Isostearic acid is an active component of imiquimod formulations used to induce psoriaform disease models

Topical imiquimod based creams are indicated as immune stimulants for papillomas and various skin neoplasms. Imiquimod is considered a TLR7 ligand. These creams are also used in research to induce skin inflammation in mice as a model for psoriasis. We observed that this inflammatory response was not strictly imiquimod dependent and we set out to establish which components drive the proinflammatory effects. To this end, we examined the induction response in a BALB/cJRj mouse model, in which 50 mg of cream is applied to 2 cm2 of skin (125 mg/kg imiquimod—5% W/V, and/or 625 mg/kg isostearic acid—25% W/V). Comparing cream formulations containing isostearic acid, imiquimod and the combination, we observed that isostearic acid causes skin inflammation within 2 days, whereas imiquimod requires up to 5 days for initial signs. Isostearic acid activated an inflammasome response, stimulated release of proinflammatory cytokines and upregulated the IL-23/17 axis. Animals treated with isostearic acid had enlarged livers (+ 40% weight), which was not observed with imiquimod alone. Imiquimod was readily metabolized and cleared from plasma and liver, but was maintained at high levels in the skin throughout the body (200 mM at area of application; 200 µM in untreated skin). Imiquimod application was associated with splenomegaly, cytokine induction/release and initial body weight loss over 3 days. Despite high imiquimod skin levels throughout the animal, inflammation was only apparent in the treated areas and was less severe than in isostearic acid groups. As the concentrations in these areas are well above the 10 µM required for TLR7 responses in vitro, there is an implication that skin inflammation following imiquimod is due to effects other than TLR7 agonism (e.g., adenosine receptor agonism). In brain, isostearic caused no major changes in cytokine expression while imiquimod alone sightly stimulated expression of IL-1β and CCL9. However, the combination of both caused brain induction of CCL3, -9, CXCL10, -13, IL-1β and TNFα. The implication of these data is that isostearic acid facilitates the entry of imiquimod or peripherally secreted cytokines into the brain. Our data suggest that psoriaform skin responses in mice are more driven by isostearic acid, than generally reported and that the dose and route used in the model, leads to profound systemic effects, which may complicate the interpretation of drug effects in this model.


Introduction
Psoriasis is a chronic skin disease, that causes visible and painful inflammation and hyperproliferation of skin. Its etiology is not fully understood but includes genetic, environmental, infectious and lifestyle factors (Global Report on Psoriasis 2023; Girolomoni et al. 2017). In Western countries around 1-2% of the population is affected by various degrees of psoriasis (Nakajima and Sano 2018). Treatment options include topical, systemic, and photo-therapy (Global Report on Psoriasis 2023). Systemic treatments include antibodies (e.g. inhibitors of interleukin-23: ustekinumab, guselkumab, tildrakizumab, and Risankizumab; TNFα: etanercept, adalimumab; IL-17: secukinumab (AIN457), ixekizumab (LY2439821), and brodalumab (AMG827) (Yang et al. 2021;Zhang et al. 2022) as well as small molecules (e.g., glucocorticoids, methotrexate, cyclosporine and fumarates). The central goal of therapy, is to reduce psoriatic activity without generally suppressing the immune system. In this regard, the introduction of therapy directed to IL-17 and IL-23 has led to substantial improvements in quality of life.
Finding improved therapies for psoriasis requires disease models, that recapitulate the disease mechanisms either in combination or alone. The involvement of TLR receptor, therefore, provides at least one means to easily mimic disease in WT animals. Van der Fits et al. first proposed in 2009 a murine model for psoriasis, based on repeated topical application of Imiquimod (IMQ) formulated as Aldara ® (Fits et al. 2009), which activates murine TLR7, the proposed target of IMQ. This activation leads to a proinflammatory response similar to human acute phase psoriasis. IMQ containing cream (clinical formulation of Aldara ® ), is applied on the depilated dorsal skin of mice on consecutive days to induce the IL-17/23 axis (Fits et al. 2009;Brück et al. 2018;Flutter and Nestle 2013). Clinical signs include local thickening, scaling, redness and inflammation of the skin. This leads to psoriasis like lesions: thickening of the epidermis through hyperproliferation of keratinocytes, a reduced layer of granules (hypogranulosis), redness of skin because of dilation of blood vessels, accumulation of neutrophils in stratum corneum (Munro's microabscesses), and infiltration from clusters of CD4 + , CD8 + and antigen presenting DCs in dermis and epidermis (Nakajima and Sano 2018;Brück et al. 2018). The Aldara ® induced psoriasis model is widely used because of its rapid development, low cost and simple read out. However, although operationally straightforward, the molecular biology and immunology of the model is less simple (Walter et al. Mar. 2013;Horváth et al. 2019;Hawkes et al. 2017;Luo et al. 2016). IMQ also acts systemically causing splenomegaly, lymphadenopathy and possibly elevated systemic cytokines (Hawkes et al. 2017) and is likely to also be an adenosine receptor inhibitor (Fits et al. 2009;Schön and Schön 2007;Schön et al. 2006). Walter et al. (2013) showed that isostearic acid, (25% by weight of Aldara ® ), also has proinflammatory effects in vitro and in vivo (Walter et al. 2013). Although isostearic acid is thought to be included for its physical formulation properties (it is a weak acid while imiquimod is a weak base), it appears to be immunologically and systemically active. Thus, isostearic acid present in the formulation of Aldara ® may have a functional role irrespective of the initial intentions behind its inclusion.
We have been employing the model to study the effects of anti-inflammatory drug candidates for pharmaceutical developers. In the course of such studies, we have observed that compounds that should influence the model, were sometimes more or less active than expected, depending on the mode of action. We also observed that in topical formulations of certain compounds, there were cases of significant effects of the vehicle on the performance of Aldara ® as an inducer of disease. To better control the model, we then set out to create a "sham cream" that did not contain IMQ, that could be used as a better control vs. the IMQ induced disease. We used this preparation, in which IMQ could be added to a final concentration of 5% W/V which is the same as that in the original imiquimod containing Aldara ® . Studies with this material showed that it produced a different disease pattern to that of Aldara ® . Further studies demonstrated that isostearic acid was active alone in this formulation, and in some ways synergistic to the effects of IMQ, especially locally. The formulated in-house creams are based on DAC cream (40% H 2 O, 25.5% Vaseline, 10% propylene glycol, 7.5% triglyceride, 7% Macrogol-20-glycerol-monostearate, 6% cetyl alcohol, 4% 60-glycerol-monosterate) (Höfel and "Basiscreme DAC" 2018).
Here we demonstrate that the isostearic acid is a major contributor to the local skin inflammatory reaction, whilst IMQ tends to act only in combination with isostearic acid. This gives a better understanding of the Aldara ® induced psoriasis model and the modes of action that may be active in the model. We demonstrate a clear effect of isostearic acid and/or IMQ to inflammatory response, spleen and liver weight and scaling, redness and thickness of the skin of female BALB/cJRj mice. Finally, we quantified the effects of the combination on uptake to skin and distribution systemically. These observations may be useful to researchers in the field attempting to interpret results obtained with this in vivo model. In particular, the knowledge that the model is based on multiple disease inducers may explain cases where treatments with a very specific mode of action do not fully suppress disease signs.

