- Email: [email protected]
The role of alkaline phosphatase in intracellular lipid accumulation in the human hepatocarcinoma cell line, HepG2
Experimental and Molecular Pathology 102 (2017) 224–229
Contents lists available at ScienceDirect
Experimental and Molecular Pathology journal homepage: www.elsevier.com/locate/yexmp
The role of alkaline phosphatase in intracellular lipid accumulation in the human hepatocarcinoma cell line, HepG2 George M Chirambo a,b, Chantal van Niekerk a,1, Nigel J Crowther a,⁎ a b
Department of Chemical Pathology, National Health Laboratory Service, University of Witwatersrand Medical School, Johannesburg, South Africa Department of Biochemistry, College Of Medicine, University of Malawi, Blantyre, Malawi
a r t i c l e
i n f o
Article history: Received 12 October 2016 Accepted 11 February 2017 Available online 13 February 2017 Keywords: Alkaline phosphatase HepG2 Lipid droplet PPARgamma Adipogenesis
a b s t r a c t Inhibition of tissue non-specific alkaline phosphatase (TNALP) decreases intracellular lipid accumulation in human preadipocytes and the murine preadipocyte cell line, 3T3-L1. Therefore, the current study was performed to determine if TNALP is required for intracellular lipid deposition in the human hepatocyte cell line, HepG2. Intracellular lipid accumulation, TNALP activity and peroxisome proliferator activated receptor (PPAR) γ gene expression were measured in HepG2 and 3T3-L1 cells in the presence and absence of the TNALP inhibitors levamisole and histidine. Sub-cellular TNALP activity was localized using cytochemical analysis. Both PPARγ gene expression and TNALP activity increased during intracellular lipid accumulation in HepG2 and 3T3-L1 cells. Inhibition of TNALP blocked intracellular lipid accumulation but did not alter expression of the PPARγ gene. In HepG2 cells, TNALP co-localized with adipophilin on the lipid droplet membrane. These data suggest a role for TNALP in lipid droplet formation, possibly downstream from PPARγ, within HepG2 and 3T3-L1 cells. © 2017 Elsevier Inc. All rights reserved.
1. Introduction Alkaline phosphatases (ALPs) are a group of membrane-bound glycoproteins that hydrolyze a broad range of monophosphate esters at alkaline pH optima (Price, 1993). There are four ALP isoenzymes termed tissue non-specific, intestinal, placental and germ cell, each coded by a separate gene (Moss, 1992). Tissue non-specific alkaline phosphatase (TNALP) comprises of three different isoforms called liver, bone and kidney ALP. These enzymes are coded by the same gene but differ in terms of their pattern of glycosylation (Nosjean et al., 1997). Tissue non-specific alkaline phosphatase has a broad tissue distribution however, its precise function and mode of action are not fully understood. Alkaline phosphatase activity is found in human blood and is a commonly measured analyte in clinical chemistry laboratories. Obstruction of the bile duct is one of the main causes of elevated ALP levels in human serum. Also, increased levels of ALP may be observed in serum during active bone formation (Neuschhwander-Terti, 1995; Wiwanitkit, 2001). Therefore, the principal clinical and diagnostic
⁎ Corresponding author at: Department of Chemical Pathology, National Health Laboratory Service, University of Witwatersrand Medical School, 7 York Rd., Parktown 2193, Johannesburg, South Africa. E-mail addresses: [email protected] (G.M. Chirambo), [email protected] (C. van Niekerk), [email protected] (N.J. Crowther). 1 Current address: Department of Chemical Pathology, National Health Laboratory Service, University of Pretoria, Pretoria, South Africa.
http://dx.doi.org/10.1016/j.yexmp.2017.02.007 0014-4800/© 2017 Elsevier Inc. All rights reserved.
reasons for measuring blood ALP concentration is to confirm the presence of bone or liver pathology. Studies have shown that TNALP is expressed in human preadipocytes (Ali et al., 2006a) and in the murine preadipocyte cell lines, 3T3-L1 (Ali et al., 2005) and 3T3-F442A (Hernández-Mosqueira et al., 2015). The 3T3-L1 cell line has proven to be an excellent model for studying preadipocyte differentiation and maturation (Ntambi and Young-Cheul, 2000). The proliferation and determination of preadipocytes is associated with the specific activation of a number of transcriptional factors and an upward regulation of adipogenic genes (Cornelius et al., 1994). The hallmark of adipocyte maturation is intracellular accumulation of lipid within membrane bound lipid droplets (Gregoire et al., 1998), with the peroxisome proliferator activated receptor (PPAR) γ transcription factor acting as a prime regulator of adipogenesis (Spiegelman, 1998). A role for TNALP in intracellular lipid accumulation in 3T3-L1 cells and human preadipocytes has been proven using inhibitors of TNALP (Ali et al., 2006a; Ali et al., 2006b; Ali et al., 2005), and studies have shown that this enzyme is localized to the membrane surrounding the lipid droplet (Ali et al., 2006a; Ali et al., 2005). However, the exact role of TNALP in adipogenesis is not yet understood. It is known that TNALP is expressed in hepatocytes (Price, 1993), and that these cells are also able to accumulate intracellular lipid within membranebound droplets. These organelles express a number of proteins on their surfaces that are common to both adipocytes and hepatocytes (Okumura, 2011). Excessive acquisition of lipid within hepatocytes is
G.M. Chirambo et al. / Experimental and Molecular Pathology 102 (2017) 224–229
observed in alcoholic and non-alcoholic steatohepatitis (NASH) with the latter now being one of the most common liver pathologies observed in developed countries, a probable consequence of the high prevalence of obesity (Preiss and Sattar, 2008). The presence of TNALP in preadipocytes and hepatocytes, the ability of both cell types to accumulate lipid within membrane-bound lipid droplets, and the evidence that TNALP is involved in intracellular lipid storage within preadipocytes forms the basis of our hypothesis that TNALP plays a role in lipid accumulation within hepatocytes. This hypothesis was tested by inducing lipid accumulation in the human hepatoma cell line, HepG2 and 3T3-L1 cells in the presence and absence of the TNALP inhibitors, levamisole and histidine. The expression of the adipogenic regulator, PPARγ was also assessed and the sub-cellular localization of TNALP was analysed via cytochemical analysis. 2. Materials and methods 2.1. Culture of HepG2 and 3T3-L1 cells All tissue culture reagents were obtained from BioWhittaker (Walkersville, MD, USA) and GibCo Invitrogen (Paisley, Scotland) unless otherwise stated. HepG2 and 3T3-L1 cells were obtained from the American Type Culture Collection (ATCC). HepG2 cells were grown in Earle's minimum essential medium (EMEM) supplemented with 10% fetal bovine serum, 100 U/ml penicillin, 100 μg/ml streptomycin, 2 mmol/l glutamine, non-essential amino acids, 1 mmol/l sodium pyruvate (maintenance medium) and oleic acid coupled to albumin (Sigma-Aldrich, St Louis, MO, USA) at a final concentration of 400 μmol/l of oleic acid, to induce intracellular lipid accumulation (Brasaemle et al., 1997). The maintenance medium was replaced every third day until the cells were ready for experiments. The 3T3-L1 cells were maintained in Dulbeco's modified eagle's medium (DMEM) supplemented with 10% fetal bovine serum, 100 U/ml penicillin, 100 μg/ml streptomycin and 2 mmol/l glutamine. When the cells reached confluence, 1.70 μmol/l insulin (Novo-Nordisk. Denmark), 0.50 mmol/l 3-isobutyl-1-methylxanthine (IBMX, Sigma-Aldrich, St Louis, MO, USA) and 0.22 μmol/l dexamethasone (Sigma-Aldrich, St Louis, MO, USA) were added to the culture medium to initiate adipogenesis. This transformation medium was removed after 3 days and replaced with similar medium containing insulin only. Following a further 3 days of culture this medium was replaced with the original maintenance medium which was changed every 3 days. 2.2. Use of TNALP inhibitors In HepG2 cells levamisole was used at a final concentration of 3.0 mmol/l whilst histidine was used at a concentration of 75.0 mmol/l. In 3T3-L1 cells levamisole was used at final concentration of 2.0 mmol/l whilst histidine was used at a concentration of 50.0 mmol/l. The concentrations of the TNALP inhibitors were derived from the literature for the 3T3-L1 cells (Ali et al., 2005) whilst dose response curves were performed for the HepG2 cells based on the concentrations used for the preadipocyte cell line. The final concentrations chosen were those that gave maximal inhibition. The inhibitors were added to the culture medium simultaneous with the induction of lipid droplet formation. 2.3. Measurements of intracellular lipid accumulation, TNALP activity and total protein Intracellular lipid accumulation was measured at baseline (day 0) and at 4, 7 and 11 days after the induction of intracellular lipid formation in HepG2 and 3T3-L1 cells using the lipid-specific dye Oil Red O (Laughton, 1986). TNALP activity was determined at the same time points as the cellular lipid levels using a previously published method
225
(Ali et al., 2005). The cellular protein content was analysed using the Bradford method (Bradford, 1976).
