The fatty acid pattern of membrane phospholipids is suggested to affect membrane fluidity and epithelial barrier function as a result of membrane fatty acid unsaturation. The incorporation of n-3 polyunsaturated fatty acids (PUFAs) into membrane phospholipids may diminish inflammatory potential in patients with gastrointestinal diseases. The aim of this study was to improve the fatty acid profile of erythrocyte membrane phospholipids after oral supplementation of specific fatty acids in patients with maldigestion and/or malabsorption.
We conducted a randomized, double-blind, controlled trial. A total of 48 patients with gastrointestinal diseases received either fat-soluble vitamins A,D,E,K (ADEK) or ADEK plus fatty acids α-linolenic acid (ALA), docosahexaenoic acid (DHA) and medium-chain triglycerides (FA-ADEK) for 12 weeks. The fatty acid profile of erythrocyte membrane phospholipids, dietary intake, plasma antioxidant vitamins and serum γ-glutamyl transferase (GGT) were evaluated at baseline, 8 and 12 weeks after supplementation.
Supplementation with FA-ADEK increased ALA, DHA and eicosapentaenoic acid (EPA) concentrations of erythrocyte membrane phospholipids by 0.040, 1.419 and 0.159%, respectively, compared with ADEK supplementation (−0.007, 0.151 and 0.002%, respectively) after 12 weeks (all P0.001). Serum GGT activity decreased in patients receiving FA-ADEK compared with those receiving ADEK with a significant difference after 8 weeks.
The significant change in erythrocyte membrane fatty acid pattern demonstrates the incorporation of orally administered n-3 PUFA in patients with maldigestion and malabsorption. The increase in ALA and DHA, as well as the conversion of ALA to EPA is attributed to the supplementation of sufficient amounts of ALA and DHA, respectively. Serum GGT activity decreased in response to decreased oxidative stress.
Gut homeostasis is achieved, in particular, by the regulation of gastric acid, mucous and epithelial barrier function. Intestinal epithelial barrier dysfunction has been recognized as a link between gastrointestinal diseases associated with maldigestion and malabsorption. An altered barrier function is accompanied by oxidative stress, inflammation and disturbances in intestinal permeability, followed by mucosal damage (Ma, 1997; DeMeo et al., 2002; Kruidenier et al., 2003). Within the scope of maldigestion, these disorders may result in a loss of exocrine pancreatic function, which is verified by a reduction in fecal elastase-1 (Carroccio et al., 2001; Mann et al., 2003). These pathomechanisms occur concomitantly with diseases, such as cystic fibrosis (CF), chronic pancreatitis and liver diseases. Inflammatory bowel diseases, celiac disease, small-bowel resections, biliary salt deficiency, bacterial overgrowth, digestive fistulas and side effects of medications contribute to decreased absorption, particularly of fat-soluble vitamins and may lead to the malabsorption of long-chain polyunsaturated fatty acids (PUFAs).
The integrity of intestinal barrier function is substantially affected by the fatty acid profile of membrane phospholipids (Whiting et al., 2005; Zhao et al., 2008). The proportion of saturated to unsaturated fatty acids and the ratio between n-6 and n-3 PUFA exert a substantial effect on membrane fluidity and epithelial barrier function. A favorable ratio between pro-inflammatory arachidonic acid (AA) and anti-inflammatory eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) may reduce mucosal damage, diminish the disturbance and alter barrier function (Mills et al., 2005; Calder, 2008). It is assumed that n-3 PUFA intake ameliorates the quality of membrane fatty acid profile and diminishes inflammatory potential when antioxidative capacity is sufficient (D’Odorico et al., 2001; Koutroubakis et al., 2004; Trebble et al., 2004; Guebre-Egziabher et al., 2008). However, there is concern that PUFA could increase lipid peroxidation, resulting in an increased susceptibility to oxidative stress (Harats et al., 1991). Lipid peroxidation has been shown to be indicated by elevated concentrations of F2-isoprostanes, which are prostaglandin-like products of nonenzymatic peroxidation of AA (Morrow et al., 1999). A trial conducted in postmenopausal women showed no significant increase in plasma free F2-isoprostanes, even when normalized to plasma AA concentrations, after supplementation with fish-oil long-chain PUFA (Higdon et al., 2000).
