Abstract
Conventional peritoneal dialysis solution (PDS) relaxes visceral and parietal peritoneal arterioles (microvessels) by unclear mechanisms. The present study was originally designed to investigate the mechanisms of PDS-mediated vascular reactivity. Surprisingly, our preliminary data indicated that PDS induces contraction in large vessels such as the aorta. That result contrasts with the relaxation observed in the microvasculature. We therefore extended the study to (1) determine the effect of PDS on the superior mesenteric artery (SMA), (2) confirm the PDS-induced contraction in the aorta, and (3) determine if a prostanoid and nitric oxide are involved in the observed PDS-induced vessel response.
Rat SMA rings with intact endothelium and aortic rings with and without endothelium were prepared and placed in baths filled with a non vasoactive physiologic salt solution (PSS), or with PSS plus mefenamic acid (MFA, a cyclo-oxygenase inhibitor), or PSS plus NG-monomethyl-l-arginine (L-NMMA, an inhibitor of nitric oxide synthase) under a force transducer. We recorded changes in tension throughout the protocols. After equilibration, the baths were filled with a conventional glucose-based PDS (Delflex 2.5%: Fresenius Medical Care, Bad Homburg, Germany) with and without MFA or L-NMMA for 30 minutes. The rings were then washed, contracted with phenylephrine, and relaxed with acetylcholine to verify the presence or absence of endothelium.
In both SMA and aorta, PDS induced contraction. That contraction was suppressed by MFA [SMA: 0.57 g vs. 0.13 g (± 0.035 g); aorta: 0.88 g vs. 0.27 g (± 0.035 g); p < 0.05 by analysis of variance (ANOVA)]. Aortic contraction induced by PDS was not altered by L-NMMA.
Conventional PDS induces contraction in large vessels, in contrast to its action of relaxation in microvessels. Vascular reactivity in large vessels involves the production of a constrictor prostanoid in the vascular smooth muscle. Peritoneal dialysis solutions do not induce NO in aortic endothelium. Peritoneal dialysis solution–induced, prostanoid-mediated contraction of smooth muscle may contribute to a worsening of hypertension and the premature uterine contractions observed in the rare cohort of pregnant uremic patients on peritoneal dialysis.
Keywords: Peritoneal dialysis solution, vascular rings, vascular reactivity
Introduction
Conventional peritoneal dialysis solution (PDSs) dilates the visceral and parietal microvasculature by mechanisms that are possibly related to the hyperosmolality, low pH, and buffer anion system of the solution (1,2). Our recent intravital microscopy studies indicated that conventional PDS produces an instantaneous and sustained near-maximal vascular relaxation at all levels of the intestinal (visceral) microvasculature (3). That generalized vascular relaxation was independent of solution pH and similar in magnitude at all arteriolar levels (7 – 100 μm in diameter).
Recent intravital microscopy study (4) of rat mesenteric arteries (250 – 350 μm) has shown that conventional lactate-buffered PDS dilates mesenteric arteries at a magnitude similar to that observed in the cremaster muscle and the cecal and intestinal arterioles (1-3). However, no significant vascular reactivity was observed when mesenteric arteries were suffused with a pH-neutral, bicarbonate-buffered PDS with low glucose degradation products (4). Thus, the authors concluded that the vasoactive components of the conventional PDS are the glucose degradation products and, to a lesser extent, the lactate buffer anion system (4).
The present study was originally designed to investigate the molecular mechanisms responsible for the vascular relaxation induced by conventional PDS. The experiments were conducted with vascular rings from the aorta and superior mesenteric artery, studied in a standard tissue bath.
Vascular ring studies have the advantage of using multiple rings from each animal, allowing for the study of multiple inhibitors in a paired manner, and such studies are considered a “gold standard” for exploring the molecular mechanisms involved in vascular control.
The results of our current studies were completely surprising. In contrast to the data found in the literature on conventional PDS–mediated visceral and parietal microvascular relaxation, our data show that PDS produces contraction in large arteries by a vascular smooth muscle–derived constrictor mechanism.
Material and methods
General animal care
Male Sprague–Dawley rats (Harlan Industries, Indianapolis, IN, U.S.A.) were housed in facilities approved by the American Association for Accreditation of Laboratory Animal Care and were maintained on standard rat diet and water ad libitum for at least 1 week before use. All animal care and experimental procedures conformed to Principles of Laboratory Animal Care by the National Society for Medical Research and the Guide for the Care and Use of Laboratory Animals by the U.S. National Academy of Science as published by the National Institutes of Health (publication number 80-23, revised 1987) and were approved in advance by the Institutional Animal Care and Use Committee of the University of Louisville and the Louisville Veterans Administration. Experiments were performed on rats (215 – 235 g) that had been fasted overnight.
