Microtubule Involvement in Translational Regulation of Fibronectin Expression by Light Chain 3 of Microtubule-Associated Protein 1 in Vascular Smooth Muscle Cells

From the Division of Cardiovascular Research, Research Institute, the Hospital for Sick Children and the Departments of Pediatrics and Pathology, University of Toronto, Toronto, Ontario, Canada.

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Abstract—Our previous studies suggested that enhanced fibronectin mRNA translation in ductus arteriosus compared with aortic smooth muscle cells is related to increased expression of light chain 3 (LC3) of microtubule-associated protein 1, which binds an AU-rich element in the 3' untranslated region of fibronectin mRNA. We therefore hypothesized that microtubules are involved in LC3-mediated fibronectin mRNA translational regulation. In this study we show that disruption of microtubules by colchicine inhibits fibronectin mRNA translation in cultured ductus arteriosus smooth muscle cells. We proposed that the mechanism might be related to decreased docking of fibronectin mRNA on the translational machinery, ie, membrane-bound polysomes on rough endoplasmic reticulum, and confirmed this by Northern blot analysis. To investigate the mechanism further, we carried out polysome analysis using sucrose gradient centrifugation and fractionation and studied the polysomal distribution of fibronectin mRNA and LC3 protein in the sucrose gradient by using RNase protection assay and Western immunoblotting, respectively. Colchicine treatment shifts fibronectin mRNA from the fractions containing membrane-bound polysomes to the fractions carrying free polysomes and concomitantly decreases the amount of LC3 protein in the fractions containing membrane-bound polysomes. Furthermore, an EDTA-release experiment demonstrates that LC3 protein associates with the 60S ribosomal subunit. Our data support the concept that microtubules may function with LC3 to facilitate sorting of fibronectin mRNA onto rough endoplasmic reticulum and translation.

Key Words: fibronectin • microtubule • light chain 3 • translation • smooth muscle cell

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Fibronectin is a large extracellular matrix glycoprotein involved in cell adhesion, migration, and differentiation.¹ Its synthesis is regulated at both the transcriptional and posttranscriptional levels. Posttranscriptional mechanisms include enhanced mRNA translation² or increased mRNA stability.^{3 4} Our previous studies have shown that increased fibronectin synthesis is necessary for the migratory phenotype of ductus arteriosus smooth muscle cells (SMCs).^{5 6} The increased fibronectin synthesis by ductus arteriosus compared with aortic SMCs results from more efficient mRNA translation.⁷ The underlying mechanism has been related to an AU-rich element (ARE) in the 3' untranslated region (3'UTR) of fibronectin mRNA and an RNA-binding protein that we have purified and identified as the light chain 3 (LC3) of microtubule-associated protein 1.⁸

LC3 is a basic protein with arginine-rich sequences similar to those in arginine-rich RNA-binding proteins previously described.⁹ LC3 binds to the ARE in the 3'UTR of fibronectin mRNA in vitro.⁸ Several lines of evidence suggest that their interaction in vivo may facilitate fibronectin mRNA translation in vascular SMCs.⁸ First, in cultured ductus arteriosus and aortic SMCs, mutation of the ARE downregulates the translation of a chimeric mRNA containing the chlorampenicol acetyltransferase coding region and downstream fibronectin mRNA 3'UTR. Second, SMCs transfected with the chlorampenicol acetyltransferase–fibronectin 3'UTR construct bearing a wild-type but not mutated ARE decrease endogenous fibronectin synthesis, presumably because of the decoy of cytoplasmic ARE-binding factors by the vector ARE. Third, the increased amount of LC3 protein correlates with the increased ARE-binding activities in cultured ductus arteriosus compared with aortic SMCs, and LC3 protein colocalizes with fibronectin protein in the migratory SMCs in ductus arteriosus tissue during the early formation of the intimal cushion. Last, overexpression of LC3 in aortic SMCs optimizes fibronectin mRNA translation to the levels observed in ductus arteriosus SMCs.

Microtubules have been implicated in the storage, sorting, and translational control of mRNAs in a variety of cells.^{10 11 12 13 14} Signals that direct intracellular localization of mRNAs via microtubules are found within the 3'UTR of mRNAs.^{15 16 17 18} Several trans-acting factors that recognize cis elements in the 3'UTR of RNA, as well as bind to microtubules, have been reported.^{19 20 21 22} Although most of the studies are related to localized mRNAs that encode mostly cytoskeletal or cytoplasmic proteins, the functional involvement of microtubules in the sorting and translation of secreted proteins, eg, fibronectin, has not been explored. Since LC3 was initially cloned from a rat brain cDNA library, colocalizes with microtubules in cultured rat neuronal cells, and coprecipitates with in vitro assembled microtubules,²³ the dual function of LC3 as an RNA-binding protein, as well as a microtubule-associated protein, suggests that microtubules may play a role in translational regulation of fibronectin mRNA by LC3.

In the present study, we therefore address the role of microtubules in the context of LC3 in regulating fibronectin mRNA translation. We show that disruption of microtubules in cultured ductus arteriosus SMCs by colchicine treatment inhibits fibronectin mRNA translation. Analysis of intracellular distribution of fibronectin mRNA and LC3 protein by cellular fractionation and polysomal profile analysis demonstrates a concomitantly decreased association of fibronectin mRNA and LC3 protein with rough endoplasmic reticulum (RER) membrane-bound polysomes in cells after colchicine treatment. An EDTA-release experiment illustrates association of the LC3 protein with the 60S ribosomal subunit. These data support a role for microtubules in facilitating LC3-mediated enhancement of fibronectin mRNA RER sorting and translation.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Cell Culture
Primary ductus arteriosus SMC culture was prepared from fetal Rambouillet lambs (Renwick Farms, Clifford, Ontario) on day 100 of a 145-day timed gestation period, as previously described.²⁴ Cells were propagated by tissue explant after removing endothelial layers and adventitia and cultured in medium 199 with 10% FBS (GIBCO-BRL). In the experiments to examine the effect of microtubule structure on fibronectin synthesis, 2x10⁶ ductus arteriosus SMCs were seeded into 6-well dishes and cultured for 24 hours, followed by a 2-hour treatment with 10 µmol/L colchicine, a microtubule-depolymerizing agent. After the cells were washed with methionine-free media, they were labeled with [³⁵S]methionine (100 µCi/mL, Amersham Corp) in 1 mL serum- and methionine-free media containing colchicine in the above concentrations for an additional 2 hours. At the end of labeling, the cells were photographed under a phase-contrast microscope before removal of the conditioned media. The conditioned media were collected for assessment of total protein synthesis and fibronectin production. Cells were lysed and extracted for total RNA.

