This cross-sectional study included 21 Egyptian patients with active UC that were recruited from the gastroenterology outpatient clinic of the Alexandria Main University Hospital, as well as from its gastroenterology inpatient ward, and 20 Egyptian healthy subjects with matched age, sex, and body mass index, as the control group. The dietary habits of all participants were assessed through a food frequency questionnaire.

Malignancy; recent surgical intervention of the small or large intestine within the last 6 months; infectious diarrhea including bacterial, viral, and parasitic diarrhea excluded by stool analysis, culture, and polymerase chain reaction (PCR) for certain pathogens; a history of prolonged use of antibiotics; nonsteroidal anti-inflammatory drug abuse; a history of corticosteroid, methotrexate, azathioprine, or 6-mercaptopurine use for the past 3 months; biologic therapy; other autoimmune diseases; pregnancy; severe burn; sepsis; chronic renal and liver diseases; diabetes mellitus; pure vegetarianism; alcohol or substance addiction; inability to give consent due to a mental disorder; and children. Patients were required not to have taken any active drugs for the treatment of UC, such as 5-aminosalicylic acid or salazosulfapyridine, one week before entering the study.

The study follows the principles of the Declaration of Helsinki (1964) and was approved by the Medical Research Ethics Committee of the Alexandria Faculty of Medicine, Egypt.

Signed statements of informed consent were obtained from all patients, expressing their acceptance to participate in the study and giving their permission for the results to be published.

All patients and controls had their complete medical histories taken and a full clinical examination. UC was diagnosed, based on clinical, radiologic, endoscopic, and histologic examinations11. UC activity was determined, according to a composite of clinical and endoscopic indexes (the Mayo Clinic Index and the Disease Activity Index), with scores ranging from 0 to 12 points (from no activity to the most severe activity) and sub-scores (combining rectal bleeding, stool frequency, and the Physician’s Global Assessment or the Endoscopy sub-score)12.

Routine laboratory workup was performed, including complete blood count, serum liver enzymes (alanine aminotransferase and aspartate aminotransferase), albumin, bilirubin, HBsAg, and HCVAb, fasting and postprandial blood sugar, serum cholesterol, triglycerides, blood urea nitrogen, creatinine, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), fecal calprotectin, and pANCA.

Colonoscopy was carried out on all patients. Bowel preparation was a low-fiber diet on the day preceding the colonoscopy and a split regimen of 4 L of polyethylene glycol (PEG) solution (or a same-day regimen in the case of afternoon colonoscopy), no longer than 4 h before the colonoscopy. Four biopsy specimens from 5 sites, including the ileum and rectum, were obtained from both affected and normal-appearing mucosa; specimens from different locations were labeled and submitted separately.

The biopsy specimen underwent histopathologic examination.

Fresh stool samples were collected from cases and controls and stored in aliquots at −80 °C for further processing.

DNA was extracted from 180 to 220 mg stool samples using QIAamp® Fast DNA Stool Mini Kit (Qiagen, Germany), according to the manufacturer’s instructions. DNA extracts were stored at −80 °C until PCR testing. Two µl of DNA extract were used in the PCR reaction.

The real-time PCR protocol was performed, as previously described by Tomova et al. (2015)13. Specific PCR primers were used to target selected phyla, genera, or species constituting the gut microbiota (Bacteroides, Prevotella, Ruminococcus, Bacteroidetes, Firmicutes, Bifidobacterium, Lactobacillus, Clostridioides difficile, A. muciniphila and F. prausnitzii), in addition to a broad-range primer targeting the conserved 16SrRNA gene sequences of all the bacteria, whose amplification served as the denominator against which the amplification of other bacteria was estimated.

Amplification was performed in a real-time PCR cycler, the Rotor-Gene Q (Qiagen, Germany) using a SensiFAST™ SYBR® No-ROX PCR kit (Bioline Co., UK). It was performed in 20 µl reaction volumes, containing 4 pmols of each primer. The reaction consisted of initial denaturation at 95 °C for 10 min, followed by 40 cycles of denaturation at 95 °C for 30 s, annealing at 60 °C for 30 s, and extension at 72 °C for 30 s. Melting curve analysis was performed to check the specificity of the amplified products. Quantification of specific bacterial DNA was expressed as relative quantification (the cycle threshold [Ct] at which DNA for a specific target was detected, relative to the Ct at which universal bacterial DNA was detected). Relative quantification was calculated automatically by the Rotor Gene software and expressed as relative fold change13.

The enterotype of all participants was determined, according to the dominant type present in each of the 3 bacteria: Bacteroides (enterotype 1), Prevotella (enterotype 2) or Ruminococcus (enterotype 3).

Data were entered into the computer and analyzed using the IBM SPSS software package version 20.0. (Armonk, NY: IBM Corp., USA). The qualitative data were described, using number and percentage, and the Kolmogorov-Smirnov test was used to verify the normality of distribution. The quantitative data were described, using range (minimum and maximum), mean, standard deviation, median, and interquartile range (IQR). Significance of the results was judged at the 5% level.

To compare different groups, the chi-square test was used for the categorical variables. For the 2 study groups, the normally distributed quantitative variables were compared, utilizing the Student’s t test, and the abnormally distributed quantitative variables were compared, using the Mann-Whitney U test. To compare more than 2 study groups, the Kruskal-Wallis test was utilized for the abnormally distributed quantitative variables. The Pearson coefficient was used to correlate 2 normally distributed quantitative variables and regression was utilized to detect the most independent factor that affected Ruminococcus.

