INTRODUCTION

The renin-angiotensin system (RAS) is a critical regulator ofblood pressure but it has several other functions bothsystemically and locally. For example, RAS activation promotesinflammation and fibrosis in many tissues, including theintestine (1, 2). The majority of RAS’s blood-pressure elevatingand pro-inflammatory properties are mediated by angiotensin II(Ang II) and its receptor, angiotensin II receptor subtype 1(AT1R). Ang II is produced from angiotensin I mainly by theactivity of angiotensin-converting enzyme (ACE). RASparticipates in the regulation of gastrointestinal inflammation (1,3) and has been linked to human inflammatory bowel diseases(IBD) (4-7). The expression of angiotensinogen is reduced inactive disease sites in ileal Crohn’s disease (6) but certainangiotensinogen (AGT) and ACE gene variants are moreprevalent in IBD patients (5, 8). The levels of angiotensin I andII are increased in the colons of Crohn’s disease patients (4),which could contribute to fibrosis in IBD via Ang II-mediatedinduction of transforming growth factor beta1 (TGF-b1) (2). Thepro-inflammatory ACE - Ang II - AT1R arm is generally referredto as the classical RAS. Components of an alternative RAS,ACE2 - angiotensin 1-7 (Ang 1-7) - Mas receptor, are considered

anti-inflammatory, and their levels are elevated in the blood ofIBD patients (9).

The chemically-induced inflammation in DSS colitis isclaimed to resemble human ulcerative colitis (UC)morphologically and symptomatically and offers interestinginsights to IBD pathogenesis, nevertheless also with markeddifferences (10). Pharmacological inhibition of classical RAS iscapable of alleviating even severe colitis in experimental animalmodels. Several drugs which were originally developed asantihypertensives, including ACE inhibitors, e.g. enalapril (11),enalaprilat (12-15), captopril (16, 17) and lisinopril (17) as wellas angiotensin receptor blockers (ARBs), valsartan (18),candesartan (19, 20), losartan and its analogs (15, 20-24),olmesartan (25) and telmisartan (26) have been demonstrated toprevent and alleviate dextran sodium sulfate (DSS) andtrinitrobenzene sulfonic acid (TNBS) -induced colitis in miceand rats. In addition, knockout of angiotensinogen (21) orangiotensin II receptor 1 (AT1R) (27) has been demonstrated toprotect mice from colitis, whereas mice overexpressing reninare more susceptible to TNBS-induced colitis (24).Furthermore, Ang 1-7 administration has exerted beneficialeffects on DSS-induced colitis (28) and in experimental modelof gastric ulcers (29).

JOURNAL OF PHYSIOLOGY AND PHARMACOLOGY 2018, 69, 4www.jpp.krakow.pl | DOI: 10.26402/jpp.2018.4.07

H. SALMENKARI1, L. PASANEN1, J. LINDEN2, R. KORPELA1, H. VAPAATALO1

BENEFICIAL ANTI-INFLAMMATORY EFFECT OF ANGIOTENSIN-CONVERTINGENZYME INHIBITOR AND ANGIOTENSIN RECEPTOR BLOCKER

IN THE TREATMENT OF DEXTRAN SULFATE SODIUM-INDUCED COLITIS IN MICE

1Faculty of Medicine, Pharmacology, University of Helsinki, Helsinki, Finland; 2Faculty of Veterinary Medicine, Department of Basic Veterinary Sciences, University of Helsinki, Helsinki, Finland

The renin-angiotensin system (RAS) in the intestine is involved in the regulation of inflammation, apoptosis and tissuefibrosis in experimental models of colitis; the inhibition of local RAS by pharmacologic interventions has been claimed toprevent and alleviate colitis. In this study, we compared the benefits of an angiotensin-converting enzyme (ACE) inhibitor,enalapril, an angiotensin receptor blocker, losartan and their combination in dextran sodium sulfate (DSS)-induced colitisin mice by assessing the histopathological and macroscopic changes in the colon, and by measuring the expression of thepro-inflammatory interleukin 1beta (IL-1b) and tumor necrosis factor alpha (Tnf-a) genes. We also examined theconsequences of these interventions on colonic angiotensin-converting enzyme protein and its ectodomain shedding as wellas gene expression of RAS components, Agt and Ace, and corticosterone synthesis and its components, Lrh-1 and Cyp11b1.Both enalapril and losartan alleviated colitis by reducing the inflammatory cell infiltrate in colon. In addition, enalaprildownregulated the pro-inflammatory IL-1b expression whereas losartan treatment resulted in lower macroscopic scores,but the effects of the medications were not synergistic when the drugs were combined. ACE-ectodomain shedding wasenhanced in the distal colon in DSS colitis. We found no evidence that ACE inhibition or angiotensin receptor blockadealtered intestinal RAS or corticosterone synthesis. We conclude that some of the benefits of ACE inhibition and angiotensinreceptor blockade might differ in the treatment of colitis, but their combination is unlikely to confer additional benefits.

K e y w o r d s : renin-angiotensin system, angiotensin-converting enzyme, angiotensin receptor, proinflammatory cytokines,angiotensin-converting enzyme ectodomain shedding, dextran sodium sulfate, colitis, enalapril, losartan

Despite the evidence emerging from preclinical studieshighlighting the benefits of ACE inhibitors and ARBs inexperimental colitis, the use of either drug as a treatment for IBDhas not been investigated to any significant extent. There are ratherfew studies describing human data and none of them arerandomized controlled trials. In one study, IBD patients on ARBtherapy exhibited a lower expression of several pro-inflammatorycytokines and chemokines in their intestinal mucosa compared tothose who did not receive ARBs (24). In contrast, in anotherretrospective study, presented as a conference abstract, Jacobs et al.(30) noted that in IBD patients, ACE inhibitor and ARB use wasassociated with higher all-cause hospitalization and highercorticosteroid use. In addition to inflammation, IBD can lead tosevere fibrosis, and therefore a reduction of angiotensin IIsignalling could offer additional benefits for IBD patients byreducing scarring and the formation of strictures (2, 31). Thisproposal is supported by preclinical studies in which both ACEinhibitors and ARBs were able to prevent colitis-inducedextracellular matrix accumulation and fibrosis in rats (16, 22). Inaddition, intestinal epithelial cell apoptosis has been indicated inIBD (32, 33), and ACE inhibitors and ARBs have been reported toreduce apoptosis rate in intestinal epithelium in preclinical colitisstudies (11, 12). Combination therapy of ACE inhibitors and ARBscan potentially have synergistic effects and strengthen the anti-inflammatory and tissue-protective properties of these medicationsin alleviation of intestinal inflammation. Pharmacological andbiological treatments for IBD have their own shortcomings, i.e.adverse effects and the development of drug resistance, leading toa declining effect and relapse (34, 35). Therefore, preclinicalstudies aimed at the modulation of intestinal RAS remain animportant study target while paving the way to clinical studies.

