Colon diverticulosis is a frequent condition in adults in Western countries and several patients report clinical symptoms (abdominal pain, bloating, and altered bowel habits) even when diverticulosis is not complicated by perforation, fistula, obstruction, and/or bleeding: this condition being referred to as symptomatic uncomplicated diverticular disease (SUDD).1-3
Impaired intestinal motility seems to play an important role in this condition, although it is unclear if the altered colonic motor activity is the first event, which leads to the formation of diverticula, or if it is a consequence.2-5
Colonic transit studies reported intestinal higher pressures and raised electrical activity and an abnormal response to cholinergic stimulus in the left colon in most of SUDD patients.2,4-6 Therefore, data from in vitro studies mainly reported an increased excitatory response that might account for the spasticity observed in these patients.2
In SUDD patients, undergoing surgery with uncomplicated disease, it has been reported an incremented in vitro sensitivity of colonic circular smooth muscle to exogenous acetylcholine (ACh)5,7 and tachykinergic agonists6 in combination with a reduction of smooth muscle choline acetyltransferase activity and upregulation of postsynaptic muscarinic M3 receptors. These results raised the hypothesis of a type of hypersensitivity to cholinergic denervation in SUDD,7 which is a mechanism also recognized following a damage to skeletal muscle motor innervation. Nevertheless, Tomita et al8 did not show a significant increase in sensitivity to exogenous ACh of longitudinal muscle strips (MS), although observing an up-regulation of muscarinic M3 receptors in circular and longitudinal muscles. On the other hand, most of these in vitro studies were performed in patients undergoing surgery for diverticulitis,6,9 thus suggesting the doubt that the effect could be related to the inflammatory state. Furthermore, the current mechanisms underlying these alterations have not been fully clarified yet and data on smooth muscle cells (SMCs), which are the last effectors of enteric neuromuscular units, are lacking in diverticular disease (DD) patients undergoing surgery for diverticulitis.9 A recent paper explored the expression patterns of molecular factors involved in the contractile activity of SMCs of colonic samples from patients with complicated DD, focusing on the role of Ras homolog family member A (RhoA)/Rho-associated protein kinase (ROCK), and protein kinase C/C-kinase-potentiated protein phosphatase 1 inhibitor of 17 kDa that seems to be strategic in the control of the intestinal SMC contraction. The results of this study showed an altered phenotype compatible with an inhibition of the RhoA/ROCK pathway, such a condition has been shown to directly affect the relaxation of SMCs.9
Moreover, previous studies have demonstrated, in case of sepsis, the role of macrophages present in the intestinal muscularis externa that, after the identification of the bacterial endotoxin lipopolysaccharide (LPS) of Gram-negative bacteria, being due primarily to Toll-like receptor 4 (TLR4) activation, produce inflammatory cytokines responsible for the impairment of smooth muscle contractility10,11 and increase trans-membrane permeability. In a recent investigation, it has been demonstrated that SUDD causes significant modifications of TLR2 and TLR4 expression on various immune cell subpopulations obtained by both peripheral blood and affected mucosa.12
In SUDD, it cannot be excluded that neuromuscular changes could be associated to an impaired muscle contractility due to products released by the local colonic bacterial population. Therefore, in this study the contractility patterns of colonic SMCs and MS isolated from human circular colonic muscle layers of SUDD patients were explored and their mucosal permeability, trans-epithelial electrical resistance and tight junction (TJ) protein complex integrity were assessed, as a measurable feature of the intestinal barrier function, as well as the expression of TLRs.
Specimens of sigmoid colon were obtained from 9 patients with SUDD undergoing elective left hemi-colectomy for uncomplicated colonic cancer; all patients had previously reported symptoms as abdominal pain and/or alterations of bowel habits (SUDD group). Documented evidence of DD was detected with CT scan or during endoscopy. Specimens of sigmoid were taken also from 9 patients undergoing surgery for uncomplicated colonic cancer as controls (control group). In both groups, samples were obtained at least 10 cm away from any macroscopically evident alteration due to inflammation or cancer. The average age in the control group was 63 years (range 47-76 years). The average age for SUDD patients was 76 years (range 68-83 years,
Informed consent was obtained from all patients and the clinical protocol was approved by the Ethical Committee (Ref. No. 29/14 PAR). Exclusion criteria were: previous radiotherapy or chemotherapy; treatment with calcium channel blockers or opioids. Patients had not been treated with steroids or opioids before surgery. The pathologist reports were reviewed to be sure regarding the absence/presence of colonic diverticula and to exclude samples with other histological abnormalities or inflammation.
Fresh samples were transported to the laboratory in chilled oxygenated Krebs solution (consisting of 116.6 NaCl, 21.9 NaHCO3, 1.2 KH2PO4, 5.4 glucose, 1.2 MgCl2, 3.4 KCl, and 2.5 CaCl2 in mM).