It has been suggested that a change in membrane fatty acid patterns due to oral supplementation of a specific fatty acid composition decreases AA concentrations of tissue and cell membrane phospholipids as a result of a reduced ratio of n-6 to n-3 PUFA (Hillier et al., 1991; Cao et al., 2006). The concomitant medium-chain triglyceride component is supposed to prevent a diminished incorporation of long-chain PUFA into the phospholipids of the erythrocyte membrane (Periago et al., 1990).
Although clinical evidence suggests that the increased intake of n-3 PUFA enhances anti-inflammatory capacity and diminishes disease activity in patients with maldigestion and malabsorption (Belluzzi et al., 1996; Romano et al., 2005), the responsible mechanism of action remains unclear. The aim of this study was to assess the hypothesis that oral supplementation of specific fatty acids (such as medium-chain triglycerides and selected n-3 PUFAs) increases their incorporation into erythrocyte membrane phospholipids in patients with maldigestion and/or malabsorption.
Subjects and methods
The prospective, randomized, controlled, double-blind trial was conducted in patients admitted to the University Hospital of Bonn. A total of 99 subjects were screened, 53 were randomized to receive either fat-soluble vitamins A,D,E,K (ADEK) or ADEK plus fatty acids α-linolenic acid (ALA), DHA and medium-chain triglycerides (FA-ADEK) and 48 patients completed the trial. The trial profile is shown in Figure 1. The participants consisted of 17 men (aged 21–76 years) and 31 women (aged 20–76 years) with diagnosed maldigestion and/or malabsorption. The characteristics of the patients are shown in Table 1. Inclusion criteria were fecal elastase-1 200 μg/g and/or clinical findings. Exclusion criteria were renal replacement therapy, insulin-dependent diabetes mellitus with metabolic complications, malignancies and pregnancy. All fatty acid and fat-soluble vitamin supplements other than trial supplements were excluded for the duration of the study, that is, at least 4 weeks before run-in until the end of the study. The study was approved by the ethics committee of the Medical Faculty of the University of Bonn. Informed consent was obtained from each participant before trial onset.
All subjects underwent a run-in period of 2 weeks during which they received fat-soluble vitamins (ADEK) at a daily dose of 1.5 mg per day vitamin A (RE), 9.6 μg per day vitamin D3, 198 mg per day vitamin E (TE) and 0.09 mg per day vitamin K (Table 2). The run-in phase was necessary to avoid any deficiency of fat-soluble vitamins during the study period and to standardize the dietary intake of fat-soluble vitamins A, D, E and K of the patients. After these 2 weeks, baseline blood samples were collected and dietary record and clinical data were assessed. Subjects were then randomly assigned either to continue intake of ADEK or to commence intake of ADEK plus fatty acids (900 mg per day α-linolenic acid (ALA), 375 mg per day DHA, 225 mg per day medium-chain triglycerides) (FA-ADEK) for 12 weeks. The daily supplementation in the ADEK and FA-ADEK groups is shown in Table 2. Supplements were administered three times a day with main meals with a total of six capsules. Whereas the FA-ADEK group received xylose-hardened (enteric-coated) capsules, the ADEK group received nonenteric coated capsules. Adherence to the study drug as assigned was monitored by capsule counts at 8 and 12 weeks from baseline. Supplements were obtained from Meduna Ltd, Isernhagen, Germany.
Patients were centrally randomized using a telephone-based block randomization pattern and stratified according to sex and smoking behavior. Collection of blood samples, dietary records, clinical and biochemical data was repeated 8 and 12 weeks after supplementation. The primary end point of the study was the change in DHA concentration of erythrocyte membrane phospholipids. The planned sample size of 27 patients per group was based on a power of 80% to detect a difference of at least 20% between both treatment groups at an overall 5% level of significance.
Venous blood samples were obtained by trained medical staff after a nonsmoking and overnight fasting period of at least 12 h. Weight was recorded on a digital scale and height was measured. Dietary intake data (3-day weighed food record) was analyzed using the PRODI 4.2 software (WVG, Stuttgart, Germany) with database BLS II.3. The average of 3 days was assessed. Analysis of clinical chemistry, standard biochemical blood parameters and serum γ-glutamyl transferase (GGT) activity, a sensitive marker of oxidative stress, was performed by routine methods.
