Get 20M+ Full-Text Papers For Less Than $1.50/day. Subscribe now for You or Your Team.

Learn More →

STAT3-activated CD36 facilitates fatty acid uptake in chronic lymphocytic leukemia cells

STAT3-activated CD36 facilitates fatty acid uptake in chronic lymphocytic leukemia cells www.oncotarget.com Oncotarget, 2018, Vol. 9, (No. 30), pp: 21268-21280 Research Paper STAT3-activated CD36 facilitates fatty acid uptake in chronic lymphocytic leukemia cells 1,2 1 1 1 1 Uri Rozovski , David M. Harris , Ping Li , Zhiming Liu , Preetesh Jain , Alessandra 1 1 1 1 1 Ferrajoli , Jan Burger , Phillip Thompson , Nitin Jain , William Wierda , Michael J. 1 1 Keating and Zeev Estrov Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA Institute of Hematology, Davidoff Cancer Center, Rabin Medical Center, Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel Correspondence to: Zeev Estrov, email: zestrov@mdanderson.org Keywords: CLL; CD36; metabolism; STAT3 Received: January 31, 2018 Accepted: March 21, 2018 Published: April 20, 2018 Copyright: Rozovski et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License 3.0 (CC BY 3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. ABSTRACT Although several studies established that unlike normal B cells chronic lymphocytic leukemia (CLL) cells metabolize fatty acids (FA), how CLL cells internalize FA is poorly understood. Because in various cell types CD36 facilitates FA uptake, we wondered whether a similar mechanism is operative CLL. We found that CD36 levels are higher in CLL cells than in normal B cells, and that small interfering RNA, CD36 neutralizing antibodies or sulfosuccinimidyl oleate (SSO) that inhibits CD36 significantly reduced the oxygen consumption of CLL cells incubated with FA. Because CD36 is oeverexpressed and STAT3 is constitutively activated in CLL cells, we wondered whether STAT3 induces CD36 expression. Sequence analysis identified putative STAT3 binding sites in the CD36 gene promoter. Chromatin immunoprecipitation and an electrophoretic mobility shift assay revealed that STAT3 binds to the CD36 gene promoter. A luciferase assay and STAT3-small hairpin RNA, that significantly decreased the levels of CD36 in CLL cells, established that STAT3 activates the transcription of the CD36 gene. Furthermore, SSO induced a dose-dependent apoptosis of CLL cells. Taken together, our data suggest that STAT3 activates CD36 and that CD36 facilitates FA uptake in CLL cells. Whether CD36 inhibition would provide clinical benefits in CLL remains to be determined. labeling studies it is estimated that approximately 1% of INTRODUCTION the CLL clone expands daily [5], an expansion rate that is remarkably similar to that of normal adipocytes [6]. Chronic lymphocytic leukemia (CLL) is We and others [7, 8] have recently shown that, characterized by a gradual increase in the number unlike normal B cells, CLL cells adopt metabolic of circulating mature appearing lymphocytes. For pathways that are operative in adipocytes and myocytes. more than four decades CLL was viewed as a disease Specifically, we found that CLL cells store lipids in characterized by an accumulation of long-lived mature intracytoplasmic vacuoles and, by aberrantly expressing looking neoplastic B-lymphocytes that do not proliferate lipoprotein lipase (LPL), utilize fatty acids (FA) for the and do not die. Studies conducted in recent years clearly production of chemical energy [8, 9]. We have also shown demonstrated that CLL cells do proliferate [1–3] and that LPL expression and FA metabolism in CLL is driven undergo spontaneous apoptosis [4]. The proliferation by constitutive activation of the signal transducer and rate of CLL cells varies, however based on heavy water www.oncotarget.com 21268 Oncotarget activator of transcription (STAT)-3. However, what is the (Figure 2B). To confirm that these DNA fragments bind source of FA and what mechanisms are recruited to enable STAT3 we performed EMSA using biotinylated DNA FA entry into CLL cells is not completely understood. probes corresponding to the binding sites +187 bp – +196 CD36, also known as fatty acid translocase protein, bp, +363 bp – +374 bp, and +382 bp – +391 bp, and -93 is a multi-ligand glycoprotein that is expressed on the bp – -83 bp. Confirming the ChIP data, we found that extracellular membrane and facilitates FA uptake in CLL cell nuclear protein extracts from 2 different patients various cells such as myocytes and adipocytes [10] in formed complexes with STAT3 binding sites +187 bp – +196 bp, +363 bp – +374 bp, and +382 bp – +391 bp, but which lipid uptake is markedly impaired when CD36 not with the putative binding site -93 bp – -83 bp and that function is compromised [11–13]. Because CLL cells’ metabolic pathways resembles those of adipocytes [14] the binding was significantly attenuated by the addition and the CD36 gene harbors putative STAT3-binding sites, of excess unlabeled probe or by anti-STAT3 antibodies we hypothesized that STAT3 induces CD36 cell surface (Figure 2C). To further delineate these findings we expression and that CD36 facilitates FA uptake in CLL performed a luciferase assay of MM1 cells in which we cells. concentrated on putative STAT3 binding sites that were in close proximity to the CD36 gene start codon. We found the highest luciferase activity in IL-6 stimulated MM1 RESULTS cells transfected with the promoter fragment that included the 3 active binding sites (Figure 2D). Furthermore, we CLL cells express high CD36 protein levels of found that STAT3-shRNA downregulated mRNA levels both of STAT3 and CD36 by 9 and 6-fold, respectively, To assess the levels of CD36 protein in CLL cells we and as a result, significantly reduced STAT3 and CD36 performed Western blot analysis of PB-derived CLL cells protein levels, confirming that STAT3 induces the from 6 randomly chosen CLL patients and, as control, expression of CD36 in CLL cells (Figure 2E). CD19+ B cells from the PB of two healthy individuals. As shown in Figure 1A, protein levels of CD36 were readily detected in all CLL-patient samples and their levels were CD36 facilitates FA intake and metabolism in significantly higher than those of CD19+ B cells obtained CLL cells from healthy individuals. As in previous studies [15], Because CLL cells utilize FA and CD36 plays a key phosphoserine STAT3 was detected in CLL cells but its role in FA uptake in various cell types [18, 19], we sought levels were undetectable in normal CD19+ B cells. To to determine whether CD36 contributes to FA uptake in confirm that CLL cells express CD36 cell membrane CLL cells. We cultured CLL cells in tightly closed flaks in protein we performed flow cytometry and detected serum- and glucose-free medium and measured the oxygen CD36 protein on the surface of PB CLL cells from 6 concentration prior to and after adding FA, assuming that randomly selected CLL patients (Figure 1B). However, the if the cells consume the FA, oxygen levels in the culture percentages of cells co-expressing CD19, CD5, and CD36 medium will drop. As expected, when FA were added to antigens varied among the different patients and were 14, culture the levels of oxygen dissolved in the culture media 25, 32.9, 37, 41, and 51, respectively. To confirm these were markedly reduced whereas the dO levels of CLL cells findings we used confocal microscopy and found that, like transfected with CD36 siRNA and incubated with oleic in other cell types, CD36 is detected on the surface but acid, remained significantly higher than the dO levels in not in the cytoplasm or nucleus of CLL cells (Figure 1C). 2 the medium of non-transfected or GPDH-transfected CLL cells (Figure 3A). Similarly, the dO levels of CLL cells STAT3 binds to the CD36 gene promoter and 2 incubated with oleic acid or palmitic acid in the absence activates the CD36 gene of CD36 neutralizing antibodies significantly dropped whereas dO levels of CLL cells incubated with oleic Because STAT3 is constitutively activated in CLL acid or palmitic acid in the presence of CD36 neutralizing cells [15], we wondered whether overexpression of CD36 antibodies remained unchanged (Figure 3B). Likewise, is driven by STAT3. Using the TFSEARCH database [16] the irreversible CD36 inhibitor SSO and the LPL inhibitor we identified 4 GAS-like elements, known as putative orlistat significantly reduced CLL cell O consumption STAT3-binding sites [17], within 600 bp upstream of and the effect of both inhibitors was significantly additive the CD36 gene start codon (Figure 2A). To determine (Figure 3C). Furthermore, SSO induced apoptosis of CLL whether STAT3 binds to any of these GAS-like elements cells in a dose-dependent manner (Figure 3D). we performed ChIP. Our ChIP analysis revealed that 3 DNA fragments, whose primers amplified the regions of the putative STAT3 binding sites +187 bp – +196 bp, DISCUSSION +363 bp – +374 bp, and +382 bp – +391 bp, but not the region of the putative STAT3 binding site -93 bp – -83 Here we show that CLL cells express high levels of bp, co-immunoprecipitated with anti-STAT3 antibodies CD36 cell surface protein, that overexpression of CD36 is www.oncotarget.com 21269 Oncotarget driven by STAT3-mediated activation of the CD36 gene, with proliferation capacity [8, 21–23]. The common and that CD36 facilitates uptake of FA by CLL cells. constitutive activation of STAT3 in all CLL patients Although CLL is a heterogeneous disease, in all suggests that STAT3 is involved in a core function that patients, regardless of clinical characteristics, disease affects CLL cell survival. burden, cytogenetic abnormalities or IgHv gene mutation Several observations linked STAT3 to cellular status, STAT3 is constitutively phosphorylated on serine metabolism in malignant and non-malignant cells [24]. In 727 residues [15, 20]. Phosphoserine STAT3 activates a cancer cells STAT3 emerged as a as a key regulator of diverse repertoire of coding and non-coding genes which metabolism that integrates signals via both