International Journal of Molecular Sciences Article The Local Activation of Toll-like Receptor 7 (TLR7) Modulates Colonic Epithelial Barrier Function in Rats Javier Estévez 1 and Vicente Martínez 1,2,3, * 1 Department of Cell Biology, Physiology and Immunology, Universitat Autònoma de Barcelona, 08193 Barcelona, Spain 2 Neuroscience Institute, Universitat Autònoma de Barcelona, 08193 Barcelona, Spain 3 Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd), Instituto de Salud Carlos III, 28049 Madrid, Spain * Correspondence: vicente.martinez@uab.es; Tel.: +34-093-581-3834 Abstract: Toll-like receptors (TLRs)-mediated host–bacterial interactions participate in the microbial regulation of gastrointestinal functions, including the epithelial barrier function (EBF). We evaluated the effects of TLR7 stimulation on the colonic EBF in rats. TLR7 was stimulated with the selective agonist imiquimod (100/300 µg/rat, intracolonic), with or without the intracolonic administration of dimethyl sulfoxide (DMSO). Colonic EBF was assessed in vitro (electrophysiology and permeability to macromolecules, Ussing chamber) and in vivo (passage of macromolecules to blood and urine). Changes in the expression (RT-qPCR) and distribution (immunohistochemistry) of tight junction- related proteins were determined. Expression of proglucagon, precursor of the barrier-enhancer factor glucagon-like peptide 2 (GLP-2) was also assessed (RT-qPCR). Intracolonic imiquimod enhanced the EBF in vitro, reducing the epithelial conductance and the passage of macromolecules, thus indicating a pro-barrier effect of TLR7. However, the combination of TLR7 stimulation and DMSO had a detrimental effect on the EBF, which manifested as an increased passage of macromolecules. DMSO alone had no effect. The modulation of the EBF (imiquimod alone or with DMSO) was not associated with changes in gene expression or the epithelial distribution of the main tight junction-related proteins (occludin, tricellulin, claudin-2, claudin-3, junctional adhesion molecule 1 and Zonula occludens-1). No changes in the proglucagon expression were observed. These results show that Citation: Estévez, J.; Martínez, V. The TLR7 stimulation leads to the modulation of the colonic EBF, having beneficial or detrimental effects Local Activation of Toll-like Receptor depending upon the state of the epithelium. The underlying mechanisms remain elusive, but seem 7 (TLR7) Modulates Colonic independent of the modulation of the main tight junction-related proteins or the barrier-enhancer Epithelial Barrier Function in Rats. factor GLP-2. Int. J. Mol. Sci. 2023, 24, 1254. https://doi.org/10.3390/ Keywords: epithelial barrier function; host–bacterial interactions; imiquimod; intestinal permeability; ijms24021254 tight junctions; TLR; Toll-like receptors Academic Editor: Nitin Telang Received: 9 December 2022 Revised: 30 December 2022 1. Introduction Accepted: 4 January 2023 Published: 9 January 2023 The disruption of the epithelial barrier function (EBF) with increased intestinal perme- ability is a common finding in several gastrointestinal disorders, including inflammatory bowel disease (IBD) and irritable bowel syndrome (IBS) [1–4]. The intestinal epithelium rep- resents a functional barrier between the lumen and the internal milieu, allowing selective Copyright: © 2023 by the authors. exchange processes (secretory and absorptive) and acting at the same time as a protective Licensee MDPI, Basel, Switzerland. barrier against luminal antigens [5]. Gut commensal microbiota, considered for long time to This article is an open access article have been a passive component of the gastrointestinal tract, is now considered as a dynamic distributed under the terms and player in intestinal homeostasis [6–9]. Gut microbiota and microbial-derived products act conditions of the Creative Commons as luminal factors that interact continuously with the host. In physiological conditions, Attribution (CC BY) license (https:// these interactions contribute to the maintenance of intestinal homeostasis. However, in creativecommons.org/licenses/by/ states of dysbiosis, host–microbial interactions change and might lead to the activation of 4.0/). Int. J. Mol. Sci. 2023, 24, 1254. https://doi.org/10.3390/ijms24021254 https://www.mdpi.com/journal/ijms Int. J. Mol. Sci. 2023, 24, 1254 2 of 17 the immune system and the generation of a persistent inflammatory state with functional changes, including alterations in the EBF [10]. In this sense, dysbiosis and altered epithelial permeability are co-existent features of IBS and IBD [1,3,11–15]. Host–microbial interactions are largely mediated through pattern recognition receptors (PRRs). PRRs are expressed within the gut and recognize conserved microbial components named pathogen-associated molecular patterns (PAMPs) [16]. In particular, recognition of gut microbiota depends largely on the interaction with a subgroup of PRRs named Toll-like receptors (TLRs), with up to 13 members described in mammals (TLR1–13) [17,18]. The relative expression of the different TLRs differs depending upon the cell type and the cellular localization [18,19]. Within the gut, some of the predominant TLRs are TLR2, TLR3, TLR4, TLR5 and TLR7, being present in several cell types, including the epithelial cells [17]. TLRs directly interact with luminal microbial components and, upon activation, elicit neuroimmune responses that affect intestinal functions, including the EBF [20–25]. For instance, we have shown that during antibiotic-induced dysbiosis, host–bacterial interactions are altered, including changes in the expression of TLRs, and there are local changes in immune- and sensory-related systems within the gut [26,27]. Similar alterations might be elicited when TLRs are over-stimulated, mimicking a state of dysbiosis with altered host–bacterial interactions. In this sense, we have shown that direct stimulation of colonic TLR7 with the selective agonist imiquimod leads to immune- and sensory-related changes similar to those observed during intestinal dysbiosis [28,29]. Moreover, imiquimod reduced the serum levels of inflammatory cytokines and ameliorated clinical signs and intestinal lesions during DSS-induced colitis in mice [30]. Taking these observations into account, the aim of the present work was to directly assess the effects of colonic TLR7 stimulation on the EBF in rats. For this, we assessed the effects of the over-stimulation of colonic TLR7 with the selective agonist imiquimod on epithelial electrical parameters and on epithelial permeability to macromolecules in in vitro (the Ussing chamber system) and in vivo conditions. In order to characterize the underlying mechanisms mediating potential TLR7-dependent modulation of the EBF, we also assessed changes in the expression and epithelial distribution of the main tight- junction (TJ)-related proteins (occludin, tricellulin, claudin-2, claudin-3, junctional adhesion molecule 1 -JAM-1- and zonula occludens-1 -ZO-1-). Moreover, changes in the expression of the barrier-enhancer factor glucagon-like peptide 2 (GLP-2, assessed through the expression of its precursor, proglucagon) [31] were also assessed. Finally, the induction of a colonic inflammatory-like state associated with the over-stimulation of TLR7 was assessed at the macroscopical and molecular levels. 