Welcome to the new academic journal OBM Hepatology and Gastroenterology. Recent progress in understanding liver, biliary, pancreatic and gastro-intestinal diseases and their treatments has been observed in the world. OBM Hepatology and Gastroenterology publishes interesting and informative reviews, original articles, and invaluable case reports in this area. We also publish basic research as well as clinical research.

Hepatitis A virus (HAV), HBV, HCV, HDV, and HEV are still a serious issue worldwide. Treatments on these viruses have recently improved. However, liver fibrosis, cirrhosis and hepatocellular carcinoma are still critical conditions. We focus on all of these liver diseases. We also focus on broad-spectrum of gastro-intestinal diseases in this journal.

Please accept our special thanks for choosing to publish in the OBM Hepatology and Gastroenterology. We are looking forward to your submissions for OBM Hepatology and Gastroenterology.

Free Publication in 2019
Current Issue: 2019  Archive: 2018 2017
Open Access Research Article
NF-κB; the Critical Link between Immune and Metabolic Pathways: Could NF-κB be Used as a Novel Diagnostic and Prognostic Biomarker for Non-Alcoholic Steatohepatitis?

Athina Chasapi 1, †, *, Konstantinos Balampanis 1, 2, 6, †, Anna Tanoglidi 3, Eleni Kourea 1, George I. Lambrou 4, Vaia Lambadiari 2, Fotios Kalfarentzos 5, Erifili Hatziagelaki 2, Maria Melachrinou 1, Georgia Sotiropoulou-Bonikou 6

1. Department of Pathology, Medical School, University of Patras, 26500 Patras, Greece

2. Second Department of Internal Medicine, Research Unit and Diabetes Center, Attikon University Hospital, National and Kapodistrian University of Athens, Medical School, Rimini 1, Haidari, 12462 Athens, Greece

3. Department of Clinical Pathology, Akademiska University, Uppsala, Sweden

4. First Department of Pediatrics, Choremeio Research Laboratory, National and Kapodistrian University of Athens, Medical School, Thivon & Levadeias 8, 11527, Goudi, Athens, Greece

5. First Department of Propaedeutic Medicine, National and Kapodistrian University of Athens Medical School, Laiko General Hospital, 17, Ag. Thoma St, 11527 Athens, Greece

6. Department of Anatomy and Histology-Embryology, Medical School, University of Patras, 26500 Patras, Greece

† These authors contributed equally to this work.

Correspondence: Athina Chasapi

Academic Editor: Matthias Bartneck

Special Issue: Exploring novel treatment options for liver fibrosis: can anti-angiogenics succeed?

Received: January 28, 2019 | Accepted: May 6, 2019 | Published: May 22, 2019

OBM Hepatology and Gastroenterology 2019, Volume 3, Issue 2, doi:10.21926/obm.hg.1902020

Recommended citation: Chasapi A, Balampanis K, Tanoglidi A, Kourea E, Lambrou GI, Lambadiari V, Kalfarentzos F, Hatziagelaki E, Melachrinou M, Sotiropoulou-Bonikou G. NF-κB; the Critical Link between Immune and Metabolic Pathways: Could NF-κB be Used as a Novel Diagnostic and Prognostic Biomarker for Non-Alcoholic Steatohepatitis? OBM Hepatology and Gastroenterology 2019;3(2):29; doi:10.21926/obm.hg.1902020.

© 2019 by the authors. This is an open access article distributed under the conditions of the Creative Commons by Attribution License, which permits unrestricted use, distribution, and reproduction in any medium or format, provided the original work is correctly cited.


Background: A great number of inflammatory mediators and metabolic biomarkers have been shown to contribute to the development and progression of obesity-induced pathologies, including insulin resistance and non-alcoholic fatty liver disease (NAFLD). Many of those mediators are either targets or activators of Nuclear Factor-κappa B (NF-κB), which is a key transcription factor that plays a pivotal role in the homeostasis and regulation of inflammatory signaling pathways in the liver.

Methods: Our study population consists of 50 morbidly obese patients undergoing planned bariatric surgery, during which biopsies were taken from the visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), skeletal muscle (SM), extramyocellular adipose tissue (EMAT) and the liver. We evaluated the differential protein expression of NF-κB, ERβ (Estrogen Receptor β), NFATc1 (Nuclear factor of activated T-cells) and PGC1a (peroxisome proliferator-activated receptor gamma coactivator 1-alpha) by immunohistochemistry.

Results: We found that NF-κB is the key biomarker in a complicated intra- and inter-tissue co-expression network that interconnects metabolic and immune signaling pathways. We also demonstrated a possible role in lipid metabolism and the development of skeletal muscle insulin resistance and non-alcoholic steatohepatitis (NASH).

Conclusions: Our findings suggest that NF-κB may be the critical link between immune and metabolic pathways and could represent a future preventive and therapeutic target against obesity-induced metabolic diseases. We hope that our study will contribute to the better understanding of the complex inter-tissue connections that are disrupted in obesity and its associated comorbidities.

Graphical abstract


Obesity, diabetes, NAFLD, metabolic syndrome, ΝF-κB, PGC1, NFATc1, ERβ

1. Introduction

Metabolic disorders including obesity, type II diabetes, non-alcoholic fatty liver disease (NAFLD) and atherosclerosis, are nowadays considered part of an extended disease network [1]. Moreover, it has already been proven that chronic low-grade inflammation plays a pivotal role in the development of cardiometabolic diseases [2,3,4].

Non-alcoholic fatty liver disease (NAFLD) is considered a global epidemic which can progress to non-alcoholic steatohepatitis (NASH) and even cirrhosis in a minority of patients. Steatosis, namely the accumulation of fat above the physiological level (<5%) in the hepatocytes, is the first step in the development of NASH. Although steatosis is present in the vast majority of obese subjects, not all of them will eventually develop NASH. On the other hand, many lean individuals may also have the disease. This finding suggests that several factors apart from obesity and insulin resistance are involved in the progression of NAFLD [5]. The pathogenesis of NASH is complicated and requires cell interactions between liver and immune cells populations, as well as crosstalk between liver and other organs such as adipose tissue and skeletal muscle. Currently, there are limited pharmacological treatments for NASH and therefore understanding the underlying pathophysiology is of critical significance [6,7,8].

Previous studies have implicated the role of NF-κB in the development of numerous inflammatory diseases, regarding inflammation as the key factor in their aetiology and progression. The NF-κB family consists of five members: p65 (RelA), RelB, c-Rel, NF-κB1 (p50/p105), and NF-κB2 (p52/p100). All members contain a N-terminal Rel-homology domain, which is required for homo- or heterodimerization, sequence-specific DNA binding, and nuclear translocation of NF-κB via its nuclear localization signal (NLS). As we have mentioned previously, “all NF-κB family members are able to form homo- or heterodimers directing the dimer to a specific set of target genes. NF-κB is deemed the key modulator of inflammation because its activity level is increased and its mediated signalling is associated with a long list of inflammatory and pathological entities” [9,10,11].

NF-κB promotes immunity by regulating the expression of genes involved in inflammatory process. Transcription factor NF-κB was first described in the hematopoietic system, but it has now been proven that it also plays significant roles and mediates inflammation-associated signalling pathways in the liver, adipose tissue and central nervous system [12,13]. In the resting state, NF-κB remains inactive in the cytoplasm, attached to nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor, alpha (IkBa) and as we have similarly reported previously, “this binding masks its nuclear localization sequence (NLS). When cells are stimulated by growth factors, cytokines, pathogens or other stimuli, the IkB kinase (IKK) complex is activated and IKKβ phosphorylates Ser32 and 36 of IκBα. This phosphorylation promotes the degradation of IκBα, exposing the NLS of NF-κB and subsequently causing NF-κB to move to the nucleus where it induces the gene expression of inflammatory mediators. “Numerous NF-κB target genes, such as the tumour necrosis factor a (TNFα), interleukin 6 (IL-6) and monocyte chemoattractant protein-1 (MCP-1), are also involved in the development of obesity-induced insulin resistance” [10,11,14].

It has been demonstrated that obesity promotes inflammation via the IKKβ/NF-κB pathway and that inhibition of this pathway either by genetic deletion of IKKβ or by pharmacological inhibitors, such as high dose of salicylates or aspirin, ameliorates insulin resistance [14,15]. Moreover, it has been established that when inflammation in insulin-resistant or diabetic patients was pharmacologically repressed, both the glycemic control of the patients and the concomitant inhibition of NF-κB activity were improved [16]. These findings indicate that suppressing the IKKβ/NF-κB-dependent inflammatory pathways ameliorates insulin resistance and type II diabetes. Madonna et al has demonstrated that cardiac impairment in obese and diabetic animals is also mediated by NF-κB via CD1d-associated expression [17]. Therefore, it looks like IKKβ is not only a pivotal mediator of obesity-induced diabetes, but could also represent a future therapeutic target for the treatment of diabetes. Moreover, there is enough evidence to suggest that NF-κB activity may be directly responsible for diet-induced insulin resistance [18] and inflammation, which is probably mediated by the Toll-like receptors (TLR) signalling pathway and cytokine receptors [19].

NF-κB has a pivotal role in obesity mediating white adipose tissue (WAT) remodelling and adipokine secretion [20]. It appears that tissue hypoxia activates inflammation with subsequent chronic damage and cell death or repair leading to vessel proliferation and fibrosis. This pathway is mediated by key transcription factors such as NF-κB and is implicated in several common diseases including cancer, myocardial infarction, stroke, diabetes and obesity. Interestingly enough, pro-inflammatory genes can be expressed earlier in hypoxic tissues and in activated resident or recruited leukocytes [21].