mitochondrial protect CLL cells from apoptosis and provide the cells and nuclear activities [25]. We have recently shown that www.oncotarget.com 21270 Oncotarget Figure 1: CLL cells express high levels of CD36. (A) Western blot analysis of CLL cells from 6 different patients and CD19+ normal B cells from 2 normal controls. In all patients samples CD36 was detected at higher levels than in normal B cells. Phosphoserine STAT3 was detected in all patient samples but not in CD19+ normal B cells and levels of unphosphorylated STAT3 were higher in CLL patient samples than in CD19+ normal B cells. Actin was used as loading control and densitometry units were normalized to actin units in the corresponding lanes. (B) CLL cells co-express CD19, CD5 and CD36. Left panel: A representative figure of flow cytometry analysis performed on CLL cells from 6 different patients is depicted. Right panel: Isotype controls. The black line is Isotype control and the red line is the corresponding antibody. Percentages represent the differences between the isotype and the specific antibody. The precentages are the (C) Confocal microscopic images (×400) of freshly isolated CLL cells (right) and normal B cells (left) stained with anti-CD36 antibodies (green) and Evans blue (red) showing CD36 on the cell surface of CLL cells but not on normal B cells. by inducing LPL production, STAT3 promotes uptake FA uptake was found to promote metastatic development and metabolism of lipid particles [8]. In various cell in a subpopulation of human oral carcinoma cells that types including CLL cells, lipid particles are stored in express high levels of CD36 [28]. vacuoles and following additional processing triglycerides To evaluate the role of CD36 in CLL cells we used undergo hydrolysis into FA generating substrates that enter three different strategies. We downregulated CD36 mRNA the Krebs cycle. Here we show that CLL cells utilize a levels by transfecting CLL cells with CD36 siRNA and, complementary strategy for a direct uptake of circulating in other experiments, blocked CD36 activity with CD36- FA by STAT3-mediated expression of CD36. neutralizing antibodies or SSO. Using these three different Studies in CD36-null mice identified CD36 as a methods we consistently found that inhibition of CD36 facilitator of FA uptake [26]. In addition, several groups disrupted FA metabolism, confirming that in CLL cells reported that CD36-mediated FA uptake plays a role in CD36 is required for FA uptake. the pathogenesis of various neoplasms. For example, Nath Sulfo-N-succinimidyl oleate (SSO) attaches to et al. showed that in hepatocellular carcinoma CD36- the CD36 binding pocket and, by doing so, it inhibits mediated uptake of FA induces epithelial to mesenchymal CD36-mediated FA uptake [29]. Therefore SSO has been transition and metastasis [27]. Likewise, CD36-mediated extensively used to study the role of CD36. For example, www.oncotarget.com 21271 Oncotarget www.oncotarget.com 21272 Oncotarget Figure 2: STAT3 binds to and activates the CD36 gene promoter. (A) Sequence analysis of the CD36 gene promoter upstream of the start codon (green) revealed 4 putative STAT3 binding sites (red). Upstream putative STAT3 biding sites are not shown because in most experiments we concentrated our efforts on studying binding sites that are in proximity to the CD36 start codon. (B) Using the ChIP method, CLL-cell chromatin fragments pulled down by anti-STAT3 antibodies were analyzed by PCR using primers directed at the 4 putative STAT3 binding sites upstream the CD36 gene start codon. As shown, anti-STAT3 antibodies co-immunoprecipitated the DNA detected by primers +236 bp – +406 bp, +120 bp – +256 bp (amplifying regions of the STAT3 putative binding sites +187 bp – +196 bp, +363 bp – 374 bp, and +382 bp – +391 bp), and -527 bp – -355 bp but not by primers -203 bp – +1 bp (amplifying the region of the putative binding site -93 bp – 83 bp). (C) Using EMSA, biotin-labeled CD36-DNA probes were incubated with CLL cells’ protein extract from 2 patients. The EMSA demonstrated that CLL cell nuclear protein extracts bound to the CD36 gene promoter at regions that include the putative STAT3 binding sites +187 bp – +196 bp, +363 bp – +374 bp, and +382 bp – +391 bp, but not the region that includes the putative binding site -93 bp – 83 bp, and that the addition of excess unlabeled probe or anti-STAT3 antibodies attenuated the binding. (D) The luciferase activity of IL-6-stimulated MM-1 cells was assessed 24 hours after transfection with the 3 depicted DNA fragments containing putative STAT3 binding sites in close proximity to the start codon is shown. The luciferase activity of each of the human CD36 promoter constructs was determined by calculating the constructs’ luciferase activity relative to the activity of the Renilla luciferase produced by the pRL-SV40 control vector. The luciferase activity of unstimulated MM1 cells (not shown) was similar to that of the pRL-SV40 control vector. The highest luciferase activity, compared to the pGL4.17 (control), in IL-6-stimulated MM1 cells transfected with the promoter fragment that included the 3 active binding sites (-203 bp; P = 0.0002). A lower albeit increased activity was observed in cells transfected with the promoter fragment that included 2 active binding sites (+236 bp; P = 0.009). There was no significant difference in the luciferase activity of fragments +122 bp and +236 bp (P = 0.12). (E) Infection of CLL cells with STAT3-shRNA, downregulated mRNA levels both of STAT3 and CD36 by 9 and 6 fold, respectively (left panel), and significantly reduced protein levels of STAT3, phsophoserine STAT3 and CD36 (right panel). The figure depicts representative results of 3 different experiments. www.oncotarget.com 21273 Oncotarget www.oncotarget.com 21274 Oncotarget Figure 3: CLL cell FA intake and metabolism is CD36-dependent. (A) CLL cells were transfected with CD36-siRNA or GAPDH and incubated with 80 mM oleic acid in a serum-free, glucose-free medium in tightly sealed flask for 48 hours. Transfection (at a transfection efficiency of 30% as assessed by flow cytometry; left panel) with CD36-siRNA, but not GAPDH, significantly reduces CD36 mRNA levels compared to transfection with GAPD (CTRL), and the dO levels were significantly reduced in the medium of non- transfected or GAPDH-transfected CLL cells but not in the medium of CD36-siRNA-transfected cells (right panel). Representative results from 3 different experiments are depicted. (B) CD36 neutralizing antibodies inhibited FA metabolism in CLL cells. CLL cells from two patients were incubated in the presence or absence of 80 mM palmitic acid (PA) or oleic acid (OA) with or without CD36 neutralizing antibodies (Ab) in sealed tissue culture flasks. The relative dO in the culture media was assessed prior to and after 48 hours of incubation. All control cultures contained ethanol at the same concentration as in palmitic acid and oleic acid. To compare dO we assessed the difference in O concentration before and after the incubation. The relative difference in dO before after adding FA is depicted after the 2 2 dO in the controls were set to 1. As shown, the CD36-neutralizing antibodies did not affect the cells’ oxygen consumption whereas in the presence of PA or OA the cells consumed O , and this effect was reversed when neutralizing antibodies were added to culture. (C) CLL cells from 2 patients were incubated for 48 hours in the presence or absence of 80 mM palmitic acid (PA) with or without the LPL inhibitor orlistat, the CD36 inhibitor SSO or both. The culture media concentration was assessed prior to and after 48 hours of incubation. As shown in the presence of PA dO was reduced, but less when SSO or orlistat were added to culture and the effect of SSO and orlistat was additive, indicating that these agents reduced FA-dependent metabolism. (D) Apoptosis was assessed in CLL cells incubated with or without increasing concentrations of SSO using flow cytometry with annexin V/PI staining. Upper panel: Representative flow cytometry plots of the below depicted three experiments. Lower panel: Results of 3 different experiments, using cells from 3 different patients are depicted as percent of the apoptosis rate in cells incubated without SSO (control). SSO was found to effectively block CD36-mediated In conclusion, unlike their normal resting B cell fatty acid uptake into cardiomyocytes [11, 30, 31] and counterparts, CLL cells utilize FA. FA uptake in CLL to restore of diabetic heart’s function following hypoxia/ cells is facilitated by STAT3-enhanced CD36 expression. reoxygenation [32]. However, SSO is chemically instable Whether inhibition of CD36-dependent FA uptake might and therefore cannot be used in its current form for long- have therapeutic benefit in CLL remains to be determined. term experiments or in vivo studies [11]. The use of CD36 neutralizing antibodies is an attractive alternative that has MATERIALS AND METHODS been used in animal models [31, 33]. CD36 neutralizing antibodies were shown to eliminate lymph node metastasis Patients’ characteristics in mice that were inoculated with tumor cells [28], suggesting that it may also be applicable in treatment of Peripheral blood samples were obtained from 19 CLL-affected lymph nodes. patients with CLL who were treated at The University of www.oncotarget.com 21275 Oncotarget Texas MD Anderson Cancer Center Clinic. The study was Confocal microscopy approved by our Institutional Review Board and patients’ A previously described method was used [35]. Informed Consent was obtained prior to sample collection. CLL low-density cells were incubated in microtubes Clinical characteristics of all patients that participated in in PBS supplemented with 5% bovine serum albumin this study are depicted in Table 1. (Cell Signaling Technology,). After 1 hour of incubation, the cells were washed three times with PBS and then Cell fractionation incubated with mouse anti-CD36 antibodies (Novus Scientific, Littleton, CO) for 1 hour, washed in PBS, re- PB cells were fractionated using Ficoll Hypaque suspended in a 0.1% solution of Evans blue dye (Sigma- 1077 (Sigma, St. Louis, MO). The low-density Aldrich) for 5 minutes, and washed in PBS to remove cellular fraction was used immediately or frozen for unbound dye. The cells were resuspended in PBS and additional studies. More than 95% of the peripheral 0.4 placed into µ-slide VI chamber slides (ibidi LLC, blood lymphocytes obtained from these patients were Verona, WI) for microscopic analysis. The slides were CD19+/CD5+, as assessed by flow cytometry (Becton, viewed using an Olympus FluoView 500 Laser Scanning Dickinson and Company, Franklin Lakes, NJ). As control Confocal Microscope (Olympus America, Center Valley, studies we obtained from the Central Blood Bank left- PA), and images were analyzed using the FluoView over buffy coats of healthy blood donors. After Ficoll- software (Olympus America). Hypaque fractionation, the donors’ B cells were isolated using Miltenyi CD19-coated beads according to the manufacturer’s instructions (Miltenyi Biotec, Bergisch Chromatin immunoprecipitation (ChIP) assay Gladbach, Germany). A chromatin immunoprecipitation (ChIP) assay was performed using a SimpleChIP Enzymatic Chromatin IP Western immunoblotting Kit (Cell Signaling Technology, Boston, MA) according to the manufacturer’s instructions. Briefly, cells were Western immunoblotting was performed as cross-linked with 1% formaldehyde for 10 minutes at previously described [34]. Briefly, CLL cell extract was room temperature and then harvested and incubated on ice prepared. The protein concentration was determined for 10 minutes in lysis buffer. Nuclei were pelleted and using a Micro BCA protein assay reagent kit (Thermo digested with micrococcal nuclease. Following sonication Scientific, Pierce, Rockford, IL). Cell lysates were and centrifugation, sheared chromatin was incubated with denatured and following electrophoresis transferred anti-STAT3 or rabbit serum (negative control) overnight at to a nitrocellulose membranes. The membranes were 4° C. Then, protein G beads were added, and the chromatin incubated with either monoclonal mouse anti-human was incubated for 2 hours in rotation. Antibody-bound STAT3 (BD Bioscience, Palo Alto, CA), polyclonal protein-DNA complexes were eluted and subjected to rabbit anti-human phosphoserine (serine 727) STAT3 PCR analysis. The primers to amplify the human CD36 (Cell Signaling Technology, Beverly, MA), or rabbit promoter were F’: -201 and R: +1, which generate a 200- anti-human CD36 (Pierce, Thermo Scientific, Waltham, bp product that covers the GAS binding site-93 to -83 bp MA) antibodies and Horseradish peroxidase-conjugated upstream of CD36; F’: +120 and R’: +256, which generate secondary antibodies (GE Healthcare, Amersham, a 136-bp product that covers the GAS binding site +187 to Buckinghamshire, UK). Proteins were visualized via +196 bp upstream of CD36; F’: +236 and R’: +406, which an enhanced chemiluminescence detection system (GE generate a 170-bp product that covers the GAS binding Healthcare). site +363 to +374 bp upstream of CD36; and F’: +355 and R’: +427, which generate a 72-bp product that covers the Flow cytometry analysis GAS binding site +382 to +391. CLL cells were fixed in 2% paraformaldehyde for 10 minutes and permeabilized overnight at –20° C. Electrophoretic mobility shift assay (EMSA) Before staining cells were washed three times in PBS Non-denatured cellular nuclear extracts were with 2% FBS. Cells were then stained with CD19 (BD prepared using a NE-PER extraction kit (Thermo Biosciences, San Jose CA), CD5 (BD Biosciences), Scientific Pierce, Rockford, IL). Nuclear protein extracts and CD36 (BD Biosciences) as well as appropriate were incubated with biotin-labeled CD36 promoters’ isotypic controls. Cells were analyzed on a FacsCaliber DNA probes in binding buffer for 30 minutes on ice. All flow cytometer (BD Biosciences) and data analysis probes which target GAS binding sites were synthesized performed using CellQuest software (BD Biosciences). by Sigma-Genosys (The Woodlands, TX). Following Graphics were created with CellQuest (BD Biosciences) incubation, the samples were separated on a 5% and WinList (Verity Software House, Topsham, ME) polyacrylamide gel, transferred onto a nylon membrane, software. www.oncotarget.com 21276 Oncotarget Table 1: Patient characteristics (n = 25) Characteristic Measure/Category Overall Age, years Median (range) 58 (40–78) WBC × 10 /L Median (range) 27 (7–55) ALC × 10 /L Median (range) 22 (1–48) Rai stage (0, 1–2/3–4) (%) (95/5) CD38 ≤30/>30 (%) 84/16 (%) Zap-70 Negative/Positive 78/22 (%) β2M (mg/L) (</≥ 4 mg/L) 95/5 (%) IGHV mutation (M/UM) 78/22 (%) FISH result, n del17p/11q/T12/13q/Negative 1/3/3/10/2 Karyotype, n Diploid/Non-diploid/Complex/not done 11/7/0/1 Survival status, (n) Alive/Dead 19/0 WBC, white blood cell count; ALC, absolute lymphocyte count; β2M, beta-2 microglobulin; M, mutated; UM, unmutated; FISH, fluorescence in situ hybridization. and fixed on the membrane via ultraviolet cross-linking. Inghirami, (department of pathology, University of The biotin-labeled probe was detected with strepavidin- Torino, Italia) using the Superfect transfection reagent horseradish peroxidase (Gel-Shift Kit; Panomics, Fremont, (QIAGEN Inc.) as previously described [15]. The 293T CA). The control consisted of 7-fold excess unlabeled cold cell culture medium was replaced after 16 hours and probe. To test the effect of STAT3, anti-STAT3 antibodies collected after 48 hours. Then, the culture medium was (BD Bioscience) mouse IgG1 (BD Bioscience) were filtered through a 45-μm syringe filter to remove floating added with the nuclear extracts [15, 36]. cells, the lentivirus was concentrated by filtration through an Amicon Ultra centrifugal filter device (EMD), and the concentrated supernatant was used to transfect CLL cells. Transfection of MM-1 cells with CD36 gene CLL cells (5 × 10 /ml) were incubated in six-well plates promoter fragments and luciferase assay (Becton, Dickinson and Company, Franklin Lakes, NJ) Four different CD36 promoter fragments, in 2 mL of Dulbecco’s modified Eagle medium (DMEM) corresponding to the above described putative STAT3- supplemented with 10% fetal bovine serum (FBS) binding sites, were transfected into MM-1 cells by using and were transfected with 100 μl of viral supernatant. electroporation. MM-1 cells were used because in these Polybrene (10 ng/mL) was added to the viral supernatant cells STAT3 phosphorylation is induced by extracellular at a ratio of 1:1000 (v/v). Transfection efficiency was signals such as IL-6. Each construct harbored a luciferase measured after 48 hours and ranged between 40% and 50% reporter gene and a CD36 promoter fragment that included (calculated on the basis of the ratio of propidium iodide– at least 1 γ-interferon activation sequence (GAS)-like negative and GFP-positive cells). These experiments elements. The luciferase activity of unstimulated or IL- were conducted using an upgraded FACSCalibur flow 6-stimulated MM-1 cells was assessed 24 hours after cytometer (Becton, Dickinson). transfection using a Dual-Luciferase Reporter Assay System (Promega) and a Sirius luminometer V3.1 Transfection of CLL cells with CD36 small (Berthold Detection Systems, Pforzheim, Germany). The interfering RNA (siRNA) luciferase activity of each of the human CD36 promoter constructs was determined by calculating the constructs’ Hundred μM human CD36-siRNA, 10 μl FAM- luciferase activity relative to the activity of the Renilla labeled siRNA, targeting the human glyceraldehyde luciferase produced by the pRL-SV40 control vector. 3-phosphate dehydrogenase, or a scrambled control (Applied Biosystems, Foster City, CA) were added to 10 μl siPORT NeoFX transfection reagent diluted in 50 µl Generation of GFP-conjugated lentiviral STAT3 Opti-MEMI reduced serum medium (Thermo Fisher short hairpin RNA (shRNA) and infection of Scientific), and incubated at room temperature for 10 CLL cells minutes. The transfection agents were incubated at room temperature with 1 × 10 CLL cells suspended in 0.2 293T cells were co-transfected with GFP-conjugated ml Opti-MEM I medium. After 1 hour of incubation, lentiviral STAT3 shRNA or a GFP-conjugated empty electroporation (Bio-Rad Laboratories) was performed and lentiviral vector and with packaging vectors (pCMV delta the cells were incubated in RPMI supplemented with 10% R8.2 and pMDG generously provided by Dr. Giorgio www.oncotarget.com 21277 Oncotarget FBS for 24 hours. Transfection efficiency of the FAM- CONFLICTS OF INTEREST conjugated siRNA was assessed on a FACSCalibur flow cytometer (Becton Dickinson Biosciences). All authors report no conflicts of interest. Measurement of cellular O consumption FINANCIAL SUPPORT Because FA metabolism increases O consumption, This study was supported by the Cancer Center palmitic acid and oleic acid utilization was assessed Support Grant from the NIH/NCI, P30 CA016672 grant. by measuring the level of dissolved O (dO ) using the 2 2 SevenGo pro Dissolved Oxygen Meter (Mettler Toledo, REFERENCES Worthington Columbus, OH). Preliminary experiments designed to test FA 1. Bueso-Ramos CE, Ferrajoli A, Medeiros LJ, Keating MJ, consumption that used palmitic acid or oleic acid Estrov Z. Aberrant morphology, proliferation, and apoptosis dissolved in ethanol, determined that the O consumption of B-cell chronic lymphocytic leukemia cells. Hematology. with 80 mM oleic acid or 80 mM palmitic acid (both 2004; 9:279–86. from Sigma-Aldrich) each is maximal therefore we used 2. Chiorazzi N, Ferrarini M. Evolving view of the in-vivo these concentrations in the following experiments. In kinetics of chronic lymphocytic leukemia B cells. Hematology each experiment we used CLL cells at a concentration 6 Am Soc Hematol Educ Program. 