Heart-type fatty-acid-binding protein (FABP3) is a lysophosphatidic acid-binding protein in human coronary artery endothelial cells

Highlights • We have identified FABP3 as an LPA carrier protein in human coronary artery endothelial cells (HCAECs).• Administration of LPA to HCAECs resulted in a dose-dependent increase in PPARγ activation.• LPA-induced PPARγ activation was abolished when the FABP3 expression was reduced using small interfering RNA (siRNA).• FABP3 governs the transcriptional activities of LPA by targeting them to cognate PPARγ in the nucleus.


Introduction
Fatty-acid-binding protein 3, muscle and heart (FABP3) is one of nine known cytosolic FABPs ranging in size from 14 to 15 kDa. It is most ubiquitously expressed in heart, skeletal muscle, and other tissues [1,2]. The protein is involved in maintaining an energy supply to the heart [3] and other body parts [4], as well as in regulating the intramuscular fat content [5] and improving insulin sensitivity [6]. Shan et al. reported FABP3 works as being the response gene of peroxisome proliferator-activated receptor gamma (PPARc) [7]. PPARc is a member of the nuclear hormone receptor superfamily, many members of which function as lipid-activated transcription factors [8]. These gene regulatory factors have a direct impact on human diseases such as atherosclerosis, obesity, diabetes, and can-cer [9][10][11][12]. Gene transcription by PPARc has been shown to be regulated by ligands such as retinoic acid and long-chain fatty acids [13], which bind in cells to members of the family of intracellular lipid binding proteins, including FABPs. Published X-ray crystal structures of FABPs have revealed that the entrance to the ligand-binding pockets is flanked by two a-helices that appear to limit access to the binding site [14]. In this study, we show that lysophosphatidic acid (LPA)-bound FABP3 can be displaced by LPA but not by lysophosphatidylcholine (LPC). Previous studies have provided evidence that LPA is an intracellular agonist for PPARc, and might enter cells to activate the PPARc. In the inactivated state, PPARc is believed to exist as complexes, bound with corepressor proteins such as nuclear receptor corepressor and silencing mediator of retinoid [8]. In this state, PPARc may have a cytoplasmic rather than a nuclear location. Upon ligand activation, PPARc dissociates from the corepressor and recruits coactivators [15]. However, the important question that remains to be answered is how lipids like the poorly water-soluble LPA reach the nucleus to bind to PPARc and initiate transcriptional activity. Here, we present evidence that FABP3 functions as an intracellular carrier for LPA.

Reagents and antibodies
LPA-coated agarose beads (L-6101) and control beads (P-B000) were purchased from Echelon Biosciences (Salt Lake City, UT, USA). The tissue culture media were obtained from Nacalai Tesque (Kyoto, Japan). Rabbit polyclonal antibody against cardiac FABP (ab45966) was purchased from Abcam (Cambridge, UK), and mouse polyclonal antibody against b-actin (sc-47778) was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA).

Cell culture
Primary HCAECs (Lonza, Walkersville, MD, USA) were propagated in endothelial cell growth medium (EGM-2) containing 5% fetal bovine serum, manufacturer-recommended supplemental growth factors (EGM-2 BulletKit), antibiotics, and antimycotics. All assays were performed using cells taken between passages 3 and 12 at 60-80% confluence, and were repeated at least three times in duplicate or triplicate. 2.4. In-gel digestion and protein identification by matrix-assisted laser desorption/ionization-time-of-flight mass spectrometry In-gel digestion of gel bands was performed as described previously [16]. Briefly, protein spots excised from the gel were destained with 100 mM ammonium bicarbonate in 50% acetonitrile. The gel pieces were dried and digested with sequencinggrade modified trypsin (Promega). The peptide solution was recovered, and residual peptides were extracted by shaking the solution with 5% trifluoroacetic acid (TFA) in 50% acetonitrile. The combined solutions were concentrated using a lyophilizer. The tryptic peptides, which were dissolved in 0.1% TFA, were desalted with Zip-Tip (Millipore, Billerica, MA, USA) according to the manufacturer's instructions, mixed with an equal volume of a matrix solution (10 mg/mL a-cyano-4-hydroxycinnamic acid in 50% acetonitrile/ 0.1% TFA), and applied to a target plate. MS/MS analyses were performed using the AB SCIEX TOF/TOF™ 5800 System (AB SCIEX, Foster City, CA, USA). Protein identification was performed with the ProteinPilot™ software (AB Sciex) using the UniProt database.