2.4. PPARγ gene expression in HepG2 and 3T3-L1 cells Total RNA was isolated from cell cultures on day 0, 4 and 7 post-induction of lipid droplet formation using an RNeasy Mini kit (Qiagen, Hilden, Germany). Synthesis of cDNA from the total RNA was then carried out using an oligo dT primer (Inqaba Biotech, Pretoria, South Africa), and Moloney murine leukemia virus reverse transcriptase (Promega, Madison WI, USA). Quantitative real-time PCR was then performed with the TATA-box binding protein (TBP) gene used as an endogenous internal control. Primers were designed using the GeneRunner Software (Hasting Software Inc., Las Vegas, USA) and were synthesized by Inqaba Biotech (Pretoria, South Africa). The TBP primer sequences for the HepG2 cells were: forward, 5′-CAG TGA CCC AGC AGC ATC -3′ and reverse, 5′-GTC AGT CCA GTG CCA TAA GG-3′, producing an amplicon of 277 bp. The sequences for the 3T3-L1 TBP primers were: forward, 5′-ACC CTT CAC CAA TGA CTC CTA TG -3′ and reverse, 5′-ATG ATG ACT GCA GCA AAT CGC-3′, with an amplicon length of 189 bp. The primer sequences of the PPARγ gene for the HepG2 cells were as follows: forward, 5′-GGT TGA CAC AGA GAT GCC A-3′ and reverse, 5′-CAA AGG AGT GGG AGT GGT C-3′, producing an amplicon of 88 bp. The sequences for the 3T3-L1 PPARγ primers were: forward, 5′CCA GAG CAT GGT GCC TTC GCT-3′ and reverse, 5′-CAG CAA CCA TTG GGT CAG CTC-3′ producing an amplicon of 240 bp. Real-time quantitative PCR of the PPARγ and TBP cDNA was carried out using a Rotorgene 6000 (Qiagen, Hilden, Germany) thermocycler and a SensiMix SYBR kit (Quantance Ltd., London, UK). The amplification conditions for the PPARγ and TBP gene products from the HepG2 cells comprised of an annealing temperature of 57 °C and an elongation time of 20 s, whilst for the 3T3-L1 cells, the annealing temperature was 60 °C, with the same annealing time as for the HepG2 cells. A 5-point standard curve was set up using cDNA isolated from the respective cell line and ranging in concentration from 14 to 272 ng/μl. The cDNA concentration was measured using a NanoDrop 1000 spectrophotometer (Thermo Scientific, Wilmington, DE, USA).
2.5. Sub-cellular localization of TNALP The HepG2 and 3T3-L1 cells were cultured on cover slips coated in gelatin (BDH, Poole, UK), and induced to accumulate lipid using the methods described in Section 2.1. Cells were harvested between days 4–8 of culture, at which point they were fixed using a 3% >paraformaldehyde (Sigma-Aldrich, St Louis, MO, USA) solution and permeabilized with a 0.01% saponin (Sigma-Aldrich, St Louis, MO, USA) solution. The 3T3-L1 cells were incubated with a rabbit antiperilipin antibody (Cell Signaling Technology Beverly, MA, USA) whilst the HepG2 cells were incubated with a guinea pig antiadipophilin antibody (Progen Biotechnik, Heidelberg, Germany), overnight at 4 °C. The 3T3-L1 cells were incubated with a goat antirabbit immunoglobulin G antibody labeled with Alexa Fluor 594, and the HepG2 cells were incubated with a goat anti-guinea pig immunoglobulin G antibody labeled with Alex Fluor 488 (both secondary antibodies supplied by Life Technologies, Grand Island, NY, USA). The incubation period was 1 h, in the dark at room temperature. Both cell types were then stained for TNALP activity using a synthetic substrate, ELF97 (Invitrogen Molecular Probes, Leiden, Holland) that gives rise to a fluorescent yellow/green end product that precipitates at the site of TNALP activity. The incubation with ELF97 lasted for 15 min, at room temperature. The cover slips were then mounted onto a microscope slide and visualized at a magnification of 100 using an Olympus 1 × 71 microscope fitted with an Olympus XM10 camera (Olympus, Hamburg, Germany).
226
G.M. Chirambo et al. / Experimental and Molecular Pathology 102 (2017) 224–229
2.6. Statistical analysis All experiments were repeated 3 times unless otherwise stated. Mean and standard error of the mean (SEM) were calculated for each of the variables under study. The Students paired t-test was used to compare mean values at each time point with the baseline value and to compare the mean values at each time point, obtained in the presence of TNALP inhibitors, to the value obtained at the same time point in the absence of inhibitors. An ANOVA was used for comparing the areas-under-the-curve (AUCs; calculated using the trapezoid rule) for TNALP activity, lipid accumulation and PPARγ gene expression across the 3 different treatment groups.
The data in Fig. 1B demonstrates that in the 3T3-L1 cells that were not treated with TNALP inhibitors, TNALP activity rose to a peak on day 7, and then fell back to a near-baseline level by day 11, with values at all 3 time points being significantly higher than the level at baseline (P b 0.05 for all). In cells treated with levamisole or histidine, TNALP levels stayed close to baseline values with levels being significantly lower than those in untreated cells at all time points (P b 0.05 for all) for the histidine treated cells, and at day 4 (P b 0.005) for the levamisole treated cells. The AUC values were 24.3 ± 1.77, 4.68 ± 1.19 and 1.76 ± 1.39 for non-treated, levamisole treated and histidine treated cells, respectively, with the values for the non-treated cells being significantly different from both treated groups (P b 0.0001 for both comparisons).
3. Results 3.2. Effect of levamisole and histidine on intracellular lipid accumulation 3.1. Effect of levamisole and histidine on TNALP activity In the HepG2 cells, in the absence of TNALP inhibitors, the activity of the enzyme significantly rose from day 0 to day 4 (P b 0.05) and then fell almost to the baseline level by day 11 (Fig. 1A). This trend was repeated in cells treated with levamisole but at days 4, 7 and 11, the ALP activity was lower than that observed in the untreated cells. The cells treated with histidine had TNALP activity levels that were lower than those observed in the untreated and levamisole-treated cells at all time points, with the level on day 4 being statistically significantly lower than in the untreated cells (P b 0.05). The AUC was derived from the data in Fig. 1A and the results were as follows (values are mean ± SEM): non-treated, 43.0 ± 5.17; levamisole treated, 26.1 ± 0.52; histidine treated, 7.73 ± 1.16. All values were significantly different from each other (P b 0.01 for all comparisons).
Fig. 1. Intracellular TNALP levels during adipogenesis in HepG2 (A) and 3T3-L1 (B) cells. Each marker and bar represents the mean and SEM respectively, of 3 experiments; the solid line represents incubations performed in the absence of TNALP inhibitors, the long dashes represent incubations performed in the presence of levamisole, and the short dashes represent incubations performed in the presence of histidine; *P b 0.05 versus day 0; ##P b 0.005 versus levamisole; + P b 0.05 versus histidine.
The data in Fig. 2A shows that in non-treated HepG2 cells, intracellular lipid levels rose from day 0 to day 11, with values being significantly higher (P b 0.05) than baseline levels at days 7 and 11. Lipid levels also rose in the levamisole treated cells but the values remained below those for the non-treated cells, being significantly lower at day 11 (P b 0.05). The histidine treated HepG2 cells had the lowest intracellular lipid levels, being greater than the baseline values only at day 4 (P b 0.05) and being lower than non-treated cells at days 7 and 11 (P b 0.05 for both). The AUCs derived from Fig. 2A were as follows: 12.6 ± 1.7, 8.81 ± 1.06 and 6.70 ± 0.71 for non-treated, levamisole treated and histidine treated, respectively. The AUC levels were significantly higher in
Fig. 2. Intracellular lipid accumulation during adipogenesis in HepG2 (A) and 3T3-L1 (B) cells. Each marker and bar represents the mean and SEM respectively, of 3 experiments; the solid line represents incubations performed in the absence of TNALP inhibitors, the long dashes represent incubations performed in the presence of levamisole, and the short dashes represents incubations performed in the presence of histidine; *P b 0.05, **P b 0.005 versus day 0; #P b 0.05, ##P b 0.005 versus levamisole; + P b 0.05 versus histidine.
G.M. Chirambo et al. / Experimental and Molecular Pathology 102 (2017) 224–229
the non-treated compared to the histidine treated cells (P b 0.05), and also tended to be higher in the former group when compared to the levamisole treated cells (P = 0.06). Similar trends to those in the HepG2 cells were observed for intracellular lipid levels in the 3T3-L1 cells (Fig. 2B). Thus, lipid levels were higher in non-treated cells on days 7 (P b 0.05) and 11 (P b 0.005) when compared to baseline and when compared to the histidine treated cells (P b 0.05 for both days). On day 4, lipid levels were higher in non-treated compared to levamisole treated cells (P b 0.05). The AUC values for lipid accumulation in the 3T3-L1 cells were 10.9 ± 0.88 for untreated cells, 6.47 ± 1.60 for levamisole treated cells and 4.30 ± 1.25 for histidine treated cells, with values being significantly higher for both groups of TNALP inhibitor treated cells when compared to the control group (P b 0.05 for both).
3.3. PPARγ gene expression during intracellular lipid accumulation Fig. 3A shows that PPARγ gene expression in HepG2 cells rose to a peak on day 4, at which point the level was significantly higher in all the groups of cells when compared to baseline levels (P b 0.05 for all comparisons). A similar trend was observed in the 3T3-L1 cells with statistically significant differences between day 4 and baseline levels observed in the non-treated cells (Fig. 3B), with a close-to-significant effect seen in the levamisole treated cells (P = 0.053). There were no significant differences in AUC levels for PPARγ gene expression across the treatment groups for either the HepG2 or the 3T3-L1 cells (data not shown).
Fig. 3. PPARγ gene expression during adipogenesis in HepG2 (A) and 3T3-L1 (B) cells. Each marker and bar represents the mean and SEM respectively, of 3 experiments; the solid line represents incubations performed in the absence of TNALP inhibitors, the long dashes represent incubations performed in the presence of levamisole, and the short dashes represent incubations performed in the presence of histidine; *P b 0.05 versus day 0.
227
3.4. Sub-cellular localization of TNALP activity in HepG2 cells The photomicrographs in Fig. 4 demonstrate that the staining for perilipin in 3T3-L1 cells and adipophilin in HepG2 cells co-localises with TNALP activity.