We have previously reported that the ACE ectodomain isshed from the intestine during DSS-induced colitis (36).Furthermore, the ACE inhibitor, captopril, reduced the geneexpression of angiotensinogen, Ace and Cyp11b1, the genecoding for the rate-limiting enzyme of corticosterone synthesis,in colon (37), and intestinal corticosterone synthesis wasincreased by Ang II (36). However, we could not detect anyamelioration of colitis after oral captopril treatment (37).Interestingly, we have found that mesenchymal stem cell therapyreduces protein levels of pro-inflammatory ACE in colon ofmice in the DSS model (Salmenkari et al., unpublished data).

Intestinal RAS clearly plays a part in human andexperimental colitis. In this preclinical study, our aim was tocompare the efficacy of orally given ACE inhibitor, enalapril, theARB, losartan, and their combination in alleviating DSS-induced colitis and local production of anti-inflammatorycorticosterone production in mice. We measured the geneexpression of several RAS components and ACE protein andcorticosterone synthesis in the colons of the mice.

MATERIALS AND METHODS

Animals

The study was approved by National Animal ExperimentationCommittee of Finland (ESAVI/114/04.10.07/2015) according toEC Directive 86/609/ECC and Finnish Experimental Animal Act62/2006.

A total of 40 Balb/C male mice were obtained from Envigo(Horst, Netherlands) at the age of 7 weeks. The mice were

2

Fig. 1. Schematic illustrating the experimental design. (A): Schematic presenting the dosing of medications in the intervention groupsduring the run-in and induction of colitis periods. The healthy controls received pure water throughout the experiment and the DSScontrols were provided with water during the run-in period and 3% DSS without medications during the induction of colitis. (B):Schematic of the sample collection and allocation from the excised colons.

housed in individually-ventilated cages in specific-pathogen freelaboratory conditions with 12-h light/dark cycle. The animalswere provided with standard rodent feed, 2018 Teklad Global18% Protein Rodent Diet (Harlan Laboratories, Indianapolis, IN,USA) and drinking water ad libitum.

Administration of medications and induction of colitis

The mice were assigned into five experimental groups (n = 8in each group) with similar starting weights (23.5 ± 0.1 g). At thebeginning of the experiment, the mice in the intervention groupswere given orally in drinking water either 11 mg/l enalapril(PHR1289, Sigma Aldrich, St. Louis, MO, USA), 37 mg/llosartan (PHR1602, Sigma Aldrich), corresponding to daily dosesof approximately 1.3 mg/kg/d of enalapril and 4.4 mg/kg/d oflosartan, respectively, or a combination of both drugs for 5 daysto acquaint the animals to the medications. Subsequently, 3%(w/vol) 40 kDa DSS (TdB Consultancy, Uppsala, Sweden) wasadded to the drinking water of the three intervention groups anda DSS-control group for 7 days, and the dosing of medications inthe intervention groups were concurrently increased to 25 mg/lenalapril and 83 mg/l losartan, corresponding to daily doses ofapproximately 3 mg/kg/d enalapril and/or 10 mg/kg/d losartan,respectively. These doses are based on effective doses reported inother publications (11, 23). The study design is presented in Fig.1A. The healthy controls received water throughout theexperiment. The animals were weighed daily throughout theexperiment. The daily fluid intake was measured, and the intakewas not affected by addition of either the drugs or DSS. At theend of the experiment, the animals were sacrificed byexsanguination under isoflurane anaesthesia.

Macroscopic evaluation of colitis

The macroscopic scoring criteria was adapted from Holma etal., and Melgar et al., (38, 39). The colons of mice were excised,and the lengths were measured. The colons were openedlongitudinally and the stool consistency was evaluated on a scalefrom 0 to 4 (0 = normal, 1 = moist, 2 = loose, 3 = liquid, 4 =none) and the presence of blood was scored on a scale from 0 to3 (0 = none, 1 = streaks, 2 = clearly visible, 3 = heavy bleeding).The colons were cleaned of intestinal contents and edema wasscored on a scale of 0 to 3 (0 = normal, 1 = mild, 2 = moderateand 3 = marked). These scores were combined as a totalmacroscopic score (scale 0 – 10). The sample collectionschematic is presented in Fig. 1B.

Histopathologic evaluation of colitis

Pieces of the most distal part of colon were fixed in 10%neutral-buffered formalin (Sigma Aldrich) for 24 hours,embedded longitudinally into paraffin blocks, sectioned at 4 µmthickness, and stained with hematoxylin and eosin (H&E).Histopathological changes were assessed in a partly blindedmanner; the pathologist (JL) was aware of the group of eachanimal, but unaware of the group identities and treatments.Briefly, surface epithelial injury, crypt epithelial injury, and cryptdilatation and distortion were graded on a scale of 0 – 3 each andcombined into an injury score (scale 0 – 9). The presence ofgranulocytes was graded on a scale of 0 – 3, the presence oflymphocytes and macrophages on a scale of 0 – 4, and thenumber and extent of granulomatous inflammatory foci(granulomas) on a scale of 0 – 2 and combined into aninflammation score (scale 0 – 9). Finally, the injury andinflammation scores were combined as the histopathologyscores (scale 0 – 18). The grading was simplified from a gradingsystem used in a study examining chemotherapy-induced

gastrointestinal toxicity in rats (40) and employed lesiondefinitions generally used in toxicological pathology (41).Criteria for grading are detailed in Table 1.

Ex-vivo incubation samples of colon

Ex-vivo incubation samples were prepared as previouslydescribed (37). Briefly, pieces of distal colon, adjacent to thehistological samples, were incubated in a pre-oxygenated Krebsbuffer (119 mmol/l NaCl, 25 mmol/l NaHCO3, 15 mmol/l KCl,11 mmol/l glucose, 1.6 mmol/l CaCl2, 1.2 mmol/l KH2PO4, 1.2mmol/l MgSO4) for 90 min at 37°C with gentle agitation. Thesamples were centrifuged at 13,300 g for 3 min. The supernatantwas used for ACE protein (#DY1513 R&D System, Minnesota,MO, USA) and corticosterone (#501320 Cayman Chemical,Michigan, MI, USA) assays, and the tissue pieces for proteinquantification (Pierce™ BCA Protein Assay Kit, ThermoScientific).

Angiotensin-converting enzyme protein quantification fromcolonic tissue

Tissue pieces of mid-colon were homogenized in 100 mMTris - 150 mM NaCl, pH 8.3 using a Precellys 24 cryobeadhomogenizer (Bertin Technologies, Montigny le Bretonneux,France). The samples were centrifuged at +4°C for 20 min andACE protein and total protein were quantified from thesupernatant (#DY1513 R&D System and Pierce™ BCA ProteinAssay Kit, Thermo Scientific).

Reverse-transcription quantitative PCR

RNA was extracted from mid-distal colon using NucleoSpinRNA kit (#740955, Macherey Nagel, Duren, Germany), andreverse transcribed to cDNA using iScript™ cDNA SynthesisKit (#1708891, BioRad, Hercules, CA, USA). RT-qPCR was runusing LightCycler® 480 SYBR Green I Master (#04707516001,Roche Diagnostics Corp., Indianapolis, IN, USA). The primerswere ordered from Sigma-Aldrich (St- Louis, MO, USA).