Indirect Immunofluorescence
For double-immunofluorescence staining of LC3 and tubulin, 10⁵ cells were plated on coverslips and cultured for 3 days. After treatment with colchicine for 2 hours, cells were fixed by 4% paraformaldehyde in microtubule-stabilizing buffer (100 mmol/L PIPES [pH 6.9], 4 mmol/L MgSO₄, and 1 mmol/L EGTA) for 20 minutes at room temperature and permeabilized with 0.1% Triton X-100 in PBS for 2 minutes on ice. Fluorescent labeling for tubulin and LC3 was carried out using a monoclonal mouse anti–-tubulin IgG (1:1000; Sigma Chemical Co) and a rabbit anti-LC3 antiserum (1:100, a gift from Dr J. Hammarback, Department of Neurobiology and Anatomy, Bowman Gray School of Medicine, Winston-Salem, NC).²³ Secondary antibodies used were fluorescein-conjugated goat anti-mouse IgG and Texas red–conjugated goat anti-rabbit IgG (all dilutions at 1:100). Normal rabbit IgG and secondary antibody alone were used as controls. Cell nuclei were stained with 1:10 000 diluted DAPI (Sigma) in PBS for 10 minutes.

Total Protein and Fibronectin Synthesis
Measurement of total protein synthesis was carried out by collecting conditioned media from normal or colchicine-treated cultures. Cell numbers were then counted after trypsinization. Secreted total proteins in aliquots (50 µL) of conditioned media were precipitated with trichloroacetic acid (TCA), and radioactivity reflecting newly synthesized and secreted protein was counted by liquid scintillation spectrometry and normalized for cell number.

Assessment of fibronectin protein production was performed as previously described²⁵ by incubating conditioned media containing equal counts (5x10⁶ cpm) of total TCA-precipitated protein with 50 µL of Gelatin 4B Sepharose (Pharmacia Biotech Inc). Fibronectin retained on the beads was eluted by boiling for 5 minutes in 100 µL SDS sample buffer and resolved by 6% SDS-PAGE. Gels were prepared for fluorography by treatment with En³Hance (DuPont-NEN), dried, and exposed to the film. With the autoradiograph used as a template, the corresponding bands were cut from the gel, and the radioactivity was determined by liquid scintillation spectrometry.

RNA Extraction and Northern Blot
Total RNA was extracted using Trizol reagent (GIBCO) according to the manufacturer's instructions. Extracted total RNA (10 µg) from each condition was separated by 1% agarose gel, transferred onto a nylon membrane (Amersham) by capillary elution, and fixed by UV irradiation. After the blocking procedure, the membranes were probed with a [³²P]dCTP random-labeled 1.4-kb human fibronectin (GIBCO) or 0.4-kb human cDNA probe for human c-myc (Oncogene) at a concentration of 10⁶ cpm/mL overnight at 50°C, followed by 2 washes with 2x SSC/0.1% SDS at 55°C for 30 minutes and 1 wash with 0.2x SSC/0.1% SDS at 65°C for 1 hour. Autoradiographs of Northern blots were analyzed by densitometry. Ethidium bromide staining of 28S and 18S RNAs served to control loading conditions.

Cell Fractionation
Cultured ductus arteriosus SMCs were washed in PBS and scraped into 14-mL Falcon tubes in 5 mL PBS. After centrifugation at 500g for 5 minutes, the cell pellets were extracted for 5 minutes on ice with 1 mL lysis buffer (10 mmol/L Tris-HCl [pH 7.4], 0.15 mol/L NaCl, 5 mmol/L MgCl₂, 0.1 mmol/L phenylmethylsulfonyl fluoride, 10 mmol/L dithiothreitol, 5 µg/mL cycloheximide, 100 U/mL RNAguard [Pharmacia], and 0.5% Nonidet P-40 [NP-40] [vol/vol]) and lysed with a Dounce homogenizer, type B pestle. After removal of nuclei by centrifugation at 1000g for 5 minutes, the lysate was centrifuged at 16 000g for 30 minutes. RNA was extracted from the supernatant (free and cytoskeletal-bound polysomes) and the pellet (membrane-bound polysomes)²⁶ and analyzed for fibronectin mRNA by Northern blot.

Polysome Profile Analysis
Analyses of polysome profiles by sucrose gradient fractionation were performed with the use of 2x10⁷ confluent cells as previously described.²⁷ Cells were harvested and homogenized with lysis buffer (10 mmol/L Tris-HCl [pH 7.4], 100 mmol/L NaCl, 5 mmol/L MgCl₂, 200 U/mL RNAguard, 100 µg/mL cycloheximide, and 0.5% NP-40). Nuclei were pelleted by centrifugation at 4°C and 12 000g for 2 minutes. The resulting supernatant (200 µL) was layered on 4.5 mL of a 15% to 40% sucrose gradient and centrifuged at 42 000 rpm for 2 hours at 4°C in a Beckman SW 50.1 rotor. After centrifugation, 17 fractions (0.25 mL each) were collected from top to bottom, and the absorbency of the fractions was measured at 254 nm. Proteins and RNA were extracted from each fraction using Trizol reagent (GIBCO) according to the manufacturer's instructions. Further analyses of the distributions of LC3 protein and fibronectin or c-myc mRNA were performed by Western immunoblotting and RNase protection assays or Northern blot analysis, respectively. The results were analyzed by the NIH image program and used for graphic presentation.

To study the potential association of LC3 protein with ribosome subunits, total cell lysates were treated with EDTA at a concentration of 10 µmol/L for 30 minutes on ice to release ribosome subunits from polysomes before the sucrose gradient centrifugation for 4 hours as described previously.²⁸

RNase Protection Assays
The construct used in in vitro synthesis of the sheep-specific antisense fibronectin RNA probe for the RNase protection assay was made using a standard reverse transcription–polymerase chain reaction method. One microgram of total RNA harvested from cultured lamb aortic SMCs was primed with a synthetic oligonucleotide, 5'-TGTTCGGTAATTAATGGAAATTGG-3', which complemented the 3' end of human fibronectin exon 7, and the reaction was driven by the MMLV reverse transcriptase (GIBCO). The polymerase chain reaction was then carried out by adding the 5' primer, 5'-TTTCTGATGTTCCGAGGGACC-3', which is homologous to the 5' end of human fibronectin exon 6. The 221-bp polymerase chain reaction product containing sheep fibronectin exons 6 and 7 were cloned into a pCR2.1 vector (Invitrogen Co). After linearizing by SacI, a 329-nt antisense RNA was generated by an in vitro transcription reaction using T7 RNA polymerase. The expected protected fibronectin mRNA was 221 nt.