It is widely known, which we have also previously mentioned, that “adipose tissue is an important endocrine and metabolic organ that is actively involved in functional ‘cross-talk’ with peripheral organs such as skeletal muscle, liver and pancreas [22]. Visfatin is a recently discovered adipokine with pleiotropic functions. Oita et al published that phosphorylation of NF-κB was associated with visfatin-mediated production of Reactive Oxygen Species (ROS) and blockade of this pathway via selective IKK inhibition led to a partial reduction in oxidative stress. Moreover, the generation of ROS after treatment with visfatin was found to be dependent on de novo transcription and translation. “Taken together, these novel findings provide us with useful information regarding the key role of visfatin and NF-κB in skeletal muscle and therefore in the pathophysiological mechanism of insulin resistance” [10,11,23].

There is no doubt that skeletal muscle is an important insulin responsive tissue, despite the finding that active IKKβ can cause severe muscle wasting without change in insulin signalling [24]. However, it has been published that deletion of IKKβ in macrophages protected mice from the development of obesity-induced insulin resistance [25]. Moreover, it has been indicated that NF-κB activation in macrophages of both muscle and liver adipose tissue can contribute to the development of insulin resistance in these tissues [12].

The present study will examine the role of the complex inter- and intra-tissue signalling pathways in metabolic diseases, focusing on the master regulator of inflammation, NF-κB. The NF-κB pathway merges the inflammatory and metabolic responses and represents the key transcription factor for understanding and eventually treating the metabolic diseases. Our aim is to morphologically evaluate the effect of the inflammatory regulator NF-κB on ERβ and NFAT signalling pathways and the cross-talk with the co-regulator PGC1a in morbid obesity. We have already demonstrated that PGC1a plays an essential role in human metabolism [10,11] and there is plenty of data in the literature suggesting that the underlying inflammation in morbidly obese individuals is mediated via NF-κB signalling pathway. For our work, we applied the well-established method of immunohistochemistry in five different tissues of patients with morbid obesity who underwent planned bariatric surgery and we also performed an “in situ” assessment of the subcellular localization of the above mentioned biomarkers. Thus, the aim of our study is to identify all possible interactions between the investigated biomarkers in the examined tissues and determine possible associations with the development of type II diabetes and NASH, as well as with important demographic characteristics and clinical parameters.

We hope that our study will shed light on the pathophysiology of obesity and its associated metabolic diseases and that the investigated biomarkers could represent novel therapeutic targets against obesity, cardiovascular disease, diabetes, NAFLD and certain types of cancer [26].

2. Materials and Methods

All patients, were asked to provide written consent prior to participation. The present study was approved by the Institutional Review Board of the University Hospital of Patras and the ethics procedures were according to the Declaration of Helsinki (1975, review 2000).

2.1 Tissue Samples and Patients

In previous studies, we have extensively described our patient population [10,11], which consists of 50 morbidly obese individuals who underwent planned bariatric surgery [10,11]. Inclusion criteria have been described previously. Briefly, patients older than 18 years of age were included, which manifested significant obesity (BMI>40) as well as manifested a clear indication for surgery. Further on, another important inclusion criterion was the absence of a severe condition (e.g. renal failure, heart condition, cancer, known chronic infectious or autoimmune disorders). Patients were asked to provide a written consent for participating in the study. Respectively, exclusion criteria included increased alcohol consumption or alcohol abuse (defined as >20g/day), viral hepatitis or other known chronic liver disease or administration of long-term medical treatments leading to or having caused liver injury. Patient study cohort underwent Roux-en-Y gastric by-pass surgical operation, through which biopsies were obtained from abdominal visceral adipose tissue (omentum) (tissue α, VAT), abdominal subcutaneous adipose tissue (tissue γ, SAT), skeletal muscle from rectus abdominis (tissue δ.m, SM) with its extramyocellular fat (tissue δ.ad, EMAT) and liver (tissue ε).

In concordance with both our previous [10,11] and other researchers’ studies [27,28], we used the term extramyocellular adipose tissue (EMAT) for visualizing intra-muscular adipocyte lipid storage located (also termed intramuscular fat: IMAT) and muscles (intermuscular or perimuscular fat: PMAT). Further on, biopsies were obtained in order to identify smaller lipid groups, which are known to be stored within the muscle fibers, known as lipid droplets or intramyocellular lipids (IMCL). Biopsies samples were fixed at the Pathology Department of the University Hospital of Patras and then were paraffin embedded. Biopsies’ serial sections were cut at a thickness of 4um and placed on gelatin-coated glass slides. All biopsies used were collected from August 2005 until December 2006.

Moreover, anthropometric data were collected such as sex, age, BMI, body fat percentage, as well as serum biomarkers such as lipid (total cholesterol: TC, HDL, LDL, TGs) and transaminase (SGOT: serum glutamic-oxaloacetic transaminase, SGPT: serum glutamic-pyruvic transaminase) levels were estimated.

2.2 Immunohistochemistry

Immunohistochemical analysis from formalin fixed, paraffin embedded (FFPE) blocks has been previously described [10,11]. We used the following primary antibodies: pre-diluted rabbit polyclonal antibody against ERβ (AR385-5R, Biogenex, CA,USA), mouse monoclonal antibody against NFATc1 (1:50 dilution, NFATc1: H-10, sc 17834 , Santa Cruz Biotechnology), mouse monoclonal against NF-κB (1:50, IgG1 anti-NFκB subunit p65 ,F-6, Santa Cruz Biotechnology, USA) and rabbit polyclonal antibody against PCG-1 (1:50 dilution, IHC-00029, BETHYL Laboratories, TX, USA).

Briefly, we adhered to the commonly used immunohistochemistry protocol: section deparrafinazation in xylene, rehydration in a series of graded ethanol solutions and incubation with 0.3% hydrogen peroxide for 15 min at room temperature for blocking endogenous peroxidase activity. Then, slides were ready for performing antigen retrieval. Sections were heated either in 10 mM citrate buffer, pH 6 (for ERβ, NF-κB) or 1 mM ethylenediamine tetraacetic acid (EDTA)-NaOH, pH 8 (for NFATc1, PGC1a) for 15 min in the microwave and then left to cool at room temperature. Immunohistochemical procedure was performed as previously described [10,11]. Sections from breast, prostatic and colon carcinoma were used as positive control slides for ERβ, NF-κB and PGC1a, respectively, and lymphoma for NFATc1. We followed the same process for negative control slides, but 1% TBS was used instead of the primary antibody.

2.3 Staining and Evaluation

Hematoxylin-Eosin (H&E) staining was used in order to evaluate the presence and severity of NAFLD according to Kleiner’s histological scoring system [29]. In several cases, additional histochemical stains were applied, such as Masson’s Trichrome, for estimating liver fibrosis.

Slides were scored by two independent observers blinded to all clinical data [10,11], using an Olympus light microscope. Initial microscopical evaluation was performed by choosing representative areas at a 100x magnification, whereas cell count was performed at high magnification (400x). The number of positively stained cells along with the total number of cells in 10 different, non-overlapping fields per section were counted, and thus the percentage of positive stained cells for each section was calculated (positive stained cells %). For PGC1a and NF-κB, both nuclear or cytoplasmic localization was considered positive whereas for ERβ and NFATc1, only the nuclear immunostaining was taken into account. The density of the staining was not taken into consideration for any of the investigated biomarkers. We did not perform quantitative assessment for liver sections, but just categorized the expression of biomarkers into low and high, due to the great number of cells per field.

Tissue preparation was supervised by a consultant pathologist. Our study is a morphological assessment of protein expression and their subcellular localization in the investigated tissues. All slides were evaluated by two independent, experienced observers and in case of discrepancy or doubt, a double-headed microscope was used until a consensus was reached. In the liver, the different types of cells can be easily differentiated from each other only by their morphology and location. Kupffer cells for example, have a characteristic shape and an irregular surface and are located at the luminal side of liver sinusoidal endothelium or extending into the Disse space through the fenestrae. Thus, they cannot be mistaken for hepatocytes, which are square, thick, parenchymal cells with a central nucleus. However, in order to better evaluate NASH and assess inflammation in all of the investigated tissues, we had previously used two additional biomarkers: CD68 and CD3 (data not shown here). CD68 marks macrophages including Kupffer Cells and CD3 marks lymphocytes as previously reported [10,11]. These biomarkers proved to be more useful in differentiating cells when evaluating adipose tissue.

2.4 Statistical Analysis

The multi-parameter analyses were performed with the MATLAB® simulation environment (The Mathworks, Inc., Natick, MA). The level of statistical significance was set at p-value <0.05.

Power analysis was performed according to which estimated sample size was 47 clinically obese patients. We conducted power analysis based on the hypotheses that body weight follows normal distribution and obese subjects (BMI>30) consist of 12.5% of the population aged 20 to 59 years [30,31]. We correlated all biomarkers’ expression levels with the collected anthropometric and clinical parameters as well as the histopathological findings of our study population. Final endpoint of our study is to investigate all possible correlations between biomarkers’ expression, clinical parameters, the development of diabetes and the presence of non-alcoholic steatohepatitis (NASH) in liver biopsy.

Nominal variables were described as absolute (N) and relative frequencies (%), whereas continuous variables are described as the mean±standard deviation (M±SD) or as the median (MED) and the interquartile range (IQR) in case of parametric and non-parametric variables respectively. Normal distribution was assessed with Shapiro-Wilk test.

