Metabolic Syndrome in Inflammatory Bowel Disease: Association with Genetic Markers of Obesity and Inflammation

Abstract

Background:

This study analyzed poorly understood relationship of two overlapping conditions: metabolic syndrome (MeS) and inflammatory bowel disease (IBD), both associated with inflammation in the visceral adipose tissue.

Methods:

Newly diagnosed 104 IBD patients, of which 50 Crohn's disease (CD) and 54 ulcerative colitis (UC), and 45 non-IBD controls were examined for MeS-related obesity and lipid markers. Th-17 immune genes IL17A, IL17F, IL23A, and TLR9 mRNAs were measured in intestinal mucosa by qRT-PCR. Subjects were genotyped for obesity-associated FTO variant rs9939609 by polymerase chain reaction–amplification refractory mutation system.

Results:

CD was associated with MeS (P = 0.01), while both CD and UC were associated with central obesity (P = 10⁻⁵, P = 0.002, respectively) and low levels of high-density lipoprotein (HDL) cholesterol (P = 5 × 10⁻⁶, P = 6 × 10⁻⁶, respectively). IBD lipid profile was characterized by decreased total and HDL cholesterol, while low-density lipoprotein cholesterol was reduced only in CD. Negative correlations were found between total cholesterol and CD activity index (P = 0.005), waist circumference and IL17A as well as IL17F mRNA levels in inflamed CD colon (P = 0.003, P = 0.001, respectively). Carriers of FTO rs9939609 AA genotype showed increased risk of CD (OR 2.6, P = 0.01).

Conclusions:

MeS, central obesity, and dyslipidemia could be important for IBD pathogenesis. This could influence therapeutic approaches and prevention strategies in high-risk groups.

Introduction

Central obesity, dyslipidemia, impaired glucose metabolism, and arterial hypertension are components of metabolic syndrome (MeS). Recent findings indicate association between MeS and inflammatory bowel disease (IBD), which share common pathophysiological features such as immune imbalance, chronic inflammation, disturbed secretion of adipokines, and increased risk of cardiovascular disease.¹ However, common underlying mechanism is not yet fully understood. It is shown that development of IBD and other MeS-associated diseases such as obesity, insulin resistance, and type 2 diabetes mellitus (DM2) is influenced by altered gut microbiota–host interaction, making intestinal dysbiosis as one of possible cross-links between IBD and MeS.^1

–7 Moreover, data regarding the prevalence of MeS among IBD patients are still scarce.^8,9

Contrary to acquainted assumptions that IBD patients are malnourished, prevalence of obese and overweight patients with Crohn's disease (CD) and ulcerative colitis (UC) is estimated up to 40%.^10

–14 Central obesity, characterized by excess of visceral adipose tissue (VAT) surrounding the intra-abdominal organs, is described in IBD as a local hyperplasia of the mesenteric fat next to the inflamed intestine. This so-called “creeping fat” is associated with transmural inflammation, fibrosis, and stricture formation.¹⁵ VAT expresses and secretes bioactive molecules and hormones, including adiponectin, leptin, omentin, tumor necrosis factor (TNF)-α, resistin, and interleukin (IL)-6. Visceral excessive fat accumulation is associated with features of MeS, prothrombotic, and proinflammatory states.^10,16
–18

It is long known that inflammation may induce dyslipidemia through downregulation of lipoprotein lipase enzyme affected by the action of proinflammatory cytokines TNF-α, IL-6, and interferon (IFN)-γ.^19
–21 In IBD and other chronic inflammatory diseases such as systemic lupus erythematosus (SLE), altered lipoprotein level is a commonly seen condition. Unlike SLE, in which lipoproteins have specific “lupus pattern” mostly depicted with high triglycerides (TGs) and low levels of high-density lipoprotein cholesterol (HDL)-C, reported IBD dyslipidemia profiles have varied across different studies.^19

–22 Moreover, the fact that HDL performs many anti-inflammatory activities indicates that alternation in its level could not be only the effect but also the cause of intestinal chronic inflammation.^23,24

The aim of this study was to analyze association of MeS with IBD occurrence and disease activity. Moreover, we examined correlations of obesity and lipid serum markers with expression of Th17-related genes IL17A, IL17F, IL23A, and TLR9 in intestinal patient samples, as they are shown to be the indicators of IBD-inflamed mucosa.²⁵ The additional value of this study is the novel association analysis of IBD with genetic variant FTO rs9939609, recognized before as the risk factor not only for increased fat mass and obesity but also for alternation of metabolic traits.^26
–28 This nutrigenetic approach could help understanding MeS- and obesity-associated complex diseases and improve disease risk stratification, considering that genetic markers are not influenced by confounding factors such as education, physical activity, social-economic status, and diet.

Materials and Methods

Subjects

The cross-sectional study included 104 newly diagnosed IBD patients successively recruited at the Clinic for Gastroenterology and Hepatology, Clinical Center of Serbia. All patients fulfilled the conventional criteria for IBD diagnosis.^29,30 Patients included in the study have received none of the following drugs: corticosteroids, azathioprine/6-mercaptopurine, methotrexate, and biologic agents. Montreal's classification was used to determine IBD phenotype (Table 1). Patients with indeterminate IBD and <18 years of age were not included in the study. Control group involved 45 non-IBD participants who underwent ileocolonoscopy during the same period due to positive fecal occult blood test with hemorrhoidal disease as the final diagnosis.