RNase protection assays were carried out using a Ribonuclear Protection Kit II (Ambion) according to the manufacturer's instructions. For each reaction, 8x10⁴-cpm probes were hybridized with RNA extracted from each fraction of sucrose gradients overnight at 45°C. Yeast RNA (10 µg) was used as a negative control. The protected probes were separated by 6% acrylamide/8 mol/L urea gel after RNase T1/RNase A digestion.

Western Immunoblotting
Western immunoblot analysis was performed using a standard protocol. Samples were separated by SDS-PAGE and transferred onto a polyvinylidene fluoride membrane. The blot was blocked for 1 hour at room temperature in Tris-buffered saline containing 0.5% Tween 20 and 5% nonfat milk and then probed with rabbit anti-LC3 antiserum (1:2000 dilution). The blot was washed 4 times for 5 minutes each with Tris-buffered saline containing 0.5% Tween 20, followed by incubation with horseradish peroxidase–conjugated goat anti-rabbit IgG (1:3000; Bio-Rad Laboratories) for 1 hour at room temperature. The blots were washed 4 times for 5 minutes each with Tris-buffered saline containing 0.5% Tween 20 and developed with the use of an enhanced chemiluminescence kit (ECL, Amersham) according to the supplier's instructions.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Intact Microtubule Structures Are Important for Fibronectin Synthesis
Colchicine was applied to disrupt microtubules in cultured ductus arteriosus SMCs. Fibronectin synthesis in normal-cultured and colchicine-treated cells was examined. Treatment of cells with 10 µmol/L colchicine for 2 hours induced changes in cell shape related to alterations in microtubule structure (Figure 1). Those treated with colchicine looked foreshortened, with fragmentation of microtubules, which appeared as condensations in the perinuclear region as well as at the cell edge (Figure 1B versus 1A). The distribution of LC3 in colchicine-treated and normal-cultured cells was similar in its perinuclear distribution and punctate appearance. Cells probed with normal IgG instead of primary antibody were used as negative controls and showed only nuclear staining of DAPI (Figure 1C). Cells were then labeled with [³⁵S]methionine for 2 hours, and conditioned media were collected for fibronectin extraction. Although only a slight statistically insignificant decrease (<10%) in synthesis of total secreted proteins was observed after colchicine treatment (Figure 2a), fibronectin extracted from conditioned media containing an equal amount of total TCA-precipitated protein was reduced by 80% in cells treated with colchicine compared with control cells (Figure 2b and 2c, P<0.01). Although we cannot exclude the possibility that synthesis of some other proteins was also altered, the reduction in fibronectin was not the byproduct of a generalized effect of colchicine on protein synthesis. Northern blot analysis demonstrated that the steady-state levels of fibronectin mRNA were not changed in cells treated with colchicine compared with control cells (Figure 2d). This is consistent with the reduction of fibronectin synthesis by disruption of microtubules being related to a suppression of mRNA translation.

View larger version (81K): [in this window] [in a new window]   Figure 1. Disruption of microtubules alters SMC shape. Cultured ductus arteriosus SMCs were treated with colchicine for 2 hours, fixed, and immunostained for tubulin and LC3 using a monoclonal mouse anti–-tubulin IgG and a rabbit anti-LC3 antiserum as described in Materials and Methods. Secondary antibodies used were FITC-conjugated goat anti-mouse IgG and Texas red–conjugated goat anti-rabbit IgG. Cell nuclei were stained with DAPI. Compared with normal cultured cells (A), cells treated with colchicine (B) are foreshortened, with fragmentation of microtubules (FITC staining), which appear as prominent condensations at the cell edge. LC3 codistributes with tubulin in the perinuclear region and appears as particulate with yellow staining, resulting from overlap of Texas-red with FITC. A control specimen (C) using normal IgG shows only DAPI staining for nuclei. Bar=20 µm.

View larger version (33K): [in this window] [in a new window]   Figure 2. Disruption of microtubule by colchicine inhibits fibronectin synthesis. a, Quantitative study of total protein synthesis shows similar values in normal and colchicine-treated cultures from 3 different experiments. b, Quantitative analysis from 4 different experiments shows that fibronectin synthesis was significantly decreased in cells treated with colchicine compared with normal cultured cells (*P<0.01 by Student's t test). c, Representative autoradiograph of newly synthesized fibronectin (arrowhead) extracted from conditioned media of normal cultures (lane 1) or colchicine-treated cultures (lane 2). Equal amounts of total secreted proteins, as judged by TCA-precipitated counts, were used for fibronectin extraction. d, Representative Northern blot analysis shows that the steady-state level of fibronectin mRNA (arrowhead) remains the same when normal (lane 1) and colchicine-treated (lane 2) SMCs are compared. 28S and 18S RNA are shown to confirm equal loading conditions. Three different experiments showed similar results.

We confirmed that cells were seeded at the same density, and cell lifting and loss were not observed during the treatment of colchicine when the cells were examined and photographed under a phase-contrast microscope (Figure 3). Since the time frame of the experiment was only 4 hours but the doubling time of these primary ductus arteriosus SMCs is 72 hours, the selective reduction in fibronectin secretion after colchicine was unlikely due to mitotic arrest. In addition, there was no retention of newly synthesized fibronectin within the cells, ie, blocked secretion (data not shown). We also confirmed, by Western immunoblot, that the decreased fibronectin synthesis after colchicine treatment was not reflected in a decrease in LC3 protein (data not shown).

View larger version (146K): [in this window] [in a new window]   Figure 3. Phase-contrast light microscopy of ductus arteriosus SMCs treated with colchicine. Compared with normal cultures (A), there is no lifting or loss of cells 4 hours after colchicine treatment (B), although the cells appear somewhat less elongated after colchicine treatment. Bar=40 µm.