2. Results 2.1. The Acute Addition of Imiquimod to the Ussing Chamber Did Not Affect Colonic Electrical Parameters or Permeability to Macromolecules After stabilization, the short circuit current (Isc), conductance (G) and potential dif- ference (PD) had a value of −26.1 ± 2.4 µA cm−2 , 23.3 ± 1.7 ms cm−2 and −1.2 ± 0.1 mV, respectively (n = 24 tissue samples from 5–7 animals). These values are similar to those previously described by us in equivalent experimental conditions [12,32]. Neither the apical nor the basolateral addition of imiquimod (125 µM) to the Ussing chamber affected the basal electrical parameters (Figure 1). Similarly, the passage of FD4 was not affected by IMQ (125 µM, either basolateral or apical; Figure 1). Int. J. Mol. Sci. 2023, 23, x FOR PEER REVIEW 3 Int. J. Mol. Sci. 2023, 24, 1254 3 of 17 6 A 4 B 4 2 Isc (A/cm2)  G (ms/cm2) 2 0 0 -2 -2 -4 -6 -4 Veh-Ap IMQ-Ap Veh-Bl IMQ-Bl Veh-Ap IMQ-Ap Veh-Bl IMQ-Bl 1.0 C D 0.07 Passage of FD4 (%) 0.06 Veh-Ap 0.5 0.05 Veh-Bl  PD (mV) 0.04 IMQ-Ap 0.0 IMQ-Bl 0.03 -0.5 0.02 0.01 -1.0 0.00 Veh-Ap IMQ-Ap Veh-Bl IMQ-Bl 0 20 40 60 80 100 Time (min) Figure 1. The effects of the acute addition of imiquimod (125 µM, IMQ) or its vehicle (0.5% hydrox- Figure 1. The ypropylmethyl cellulose, Veh)effects to the of the acute Ussing addition chamber, of to either imiquimod the apical(125 (Ap)µorM,the IMQ) or its vehicle basolateral (Bl) (0.5% hy ypropylmethyl cellulose, Veh) to the Ussing chamber, either to the side, on the colonic epithelial electrical parameters and permeability to macromolecules. (A–C) the apical (Ap) or the basolater side, on the colonic epithelial electrical parameters and permeability to macromolecules. (A– net change in epithelial electrical parameters: short circuit current (Isc , panel A), conductance (G, net change in epithelial electrical parameters: short circuit current (Isc, panel A), conductan panel B), potential difference panel (PD, panel B), potential C). Each difference (PD,point panelrepresents a colonic C). Each point sheet, and represents the horizontal a colonic sheet, and the hori line with errors line shows the mean ± SEM (n = 5–7 colonic sheets from 5 animals with errors shows the mean ± SEM (n = 5–7 colonic sheets from 5 animals per group). Thepernet group). T change (∆) was change obtained as the difference between the value before and the value after the (∆) was obtained as the difference between the value before and the value after the 660 min exposure to imiquimod. exposure to (D) Effects of (D) imiquimod. imiquimod Effects ofon epithelialon imiquimod permeability to macromolecules. epithelial permeability to macromolecule The graph shows graph shows the the mucosal tomucosal to basolateral basolateral passage (as passage (as a percentage a percentage of theadded of the amount amount to added the to the sal of mucosal reservoir) reservoir) fluorescein of isothiocyanate-dextran fluorescein isothiocyanate-dextran 4 kD (FD4) in4tissues kD (FD4) in tissues exposed to Stheexposed vehicle to Sthe v (Veh)(125 (Veh) or imiquimod or imiquimod µM, IMQ), (125 µ M, either IMQ), to the either apical to or (Ap) thethe apical (Ap) or (Bl) basolateral the side. basolateral Data are (Bl) side. Da mean ± SEM, n = 4–7 colonic sheets from 3–6 animals per group. mean ± SEM, n = 4–7 colonic sheets from 3–6 animals per group. 2.2. The Over-Stimulation 2.2. The Over-Stimulation of Colonic TLR7ofwith Colonic TLR7 with Imiquimod DoesImiquimod Not TriggerDoes an Not Trigger an Inflamm Inflammatory-like Response within the Colon like Response within the Colon Regardless of the experimental Regardless of thegroup, no signs of experimental colonic group, noinflammation wereinflammation signs of colonic observed wer upon examination of the colon at the time of necropsy. Similarly, no change served upon examination of the colon at the time of necropsy. Similarly, in colonic no chan relative weightcolonic or the colonic