INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Pulmonary hypertension occurs commonly in patients with chronic hypoxic lung disease (1). When pulmonary hypertension develops in the face of hypoxia there is remodeling of all three layers of the pulmonary arteries associated with increased replication of the cells that constitute the arterial wall (2), but the cause of this increased replication is not understood.

One of the most interesting paradoxes in biology, known since 1946, is that the pulmonary vessels constrict to hypoxia, whereas those in the systemic circulation dilate (3). The effect of hypoxia on replication of pulmonary vascular and systemic vascular cells is not understood. We have been interested in the effects of hypoxia on pulmonary vascular fibroblasts because it is the matrix proteins generated by these cells in the media and adventitia that render the vessels indistensible by vasodilators, the consequence being a fixed state of pulmonary hypertension. We have previously shown that hypoxia increases the rate of replication of pulmonary artery fibroblasts both directly (4) and indirectly via an effect on the endothelium (5). We therefore wished to compare the response of pulmonary artery fibroblasts with those derived from the systemic circulation to establish whether hypoxia had a different effect on the replication of systemic artery cells, i.e., that there was a link between the observed physiological vasoconstriction and vascular remodeling. We hypothesized that hypoxia would increase replication of pulmonary vascular cells but not those derived from systemic vessels. Of available systemic vascular cells we chose those from the mesentery because previous work has suggested that hypoxia affects ion exchange (potassium currents) differently in mesenteric and pulmonary artery cells (6). We also wished to determine whether a cell-signaling pathway intimately associated with cell replication and known to be activated by hypoxia (7)the inositol 1,4,5-triphosphate (IP₃) pathwaywas affected differently by hypoxia in the two types of cells.

To test this hypothesis, we obtained bovine pulmonary artery and bovine mesenteric artery fibroblasts from explants and carried out growth and cell-signaling studies under normoxic and hypoxic conditions.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Materials

All general-purpose compounds were of an analar grade and were obtained from Sigma (Poole, Dorset, UK). [³H]Thymidine was purchased from DuPont (Stevenage, Hertfordshire, UK). All tissue culture flasks and media were obtained from GIBCO (Paisley, Renfrewshire, UK). Fetal calf serum was obtained from Imperial Laboratories (Andover, Hants, UK).

Primary Culture of Bovine Pulmonary Artery Fibroblasts (BPAF)

Freshly excised bovine lung was obtained from the local abbatoir. Lobar pulmonary artery was dissected free from the lung and a section of artery (5 mm in diameter) located toward the apex of the lobe was cut longitudinally and opened into a flat sheet. Pulmonary artery fibroblasts were prepared using the technique of Freshney (8), with some modifications. Briefly, muscular tissue and endothelial cell layers were removed by gentle abrasion of the vessel. The remaining tissue (adventitia) was then dissected into 1 mm³ portions. Approximately 25 portions of tissue were evenly distributed over the base of a 25-cm² culture flask containing 2 ml of DMEM with 20% fetal calf serum (FCS), supplemented with penicillin/streptomycin (400 IU/ml and 400 µg/ml) and amphotericin B (5 µg/ml). The explants were incubated in a humidified atmosphere of 5% CO₂ in air at 37° C. We have previously shown by staining for actin (9) that this technique provides a pure culture of fibroblasts (10). Cells were maintained in DMEM containing 10% FCS, supplemented with penicillin/streptomycin (200 IU/ ml and 200 µg/ml) and L-glutamine (27 mg/ml) and used between passages 3 to 10.

Primary Culture of Bovine Mesenteric Artery Fibroblasts (BMAF)

Approximately 1 m of freshly excised bovine small intestine was obtained from the local abbatoir. A mesenteric artery of the same diameter as that of the pulmonary artery was located in the fatty tissue connected to the intestine. Once isolated, the procedure for obtaining fibroblasts was the same as for the BPAF.

