Internalization of eNOS via caveolae regulates PAF-induced inflammatory hyperpermeability to macromolecules

Fabiola A. Sánchez, David D. Kim, Ricardo G. Durán, Cynthia J. Meininger, Walter N. Durán


Endothelial nitric oxide (NO) synthase (eNOS) is thought to regulate microvascular permeability via NO production. We tested the hypotheses that the expression of eNOS and eNOS endocytosis by caveolae are fundamental for appropriate signaling mechanisms in inflammatory endothelial permeability to macromolecules. We used bovine coronary postcapillary venular endothelial cells (CVECs) because these cells are derived from the microvascular segment responsible for the transport of macromolecules in inflammation. We stimulated CVECs with platelet-activating factor (PAF) at 100 nM and measured eNOS phosphorylation, NO production, and CVEC monolayer permeability to FITC-dextran 70 KDa (Dx-70). PAF translocated eNOS from plasma membrane to cytosol, induced changes in the phosphorylation state of the enzyme, and increased NO production from 4.3 ± 3.8 to 467 ± 22.6 nM. PAF elevated CVEC monolayer permeability to FITC-Dx-70 from 3.4 ± 0.3 × 10−6 to 8.5 ± 0.4 × 10−6 cm/s. The depletion of endogenous eNOS with small interfering RNA abolished PAF-induced hyperpermeability, demonstrating that the expression of eNOS is required for inflammatory hyperpermeability responses. The inhibition of the caveolar internalization by blocking caveolar scission using transfection of dynamin dominant-negative mutant, dyn2K44A, inhibited PAF-induced hyperpermeability to FITC-Dx-70. We interpret these data as evidence that 1) eNOS is required for hyperpermeability to macromolecules and 2) the internalization of eNOS via caveolae is an important mechanism in the regulation of endothelial permeability. We advance the novel concept that eNOS internalization to cytosol is a signaling mechanism for the onset of microvascular hyperpermeability in inflammation.

  • endothelial nitric oxide
  • endothelial cells
  • microvascular permeability
  • endothelial nitric oxide synthase translocation
  • acetylcholine
  • platelet-activating factor

increased microvascular permeability (hyperpermeability) is a hallmark of inflammation. Hyperpermeability to macromolecules is a process that occurs in vivo mainly at postcapillary venules. Whereas the significance of endothelial nitric oxide (NO) synthase (eNOS)-derived NO in the regulation of microvascular transport across postcapillary venules is an experimentally supported emerging concept (16, 27, 34, 36), the mechanisms by which eNOS controls microvascular permeability are poorly understood.

Proinflammatory agonists that increase permeability, such as vascular endothelial growth factor (VEGF), bradykinin, and platelet-activating factor (PAF), stimulate signaling cascades that converge on eNOS (20), activate the enzyme, and cause NO production (9). The activation of eNOS proceeds mainly through the phosphorylation of the enzyme at serine-1177 (Ser1177) (7, 12) and the dephosphorylation at threonine-495 (Thr495) (8). Interestingly, acetylcholine (ACh), an agent that causes vasodilation but does not alter microvascular permeability (27), induces exactly the same changes in phosphorylation of eNOS (30). These observations suggest that phosphorylation of eNOS per se is not a determinant of the functional consequences of eNOS-derived NO.

We reasoned that the location of eNOS may contribute to determine the functional significance of eNOS-derived NO. eNOS is found in plasma membrane (mostly in caveolae) and Golgi in control endothelial cells (ECs). It is established that eNOS can produce NO regardless of its location, even though its efficiency may vary (5, 37). Also, eNOS can translocate, upon stimulation, from plasma membrane to intracellular compartments (10, 23, 30). This movement has been associated with the depalmitoylation of eNOS (35), and more recently, it has been suggested that caveolae may serve as a vehicle for eNOS translocation (4).

The functional significance of eNOS internalization or traffic, if any, has not been previously investigated. Work from Michel and colleagues (10) shows eNOS movement to the cytosol after VEGF application, an agonist that induces hyperpermeability, and we demonstrated that PAF-induced eNOS preferential internalization to cytosol was associated with hyperpermeability to macromolecules (30). Based on these and the above-mentioned observations, we hypothesized that eNOS expression is required for the development of hyperpermeability and that eNOS internalization to NO receptors serves to determine the functional significance of eNOS-derived NO in response to PAF, a proinflammatory autacoid. The experiments reported herein were designed to test these hypotheses in bovine coronary ECs derived from postcapillary venules (CVECs) (31). We confirm in ECs that eNOS expression is required for an increase in permeability and propose, for the first time, that eNOS internalization via caveolae plays a functional role in determining PAF-induced hyperpermeability to macromolecules.



