Mechanisms underlying early development of pulmonary vascular obstructive disease in Down syndrome: An imbalance in biosynthesis of thromboxane A₂ and prostacyclin^†

INTRODUCTION

About one-half of Down syndrome (DS) patients have congenital heart defects, most of which are associated with a left-to-right shunt such as a complete atrioventricular septal defect, a ventricular septal defect, a patent ductus arteriosus, or an atrial septal defect. Whereas a large left-to-right shunt leads to increased pulmonary blood flow and, hence, pulmonary hypertension (PH) and eventually pulmonary vascular obstructive disease (PVOD), it has long been suggested that DS patients with a left-to-right shunt develop PVOD earlier in their clinical course than non-DS patients with a left-to-right shunt [Chi and Krovetz, 1975; Clapp et al., 1990; Suzuki et al., 2000]. Of particular importance, infants with DS have been suggested to be at risk of developing persistent pulmonary hypertension of the newborn even in the absence of congenital heart defects [Shah et al., 2004], implying a functional abnormality of the pulmonary vessels that is specific to DS.

Two vasoactive eicosanoids, thromboxane A₂ (TXA₂) and prostacyclin (PGI₂), have been postulated to play important roles in maintaining pulmonary vascular function. TXA₂ is synthesized mainly by activated platelets and acts as a vasoconstrictor, a promoter of platelet aggregation [Hamberg et al., 1975], and a stimulator of vascular smooth muscle cell proliferation [Sachinidis et al., 1995; Pakala et al., 1997]. PGI₂ is mainly synthesized by endothelial cells, including pulmonary arterial endothelial cells, and acts as a vasodilator, an inhibitor of platelet aggregation [Moncada et al., 1976], and an inhibitor of vascular smooth muscle cell proliferation [Sinzinger et al., 1987; Wharton et al., 2000]. Thus, TXA₂ and PGI₂ have opposing effects and the balance of their biosynthesis plays an important role in the modulation of vascular tone and platelet aggregation. Actually, the TXA₂:PGI₂ biosynthesis ratio is reportedly high in patients with primary (idiopathic) and secondary pulmonary hypertension, for example, collagen vascular diseases [Christman et al., 1992]. A few reports have investigated the biosynthesis of vasoactive eicosanoids among children with congenital heart disease, and an elevated TXA₂ level has been reported in such children [Adatia et al., 1993]. However, the TXA₂:PGI₂ biosynthesis ratio has never been investigated in children with DS.

TXA₂ and PGI₂ have very short half-lives in the blood circulation, and the biosynthesis of both TXA₂ and PGI₂ exhibits circadian variations. Therefore, only the stable breakdown products or metabolites of these eicosanoids, and not the eicosanoids themselves, can be measured accurately. Furthermore, stimulation of the endothelium and platelets during blood sampling can induce artifactual increases in the concentrations of TXA₂ and PGI₂. For these reasons, we measured the urinary excretion of 11-dehydro-TXB₂ and 2,3-dinor-6-keto-prostaglandin (PG) F_1α in 24-hr urine samples as indices of extrarenal TXA₂ and PGI₂ biosynthesis, respectively. The purpose of the present study was to investigate the possible existence of an imbalance in the biosynthesis of TXA₂ and PGI₂ in DS patients with congenital heart defect and a left-to-right shunt and to evaluate whether the biosynthesis of TXA₂ and PGI₂ may differ in DS patients and non-DS patients by measuring the urinary excretion of 11-dehydro-TXB₂ and 2,3-dinor-6-keto-PGF_1α in 24-hr urine samples.

MATERIALS AND METHODS

RESULTS

The clinical findings and eicosanoid excretion levels in the D-PH group and the ND-PH group are shown in Table I. The background characteristics (age and sex) of both groups were comparable. The pulmonary arteriolar resistance was higher in the D-PH group than in the ND-PH group (4.5 ± 1.6 vs. 3.5 ± 1.6 Wood units m², P = 0.04), whereas the mean pulmonary artery pressures were not significantly different (53.6 ± 12.5 in the D-PH vs. 44.3 ± 8.5 mmHg in the ND-PH group, P = 0.07).

