The Role of Sugars in Fat Accumulation in the Liver

1. INTRODUCTION Non-alcoholic fatty liver disease (NAFLD) is currently the leading cause of chronic liver disease, it affects approximately one-quarter of the global population and the incidence of the disease is still rising (doi: 10.1210/endrev/bnz009). The occurrence of NAFLD is tightly associated with the obesity, insulin resistance and/or type 2 diabetes (doi: 10.7759/cureus.20776). Some authors even consider NAFLD to be hepatic manifestation of metabolic syndrome or even the precursor of metabolic syndrome development (doi: 10.1016/j.dld.2014.09.020).

   The growing incidence of NAFLD has been attributed to sedentary life-style, energy intake inadequate to energy expenditure and also to increased intake of simple carbohydrates (added sugars), and specifically, increased intake of fructose (both as a part of sucrose molecule and also, especially in the United States (US), as a component of high-fructose corn syrup (HFCS)). Over the last decades, a number of papers have been presented which have shown a link between increased consumption of added sugars (especially of fructose) and the obesity and insulin resistance pandemic (doi: 10.1097/MOL.0b013e3283613bca, doi: 10.1007/s00394-018-1711-4, doi: 10.1016/j.nut.2013.08.014).

   However, recent meta-analyses have not supported the role of increased added sugar or fructose intake in development of obesity, insulin resistance, metabolic syndrome and type 2 diabetes (doi: 10.1017/S0954422414000067, doi: 10.1007/s00394-016-1345-3). They rather suggest that the added sugars are detrimental if associated with inadequately increased energy intake. It was even suggested that increased fructose/added sugars consumption may be a marker of unhealthy life-style. On the other hand, not so many data are available with respect to the role of added sugars and, specifically, fructose, in the development of fatty liver.

   This may be due to the fact that the metabolism of glucose and fructose metabolism in the liver is fundamentally different (doi: 10.1016/j.jada.2010.06.008, doi: 10.3390/nu9091026, doi: 10.1016/j.clnu.2020.12.022). Contrary to glucose, which is largely metabolized by extrahepatic tissues, most of fructose is directly uptaken and metabolized by the liver. Moreover, phosphorylation of fructose to fructose-1-phosphate catalyzed by fructokinase/ketohexokinase is not under any feedback control and results in uncontrolled production of trioso-3-phosphates. These can enter gluconeogenesis pathway and be further used to replete glycogen (in state of glycogen depletion) but they can also provide a substrate for de novo lipogenesis (DNL) and, at the same time, inhibit fatty acid oxidation. Fructose seems to be also a potent activator of carbohydrate-responsive element-binding protein (ChREBP) regulatory pathway that together with sterol regulatory element-binding protein (SREBP) promotes DNL in the liver.

   Importantly, in spite of the fact that fatty acids released from adiposse tissue are the major source of the liver fat, the contribution of DNL to liver fat accumulation significantly increases specifically in patients with NAFLD (doi: 10.1172/JCI23621, doi: 10.1053/j.gastro.2013.11.049).

   Studies looking at the effect of fructose consumption on hepatic fat content (HFC) have not reach a clear conclusion. In healthy subjects, fructose administration under isocaloric conditions does not seem to have any impact on HFC (doi: 10.1093/ajcn/84.6.1374, doi: 10.1093/ajcn/nqz271). Fat accumulated in the livers of healthy subjects only when exposed to a hypercaloric high-fructose diet (doi: 10.3945/ajcn.2008.27336). On the contrary, abdominally obese men accumulated more fat in the liver even on an isocaloric high-fructose diet (doi: 10.1111/joim.12632). Interestingly, in the study comparing a high-glucose and high-fructose diet under both iso- and hypercaloric conditions, overweight men accumulated liver fat in the liver only under hypercaloric conditions but no difference in the effect on HFC was found between glucose and fructose (doi: 10.1053/j.gastro.2013.07.012).