Materials and methods
All chemicals, consumables and supply products were purchased from commercial sources and used as received. All study protocols were approved by the local Animal Care and Ethics Committee (Federal government ethics committee, Tübingen, Germany).

HPLC-MS/MS and MS/MS studies
Quantification of analytes (imiquimod and isostearic acid; MS tunes in Table 1) was performed with an Agilent 1260 Infinity system coupled to a triple quadrupole Sciex API 4000 MS/MS detector. An Agilent C 18 Poroshell 120 column (4.6 × 50 mm, 2.7 µm) was used for separation. The mobile phase was composed of water containing 0.1% formic acid (eluent A) and acetonitrile containing 0.1% formic acid (eluent B). Gradient used was: 10% B for 1 min, to 100% B in 2 min, 100% B for 3 min, to 5% B in 1 min, 5% B for 3 min.

Real-time polymerase chain reaction (qPCR)
Analysis of gene expression was done by isolating RNA from flash frozen skin samples, using the Qiagen RNeasy ® MiniKit (Qiagen). 3.6 µg of RNA was used for reverse transcription into complementary DNA (cDNA) using Per-feCTa DNase I (Quanta Bioscience) and 5× PrimeScript RT Mastermix (Takara). Gene-specific primers used for qPCR are listed in Table 2. The qPCR reaction was performed in duplicates with Blue S'Green qPCR Mix Separate ROX (Biozym) in 96-well plates. Amplification was carried out in the QuantStudio ® 3 qPCR system (ThermoFisher Scientific/ Quantstudio™ Design & Analysis Software v.1.4.3). Cycle conditions were set at 95 °C for 2 min followed by 40 cycles of 95 °C for 5 s and 60 °C for 17 s; a melt-curve analysis was performed for each qPCR reaction. Primer efficiency was determined for each primer pair to ensure linear standard curve and high amplification efficiency. Threshold cycle (C t ) values were used for the calculation of the relative gene expression levels. For this, all values were first normalized to respective C t values of the HPRT (hypoxanthine phosphoribosyltransferase) housekeeping gene. Then, normalized ΔC t values of the different treatment groups were compared to the ΔC t levels of the vehicle treated control group which was set to 1. Normalized expression ratios (2(− ∆∆C t )) are given for all treatment groups relative to the vehicle group. C t values of all samples were normalized to the respective C t values obtained for HPRT housekeeping gene, resulting in the ∆C t value for each sample: ∆C t (gene x) = C t (gene x) -C t (HPRT). To calculate the ∆∆C t value, the ∆C t average of the control condition (vehicle) was used. This analysis was also performed for the individual control condition samples to determine variance within the control group. Subsequently, ∆∆C t values of each sample were transformed into normalized expression ratio which gives the fold increase (or decrease) of the target gene expression compared to the control condition (vehicle) and normalized to the reference gene (HPRT). Normalized expression ratio (fold change) = 2− ∆∆C t . Data are shown as scatter plots using GraphPad Prism 9.1.0.