We used mouse anti-human eNOS, mouse anti-phospho-human eNOS (Ser1177), mouse anti-phospho-human eNOS (Thr495), and rabbit anti-caveolin antibodies from BD Biosciences (San Jose, CA). All the above-mentioned mouse antibodies recognize the corresponding eNOS epitopes in the bovine CVECs.

Cell culture and transfection.

We grew CVECs in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum; 20 U/ml heparin sodium, 50 μg/ml penicillin, 50 μg/ml streptomycin, and 10 μg/ml neomycin. All the experiments were performed using passage 3–7 CVECs. Using electroporation, we transfected CVECs with cDNA for the dominant-negative mutant of dynamin-2 coupled to green fluorescent protein (dyn2K44A; kindly provided by Dr. Mark McNiven, Mayo Clinic College of Medicine, Rochester, MN). Briefly, we grew cells to 70–80% confluence, trypsinized them, and resuspended them in 100 μl of nucleofection solution. We then added 1 μg of dyn2K44A cDNA to the cells. We electroporated CVECs using a basic nucleofector kit for primary mammalian ECs. We applied program T23 from Amaxa Biosystems (Gaithersburg, MD). We electroporated CVECs with nucleofection solution and with the empty vector as a control. The experiments were carried out 72 h after transfection. We verified the expression of transfected dyn2K44A by the cellular fluorescence.

eNOS siRNA transfection.

We used specific siRNA (Ambion, Austin, TX), as designed by Zhang et al. (37), to deplete eNOS. We transfected CVECs with small interfering (si)RNA using lipofectin and following the protocol recommended by the manufacturer (Invitrogen, Carlsbad, CA). We lysed the cells at predetermined times and performed Western blot analysis against eNOS to confirm the depletion of the protein.

NO measurements.

We measured NO production using NO-sensitive recessed-tip microelectrodes (1). We used 100% nitrogen and 400 and 800 parts/million NO in nitrogen to establish a calibration range for NO of 0–600-1,200 nM (in saline). We placed coverslips containing confluent CVECs in a perfusion chamber. We superfused the cells with media supplemented with l-arginine at a rate of 1 ml/min. We administered PAF (Sigma Chemicals, St. Louis, MO) through a side port in the perfusion tubing to achieve a PAF concentration of 100 nM in the chamber.

Immunofluorescence microscopy.

We fixed confluent monolayers of CVECs grown on glass coverslips with 3% paraformaldehyde for 15 min, permeabilized them with 0.5% PBS-Triton for 5 min, blocked them in 1% PBS-BSA for 30 min, and then incubated them with anti-eNOS antibodies and Alexa Fluor secondary antibodies. We examined the CVECs with an inverted fluorescence microscope (Axiovert-200 M; Zeiss).

Detergent-free purification of caveolae-enriched membrane fractions.

We grew CVECs in 100 mm tissue culture dishes (2 plates per treatment). After stimulation with 100 nM PAF, we homogenized and sonicated the cells according to described protocols (21). We placed cell lysates adjusted to 45% sucrose in a volume of 1.5 ml adding buffer 2 [containing 90% sucrose, 25 mM MES (pH 6.5), 150 mM NaCl, and inhibitors of proteases and phosphatases] at the bottom of a 4.5-ml centrifuge tube. We layered 1.5 ml of each buffer 3 [containing 35% sucrose, 250 mM sodium carbonate (pH 11), 25 mM MES, and 150 mM NaCl] and buffer 4 [containing 5% sucrose, 250 mM sodium carbonate (pH 11), 25 mM MES, and 150 mM NaCl] on top of it and centrifuged the samples at 44,000 rpm for 18 h in a Beckman L8-70M ultracentrifuge equipped with a SW60Ti rotor. We collected 12 fractions from the top to the bottom of each tube. An equal volume of each fraction was used for Western blot analysis and probed for eNOS and caveolin-1.