2006; 2006: 273–8, 512. of 2 × 10 cells/ml. The cells were incubated with a 3. Gibson SE, Leeman-Neill RJ, Jain S, Piao W, Cieply KM, minimum essential medium (MEM) with Hank’s salts Swerdlow SH. Proliferation centres of chronic lymphocytic and L-glutamine (Life Technologies, Carlsbad, CA) or leukaemia/small lymphocytic lymphoma have enhanced with phosphate buffered saline medium (Invitrogen) for expression of MYC protein, which does not result from 48 to 72 hours in tightly sealed T25 tissue culture flasks rearrangement or gain of the MYC gene. Br J Haematol. (Corning, Tewksbury, MA ) at 37° C in the presence 2016; 175:173–5. or absence of palmitic acid or oleic acid. The O meter 4. Rozovski U, Harris DM, Li P, Liu Z, Wu JY, Grgurevic S, probe was placed in the flask and the reading allowed Faderl S, Ferrajoli A, Wierda WG, Martinez M, stabilizing. Then, the dissolved O (dO ) level was 2 2 Verstovsek S, Keating MJ, Estrov Z. At High Levels, recorded. The probe was cleaned before it was re-used. Constitutively Activated STAT3 Induces Apoptosis of Measurements of dO were repeated at least three times Chronic Lymphocytic Leukemia Cells. J Immunol. 2016; for every data point. The dO of CLL cells transfected 196:4400–9. with CD36 siRNA, or incubated either with 2500 mg/ 5. Chiorazzi N. Cell proliferation and death: forgotten features ml CD36-neutralizing antibodies (Abcam), with 100 of chronic lymphocytic leukemia B cells. Best Pract Res µM of the LPL inhibitor orlistat, or 100 µM of the CD36 Clin Haematol. 2007; 20:399–413. inhibitor sulfosuccinimidyl oleate sodium (SSO) was assessed. 6. Neese RA, Misell LM, Turner S, Chu A, Kim J, Cesar D, We used the Student t-test when one experimental Hoh R, Antelo F, Strawford A, McCune JM, Christiansen M, condition was compared to non-treated (controls) and Hellerstein MK. Measurement in vivo of proliferation rates one-way ANOVA when 2 experimental conditions were of slow turnover cells by 2H2O labeling of the deoxyribose compared to controls. Statistical analyses were performed moiety of DNA. Proc Natl Acad Sci U S A. 2002; using GraphPad version 5 (San Diego, California, USA). 99:15345–50. 7. McCaw L, Shi Y, Wang G, Li YJ, Spaner DE. Low Density Annexin V/propidium iodide apoptosis assay Lipoproteins Amplify Cytokine-signaling in Chronic Lymphocytic Leukemia Cells. EBioMedicine. 2017; 15:24–35. The rate of cellular apoptosis was analyzed using 8. Rozovski U, Grgurevic S, Bueso-Ramos C, Harris DM, double staining with a Cy5-conjugated annexin V kit Li P, Liu Z, Wu JY, Jain P, Wierda W, Burger J, O'Brien S, and propidium iodide (PI; BD Biosciences) according Jain N, Ferrajoli A, et al. Aberrant LPL Expression, Driven to the manufacturer’s instructions. Briefly, 1 × 10 CLL by STAT3, Mediates Free Fatty Acid Metabolism in CLL cells were incubated for 24 hours in glucose-free MEM Cells. Mol Cancer Res. 2015; 13:944–53. (Life Technologies) supplemented with 10% FCS. After 9. Rozovski U, Hazan-Halevy I, Barzilai M, Keating MJ, incubation the cells were washed once with phosphate- Estrov Z. Metabolism pathways in chronic lymphocytic buffered saline and resuspended in 200 µL binding buffer leukemia. Leuk Lymphoma. 2016; 57:758–65. with 0.5 µg/ml annexin V-Cy5 and 2 µg/ml propidium 10. Martin CA, Longman E, Wooding C, Hoosdally SJ, iodide (PI). After incubation for 10 minutes in the dark Ali S, Aitman TJ, Gutmann DA, Freemont PS, Byrne B, at room temperature, the samples were analyzed on Linton KJ. Cd36, a class B scavenger receptor, functions a FACSCalibur flow cytometer (Becton Dickinson as a monomer to bind acetylated and oxidized low-density Biosciences). Cell viability was calculated as the lipoproteins. Protein Sci. 2007; 16:2531–41. percentage of annexin V-positive cells. www.oncotarget.com 21278 Oncotarget 11. Coort SL, Willems J, Coumans WA, van der Vusse GJ, Burger J, O'Brien S, Faderl S, et al. STAT3-activated Bonen A, Glatz JF, Luiken JJ. Sulfo-N-succinimidyl esters GM-CSFRalpha translocates to the nucleus and protects CLL of long chain fatty acids specifically inhibit fatty acid cells from apoptosis. Mol Cancer Res. 2014; 12:1267–82. translocase (FAT/CD36)-mediated cellular fatty acid uptake. 23. Li P, Harris D, Liu Z, Liu J, Keating M, Estrov Z. Stat3 Mol Cell Biochem. 2002; 239:213–9. activates the receptor tyrosine kinase like orphan receptor-1 gene in chronic lymphocytic leukemia cells. PLoS One. 12. Kerkhoff C, Sorg C, Tandon NN, Nacken W. Interaction 2010; 5:e11859. of S100A8/S100A9-arachidonic acid complexes with the 24. Demaria M, Camporeale A, Poli V. STAT3 and metabolism: scavenger receptor CD36 may facilitate fatty acid uptake how many ways to use a single molecule? Int J Cancer. by endothelial cells. Biochemistry. 2001; 40:241–8. 2014; 135:1997–2003. 13. Pohl J, Ring A, Korkmaz U, Ehehalt R, Stremmel W. FAT/ 25. Shulga N, Pastorino JG. GRIM-19-mediated translocation CD36-mediated long-chain fatty acid uptake in adipocytes of STAT3 to mitochondria is necessary for TNF-induced requires plasma membrane rafts. Mol Biol Cell. 2005; necroptosis. J Cell Sci. 2012; 125:2995–3003. 16:24–31. 26. Febbraio M, Abumrad NA, Hajjar DP, Sharma K, Cheng W, 14. Bilban M, Heintel D, Scharl T, Woelfel T, Auer MM, Pearce SF, Silverstein RL. A null mutation in murine CD36 Porpaczy E, Kainz B, Krober A, Carey VJ, Shehata M, reveals an important role in fatty acid and lipoprotein Zielinski C, Pickl W, Stilgenbauer S, et al. Deregulated metabolism. J Biol Chem. 1999; 274:19055–62. expression of fat and muscle genes in B-cell chronic 27. Nath A, Li I, Roberts LR, Chan C. Elevated free fatty acid lymphocytic leukemia with high lipoprotein lipase uptake via CD36 promotes epithelial-mesenchymal transition expression. Leukemia. 2006; 20:1080–8. in hepatocellular carcinoma. Sci Rep. 2015; 5:14752. 15. Hazan-Halevy I, Harris D, Liu Z, Liu J, Li P, Chen X, 28. Pascual G, Avgustinova A, Mejetta S, Martin M, Castellanos Shanker S, Ferrajoli A, Keating MJ, Estrov Z. STAT3 is A, Attolini CS, Berenguer A, Prats N, Toll A, Hueto JA, constitutively phosphorylated on serine 727 residues, binds Bescos C, Di Croce L, Benitah SA. Targeting metastasis- DNA, and activates transcription in CLL cells. Blood. 2010; initiating cells through the fatty acid receptor CD36. Nature. 115:2852–63. 2017; 541:41–5. 16. Heinemeyer T, Wingender E, Reuter I, Hermjakob H, Kel AE, 29. Kuda O, Pietka TA, Demianova Z, Kudova E, Cvacka J, Kel OV, Ignatieva EV, Ananko EA, Podkolodnaya OA, Kopecky J, Abumrad NA. Sulfo-N-succinimidyl oleate Kolpakov FA, Podkolodny NL, Kolchanov NA. Databases (SSO) inhibits fatty acid uptake and signaling for on transcriptional regulation: TRANSFAC, TRRD and intracellular calcium via binding CD36 lysine 164: SSO COMPEL. Nucleic Acids Res. 1998; 26:362–7. also inhibits oxidized low density lipoprotein uptake by 17. Aaronson DS, Horvath CM. A road map for those who don't macrophages. J Biol Chem. 2013; 288:15547–55. know JAK-STAT. Science. 2002; 296:1653–5. 30. Kusaka Y, Tanaka T, Okamoto F, Terasaki F, Matsunaga Y, 18. Abumrad NA, el-Maghrabi MR, Amri EZ, Lopez E, Miyazaki H, Kawamura K. Effect of sulfo-N-succinimidyl Grimaldi PA. Cloning of a rat adipocyte membrane protein palmitate on the rat heart: myocardial long-chain fatty acid implicated in binding or transport of long-chain fatty uptake and cardiac hypertrophy. J Mol Cell Cardiol. 1995; acids that is induced during preadipocyte differentiation. 27:1605–12. Homology with human CD36. J Biol Chem. 1993; 31. Glatz JF, Angin Y, Steinbusch LK, Schwenk RW, Luiken JJ. 268:17665–8. CD36 as a target to prevent cardiac lipotoxicity and insulin 19. Ibrahimi A, Sfeir Z, Magharaie H, Amri EZ, Grimaldi P, resistance. Prostaglandins Leukot Essent Fatty Acids. 2013; Abumrad NA. Expression of the CD36 homolog (FAT) in 88:71–7. fibroblast cells: effects on fatty acid transport. Proc Natl 32. Mansor LS, Sousa Fialho MDL, Yea G, Coumans WA, West Acad Sci U S A. 1996; 93:2646–51. JA, Kerr M, Carr CA, Luiken JJFP, Glatz JFC, Evans RD, 20. Frank DA, Mahajan S, Ritz J. B lymphocytes from Griffin JL, Tyler DJ, Clarke K, Heather LC. Inhibition of patients with chronic lymphocytic leukemia contain signal sarcolemmal FAT/CD36 by sulfo-N-succinimidyl oleate transducer and activator of transcription (STAT) 1 and rapidly corrects metabolism and restores function in the STAT3 constitutively phosphorylated on serine residues. J diabetic heart following hypoxia/reoxygenation. Cardiovasc Clin Invest. 1997; 100:3140–8. Res. 2017; 113:737–48. 33. Greenwalt DE, Scheck SH, Rhinehart-Jones T. Heart CD36 21. Rozovski U, Calin GA, Setoyama T, D'Abundo L, expression is increased in murine models of diabetes and in Harris DM, Li P, Liu Z, Grgurevic S, Ferrajoli A, Faderl S, mice fed a high fat diet. J Clin Invest. 1995; 96:1382–8. Burger JA, O'Brien S, Wierda WG, et al. Signal transducer and activator of transcription (STAT)-3 regulates microRNA 34. Ferrajoli A, Faderl S, Van Q, Koch P, Harris D, Liu Z, gene expression in chronic lymphocytic leukemia cells. Mol Hazan-Halevy I, Wang Y, Kantarjian HM, Priebe W, Cancer. 2013; 12:50. Estrov Z. WP1066 disrupts Janus kinase-2 and induces 22. Li P, Harris D, Liu Z, Rozovski U, Ferrajoli A, Wang Y, caspase-dependent apoptosis in acute myelogenous Bueso-Ramos C, Hazan-Halevy I, Grgurevic S, Wierda W, leukemia cells. Cancer Res. 2007; 67:11291–9. www.oncotarget.com 21279 Oncotarget 35. Harris DM, Hazan-Haley I, Coombes K, Bueso-Ramos 36. Liu Z, Hazan-Halevy I, Harris DM, Li P, Ferrajoli A, C, Liu J, Liu Z, Li P, Ravoori M, Abruzzo L, Han L, Faderl S, Keating MJ, Estrov Z. STAT-3 activates Singh S, Sun M, Kundra V, et al. Transformation of human NF-kappaB in chronic lymphocytic leukemia cells. Mol mesenchymal cells and skin fibroblasts into hematopoietic Cancer Res. 2011; 9:507–15. cells. PLoS One. 2011; 6:e21250. www.oncotarget.com 21280 Oncotarget http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Oncotarget Pubmed Central

STAT3-activated CD36 facilitates fatty acid uptake in chronic lymphocytic leukemia cells

Loading next page...