Cream-induced psoriatic murine model
All study protocols were approved by the local Animal Care and Ethics Committee (Federal government ethics committee, Tübingen, Germany under the license 35/9185.81-7/ SYN 07/18). A total of 88 mice (BALB/cJRj, female, 8 weeks old) were used for the studies performed (divided into two separate studies. First study with 56 animals and second study with 32 animals) were purchased from Janvier Labs (Le Genest-Saint-Isle). Animals were acclimatized for at least one week with standard chow (Mouse Maintenance, V1534-000, Ssniff Spezialdiäten GmbH, Germany) and drinking water ad libitum before the start of the experiment. Animals were housed in type IV cages, with bedding and enrichment material and kept at 23 °C (± 1 °C), 50-60% humidity, with 12/12 h dark/light cycles period. Animals were monitored daily and at the end of the experiment, terminated painlessly with overflow of CO 2 . For the first optimization study, mice were grouped in 8 animals per cage. Induction was performed with following creams: Aldara ® (MEDA Pharma GmbH & Co. KG; Aldara ® here referred as clinical cream or clinical formulation), Cetyl alcohol + isostearic acid, 5% IMQ + cetyl alcohol, 7% IMQ + cetyl alcohol, 7% IMQ + cetyl alcohol + isostearic acid, 5% IMQ + DMSO or 5% IMQ + DMSO + isostearic acid (see also  Table 4). For the second optimization study, mice were grouped in 4 animals per cage and the following creams were used: vehicle DAC base cream, 5% IMQ + DMSO, Isostearic acid or 5% IMQ + DMSO + isostearic acid. On day 0, prior to psoriasis induction, dorsal skin was removed using a clipper, followed by application of depilatory cream for complete fur removal. From day 1 to day 6 (first study) or 7 (second study) 50 mg of each induction cream (see cream formulation section) was applied to 2 cm 2 of the depilated skin of the animals. The required amount of cream was weighed and applied using a spreader to achieve homogeneity. During the experiment, daily measured parameters were: body weight (BW), skin thickness, skin redness and scaling (0 = no change, 1 = marginal effects, 2 = moderate effects, 3 = strong effects, 4 = maximum). At termination, pictures of dorsal skin were taken (ante mortem), skin samples (for histology and gene expression) were collected and the weights of spleen and liver were recorded. Additional organ samples, including skin, were collected, washed to remove remnant cream, flash frozen, and kept at -20 °C until processing to determine tissue distribution of imiquimod and isostearic acid. Skin samples were kept in 4% paraformaldehyde solution until processing.

Sample preparation for pharmacokinetic assessment
Frozen organ samples were thawed and treated as described earlier (Straß et al. 2021). Briefly, samples were digested with Proteinase K in a buffer volume in µL corresponding to sample weight in mg (Proteinase K, VWR, diluted 1:40 in phosphate buffer 20 mM) for 1 h

Histology
Paraffin-embedded skin sections were stained using the hematoxylin and eosin (H&E) method. Briefly, skin samples were fixed for 24 h in 4% PFA and then transferred to 70% ethanol until tissue preparation took place. Skin samples were placed into cassettes for the subsequent embedding process. Dehydration and clearance were by sequential incubation steps: 90% ethanol for 45 min, 100% ethanol for 45 min (2×), 100% ethanol for 60 min, 100% isopropanol for 60 min (2×), 100% isopropanol, overnight. Next day, organs were transferred into liquid paraffin for 30 min (55-65 °C) and cooled overnight at room temperature, to complete the embedding in paraffin. After deparaffinizing and hydration steps from xylol to several ethanol dilutions and H 2 O, 8 µm sections of the tissue samples were stained with Mayer's hematoxylin to stain nuclei and Eosin Y 1% aqueous solution to stain cytoplasm (H&E staining). H&E-stained tissue was used to perform histopathological evaluation and quantitative analysis (derma and submucosa depth) using QuPath 3.0 software. Ki67 immunofluorescent staining was performed on formalin-fixed and paraffin-embedded skin samples previously cut on a microtome into 5 µm thick sections. The sections were deparaffinized in xylol and hydrated in decreasing concentrations of ethanol. Antigen retrieval was performed in heated citrate buffer (pH 6.0) and, after cooling to room temperature, Sudan Black buffer (Sigma-Aldrich 199664-25G; 0.1% in 10% ethanol; pH 7.4) was used to inhibit erythrocytes autofluorescence and goat serum (Vector Laboratories) was used as blocking solution. After sections were washed in 1× PBS (pH 7.4) 3 times, 5 min each, they were incubated with specific primary unconjugated antibody (BioLegend, rat anti-mouse, 652402), 2h, 37 °C. After sections were washed in 1× PBS (pH 7.4) 3 times, 5 min each, the specific secondary antibody (NovusBio, TX-RED, NBP1-731 49) was used in the reaction and, after nuclear counterstaining with DAPI glue mounting step, skins samples were scanned with KEYENCE BZ-X810 fluorescence microscope and investigated for further analysis. The positive stained cells were counted (per µm 2 ) for total epidermal immunostaining in the same 20× magnification field area and the% of the intensity was calculated vs the DAPI counter staining.

