**E**ff**ects of a Low Dose of Ca**ff**eine Alone or as Part of a Green Co**ff**ee Extract, in a Rat Dietary Model of Lean Non-Alcoholic Fatty Liver Disease without Inflammation**

**Ana Magdalena Velázquez 1,**†**, Núria Roglans 1,2,3,**†**, Roger Bentanachs 1, Maria Gené 1, Aleix Sala-Vila 4,5, Iolanda Lázaro 4, Jose Rodríguez-Morató 3,4,6, Rosa María Sánchez 1,2,3, Juan Carlos Laguna 1,2,3,\* and Marta Alegret 1,2,3,\***


Received: 21 September 2020; Accepted: 21 October 2020; Published: 23 October 2020

**Abstract:** Non-alcoholic fatty liver disease is a highly prevalent condition without specific pharmacological treatment, characterized in the initial stages by hepatic steatosis. It was suggested that lipid infiltration in the liver might be reduced by caffeine through anti-inflammatory, antioxidative, and fatty acid metabolism-related mechanisms. We investigated the effects of caffeine (CAF) and green coffee extract (GCE) on hepatic lipids in lean female rats with steatosis. For three months, female Sprague-Dawley rats were fed a standard diet or a cocoa butter-based high-fat diet plus 10% liquid fructose. In the last month, the high-fat diet was supplemented or not with CAF or a GCE, providing 5 mg/kg of CAF. Plasma lipid levels and the hepatic expression of molecules involved in lipid metabolism were determined. Lipidomic analysis was performed in liver samples. The diet caused hepatic steatosis without obesity, inflammation, endoplasmic reticulum stress, or hepatic insulin resistance. Neither CAF nor GCE alleviated hepatic steatosis, but GCE-treated rats showed lower hepatic triglyceride levels compared to the CAF group. The GCE effects could be related to reductions of hepatic (i) mTOR phosphorylation, leading to higher nuclear lipin-1 levels and limiting lipogenic gene expression; (ii) diacylglycerol levels; (iii) hexosylceramide/ceramide ratios; and (iv) very-low-density lipoprotein receptor expression. In conclusion, a low dose of CAF did not reduce hepatic steatosis in lean female rats, but the same dose provided as a green coffee extract led to lower liver triglyceride levels.

**Keywords:** caffeine; coffee; dietary supplements; hepatic steatosis; non-alcoholic fatty liver disease

#### **1. Introduction**

Non-alcoholic fatty liver disease (NAFLD) is a spectrum of alterations ranging from simple hepatic steatosis to non-alcoholic steatohepatitis (NASH), cirrhosis, and hepatocellular carcinoma. Hepatic steatosis, defined as the accumulation of triglycerides (TGs) in lipid droplets in at least 5% of the hepatocytes, is the initial reversible phase of NAFLD, affecting around 33% of adults in the US [1]. Although NAFLD is usually associated with obesity, this condition might also be present in individuals with a body mass index in the normal range, which is referred to as lean or non-obese NAFLD [2]. Compared to obese NAFLD, lean individuals with NAFLD are more commonly female and exhibit a lower prevalence of insulin resistance [2,3].

Consumption of sweetened beverages with a high fructose content is one of the main dietary triggers of NAFLD [4]. Despite the implementation of public policies that aim to reduce their consumption, a recent study on diet population trends showed that 42% of energy intake in US adults still comes from low-quality carbohydrates, including fruit juices and added sugars in beverages [5]. Moreover, the consumption of saturated fats, another dietary factor associated with NAFLD, still remains above the recommended maximal intake of 10% of the energy intake [5].

Given the difficulty to avoid excessive consumption of simple sugars and fats in the population, one strategy to fight NAFLD is the inclusion in the usual diet of functional foods or dietary supplements that could be effective to prevent or reduce hepatic lipid accumulation. Several meta-analysis of randomized clinical trials showed that compounds such as resveratrol, silymarin, vitamin E or D, and curcumin, exert positive effects on NAFLD, which might be attributed to their antioxidant or anti-inflammatory properties [4]. However, not all evidences showed clinical efficacy, which could be related to the different doses, formulation issues, or duration of studies [6–8]. Coffee was reported to exert beneficial effects on liver-related disorders [9], including a reduced risk of NAFLD and of liver fibrosis in NAFLD patients, as revealed by a recent meta-analysis [10]. Effects of coffee on NAFLD development were mainly ascribed to its caffeine content. Several studies indicated that caffeine reduces intrahepatic fat accumulation in mice and rats, however, these studies did not specify the dose of caffeine based on animal weight or they used a dose close to the maximal one admitted in humans after interspecies conversion [11–13]. Moreover, coffee contains more than one hundred compounds besides caffeine, and it is especially rich in polyphenols such as chlorogenic acids [14], which might also be responsible for its beneficial effects.