Table 1.

Montreal Classification Data of Analyzed Inflammatory Bowel Disease Patients

Montreal classification	CDi (n = 26)	CDc (n = 24)	UC (n = 54)
Localization of disease
L1^a/E1^b (%)	16 (61.5)	NA	16 (29.6)
L2^a/E2^b (%)	NA	15 (62.5)	19 (35.2)
L3^a/E3^b (%)	10 (38.5)	9 (37.5)	19 (35.2)
Phenotype of disease
B1^c/S1^d (%)	11 (42.3)	15 (62.5)	36 (66.7)
B2^c/S2^d (%)	11 (42.3)	6 (25.0)	15 (27.8)
B3^c/S3^d (%)	4 (15.4)^*	3 (12.5)^#	3 (5.6)

CDi—samples obtained from ileal CD mucosa, CDc—samples obtained from colonic CD mucosa, UC—samples obtained from colonic UC mucosa.

All penetrating CDi patients had perianal modifier.

Two CDc patients with penetrating phenotype had also perianal disease.

Localization of CD according to Montreal classification, L1—ileum, L2—colon, L3—ileocolon.

Extent of UC according to Montreal classification, E1—ulcerative proctitis, E2—left side UC, E3—extensive UC.

Behavior of CD according to Montreal classification, B1—nonstricturing, nonpenetrating disease, B2—stricturing, B3–penetrating.

Severity of UC according to Montreal classification, S1—mild, S2—moderate, S3—severe.

CD, Crohn's disease; UC, ulcerative colitis; NA, not applicable.

For the purpose of genetic association analysis of FTO variant, IBD group was expanded to 94 CD and 98 UC patients aiming to increase statistical power of this part of the study. Control group was also expanded to total of 91 participants; of which 38 subjects from the non-IBD group with available DNA material were included, while 53 subjects were selected from general population declaring not having gastrointestinal, autoimmune, or other inflammatory issues (median age 30 years, range 17–76 years, males 42%). Considering that IBD incidence is relatively low (∼10/100,000), subjects from general population were considered as healthy and nonaffected with IBD.

This study was conducted after the approval of The Ethic Committee of Clinical Centre of Serbia, School of Medicine, University of Belgrade, in accordance with the Helsinki Declaration. Written informed consent was obtained from all participants.

Metrics and laboratory data

From all 104 IBD and 45 non-IBD subjects, metric data, as well as peripheral blood and fecal samples, were collected at the time of patients' diagnosis or in case of the control group, diagnosis exclusion.

Medical records were reviewed for the presence of MeS, body mass index (BMI), waist circumference (WC), lipid status (level of total cholesterol (TCh), low-density lipoprotein cholesterol (LDL-C), HDL-C and TG, C-reactive protein (CRP), and fecal calprotectin (FC) levels. Clinical disease activity of IBD patients was assessed by Crohn's Disease Activity Index (CDAI) and Mayo score (MS). The International Diabetes Federation and Adult Treatment Panel (ATP III) diagnostic criteria for MeS included central obesity, defined by specific WC beyond the ethnic standard (Europids: WC in male ≥94 cm; female ≥80 cm), and the presence of two or more of the following: dyslipidemia (TG level ≥1.7 mM; HDL-C < 1.03 mM in males, <1.29 mM in females, or specific treatment of lipid abnormality), impaired glucose tolerance (fasting plasma glucose [FPG] ≥5.6 mM, or treated for DM2), and high blood pressure (BP) (systolic BP ≥130 mmHg or diastolic BP ≥85 mmHg, or treated for arterial hypertension).³¹

Serum CRP levels and FC at the time of IBD diagnosis were used for biochemical disease activity. Using latex particle enhanced immunoturbidimetric assay (PETIA), CRP cutoff values of 5 mg/L were adopted for detection of an inflammatory event. Fecal samples were provided from the first bowel movement in the morning and extracted in accordance with the instructions of commercially available PETIA (BÜHLMANN FCAL turbo test, Bühlmann Laboratories, Switzerland) for quantification of calprotectin, with cutoff values for FC equal to 200 μg/g before bowel preparation and endoscopy as part of the diagnostic workup for IBD.

Endoscopic investigations

Mucosal intestinal specimens from 104 IBD patients were taken depending on the localization of the disease, during ileocolonoscopy performed at diagnosis. From each patient, mucosal samples were collected from two sites, area with macroscopically evident endoscopic inflammation as well as macroscopically unaffected region. In our study, term inflamed mucosa was further used as active disease endoscopically and histologically defined by presence of neutrophilic inflammation, cryptitis, or crypt abscesses, while noninflamed mucosa presented biopsy samples free from endoscopically and histologically active or chronic inflammation. Biopsies collected for mRNA analysis were immediately submerged in All Protect Tissue Reagent (Qiagen, Germany), transported to the Laboratory for Molecular Biomedicine Institute of Molecular Genetics and Genetic Engineering, and stored at −20°C until processing.

Endoscopic activity was determined by Simple Endoscopic Score for Crohn's disease (SES-CD) and Mayo scoring system for UC based on the assessment of granulation scattering, vascular pattern, vulnerability, and mucosal damage of the bowel.