Statistics
All experimental results were first tested for normal distribution using Shapiro-Wilk for normality. Statistical analysis between multiple groups was done using two-way ANOVA followed by Bonferroni correction. Kruskal-Wallis test was used for non-parametric data. Non-marked bars are considered as not significant by which statistical significance was considered at the level of p value < 0.05. Calculations of e.g., half-life were made with GraphPad Prism 9.3.1.

Differences to clinical products in induction of skin inflammation
The widely used Aldara ® induced psoriatic skin model is not fully defined or understood (Horváth et al. 2019;Hawkes et al. 2017;Luo et al. 2016;Walter et al. 2013). We focused on the influence of components of imiquimod cream formulations. To study this, we induced skin inflammation in BALB/cJRj mice with different induction creams (Tables 3,  4). Isostearic acid may promote inflammasome activation, IL-1 release and apoptosis (Walter et al. 2013). Although DMSO is known to increase skin permeation, making substances more systemically available, this effect is not seen with IMQ (Telò et al. 2016). DMSO is used here to facilitate the incorporation of the substance into the cream and to overcome the poor solubility of IMQ (DMSO is permitted in human use up to 10% in creams).
In a first study, we tested different formulations for their ability to induce a psoriatic-like skin disease in mice. All of the creams used were mixed with inducing substances (either IMQ, isostearic acid or a mixture of the two) and none of the creams are considered to be vehicle creams. The usual concentration of IMQ in Aldara ® or similar products is 5% W/V, however, we also tested higher concentrations (7% IMQ) to assess the effect on rate of onset of skin signs. Creams were applied 25 mg/cm 2 (50 mg total) corresponding to a total of 125 mg/kg IMQ (5%) or 175 mg/kg IMQ (7%) daily over seven consecutive days. In this initial experiment with different cream formulations (Table 3) the first changes in skin score were observed from day three on (Fig. 1A). Groups receiving isostearic acid were the first to show clinical signs in skin. Animals induced with Aldara ® had a score plateau starting at day 5 (day 5-7 score of 5.9), while other groups induced with isostearic acid containing creams did not reach a plateau, but showed a reduced rate in score development (cetyl alcohol + isostearic acid: 5.5, 6.6 and 7.0; 7% IMQ + cetyl alcohol + isostearic acid: 5.6, 6.9 and 7.1; 5% IMQ + DMSO + isostearic acid: 5.3, 6.8 and 7.3 at days 5-7).
The cream containing isostearic acid and cetyl alcohol, but not IMQ showed a similar increase in score (2.6, 4.3, 5.5, 6.6, 7.0 from days 3-7), whilst the other groups treated with creams containing IMQ, but not isostearic acid, showed only minor signs in the first five to six days and a minor increase in skin score by day 7 (5% IMQ + cetyl alcohol: 3.4; 7% IMQ + cetyl alcohol: 2.8; 5% IMQ + DMSO + isostearic acid: 2.1 at day 7). Variation of cetyl alcohol or DMSO and the amount of IMQ did not change the scores to a significant extent. Consistent with literature reports, IMQ treatment in our studies also leads to a loss in body weight (Fig. 1B) (Horváth et al. 2019) peaking around day 3 (clinical cream 94%; 5% IMQ + cetyl alcohol: 98%; 7% IMQ + cetyl alcohol: 93%; 7% IMQ + cetyl alcohol + isostearic acid: 98%; 5% IMQ + DMSO: 98%; 5% IMQ + DMSO + isostearic acid: 101%) and suggesting that there was no difference in the availability of IMQ from the various formulations (see Fig. 1E, for lesions). Isostearic acid caused a severe reaction in terms of redness, scaling and thickness, however, without major loss in BW (tendency of loss in BW from day 4 on). IMQ also caused splenomegaly (Fig. 1D) with Aldara ® induced animals, having the largest spleens (avg. spleen weights: cetyl alcohol + isostearic acid 102 mg; 7% IMQ + cetyl alcohol 206 mg; 5% IMQ + DMSO 212 mg; 5% IMQ + cetyl alcohol 214 mg; 5% IMQ + DMSO + isostearic acid 216 mg; 7% IMQ + cetyl alcohol + isostearic acid 224 mg; Aldara ® 272 mg). Increased spleen size is linked to IMQ containing cream and is probably due to systemic inflammatory reactions (Fits et al. 2009;Horváth et al. 2019). Larger liver size (hepatomegaly) was observed for animals receiving creams containing isostearic acid (Fig. 1C). Creams without isostearic acid had liver weights similar to those in control animals. Mean liver weight of the Aldara ® -treated group was between the two clusters (avg. liver weights: 7% IMQ + cetyl alcohol 1044 mg; 5% IMQ + DMSO 1065 mg; 5% IMQ + cetyl alcohol 1126 mg; Aldara ® 1307 mg; cetyl alcohol + isostearic acid 1494 mg; 5% IMQ + DMSO + isostearic acid 1528 mg; 7% IMQ + cetyl alcohol + isostearic acid 1568 mg). Isostearic acid in the induction cream showed a direct effect on the weight of the liver independent of BW suggesting that it has a systemic effect, either via absorption by the topical route or via ingestion during grooming. The effects of IMQ on body weight were lost by day 7, consistent with a continuous attenuation of cytokine response after repeated application of IMQ or Resiquimod (data not shown) we observed in other studies. We assume that the observed In graphs A and B groups were compared against Aldara ® and tested with ordinary two-way ANOVA with Bonferroni's comparison. In graphs C and D Kruskal-Wallis test with Dunn's comparison was used after Shapiro-Wilk test for normality. Data represented as means with error bars as SD. p values ≤ 0.0001 are indicated with ****p ≤ 0.001 are indicated as ***p ≤ 0.01 are indicated as **p ≤ 0.05 are indicated as * body weight pattern is an effect of the cytokine attenuation and the well-known reduction in TLR stimulation responses with time (Horváth et al. 2019). The data from this study suggests that isostearic acid is playing a major role in the induction of skin inflammation in mice. To clarify aspects of the induction of psoriasis-like inflammation further, we investigated the key parameters in more detail in a second study.