 
/lp/pubmed-central/stat3-activated-cd36-facilitates-fatty-acid-uptake-in-chronic-XSPwDIqDUK

References (39)

Publisher
Pubmed Central
Copyright
Copyright: © 2018 Rozovski et al.
ISSN
1949-2553
eISSN
1949-2553
DOI
10.18632/oncotarget.25066
Publisher site
See Article on Publisher Site

Abstract

www.oncotarget.com Oncotarget, 2018, Vol. 9, (No. 30), pp: 21268-21280 Research Paper STAT3-activated CD36 facilitates fatty acid uptake in chronic lymphocytic leukemia cells 1,2 1 1 1 1 Uri Rozovski , David M. Harris , Ping Li , Zhiming Liu , Preetesh Jain , Alessandra 1 1 1 1 1 Ferrajoli , Jan Burger , Phillip Thompson , Nitin Jain , William Wierda , Michael J. 1 1 Keating and Zeev Estrov Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA Institute of Hematology, Davidoff Cancer Center, Rabin Medical Center, Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel Correspondence to: Zeev Estrov, email: zestrov@mdanderson.org Keywords: CLL; CD36; metabolism; STAT3 Received: January 31, 2018 Accepted: March 21, 2018 Published: April 20, 2018 Copyright: Rozovski et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License 3.0 (CC BY 3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. ABSTRACT Although several studies established that unlike normal B cells chronic lymphocytic leukemia (CLL) cells metabolize fatty acids (FA), how CLL cells internalize FA is poorly understood. Because in various cell types CD36 facilitates FA uptake, we wondered whether a similar mechanism is operative CLL. We found that CD36 levels are higher in CLL cells than in normal B cells, and that small interfering RNA, CD36 neutralizing antibodies or sulfosuccinimidyl oleate (SSO) that inhibits CD36 significantly reduced the oxygen consumption of CLL cells incubated with FA. Because CD36 is oeverexpressed and STAT3 is constitutively activated in CLL cells, we wondered whether STAT3 induces CD36 expression. Sequence analysis identified putative STAT3 binding sites in the CD36 gene promoter. Chromatin immunoprecipitation and an electrophoretic mobility shift assay revealed that STAT3 binds to the CD36 gene promoter. A luciferase assay and STAT3-small hairpin RNA, that significantly decreased the levels of CD36 in CLL cells, established that STAT3 activates the transcription of the CD36 gene. Furthermore, SSO induced a dose-dependent apoptosis of CLL cells. Taken together, our data suggest that STAT3 activates CD36 and that CD36 facilitates FA uptake in CLL cells. Whether CD36 inhibition would provide clinical benefits in CLL remains to be determined. labeling studies it is estimated that approximately 1% of INTRODUCTION the CLL clone expands daily [5], an expansion rate that is remarkably similar to that of normal adipocytes [6]. Chronic lymphocytic leukemia (CLL) is We and others [7, 8] have recently shown that, characterized by a gradual increase in the number unlike normal B cells, CLL cells adopt metabolic of circulating mature appearing lymphocytes. For pathways that are operative in adipocytes and myocytes. more than four decades CLL was viewed as a disease Specifically, we found that CLL cells store lipids in characterized by an accumulation of long-lived mature intracytoplasmic vacuoles and, by aberrantly expressing looking neoplastic B-lymphocytes that do not proliferate lipoprotein lipase (LPL), utilize fatty acids (FA) for the and do not die. Studies conducted in recent years clearly production of chemical energy [8, 9]. We have also shown demonstrated that CLL cells do proliferate [1–3] and that LPL expression and FA metabolism in CLL is driven undergo spontaneous apoptosis [4]. The proliferation by constitutive activation of the signal transducer and rate of CLL cells varies, however based on heavy water www.oncotarget.com 21268 Oncotarget activator of transcription (STAT)-3. However, what is the (Figure 2B). To confirm that these DNA fragments bind source of FA and what mechanisms are recruited to enable STAT3 we performed EMSA using biotinylated DNA FA entry into CLL cells is not completely understood. probes corresponding to the binding sites +187 bp – +196 CD36, also known as fatty acid translocase protein, bp, +363 bp – +374 bp, and +382 bp – +391 bp, and -93 is a multi-ligand glycoprotein that is expressed on the bp – -83 bp. Confirming the ChIP data, we found that extracellular membrane and facilitates FA uptake in CLL cell nuclear protein extracts from 2 different patients various cells such as myocytes and adipocytes [10] in formed complexes with STAT3 binding sites +187 bp – +196 bp, +363 bp – +374 bp, and +382 bp – +391 bp, but which lipid uptake is markedly impaired when CD36 not with the putative binding site -93 bp – -83 bp and that function is compromised [11–13]. Because CLL cells’ metabolic pathways resembles those of adipocytes [14] the binding was significantly attenuated by the addition and the CD36 gene harbors putative STAT3-binding sites, of excess unlabeled probe or by anti-STAT3 antibodies we hypothesized that STAT3 induces CD36 cell surface (Figure 2C). To further delineate these findings we expression and that CD36 facilitates FA uptake in CLL performed a luciferase assay of MM1 cells in which we cells. concentrated on putative STAT3 binding sites that were in close proximity to the CD36 gene start codon. We found the highest luciferase activity in IL-6 stimulated MM1 RESULTS cells transfected with the promoter fragment that included the 3 active binding sites (Figure 2D). Furthermore, we CLL cells express high CD36 protein levels of found that STAT3-shRNA downregulated mRNA levels both of STAT3 and CD36 by 9 and 6-fold, respectively, To assess the levels of CD36 protein in CLL cells we and as a result, significantly reduced STAT3 and CD36 performed Western blot analysis of PB-derived CLL cells protein levels, confirming that STAT3 induces the from 6 randomly chosen CLL patients and, as control, expression of CD36 in CLL cells (Figure 2E). CD19+ B cells from the PB of two healthy individuals. As shown in Figure 1A, protein levels of CD36 were readily detected in all CLL-patient samples and their levels were CD36 facilitates FA intake and metabolism in significantly higher than those of CD19+ B cells obtained CLL cells from healthy individuals. As in previous studies [15], Because CLL cells utilize FA and CD36 plays a key phosphoserine STAT3 was detected in CLL cells but its role in FA uptake in various cell types [18, 19], we sought levels were undetectable in normal CD19+ B cells. To to determine whether CD36 contributes to FA uptake in confirm that CLL cells express CD36 cell membrane CLL cells. We cultured CLL cells in tightly closed flaks in protein we performed flow cytometry and detected serum- and glucose-free medium and measured the oxygen CD36 protein on the surface of PB CLL cells from 6 concentration prior to and after adding FA, assuming that randomly selected CLL patients (Figure 1B). However, the if the cells consume the FA, oxygen levels in the culture percentages of cells co-expressing CD19, CD5, and CD36 medium will drop. As expected, when FA were added to antigens varied among the different patients and were 14, culture the levels of oxygen dissolved in the culture media 25, 32.9, 37, 41, and 51, respectively. To confirm these were markedly reduced whereas the dO levels of CLL cells findings we used confocal microscopy and found that, like transfected with CD36 siRNA and incubated with oleic in other cell types, CD36 is detected on the surface but acid, remained significantly higher than the dO levels in not in the cytoplasm or nucleus of CLL cells (Figure 1C). 2 the medium of non-transfected or GPDH-transfected CLL cells (Figure 3A). Similarly, the dO levels of CLL cells STAT3 binds to the CD36 gene promoter and 2 incubated with oleic acid or palmitic acid in the absence activates the CD36 gene of CD36 neutralizing antibodies significantly dropped whereas dO levels of CLL cells incubated with oleic Because STAT3 is constitutively activated in CLL acid or palmitic acid in the presence of CD36 neutralizing cells [15], we wondered whether overexpression of CD36 antibodies remained unchanged (Figure 3B). Likewise, is driven by STAT3. Using the TFSEARCH database [16] the irreversible CD36 inhibitor SSO and the LPL inhibitor we identified 4 GAS-like elements, known as putative orlistat significantly reduced CLL cell O consumption STAT3-binding sites [17], within 600 bp upstream of and the effect of both inhibitors was significantly additive the CD36 gene start codon (Figure 2A). To determine (Figure 3C). Furthermore, SSO induced apoptosis of CLL whether STAT3 binds to any of these GAS-like elements cells in a dose-dependent manner (Figure 3D). we performed ChIP. Our ChIP analysis revealed that 3 DNA fragments, whose primers amplified the regions of the putative STAT3 binding sites +187 bp – +196 bp, DISCUSSION +363 bp – +374 bp, and +382 bp – +391 bp, but not the region of the putative STAT3 binding site -93 bp – -83 Here we show that CLL cells express high levels of bp, co-immunoprecipitated with anti-STAT3 antibodies CD36 cell surface protein, that overexpression of CD36 is www.oncotarget.com 21269 Oncotarget driven by STAT3-mediated activation of the CD36 gene, with proliferation capacity [8, 21–23]. The common and that CD36 facilitates uptake of FA by CLL cells. constitutive activation of STAT3 in all CLL patients Although CLL is a heterogeneous disease, in all suggests that STAT3 is involved in a core function that patients, regardless of clinical characteristics, disease affects CLL cell survival. burden, cytogenetic abnormalities or IgHv gene mutation Several observations linked STAT3 to cellular status, STAT3 is constitutively phosphorylated on serine metabolism in malignant and non-malignant cells [24]. In 727 residues [15, 20]. Phosphoserine STAT3 activates a cancer cells STAT3 emerged as a as a key regulator of diverse repertoire of coding and non-coding genes which metabolism that