4. Discussion The current study is the first to demonstrate that TNALP activity increases within a hepatocyte cell line during intracellular lipid accretion. It is also the first study to show that inhibition of TNALP activity blocks cytoplasmic lipid accumulation in these cells. Previous studies have shown that TNALP inhibition blocks intracellular lipid accumulation in human preadipocytes (Ali et al., 2006a) and in the murine preadipocyte cell line, 3T3-L1 (Ali et al., 2005). Furthermore, TNALP activity has been identified in lipid-storing cell types from a number of different species i.e. rat adipose tissue (Wallach and Ko, 1964), adipocyte precursor cells in human bone marrow (Bianco et al., 1988), rabbit preadipocytes (Lecoeur and Ouhayoun, 1997), 3T3-L1 (Ali et al., 2005) and 3T3-F442A (Hernández-Mosqueira et al., 2015) murine preadipocyte cell lines, human preadipocytes (Ali et al., 2006a), murine bone marrow preadipocytes (Kodama et al., 1983) and adipocytes from chicken bone marrow (Yoshida and Yumoto, 1987). Also, TNALP is expressed in lactating rat mammary gland epithelial cells (Leung et al., 1989) where it is located on the membrane enclosing the cytoplasmic lipid droplets that are secreted and form the fat fraction of milk. TNALP has also been observed on the membrane of the milk fat globules present in bovine milk (Bingham and Malin, 1992). Within human preadipocytes (Ali et al., 2006a) and murine 3T3-L1 preadipocyte cells (Ali et al., 2005), TNALP is also found on the membrane surrounding the intracellular lipid droplets, and this sub-cellular location was also confirmed for TNALP in HepG2 cells in the present study. These data suggest that TNALP expression is a characteristic of lipid storing cell types. Furthermore, its location on the lipid droplet membrane and the ability of TNALP inhibitors to block intracellular lipid accumulation suggest that it is intrinsically involved in the control of cellular lipid storage. This function of TNALP is further confirmed by the lack of fat tissue observed in mice harbouring a TNALP gene-knockout (Narisawa et al., 1997) and by a study demonstrating that polymorphisms in the TNALP gene are associated with body fat distribution in humans (Korostishevsky et al., 2010). The role of TNALP within the many cell types in which it is expressed, with the exception of osteoblasts, is not known. However, the current study now shows that in lipid storing cells, TNALP acts to control intracellular lipid accumulation. The TNALP inhibitors used in this study do have non-TNALP mediated sub-cellular effects (Hsu, 1980; Basi et al., 1994; Rae et al., 2003). It is therefore possible that the inhibition of cellular lipid accumulation observed in this study is related to non-TNALP effects of levamisole and histidine. However, the use of small interfering (si) RNAs specific for TNALP in 3T3-L1 and HepG2 cells recapitulated the effects seen in the present study (Chirambo, 2012), strongly suggesting that the TNALP inhibitors used in the current investigation are working through TNALP. Furthermore, the localization of TNALP to lipid droplets (Ali et al., 2005; Ali et al., 2006a; Leung et al., 1989) and its presence in a number of different lipid-storing cell types (Ali et al., 2005; Ali et al., 2006a; Bianco et al., 1988; Hernández-Mosqueira et al., 2015; Kodama et al., 1983; Lecoeur and Ouhayoun, 1997; Leung et al., 1989; Wallach and Ko, 1964; Yoshida and Yumoto, 1987) further suggest that the most likely intermediary for the effects of these TNALP inhibitors is TNALP itself. The TNALP activity levels peak in the 3T3-L1 cells after 7 days but peak after 4 days in the HepG2 cells. The reasons for this are not known however, differences in the time course of gene expression should not be unexpected since these are 2 very different cell types. Thus, the HepG2 cell is representative of a mature human hepatocyte whilst the 3T3-L1 cell line is representative of an immature murine
228
G.M. Chirambo et al. / Experimental and Molecular Pathology 102 (2017) 224–229
Fig. 4. Cytochemical analysis of HepG2 and 3T3-L1 cells for intra-cellular localization of TNALP. In HepG2 cells, ALP (A) and adipophilin (B) staining were co-localized, thus suggesting that TNALP is present on the lipid droplet membrane. In the 3T3-L1 cells, TNALP (D) and perilipin (E) staining were also found to co-localize, once again demonstrating that TNALP is situated on the lipid droplet membrane. Plates C and F show phase contrast pictures of HepG2 and 3T3-L1 cells respectively, taken under transmitted light. Magnification was ×100 for all images.
adipocyte. Furthermore, each cell type has very different functions with lipid storage being the prime purpose of adipocytes but not hepatocytes. In both cell lines intracellular lipid accumulation was still increasing on day 11 of culture despite falling levels of TNALP activity (see Figs. 1 and 2). Furthermore, treatment of both HepG2 and 3T3-L1 cells with TNALP inhibitors had no effect on PPARγ expression levels (Fig. 3). These data suggest that TNALP is required for the initiation of lipid droplet formation rather than for maintenance, and that it is not involved in the cascade of transcription factor expression that activates the process of adipogenesis (Cornelius et al., 1994), of which PPARγ is a major component (Spiegelman, 1998). Thus, TNALP is not a modulator of adipogenesis but is rather an inducer of cytoplasmic lipid storage. Our data suggests that TNALP may play a role in intracellular lipid accumulation in human hepatocytes. Unfortunately, no immunohistochemical studies have analysed whether TNALP is present within lipid-laden hepatocytes in tissue sections taken from the livers of individuals with NASH. Such investigations must be undertaken to confirm our hypothesis. However, it is interesting to note that studies have demonstrated elevated serum levels of alkaline phosphatase in subjects with histologically-proven NASH (Neuschwander-Tetri et al., 2010; Pirvulescu et al., 2012), and in subjects with NASH-related liver fibrosis (Singh et al., 2008; Kocbay et al., 2011; Bambha et al., 2012). The HepG2 cell line contains a mutation in the gene encoding β-catenin that leads to increased expression of the protein within the nucleus and higher activity of the Wnt/β-catenin signalling pathway (de La Coste et al., 1998). This is a characteristic feature of human liver tumours (de La Coste et al., 1998; Takigawa and Brown, 2008). The Wnt/β-catenin signalling pathway is known to suppress adipogenesis in mesenchymal stem cells (Kang et al., 2007) and may also block intracellular lipid accumulation in hepatocytes (Behari et al., 2010). These data suggest that the presence of the activating mutation in the β-catenin gene in HepG2 cells cannot explain the lipogenic activity of TNALP in the current study. This is further confirmed by the ability of TNALP to induce lipid accumulation in primary human preadipocytes (Ali et al., 2006a) and within the 3T3-L1 cell line (Ali et al., 2006b), which exhibits normal expression of the Wnt/β-catenin signalling pathway, which is suppressed during adipogenesis (Bennett et al., 2002). The ability of inhibitors of TNALP to block lipid accumulation within a human hepatocyte cell line suggests that this may represent a
therapeutic modality for steatohepatitis. However, TNALP is expressed in many different tissues and it has functional activity within the liver and therefore inhibition of its activity, even if confined to hepatic tissue, may have deleterious side effects. This may be averted by undertaking further studies on the precise role of TNALP in intracellular lipid accumulation. This would allow identification of the molecular events downstream of TNALP that are specific to cellular lipid storage and which may represent valid targets for future therapies for steatohepatitis. The aim of the current study was not to determine the function of TNALP in intracellular lipid accumulation. However, it is possible to hypothesise what the role of TNALP may be within this sub-cellular process. It is known that TNALP plays an important role in bone mineralization where it works to convert inorganic pyrophosphate, a potent inhibitor of the mineralization process, to phosphate, a known activator of mineralization. Within osteoblasts, pyrophosphate is generated by the enzyme ectonucleotide pyrophosphatase/phosphodiesterase (ENPP) 1. Thus, the level of bone mineralization is determined by the balance between TNALP and ENPP1 activity (Hessle et al., 2002). It is noteworthy that ENPP1 expression has been detected in 3T3-L1 cells, where over expression reduces the level of adipogenesis (Liang et al., 2007). This enzyme is also expressed in liver tissue (Zhou et al., 2009) and a positive association between adipocyte expression of ENPP1 and hepatic triglyceride levels has been demonstrated in humans (Chandalia et al., 2012). Furthermore, it has been shown that ENPP1 and TNALP gene polymorphisms associate with anthropometric markers of obesity (Korostishevsky et al., 2010). It is therefore tempting to suggest that TNALP and ENPP1 may function in adipocytes to control intracellular levels of pyrophosphate and phosphate. However, it is not known whether pyrophosphate (or phosphate) has any effect on intracellular lipid accumulation, although it is known that pyrophosphate is an inhibitor of lipoprotein lipase (Krauss et al., 1973), a key enzyme for the uptake of free fatty acids into adipocytes. In summary, the present study has shown that intracellular lipid accumulation and TNALP activity are closely linked in both HepG2 and 3T3-L1 cells, suggesting that TNALP expression maybe a common requirement for cells that accumulate cytoplasmic lipid droplets. The localization of TNALP to the surface of the lipid droplet membrane in both cell lines, as well as in the epithelial cells of lactating rat mammary