Growth of Cells in a Hypoxic Environment

A humidified temperature-controlled incubator (Model GA156; LEEC, Colwick, Nottingham, UK) was used as a hypoxic chamber. This incubator allows control of internal oxygen levels between 0 and 21% while the CO₂ level is simultaneously controlled at 5%. For these experiments, cells were transferred to 24-well plates. From our previous work (4) we elected to use a tissue culture supernatant PO₂ of 35 mm Hg, which could be obtained by maintaining an atmosphere of 2% oxygen. We have measured the PO₂ within and between the wells of the plates and found the PO₂ to be constant.

We discovered that, because of the slow speed of gaseous diffusion, a decrease in atmospheric oxygen did not have an immediate effect on the PO₂ of the supernatant and that a period of preincubation in hypoxia was necessary so that the desired supernatant PO₂ could be achieved at the start of the 24-h incubation period to which all cells to be made hypoxic were subjected.

Preincubation time. To establish the desired length of this preincubation period we measured both PO₂ in the supernatant and also the effects on cellular replication as the preincubation period was varied between 0 and 24 h. We found that after 6 h of preincubation the PO₂ had reached the desired level of 35 mm Hg (see RESULTS).

Effects of hypoxia on replication of pulmonary and mesenteric fibroblasts. After a 6-h preincubation period, we subjected both BPAF and BMAF cells to 24 h of hypoxia in the presence or absence of 0, 0.1, 0.5, 1, and 2% serum and measured replication by [³H]thymidine incorporation assay as described below. We then assessed the effect on proliferation, under both normoxic and hypoxic conditions, of stimulating the cells through different receptors. We used a G-protein-linked receptor agonist, endothelin-1 (ET-1) (10⁷ M) and a tyrosine kinase receptor agonist, platelet-derived growth factor (PDGF). Different doses of PDGF (0.1, 3, and 10 ng/ml) were used to elucidate whether hypoxia increased the maximal replicative ability of the cells or simply sensitized them to grow more readily in response to a known agonist.

[³H]Thymidine Incorporation

Bovine pulmonary and mesenteric artery fibroblast cells were grown to approximately 60% confluency (to allow room for replication) in 24-well plates at 37° C.

To measure cell replication fibroblasts were pulsed with [³H]thymidine (0.1 µCi/well) 4 h before the end of the 24 h of stimulation. At 24 h the medium was removed and the cells were washed twice with 0.5 ml phosphate-buffered saline (PBS). Cellular proteins were precipitated by washing with 5% trichloroacetic acid (TCA), and lipid fractions were solubilized by washing with 100% ethanol. The remaining cell contents were dissolved in 0.3 M NaOH. The contents of each well were transferred to scintillation vials, to which was added 5 ml of Ecosint A scintillation fluid. Vials were vortexed thoroughly before radioactive counts were measured by scintillation counter. Counts were measured in DPM (disintegrations per minute).

Effect of Hypoxia on IP₃ Generation

IP₃ is a second messenger known to mobilize intracellular stores of calcium, and hence of possible importance in cell growth. We studied the effects of hypoxia on IP₃ generation in BMAF and BPAF cells in the presence or absence of ET-1 and PDGF.

Mass measurement of IP₃. IP₃ was measured using the technique of Palmer and Wakelam (11). Briefly, the technique is as follows.