2. PILOT DATA Glucose and fructose can have different effects in the accumulation of fat in the liver which we have demonstrated in our studies of acute changes of HFC after nutrient administration (doi: 10.1093/ajcn/nqy386, doi: 10.26402/jpp.2021.1.05). In these studies we developed and introduced a functional test that allowed us to monitor acute changes in hepatic fat after a challenge with high-fat and/or simple carbohydrate load. In the test, the 150 g of fat is administered to subjects at the beginning of experiment, whereas carbohydrates (50 g of fructose or glucose) are administered every 2 hours to efficiently suppress the lipolysis in adipose tissue as documented by suppression of non-esterified fatty acids (NEFA) concentration (doi: 10.1093/ajcn/nqy386). In this way we can limit the impact of fatty acids from the adipose tissue on accumulation of hepatic fat and the resulting changes should be explained by contribution of dietary fat or DNL.

   Importantly, in this model we have shown that 6 hours after consumption of 150 g of fat (cream), HFC increases in both healthy non-obese subjects with normal fat content (doi: 10.1093/ajcn/nqy386) and also in non-obese subjects with hepatosteatosis (doi: 10.26402/jpp.2021.1.05). Liver fat increases also when fructose is co-administered with fat. On the contrary, glucose co-administration with fat prevents such an increase in HFC (doi: 10.1093/ajcn/nqy386, doi: 10.26402/jpp.2021.1.05).

3. THE RATIONALE OF THE PROJECT The added sugars represent more than 12% of daily energy intake. Their role in accumulation of liver fat has been studied so far with pure glucose or pure fructose. However, in real life people do not consume pure glucose or fructose, they consume fructose together with glucose both as a part of sucrose or in high-fructose corn syrup. It is not entirely clear how the hepatic fat content will respond to the mixture of both sugars. The question how concurrent consumption of both hexoses affect the accumulation of fat in the liver may be of great practical significance and answering this question might greatly contribute to the modification of dietary recommendations for NAFLD treatment.

   Importantly, as we were unable to demonstrate the difference between the response of HFC to glucose and fructose administration in non-obese non-diabetic subjects (doi: 10.26402/jpp.2021.1.05), it would be preferable to study the acute effects of these three sugars on HFC after their co-administration with high-fat load.

   The number of studies clearly documented adverse effects of excessive fructose consumption on metabolic health. On the other hand, if fructose does harm human health, it would be important to know whether fructose restriction has an effect on HFC. Indeed, fructose restriction led to a significant reduction in liver fat after 9 days in obese children (doi: 10.1053/j.gastro.2017.05.043) and after 6 weeks in obese individuals (doi: 10.1093/ajcn/nqaa332). Based on our previous results, it can be hypothesized that eliminating fructose from the diet and replacing it by starch or glucose should effectively suppress hepatic fat content even in a few days.

4. HYPOTHESES

   Two principal hypotheses will be tested:

   • The increase of hepatic fat content after high-fat load is blunted by glucose but not fructose neither sucrose (Part A of the project).

   • The short-term (7 days) dietary fructose restriction results in a decrease of hepatic fat content (Part B of the project)

5. EXPERIMENTAL DESIGN Part A The design of this part of the study will be the same as that in our previous studies (doi: 10.1093/ajcn/nqy386, doi: 10.26402/jpp.2021.1.05) except that only two measurements of hepatic fat content (HFC) by magnetic resonance spectroscopy (MRS) will be carried out during each experiment. The three examinations lasting approximately 8 hours will be carried out in each subject in the study. First, hepatic fat content (HFC) will be measured using MRS (30-45 mins). The cannula will be then inserted into the antecubital vein and the first blood draw will be carried out. The subjects will then receive 460 ml of cream (high fat load - 150g of fat in total) with fruit tea sweetened with 50 g of one of the sugars (glucose/fructose/sucrose). Two and four hours later they will receive again fruit tea sweetened with the same sugar as at time 0 h. The blood for biochemical analyses will be then taken at 0.5, 1, 2 (before tea), 2.5, 3, 4 (before tea), 4.5, 5 and 6 hours. During the day, the subjects will not receive any food and they will be allowed to drink only water or fruit tea. The measurement of hepatic fat content will be repeated at time 6 hours. During magnetic resonance (MR) examination the volume of visceral fat will be also measured and body composition will be determined. The order of examinations will be randomized and they will be carried out in at least two-week intervals.