Analysis of fatty acids, vitamins and isoprostanes
Venous blood samples (using EDTA tubes) were collected after an overnight fast and immediately centrifuged for separation of plasma and blood cells (2800 × g, 10 min, 4 °C). The plasma was frozen in aliquots at −80 °C until analysis. The white buffy coat was aspirated from blood cells, and red blood cells were washed twice with EDTA/saline. A few drops of butylhydroxytoluol (0.01%) in methanol were added to avoid oxidation. Red blood cells were stored under nitrogen atmosphere at −80 °C until analysis.
Determination of tocopherol and retinol
Plasma samples were deproteinized by the addition of ethanol (containing apocarotinal as internal standard, 5 μmol/l). Fat-soluble vitamins were extracted with n-hexane and analyzed using normal-phase high performance liquid chromatography (column: Nucleosil 100-5 CN, 250 × 4.0 mm, Machery and Nagel, Düren, Germany) and ultraviolet detection (292 nm) (Vuilleumier et al., 1983) (coefficient of variation (CV): retinol 4.3%, tocopherol 4.1%). Results were expressed in μmol/l of plasma.
Determination of vitamin D
25-Hydroxycholecalciferol was analyzed in plasma samples using an ELISA (enzyme-linked immunosorbent assay) kit (IDS, Mainz, Germany) (CV: 5.6%).
Determination of 8-isoprostane
The total concentration of 8-isoprostanes (8-iso-PGF2α) was determined after alkaline hydrolysis of esterified isoprostanes and purification by solid-phase extraction using a C-18 cartridge (Cayman Chemical, IBL, Hamburg, Germany) using an ELISA kit (Cayman Chemical) according to the manufacturer's instructions (CV: 15%).
Fatty acid composition in phospholipids of erythrocyte membranes
Erythrocytes were defrosted and 1 ml of it was hemolyzed by adding 1 ml of distilled water. Before extraction, 340 mg of NaCl was added. The lipid fraction was extracted using methanol/chloroform (1/2, v/v) according to a modified Folch method (Folch et al., 1957) and the lipid classes were separated using thin-layer chromatography (Christophe and Matthijs, 1967). After scraping off the phospholipid band, while looking under ultraviolet light, the phospholipid fraction was methylated using methanol/HCl (100/4, v/v) (incubation at 95 °C for 4 h). The methyl esters were extracted with petroleum ether and analyzed by temperature-programmed capillary gas chromatography (column: CP-SIL 88, 50 m × 0.25 mm, Chrompack, Middelburg, The Netherlands, flame ionization detection (FID)). Results were expressed in weight percentages (CV: <4%).
Determination of fecal elastase-1
Feces were homogenized and a sample was preserved at −20 °C. Elastase-1 concentration was determined by a commercially available ELISA kit, using two monoclonal antibodies against two distinct specific epitopes of human pancreatic elastase-1 (Schebo-Biotec, Giessen, Germany). Results were expressed as μg/g of stool (CV: 5.8%).
Statistical comparisons between groups were performed using the Mann–Whitney U-test for nonparametric unpaired data. The nonparametric Wilcoxon matched-pairs signed-rank test was used for data comparison at different time points within the study groups. All statistical tests were two-sided. Missing data were not replaced. Differences were considered significant at P<0.05. Data are reported as mean± s.d. All analyses were conducted using SPSS version 14.0 (SPSS, Chicago, IL, USA).
The characteristics of the patients are shown in Table 1. No differences were found between the FA-ADEK and ADEK groups regarding distribution of gender, age and anthropometric parameters. Underweight patients were not present in the study population. Both smokers and nonsmokers were evenly distributed among the FA-ADEK and ADEK groups.
Reduced fecal elastase-1 was used as a diagnostic criterion for maldigestion. Clinical classification of patients with exocrine pancreatic insufficiency corresponded to the number of patients with reduced fecal elastase-1.
Clinical chemistry and biochemical parameters
Clinical chemistry and biochemical parameters of the study population were within normal ranges (Table 3). The study population showed neither symptoms of protein-energy malnutrition nor signs of acute inflammation. Oral supplementation resulted in an adequate supply with fat-soluble vitamins A, D and E. No significant differences in 8-isoprostanes were found between both groups. However, FA-ADEK led to a decrease in serum GGT activity with a significant difference after 8 weeks.
An analysis of dietary records showed no clinically relevant differences between both groups regarding nutrient intake in the course of the study (data not shown).
