The animals used in this study were female BALB/c mice (aged 6–10 weeks old with a mean weight of 20 g) that were supplied by the Animal Resource Center of the Health Sciences Center at Kuwait University. The mice were kept under standard conditions, including controlled temperature (25°C ± 1°C), a 12-h light–dark cycle, and free access to food and drinking water. All experiments were approved by the Animal Care Committee at Kuwait University Health Sciences Center and conformed to their rules and regulations as described previously (protocol approval number: P11613PT01) (Khajah et al., 2016a).

Colitis was induced in mice by mixing DSS polymers with the drinking water (3.5% w/v, Cat # 160110, MP Biomedicals) given ad libitum (Wirtz et al., 2007; Khajah et al., 2016a). Control, untreated (UT) mice received tap water. Daily weight changes were determined as a loss of baseline body weight. Colon length and maximal bowel thickness (in millimeters) were also determined (Khajah et al., 2016a). Two approaches were used in this study: a) prophylactic and b) treatment. For the prophylactic approach, mice received daily intra-peritoneal (i.p) injections of minocycline (10–40 mg/kg, Cat# M9511, Sigma) or vehicle (saline) along with DSS administration for 5 days. For the treatment approach, mice received daily DSS administration for 4 days (to first establish a state of active colitis), followed by daily i.p injections of minocycline (1–40 mg/kg), budesonide (0.1–100 μg/kg, Cat # 01901680540,365, AstraZeneca), methylprednisolone (0.01–10 mg/kg, Cat #P6004, Sigma), a combination regimen of minocycline plus either budesonide or methyl prednisolone, or vehicle (saline) for 3 days (without DSS administration). After that, mice were sacrificed (on day 5 after the first administration of DSS for the prophylactic approach and on day 7 after the first administration of DSS for the treatment approach), and the severity of colitis was determined by macroscopic (gross) and microscopic (histologic) assessments.

The macroscopic assessment for colitis severity in all treatment groups was performed as previously described (Khajah et al., 2016a). The macroscopic assessment data are presented in tables as the percentage (%) of mice in each treatment group, showing various features of colitis (edema, erythema, diarrhea, blood in stool, anorectal bleeding, and adhesion).

For the microscopic assessment of colitis severity, formalin-fixed colon tissues (the whole colon) were processed and (blindly) scored by two observers using a standard semi-quantitative histology scoring system, as previously described (Khajah et al., 2016a).

The colonic (descending part) protein expression of phospho/total ERK1/2 (Cat # 4370 and 9102), p38 MAPK (Cat # 9216), phospho-AKT (Cat # 9271), AKT (Cat # 9272), COX-2 (Cat # 4842), SRC (Cat # 5473), BAK (Cat # 12105), caspase-3 (Cat # 9662), the granulocyte marker LY6G (Cat # 87048), E-cadherin (Cat # 3195), and actin (loading control, Cat # 8457) (all from Cell Signaling, United States, 1:1000 dilution) were determined from all groups of mice by Western blotting, as described previously (Khajah et al., 2016b). In brief, colonic tissue was snap-frozen in liquid nitrogen and stored at −80°C until it was needed for further experiments. The colonic tissue samples were pulverized while frozen and then transferred to lysis buffer (pH 7.6) containing 10 mM Tris-base, 140 mM NaCl, 10 mM Na₄P₂O₇, 1 mM NaF, 1 mM CaCl₂, 1 mM MgCl₂, 10% glycerol, 1% NP40 (Tergitol-type NP-40 and nonyl phenoxypolyethoxylethanol), 2 mM Na₃VO₄, and a protease inhibitor cocktail. The tissues were homogenized using Polytron PT 4000 (Switzerland), and the samples were left to lyse completely by incubating them on an ice-cold shaker for 30 min. Subsequently, they were centrifuged at 14,000 rpm for 20 min at 4°C, and the supernatants were collected. Protein concentrations in the supernatants were measured using a Lowry/Bradford protein assay. Aliquots containing equal amounts of protein (10 µg) were subjected to gel electrophoresis and transferred onto nitrocellulose/PVDF membranes. Membranes were then subjected to BSA (2%–4%) for 1 h and then incubated with the appropriate antibodies and subsequently with appropriate secondary antibodies conjugated to horseradish peroxidase. Immunoreactive bands were detected with the SuperSignal chemiluminescent substrate (Pierce, United Kingdom) using Kodak autoradiography film (G.R.I., Rayne, United Kingdom).