integrates signals via both mitochondrial protect CLL cells from apoptosis and provide the cells and nuclear activities [25]. We have recently shown that www.oncotarget.com 21270 Oncotarget Figure 1: CLL cells express high levels of CD36. (A) Western blot analysis of CLL cells from 6 different patients and CD19+ normal B cells from 2 normal controls. In all patients samples CD36 was detected at higher levels than in normal B cells. Phosphoserine STAT3 was detected in all patient samples but not in CD19+ normal B cells and levels of unphosphorylated STAT3 were higher in CLL patient samples than in CD19+ normal B cells. Actin was used as loading control and densitometry units were normalized to actin units in the corresponding lanes. (B) CLL cells co-express CD19, CD5 and CD36. Left panel: A representative figure of flow cytometry analysis performed on CLL cells from 6 different patients is depicted. Right panel: Isotype controls. The black line is Isotype control and the red line is the corresponding antibody. Percentages represent the differences between the isotype and the specific antibody. The precentages are the (C) Confocal microscopic images (×400) of freshly isolated CLL cells (right) and normal B cells (left) stained with anti-CD36 antibodies (green) and Evans blue (red) showing CD36 on the cell surface of CLL cells but not on normal B cells. by inducing LPL production, STAT3 promotes uptake FA uptake was found to promote metastatic development and metabolism of lipid particles [8]. In various cell in a subpopulation of human oral carcinoma cells that types including CLL cells, lipid particles are stored in express high levels of CD36 [28]. vacuoles and following additional processing triglycerides To evaluate the role of CD36 in CLL cells we used undergo hydrolysis into FA generating substrates that enter three different strategies. We downregulated CD36 mRNA the Krebs cycle. Here we show that CLL cells utilize a levels by transfecting CLL cells with CD36 siRNA and, complementary strategy for a direct uptake of circulating in other experiments, blocked CD36 activity with CD36- FA by STAT3-mediated expression of CD36. neutralizing antibodies or SSO. Using these three different Studies in CD36-null mice identified CD36 as a methods we consistently found that inhibition of CD36 facilitator of FA uptake [26]. In addition, several groups disrupted FA metabolism, confirming that in CLL cells reported that CD36-mediated FA uptake plays a role in CD36 is required for FA uptake. the pathogenesis of various neoplasms. For example, Nath Sulfo-N-succinimidyl oleate (SSO) attaches to et al. showed that in hepatocellular carcinoma CD36- the CD36 binding pocket and, by doing so, it inhibits mediated uptake of FA induces epithelial to mesenchymal CD36-mediated FA uptake [29]. Therefore SSO has been transition and metastasis [27]. Likewise, CD36-mediated extensively used to study the role of CD36. For example, www.oncotarget.com 21271 Oncotarget www.oncotarget.com 21272 Oncotarget Figure 2: STAT3 binds to and activates the CD36 gene promoter. (A) Sequence analysis of the CD36 gene promoter upstream of the start codon (green) revealed 4 putative STAT3 binding sites (red). Upstream putative STAT3 biding sites are not shown because in most experiments we concentrated our efforts on studying binding sites that are in proximity to the CD36 start codon. (B) Using the ChIP method, CLL-cell chromatin fragments pulled down by anti-STAT3 antibodies were analyzed by PCR using primers directed at the 4 putative STAT3 binding sites upstream the CD36 gene start codon. As shown, anti-STAT3 antibodies co-immunoprecipitated the DNA detected by primers +236 bp – +406 bp, +120 bp – +256 bp (amplifying regions of the STAT3 putative binding sites +187 bp – +196 bp, +363 bp – 374 bp, and +382 bp – +391 bp), and -527 bp – -355 bp but not by primers -203 bp – +1 bp (amplifying the region of the putative binding site -93 bp – 83 bp). (C) Using EMSA, biotin-labeled CD36-DNA probes were incubated with CLL cells’ protein extract from 2 patients. The EMSA demonstrated that CLL cell nuclear protein extracts bound to the CD36 gene promoter at regions that include the putative STAT3 binding sites +187 bp – +196 bp, +363 bp – +374 bp, and +382 bp – +391 bp, but not the region that includes the putative binding site -93 bp – 83 bp, and that the addition of excess unlabeled probe or anti-STAT3 antibodies attenuated the binding. (D) The luciferase activity of IL-6-stimulated MM-1 cells was assessed 24 hours after transfection with the 3 depicted DNA fragments containing putative STAT3 binding sites in close proximity to the start codon is shown. The luciferase activity of each of the human CD36 promoter constructs was determined by calculating the constructs’ luciferase activity relative to the activity of the Renilla luciferase produced by the pRL-SV40 control vector. The luciferase activity of unstimulated MM1 cells (not shown) was similar to that of the pRL-SV40 control vector. The highest luciferase activity, compared to the pGL4.17 (control), in IL-6-stimulated MM1 cells transfected with the promoter fragment that included the 3 active binding sites (-203 bp; P = 0.0002). A lower albeit increased activity was observed in cells transfected with the promoter fragment that included 2 active binding sites (+236 bp; P = 0.009). There was no significant difference in the luciferase activity of fragments +122 bp and +236 bp (P = 0.12). (E) Infection of CLL cells with STAT3-shRNA, downregulated mRNA levels both of STAT3 and CD36 by 9 and 6 fold, respectively (left panel), and significantly reduced protein levels of STAT3, phsophoserine STAT3 and CD36 (right panel). The figure depicts representative results of 3 different experiments. www.oncotarget.com 21273 Oncotarget www.oncotarget.com 21274 Oncotarget Figure 3: CLL cell FA intake and metabolism is CD36-dependent. (A) CLL cells were transfected with CD36-siRNA or GAPDH and incubated with 80 mM oleic acid in a serum-free, glucose-free medium in tightly sealed flask for 48 hours. Transfection (at a transfection efficiency of 30% as assessed by flow cytometry; left panel) with CD36-siRNA, but not GAPDH, significantly reduces CD36 mRNA levels compared to transfection with GAPD (CTRL), and the dO levels were significantly reduced in the medium of non- transfected or GAPDH-transfected CLL cells but not in the medium of CD36-siRNA-transfected cells (right panel). Representative results from 3 different experiments are depicted. (B) CD36 neutralizing antibodies inhibited FA metabolism in CLL cells. CLL cells from two patients were incubated in the presence or absence of 80 mM palmitic acid (PA) or oleic acid (OA) with or without CD36 neutralizing antibodies (Ab) in sealed tissue culture flasks. The relative dO in the culture media was assessed prior to and after 48 hours of incubation. All control cultures contained ethanol at the same concentration as in palmitic acid and oleic acid. To compare dO we assessed the difference in O concentration before and after the incubation. The relative difference in dO before after adding FA is depicted after the 2 2 dO in the controls were set to 1. As shown, the CD36-neutralizing antibodies did not affect the cells’ oxygen consumption whereas in the presence of PA or OA the cells consumed O , and this effect was reversed when neutralizing antibodies were added to culture. (C) CLL cells from 2 patients were incubated for 48 hours in the presence or absence of 80 mM palmitic acid (PA) with or without the LPL inhibitor orlistat, the CD36 inhibitor SSO or both. The culture media concentration was assessed prior to and after 48 hours of incubation. As shown in the presence of PA dO was reduced, but less when SSO or orlistat were added to culture and the effect of SSO and orlistat was additive, indicating that these agents reduced FA-dependent metabolism. (D) Apoptosis was assessed in CLL cells incubated with or without increasing concentrations of SSO using flow cytometry with annexin V/PI staining. Upper panel: Representative flow cytometry plots of the below depicted three experiments. Lower panel: Results of 3 different experiments, using cells from 3 different patients are depicted as percent of the apoptosis rate in cells incubated without SSO (control). SSO was found to effectively block CD36-mediated In conclusion, unlike their normal resting B cell fatty acid uptake into cardiomyocytes [11, 30, 31] and counterparts, CLL cells utilize FA. FA uptake in CLL to restore of diabetic heart’s function following hypoxia/ cells is facilitated by STAT3-enhanced CD36 expression. reoxygenation [32]. However, SSO is chemically instable Whether inhibition of CD36-dependent FA uptake might and therefore cannot be used in its current form for long- have therapeutic benefit in CLL remains to be determined. term experiments or in vivo studies [11]. The use of CD36 neutralizing antibodies is an attractive alternative that has MATERIALS AND METHODS been used in animal models [31, 33]. CD36 neutralizing antibodies were shown to eliminate lymph node metastasis Patients’ characteristics in mice that were inoculated with tumor cells [28], suggesting that it may also be applicable in treatment of Peripheral blood samples were obtained from 19 CLL-affected lymph nodes. patients with CLL who were treated at The University of www.oncotarget.com 21275 Oncotarget Texas MD Anderson Cancer Center Clinic. The study was Confocal microscopy approved by our Institutional Review Board and patients’ A previously described method was used [35]. Informed Consent was obtained prior to sample collection. CLL low-density cells were incubated in microtubes Clinical characteristics of all patients that participated in in PBS supplemented with 5% bovine serum albumin this study are depicted in Table 1. (Cell Signaling Technology,). After 1 hour of incubation, the cells were washed three times with PBS and then Cell