The primer sequences used for the study are:

Rplp0 F: 5’-TAACCCTGAAGTGCTCGACA-3’, R: 5’-GGTACCCGATCTGCAGACA-3’;

S18 F: 5’-AACGAACGAGACTCTGGCAT-3’, R: 5’-ACGCCACTTGTCCCTCTAAG-3’;

Eef2 F: 5’-TGTCAGTCATCGCCCATGTG-3’, R: 5’-CATCCTTGCGAGTGTCAGTGA-3’;

IL-1b F: 5’-CTCCAGCCAAGCTTCCTTGT-3’,R: 5’-TCATCACTGTCAAAAGGTGGCA-3’; Tnf-a F: 5’-GCCTCTTCTCATTCCTGCTTG-3’, R: 5’- CTGATGAGAGGGAGGCCATT-3’ (42);

Ace F: 5’-GCTGGAGGGTCTTTGATGGA-3’, R: 5’-AGTCACCTTGGGATCTTGGC-3’;

Agt F: 5’-CTTCCAAGGAACGATGAGAGGTT-3’, R: 5’-ACAGACACCGAGATGCTGTT-3’;

Cyp11b1 F: 5’-GGAACCCACCATCAGTAAGGA-3’, R: 5’-TCTTCCCTCACGCATGACAA-3’;

Lrh-1 F: 5’-GCATGGGAAGGAAGGGACAA-3’, R: 5’-CGCTGATCGAACTGAAGGGA-3’.

Target gene expression was normalized to the geometricmean of Rplp0, S18 and Eef2 expression.

Statistical analysis

RT-qPCR data is presented in graphs as individual values andgeometric mean, and in the text, as percent increase or decrease of

3

the geometric means. All other parametric data is presented asmean ± SD in the text and individual values and mean in thegraphs. Non-parametric data is presented as median andinterquartile range (IQR) in the tables. Statistical analyses wereconducted with SPSS versions 22 or 23. The differences betweengroups were analyzed using either one-way ANOVA with Tukey’stest or Kruskal-Wallis and Mann-Whitney U test. P-values lowerthan 0.05 were considered statistically significant.

RESULTS

Macroscopic disease activity is reduced by losartan

DSS induced a relatively mild colitis during the 7-day DSSadministration. During that time, the body weight of the miceremained close to the initial weight in all groups and by the endof the trial, the weight had declined only slightly in the colitisgroups (Fig. 2A). DSS induced diarrhea, swelling of the colonand colonic bleeding in all colitis groups, which were measuredas macroscopic scores (Table 2). All colitis groups had elevatedmacroscopic scores (P < 0.01 for all groups) when comparedwith the healthy mice. The macroscopic scores were lower in thelosartan group compared with the DSS-control group (P < 0.05)but otherwise the colitis groups did not statistically significantlydiffer from each other (Table 2).

Alleviation of histopathologic colitis by enalapril and losartan

The histopathological changes induced by DSS wererelatively mild (Fig. 2B-2F, Table 2). Injury, inflammation and

histopathology scores were higher in all colitis groups (P < 0.001for all groups) when compared with healthy controls (Table 2).Both enalapril (P = 0.007) and losartan (P = 0.038) alone, but nottheir combination, reduced the inflammatory infiltrate i.e. theinflammation score in colon when compared with the DSScontrol group (Table 2). The lower inflammation scores weremainly due to reduced number of neutrophils both in enalapril-and losartan-treated mice, and by a lower number ofgranulomatous foci in enalapril-treated mice. The treatments hadno statistically significant impact on the injury scores. Overall,the reduction in histopathology score remained significant in theenalapril group when compared with DSS controls (P = 0.021)but the reduction attributable to either losartan or thecombination of enalapril and losartan was not statisticallysignificant. Furthermore, the differences between enalapril andlosartan or the combination group were not statisticallysignificant.

Enalapril reduces colonic IL-1b following dextran sodiumsulfate colitis

In the next part of the experiment, we evaluated whether themedications influenced any of the molecular markers ofinflammation by measuring the levels of Il-1b and Tnf-a mRNAin the colons of the mice (Fig. 3A). DSS increased the expressionof IL-1b mRNA by 173% compared with healthy controls(almost statistically significant, P = 0.060). The expression wasalso increased by 149% in the losartan and by 254% in thecombination group (P = 0.040 and 0.002, respectively)compared with healthy controls. In comparison with the DSScontrols, enalapril, but not losartan or the combination treatment,

4

Mucosal injury grading Grade Surface epithelial injury Crypt epithelial injury Crypt distortion

0 Absent

Absent Straight crypts with inconspicuous lumen

1 Mild and focal to multifocal degeneration

Mild epithelial degeneration; single swollen goblet cells

Minimal to mild and focal crypt tortuousness and distension

2 Moderate and multifocal degeneration

Moderate degeneration; swollen goblet cells

Moderate and multifocal distortion and distension

3 Moderate to marked and widespread to diffuse degeneration; desquamation

Marked and diffuse degeneration; ruptured goblet cells; crypt destruction

Widespread to diffuse and marked distortion and distension

Inflammation grading Grade Granulocytes a Mononuclear inflammatory

cells b Granulomatous foci c

0 Neutrophils absent; few eosinophils

Few macrophages; some lymphocytes

Absent

1 Mild and focal or multifocal

Mild and focal or multifocal in basal lamina propria

One granuloma in the mucosa

2 Moderate and multifocal Moderate and multifocal to widespread in basal lamina propria

Two or more foci or at least one focus extends into the tela submucosa

3 Marked and diffuse

Marked and widespread, extends to superficial lamina propria

4

Severe and widespread throughout lamina propria

a Neutrophils and eosinophils were graded together as granulocytes. Eosinophil numbers were unaffected by the DSS treatment.b Macrophages, lymphocytes and plasma cells (very few) were graded together as mononuclear inflammatory cells.c Granulomatous foci (granulomas) consisted of a dense aggregation of macrophages encircled basally and laterally by lymphocytes.The macrophages were superficially admixed with variable number of neutrophils infiltrating into the surface epithelium. Some fociwere ulcerated.

Table 1. Criteria for grading mucosal injury and inflammation.

reduced IL-1b expression by 72% (P = 0.020). The level of IL-1b expression was similar between DSS controls, losartan andthe combination groups, but the expression was lower inenalapril group than in the losartan (P = 0.012) or thecombination groups (P = 0.001). The expression pattern of Tnf-a was similar to IL-1b, but there were no statistically significantdifferences between the groups (Fig. 3B). The geometric meanwas 376% higher in the DSS group than in the healthy group and37% and 40% lower in the enalapril and the combination groups

when compared with the DSS-control group, respectively. Theexpression was 37% higher in the losartan group than the DSScontrols.