Real-time RT-PCR was used to quantify the mRNA of the cytokines, TNF-α, IL-1β, and IL-10, in the colon and blood. The descending part of the colon was dissected and snap-frozen in liquid nitrogen. On the day of RNA extraction, the descending part of the colon was homogenized in 1.5 mL of lysis buffer from the RNeasy^® Mini Kit, containing 15 μL of β-mercaptoethanol. Blood was extracted from the mice by cardiac puncture and added to the RNAprotect Animal Blood Tube (100 μL). The tube with blood was inverted eight to ten times and left standing at 15–25°C for 2 h. The contents were transferred to a 1.5-mL Eppendorf tube. This tube was centrifuged for 3 min at 5,000 × g at 20°C, the supernatant was discarded, and 1 mL of RNase-free water was added and vortexed with the pellet; this was centrifuged for 3 min at 5,000 × g at 20°C, and the supernatant was discarded. Buffer RSB (240 μL) from the RNeasy^® Protect Animal Blood Kit was added to the pellet, vortexed until dissolved, and used for RNA extraction. Total RNA was extracted from homogenized colon samples and prepared into blood samples using the RNeasy^® Mini Kit and RNeasy^® Protect Animal Blood Kit, respectively, following the manufacturer’s protocols, and quantified by spectrometry. The total RNA (500 ng) was used to synthesize cDNA, as previously described (Masocha, 2009). Real-time RT-PCR amplification was performed in triplicate using an ABI Prism^® 7500 Sequence Detection System (Applied Biosystems). A measure of 1 µL of cDNA was added to a 19 µL reaction mixture containing 1× Platinum^® SYBR^® Green qPCR Supermix-UDG, ROX dye, and forward and reverse primers (125–500 nM), and amplified as previously described (Masocha, 2009). The primer sequences that were used (Table 1) were extracted from previous publications from other laboratories (Li et al., 2017) and our laboratory (Parvathy and Masocha, 2013), checked for specificity by BLAST in the GenBank database, and ordered from Invitrogen by Life Technologies. The housekeeping genes 18S ribosomal RNA (18s) and cyclophilin A (Ppia) were used to normalize the amount of transcripts in individual animal samples (ΔCT; n = 3 to 5 per group for blood and n = 7 to 19 for colon). The relative amount of target gene mRNA in the blood and colons was calculated using the previously described 2^−ΔΔCT method (Livak and Schmittgen, 2001).

Data were analyzed using GraphPad InStat and Prism software programs (California, United States). Differences between groups were assessed using the Student unpaired t-test, Mann–Whitney test, one-way ANOVA followed by a Bonferroni post hoc test, or Kruskal–Wallis test, followed by Dunn’s multiple comparisons test, with p < 0.05 being regarded as significant.

A significant drop in body weight from the baseline level (7%) was observed in mice that received DSS plus vehicle compared to UT healthy mice (that received tap water only) (p < 0.05; Figure 1A). Treatment with minocycline (10–40 mg/kg) prevented the DSS-induced weight loss, and the groups treated with minocycline had body weight similar to that of the UT group. A significant decrease in colon length (Figure 1B) and increase in colon thickness (Figure 1C) were observed in mice treated with DSS plus vehicle compared to UT mice (p < 0.05). Treatment with minocycline (20–40 mg/kg) prevented the DSS-induced changes in colon length and width, and the groups treated with minocycline had a colon length and width similar to those of UT mice. No signs of inflammation at the gross level were observed in UT healthy mice (Table 2). DSS plus vehicle administration resulted in edema (100%), erythema (28%), diarrhea (57%), anorectal bleeding (28%), and adhesions (100%). Minocycline treatment at doses of 30–40 mg/kg significantly reduced the presence of edema. Treatment with minocycline at doses of 20–40 mg/kg prevented the development of DSS-induced erythema, diarrhea, anorectal bleeding, and adhesions. Administration of DSS caused a significant increase in the histological score of colitis (Figure 1D) and percentage of ulceration in the whole colon (Figure 1E) compared to UT (p < 0.05). Treatment with DSS plus minocycline (at all doses, but specifically with 20–40 mg/kg) resulted in a significantly reduced histological score of colitis and percentage of ulceration compared to treatment with DSS plus vehicle. Figure 1F shows examples of colon sections taken from UT, DSS plus vehicle, and DSS plus minocycline-treated mice. UT mice have normal colon architecture. DSS plus vehicle resulted in significant destruction of mucosal architecture, immune cell infiltration, submucosal edema, and muscle wall thickness. Treatment with DSS plus minocycline prevented the destruction of the mucosal architecture and muscle thickening, and significantly reduced the degree of immune cell infiltration and submucosal edema.