ERβ and ΝFAT expression levels were first analyzed as continuous variables, but we eventually dichotomized their expression into low and high in order to facilitate statistical analysis and comparisons with other biomarkers. Due to low levels of PGC1a and NF-κB, the expression of these two biomarkers was dichotomized into “positive” and “negative” staining and therefore analyzed as nominal values. The cut-off points for ERβ and NFAT expression were used values between the mean and median expression of each biomarker per tissue [32] in a way so as the 2 groups would contain similar number of individuals. The used cut-off values are displayed in Table 1. All possible associations between two continuous variables were evaluated by using Pearson’s (r) and Spearman’s (ρ) rank correlation tests, depending on the symmetry of the distribution. The correlations between a nominal and a continuous variable were examined with t-test, Wilcoxon and the non-parametric Kruskal-Wallis test. Finally, comparisons between two nominal variables were performed by using Pearson’s Chi square (x2) or Fisher test.

For assessing liver fibrosis, we used Kleiner histological scoring system. For avoiding sample fragmentation, we excluded stages 0 and 4 from further statistical analysis and we did not use the subcategories for stage 1. All raw data is available upon reasonable request.

Table 1 Cut-off values used in our study for dichotomizing biomarkers’ expression in low and high expression (*For NF-κB and PGC1a, we only dichotomized their expression into “negative” and “positive” due to their low detected levels).

3. Results

3.1 Descriptive Statistics of Demographic and Clinical Characteristics

We included 50 morbidly obese patients in our study who underwent planned bariatric surgery. All their anthropometric and clinical characteristics have been described previously [10,11]. Briefly, age was 38.62±10.48 years, BMI 58.6±8.94, body fat (%) 49.63±5.02, SGOT (mg/dl) 27.92±15.23, SGPT (mg/dl) 38.06±20.24, cholesterol (CHOL) (mg/dl) 196.8±38.37, Low Density Lipoprotein (LDL) (mg/dl) 123.4±30.72, High Density Lipoprotein (HDL) (mg/dl) 46.3±13.25 and triglycerides (mg/dl) 156.3±74.92.

Further clinical variables are presented in Table 2 for continuous and nominal values respectively. According to Shapiro-Wilk test, only total cholesterol and LDL levels were normally distributed.

Interestingly enough, our population consists mostly of young women, with less than a third of the subjects suffering from impaired glucose tolerance or full-blown type II diabetes. However, 75% had evidence of non-alcoholic fatty liver disease (NAFLD), while the vast majority had some degree of fibrosis as assessed by liver biopsy.

Body fat content appeared to be higher in women compared to men, as expected, but BMI failed to show statistical difference between the two sexes. Moreover, there was no statistically significant correlation between sex, hypertension, BMI, body fat percentage, total cholesterol, LDL and HDL levels with the presence of diabetes or NASH in our study population. However, transaminases’ levels were positively associated with the presence of NASH. Notably, we detected statistical trends between both intramuscular fat (p=0.059) and steatosis (p=0.063) with the presence of type II diabetes.

Table 2 Absolute (N) and relative (%) frequencies for the nominal variables of our study population (Ν=50) (¥According to Kleiner histological scoring system [29], there are 5 stages of liver fibrosis and stage 1 is subdivided in other 3 categories (0: none, 1a: mild zone 3 perisinusoidal fibrosis, 1b: moderate zone 3 perisinusoidal fibrosis, 1c: portal/periportal fibrosis only, 2: perisinusoidal and portal/periportal, 3: bridging fibrosis, 4: cirrhosis). For avoiding sample fragmentation, we excluded stages 0 and 4 from further statistical analysis and we did not use the subcategories for stage 1. *In liver biopsy. ¥ For 1 of the patients liver biopsy was not technically feasible and thus are 49 instead of 50 tissue samples).

3.2 Descriptive Statistics of Biomarkers’ Expression

All tissues were examined for the expression of ERβ, NFATc1, NF-κB and PGC1a by immunohistochemistry. The investigated biomarkers were expressed in all tissues, a finding that implies their pivotal role in human metabolism. In the present study, we will focus primarily on the expression of NF-κB and we will endeavor to uncover its role in the metabolic inter- and intra-tissue co-expression networks. The absolute and relative frequencies of the biomarkers’ immunostaining in adipocytes, skeletal muscle cells and hepatocytes are displayed in Table 3.

3.3 NF-κB

The inactive form of NF-κB is predominantly localized in the cytoplasm. Several stimuli including inflammatory cytokines can promote the degradation of IκB (NF-κB inhibitor) and subsequently allow nuclear deposition of NF-κB and regulation of specific gene expression. As for PGC1a [10,11,33,34], there is evidence to suggest that NF-κB and IκBa (but not IκBβ) are also located in the mitochondria and that NF-κB can negatively regulate mitochondrial mRNA expression [35,36]. Thus, NF-κB signaling pathways may be involved in cell growth and apoptosis. In VAT, SAT and EMAT, NF-κB was expressed in adipocytes, vascular endothelial and some inflammatory cells including macrophages and lymphocytes. In SM, it was expressed in skeletal muscle cells and in the liver in hepatocytes, vascular endothelial, biliary epithelial, Kupffer cells and some inflammatory cells in cases with documented NASH. In the present study, we only semi-quantitatively evaluated and statistically analyzed its expression in adipocytes, skeletal muscle cells and hepatocytes.

Table 3 Absolute (N) and Relative (%) frequencies of the recorded NF-κB, PGC1a, ERβ and NFATc1 immunostaining in adipocytes, skeletal muscle cells and hepatocytes (*(-): negative expression, #(+): positive expression, low: low expression, high: high expression).

The absolute and relative frequencies of the biomarkers’ immunostaining in adipocytes, skeletal muscle cells and hepatocytes are displayed in Table 3. Representative images of NF-κB immunostaining are shown in Figure 1.

Figure 1 Representative images of NF-κB immunostaining in VAT (Α), SAT (Β), SM (C, D), EMAT (E), liver (F) (original magnifications 400x).

3.4 Correlations between Biomarkers’ Expression and Clinical Characteristics of our Study Population

We found that NF-κB expression in VAT is significantly negatively correlated to body fat content (p=0.0077) (Figure 2A). The median body fat content was 54.65% and 49.1% in patients with negative and positive NF-κB expression in VAT, respectively (Figure 2A). Moreover, we detected a strong negative link between NF-κB expression in VAT and plasma LDL levels (p=0.017) (Figure 2B). Specifically, mean LDL levels were 149.6 mg/dl and 118.4 mg/dl in subjects with negative and positive expression, respectively (Figure 2B). Finally, we noticed a positive statistically significant association between NF-κB expression in the liver and the presence of intramuscular fat (p=0.027) (Figure 2C) and a borderline link with the detection of IMCL (p=0.0521) (data not shown on the graph). The majority of patients with confirmed NASH on liver biopsy had negative hepatic NF-κB staining, whereas most patients with borderline or without NASH had positive hepatic NF-κB staining (Figure 2D).

Figure 2 Diagrams showing the correlation between NF-κB expression in VAT (tissue α) with body fat content (A) and with plasma LDL levels (B). Further on, the frequency of hepatic NF-κB (tissue ε) positive and negative staining is presented in correlation to the presence or absence of intramuscular fat. In particular, it appeared that patients with positive hepatic NF-κB staining and intramuscular fat manifested higher frequency compared to those with positive hepatic NF-κB staining and absence of intramuscular fat (p=0.027). In other words, patients with positive hepatic NF-κB staining were more likely to have intramuscular fat as well (C). The majority of patients with confirmed NASH on liver biopsy had negative hepatic NF-κB staining, whereas most patients borderline or without NASH had positive hepatic NF-κB staining (D) (Legend: NF-κB.α: NF-κB expression in tissue α (VAT), NF-κB.ε: NF-κB expression in tissue ε (liver), Negative: Negative Expression, Positive: Positive Expression).

Furthermore, there was a borderline negative correlation between NF-κB expression in VAT and total cholesterol levels (p=0.055) (Data not shown).

3.5 Correlations between Biomarkers’ Expression and Study Endpoints

The most striking finding was the statistically significant association between NF-κB expression in the liver and the presence of NASH (p=0.049) (Figure 2D). Specifically, among patients with negative NF-κB expression, 15 (30.6% of all enrolled patients) had a definite diagnosis of NASH, 4 (8.2%) had a borderline diagnosis of NASH and 2 (4.1%) did not meet the pathological criteria for steatohepatitis. On the other hand, amongst subjects with positive NF-κB expression in the liver, 11 (22.45%) were diagnosed with NASH, 7 (14.3%) were borderline and 10 (20.41%) did not meet the criteria for such a diagnosis. Although there is no doubt that there exists a strong link between NF-κB expression in the liver and the development of NASH, our findings are not enough to establish a positive association between these two parameters. Representative images of NF-κB immunostaining in patients with, without and borderline NASH are presented in Figure 3.

No strong relationship was detected between NF-κB expression in any of the tissues with the presence of type II diabetes.

3.6 Inter- and Intra-Tissue Co-Expression Networks

In the present study we focused on the co-expression networks that involve NF-κB and we demonstrated a wide intra- and inter-tissue co-expression net, indicative of the complex, intercommunicating signaling pathways in human metabolism. In particular, we have found that significantly more patients with both low and high NFATc1 expression in subcutaneous adipose tissue manifested positive staining for NF-κB in SM (p=0.017) (Figure 4A), while significantly more patients with low ERβ expression in the liver were also negative for NF-κB expression in EMAT (p=0.024) (Figure 4B). Furthermore, the majority of the patients with low ERβ expression in extramuscular adipose tissue also had negative PGC1 staining in the liver as compared to all other subgroups (Figure 4C). The key point in our observation is that all patients with positive NF-κB expression in SM also had high NFATc1 expression in SAT (Figure 4A). We found that all patients with negative NF-κB expression in SM also have low ERβ expression in the liver (Figure 4B). Interestingly, most patients with negative PGC1a expression in the liver also had low ERβ expression in EMAT (Figure 4C).

