If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Correspondence: Claudia Günther, Department of Dermatology, University Hospital, Technical University Dresden, Fetscherstrasse 74, D-01307 Dresden, Germany.
Innate immune processes are central in the development of the chronic inflammatory skin disease psoriasis. Studying stimulation of keratinocytes, monocytes, and dendritic cells by type I interferons or ligation of Toll-like receptors 1/2, 2/6, or 7, but not 7/8, resulted in enhanced surface expression and secretion of CXC chemokine ligand (CXCL) 16. The corresponding CXC chemokine receptor 6 was expressed on neutrophils whose recruitment into skin is important, especially in early psoriatic disease. Using the recently developed technique real-time deformability cytometry demonstrated that CXCL16 and IL-8 decreased the stiffness and enhanced deformation of neutrophils facilitating transmigration through vessel wall. In addition, CXCL16 potently induced migration of neutrophils and enhanced the chemotactic effect of IL-8. The positive feedback loop was supported by IL-8 enhancing CXCL16 production of neutrophils. Blocking of CXCL16 expression by effective treatment of psoriasis patients with tumor necrosis factor−α blockers further supported the pathogenic role of this chemokine. In summary, the data link innate immune stimulation to CXCL16 upregulation and neutrophil infiltration into skin. CXCL16 could therefore represent a potent future target for treatment of psoriasis.
). The complex pathogenesis of psoriasis is based on a genetic predisposition that can give rise to chronic scaling cutaneous plaques with variable joint and cardiovascular involvement (
). Psoriatic skin lesions are characterized by a mixed inflammatory infiltrate consisting of neutrophils, dendritic cells (DCs), macrophages, and T cells (
). Macrophages and DCs are of special immanent importance as antigen-presenting cells linking innate immune sensing and stimulation of adaptive immune responses (
A novel dendritic cell population in human blood: one-step immunomagnetic isolation by a specific mAb (M-DC8) and in vitro priming of cytotoxic T lymphocytes.
). They stand out by their capacity to secrete proinflammatory cytokines and stimulate T helper 1/T helper 17 T-cell responses, which drive hyperproliferation of keratinocytes and psoriatic inflammation (
). We have recently described that CXC chemokine ligand (CXCL) 16 is upregulated on keratinocytes and myeloid antigen-presenting cells in psoriatic skin and has the capacity to induce migration of CD8+ T cells in vivo (
). The expression of the CXCL16 receptor (CXCR) 6 is not only restricted to CD8+ T cells, but is also found on type 1 polarized peripheral blood CD4+ T cells, γδ T cells, natural killer cells, natural killer T cells, monocytes, and neutrophils (
Human trophoblasts recruited T lymphocytes and monocytes into decidua by secretion of chemokine CXCL16 and interaction with CXCR6 in the first-trimester pregnancy.
). Neutrophils might be especially important in the induction phase of psoriasis and contribute by secretion of proinflammatory cytokines, including IL-17, and the formation of neutrophil extracellular traps, which are rich in self-nucleic acids (NAs) (
). The impact of CXCL16 on neutrophil recruitment in psoriasis has not yet been specifically addressed. Neutrophils are effector cells of the innate immune system and stimulation of innate immunity has increasingly been recognized as a potential initial trigger for psoriasis. It is known that psoriasis can be provoked or exacerbated by a variety of environmental factors, including the microbiome consisting of bacterial and viral pathogens colonizing the skin (
). Toll-like receptors (TLRs) recognize bacterial wall components or NAs. TLRs binding NAs are located in endosomal compartments to avoid unnecessary immune activation to self-NAs that are continuously released by dying host cells or extruded by activated neutrophils (
). In psoriasis, the high prevalence of antimicrobial peptides can break protective mechanisms by complexing and stabilizing DNA and RNA not only from pathogens, but also from self-originating NAs. This activation has been recognized as a major innate immune pathway triggering psoriasis (
We therefore aimed at investigating the effect of TLR stimulation on the regulation of CXCL16 expression. Hereby we could find a strong induction of CXCL16 after TLR ligation that led to an induction of neutrophil shape changes and deformation contributing to cell activation and recruitment.