Western blotting
The cytoplasmic and nuclear extracts were isolated using NE-PER nuclear and cytoplasmic extraction reagents (Pierce, Rockford, IL, USA). The protein extracts were then separated on 5-20% SDS-PAGE gels (e-PAGEL; ATTO, Tokyo, Japan) and electrotransferred to Immobilon-P membranes (Millipore). The membranes were blocked in Block Ace (DS Parma Biomedical Co. Ltd., Osaka, Japan) for 1 h and then incubated with a primary antibody in Tris-buffered saline-Tween 20 with 5% Block Ace for 12 h at 4°C. Bands were visualized with EzWestLumi plus (ATTO).

Analysis of LPA (18:1) by liquid chromatography-tandem mass spectrometry
Total lipids were extracted from whole cells by the use of butanol. The samples were diluted to 1:1000 in methanol/water (95:5, v/v) containing 5 mM ammonium formate, and the amount of LPA (18:1) was measured by liquid chromatography-tandem mass spectrometry (LC-MS/MS). The LC-MS/MS was performed using a quadrupole-linear ion trap hybrid MS, 5500 QTRAP (Applied Biosystems/AB Sciex, Concord, ON, Canada), with an Agilent 1200 LC system combined with an autosampler (Agilent Technologies, Wilmington, DE, USA). The LPA (18:1) was separated on an Inertsil Ò ODS-3 column (150 mm Â 2.1 mm; with 3 lm silica particles; GL Sciences Inc., Tokyo, Japan), which was eluted with methanol/ water (95:5, v/v) containing 5 mM ammonium formate at a flow rate of 0.2 mL/min. Aliquots (10 lL) of the test solutions were applied to the mass spectrometer for analysis. The LPA (18:1) was analyzed by multiple reaction monitoring in the negativeion mode, with Q3 (product ion) set at m/z 281.2 in combination with the deprotonated molecular ion as Q1 (m/z 435.16). The relative amount of LPA (18:1) was calculated from the ratios of the peak areas using Analyst Ò software (AB SCIEX). Supernatants were analyzed by immunoblotting, using an antibody against FABP3. Incubation with an anti-b-actin antibody was used as a protein-loading control.

Quantitative real-time polymerase chain reaction (PCR) analysis
Total RNA was prepared from HCAECs cells using NucleoSpin Ò RNA II (Takara, Shiga, Japan). Then, 0.5 lg of total RNA was used for the subsequent synthesis of cDNA using the ReverTra Ace qPCR RT Kit (Toyobo, Osaka, Japan) as recommended by the manufacturer. mRNA levels were quantified by using an ECO Real-Time PCR system (Illumina, Inc., San Diego, CA, USA) and the SYBR Green Realtime PCR Master Mix Plus (Toyobo) with the following primers: CD36, 5 0 -GGCTGTGACCGGAACTGTG-3 0 (F) and 5 0 -AGGTCT CCAACTGGCATTAGAA-3 0 (R), and CYP27A1, 5 0 -CGGCAACGGAG CTTAGAGG-3 0 (F) and 5 0 -GGCATAGCCTTGAACGAACAG-3 0 (R) Primers from a human housekeeping gene primer set (Takara) were used as the internal control; the most suitable housekeeping genes were determined according to the manufacturer's instructions. All PCRs were performed in 10-lL volumes using 48-well PCR plates (Illumina, Hayward, CA, USA). The cycling conditions were 95°C for 10 min (polymerase activation) followed by 40 cycles of 95°C for 15 s, 55°C for 15 s, and 72°C for 30 s. The relative mRNA quantification was calculated using the arithmetic formula 2 ÀDDCq , where DCq is the difference between the threshold cycle of a given target cDNA and an endogenous reference cDNA.