Moreover, we observed strong negative correlations between NF-κB and PGC1a expression in the following biomarker-tissue pairs: (i) NF-κB and PGC1a in VAT (rho=-0.33, p=0.03), (ii) NF-κB and PGC1a in SAT (rho=-0.30, p=0.04), (iii) NF-κB in VAT and PGC1a in EMAT (rho=-0.58, p=0.00064), (iv) NF-κB in SAT and PGC1a in EMAT (rho=-0.42, p=0.018), (v) NF-κB in VAT and PGC1a in SM (rho=-0.35, p=0.025), (vi) NF-κB in SAT and PGC1a in SM (rho=-0.30, p=0.04), (vii) NF-κB in VAT and PGC1a in SAT (rho=-0.35, p=0.025) and (viii) NF-κB in SAT and PGC1a in VAT (rho=-0.31, p=0.038). We also demonstrated statistically significant positive links between the following biomarker-tissue pairs: (i) NF-κB in SAT and NF-κB in VAT (rho=0.56, p=0.003), (ii) NF-κB in SAT and NF-κB in SM (rho=0.38, p=0.03) and (iii) NF-κB in VAT and NF-κB in SM (rho=0.36, p=0.021). Correlation results are summarized in Figure 5.

Furthermore, Figure 6 displays characteristic images of ERβ, NFATc1 and PGC1a immunostaining, based on the statistically significant correlations presented in Figure 4.

Figure 3 Representative images of negative and positive NF-κB hepatic immunostaining in patients with (A, B) without (C, D) and borderline NASH (E, F) (original magnifications 400x). (A): NASH with negative NF-κB immunostaining, (B): NASH with positive NF-κB immunostaining, (C): No NASH with negative NF-κB immunostaining, (D): No NASH with positive NF-κB immunostaining, (E): Borderline NASH with negative NF-κB immunostaining, (F): Borderline NASH with positive NF-κB immunostaining.

Figure 4 Bar plots showing the correlation between (A) NF-κB expression in SM (tissue δ) and NFATc1 expression in SAT (tissue γ), (B) NF-κB expression in EMAT (tissue δ.adipose) and ERβ in the liver (tissue ε) and (C) PGC1a expression in the liver (tissue ε) and ΕRβ in EMAT (tissue δ.adipose). In particular, all patients with high NFATc1 expression in SAT also had positive NF-κB staining in SM (A). In addition, the majority of the patients with low ERβ expression in the liver also had negative NF-κB expression in EMAT (B). Finally, significantly more patients with low ERβ expression in ΕΜΑΤ also had negative PGC1 staining in the liver (C) (Legend: tissue γ: SAT, tissue δ.m: SM, tissue δ.adipose: EMAT, tissue ε: liver, Negative: Negative Expression, Positive: Positive Expression, Low: Low Expression, High: High Expression).

Figure 5 Correlation matrix that gives a comprehensive view of all inter- and intra-tissue correlations between NF-κB and PGC1a expression. All statistically significant correlations (>30%) are shown in grey, whereas white indicates no strong link. The density of the color increases with the significance of the correlation. Negative sign (-) in front of the number indicates a negative correlation between the 2 parameters and no sign indicates a positive one (Legend: a: VAT, γ: SAT, δ.muscle: SM, δ.adipose: EMAT, e: liver).

Figure 6 Representative images of ERβ, NFATc1 and PGC1a immunostaining based on the statistically significant correlations displayed on the diagrams in Figure 4. (A) NFATc1 immunostaining in SAT (B) ERβ immunostaining in the liver (C) ERβ immunostaining in EMAT (D) PGC1a immunostaining in the liver (original magnifications 400x).

4. Discussion

4.1 NF-κB in VAT Negatively Regulates Body Fat Content and LDL Levels

In the present study, we demonstrated strong negative links between NF-κB expression in VAT with body fat content (p=0.0077) and LDL plasma levels (p=0.017) and borderline association with total cholesterol levels (p=0.055). There are a few studies in the literature that correlate lipid metabolism and energy expenditure with inflammation and NF-κB and subsequently partly support our findings. Increased energy expenditure can often be translated in practical terms as reduced body fat content and protection from diet-induced obesity. To begin with, Tang et al (2009) studied extensively the transcriptional activity of NF-κB by using two different animal models [over-expression of NF-κB p65 (RelA) in aP2-p65 mice and inactivation of NF-κB p50 (NF-κB1) through gene knockout] and found it increased in both groups. Both animal models had increased energy expenditure without changing their eating habits and did not develop adulthood or diet-induced obesity. The authors suggested that NF-κB induces energy expenditure and inhibits adipose tissue expansion and recommended that inflammation may reduce lipid accumulation and subsequently prevent insulin resistance [37]. Furthermore, it has already been established that NF-κB activation is associated with energy expenditure in cachexia [38] and that it inhibits both differentiation and function of adipocytes in the signaling pathway of TNF-α [39]. The underlying mechanism involves suppression of PPARγ by NF-κB and this negative regulation may lead to inhibition of adipocyte differentiation and adipose tissue expansion in transgenic mice, increase in energy expenditure and eventually protection from diet-induced obesity [37]. Moreover, the authors claimed that the transgenic mice gained less weight and had less fat content compared to their control littermates, implying that the genetic modification of NF-κB p65 may play a vital role in preventing obesity [37]. In summary, their findings suggest that NF-κB mediated inflammation increases energy expenditure and may finally protect from obesity and insulin resistance, which is in agreement with our results [37].

In the same context, Calleros et al (2006) published that significant reduction in cholesterol levels enhances NF-κB transcriptional activity by several mechanisms including IκBa degradation and translocation of p65/NF-κB to the nucleus and p65/NF-κB transactivating potential regulation through a p38 MAPK/MSK1 mediated pathway [42]. Finally, Rudofsky et al (2012) published that LDL reduction with high dose simvastatin reduces both proinflammatory NF-κB binding activity and high sensitive CRP (hsCRP) levels, while combination treatment with low dose simvastatin and ezetimibe resulting in a similar LDL-reduction does not have such effects [40].

Dyslipidemia is a disorder of lipid metabolism and is characterized by increased levels of total cholesterol, LDL cholesterol and triglycerides and decreased levels of HDL cholesterol. Increased plasma cholesterol levels is detrimental due to lipid peroxidation, a process that produces Reactive Oxygen Species (ROS) and causes tissue damage, inflammation and endothelial dysfunction. ROS has also been proven to activate NF-κB, which is involved in several biological functions including inflammation [41]. However, despite the association with oxidized low-density lipoprotein (ox-LDL) circulating levels, some researchers failed to demonstrate a direct link between NF-κB inflammatory pathway and plasma lipid levels. More specifically, Cominacini et al (2005) examined the results of ox-LDL circulating levels on NF-κB activation in peripheral blood mononuclear cells of patients with unstable or stable angina and control individuals [42]. They concluded that patients with unstable angina had higher levels of circulating ox-LDL and NF-κB than control group and the ones with stable angina and the increase in circulating NF-κB was mainly due to the activation of monocytes. Notably, no significant differences in plasma lipid levels were detected. Their findings imply that NF-κB activation is -at least partly- triggered by ox-LDL [42]. Nevertheless, it is known that LDL has to be modified and extensively oxidized in order to induce inflammation and become atherogenic. Oxidized LDL itself is a potent chemoattractant with direct effects on monocytes and T lymphocytes and plays a crucial role in their migration and differentiation in the intima.

4.2 Hepatic NF-κB Positively Correlates with Intramuscular Fat

We detected a strong positive link between NF-κB expression in the liver and intramuscular fat (p=0.027) and a borderline association with the presence of IMCL (p=0.0521), although we failed to show a direct statistically significant correlation between NF-κB expression and the development of type II diabetes. However, it has already been established that the presence of intramuscular fat and IMCL is a sign of systemic inflammation and thus closely related with the development of insulin resistance and type II diabetes [28,43,44].

It has been demonstrated that low grade, chronic inflammation in skeletal muscle can lead to obesity, insulin resistance and type II diabetes, which is responsible for increased morbidity and mortality. Glucose is being primarily disposed in skeletal muscle and thus impaired insulin action in this tissue drives whole body insulin resistance. Thus, preventing skeletal muscle inflammation could theoretically protect us from developing chronic diseases such as diabetes [45]. Fatty acid accumulation increases circulating levels of pro-inflammatory cytokines in obese individuals via NF-κB activation, which impairs several signaling pathways that regulate skeletal muscle insulin signaling and fatty acid oxidative capacity [46,47]. NF-κB is activated by numerous pro-inflammatory stimuli such as TNF-α, lipopolysaccharide (LPS), and free fatty acids (FFA) and therefore NF-κB signaling represents the inflammatory status in skeletal muscle [48].

Moreover, it appears that intramuscular and intrahepatic lipid accumulation via various mechanisms [47,48] is linked to high plasma FFA levels in obese individuals, which can initiate low-grade, chronic inflammation in skeletal muscle and liver via direct NF-κB activation, resulting in release of several pro-inflammatory and proatherogenic cytokines [48]. This in turn can cause insulin resistance in skeletal muscle, liver and endothelial cells, leading finally to insulin resistance, type II diabetes, hypertension, dyslipidemia, atherosclerotic vascular disease and NAFLD. Green et al published that NF-κB-p65 DNA binding activity was found to be significantly increased in myocytes from obese diabetic patients compared to lean individuals and that insulin-induced glucose uptake was inhibited in myocytes from obese individuals with impaired glucose tolerance or type II diabetes [45].