G.M. Chirambo et al. / Experimental and Molecular Pathology 102 (2017) 224–229
glands (Leung et al., 1989), also points to the involvement of TNALP in the sub-cellular processes related to intracellular lipid accumulation. Further studies are required to determine the exact role of TNALP in cytoplasmic lipid storage but should initially focus on the control of intracellular inorganic pyrophosphate and phosphate levels. Disclosure None of the authors has any conflict of interest related to this manuscript. Acknowledgements The authors thank members of the routine laboratory service of the department of Chemical pathology, NHLS, Charlotte Maxeke Hospital (Johannesburg) for performing the TNALP activity assay. Funding: The study was supported by financial assistance from the NHLS Research Trust, the National Research Foundation of South Africa and the College of Medicine (Malawi). References Ali, A.T., Penny, C.B., Paiker, J.E., van Niekerk, C., Smit, A.T., Crowther, N.J., 2005. Alkaline phosphatase is involved in the control of adipogenesis in the murine preadipoctyte cell line, 3T3-L1. Clin. Chim. Acta 351, 101–109. Ali, A.T., Penny, C.B., Paiker, J.E., Psaras, G., Ikram, F., Crowther, N.J., 2006a. The effect of alkaline phosphatase inhibitors on intracellular lipid accumulation in preadipocytes isolated from human mammary tissue. Ann. Clin. Biochem. 43, 207–213. Ali, A.T., Penny, C.B., Paiker, J.E., Psaras, G., Ikram, F., Crowther, N.J., 2006b. The relationship between alkaline phosphatase activity and intracellular lipid accumulation in murine 3T3-L1 cells and human preadipocytes. Anal. Biochem. 354, 247–254. Bambha, K., Belt, P., Abraham, M., Wilson, L.A., Pabst, M., Ferrell, L., et al., 2012. Nonalcoholic steatohepatitis clinical research network research group. Ethnicity and nonalcoholic fatty liver disease. Hepatology 55, 769–780. Basi, N., George, M., Pointer, R., 1994. Regulation of glycogen synthase activity in isolated rat adipocytes by levamisole. Life Sci. 54, 1027–1034. Behari, J., Yeh, T.H., Krauland, L., Otruba, W., Cieply, B., Hauth, B., et al., 2010. Liver-specific beta-catenin knockout mice exhibit defective bile acid and cholesterol homeostasis and increased susceptibility to diet-induced steatohepatitis. Am. J. Pathol. 176, 744–753. Bennett, C.N., Ross, S.E., Longo, K.A., Bajnok, L., Hemati, N., Johnson, K.W., et al., 2002. Regulation of Wnt signaling during adipogenesis. J. Biol. Chem. 277, 30998–31004. Bianco, P., Costantini, M., Dearden, L.C., Bonucci, E., 1988. Alkaline phosphatase positive precursors of adipocytes in the human bone marrow. Br. J. Haematol. 68, 401–403. Bingham, E.W., Malin, E.L., 1992. Alkaline phosphatase in the lactating bovine mammary gland and the milk fat globule membrane. Release by phosphatidylinositol-specific phospholipase C. Comp. Biochem. Physiol. B 102, 213–218. Bradford, M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilising the principle of protein-dye binding. Anal. Biochem. 72, 248–254. Brasaemle, D.L., Barber, T., Kimmel, A.R., Londos, C., 1997. Post-translational regulation of perilipin expression. Stabilization by stored intracellular neutral lipids. J. Biol. Chem. 272, 9378–9387. Chandalia, M., Davila, H., Pan, W., Szuszkiewicz, M., Tuvdendorj, D., Livingston, E.H., Abate, N., 2012. Adipose tissue dysfunction in humans: a potential role for the transmembrane protein ENPP1. J. Clin. Endocrinol. Metab. 97, 4663–4672. Chirambo, G.M., 2012. The role played by alkaline phosphatase in lipid droplet formation in different lipid-storing cell types. PhD thesis, University of the Witwatersrand, Department of Chemical Pathology. Cornelius, P., MacDougald, O., Lane, M., 1994. Regulation of adipocyte development. Annu. Rev. Nutr. 14, 99–129. de La Coste, A., Romagnolo, B., Billuart, P., Renard, C.A., Buendia, M.A., Soubrane, O., et al., 1998. Somatic mutations of the beta-catenin gene are frequent in mouse and human hepatocellular carcinomas. Proc. Natl. Acad. Sci. U. S. A. 95, 8847–8851. Gregoire, F., Smas, C., Sul, H., 1998. Understanding of adipocyte differentiation. Physiol. Rev. 78, 783–809. Hernández-Mosqueira, C., Velez-delValle, C., Kuri-Harcuch, W., 2015. Tissue alkaline phosphatase is involved in lipid metabolism and gene expression and secretion of adipokines in adipocytes. Biochim. Biophys. Acta 1850, 2485–2496.
229
Hessle, L., Johnson, K.A., Anderson, H.C., Narisawa, S., Sali, A., Goding, J.W., et al., 2002. Tissue-nonspecific alkaline phosphatase and plasma cell membrane glycoprotein-1 are central antagonistic regulators of bone mineralization. Proc. Natl. Acad. Sci. U. S. A. 99, 9445–9449. Hsu, W., 1980. Toxicity and drug interactions of levamisole. J. Am. Vet. Med. Assoc. 176, 1166–1169. Kang, S., Bennett, C.N., Gerin, I., Rapp, L.A., Hankenson, K.D., Macdougald, O.A., 2007. Wnt signaling stimulates osteoblastogenesis of mesenchymal precursors by suppressing CCAAT/enhancer-binding protein alpha and peroxisome proliferator-activated receptor gamma. J. Biol. Chem. 282, 14515–14524. Kocbay, G., Telci, A., Tutuncu, Y., Tiryaki, B., Ozel, S., Cevikbas, U., et al., 2011. Alkaline phosphatase: can it be considered as an indicator of liver fibrosis in non-alcoholic steatohepatitis with type 2 diabetes? Bratisl. Lek. Listy 112, 626–629. Kodama, H., Koyama, H., Sudo, H., Kasai, S., 1983. Adipose conversion of mouse bone marrow fibroblasts in vitro: their alkaline phosphatase activity. Cell Struct. Funct. 8, 19–27. Korostishevsky, M., Cohen, Z., Malkin, I., Ermakov, S., Yarenchuk, O., Livshits, G., 2010. Morphological and biochemical features of obesity are associated with mineralization genes' polymorphisms. Int. J. Obes. 34, 1308–1318. Krauss, R.M., Herbert, P.N., Levy, R.I., Fredrickson, D.S., 1973. Further observations on the activation and inhibition of lipoprotein lipase by apolipoproteins. Circ. Res. 33, 403–411. Laughton, C., 1986. Measurement of the specific lipid content of attached cells in microtiter cultures. Anal. Biochem. 156, 307–314. Lecoeur, L., Ouhayoun, J.P., 1997. In vitro induction of osteogenic differentiation from nonosteogenic mesenchymal cells. Biomaterials 18, 989–993. Leung, C.T., Maleeff, B.E., Farrell Jr., H.M., 1989. Subcellular and ultrastructural localization of alkaline phosphatase in lactating rat mammary glands. J. Dairy Sci. 72, 2495–2509. Liang, J., Fu, M., Ciociola, E., Chandalia, M., Abate, N., 2007. Role of ENPP1 on adipocyte maturation. PLoS One 2, e882. Moss, D.W., 1992. Perspectives in alkaline phosphatase research. Clin. Chem. 38, 2486–2492. Narisawa, S., Fröhlander, N., Millán, J.L., 1997. Inactivation of two mouse alkaline phosphatase genes and establishment of a model of infantile hypophosphatasia. Dev. Dyn. 208, 432–446. Neuschhwander-Terti, B.A., 1995. Common blood tests for liver disease. Which ones are most useful? Overseas Postgrad. Med. J. 98, 49–56. Neuschwander-Tetri, B.A., Clark, J.M., Bass, N.M., Van Natta, M.L., Unalp-Arida, A., Tonascia, J., et al., 2010. NASH Clinical Research Network. Clinical, laboratory and histological associations in adults with nonalcoholic fatty liver disease. Hepatology 52, 913–924. Nosjean, O., Koyama, I., Goseki, M., Roux, B., Komoda, T., 1997. Human tissue non-specific alkaline phosphatases: sugar-moiety-induced enzymic and antigenic modulations and genetic aspects. Biochem. J. 321, 297–303. Ntambi, J.M., Young-Cheul, K., 2000. Adipocyte differentiation and gene expression. J. Nutr. 130, 3122S–3126S. Okumura, T., 2011. Role of lipid droplet proteins in liver steatosis. J. Physiol. Biochem. 67, 629–636. Pirvulescu, I., Gheorghe, L., Csiki, I., Becheanu, G., Dumbravă, M., Fica, S., et al., 2012. Noninvasive clinical model for the diagnosis of nonalcoholic steatohepatitis in overweight and morbidly obese patients undergoing bariatric surgery. Chirurgia (Bucur) 107, 772–779. Preiss, D., Sattar, N., 2008. Non-alcoholic fatty liver disease: an overview of prevalence, diagnosis, pathogenesis and treatment considerations. Clin. Sci. 115, 141–150. Price, C.P., 1993. Multiple forms of human serum alkaline phosphatase: detection and quantitation. Ann. Clin. Biochem. 30, 355–372. Rae, C., Hare, N., Bubb, W., 2003. Inhibition of glutamine transport depletes glutamate and GABA neurotransmitter pools: further evidence for metabolic compartmentation. J. Neurochem. 85, 503–514. Singh, D.K., Sakhuja, P., Malhotra, V., Gondal, R., Sarin, S.K., 2008. Independent predictors of steatohepatitis and fibrosis in Asian Indian patients with non-alcoholic steatohepatitis. Dig. Dis. Sci. 53, 1967–1976. Spiegelman, B.M., 1998. PPAR-γ, adipogenic regulator and thiazolidinedione receptor. Diabetes 47, 507–514. Takigawa, Y., Brown, A.M., 2008. Wnt signaling in liver cancer. Curr. Drug Targets 9, 1013–1024. Wallach, D., Ko, H., 1964. Some properties of an alkaline phosphatase from rat adipose tissue. Can. J. Biochem. 42, 1445–1457. Wiwanitkit, V., 2001. High serum alkaline phosphatase levels, a study in 181 Thai adult hospitalized patients. BMC Fam. Pract. 2, 2. Yoshida, H., Yumoto, T., 1987. Alkaline phosphatase-positive reticular cells of chicken bone marrow – in vivo and in vitro studies. Int. J. Cell Cloning 5, 35–54. Zhou, H.H., Chin, C.N., Wu, M., Ni, W., Quan, S., Liu, F., et al., 2009. Suppression of PC-1/ ENPP-1 expression improves insulin sensitivity in vitro and in vivo. Eur. J. Pharmacol. 616, 346–352.