In the present study, we investigated the effects of a moderate dose of caffeine (5 mg/kg/day, alone or as part of a green coffee extract) in a model of hepatic steatosis without obesity and without inflammation, induced in female rats by feeding a cocoa butter-rich, high-fat diet, together with liquid fructose. Female rats were used, as non-obese steatosis is more frequent in females than in males [3]. The aims of the study were to determine whether caffeine at this low dose reversed hepatic steatosis in this model, whether there were different effects when the same dose of caffeine was administered in the form of a coffee extract, and to explore the mechanisms involved.

#### **2. Materials and Methods**

#### *2.1. Animals*

Female Sprague Dawley rats were purchased from Envigo (Barcelona, Spain). Animals were maintained under conditions of constant humidity (40–60%) and temperature (20–24◦C), with a light/dark cycle of 12 h (2 rats/cage). Studies were conducted in accordance with the principles and procedures outlined in the guidelines established by the Bioethics Committee of the University of Barcelona (Autonomous Government of Catalonia Act 5/21 July 1995). The Animal Experimentation Ethics Committee of the University of Barcelona approved all experimental procedures involving animals (approval no. 10106).

#### *2.2. Dosage Regimen*

Forty-eight female rats aged 8 weeks were randomly assigned into 4 groups (*n* = 12 in each), which received: (1) standard rodent chow (control group, CT); (2) high-fat diet, and 10% w/v fructose in the drinking water (high-fat-high-fructose group, HF-HFr); (3) high-fat diet containing caffeine (from Sigma–Aldrich, St. Louis, MO, USA, 0.18 g/kg of diet) and 10% w/v fructose in drinking water (caffeine group, CAF); or (4) high-fat diet containing a green coffee extract providing 0.18 g of caffeine/kg of diet, and 10% w/v fructose in the drinking water (green coffee extract group, GCE). Groups 1 and 2 received their respective diets for 3 months. Groups 3 and 4 received the HF-HFr diet for 2 months, with the caffeine or green coffee extract supplied to the rats incorporated in the high-fat diet pellets during the third month of the protocol. The green coffee extract (a generous gift from Applied Food Science Inc., Austin, TX, USA) was obtained by extraction with 70:30 ethanol: water mixture, and then the extract was filtered, evaporated, and spray dried. The compositions of the control diet (2018 Teklad Global 18% protein) and the high-fat diet (Teklad Custom Diet TD.180456) are detailed in Supplementary material Table S1. Diets containing caffeine and green coffee extract were prepared by Envigo (Madison, WI, USA), by mixing the compounds with the different ingredients of the high-fat diet and pelleting. Fructose solutions were changed every two days. Throughout the treatment, solid food and liquid consumption was controlled three times a week, and body weight was recorded once a week. Based on the amount of diet consumed and the body weight of each rat, the amount of caffeine ingested in both the CAF and GCE groups was 5.0 ± 0.8 mg/kg/day. The human equivalent dose based on body surface area (Km value for humans = 37 and for rats weighing 250 g = 7) was 0.95 mg/kg/day [15].

### *2.3. Open Field Test*

In the last week of the treatment, an open field test (OFT) was performed to study locomotor activity in the control and treated rats. Rats were placed in the middle of a black box (40 × 40 × 40 cm), under a low illumination of 12 lux. Rats underwent habituation sessions for two consecutive days. On the third day, the distance traveled by each rat was monitored during 60 min (SMART® version 3.0 software, Panlab SL, Barcelona, Spain). The OFT apparatus was cleaned with 10% ethanol solution, before using it with another rat.

#### *2.4. Oral Glucose Tolerance Test*

An oral glucose tolerance test (OGTT) was performed in the last week of the treatment, one day after the OFT test. Rats were fasted for 6 h, and a sample of blood was collected from the tail vein (time 0). A glucose solution of 2 g/kg of body weight was then administered by oral gavage, and blood samples were collected from the tail vein at 15, 30, 60, and 120 min after glucose administration. Glucose levels were determined in all blood samples using a hand-held glucometer (Accutrend® Plus System, Cobas, Roche Farma, Barcelona, Spain). Plasma was obtained from blood samples collected at 0, 15, and 120 min, and insulin levels were measured using a rat insulin enzyme-linked immunosorbent assay (ELISA) kit (Millipore, Billerica, MA, USA).