Direct influence of isostearic acid on development of psoriasis in mice
We then directly compared IMQ (5%) containing cream with isostearic acid (25%) containing cream and a cream containing both substances versus a vehicle cream based on DAC (Table 4). Female BALB/cJRj mice were again treated with induction creams on seven consecutive days. Each animal (n = 8 per group) received 50 mg of cream dorsally on depilated skin (2 cm 2 ). Consistent with the results of the first study we observed changes in the skin areas treated with isostearic acid containing creams. Isostearic acid alone was sufficient to induce visible signs of skin inflammation ( Fig. 2A, F). The onset of the disease in the isostearic acid containing creams was again at the third day and reached a plateau at day five until end of study (isostearic acid 4.4, 4.4, 3.8; 5% IMQ + DMSO + isostearic acid 4.3, 4.6, 4.8 at days 5-7). Scores for scaling in group 3 (isostearic acid alone) were reduced on day seven, when compared to day six (from 4.4 to 3.8) This was not observed for the IMQ/isostearic combination, which had a slightly higher total score on the final study day. Both isostearic acid containing creams showed a very similar course of induction. In animals treated with vehicle cream no changes were observed. Group 2 (IMQ containing cream) showed almost no changes over the course of the study. Individual animals of this group showed very slight thickening and redness of the skin, but significantly less than in the isostearic acid groups. As observed in our first study it appears that isostearic acid is the main stimulant of changes in skin and the development of lesions. This can be observed for all three parameters (thickness, redness and scaling). However, in contrast to the first study, no additive induction with IMQ was found here. In general, this study showed a milder course of the disease. A maximum score of 4.8 was observed for the IMQ/isostearic acid combination (isostearic acid alone 3.8 at final day). In the first study isostearic acid reached 7.0, Aldara 5.9, and 7.2 (combination of 5% IMQ + DMSO + isostearic acid) and 7.1 (7% IMQ + cetyl alcohol + isostearic acid) for the IMQ/isostearic acid combinations, respectively. Photographs taken on the final day of the studies (Figs. 1E, 2E) provide further qualitative impressions of the signs at termination.
Increased liver weights in groups that received isostearic acid and increased spleen weights in groups that were treated with IMQ containing creams (Fig. 2C, D), confirmed observations from the first study. Body weight distribution (Fig. 2B) did not follow a clear trend. Initially, the animals induced with IMQ tend to lose body weight (95% at day 3), but a slight loss is also seen in the vehicle animals (96% BW at day 4). After four to five days, animals from IMQ induced group recover BW (97% at day 4 and 103% at day 5), while the animals induced with isostearic acid start to lose weight continuously displaying the lowest body weight values at the end of the study (95% at final day for isostearic acid group and 90% for IMQ and isostearic acid group). The same trend in weight loss (from day 4) was observed for animals induced with isostearic acid in the first study.
Differences in the development of skin parameters are apparent by histology (Fig. 2H). Skin treated with isostearic acid showed significant thickening in several layers. These trends are most notable in the thickness of the dermis, but not in the submucosa (Fig. 2G). Immunohistochemical staining of Ki67, a marker of cellular proliferation, was performed with skin samples from the final day of the study (Fig. 2H). Ki67 staining suggested lower proliferation for groups receiving isostearic acid.
These data suggest that visible skin lesions and hepatomegaly in BALB/cJRj mice are mostly driven by isostearic acid. This is surprising given that it is generally considered to be safe in most reports (CIR 2013) (up to 37.4% in rinseoff products and 21% in leave-on products) and may suggest a specific effect in mice or a consequence of the high concentration and/or overall dose-625 mg/kg or 6.25 mg/ cm 2 -in a small animal study in proportion to the skin area affected. Splenomegaly and proinflammatory cytokines appear to be due to TLR7 activation through IMQ (Horváth et al. 2020;Jabeen et al. 2020;Butchi et al. 2008). While IMQ and isostearic acid contribute to the disease model, the effects on the conventional signs of keratinocyte proliferation may be more related to isostearic acid in mice.