Anti-inflammatory corticosterone production was not affectedby colitis or interventions in distal colon

In order to investigate whether ACE inhibition or AT1Rblockade affects colonic corticosterone synthesis pathway, we

5

Fig. 2. Enalapril suppresses DSS-induced colonic damage. (A): Body weight change from the beginning of DSS administration. (B-F): Hematoxylin and eosin-stained micrographs of colon after 7 days of DSS administration in healthy, DSS, enalapril, losartan andcombination group colons. Abbreviations: DSS, dextran sodium sulfate, E + L, enalapril + losartan combination group.

measured corticosterone production and the gene expression ofthe transcription factor liver receptor homolog-1 (Lrh-1), whichdrives intestinal glucocorticoid synthesis, and the rate-limitingenzyme of corticosterone synthesis Cyp11b1 in the colons of themice (Fig. 3C-3E). Corticosterone production in distal colonwas measured from intestinal incubation supernatants and theexpression levels of Lrh-1 and Cyp11b1 were measured from theadjacent mid-distal colon segment (Fig. 1B). Neither DSS colitisnor the medications exerted any influence on corticosteronesecretion (Fig. 3C) or Lrh-1 (Fig. 3D) or Cyp11b1 (Fig. 3E) geneexpression.

Dextran sodium sulfate induces angiotensin-converting enzymeshedding in distal colon

ACE protein was quantified from incubation supernatants ofdistal colon (Figs. 1B and 4A) and tissue segments of mid-colon(Figs. 1B and 4B) in order to examine whether the medicationshad influenced the levels of ACE in colon while Ace geneexpression was quantified in mid-distal colon (Figs. 1B and 4C).The ACE-ectodomain shedding was increased in all colitisgroups (9.8 ± 4.7 ng/mg protein, P = 0.032 in DSS-control, 13.1± 4.6 ng/mg, P < 0.001 in enalapril, 12.1 ± 3.0 ng/mg, P = 0.001in losartan and 11.3 ± 2.5 ng/mg, P = 0.005 in the combinationgroup samples) compared with healthy controls (3.8 ± 1.5ng/mg), and it was not altered by any of the interventions (Fig.4A). In contrast, Ace expression and tissue-bound ACE levelswere not modulated by DSS or by any of the treatments (Fig. 4B,4C). Furthermore, the gene expression of angiotensinogen didnot differ between the groups (Fig. 4D). Additionally, wemeasured ACE shedding in distal ileum, but this was not affectedby DSS or the medications (data not shown).

DISCUSSION

Modulation of RAS activity has been beneficial in thetreatment of experimental colitis (11-23, 25, 26). Therefore, wecompared two common drugs, the ACE inhibitor, enalapril, andthe ARB, losartan, and their combination in alleviation of DSS-induced colitis in mice. Several ACE inhibitors and ARBs anddifferent administration regimens have been evaluated in earlierstudies, but as far as we are aware, previously they have not beenmutually compared or examined in combination in the treatmentof colitis. We observed an amelioration of colitis by both ACEinhibition and ARB alone but not by their combination. Ourfindings support the beneficial role of ACE inhibitors and ARBsin murine intestinal inflammation and reveal some potential

differences in their disease-alleviating outcomes but indicatethere are no added benefits in combining these drugs, in fact,there may be disadvantages.

The histopathological changes in DSS colitis in the acutephase are described to encompass goblet cell depletion, erosionand ulceration as well as infiltration of granulocytes into thelamina propria and submucosa (10, 43, 44). Our resultsdemonstrate that both pharmacological ACE inhibition as wellas angiotensin receptor blockade can reduce the severity ofcolitis by reducing colonic inflammatory cell infiltration. Morespecifically, both drug classes reduced the number of infiltratingneutrophils and enalapril reduced the number of granulomas.One possible explanation is that there is a reduction in leukocyterecruitment. Ang II promotes leukocyte homing into manyinflamed tissues including the intestine, which occurs via AT1Rsignalling followed by a subsequent increase in cell adhesionmolecules (19, 45). Protein expression of MAdCAM-1, animportant adhesion molecule in the pathogenesis of human IBD,was shown to be inhibited in AT1R knockout mice during DSS-induced colitis (19).

In acute disease model studies, drugs are commonlyadministered by injections or oral gavage because of thepossibility of exact dosing of compounds to each animal and toavoid any possible risk of interaction with for example DSS.Multiple oral gavage treatments, however, can cause weight loss,esophageal tearing, or lowering of white blood cell counts of theanimals (46), which would be problematic and confounding incolitis studies. Administration in the drinking fluid has thebenefit of stable intake in case of drugs with short half-life likelosartan. Therefore, we chose to investigate the effect of thedrugs when given orally in drinking water. Nevertheless, thestability of the drugs in fluid and interactions with DSS is ofconcern and must be tested with every individual compound.Losartan has been previously administered in drinking water(23, 24) in TNBS model. Valsartan, another angiotensin receptorblocker with a similar chemical formula as losartan, has beenemployed together with DSS in drinking water by Santiago et al.(18). Enalapril was used successfully in drinking water in adiabetic kidney model by Sasaki et al. (47). In this study, we canconclude that both drugs had an anti-inflammatory effect whenadministered together with DSS.

The comparison of ACE inhibition and AT1R blockade madeit possible to reveal the differences between the drugs in howthey alleviate the histopathological, macroscopic andbiochemical symptoms of colitis, as well as the apparent self-regulatory role of intestinal RAS observed in our earlier studies(37). On the other hand, we chose an experimental design withonly a single dose, which does not allow a full pharmacological

6

Group

Colon length (cm)

Mean ± SD

Macroscopic score (0-10)

Median (IQR)

Injury score (0-9)

Median (IQR)

Inflammation score (0-9)

Median (IQR)

Histopathology score (0-18)

Median (IQR)

Healthy 7.7 ± 0.9 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) DSS 6.7 ± 0.6 5.0 (4.8) # 5.5 (6.0) † 5.0 (1.5) † 10.5 (6.5) † Enalapril 7.2 ± 0.7 4.5 (5.5) # 4.0 (3.5) † 2.0 (2.8) **† 6.0 (5.8) *† Losartan 7.0 ± 1.1 2.5 (4.3) *# 4.0 (5.3) † 2.8 (2.8) *† 6.5 (8.1) † E + L 7.2 ± 0.5 5.0 (5.3) # 4.0 (3.5) † 3.0 (4.8) † 6.5 (8.3) †

Table 2. Inflammation is reduced by RAS inhibition. Macroscopic and histopathology scores (n = 8 in each group). * denotes P < 0.05compared to DSS control group, ** denotes P < 0.01 compared to DSS control group, # denotes P < 0.01 compared to healthy controlgroup, † denotes P < 0.001 compared to healthy control group. Abbreviations: DSS, dextran sodium sulfate, E + L, enalapril + losartancombination group.

