Correlation between changes in gut flora and serum inflammatory factors in children with noninfectious diarrhea.

Introduction

Diarrhea is the leading cause of illness and death among infants and children and thus represents a major threat to pediatric health, particularly in developing countries.[1] Diarrhea is caused by multiple factors and pathogens and is characterized by the following clinical features: fluid and electrolyte imbalances, frequent stools, and loose stool for >3 days.[2] Diarrhea can be classified as acute, prolonged, or persistent according to the disease course[3] and can be of either noninfectious or infectious origin. Noninfectious diarrhea is common in infants[4,5] because their digestive system development is immature, gastric acid and digestive enzyme activity is low, and the ability to adapt to food changes is poor. Furthermore, because children grow rapidly, their nutrient needs are relatively high and their digestive organs are under strain. For all these reasons, children easily develop digestive function disorders.

There are many causes of noninfectious diarrhea in children and infants. When infants are fed inappropriately (mostly from artificial feeding), they can develop diarrhea.[6,7] Moreover, because the development of the infant digestive system is incomplete, optimal digestion and absorption are not achieved, thus increasing pressure in the gut cavity and leading to diarrhea.[8] In infants with chronic diarrhea, congenital or heredity factors result in trait diarrhea, anaphylactic diarrhea, glucose-galactose malabsorption diarrhea, dyslipidemia diarrhea, endocrine diarrhea, congenital chloride diarrhea, and immunodeficiency diarrhea. Gut flora and serum inflammatory factors play important roles in the pathogenesis of diarrhea.[9] There is no specific treatment at present for noninfectious diarrhea, and alleviating symptoms is generally the clinical goal.

Clinically, to diagnose noninfectious diarrhea, physicians may request a stool sample to be examined in a laboratory under a microscope to detect the presence of bacteria, parasites or other microorganisms. Traditional testing methods for microbial pathogens, such as stool culture, antigen testing, and microscopy, have not generally been successful in patients with noninfectious diarrhea. Standard laboratory diagnosis of diarrheal illness has relied on microscopy, antigen detection, and culture. However, these methods are often time-consuming, require multiple orders and tests, and interpretation is often subjective. Notably, PCR and sequencing have been increasingly applied in recent years for detection of enteric pathogens, including bacteria and associated toxins, parasites, and viruses, with dramatically improved accuracy and sensitivity.[10,11] A recent study demonstrated that comprehensive molecular testing using the BioFire FilmArray gastrointestinal PCR panel could rapidly identify a broad range of pathogens and determine the clinical impact of the most common agents of infectious diarrhea in approximately 1 hour.[12]

An understanding of diarrhea in terms of its genetic basis and other pathophysiological mechanisms, via fecal analysis and blood tests, will enable doctors to assess children with diarrhea and administer treatments accordingly. In this study, gut flora and serum inflammatory factors in children with noninfectious diarrhea were studied, and the correlations between them were also analyzed.

Materials and methods

Patients and groups

From February 2014 to June 2017, 93 children diagnosed with and treated for noninfectious diarrhea at our hospital and 80 healthy children were enrolled in this study. Participants included 90 males and 83 females with ages ranging from 4 months to 7 years (average: 3.98 ± 1.23 years). This study was approved by the Ethics Committee of Liaocheng Second People's Hospital. The parents or legal guardians of participants provided informed consent to participate in the study.

Inclusion and exclusion criteria

The inclusion criteria were as follows: (1) following admission, the child was diagnosed with noninfectious diarrhea; (2) the number of stools per day was >5; and (3) stool traits changed according to medical history, clinical manifestation, and related auxiliary examinations. The exclusion criteria were as follows: (1) diagnosis of infectious diarrhea; (2) severe vomiting and dehydration; or (3) severe heart, lung, liver, or kidney dysfunction or other serious diseases.

