Immunostimulatory Endogenous Nucleic Acids Perpetuate Interface Dermatitis-Translation of Pathogenic Fundamentals Into an In Vitro Model

Abstract

Interface dermatitis is a histopathological pattern mirroring a distinct cytotoxic immune response shared by a number of clinically diverse inflammatory skin diseases amongst which lichen planus and cutaneous lupus erythematosus are considered prototypic. Interface dermatitis is characterized by pronounced cytotoxic immune cell infiltration and necroptotic keratinocytes at the dermoepidermal junction. The initial inflammatory reaction is established by cytotoxic immune cells that express CXC chemokine receptor 3 and lesional keratinocytes that produce corresponding ligands, CXC motif ligands 9/10/11, recruiting the effector cells to the site of inflammation. During the resulting anti-epithelial attack, endogenous immune complexes and nucleic acids are released from perishing keratinocytes, which are then perceived by the innate immune system as danger signals. Keratinocytes express a distinct signature of pattern recognition receptors and binding of endogenous nucleic acid motifs to these receptors results in interferon-mediated immune responses and further enhancement of CXC chemokine receptor 3 ligand production. In this perspective article, we will discuss the role of innate nucleic acid sensing as a common mechanism in the perpetuation of clinically heterogeneous diseases featuring interface dermatitis based on own data and a review of the literature. Furthermore, we will introduce a keratinocyte-specific in vitro model of interface dermatitis as follows: Stimulation of human keratinocytes with endogenous nucleic acids alone and in combination with interferon gamma leads to pronounced production of distinct cytokines, which are essential in the pathogenesis of interface dermatitis. This experimental approach bears the capability to investigate potential therapeutics in this group of diseases with unmet medical need.

Keywords: damage associated molecular patterns (DAMPs); dermatomyositis; in vitro model; interface dermatitis/lichenoid tissue reaction; lichen planus; lupus erythematosus; nucleic acid sensing; type I immunity.

Figure 1

(A) Representative histological findings in interface dermatitis and healthy skin. Representative findings of DNA, MxA, CXCL10, and CXCR3 immunostaining in healthy skin and interface dermatitis (lichen planus). Original magnification x200 (x400 concerning DNA). Black arrows highlight extranuclear localization of DNA motifs. Boxes highlight digitally enlarged aspects. (B) Schematic presentation of assumed etiopathological mechanisms of interface dermatitis (as reviewed above). ID is characterized by a Th1 cytokine milieu in which endogenous nucleic acids activate PRRs. Downstream signaling unleashes cell death cascades (cdc) and leads to production of type-I and -III IFNs and pro-inflammatory cytokines as well as inflammasome activation. Interferon-inducible chemokines (produced by keratinocytes upon autocrine IFN-stimulation) recruit CXCR3 + effector cells into lesional skin, which induce keratinocyte perishing and thus release of pro-inflammatory cell components. (C) Schematic presentation of our in vitro model of interface dermatitis. Nucleic acids extracted from keratinocytes and IFNy are administered to different keratinocyte models (HaCaT, HEK, epiCS) as an ID-like stimulus. Genetic modification of the cells of interest can be made prior to stimulation in order to evaluate specific components of ID pathogenesis. Furthermore, the effect of innovative pharmaceuticals on ID-like stimulation can be analyzed. Super-/undernatants and the cellular compartment are available to read out methods.

Figure 2

(A) Expression of upregulated genes involved in ID pathogenesis in HEK cells stimulated with eNA (12,5 µg/ml) for 24 h compared to HEK cells solely exposed to medium (control), (n = 4, fold change > 2, p < 0.01, Partek^® Flow^®). (B) CXCL10 levels within the supernatant of HaCaT cells stimulated with eNA (5 µg/ml) or IFNy (1 x 10^3 U/ml) or the combination of both for 20 h compared to HaCaT cells solely exposed to medium (control), MTT assay executed on the cells corresponding with the respective supernatant, mean of controls defined as 100%. Given are respective means with standard deviations indicated by error bars (n = 4, * indicates significance (p < 0.05), Mann Whitney test). (C) CXCL10 levels within the supernatant of HEK cells stimulated with the combination of eNA (5 µg/ml) and IFNy (1 x 10^3 U/ml) for 6.5 h compared to HEK cells solely exposed to medium (control), MTT assay executed on the cells corresponding with the respective supernatant, mean of controls defined as 100%. Given are respective means with standard deviations indicated by error bars (n = 4, * indicates significance (p < 0.05), Mann Whitney test). (D) Representative findings in 3D epidermis equivalents upon control (medium) settings and stimulation with eNA (5 µg/ml) and IFNy (1 x 10^3 U/ml) for 22 h. Hematoxylin and eosin stain, MxA, CXCL10. Original magnification ×400. (n = 3).