Results
TLR-mediated expression of CXCL16 on monocytes and slanDCs
We have previously shown that the chemokine CXCL16 is upregulated in psoriatic skin and mainly expressed on CD163+ macrophages and CD11c+ dendritic cells in the dermis (
). To further analyze the role of CXCL16 in the pathogenesis of psoriasis, we determined the CXCL16 expression on peripheral blood mononuclear cells (PBMCs) by flow cytometry. CXCL16 was mainly expressed on CD14+CD16low and CD14lowCD16+ monocytes and slanDCs (Figure 1a and Supplementary Figure S1 online). Interestingly, antigen-presenting cells from psoriasis patients showed a higher surface expression of CXCL16 compared with healthy controls (Figure 1a). This expression is sustained during migration in the skin, as we could detect CXCL16-expressing CD163+ monocytes/macrophages (
Figure 1CXCL16 expression and secretion by monocytes and slanDCs is regulated by TLR stimulation. (a) Flow cytometric analysis of CXCL16 expression on freshly blood-isolated antigen-presenting cells. Mean ± standard deviation (SD) values of six psoriasis patients and six healthy controls are shown; unpaired Student t test. (b−f) Expression of CXCL16 after 24 hours culture in medium compared to freshly isolated cells (w/o) (b) and compared to TLR stimulation on CD14+CD16low (c), CD14lowCD16+ monocytes (d), and slanDCs (e). Concentration of CXCL16 in PBMC supernatants measured by ELISA (f). (g) CXCL16 expression and production of PBMCs from psoriasis patients compared to healthy controls (24 hours). Mean ± SD values of at least five independent donors are shown; paired Student t test (b), one-way analysis of variance with Bonferroni’s multiple comparison test (c−f), one-tailed unpaired Student t test (g). CXCL, CXC chemokine ligand; DC, dendritic cell; LPS, lipopolysaccharide; MFI, mean fluorescence intensity; PBMC, peripheral blood mononuclear cell; slan, 6-sulfo LacNAc; TLR, Toll-like receptor.
Given the elevated expression of CXCL16 on antigen-presenting cells in psoriasis patients, we asked which trigger factors could be responsible for induction of CXCL16. We observed that already spontaneous maturation in medium for 24 hours can upregulate expression of CXCL16 on CD14+CD16low monocytes (8.4-fold), CD14lowCD16+ monocytes (2.6-fold), and slanDCs (4-fold) compared with freshly isolated cells (Figure 1b). In psoriasis, activation of highly expressed TLRs by bacterial or viral infections can contribute to the initiation and maintenance of the disease (
Toll-like receptor (TLR) 2 is upregulated on peripheral blood monocytes of patients with psoriatic arthritis: a role for a gram-positive inflammatory trigger?.
). Stimulation of differentiated mononuclear cells from psoriasis patients with the synthetic TLR ligands Pam2 (TLR2/6), Pam3 (TLR2/1), and R837 (TLR7) significantly enhanced the expression level of CXCL16 on the monocyte subpopulations CD14+CD16low (1.6-fold; Figure 1c) and CD14lowCD16+ (2-fold; Figure 1d) compared to unstimulated cells. This effect was not observed on slanDCs (Figure 1e). CXCL16 expression levels after TLR ligation were increased compared to healthy controls (Figure 1g). In contrast, stimulation with lipopolysaccharide (LPS; TLR4) and R848 (TLR8/7) inhibited CXCL16 expression by 60% on monocytes and DCs compared to unstimulated cells from psoriasis patients (Figure 1c−1e).
Chemotactic effects of CXCL16 are exerted by the soluble chemokine upon enzymatic cleavage of membrane bound CXCL16. Likewise to transmembrane chemokine expression, Pam2, Pam3, and R837 increased the CXCL16 secretion and LPS and R848 inhibited the release of CXCL16 (Figure 1f, 1g).
