Introduction
Inflammatory bowel diseases (IBD), which include Crohn’s disease (CD) and ulcerative colitis (UC), represent an important diagnostic and therapeutic problem in gastroenterology. Traditionally, these two entities are distinguished as UC being limited to the mucosa in the large intestine and CD involving the full thickness of the wall and occupying any sections of the gastrointestinal tract. The increasingly younger age of onset and the serious therapeutic challenges make the search for answers about the pathogenesis and treatment of IBD an ever-present and important issue [1, 2].
In recent years, there has been significant progress in research on the pathogenesis of IBD. The results of many studies indicate that gut microbiota and immune dysregulation play relevant and disease activity-related roles in the development and flare of inflammation during IBD [3, 4]. It has been hypothesized that the disease may result from an abnormal immune response directed against the normal intestinal microbiota or be a consequence of a loss of mucosal barrier integrity, whereby commensal bacteria passing through the mucosa stimulate an inflammatory response [5].
The mucosal intestinal barrier, which is formed, among other things, by tightly adherent epithelial cells and a layer of mucus, limits the communication of the intestinal microbiota with the host immune system. Disruption of this “buffer zone” might promote “leaky gut”. The tight junctions (TJ) composed of proteins including claudins, occludins, zonula occludens (Zos), cingulin, junctional adhesion molecules (JAMs) and actin are crucial for proper barrier integrity. Up- or down-regulation of some proteins constituting TJ can contribute to bacterial translocation in IBD [6, 7]. This is reflected in the release of bacterial products such as endotoxins – lipopolysaccharides (LPS), Gram-negative bacteria components – that penetrate into the circulation. In fact, LPS are the first bacterial particles to interplay with the immune system [8]. Consequently, the cascade of the inflammatory process is started, and B and T helper lymphocytes are activated, leading to the increase in the synthesis of immunoglobulins and the production of cytokines. The disturbance of the balance between pro- and anti-inflammatory cytokines further intensifies the inflammatory process. LPS are important components in triggering a vicious circle of inflammation in the gut. Among other things, LPS promotes the expression of interleukin (IL)-8 and, through the interaction with toll-like receptor 4 (TLR4), triggers the expression of nuclear factor kappa light chain enhancer of activated B cells (NF-κB), which then triggers a further chain of inflammatory reactions involving tumor necrosis factor α (TNF-α) and IL-12 [9, 10].
Several clinical indexes are used to assess IBD activity, but these do not always reflect the true severity of inflammation. The only gold standard for assessing deep histological remission is gastrointestinal endoscopy with biopsy. Unfortunately, endoscopic examinations are invasive, uncomfortable for patients and expensive, which makes them a less than ideal method for monitoring disease progression. Systemic, non-specific inflammatory markers such as C-reactive protein (CRP), erythrocyte sedimentation rate (ESR) in blood or calprotectin in stool, which are all increased in IBD patients, are among the commonly used markers [11]. The search is still on for simple and non-invasive markers that would additionally allow the assessment of the current inflammatory state in IBD.
The question is, therefore: What is the usefulness of assessing markers of bacterial translocation, and can their monitoring be applied in clinical practice?
Aim
The study aimed to determine the correlation between markers of bacterial translocation and the proinflammatory response in patients with active, newly diagnosed IBD and at the early stage of disease remission.
Material and methods
Patients
Patients hospitalized in the Department of Pediatrics, Gastroenterology and Nutrition were eligible for the study. The inclusion criteria were as follows: age over 2 years, confirmed new diagnosis of IBD, excluded bacterial infections in the gastrointestinal tract (i.e. negative antigen tests and/or cultures for Clostridium difficile, Yersinia enterocolitica, Salmonella and Shigella sp., enteropathogenic E. coli strains and Campylobacter), no confirmed immune deficiencies or other inflammatory diseases, and informed consent to participate in the study. The study group consisted of pediatric patients aged 5 to 18 years newly hospitalized due to active IBD, both CD and UC. The diagnosis was made based on the clinical picture, endoscopic and histopathological findings and imaging studies. The clinical activity of the illnesses was assessed based on the Pediatric Crohn’s Disease Activity Index (PCDAI) and the Pediatric Ulcerative Colitis Activity Index (PUCAI) accordingly. A value of less than 10 points was taken as remission of clinical activity.