Ex vivo experiments on human colonic muscle strip motility
Ex vivo experiments on human colonic smooth muscle cell motility
Subjects for Ussing Chambers experiments
Human biopsy samples were obtained from distal colon of 9 consecutive patients with SUDD, who were consecutively recruited among the outpatients and inpatients attending the Gastrointestinal Disease Unit of Campus Bio-Medico. Control group was composed of 9 healthy volunteers age- and gender-matched, undergoing colonoscopy for colorectal cancer screening or follow-up of polyposis. Exclusion criteria were: assumption of antibiotics or probiotics/prebiotics within the last 4 weeks, current infectious diarrhea, use of drugs altering gastrointestinal motility and secretions (anticholinergics and prokinetics agents) and recent NSAIDs use. After obtaining informed consent, colonic mucosa integrity and function was assessed by collecting 4 biopsies for each patient from the sigmoid colon. Abdominal pain, meteorism, comorbidity, and smoking habits of healthy subjects or patients with SUDD were recorded.
Ussing Chambers experiments for mucosal integrity
Mucosal integrity of colonic biopsy specimens was analyzed in Ussing Chambers (Mussler Scientific Instruments, Aachen, Germany) by detecting and comparing paracellular permeability to fluorescein isothiocyanate-dextran (FD-4) and transepithelial electrical resistance (TEER) between 9 SUDD patients and 9 healthy controls. Transmural alternating current impedance analysis was carried out as described earlier.17,18 Biopsies were positioned in modified 3 mL Ussing Chambers as previously described,19 exposing a surface of 0.017 cm2. Briefly, mucosal and serosal layers were filled with 3 mL of 10 mM mannitol and 10 mM glucose in Krebs–Ringer bicarbonate buffer at pH 7.4, respectively. Solutions were maintained at 37℃ and continuously carbogenated with O2/CO2 (95%/5%). After reaching a stable TEER baseline, the mucosal side of the biopsies was exposed to FD-4 (molecular mass = 4000 Da, 1 mg/mL; Sigma-Aldrich, St Louis, MO, USA) to measure paracellular permeability. Samples (300 µL/each) were collected from the serosal side during 2 hours at 30-minute intervals, the fluorescence level of which was detected using a fluorescence reader (Tecan Infinite M200-Pro multiplate reader; Tecan, Männedorf, Switzerland). Fluorescence values were then converted in concentrations of fluorescein (pmol) through a standard curve. Time points 0 minute and 30 minutes were excluded from the analysis, as the paracellular probe takes time to accumulate on the serosal side. The average of time points 60, 90, and 120 minutes of the biopsy specimens used in each experiment was calculated. TEER was measured from the voltage deflections induced by bipolar constant-current pulses of 16 mA every 60 seconds with duration of 200 milliseconds and was recorded every 30 minutes over 2 hours. The average of all time points (from 30 minutes to 120 minutes) of the biopsy samples used in each experiment was calculated and results were expressed in Ω × cm2. Each experiment was performed in quadruplicate.
Zonulin, claudin, and TLR4 expression were studied by immunohistochemistry performed on 3-µm thick consecutive sections obtained from formalin-fixed tissue embedded in paraffin. Sixteen samples from SUDD patients and 16 samples from control group patients were analyzed and quantified for TLR4 and TJs. After antigen retrieval with citrate buffer (pH 6.0), sections were incubated at room temperature with the following primary antibodies: mouse monoclonal antibody anti-claudin diluted 1:100 (sc81796; Santa Cruz, Dallas, TX, USA), rabbit polyclonal antibody anti-zonulin diluted 1:100 (21773-I-AP; Proteintech, Rosemont, IL, USA), and mouse monoclonal antibody anti-TLR4 (clone C-18) diluted 1:100 (sc8594; Santa Cruz). Negative controls were obtained by omitting primary antibodies. Immunohistochemical reactions were visualized by 3’3-diaminobezidine tetrahydrochloride as the chromogen from MACH 1 Universal HRP-Polymer Detection (Biocare Medical, Concord, MA, USA). Hematoxylin was used as counterstain in order to visualize nuclei.
Immunoreactivity in unaffected mucosa of SUDD patients and control group subjects was scored. Images were acquired with an automated digital scanning system (Hamamatsu NanoZoomer 2.0 RS, Hamamatsu, Hamamatsu City, Japan). The TLR4 and TJ immunostaining were analyzed in 3 fields for each sample (magnification ×100), and a specific Image J (National Institutes of Health, Bethesda, Maryland, USA) plugin (IHC Profiler as described in Varghese et al20) was used to quantify the immunoreactivity. For correlative purpose, an average score was derived for each sample.