Gene expression analysis

Data regarding IL17A, IL17F, IL23A, and TLR9 mRNA levels formerly examined by our group in IBD patients' noninflamed and inflamed intestinal biopsy samples were included in assessment. The total mRNA material was isolated from mucosal biopsies and analyzed by qRT-PCR method using GAPDH mRNA level as a reference, as previously described.²⁵ For each patient, target mRNA level in inflamed mucosa was normalized with target mRNA level in noninflamed mucosa and calculated as relative quantity (RQ) using 2^−ddCt method as follows: dCt (target inflamed) = Ct (target inflamed) – Ct (reference inflamed), dCt (target noninflamed) = Ct (target noninflamed) – Ct (reference noninflamed), ddCt = dCt (target inflamed) – dCt (target noninflamed), RQ (target) = 2^−ddCt.

FTO genotyping

DNA material from 94 CD, 98 UC, and 91 controls were obtained from peripheral blood or buccal swab samples using QIAamp Blood DNA Mini Kit (Qiagen, Germany). IBD patients and controls were analyzed for FTO rs9939609 variant using polymerase chain reaction–amplification refractory mutation system method, as previously described.³²

Statistical analysis

Differences in the level of continuous variables were tested using Mann–Whitney and Kruskal–Wallis U tests, while Pearson's chi-square test was used for analysis of distribution differences of discrete variables. Logistic regression model was applied to assess associations between tested parameters and clinical outcomes controlled for confounders' effect. Correlation between continuous variables was described by Spearman's rank correlation coefficient. All performed tests were two sided, and threshold for statistical significance was defined using appropriate multiple comparisons correction methods, such as Bonferroni. If the number of comparisons was high, Benjamin-Hochberg false discovery rate (FDR) was applied as less stringent method to decrease the number of false-negative results. Statistical analyses were performed using SPSS 21.0 software (IBM, Armonk, NY).

Results

Study group data

The study analyzed 104 newly diagnosed IBD patients and 45 non-IBD controls. Study group data are presented in Table 2.

Table 2.

Descriptive Characteristics and Metabolic Data of the Study Participants

Subjects' parameters				CD vs. non-IBD	UC vs. non-IBD	CD vs. UC
	CD (n = 50)	UC (n = 54)	non-IBD (n = 45)	CD vs. non-IBD	UC vs. non-IBD	CD vs. UC
	CD (n = 50)	UC (n = 54)	non-IBD (n = 45)	P
Age, years	35 (19–72)	43.5 (20–78)	42 (18–78)	0.06	0.5	0.008
Females, n (%)	17 (34.0)	30 (55.6)	22 (48.9)	0.1	0.5	0.027
Smoking, n (%)	19 (38.0)	21 (38.9)	18 (40.0)	0.5	0.9	0.9
BMI male, kg/m²	24.6 [22.1–26.5]	24.2 [22.3–26.7]	23.6 [22.1–24.2]	0.1	0.1	0.9
BMI female, kg/m²	24.6 [21.9–26.9]	22.6 [20.3–24.4]	23.3 [22.2–24.8]	0.2	0.2	0.02
WC males, cm	94.0 [80.0–95.5]	87.0 [80.0–95.0]	87.0 [81.0–92.0]	0.1	0.6	0.5
WC females, cm	89.0 [82.0–89.5]	80.5 [66.7–84.0]	72.0 [67.7–79.2]	3 × 10^–6	0.1	0.001
TCh, mM	4.4 [3.6–5.1]	4.2 [3.5–5.1]	4.8 [4.3–5.6]	0.006	0.001	0.3
HDL-C, mM	0.9 [0.5–1.4]	0.9 [0.6–1.4]	1.8 [1.2–2.1]	10⁻⁵	2 × 10⁻⁵	0.6
LDL-C, mM	2.1 [1.4–2.4]	2.4 [1.8–2.8]	2.6 [2.1–3.2]	0.002	0.07	0.06
TG, mM	1.2 [0.8–1.8]	1.1 [0.7–1.8]	1.1 [0.7–1.5]	0.2	0.8	0.4
CRP, mg/L	12.0 [8.0–22.0]	12.0 [8.0–18.7]	2.8 [1.1–4.4]	10⁻¹³	2 × 10⁻¹³	0.7
FC, mg/kg	1100.0 [633.7–1500.0]	1100.0 [785.7–1515.0]	NA	NA	NA	0.9
CDAI/Mayo	180.0 [153.5–231.0]	6.0 [4.0–8.0]	NA	NA	NA	NA
Endoscopic score	7.0 [5.0–7.0]	2.0 [1.0–2.0]	NA	NA	NA	NA

Data are presented as median values with the corresponding interquartile ranges, except for age where median value was presented with min-max range. Associations were tested using chi-square test for discrete and Mann–Whitney U-test for continuous variables. Bonferroni correction for multiple testing introduced new significance threshold level P < 0.016. Bolded values indicate significant differences after Bonferroni correction.

BMI, body mass index; WC, waist circumference; TCh, total cholesterol; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; TG, triglycerides; CRP, C-reactive protein; FC, fecal calprotectin; CDAI, Crohn's disease activity index was considered significant after Bonferroni correction; IBD, inflammatory bowel disease; NA, not applicable.