Sorting of Fibronectin mRNA to RER Membrane-Bound Polysomes Requires Microtubules
Since fibronectin is a secreted glycoprotein, the signal peptide would direct fibronectin mRNA to the RER, where fibronectin mRNA is translated. We therefore examined whether disruption of microtubules alters fibronectin mRNA distribution, eg, its association with RER membrane-bound polysomes. Cultured ductus arteriosus SMCs were lysed with nonionic detergent, NP-40. Free and membrane-bound polysomes were fractionated by 1-step centrifugation as described in Materials and Methods. The supernatant of the cytoplasmic extract contains free polysomes and polysomes released from the cytoskeleton, whereas the pellet contains RER membrane-associated polysomes. RNA was extracted from both portions, and distribution of fibronectin mRNA was determined by Northern blot analysis. As shown in Figure 4, in normal cultured cells, approximately equal amounts of fibronectin mRNA were found associated with membrane-bound polysomes (lane 1) and free polysomes (lane 2). However, in cells treated with colchicine, fibronectin mRNA associated with membrane-associated polysomes was reduced (lane 3) and appeared to be "shifted" or increased in inverse proportion into the free polysome fraction (lane 4). Thus, the reduction of fibronectin mRNA in RER membrane-bound polysomes by colchicine treatment implies that microtubules are involved in fibronectin mRNA sorting.

View larger version (29K): [in this window] [in a new window]   Figure 4. Alteration of fibronectin mRNA distribution in cultured ductus arteriosus SMCs by disrupting microtubule structure. a, Northern blot analysis of fibronectin mRNA distribution. Ductus arteriosus SMCs were cultured under normal conditions (lanes 1 and 2) or colchicine-treated conditions (10 µmol/L for 4 hours) (lanes 3 and 4). Cells were extracted with lysis buffer containing 0.5% NP-40; mRNA associated with membrane-bound polysomes (lanes 1 and 3) was separated from mRNA associated with cytoskeleton or free ribosomes (lanes 2 and 4) by 1-step centrifugation at 16 000g for 30 minutes. Distribution of fibronectin mRNA was then assessed by Northern blot analysis. b, Corresponding graphic presentation of distribution of fibronectin mRNA (% of total cellular fibronectin mRNA) in free ribosomes or cytoskeletal-bound polysomes (F) versus membrane-bound polysome fractions (M) under normal or colchicine-treated conditions. Two different experiments showed reproducible results.

Polysomal Distribution of Fibronectin mRNA and LC3 Protein
To determine whether the effect of microtubules on fibronectin mRNA translation likely involves LC3, we applied a polysome profile analysis to compare the distribution of fibronectin mRNA and LC3 protein in normal-cultured cells with that of colchicine-treated cells. In normal cultures (Figure 5a, arrowhead indicates the position of 80S monosome), all the fibronectin mRNA was found in the polysome region, and there was no detectable fibronectin mRNA associated with ribosome subunits. The polysome-associated fibronectin mRNA in normal cultures was largely concentrated in the last 5 fractions (normal, fractions 13 to 17), which were located at the bottom of the sucrose gradient containing heavy polysomes. This suggests that most of the fibronectin mRNA is associated with RER membrane-bound polysomes and is translationally activated. In contrast, the distribution of fibronectin mRNA in colchicine-treated cells was shifted into fractions in the sucrose gradient, which contained lighter or free polysomes (colchicine, fractions 9 to 15). These data were further illustrated graphically by densitometric analysis.

View larger version (21K): [in this window] [in a new window]   Figure 5. a and b, Polysomal distribution of fibronectin mRNA (a) and LC3 protein (b) is shown in normal or colchicine-treated ductus arteriosus SMCs. Postnuclei cell lysates were layered on a 15% to 40% sucrose gradient, centrifuged at 42 000 rpm for 2 hours at 4°C in a Beckman SW 50.1 rotor, and fractionated from top to bottom. Analyses of the distributions of fibronectin mRNA and LC3 protein were performed by RNase protection assay and Western immunoblotting, respectively, as described in Materials and Methods. One of 2 reproducible studies is shown as an autoradiograph at the top of panels a and b (arrowhead indicates the position of 80S monosome). The corresponding densitometric analysis using the NIH image program is graphically represented at the bottom of these panels. c, Ribosomal distribution is assessed by the optical density (OD) at 254 nm and is similar in normal and colchicine-treated cells.

The distribution of LC3 protein was also addressed by Western immunoblot, as shown in Figure 5b. In normal cultures, LC3 appeared in most fractions and had a dual distribution. The majority of LC3 was found at the top of the gradient (fractions 1 to 3), which is consistent with LC3 as a microtubule-associated protein, since microtubules are dissociated as soluble tubulin monomers during the process of homogenization of cells. A considerable amount of LC3 protein, however, also appeared in polysomal fractions, which contained the majority of fibronectin mRNA (fractions 12 to 17). It is of further interest that LC3 is frequently seen as a doublet and that the higher molecular weight component appears increased in the fractions containing heavy polysomes, and this will be discussed. However, LC3 was found in fractions carrying only ribosome subunits or ribonucleoproteins in the colchicine-treated cultures (fractions 1 to 4). In addition, the overall distribution of polysomes was not changed in colchicine-treated compared with normal-cultured SMCs (Figure 5c). These data are consistent with the previous 1-step free versus membrane-bound polysome fractionation study of fibronectin mRNA distribution using Northern blot analysis. They further suggest that disruption of microtubules interferes with the interaction of fibronectin mRNA with LC3 and impedes the docking of fibronectin mRNA onto RER membrane-bound polysomes and the formation of heavier polysomes.

Using Northern blot analysis, we then investigated the polysomal distribution of c-myc mRNA, which also has ARE in its 3'UTR. The ARE in the 3'UTR of c-myc mRNA is involved in mRNA stability and interacts with several RNA-binding proteins different from LC3.^{29 30 31 32 33} Unlike fibronectin mRNA, the overall distribution of c-myc mRNA, which was assessed in our system by Northern blot analysis, was not altered by colchicine treatment, although in the final fraction (fraction 20), there did appear to be slightly less c-myc mRNA in the colchicine-treated experiment (Figure 6). This suggests that colchicine treatment affects selectively the polysomal distribution of fibronectin mRNA.

View larger version (59K): [in this window] [in a new window]   Figure 6. a, Northern blot analysis of polysomal distribution of c-myc mRNA in normal and colchicine-treated ductus arteriosus SMCs. Similar distribution patterns for c-myc, as well as polysome profiles, assessed by ethidium bromide staining of agarose gel, were observed under both conditions. b, Corresponding densitometric analysis of c-myc mRNA using the NIH image program. Fraction 15 from the normal culture and fraction 7 from the colchicine-treated culture were missing or underloaded.