The control group consisted of non-IBD, healthy children without any known inflammatory or chronic condition and without any symptoms from the gastrointestinal tract.
Methods
A marker of bacterial translocation (LPS), proinflammatory cytokines (IL-8, IL-12 and TNF-α) and antimicrobial proteins (β-defensins) were obtained at the outbreak of the illness and then 1 and 3 month after introduction of the treatment.
Blood samples were collected at defined periods. After 30 min, when the clot was created, the samples were centrifuged for 15 min at 1000 x g. The serum was stored at –80°C until measurements.
The IL-8, IL-12 and TNF-α concentrations were measured using commercially available enzyme-linked immunosorbent assay (ELISA) produced by R&D Systems, Inc (MN, USA, catalog number: D8000C, DP400, HSTA00D). For bacterial lipopolysaccharides (LPS) and human β-defensins, ELISA kits, produced by MyBiosource, Inc (CA, USA, MBS702450), were used. All assays were carried out in accordance with the manufacturer’s instructions.
CRP (Vitros, Ortho Clinical Diagnostics, NY, USA) levels and ESR (Sediplus S100, Sarstedt Ag & Co, Germany) were obtained simultaneously with the samples collected for measurements of detected inflammatory biomarkers.
Statistical analysis
Statistical analysis was performed with Statistica 13.0 software (TIBCO Software Inc., Palo Alto, CA, USA) using the Mann-Whitney U test, the Kruskal-Wallis rank test to compare independent variables, the Friedman test with Kendall’s coefficient of concordance and Spearman’s correlation rank test. P-value < 0.05 was considered statistically significant. Cytokine concentrations were expressed as median values and ranges.
Results
There were 33 participants included in the study (15 girls), with a mean age of 169.5 ±36.2 months, 27 of them with active IBD (CD = 17, UC = 10) and 6 controls. Throughout the study period, all patients with CD were treated with 5-ASA and azathioprine. The same treatment was given to 2 patients in the UC group, and the others were treated exclusively with 5-ASA. None of the patients in the study were treated with corticosteroids or biologics. Table I contains clinical and demographic data of the study group.
Table I
Descriptive characteristics of study group
During the active phase of the disease, the levels of the inflammatory markers were higher than in the control group. Statistically significant differences were found between IBD patients and controls for LPS, TNF-α and IL-12 (21.83 pg/ml vs. 10.26 pg/ml, 1.74 ng/ml vs. 0.83 ng/ml, 122.71 vs. 65.3 pg/ml, p < 0.05 respectively). During the first 3 months of treatment, all patients showed decreased disease activity as assessed by clinical scales appropriate to the diagnosis. Patients with CD had a mean ± SD PCDAI of 23.7 ±13.4 points at diagnosis, decreasing to 8.2 ±11.0 and 2.5 ±7.0 points, respectively, at subsequent measurements. In the UC group, the mean ± SD PUCAI score was 53.1 ±16.0 points at diagnosis, then 11.5 ±17.6 and 4.0 ±7.7 points at subsequent measurements. One patient in each group did not achieve clinical remission. There was also a significant decrease in ESR and CRP values (Table II).
Table II
Lipopolysaccharide and cytokine concentrations in subgroups. Results are expressed as median values and ranges.
All proinflammatory cytokines decreased, but significant down-regulation was observed only in relation to IL-12 (129.21 vs. 82.98 pg/ml, p < 0.05) in CD and IL-8 (32.72 vs. 20.97 pg/ml, p < 0.05) in UC patients. Patients with CD also showed a significant change in β-defensin levels (24.96 vs. 19.97 pg/ml, p < 0.05). After a 3-month treatment period, TNF-α levels decreased but did not reach values in healthy children. Interestingly, at the same time, LPS concentration further increased in both IBD groups (Table II). Furthermore, it was found that in patients with a baseline high ESR, LPS concentration increased in subsequent measurements.