   Subjects: Two groups of male volunteers will be studied in this part of the project

   • 15 nonobese subjects (Body Mass Index (BMI) < 30 kg/m2) and 15 obese subjects (BMI ˃ 30 kg/m2). The subjects will be 18 - 70 years old and groups will be age-matched. Based on our previous studies with healthy male subjects the number of 13 subjects in one group should be sufficient to detect 10 ± 10% increment in HFC after the fat load at α = 0.05 (two-tailed paired t-test) and β = 0.10. The slightly higher number per group covers the risks associated with the exclusion or withdrawal of some subjects during the study.

   The exclusion criteria for both groups will be diabetes mellitus (fasting glucose above 7 mmol/l, 2-hour glucose after oGTT ˃ 11.1 mmol/l, or antidiabetic treatment) or other serious illnesses (cardiovascular disease, cancer, etc.), alcohol consumption ˃ 30 g/day, fructose intolerance, use of drugs affecting lipid metabolism and intolerance of MR examination (claustrophobia, metal implants, etc.). Before the entry into the study, subjects will sign the informed consents with participation in the study and genetic tests and before each experiment they will sign the informed consent with MR examination.

   Part B Subjects selected for this part of the study will first complete a detailed three-day dietary record allowing us to quantify their fructose consumption among other things. They will then be instructed in detail how to change their diet to eliminate fructose and receive these recommendations also in writing. On the first day of the seven-day dietary intervention, they will have fasting blood drawn and then the hepatic fat content will be measured by MRS. After that they will receive a package of glucose and selected foods without fructose to supplement their diet. For seven days they then will adhere to the diet without fructose and keep the detailed dietary record. On the last day of the intervention, they will have fasting blood drawn and HFC measured by MRS again (after 168 hours exactly).

   Subjects: Fifteen subjects 18 to 70 years old will be included in the study. Similarly as in Part A, such a number of subjects is slightly higher than 13 - a number sufficient to detect 10 ± 10% decrease in HFC at α = 0.05 (two-tailed paired t-test) and β = 0.10.

   To be included, the hepatic fat content in the subjects must be higher than 6.2% and less than 16.5% which corresponds to steatosis grade 2 (doi: 10.1007/s10334-011-0264-9). The participants of Part A of the study meeting this criterion can be also included in Part B. The exclusion criteria will be the same as in Part A. Before the entry into the study, subjects will sign the informed consents with participation in the study and genetic tests and before each experiment they will sign the informed consent with MR examination.

6. METHODS MR spectroscopy (MRS). Standard single voxel spectroscopy sequence with breath holding (3 Tesla VIDA Siemens magnetic resonance (MR) system equipped with multi-channel surface body coil) will be used to measure hepatic fat content (HFC). MR images in three anatomical orientations will be used to set the volume of interest of 40x30x25 mm. The correction of signal intensities to T2 relaxation will be done.

   To get information about lipid profile of the liver triglycerides (TG) the MRS spectra will be analyzed as described earlier (doi: 10.3390/metabo11090625.).

   HISTO MR imaging (MRI). Proton Density Fat Fraction (PDFF) imaging will be measured by HISTO Siemens protocol approved by Food and Drug Administration (FDA). It includes spectroscopic, Dixon and multiecho Dixon sequences.

   Data evaluation. MR spectra will be evaluated using the LCModel (doi: 10.1002/nbm.698) and percentage of HFC assessed by MRS will be calculated according to Longo (doi: 10.1002/jmri.1880050311).