Treatment with minocycline (10–40 mg/kg) reversed both the DSS-induced drop in body weight (p < 0.05; Figure 2A) and increase in colon thickness (p < 0.05; Figure 2C) but had no significant effects on colon length (Figure 2B). Minocycline treatment at doses of 20–30 mg/kg significantly reduced DSS-induced edema, erythema, diarrhea, blood in stool, anorectal bleeding, and adhesions (Table 3). These signs of DSS-induced inflammation were absent in mice treated with minocycline at a dose of 40 mg/kg. Minocycline treatment (at all doses) significantly reduced the DSS-induced histological score of colitis by 50%–60% (Figure 2D), and a minocycline dose of 40 mg/kg reduced the percentage of ulceration by 60% (Figure 2E). Figure 2F shows examples of colon sections taken from different treatment groups. Minocycline treatment (specifically at 30–40 mg/kg doses) significantly restored mucosal architecture and muscle thickening, with a significant reduction in the degree of immune cell infiltration and submucosal edema.

Budesonide treatment (at all doses) reversed both the DSS-induced drop in body weight (p < 0.05; Figure 3A) and decrease in colon length (p < 0.05; Figure 3B). Treatment with budesonide at doses from 0.5 to 100 μg/kg reversed the DSS-induced increase in colon thickness (p < 0.05; Figure 3C). Budesonide treatment at doses of 0.5–1 μg/kg significantly decreased DSS-induced edema, erythema, diarrhea, blood in stool, anorectal bleeding, and adhesions (Table 4), and these signs of inflammation were absent at budesonide doses of 10–100 μg/kg. Budesonide treatment significantly reduced the DSS-induced histological score of colitis and percentage of ulceration by approximately 50% at a dose of 0.5 μg/kg and by 80%–98% at higher doses of 10–100 μg/kg (Figures 3D, E). Figure 3F shows examples of colon sections taken from different treatment groups. Treatment with budesonide at doses of 1–100 μg/kg significantly restored all these parameters.

Sub-optimal doses of minocycline and budesonide, 10 mg/kg minocycline and 0.5 μg/kg budesonide, were used to study whether there is a synergistic effect between the two drugs in reducing symptoms of colitis in mice. Treatment with minocycline, budesonide, or their combination reversed the DSS-induced drop in body weight (p < 0.05; Figure 4A), decrease in colon length (p < 0.05; Figure 4B), and increase in colon thickness (p < 0.05; Figure 4C). Treatment with individual drugs significantly reduced the DSS-induced edema, erythema, diarrhea, blood in stool, anorectal bleeding, and adhesions, but most importantly, the combination regimen completely reversed all these DSS-induced symptoms (Table 5). Treatment with minocycline or budesonide alone significantly reduced the DSS-induced histological score of colitis and percentage of ulceration, but most importantly, the combination regimen produced a more robust and significant reduction of these parameters than monotherapy (p < 0.05; Figures 4D, E). Figure 4F shows examples of colon sections taken from different treatment groups. Treatment with individual drugs reduced the various inflammatory parameters, but the combination regimen produced a higher degree of these parameters.

Treatment with another corticosteroid, methylprednisolone, at doses of 0.1–5 mg/kg partially reversed both DSS-induced drop in body weight, while a higher dose of 10 mg/kg restored the body weight to levels similar to those of UT mice (Figure 5A). Treatment with methylprednisolone at 10 mg/kg reversed both the DSS-induced decrease in colon length (p < 0.05; Figure 5B) and the increase in colon thickness (p < 0.05; Figure 5C). Methylprednisolone treatment (at doses of 0.1–5 mg/kg) significantly decreased the DSS-induced edema, erythema, diarrhea, blood in stool, anorectal bleeding, and adhesions, and these signs of inflammation were absent at a higher methylprednisolone dose of 10 mg/kg (Table 6). Treatment with methyl prednisolone at doses of 5–10 mg/kg significantly reduced the DSS-induced histological score of colitis and % of ulceration Figure 5D, E. Figure 5F shows examples of colon sections taken from different treatment groups. The various inflammatory parameters were significantly improved with methylprednisolone treatment, and the higher dose of 10 mg/kg eliminated the signs of inflammation.