In patients with UC, a significant correlation was found between baseline disease activity (PUCAI) and IL-8 (p = 0.027). There was no other correlation between the change in PCDAI or PUCAI and the other inflammatory markers tested. Only in the group of patients with CD was there a statistically significant correlation between ESR and CRP at each measurement point (p = 0.01, p < 0.001, p = 0.01, respectively). Also, a statistically significant relationship was found between CRP and PCDAI at three consecutive measurement points (p = 0.03, p = 0.01, p = 0.005, respectively). However, we found no correlation between ESR and CRP and the interleukins tested.
Discussion
In the study group, we found high TNF-α, IL-8 and IL-12, which were progressively reduced with improved clinical status as assessed by PCDAI and PUCAI. In contrast, LPS was still elevated in patients in early remission.
Cytokines are important players contributing to both the intensification and inhibition of the inflammatory cascade in all inflammatory diseases. Proinflammatory cytokines are also a target for the treatment of many diseases, including IBD. TNF-α plays an important role in immune system development and inflammation, and anti-TNF treatment was the first successfully introduced biological therapy in IBD patients [12, 13]. TNF-α is produced by tumor cells and epithelial, endothelial and immune cells. It is up-regulated in both UC and CD and, to some extent, correlates with disease activity, although its isolated evaluation is less reliable than in combination with other cytokines. Another of the cytokines studied, IL-8, belongs to the chemokine family and is secreted by monocytes, neutrophils, fibroblasts, mast cells and endothelial cells, as well as many others. This cytokine has proinflammatory effects on immune cells: macrophages/monocytes microbial exposure results in IL-8 release [14]. In response, IL-8 pulls inflammatory cells (neutrophils and lymphocytes) to the site of inflammation, increasing the severity of mucosal damage and thus increasing the severity of inflammation [15]. The latest ex vivo research studies confirm the increase of IL-8 in the intestinal mucosa under the influence of pathogenic bacterial flora [16]. In our study, like Bourgonje et al., we observed higher IL-8 levels at diagnosis in patients with UC but not with CD [17]. There was also a significant correlation between baseline disease activity (PUCAI) and IL-8 (p = 0.027). Furthermore, we found significant down-regulation of IL-8 during remission induction in UC patients. Also, Bertani et al. confirmed that in vedolizumab-treated UC patients, a concomitant decrease in IL-6 and IL-8 at week 6 of treatment predicted mucosal healing after 1 year of endoscopic assessment [18]. IL-12 is a natural killer cell stimulatory factor. It is produced by antigen-presenting cells: monocytes/macrophages, dendritic cells and B lymphocytes. IL-12 has multiple effects on T lymphocytes and natural killer cells: it stimulates cytotoxicity, proliferation and cytokine production, including interferon-γ and TNF-α. IL-12 is also of interest because it shares a common subunit with IL-23: p-40. The monoclonal anti-p40 antibody ustekinumab blocks IL12/Il23 and successfully treats UC and CD [19–21]. Our study showed significant down-regulation of IL-12 during remission induction in CD patients. High concentrations of IL-12 were observed primarily in the early, active phase of CD [22]. This is also confirmed by the results of our study, as the highest IL-12 levels were observed in newly diagnosed, untreated CD patients. Patients with UC also had elevated levels of IL-12, although to a lesser extent than in CD. In a study assessing the usefulness of different cytokines in clinical practice, Obraztsov et al. found that a combination of 7 cytokines (TNF-α, IL-12, IL-8, IL-2, IL-5, IL-1β, and IFN-γ) is a strong predictor of clinical response to anti-TNF induction therapy in patients with UC [23]. Also, in our study, we observed a reduction in clinical activity and levels of TNF-α, IL-12, and IL-8, indicating the usefulness of these cytokines as potential markers to assess response to treatment.