relative expression weight or of inflammatory the markersof(IL-6 colonic expression and IFNα1)markers inflammatory was (IL- observed among groups. Likewise, regardless of the treatments applied, IFNα1) was observed among groups. Likewise, regardless of the treatmentsno changes in applie TLR7 expression were detected changes in TLR7(relative expression: expression vehicle, 1.07 were detected ± 0.22;expression: (relative imiquimod 100 µg: 1.07 ± vehicle, 1.73 ± 0.48; imiquimod 300 µg, 0.78 ± 0.09; p = 0.601, ANOVA). imiquimod 100 µ g: 1.73 ± 0.48; imiquimod 300 µ g, 0.78 ± 0.09; p = 0.601, ANOVA). 2.3. Pretreatment with Imiquimod Altered the Colonic Epithelial Conductance and Reduced 2.3. Pretreatment Paracellular Permeability In Vitrowith Imiquimod Altered the Colonic Epithelial Conductance and Reduced Paracellular Permeability In Vitro In colonic samples obtained from animals exposed to imiquimod (300 µg/rat, intra- In colonic samples colonic) the tissue conductance obtained was slightly from animals decreased, exposed although to imiquimod statistical (300 µ g/rat, significance was not reached (p = 0.087 vs. vehicle-treated tissues). A lower dose (100 µg/rat, intra- signifi colonic) the tissue conductance was slightly decreased, although statistical was noteffect. colonic) was without reached (p =epithelial Other 0.087 vs.electrical vehicle-treated tissues). parameters wereAnotlower dose by affected (100 theµ g/rat, in pre-treatment lonic) was without with imiquimod effect.2A–C). (Figure Other epithelial electrical parameters were not affected b pre-treatment with imiquimod (Figure 2A–C). respectively, when compared with vehicle pre-exposed tissues (Figure 1D). In general, an increase in the short circuit current (Isc) in response to a single concen- tration of carbachol (CCh) (100 µ M), added at the end of the experimental protocol, was observed regardless of the treatment applied. Occasionally (less than 10% of the prepara- Int. J. Mol. Sci. 2023, 24, 1254 tions) the tissues were discarded because of the lack of response to CCh, as indicative 4 of 17 of tissue damage. Figure 2. Effects of pre-exposure to imiquimod (IMQ) (intracolonic) on the colonic epithelial electrical Figure 2.and parameters Effects of pre-exposure permeability to imiquimod to macromolecules. (IMQ) (A–C) (intracolonic) Epithelial on the electrical colonic epithelial parameters: electri- conductance cal parameters and permeability to macromolecules. (A–C) Epithelial electrical parameters: con- (G, panel A), potential difference (PD, panel B) and short circuit current (Isc , panel C). Each point ductance (G, panel A), potential difference (PD, panel B) and short circuit current (Isc, panel C). Each represents a colonic sheet, and the horizontal line with errors shows the mean ± SEM (n = 15–28 point represents a colonic sheet, and the horizontal line with errors shows the mean ± SEM (n = 15– colonic sheetssheets 28 colonic from 9–16 fromanimals per group). 9–16 animals (D) Effects per group). of imiquimod (D) Effects of imiquimodon epithelial permeability on epithelial to permeability macromolecules. The graph shows the mucosal to basolateral passage (as percentage to macromolecules. The graph shows the mucosal to basolateral passage (as percentage of the of the amount added to theadded amount mucosal reservoir) to the mucosal ofreservoir) fluoresceinofisothiocyanate-dextran 4 kD (FD4) in 4control fluorescein isothiocyanate-dextran conditions kD (FD4) in control conditions (vehicle (vehicle and pre-treatment) pre-treatment) and in tissues in tissues pre-exposed pre-exposed to imiquimod. to imiquimod. Data ± SEM, are the meanData are the mean ± n = 6–12 SEM,sheets colonic n = 6–12 colonic from sheets from 3–6 animals 3–6 animals