 In addition, there is data that lipid accumulation in the liver drives hepatic inflammation and subsequent hepatic insulin resistance, via NF-κB activation and downstream cytokine release [48,49]. Several studies have demonstrated that TLRs (particularly TLR4), also via NF-κB activation, play a crucial role in mediating pro-inflammatory effects of saturated fatty acids, further contributing to insulin resistance. TLR4 expression is enhanced in obese subjects and its depletion is related to adipose tissue and skeletal muscle inflammation [50].

In summary, it seems that cellular NF-κB has effects to intramuscular fat and insulin sensitivity and there is plenty of data in the literature that support our findings. We believe that NF-κB could represent a novel therapeutic target for treating metabolic diseases such as type II diabetes and NASH, as it is a critical link between inflammation and insulin resistance.

4.3 Hepatic NF-κB Mediates the Initiation and Progression of NASH

We detected a statistically significant relationship between NF-κB expression in the liver and the presence of NASH (p=0.049), although we were unable to unveil the exact pattern of this link. Other researchers have also published similar correlations between NF-κB expression and NASH, backing up our results. According to Videla et al (2009), hepatic NF-κB expression is increased in obese individuals with NASH and is significantly associated to oxidative stress and insulin resistance [51]. There is evidence that the IKK-β/NF-κB signaling pathway, which is activated by obesity and high-fat diet, is associated with the chronic inflammation that occurs in hepatic steatosis. In addition, hepatic release of pro-inflammatory cytokines such as TNF-α, IL-6 and IL-1β is increased, suggesting that lipid accumulation in the liver drives subacute inflammation via NF-κB activation and cytokine production [15,52]. We also know that activation of NF-κB demands ceramide biosynthesis and that activation of pro-inflammatory pathways in the adipose tissue requires increased availability of free fatty acids, generation of oxidative stress-related products [53], and endoplasmic reticulum stress [54].

Moreover, Locatelli et al (2013) [55] published that NF-κB activation contributes to the pathogenesis of NASH by favoring natural killer T cells recruitment. Among the NF-κB subunits, it appears that p50/NF-κB1 has regulatory activities, down-modulating NF-κB-mediated responses. Moreover, Zhou et al (2015) [56] demonstrated that activating the cholinergic anti-inflammatory pathway inhibits the NASH-related inflammatory reaction and the mechanism for this inhibition involves the NF-κB signaling pathway.

Of all the mechanisms related to NASH, oxidative stress has been most widely studied. Oxidative stress is triggered by an imbalance between pro-oxidants and antioxidants. It is now clear that oxidative stress can provoke liver damage either directly by causing cell injury or indirectly by changing signaling pathways. For example, ROS activate NF-κB, which regulates the production of several pro-inflammatory cytokines such as IL-1β, TNFα, and IL-6. Hepatic inhibition of NF-κB is expected to improve diet-induced inflammation in the liver [25,57]. According to Zhao et al (2012), one of the possible therapeutic mechanisms of pioglitazone in NASH involves down-regulation of hepatic NF-κB and COX-2 expression [58].

It has also been suggested that NF-κB represents the link between hepatic damage, fibrosis and hepatocellular carcinoma (HCC) [59]. Moreover, it has been demonstrated that inflammatory microniches may facilitate the survival and proliferation of malignant hepatocyte progenitor cells before they were able to form tumors, a process which is related to autocrine production of cytokines and depends on NF-κB activation and the presence of T cells [60,61]. Luedde et al (2007) published that NEMO (NF-κB essential modulator)-mediated NF-κB activation in the liver can prevent the development of steatohepatitis and HCC, implying that NEMO acts as a hepatic tumor suppressor [62]. It has already been established that NF-κB prevents cell death in the fetus, but little is known about the hepatic function of NF-κB in adults [63]. However, there is contradictory data in the literature regarding the role of NF-κB in hepatic cancer, as it can act both as a tumor promoter and a tumor suppressor [62,64]. This inconsistency in researchers’ findings may just represent liver’s unique characteristics, which is capable of regenerating after injury via hepatocellular proliferation. NF-κB may be the key molecule that connects inflammatory, survival, and metabolic pathways in the liver that link NASH to liver cancer. There is no doubt that extensive research is needed regarding the potential side effects of drugs targeting the IKK/NF-κB pathway.

Taken together, these data highlight the significant role of NF-κB in metabolic diseases such as insulin resistance and NASH [54]. Many researchers have been published, in line with our own findings, that NF-κB pathway is mediated in NASH development and progression. However, the exact role of NF-κB in the pathophysiology of NASH is more complicated than we originally thought. Interestingly enough, there is evidence to suggest that inflammation is necessary for liver regeneration, which is mediated through anti-apoptosis and pro-proliferative characteristics of NF-κB [65]. The characteristics of chronic inflammation that induce NASH are not clear. Liang et al investigated the effects of non-metabolic (LPS, IL-1β) and metabolic dietary trigger factors (carbohydrate, cholesterol) of inflammation on the transition of simple steatosis to NASH and concluded that high-fat diet followed by NF-κB activation per se is not enough for progression to NASH. On the other hand, when is being followed by metabolically provoked inflammation, it activates additional inflammatory pathways and can progress to NASH [66].

In summary, NF-κB has both favorable and unfavorable functions and thus its inhibition apart from the benefits may also have negative impacts on hepatocyte viability. Identifying novel targets and selective therapeutic agents with tissue-specific actions is crucial in order to avoid liver injury associated with complete NF-κB blockade in hepatocytes [67].

4.4 NF-κB Co-Activates the Transcriptional Activities of ERβ and ΝFATc1 in a Tissue-Specific Manner

It seems that NF-κB is a master regulator of immunity, stress responses, apoptosis and differentiation. Several stimuli can influence NF-κB activation, mediating different transcriptional programs. As a result, NF-κB-mediated transcription is positively and negatively regulated and tightly coordinated with other signaling pathways. This complicated cross-talk network is necessary for shaping the different biological functions of NF-κB into cell type- and context-specific responses [13]. NF-κB mediated transcription can also be affected by its crosstalk with other transcription factors and co-regulators, which can be subdivided to co-activators and co-repressors. The co-regulators can either activate or inhibit the transcriptional activity of the enhanceosome (i.e., the multiprotein complex mediating promoter activation and gene transcription) or alter the chromatin structure [9].

In the present study, we uncovered a complex and extensive intra- and inter-tissue co-expression network with NF-κB being at the center of our interest. Regarding NF-κB expression alone, we identified strong positive links between specific tissue pairs including (i) VAT and SAT (p=0.00066) (ii) SAT and SM (p=0.017) (iii) VAT and SM (p=0.026). This finding is meaningful, as it establishes the activity of the biomarker in the above mentioned tissues and is suggestive of its prominent role in human metabolism and the underlying pathophysiological mechanisms in morbid obesity.

Moreover, we found that NF-κB expression in SM is positively correlated to NFATc1 expression in SAT (p=0.003) and NF-κB expression in EMAT with ERβ expression in the liver (p=0.0119). ERβ could also be co-activated by PGC1a cofactor, as we detected a significant positive link between PGC1a expression in the liver and ERβ expression in EMAT (p=0.0191).

There is plenty of evidence in the literature supporting the cross-talk between NFAT and NF-κB families in human metabolism and several pathologies including postmenopausal osteoporosis and rheumatoid joint destruction [68], osmotic stress [69] and certain developmental and inflammatory processes [70]. However, there is still lack of data regarding co-expression networks in the metabolic tissues and obesity induced diseases and thus further research is needed in this field. Cai et al. (2011) showed a connection between these 2 families and carcinogenesis. In particular, it seems that ROS generation due to nickel exposure is involved in the regulation of signaling leading to NFAT and NF-κB activation, the COX-2 induction and finally lung carcinogenesis [71]. Moreover, a tight connection has been demonstrated between NFATc1 and NF-κB in B [72] and T lymphocytes regulation [73,74]. Finally, there is limited data that NF-κB NFATc1 cross talk is involved in obesity-induced white adipose tissue inflammation, contributing to the pathogenesis of metabolic syndrome, atherosclerosis and insulin resistance [75].

ΝFκB-ERβ cross-talk, has been studied mostly in cancer including prostate [76], breast, colon and CNS malignancies [77]. This interaction also plays an important role in human endometrial epithelial cells and alterations in the cross-talk between the ER and NF-κB signaling pathways may contribute to early pregnancy loss and infertility [78]. Estrogens are known to have anti-inflammatory actions, whereas activated NF-κB mediates inflammatory responses [79]. There is data in the literature regarding the negative cross-talk between NF-κB and ERs both in vitro [80] and in vivo [81]. For example, direct communication between ERα or ERβ and p65 [80,82], as well as cofactor-mediated cross talk and modulation of IκBa expression by estradiol have been reported. However, other studies have demonstrated a positive interaction between NF-κB and ER signaling [78]. For example, Adamson et al (2008) [83] published that treatment with estradiol and TNFa had a synergistic effect on estrogen response element (ERE)-dependent stimulation of the human prolactin gene in GH3 pituitary cells. Additionally, pro-inflammatory cytokines (TNFa, IL-1b or IL-6) and estradiol can also act synergistically to up-titrate expression of prostaglandin E synthase (PGES) mRNA levels in MCF7 breast cancer cells [84].