There was no significant difference in BMI regarding gender between studied groups. However, CD females had significantly higher WC than non-IBD and UC females (P = 3 × 10⁻⁶ and P = 0.001, respectively).

Compared with non-IBD subjects, both CD and UC patients had significantly lower levels of TCh (P = 0.006, P = 0.001, respectively), HDL-C (P = 10⁻⁵, P = 2 × 10⁻⁵, respectively), while LDL-C was significantly lower only in CD patients (P = 0.002). TG levels were similar among CD, UC, and non-IBD groups.

MeS and IBD

MeS was more frequent in CD (36%) and UC (29.6%) than in non-IBD control group (13.3%); however, statistically significant association was reached only between MeS and CD (P = 0.01). We found that central obesity was significantly more frequent in both CD and UC compared with control group, 66%, 51.9%, and 20% (P = 10⁻⁵, P = 0.002), respectively. Also, there were more participants with HDL-C level below the cutoff value for MeS diagnosis in CD and UC than in control group, 70%, 66.7%, and 22.2% (P = 5 × 10⁻⁶, P = 6 × 10⁻⁶), respectively. There were no patients with DM2, and frequencies of patients with hypertension were similar in all studied groups (Table 3). All tested associations were adjusted for age, gender, and smoking in logistic regression model.

Table 3.

Association of Metabolic Syndrome and Its Components with Inflammatory Bowel Disease Occurrence

				CD vs. non-IBD	UC vs. non-IBD	CD vs. UC
	CD (n = 50)	UC (n = 54)	Non-IBD (n = 45)	CD vs. non-IBD	UC vs. non-IBD	CD vs. UC
	CD (n = 50)	UC (n = 54)	Non-IBD (n = 45)	P
MeS	18 (36.0%)	16 (29.6%)	6 (13.3%)	0.01	0.05	0.4
Central obesity, WC (M: ≥94, F: ≥80 cm)	33 (66.0%)	28 (51.9%)	9 (20.0%)	10⁻⁵	0.002	0.03
High TG (>1.7 mM)	14 (28.0%)	16 (29.6%)	7 (15.6%)	0.2	0.1	0.9
Low HDL-C (M < 1.03 mM, F < 1.29 mM)	35 (70.0%)	36 (66.7%)	10 (22.2%)	5 × 10⁻⁶	6 × 10⁻⁶	0.8
Diabetes mellitus (FPG ≥5.6 mM)	0 (0.0%)	0 (0.0%)	4 (8.9%)	1	1	NA
Hypertension (sBP ≥130; dBP ≥85 mm Hg)	14 (28.0%)	17 (31.5%)	14 (31.1%)	0.6	0.8	1

Associations were tested using logistic regression model adjusted for age, gender, and smoking. Bonferroni correction introduced new significance threshold level P < 0.016. Bolded values indicate significant differences after Bonferroni correction.

MeS, metabolic syndrome; FPG, fasting plasma glucose; sBP, systolic blood pressure; dBP, diastolic blood pressure; NA, not applicable.

Correlations of metric and metabolic parameters with IBD inflammation markers

We investigated correlation of TCh, HDL-C, LDL-C, TG, BMI, and WC with different inflammation level indicators, such as disease activity scores (CDAI, MS), endoscopic activity scores (SES-CD, MS), CRP, FC, and mRNA levels of IL17A, IL17F, IL23A, and TLR9 genes in intestinal biopsies of IBD patients.

In CD patients, we observed negative correlation between TCh level and CDAI score (rs = −0.39, P = 0.005) and positive correlation of LDL-C with CRP level (rs = 0.29, P = 0.04). In UC patients, HDL-C level was negatively correlated with CRP, FC, and MS (rs = −0.29 P = 0.03, rs = −0.27 P = 0.04 and rs = −0.31 P = 0.018, respectively).

In inflamed CD colonic mucosa, positive correlation was found between TCh and IL23A mRNA levels (rs = 0.42, P = 0.04), while negative correlations were observed between WC and IL17A (rs = −0.57, P = 0.003) and IL17F mRNA levels (rs = −0.64, P = 0.001). In inflamed UC mucosa, negative correlations were found between LDL-C and IL23A mRNA levels (rs = −0.29, P = 0.03) as well as TG and IL17F mRNA levels (rs = −0.27, P = 0.04).

After correction for multiple testing, only negative correlation between TCh level and CDAI score, as well as negative correlations between WC and inflamed colon IL17A and IL17F mRNA levels, remained significant (Table 4).

Table 4.

