Mitogen-Activated Protein Kinases Regulate Platelet-Activating Factor-Induced Hyperpermeability

Abstract

Objective

The authors tested the hypothesis that p42/44- (ERK-1/2) and/or p38-mitogen-activated protein kinases (MAPK) are in vivo regulatory elements in the platelet-activating factor (PAF) activated signaling cascade that stimulates microvascular hyperpermeability.

Methods

FITC-dextran 70 was used as the macromolecular tracer for microvascular permeability in the mouse mesenteric fat tissue. Interstitial integrated optical intensity (IOI) was used as an index of permeability.

Results

An application of 10⁻⁷ MPAF increased IOI from 23.1 ± 3.6 to 70.8 ± 7.4 (mean ± SEM). Inhibition of ERK-1/2 with 3 μM and 30 μM AG126 reduced IOI to 32.3 ± 2.5. Similarly, inhibition of p38-MAPK with 6 nM, 60 nM and 600 nM SB203580 lowered IOI to 29.1 ± 2.4.

Conclusions

The results demonstrate that ERK-1/2 and p38MAPK participate in the signaling cascade that regulates PAF-induced microvascular hyperpermeability in vivo.

Hyperpermeability is an important hallmark of the inflammatory reaction that characterizes vascular disease and wound healing. However, the signaling pathways that regulate microvascular permeability are only partially understood (1). We and other investigators have documented in vivo and in vitro that endothelial nitric oxide synthase (eNOS, through production of nitric oxide [NO]), protein kinase C (PKC), PKB/akt, and phospholipase C (PLC) are elements of the signaling cascade triggered by platelet-activating factor (PAF), vascular endothelial growth factor (VEGF), and other pro-inflammatory agonists (2-18,40).

We have shown in vitro that p42/44 (ERK-1/2) and p38MAPK (mitogen-activated protein kinases) mediate VEGF-induced hyperpermeability in human umbilical vein endothelial cells (HUVEC) (2,3,19). The purpose of this study was to test the hypothesis that p42/44 and/or p38 MAPK regulates the PAF-stimulated signaling cascade, leading to microvascular hyperpermeability in vivo.

MATERIALS AND METHODS

Animals

Thirty-five male wild-type C57BL-6J mice (Japan Biological Material Center, Saitama, Japan), 10–15 weeks old, 25–35 g in body weight, were used in this study. The study was approved by the Animal Care and Use Committee at the University of Tokyo, School of Medicine, and complied with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Intravital Microscopy and Digital Image Analysis

A U-LH 100HG microscope (Olympus Optical CO, LTD, Japan) was used for intravital microscopy. The image was transferred through a CCD camera C6790-81 (Hamamatsu Photonics, Japan) to a computer with digital image analysis software “U4469-01 Argus/Hisca Software” (Hamamatsu Photonics) for analysis.

Microvascular permeability was measured as interstitial integrated optical intensity (IOI) (20,21), which was calculated by integrating gray-scale values of 512 × 512-pixel windows at each selected area. Five areas in the fat tissue interphase of the mesentery were randomly selected. The same areas were recorded and analyzed after the experimental interventions. Net IOI, used in the evaluation of permeability changes, was defined as the difference between experimental IOI value and average of baseline IOI value.

Surgical Technique

Under general anesthesia (50–60 mg/kg of sodium pentobarbital, IP), the anterior neck and abdomen were prepared with an electronic shaver. The mouse was placed on a heating pad (temperature was maintained at 37°C) under the dissecting microscope in the supine position, and the forelegs, hind legs, and jaws were mildly stretched to facilitate the operation. Tracheotomy was performed with an Intramedic polyethylene tube PE-50 (Becton Dickinson, Japan) to ensure clear airway passage. The left jugular vein was cannulated with PE-10 for the administration of supplemental doses of anesthetic.

After full midline incision, the mouse was placed in the prone position on the saline-filled mesentery chamber with the mesentery gently splayed to fit in the chamber. The mesenteric preparation was super-perfused with bicarbonate buffer (adjusted to pH 7.4 and continuously equilibrated with 95% nitrogen/5% carbon dioxide).