Gut flora

A sterile container was used to collect the stool samples. Following serial dilution, the samples were inoculated on aerobic culture medium to culture Escherichia coli and Enterococcus and on an anaerobic medium to culture lactic acid bacteria and Bifidobacterium. Following incubation for 3 days at 37°C, the number of colonies was counted. The results were expressed as the logarithmic value of colony-forming units (CFUs) per gram of wet feces weight (log CFU/g).

Determination of serum interleukin (IL)-1, -6, and -17 and tumor necrosis factor (TNF)-α levels

Levels of serum inflammatory factors, including IL-1, IL-6, IL-17, and TNF-α, were measured using an ELISA kit (Abcam, UK) according to the manufacturer’s instructions. Briefly, a standard curve was constructed from a 10-fold dilution series in phosphate-buffered saline. The samples were collected from the patients’ arm veins prior to hospital admission and serum was separated by centrifugation. For ELISA, 100 µL of the standard material and samples were added to wells of a 96-well plate coated with the antibody and incubated at 37°C for 90 minutes. Then, the liquid was discarded and 100 µL of biotin-labeled antibody was added. After washing each well three times with 250 µL of wash buffer, 100 µL of working solution was added, and the plate was incubated for 30 minutes at 37°C. Then, 90 µL of developing solution containing 3,3′,5,5′-tetramethylbenzidine (TMB) was added to each well. The plate was incubated for 20 minutes at 37°C in the dark. Next, 100 µL of TMB termination solution was added, and the mixture was evenly mixed. Finally, the concentrations of IL-1, IL-6, IL-17, and TNF-α were calculated by measuring the optical density at 450 nm using a microplate reader.

Statistical analysis

Data were analyzed using SPSS version 17.0 software (SPSS Inc., Chicago, IL, USA). All measurement data were expressed as means ± standard deviations. The paired t-test was used for intergroup comparison, the chi-square test was used for categorical data, and analysis of variance was used for multigroup comparison. Values of p < 0.05 were considered statistically significant.

Results

General characteristics of each group

The sex ratio, age, body mass index, and other characteristics of both groups were compared (Table 1). The abundance of E. coli and Enterococcus in feces was significantly higher in the diarrhea group than in the healthy control group (p < 0.05) (Figure 1a-b). The abundance of Bifidobacterium and lactic acid bacteria in feces was significantly lower in the diarrhea group than in the healthy control group (p < 0.05) (Figure 1c-d). Serum IL-1, IL-6, IL-17, and TNF-α levels in the diarrhea group were significantly higher than those in the healthy control group (Figure 2a-d). The abundance of E. coli and Enterococcus was positively correlated with serum IL-1, IL-6, IL-17, and TNF-α levels, whereas the abundance of Bifidobacterium and lactic acid bacteria was negatively correlated with serum IL-1, IL-6, IL-17, and TNF-α levels. All of these correlations were statistically significant (p < 0.05) (Table 2)

Table 1.

General characteristics of the diarrhea and healthy control groups.

Groups	Cases	Sex, M/F	Age (years)	Body mass index (kg/m2)
Diarrhea	93	46/47	3.24 ± 2.12	22.03 ± 1.14
Healthy control	80	44/36	3.87 ± 1.49	21.97 ± 2.08
Chi-square statistic	–	0.062	0.038	0.087
P value	–	n.s.	n.s.	n.s.

M, male; F, female; n.s., not significant.

Figure 1.

Comparison of the abundance of bacterial cells in gut flora between the diarrhea group and the healthy control group. (a) E. coli, (b) Enterococcus, (c) Bifidobacterium, and (d) lactic acid bacteria. The abundance of E. coli and Enterococcus in the diarrhea group was significantly higher than in the healthy control group. The abundance of Bifidobacterium and lactic acid bacteria in the diarrhea group was significantly lower than in the healthy control group (*p < 0.05).

Figure 2.

Comparison of serum inflammatory factor levels between the diarrhea and healthy control groups by ELISA. (a) Serum IL-1 level, (b) serum IL-6 level, (c) serum IL-17 level, and (d) TNF-α level. Serum IL-1, IL-6, IL-17, and TNF-α levels in the diarrhea group were significantly higher than those in the healthy control group (*p < 0.05). IL, interleukin; TNF, tumor necrosis factor.