1. Principle of IP₃ mass assay. Samples containing unknown amounts of IP₃ are placed in direct competition with [³H]IP₃ for binding sites on a protein isolated from bovine adrenal glands. Low associated radioactivity in the assay sample therefore reflects a high level of IP₃ in the sample. In order to establish the relationship between measured radioactivity and IP₃ concentrations a standard curve was constructed using known concentrations of IP₃.
2. Sample preparation. Cells were grown to 100% confluency (since no space was required for growth and we needed to achieve an assayable quantity of IP₃ for measurement) on 24-well plates and quiesced in serum-free medium for 48 h under normoxic conditions. The reaction was started by the addition of agonists, endothelin-1 (10⁷ M) or PDGF (30 ng/ml). The reaction was stopped by the addition of 25 µl ice-cold 10% perchloric acid (PCA) wt/vol at different time points (0, 5, 10, 20, 30, 60, 120, and 300 s). Cells were mechanically disrupted by scraping, and the sample was then centrifuged. The resulting supernatant was stored at 20° C until time of assay.
3. The binding assay. Fifty microliters of each sample were added to 25 µl of diluted [³H]IP₃ (3,000 cpm/25 µl) and 25 µl of incubation buffer. The reaction was then started by the addition of 25 µl of diluted IP₃ binding protein previously prepared from bovine adrenal glands. Samples were centrifuged and the supernatant was discarded. The radioactivity of the resultant pellet was measured in a scintilation counter and compared with a standard curve generated using known concentrations of IP₃. The associated radioactivity was then measured in a scintillation counter.

Statistics

Each result represents the mean of four experiments performed on cells from the same animal. All results shown were confirmed in additional experiments on four different animals. Data in Table 1 were analyzed by analysis of variance (ANOVA) and data in Figures 1 and 3 by two way ANOVA. Results were considered significant at p < 0.05. Comparison between means (Figure 2) was determined using Student's t test (significant at p < 0.05).

View larger version (11K): [in this window] [in a new window]   Figure 1.   Effect of hypoxia on serum-induced BPAF and BMAF cell proliferation. BPAF and BMAF cells were grown to 60% confluency in normoxia and 10% serum in 24-well plates and then quiesced in 0% serum for 24 h before stimulation with 0, 0.1, 0.5, 1, or 2% serum. The cells were then allowed to grow in either normoxic or hypoxic conditions for a period of 24 h. Cell growth was assessed by [3H]thymidine uptake. Hypoxia caused significant increases in [3H]thymidine uptake in BPAF cells in the presence of serum (a), but had no effect on BMAF (b). Values shown are the mean ± SD for four replicate plates from the same animal. The experiment was repeated in four per minute. Norm = normoxic; Hyp = hypoxic. Asterisks indicate value for hypoxia significantly greater than those for normoxia.

View larger version (11K): [in this window] [in a new window]   Figure 3.   Effect of hypoxia on PDGF-stimulated proliferation of BPAF and BMAF cells. BPAF (a) and BMAF (b) cells were grown to 60% confluency in 24-well plates. The cells were then allowed to grow in both normoxic and hypoxic conditions for a period of 24 h in the presence of varying concentrations of PDGF (0.1, 3, and 30 ng/ml). Cell growth was assessed by [3H]thymidine uptake. Hypoxia does not increase the maximal replicative ability of either BPAF or BMAF cells, but sensitized the BPAF cells (but not the BMAF cells) to the replicative effects of PDGF (p < 0.05). Values shown are the mean ± SD for four replicate plates from the same animal. The experiment was repeated in four different animals and the results shown are typical of those obtained. DPM = disintegrations per minute; Norm = normoxic; Hyp = hypoxic; PDGF = platelet-derived growth factor. Asterisks indicate values for hypoxia significantly greater than those for normoxia.

View larger version (15K): [in this window] [in a new window]   Figure 2.   Effect of hypoxia on endothelin-1 and PDGF-induced stimulation of BPAF and BMAF cell proliferation. BPAF (a) and BMAF (b) cells were grown to approximately 60% confluency in 24-well plates. The cells were then allowed to grow in either normoxic or hypoxic conditions for a period of 24 h in the presence or absence of ET-1 (107 M) or PDGF (3 ng/ml). Cell growth was assessed by [3H]thymidine uptake. Endothelin-1 did not stimulate either type of cell above control level under either normoxic or hypoxic conditions. PDGF had a marked effect on both BPAF and BMAF cells, but the effect on BPAF (though not BMAF) cells was enhanced by hypoxia (p < 0.05). Values shown are the mean ± SD for four replicate plates from the same animal. The experiment was repeated in four different animals and the results shown are typical of those obtained. DPM = disintegrations per minute; Norm = normoxic; Hyp = hypoxic, ET-1 = endothelin 1; PDGF = platelet-derived growth factor.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Preincubation Time