Experimental Design

A period of 60 min was allowed for stabilization. Subsequently, FITC-dx 70 was administered intravenously 15 min before baseline measurements. Pretreatment with topical application of MAPK inhibitors (AG126 or SB203580) was initiated 10 min before PAF application and maintained throughout the experiment. PAF was topically applied to the mesentery fat tissue for 3 min. The intravital microscopic image was captured by a TV camera and transferred to the computer. Microvascular transport was measured using a computer-assisted digital image analysis system showing the image and the value of IOI. Baseline data was acquired every 10 min three times and experimental data was recorded every 5 min for 60 min.

Chemicals and Agents

All chemicals were purchased from Sigma-Aldrich (St Louis, MO) unless otherwise noted. PAF (dl-α-phosphatidylcholine, β-acetyl-γ-o-hexadecyl) at 10⁻⁷ M was used as a pro-inflammatory agonist. Preliminary data demonstrated that 10⁻⁷ M PAF is an effective dose to induce permeability changes in the mouse mesentery. This concentration of PAF is in agreement with our earlier data in the hamster cheek pouch (22). FITC-dx 70 (fluorescein isothiocyanatedextran 70 kDa) served as the macromolecular tracer for microvascular permeability. AG126 (Tyrphostin AG 126, 3-hydroxy-4-nitrobenzylidene) at 3 and 30 μM was used to inhibit ERK-1/2. SB203580 [4-(4-fluorophenyl)-2-(4-methylsulfinyl-phenyl)-5-(4-pyridyl)-1h-imidazole] at 6, 60, and 600 nM was used to selectively block p38MAPK.

The composition of bicarbonate buffer (pH 7.4) was (in mM) 131.9 NaCl, 4.7 KCl, 2.0 CaCl₂, 1.2 MgSO₄, and 18.0 NaHCO₃. PAF, AG126, and SB203580 were dissolved in dimethyl sulfoxide (DMSO) to stock concentrations. Immediately before topical application, a PAF aliquot was diluted to 10⁻⁷ M with bicarbonate buffer containing 1.5% bovine serum albumin. FITC-dextran 70 was injected initially intravenously as a bolus (100 mg/kg), and then infused continuously throughout the experiment at the rate of 0.15 mg/kg/min via a jugular vein. The tracer solution was prepared by diluting FITC-dextran 70 in 0.9% NaCl. Bovine serum albumin (BSA; 1.5% solution) was also dissolved in 0.9% NaCl.

Statistical Analysis

IOI data are presented as means ± SEM for each time-point. Time-related changes in IOI and differences of IOI between the control group and experimental group with MAPK inhibitors were analyzed with one-way ANOVA (SPSS), followed by Tukey's test. Statistical significance was accepted at p < .05.

RESULTS

PAF-Induced Hyperpermeability

Topical application of 10⁻⁷ MPAF induced large and significant changes in microvascular permeability relative to control, as indicated by analysis of IOI (Figure 1). An increase in permeability was detected in the first sample (5 min) after PAF and remained elevated for the remainder of the experiment. PAF at 10⁻⁷ M increased IOI from 23.1 ± 3.6 to 70.8 ± 7.4 (mean ± SEM).

Figure 1.

PAF (10⁻⁷ M) increases microvascular permeability in mesenteric tissue–fat interphase. The panel shows the time course of microvascular permeability as assessed by changes in net IOI. PAF significantly increased IOI relative to control. Data are means ± SEM (n = 5 for control and PAF experiments; p < .05 at t = 5 min; p < .001 thereafter).

ERK-1/2 and p38MAPK Regulate PAF-Induced Microvascular Permeability

The selective ERK-1/2 inhibitor AG126 at 3 and 30 μM significantly attenuated the microvascular hyperpermeability induced by 10⁻⁷ M PAF to 32.3 ± 2.5 (Figure 2). There was no significant difference in the degree of attenuation between AG126 at 3 and 30 μM. AG126 at 0.3 μM had no inhibitory effect. The statistical analysis of net IOI of PAF-induced permeability changes with or without p42/44 MAPK inhibitor yielded the following values for significance: no AG vs. AG 3 μM: p = .003; no AG vs. AG 30 μM: p = .001; AG 3 μM vs. AG 30 μM: p = .792

Figure 2.

Inhibition of p42/p44 MAPK, with AG126, reduces PAF-induced hyperpermeability. The panel displays the time course of IOI as an index of microvascular permeability to FITC-dextran 70. Inhibitor was applied topically via superfusate, starting 10 min before topical application of 10⁻⁷ M PAF and continued throughout the remainder of the experiment. AG126 significantly inhibited PAF-induced hyperpermeability. Data are means ± SEM (n = 5 for PAF and each AG126 concentration; p < .01 at t ≥ 20 min).

