This study is the first to elucidate the acute effects of long-term IMQ exposure using a multimodal approach in healthy volunteers. We have shown that compared to short exposure, long exposure to IMQ results in a stronger immunological response as evidenced by additional enriched pathways such as TLR-induced IRF7 signalling, more prominent TNF signalling via NF-κB along with downstream effects such as induction of type I and type II interferons leading to activation of the JAK-STAT pathway. Furthermore, increased complement gene expression was identified upon long exposure to IMQ[7, 22]. Although imaging and biophysical measurements showed no significantly enhanced response after long IMQ application compared to short exposure, a strong cellular infiltration boost was observed. At the transcriptomic level, this was demonstrated by an increased abundance of M1 macrophages, resting NK cells, and resting and activated CD4 + memory T cells. While a statistically significant increase was observed in M2 macrophages and CD8+ T cells following IMQ application, no difference was observed between long and short exposure. The increased abundance of naïve B cells and the appearance of T cells is indicative of both innate and adaptive immune responses involvement. The transcriptomic profile partially aligns with the IHC-based cellular infiltration of macrophages, NK cells and CD4+ T cells, demonstrating clear time-dependent effects with increased infiltration after long IMQ exposure. Additionally, IMQ increased the expression of type II interferon-related genes, which aligns with the IHC observation of CD8+ cell influx. These cellular findings are consistent with classical TLR signalling. Activation of TLRs is also known to trigger MyD88, IRAK1 and IRAK4, leading to IRF7 and NF-κB signalling, which is in line with our findings. These pathways result in upregulation of transcription factors for several cytokines including type I interferons, TNF, IL-2, IL-6, IL-8, IL-12, IFN-α and chemokines such as macrophage inflammatory protein (MIP)-1α, MIP-1β and monocyte chemotactic protein-1[24].
Another challenge agent that we use to effectively induce an in vivo TLR response in men is lipopolysaccharide (LPS)[25–28]. Intradermal injection of LPS triggers an acute inflammatory response via TLR4, leading to increased innate immune cell populations including neutrophils, monocytes and dendritic cells. Furthermore, LPS elicits an adaptive immune response, as evidenced by the presence of B and T cells. Elevated levels of IL-6, IL-8, IL-1β and TNF following LPS injection indicate NF-κB involvement. The current study showed moderate activation of NF-κB signalling after long IMQ application, supported by upregulated expression of NF-κB1, NF-κB2, IL-6, CXCL8, CCL2, IL-17C and IL-23A at the transcriptomic level. This contradicts previous studies, as no significant IL-6 and IL-8 responses were observed 72 hours after IMQ application, suggesting only mild NF-κB involvement. However, the release of Mx-A (a downstream marker indicative of IFN-α activation through IRF7) was evident[7, 8]. Our current data reinforces this finding, as it suggests both My-D88 transcript expression and downstream IRF7 and Mx-1 activation, which increases upon prolonged IMQ application. The activation of interferons leads to the engagement of their respective receptors, which in turn triggers the JAK - STAT pathway culminating in the release of proinflammatory cytokines[29, 30]. The JAK - STAT pathway was more enriched, with chemokines such as CXCL9, CXCL10 and CXCL11 significantly overexpressed after prolonged IMQ exposure. Recently, Chen et al. provided an overview of studies examining DNA and RNA specific profiles in cutaneous lupus erythematosus (CLE) patients, which indicates an upregulation of innate immune response functions including JAK – STAT signalling, TLR signalling, and pattern recognition receptors. Furthermore, there was a notable increase in the expression of type 1 interferons, along with an upregulated expression of chemokines CXCL9, CXCL10, and CXCL11, which are recognized as characteristic indicators for CLE[31–34]. Although a direct comparison of our data with the existing RNA datasets of CLE patients was not conducted, analysis of pathway activity and overexpressed genes detected upon prolonged IMQ application allows us to conclude that our current model aligns more closely with CLE characteristics than with psoriasiform lesions[7]. In addition, the histopathological changes of a vacuolar interface dermatitis with adnexal involvement were also reminiscent of CLE. These observations differ from the prevalent use of the model in preclinical studies, where the murine IMQ model is typically used to investigate psoriasis-like conditions[35].
In contrast to our cellular observations, imaging and biophysical measurements showed no significantly enhanced response after long IMQ application compared to short exposure. This may be because Aldara (besides its role as a TLR7 agonist) may also act as an exogenous mediator by enhancing transient receptor potential vanilloid 1 channel activity on the primary afferent sensory neuron[36]. Activation of this channel leads to the release of bioactive substances such as nitric oxide (NO). NO can then interact with target cells in the surrounding tissue, including vascular smooth muscle cells. The interaction of NO with smooth muscle cells leads to vasodilation, resulting in increased blood perfusion and erythema[40]. Our observations suggest that the vascular response is independent of the inflammatory process. The role of bioactive, vasodilating substances in the IMQ-induced erythema and perfusion response remains to be further elucidated.
Our second objective was to explore the translational value of the IMQ model. In contrast to the mouse data, IHC staining did not show involvement of complement in the human IMQ response. We hypothesize that this may be due to the difference in severity of the hit, as in mice, the entire surface area of the back is challenged, whereas in humans IMQ is applied to a much smaller relative surface area. Therefore, it is currently unknown whether the observed differences are a result of the magnitude of TLR activation or can be contributed to species differences[1, 2]. However, at the transcriptome level, classical, and alternative pathway genes were enriched, particularly after prolonged exposure to IMQ, suggesting complement involvement in the human IMQ response. It is unclear how these sequencing results translate to the protein level, or if the transient nature of complement explains the lack of IMQ-driven complement responses in IHC analysis. The same holds true for the observed lack of neutrophils and CXCL8 expression, which emphasizes that the role of neutrophils in the human IMQ response needs to be elucidated further[6]. The observed differences between preclinical animal models and the human response may partially be explained by species differences in TLR7 expression[37]. For instance, Bhagchandani et al. described that the expression of TLR7 on neutrophils is higher in mouse than in men[37, 38]. Moreover, expression patterns of TLR7 within a given cell type may differ across tissues and across activation status of the cell[39], further complicating the translational interpretation both across and within species. These findings highlight the complementary value of human challenge models in the development of immune-targeting compound development.
In conclusion, our study provides a comprehensive characterization of the cutaneous response to both short and prolonged IMQ exposure in healthy volunteers by using a multimodal approach. We have demonstrated that prolonging the IMQ exposure has added value by enhancing cellular responses and increasing abundance of specific immune cell types along with stronger activation of a diverse set of pathways, particularly those driven by IRF and related to complement. We also argue that prolonged IMQ application results in a CLE-like cutaneous inflammation, both at the transcriptomic level and from a histopathological perspective. Our results suggest that biophysical and vascular responses are not exclusively driven by cutaneous inflammation. The described discrepancies between preclinical and clinical results, most notably the neutrophil response, illustrate the complementary value of human challenge models in the development of compounds targeting the immune system. This in vivo immune challenge model is of value for future early clinical evaluation of topically or systemically applied anti-inflammatory or immunomodulatory compounds, particularly compounds targeting IRF and JAK-STAT signalling.