Tumor necrosis factor−α antagonists downregulate CXCL16 in vitro and in vivo
The pro-inflammatory cytokine tumor necrosis factor−α (TNF-α) is highly expressed in psoriasis (
) and is a strong inductor of CXCL16 expression and secretion in monocytes and slanDCs (Figure 2a, 2b ). Blockade of TNF-α by treatment with etanercept completely inhibited TNF-α−induced CXCL16 expression and secretion in monocytes and slanDCs from psoriasis patients (Figure 2a, 2b). On mononuclear cells cultured in medium for 24 hours, the TNF-α blocker etanercept and adalimumab ameliorated the CXCL16 surface expression by 60% on slanDCs and by 40% on monocytes (Figure 2c). In addition, upon Pam2 or R837 stimulation, both inhibitors significantly decreased the CXCL16 surface expression on monocyte subpopulations (Figure 2c). In the PBMC supernatant, concentration of soluble CXCL16 was reduced by 20% after preincubation with TNF-α blockers (Figure 2c). In all experiments, the effects of etanercept and adalimumab were similar. To validate our findings in vivo, we analyzed skin biopsies from patients treated with adalimumab. We detected a decrease in CXCL16 expression in lesional skin after a 4-week treatment period (Figure 2d and Supplementary Table S1 online). This was comparable to the CXCL16 reduction by etanercept observed previously (
Figure 2TNF-α blockade reduces CXCL16 expression in vitro and in the skin of psoriasis patients. (a−c) PBMCs were cultivated in medium with TNF-α, Pam2, or R837 upon treatment with etanercept or adalimumab for 24 hours. CXCL16 expression was analyzed by flow cytometry (a, c) and CXCL16 production by ELISA (b, c). Three individual experiments (a) performed in duplicates (b) and 6 (mean fluorescence intensity CXCL16) or at least five (concentration CXCL16) individual experiments (c) are shown; one-tailed Mann-Whitney U test (a) and one-way analysis of variance with Bonferroni’s multiple comparison test (b, c). (d) Immunofluorescence staining of CXCL16 (green) in psoriatic skin before treatment with adalimumab and at week 4 (n = 3). The basement membrane zone is marked by the solid line and the epidermis is indicated by “e.” Scale bar = 50 μm. CXCL, CXC chemokine ligand; LPS, lipopolysaccharide; MFI, mean fluorescence intensity; PBMC, peripheral blood mononuclear cell; TNF-α, tumor necrosis factor−α.
Keratinocytes from psoriasis patients secrete more CXCL16 upon TLR and TNF-α stimulation
Staining of CXCL16 in psoriatic lesions demonstrates chemokine expression not only in dermal inflammatory cells, but also a bright expression in the epidermis, which can result from CXCL16 expression and secretion by lesional keratinocytes (Figure 2d) (
). Analyzing primary keratinocytes from psoriatic lesions in comparison to healthy controls did not demonstrate significantly enhanced surface expression of CXCL16 (Figure 3a). However, the amount of CXCL16 secreted by keratinocytes from psoriatic patients after 24 hours of culture conditions and the CXCL16 protein levels analyzed by Western blot were significantly enhanced in psoriatic keratinocytes compared to healthy controls (Figure 3b, 3c). This suggests that production of CXCL16 in psoriatic epidermis is enhanced, resulting in a high turnover rate of transmembrane-expressed chemokine. Due to enhanced levels of ADAM10 in psoriatic epidermis (
), CXCL16 will be rapidly cleaved and secreted as indicated by the enhanced epidermal staining in lesional skin (Figure 2d).
Figure 3CXCL16 expression and secretion by primary keratinocytes of psoriasis patients upon TLR and TNF-α stimulation (24 hours). (a) Expression and (b) secretion of CXCL16 with and without poly(IC) stimulation compared to healthy controls analyzed by flow cytometry and ELISA. (c) CXCL16/β-actin protein expression ratio of in medium cultured psoriatic keratinocytes (P1−P3) compared to healthy controls (H1−H3) detected by western blot. C = chemokine domain. (d) CXCL16 expression and (e) secretion with and without TLR stimulation evaluated by flow cytometry and ELISA. CXCL16 secretion upon TNF-α stimulation is inhibited by etanercept (f). Three individual experiments (a, c, d) performed in duplicate (b, e, f) are shown; Friedman test (d), Mann-Whitney U test (a, c), one-way analysis of variance with Bonferroni’s multiple comparison test (e, f), and unpaired Student t test (d). CXCL, CXC chemokine ligand; poly(IC), polyinosinic-polycytidylic acid; TLR, Toll-like receptor; TNF-α, tumor necrosis factor−α.
Expression, subcellular localization and cytokinic modulation of Toll-like receptors (TLRs) in normal human keratinocytes: TLR2 up-regulation in psoriatic skin.