fractionation incubated with mouse anti-CD36 antibodies (Novus Scientific, Littleton, CO) for 1 hour, washed in PBS, re- PB cells were fractionated using Ficoll Hypaque suspended in a 0.1% solution of Evans blue dye (Sigma- 1077 (Sigma, St. Louis, MO). The low-density Aldrich) for 5 minutes, and washed in PBS to remove cellular fraction was used immediately or frozen for unbound dye. The cells were resuspended in PBS and additional studies. More than 95% of the peripheral 0.4 placed into µ-slide VI chamber slides (ibidi LLC, blood lymphocytes obtained from these patients were Verona, WI) for microscopic analysis. The slides were CD19+/CD5+, as assessed by flow cytometry (Becton, viewed using an Olympus FluoView 500 Laser Scanning Dickinson and Company, Franklin Lakes, NJ). As control Confocal Microscope (Olympus America, Center Valley, studies we obtained from the Central Blood Bank left- PA), and images were analyzed using the FluoView over buffy coats of healthy blood donors. After Ficoll- software (Olympus America). Hypaque fractionation, the donors’ B cells were isolated using Miltenyi CD19-coated beads according to the manufacturer’s instructions (Miltenyi Biotec, Bergisch Chromatin immunoprecipitation (ChIP) assay Gladbach, Germany). A chromatin immunoprecipitation (ChIP) assay was performed using a SimpleChIP Enzymatic Chromatin IP Western immunoblotting Kit (Cell Signaling Technology, Boston, MA) according to the manufacturer’s instructions. Briefly, cells were Western immunoblotting was performed as cross-linked with 1% formaldehyde for 10 minutes at previously described [34]. Briefly, CLL cell extract was room temperature and then harvested and incubated on ice prepared. The protein concentration was determined for 10 minutes in lysis buffer. Nuclei were pelleted and using a Micro BCA protein assay reagent kit (Thermo digested with micrococcal nuclease. Following sonication Scientific, Pierce, Rockford, IL). Cell lysates were and centrifugation, sheared chromatin was incubated with denatured and following electrophoresis transferred anti-STAT3 or rabbit serum (negative control) overnight at to a nitrocellulose membranes. The membranes were 4° C. Then, protein G beads were added, and the chromatin incubated with either monoclonal mouse anti-human was incubated for 2 hours in rotation. Antibody-bound STAT3 (BD Bioscience, Palo Alto, CA), polyclonal protein-DNA complexes were eluted and subjected to rabbit anti-human phosphoserine (serine 727) STAT3 PCR analysis. The primers to amplify the human CD36 (Cell Signaling Technology, Beverly, MA), or rabbit promoter were F’: -201 and R: +1, which generate a 200- anti-human CD36 (Pierce, Thermo Scientific, Waltham, bp product that covers the GAS binding site-93 to -83 bp MA) antibodies and Horseradish peroxidase-conjugated upstream of CD36; F’: +120 and R’: +256, which generate secondary antibodies (GE Healthcare, Amersham, a 136-bp product that covers the GAS binding site +187 to Buckinghamshire, UK). Proteins were visualized via +196 bp upstream of CD36; F’: +236 and R’: +406, which an enhanced chemiluminescence detection system (GE generate a 170-bp product that covers the GAS binding Healthcare). site +363 to +374 bp upstream of CD36; and F’: +355 and R’: +427, which generate a 72-bp product that covers the Flow cytometry analysis GAS binding site +382 to +391. CLL cells were fixed in 2% paraformaldehyde for 10 minutes and permeabilized overnight at –20° C. Electrophoretic mobility shift assay (EMSA) Before staining cells were washed three times in PBS Non-denatured cellular nuclear extracts were with 2% FBS. Cells were then stained with CD19 (BD prepared using a NE-PER extraction kit (Thermo Biosciences, San Jose CA), CD5 (BD Biosciences), Scientific Pierce, Rockford, IL). Nuclear protein extracts and CD36 (BD Biosciences) as well as appropriate were incubated with biotin-labeled CD36 promoters’ isotypic controls. Cells were analyzed on a FacsCaliber DNA probes in binding buffer for 30 minutes on ice. All flow cytometer (BD Biosciences) and data analysis probes which target GAS binding sites were synthesized performed using CellQuest software (BD Biosciences). by Sigma-Genosys (The Woodlands, TX). Following Graphics were created with CellQuest (BD Biosciences) incubation, the samples were separated on a 5% and WinList (Verity Software House, Topsham, ME) polyacrylamide gel, transferred onto a nylon membrane, software. www.oncotarget.com 21276 Oncotarget Table 1: Patient characteristics (n = 25) Characteristic Measure/Category Overall Age, years Median (range) 58 (40–78) WBC × 10 /L Median (range) 27 (7–55) ALC × 10 /L Median (range) 22 (1–48) Rai stage (0, 1–2/3–4) (%) (95/5) CD38 ≤30/>30 (%) 84/16 (%) Zap-70 Negative/Positive 78/22 (%) β2M (mg/L) (</≥ 4 mg/L) 95/5 (%) IGHV mutation (M/UM) 78/22 (%) FISH result, n del17p/11q/T12/13q/Negative 1/3/3/10/2 Karyotype, n Diploid/Non-diploid/Complex/not done 11/7/0/1 Survival status, (n) Alive/Dead 19/0 WBC, white blood cell count; ALC, absolute lymphocyte count; β2M, beta-2 microglobulin; M, mutated; UM, unmutated; FISH, fluorescence in situ hybridization. and fixed on the membrane via ultraviolet cross-linking. Inghirami, (department of pathology, University of The biotin-labeled probe was detected with strepavidin- Torino, Italia) using the Superfect transfection reagent horseradish peroxidase (Gel-Shift Kit; Panomics, Fremont, (QIAGEN Inc.) as previously described [15]. The 293T CA). The control consisted of 7-fold excess unlabeled cold cell culture medium was replaced after 16 hours and probe. To test the effect of STAT3, anti-STAT3 antibodies collected after 48 hours. Then, the culture medium was (BD Bioscience) mouse IgG1 (BD Bioscience) were filtered through a 45-μm syringe filter to remove floating added with the nuclear extracts [15, 36]. cells, the lentivirus was concentrated by filtration through an Amicon Ultra centrifugal filter device (EMD), and the concentrated supernatant was used to transfect CLL cells. Transfection of MM-1 cells with CD36 gene CLL cells (5 × 10 /ml) were incubated in six-well plates promoter fragments and luciferase assay (Becton, Dickinson and Company, Franklin Lakes, NJ) Four different CD36 promoter fragments, in 2 mL of Dulbecco’s modified Eagle medium (DMEM) corresponding to the above described putative STAT3- supplemented with 10% fetal bovine serum (FBS) binding sites, were transfected into MM-1 cells by using and were transfected with 100 μl of viral supernatant. electroporation. MM-1 cells were used because in these Polybrene (10 ng/mL) was added to the viral supernatant cells STAT3 phosphorylation is induced by extracellular at a ratio of 1:1000 (v/v). Transfection efficiency was signals such as IL-6. Each construct harbored a luciferase measured after 48 hours and ranged between 40% and 50% reporter gene and a CD36 promoter fragment that included (calculated on the basis of the ratio of propidium iodide– at least 1 γ-interferon activation sequence (GAS)-like negative and GFP-positive cells). These experiments elements. The luciferase activity of unstimulated or IL- were conducted using an upgraded FACSCalibur flow 6-stimulated MM-1 cells was assessed 24 hours after cytometer (Becton, Dickinson). transfection using a Dual-Luciferase Reporter Assay System (Promega) and a Sirius luminometer V3.1 Transfection of CLL cells with CD36 small (Berthold Detection Systems, Pforzheim, Germany). The interfering RNA (siRNA) luciferase activity of each of the human CD36 promoter constructs was determined by calculating the constructs’ Hundred μM human CD36-siRNA, 10 μl FAM- luciferase activity relative to the activity of the Renilla labeled siRNA, targeting the human glyceraldehyde luciferase produced by the pRL-SV40 control vector. 3-phosphate dehydrogenase, or a scrambled control (Applied Biosystems, Foster City, CA) were added to 10 μl siPORT NeoFX transfection reagent diluted in 50 µl Generation of GFP-conjugated lentiviral STAT3 Opti-MEMI reduced serum medium (Thermo Fisher short hairpin RNA (shRNA) and infection of Scientific), and incubated at room temperature for 10 CLL cells minutes. The transfection agents were incubated at room temperature with 1 × 10 CLL cells suspended in 0.2 293T cells were co-transfected with GFP-conjugated ml Opti-MEM I medium. After 1 hour of incubation, lentiviral STAT3 shRNA or a GFP-conjugated empty electroporation (Bio-Rad Laboratories) was performed and lentiviral vector and with packaging vectors (pCMV delta the cells were incubated in RPMI supplemented with 10% R8.2 and pMDG generously provided by Dr. Giorgio www.oncotarget.com 21277 Oncotarget FBS for 24 hours. Transfection efficiency of the FAM- CONFLICTS OF INTEREST conjugated siRNA was assessed on a FACSCalibur flow cytometer (Becton Dickinson Biosciences). All authors report no conflicts of interest. Measurement of cellular O consumption FINANCIAL SUPPORT Because FA metabolism increases O consumption, This study was supported by the Cancer Center palmitic acid and oleic acid utilization was assessed Support Grant from the NIH/NCI, P30 CA016672 grant. by measuring the level of dissolved O (dO ) using the 2 2 SevenGo pro Dissolved Oxygen Meter (Mettler Toledo, REFERENCES Worthington Columbus, OH). Preliminary experiments designed to test FA 1. Bueso-Ramos CE, Ferrajoli A, Medeiros LJ, Keating MJ, consumption that used palmitic acid or oleic acid Estrov Z. Aberrant morphology, proliferation, and apoptosis dissolved in ethanol, determined that the O consumption of B-cell chronic lymphocytic leukemia cells. Hematology. with 80 mM oleic acid or 80 mM palmitic acid (both 2004; 9:279–86. from Sigma-Aldrich) each is maximal therefore we used 2. Chiorazzi N, Ferrarini M. Evolving view of the in-vivo these concentrations in the following experiments. In kinetics of chronic lymphocytic leukemia B cells. Hematology each experiment we used CLL cells at a concentration 6 Am Soc Hematol Educ Program. 