The mean urinary 11-dehydro-TXB₂ excretion levels in the D-PH, D-C, ND-PH, and ND-C groups were 7,985 ± 4,378, 3,504 ± 1,400, 5,101 ± 2,687, and 1,663 ± 619 ng/g of creatinine, respectively (Fig. 1a). The urinary 11-dehydro-TXB₂ excretion level was higher both in the presence of DS (P < 0.0001) and in the presence of PH (P < 0.0001). No significant interaction between the presence of DS and the presence of PH was observed (P = 0.326), thus the presence of DS and PH affected urinary 11-dehydro-TXB₂ excretion levels in independent manners.

The mean urinary 2,3-dinor-6-keto-PGF_1α excretion levels in the D-PH, D-C, ND-PH, and ND-C groups were 1,371 ± 1,003, 876 ± 571, 2,466 ± 1,003, and 810 ± 294 ng/g of creatinine, respectively (Fig. 1b). A significant interaction between the presence of DS and the presence of PH (P = 0.018) was observed, and the effect of PH on the urinary 2,3-dinor-6-keto- PGF_1α excretion levels was less pronounced in the presence of DS. The urinary 2,3-dinor-6-keto-PGF_1α excretion level was lower in the presence of DS (P = 0.012) and higher in the presence of PH (P < 0.0001).

The urinary 11-dehydro-TXB₂:2,3-dinor-6-keto-PGF_1α excretion ratios in the D-PH, D-C, ND-PH, and ND-C groups were 7.69 ± 6.19, 4.71 ± 2.00, 2.10 ± 0.93, and 2.27 ± 1.00, respectively (Fig. 1c). The ratio of 11-dehydro-TXB₂ to 2,3-dinor-6-keto-PGF_1α was higher in the presence of DS (P < 0.001) and independent of the presence of PH (P = 0.297). The P-value of the interaction analysis between the presence of DS and the presence of PH was 0.141, implying that an interaction was present. In a subgroup analysis among infants with DS, patients with PH tended to have a higher ratio than infants without congenital heart disease, even though statistical significance was not achieved (P = 0.086).

DISCUSSION

The present study demonstrated that the urinary 11-dehydro-TXB₂ excretion levels and the 11-dehydro-TXB₂:2,3-dinor-6-keto-PGF_1α excretion ratio were higher in DS infants than in non-DS infants, irrespective of the presence or absence of PH. Of note is that among the non-DS infants, the urinary 11-dehydro-TXB₂:2,3-dinor-6-keto-PGF_1α excretion ratio was comparable between the ND-PH group and the ND-C group, whereas both the urinary 11-dehydro-TXB₂ level and the 2,3-dinor-6-keto-PGF_1α excretion level were elevated in association with PH. This seemingly “compensatory” elevation in the 2,3-dinor-6-keto-PGF_1α excretion level in non-DS infants with PH agrees with the experimental findings that TXA₂ receptor stimulation upregulates PGI₂ biosynthesis [Jeremy et al., 1985] and that shear stress induces the biosynthesis of both TXA₂ and PGI₂ [Doroudi et al., 2000]. In DS infants with PH, on the other hand, the urinary 2,3-dinor-6-keto-PGF_1α excretion level was elevated to a lesser degree than the marked elevation of the urinary 11-dehydro-TXB₂ excretion level; thus, the urinary 11-dehydro-TXB₂:2,3-dinor-6-keto-PGF_1α excretion ratio was also significantly elevated. This observation strongly indicates an imbalance in the biosynthesis of TXA₂ and PGI₂ that may participate in vasoconstriction and vascular smooth muscle cell proliferation in the lungs.

As for the possible mechanisms by which the presence of an extra copy of chromosome 21 leads to an imbalance in the biosynthesis of vasoactive eicosanoids, we suspect that the presence of an extra copy of the Cu,Zn-superoxide dismutase (SOD) gene on the long arm of chromosome 21 and the overexpression of its protein may account for an imbalance in TXA₂ and PGI₂ biosynthesis. Increased pulmonary blood flow is known to be associated with the increased biosynthesis of PGH₂ from arachidonic acid [Smith et al., 1996; Dubois et al., 1998], which then leads to a balanced increase in both TXA₂ and PGI₂ biosynthesis. However, in DS infants, both peroxide and peroxynitrate biosynthesis are likely to be up-regulated in the presence of increased SOD activity [De La Torre et al., 1996; Pastor et al., 1998; Pastore et al., 2003]. While peroxide promotes PGH₂ biosynthesis [Smith et al., 1996; Dubois et al., 1998], peroxynitrate inhibits the biosynthesis of PGI₂, but not TXA₂, from PGH₂ [Zou and Ullrich, 1996; Zou et al., 1997] consequently, an imbalance in TXA₂ and PGI₂ biosynthesis arises that eventually leads to vasoconstriction and vascular smooth muscle cell proliferation [Ullrich et al., 2001].