4.5 NF-κB -Mediated Transcription is Co-Repressed by Pgc1a in all Investigated Metabolic Tissues

We identified significant negative correlations between NF-κB and PGC1a expression in several tissue pairs as described extensively above. There is evidence of cross-talk between NF-κB and PGC1a in human metabolism and inflammatory diseases, but we still need to clarify their exact role in obesity and its related commorbidities. For example, Alvareez-Gardia et al. (2010) published that the p65 subunit of NF-κB binds to PGC-1a, linking inflammation and metabolic disturbances in the heart [85]. NF-κB activation is triggered by several factors, including high glucose levels and pro-inflammatory cytokines and is associated with cardiac hypertrophy and heart failure. It has been reported that NF-κB-mediated suppression of PGC-1a might explain the change in glucose metabolism seen in heart diseases, which is caused by pro-inflammatory stimuli. The underlying mechanism involves exposure to TNFa, which results in PGC1a down-regulation in cardiac cells, leading to a reduction in pyruvate dehydrogenase kinase 4 (PDK4) expression and finally to the metabolic dysregulation seen in cardiac conditions [85].

Moreover, it has been demonstrated that PGC1 coactivators inhibit the transcriptional activity of NF-κB in skeletal muscle [86,87]. PGC1s are metabolic coactivators whose function is impaired in skeletal muscle in numerous chronic diseases and they also regulate the NF-κB pathway and thereby the inflammatory process. These findings represent an exquisite example of cross-talk between metabolic and immune pathways in skeletal muscle. Moreover, a bidirectional negative regulation of PGC1α and NF-κB may be the key for the mutual antagonism between oxidative metabolism and inflammation in skeletal muscle [88]. In conclusion, targeting PGC1 in chronic diseases such as obesity, type II diabetes and NASH, might prevent inflammation and disease progression and therefore improve metabolic health [87].

4.6 Study Limitations

One of the limitations of our study is the lack of a lean control group that would allow for direct comparisons between biomarkers’ expression in the two categories. Although there is limited data in the literature regarding the expression of the investigated biomarkers in healthy, lean individuals [89,90,91,92,93,94], we are still far from understanding the tissue-specific differences in biomarkers’ expression between obese and non-obese subjects. Further research is required for unmasking the distinct biomarkers’ roles as well as their communication networks in human homeostasis and the subsequent alterations in obesity-induced diseases. Moreover, quantitative methods would have been helpful for confirming our immunohistochemical findings and strengthening our study results.

5. Conclusions

In summary, NF-κB appears to be a master regulator of human metabolism which acts in a tissue specific manner and links metabolic and immune signaling pathways. Our findings represent a prime example of such a cross-talk and demonstrate that NF-κB dysregulation may be implicated in obesity and its associated comorbidities, including insulin resistance and NASH.

While weight loss and physical exercise have provided clear benefits, targeted therapy remains still an unmet need. The development of novel pharmacological agents requires a detailed and in-depth knowledge and understanding of the underlying, complex pathophysiological mechanisms. It is clear that NF-κB inflammatory pathways induce metabolic disease and successful therapeutic agents may include directly targeting NF-κB genes. Although inhibitors of specific NF-κB pathways are not yet clinically available, there is progress in the development of more selective anti-NF-κB pharmaceutical agents. We hope that our findings, along with the explosion of information witnessed in this field during the last few years, will contribute to unmasking the systemic details of obesity and eventually lead to new, targeted and personalized therapies. There is no doubt that further research in this field of interest and larger scale studies are needed in order to confirm our findings.


Not applicable.

Author Contributions

AC: Designed the study and interpreted the results, drafted the manuscript, performed immunohistochemistry staining, performed slides evaluation & score, performed statistical analysis, collected samples, collected biopsies, processed samples, performed experiments. KB: Collected tissues after surgery, took blood samples and evaluated the results, performed slides evaluation, recruited and informed patients, received written consent for participation in the study, processed samples. AT: slides evaluation & score. EK: Performed immunohistochemistry staining, performed slides evaluation & scoring, interpreted the results and provided clinical insight. GIL: proof-read the manuscript, reviewed the manuscript, performed and provided critical insight in data analysis. VL: Performed statistical analysis, revised the manuscript and interpreted the results, provided clinical insight. FK: Co-designed the study, performed biopsies of the investigated tissues during surgery, provided clinical insight. EH: critical review. MM: Performed slides evaluation in case of discrepancy, proof-read the manuscript, provided clinical insight, GSB: Co-designed the study and interpreted the results, performed slides evaluation in case of discrepancy, critical proof-reading of the manuscript, provided critical insight, gave final permission for submission.


Not applicable

Competing Interests

The authors have declared that no competing interests exist.