Western blot analysis.

We grew CVECs to confluence in 100-mm plates. Immunoblot analyses of protein expression and phosphorylation were assessed as we described previously (2, 30). Densitometric analyses of Western blots were performed using the National Institutes of Health ImageJ program.

Measurement of monolayer permeability.

We determined control and PAF-stimulated permeability to fluorescein isothiocyanate dextran 70 KDa (FITC-Dx-70, a macromolecule that mimics albumin) across confluent CVEC monolayers using an established method (2, 30). We obtained samples for baseline permeability every 15 min from the abluminal chamber for a period of 60 min. After the addition of PAF to both sides of the chamber (final concentration = 100 nM), we collected samples for an additional 60 min.

Statistical analysis.

Data are presented as means ± SE. Groups were analyzed for differences by one-way ANOVA followed by Tukey-Kramer's test. Significance was accepted at P < 0.05.


PAF stimulates eNOS phosphorylation and NO production in CVECs.

Because we used ECs derived from postcapillary venules, our first step was to confirm that CVECs express eNOS and that the enzyme can be activated to produce NO. We verified the expression of eNOS in CVECs by immunofluorescence. Figure 1A shows immunofluorescent localization of eNOS preferentially in the plasma membrane and Golgi, in agreement with the typical distribution of eNOS described in ECs (13, 14, 30). We corroborated eNOS activation in response to PAF by measuring NO production and eNOS phosphorylation. The application of 100 nM PAF rapidly increased NO production by CVECs from 4.3 ± 3.8 to 467.0 ± 22.6 nM (means ± SE; P < 0.05; Fig. 1B). After the quantification of eNOS phosphorylation, we calculated the ratio of phosphorylated eNOS to total eNOS at control (time = 0), 0.5, 1.0, and 3.0 min after PAF. Figure 1C, top, shows that PAF significantly increased phosphorylation of eNOS at Ser1177 as early as 0.5 min. The phosphorylation at Ser1177 remained elevated up to minute 3.0. PAF-induced dephosphorylation of eNOS at Thr495 was significant 1 min after the application of the autacoid and showed a trend for returning toward baseline levels at 3 min. Figure 1C, bottom, displays Western blots illustrating changes in eNOS phosphorylation at Ser1177 and Thr495. Interestingly, NO production reached a maximum ∼3 min after the application of PAF, indicating a close temporal correlation between eNOS phosphorylation and NO production in CVECs.

Fig. 1.

Platelet-activating factor (PAF; 100 nM) induces endothelial nitric oxide (NO) synthase (eNOS) phosphorylation and NO production in postcapillary venular endothelial cells (CVECs). A: immunofluorescence shows eNOS expression in cell membrane and Golgi in control CVECs. B: PAF significantly stimulated a robust increase in pericellular NO concentration. The induced response was rapid, peaked at about 3 min, and returned to basal levels ∼10 min after PAF application (means ± SE, n = 6 experiments, P < 0.05). C: PAF significantly increased phosphorylation of eNOS at Ser1177 and decreased eNOS phosphorylation at Thr495 as a function of time, as indicated by the ratio of phosphorylated (p-eNOS) to total eNOS (*P < 0.05, n = 3 experiments). The Western blots illustrate typical examples of changes in eNOS phosphorylation at Ser1177 and Thr495.

PAF-stimulated hyperpermeability is regulated by eNOS-derived NO in CVECs.

To correlate molecular signals with functional end points, we investigated PAF-induced hyperpermeability to FITC-Dx-70 in CVEC monolayers. PAF significantly increased permeability to FITC-Dx-70 from 3.4 ± 0.3 × 10−6 to 8.5 ± 0.4 × 10−6 cm/s (Fig. 2A). This change was rapid inasmuch as a change in the slope of flux of FITC-Dx-70 was detected in the first sample taken at 5 min after PAF application. To test that the increase in permeability is mediated by eNOS, we depleted eNOS from CVECs using specific siRNA (37). Figure 2B shows that eNOS expression decreases as early as 16 h after transfection of CVECs with eNOS siRNA and illustrates that the maximal depletion of eNOS occurred at 72 h using 30 nM eNOS siRNA. We chose this maximal depletion time point and siRNA concentration to perform our measurements of CVEC monolayer permeability to FITC-Dx-70. Scrambled siRNA, used as control, does not decrease the normal expression of eNOS in CVECs. Figure 2C shows that the basal permeability is normal in eNOS-depleted CVECs. However, the hyperpermeability response to PAF is completely absent in eNOS-depleted CVECs (Fig. 2C).