Tissue preparation
superior mesenteric artery rings
Each rat was anesthetized with an intraperitoneal injection of sodium pentobarbital (60 mg/kg). The superior mesenteric artery (SMA) was exposed and cleaned of adherent tissue. The straight segment of the SMA from its aortic origin to its first branch to the intestine—a portion approximately 5 – 7 mm long—was harvested. A mid-portion of the segment was cut to make a pair of 2-mm rings. Two stainless steel wires (each 0.3 mm in diameter) were passed through the lumen of each ring, and closed on themselves to form two wire triangles. One wire triangle was attached to a fixed glass hook and the other wire triangle was attached to a force transducer (Micro-Med, Louisville, KY, U.S.A.). The transducer was connected to a microprocessor-based tissue force analyzer (Micro-Med).
Each SMA ring with its two triangle attachments was suspended in a 17-mL tissue bath filled with a physiologic salt solution (PSS: 118 mmol/L NaCl, 4.7 mmol/L KCl, 2.5 mmol/L CaCl2, 1.2 mmol/L KH2PO4, 1.2 mmol/L MgSO4, 12.5 mmol/L NaHCO3, and 11.1 mmol/L glucose). The PSS was maintained at 37°C and was bubbled with a 95% O2/5% CO2 mixture via a frit tube to yield a pH of 7.4.
aortic rings
Immediately following SMA harvest, the thoracic aorta was excised from each rat and was cleaned of adherent extravascular tissue. The response of the aortic segments closer to the aortic arch and closer to the diaphragm to several agonists differs from the response of the middle segments (Unpublished data). We therefore used only the middle 8 mm of the thoracic aorta in the present experiments. One half of this thoracic aorta segment was denuded of endothelium by passing a fine glass rod about the size of the inner diameter of the aorta to and fro once through the lumen (endothelium denuded group). The other half of the thoracic aorta segment was not so treated (endothelium intact group). The presence or absence of viable endothelium in the aortic rings was tested by endothelium-dependent acetylcholine (ACH) relaxation.
The endothelium-intact and endothelium-denuded segments of the aorta were each cut to produce a pair of 2-mm rings. The aortic rings were mounted in a tissue bath in the same manner as the SMA rings were.
Experimental protocols
protocol 1: superior mesenteric artery rings
The SMA rings were stretched under an initial tension (preload) of 0.5 g in baths filled with PSS. Mefenamic acid (MFA) was added to the solution bathing one of each pair of rings to give final concentration of 40 μmol/L. The MFA remained in the solution from that point until the end of the protocol. The other ring of the pair acted as a “no inhibitor” control, but was otherwise treated identically to the ring with the added inhibitor.
The preload was adjusted every 10 minutes for 45 minutes for equilibration of preload and inhibitor. After 45 minutes of equilibration, the PSS or PSS+MFA was replaced with PDS or PDS+MFA. The changes in force level resulting from exposure to the PDS were recorded for 30 minutes. The rings were then washed with PSS three times over 25 minutes to bring the ring tension down to the level of the original preload.
Next, the rings were contracted with 1.0 μmol/L phenylephrine (PHE) for 15 minutes and relaxed with 3.0 μmol/L ACH for 10 minutes to demonstrate ACH-induced, endothelium-dependent relaxation as evidence for a viable endothelium.
protocol 2: aortic rings
The aortic rings were stretched to a preload of 1.0 g in baths filled with PSS. Either MFA or NG-monomethyl-l-arginine (L-NMMA) was added to the solution bathing one of the paired rings to give a final concentration of either 40 μmol/L (MFA) or 60 μmol/L (L-NMMA). The relevant inhibitor remained in the solution from that point until the end of the protocol. The other ring of the pair acted as a “no inhibitor” control, but was otherwise treated identically to the ring with the added inhibitor.
After 45 minutes of equilibration of the tissues to preload and inhibitor, the PSS and PSS+MFA or PSS+L-NMMA was replaced with PDS and PDS+MFA or PDS+L-NMMA. The changes in force level resulting from exposure to the PDS were recorded for 30 minutes. The rings were then washed with PSS, contracted with PHE, and relaxed with ACH to demonstrate ACH-induced relaxation as evidence for the presence (in endothelium-intact rings) or absence (in endothelium-denuded rings) of viable endothelium.
Experimental groups
A total of 12 SMA rings (6 pairs) and 36 aortic rings (18 pairs) were used. The 6 adjacent pairs of endothelium-intact SMA rings, the 12 adjacent pairs of endothelium-intact aortic rings, and the 6 adjacent pairs of endothelium-denuded aortic rings were divided into 4 groups, as follows:
6 pairs of endothelium-intact SMA rings were studied with and without MFA. [One ring from each pair was treated with MFA in the bath; the other was a “no inhibitor” control (protocol 1).]
