Remodeling of lung interstitium but not resistance vessels in canine pacing-induced heart failure.

We previously showed that pacing-induced heart failure in dogs results in an enhancement of pulmonary vascular reactivity. In the present study we hypothesized that enhanced matrix deposition and structural remodeling of lung resistance microvessels would underlie these functional changes. Using biochemical measures, we found no difference in the normalized lung content of hyaluronan, uronic acid, and collagen between control dogs and dogs paced for 1 mo, although lung dry weight and noncollagen protein content increased significantly in the paced group (P < 0.05). From separate Formalin-fixed lung lobes, 5-microm frozen sections were prepared and stained with Masson's trichrome, and vascular structure was evaluated using standard morphometric techniques. When perivascular fluid cuffs were excluded from the measure of wall thickness, collagen and media volume fractions in any size range did not differ between paced and control groups. Similarly, in the paced group, medial thickness in <400-microm arterial or venular microvessels did not vary significantly from that in the controls. In contrast, the relationship of interstitial fluid pressure to lung water was significantly shifted to the right in the paced group, such that normal tissue pressures were observed, despite the increased water content. We conclude that although 1 mo of pacing-induced heart failure results in altered interstitial function, the attendant pulmonary hypertension and/or hormonal responses are insufficient to induce medial hypertrophy or other remodeling of the extra-alveolar microvasculature.

Phosphotyrosine phosphatase and tyrosine kinase inhibition modulate airway pressure-induced lung injury.

We determined whether drugs which modulate the state of protein tyrosine phosphorylation could alter the threshold for high airway pressure-induced microvascular injury in isolated perfused rat lungs. Lungs were ventilated for successive 30-min periods with peak inflation pressures (PIP) of 7, 20, 30, and 35 cmH2O followed by measurement of the capillary filtration coefficient (Kfc), a sensitive index of hydraulic conductance. In untreated control lungs, Kfc increased by 1.3- and 3.3-fold relative to baseline (7 cmH2O PIP) after ventilation with 30 and 35 cmH2O PIP. However, in lungs treated with 100 microM phenylarsine oxide (a phosphotyrosine phosphatase inhibitor), Kfc increased by 4.7- and 16.4-fold relative to baseline at these PIP values. In lungs treated with 50 microM genistein (a tyrosine kinase inhibitor), Kfc increased significantly only at 35 cmH2O PIP, and the three groups were significantly different from each other. Thus phosphotyrosine phosphatase inhibition increased the susceptibility of rat lungs to high-PIP injury, and tyrosine kinase inhibition attenuated the injury relative to the high-PIP control lungs.

Involvement of cytochrome P-450 enzyme activity in the control of microvascular permeability in canine lung.

Products of cytochrome P-450 enzymes may play a role in capacitative Ca2+ entry in endothelial cells, which can promote a rise in vascular permeability. Thapsigargin (150 nM) stimulated capacitative Ca2+ entry and increased the capillary filtration coefficient (Kf,c) in isolated normal canine lung lobes. Pretreatment of the lobes with cytochrome P-450 inhibitors clotrimazole (10 microM) or 17-octadecynoic acid (5 microM) abolished the thapsigargin-induced increases in Kf,c. Because clotrimazole also blocks Ca2+-activated K+ channels, the K+-channel blocker tetraethylammonium (10 mM) was used to ensure that permeability was not influenced by this mechanism. Tetraethylammonium did not affect thapsigargin-induced permeability. The effects of the cytochrome P-450 arachidonic acid metabolite 5,6-epoxyeicosatrienoic acid (EET) were also investigated in lobes taken from control dogs and dogs with pacing-induced heart failure (paced at 245 beats/min for 4 wk). 5,6-EET (10 microM) significantly increased Kf,c in lobes from the control but not from the paced animals. We conclude that cytochrome P-450 metabolites are involved in mediating microvascular permeability in normal canine lungs, but an absence of 5,6-EET after heart failure does not explain the resistance of lungs from these animals to permeability changes.

Ablation of lung endothelial injury after pacing-induced heart failure is related to alterations in Ca2+ signaling.

We have previously shown that ANG II increases microvascular permeability in normal dog lungs but not after pacing-induced heart failure. This study investigated how ANG II induces permeability in isolated blood-perfused canine lung lobes and what alterations occur during heart failure. In normal lobes, the protein kinase C (PKC) inhibitors staurosporine (500 nM) or chelerythrine (10 microM) did not modify ANG II-induced increases in the capillary filtration coefficient (Kf,c, ml . min-1 . cmH2O-1 . 100 g-1; an index of microvascular permeability), suggesting that PKC is not involved. Thapsigargin (150 nM) was used to stimulate capacitative Ca2+ entry in lobes from control dogs and dogs paced at 245 beats/min for 4 wk to induce heart failure. In control lobes, Kf,c rose after thapsigargin, from 0.06 +/- 0.01 to 0.17 +/- 0.03 ml . min-1 . cmH2O-1 . 100 g-1 (mean +/- SE, P < 0.05) but did not change in the paced group. A Ca2+ ionophore, A-23187, increased Kf,c in both control (10 microM; 0.05 +/- 0.01 to 0.17 +/- 0.05 ml . min-1 . cmH2O-1 . 100 g-1, P < 0.05) and pace (5 microM; 0.06 +/- 0.01 to 0. 21 +/- 0.07 ml . min-1 . cmH2O-1 . 100 g-1, P < 0.05) lobes, indicating that increasing intracellular Ca2+ is sufficient to induce pulmonary microvascular permeability after pacing. We conclude that during heart failure, Ca2+ signaling within the pulmonary microvascular endothelium is altered.