). Stimulating keratinocytes with the synthetic TLR agonists polyinosinic-polycytidylic acid (poly[IC]; TLR3), Pam2 (TLR2/6), Pam3 (TLR2/1), or LPS (TLR4) for 24 hours did not influence cell surface expression of CXCL16 (Figure 3d), but stimulation with poly(IC) or Pam2 led to an increase in CXCL16 secretion (Figure 3e), indicating the influence of external trigger factors in CXCL16-mediated inflammation. The response after poly(IC) stimulation was significantly higher in keratinocytes from psoriasis patients (Figure 3b).
Besides monocytes and slanDCs, keratinocytes from psoriatic lesions also exhibited an increased production of soluble CXCL16 upon stimulation with the proinflammatory cytokine TNF-α for 24 hours. This effect was completely inhibited by the TNF-α blocker etanercept (Figure 3f).
TLR-dependent CXCL16 regulation is enhanced by type I IFN and inhibited by IL-10
TLR-mediated detection of NAs and production of type I IFN are linked to the induction of psoriasis (
). Because stimulation with the RNA ligands R837 (TLR7) on monocytes and poly(IC) (TLR3) on keratinocytes induces a signaling pathway that can lead to the production of type I IFN (
), we first measured the concentration of type I IFN in the supernatants of PBMCs and keratinocytes by IFN reporter assay. Compared to unstimulated cells, high levels of type I IFN were detectable in supernatants of R837-stimulated PBMCs (Figure 4a) and poly(IC)-stimulated keratinocytes (Figure 4b) of psoriasis patients. Stimulation with IFN-beta resulted in an upregulation of CXCL16 on CD14lowCD16+ monocytes and slanDCs (Figure 4c), as well as an increased production of soluble CXCL16 by keratinocytes (Figure 4d) that was higher in patients with psoriasis compared with healthy controls suggesting cell intrinsic priming.
Figure 4CXCL16 expression and secretion is upregulated by type I IFN in psoriasis patients (24 hours). (a, b) Type I IFN concentrations in supernatant of R837-stimulated PBMCs (a) and poly(IC)-stimulated keratinocytes (b) analyzed by IFN reporter assay. Mean ± standard deviation (SD) values of three individual experiments of three donors are shown; paired Student t test. (c, d) CXCL16 expression on CD14lowCD16+ monocytes and slanDCs (c) and CXCL16 secretion by keratinocytes upon IFN-beta stimulation compared to healthy controls (d) detected by flow cytometry and ELISA. Data are given as mean ± SD of three independent donors (c) performed in duplicates (d); one-tailed Mann-Whitney U test (c), paired and unpaired Student t test (d). CXCL, CXC chemokine ligand; DC, dendritic cell; MFI, mean fluorescence intensity; PBMC, peripheral blood mononuclear cell; poly(IC), polyinosinic-polycytidylic acid; slan, 6-sulfo LacNAc.
To further explore the mechanism of CXCL16 inhibition after stimulation of TLR4 and TLR7/8 on mononuclear cells, we analyzed supernatants of PBMCs from patients with psoriasis for secretion of IL-10, which has been suggested to downregulate CXCL16 (
Monocytes, peripheral blood mononuclear cells, and THP-1 cells exhibit different cytokine expression patterns following stimulation with lipopolysaccharide.
). Indeed, we detected elevated concentrations of IL-10 after LPS (TLR4) or R848 (TLR7/8) stimulation (Supplementary Figure S3a online). The inhibitory effect on CXCL16 was confirmed by dose-dependent stimulation of PBMCs with IL-10 for 24 hours. IL-10 (5 ng/ml) significantly decreased the secretion of soluble CXCL16 by 18% and the expression of transmembrane CXCL16 on CD14+CD16low monocytes by 44%, CD14lowCD16+ monocytes by 31% and slanDCs by 38% (Supplementary Figure S3b, S3c).
CXCL16 enhances migratory and mechanical properties of CXCR6 and CXCL16 expressing neutrophils
Upregulation of CXCL16 in psoriasis can facilitate the recruitment of CXCR6+ CD8+ T cells into the skin (
). Besides T cells, neutrophils are important effector cells in psoriasis profoundly infiltrating into skin (Supplementary Figure S4 online), which secrete proinflammatory cytokines such as IL-1β, IL-6, IL-17, and IL-23 (
Role of neutrophils in induction of acute inflammation in T-cell-mediated immune dermatosis, psoriasis: a neutrophil-associated inflammation-boosting loop.