As a gut microbiome appears to play a role in both initiating and extinguishing the inflammatory process, we attempted to assess whether LPS concentrations, indicative of bacterial endotoxin passage through the leaky mucosa, and β-defensins, proteins with antimicrobial properties, correlate with disease activity and other inflammatory markers. In the investigated group of patients, despite a decrease in the levels of the cytokines and a decrease in clinical disease activity, we observed an increase in LPS levels. LPS play an important role in activating the inflammatory process and damaging the intestinal barrier [24]. It has been shown that, in the presence of LPS, concentrations of important proteins such as occludin and claudins, which affect the integrity of the intestinal barrier, are reduced. In addition to decreasing the expression of intestinal barrier proteins, LPS increases the concentration of IL-8 [25]. However, we were not able to identify this relationship in our patients. This may be because the changes occurring in the mucosa are exactly where they are most strongly expressed, and evaluation in blood serum is an under-sensitive method. LPS also stimulates other inflammatory factors such as TNF-α, IL-6, and IL-1β [26]. It is, therefore, likely that the lack of full normalization of TNF-α in the study group was due to increased concentrations of LPS. The constant presence of LPS in the circulation, even at low concentrations, may be responsible for the persistence of inflammation, which consequently leads to the development of many diseases, including obesity, metabolic syndrome, and autoimmune diseases [24]. The results of some studies also suggest that an increase in plasma LPS levels, which promotes endotoxemia, contributes to the occurrence and development of colorectal cancer [27]. This observation is of particular relevance for patients with IBD, as long-standing disease increases the risk of developing colorectal cancer. The role of LPS as a factor in initiating or maintaining inflammation in the gastrointestinal tract has not yet been thoroughly explained. As LPS is involved in activating a number of pro-inflammatory pathways, interrupting them by targeted therapy may be beneficial in improving mucosal barrier integrity or maintaining remission.
β-defensins belong to a family of proteins known as antimicrobial proteins. Among other locations, they are present in the epithelium of the small and, in high numbers, in the large intestine. They possess antimicrobial properties against Gram-negative and Gram-positive bacteria and promote the immune response. β-defensins are weakly present in the colon without inflammation but are induced during inflammation, activated, among other triggers, by LPS [28]. Patients with CD show lower expression than patients with UC due to the different mucosal defense responses in the two diseases [29]. In addition, intestinal defensins are upregulated as an early response to inflammation, which was also observed in our study. This was especially true for UC patients. β-defensins exhibit anti-inflammatory effects through effects on mucosal tightness and inhibition of the NF-kappa B pathway [30]. Colonic CD is suspected to be associated with impaired production of β-defensins [31]. In the patient group studied, β-defensin levels were lower in patients with CD than UC, but the differences were not statistically significant. This may be related to the small size of the two groups and the fact that the CD group did not include patients exclusively with colonic involvement.
Enhancing the role of β-defensins may have a role in the treatment and maintenance of remission, but assessing their blood levels appears to have little clinical utility.
The weaknesses of our study are the observation time, which is too short to assess the longevity of remission, and the small group of patients. Another weakness of our study is that we were able to correlate the investigated markers of bacterial translocation with inflammatory markers in the blood (ESR, CRP), but not with fecal inflammatory markers such as calprotectin or lactoferrin. However, we believe that finding a sensitive, non-invasive marker or combination of markers that, when appropriately validated, will help to assess disease activity and response to treatment is of paramount importance, especially for children with IBD [32]. We also believe that the results we obtained are pilot in nature and require a larger-scale clinical study involving larger groups of patients.
Conclusions
Our study indicates that IL-12 in CD and IL-8 in UC can be considered markers of inflammation in non-invasive monitoring of disease activity. However, our results suggest that clinical improvement and reduction of systemic inflammatory markers do not necessarily imply a complete cessation of the inflammatory cascade. Elevated levels of TNFα and LPS found in patients in early remission (higher than in healthy children) may be a marker of subclinical inflammation. This requires further studies involving a larger group of patients and a longer follow-up time. It can also be speculated that, perhaps in the future, LPS, like TNF-α, may become a therapeutic target in the treatment of IBD in children.