SB203580, a selective inhibitor of p38MAPK, significantly reduced the increase in IOI induced by PAF at 10⁻⁷ M to 29.1 ± 2.4 (Figure 3). All concentrations of SB203580, 6, 60, and 600 nM, produced similar reductions in PAF-stimulated hyperpermeability (p > .05), while 0.6 nM SB203580 had no effect. The statistical analysis of net IOI of PAF-induced permeability changes with or without SB203580 (p38 MAPK inhibitor) yielded the following values for significance: no SB vs. SB 6 nM: p = .006; no SB vs. SB 60 nM: p = .011; no SB vs. SB 600 nM: p = .009; SB 60 nM vs. 6nM: P = .953; SB 600 nM vs. 6 μM: p = .826; SB 60 nM vs. 600 nM: p = .821.

Figure 3.

Inhibition of p38-MAPK, with SB203580, reduces PAF-induced hyperpermeability. The panel displays the time course of IOI as an index of microvascular permeability to FITC-dextran 70. Inhibitor was applied topically via superfusate, starting 10 min before topical application of 10⁻⁷ M PAF and continued throughout the remainder of the experiment. SB203580 significantly inhibited PAF-induced hyperpermeability. Data are means ± SEM (n = 5 for PAF and each SB203580 concentration; p < .01 at t ≥ 20 min).

DISCUSSION

Our experiments demonstrate in vivo that ERK-1/2 and p38 MAPK are operational elements of the signaling pathways that regulate PAF-stimulated microvascular hyperpermeability. The conclusion rests on evidence from other laboratories supporting the concept that AG126 (23) and SB203580 (24,25) are specific inhibitors of ERK-1/2 and p38MAPK, respectively. Our observations and conclusions are consistent with our previous in vitro study where inhibition of either ERK-1/2 or p38MAPK significantly reduced VEGF-induced hyperpermeability in HUVEC (2,3). These studies, in vivo and in vitro, provide strong evidence that both ERK-1/2 and p38 MAPK are involved in the signal transduction regulating microvascular permeability.

The basic elements of the signaling pathways initiating the control of microvascular permeability have been identified through work from several laboratories, including our own. Elements that have been confirmed in vitro and in vivo include specific receptors (usually G-protein coupled receptors), PLC, PKB/akt, PKC, eNOS, ERK-1/2, and p38MAPK (1-12,14,15,17-19,40). The precise sequence of signaling still remains to be elucidated. Whether all these elements work in series or some are arranged in parallel is an open question. While there is consensus that PKB/akt phosphorylates eNOS at serine 1179, causing release of NO (18), there are presently two opposing views regarding the sequence of biochemical events involving MAPK and eNOS. One view is that the activity of eNOS releases NO, which increases the production of cGMP by soluble guanylyl cyclase and activates the cGMP-dependent kinase, PKG. In turn, PKG activates MAPK. Thus, MAPK may be considered a signal element downstream of eNOS. The opposing view is that MAPK are upstream of eNOS. According to this view, MAPK are activated by an agonist (such as bradykinin) and phosphorylate eNOS (26). Whether MAPK are upstream or downstream of eNOS or represent a parallel pathway as well as the nature of possible cross-talks (2) in the regulation of microvascular permeability remains unresolved. However, it is noteworthy that inhibition of ERK-1/2 decreases baseline permeability in HUVEC (19). In this regard, ERK-1/2 shares this capability with PLC (27) and myosin light-chain kinase (28). Even though this regulatory capacity is difficult to demonstrate in vivo (7), it is plausible and reasonable to envision a signaling pathway that can lower baseline microvascular permeability. MAPK seem to be a signaling element involved in maintenance of baseline microvascular integrity.

In addition to controlling microvascular permeability, the MAPK superfamily has roles in cellular differentiation and proliferation (29), apoptosis (30-33), and expression of the proinflammatory phenotype in innate immune cells, including monocytes, endothelial cells, and neutrophils (34-39). The answer to the crucial question regarding how the MAPK pathway manages and codes the signaling information to precisely deliver signals to the downstream molecules and to finally elicit specific biological responses remains open to active investigation.