All statistical analyses were performed by GraphPad Prism ver. 4.02 (GraphPad Software Inc, CA, USA). Variables are expressed as median and 25%-75% interquartile range. Differences were evaluated by the Kruskal–Wallis test followed by
Histopathological analysis, carried out by conventional hematoxylin and eosin staining, did not reveal significant changes in the healthy mucosa between patients with and without diverticula. TLR4 immunohistochemical expression was localized in epithelial cells and endocrine-like cells, present in the intestinal glands and in inflammatory cells present in lamina propria.
Immunohistochemical expression of TJ protein complex (claudin and zonulin) was lower in the colonic healthy mucosa of SUDD patients compared to control group subjects. Claudin immunohistochemical expression was 41.62 (21.02-89.67) in SUDD vs 72.24 (46.61-108.88) in the control group (
The average score of TLR4 immunohistochemical expression positivity did not show significant difference between SUDD patients, 133.82 (113.29-167.47) and control group, 143.66 (125.28-153.81) (
Following 60 minutes of stabilization, colonic MS of SUDD patients and controls developed basal tone and spontaneous activity, consisting in small amplitude rhythmic contractions of myogenic origin (Fig. 3). In SUDD group we found a significant reduced basal tone and ACh-elicited contraction over basal compared to the control group being 9.5 g (minimum 4.2-maximum 21 g), and 47.0% in the SUDD group, and 14.16 g (minimum 7.2-maximum 25.77 g), and 69.0% in the control group,
SMCs of control group presented a resting cell length of 178.7 ± 12.1 µm and a maximal contractile response to ACh of 16.49% ± 1.2 over basal. We found that in the SUDD group the resting cell length was higher, even if not statistically different from the controls of 184.8 ± 12.9. Similarly to MS, the ACh-elicited contraction of SMCs was significantly reduced when compared to controls, being 8.83% ± 1.8 (Fig. 5).
Compared to healthy controls, patients with SUDD showed lower TEER (median 15.0 Ω × cm2, minimum 11.0-maximum 18.7 in SUDD vs 20.3 Ω × cm2, minimum 13.9-maximum 39.3 in the control group,
In the present study, the contractile activity of SUDD colonic SM strips showed a significant alteration of both basal tone and ACh-elicited contraction, compared to controls. Previous studies support the evidence that, in patients with DD, the colon exhibits an altered motor pattern showing an increased tone and enhanced postprandial motility with the presence of spastic motor activity and excessive segmentary contractions.21,22 It seems that the predisposition to mucosal herniation at areas of weakness in the colonic wall could be determined by these consistent high-pressure contractions.
Differently from our investigation, which analyze the isolated colonic muscle contraction, most of previous studies have investigated the longitudinal muscle contraction or were mainly conducted by in vivo evaluation using manometric techniques, thus analyzing the whole colonic motility.21,23 These conditions could explain the apparent disagreement with the results of our study, which analyzed circular muscle specimens in an in vitro model. Our data are in accordance with an interesting investigation, conducted by Gallego et al24 in 2013, which demonstrated a significant altered in vitro motor pattern, in sigmoid and left colon strips from patients with SUDD, showing a decrease in the amplitude and area under the curve of rhythmic phasic contractions.24
Colonic motor alterations in DD have been suggested to be due to changes in enteric nerves, smooth muscle function and/or mucosal and myenteric inflammation.25 An interesting study, which demonstrated some structural and functional abnormalities of the enteric musculature in DD patients, consisted in a disturbed muscular architecture, connective tissue replacement and loss of specific myofilaments, suggests a primitive muscle disorder responsible for the colonic contractile alteration, which could be involved in the pathogenesis and development of DD.26
Moreover, an immunohistochemical analysis, performed in colonic samples from patients with DD who underwent elective abdominal surgery following the third or fourth episode of diverticulitis, revealed significant abnormalities in both the expression and distribution patterns of several proteins involved in the functions of gap junctions and the contractile activity of SMCs; the specimens analyzed did not have microscopically abnormalities and they did not show the presence of leukocyte infiltrates.9
Supporting this hypothesis, we found that, similarly to the contractile impairment of MS, the analysis performed on colonic SMCs showed a reduced ACh-elicited contraction. Most of the in vitro studies conducted on colonic smooth muscle tissue of DD patients regards DD, complicated with acute diverticulitis; interestingly the present study is focused on SUUD, for which little evidence is present regarding the possible mechanism responsible for the presence of symptoms in the absence of macroscopically alterations of the intestinal mucosa.