Table 2.

Correlations between gut flora and serum levels of IL-1, IL-6, IL-17, and TNF-α.

Groups	IL-1		IL-6		IL-17		TNF-α
	r	P	r	P	r	P	r	P
Escherichia coli	0.302	<0.05	<0.05	<0.05	0.462	<0.05	0.257	<0.05
Enterococcus	0.238	<0.05	0.329	<0.05	0.318	<0.05	0.302	<0.05
Bifidobacterium	−0.225	<0.05	−0.356	<0.05	−0.303	<0.05	−0.291	<0.05
Lactic acid bacteria	0.338	<0.05	−0.299	<0.05	−0.421	<0.05	−0.424	<0.05

IL, interleukin; TNF, tumor necrosis factor.

Gut microflora

The number of gut microflora in the fecal samples from the two groups was compared. The mean numbers of E. coli and Enterococcus in the diarrhea group were 8.27 ± 0.13 log CFU/g and 7.81 ± 0.08 log CFU/g, respectively, significantly higher (p < 0.05) than in the healthy control group (6.88 ± 0.21 log CFU/g and 5.93 ± 0.16 log CFU/g, respectively) (Figure 1a-1b). The numbers of Bifidobacterium and lactic acid bacteria in the diarrhea group were 6.02 ± 0.60 log CFU/g and 5.89 ± 0.46 log CFU/g, respectively, significantly lower (p < 0.05) than in the healthy control group (7.87 ± 0.47 log CFU/g and 7.33 ± 0.25 log CFU/g, respectively) (Figure 1c-1d).

Comparison of inflammatory factor levels in serum

Levels of the inflammatory factors IL-1 (pg/mL), IL-6 (pg/mL), IL-17 (pg/mL), and TNF-α (ng/mL) in the sera of both groups were compared. Serum IL-1, IL-6, IL-17, and TNF-α levels in the diarrhea group were 83.62 ± 3.98 pg/mL, 54.68 ± 4.92 pg/mL, 52.72 ± 4.76 pg/mL, and 63.41 ± 5.09 ng/mL, respectively, significantly higher (p < 0.05) than levels in the healthy control group (13.09 ± 5.81 pg/mL, 13.09 ± 5.81 pg/mL, 28.93 ± 5.02 pg/mL, and 10.28 ± 4.22 ng/mL, respectively) (Figure 2a–d).

Correlation between gut flora and serum inflammatory factors in children with noninfectious diarrhea

In the diarrhea group, Pearson’s correlation coefficient showed that the abundance of E. coli was positively correlated with serum IL-1, IL-6, IL-17, and TNF-α levels (r = 0.302, 0.294, 0.462, and 0.257, respectively). The abundance of Enterococcus was also positively correlated with serum IL-1, IL-6, IL-17, and TNF-α levels (r = 0.238, 0.329, 0.318, and 0.302, respectively). In contrast, the abundance of Bifidobacterium was negatively correlated with serum IL-1, IL-6, IL-17, and TNF-α levels (r = −0.225, −0.356, −0.303, and −0.291, respectively), as was the abundance of lactic acid bacteria (r = −0.338, −0.299, −0.421, and −0.424, respectively). All of these correlations were statistically significant (p < 0.05) (Table 2).

Discussion

There are hundreds of species of bacteria in the human digestive tract and thousands of microbes per gram of feces in the large intestine; these include anaerobic, facultatively anaerobic, and aerobic bacteria. The human gut flora is established within 1 to 2 years of birth and is critical to gastrointestinal function and human health.[13] When the gut flora becomes imbalanced, this stimulates the release of inflammatory factors, which can be harmful to the human body.[14] It has been reported that the occurrence of gut disease is closely related to gut microecology, and the role of gut flora imbalances in the development of gut diseases is becoming more and more appreciated.[15] The main microbes in the small intestine are Gram-positive cocci and rod-shaped bacteria, which contribute extensively to gut function; however, excessive microbiota growth leads to gut dysfunction.[16] The largest bacterial ecosystem in the human body is in the large intestine, and is mainly composed of specific anaerobes such as Bacillus and Bifidobacterium. Our study showed that in comparison with the healthy control group, abundance of E. coli and Enterococcus was increased significantly and abundance of Bifidobacterium and lactic acid bacteria decreased significantly in the diarrhea group. Thus, children with noninfectious diarrhea experience gut dysbacteriosis in vivo, with levels of probiotics decreasing and those of harmful bacteria increasing.