#### *2.5. Sample Preparation*

At the end of the treatment, rats were fasted for 2 h and blood samples were obtained from the tail vein to measure TG, cholesterol, and glucose levels, using an Accutrend® Plus system glucometer (Cobas, Roche Farma, Barcelona, Spain). The rats were then immediately anesthetized with ketamine/xylazine (9 mg/40 μg per 100 g of body weight, respectively) and blood was collected into micro-tubes (Sarstedt AG & Co., Nümbrecht, Germany) through cardiac puncture and centrifuged at 10,000× *g* for 5 min, at room temperature. Rats were euthanized by exsanguination, and the liver and visceral white adipose tissue (vWAT) were collected and weighed. For the histological studies, samples of the liver of each animal were fixed in buffered formalin or were embedded in OCT, frozen quickly in liquid nitrogen, and stored at −80 ◦C. The remaining liver tissues were immediately frozen in liquid nitrogen and stored at −80 ◦C until needed for biomolecular assays.

#### *2.6. Plasma Analysis*

Plasma samples were assayed in duplicates. Insulin and adiponectin concentrations were determined using specific ELISA kits (Millipore, Billerica, MA, USA). Alanine aminotransferase (ALT) activity was determined using an ALT/GPT enzymatic assay kit (Spinreact, Girona, Spain). Insulin sensitivity index (ISI) was calculated as 2/[plasma insulin (nM) × blood glucose (μM) + 1].

#### *2.7. Histological Studies*

Liver samples were dehydrated and paraffin embedded using a Leica TP1020 automatic tissue processor and a Leica EG1150 H Paraffin Embedding Module (Leica Microsistemas, Barcelona, Spain). Samples were cut to 5 microns and stained with hematoxylin and eosin (H&E). Lipid accumulation was analyzed in OCT-embedded liver sections stained with Oil-Red O (ORO). Images were acquired with a Leica DMSL microscope equipped with a DP72 camera (Leica Microsistemas, Barcelona, Spain) and analyzed using Image J 1.49 software (National Institutes of Health, Bethesda, MD, USA). The area of positive ORO staining was calculated as the positively stained area per total area. All procedures were carried out in the Animal Histopathology Laboratory at the University of Barcelona.

#### *2.8. Liver Assays*

Liver TGs were extracted as described by Qu et al. [16] and determined using a TG colorimetric assay kit (Spinreact, Girona, Spain). Total hepatic fatty acid β-oxidation was determined in rat livers, as described by Lazarow [17], using 30 μg of postnuclear supernatant.

#### 2.8.1. Measurement of Fatty Acid Methyl Esters in Liver TGs

Fatty acid methyl esters (FAMEs) from liver TGs were determined by gas chromatography/electron ionization mass spectrometry as described in the Supplementary Methods and Table S3.

#### 2.8.2. Lipidomic Analysis in Rat Liver Homogenates

Levels of diacylglycerols [DAG], ceramides [Cer], and hexosylceramides [HexCer] in rat livers were determined by liquid chromatography-tandem mass spectrometry (LC–MS/MS) system, as described in Supplementary Methods and Table S4.

#### *2.9. RNA Preparation and Analysis*

Total RNA was isolated from the liver samples using Trizol® (Invitrogen, Carlsbad, CA, USA), cDNA was synthesized by reverse transcription and specific mRNAs were assessed by real-time reverse transcription polymerase chain reaction (RT-PCR), as described previously [18]. TBP (TATA-box-binding protein) was used as an internal control. The primer sequences and PCR product lengths are listed in Supplementary Material (Table S2).

#### *2.10. Preparation of Protein Extracts*

Liver samples were homogenized with a Polytron PT 1200E in lysis buffer containing proteases, phosphatases, and deacetylase inhibitors, and incubated for 1.5 h at 4 ◦C. Samples were then centrifuged at 15,000× *g* for 15 min at 4 ◦C, and the supernatants were collected. To obtain hepatic nuclear extracts, samples were homogenized with a homogenization buffer, kept on ice for 10 min, and centrifuged at 1000× *g* for 10 min at 4 ◦C. Lysis buffer was added to the obtained pellet and samples were incubated for 1.5 h at 4 ◦C, before being centrifuged at 25,000× *g* for 30 min at 4 ◦C. The resulting supernatants were then collected. Protein concentrations were determined by the Bradford method [19].