   Visceral fat content measurement by MR imaging. The content of visceral fat will be measured by turbo spin echo sequence before or after PDFF measurement. To evaluate the volume of visceral and subcutaneous fat software written in MATLAB will be used. To reduce the impact of the operator, visceral fat areas will be evaluated by two independent experts.

   Body composition. The body composition during the screening will be determined using InBody.

   Biochemical analysis. The blood will be collected into tubes containing all necessary inhibitors, immediately placed on the ice and centrifuged to collect plasma within 15 min. Glucose, triglyceride (TG), non-esterified fatty acids (NEFA), insulin, and β-hydroxybutyrate will be measured in all the samples obtained in the study. All the necessary biochemical, Radioimmunoassay (RIA) and enzyme-linked immunosorbent assay (ELISA) methods for determination of these parameters are currently in use in the laboratories of applicants.

   Lipidomics. The VLDL will be isolated by ultracentrifugation, and the analysis of fatty acid profile of VLDL-triglycerides and plasma free fatty acids will be carried out by liquid chromatography - mass spectrometry (LC-MS) in Metabolomics department of Institute of Physiology, Czech Academy of Sciences (paid service).

   Genetics. The polymorphisms of genes (PNPLA, TM6SF2, and others) known to affect accumulation of liver fat will be determined.

   Statistics. Statistical data analysis and correlations will be performed with PRISM 9.1 or JMP software.

7. TIME SCHEDULE OF THE PROJECT 2024: Part A: Preparation of study protocols, search and screening of subjects, start of the study examinations in first subjects - approximately 20 whole-day examinations will be carried out of 90 examinations total (3 examinations per each of 30 subjects). The number of subjects that can be studied is limited by the availability of MR tomograph for research studies (1 day a week) and complexity of examinations. Basic biochemical analyses will be carried out continuously but all the samples from each subject (30 total) will be stored at -80°C and analyzed simultaneously.

Part B: Preparation of study protocols and dietary records that can be used for quantification of fructose consumption. Search and screening of subjects.

2025: Part A: Screening of subjects, study examinations - min. 30 whole-day examinations will be carried out. Continuation of biochemical analyses.

Part B: Screening of subjects (subjects with hepatosteatosis studied in Part A can be also included in Part B). Dietary intervention experiments (approx. n=20) in approx. 10 subjects (1 subject a week can be intervened, including MR examination and overnight isolation of VLDL on the first and the last day of intervention). Biochemical analyses.

2026: Part A: Screening of subjects, study examinations - min. 30 whole-day examinations will be carried out. Continuation of biochemical analyses. Start of intensive data analysis.

Part B: Screening of subjects. Dietary intervention experiments (n = 10) in remaining subjects on the first and the last day of intervention. Biochemical analyses. Specialized biochemical analyses (for example, ELISA for fibroblast rowth factor 19 (FGF-19), selected hepatokines or other proteins that can be involved in regulation of de novo lipogenesis) and lipidomic analysis of fatty acid profile in free fatty acid (FFA) pool and TG in VLDL. Analysis of the data and publication of results.

2027: Part A: Study examinations in remaining subjects - max. 10 whole-day examinations will be carried out. Continuation of routine biochemical analyses. Specialized biochemical analyses (see above). Analysis of the data and publication of results.

Part B: Final analysis of the data and publication of results.

8. ANTICIPATED RESULTS This project will provide new insights into the mechanisms by which simple carbohydrates affect the accumulation of liver fat. Of particular importance will be comparison the role of fructose and glucose (which are intensively studied but not used by population) and that of sucrose, the sugar generally used to sweeten by the whole population. Moreover, the experiment with fructose restriction from the diet should determine whether such an intervention can be useful for treatment of hepatosteatosis. On the whole, both parts of the projects should provide the results that can be implemented in practical dietary recommendations and improve the strategies for management of NAFLD.