To address the relationship of LC3 protein with ribosomes, we carried out a 4-hour rather than a 2-hour sucrose gradient centrifugation. LC3 protein codistributes with both the 40S and 60S subunits and continuously spreads toward the bottom of sucrose gradient–containing monosomes or polysomes (Figure 7). To substantiate this finding further, we carried out an EDTA-release experiment to dissociate 40S and 60S subunits from assembled monosomes and polysomes. LC3 protein clearly codistributes with 60S RNA, but there appears to be degradation of 40S RNA, and this is associated with reduced LC3 protein in these fractions.

View larger version (46K): [in this window] [in a new window]   Figure 7. Association of LC3 protein with the 60S ribosomal subunit. a, Polysomal distribution of LC3 protein was studied by Western immunoblot analysis after EDTA treatment and a prolonged centrifugation, as described in Materials and Methods. Ethidium bromide staining of agarose gels indicates the distribution of ribosomal subunits, which show some separation with regard to 40S and 60S that is due to prolonged centrifugation compared with Figure 6. After EDTA release, the majority of 60S subunits are shifted into 3 fractions (fractions 9 to 11). In the control with Mg2+, although most of LC3 protein is located in the top of sucrose gradient (fractions 1 to 3), a portion of LC3 protein exists in the fractions containing 60S ribosomal subunits. After releasing ribosomes by EDTA treatment, LC3 protein is seen codistributed with 60S in fractions 9 to 11. The possible mechanism accounting for the alteration in LC3 from a single band to a doublet is described in the text. b, Corresponding densitometric analysis of LC3 protein distribution is graphically represented. The location of LC3 in the gradient after EDTA release is indicated by the bold outline.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
We have reported that a fibronectin mRNA ARE-binding protein is LC3 of microtubule-associated protein 1.⁸ The work described in the present study was therefore undertaken to examine whether the mechanism by which LC3 regulates fibronectin mRNA translation involves microtubules. We have demonstrated that disruption of microtubules by colchicine selectively results in the repression of fibronectin mRNA translation in ductus arteriosus SMCs. We further demonstrate by cellular fractionation and polysome profile analysis of fibronectin mRNA and LC3 distribution that facilitation of fibronectin mRNA docking onto RER membrane-bound polysomes is likely the mechanism by which microtubules function in enhanced fibronectin mRNA translation mediated by LC3.

To our knowledge, our findings provide the first evidence of a coordinated function of microtubules and a specific microtubule-associated protein (LC3) in regulating translation of a translationally activated mRNA encoding a secreted glycoprotein. It has previously been shown that mRNAs anchored to microtubules or other cytoskeletal elements encode nonsecretory proteins and are spatially localized.^{10 11 12 13 14} Signals that direct intracellular transport and localization of mRNAs are within the 3'UTR of mRNAs. For example, in Xenopus oocytes, Vgl RNA-binding protein mediates the association of Vgl mRNA with microtubules, and Vgl RNA-binding protein itself binds to microtubules.²¹ In male germ cells, spermatid perinuclear RNA-binding protein, which binds to the 3'UTR of spermatid-specific protamine-1 mRNA, is localized to microtubules and is proposed to be functionally involved in RNA transport or translational activation.²² In mice, a testis/brain RNA-binding protein attaches to translationally repressed transcripts, such as protamine-2, tau, and myelin basic protein containing conserved sequences (Y and H elements).²⁰

Colchicine has been used previously to depolymerize microtubules to establish their pivotal role in the transport or localization of mRNA. In our studies, we used colchicine to determine that microtubules are critical to fibronectin mRNA translation. Others have reported that colchicine treatment inhibits synthesis of collagen,³⁴ stearoyl coenzyme A desaturase, and fatty acid synthetase³⁵ without affecting total protein synthesis, but the mechanism was unexplored. Colchicine has also been demonstrated to inhibit general translation of mRNA in hepatectomized rats without affecting cell proliferation by inducing a global dissociation of membrane-bound polysomes.^{36 37} However, this mechanism cannot explain the selective inhibition of fibronectin mRNA translation by colchicine in the present study, since the overall polysome profile was unchanged in colchicine-treated cells compared with normal cultures, yet the association with fibronectin mRNA was altered. The difference may be related to the experimental model: whole hepatectomized animal and rapidly proliferating hepatic cells versus, in the present study, relatively quiescent cultured SMCs.

It is also interesting that the staining of LC3 in the SMC appears "particulate" around the perinuclear region, even in non–colchicine-treated ductus arteriosus cells. This suggests that LC3 may associate with fibronectin mRNA as a ribonucleoprotein complex that contains segregated microtubule elements as described previously by others.¹² Since there were no changes in LC3 protein or fibronectin mRNA levels after disruption of microtubules by colchicine or in the LC3 perinuclear distribution pattern, we speculate that the decreased translation of fibronectin mRNA is due to an inability of LC3, in the absence of microtubules, to influence the sorting of fibronectin mRNA onto membrane-bound polysomes.

We attempted to elucidate the mechanism by which LC3 regulates fibronectin mRNA translation in the context of microtubules by comparing intracellular distribution of both fibronectin mRNA and LC3 protein in normal cultured or colchicine-treated ductus arteriosus SMCs. Colchicine treatment diminished membrane-associated fibronectin mRNA, suggesting that sorting of fibronectin mRNA onto the RER membrane was impaired. Further polysome profile studies indicated that the fibronectin mRNA is translationally activated in ductus arteriosus SMCs, whereas the distribution of LC3 protein exhibited a biphasic pattern. The majority of LC3 protein was distributed with the translationally inactive subunits, which would likely be due to the dissociation of microtubules after homogenization of the cells on ice. A small fraction of LC3, however, codistributed with fibronectin mRNA on the heavy polysomes. The overlap of distribution of fibronectin mRNA and LC3 protein in ductus arteriosus SMCs suggests that LC3 is associated with ribosomes through the binding of fibronectin mRNA.⁸ However, we could not exclude the possibility that LC3 is associated with other mRNAs containing AREs in the complex of ribonucleoproteins or is directly associated with ribosomal subunits.