#### *2.11. Western Blot Analysis*

Western blots were performed using three samples per group, each sample was pooled from two animals. A total of 20–30 μg of protein extracts were subjected to SDS-polyacrylamide gel electrophoresis. Proteins were then transferred onto Immobilon polyvinylidene difluoride transfer membranes (Millipore, Billerica, MA, USA), and blocked for 1 h at room temperature, with 5% non-fat milk solution in Tris-buffered saline (TBS) containing 0.1% Tween-20. Membranes were then incubated with specific primary antibodies. Detection was performed using the Immobilion Western HRP substrate Peroxide Solution® (Millipore, Billerica, MA, USA). To confirm the uniformity of protein loading, blots were incubated with anti-β-actin or anti-β-tubulin antibody (Sigma–Aldrich, St. Louis, MO, USA) as a control for total protein extracts, and with anti-TBP antibody (AbCam, Cambridge, UK) for nuclear protein extracts.

#### *2.12. Statistical Analysis*

The results are expressed as mean ± standard deviation (SD). Significant differences between the groups were established by one-way ANOVA and Šidák's post-hoc test for selected comparisons (GraphPad Software version 8, San Diego, CA, USA). When the SD of the groups was different according to Bartlett's test, the data were transformed into their logarithms and ANOVA was rerun, or the corresponding non-parametric test was applied. The OGTT curves for glucose and insulin were analyzed by two-way ANOVA. The level of statistical significance was set at *p* ≤ 0.05.

#### **3. Results**

#### *3.1. The HF-HFr Diet Does Not Induce Obesity or Gluconeogenic Gene Expression*

As shown in Table 1, although the HF-HFr diet induced a 1.8-fold increase in total caloric intake, the final body weight and vWAT weight were not significantly modified. Only the liver weight/body weight ratio showed a significant increase in response to the HF-HFr diet. Locomotor activity (measured as the total distance travelled in the open field test) was not significantly affected by the diet or treatments.


**Table 1.** Zoometric parameters, blood analytes, and open field test results.

Values are expressed as mean ± SD (*n* = 10–12). ALT: alanine aminotransferase; AUC: area under the curve; CAF: caffeine; CT: control; GCE: green coffee extract; HF-HFr: high-fat-high-fructose; ISI: insulin sensitivity index, calculated as 2/(plasma insulin (nM) × blood glucose (μM) + 1); OFT: open field test; vWAT: visceral white adipose tissue. \* *p* < 0.05; \*\* *p* < 0.01; \*\*\* *p* < 0.001 vs. control.

Basal blood glucose and insulin levels were similar across the different groups (Table 1). After a glucose challenge in the OGTT, all groups on the HF-HFr diet exhibited higher glucose levels than the CT group at the shortest time points (Figure 1A). However, no differences were observed in the integrated glucose concentration, which was calculated as the area under the curve (AUC) (Figure 1B). Both the insulin levels (Figure 1C) and the corresponding AUC (Figure 1D) were significantly increased by the HF-HFr diet, with neither CAF nor GCE attenuating this increase. Accordingly, the ISI was

significantly reduced in the HF-HFr group and none of the treatments reversed this decrease (Table 1). The mRNA levels of the insulin-responsive gluconeogenic genes phosphoenolpyruvate carboxykinase (*Pepck*) and glucose-6 phosphatase (*G6Pase*) decreased in the rats fed the HF-HFr diet (Figure 1E–F).

**Figure 1.** Blood glucose (**A**), area under the curve (AUC) for glucose (**B**), plasma insulin (**C**), and AUC for insulin (**D**) at different times after oral administration of a glucose solution (2 g/kg body weight). Results are the mean ± SD of values from 10–12 animals/group. Bar plots representing the mean ± SD mRNA levels corresponding to liver *Pepck* (**E**) and *G6Pase* (**F**) genes from CT (*n* = 5), HF-HFr (*n* = 6), CAF (*n* = 6) and GCE (*n* = 6) experimental groups. \* *p* < 0.05; \*\*\* *p* < 0.001 vs. control.