per group. per vs. *: p < 0.05 group. *: p < 0.05 vs. Vehicle. Vehicle. InImiquimod 2.4. imiquimod pre-exposed Enhances Colonictissues, the apical-to-basolateral Permeability flux after Disruption of the of FD4 was Epithelial reduced Barrier in In in aVivo dose-related Conditionsmanner. At the end of the experimental time (1 h) the passage of FD4 was reduced byconditions In basal 28.5% (p >(animals 0.05) andtreated 62.2% intracolonically (p < 0.05) for thewith doses theofvehicle), 100 µg and 300 µg, of the passage respectively, when compared with vehicle pre-exposed tissues (Figure 1D). FD4 to blood and urine, as well as the accumulation in the colonic wall, was detected in a In general, an reproducible increase manner. in the short with Pre-treatment circuit current (Iscimiquimod intracolonic ) in response to aµ single (300 concen- g/rat) did not af- tration of carbachol (CCh) (100 µM), fect the passage of FD4 (Figure 3). added at the end of the experimental protocol, was observed regardless of the treatment applied. Occasionally (less than 10% of the prepara- Intracolonic imiquimod followed by a challenge of the colonic mucosa with DMSO tions) the tissues were discarded because of the lack of response to CCh, as indicative of resulted in an enhanced accumulation of FD4 in the colonic wall, with a 2-fold increase in tissue damage. plasma levels (p < 0.05 vs. animals without DMSO) and an 8.5-fold increase in the FD4 2.4. Imiquimod Enhances Colonic Permeability after Disruption of the Epithelial Barrier in In Vivo Conditions In basal conditions (animals treated intracolonically with the vehicle), the passage of FD4 to blood and urine, as well as the accumulation in the colonic wall, was detected in a reproducible manner. Pre-treatment with intracolonic imiquimod (300 µg/rat) did not affect the passage of FD4 (Figure 3). concentration in urine (p < 0.05 vs. animals without DMSO) (Figure 2). DMSO, per se, re- sulted in a slight, non-significant, increase in the passage of FD4 to blood and urine (Fig- Int. J. Mol. Sci. 2023, 24, 1254 5 of 17 ure 3). 19 A Percentage of FD4 (%) 16 13 10 7 4 1 Veh IMQ Veh IMQ SSF DMSO Concentration of FD4 (µg/mL serum) 1.5 B ** 1.0 0.5 0.0 Veh IMQ Veh IMQ SSF DMSO Concentration of FD4 (µg/mL urine) 100 C ** 80 60 40 20 0 Veh IMQ Veh IMQ SSF DMSO Figure 3. Effects of imiquimod (IMQ) on colonic permeability to FD4 in in vivo conditions. Data Figure show3.the Effects of imiquimod accumulation of FD4(IMQ) on theoncolonic colonic permeability wall (A) and thetopassage FD4 in of in FD4 vivofrom conditions. colon toData blood show (µg of FD4/mL serum, (B)) and urine (µg of FD4/mL urine, (C)) during a 30-min periodtoinblood the accumulation of FD4 on the colonic wall (A) and the passage of FD4 from colon control (µ g of FD4/mL serum, (B)) and urine (µ g of FD4/mL urine, (C)) during a 30-min period in control conditions and after treatment with imiquimod with or without DMSO. Data are mean ± SEM, conditions and after treatment with imiquimod with or without DMSO. Data are mean ± SEM, n = n = 9–11 animals per group. **: p < 0.01 vs. other groups (ANOVA). Veh: Vehicle. SSF: Saline. DMSO: 9–11 animals per group. **: p < 0.01 vs. other groups (ANOVA). Veh: Vehicle. SSF: Saline. DMSO: Dimethyl Dimethyl sulfoxide. sulfoxide. Intracolonic imiquimod followed by a challenge of the colonic mucosa with DMSO 2.5. Effects of Imiquimod on Tight Junction-Related Proteins and Barrier-Modulators resulted in an enhanced accumulation of FD4 in the colonic wall, with a 2-fold increase The expression in plasma levels (pof the main < 0.05 TJ-related vs. animals proteins without (occludin, DMSO) and antricellulin, claudin-2, 8.5-fold increase clau- in the FD4 din-3, JAM-1 andinZO-1) concentration urinewas (p