Upregulation of cytokine markers in course of disease
To investigate how the components of the induction creams influence the cytokine response we collected samples of brain and skin from animals at day 7 of both studies. Figure 3 displays results from skin samples collected in the first study (other results of this study are displayed in Fig. 1). Figure 4 displays results from skin samples collected at final day of second study (other results of this study are displayed in Fig. 2). Brain samples were collected to check systemic inflammation response of chemokines (Fig. 5) (McColl et al. 2016;Nerurkar et al. 2017). Samples gathered in first study were collected on day seven, 27 h after last application of creams when induction effects may have been reduced (Fig. 3). Fig. 2 A Accumulated skin score based on the parameters: redness, thickness and scaling (scores each between 0 and 4, with 4 being most severe). BALB/cJRj females (8 weeks old, n = 8 per group) were treated with different formulations of induction cream (50 mg) for seven consecutive days. Scored was dorsal skin and first changes were observed on day 3. B Change of body weight during study vs day 0 (n = 8). C Weights of livers at day of termination (n = 8). D Weights of spleens at day of termination (n = 8). E Pictures of dorsal skin of one representative animal of each group at day 7 ante mortem (scale bar represents 100 µm). F Skin score of parameters redness, thickness and scaling (scores each between 0 and 4, with 4 being most severe) (n = 4). G Analyzed histological parameters from dorsal skin collected at final day of study (n = 4). H H&E staining of skins collected at final day of study. I Analyzed Ki67 positive cells in dorsal skin collected at final day of study (n = 4). J Ki67 staining (TxRed) on DAPI nuclear counterstaining captured at ×40 (scale bar represents 100 µm). In graphs A, B and F groups were compared against vehicle cream and tested with ordinary two-way ANOVA with Bonferroni's comparison. In graphs C and D Kruskal-Wallis test with Dunn's comparison was used after Shapiro-Wilk test for normality. Data represented as means with error bars as SD. p values ≤ 0.0001 are indicated with ****p ≤ 0.001 are indicated as ***p ≤ 0.01 are indicated as **p ≤ 0.05 are indicated as *Groups were compared with vehicle cream Although isostearic acid caused the highest skin score a higher inflammatory reaction at a gene expression level was observed in IMQ induced dorsal skin samples in this experiment, and was highest in 7% IMQ induced samples where a dose dependence between cytokines and IMQ concentration can be observed. Addition of DMSO had no effect on cytokine expression. Clinical cream was more effective on IL-17A and IL-17F cytokine expression than other 5% IMQ formulations without increasing TNFα or IL-6 signal (Fits et al. 2009). These results show a tendency to induce psoriasis relevant proinflammatory cytokines, which indicates a role of IMQ in this model. However, effects may depend on the time between substance application and measurement. In this study, we chose a period of 27 h after the last application (overnight) which reflects conditions in published models and potentially a means to observe the "chronic" or "trough" expression levels following repeated stimulus.
In the second study, we sampled 2 h after the last application of cream and we assessed a wider range of relevant targets, (Fig. 4) including inflammasome related markers, IL-1β and NLRP3 which are influenced in the model (Walter et al. 2013). TNFα and IL-6 were slightly upregulated while other markers (IL-17A, IL-17F, IL-23A and IL-22) were significantly upregulated. The IL-17 family cytokines and IL-22 are known to play a central role in clinical psoriasis (Yang et al. 2021). The relative effects of IMQ and isostearic acid were differed as in previous reports (see Fig. 3 and Fits et al. 2009), however, these samples were taken 2 h after the final application of the cream which may increase effects of isostearic acid and explain the variations in the results. The implication is that responses to isostearic acid are shorter Fig. 3 Levels of IL-17A, IL-17F, TNFa and IL-6 mRNA in dorsal skin samples collected 27 h after the last application of cream. Fold change compared to untreated control. Fold changes were compared to Aldara ® . Kruskal-Wallis test with Dunn's comparison was used after Shapiro-Wilk test for normality. Data represented as means with error bars as SD (n = 8). p values ≤ 0.0001 are indicated with ****p ≤ 0.001 are indicated as ***p ≤ 0.01 are indicated as **p ≤ 0.05 are indicated as * in duration than those to IMQ and possibly indicate a more local proinflammatory reaction.
In addition to skin samples, we analyzed samples from brain (Fig. 5) collected at the same time as skin samples (Fig. 4). However, unlike skin, changes in gene expression in the brain can only be seen in animals receiving the combination cream. Systemic IMQ has been shown to influence chemokines in brain tissue (Nerurkar et al. 2017), but here we show that only the combination is able to induce these chemokines via the topical route. Disruption of the blood-brain-barrier by one of the compounds could play a role (Nerurkar et al. 2017;Bourgognon et al. 2019), however, substance concentrations in brain tissue are neither elevated for IMQ nor isostearic acid in groups receiving the combination (Fig. 6) suggesting that disruption is unlikely. The main markers that we found to upregulated in brain (chemokines, IL-1β, TNFα) were not those that were upregulated in skin (IL-22, IL-23). The chemokines are closely connected with virus-induced neuroinflammation (Butchi et al. 2008). Nerurkar et al. (2017) were able to measure elevated levels of chemokine markers after topical treatment with Aldara ® . They explained their results by a priming phenomenon due to peripheral immune stimulation with IMQ. The concentration of IMQ in the brain tissue that we report here was low, compared with a study published by Butchi et al. (2008), where IMQ was directly injected into the brain. These data suggest the hypothesis, that the combination of IMQ elevates chemokine levels without overcoming the blood-brain-barrier (see Fig. 6 for IMQ brain concentration).