Fig. 2.

Depletion of eNOS inhibits PAF-induced hyperpermeability. A: 100 nM PAF increases permeability to FITC-dextran 70 KDa (Dx-70) in CVECs. Data are expressed as permeability coefficients (means ± SE). *P < 0.05, n = 5 experiments. B: Western blots show depletion of eNOS as a function of small interfering (si)RNA concentration and time. C: depletion of endogenous eNOS in CVECs abrogates the development of PAF-induced hyperpermeability. CVECs transfected with scrambled (sc) siRNA served as a control. The increase in permeability elicited by 100 nM PAF is significant compared with all other interventions (means ± SE; *P < 0.05, n = 5 experiments).

PAF induces eNOS trafficking in CVECs.

Having demonstrated that eNOS is required for the development of an increase in CVEC monolayer permeability to macromolecules, we investigated whether or not the translocation of eNOS is also required for PAF-induced hyperpermeability. As a first approach, we verified eNOS translocation in CVECs using immunofluorescence microscopy and purification of caveolae-enriched membrane fractions. Figure 3A shows that eNOS is located in the plasma membrane and Golgi in control CVECs and illustrates that PAF stimulates eNOS translocation from plasma membrane. Since caveolin is a widely used marker for caveolae, we confirmed the information provided by microscopy by isolating caveolin-containing lipid rafts fractions and probing for eNOS and caveolin by Western blot analysis. Figure 3B shows that, under baseline conditions, eNOS is distributed in fractions 7–12 and is greatly enriched in fractions 7 and 8. Baseline caveolin is enriched in the same fractions. Three minutes after PAF application, eNOS is found mainly, if not only, in heavier fractions (particularly in fraction 12). Taken together, the microscopy and lipid rafts data indicate that PAF induces eNOS translocation in CVECs.

Fig. 3.

PAF induces eNOS translocation in CVECs. A: immunofluorescence images of CVECs were obtained in control cells and after 100 nM PAF treatment. The image shows that PAF induces the disappearance of eNOS from plasma membrane and its appearance in a diffuse fashion in cytosol. The images are representative of 3 independent experiments. B: Western blots of isolated lipid rafts in control and PAF-treated cells (fraction 1, lightest; fraction 12, heaviest). Fractions were probed against eNOS and caveolin.

eNOS internalization via caveolae and PAF-induced hyperpermeability.

Because PAF-induced eNOS preferential translocation to cytosol is associated with PAF-induced hyperpermeability (30), we hypothesized that stimulated trafficking of eNOS-containing caveolae may serve to deliver eNOS to subcellular effectors which determine the functional outcome of the initial stimulus. To test this hypothesis, we inhibited eNOS internalization by transfecting CVECs with dyn2K44A, a dominant-negative mutant of dynamin-2 (18, 24, 25). Figure 4A shows the expression of dyn2K44A in CVECs. The efficiency of transfection, assessed by cell counting (fluorescent vs. total cells), was 60%. Transfection of CVECs with dyn2K44A completely prevented PAF-induced hyperpermeability, whereas CVEC transfected with the empty vector (used to deliver dyn2K44A) developed a hyperpermeability response to PAF (Fig. 4B), showing that the transfection of CVECs with the empty vector did not inhibit PAF-induced hyperpermeability.

Fig. 4.

Inhibition of caveolar internalization decreases PAF-induced hyperpermeability in CVECs. A: expression of dyn2K44A in transfected CVECs. B: impact of CVEC transfection with dyn2K44A on permeability to FITC-DX-70. CVECs transfected with the corresponding empty vector served as control. PAF induced a robust hyperpermeability in the control CVECs. Transfection of CVEC monolayers with dyn2K44A significantly inhibited the PAF-induced hyperpermeability to FITC-DX-70 (*P < 0.05 compared with control and interventions, n = 5 experiments).