). Staining of neutrophils infiltrating psoriatic lesions revealed expression of the CXCL16 receptor CXCR6 on CD66b+ neutrophils (see Supplementary Figure S4a). In addition, neutrophils isolated from the periphal blood of psoriasis patients expressed CXCR6 on their surface (see Supplementary Figure S4b), which was enhanced upon stimulation with phorbol 12-myristate 13-acetate.
After confirming the expression of CXCR6, we analyzed chemotaxis of isolated neutrophils in response to CXCL16 in vitro using a transwell migration assay. CXCL16 induced a dose-dependent migration of neutrophils in patients with psoriasis (Figure 5a), as well as healthy controls (data not shown). An important chemotactic factor for neutrophil attraction into psoriatic lesions is IL-8 (also known as CXCL8) (
). Because chemotaxis of inflammatory cells is a redundant process, we investigated whether CXCL16 can potentiate IL-8−induced chemotaxis. Indeed, the simultaneous stimulation by CXCL16 and IL-8 resulted in an enhanced migratory response of neutrophils (Figure 5a).
Figure 5CXCL16 mediates chemotaxis of CXCL16 expressing neutrophils from psoriasis patients. (a) Migrating neutrophils in response to CXCL16 and/or IL-8 analyzed by transwell migration assay. Mean ± standard deviation (SD) of at least five independent donors are shown; one-way analysis of variance with Bonferroni’s multiple comparison test. (b, c) Immunofluorescence staining of CXCL16 (green) by neutrophils (CD66b, red; overview: scale bar = 30 μm, detail marked by square: scale bar = 10 μm; b) and in IL-8-expressing dermal vessels (red; scale bar = 10 μm; c) in psoriasis vulgaris. (d) Western blot of CXCL16 protein expression. (e) CXCL16 secretion after stimulation with IL-8 (50 ng/ml) for 24 hours examined by ELISA. Mean ± SD values of three individual experiments performed in duplicate are shown; paired Student t test. C, chemokine domain; CXCL, CXC chemokine ligand; PBMC, peripheral blood mononuclear cell; PMN, polymorphonuclear neutrophils; T, transmembrane full-length form; S, soluble form.
In addition, immunohistochemical staining of psoriatic lesions revealed expression of CXCL16 by neutrophils in the upper dermis (Figure 5b) and in dermal located blood vessels (Figure 5c) (see also Supplementary Table S1). Western blot analysis of lysed neutrophils isolated from blood of patients with psoriasis confirmed expression of CXCL16 similar to the known expression of CXCL16 on antigen-presenting mononuclear cells (Figure 5d). Stimulation of neutrophils with IL-8 for 24 hours induced secretion of soluble CXCL16 (Figure 5e), suggesting a positive feedback loop.
As transmigration into tissue requires activation and mechanical deformation of neutrophils, we further investigated the effect of the chemoattractant IL-8 and CXCL16 on mechanical properties of neutrophils using a recently described technique called real-time deformability cytometry (
). Hereby, cells are deformed in a microfluidic channel constriction and analysis of morphological and mechanical parameters is performed in real-time at rates up to 1,000 cells/s (
). First, we analyzed size (projected area of the cell) and shape of neutrophils. The latter is estimated by the area ratio of convex hull and contour of the cell. Compared to healthy neutrophils, untreated neutrophils from psoriasis patients were larger and had an increased area ratio indicating an irregular shape (Figure 6a, 6b ). This can be interpreted as sign of pre-activation. Next we analyzed the deformation D, as described in
. Untreated neutrophils from psoriasis patients showed a higher deformation (Figure 6a, 6b). IL-8 (50 ng/ml) induced a significant increase in area, area ratio, and deformation of neutrophils that was observed to a similar extent in neutrophils isolated from healthy controls or patients with psoriasis (Figure 6a, 6b). Instead, an increased deformation upon CXCL16 ligation (100 ng/ml) was recorded on healthy neutrophils only, but was less strong compared with IL-8 (Figure 6a, 6b). The increased deformation of round, inactivated neutrophils (area ratio of 1.0:1.05) from healthy donors upon CXCL16 and IL-8 stimulation indicates cell softening (Figure 6c). Simultaneous stimulation with both chemokines had no additional effect on activation or deformation (Figure 6b, 6c).