Contents lists available at ScienceDirect
Experimental and Molecular Pathology journal homepage: www.elsevier.com/locate/yexmp
The role of alkaline phosphatase in intracellular lipid accumulation in the human hepatocarcinoma cell line, HepG2 George M Chirambo a,b, Chantal van Niekerk a,1, Nigel J Crowther a,⁎ a b
Department of Chemical Pathology, National Health Laboratory Service, University of Witwatersrand Medical School, Johannesburg, South Africa Department of Biochemistry, College Of Medicine, University of Malawi, Blantyre, Malawi
a r t i c l e
i n f o
Article history: Received 12 October 2016 Accepted 11 February 2017 Available online 13 February 2017 Keywords: Alkaline phosphatase HepG2 Lipid droplet PPARgamma Adipogenesis
a b s t r a c t Inhibition of tissue non-specific alkaline phosphatase (TNALP) decreases intracellular lipid accumulation in human preadipocytes and the murine preadipocyte cell line, 3T3-L1. Therefore, the current study was performed to determine if TNALP is required for intracellular lipid deposition in the human hepatocyte cell line, HepG2. Intracellular lipid accumulation, TNALP activity and peroxisome proliferator activated receptor (PPAR) γ gene expression were measured in HepG2 and 3T3-L1 cells in the presence and absence of the TNALP inhibitors levamisole and histidine. Sub-cellular TNALP activity was localized using cytochemical analysis. Both PPARγ gene expression and TNALP activity increased during intracellular lipid accumulation in HepG2 and 3T3-L1 cells. Inhibition of TNALP blocked intracellular lipid accumulation but did not alter expression of the PPARγ gene. In HepG2 cells, TNALP co-localized with adipophilin on the lipid droplet membrane. These data suggest a role for TNALP in lipid droplet formation, possibly downstream from PPARγ, within HepG2 and 3T3-L1 cells. © 2017 Elsevier Inc. All rights reserved.
1. Introduction Alkaline phosphatases (ALPs) are a group of membrane-bound glycoproteins that hydrolyze a broad range of monophosphate esters at alkaline pH optima (Price, 1993). There are four ALP isoenzymes termed tissue non-specific, intestinal, placental and germ cell, each coded by a separate gene (Moss, 1992). Tissue non-specific alkaline phosphatase (TNALP) comprises of three different isoforms called liver, bone and kidney ALP. These enzymes are coded by the same gene but differ in terms of their pattern of glycosylation (Nosjean et al., 1997). Tissue non-specific alkaline phosphatase has a broad tissue distribution however, its precise function and mode of action are not fully understood. Alkaline phosphatase activity is found in human blood and is a commonly measured analyte in clinical chemistry laboratories. Obstruction of the bile duct is one of the main causes of elevated ALP levels in human serum. Also, increased levels of ALP may be observed in serum during active bone formation (Neuschhwander-Terti, 1995; Wiwanitkit, 2001). Therefore, the principal clinical and diagnostic
⁎ Corresponding author at: Department of Chemical Pathology, National Health Laboratory Service, University of Witwatersrand Medical School, 7 York Rd., Parktown 2193, Johannesburg, South Africa. E-mail addresses: [email protected] (G.M. Chirambo), [email protected] (C. van Niekerk), [email protected] (N.J. Crowther). 1 Current address: Department of Chemical Pathology, National Health Laboratory Service, University of Pretoria, Pretoria, South Africa.
http://dx.doi.org/10.1016/j.yexmp.2017.02.007 0014-4800/© 2017 Elsevier Inc. All rights reserved.
reasons for measuring blood ALP concentration is to confirm the presence of bone or liver pathology. Studies have shown that TNALP is expressed in human preadipocytes (Ali et al., 2006a) and in the murine preadipocyte cell lines, 3T3-L1 (Ali et al., 2005) and 3T3-F442A (Hernández-Mosqueira et al., 2015). The 3T3-L1 cell line has proven to be an excellent model for studying preadipocyte differentiation and maturation (Ntambi and Young-Cheul, 2000). The proliferation and determination of preadipocytes is associated with the specific activation of a number of transcriptional factors and an upward regulation of adipogenic genes (Cornelius et al., 1994). The hallmark of adipocyte maturation is intracellular accumulation of lipid within membrane bound lipid droplets (Gregoire et al., 1998), with the peroxisome proliferator activated receptor (PPAR) γ transcription factor acting as a prime regulator of adipogenesis (Spiegelman, 1998). A role for TNALP in intracellular lipid accumulation in 3T3-L1 cells and human preadipocytes has been proven using inhibitors of TNALP (Ali et al., 2006a; Ali et al., 2006b; Ali et al., 2005), and studies have shown that this enzyme is localized to the membrane surrounding the lipid droplet (Ali et al., 2006a; Ali et al., 2005). However, the exact role of TNALP in adipogenesis is not yet understood. It is known that TNALP is expressed in hepatocytes (Price, 1993), and that these cells are also able to accumulate intracellular lipid within membranebound droplets. These organelles express a number of proteins on their surfaces that are common to both adipocytes and hepatocytes (Okumura, 2011). Excessive acquisition of lipid within hepatocytes is
G.M. Chirambo et al. / Experimental and Molecular Pathology 102 (2017) 224–229
observed in alcoholic and non-alcoholic steatohepatitis (NASH) with the latter now being one of the most common liver pathologies observed in developed countries, a probable consequence of the high prevalence of obesity (Preiss and Sattar, 2008). The presence of TNALP in preadipocytes and hepatocytes, the ability of both cell types to accumulate lipid within membrane-bound lipid droplets, and the evidence that TNALP is involved in intracellular lipid storage within preadipocytes forms the basis of our hypothesis that TNALP plays a role in lipid accumulation within hepatocytes. This hypothesis was tested by inducing lipid accumulation in the human hepatoma cell line, HepG2 and 3T3-L1 cells in the presence and absence of the TNALP inhibitors, levamisole and histidine. The expression of the adipogenic regulator, PPARγ was also assessed and the sub-cellular localization of TNALP was analysed via cytochemical analysis. 2. Materials and methods 2.1. Culture of HepG2 and 3T3-L1 cells All tissue culture reagents were obtained from BioWhittaker (Walkersville, MD, USA) and GibCo Invitrogen (Paisley, Scotland) unless otherwise stated. HepG2 and 3T3-L1 cells were obtained from the American Type Culture Collection (ATCC). HepG2 cells were grown in Earle's minimum essential medium (EMEM) supplemented with 10% fetal bovine serum, 100 U/ml penicillin, 100 μg/ml streptomycin, 2 mmol/l glutamine, non-essential amino acids, 1 mmol/l sodium pyruvate (maintenance medium) and oleic acid coupled to albumin (Sigma-Aldrich, St Louis, MO, USA) at a final concentration of 400 μmol/l of oleic acid, to induce intracellular lipid accumulation (Brasaemle et al., 1997). The maintenance medium was replaced every third day until the cells were ready for experiments. The 3T3-L1 cells were maintained in Dulbeco's modified eagle's medium (DMEM) supplemented with 10% fetal bovine serum, 100 U/ml penicillin, 100 μg/ml streptomycin and 2 mmol/l glutamine. When the cells reached confluence, 1.70 μmol/l insulin (Novo-Nordisk. Denmark), 0.50 mmol/l 3-isobutyl-1-methylxanthine (IBMX, Sigma-Aldrich, St Louis, MO, USA) and 0.22 μmol/l dexamethasone (Sigma-Aldrich, St Louis, MO, USA) were added to the culture medium to initiate adipogenesis. This transformation medium was removed after 3 days and replaced with similar medium containing insulin only. Following a further 3 days of culture this medium was replaced with the original maintenance medium which was changed every 3 days. 2.2. Use of TNALP inhibitors In HepG2 cells levamisole was used at a final concentration of 3.0 mmol/l whilst histidine was used at a concentration of 75.0 mmol/l. In 3T3-L1 cells levamisole was used at final concentration of 2.0 mmol/l whilst histidine was used at a concentration of 50.0 mmol/l. The concentrations of the TNALP inhibitors were derived from the literature for the 3T3-L1 cells (Ali et al., 2005) whilst dose response curves were performed for the HepG2 cells based on the concentrations used for the preadipocyte cell line. The final concentrations chosen were those that gave maximal inhibition. The inhibitors were added to the culture medium simultaneous with the induction of lipid droplet formation. 2.3. Measurements of intracellular lipid accumulation, TNALP activity and total protein Intracellular lipid accumulation was measured at baseline (day 0) and at 4, 7 and 11 days after the induction of intracellular lipid formation in HepG2 and 3T3-L1 cells using the lipid-specific dye Oil Red O (Laughton, 1986). TNALP activity was determined at the same time points as the cellular lipid levels using a previously published method
225
(Ali et al., 2005). The cellular protein content was analysed using the Bradford method (Bradford, 1976).