To further verify the efficacy of dyn2K44A in preventing scission of caveolae from plasma membrane in CVECs, we collected, at the end of the experiment, the snapwell inserts containing the CVEC monolayers used to measure transport of FITC-DX-70 and examined the location of eNOS by indirect immunofluorescence microscopy. Figure 5 displays the images obtained from the experimental snapwells. Control CVECs show eNOS located in the cell membrane and Golgi. Nontransfected cells show that 100 nM PAF causes translocation of eNOS to subcellular locations, since eNOS appears diffuse in the cytosol. Importantly, CVECs transfected with dyn2K44A demonstrate that eNOS is located at the cell membrane and Golgi after application of 100 nM PAF, indicating that dyn2K44A is highly efficacious in preventing the movement of eNOS from the cell membrane (caveolae) to cytosol.

Fig. 5.

Dyn2K44A inhibits eNOS traffic in CVECs. eNOS inmunofluorescence images taken from CVEC monolayers at the end of the measurement of permeability. In the control (baseline) CVEC, eNOS is distributed in the cell membrane and in the Golgi area. Nontransfected CVECs show eNOS translocation to cytosol (diffuse material) after challenge with PAF (middle). In contrast, eNOS is efficaciously retained in the plasma membrane in the dyn2K44A-transfected CVECs after administration of 100 nM PAF.


Our results demonstrate that 1) eNOS expression is required for robust hyperpermeability response to PAF, a proinflammatory agent; 2) internalization of eNOS is an important element of the signals determining the permeability-enhancing function of eNOS-derived NO; and 3) traffic of eNOS via caveolae is associated with the onset of hyperpermeability.

The translocation of eNOS from plasma membrane to subcellular compartments in response to agonists has been reported to occur via enzyme depalmitoylation (29, 35) and/or in association with the internalization via caveolae (4) in ECs derived from large vessels. A few reports have advanced speculations concerning the functional consequences of eNOS translocation (15, 37). Our work is the first to suggest a close association between eNOS translocation via caveolae and an increase in endothelial monolayer permeability to macromolecules. In addition, our data strongly indicate that eNOS translocation is required to stimulate the onset of PAF-induced hyperpermeability.

Although our results are strengthened by the fact that we obtained them in ECs derived from the microvascular segment normally involved in enhancing the transport of macromolecules in inflammation, the preferential translocation of eNOS to the cytosol by agents that cause hyperpermeability appears to be a property shared by other ECs in culture. VEGF (10) and bradykinin (23), two well-characterized hyperpermeability-enhancing agents, cause eNOS translocation to cytosol. In contrast, acetylcholine, an agent that causes vasodilation via eNOS-derived NO but does not alter permeability, induces preferential translocation of eNOS to the Golgi region (30). The consistency of eNOS translocation to defined subcellular locations in response to agents that induce different functional end points argues against the interpretation that eNOS movement reflects regular protein traffic and argues in favor of a response to specific external stimuli or agonists.

The pharmacological inhibition of eNOS blocks hyperpermeability induced by PAF, VEGF, and histamine in vivo and in cultured ECs (2, 20, 27, 33, 36). In mice, the deletion of the gene encoding for eNOS leads to a loss of microvascular hyperpermeability responses to PAF (16) and to VEGF (11). We confirm in CVEC monolayers that eNOS expression is an absolute requirement for the development of increased endothelial permeability to macromolecules in response to PAF. Interestingly, the loss or depletion of eNOS does not influence baseline permeability in mice and in CVECs. It seems that baseline permeability is maintained at a set point by a number of redundant mechanisms, whereas the ability to respond to proinflammatory agents (PAF and VEGF) is exquisitely sensitive to or dependent on mechanisms based on functional eNOS.

Based on our results, we propose eNOS internalization via caveolae as a novel mechanism in endothelial regulation of microvascular permeability to macromolecules in response to PAF, a well-known proinflammatory agent. Our data support the concept that eNOS-derived NO serves as an onset signal for hyperpermeability. This assessment is based on the close temporal correlation among the changes induced by PAF on eNOS activation (Ser1177 phosphorylation at 0.5–1.0 min; Thr495 dephosphorylation at 1.0 min), NO production (peaking at about 1–3 min), and increases in permeability to FITC-Dx-70 detected shortly after the application of PAF.