Additional Materials

List of Abbreviations


Bioelectrical Impedance Analysis


Body-Mass Index


Extramyocellular Adipose Tissue


Formalin Fixed, Paraffin Embedded




Hepatocellular Carcinoma


High Density Lipoprotein


Intra-Muscular Fat


Intra-Myocellular Lipids


Intra-Myocellular Lipids


Interquartile Range


Low Density Lipoprotein


Non Alcoholic Fatty Liver Disease


Non Alcoholic Steatohepatitis


Oral Glucose Tolerance Test


Perimuscular Fat


Quantile-Quantile Plot


Subcutaneous Adipose Tissue


Standard Deviation


Skeletal Muscle


Visceral Adipose Tissue


  1. Barabasi AL. Network medicine--from obesity to the "diseasome". N Engl J Med. 2007; 357: 404-407. [CrossRef]
  2. Lackey DE, Olefsky JM. Regulation of metabolism by the innate immune system. Nat Rev Endocrinol. 2016; 12: 15-28. [CrossRef]
  3. Osborn O, Olefsky JM. The cellular and signaling networks linking the immune system and metabolism in disease. Nat Med. 2012; 18: 363-374. [CrossRef]
  4. Tordjman J, Guerre-Millo M, Clement K. Adipose tissue inflammation and liver pathology in human obesity. Diabetes Metab. 2008; 34: 658-663. [CrossRef]
  5. Tran A, Gual P. Non-alcoholic steatohepatitis in morbidly obese patients.Clin Res Hepatol Gas. 2013; 37: 17-29. [CrossRef]
  6. Gruben N, Shiri-Sverdlov R, Koonen DP, Hofker MH. Nonalcoholic fatty liver disease: A main driver of insulin resistance or a dangerous liaison? Biochim Biophys Acta. 2014; 1842: 2329-2343. [CrossRef]
  7. Vonghia L, Francque S. Cross talk of the immune system in the adipose tissue and the liver in non-alcoholic steatohepatitis: Pathology and beyond. World JHepatol. 2015; 7: 1905-1912. [CrossRef]
  8. Magee N, Zou A, Zhang Y. Pathogenesis of Nonalcoholic Steatohepatitis: Interactions between Liver Parenchymal and Nonparenchymal Cells. Biomed Res Int. 2016; 2016: 5170402. [CrossRef]
  9. Beck IM, Vanden Berghe W, Vermeulen L, Yamamoto KR, Haegeman G, De Bosscher K. Crosstalk in inflammation: The interplay of glucocorticoid receptor-based mechanisms and kinases and phosphatases. Endocr Rev. 2009; 30: 830-882. [CrossRef]
  10. Balampanis K, Chasapi A, Kourea E, Tanoglidi A, Hatziagelaki E, Lambadiari V, et al. Inter-Tissue expression patterns of the key metabolic biomarker Pgc1a in severely obese individuals: Implication in obesity-induced disease. HellJCardiol. 2018; pii: S1109-9666(18)30252-5. [CrossRef]
  11. Chasapi A, Balampanis K, Kourea E, Kalfaretzos F, Lambadiari V, Lambrou GI, et al. Can obesity-induced inflammation in skeletal muscle and intramuscular adipose tissue accurately detect liver fibrosis? J Musculoskelet Neuronal Interact. 2018; 18: 509-524.
  12. Baker RG, Hayden MS, Ghosh S. NF-kappaB, inflammation, and metabolic disease. Cell Metab. 2011; 13: 11-22. [CrossRef]
  13. Oeckinghaus A, Hayden MS, Ghosh S. Crosstalk in NF-kappaB signaling pathways. Nat Immunol. 2011; 12: 695-708. [CrossRef]
  14. Lee BC, Lee J. Cellular and molecular players in adipose tissue inflammation in the development of obesity-induced insulin resistance. Biochim Biophys Acta. 2014; 1842: 446-462. [CrossRef]
  15. Yuan M, Konstantopoulos N, Lee J, Hansen L, Li ZW, Karin M, et al. Reversal of obesity- and diet-induced insulin resistance with salicylates or targeted disruption of Ikkbeta. Science. 2001; 293: 1673-1677. [CrossRef]
  16. Goldfine AB, Fonseca V, Jablonski KA, Pyle L, Staten MA, Shoelson SE. The effects of salsalate on glycemic control in patients with type 2 diabetes: a randomized trial. Ann Intern Med. 2010; 152: 346-357. [CrossRef]
  17. Madonna R, Wu H, Shelat H, Geng YJ. CD1d-associated expression of NF-kB and cardiac dysfunction in diabetic and obese mice. Int J Immunopathol Pharmacol. 2013; 26: 59-73. [CrossRef]
  18. Hayden MS, Ghosh S. Shared principles in NF-kappaB signaling. Cell. 2008; 132: 344-362. [CrossRef]
  19. Beutler BA. TLRs and innate immunity. Blood. 2009; 113: 1399-1407. [CrossRef]
  20. Dranse HJ, Muruganandan S, Fawcett JP, Sinal CJ. Adipocyte-secreted chemerin is processed to a variety of isoforms and influences MMP3 and chemokine secretion through an NFkB-dependent mechanism. Mol Cell Endocrinol. 2016; 436: 114-129. [CrossRef]
  21. Russo MA, Sansone L, Carnevale I, Limana F, Runci A, Polletta L, et al. One special question to start with: Can HIF/NFkB be a target in inflammation? Endocr Metab Immune Disord Drug Targets. 2015; 15: 171-185. [CrossRef]
  22. Cao H. Adipocytokines in obesity and metabolic disease. J Endocrinol. 2014; 220: T47-59. [CrossRef]
  23. Oita RC, Ferdinando D, Wilson S, Bunce C, Mazzatti DJ. Visfatin induces oxidative stress in differentiated C2C12 myotubes in an Akt- and MAPK-independent, NFkB-dependent manner. Pflugers Arch. 2010; 459: 619-630. [CrossRef]
  24. Cai D, Frantz JD, Tawa NE, Jr., Melendez PA, Oh BC, Lidov HG, et al. IKKbeta/NF-kappaB activation causes severe muscle wasting in mice. Cell. 2004; 119: 285-298. [CrossRef]
  25. Arkan MC, Hevener AL, Greten FR, Maeda S, Li ZW, Long JM, et al. IKK-beta links inflammation to obesity-induced insulin resistance. Nat Med. 2005; 11: 191-198. [CrossRef]
  26. Feng Q, O'Malley BW. Nuclear receptor modulation--role of coregulators in selective estrogen receptor modulator (SERM) actions. Steroids. 2014; 90: 39-43. [CrossRef]
  27. Addison O, Marcus RL, Lastayo PC, Ryan AS. Intermuscular fat: A review of the consequences and causes. Int J Endocrinol. 2014; 2014: 309570. [CrossRef]
  28. Khan IM, Perrard XY, Brunner G, Lui H, Sparks LM, Smith SR, et al. Intermuscular and perimuscular fat expansion in obesity correlates with skeletal muscle T cell and macrophage infiltration and insulin resistance. Int J Obes (Lond). 2015; 39: 1607-1618. [CrossRef]
  29. Kleiner DE, Brunt EM, Van Natta M, Behling C, Contos MJ, Cummings OW, et al. Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology. 2005; 41: 1313-1321. [CrossRef]
  30. Filippidis FT, Tzavara C, Dimitrakaki C, Tountas Y. Compliance with a healthy lifestyle in a representative sample of the Greek population: Preliminary results of the Hellas Health I study. Public Health. 2011; 125: 436-441. [CrossRef]
  31. Kapantais E, Tzotzas T, Ioannidis I, Mortoglou A, Bakatselos S, Kaklamanou M, et al. First national epidemiological survey on the prevalence of obesity and abdominal fat distribution in Greek adults. Ann Nutr Metab. 2006; 50: 330-338. [CrossRef]
  32. Grivas PD, Tzelepi V, Sotiropoulou-Bonikou G, Kefalopoulou Z, Papavassiliou AG, Kalofonos H. Expression of ERalpha, ERbeta and co-regulator PELP1/MNAR in colorectal cancer: Prognostic significance and clinicopathologic correlations. Cell Oncol. 2009; 31: 235-247.
  33. Jornayvaz FR, Shulman GI. Regulation of mitochondrial biogenesis. Essays Biochem. 2010; 47: 69-84. [CrossRef]
  34. Scarpulla RC. Metabolic control of mitochondrial biogenesis through the PGC-1 family regulatory network. Biochim Biophys Acta. 2011; 1813: 1269-1278. [CrossRef]
  35. Cogswell PC, Kashatus DF, Keifer JA, Guttridge DC, Reuther JY, Bristow C, et al. NF-kappa B and I kappa B alpha are found in the mitochondria. Evidence for regulation of mitochondrial gene expression by NF-kappa B. J Biol Chem. 2003; 278: 2963-2968. [CrossRef]
  36. Dahl R, Larsen S, Dohlmann TL, Qvortrup K, Helge JW, Dela F, et al. Three-dimensional reconstruction of the human skeletal muscle mitochondrial network as a tool to assess mitochondrial content and structural organization. Acta Physiol. 2015; 213: 145-155. [CrossRef]
  37. Tang T, Zhang J, Yin J, Staszkiewicz J, Gawronska-Kozak B, Jung DY, et al. Uncoupling of inflammation and insulin resistance by NF-kappaB in transgenic mice through elevated energy expenditure. J Biol Chem. 2010; 285: 4637-4644. [CrossRef]
  38. Strasser F. Appraisal of current and experimental approaches to the treatment of cachexia. Curr Opin Support Pa. 2007; 1: 312-316. [CrossRef]
  39. Ruan H, Pownall HJ, Lodish HF. Troglitazone antagonizes tumor necrosis factor-alpha-induced reprogramming of adipocyte gene expression by inhibiting the transcriptional regulatory functions of NF-kappaB. J Biol Chem. 2003; 278: 28181-28192. [CrossRef]
  40. Rudofsky G, Reismann P, Groener JB, Djuric Z, Fleming T, Metzner C, et al. Identical LDL-cholesterol lowering but non-identical effects on NF-kappaB activity: High dose simvastatin vs combination therapy with ezetimibe. Atherosclerosis. 2012; 223: 190-196. [CrossRef]
  41. Wihastuti TA, Sargowo D. The comparation of activation Nuclear Factor Kappa Beta (NFKB) at rattus novergicus strain wistar induced by various duration High Fat Diet (HFD). Int J Med Health Biomed Bioeng Pharm Eng. 2012; 6: 423-425.
  42. Cominacini L, Anselmi M, Garbin U, Fratta Pasini A, Stranieri C, Fusaro M, et al. Enhanced plasma levels of oxidized low-density lipoprotein increase circulating nuclear factor-kappa B activation in patients with unstable angina. J Am Coll Cardiol. 2005; 46: 799-806. [CrossRef]
  43. Hegarty BD, Furler SM, Ye J, Cooney GJ, Kraegen EW. The role of intramuscular lipid in insulin resistance. Acta Physiol Scand Suppl. 2003; 178: 373-383. [CrossRef]
  44. Dube J, Goodpaster BH. Assessment of intramuscular triglycerides: Contribution to metabolic abnormalities. Curr Opin Clin Nutr Metab Care. 2006; 9: 553-559. [CrossRef]
  45. Green CJ, Pedersen M, Pedersen BK, Scheele C. Elevated NF-kappaB activation is conserved in human myocytes cultured from obese type 2 diabetic patients and attenuated by AMP-activated protein kinase. Diabetes. 2011; 60: 2810-2819. [CrossRef]
  46. Barma P, Bhattacharya S, Bhattacharya A, Kundu R, Dasgupta S, Biswas A, et al. Lipid induced overexpression of NF-kappaB in skeletal muscle cells is linked to insulin resistance. Biochim Biophys Acta. 2009; 1792: 190-200. [CrossRef]
  47. Boden G. Fatty acid-induced inflammation and insulin resistance in skeletal muscle and liver. Curr Diabetes Rep. 2006; 6: 177-181. [CrossRef]