7

Fig. 3. Inflammatory markers and corticosterone synthesis. (A): IL-1b gene expression (n = 4 – 7 in each group) was normalized byenalapril. (B): Tnf-a gene expression was not changed by the treatments. (C): Corticosterone secretion (n = 7 – 8 in each group)expressed as pg/mg protein. (D): Lrh-1 (n = 4 – 7 in each group) and (E): Cyp11b1 gene expression (n = 4 – 7 in each group) were notaffected by DSS or modulation or RAS in distal colon. * denotes P < 0.05, ** denotes P < 0.01, Abbreviations: DSS, dextran sodiumsulfate, E + L, enalapril + losartan combination group.

comparison of the therapeutic effect between ACE inhibition andAT1R blockade. Other groups have experimented on the dosagesof both enalapril and losartan. Oral doses of 1 and 5 mg/kg/d ofenalapril were tested by Lee et al., (11), and 5 mg/kg dose wasmore effective in the treatment of colitis. Oral doses of 7 (22), 10(23) and 15 mg/kg/d (24) losartan were all effective againstTNBS colitis. In our study, 3 mg/kg/d enalapril and 10 mg/kg oflosartan represent effective but submaximal doses for mice,which could have therefore had synergistic effects in treatmentof colitis. Nevertheless, enalapril treatment resulted in alowering of the overall histopathologic scores and IL-1bexpression, whereas administration of losartan reduced themacroscopic and histopathologic inflammation scores. This datademonstrates that both enalapril and losartan can alleviate colitisbut suggest that the outcomes and obviously the are notidentical. ACE inhibition exerted a stronger anti-inflammatoryeffect, but only ARB reduced the macroscopic manifestations(stool consistency, colonic edema and bleeding) of DSS colitis.Neither of the drug was able to prevent mucosal injury. A recent

study found that losartan reduces the angiotensin II-dependentcontractibility of both small and large intestine ex vivo (48). Thiscould in part explain the beneficial effect of losartan on themacroscopic scores (especially stool consistency) in our study. Itis possible that the dose of enalapril was more potent inalleviating DSS colitis than that of losartan, but losartan mighthave the additional benefit of reducing smooth musclecontractibility.

Both drug classes inhibit the classical proinflammatory RASpathway, but due to the complexity of the RAS network thedrugs might cause a shift in production of other angiotensinmetabolites than Ang II, such as Ang 1-7 (3). ACE inhibitionreduces Ang II formation, and to some degree, the breakdown ofthe anti-inflammatory Ang 1-7 (49). Ang II mainly acts viaAT1R, although AT2R is constitutively expressed in theintestine, albeit to a lesser degree (50). AT1R signalling is amongother things pro-inflammatory, and pro-apoptotic, as well asinduces contractions of the smooth muscle cells both in theintestinal and vascular wall (1, 51). AT2R and the Mas receptor,

8

Fig. 4. DSS-induced colitis triggered ACE shedding in distal colon. (A): ACE shedding measured in ex vivo-incubation supernatantsof colon (n = 7 – 8 in each group), (B): Tissue-bound ACE in mid colon (n = 6 in each group), (C): Ace (n = 4 – 7 in each group) and(D): Agt gene expression (n = 4 – 7 in each group). * denotes P < 0.05, ** denotes P < 0.01, *** denotes P < 0.001.Abbreviations: DSS, dextran sodium sulfate, E + L, enalapril + losartan combination group.

the receptor for Ang 1-7, act in opposite or counteract the effectsof AT1R signalling (1). Small angiotensinogen-derived peptides(e.g. from Ang 1-5 to Ang 1-2) have recently been found toactivate Mas receptor (52). The complex network of peptidesand receptors might be the underlying cause of the smalldifferences between ACE inhibitors and ARBs.

Angiotensin receptor blockers have been reported to lowerthe colonic Il-1b and Tnf-a expression in several TNBS-colitisstudies (21, 23, 25). In the DSS model, 10-fold higher doses thanin our study, administered per rectum reduced proinflammatorycytokine expression (20), but we found no other mention oflosartan’s or other ARB’s effects on colonic Il-1b in the DSSmodel. Some studies report ARBs decrease TNF-a levels in DSSmodel (18, 25). The relatively low dose of losartan might explainthe lack of Il-1b reduction in our study.

We believed that simultaneous inhibition of ACE and AT1Rblockade would lead to a more potent inhibition of AT1Rsignalling, and thus more efficient alleviation of colitis.However, the benefits of the drugs were lost when they werecombined, suggesting that either the combination introducesadditional adverse effects due to excessive RAS inhibition or,more unlikely, the drugs have some unknown interaction whichcancels out the beneficial effects. Although we chose doses thatwere lower than the biggest effective doses for colitis found inliterature (11, 24), reducing doses in combination therapymight have yielded better results similar to single medications.ACE inhibitors and ARBs are important antihypertensivedrugs. We did not measure the blood pressure of the mice, butaccording to literature the chosen doses of enalapril andlosartan are unlikely to cause hypotension for normotensivemice (53, 54). The combination of an ACE inhibitor and anARB has not been studied in the context of experimental colitisor human IBD, but several clinical trials have been conductedregarding post-myocardial infarction therapy and diabeticnephropathy (55-58). Several studies reported that thecombination therapy increased the occurrence of potentiallyserious adverse effects (56-58), including hyperkalemia andkidney injury, and only one reported that combinationtreatment reduced the risk of cardiovascular events (55). Theseresults advocate caution for use in other disease areas likegastrointestinal medicine. However, several factors mightdistinguish the actions of either enalapril or losartan indifferent tissues, and therefore the results of cardiovascularstudies might not directly apply in gastrointestinal tract. Thedegree and duration of ACE inhibition can differ betweentissues, as described using lisinopril (59). Activation of AT1Rmight have different outcomes in the intestine than in thevessels due to selective G-protein coupling (60). In addition,some ACE inhibitors also inhibit other peptidases of the gut tosome extent (61, 62), and together this could lead to reducedeffect during combination treatment. However, our resultsshow that combination therapy is unlikely to confer additionalbenefits in the treatment of colitis.

We have previously reported that the ACE inhibitor captoprillowered the gene expression of RAS components, Ace andangiotensinogen, and the rate-limiting enzyme in corticosteronesynthesis, Cyp11b1 (37). In this study, we wanted to investigatewhether a similar effect could be obtained with another ACEinhibitor, enalapril, or with an ARB. However, neither enalaprilnor losartan exerted a similar effect on RAS modulation ascaptopril in acute/subacute DSS-induced colitis and thetreatments did not affect the expression levels of these genes.There could be several reasons for these discrepancies;differences in the severity of colitis, differences in the effectivedose of ACE inhibitors, or structural differences between theACE inhibitors, e.g. captopril’s sulfhydryl-moiety. There was arelatively mild colitis applied in this study, i.e. the animals

experienced minimal weight loss, moderate macroscopic andhistopathology scores and there was a lack of induction ofcorticosterone synthesis. Corticosterone is an anti-inflammatoryhormone which is produced primarily in the adrenal gland butalso locally throughout the intestine (63). In the intestine,glucocorticoid synthesis is regulated by the transcription factor,LRH-1 (64). During acute intestinal inflammation, TNF-aactivates LRH-1 to promote the transcription of the genes ofglucocorticoid synthesis e.g. the rate-limiting enzyme Cyp11b1but the signal becomes suppressed during chronic inflammation(65-67). Here, despite the evident increase in the levels of pro-inflammatory IL-1b and the disease scores provoked by DSS,the colitis might have been too mild to induce corticosteronesynthesis. Similarly, the gene expression levels of Lrh-1 andCyp11b1 were not elevated by the DSS treatment.

