages and dendritic cells residing in the eye [
4]. Therefore, studying TLRs on iris pigment epithelial cells (IPE) and their response to PAMPs may provide an insight into pathogenesis of ocular inflammation, particularly anterior uveitis.
The iris pigment epithelium is the layer of pigmented cells forming the posterior layer of the iris [
5]. There is a remarkable resemblance between IPE and retinal pigment epithelial cells (RPE) due to their shared embryonic development [
6‐
8].
In vitro, IPE and RPE share functional properties such as phagocytosis and synthesis of cytokines and growth factors [
7,
9]. Rezai and colleagues showed that IPE elicited phagocytic activity similar to RPE [
7]. Non-immune cells, such as IPE and RPE, form an interface between the eye and the environment that is not readily accessible to myeloid cells. By virtue of their ability to detect signals via innate immune receptors, such as toll-like receptors, they are able to recruit myeloid cells, such as neutrophils and macrophages to the site of injury and induce inflammation.
Expression of TLRs has been reported in a number of ocular tissues such as cornea, conjunctiva, sclera and retina [
10‐
14]. Studies have emphasized the importance of the LPS receptor complex (TLR4 and co-receptors CD14 and MD2) expression in ocular tissues and cells such as corneal epithelial cells, cornea stroma fibroblasts, human ciliary body, human iris endothelial cells (TLR4 only), RPE and resident antigen presenting cells in human uvea [
14‐
18]. It has been shown that human RPE express TLRs and are considered to play an important role in posterior ocular inflammation due to their ability to secrete several inflammatory mediators [
13,
19]. However, little is known about the distribution of TLRs in the uvea, especially the iris. In this study, we investigated the expression pattern and functional significance of TLRs in human ocular pigment epithelial cells (IPE and RPE). This study demonstrated that human IPE and RPE secrete IL-8 and MCP-1 in response to PAMPs, which was partially mediated through TLR activation.
Discussion
This is the first report to characterise the expression profile of TLRs in human IPE. IPE and RPE expressed TLR1-6 and 8–10 transcripts; TLR1-6 and 9 proteins (see Additional file
2) however TLR7, 8 and 10 proteins were not detected. Real-time PCR was performed to further confirm whether there is differential expression of TLR2, −3, −4 and −6 genes in both cell types (Figure
1). Additionally, IPE and RPE expressed higher TLR4 levels than TLR2 and this may account for its higher responsiveness to LPS than Pam
3CSK
4.3HCl stimulations (Figures
1 and
2). IPE may have been more responsive to LPS stimulation than RPE due to the fact that they expressed higher TLR4 levels than RPE (Figures
1 and
2). The level of TLR3 mRNA and proteins was not significantly different between the two cell types under basal medium (Figure
1 A). However, IPE and RPE showed different sensitivity in response to Poly(I:C) stimulation. In addition, both cells showed higher expression of TLR4 mRNA and proteins than that of TLR3. They secreted higher level of IL-8 in response to Poly(I:C), compared to LPS stimulation (Figure
2). It is noteworthy that the experiments conducted for Figure
1 were performed with basal medium. Hence, further study is needed to investigate the level of TLR2, −3, −4, −6 mRNA and proteins following ligand stimulation. Nevertheless the presence of TLR proteins detected by Western blotting may not necessarily reflect on the amount of the proteins involved in the signalling. Some of the proteins may be degraded or undergo post-translational modification for functioning. Interestingly, TLR7 mRNA was not detected in either cell type and this may well explain the absence of TLR7 proteins. The lack of TLR8 and 10 proteins in ocular pigment epithelial cells may reflect the low expression levels, which in turn may influence their response to PAMPs.
IPE express TLR4
in vitro consistent with our earlier findings that these cells express a functional LPS receptor complex (TLR4, MD-2, and CD14)
in vitro and secreted several pro-inflammatory cytokines including IL-6, IL-8, MCP-1, IP-10, MIP-1-beta and RANTES in response to LPS stimulation [
23]. Here, we showed that RPE expressed all TLR transcripts except TLR7. In contrast, Kumar and colleagues reported that human RPE express all TLR mRNAs except TLR8 [
13]. This difference may be due to donor variation, different culture conditions or different time in culture (passage 2–5 in our study versus passage 7–12). In our studies, both IPE and RPE became less responsive to TLR ligands with increasing passage and failed to respond beyond generation 5 and 6 respectively (results not shown). Our results seem more reliable as low passage number cultures and donor-matched primary IPE and RPE were used, and the medium was formulated to promote optimal epithelial growth.
Previous studies used a well-characterised cell line, ARPE-19 (spontaneous arising retinal pigment epithelial cells), as representative of retinal pigment epithelial cells (RPE). In our previous study, we showed that there is a differential expression of CD14, a co-receptor of TLR4, between IPE and ARPE-19 [
23]. However, expression of CD14 from donor-matched primary IPE and RPE was not significantly different between the two cell types (data not shown). In addition, it has been shown that an elevated expression of proteins associated with microtubule cytoskeleton and IL-18 production was observed in ARPE-19 in comparison to RPE [
28,
29]. Therefore, the results generated from ARPE-19 should be interpreted with caution, as they may not be a reliable substitute for the primary cultured cells.
