Main

One of several control mechanisms to mediate postnatal circulatory adaptations may be local factors; i.e. the capacity of the vessel wall itself to produce and react to vasoactive substances. There are indications of maturational changes in the effect of endothelium on both relaxant and contractile responses after birth. Studies on isolated pulmonary vessels have shown developmental changes in endothelium-dependent relaxation in sheep and piglets(1–4). An increasing relaxant effect of ACh from birth toward the second week of life was reported in porcine pulmonary arteries(3, 4). A recent study on isolated intrapulmonary arteries from newborn piglets revealed a developmental change in the effect of endothelium on the contractile responses to various agonists within the first 10 d of life(5).

Vascular endothelium plays a crucial role in mediating the dilator response to ACh and a number of other vasoactive substances via NO and an increase in cGMP(6–8). A hyperpolarizing effect of ACh was also recognized previously, based on the use of microelctrode technique to record membrane potential(9). The existence of an EDHF[for review, see Garland et al.(10)] different from NO was based on the findings of an inability of NOS inhibitors(e.g. NG-monomethyl-L-arginine/L-NMMA) and NO capturing substances such as hemoglobin, to block both the relaxant and hyperpolarizing effect of ACh(11–14). In addition, an endothelial-derived relaxing factor different from NO, a prostanoid, or EDHF is suggested to account for a significant part of ACh-induced relaxation in rat mesenteric arteries(15).

We wanted to investigate the relative importance of NO in mediating endothelium-dependent relaxation in the femoral artery from piglets in the first week of life. In an in vitro model of isolated femoral artery ring segments from newborn and 1-wk-old piglets, we studied responses to endothelium-dependent and -independent agonists and the effect of NOS inhibition on these responses.

METHODS

Tissue preparation and mechanical measurements. Male piglets newborn (10-22-h) and 7 d (7-10-d)-old were anesthesized with sodium thiopental 50 mg/kg i.v. and exsanguinated. The femoral arteries were immediately removed by careful dissection and kept at 4 °C in Krebs buffer of the following composition (mM): NaCl 118.9, NaHCO3{minus] 20.0, glucose 11.0, KCl 4.6, MgCl2 1.2, NaH2PO4 1.0, and CaCl2 1.5. The arteries were cut into cylindrical segments 2-4 mm in length and mounted in organ chambers for measuring isometric tension. Two L-shaped stainless steel holders (0.15 mm) were inserted into the segment's lumen; one holder was connected to a Grass FT03C force-displacement transducer and the other to a micrometer. The transducers were linked to an Analog Digital Instruments MacLab Analog Digital Converter through a MacLab Bridge Amplifier. The digitalized data were analyzed by an Apple Macintosh LC 475 8/160 computer.

The holders were immersed in organ baths containing 5 mL of buffer solution at a temperature of 37.5 ± 0.5 °C, continuously bubbled with 5% CO2 and 95% O2 to maintain a pH of 7.3-7.4. In some segments the endothelium was removed by rubbing the luminal surface with a stainless steel wire (0.5 mm) while rolling the vessel on thin tissue wetted with buffer solution. The removal of endothelium was considered successful if the segment relaxed less than 10% when exposed to ACh.

Immediately after mounting, the rings were allowed to rest at low tension(100-200 mg) for 15 min. The rings were stretched in 200-mg increments every 15 min to a resting tension of 500 mg and allowed to equilibrate for 45-90 min. The segments were subsequently exposed to a high potassium buffer solution (60 mM K+, prepared by isoosmolar substitution of NaCl with KCl) until a stable contraction was achieved (2-4 times). At least 30 min was allowed after washout before a new contraction. After another 30 min of equilibration, the experiments were started. In four animals, a complete concentration-response curve to KCl (10-85 mM) was established. These rings were not used in the relaxation experiments.

Experimental design. Two series of relaxation experiments were performed. In the first series, four rings from each animal (newborn:n = 6, 7-d-old: n = 8) were studied in parallel; one without endothelium, one control ring with intact endothelium, and two endothelium-intact rings, one incubated with indomethacin (10 μM) and one incubated with the NOS inhibitor L-NMMA (0.3 mM). All rings were precontracted with phenylephrine in concentrations needed to reach a contraction corresponding to approximately 80% of the 60 mM K+-induced contraction. After a stable contraction had been achieved, cumulative concentration-response studies to ACh (1 nM to 1 μM) were performed. The rings were then allowed to rest for at least 90 min after washout, before repeating the incubation and precontraction procedures, and performing cumulative concentration-response curves to SNP (1 nM to 10 μM).