We have made a similar observation in a natural LC3 "knockout" cell line, HT1080 human fibrosarcoma cell. Expression of LC3 in those cells increases fibronectin mRNA translation due to enhanced recruitment of ribosomes, and LC3 protein and fibronectin mRNA codistribute with heavy polysomes in the sucrose gradient.³⁸ Others have also shown that ectopic expression of another RNA binding protein, Hel-N1, in 3T3-L1 preadipocytes increases glucose transporter (GLUT1) expression as a result of the acceleration of recruitment and progression from the preinitiation complex to heavy polysomes.³⁹ However, the association of Hel-N1 with translational machinery (polysome profile) was not addressed in that study.

The polysomal distribution of fibronectin mRNA and LC3 protein in colchicine-treated cells provides additional experimental evidence that intact microtubule structures are required for the interaction of LC3 and fibronectin mRNA and for the docking of fibronectin mRNA onto polysomes. Disruption of microtubules by colchicine shifts the fibronectin mRNA into fractions in the sucrose gradient associated with fewer polysomes, and LC3 was detected in only the translationally inactive fractions. This finding strongly suggests that there are 2 functionally distinct pools of LC3: one binds microtubules, and the other binds fibronectin mRNA and associates with 60S ribosomal subunits. Microtubules may affect fibronectin mRNA translation by providing a "track" for mRNA sorting to RER (Figure 8 shows the proposed schema), thus incorporating the dual function of LC3 both as a microtubule- and as an mRNA-binding protein. The association of LC3 with ribosomal subunits, mainly 60S subunits, also implies a function related to translational initiation as has been described for other RNA-binding proteins.^{28 39} The LC3 in the translationally inactive compartment may represent a storage pool, and modification of microtubule-associated LC3 by phosphorylation⁴⁰ is necessary for its function in promoting membrane-bound polysome recruitment of fibronectin mRNA. This also appears to explain the LC3 doublet observed in the polysome assay. Alterations in phosphorylation status are of functional significance in other ARE-binding proteins⁴¹ and in the testis/brain RNA binding protein, which also binds microtubules.^{19 20}

View larger version (22K): [in this window] [in a new window]   Figure 8. Proposed model of translational control of fibronectin expression by mRNA sorting. Fibronectin mRNA is sorted onto RER by nascent signal peptide, and translation is then initiated. The dual function of LC3, ie, binding of LC3 to fibronectin mRNA through the ARE at the 3'UTR and to microtubules, may facilitate sorting of fibronectin mRNA onto RER via microtubules. Alternatively, LC3 may dissociate from microtubules and link fibronectin mRNA to ribosomal complex by binding to the ARE and to an unknown factor (designated as ?) in the 60S ribosomal subunit.

Although we have focused on the relationship between LC3 and fibronectin mRNA translation in the context of microtubules, it is likely that LC3 regulates other mRNAs, particularly those with AREs, which may also influence the migratory cell phenotype in ductus arteriosus SMCs or which are associated with neointimal formation in vascular diseases.^{42 43 44 45 46} Therefore, we also examined the polysome distribution pattern for c-myc, a proto-oncogene with an ARE in its 3'UTR. However, the polysomal distribution pattern for c-myc mRNA did not change after colchicine treatment in the manner observed for fibronectin mRNA. This may reflect the interaction of c-myc with different RNA-binding proteins and the influence of c-myc ARE on a different posttranscriptional mechanism, ie, mRNA stability versus translation.^{28 29 30 31 32 33} Alternatively, unlike fibronectin mRNA, c-myc mRNA encodes a nuclear protein that is not translated by RER membrane-bound polysomes. In summary, our data show that the effect of intact microtubules on translational regulation of fibronectin mRNA mediated by LC3 is likely at the level of sorting of mRNA onto RER. How binding of LC3 to 3'UTR of fibronectin mRNA impacts on ribosome recruitment and how microtubules influence the interaction of LC3 and fibronectin mRNA will be of great interest in future studies.

	Acknowledgments

 
This study was supported by a program grant from the Medical Research Council of Canada (MT8546) to Dr Rabinovitch. Dr Rabinovitch is a Career Investigator of the Heart and Stroke Foundation of Ontario. The authors are most grateful to Dr J. Hammarback in the Department of Neurobiology and Anatomy, Bowman Gray School of Medicine, Winston-Salem, NC, for the anti-LC3 antiserum, and Susy Taylor and Joan Jowlabar for secretarial assistance in preparing the manuscript.

	Footnotes

 
Reprint requests to Marlene Rabinovitch, MD, Division of Cardiovascular Research, Hospital for Sick Children, 555 University Ave, Toronto, ON, Canada M5G 1X8.

Received May 4, 1998; accepted June 23, 1998.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 
1. Hynes RO, Lander AD. Contact and adhesive specificities in the associations, migrations, and targetting of cells and axons. Cell. 1992;68:303–322.[Medline] [Order article via Infotrieve]

2. Rasoamanantena P, Thweatt R, Labat-Robert J, Goldstein S. Altered regulation of fibronectin gene expression in Werner syndrome fibroblasts. Exp Cell Res. 1994;213:121–127.[Medline] [Order article via Infotrieve]

3. Chou L, Firth JD, Uitto V-J, Brunette DM. Substratum surface topography alters cell shape and regulates fibronectin mRNA level, mRNA stability, secretion and assembly in human fibroblasts. J Cell Sci. 1995;108:1563–1573.[Abstract]

4. Dean DC, Newby RF, Bourgeois S. Regulation of fibronectin biosynthesis by dexamethasone, transforming growth factor ß, and cAMP in human cell lines. J Cell Biol. 1988;106:2159–2170.[Abstract/Free Full Text]

5. Boudreau N, Rabinovitch M. Developmentally regulated changes in extracellular matrix in endothelial and smooth muscle cells in the ductus arteriosus may be related to intimal proliferation. Lab Invest. 1991;64:187–199.[Medline] [Order article via Infotrieve]

6. Boudreau N, Turley EA, Rabinovitch M. Fibronectin, hyaluronan, and a hyaluronan binding protein contribute to increased ductus arteriosus smooth muscle migration. Dev Biol. 1991;143:235–247.[Medline] [Order article via Infotrieve]

7. Boudreau N, Clausell N, Boyle J, Rabinovitch M. Transforming growth factor-beta regulates increased ductus arteriosus endothelial glycosaminoglycan synthesis and a post-transcriptional mechanism controls increased smooth muscle fibronectin, features associated with intimal proliferation. Lab Invest. 1992;67:350–359.[Medline] [Order article via Infotrieve]