It can be seen in Table 1 that a 6-h preincubation period was required for the bathing medium of the cells to reach the desired level of hypoxia (35 mm Hg). This time period also coincided with the largest difference in incorporation between unstimulated normoxic (PO₂, 147 mm Hg) and hypoxic BPAF, a longer period of preincubation diminished thymidine incorporation, at least in the hypoxic cells.

Effects of Hypoxia on Replication of Pulmonary and Mesenteric Fibroblasts

Hypoxia alone enhanced the replication of BPAF cells (Figures 1a and 2a). This response, however, was not seen with the BMAF cells (Figures 1b and 2b). Hypoxia also enhanced the replication of BPAF cells (Figure 1a) in the presence of serum (p < 0.05), but this response was not seen with the BMAF cells (Figure 1b).

ET-1 (10⁷ M) did not stimulate either type of cell above basal levels in normoxic conditions. The apparent stimulation in hypoxic conditions was no greater than that seen with hypoxia alone (Figures 2a and 2b). PDGF caused a marked increase in replication under both normoxic and hypoxic conditions (p < 0.05). The effect was greater in conditions of hypoxia for the BPAF cells (Figure 2a) than for the BMAF cells (Figure 2b). PDGF caused a dose-dependent increase in cell replication for both types of cell, but whereas hypoxia shifted this dose-dependent curve (but not maximal response) to the left in BPAF cells (Figure 3a), this effect was not seen in BMAF cells (Figure 3b).

Effect of Hypoxia on IP₃ Generation

PDGF (30 ng/ml) did not increase the generation of IP₃ by BPAF cells under normoxic conditions, but it caused a large rise in IP₃ generation in these cells under hypoxic conditions (Figure 4a). There was no stimulation of IP₃ by PDGF in BMAF cells under either normoxic or hypoxic conditions (Figure 4b). Hypoxia did not affect ET-1-induced stimulation of IP₃ generation in either type of cell.

View larger version (12K): [in this window] [in a new window]   Figure 4.   Effects of hypoxia on ET-1 and PDGF-induced generation of IP3 in BPAF and BMAF cells. BPAF and BMAF cells were grown to 100% confluency in 24-well plates and quiecsed in serum-free medium for 24 h under normoxic or hypoxic conditions. ET-1 (107 M) or PDGF (30 ng/ml) was added and IP3 generation was measured at 0, 5, 10, 20, 30, 60, 120, and 300 s. Hypoxia significantly (p < 0.05) enhanced the effects of PDGF on IP3 generation on BPAF cells (a) but not in BMAF cells (b). Hypoxia had no effect on ET-1-induced IP3 generation in either cell type. Values shown are the mean ± SD. There were four replicate plates from the same animal on different days. Norm = normoxic, Hyp = hypoxic; ET-1 = endothelin-1; PDGF = platelet-derived growth factor.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study showed that hypoxia alone increased proliferation of pulmonary artery fibroblasts but did not affect the replicative ability of fibroblasts harvested from the mesentery. Furthermore, the effects of serum and PDGF on replication were enhanced by hypoxia only the pulmonary artery cells. The effect of hypoxia on serum and PDGF-induced cell replication appeared to be an effect on sensitivity to the agonist rather than the maximal responses since the dose-response curve in normoxia had not plateaued. Superficially, it may appear that differences between the responses of BMAF and BPAF to serum were due to a poor response of BPAF in normoxia, but absolute levels of cell replication as measured by [³H]thymidine incorporation cannot be compared between cell types. More important, we believe, is the difference in response to the different environments in the same cell types. The measured increase in proliferation with hypoxia was associated with an increased IP₃ generation in the pulmonary artery cells but not in the mesenteric artery cells. The effects of ET-1 on either cell proliferation or IP₃ generation were unaffected by hypoxia in either cell type. The effect of hypoxia on PDGF stimulation suggests that both the replicative response and IP₃ generation were likely to have been a consequence of tyrosine kinase activation.

