Related Experiment Video
Updated: Jul 7, 2026
Xenograft Skin Model to Manipulate Human Immune Responses In Vivo
Published on: June 29, 2022
CXCR3 <-> ligand-mediated skin inflammation in cutaneous lichenoid graft-versus-host disease
Joerg Wenzel1, Svenja Lucas, Sabine Zahn
1Department of Dermatology at the Universitie of Bonn, Bonn, Germany. joerg.wenzel@ukb.uni-bonn.de
Background:
Lichenoid graft-versus-host disease (liGVHD) histologically shares several common features with other lichenoid dermatoses, such as cutaneous lupus erythematosus and lichen planus (LP), which collectively show a junctional infiltrate of cytotoxic lymphocytes with liquefaction of the basal layer ("interface dermatitis"). Because recent studies have shown a role for type I interferon (IFN)-associated inflammation, including lymphocyte recruitment via CXCR3 <-> ligand interaction in cutaneous lupus erythematosus and LP, we hypothesized that similar mechanisms might also be involved in liGVHD.
Methods:
Ten representative lesional skin biopsies taken from patients with different subsets of chronic cutaneous graft versus host disease (GvDH) were recovered from the authors' archives. Eight LP specimens and 5 punch biopsies taken from healthy skin were analyzed for control purposes. Immunohistochemistry was performed to characterize the lesional infiltrate (CD3, CD4, CD8, CD20, CD56, or CD68), to analyze type I IFN signaling (MxA), and to investigate expression of the IFN-inducible chemokines CXCL9 and CXCL10 and their ligand CXCR3. In situ hybridization was performed to visualize IFNalpha expression on the mRNA level.
Results:
Our analyses revealed striking similarities between the inflammatory pattern seen in LP and liGVHD. Both disorders presented with a predominantly T-cellular inflammation with CD8(+) lymphocytes affecting the basal epidermal layer. The majority of lesional lymphocytes expressed the chemokine receptor CXCR3. The corresponding chemokines CXCL9 and CXCL10 were found in the epidermis and within the inflammatory infiltrate. Analyses of MxA and IFNalpha mRNA expression supported a role for type I IFNs in these conditions.
Limitations:
This study was limited by the number of well characterized cases in our archives. In situ hybridization was realizable only in single cases.
Conclusion:
Our results support the hypothesis that CXCR3 <-> ligand-mediated lymphocyte recruitment is involved in cutaneous liGVHD. The fact that CXCL10 was seen in precisely those areas with extensive liquefaction of the basal epidermis supports a role of this chemokine for the development of the typical histologic "interface" pattern.