In the second series, femoral arteries from six additional animals(n = 3 in each age group) were mounted as described above. After equilibration and exposure to repeated high potassium (60 mM) buffer solution, two rings were incubated with LY83583, a putative superoxide anion generator(16, 17), and two rings served as controls. After phenylephrine addition, cumulative concentration-response curves to ACh (1 nM to 1 μM) and SNP (1 nM to 10 μM) were obtained. In these experiments indomethacin (10 μM) was present in all baths. Before ending the experiments, rings were again exposed to 60 mM potassium-buffer solution.

Histology. Rings with and without intact endothelium were fixed in 10% buffered formalin solution, embedded in paraffin, cut into 3-μm sections, and stained with hematoxylin and eosin. The integrity of the internal elastic lamina and smooth muscle in all rings, and the presence or lack of an intact endothelium in unrubbed and rubbed segments, respectively, was verified by light microscopy.

Drugs. The following drugs were used: ACh chloride, L-phenylephrine hydrochloride, L-NMMA, L-arginine, SNP dihydrate, indomethacin, LY83583 (Sigma Chemical Co., St. Louis, MO). Indomethacin was dissolved in a few drops of NaOH and diluted in demineralized water to a stock solution of 10 mM, which was made fresh weekly. SNP was dissolved in water immediately before use. LY83583 was dissolved in demineralized water and stored at 4 °C. All other substances were prepared in demineralized water as stock solutions, frozen at -70 °C, thawed, and diluted further on the day of the experiment.

Calculations and analysis of results. Responses to ACh and SNP were calculated as percentage relaxation of phenylephrine-induced contraction. Results are expressed as means ± SEM. In all cases, n refers to the number of animals used in each experimental group.

A t test, paired and unpaired comparisons, was used to compare differences between means. Nonlinear regression analysis(18) was used to calculate pD2 values(negative log of the concentration of a drug that produces half its maximal response, -log EC50) with a 95% confidence interval from SNP concentration-response experiments. ACh induced constriction at higher concentrations and did, therefore, not produce saturable concentration-response curves. This made it impossible to calculate reliable pD2 values. In the case of ACh, analysis of variance for repeated measures with Greenhouse-Geisser correction was used instead to compare differences between groups. An unpaired t test with Bonferroni's correction was used to identify at which concentrations the changes occurred. A p value less than 0.05 was considered statistically significant.

RESULTS

Potassium-induced contraction. When normalized to maximal contraction, contractile responses to KCl (10-85 mM) were identical in the two age groups (Fig. 1). Responses to high potassium buffer solution (60 mM K+) remained stable throughout experiments, and no sign of fatigue or loss of ability to contract was observed (last contraction being 100.7 ± 1.01% of reference contraction).

Figure 1
figure 1

Potassium chloride-induced contraction. Cumulative concentration-response curves to potassium chloride in femoral artery rings from 0-d (10-22-h) and 7-d (7-10-d)-old piglets. Data are expressed as the percentage of change of maximal contraction in tension developed at each concentration. Data are expressed as means ± SEM (four arterial rings from two animals at each age).

Full size image

Effect of L-NMMA. With the rings passively stretched to approximately 500 mg, the addition of L-NMMA (0.3 mM) to the organ baths induced a small contraction in femoral artery rings from both newborn and 7-d-old piglets. Expressed as a percentage of the contraction induced by high potassium buffer solution, the increase in tension from baseline was larger in the newborn than in the 7-d-old piglets (9.0 ± 1.4% versus 5.5 ± 0.7%, respectively, p < 0.05). The concentration of phenylephrine needed to achieve approximately 80% of potassium-induced contraction was significantly smaller in rings incubated with L-NMMA compared with controls; 0.12 ± 0.03 μM versus 0.23 ± 0.03μM in newborn and 0.13 ± 0.03 μM versus 0.29 ± 0.07 μM in 7-d-old piglets, p < 0.05.

ACh-induced relaxation. Rings from both age groups relaxed when exposed to low concentrations of ACh (1 nM to 30 nM) in the presence of intact endothelium. The onset of relaxation occurred within 15-40 s and, in most cases, reached a steady state after approximately 60 s. After steady state was reached, some rings started to contract, whereas other rings remained in a relaxed state until washout, or a new dose was added. At higher Ach concentrations (0.1-1 μM) no further relaxation took place, and only contraction was seen. Segments with endothelium removed mechanically showed either no relaxation or an increase in tension in response to increasing concentrations of ACh (not shown).