8. Zhou B, Boudreau N, Coulber C, Hammarback J, Rabinovitch M. Microtubule-associated protein 1 light chain 3 is a fibronectin mRNA-binding protein linked to mRNA translation in lamb vascular smooth muscle cells. J Clin Invest. 1997;100:3070–3082.[Medline] [Order article via Infotrieve]

9. Burd CG, Greyfuss G. Conserved structures and diversity of function of RNA-binding proteins. Science. 1994;265:615–621.[Abstract/Free Full Text]

10. Singer RH. The cytoskeleton and mRNA localization. Curr Opin Cell Biol. 1992;3:719–721.

11. Steward O, Banker GA. Getting the message from the gene to the synapse: sorting and intracellular transport of RNA in neurons. Trends Neurosci. 1992;15:180–186.[Medline] [Order article via Infotrieve]

12. Suprenant KA. Microtubules, ribosomes, and RNA: evidence for cytoplasmic localization and translational regulation. Cell Motil Cytoskeleton. 1993;25:1–9.[Medline] [Order article via Infotrieve]

13. St Johnston D. The intracellular localization of messenger RNAs. Cell. 1995;81:161–170.[Medline] [Order article via Infotrieve]

14. Wilhelm J, Vale R. RNA on the move: the mRNA localization pathway. J Cell Biol. 1993;123:269–274.[Free Full Text]

15. Macdonald PM, Struhl G. Cis-acting sequences responsible for anterior localization of bicoid mRNA in Drosophila embryos. Nature. 1988;336:595–598.[Medline] [Order article via Infotrieve]

16. Gottlieb E. The 3' untranslated region of localized maternal messengers contains a conserved motif involved mRNA localization. Proc Natl Acad Sci U S A.. 1992;89:7164–7168.[Abstract/Free Full Text]

17. Gavis ER, Lehmann R. Localization of nanos RNA controls embryonic polarity. Cell. 1992;71:301–313.[Medline] [Order article via Infotrieve]

18. Mowry KL, Melton DA. Vegetal messenger RNA localization directed by a 340-nt RNA sequence element in Xenopus oocytes. Science. 1992;255:991–994.[Abstract/Free Full Text]

19. Kwon YK, Hecht NB. Binding of a phosphoprotein to the 3' untranslated region of the mouse protamine 2 mRNA temporally represses its translation. Mol Cell Biol. 1993;13:6547–6557.[Abstract/Free Full Text]

20. Han JR, Yiu GK, Hecht NO. Testis/brain RNA-binding protein attaches translationally repressed and transported mRNAs to microtubules. Proc Natl Acad Sci U S A.. 1995;92:9550–9554.[Abstract/Free Full Text]

21. Elisha Z, Havin L, Ringel I, Yisraeli JK. Vg1 RNA binding protein mediates the association of Vg1 mRNA with microtubules in Xenopus oocytes. EMBO J. 1995;14:5109–5114.[Medline] [Order article via Infotrieve]

22. Schumacher JM, Lee K, Edelhoff S, Braun RE. Spnr, a murine RNA-binding protein that is localized to cytoplasmic microtubules. J Cell Biol. 1995;129:1023–1032.[Abstract/Free Full Text]

23. Mann SS, Hammarback JA. Molecular characterization of light 3: a microtubule binding subunit of MAP1A and MAP1B. J Biol Chem. 1994;269:11492–11497.[Abstract/Free Full Text]

24. Rabinovitch M, Beharry S, Bothwell T, Jackowski G. Qualitative and quantitative differences in protein synthesis comparing fetal lamb ductus arteriosus endothelium and smooth muscle with cells from adjacent vascular sites. Dev Biol. 1988;130:250–258.[Medline] [Order article via Infotrieve]

25. Wrana JL, Maeno M, Hawrylyshyn B, Yao KL, Domenicucci C, Sodek J. Differential effects of transforming growth factor-beta on the synthesis of extracellular matrix proteins by normal fetal rat calvarial bone cell populations. J Cell Biol. 1988;106:915–924.[Abstract/Free Full Text]

26. Hesketh JE, Pryme IF. Interaction between mRNA, ribosomes and the cytoskeleton. Biochem J. 1991;277:1–10.

27. Chen C-YA, Xu N, Shyu A-B. mRNA-decay mediated by two distinct AU-rich elements from c-fos and granulocyte-macrophage colony stimulating factor transcripts: different deadenylation kinetics and uncoupling from translation. Mol Cell Biol. 1995;15:5777–5788.[Abstract]

28. Siomi MC, Zhang Y, Siomi H, Dreyfuss G. Specific sequences in the fragile X syndrome protein FMR1 and the FXR proteins mediate their binding to 60s ribosomal subunits and the interactions among them. Mol Cell Biol. 1996;16:3825–3832.[Abstract]

29. Zhang W, Wagner BJ, Ehrenman K, Schaefer AW, DeMaria CT, Crater D, DeHaven K, Long L, Brewer G. Purification, characterization, and cDNA cloning of an AU-rich element RNA-binding protein, AUF1. Mol Cell Biol.. 1993;13:7652–7665.[Abstract/Free Full Text]

30. DeMaria CT, Brewer G. AUF1 binding affinity to A+U-rich elements correlates with rapid mRNA degradation. J Biol Chem. 1996;271:12179–12184.[Abstract/Free Full Text]

31. Levine TD, Gao F, King PH, Andrews LG, Keene JD. Hel-N1: an autoimmune RNA-binding protein with specificity for 3' uridylate-rich untranslated regions of growth factor mRNAs. Mol Cell Biol. 1993;13:3494–3504.[Abstract/Free Full Text]

32. Nakagawa J, Waldner H, Meyer-Monard S, Hofsteenge J, Jeno P, Moroni C. AUH, a gene encoding an AU-specific RNA binding protein with intrinsic enoyl-CoA hydratase activity. Proc Natl Acad Sci U S A.. 1995;92:2051–2055.[Abstract/Free Full Text]

33. Huang LY, Tholanikunnel BG, Vakalopoulou E, Malbon CC. The M(r) 35,000 beta-adrenergic receptor mRNA-binding protein induced by agonists requires both an AUUUA pentamer and U-rich domains for RNA recognition. J Biol Chem. 1993;268:25768–25775.