To evaluate the degree of variation of the microbial community structure within a sample, we measured the alpha diversity by employing the Shannon diversity index14, and to evaluate the degree of similarity between cases of UC and their siblings or controls, we employed the Bray-Curtis similarity index15.

The present study was carried out on 21 UC patients. Seventeen of the patients were males and 4 were females, with a 4.3:1 male-to-female ratio. Patient age ranged from 18 to 43 years, with a mean age of 28.90 ± 6.55 years. The control group was comprised of 20 cases; 11 males and 9 females, and their age ranged from 20 to 42 years, with a mean age of 29.75 ± 5.17 years.

None of the patients had a positive family history of UC. Disease duration ranged from 1 to 7 years, with a mean of 1.86 ± 1.56 years. The total Mayo Score ranged from 4 to 12, producing a mean of 9.05 ± 2.38, with 2 (9.5%) mild, 12 (57.1%) moderate, and 7 (33.4%) severe cases. Table 1 shows the different items of the Mayo Score.

In addition, fecal calprotectin ranged from 98-2,810, with a mean of 533.95 ± 714.20, and CRP ranged from 7-102 with a mean of 40.43 ± 34.94. Seventeen of the 21 patients (81%) were negative for pANCA and 4/21 (19%) were positive.

Conventional colonoscopic examination was carried out on all patients. A decrease in the normal vascular pattern and mild erythema was found in 2 patients and a complete loss of the vascular pattern and marked erythema were found in 19 patients. Mucosal friability was mild in 7 patients and severe in 14 patients. Erosions were seen in 12 patients. Severe ulcerations and spontaneous bleeding were found in 7 patients. Pseudopolyps were seen in 3 patients. Table 1 shows the macroscopic and microscopic findings.

Quantification of specific bacterial DNA was not expressed as an absolute number, but rather was relative to the total bacterial DNA present in the stool sample. The mean of the relative difference values of the various bacteria were shown at instances in which the decimal value was low, such as E-05 (4.75 × 10-5 is shown as 4.75E-05).

The bacterial phylum analysis showed that patients with UC had a statistically significant decrease in Firmicutes (p < 0.001), and although Bacteroidetes was decreased, the difference was not statistically significant, when compared with the control group. In addition, although the Firmicutes/Bacteroidetes ratio was lower in the UC cases (1.65), when compared with the healthy controls (2.93), the difference was not statistically significant (Table 2) (Fig. 1).

Figure 1.

Box and whisker graph of the gut microbiome in the study groups. The thick line in the middle of the box represents the median, the box represents the interquartile range (from the 25th to 75th percentiles), the whiskers represent the minimum and maximum: UC: control.

(0.54MB).

Patients with UC showed a statistically significant decrease in Ruminococcus, when compared with the control group (p = 0.001), whereas there was no statistically significant difference between the UC patients and the controls, regarding Bacteroides and Prevotella (Table 2) (Fig. 1). There was also no statistically significant difference between the UC group and the control group, with respect to the Prevotella/Bacteroides ratio (Table 2, Fig. 1).

Regarding the beneficial bacteria, patients with UC showed a statistically significant decrease in F. prausnitzii, in comparison to the control group (p < 0.001). However, there was no statistically significant difference between the UC patients and the control cases, with respect to A. muciniphila, Bifidobacteria, and Lactobacilli (Table 2) (Fig. 1).

For the pathogenic bacterium, Clostridioides difficile, none of the control cases were positive and only one UC case was positive, with a relative abundance of 2.10E-6.

In relation to the 3 severity groups of the UC cases, although the beneficial bacteria (A. muciniphila, Bifidobacterium, F. prausnitzii, and Firmicutes) were higher in the mild cases, than in the moderate and severe cases, the difference was not statistically significant.

When studying the correlation between the relative abundance of the different gut microbiota and disease duration, the total Mayo Score and its items, fecal calprotectin, and pANCA, no statistically significant correlations were detected, except for a significant positive correlation between A. muciniphila and fecal calprotectin (r = 0.613, p = 0.003). There was also a significant positive correlation between Ruminococcus and fecal calprotectin (r = 0.448, p = 0.041) and disease duration (r = 0.467, p = 0.033).

A multivariate linear regression analysis of the parameters affecting Ruminococcus, in the UC cases, produced a statistically significant correlation between the endoscopic score and fecal calprotectin, whereas there was no statistically significant correlation with disease duration (Table 3).

The Shannon diversity index takes both species richness and evenness into account and demonstrated a lower degree of microbial diversity in the UC cases, than in the healthy controls. The median diversity index value for UC was 1.41 (mild cases = 1.24, moderate = 1.42, severe = 1.44) and it was 1.48 in the healthy controls. No statistically significant difference was observed between the UC cases (Table 4).

The Bray-Curtis Similarity Index was performed for assessing the similarity and dissimilarity between the UC cases and healthy controls. Compared with the healthy controls, the mean of dissimilarity (difference) for UC was 57%, ranging from 20-91%.

The mild, moderate, and severe UC cases showed an average dissimilarity, in relation to the healthy controls, of 41, 60, and 58%, respectively. However, no statistically significant difference in severity was observed.

Ten (47.6%) of the UC patients were assigned to enterotype 1, 10 (47.6%) to enterotype 2, and only 1 to enterotype 3. On the other hand, 13 (65%) of the control cases were assigned to enterotype 1, 7 (35%) to enterotype 2, and none to enterotype 3. No statistically significant difference was detected between the 2 groups, regarding enterotype distribution (p = 0.436).