2.4. PPARγ gene expression in HepG2 and 3T3-L1 cells Total RNA was isolated from cell cultures on day 0, 4 and 7 post-induction of lipid droplet formation using an RNeasy Mini kit (Qiagen, Hilden, Germany). Synthesis of cDNA from the total RNA was then carried out using an oligo dT primer (Inqaba Biotech, Pretoria, South Africa), and Moloney murine leukemia virus reverse transcriptase (Promega, Madison WI, USA). Quantitative real-time PCR was then performed with the TATA-box binding protein (TBP) gene used as an endogenous internal control. Primers were designed using the GeneRunner Software (Hasting Software Inc., Las Vegas, USA) and were synthesized by Inqaba Biotech (Pretoria, South Africa). The TBP primer sequences for the HepG2 cells were: forward, 5′-CAG TGA CCC AGC AGC ATC -3′ and reverse, 5′-GTC AGT CCA GTG CCA TAA GG-3′, producing an amplicon of 277 bp. The sequences for the 3T3-L1 TBP primers were: forward, 5′-ACC CTT CAC CAA TGA CTC CTA TG -3′ and reverse, 5′-ATG ATG ACT GCA GCA AAT CGC-3′, with an amplicon length of 189 bp. The primer sequences of the PPARγ gene for the HepG2 cells were as follows: forward, 5′-GGT TGA CAC AGA GAT GCC A-3′ and reverse, 5′-CAA AGG AGT GGG AGT GGT C-3′, producing an amplicon of 88 bp. The sequences for the 3T3-L1 PPARγ primers were: forward, 5′CCA GAG CAT GGT GCC TTC GCT-3′ and reverse, 5′-CAG CAA CCA TTG GGT CAG CTC-3′ producing an amplicon of 240 bp. Real-time quantitative PCR of the PPARγ and TBP cDNA was carried out using a Rotorgene 6000 (Qiagen, Hilden, Germany) thermocycler and a SensiMix SYBR kit (Quantance Ltd., London, UK). The amplification conditions for the PPARγ and TBP gene products from the HepG2 cells comprised of an annealing temperature of 57 °C and an elongation time of 20 s, whilst for the 3T3-L1 cells, the annealing temperature was 60 °C, with the same annealing time as for the HepG2 cells. A 5-point standard curve was set up using cDNA isolated from the respective cell line and ranging in concentration from 14 to 272 ng/μl. The cDNA concentration was measured using a NanoDrop 1000 spectrophotometer (Thermo Scientific, Wilmington, DE, USA).
2.5. Sub-cellular localization of TNALP The HepG2 and 3T3-L1 cells were cultured on cover slips coated in gelatin (BDH, Poole, UK), and induced to accumulate lipid using the methods described in Section 2.1. Cells were harvested between days 4–8 of culture, at which point they were fixed using a 3% >paraformaldehyde (Sigma-Aldrich, St Louis, MO, USA) solution and permeabilized with a 0.01% saponin (Sigma-Aldrich, St Louis, MO, USA) solution. The 3T3-L1 cells were incubated with a rabbit antiperilipin antibody (Cell Signaling Technology Beverly, MA, USA) whilst the HepG2 cells were incubated with a guinea pig antiadipophilin antibody (Progen Biotechnik, Heidelberg, Germany), overnight at 4 °C. The 3T3-L1 cells were incubated with a goat antirabbit immunoglobulin G antibody labeled with Alexa Fluor 594, and the HepG2 cells were incubated with a goat anti-guinea pig immunoglobulin G antibody labeled with Alex Fluor 488 (both secondary antibodies supplied by Life Technologies, Grand Island, NY, USA). The incubation period was 1 h, in the dark at room temperature. Both cell types were then stained for TNALP activity using a synthetic substrate, ELF97 (Invitrogen Molecular Probes, Leiden, Holland) that gives rise to a fluorescent yellow/green end product that precipitates at the site of TNALP activity. The incubation with ELF97 lasted for 15 min, at room temperature. The cover slips were then mounted onto a microscope slide and visualized at a magnification of 100 using an Olympus 1 × 71 microscope fitted with an Olympus XM10 camera (Olympus, Hamburg, Germany).
226
G.M. Chirambo et al. / Experimental and Molecular Pathology 102 (2017) 224–229
2.6. Statistical analysis All experiments were repeated 3 times unless otherwise stated. Mean and standard error of the mean (SEM) were calculated for each of the variables under study. The Students paired t-test was used to compare mean values at each time point with the baseline value and to compare the mean values at each time point, obtained in the presence of TNALP inhibitors, to the value obtained at the same time point in the absence of inhibitors. An ANOVA was used for comparing the areas-under-the-curve (AUCs; calculated using the trapezoid rule) for TNALP activity, lipid accumulation and PPARγ gene expression across the 3 different treatment groups.
The data in Fig. 1B demonstrates that in the 3T3-L1 cells that were not treated with TNALP inhibitors, TNALP activity rose to a peak on day 7, and then fell back to a near-baseline level by day 11, with values at all 3 time points being significantly higher than the level at baseline (P b 0.05 for all). In cells treated with levamisole or histidine, TNALP levels stayed close to baseline values with levels being significantly lower than those in untreated cells at all time points (P b 0.05 for all) for the histidine treated cells, and at day 4 (P b 0.005) for the levamisole treated cells. The AUC values were 24.3 ± 1.77, 4.68 ± 1.19 and 1.76 ± 1.39 for non-treated, levamisole treated and histidine treated cells, respectively, with the values for the non-treated cells being significantly different from both treated groups (P b 0.0001 for both comparisons).
3. Results 3.2. Effect of levamisole and histidine on intracellular lipid accumulation 3.1. Effect of levamisole and histidine on TNALP activity In the HepG2 cells, in the absence of TNALP inhibitors, the activity of the enzyme significantly rose from day 0 to day 4 (P b 0.05) and then fell almost to the baseline level by day 11 (Fig. 1A). This trend was repeated in cells treated with levamisole but at days 4, 7 and 11, the ALP activity was lower than that observed in the untreated cells. The cells treated with histidine had TNALP activity levels that were lower than those observed in the untreated and levamisole-treated cells at all time points, with the level on day 4 being statistically significantly lower than in the untreated cells (P b 0.05). The AUC was derived from the data in Fig. 1A and the results were as follows (values are mean ± SEM): non-treated, 43.0 ± 5.17; levamisole treated, 26.1 ± 0.52; histidine treated, 7.73 ± 1.16. All values were significantly different from each other (P b 0.01 for all comparisons).
Fig. 1. Intracellular TNALP levels during adipogenesis in HepG2 (A) and 3T3-L1 (B) cells. Each marker and bar represents the mean and SEM respectively, of 3 experiments; the solid line represents incubations performed in the absence of TNALP inhibitors, the long dashes represent incubations performed in the presence of levamisole, and the short dashes represent incubations performed in the presence of histidine; *P b 0.05 versus day 0; ##P b 0.005 versus levamisole; + P b 0.05 versus histidine.
The data in Fig. 2A shows that in non-treated HepG2 cells, intracellular lipid levels rose from day 0 to day 11, with values being significantly higher (P b 0.05) than baseline levels at days 7 and 11. Lipid levels also rose in the levamisole treated cells but the values remained below those for the non-treated cells, being significantly lower at day 11 (P b 0.05). The histidine treated HepG2 cells had the lowest intracellular lipid levels, being greater than the baseline values only at day 4 (P b 0.05) and being lower than non-treated cells at days 7 and 11 (P b 0.05 for both). The AUCs derived from Fig. 2A were as follows: 12.6 ± 1.7, 8.81 ± 1.06 and 6.70 ± 0.71 for non-treated, levamisole treated and histidine treated, respectively. The AUC levels were significantly higher in
Fig. 2. Intracellular lipid accumulation during adipogenesis in HepG2 (A) and 3T3-L1 (B) cells. Each marker and bar represents the mean and SEM respectively, of 3 experiments; the solid line represents incubations performed in the absence of TNALP inhibitors, the long dashes represent incubations performed in the presence of levamisole, and the short dashes represents incubations performed in the presence of histidine; *P b 0.05, **P b 0.005 versus day 0; #P b 0.05, ##P b 0.005 versus levamisole; + P b 0.05 versus histidine.
G.M. Chirambo et al. / Experimental and Molecular Pathology 102 (2017) 224–229
the non-treated compared to the histidine treated cells (P b 0.05), and also tended to be higher in the former group when compared to the levamisole treated cells (P = 0.06). Similar trends to those in the HepG2 cells were observed for intracellular lipid levels in the 3T3-L1 cells (Fig. 2B). Thus, lipid levels were higher in non-treated cells on days 7 (P b 0.05) and 11 (P b 0.005) when compared to baseline and when compared to the histidine treated cells (P b 0.05 for both days). On day 4, lipid levels were higher in non-treated compared to levamisole treated cells (P b 0.05). The AUC values for lipid accumulation in the 3T3-L1 cells were 10.9 ± 0.88 for untreated cells, 6.47 ± 1.60 for levamisole treated cells and 4.30 ± 1.25 for histidine treated cells, with values being significantly higher for both groups of TNALP inhibitor treated cells when compared to the control group (P b 0.05 for both).
3.3. PPARγ gene expression during intracellular lipid accumulation Fig. 3A shows that PPARγ gene expression in HepG2 cells rose to a peak on day 4, at which point the level was significantly higher in all the groups of cells when compared to baseline levels (P b 0.05 for all comparisons). A similar trend was observed in the 3T3-L1 cells with statistically significant differences between day 4 and baseline levels observed in the non-treated cells (Fig. 3B), with a close-to-significant effect seen in the levamisole treated cells (P = 0.053). There were no significant differences in AUC levels for PPARγ gene expression across the treatment groups for either the HepG2 or the 3T3-L1 cells (data not shown).
Fig. 3. PPARγ gene expression during adipogenesis in HepG2 (A) and 3T3-L1 (B) cells. Each marker and bar represents the mean and SEM respectively, of 3 experiments; the solid line represents incubations performed in the absence of TNALP inhibitors, the long dashes represent incubations performed in the presence of levamisole, and the short dashes represent incubations performed in the presence of histidine; *P b 0.05 versus day 0.
227
3.4. Sub-cellular localization of TNALP activity in HepG2 cells The photomicrographs in Fig. 4 demonstrate that the staining for perilipin in 3T3-L1 cells and adipophilin in HepG2 cells co-localises with TNALP activity.