34. Mansour MM, Dunn MA, Salah LA. Effect of colchicine on collagen synthesis by liver fibroblasts in murine schistosomiasis. Clin Chim Acta. 1988;177:11–20.[Medline] [Order article via Infotrieve]

35. Kasturi R, Joshi VC, Wakil SJ. Colchicine inhibition of insulin induction of stearoyl-CoA desaturase and fatty acid synthetase in cultured avian liver explants. Arch Biochem Biophys. 1984;233:530–539.[Medline] [Order article via Infotrieve]

36. Walker PR, Whitfield JF. Colchicine prevents the translation of mRNA molecules transcribed immediately after proliferative activation of hepatocytes in regenerating rat liver. J Cell Physiol. 1984;118:179–185.[Medline] [Order article via Infotrieve]

37. Walker PR, Whitfield JF. Cytoplasmic microtubules are essential for the formation of membrane-bound polysomes. J Biol Chem. 1985;260:765–770.[Abstract/Free Full Text]

38. Zhou B, Hammarback J, Rabinovitch M. Light chain 3 of microtubule-associated protein 1, an RNA-binding protein, increases fibronectin mRNA translation in human fibrosarcoma cells [abstract]. Mol Biol Cell. 1997;8S:322a.

39. Jain RG, Andrews LG, McGowan KM, Pekala PH, Keene JD. Ectopic expression of Hel-N1, an RNA-binding protein, increases glucose transporter (Glut1) expression in 3T3–L1 adipocytes. Mol Cell Biol. 1997;17:954–962.[Abstract]

40. Mason C, Rabinovitch M. Post-transcriptional regulation of fibronectin (FN) synthesis by nitric oxide (NO) in motile vascular smooth muscle cells (SMC) is mediated by phosphorylation of a microtubule-associated fibronectin mRNA binding protein, LC3 [abstract]. Mol Biol Cell. 1997;8S:322a.

41. Malter JS, Hong Y. A redox switch and phosphorylation are involved in the post-translational up-regulation of the adenosine-uridine binding factor by phorbol ester and ionophore. J Biol Chem. 1991;266:3167–3171.[Abstract/Free Full Text]

42. Caput D, Beutler B, Hartog K, Brown-Shimer S, Cerami A. Identification of a common nucleotide sequence in the 3'-untranslated region of mRNA molecules specifying inflammatory mediator. Proc Natl Acad Sci U S A.. 1986;83:1670–1675.[Abstract/Free Full Text]

43. Clausell N, Molossi S, Rabinovitch M. Increased interleukin-1 beta and fibronectin expression are early features of the development of the postcardiac transplant arteriopathy in piglets. Am J Pathol. 1993;142:1772–1782.[Abstract]

44. Molossi S, Clausell N, Rabinovitch M. Reciprocal induction of tumor necrosis factor-alpha and interleukin-1 beta activity mediates fibronectin synthesis in coronary artery smooth muscle cells. J Cell Physiol. 1995;163:19–29.[Medline] [Order article via Infotrieve]

45. Bennett MR, Anglin S, McEwan JR, Jagoe R, Newby AC, Evan GI. Inhibition of vascular smooth muscle cell proliferation in vitro and in vivo by c-myc antisense oligodeoxynucleotides. J Clin Invest. 1994;93:820–828.

46. Shi Y, Fard A, Galeo A, Hutchinson HG, Vermani P, Dodge GR, Hall DJ, Shaheen F, Zalewski A. Transcatheter delivery of c-myc antisense oligomers reduces neointimal formation in a porcine model of coronary artery balloon injury. Circulation. 1994;90:944–951.[Abstract/Free Full Text]

This article has been cited by other articles:

				L. Ying, A. Lau, C. M. Alvira, R. West, G. M. Cann, B. Zhou, C. Kinnear, E. Jan, P. Sarnow, M. Van de Rijn, et al. LC3-mediated fibronectin mRNA translation induces fibrosarcoma growth by increasing connective tissue growth factor J. Cell Sci., May 1, 2009; 122(9): 1441 - 1451. [Abstract] [Full Text] [PDF]

				J. Reese Death, dying, and exhaustion in the ductus arteriosus: prerequisites for permanent closure Am J Physiol Regulatory Integrative Comp Physiol, February 1, 2006; 290(2): R357 - R358. [Full Text] [PDF]

				J. Fricke, C. Voss, M. Thumm, and G. Meyers Processing of a Pestivirus Protein by a Cellular Protease Specific for Light Chain 3 of Microtubule-Associated Proteins J. Virol., June 1, 2004; 78(11): 5900 - 5912. [Abstract] [Full Text] [PDF]

				H. He, Y. Dang, F. Dai, Z. Guo, J. Wu, X. She, Y. Pei, Y. Chen, W. Ling, C. Wu, et al. Post-translational Modifications of Three Members of the Human MAP1LC3 Family and Detection of a Novel Type of Modification for MAP1LC3B J. Biol. Chem., August 1, 2003; 278(31): 29278 - 29287. [Abstract] [Full Text] [PDF]

				S. Mochizuki, B. Brassart, and A. Hinek Signaling Pathways Transduced through the Elastin Receptor Facilitate Proliferation of Arterial Smooth Muscle Cells J. Biol. Chem., November 15, 2002; 277(47): 44854 - 44863. [Abstract] [Full Text] [PDF]

				C. A. E. MASON, P. CHANG, C. FALLERY, and M. RABINOVITCH Nitric oxide mediates LC-3-dependent regulation of fibronectin in ductus arteriosus intimal cushion formation FASEB J, August 1, 1999; 13(11): 1423 - 1434. [Abstract] [Full Text]

				S. H. Platts, J. C. Falcone, W. T. Holton, M. A. Hill, and G. A. Meininger Alteration of microtubule polymerization modulates arteriolar vasomotor tone Am J Physiol Heart Circ Physiol, July 1, 1999; 277(1): H100 - H106. [Abstract] [Full Text] [PDF]

				C. A. E. O'Blenes, C. Kinnear, and M. Rabinovitch Tumor Necrosis Factor-{alpha} Induces Fibronectin Synthesis in Coronary Artery Smooth Muscle Cells by a Nitric Oxide-Dependent Posttranscriptional Mechanism Circ. Res., July 6, 2001; 89(1): 26 - 32. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Zhou, B. Articles by Rabinovitch, M. Search for Related Content PubMed PubMed Citation Articles by Zhou, B. Articles by Rabinovitch, M.