Erythrocyte membrane fatty acids
Fatty acid concentrations of the erythrocyte membrane in the ADEK and FA-ADEK groups during the study are presented in Table 4. At baseline, EPA concentration of the erythrocyte membrane differed significantly between the FA-ADEK and ADEK groups. In the course of the study, a significant increase in EPA concentration in the FA-ADEK group was observed. Moreover, an increase in ALA and DHA, as well as a mild decrease in AA concentrations were found in the FA-ADEK group at weeks 8 and 12, compared with baseline. The sum of EPA and DHA, the sum of ALA, EPA and DHA, as well as the corresponding ratios clearly demonstrate the effect of oral supplementation with fatty acids on changes in PUFA patterns of erythrocyte membrane phospholipids in terms of an anti-inflammatory profile.
The significant mean absolute changes in the fatty acid profile of erythrocyte membrane phospholipids are confirmed by relative changes with reference to baseline (Figure 2). Oral n-3 PUFA supplementation resulted in a significant increase in ALA by 49% after 12 weeks (P<0.001). Although the supplement did not contain EPA, the EPA concentration of erythrocyte membrane phospholipids increased by 29 (P=0.001) and 37% (P=0.001) after weeks 8 and 12, respectively. In addition, DHA concentration increased significantly by 37 (P<0.001) and 48% (P<0.001) after weeks 8 and 12, respectively. The increase in DHA was found to be similar to that of EPA.
Oral fatty acid supplementation resulted in a clear shift from pro-inflammatory to anti-inflammatory precursors of eicosanoids, with a significant change in DHA and EPA concentrations (Table 4). After only 8 weeks, fatty acid supplementation led to a significant increase in omega-3-index (EPA and DHA) to above 5% in the FA-ADEK group, while an even more pronounced difference was observed after 12 weeks of supplementation. This finding is supported by the sum of ALA, EPA and DHA. The significant amelioration of the ratio of AA to DHA, as well as that of AA to EPA and DHA suggests a reduction in the pro-inflammatory potential. Evidence is supported by changes in other ratios, with the exception of EPA to DHA. Although the ratio between EPA and DHA decreased, no statistically significant difference was observed.
The goal of specific fatty acid supplementation in patients with maldigestion and malabsorption is the incorporation of n-3 PUFA into membrane phospholipids to ameliorate gastrointestinal barrier function and to minimize oxidative stress and pro-inflammation in the gastrointestinal tract (Hillier et al., 1991; Belluzzi et al., 1996; Calder, 2008). Timely supplementation avoids the shift in membrane fatty acid pattern demonstrated in healthy subjects and patients with Crohn's disease without diagnosed malabsorption (Belluzzi et al., 1996; Cao et al., 2006). Although a reduced fat intake is no longer at the center of medical nutrition in patients with maldigestion and malabsorption, intestinal fat malabsorption often persists despite pancreatic enzyme substitution. This may also be caused by defects in mucosal mechanisms (Peretti et al., 2006).
In this study, increases in DHA and ALA in erythrocyte membrane phospholipids demonstrate the incorporation of orally administered n-3 PUFA even in patients with different gastrointestinal diseases associated with maldigestion and malabsorption. These increases in DHA and ALA are attributed to the supplementation of sufficient amounts of DHA and ALA. Although EPA was not supplemented in this study, a significant increase in erythrocyte membrane EPA was observed. These results suggest that the conversion from ALA to EPA occurred within the study duration of 12 weeks. Clearly, n-3 fatty acids in combination with medium-chain triglycerides and fat-soluble vitamins improved the fatty acid status of patients with maldigestion and malabsorption in terms of anti-inflammatory potential and amelioration of membrane fluidity. Moreover, administration of enteric-coated capsules without unpleasant taste and side effects resulted in a high compliance of patients. The content of the capsules is delivered directly into the small intestine (Belluzzi et al., 1996; Romano et al., 2005). Our finding of a substantial increase in EPA in membrane phospholipids is consistent with previous reports on ALA supplementation. Hussein et al. (2005) and Barceló-Coblijn et al. (2008) showed that dietary ALA intake significantly increases the concentration of erythrocyte EPA in healthy adults.