Correlation Analysis of Metabolic Syndrome Parameters with Inflammatory Bowel Disease Inflammation Markers' Levels

Metabolic parameter	CD		UC
Metabolic parameter	Inflammation marker	Correlation	Inflammation marker	Correlation
TCh	CDAI	rs = −0.389, P = 0.005	Inflamed colon TLR9 mRNA	rs = 0.234, P = 0.09
	Inflamed colon IL23A mRNA	rs = 0.420, P = 0.04	Inflamed colon TLR9 mRNA	rs = 0.234, P = 0.09
	Inflamed colon IL17A mRNA	rs = 0.384, P = 0.06
	SES-CD	rs = −0.280, P = 0.07
HDL-C			CRP	rs = −0.292, P = 0.03
			FC	rs = −0.271, P = 0.04
			Mayo endoscopic score	rs = −0.310, P = 0.02
			Mayo score	rs = −0.237, P = 0.08
			inflamed colon IL23A mRNA	rs = −0.228, P = 0.09
LDL-C	CRP	rs = 0.287, P = 0.04	inflamed colon IL23A mRNA	rs = −0.290, P = 0.03
	Inflamed ileum IL23A mRNA	rs = 0.362, P = 0.07	inflamed colon IL17A mRNA	rs = −0.228, P = 0.09
			inflamed colon TLR9 mRNA	rs = −0.230, P = 0.09
TG	Inflamed colon IL17A mRNA	rs = −0.343, P = 0.09	inflamed colon IL17F mRNA	rs = −0.274, P = 0.04
TG			inflamed colon IL17A mRNA	rs = −0.243, P = 0.07
BMI			FC	rs = −0.269, P = 0.05
WC	Inflamed colon IL17A mRNA	rs = −0.569, P = 0.003	inflamed colon IL17A mRNA	rs = 0.264, P = 0.05
	Inflamed colon IL17F mRNA	rs = −0.637, P = 0.001
	Inflamed ileum IL23A mRNA	rs = −0.395, P = 0.05

Correlations between IL17A, IL17F, IL23A, and TLR9 RQ values and markers of inflammation were assessed using Spearman's test. The table shows correlations that demonstrate statistical trend (P < 0.1). Bolded correlations were significant after Benjamini–Hochberg correction for multiple testing (FDR of 0.15). Total number of CD patients was n = 50 (ileal mucosal samples n = 25, colonic mucosal samples n = 25), while total number of UC patients was n = 54 (colonic mucosal samples n = 54). Number of operated CD patients was 9 and for those patients, SES-CD score was not available.

FTO rs9939609 and IBD

We analyzed distribution of genetic variant FTO rs9939609, previously associated with obesity. This part of the study included the following: 94 CD patients, median age 35 (range 19–72), 36 females (38.3%); 98 UC patients, median age 40 (range 20–78), 48 females (49%); 38 non-IBD controls from the Clinic, median age 50 (range 21–74), 22 females (57.9%); and 53 healthy controls from general population, median age 30 (range 17–76), 27 females (50.9%). The genotype distributions were in Hardy–Weinberg equilibrium in each and total analyzed group. We did not observe significant differences in genotype frequencies between non-IBD group selected at the Clinic and healthy subjects from general population, and therefore, they were considered as total control group for further comparisons with CD and UC groups.

Results showed that carriers of FTO AA genotype were more frequent in CD than in UC and control group, 29.8%, 23.5%, and 14.3%, respectively. It has been demonstrated that the presence of AA genotype was significant predictor of CD occurrence (P = 0.01), adjusted for age and gender in the logistic regression model. BMI as an additional covariate in the model did not change the result (P = 0.02), indicating that FTO has been a significant predictor of CD independently from the BMI status. Compared with TT and TA carriers, carriers of AA genotype had 2.5 higher odds for CD development (OR = 2.5 95% CI [1.2–5.3]) (Table 5).

Table 5.

Association Analysis of FTO Genetic Variant with Crohn's Disease and Ulcerative Colitis Using Logistic Regression Model

FTO rs9939609	CD (n = 94)	UC (n = 98)	CTRL (n = 91)	CD vs. CTRL -adjusted for age, gender adjusted for BMI, age, gender	UC vs. CTRL adjusted for age, gender adjusted for BMI, age, gender
TT	25 (26.6%)	30 (30.6%)	26 (28.6%)	P = 0.01 OR 2.6 95% CI [1.2–5.4]	P = 0.06OR 2.195% CI [1–4.6]
TA	41 (43.6%)	45 (45.9%)	52 (57.1%)
AA	28 (29.8%)	23 (23.5%)	13 (14.3%)
T (%)	48	54	57	P = 0.02 OR 2.5 95% CI [1.2–5.3]	P = 0.1OR 1.995% CI [0.8–4.3]
A (%)	52	46	43	P = 0.02 OR 2.5 95% CI [1.2–5.3]	P = 0.1OR 1.995% CI [0.8–4.3]

In logistic regression model, carriers of FTO AA genotype were compared with carriers of “referent” TT+TA genotypes (recessive genetic model for the A allele). Logistic regression model was used to adjust for confounding variables (BMI at disease onset, age, gender). P < 0.025 was considered significant after Bonferroni correction and bolded.

Discussion

Our study examined MeS in association with IBD occurrence and the level of inflammation and disease activity as well as Th-17 immune genes' mRNAs in patients' intestinal mucosa. Also, we analyzed association between obesity-related FTO genetic variant and IBD development.