Although IPE and RPE possess a similar TLR expression profile, IPE appears to be more sensitive to PAMPs than RPE in culture and this observation was consistent across all three IPE and RPE matched donors (Figure
2). Culture supernatants confirmed that both IPE and RPE secreted a similar amount of total protein. IPE secreted maximal levels of IL-8 at 50 μg/ml following Poly(I:C) stimulation, whereas RPE required significantly higher dose of poly(I:C) (>100 μg/ml) for maximal IL-8 secretion. It was noted that high dose of LPS stimulation (10 μg/ml) was needed for a significant response in RPE. A recent finding suggests that LPS can be recognised in a TLR4-independent, caspase 11-dependent manner in mice and this activation was serotype specific, occurring in response only to
E. coli serotype O111:B4 [
30]. In our study the same serotype of LPS from
E. coli was used. Therefore it cannot be ruled out the possibility that LPS mediated IL-8 secretion in RPE may be TLR4-independent.
The pro-inflammatory cytokines, IL-8, was chosen for this study based on our previous findings: (i) that IL-8, MCP-1, IP-10, RANTES and MIP-1
b were significantly increased in aqueous humour of patients with active stage of anterior uveitis and this correlated with the clinical severity of the disease [
31]; (ii) IL-8 has been shown to contribute to the chemotactic signal for the recruitment of leukocytes in EIU [
32]. Therefore, IL-8 is an important mediator of the inflammatory response in clinical settings and in experimental animal models of uveitis.
MCP-1 is one of the key chemokines that regulate migration and infiltration of monocytes\macrophages to the sites of inflammation due to infection or tissue injury. The fact that IPE and RPE secreted both IL-8 (neutrophil chemoattractant) and MCP-1 in response to PAMPs are consistent with their role in innate immune responses and inflammation. IPE and RPE are regarded as “guardians” of the eye as they sense danger signals and consequent initiation of an inflammatory response. Whether the differences in their responsiveness to PAMPs is due to the nature of the cells; differential expression of TLRs and/or their co-receptors or different down-stream TLR signalling pathways or mRNA and protein stability, remains to be ascertained. Differential expression of TLRs and their co-receptors in ocular pigment epithelial cells may influence their response to PAMPs.
The role of TLR2, −3 and −4 in mediating cytokine response in Poly(I:C), LPS or MALP-2-treated IPE and RPE was also investigated using TLR inhibitors (OxPAPC, CI-095 and chloroquine; Figure
4). OxPAPC is an inhibitor of TLR2 and -4 signaling by competing with CD14, LBP and MD2, the accessory molecules that interact with bacterial lipids [
24]. Interestingly, IPE and RPE showed different sensitivity to OxPAPC. For example, IPE were more sensitive to OxPAPC in suppressing LPS mediated IL-8 secretion than were RPE cells (Figure
4A and E), but less sensitive to suppress MALP-2 mediated IL-8 secretion than RPE (Figure
4C and G). The difference could be due to different levels of accessory molecules of TLR4, such as CD14, LBP and MD2 on cell surface between the two cell types. CI-095 (also known as TAK-242) suppresses TLR4 signaling via its action on the intracellular domain of TLR4 and inhibits the production of nitric oxide and pro-inflammatory cytokines [
25,
26]. Chloroquine is a lysosomotropic agent that prevents endosomal acidification thus blocks signalling of intracellular TLRs [
27]. Both IPE and RPE showed similar response/sensitivity to CI-095 and chloroquine, which are potent inhibitors for TLR4 and TLR3, respectively. The fact that OxPAPC, CI-095 and chloroquine inhibited IL-8 secretion from LPS, MALP-2 or Poly(I:C)-treated IPE and RPE, suggests a role for TLR4 and possibly TLR2, TLR3 in LPS, MALP-2, Poly(I:C) mediated inflammatory responses.
A limitation of our study is that we employed a cell culture system. Nevertheless, animal studies support the concept that TLRs play a role in the pathogenesis of experimental autoimmune uveitis and EIU [
4,
33‐
35]. In addition, animal studies also provide an insight into physiological relevance of pro-inflammatory cytokine production
in vivo. For example, Allensworth et al. showed the potential of TLRs to trigger uveitis in mice [
33]. They concluded that all TLR agonists tested induced inflammation in the mouse eye with a marked increase of TNF-α, IL-6, IP-10, MCP-1 and KC and relatively little production of IFN-γ, IL-10, IL-12, IL-17, IL-1β, IL-4 or RANTES. In the current study, we have shown for the first time that cultured IPE cells express functional TLRs and respond to PAMPs through activation of TLRs, particularly TLR2, TLR3 and TLR4. This study extends the current knowledge of the role of TLR activation in iris culture and uveal innate immune mechanism in the pathogenesis of ocular inflammation.
Competing interests
Authors declare no conflict of interest.
Authors' contributions
KM designed, performed and analysed all the experiments and prepared the manuscript. JJYC assisted in data analyses, presentation and helped to draft the manuscript. NDG, PJM and DW designed experiments, supervised all aspects of the project and revised the manuscript. All authors read and approved the final manuscript.