  48. Shoelson SE, Lee J, Yuan M. Inflammation and the IKK beta/I kappa B/NF-kappa B axis in obesity- and diet-induced insulin resistance. Int J Obes Relat Metab Disord. 2003; 27: S49-52. [CrossRef]
  49. Ahn JH, Park TJ, Jin SH, Kang HY. Human melanocytes express functional Toll-like receptor 4. Exp Dermatol. 2008; 17: 412-417. [CrossRef]
  50. Reyna SM, Ghosh S, Tantiwong P, Meka CS, Eagan P, Jenkinson CP, et al. Elevated toll-like receptor 4 expression and signaling in muscle from insulin-resistant subjects. Diabetes. 2008; 57: 2595-2602. [CrossRef]
  51. Videla LA, Tapia G, Rodrigo R, Pettinelli P, Haim D, Santibanez C, et al. Liver NF-kappaB and AP-1 DNA binding in obese patients. Obesity. 2009; 17: 973-979. [CrossRef]
  52. Cai D, Yuan M, Frantz DF, Melendez PA, Hansen L, Lee J, et al. Local and systemic insulin resistance resulting from hepatic activation of IKK-beta and NF-kappaB. Nat Med. 2005; 11: 183-190. [CrossRef]
  53. Marra F. Nuclear factor-kappaB inhibition and non-alcoholic steatohepatitis: inflammation as a target for therapy. Gut. 2008; 57: 570-572. [CrossRef]
  54. Petta S, Amato MC, Di Marco V, Camma C, Pizzolanti G, Barcellona MR, et al. Visceral adiposity index is associated with significant fibrosis in patients with non-alcoholic fatty liver disease. Aliment Pharmacol Ther. 2012; 35: 238-247. [CrossRef]
  55. Locatelli I, Sutti S, Vacchiano M, Bozzola C, Albano E. NF-kappaB1 deficiency stimulates the progression of non-alcoholic steatohepatitis (NASH) in mice by promoting NKT-cell-mediated responses. Clin Sci Mol Med. 2013; 124: 279-287. [CrossRef]
  56. Zhou Z, Chen X, Li F, Tang C. [Effects and mechanisms of the inflammatory reaction related to NASH and induced by activation of the cholinergic anti-inflammatory pathway. Chinese J Hepatol. 2015; 23: 64-68.
  57. Li XH, McGrath KC, Nammi S, Heather AK, Roufogalis BD. Attenuation of liver pro-inflammatory responses by Zingiber officinale via inhibition of NF-kappa B activation in high-fat diet-fed rats. Basic Clin Pharmacol Toxicol. 2012; 110: 238-244. [CrossRef]
  58. Zhao JS, Zhu FS, Liu S, Yang CQ, Chen XM. Pioglitazone ameliorates nonalcoholic steatohepatitis by down-regulating hepatic nuclear factor-kappa B and cyclooxygenases-2 expression in rats. Chin Med J (Engl). 2012; 125: 2316-2321.
  59. Sharma M, Mitnala S, Vishnubhotla RK, Mukherjee R, Reddy DN, Rao PN. The riddle of nonalcoholic fatty liver disease: Progression from nonalcoholic fatty liver to nonalcoholic steatohepatitis. J Clin Exp Hepatol. 2015; 5: 147-158. [CrossRef]
  60. Finkin S, Yuan D, Stein I, Taniguchi K, Weber A, Unger K, et al. Ectopic lymphoid structures function as microniches for tumor progenitor cells in hepatocellular carcinoma. Nat Immunol. 2015; 16: 1235-1244. [CrossRef]
  61. Villanueva A, Luedde T. The transition from inflammation to cancer in the liver. Dig Liver Dis. 2016; 8: 89-93. [CrossRef]
  62. Luedde T, Beraza N, Kotsikoris V, van Loo G, Nenci A, De Vos R, et al. Deletion of NEMO/IKKgamma in liver parenchymal cells causes steatohepatitis and hepatocellular carcinoma. Cancer Cell. 2007; 11: 119-132. [CrossRef]
  63. Tanaka M, Fuentes ME, Yamaguchi K, Durnin MH, Dalrymple SA, Hardy KL, et al. Embryonic lethality, liver degeneration, and impaired NF-kappa B activation in IKK-beta-deficient mice. Immunity. 1999; 10: 421-429. [CrossRef]
  64. Pikarsky E, Porat RM, Stein I, Abramovitch R, Amit S, Kasem S, et al. NF-kappaB functions as a tumour promoter in inflammation-associated cancer. Nature. 2004; 431: 461-466. [CrossRef]
  65. Chaisson ML, Brooling JT, Ladiges W, Tsai S, Fausto N. Hepatocyte-specific inhibition of NF-kappaB leads to apoptosis after TNF treatment, but not after partial hepatectomy. J Clin Invest. 2002; 110: 193-202. [CrossRef]
  66. Liang W, Lindeman JH, Menke AL, Koonen DP, Morrison M, Havekes LM, et al. Metabolically induced liver inflammation leads to NASH and differs from LPS- or IL-1beta-induced chronic inflammation. Lab Invest. 2014; 94: 491-502. [CrossRef]
  67. Luedde T, Schwabe RF. NF-kappaB in the liver--linking injury, fibrosis and hepatocellular carcinoma. Nat Rev Gastroenterol Hepatol. 2011; 8: 108-118. [CrossRef]
  68. Yamashita T, Yao Z, Li F, Zhang Q, Badell IR, Schwarz EM, et al. NF-kappaB p50 and p52 regulate receptor activator of NF-kappaB ligand (RANKL) and tumor necrosis factor-induced osteoclast precursor differentiation by activating c-Fos and NFATc1. J Biol Chem. 2007; 282: 18245-18253. [CrossRef]
  69. Lopez-Rodriguez C, Aramburu J, Jin L, Rakeman AS, Michino M, Rao A. Bridging the NFAT and NF-kappaB families: NFAT5 dimerization regulates cytokine gene transcription in response to osmotic stress. Immunity. 2001; 15: 47-58. [CrossRef]
  70. Yang G, Yu F, Fu H, Lu F, Huang B, Bai L, et al. Identification of the distinct promoters for the two transcripts of apoptosis related protein 3 and their transcriptional regulation by NFAT and NFkappaB. Mol Cell Biochem. 2007; 302: 187-194. [CrossRef]
  71. Cai T, Li X, Ding J, Luo W, Li J, Huang C. A cross-talk between NFAT and NF-kappaB pathways is crucial for nickel-induced COX-2 expression in Beas-2B cells. Curr Cancer Drug Targets. 2011; 11: 548-559. [CrossRef]
  72. Muhammad K, Alrefai H, Marienfeld R, Pham DA, Murti K, Patra AK, et al. NF-kappaB factors control the induction of NFATc1 in B lymphocytes. Eur J Immunol. 2014; 44: 3392-3402. [CrossRef]
  73. Khalaf H, Jass J, Olsson PE. The role of calcium, NF-kappaB and NFAT in the regulation of CXCL8 and IL-6 expression in Jurkat T-cells. Int J Biochem Cell B. 2013; 4: 150-156.
  74. Fisher WG, Yang PC, Medikonduri RK, Jafri MS. NFAT and NFkappaB activation in T lymphocytes: a model of differential activation of gene expression. Ann Biomed Eng. 2006; 34: 1712-1728. [CrossRef]
  75. Ahmad F, Chung YW, Tang Y, Hockman SC, Liu S, Khan Y, et al. Phosphodiesterase 3B (PDE3B) regulates NLRP3 inflammasome in adipose tissue. Sci Rep-UK. 2016; 6: 28056. [CrossRef]
  76. Miao L, Yun X, Tao R, Wang Y, Fan G, Zhu Y, et al. Bakuchiol exhibits anti-metastasis activity through NF-κB cross-talk signaling with AR and ERβ in androgen-independent prostate cancer cells PC-3. J Pharmacol Sci. 2018; 138: 1-8. [CrossRef]
  77. Dey P, Barros RP, Warner M, Strom A, Gustafsson JA. Insight into the mechanisms of action of estrogen receptor beta in the breast, prostate, colon, and CNS. J Mol Endocrinol. 2013; 51: T61-74. [CrossRef]
  78. King AE, Collins F, Klonisch T, Sallenave JM, Critchley HO, Saunders PT. An additive interaction between the NFkappaB and estrogen receptor signalling pathways in human endometrial epithelial cells. Hum Reprod. 2010; 25: 510-518. [CrossRef]
  79. Harnish DC. Estrogen receptor ligands in the control of pathogenic inflammation. Curr Opin Investig Drugs. 2006; 7: 997-1001.
  80. Quaedackers ME, van den Brink CE, van der Saag PT, Tertoolen LG. Direct interaction between estrogen receptor alpha and NF-kappaB in the nucleus of living cells. Mol Cell Endocrinol. 2007; 273: 42-50. [CrossRef]
  81. Evans MJ, Eckert A, Lai K, Adelman SJ, Harnish DC. Reciprocal antagonism between estrogen receptor and NF-kappaB activity in vivo. Circ Res. 2001; 89: 823-830. [CrossRef]
  82. Xing D, Oparil S, Yu H, Gong K, Feng W, Black J, et al. Estrogen modulates NFkappaB signaling by enhancing IkappaBalpha levels and blocking p65 binding at the promoters of inflammatory genes via estrogen receptor-beta. PloS One. 2012; 7: e36890. [CrossRef]
  83. Adamson AD, Friedrichsen S, Semprini S, Harper CV, Mullins JJ, White MR, et al. Human prolactin gene promoter regulation by estrogen: convergence with tumor necrosis factor-alpha signaling. Endocrinology. 2008; 149: 687-694. [CrossRef]
  84. Frasor J, Weaver AE, Pradhan M, Mehta K. Synergistic up-regulation of prostaglandin E synthase expression in breast cancer cells by 17beta-estradiol and proinflammatory cytokines. Endocrinology. 2008; 149: 6272-6279. [CrossRef]
  85. Alvarez-Guardia D, Palomer X, Coll T, Davidson MM, Chan TO, Feldman AM, et al. The p65 subunit of NF-kappaB binds to PGC-1alpha, linking inflammation and metabolic disturbances in cardiac cells. Cardiovasc Res. 2010; 87: 449-458. [CrossRef]
  86. Bakkar N, Ladner K, Canan BD, Liyanarachchi S, Bal NC, Pant M, et al. IKKalpha and alternative NF-kappaB regulate PGC-1beta to promote oxidative muscle metabolism. J Cell Sci. 2012; 196: 497-511. [CrossRef]
  87. Eisele PS, Salatino S, Sobek J, Hottiger MO, Handschin C. The peroxisome proliferator-activated receptor gamma coactivator 1alpha/beta (PGC-1) coactivators repress the transcriptional activity of NF-kappaB in skeletal muscle cells. J Biol Chem. 2013; 288: 2246-2260. [CrossRef]
  88. Eisele PS, Furrer R, Beer M, Handschin C. The PGC-1 coactivators promote an anti-inflammatory environment in skeletal muscle in vivo. Biochem Biophys Res Commun. 2015; 464: 692-697. [CrossRef]
  89. Erkan G, Yilmaz G, Konca Degertekin C, Akyol G, Ozenirler S. Presence and extent of estrogen receptor-alpha expression in patients with simple steatosis and NASH. Pathol Res Pract. 2013; 209: 429-432. [CrossRef]
  90. Wiik A, Glenmark B, Ekman M, Esbjornsson-Liljedahl M, Johansson O, Bodin K, et al. Oestrogen receptor beta is expressed in adult human skeletal muscle both at the mRNA and protein level. Acta Physiol Scand Suppl. 2003; 179: 381-387. [CrossRef]
  91. Thul PJ, Akesson L, Wiking M, Mahdessian D, Geladaki A, Ait Blal H, et al. A subcellular map of the human proteome. Science. 2017; 356: eaal3321. [CrossRef]
  92. Uhlen M, Fagerberg L, Hallstrom BM, Lindskog C, Oksvold P, Mardinoglu A, et al. Proteomics. Tissue-based map of the human proteome. Science. 2015; 347: 1260419. [CrossRef]
  93. Xie J, Yang Y, Li J, Hou J, Xia K, Song W, et al. Expression of tmp21 in normal adult human tissues. Int J Clin Exp Med. 2014; 7: 2976-2983.
  94. Crandall DL, Busler DE, Novak TJ, Weber RV, Kral JG. Identification of estrogen receptor beta RNA in human breast and abdominal subcutaneous adipose tissue. Biochem Biophys Res Commun. 1998; 248: 523-526. [CrossRef]
Download PDF