Gadolinium prevents high airway pressure-induced permeability increases in isolated rat lungs.

To determine the initial signaling event in the vascular permeability increase after high airway pressure injury, we compared groups of lungs ventilated at different peak inflation pressures (PIPs) with (gadolinium group) and without (control group) infusion of 20 microM gadolinium chloride, an inhibitor of endothelial stretch-activated cation channels. Microvascular permeability was assessed by using the capillary filtration coefficient (Kfc), a measure of capillary hydraulic conductivity. Kfc was measured after ventilation for 30-min periods with 7, 20, and 30 cmH2O PIP with 3 cmH2O positive end-expiratory pressure and with 35 cmH2O PIP with 8 cmH2O positive end-expiratory pressure. In control lungs, Kfc increased significantly to 1.8 and 3.7 times baseline after 30 and 35 cmH2O PIP, respectively. In the gadolinium group, Kfc was unchanged from baseline (0.060 +/- 0.010 ml . min-1 . cmH2O-1 . 100 g-1) after any PIP ventilation period. Pulmonary vascular resistance increased significantly from baseline in both groups before the last Kfc measurement but was not different between groups. These results suggest that microvascular permeability is actively modulated by a cellular response to mechanical injury and that stretch-activated cation channels may initiate this response through increases in intracellular calcium concentration.

Isoproterenol attenuates high vascular pressure-induced permeability increases in isolated rat lungs.

To separate the contributions of cellular and basement membrane components of the alveolar capillary barrier to the increased microvascular permeability induced by high pulmonary venous pressures (Ppv), we subjected isolated rat lungs to increases in Ppv, which increased capillary filtration coefficient (Kfc) without significant hemorrhage (31 cmH2O) and with obvious extravasation of red blood cells (43 cmH2O). Isoproterenol (20 microM) was infused in one group (Iso) to identify a reversible cellular component of injury, and residual blood volumes were measured to assess extravasation of red blood cells through ruptured basement membranes. In untreated lungs (High Ppv group), Kfc increased 6.2 +/- 1.3 and 38.3 +/- 15.2 times baseline during the 31 and 43 cmH2O Ppv states. In Iso lungs, Kfc was 36.2% (P < 0.05) and 64.3% of that in the High Ppv group at these Ppv states. Residual blood volumes calculated from tissue hemoglobin contents were significantly increased by 53-66% in the high Ppv groups, compared with low vascular pressure controls, but there was no significant difference between High Ppv and Iso groups. Thus isoproterenol significantly attenuated vascular pressure-induced Kfc increases at moderate Ppv, possibly because of an endothelial effect, but it did not affect red cell extravasation at higher vascular pressures.

Neutrophil chemoattractants generated in two phases during reperfusion of ischemic myocardium in the rabbit. Evidence for a role for C5a and interleukin-8.

The neutrophil chemoattractants generated in a model of myocardial infarction in the anesthetized rabbit were investigated. Coronary artery occlusion was followed by reperfusion for periods from 5 min to 4.5 h. Extracts of myocardial tissue in normal and post-ischemic zones were tested for C5a and interleukin-8 (IL-8) using specific radioimmunoassays. In the post-ischemic zone, immunoreactive C5a was detected within 5 min and rose progressively to reach a plateau at 3-4.5 h. In contrast, immunoreactive IL-8 concentrations rose after a delay and were highest at the last time point tested, 4.5 h. Myeloperoxidase activity levels, an index of neutrophil accumulation, rose progressively as the concentrations of chemoattractants increased. Using cation exchange and reversed phase HPLC, immunoreactive C5a and IL-8 co-eluted with authentic standards. Fractions taken at the C5a and IL-8 peaks from reversed phase HPLC exhibited neutrophil aggregating activity which was neutralized by the respective antibody used in the radioimmunoassays. Depletion of circulating neutrophils virtually abolished immunoreactive IL-8 in the post-ischemic myocardial tissue. These observations suggest a sequential release of chemoattractants: the first, C5a is generated in interstitial fluid, followed by IL-8 generated by infiltrating neutrophils. Thus, over the time period studied, IL-8 generation would be expected to be indirectly dependent on C5a production.