Figure 6Morphological and mechanical properties of neutrophils analyzed by real-time deformability cytometry. Representative scatter and contour plots (a) and statistical analysis (b) of area, area ratio, and deformation of neutrophils from healthy controls and patients with psoriasis after stimulation with or without CXCL16 or IL-8 for 1 hour (area ratio 1.0:1.2; 0.04 μl/s flow rate). (c) Deformation of round, inactivated neutrophils (area ratio 1.0:1.05) from healthy donors. Results of three independent donors are shown; linear mixed-effects models. CXCL, CXC chemokine ligand.
Chemokines orchestrate the overboarding immune activation in psoriasis, which leads to hyperproliferation of keratinocytes accompanied by infiltrating immune cells. Psoriasis can be provoked by a variety of environmental factors, which include the microbiome colonizing the skin. TLRs are the most important class of innate immune receptors recognizing pathogen-associated molecular patterns (
). Here, we demonstrate that ligation of TLR2/6 and TLR2/1 on monocytes and keratinocytes upregulate CXCL16 secretion. TLR2/6 can be engaged by diacylated lipopeptides from Gram-positive bacteria, such as Staphylococcus aureus (
). TLR2 can also be engaged by other Gram-positive bacteria, such as streptococcal organism. This could be pathogenically relevant in cases of psoriasis induced by Streptococcus pyogenes throat infection. Invading pathogens could access antigen-presenting cells in the tonsils, which migrate through lymphatics and blood vessels into the skin and activate the immune response (
Besides exogenous factors, there are many endogenous TLR ligands, such as antimicrobial peptides and proteins, overexpressed in psoriatic skin. This overexpression may trigger inflammation in psoriasis by activation of proinflammatory cytokines and chemotaxis toward immune cells (
). S100 proteins, such as S100A7, S100A8, and S100A9, as wells as human β-defensin 2 and 3, antimicrobial ribonuclease RNase 7, lysozyme, and cathelicidin LL37, were identified to be highly produced by keratinocytes from patients with psoriasis vulgaris (
). We have shown that stimulation of TLR3 and TLR7 can lead to CXCL16 upregulation. TLR3 is thought to recognize double-stranded RNA, while TLR7 has been implicated in single-stranded RNA recognition. These ligands could originate from bacteria or self-NAs. Cationic antimicrobial peptides, such as LL37 and β-defensin, which are overexpressed in psoriatic skin (
) can form complexes with free self-DNA and self-RNA. These self-RNA−antimicrobial peptide complexes protect NAs from degradation and can activate human cells through TLR7 (
). The importance of TLR7 ligation for triggering psoriasis has been demonstrated best by the observation that topical application of the TLR7 agonist imiquimod to the skin exacerbates human psoriasis (
). TLR7 and TLR9 are also expressed by plasmacytoid DCs, which sense RNA and DNA released by dying bacteria and host cells coupled with antimicrobial peptides resulting in type I IFN secretion (
). Plasmacytoid DCs have been described as important innate immune cells in the initiation phase of psoriasis and are well known as the most potent type I IFN producing immune cells (
). Interestingly, our data revealed that CXCL16 can be induced by type I IFN, which implicates its importance in an early phase of psoriasis induction. Type I IFN is induced IRF-3/7−dependent after ligation of TLR3 and TLR7, whereas stimulation of TLR2/1 and TLR2/6 activates the MyD88- and NF-κB−dependent pathway (
) and could explain the TLR2-mediated upregulation of CXCL16 in keratinocytes (see Supplementary Figure S5 online). CXCL16 could be induced by both TLR signaling pathways highlighting its importance in the innate immune response against pathogen-associated molecular patterns, including microbes and self-NAs. Importantly, both types of TLR signaling leading to CXCL16 upregulation were induced in keratinocytes. Their function as epidermal barrier with first contact to the microbiome further implicates CXCL16 upregulation as an initial inflammatory mechanism in psoriasis. Furthermore, TNF-α is a potent stimulator of CXCL16 expression (
), and we could show that TNF-antagonizing drugs ameliorate CXCL16 expression in vivo and in vitro even upon TLR2 and TLR7 ligation supporting the role of CXCL16 in the pathogenesis of the disease. In rare cases, triggering psoriasis under TNF blockade can be observed (
), which could be explained by type I IFN triggering the disease, including an upregulation of CXCL16 upon TLR2/1, 2/6, or TLR7 stimulation.
