Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans

Fredrik Rosqvist; David Iggman; Joel Kullberg; Jonathan Cedernaes; Hans-Erik Johansson; Anders Larsson; Lars Johansson; Håkan Ahlström; Peter Arner; Ingrid Dahlman; and Ulf Risérus


Diabetes. 2014;63(7):2356-2368. 

In This Article

Research Design and Methods


Healthy, normal-weight men and women were recruited by local advertising. Inclusion criteria were age 20–38 years, BMI 18–27 kg/m2, and absence of diabetes and liver disease. Exclusion criteria included abnormal clinical chemistry, alcohol or drug abuse, pregnancy, lactation, claustrophobia, intolerance to gluten, egg, or milk protein, use of drugs influencing energy metabolism, use of n-3 supplements, and regular heavy exercise (>3 h/week). Subjects were instructed to maintain their habitual diet and physical activity level throughout the study. Subjects were fasted for 12 h before measurements and discouraged from physical exercise or alcohol intake 48 h before measurements.

Study Design

The LIPOGAIN study was a 7-week, double-blind, randomized, controlled trial with parallel group design in free-living subjects. The study was carried out from August through December 2011 at the Uppsala University Hospital, Uppsala, Sweden. Subjects were randomized by drawing lots, with a fixed block size of 4 and allocation ratio 1:1. Subjects were stratified by sex and were unaware of the block size. The allocation sequence was only known by one of the researchers (F.R.) but concealed from all other investigators and participants. Double-blinding was ensured by labeling, and the code was concealed from all investigators until the study was finalized.

Dietary Intervention

Forty-one participants were randomized to eat muffins containing either sunflower oil (high in the major dietary PUFA linoleic acid, 18:2 n-6) or palm oil (high in the major SFA palmitic acid, 16:0). Both oils were refined. Body weight was measured, and muffins were provided to participants weekly at the clinic. Muffins were baked in large batches under standardized conditions in a metabolic kitchen at Uppsala University. Muffins were added to the habitual diet, and the amount was individually adjusted to achieve a 3% weight gain. The amount of muffins consumed per day was individually adjusted weekly (i.e., altered by ±1 muffin/day depending on the rate of weight gain of the individual). Subjects were allowed to eat the muffins anytime during the day. Except for fat quality, the muffins were identical with regard to energy, fat, protein, carbohydrate, and cholesterol content, as well as taste and structure. The composition of the muffins provided 51% of energy from fat, 5% from protein, and 44% from carbohydrates. The sugar to starch ratio was 55:45. We chose palm oil as the source of SFA for several reasons; it is particularly high in palmitic acid and low in linoleic acid and is widely used in various foods globally. Sunflower oil was chosen as the source of PUFA because it is high in linoleic acid (the major PUFA in Western diet) but low in palmitic acid. Both oils were devoid of cholesterol and n-3 PUFAs, thus avoiding potential confounding of these nutrients.

Outcome Measures

The primary outcome of this study was liver fat content (determined by magnetic resonance imaging [MRI]). Secondary outcomes included other body fat depots (MRI and Bod Pod; COSMED, Fridolfing, Germany), total body fat (MRI and Bod Pod), and lean tissue (MRI and Bod Pod). All outcome measures were measured at two time points: at baseline and at the end of the intervention. MRI was the primary assessment method.

Assessments of Liver Fat, Pancreatic Fat, and Body Composition

Liver fat content, pancreas fat content, and body composition were assessed by MRI using a 1.5T Achieva clinical scanner (Philips Healthcare, Best, the Netherlands) modified to allow arbitrary table speed. Collection and analyses of the MRI data were performed by two operators at one center under blinded conditions. The coefficients of variation between the two operators were 2.14 ± 2.14%, and the results from the two operators did not differ significantly (P > 0.4). The average from the two operators was used. Body composition was also measured using whole-body air displacement plethysmography (Bod Pod) according to the manufacturer's instructions. Pancreas fat content was assessed by duplicate measurements (SD 0.36%), and the average was used. The same images were used as from the liver fat measurements. The operator was trained by an experienced radiologist. Total-body water content was measured by bioelectrical impedance analysis (Tanita BC-558; Tanita Corporation, Tokyo, Japan).

Global Transcriptome Analysis of Adipose Tissue

Adipose tissue biopsies were taken subcutaneously, 3 to 4 cm below and lateral to the umbilicus by needle aspiration under local anesthesia (1% lidocaine). The samples were washed with saline, quickly frozen in dry ice covered with ethanol, and stored at −70°C until analysis. Hybridized biotinylated complementary RNA was prepared from total RNA and hybridized to a GeneChip Human Gene 1.1 ST Array (Affymetrix Inc., Santa Clara, CA) using standardized protocols. The microarray data have been submitted to the Gene Expression Omnibus in a Minimum Information About a Microarray Experiment–compliant format (accession number GSE43642).

Assessment of Fat Oxidation

D-3-hydroxybutyrate was analyzed as a marker of hepatic β-oxidation using a kinetic enzymatic method utilizing Ranbut reagent (RB1008; Randox Laboratories, Crumlin, U.K.) on a Mindray BS-380 chemistry analyzer (Shenzhen Mindray Bio-Medical Electronics, Shenzhen, China). All samples were analyzed in a single batch.

Dietary Assessment, Physical Activity, and Compliance

Dietary intake was assessed by 4-day weighed food records (at baseline and week 7), and processed with Dietist XP version 3.1 dietary assessment software. During these 4-day periods, subjects wore accelerometers (Philips Respironics, Andover, MD) on their right ankle to assess physical activity. Food craving, hunger, and satiety were assessed in the morning (only at week 7) by the Food Craving Inventory and Visual Analog Scales, respectively. Fatty acid composition was measured in the intervention oils as well as in plasma cholesterol esters and adipose tissue triglycerides by gas chromatography as previously described.[22,23] Hepatic stearoyl-CoA desaturase-1 (SCD-1) activity was estimated as the 16:1n-7/16:0 ratio in cholesterol esters.[22]

Biochemical Measures

Fasting concentrations of plasma glucose and serum insulin were measured as previously described,[22] and homeostasis model assessment of insulin resistance was calculated.[24] Plasma total adiponectin concentrations were measured by ELISA (10–1193–01; Mercodia, Uppsala, Sweden).

Statistical Analysis

Based on previous data,[22] 22 subjects per group were needed to detect a 1.5% difference between groups in liver fat with α = 0.05 and β = 0.2. Differences in changes between groups were analyzed per protocol with the Student t test. Nonparametric variables were log-transformed or analyzed nonparametrically (e.g., liver fat) with a Mann-Whitney U test if normality was not attained by the Shapiro-Wilk test and Q-Q plots. CIs were, however, obtained using t test calculations for all variables. Data are given as mean ± SD or median (interquartile range [IQR]). Correlations between outcome variables and fatty acids are given as Pearson r or Spearman ρ. A P value <0.05 was considered statistically significant. SPSS version 21 (SPSS Inc.) and JMP version 10.0.0 were used for analyzing data. Significance analysis of microarrays (SAM) was used to compare gene expression between groups.


This study was conducted in accordance with the Declaration of Helsinki. All subjects gave written informed consent prior to inclusion, and the study was approved by the regional ethics committee.