Retention of substances in skin
We assessed animals for distribution of IMQ and isostearic acid 2 or 27 h after last application of creams (both after 7 days of consecutive application of cream) on dorsal skin using HPLC-MS/MS. Previous studies showed that IMQ is rapidly distributed to tissue with a particular affinity to skin (oral and dermal application) (Jabeen et al. 2020;Pharmaceuticals 2009;Paula et al. 2008). Although, IMQ is quickly metabolized by a first-pass effect, topical application prolongs systemic exposure (Pharmaceuticals 2009). IMQ is able to penetrate into skin and it is retained in the stratum corneum (Jabeen et al. 2020;Paula et al. 2008). The local concentration can increase over six days with continued application (Jabeen et al. 2020).
In dorsal inflamed skin we observed high concentrations of IMQ (20 mM in the IMQ group and 13 mM in IMQ/ isostearic acid group) 2 h after cream application (Fig. 6).
Lower concentrations were also detected in healthy (noninflamed) ventral skin (0.4 mM in the IMQ group and 1 mM in IMQ/isostearic acid group). For IMQ/isostearic acid cream, concentrations were also measured 27 h after application. IMQ concentrations were significantly lower in both, dorsal (0.2 mM) and ventral skin (0.15 mM). Concentrations in the two areas almost equalized over this period. Other organs showed IMQ concentrations in lower µM ranges. Ranges, nonetheless, high enough to stimulate proinflammatory responses based on known TLR7 affinity. No major changes were observed in IMQ levels in organs other than skin over the course of time. The skin might act as a IMQ depot and release it over time and thus generates a prolonged systemic stimulation. The fact that levels in ventral skin were as high as in dorsal skin (site of application) after 27 h, suggest that either it was re-distributing within skin or via grooming, or that it was accumulating from the systemic circulation. The levels in the ventral skin were not associated with any form of obvious "psoriatic" transformation and this is, in turn, consistent with the observation that IMQ itself is less effective in causing psoriaform lesions. Quantification of isostearic acid (Fig. 6, lower panel) showed ca. 70 mM in dorsal skin where the combination cream was applied. Local concentrations in dorsal skin are elevated 2 h after application. The contribution of remnant cream on the skin is difficult to assess because overly vigorously washing could also leach material from the skin itself, nonetheless, rinsing steps ensure that detected levels are from skin associated material. Absorbed cream is distributed to organs, like spleen and liver, where concentrations reflect the treatment type. Naturally occurring "stearic acid isomers" were detected in nontreated animals (parent masses and retention times being similar). Brain levels of isostearic acid were not elevated in combination with IMQ. The high levels of Isostearic acid in dorsal skin could act as a local proinflammatory stimulus as concentrations are substantially elevated for hours.