Neutrophils are important effector cells in psoriasis that characteristically invade the stratum corneum and lead to inflammation by secretion of neutrophil extracellular traps, whereby they extrude web-like chromatin strands complexed with antimicrobial peptides, including β-defensin and LL37 (
). The attraction of neutrophils to the skin is mediated by a network of chemotactic factors secreted by keratinocytes and immune cells, of which IL-8 is one of the most potent regarding neutrophil activation. However, blocking of IL-8 alone in a clinical trial did not sufficiently control the disease, indicating that a concerted action of chemoattractants might be responsible for inducing neutrophil migration (
). Here, we demonstrated that CXCL16 can stimulate chemotaxis of neutrophils and additionally enhance the effect of IL-8 on neutrophil attraction. These findings reveal CXCL16 as an important chemokine mediating neutrophil migration through the tissue toward the epidermis of early psoriatic lesions (
During the process of tethering and rolling at the vessel wall, neutrophils become activated and eventually transmigrate through the endothelium accompanied by squeezing through microvascular constriction (
). Using real-time deformability cytometry that permits short time-scales deformation (∼1 ms) imitating the rapid transmigration process across narrow constrictions we could conclude that neutrophils from blood of patients with psoriasis are pre-activated and more prone to deformation upon shear stress. This could be an advantage for efficient transmigration into the skin additionally supported by IL-8 and therefore enhance the inflammatory process of psoriasis. Shape changes and increased deformation induced by CXCL16 and IL-8 stimulation of neutrophils from healthy controls directly implicate an effect on the cellular cytoskeleton (
Interleukin-8 induces changes in human neutrophil actin conformation and distribution: relationship to inhibition of adhesion to cytokine-activated endothelium.
), this could promote entering narrow constrictions, such as the vascular endothelium. Interestingly, neutrophilic granules also contain chemotactic factors, which are released upon stimulation and contribute to the vicious cycle of inflammation (
). Our experiments demonstrated that neutrophils in psoriatic lesions contain prestored CXCL16 and stimulation of neutrophils by IL-8−induced secretion of this chemoattractant, which could further contribute to maintain psoriatic inflammation.
In conclusion, these findings suggest that upregulation of CXCL16 by innate immune sensing upon TLR ligation and type I IFN stimulation in psoriasis does not only have a stimulating effect on CXCL16-mediated migration of neutrophils, but also alters their mechanical properties facilitating transmigration into tissue. Additionally, IL-8−induced secretion of CXCL16 by neutrophils supports the positive feedback loop (see Supplementary Figure S6 online). Targeting CXCL16 could therefore be a potential therapeutic target in psoriasis.
Materials and Methods
Patients
Blood and skin samples were taken from patients with psoriasis vulgaris (mean ± standard deviation age 47.9 ± 15.1 years), who did not receive any systemic treatment with immunosuppressive drugs for at least 2 weeks. Diagnosis of psoriasis vulgaris was confirmed by clinical and histologic criteria. Blood from age-matched healthy volunteers was used as control. The investigational protocols (EK396112011 and 539122015) were approved by the Ethics Committee of the University Hospital of the Technical University Dresden according to the Declaration of Helsinki, and patients provided written informed consent for this study.
In vitro stimulation of PBMCs
PBMCs were isolated from heparinized whole blood from psoriasis patients and healthy controls by density gradient centrifugation using Biocoll separating solution (Biochrom, Berlin, Deutschland). PBMCs were cultured in RPMI 1640 medium (Gibco LifeTechnologies, Carlsbad, CA) containing 10% fetal calf serum (Sigma-Aldrich, St. Louis, MO), 2 mM l-glutamine, 1% nonessential amino acids, 100 U/ml penicillin, and 100 mg/ml streptomycin (Biochrom, Cambridge, UK) with or without therapeutic concentrations of 10 μg/ml etanercept or 5 μg/ml adalimumab at 1 × 106 cells/ml in 6-well plates for 6 hours at 37°C in 5% CO2. Synthetic TLR agonists (100 ng/ml Pam2CSK4, 100 ng/ml Pam3CSK4, 1 μg/ml imiquimod (R837), 1 μg/ml R848 (all from InvivoGen, San Diego, CA), and 100 ng/ml LPS (Sigma-Aldrich), or recombinant human proteins TNF-α (InvivoGen), IFN-beta and IL-10 (both from PeproTech, Rocky Hill, NJ) were added for additional 18 hours. Determination of transmembrane and soluble CXCL16 was performed by flow cytometry and ELISA.