Preincubation with L-NMMA (0.3 mM) for 10 min had no significant effect on relaxation achieved in response to 1 nM to 1 μM ACh in the newborn (Fig. 2A), although there was a tendency toward a rightward shift of the concentration response curve in the presence of NOS inhibition. In rings from 7-d-old pigs, L-NMMA gave a statistically significant inhibition of ACh-induced relaxation (Fig. 2B) (p < 0.001, repeated measures analysis of variance). The effect of 0.3 mM L-NMMA on ACh-induced relaxation was reversed by 1 mM L-arginine, but not by 1 mM D-arginine (n = 2,Fig. 3).

Figure 2
figure 2

(A and B) ACh-induced relaxation and inhibition with L-NMMA and indomethacin. Cumulative concentration-response curves to ACh (1 nM to 1 μM) in femoral artery rings from (A) 0-d(10-22-h) (n = 6) and (B) 7-d (7-10-d)-old (n = 8) piglets. Data are expressed as means ± SEM. *,p< 0.01 controls vs L-NMMA.

Full size image
Figure 3
figure 3

Reversal of NOS inhibition by L-arginine. Original tracings of experiment with femoral artery rings from 8-d-old piglet showing effect of L-arginine (1 mM) (top panel) and D-arginine (1 mM) (lower panel) on ACh-induced (10 nM) relaxation in phenylephrine-precontracted rings incubated with L-NMMA (0.3 mM).

Full size image

Incubation with the superoxide anion generator LY83583 (10 μM) gave an inhibition of ACh-induced relaxation very similar to what was seen with L-NMMA (Fig. 4). In the newborn, LY83583 had no effect on ACh-induced relaxation (Fig. 4A). In the 7-d-old piglet, incubation with LY83583 significantly attenuated the relaxant response to ACh (Fig. 4B) (p < 005, repeated measures analysis of variance). Incubation with indomethacin (10 μM) had no significant effect on ACh-induced relaxation in either age group (Fig. 2).

Figure 4
figure 4

(A and B) ACh-induced relaxation and inhibition with LY83583. Cumulative concentration-response curves to ACh(1 nM to 1 μM) in femoral artery rings from (A) 0-d (10-22-h) and(B) 7-d (7-10-d)-old piglets in the absence and presence of LY83583(n = 3 in all groups). Data are expressed as means ± SEM.*p < 0.05 controls vs LY83583.

Full size image

SNP-induced relaxation. Femoral artery rings from both newborn and 7-d-old piglets relaxed almost 100% to 10 μM SNP (Fig. 5). Sensitivity to SNP was significantly lower in newborn (Fig. 5A) than in the 7-d-old piglet (Fig. 5B), as expressed by a lower pD2, 5.97 (5.84-6.10) versus 6.67 (6.51-6.83).

Figure 5
figure 5

(A and B). SNP-induced relaxation. Cumulative concentration-response curves to SNP (10 nM to 1 μM) in femoral artery rings from (A) 0-d (10-22-h) and (B) 7-d(7-10-d)-old piglets in the absence and presence of L-NMMA (0.3 mM)(n = 5 in all groups). Data are expressed as means ± SEM.

Full size image

L-NMMA potentiated SNP-induced relaxation significantly both in the newborn and 7-d-old piglet. In the newborn (Fig. 5A),pD2 values were 5.97 (5.84-6.10) (controls) versus 7.07 (6.98-7.16) (L-NMMA). In the 7-d-old piglet (Fig. 5B), this potentiating effect of L-NMMA was greatly reduced;pD2 values being 6.67 (6.51-6.83) (controls) and. 7.03(6.88-7.18) (L-NMMA). In the presence of L-NMMA, the concentration response curves to SNP were almost identical. A similar potentiating effect on SNP-induced relaxation was seen in segments deliberately removed of endothelium (data not shown). The effect of both indomethacin and LY83583 treatment on SNP-induced relaxation was negligible in both age groups (data not shown).

DISCUSSION

The results from the present study indicate that the mechanisms underlying endothelium-dependent relaxation in piglet femoral artery rings undergo developmental changes between 0 and 7 d postnatally. In the presence of indomethacin, artery rings in the two age groups relaxed to a similar extent in response to ACh. In newborn piglets, a NO-independent mechanism seemed to account for the main relaxant effect of ACh, whereas in artery rings from 7-d-old piglets, NOS inhibition significantly reduced the relaxant response to ACh.