Studies have shown that regulation of gut flora and serum inflammatory factors can improve the therapeutic effect of interventions.[17,18] In gut diseases, proinflammatory cytokines and anti-inflammatory factors play a key role in pathogenesis. Yeung et al.[19] established a mouse model and found that the expression of IL-6, IL-1β, TNF-α, and other inflammatory factors increased and the expression of IL-31 decreased in lesions of the small intestine. Furthermore, changes in the expression levels of the molecules mentioned above were suppressed and inflammation of the gut mucosa was reduced following probiotic treatment. Kim et al.[20] analyzed the sera of 538 patients with diarrhea and discovered that as a diagnostic marker of inflammatory disease, serum C-reactive protein could discriminate between noninflammatory diarrhea and inflammatory diarrhea. IL-1, IL-6, TNF-α, and other inflammatory factors are considered as the key proinflammatory mediators involved in the initiation and maintenance of inflammation during disease development.[21] In diarrhea, maturation of IL-1 depends on the activation of caspase-1, and mature IL-1 aggravates the proinflammatory response.[21] At different stages of diarrhea, the expression level of IL-6 may change, informing prognosis.[22] In our study, we found that serum IL-1, IL-6, IL-17, and TNF-α levels in children with noninfectious diarrhea were significantly higher than those of healthy children, indicating that these levels may be related to the development of noninfectious diarrhea.

The close relationship between gut flora and inflammatory factors has been widely recognized by many researchers, suggesting that altering the gut flora community could represent a potential therapeutic mechanism for gut diseases. An analysis of the correlation between gut flora and inflammatory factors in patients with Crohn’s disease has been previously conducted, and it was found that serum inflammatory factors were closely associated with gut flora.[23] The results of our study showed that the abundance of E. coli and Enterococcus in children with noninfectious diarrhea was positively correlated with serum IL-1, IL-6, IL-17, and TNF-α levels, whereas abundance of Bifidobacterium and lactic acid bacteria was negatively correlated with serum IL-1, IL-6, IL-17, and TNF-α levels. This finding suggested that changes in the gut flora may affect secretion of inflammatory factors in vivo. However, a limitation in this study was that correlations between bacterial abundance (E. coli, Enterococcus, Bifidobacteria and lactic acid bacteria) without) were not assessed among healthy individuals. Although our comparative analysis of noninfectious diarrhea and inflammatory markers is potentially interesting, the same correlations should be assessed in the 80 healthy control children as well. The current assessment of the microbiota using simple agar cultures needs to be improved. Based on our initial findings, the regulation of gut microbiota and inflammatory factors is expected to become a new and powerful approach for the diagnosis and treatment of gut diseases.[24,25] Our initial findings showed an association between gut bacteria and cytokines, in line with a previous study on osmotic diarrhea suggesting that temporary changes in cytokine levels influenced the lasting IgG response against bacteria.[26] Thus, multiple factors may have potential as biomarkers for the diagnosis of noninfectious diarrhea. Nevertheless, as cytokine measurements were performed in serum/blood, this may not reflect the state of local inflammation. We propose that the overgrowth of bacteria may trigger the release of proinflammatory cytokines systemically, although the exact effects on systemic or local inflammation remain to be elucidated.

In conclusion, our data reveal changes in the abundance of certain bacteria in the feces of children with noninfectious diarrhea. These changes were correlated with abnormal increases in serum inflammatory factor levels.