We observed higher frequency of MeS among IBD patients, but the association was demonstrated as statistically significant only with CD. Also, we showed that central obesity and low HDL-C were associated with IBD. MeS prevalence is not thoroughly investigated in IBD, and such results could contribute to better understanding of underlying common mechanisms.^8,9 One large study showed that presence of MeS, low serum HDL-C, hypertriglyceridemia, and DM2 increased the rate of hospitalizations of CD patients.³³ It is suggested that inflammation arisen in VAT may play important role in chronic systemic inflammation of both MeS and IBD.^1,34,35 The VAT setting consists of hypertrophic adipocytes that secrete abnormal levels of adipokines and proinflammatory cytokines leading to infiltration of M1 macrophages and driving low-grade chronic inflammation. Besides inflammation in VAT, gut dysbiosis is probably additional factor that could contribute to altered immune-metabolic state, as it was seen in IBD and other MeS-associated diseases such as obesity, insulin resistance, and DM2.^1,3

–7

Results of our group demonstrated that IBD patients in general had significantly decreased TCh and HDL-C compared with controls, while LDL-C was significantly lower specifically among CD patients. Hypocholesteremia was previously determined as the IBD common feature, but changes in the levels of specific lipoproteins were not fully consistent.²² Our results were in concordance with decreased levels of TCh, HDL-C, and LDL-C showed by other groups.^36

–40 However, we did not show significantly elevated serum TG levels, observed in numerous inflammatory diseases, including IBD.^37
–39 The majority IBD patients in our cohorts have high CRP levels (>10 mg/L) consistent with acute phase response and systemic inflammation. When systemic inflammation is present, there is a change in lipid profile characterized by lower levels of HDL-C and LDL-C.⁴¹

According to our findings, TCh levels showed negative correlation with CDAI, which is probably due to nutrient malabsorption related to mucosal damage.⁴² Reduced cholesterol transport to liver (reverse cholesterol transport) and alteration in cholesterol metabolism associated with inflammation may also account for reduced TCh level.^12,43 A study that enrolled >350 CD patients associated low cholesterol level with more severe disease, which is in line with our results.³⁷ On the contrary, our results suggested that TCh level was positively correlated with IL17A and IL23A mRNA levels in inflamed colon of CD patients, as well as with local TLR9 mRNA of UC patients (although not statistically significant after FDR correction). In our previous study, we found that higher mRNA levels of IL17A, IL17F, IL23A, and TLR9 were associated with gut inflammation.²⁵ It is known that cholesterol accumulation in immune cells promotes inflammatory responses, including augmentation of Toll-like receptor (TLR) signaling, which might explain positive correlation of TCh with local inflammation markers.⁴³

We showed negative correlation of HDL-C level and inflammation severity parameters in IBD patients, although none of these associations remained statistically significant after FDR correction. In a large cohort of 868 CD patients, low HDL-C level was also associated with more severe disease.³³ HDL counteracts inflammation by mediating reverse transport of cholesterol from macrophages to the liver, and its low level can be regarded as a marker of inflammation. Gerster et al. in vitro and in vivo presented HDL as an inhibitor of intestinal inflammation through autophagy and potential therapeutic target in IBD.²⁴

Even though our study showed increased frequency of central obesity in IBD compared with control group, among IBD patients with higher WC we detected lower level of inflammation markers. In particular, our results showed strong negative correlation between WC and the expression level of IL17A and IL17F in inflamed colon of CD patients. We also noted negative correlation between WC as well as TG and LDL-C levels, on one hand, and the markers of local inflammation, mRNA levels of IL23A, IL17F, IL17A, and TLR9 in inflamed colon of UC patients, on the other hand (although not significant after FDR correction). Interestingly, although central obesity contributes to IBD pathogenesis,¹⁰ increased intestinal inflammation, in turn, might cause reduction in central obesity of IBD patients as noted in this and other studies,⁴⁴ probably attributed to nutrition malabsorption. Therefore, although our results showed that patients on diagnosis were more obese than controls, they could still have different levels of malnutrition among themselves, depending on the level of intestinal damage, which is related to the level of local inflammation.

The demonstrated linkage of IBD, MeS, and central obesity has led to analysis of potentially new IBD candidate gene, FTO. To the best of our knowledge, no study has yet investigated association of FTO rs9939609 variant with IBD occurrence. The minor allele frequency (A allele) observed in this study (43%) was similar to other European countries (∼41% on average), according to 1000 Human Genome Project.⁴⁵ Great interest has been recently devoted to the association of FTO gene with obesity.^27,46,47 FTO is a member of a superfamily of Fe (II)-and 2-oxoglutarate-dependent dioxygenases, and presents with a nucleic acid demethylase activity. The proposed FTO-related pathophysiological mechanism includes alterations of methylation–demethylation states of gene expression in metabolically active tissues.⁴⁷ According to the conducted studies, A-allele of FTO variant rs9939609 was determined as a risk factor for polygenetic obesity in the Caucasians, indicating that carriers have significantly higher BMI and adiposity.^28,46 One large meta-study demonstrated the association of FTO rs9939609 variant with higher odds for developing MeS, attributing the association to FTO general effect on increased adiposity and further on adverse metabolic impact.²⁸

Our novel results showed that FTO AA genotype was associated with higher probability for CD occurrence, independently from the BMI status. That is, for a given BMI value, carriers of FTO AA genotype will have 2.5 higher odds for CD development. According to this observation, we assume that effect of FTO on disease development is probably mediated by factors of increased adiposity other than BMI. This is in line with our study results, showing that central and not general obesity is a risk factor for IBD development. Further, recent prospective IBD study demonstrated that weighted genetic risk score comprised of 71 genetic loci, including FTO, which were previously associated with higher HDL-C in GWAS, was lower among IBD patients.^36,48 However, genetic risk was presented as a summary score without the data regarding single genetic variant associations. Nevertheless, these results highlighted genetic impact on the altered lipid metabolism, which could eventually lead to pathological outcomes.