The concept that eNOS internalization to subcellular compartment plays a role in determining the stimulus-initiated function serves to complement the fund of knowledge demonstrating the activation of eNOS through phosphorylation or dephosphorylation at several sites (7, 8, 12). Although phosphorylation is clearly important as a mechanism to activate eNOS, it has been difficult to ascribe functional specificity to it inasmuch as agents that stimulate different functional outcomes do phosphorylate eNOS at the same sites, as is the case for example for PAF and ACh (30). The internalization of eNOS may serve as a mechanism to bring the enzyme in close contact with soluble guanylyl cyclase (sGC), its main effector, or an as-yet unidentified subcellular effector.

Our data support the concept that eNOS translocation and its association with the elevation in permeability involves the internalization of eNOS via caveolae. This idea is based on the ability of dyn2K44A to block the scission of caveolae from the plasma membrane, an event intimately linked to the inhibition of PAF-induced hyperpermeability to macromolecules. Even though we cannot completely rule out the contributions of endocytosis via coated-pit vesicles, it is normally accepted that caveolae cover about 85% of EC membranes (26). Thus we interpret our results with dyn2K44A as an inhibition of caveolar internalization. It is known, mainly through in vitro experiments, that dynamin-2 interacts with eNOS (3). For this reason, it could be argued that dyn2K44A blocks eNOS directly. However, it has been pointed out that dynamin-2 and eNOS colocalize mainly in Golgi and that the eNOS-binding domain with dynamin resides at a location different from the K44 mutation (3, 4). Thus it appears safe to conclude that the relevant action of dyn2K44A is to prevent the separation of caveolae from the plasma membrane. Our data (Fig. 5) support this conclusion, since they show that eNOS stays at the plasma membrane after PAF challenge in dyn2K44A-transfected CVECs. However, a recent article reports that dyn2K44A can inhibit caveolar endocytosis and NO production in response to gp60 (22). This article neither addressed the mechanism by which dyn2K44A inhibits NO production nor the discrepancy that the K44A mutation site should not alter NO production. Thus it is plausible that dyn2K44A simultaneously inhibits PAF-induced permeability through anchoring caveolae to plasma membrane and inhibiting caveolae-associated NO production.

Why is eNOS internalization via caveolae necessary? There are no evidence-based answers to these questions yet. Given that NO is a highly diffusible gas, one would anticipate that location of the source would not be essential. We and others have shown that the hyperpermeability response is impaired in the presence of NOS inhibitors such as lgname and NG-monomethyl-l-arginine (2, 20, 27, 33, 36). Furthermore, in eNOS knockout mice, NO produced by other NOS isozymes does not restore the ability of striated muscle microvasculature to produce a robust hyperpermeability in response to PAF (16). Thus not only is eNOS expression needed for adequate microvascular function, but its location is also important. In regard to traffic via caveolae, we speculate that caveolae may possess a necessary target-recognizing molecule that allows eNOS to efficaciously promote the appropriate protein-protein signaling interactions in the intracellular environment. Another speculation is that internalization via caveolae may serve to protect eNOS from S-nitrosylation, which inactivates the enzyme (8). In addition, we may speculate that this traffic may serve to deliver the appropriate NO concentration to achieve the correct stimulation of sGC, the main NO receptor, a cytosolic protein. As it is established, sGC and its product (cGMP) play a necessary role in the development of hyperpermeability (32). The mechanisms by which sGC-cGMP activate hyperpermeability is unknown; however, cGMP may activate phosphodiesterase-2 and induce the degradation of cAMP (17). In turn, the degradation of cAMP may reduce the barrier properties of the microvascular ECs and thus increase permeability to macromolecules (6, 19, 28).

In conclusion, we report and propose for the first time a novel mechanism in the regulation of microvascular endothelial permeability. Our data strongly indicate that the internalization of eNOS via caveolae is required for PAF-induced hyperpermeability. Moderate increases in permeability are helpful to allow the exchange of macromolecules needed for wound healing and tissue remodeling, but a highly elevated vascular permeability may have deleterious effects. The inhibition of caveolar endocytosis may be of help to control excessive hyperpermeability in inflammation.


This work was supported by National Heart, Lung, and Blood Institute Grant 5RO1-HL-070634 and by institutional grants from the Department of Pharmacology and Physiology, the Dean's Biomedical Research Support (New Jersey Medical School), and the Foundation of the University of Medicine and Dentistry of New Jersey.


  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


View Abstract