ACE protein is shed from vascular endothelium, andprobably also the intestinal epithelium, by ADAM9 protease(68). In addition to the vascular endothelium, ACE isprominently expressed in the intestine and in macrophages. Wehave determined that DSS induces ACE shedding in jejunum andto some extent in distal colon, but the shedding is not stronglyinduced in the proximal parts of colon (36, 37). In line with theseresults, ACE shedding was prominently increased by DSS indistal parts of colon, where the DSS-induced damage is mostdistinct. In distal ileum, the basal ACE shedding was not affectedby DSS. Therefore, ACE shedding seems to take place in certainparts of the intestine in the presence of inflammatory stimuli. Wehypothesized that the role of ACE shedding could be to regulatethe local Ang II concentration either by downregulating cellularACE or by directing soluble ACE to sites of inflammation. Itwould be interesting to determine whether the sheddingincreases in other colitis models as well or is it rather anindicator of the toxic damage evoked by DSS.

In conclusion, this study strengthens the evidence for thebenefits of using ACE inhibitors or ARBs in the treatment ofexperimental colitis, but based on our results, combinationtherapy of ACE inhibitors and ARBs does not seem advisable.Although both ACE inhibition and angiotensin receptorblockade alleviate experimental colitis, their benefits seem to bepartially dissimilar. Mechanistic studies of the anti-colitogeniceffects of these two drug classes, and the possible involvementof the extended RAS, should be conducted to determine themost suitable treatment approach for gastrointestinal diseases.Clinical, and to some degree epidemiological, studies of ACEinhibitor and ARB use in IBD patients will be needed totranslate the findings from experimental models into humantherapeutics. In addition, it would be interesting to investigatethe potential of ACE inhibitors and ARBs in the alleviation ofestablished colitis.

Abbreviations: ACE, angiotensin-converting enzyme;ADAM9, A disintegrin and metalloproteinase domain-containing protein 9; Ang II, angiotensin II; Ang 1-7, =angiotensin 1-7; AGT, angiotensinogen; ARBs, angiotensinreceptor blockers; AT1R, angiotensin receptor subtype 1; AT2R,angiotensin receptor subtype 2; DSS, dextran sodium sulfate;IBD, inflammatory bowel disease; IL-1b, interleukin 1beta;LRH-1, liver receptor homolog-1; RAS, renin-angiotensinsystem; TGF-b1, transforming growth factor beta1; TNBS,trinitrobenzene sulfonic acid; TNF-a, tumor necrosis factoralpha; UC, ulcerative colitis.

Author contributions: H. Salmenkari, H. Vapaatalo andR. Korpela designed the study plan; H. Salmenkari, J. Lindenand L. Pasanen performed the experiments; H. Salmenkarianalyzed the data; H. Salmenkari, J. Linden, H. Vapaatalo andR. Korpela wrote the paper.

9

Acknowledgements: We thank Helena Taskinen at TissuePreparation and Histochemistry Unit, Medicum, Department ofAnatomy, University of Helsinki for preparation of the H&E-stained tissue slides. We are grateful to Hanna Launonen atMedicum, Department of Pharmacology, University of Helsinkifor her skilful technical assistance during this study. Wesincerely thank Dr. Ewen MacDonald for the grammar and stylerevision. The study was funded by Juhani Aho Medical ResearchFoundation, Finland, and Finska Lakaresallskapet, Einar ochKarin Stroems Stiftelse, Finland.

Conflict of interests: None declared.

REFERENCES

1. Garg M, Angus PW, Burrell LM, Herath C, Gibson PR,Lubel JS. Review article: the pathophysiological roles of therenin-angiotensin system in the gastrointestinal tract.Aliment Pharmacol Ther 2012; 35: 414-428.

2. Szabo H, Fiorino G, Spinelli A, et al. Review article: anti-fibrotic agents for the treatment of Crohn’s disease - lessonslearnt from other diseases. Aliment Pharmacol Ther 2010;31: 189-201.

3. Brzozowski T. Role of renin-angiotensin system andmetabolites of angiotensin in the mechanism of gastricmucosal protection. Curr Opin Pharmacol 2014; 19: 90-98.

4. Jaszewski R, Tolia V, Ehrinpreis MN, et al. Increasedcolonic mucosal angiotensin I and II concentrations inCrohn’s colitis. Gastroenterology 1990; 98: 1543-1548.

5. Hume GE, Fowler EV, Lincoln D, et al. Angiotensinogenand transforming growth factor beta1: novel genes in thepathogenesis of Crohn’s disease. J Med Genet 2006; 43: e51.Doi: 10.1136/jmg.2005.040477

6. Hume GE, Doecke JD, Huang N, et al. Altered expressionof angiotensinogen and mediators of angiogenesis in ilealCrohn’s disease. J Gastrointestin Liver Dis 2016; 25: 39-48.

7. Hume GE, Radford-Smith GL. ACE inhibitors and angiotensinII receptor antagonists in Crohn’s disease management. ExpertRev Gastroenterol Hepatol 2008; 2: 645-651.

8. Saibeni S, Spina L, Virgilio T, et al. Angiotensin-convertingenzyme insertion/deletion gene polymorphism ininflammatory bowel diseases. Eur J Gastroenterol Hepatol2007; 19: 976-981.

9. Garg M, Burrell LM, Velkoska E, et al. Upregulation ofcirculating components of the alternative renin-angiotensinsystem in inflammatory bowel disease: a pilot study. J ReninAngiotensin Aldosterone Syst 2015; 16: 559-569.

10. Eichele DD, Kharbanda KK. Dextran sodium sulfate colitismurine model: an indispensable tool for advancing ourunderstanding of inflammatory bowel diseases pathogenesis.World J Gastroenterol 2017; 23: 6016-6029.

11. Lee C, Chun J, Hwang SW, Kang SJ, Im JP, Kim JS.Enalapril inhibits nuclear factor-kappaB signaling inintestinal epithelial cells and peritoneal macrophages andattenuates experimental colitis in mice. Life Sci 2014; 95:29-39.

12. Spencer AU, Yang H, Haxhija EQ, Wildhaber BE, GreensonJK, Teitelbaum DH. Reduced severity of a mouse colitismodel with angiotensin converting enzyme inhibition. DigDis Sci 2007; 52: 1060-1070.