To summarize, our study showed association of MeS with CD. In general, lipid levels and obesity metrics were negatively correlated with markers of local and systemic inflammation in IBD. In addition, we identified FTO rs9939609 genetic variant as potential risk factor for CD occurrence. As the observational data obtained from patient diagnosis could be misleading due to malnutrition induced by already active disease, it is important to conduct longitudinal prospective studies with record of MeS parameters in multiple time points. Finally, our results justify future use of nutrigenomic approach in the IBD research.

Footnotes

Author Disclosure Statement

The authors declare that they have no conflict of interest. No competing financial interests exist.

Funding Information

The work is supported by Ministry of Education, Science and Technological Development, Republic of Serbia (grant no. III 41004).

References

Michalak

, Mosińska

, Fichna

. Common links between metabolic syndrome and inflammatory bowel disease: Current overview and future perspectives. Pharmacol Rep, 2016; 68:837–846.

Basso

, Fonseca

, Bonfá

, et al. Association among genetic predisposition, gut microbiota, and host immune response in the etiopathogenesis of inflammatory bowel disease. Braz J Med Biol Res, 2014; 47:727–737.

Patterson

, Ryan

, Cryan

, et al. Gut microbiota, obesity and diabetes. Postgrad Med J, 2016; 92:286–300.

Saad

, Santos

, Prada

. Linking Gut Microbiota and Inflammation to Obesity and Insulin Resistance. Physiology (Bethesda), 2016; 31:283–293.

Baothman

, Zamzami

, Taher

, et al. The role of Gut Microbiota in the development of obesity and Diabetes. Lipids Health Dis, 2016; 15:108.

Ussar

, Griffin

, Bezy

, et al. Interactions between Gut Microbiota, Host Genetics and Diet Modulate the Predisposition to Obesity and Metabolic Syndrome. Cell Metab, 2015; 22:516–530.

Komaroff

. The Microbiome and risk for obesity and diabetes. JAMA, 2017; 317:355–356.

Nagahori

, Hyun

, Totsuka

, et al. Prevalence of metabolic syndrome is comparable between inflammatory bowel disease patients and the general population. J Gastroenterol, 2010; 45:1008–1013.

Yorulmaz

, Adali

, Yorulmaz

, et al. Metabolic syndrome frequency in inflammatory bowel diseases. Saudi J Gastroenterol, 2011; 17:376–382.

10.

Singh

, Dulai

, Zarrinpar

, et al. Obesity in IBD: Epidemiology, pathogenesis, disease course and treatment outcomes. Nat Rev Gastroenterol Hepatol, 2017; 14:110–121.

11.

Flores

, Burstein

, Cipher

, et al. Obesity in inflammatory bowel disease: A marker of less severe disease. Dig Dis Sci, 2015; 60:2436–2445.

12.

Pringle

, Stewart

, Peloquin

, et al. Body mass index, genetic susceptibility, and risk of complications among individuals with Crohn's disease. Inflamm Bowel Dis, 2015; 21:2304–2310.

13.

Seminerio

, Koutroubakis

, Ramos-Rivers

, et al. Impact of obesity on the management and clinical course of patients with inflammatory bowel disease. Inflamm Bowel Dis, 2015; 21:2857–2863.

14.

Stabroth-Akil

, Leifeld

, Pfützer

, et al. The effect of body weight on the severity and clinical course of ulcerative colitis. Int J Colorectal Dis, 2015; 30:237–242.

15.

Kredel

, Siegmund

. Adipose-tissue and intestinal inflammation—visceral obesity and creeping fat. Front Immunol, 2014; 5:462.

16.

Zietek

, Rath

. Inflammation meets metabolic disease: Gut feeling mediated by GLP-1. Front Immunol, 2016; 7:154.

17.

Kredel

, Batra

, Siegmund

. Role of fat and adipokines in intestinal inflammation. Curr Opin Gastroenterol, 2014; 30:559–565.

18.

Ouchi

, Parker

, Lugus

, et al. Adipokines in inflammation and metabolic disease. Nat Rev Immunol, 2011; 11:85–97.

19.

De Carvalho

, Bonfa

, Borba

. Systemic lupus erythematosus and “lupus dyslipoproteinemia”. Autoimmunity Rev, 2008; 7:246–250.

20.

Hardardottir

, Doerrler

, Feingold

, Grunfeld

. Cytokines stimulate lipolysis and decrease lipoprotein lipase activity in cultured fat cells by a prostaglandin independent mechanism. Biochem Biophys Res Commun, 1992; 186:237–243.

21.

Gouni

, Oka

, Etienne

, Chan

. Endotoxin-induced hypertriglyceridemia is mediated by suppression of lipoprotein lipase at a post-transcriptional level. J Lipid Res, 1993; 34:139–146.

22.

Agouridis

, Elisaf

, Milionis

. An overview of lipid abnormalities in patients with inflammatory bowel disease. Ann Gastroenterol, 2011; 24:181–187.