In a previous study examining adult patients with primary (idiopathic) PH and secondary PH arising from collagen vascular diseases, the urinary 11-dehydro-TXB₂ excretion level and the urinary 11-dehydro-TXB₂:2,3-dinor-6-keto-PGF_1α excretion ratio were both elevated [Christman et al., 1992]. In the present study examining pediatric patients, whereas the urinary 11-dehydro-TXB₂ excretion level was likewise elevated in non-DS infants with PH, the urinary 11-dehydro-TXB₂:2,3-dinor-6-keto-PGF_1α excretion ratio was comparable to that in the non-DS control group because of the “compensatory” elevation in the urinary 2,3-dinor-6-keto-PGF_1α excretion level. These findings may suggest that the autonomy of the pulmonary vascular endothelial cells is preserved in non-DS infants with PH, which is reversible by cardiac operation, in contrast to the situation in adult patients with irreversible PH.

A previous study by Adatia et al. 1993 on eicosanoid biosynthesis in infants with a left-to-right shunt reported that the ratio of vasoconstrictive eicosanoids to vasodilative eicosanoids was elevated in infants with a left-to-right shunt. In the present study, this ratio was not elevated in non-DS infants with a left-to-right shunt and PH. Although our data could not clarify the reason, two reasons may account for this discordance: first, the subjects in the report by Adatia et al. consisted of both DS infants and non-DS infants. Secondly, about one-half of the infants had PH in the previous study, whereas all the infants in our study had PH.

DS patients are widely known to be at risk for the early development of severe PVOD. Several mechanisms have been proposed, including hypoplasia of the lung (including the vascular bed) [Cooney and Thurlbeck, 1982], sleep apnea resulting in hypoxic vasoconstriction of the pulmonary arteries [Loughlin et al., 1981], and a few specific pathological changes in the pulmonary arteries, including thinning of the media and fibrous intimal proliferation [Yamaki et al., 1983]. Our data indicate that the abnormal biosynthesis of vasoactive eicosanoids in the pulmonary vascular bed may also contribute to the increased vulnerability to PVOD of infants with DS.

One limitation of this study is that the controls were not strictly matched for sex and age with the cases. Nonetheless, a previous study reported that the excretion rates of two of the eicosanoid metabolites that we measured, 11-dehydro-TXB₂ and 2,3-dinor-6-keto-PGF_1α, did not differ between male and female subjects and were not dependent on age after adjustment for body surface area [Leonhardt et al., 1992]. These findings support the reliability of our results, although a formal case-controlled study would be necessary to obtain more definitive evidence of differences in the biosynthesis of TXA₂ and PGI₂ in infants with DS and those without DS. In addition, each subject was evaluated only before surgery in the present study. The longitudinal follow-up of individual children would enable the impact of corrective surgery on the balance of TXA₂ and PGI₂ biosynthesis in DS patients to be elucidated.

In summary, the predominance of TXA₂ biosynthesis over PGI₂ may lead to vasoconstriction and vascular smooth muscle cell proliferation in DS infants, thereby contributing to the development of PVOD. Of note, such imbalance is observed even in DS infants without congenital heart defect, although its clinical influence is yet to be determined. The increased vulnerability of infants with DS to PVOD may be partly accounted for by this imbalance in the biosynthesis of TXA₂ and PGI₂.

Over the past 20 years, treatment options for patients with PH have evolved [Humbert et al., 2004], and guidelines for the treatment of PH are being established [Badesch et al., 2004; Galie et al., 2004; Badesch et al., 2007; McLaughlin et al., 2009]. However, the treatment algorithms recommended in guidelines do not yet provide individualized treatment based on an underlying specific condition, for example, Down syndrome. Our data suggest that PGI₂ derivatives, such as epoprostenol, may serve as an effective treatment option for PH in patients with DS.