In vitro stimulation of keratinocytes
Primary human keratinocytes were derived from skin biopsies of psoriasis patients or healthy controls. The epidermis was separated from dermis by dispase treatment over night at 4°C. Keratinocytes were extracted by trypsin incubation at 37°C. After cultivation on a feeder layer of 3T3 fibroblasts in a complex medium containing a mixture of DMEM and Ham’s F-12, feeder cells were removed by trypsinization and the keratinocytes were precultured at 1 × 106 cells/ml in 6-well plates in DermaLife K complete medium (Lifeline Cell Technology, Frederick, MD) for 48 hours until 90% confluency was reached. Stimulation of keratinocytes was conducted with 100 ng/ml Pam2CSK4, 100 ng/ml Pam3CSK4, 1 μg/ml poly(IC) (all from InvivoGen), 100 ng/ml LPS (Sigma-Aldrich), TNF-α (InvivoGen) with or without etanercept, and IFN-beta (PeproTech) for 24 hours. Analysis of CXCL16 expression and secretion followed up using flow cytometry and ELISA.
Flow cytometry
For flow cytometric analysis, cells were incubated with fluorescent primary and secondary antibodies (Supplementary Table S2 online) and analyzed on a FACSCalibur (BD Biosciences, San Jose, CA).
ELISA
Soluble CXCL16 in cell-free supernatants was quantified using Duoset ELISA assay (R&D Systems, Minneapolis, MN).
Statistical analysis
Data are presented as bars (indicating mean ± standard deviation) or box plots (indicating medians, whiskers: 5th to 95th percentiles and standard deviations). Statistical analysis was performed using repeated-measures one-way analysis of variance (in normally distributed values tested by Shapiro-Wilk test) with Bonferroni’s multiple comparison post-hoc test, unless otherwise indicated, with the help of GraphPad Prism 6 (GraphPad Software, San Diego, CA) or using linear mixed-effects models with ShapeOut (
). In all cases, ∗P < 0.05 was considered to be statistically significant (∗∗P < 0.01, ∗∗∗P < 0.001).
Conflict of Interest
CG has received research grants from Pfizer Pharma GmbH. SS has been supported by research grants from Pfizer Pharma GmbH.
Acknowledgments
We thank Hella Luksch, Sigrun Hofmann, and Nick Zimmermann for excellent technical help and assistance. We thank Min Ae Lee-Kirsch for support establishing the IFN-reporter assay. We are grateful to Nicole Töpfner and Christoph Herold for their valuable advice and discussions. This study was supported by research grants from Pfizer Pharma GmbH (CG), grants from the Deutsche Forschungsgemeinschaft to CG (KFO 249/GU1212/1-1 and GU 1212/1-2) and JG (KFO249/GU 612/2-2) and an Alexander von Humboldt Professorship (JG).
Expression, subcellular localization and cytokinic modulation of Toll-like receptors (TLRs) in normal human keratinocytes: TLR2 up-regulation in psoriatic skin.
Toll-like receptor (TLR) 2 is upregulated on peripheral blood monocytes of patients with psoriatic arthritis: a role for a gram-positive inflammatory trigger?.
Human trophoblasts recruited T lymphocytes and monocytes into decidua by secretion of chemokine CXCL16 and interaction with CXCR6 in the first-trimester pregnancy.
Monocytes, peripheral blood mononuclear cells, and THP-1 cells exhibit different cytokine expression patterns following stimulation with lipopolysaccharide.
A novel dendritic cell population in human blood: one-step immunomagnetic isolation by a specific mAb (M-DC8) and in vitro priming of cytotoxic T lymphocytes.
Role of neutrophils in induction of acute inflammation in T-cell-mediated immune dermatosis, psoriasis: a neutrophil-associated inflammation-boosting loop.
Interleukin-8 induces changes in human neutrophil actin conformation and distribution: relationship to inhibition of adhesion to cytokine-activated endothelium.
32962-7&id=gr1.jpg){kind=link}
32962-7&id=gr2.jpg){kind=link}
32962-7&id=gr3.jpg){kind=link}
32962-7&id=gr4.jpg){kind=link}
32962-7&id=gr5.jpg){kind=link}
32962-7&id=gr6.jpg){kind=link}