4. Discussion The current study is the first to demonstrate that TNALP activity increases within a hepatocyte cell line during intracellular lipid accretion. It is also the first study to show that inhibition of TNALP activity blocks cytoplasmic lipid accumulation in these cells. Previous studies have shown that TNALP inhibition blocks intracellular lipid accumulation in human preadipocytes (Ali et al., 2006a) and in the murine preadipocyte cell line, 3T3-L1 (Ali et al., 2005). Furthermore, TNALP activity has been identified in lipid-storing cell types from a number of different species i.e. rat adipose tissue (Wallach and Ko, 1964), adipocyte precursor cells in human bone marrow (Bianco et al., 1988), rabbit preadipocytes (Lecoeur and Ouhayoun, 1997), 3T3-L1 (Ali et al., 2005) and 3T3-F442A (Hernández-Mosqueira et al., 2015) murine preadipocyte cell lines, human preadipocytes (Ali et al., 2006a), murine bone marrow preadipocytes (Kodama et al., 1983) and adipocytes from chicken bone marrow (Yoshida and Yumoto, 1987). Also, TNALP is expressed in lactating rat mammary gland epithelial cells (Leung et al., 1989) where it is located on the membrane enclosing the cytoplasmic lipid droplets that are secreted and form the fat fraction of milk. TNALP has also been observed on the membrane of the milk fat globules present in bovine milk (Bingham and Malin, 1992). Within human preadipocytes (Ali et al., 2006a) and murine 3T3-L1 preadipocyte cells (Ali et al., 2005), TNALP is also found on the membrane surrounding the intracellular lipid droplets, and this sub-cellular location was also confirmed for TNALP in HepG2 cells in the present study. These data suggest that TNALP expression is a characteristic of lipid storing cell types. Furthermore, its location on the lipid droplet membrane and the ability of TNALP inhibitors to block intracellular lipid accumulation suggest that it is intrinsically involved in the control of cellular lipid storage. This function of TNALP is further confirmed by the lack of fat tissue observed in mice harbouring a TNALP gene-knockout (Narisawa et al., 1997) and by a study demonstrating that polymorphisms in the TNALP gene are associated with body fat distribution in humans (Korostishevsky et al., 2010). The role of TNALP within the many cell types in which it is expressed, with the exception of osteoblasts, is not known. However, the current study now shows that in lipid storing cells, TNALP acts to control intracellular lipid accumulation. The TNALP inhibitors used in this study do have non-TNALP mediated sub-cellular effects (Hsu, 1980; Basi et al., 1994; Rae et al., 2003). It is therefore possible that the inhibition of cellular lipid accumulation observed in this study is related to non-TNALP effects of levamisole and histidine. However, the use of small interfering (si) RNAs specific for TNALP in 3T3-L1 and HepG2 cells recapitulated the effects seen in the present study (Chirambo, 2012), strongly suggesting that the TNALP inhibitors used in the current investigation are working through TNALP. Furthermore, the localization of TNALP to lipid droplets (Ali et al., 2005; Ali et al., 2006a; Leung et al., 1989) and its presence in a number of different lipid-storing cell types (Ali et al., 2005; Ali et al., 2006a; Bianco et al., 1988; Hernández-Mosqueira et al., 2015; Kodama et al., 1983; Lecoeur and Ouhayoun, 1997; Leung et al., 1989; Wallach and Ko, 1964; Yoshida and Yumoto, 1987) further suggest that the most likely intermediary for the effects of these TNALP inhibitors is TNALP itself. The TNALP activity levels peak in the 3T3-L1 cells after 7 days but peak after 4 days in the HepG2 cells. The reasons for this are not known however, differences in the time course of gene expression should not be unexpected since these are 2 very different cell types. Thus, the HepG2 cell is representative of a mature human hepatocyte whilst the 3T3-L1 cell line is representative of an immature murine
228
G.M. Chirambo et al. / Experimental and Molecular Pathology 102 (2017) 224–229
Fig. 4. Cytochemical analysis of HepG2 and 3T3-L1 cells for intra-cellular localization of TNALP. In HepG2 cells, ALP (A) and adipophilin (B) staining were co-localized, thus suggesting that TNALP is present on the lipid droplet membrane. In the 3T3-L1 cells, TNALP (D) and perilipin (E) staining were also found to co-localize, once again demonstrating that TNALP is situated on the lipid droplet membrane. Plates C and F show phase contrast pictures of HepG2 and 3T3-L1 cells respectively, taken under transmitted light. Magnification was ×100 for all images.
adipocyte. Furthermore, each cell type has very different functions with lipid storage being the prime purpose of adipocytes but not hepatocytes. In both cell lines intracellular lipid accumulation was still increasing on day 11 of culture despite falling levels of TNALP activity (see Figs. 1 and 2). Furthermore, treatment of both HepG2 and 3T3-L1 cells with TNALP inhibitors had no effect on PPARγ expression levels (Fig. 3). These data suggest that TNALP is required for the initiation of lipid droplet formation rather than for maintenance, and that it is not involved in the cascade of transcription factor expression that activates the process of adipogenesis (Cornelius et al., 1994), of which PPARγ is a major component (Spiegelman, 1998). Thus, TNALP is not a modulator of adipogenesis but is rather an inducer of cytoplasmic lipid storage. Our data suggests that TNALP may play a role in intracellular lipid accumulation in human hepatocytes. Unfortunately, no immunohistochemical studies have analysed whether TNALP is present within lipid-laden hepatocytes in tissue sections taken from the livers of individuals with NASH. Such investigations must be undertaken to confirm our hypothesis. However, it is interesting to note that studies have demonstrated elevated serum levels of alkaline phosphatase in subjects with histologically-proven NASH (Neuschwander-Tetri et al., 2010; Pirvulescu et al., 2012), and in subjects with NASH-related liver fibrosis (Singh et al., 2008; Kocbay et al., 2011; Bambha et al., 2012). The HepG2 cell line contains a mutation in the gene encoding β-catenin that leads to increased expression of the protein within the nucleus and higher activity of the Wnt/β-catenin signalling pathway (de La Coste et al., 1998). This is a characteristic feature of human liver tumours (de La Coste et al., 1998; Takigawa and Brown, 2008). The Wnt/β-catenin signalling pathway is known to suppress adipogenesis in mesenchymal stem cells (Kang et al., 2007) and may also block intracellular lipid accumulation in hepatocytes (Behari et al., 2010). These data suggest that the presence of the activating mutation in the β-catenin gene in HepG2 cells cannot explain the lipogenic activity of TNALP in the current study. This is further confirmed by the ability of TNALP to induce lipid accumulation in primary human preadipocytes (Ali et al., 2006a) and within the 3T3-L1 cell line (Ali et al., 2006b), which exhibits normal expression of the Wnt/β-catenin signalling pathway, which is suppressed during adipogenesis (Bennett et al., 2002). The ability of inhibitors of TNALP to block lipid accumulation within a human hepatocyte cell line suggests that this may represent a
therapeutic modality for steatohepatitis. However, TNALP is expressed in many different tissues and it has functional activity within the liver and therefore inhibition of its activity, even if confined to hepatic tissue, may have deleterious side effects. This may be averted by undertaking further studies on the precise role of TNALP in intracellular lipid accumulation. This would allow identification of the molecular events downstream of TNALP that are specific to cellular lipid storage and which may represent valid targets for future therapies for steatohepatitis. The aim of the current study was not to determine the function of TNALP in intracellular lipid accumulation. However, it is possible to hypothesise what the role of TNALP may be within this sub-cellular process. It is known that TNALP plays an important role in bone mineralization where it works to convert inorganic pyrophosphate, a potent inhibitor of the mineralization process, to phosphate, a known activator of mineralization. Within osteoblasts, pyrophosphate is generated by the enzyme ectonucleotide pyrophosphatase/phosphodiesterase (ENPP) 1. Thus, the level of bone mineralization is determined by the balance between TNALP and ENPP1 activity (Hessle et al., 2002). It is noteworthy that ENPP1 expression has been detected in 3T3-L1 cells, where over expression reduces the level of adipogenesis (Liang et al., 2007). This enzyme is also expressed in liver tissue (Zhou et al., 2009) and a positive association between adipocyte expression of ENPP1 and hepatic triglyceride levels has been demonstrated in humans (Chandalia et al., 2012). Furthermore, it has been shown that ENPP1 and TNALP gene polymorphisms associate with anthropometric markers of obesity (Korostishevsky et al., 2010). It is therefore tempting to suggest that TNALP and ENPP1 may function in adipocytes to control intracellular levels of pyrophosphate and phosphate. However, it is not known whether pyrophosphate (or phosphate) has any effect on intracellular lipid accumulation, although it is known that pyrophosphate is an inhibitor of lipoprotein lipase (Krauss et al., 1973), a key enzyme for the uptake of free fatty acids into adipocytes. In summary, the present study has shown that intracellular lipid accumulation and TNALP activity are closely linked in both HepG2 and 3T3-L1 cells, suggesting that TNALP expression maybe a common requirement for cells that accumulate cytoplasmic lipid droplets. The localization of TNALP to the surface of the lipid droplet membrane in both cell lines, as well as in the epithelial cells of lactating rat mammary