2006; 2006: 273–8, 512. of 2 × 10 cells/ml. The cells were incubated with a 3. Gibson SE, Leeman-Neill RJ, Jain S, Piao W, Cieply KM, minimum essential medium (MEM) with Hank’s salts Swerdlow SH. Proliferation centres of chronic lymphocytic and L-glutamine (Life Technologies, Carlsbad, CA) or leukaemia/small lymphocytic lymphoma have enhanced with phosphate buffered saline medium (Invitrogen) for expression of MYC protein, which does not result from 48 to 72 hours in tightly sealed T25 tissue culture flasks rearrangement or gain of the MYC gene. Br J Haematol. (Corning, Tewksbury, MA ) at 37° C in the presence 2016; 175:173–5. or absence of palmitic acid or oleic acid. The O meter 4. Rozovski U, Harris DM, Li P, Liu Z, Wu JY, Grgurevic S, probe was placed in the flask and the reading allowed Faderl S, Ferrajoli A, Wierda WG, Martinez M, stabilizing. Then, the dissolved O (dO ) level was 2 2 Verstovsek S, Keating MJ, Estrov Z. At High Levels, recorded. The probe was cleaned before it was re-used. Constitutively Activated STAT3 Induces Apoptosis of Measurements of dO were repeated at least three times Chronic Lymphocytic Leukemia Cells. J Immunol. 2016; for every data point. The dO of CLL cells transfected 196:4400–9. with CD36 siRNA, or incubated either with 2500 mg/ 5. Chiorazzi N. Cell proliferation and death: forgotten features ml CD36-neutralizing antibodies (Abcam), with 100 of chronic lymphocytic leukemia B cells. Best Pract Res µM of the LPL inhibitor orlistat, or 100 µM of the CD36 Clin Haematol. 2007; 20:399–413. inhibitor sulfosuccinimidyl oleate sodium (SSO) was assessed. 6. Neese RA, Misell LM, Turner S, Chu A, Kim J, Cesar D, We used the Student t-test when one experimental Hoh R, Antelo F, Strawford A, McCune JM, Christiansen M, condition was compared to non-treated (controls) and Hellerstein MK. Measurement in vivo of proliferation rates one-way ANOVA when 2 experimental conditions were of slow turnover cells by 2H2O labeling of the deoxyribose compared to controls. Statistical analyses were performed moiety of DNA. Proc Natl Acad Sci U S A. 2002; using GraphPad version 5 (San Diego, California, USA). 99:15345–50. 7. McCaw L, Shi Y, Wang G, Li YJ, Spaner DE. Low Density Annexin V/propidium iodide apoptosis assay Lipoproteins Amplify Cytokine-signaling in Chronic Lymphocytic Leukemia Cells. EBioMedicine. 2017; 15:24–35. The rate of cellular apoptosis was analyzed using 8. Rozovski U, Grgurevic S, Bueso-Ramos C, Harris DM, double staining with a Cy5-conjugated annexin V kit Li P, Liu Z, Wu JY, Jain P, Wierda W, Burger J, O'Brien S, and propidium iodide (PI; BD Biosciences) according Jain N, Ferrajoli A, et al. Aberrant LPL Expression, Driven to the manufacturer’s instructions. Briefly, 1 × 10 CLL by STAT3, Mediates Free Fatty Acid Metabolism in CLL cells were incubated for 24 hours in glucose-free MEM Cells. Mol Cancer Res. 2015; 13:944–53. (Life Technologies) supplemented with 10% FCS. After 9. Rozovski U, Hazan-Halevy I, Barzilai M, Keating MJ, incubation the cells were washed once with phosphate- Estrov Z. Metabolism pathways in chronic lymphocytic buffered saline and resuspended in 200 µL binding buffer leukemia. Leuk Lymphoma. 2016; 57:758–65. with 0.5 µg/ml annexin V-Cy5 and 2 µg/ml propidium 10. Martin CA, Longman E, Wooding C, Hoosdally SJ, iodide (PI). After incubation for 10 minutes in the dark Ali S, Aitman TJ, Gutmann DA, Freemont PS, Byrne B, at room temperature, the samples were analyzed on Linton KJ. Cd36, a class B scavenger receptor, functions a FACSCalibur flow cytometer (Becton Dickinson as a monomer to bind acetylated and oxidized low-density Biosciences). Cell viability was calculated as the lipoproteins. Protein Sci. 2007; 16:2531–41. percentage of annexin V-positive cells. www.oncotarget.com 21278 Oncotarget 11. Coort SL, Willems J, Coumans WA, van der Vusse GJ, Burger J, O'Brien S, Faderl S, et al. STAT3-activated Bonen A, Glatz JF, Luiken JJ. Sulfo-N-succinimidyl esters GM-CSFRalpha translocates to the nucleus and protects CLL of long chain fatty acids specifically inhibit fatty acid cells from apoptosis. Mol Cancer Res. 2014; 12:1267–82. translocase (FAT/CD36)-mediated cellular fatty acid uptake. 23. Li P, Harris D, Liu Z, Liu J, Keating M, Estrov Z. Stat3 Mol Cell Biochem. 2002; 239:213–9. activates the receptor tyrosine kinase like orphan receptor-1 gene in chronic lymphocytic leukemia cells. PLoS One. 12. Kerkhoff C, Sorg C, Tandon NN, Nacken W. Interaction 2010; 5:e11859. of S100A8/S100A9-arachidonic acid complexes with the 24. Demaria M, Camporeale A, Poli V. STAT3 and metabolism: scavenger receptor CD36 may facilitate fatty acid uptake how many ways to use a single molecule? Int J Cancer. by endothelial cells. Biochemistry. 2001; 40:241–8. 2014; 135:1997–2003. 13. Pohl J, Ring A, Korkmaz U, Ehehalt R, Stremmel W. FAT/ 25. Shulga N, Pastorino JG. GRIM-19-mediated translocation CD36-mediated long-chain fatty acid uptake in adipocytes of STAT3 to mitochondria is necessary for TNF-induced requires plasma membrane rafts. Mol Biol Cell. 2005; necroptosis. J Cell Sci. 2012; 125:2995–3003. 16:24–31. 26. Febbraio M, Abumrad NA, Hajjar DP, Sharma K, Cheng W, 14. Bilban M, Heintel D, Scharl T, Woelfel T, Auer MM, Pearce SF, Silverstein RL. A null mutation in murine CD36 Porpaczy E, Kainz B, Krober A, Carey VJ, Shehata M, reveals an important role in fatty acid and lipoprotein Zielinski C, Pickl W, Stilgenbauer S, et al. Deregulated metabolism. J Biol Chem. 1999; 274:19055–62. expression of fat and muscle genes in B-cell chronic 27. Nath A, Li I, Roberts LR, Chan C. Elevated free fatty acid lymphocytic leukemia with high lipoprotein lipase uptake via CD36 promotes epithelial-mesenchymal transition expression. Leukemia. 2006; 20:1080–8. in hepatocellular carcinoma. Sci Rep. 2015; 5:14752. 15. Hazan-Halevy I, Harris D, Liu Z, Liu J, Li P, Chen X, 28. Pascual G, Avgustinova A, Mejetta S, Martin M, Castellanos Shanker S, Ferrajoli A, Keating MJ, Estrov Z. STAT3 is A, Attolini CS, Berenguer A, Prats N, Toll A, Hueto JA, constitutively phosphorylated on serine 727 residues, binds Bescos C, Di Croce L, Benitah SA. Targeting metastasis- DNA, and activates transcription in CLL cells. Blood. 2010; initiating cells through the fatty acid receptor CD36. Nature. 115:2852–63. 2017; 541:41–5. 16. Heinemeyer T, Wingender E, Reuter I, Hermjakob H, Kel AE, 29. Kuda O, Pietka TA, Demianova Z, Kudova E, Cvacka J, Kel OV, Ignatieva EV, Ananko EA, Podkolodnaya OA, Kopecky J, Abumrad NA. Sulfo-N-succinimidyl oleate Kolpakov FA, Podkolodny NL, Kolchanov NA. Databases (SSO) inhibits fatty acid uptake and signaling for on transcriptional regulation: TRANSFAC, TRRD and intracellular calcium via binding CD36 lysine 164: SSO COMPEL. Nucleic Acids Res. 1998; 26:362–7. also inhibits oxidized low density lipoprotein uptake by 17. Aaronson DS, Horvath CM. A road map for those who don't macrophages. J Biol Chem. 2013; 288:15547–55. know JAK-STAT. Science. 2002; 296:1653–5. 30. Kusaka Y, Tanaka T, Okamoto F, Terasaki F, Matsunaga Y, 18. Abumrad NA, el-Maghrabi MR, Amri EZ, Lopez E, Miyazaki H, Kawamura K. Effect of sulfo-N-succinimidyl Grimaldi PA. Cloning of a rat adipocyte membrane protein palmitate on the rat heart: myocardial long-chain fatty acid implicated in binding or transport of long-chain fatty uptake and cardiac hypertrophy. J Mol Cell Cardiol. 1995; acids that is induced during preadipocyte differentiation. 27:1605–12. Homology with human CD36. J Biol Chem. 1993; 31. Glatz JF, Angin Y, Steinbusch LK, Schwenk RW, Luiken JJ. 268:17665–8. CD36 as a target to prevent cardiac lipotoxicity and insulin 19. Ibrahimi A, Sfeir Z, Magharaie H, Amri EZ, Grimaldi P, resistance. Prostaglandins Leukot Essent Fatty Acids. 2013; Abumrad NA. Expression of the CD36 homolog (FAT) in 88:71–7. fibroblast cells: effects on fatty acid transport. Proc Natl 32. Mansor LS, Sousa Fialho MDL, Yea G, Coumans WA, West Acad Sci U S A. 1996; 93:2646–51. JA, Kerr M, Carr CA, Luiken JJFP, Glatz JFC, Evans RD, 20. Frank DA, Mahajan S, Ritz J. B lymphocytes from Griffin JL, Tyler DJ, Clarke K, Heather LC. Inhibition of patients with chronic lymphocytic leukemia contain signal sarcolemmal FAT/CD36 by sulfo-N-succinimidyl oleate transducer and activator of transcription (STAT) 1 and rapidly corrects metabolism and restores function in the STAT3 constitutively phosphorylated on serine residues. J diabetic heart following hypoxia/reoxygenation. Cardiovasc Clin Invest. 1997; 100:3140–8. Res. 2017; 113:737–48. 33. Greenwalt DE, Scheck SH, Rhinehart-Jones T. Heart CD36 21. Rozovski U, Calin GA, Setoyama T, D'Abundo L, expression is increased in murine models of diabetes and in Harris DM, Li P, Liu Z, Grgurevic S, Ferrajoli A, Faderl S, mice fed a high fat diet. J Clin Invest. 1995; 96:1382–8. Burger JA, O'Brien S, Wierda WG, et al. Signal transducer and activator of transcription (STAT)-3 regulates microRNA 34. Ferrajoli A, Faderl S, Van Q, Koch P, Harris D, Liu Z, gene expression in chronic lymphocytic leukemia cells. Mol Hazan-Halevy I, Wang Y, Kantarjian HM, Priebe W, Cancer. 2013; 12:50. Estrov Z. WP1066 disrupts Janus kinase-2 and induces 22. Li P, Harris D, Liu Z, Rozovski U, Ferrajoli A, Wang Y, caspase-dependent apoptosis in acute myelogenous Bueso-Ramos C, Hazan-Halevy I, Grgurevic S, Wierda W, leukemia cells. Cancer Res. 2007; 67:11291–9. www.oncotarget.com 21279 Oncotarget 35. Harris DM, Hazan-Haley I, Coombes K, Bueso-Ramos 36. Liu Z, Hazan-Halevy I, Harris DM, Li P, Ferrajoli A, C, Liu J, Liu Z, Li P, Ravoori M, Abruzzo L, Han L, Faderl S, Keating MJ, Estrov Z. STAT-3 activates Singh S, Sun M, Kundra V, et al. Transformation of human NF-kappaB in chronic lymphocytic leukemia cells. Mol mesenchymal cells and skin fibroblasts into hematopoietic Cancer Res. 2011; 9:507–15. cells. PLoS One. 2011; 6:e21250. www.oncotarget.com 21280 Oncotarget

Journal

OncotargetPubmed Central

Published: Apr 20, 2018

There are no references for this article.