Sub-optimal doses of minocycline and methylprednisolone, 10 mg/kg minocycline and 5 mg/kg methyl prednisolone, were used to study whether there is a synergistic effect between the two drugs in reducing symptoms of colitis in mice. Treatment with the individual drugs or their combination reversed the DSS-induced drop in body weight (p < 0.05; Figure 6A), while only the combination reversed the DSS-induced decrease in colon length (p < 0.05; Figure 6B) and increase in colon thickness (p < 0.05; Figure 6C). Treatment with individual drugs significantly reduced DSS-induced edema, erythema, diarrhea, blood in stool, anorectal bleeding, and adhesions, but most importantly, the combination regimen completely reversed all these DSS-induced symptoms (Table 7). Treatment with minocycline or methylprednisolone alone significantly reduced the DSS-induced histological score of colitis and percentage of ulceration, but most importantly, the combination regimen produced a more robust and significant reduction of these parameters than monotherapy (p < 0.05; Figures 6D, E). Figure 6F shows examples of colon sections taken from different treatment groups. The various inflammatory parameters were significantly improved with methylprednisolone or minocycline treatment, but the combination regimen completely eliminated these signs of inflammation.

Treatment with various doses of dexamethasone (i.p. and oral gavage routes), another corticosteroid, significantly reduced signs of colitis at the gross but not at the histological level (Supplementary Figure S1; Supplementary Table S1). Therefore, we did not use dexamethasone in the combination regimen with minocycline.

The sub-optimal doses and the route of administration for the two corticosteroids in combination with minocycline are different (budesonide = 0.5 μg/kg oral gavage and methyl prednisolone = 5 mg/kg intraperitoneal injection). The degree of reduction in the histological score of inflammation in the combination regimen vs. monotherapy was more profound with methyl prednisolone (Figure 6D; approximately 69%) compared to budesonide (Figure 4D; approximately 38%). For this, the combination regimen of minocycline plus methyl prednisolone was used for the subsequent experiments to determine the effects of treatment on the colonic expression/activity profiles of various pro-inflammatory and pro-apoptotic molecules.

Next, we used Western blot analysis to determine the possible mechanism by which these drugs reduce DSS-induced inflammation in the colon of mice (Figure 7). Administration of DSS induced a significant increase in phospho-ERK1/2, the pro-inflammatory enzyme COX-2, the granulocyte marker LY6G, BAK, and caspase 3 compared to UT mice, which were all consistently reduced by treatment with the minocycline and methyl prednisolone combination regimen, while the individual drugs had variable effects. There were no differences in the levels of phospho-p38 MAPK, phospho-AKT, phospho-SRC, and E-cadherin in the colon between DSS-treated and UT mice. However, treatment with the combination regimen significantly reduced the phosphorylation of p38 MAPK, AKT, and SRC when compared to UT, vehicle, and monotherapy. The total AKT level was not modulated in all the tested groups. Finally, the combination regimen reduced the expression of E-cadherin when compared to UT, vehicle, and monotherapy.

A pilot study showed that the relative expression of the genes for the cytokines (Il1b, Tnf, and Il10) in the colon exhibited a trend to increase at day 3 after DSS/vehicle treatment, returning to UT levels by day 7 (Supplementary Figure S2).

DSS/vehicle treatment did not alter the expression of the genes of the cytokines Il1b and Tnf in the blood at days 3 and 7 after the first administration of DSS compared to the UT group (Figures 8A, B). Il10 transcripts were lowly expressed in the blood of both control untreated (three out of the five samples undetected) or DSS/vehicle-treated mice (four out of the five samples undetected), and therefore, relative gene expression could not be calculated, while its transcripts were detected in all samples of the colon.

There was a significant increase in Il1b transcript levels in the colon of DSS/vehicle-treated mice compared to UT mice on day 3, while there were no differences in the expression of Tnf and Il10 transcripts between the DSS/vehicle-treated mice and the UT mice (Figure 8C). Treatment with the combination regimen of minocycline and methylprednisolone had no significant effects on the DSS-induced changes in Il1b transcript expression but increased the Tnf and Il10 transcript levels on day 3 (Figure 8C). On day 7, there were no differences in the levels of the transcripts of Il1b, Tnf, and Il10 between the DSS/vehicle-treated mice and UT mice (Figure 8D). Treatment with minocycline and methylprednisolone had no significant effects on the DSS-induced changes in Il1b and Tnf transcript expression but increased the Il10 transcript levels on day 7 (Figure 8D).