G.M. Chirambo et al. / Experimental and Molecular Pathology 102 (2017) 224–229
glands (Leung et al., 1989), also points to the involvement of TNALP in the sub-cellular processes related to intracellular lipid accumulation. Further studies are required to determine the exact role of TNALP in cytoplasmic lipid storage but should initially focus on the control of intracellular inorganic pyrophosphate and phosphate levels. Disclosure None of the authors has any conflict of interest related to this manuscript. Acknowledgements The authors thank members of the routine laboratory service of the department of Chemical pathology, NHLS, Charlotte Maxeke Hospital (Johannesburg) for performing the TNALP activity assay. Funding: The study was supported by financial assistance from the NHLS Research Trust, the National Research Foundation of South Africa and the College of Medicine (Malawi). References Ali, A.T., Penny, C.B., Paiker, J.E., van Niekerk, C., Smit, A.T., Crowther, N.J., 2005. Alkaline phosphatase is involved in the control of adipogenesis in the murine preadipoctyte cell line, 3T3-L1. Clin. Chim. Acta 351, 101–109. Ali, A.T., Penny, C.B., Paiker, J.E., Psaras, G., Ikram, F., Crowther, N.J., 2006a. The effect of alkaline phosphatase inhibitors on intracellular lipid accumulation in preadipocytes isolated from human mammary tissue. Ann. Clin. Biochem. 43, 207–213. Ali, A.T., Penny, C.B., Paiker, J.E., Psaras, G., Ikram, F., Crowther, N.J., 2006b. The relationship between alkaline phosphatase activity and intracellular lipid accumulation in murine 3T3-L1 cells and human preadipocytes. Anal. Biochem. 354, 247–254. Bambha, K., Belt, P., Abraham, M., Wilson, L.A., Pabst, M., Ferrell, L., et al., 2012. Nonalcoholic steatohepatitis clinical research network research group. Ethnicity and nonalcoholic fatty liver disease. Hepatology 55, 769–780. Basi, N., George, M., Pointer, R., 1994. Regulation of glycogen synthase activity in isolated rat adipocytes by levamisole. Life Sci. 54, 1027–1034. Behari, J., Yeh, T.H., Krauland, L., Otruba, W., Cieply, B., Hauth, B., et al., 2010. Liver-specific beta-catenin knockout mice exhibit defective bile acid and cholesterol homeostasis and increased susceptibility to diet-induced steatohepatitis. Am. J. Pathol. 176, 744–753. Bennett, C.N., Ross, S.E., Longo, K.A., Bajnok, L., Hemati, N., Johnson, K.W., et al., 2002. Regulation of Wnt signaling during adipogenesis. J. Biol. Chem. 277, 30998–31004. Bianco, P., Costantini, M., Dearden, L.C., Bonucci, E., 1988. Alkaline phosphatase positive precursors of adipocytes in the human bone marrow. Br. J. Haematol. 68, 401–403. Bingham, E.W., Malin, E.L., 1992. Alkaline phosphatase in the lactating bovine mammary gland and the milk fat globule membrane. Release by phosphatidylinositol-specific phospholipase C. Comp. Biochem. Physiol. B 102, 213–218. Bradford, M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilising the principle of protein-dye binding. Anal. Biochem. 72, 248–254. Brasaemle, D.L., Barber, T., Kimmel, A.R., Londos, C., 1997. Post-translational regulation of perilipin expression. Stabilization by stored intracellular neutral lipids. J. Biol. Chem. 272, 9378–9387. Chandalia, M., Davila, H., Pan, W., Szuszkiewicz, M., Tuvdendorj, D., Livingston, E.H., Abate, N., 2012. Adipose tissue dysfunction in humans: a potential role for the transmembrane protein ENPP1. J. Clin. Endocrinol. Metab. 97, 4663–4672. Chirambo, G.M., 2012. The role played by alkaline phosphatase in lipid droplet formation in different lipid-storing cell types. PhD thesis, University of the Witwatersrand, Department of Chemical Pathology. Cornelius, P., MacDougald, O., Lane, M., 1994. Regulation of adipocyte development. Annu. Rev. Nutr. 14, 99–129. de La Coste, A., Romagnolo, B., Billuart, P., Renard, C.A., Buendia, M.A., Soubrane, O., et al., 1998. Somatic mutations of the beta-catenin gene are frequent in mouse and human hepatocellular carcinomas. Proc. Natl. Acad. Sci. U. S. A. 95, 8847–8851. Gregoire, F., Smas, C., Sul, H., 1998. Understanding of adipocyte differentiation. Physiol. Rev. 78, 783–809. Hernández-Mosqueira, C., Velez-delValle, C., Kuri-Harcuch, W., 2015. Tissue alkaline phosphatase is involved in lipid metabolism and gene expression and secretion of adipokines in adipocytes. Biochim. Biophys. Acta 1850, 2485–2496.
229
Hessle, L., Johnson, K.A., Anderson, H.C., Narisawa, S., Sali, A., Goding, J.W., et al., 2002. Tissue-nonspecific alkaline phosphatase and plasma cell membrane glycoprotein-1 are central antagonistic regulators of bone mineralization. Proc. Natl. Acad. Sci. U. S. A. 99, 9445–9449. Hsu, W., 1980. Toxicity and drug interactions of levamisole. J. Am. Vet. Med. Assoc. 176, 1166–1169. Kang, S., Bennett, C.N., Gerin, I., Rapp, L.A., Hankenson, K.D., Macdougald, O.A., 2007. Wnt signaling stimulates osteoblastogenesis of mesenchymal precursors by suppressing CCAAT/enhancer-binding protein alpha and peroxisome proliferator-activated receptor gamma. J. Biol. Chem. 282, 14515–14524. Kocbay, G., Telci, A., Tutuncu, Y., Tiryaki, B., Ozel, S., Cevikbas, U., et al., 2011. Alkaline phosphatase: can it be considered as an indicator of liver fibrosis in non-alcoholic steatohepatitis with type 2 diabetes? Bratisl. Lek. Listy 112, 626–629. Kodama, H., Koyama, H., Sudo, H., Kasai, S., 1983. Adipose conversion of mouse bone marrow fibroblasts in vitro: their alkaline phosphatase activity. Cell Struct. Funct. 8, 19–27. Korostishevsky, M., Cohen, Z., Malkin, I., Ermakov, S., Yarenchuk, O., Livshits, G., 2010. Morphological and biochemical features of obesity are associated with mineralization genes' polymorphisms. Int. J. Obes. 34, 1308–1318. Krauss, R.M., Herbert, P.N., Levy, R.I., Fredrickson, D.S., 1973. Further observations on the activation and inhibition of lipoprotein lipase by apolipoproteins. Circ. Res. 33, 403–411. Laughton, C., 1986. Measurement of the specific lipid content of attached cells in microtiter cultures. Anal. Biochem. 156, 307–314. Lecoeur, L., Ouhayoun, J.P., 1997. In vitro induction of osteogenic differentiation from nonosteogenic mesenchymal cells. Biomaterials 18, 989–993. Leung, C.T., Maleeff, B.E., Farrell Jr., H.M., 1989. Subcellular and ultrastructural localization of alkaline phosphatase in lactating rat mammary glands. J. Dairy Sci. 72, 2495–2509. Liang, J., Fu, M., Ciociola, E., Chandalia, M., Abate, N., 2007. Role of ENPP1 on adipocyte maturation. PLoS One 2, e882. Moss, D.W., 1992. Perspectives in alkaline phosphatase research. Clin. Chem. 38, 2486–2492. Narisawa, S., Fröhlander, N., Millán, J.L., 1997. Inactivation of two mouse alkaline phosphatase genes and establishment of a model of infantile hypophosphatasia. Dev. Dyn. 208, 432–446. Neuschhwander-Terti, B.A., 1995. Common blood tests for liver disease. Which ones are most useful? Overseas Postgrad. Med. J. 98, 49–56. Neuschwander-Tetri, B.A., Clark, J.M., Bass, N.M., Van Natta, M.L., Unalp-Arida, A., Tonascia, J., et al., 2010. NASH Clinical Research Network. Clinical, laboratory and histological associations in adults with nonalcoholic fatty liver disease. Hepatology 52, 913–924. Nosjean, O., Koyama, I., Goseki, M., Roux, B., Komoda, T., 1997. Human tissue non-specific alkaline phosphatases: sugar-moiety-induced enzymic and antigenic modulations and genetic aspects. Biochem. J. 321, 297–303. Ntambi, J.M., Young-Cheul, K., 2000. Adipocyte differentiation and gene expression. J. Nutr. 130, 3122S–3126S. Okumura, T., 2011. Role of lipid droplet proteins in liver steatosis. J. Physiol. Biochem. 67, 629–636. Pirvulescu, I., Gheorghe, L., Csiki, I., Becheanu, G., Dumbravă, M., Fica, S., et al., 2012. Noninvasive clinical model for the diagnosis of nonalcoholic steatohepatitis in overweight and morbidly obese patients undergoing bariatric surgery. Chirurgia (Bucur) 107, 772–779. Preiss, D., Sattar, N., 2008. Non-alcoholic fatty liver disease: an overview of prevalence, diagnosis, pathogenesis and treatment considerations. Clin. Sci. 115, 141–150. Price, C.P., 1993. Multiple forms of human serum alkaline phosphatase: detection and quantitation. Ann. Clin. Biochem. 30, 355–372. Rae, C., Hare, N., Bubb, W., 2003. Inhibition of glutamine transport depletes glutamate and GABA neurotransmitter pools: further evidence for metabolic compartmentation. J. Neurochem. 85, 503–514. Singh, D.K., Sakhuja, P., Malhotra, V., Gondal, R., Sarin, S.K., 2008. Independent predictors of steatohepatitis and fibrosis in Asian Indian patients with non-alcoholic steatohepatitis. Dig. Dis. Sci. 53, 1967–1976. Spiegelman, B.M., 1998. PPAR-γ, adipogenic regulator and thiazolidinedione receptor. Diabetes 47, 507–514. Takigawa, Y., Brown, A.M., 2008. Wnt signaling in liver cancer. Curr. Drug Targets 9, 1013–1024. Wallach, D., Ko, H., 1964. Some properties of an alkaline phosphatase from rat adipose tissue. Can. J. Biochem. 42, 1445–1457. Wiwanitkit, V., 2001. High serum alkaline phosphatase levels, a study in 181 Thai adult hospitalized patients. BMC Fam. Pract. 2, 2. Yoshida, H., Yumoto, T., 1987. Alkaline phosphatase-positive reticular cells of chicken bone marrow – in vivo and in vitro studies. Int. J. Cell Cloning 5, 35–54. Zhou, H.H., Chin, C.N., Wu, M., Ni, W., Quan, S., Liu, F., et al., 2009. Suppression of PC-1/ ENPP-1 expression improves insulin sensitivity in vitro and in vivo. Eur. J. Pharmacol. 616, 346–352.