Results
Using affinity chromatography with LPA beads, we successfully captured a potential target protein (15 kDa) from the HCAECs (Fig. 1A). After extensive washing, the 15-kDa protein was excised from the gel, trypsin-digested, and analyzed by peptide mass fingerprinting with MALDI-TOF-MS. Tandem mass spectrometry profiles identifying the FABP3 protein are shown in Fig. 1A. To obtain direct proof of the interaction of LPA with FABP3, the LPA beads were incubated with HCAECs extracts in the presence or absence of 0-30 lM LPA or LPC on ice. The modified FABP3 was isolated from the free LPA or LPC, and then subjected to SDS-PAGE and western blotting against FABP3 antibody. As shown in Fig. 1B, the LPA-bead-bound FABP3 decreased dose-dependently in the presence of excess LPA, but not LPC. This result suggests that LPA interacts with the binding within the FABP3. Next, to determine the role of LPA in regulating FABP3 expression, we examined the effect of LPA on FABP3 expression in HCAECs. As shown in Fig. 2A and 2B, stimulation with LPA (0-30 lM) did not up-regulate the FABP3 mRNA and protein expressions. We then verified whether the protein involved in the LPA-mediated PPARc activation was indeed FABP3. As shown in Fig. 3A, the knockdown of FABP3 expression in HCAECs using small interfering RNA (siRNA) was effective, as evidenced by real-time polymerase chain reaction (PCR) analysis and western blot analysis using an anti-FABP3 antibody. Real-time PCR analysis showed that FABP3 mRNA in siRNAtransfected cells was knocked down by 75-80% compared with the expression in control cells. As shown in Fig. 3B, the knockdown of FABP3 in HCAECs markedly decreased the ability of the LPA to induce PPARc-mediated transactivation in a dose-dependent manner. Furthermore, we selected two PPAR target genes and monitored their transcriptional regulation in HCAEC treated with LPA for 20 h. LPA stimulation activated the expression of Cd36, and Cyp27a1 by 1.5-to 2-fold (Fig. 3C).
These results suggest that FABP3 enhances the activity of PPARc. We then examined the amount of LPA present in the HCAECs. LPA was exogenously added to cells at a concentration of 10 lM, and LPA uptake was determined by LC-MS/MS. As shown in Fig. 4B, interestingly, the nuclear fraction of the FABP3-knocked- down HCAECs had significantly less nuclear LPA than that of the control cells.

Discussion
Relatively few intracellular binding partners for LPA have been recognized. Previous studies have identified some candidate proteins, including C-terminal-binding protein/brefeldin A-dependent ADP ribosylated substrate [17], liver fatty-acid-binding protein [18], and gelsolin [19]. LPA and alkyl-LPA are also ligands for PPARc [20,21]. Here, the isolation and purification of LPA-binding proteins from HCAECs were coupled to their identification by proteomics technique. As a result, we provided evidence that FABP3 is distributed from the cytosol to the nucleus in response to LPA-mediated PPARc activation. It is becoming increasingly apparent that liver specific FABPs act as solubilizing intracellular shuttles for fatty acid ligands to the nucleus, where the ligand is released to its target PPARs [22]. The low water solubility of LPA in the hydrophilic cytosol is its diffusional barrier. To obviate this, intracellular proteins such as FABPs orchestrate the movement of fatty acids inside cell. Our study suggests that FABP3 is ubiquitously expressed in the HCAECs and may therefore serve as intracellular carriers for LPA. We found that LPA is rapidly and efficiently delivered to the nucleus. Such a carrier function is expected to be more efficient than the free diffusion of LPA across the cytosol of endocytic vesicles. The function of FABP3 as an LPA carrier was revealed by the siRNA knockdown experiment. Cells transfected with FABP3 siRNA showed a relatively slow nuclear transport of LPA compared with the wild-type control cells. Furthermore, the reporter gene assay suggested that FABP3 governs the transcriptional activities of LPA by targeting them to cognate PPARc in the nucleus. FABP3 shuttles LPA to the nucleus, wherein they activate PPARc. This hypothesis is supported by a recent report that FABP5 mediated the nuclear translocation of retinoic acid and enhanced its activation of PPARs [23]. In conclusion, FABP3 represents the first protein known to transport LPA from the cytosol to the nucleus in human coronary artery endothelial cells.

Conflict of interest
The authors declare that there is no conflict of interest regarding the publication of this article.