6 pairs of endothelium-intact aortic rings were studied with and without MFA. [One ring from each pair was treated with MFA in the bath; the other was a “no inhibitor” control (protocol 2).]
6 pairs of endothelium-denuded aortic rings were studied with and without MFA. [One ring from each pair was treated with MFA in the bath; the other was a “no inhibitor” control (protocol 2).]
6 pairs of endothelium-intact aortic rings were studied with and without L-NMMA. [One ring from each pair was treated with L-NMMA in the bath; the other was a “no inhibitor” control (protocol 2).]
Protocol 1 was designed to test (A) the response of SMA to PDS, and (B) the effect of cyclo-oxygenase inhibitor (MFA) on the PDS-induced vascular response in SMA.
Protocol 2 was designed to test (A) the response of aorta to PDS, (B) the effect of endothelium on PDS-induced vessel contraction, (C) whether the effect of MFA on PDS-induced vessel contraction of aortic rings depended on endothelium, and (D) the effect of L-NMMA on PDS-induced vessel contraction in endothelium-intact aortic rings.
Drugs
We obtained PHE [phenylephrine hydrochloride: 1-(5-oxohexyl)-3,7-dimethylxanthine], ACH (acetylcholine hydrobromide), MFA, and L-NMMA from Sigma Chemical (St. Louis, MO, U.S.A.). All drugs were dissolved in distilled water except for MFA, which was dissolved in a bicarbonate buffer. The PDS (Delflex 2.5%: Fresenius Medical Care, Bad Homburg, Germany) was a conventional glucose-based solution.
Data analysis
The maximal force of contraction (Fmax in grams) with exposure to PDS, the Fmax with exposure to PHE, and the Fmax with exposure to ACH were determined for each ring from computer-stored, digitized raw data. The Fmax for PDS was used for statistical analysis. The Fmax for PHE (pre-contraction), and the maximal relaxation to ACH were used to calculate ACH relaxation as a percentage. Presence of viable endothelium was accepted at ACH-induced endothelium-dependent relaxation > 60%, and absence of viable endothelium was accepted at ACH-induced endothelium-dependent relaxation < 5%.
Statistics
All data are presented as mean ± standard error of the mean. To determine the effect of inhibitors (MFA or L-NMMA) and endothelium on PDS-induced contraction, the contraction force in the presence of PDS was compared using two-way ANOVA within each pair of adjacent rings. Statistical significance was accepted at p < 0.05.
Results
Endothelium denudation
In the endothelium-denuded aortic ring preparation, a fine glass rod (a hematocrit microtube) was used to remove the endothelium by gently passing the rod to and fro once through the lumen of the aorta. All endothelium-intact rings demonstrated > 60% relaxation (81% ± 3.19%) and all endothelium-denuded rings demonstrated < 5% relaxation (2.7% ± 1.66%) in the presence of ACH. All experiments were performed using the timeline and protocol depicted in Figure 1.
figure 1.
Vascular ring preparation and protocol. Aorta or superior mesenteric artery was harvested from animals under anesthesia. The artery was cut to make rings 2 mm in length. Endothelium was removed in some of the rings (denuded rings). Each ring was suspended in physiologic saline solution (PSS) between a transducer and a lower hook under a certain baseline tension level. Changes in vessel tension were recorded through the transducer. After equilibration, the PSS was changed to peritoneal dialysis solution (PDS) and observed for 30 minutes. Then, following washout of PDS by PSS, the presence or absence of functional endothelium was demonstrated by the addition of phenylephrine (PHE: α-1 adrenergic agonist) to the bath to induce contraction, followed by acetylcholine (ACH) to induce endothelium-dependent relaxation.
Effect of PDS on vessel tension
In SMA (Figure 2, left bar) and aorta (Figure 3, left bar) alike, PDS induced vessel contraction. Both vessel types started to contract immediately after the PSS in the bath was replaced with PDS. Contraction reached plateau in about 20 – 25 minutes. Upon washout with PSS, vessel contraction immediately subsided. The tension level returned to baseline preload level in about 15 – 20 minutes.
figure 2.
Effect of peritoneal dialysis solution (PDS) on superior mesenteric artery (SMA) rings with and without mefenamic acid (MFA: a cyclo-oxygenase inhibitor). In SMA rings with intact endothelium (ENDO), PDS induced contraction, which MFA suppressed by 75%. The line on each bar indicates the analysis of variance (ANOVA)–determined pooled standard error of the mean for statistical comparisons. The ANOVA-determined statistical significance was p < 0.05.
figure 3.
Effect of peritoneal dialysis solution (PDS) on endothelium-intact aortic rings, with and without mefenamic acid (MFA: a cyclo-oxygenase inhibitor). In aortic rings with intact endothelium (ENDO), PDS induced contraction. That contraction was suppressed by 70% by MFA. The line on each bar indicates the analysis of variance (ANOVA)–determined pooled standard error of the mean for statistical comparisons. The ANOVA-determined statistical significance was p < 0.05.