13. Koga H, Yang H, Adler J, Zimmermann EM, TeitelbaumDH. Transanal delivery of angiotensin converting enzymeinhibitor prevents colonic fibrosis in a mouse colitis model:development of a unique mode of treatment. Surgery 2008;144: 259-268.

14. Sueyoshi R, Ignatoski KM, Daignault S, Okawada M,Teitelbaum DH. Angiotensin converting enzyme-inhibitorreduces colitis severity in an IL-10 knockout model. Dig DisSci 2013; 58: 3165-3177.

15. Okawada M, Wilson MW, Larsen SD, Lipka E, Hillfinger J,Teitelbaum DH. Blockade of the renin-angiotensin systemprevents acute and immunologically relevant colitis inmurine models. Pediatr Surg Int 2016; 32: 1103-1114.

16. Wengrower D, Zanninelli G, Pappo O, et al. Prevention offibrosis in experimental colitis by captopril: the role of tgf-beta1. Inflamm Bowel Dis 2004; 10: 536-545.

17. Jahovic N, Ercan F, Gedik N, Yuksel M, Sener G, Alican I.The effect of angiotensin-converting enzyme inhibitors onexperimental colitis in rats. Regul Pept 2005; 130: 67-74.

18. Santiago OI, Rivera E, Ferder L, Appleyard CB. Anangiotensin II receptor antagonist reduces inflammatoryparameters in two models of colitis. Regul Pept 2008; 146:250-259.

19. Mizushima T, Sasaki M, Ando T, et al. Blockage ofangiotensin II type 1 receptor regulates TNF-alpha-inducedMAdCAM-1 expression via inhibition of NF-kappaBtranslocation to the nucleus and ameliorates colitis. Am JPhysiol Gastrointest Liver Physiol 2010; 298: G255-G266.

20. Okawada M, Koga H, Larsen SD, et al. Use of enterallydelivered angiotensin II type Ia receptor antagonists toreduce the severity of colitis. Dig Dis Sci 2011; 56: 2553-2565.

21. Inokuchi Y, Morohashi T, Kawana I, Nagashima Y, KiharaM, Umemura S. Amelioration of 2,4,6-trinitrobenzenesulphonic acid induced colitis in angiotensinogen geneknockout mice. Gut 2005; 54: 349-356.

22. Wengrower D, Zanninelli G, Latella G, et al. Losartanreduces trinitrobenzene sulphonic acid-induced colorectalfibrosis in rats. Can J Gastroenterol 2012; 26: 33-39.

23. Liu TJ, Shi YY, Wang EB, Zhu T, Zhao Q. AT1R blockerlosartan attenuates intestinal epithelial cell apoptosis in amouse model of Crohn’s disease. Mol Med Rep 2016; 13:1156-1162.

24. Shi Y, Liu T, He L, et al. Activation of the renin-angiotensinsystem promotes colitis development. Sci Rep 2016; 6:27552. doi: 10.1038/srep27552

25. Nagib MM, Tadros MG, ElSayed MI, Khalifa AE. Anti-inflammatory and anti-oxidant activities of olmesartanmedoxomil ameliorate experimental colitis in rats. ToxicolAppl Pharmacol 2013; 271: 106-113.

26. Arab HH, Al-Shorbagy MY, Abdallah DM, Nassar NN.Telmisartan attenuates colon inflammation, oxidativeperturbations and apoptosis in a rat model of experimentalinflammatory bowel disease. PLoS One 2014; 9: e97193.doi: 10.1371/journal.pone.0097193

27. Katada K, Yoshida N, Suzuki T, et al. Dextran sulfate sodium-induced acute colonic inflammation in angiotensin II type 1areceptor deficient mice. Inflamm Res 2008; 57: 84-91.

28. Khajah MA, Fateel MM, Ananthalakshmi KV, Luqmani YA.Anti-inflammatory action of angiotensin 1-7 in experimentalcolitis. PLoS One 2016; 11: e0150861. doi: 10.1371/journal.pone.0150861

29. Pawlik MW, Kwiecien S, Ptak-Belowska A, et al. The renin-angiotensin system and its vasoactive metaboliteangiotensin-(1-7) in the mechanism of the healing ofpreexisting gastric ulcers. The involvement of Masreceptors, nitric oxide, prostaglandins and proinflammatorycytokines. J Physiol Pharmacol 2016; 67: 75-91.

30. Jacobs J, Gulotta G, Liao C, Li YC, Bossonnette M, PekowJ. Angiotensin II blockade is associated with worse clinicaloutcomes in IBD patients. Am J Gastroenterol 2015; 110(Suppl. 1): S766-S7847.

10

31. Latella G, Sferra R, Speca S, Vetuschi A, Gaudio E. Can weprevent, reduce or reverse intestinal fibrosis in IBD? Eur RevMed Pharmacol Sci 2013; 17: 1283-1304.

32. Zeissig S, Bojarski C, Buergel N, et al. Downregulation ofepithelial apoptosis and barrier repair in active Crohn’sdisease by tumour necrosis factor alpha antibody treatment.Gut 2004; 53: 1295-1302.

33. Blander JM. Death in the intestinal epithelium-basic biologyand implications for inflammatory bowel disease. FEBS J2016; 283: 2720-2730.

34. Quetglas EG, Mujagic Z, Wigge S, et al. Update onpathogenesis and predictors of response of therapeuticstrategies used in inflammatory bowel disease. World JGastroenterol 2015; 21: 12519-12543.

35. Sales-Campos H, Basso PJ, Alves VB, et al. Classical andrecent advances in the treatment of inflammatory boweldiseases. Braz J Med Biol Res 2015; 48: 96-107.

36. Salmenkari H, Issakainen T, Vapaatalo H, Korpela R. Localcorticosterone production and angiotensin-I convertingenzyme shedding in a mouse model of intestinalinflammation. World J Gastroenterol 2015; 21: 10072-10079.

37. Salmenkari H, Holappa M, Forsgard RA, Korpela R,Vapaatalo H. Orally administered angiotensin-convertingenzyme-inhibitors captopril and isoleucine-proline-prolinehave distinct effects on local renin-angiotensin system andcorticosterone synthesis in dextran sulfate sodium-inducedcolitis in mice. J Physiol Pharmacol 2017; 68: 355-362.

38. Holma R, Salmenpera P, Riutta A, Virtanen I, Korpela R,Vapaatalo H. Acute effects of the cys-leukotriene-1 receptorantagonist, montelukast, on experimental colitis in rats. EurJ Pharmacol 2001; 429: 309-318.

39. Melgar S, Karlsson A, Michaelsson E. Acute colitis inducedby dextran sulfate sodium progresses to chronicity inC57BL/6 but not in BALB/c mice: correlation betweensymptoms and inflammation. Am J Physiol GastrointestLiver Physiol 2005; 288: G1328-G1338.

40. Forsgard RA, Korpela R, Holma R, et al. Intestinalpermeability to iohexol as an in vivo marker of chemotherapy-induced gastrointestinal toxicity in Sprague-Dawley rats.Cancer Chemother Pharmacol 2016; 78: 863-874.