Moreover, colonic SMC contractility was not previously investigated in SUDD patients, although it has been shown to be impaired in other pathological gastrointestinal conditions such as irritable bowel syndrome (IBS) and symptomatic gallstone disease.27,28 The impairment of contractility reported in these disorders has been mainly related to a state of low grade inflammation; in IBS, inflammatory mediators, produced by colonic mucosal layer, seem to be responsible for the impaired contractility, which is similar to our study as represented by a significant decrease in basal tone and ACh-elicited contraction of MS and a significant shortening and impairment of ACh-elicited contraction of SMCs.27,29 The similarity between IBS and our patients in terms of symptoms is very interesting as both conditions are characterized by absence of macroscopic or histological inflammation; this evidence could suggest a common pathogenic mechanism related to a muscle disorder.
In IBS, alterations in gut microbiota significantly modify mucosal barrier integrity, provoking an important modulation of gut neuromuscular junction probably responsible for the impairment of intestinal motility.30 In the present investigation, colonic mucosal analysis performed with the Ussing Chamber system has shown, for the first time, a significant increase of permeability in SUDD patients, compared to controls, characterized by a significant reduction of TEER and a significant increase in paracellular permeability to FD-4. Intestinal epithelial barrier function plays a key role in maintaining the right equilibrium between selective permeability of needed nutrients and the possible passage of harmful entities including microorganisms and luminal pro-inflammatory factors.30-32
Interestingly, in this investigation, alterations in TJs were detected with immunohistochemistry, supporting the findings of altered permeability detected by the Ussing Chamber system. It is well known that the alteration of TJ barrier under inflammatory conditions is strictly linked to the pathogenesis of intestinal disease. Impaired TJ barrier determines a higher epithelial permeability contributing to intestinal epithelial damage.33,34 Both in human inflammatory bowel disease and in experimental models of intestinal inflammation, similar structural and functional changes in TJ has been observed,35-36 which were associated with decreased main TJ proteins. However, no specific studies have been extensively conducted to explore TJ complex and intestinal barrier integrity in SUDD, in which mucosal alteration and inflammation have not be found with routine histological analysis, while the presence of an impaired expression of TJ protein complex was previously observed in other GI diseases in which no histological features of a significant inflammation were detected.17,19,31,32
It could be hypothesized that, in SUDD patients, in conditions of altered permeability, a low-grade translocation of intestinal microbes and immune-stimulatory bio-products from the gut lumen through the mucosa, could be responsible for a chronic immune activation that, if prolonged, could be responsible for disease progression and onset of complications.37
In conditions of severe inflammation, such as sepsis, the altered intestinal motor function resulted to be associated with inflammation-induced changes of intestinal muscle and with an increased colonic permeability.38 We have previously demonstrated, in an in vitro model, that human colonic mucosal exposure to LPS alters muscle cell contractility and that this impairment could be determined either to LPS translocation, which directly damages the smooth muscle contractility, or to the mucosal production of free radicals and inflammatory cytokines.14
It is well known that intestinal bacteria, which translocate through the mucosa, are able to activate immune cells binding TLRs on their surface and result in the activation of NF-kB with secretion of pro-inflammatory mediators. This is one of the main mechanisms that mediates the suppression of muscle cell contractility in conditions such as inflammatory bowel disease.39 It has also been previously described of a vast immunological pattern of innate and adaptive cell markers, including TLRs, which could play a role in the pathogenesis and clinical development of the DD; Cianci et al12 showed its variations before and following the antibiotic treatment with rifaximin in DD patients. Despite these evidences, in the present investigation the TLR4 mucosal expression did not reveal any differences in patients with SUDD compared to controls. These controversial results could be determined by the different disease stages of DD patients (diverticulosis without symptoms vs DD in patients with abdominal pain and diarrhea and/or diverticulitis). Indeed, unlike the present study, previous investigations on colonic MS motility of DD were mainly conducted in patients with diverticulitis in which inflammation plays a prominent role in the motor abnormalities.
In conclusion, the results of the present study support the hypothesis that SUDD, in absence of clear inflammation, could have a pathogenic mechanism more similar to functional bowel disorders, such as IBS, in which an increased intestinal mucosal permeability is related to altered colonic motility probably responsible for symptoms genesis. Further studies are needed to clarify the exact pathogenic mechanism underlying SUDD, maybe correlating the effect of Western diet and altered microbiota to intestinal permeability changes in these patients.
Conceptualization and methodology: Annamaria Altomare, Michele P L Guarino, Michele Cicala, and Silvia Cocca; data curation: Manuele Gori, Simone Carotti, Maria Francesconi, Mentore Ribolsi, and Giuseppe Perrone; writing and original draft preparation: Annamaria Altomare, Michele P L Guarino, Manuele Gori, Simone Carotti, Silvia Cocca, Sara Emerenziani, and Mentore Ribolsi; supervision: Michele Cicala and Sergio Morini; and reviewing and editing: Annamaria Altomare, Michele P L Guarino, Sergio Morini, Michele Cicala, Mentore Ribolsi, Giuseppe Perrone, Simone Carotti, and Silvia Cocca.