Studies covering the pro-inflammatory and anti-inflammatory effects of PUFA in patients with maldigestion or malabsorption demonstrated that n-6 PUFA enhance inflammation processes, whereas high rates of n-3 PUFA showed reducing effects (De Vizia et al., 2003; Meister and Ghosh, 2005; Nielsen et al., 2005). AA is the substrate for the production of a wide variety of pro-inflammatory eicosanoids (Harris et al., 2009). Although the metabolites of ALA possess anti-inflammatory actions, higher n-6 PUFA intakes can inhibit the conversion of ALA to EPA (Das, 2006; Liou et al., 2007). De Vizia et al. (2003) reported an increase in EPA and DHA in erythrocyte membrane phospholipids after supplementation with fish oil as a source of EPA and DHA in patients with CF. Moreover, specific ratios seem to be associated with a reduction in disease activity in patients with Crohn's disease (Belluzzi et al., 1996).
In this study, oral specific fatty acid supplementation showed neither inter- nor intra-individual effects on plasma 8-isoprostane concentrations, one of the more abundant F2-isoprostanes in humans. Although an increase in 8-isoprostanes was expected to result from an elevated lipid peroxidation, no change in 8-isoprostane concentration was observed during the study period. These results are in accordance with findings from other trials (Higdon et al., 2000; Mori, 2004). The lack of significant differences in 8-isoprostanes may be attributed to an adequate intake of vitamin E (Saito and Kubo, 2003).
Fatty acid supplementation resulted in a decrease in serum GGT activity with a significant difference after eight weeks. Serum GGT is an enzyme that contributes to the extracellular catabolism of glutathione. The enzyme has been widely used as an index of liver dysfunction and as a marker of alcohol intake. Apart from these factors, elevated serum GGT activity is also a sensitive marker of oxidative stress (Whitfield, 2001). Conditions that increase serum GGT lead to increased free radical production and the threat of glutathione depletion. However, products of the GGT reaction may themselves lead to increased free radical production, particularly in the presence of iron (Whitfield, 2001). The findings of this study suggest that GGT activity declined after fatty acid supplementation as a response to decreased oxidative stress.
The role of GGT as a marker of oxidative stress has been demonstrated in children with CF. A study by Hull et al. (1997) showed that children with CF and pulmonary inflammation had higher GGT activity and lipid peroxide concentrations and lower glutathione concentrations in bronchiolar lavage fluid than did CF patients without inflammation or non-CF control subjects. Recently, increased serum GGT activity has been associated with development of atherosclerosis and cardiovascular disease (Emdin et al., 2005; Meisinger et al., 2006; Shankar et al., 2008) and their risk factors, including diabetes (André et al., 2006), hypertension (Sabanayagam et al., 2009), dyslipidemia and metabolic syndrome (Rantala et al., 2000; Lee et al., 2007).
Oral supplementation with fat-soluble vitamins resulted in high plasma concentrations of vitamins A, D and E, which corresponded in both groups to the preventive range (Biesalski et al., 1997; Ehrhardt, 2002). Medical nutrition therapy is regarded as basic treatment in patients with maldigestion and malabsorption to ensure a sufficient supply of fat-soluble vitamins.
Some limitations of this study must be considered. First, the study design does not include clinical outcome measures. Any improvement in the outcome of the patients, for example, relapse rate, is difficult to assess because of the heterogeneity of the underlying gastrointestinal diseases of the study population. Further randomized controlled trials are required to confirm these results and to determine response to treatment. Second, the study period was only 12 weeks. A follow-up of at least 1 year is necessary to demonstrate the long-term effects of supplementation. Finally, further studies should be conducted to assess the effect of changes in serum GGT activity resulting from supplementation of specific fatty acids on the development of atherosclerosis and cardiovascular disease in patients with maldigestion and malabsorption.
In conclusion, oral supplementation with specific fatty acids and fat-soluble vitamins resulted in an incorporation of n-3 PUFA into erythrocyte membrane phospholipids. Irrespective of the effect of oral supplementation on the outcome of primary disease, supplementation with n-3 PUFA is considered to be essential to prevent disturbances in epithelial barrier function.
Conflict of interest
The authors declare no conflict of interest.
CM coordinated the study and acted as an expert on fatty acid metabolism; RS conducted the study; RS and CM constituted the writing group and prepared this paper; BA and PS set up and performed the laboratory method of analysis of erythrocyte membrane fatty acids, plasma fat-soluble vitamins and 8-isoprostanes; BT was responsible for patient inclusion; BJ contributed to the collection of dietary and biochemical data; and NB conducted the statistical analysis. This study was supported in part by a research grant from Meduna Ltd, Isernhagen, Germany. Professor Metzner was a consultant for Meduna Ltd, Isernhagen, Germany. All authors contributed to the final version of the paper and gave their approval to publish the final version.