Novel immunomodulators from hard ticks selectively reprogramme human dendritic cell responses

Abstract

Hard ticks subvert the immune responses of their vertebrate hosts in order to feed for much longer periods than other blood-feeding ectoparasites; this may be one reason why they transmit perhaps the greatest diversity of pathogens of any arthropod vector. Tick-induced immunomodulation is mediated by salivary components, some of which neutralise elements of innate immunity or inhibit the development of adaptive immunity. As dendritic cells (DC) trigger and help to regulate adaptive immunity, they are an ideal target for immunomodulation. However, previously described immunoactive components of tick saliva are either highly promiscuous in their cellular and molecular targets or have limited effects on DC. Here we address the question of whether the largest and globally most important group of ticks (the ixodid metastriates) produce salivary molecules that specifically modulate DC activity. We used chromatography to isolate a salivary gland protein (Japanin) from Rhipicephalus appendiculatus ticks. Japanin was cloned, and recombinant protein was produced in a baculoviral expression system. We found that Japanin specifically reprogrammes DC responses to a wide variety of stimuli in vitro, radically altering their expression of co-stimulatory and co-inhibitory transmembrane molecules (measured by flow cytometry) and their secretion of pro-inflammatory, anti-inflammatory and T cell polarising cytokines (assessed by Luminex multiplex assays); it also inhibits the differentiation of DC from monocytes. Sequence alignments and enzymatic deglycosylation revealed Japanin to be a 17.7 kDa, N-glycosylated lipocalin. Using molecular cloning and database searches, we have identified a group of homologous proteins in R. appendiculatus and related species, three of which we have expressed and shown to possess DC-modulatory activity. All data were obtained using DC generated from at least four human blood donors, with rigorous statistical analysis. Our results suggest a previously unknown mechanism for parasite-induced subversion of adaptive immunity, one which may also facilitate pathogen transmission.

Figure 1. Pretreatment of dendritic cells with Japanin inhibits their upregulation of CD86 in response to LPS.

Dendritic cells were incubated with japanin for 24 hours prior to the addition of LPS (100 ng/ml) for a further 18–20 hours. CD86 expression was then analysed by flow cytometry. (A) The results from a representative experiment using 500 ng/ml japanin. (B) Titration of Japanin concentration, showing a dose-dependent inhibition of CD86 upregulation. The range and mean of duplicate measurements from one representative experiment are shown. This experiment was performed four times, with dose-dependency demonstrated each time, but with EC₅₀ varying between donors.

Figure 2. Japanin inhibits CD86 upregulation in response to multiple DC maturation stimuli.

(A) Dendritic cells were cultured for 18–20 hours in the presence or absence of Japanin (500 ng/ml) and stimuli: (25 µg/ml Poly(I:C), via TLR3; 100 ng/ml LPS, TLR4; 4 µg/ml CL097, TLR7/8; 20 ng/ml IFNα2, IFNAR; 10–12.5 ng/ml TNFα, TNFR; 20 ng/ml IFNγ, IFNGR). CD86 expression was then assessed by flow cytometry. Modelled means ±95% confidence intervals using data from at least four experiments are shown, except for CL097 for which three experiments were performed, using cells from a total of five donors. (B) Data from all these experiments was used to assess the effect of Japanin on CD86 expression in the absence of stimuli. ** p<0.01, * p<0.05, NS p>0.05.

Figure 3. Japanin binds specifically to DC.

(A) Monocyte-derived dendritic cells, or (B and C) human blood PBMC were incubated with 100 ng/ml Japanin-DyLight 649 (filled histograms) or 340 ng/ml OmCI-DyLight 649 (open histograms), incubated on ice for 1 hour, and washed. Binding was assessed by flow cytometry. In B, major PBMC subsets are defined by surface molecule expression. In C, blood DC are defined as CD14⁻HLA-DR⁺lineage⁻CD16⁻ then further subdivided according to CD1c, CD11c, CD123 and CD141 expression. For the complete gating strategy, see figure S4. Results shown are representative of those from 6 (A) and 4 (B and C) experiments with cells from different donors.

Figure 4. Japanin modulates dendritic cell maturation, rather than simply inhibiting it.

Dendritic cells were cultured in the presence or absence of Japanin (500 ng/ml) and LPS (100 ng/ml) for 18–20 hours. (A) CD40, CD83, CD86, CD274 and HLA-DR expression were then assessed by flow cytometry, and (B)the concentration of pro-inflammatory cytokines in the culture supernatant was measured by Luminex1. Modelled means ±95% confidence intervals using data from at least four experiments are shown, except where marked ‡ where above-scale readings in the LPS-only made it impossible to calculate meaningful confidence intervals; the graphs show the lowest possible mean value (taking an above-scale value to be equal to the maximum possible on-scale value). ** p<0.01, * p<0.05, p<0.1, NS p>0.05.

Figure 5. Japanin blocks differentiation of DC from monocytes.

Monocytes were cultured with GM-CSF (1000 U/ml) and IL4 (500 U/ml) with or without Japanin (500 ng/ml). Before the culture, and again after 3 and 5 days of culture, CD1a and CD14 expression were assessed by flow cytometry, in order to monitor differentiation into CD1a⁺CD14^low dendritic cells. Data shown is from one experiment, representative of three independent experiments using cells from different donors.

Figure 6. Japanin is a lipocalin.

The mature Japanin sequence was aligned with (A) sequences of mature tick proteins with a resolved lipocalin structure, or (B) with these and additional distantly-related sequences also accepted to be tick lipocalins. In A, key residues identified by Adam and colleagues are shaded, and their nature noted below. Residue characteristics are: hydrophobic = ACFGHILMPVWY; hydrophilic = DEHKNQRSTY; charged = DEHKR; aromatic = FHWY; bulky = EFHIKLMQRWY; small = not bulky. In B, cysteine residues are highlighted green, and the conserved tick lipocalin motif is boxed red.

Figure 7. Japanin homologues modulate DC maturation.

(A) Three Japanin homologues were successfully expressed in Sf9 cells, as shown by Western blotting with an anti-His tag antibody. ∼10 ng protein was loaded per lane. (B) & (C) Dendritic cells were cultured in the presence or absence of Japanin or Japanin homologues (500 ng/ml) and LPS (100 ng/ml) for 18–20 hours. CD86 (B) and CD274 (C) were then assessed by flow cytometry. Modelled means ±95% confidence intervals using data from three (cells with LPS) or four (cells without LPS) experiments are shown. * p<0.05, as compared to cells without Japanin or a Japanin homologue.

Figure 8. Sequence alignment of Japanin and its homologues.

Alignments were generated with ClustalX and manually refined. Shading intensity indicates BLOSUM62 score. N-glycosylation sequences and conserved cysteine residues are boxed. Ra-FS-HBP2 (PDB ID 1QFT) is aligned as an example of a tick lipocalin with low sequence similarity to Japanin.

Figure 9. Japanin-like proteins form a clade within hard tick lipocalins.

(A) A phylogenetic tree derived from maximum-likelihood analysis of hard tick lipocalins (including Japanin and its homologues), as well as the soft tick lipocalins OmCI, monomine and Am182. Sequences were aligned using ClustalX and manually refined, then Mega5.1 was used to construct a phylogeny. The frequency with which associated taxa clustered together in the bootstrap test is shown. For reasons of clarity, only selected protein names and bootstrap frequency labels are shown, and the Japanin clade is shown in detail in (B), with the full tree supplied as supplementary data.