41. Nolte T, Brander-Weber P, Dangler C, et al. Nonproliferativeand proliferative lesions of the gastrointestinal tract,pancreas and salivary glands of the rat and mouse. J ToxicolPathol 2016; 29 (Suppl. 1):1S-125S.

42. Yamakawa I, Kojima H, Terashima T, et al. Inactivation ofTNF-alpha ameliorates diabetic neuropathy in mice. Am JPhysiol Endocrinol Metab 2011; 301: E844-E852.

43. Schepp-Berglind J, Atkinson C, Elvington M, Qiao F,Mannon P, Tomlinson S. Complement-dependent injury andprotection in a murine model of acute dextran sulfatesodium-induced colitis. J Immunol 2012; 188: 6309-6318.

44. Chassaing B, Aitken JD, Malleshappa M, Vijay-Kumar M.Dextran sulfate sodium (DSS)-induced colitis in mice. CurrProtoc Immunol 2014; 104: Unit 15.25. doi:10.1002/0471142735.im1525s104

45. Suzuki Y, Ruiz-Ortega M, Lorenzo O, Ruperez M, EstebanV, Egido J. Inflammation and angiotensin II. Int J BiochemCell Biol 2003; 35: 881-900.

46. Jones CP, Boyd KL, Wallace JM. Evaluation of miceundergoing serial oral gavage while awake or anesthetized.J Am Assoc Lab Anim Sci 2016; 55: 805-810.

47. Sasaki M, Uehara S, Ohta H, et al. Losartan amelioratesprogression of glomerular structural changes in diabeticKKAy mice. Life Sci 2004; 75: 869-880.

48. Patten GS, Abeywardena MY. Effects of antihypertensiveagents on intestinal contractility in the spontaneously

hypertensive rat: angiotensin receptor system downregulationby losartan. J Pharmacol Exp Ther 2017; 360: 260-266.

49. Chappell MC, Pirro NT, Sykes A, Ferrario CM. Metabolismof angiotensin-(1-7) by angiotensin-converting enzyme.Hypertension 1998; 31: 362-367.

50. Sechi LA, Valentin JP, Griffin CA, Schambelan M.Autoradiographic characterization of angiotensin II receptorsubtypes in rat intestine. Am J Physiol 1993; 265: G21-G27.

51. Fandriks L. The renin-angiotensin system and thegastrointestinal mucosa. Acta Physiol (Oxf) 2011; 201:157-167.

52. Moraes PL, Kangussu LM, da Silva LG, Castro CH, SantosRAS, Ferreira AJ. Cardiovascular effects of small peptidesof the renin angiotensin system. Physiol Rep 2017; 5:e13505. doi: 10.14814/phy2.13505

53. Harding P, Stonier C, Aber GM. Dose-dependent effects ofangiotensin converting enzyme (ACE) inhibitors onglomerular prostanoid production by normotensive rats. Br JPharmacol 1993; 108: 327-330.

54. Diop-Frimpong B, Chauhan VP, Krane S, Boucher Y, JainRK. Losartan inhibits collagen I synthesis and improves thedistribution and efficacy of nanotherapeutics in tumors. ProcNatl Acad Sci USA 2011; 108: 2909-2914.

55. McMurray JJ, Ostergren J, Swedberg K, et al. Effects ofcandesartan in patients with chronic heart failure andreduced left-ventricular systolic function taking angiotensin-converting-enzyme inhibitors: the CHARM-Added trial.Lancet 2003; 362: 767-771.

56. Pfeffer MA, McMurray JJ, Velazquez EJ, et al. Valsartan,captopril, or both in myocardial infarction complicated byheart failure, left ventricular dysfunction, or both. N Engl JMed 2003; 349: 1893-1906.

57. Phillips CO, Kashani A, Ko DK, Francis G, Krumholz HM.Adverse effects of combination angiotensin II receptorblockers plus angiotensin-converting enzyme inhibitors forleft ventricular dysfunction: a quantitative review of datafrom randomized clinical trials. Arch Intern Med 2007; 167:1930-1936.

58. Fried LF, Emanuele N, Zhang JH, et al. Combinedangiotensin inhibition for the treatment of diabeticnephropathy. N Engl J Med 2013; 369: 1892-1903.

59. Sakaguchi K, Chai SY, Jackson B, Johnston CI, MendelsohnFA. Inhibition of tissue angiotensin converting enzyme.Quantitation by autoradiography. Hypertension 1988; 11:230-238.

60. Hunyady L, Catt KJ. Pleiotropic AT1 receptor signalingpathways mediating physiological and pathogenic actions ofangiotensin II. Mol Endocrinol 2006; 20: 953-970.

61. Reinhardt D, Sigusch HH, Hensse J, Tyagi SC, Korfer R,Figulla HR. Cardiac remodelling in end stage heart failure:upregulation of matrix metalloproteinase (MMP)irrespective of the underlying disease, and evidence for adirect inhibitory effect of ACE inhibitors on MMP. Heart2002; 88: 525-530.

62. Efsen E, Saermark T, Hansen A, Bruun E, Brynskov J.Ramiprilate inhibits functional matrix metalloproteinaseactivity in Crohn’s disease fistulas. Basic Clin PharmacolToxicol 2011; 109: 208-216.

63. Cima I, Corazza N, Dick B, et al. Intestinal epithelial cellssynthesize glucocorticoids and regulate T cell activation. J ExpMed 2004; 200: 1635-1646.

64. Mueller M, Cima I, Noti M, et al. The nuclear receptor LRH-1 critically regulates extra-adrenal glucocorticoid synthesisin the intestine. J Exp Med 2006; 203: 2057-2062.

65. Noti M, Corazza N, Mueller C, Berger B, Brunner T. TNFsuppresses acute intestinal inflammation by inducing localglucocorticoid synthesis. J Exp Med 2010; 207: 1057-1066.

11

66. Noti M, Corazza N, Tuffin G, Schoonjans K, Brunner T.Lipopolysaccharide induces intestinal glucocorticoidsynthesis in a TNFalpha-dependent manner. FASEB J 2010;24: 1340-1346.

67. Huang SC, Lee CT, Chung BC. Tumor necrosis factorsuppresses NR5A2 activity and intestinal glucocorticoidsynthesis to sustain chronic colitis. Sci Signal 2014; 7: ra20.doi: 10.1126/scisignal.2004786

68. English WR, Corvol P, Murphy G. LPS activates ADAM9dependent shedding of ACE from endothelial cells. BiochemBiophys Res Commun 2012; 421: 70-75.

R e c e i v e d : April 18, 2018A c c e p t e d : August 30, 2018

Author’s address: Prof. Heikki Vapaatalo, Faculty ofMedicine, Department of Pharmacology, University of Helsinki,8 Haartmaninkatu, 00290 Helsinki, Finland.E-mail: heikki.vapaatalo@helsinki.fi

12