Discussion
There is often a degree of time sensitivity in pharmacology studies with short duration models favored over longer versions for reasons of efficiency, ethics or simplicity. Often, shorter duration is obtained with higher dosed challenges or, indeed, more complex challenges. In developing a model based only on IMQ we faced questions about not only the degree of severity (50-60% of clinical formulations) but also duration (9-10 days instead of 5-7). These data suggest that a "pure" IMQ "psoriasis" model can be conducted, it requires only more time. While it is unlikely that simply inhibiting TLR7 signaling would be sufficient for a therapeutic effect in human patients, we suspect that the transition to skin pathology with IMQ is not strictly driven by TLR7 agonism alone. Rather, we propose that the effects in skin may be related to inhibition of the adenosine receptors A(1) and A(2A) in addition to or in concert with cytokine increases related to TLR7 agonism (Schön et al. 2006;Wolff et al. 2013). Indeed, in such a mixed challenge model (IMQ/ isostearic acid), a number of pathways are activated locally and systemically. Some effects are caused by multiple pathways, and others are specifically related to one challenge agent. Thus, it should not be assumed that compounds with a "single specific mode of action" should influence all aspects of pathology unless they impact an extremely central regulator. Action on a very central regulator is not always desirable in terms of safety (see glucocorticoids). Rather, assessing a range of signals will help deconvolute which aspects a given therapy can influence. It may also lead to assessment in a more focused model where there is less complex activation of different pathways.
Finally, mouse strain selection may play a greater role than is often appreciated and while many models are conducted in the C57BL/6 background, others like this one Fig. 6 Organ samples collected at end of IMQ-induced psoriasis study (n = 8), 2 or 27 h after last cream application. Samples were analyzed with HLPC-MS/MS. Top row shows measured concentrations of IMQ in organs. Bottom row shows results for concentration of isostearic acid in organs use BALB/cJRj for the simple convenience of more easily observing the skin score and reddening. However, contrasting results have been obtained in C57BL/6 mice, including the suggestion that high doses of IMQ (156 mg/kg) cause systemic vs. local signs which are reported to be just as apparent at 50 mg/kg IMQ with the lower dose actually associated with more significant local score (Horváth et al. 2019). These data may suggest a form of "spillover" to the systemic circulation at high doses that may be in excess of that required for local effect, especially when skin appears to accumulate or retain the substance. In this context, we will continue to examine effects of the two most eliciting components of clinical imiquimod formulations in both mouse strains, taking account of the effects of melanin and other C57BL/6 immune attributes in follow-up studies.
While clinical imiquimod formulations provide a convenient means to induce psoriasiform like lesions in mice with IL-17 involvement, the composition in relation to the mode of action is potentially more complex than is widely appreciated. It is important to recall that the original proposed use was to stimulate rejection of warts, and more recently, clinical imiquimod formulations have demonstrated efficacy in rejection of early skin cancers (Rosso 2005). Biochemical investigations suggest a range of effects of IMQ, which was widely considered to be the active principle in the induction of psoriaform like lesions in mice (TLR7 agonism, adenosine receptor inhibition and interactions with the hedgehog pathway) (Schön et al. 2006;Wolff et al. 2013). However, in human use, while redness/erythema is noted, effects are rarely "psoriasiform" and isostearic acid is considered an uncontroversial ingredient in human cosmetics (CIR 2013;Bergfeld et al. 2018). The difference between human use and murine uses is almost certainly first the dose. Human use would rarely exceed one sachet or 250 mg of cream-12.5 mg of imiquimod (assuming use of the full sachet), or 0.5 mg/cm 2 or 0.18 mg/kg. In contrast, the dose commonly used to induce the model in mice is between 62.5 and 156 mg/kg-350-870-fold higher. The dose for isostearic acid would be 62.5 mg or 2.5 mg/cm 2 , or 0.9 mg/kg in human patients. In mice, it is of the order of 625 mg/kg (ca. 700fold higher) or 6.25 mg/cm 2 . These differences in overall dose probably explain both systemic spillover in mice and more profound local effects. Accounting for metabolic rate, the exposure in mouse is still at least 100-fold over that which is common in human use. On the other hand, a lower sensitivity to TLR-agonist induced inflammation can be observed in mice with doses used in murine models showing intolerable adverse effects in clinical studies (see Resiquimod) (Pockros et al. 2007). This further points to the possibility that the contributions individual components of clinical IMQ formulations have on inflammation differ significantly between a murine and human setting.
Data reported here demonstrate that isostearic acid appears to have an important role in stimulating initial skin signs in the psoriasiform model. In contrast, IMQ appears to take longer to induce skin effects and is instead rather potent in cytokine induction. Our observations suggest that the clinical formulation is, in many ways, optimal in terms of cytokine induction, and compositions with the same amount of IMQ and isostearic acid differ in subtle ways from the formulation sold as Aldara ® in disease induction. Nonetheless, should workers wish to use a defined, non-commercial, preparation, the cream formulations we describe can at least be reproduced in their functional properties with and without the key ingredients. This should make it more straightforward to deconvolute drug effects.

Conclusion
The induction of psoriaform skin inflammation in mice is not solely due to a stimulus by IMQ. Isostearic acid, one of the main components of clinical formulations is a key trigger in the disease model. Studies using clinical creams for induction need to take this into account. Further analysis, especially regarding cellular mechanisms, should elucidate the exact role of isostearic acid in vivo in mice.