The results presented in this report were repeated in cells from the same animal and also in cells from different animals. The measured differences were large, suggesting that the differences between pulmonary and mesenteric cells was genuine. The results were also consistent with differences in physiological responses to hypoxia in the intact circulation and isolated vessels and also the differential effect of hypoxia on K⁺ channel activation in isolated pulmonary and mesenteric artery cells (6).

It is not surprising that mitogens such as PDGF and serum promoted cell replication, but interesting that this replication was enhanced by hypoxia in the pulmonary cells (but not in the mesenteric cells). It may appear surprising that the other mitogen we tested, ET-1, shown by ourselves (10) to be a potent growth factor for pulmonary artery fibroblasts in the rat, did not affect bovine pulmonary artery cells in either hypoxic or normoxic conditions, but this is consistent with our previous studies where we showed that ET-1 does not promote replication in bovine cells (12). We have been unable to find any other study where bovine mesenteric artery fibroblast cell replication has been studied.

The difference in response to hypoxia between pulmonary cells and mesenteric cells may be due to a hypoxia-activated pathway that is either present only in pulmonary artery cells or is repressed in mesenteric artery cells. Activation of cell-surface receptors and their associated tyrosine kinase and phospholipases (C-, A₂, or D) leads to the generation of lipid second messengers (IP₃ and diacyl glycerol [DAG]) (13, 14). IP₃ promotes mobilization of intracellular Ca²⁺ stores (15), whereas DAG is an endogenous activator of protein kinase C (PKC) (14). It is logical to deduce that an increase in IP₃, caused in BPAF by hypoxia, would be accompanied by an increase in DAG. Thus, PKC activity would be enhanced in two ways, firstly by increasing DAG and secondly by IP₃-induced release of intracellular Ca²⁺ stores. Increases in intracellular Ca²⁺ can directly activate PKC (16). Interestingly, activators of the PKC pathways have been shown to promote many of the same cellular responses (contraction, hypertrophy, and proliferation) attributed to hypoxia (17). If this explanation of the activation of hypoxia is true, it is not clear why the same pathway is not activated in mesenteric cells.

It is possible that pulmonary artery fibroblasts have a sensing system whereby promotor genes are turned on in the face of hypoxia. There is evidence in other cell types of specific genes switched on by hypoxia, most notably the erythropoietin gene (20). It is likely that pulmonary artery cells contain genes that allow both contraction and replication to hypoxia and that these phenomena are coupled (21). It is tempting to speculate that the phosphoinositide pathway might be involved downstream from any hypoxic sensor and responsible for vasomotor cell growth coupling, i.e., IP₃ causes increased intracellular Ca²⁺, resulting in cellular contractions but also augmentations of DAG-induced cell replication. We have not measured DAG or have we measured other cell-signaling pathways that could be involved such as MAP-kinase and PI3-kinase, but the fact that IP₃ generation in our model was increased by the same conditions that initiated cell replication suggests that there might be a connection between this pathway and the stimulus to cellular replication.

The observations reported here may have clinical implications. Pulmonary hypertension and pulmonary vascular remodeling accompany nearly all forms of cardiac and respiratory disease, and there is considerable interest in drugs that might be able to prevent or reverse this process. It may be possible to intervene at the receptor, cell membrane, cytosolic, or even nuclear levels and prevent the unnecessary cell replication, providing we first understand the intracellular events leading to this replication. The differences in response between systemic and pulmonary artery cells may lead us to a better understanding of these processes, but, more importantly, may allow us to intervene in such a way that the pulmonary vascular response is modified without any simultaneous deleterious effect on the systemic circulation.