23.

Catapano

, Pirillo

, Bonacina

, et al. HDL in innate and adaptive immunity. Cardiovasc Res, 2014; 103:372–383.

24.

Gerster

, Eloranta

, Hausmann

, et al. Anti-inflammatory function of high-density lipoproteins via autophagy of IκB kinase. Cell Mol Gastroenterol Hepatol, 2014; 1:171–187.

25.

Dragasevic

, Stankovic

, Sokic-Milutinovic

, et al. Importance of TLR9-IL23-IL17 axis in inflammatory bowel disease development: Gene expression profiling study. Clin Immunol, 2018; 197:86–95.

26.

Frayling

, Timpson

, Weedon

, et al. A common variant in the FTO gene is associated with body mass index and predisposes to childhood and adult obesity. Science, 2007; 316:889–894.

27.

Peng

, Zhu

, Xu

, et al. FTO gene polymorphisms and obesity risk: A meta-analysis. BMC Med, 2011; 9:71.

28.

Freathy

, Timpson

, Lawlor

, et al. Common variation in the FTO gene alters diabetes-related metabolic traits to the extent expected given its effect on BMI. Diabetes, 2008; 57:1419–1426.

29.

Hodson

. Inflammatory bowel disease. Nature, 2016; 540:S97.

30.

Mowat

, Cole

, Windsor

, et al. IBD Section of the British Society of Gastroenterology. Guidelines for the management of inflammatory bowel disease in adults. Gut, 2011; 60:571–607.

31.

Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA, 2001; 285:2486–2497.

32.

Müller

, Hinney

, Scherag

, et al. Fat mass and obesity associated' gene (FTO): No significant association of variant rs9939609 with weight loss in a lifestyle intervention and lipid metabolism markers in German obese children and adolescents. BMC Med Genet, 2008; 9:85.

33.

Fitzmorris

, Colantonio

, Torrazza Perez

, et al. Impact of metabolic syndrome on the hospitalization rate of Crohn's disease. patients seen at a tertiary care center: A retrospective cohort study. Digestion, 2015; 91:257–262.

34.

Peyrin-Biroulet

, Chamaillard

, Gonzalez

, et al. Mesenteric fat in Crohn's disease: A pathogenetic hallmark or an innocent bystander?. Gut, 2007; 56:577–583.

35.

Olszanecka-Glinianowicz

, Handzlik-Orlik

, Orlik

, et al. Adipokines in the pathogenesis of idiopathic inflammatory bowel disease. Endokrynol Pol, 2013; 64:226–231.

36.

Voutilainen

, Hutri-Kähönen

, Tossavainen

, et al. Low childhood high density lipoprotein cholesterol levels and subsequent risk for chronic inflammatory bowel disease. Dig Liver Dis, 2018; 50:348–352.

37.

Koutroumpakis

, Ramos-Rivers

, Regueiro

, et al. Association between long-term lipid profiles and disease severity in a large cohort of patients with inflammatory bowel disease. Dig Dis Sci, 2016; 61:865–871.

38.

Levy

, Rizwan

, Thibault

, et al. Altered lipid profile, lipoprotein composition and oxidant and antioxidant status in pediatric Crohn's disease. Am J Clin Nutr, 2000; 71:807–815.

39.

Ripollés Piquer

, Nazih

, Segain

, et al. Altered lipid, apolipoprotein, and lipoprotein profiles in inflammatory bowel disease: Consequences on the cholesterol efflux capacity of serum using Fu5AH cell system. Metabolism, 2006; 55:980–988.

40.

Hrabovský

, Zadák

, Bláha

, et al. Cholesterol metabolism in active Crohn's disease. Wien Klin Wochenschr, 2009; 121:270–275.

41.

Herbert

, Erridge

. Regulation of low-density lipoprotein cholesterol by intestinal inflammation and the acute phase response. Cardiovasc Res, 2018; 114:226–232.

42.

Uchiyama

, Kishi

, Komatsu

, et al. Lipid and bile acid dysmetabolism in Crohn's disease. J Immunol Res, 2018; 2018:7270486.

43.

Tall

, Yvan-Charvet

. Cholesterol, inflammation and innate immunity. Nat Rev Immunol, 2015; 15:104–116.

44.

Karmiris

, Koutroubakis

, Kouroumalis

. Leptin, adiponectin, resistin, and ghrelin—implications for inflammatory bowel disease. Mol Nutr Food Res, 2008; 52:855–866.

45.

1000 Genomes Project Consortium, Auton

, Brooks

, et al. A global reference for human genetic variation. Nature, 2015; 526:68–74.

46.

Castellini

, Franzago

, Bagnoli

, et al. Fat mass and obesity-associated gene (FTO) is associated to eating disorders susceptibility and moderates the expression of psychopathological traits. PLoS One, 2017; 12:e0173560.

47.

Mizuno

. Fat Mass and Obesity Associated (FTO) Gene and Hepatic Glucose and Lipid Metabolism. Nutrients, 2018; 10. pii:E1600.

48.

Willer

, Schmidt

, Sengupta

, et al. Discovery and refinement of loci associated with lipid levels. Nat Genet, 2013; 45:1274–1283.