In porcine pulmonary artery rings Liu et al.(3) found a negligible relaxant response to ACh immediately after birth, increasing to a maximal response at 3-10 d followed by a decreasing response. This is consistent with the results of Zellers and Vanhoutte(4) with an increasing relaxant response to ACh up to 10 d after birth. Similar findings have been reported in pulmonary artery rings from newborn and juvenile sheep(2). These findings suggest an important role for EDRF in establishing a stable pulmonary circulation after birth. However, in opposition to these studies, ACh induced similar relaxations in femoral artery rings from newborn and 1-wk-old piglets in the present study.

The fact that incubation with LY83583 did not significantly reduce the relaxant effect of ACh in rings from newborn piglets supports the assumption that ACh-induced relaxation in this age group was mediated by a NO-independent mechanism. Several studies have shown that LY83583 has a“cGMP-lowering” effect(19, 20), most likely by the generation of a superoxide anion which inactivates the NO molecule(16, 17). Because indomethacin did not affect the ACh-induced relaxation in any age group, the NO-independent relaxing mechanism did not seem to involve a prostanoid.

The different characteristics of ACh-induced relaxation in newborn and 7-d-old piglets may be related to a developmental change in muscarinic receptor density or efficacy. It has been suggested that different subtypes of muscarinic receptors could be involved in the release of NO and EDHF(21). Interestingly, it has been demonstrated in rat cortex that a change in the relative amount of M1 and M2 receptors can take place during the first weeks postnatally(22).

Alternatively, our findings of a reduced activity of the NO-cGMP pathway in response to ACh in artery rings from newborn, compared with 7-d-old, piglets, could be attributed to a difference in basal NO release. NO can act as an inhibitor of NOS via a negative feed-back mechanism, as has been shown in cultured endothelial cells and in arterial rings in vitro(23). In case of a higher basal NO release in the newborn femoral artery, NOS would be down-regulated, and the importance of the NO/cGMP-axis in mediating ACh-induced relaxation would be minimized. This hypothesis is supported by our finding of a slightly greater effect of L-NMMA on basal tone in rings from newborn piglets. A higher basal release of NO has also been demonstrated in the porcine superior mesenteric artery of 3-d-old, compared with 35-d-old, animals(24).

The inability of L-NMMA to inhibit ACh-induced relaxation in the isolated femoral artery from newborn piglets raises the possibility that either NO is not important per se in mediating ACh-induced relaxation, or alternatively, that these arteries have an additional capacity of vasodilation in response to ACh. It has been suggested that an EDHF could serve as a“backup” when NOS is inhibited(10). Kilpatrick and Cocks(25) found endothelium-dependent relaxation in pig coronary artery to be mainly due to NO, but in case of inhibited NOS, hyperpolarization via an endothelium-derived factor different from NO could supplement this response with as much as 60-80%.

Arterial rings from newborn and 7-d-old piglets responded with an almost 100% relaxation to SNP. This is in accordance with the results of other authors who demonstrated fully functional mechanisms for cGMP-dependent relaxation in newborn pulmonary(1) and cerebral(26, 27) arteries from sheep. In both age groups a potentiating effect of NOS inhibition on the relaxant effect of SNP was observed, and with L-NMMA present, the concentration response curves to SNP were almost identical. The supersensitivity phenomenon after the removal of basal release of NO is well known(28, 29). The mechanism(s) behind this augmented response to SNP is unclear, but it is suggested to occur at the level of soluble guanylate cyclase(28, 30), with an up-regulation of guanylate cyclase when basal NO release is inhibited. We speculate that the difference in sensitivity between the two age groups and the different magnitude of the potentiating effect of L-NMMA on SNP-induced relaxation, could be related to a difference in basal NO release and the level of guanylate cyclase activity.

In conclusion, the newborn porcine femoral artery is fully capable of relaxation in response to ACh when cyclooxygenase activity is inhibited. We have demonstrated a maturational change in the effect of NOS inhibition, probably due to a change in the mechanisms mediating ACh-induced relaxation from birth to 7 d of age. Whether other endothelium-dependent factor(s) besides NO is responsible for the ACh-induced relaxant response in newborn also in a condition of noninhibited NOS, warrant further investigation.

The SNP results demonstrate that artery rings from newborn piglets are fully capable of relaxation in response to exogenously added NO. We speculate that a higher basal NO release in the newborn leads to a down-regulated guanylate cyclase. This could explain a decreased sensitivity to SNP and a greater potentiating effect of NOS inhibition on SNP-induced relaxation in artery rings from newborn piglets.