Effect of MFA and L-NMMA on PDS-induced vessel contraction
The cyclo-oxygenase inhibitor MFA suppressed PDS-induced vessel contraction by about 65% – 80% in SMA and aorta alike, and in both endothelium-intact and endothelium-denuded rings (Figures 2, 3, and 4).
figure 4.
Effect of peritoneal dialysis solution (PDS) on endothelium-denuded aortic rings, with and without mefenamic acid (MFA: a cyclo-oxygenase inhibitor). In endothelium-denuded aortic rings, PDS induced contraction. That contraction was suppressed by 65% by MFA. The line on each bar indicates the analysis of variance (ANOVA)–determined pooled standard error of the mean for statistical comparisons. The ANOVA-determined statistical significance was p < 0.05.
The L-NMMA did not alter PDS-induced vessel contraction (endothelium-intact aorta, Figure 5).
figure 5.
Effect of peritoneal dialysis solution (PDS) on endothelium-intact aortic rings, with and without NG-monomethyl-l-arginine (L-NMMA: a nitric oxide synthase inhibitor). In aortic rings with intact endothelium, PDS induced contraction. That contraction was not altered by L-NMMA. The line on each bar indicates the analysis of variance (ANOVA)–determined pooled standard error of the mean for statistical comparisons. NS = non-significant (p > 0.05) by ANOVA.
Effect of endothelium on PDS-induced vessel contraction
Endothelium denudation increased PDS-induced vessel contraction in aorta by 30% [0.88 g vs. 1.27 g (± 0.066), p < 0.05 by ANOVA). Endothelium denudation did not change the effect of MFA on PDS-induced vessel contraction.
Discussion
The results of the present study completely surprised us. In contrast to the consistent finding of a relaxing effect of conventional PDS on microcirculation, PDS predominantly constricted large arteries through the action of a vascular smooth muscle–derived constrictor prostanoid. A possible candidate for this prostanoid is thromboxane A2, a cyclo-oxygenase product that is the major component of the constrictor prostanoids and that often appears in smooth muscle contraction mechanisms. However, the mechanism by which PDS induces the production of this vascular smooth muscle–derived constrictor prostanoid remains unknown.
The unexpected vascular reactivity that we found was independent of vascular endothelium, given that the presence or absence of endothelium had no effect on the observed constriction response. Furthermore, MFA (a cyclo-oxygenase inhibitor) did not completely eliminate the PDS-mediated vascular contraction, indicating that another vascular smooth muscle–derived constrictor pathway remains active.
It is conceivable that hyperosmolality might have altered calcium mobilization in vascular smooth muscle. Previous studies by Wang and colleagues showed that vascular smooth muscle cells cultured in high concentrations of d-glucose (or its stereoisomer l-glucose) or in mannitol exhibit abnormal intracellular calcium mobilization, indicating that the observed phenomenon can be attributed to hyperosmolality and that it is not specific for a particular osmotically active solute (5).
The reasons for the disparity in the vascular response by arteries and microvessels to conventional PDS are not clear. It is well established that vascular endothelium modulates the tone of the underlying vascular smooth muscle through a balance of vasodilation mediators [NO, endothelium-derived hyperpolarizing factor (EDHF), and prostaglandin I2] and vasoconstrictors (endoperoxides, thromboxane A2, superoxide anions, and endothelin-1). Furthermore, depending on the vascular bed, vessels of similar size appear to use different mechanisms for endothelium-dependent regulation of vascular tone (6). Similarly, the relative contributions of agonist-stimulated NO and EDHF to endothelium-dependent relaxation appear to vary between the sexes (7), with arteriole size within the same vascular bed (8), and between arterioles from different vascular beds (6,9,10). Microvessels and large arteries are therefore likely to vary with regard to the mechanisms of local control of vascular tone as well as in their response to vaso-active agents.
The vascular ring technique used in the present study is a useful tool for dissecting the molecular mechanisms of vessel response, and the technique is typically used in ex vivo tissue-bath experiments. The technique does not account for neurohormonal effects on large vessels, which may override any local control of vascular tone. From this reasoning, the constriction of large arteries in response to conventional PDS may have an insignificant effect on peritoneal dialysis efficacy, but a greater effect on blood pressure and contractility of the smooth muscle of the uterus.
Given that NO and prostanoids are known to be involved in contraction–relaxation mechanisms in many vessels, including aorta and microvessels, we tested whether NO and prostanoids are involved in the PDS-induced vessel response. The fact that the observed vascular response in the present study is NO-independent and that it is exclusively accounted for by a vascular smooth muscle–derived mechanism indicates that basal production or flow-mediated release of NO from the vascular endothelia of large arteries is not a major component in the mechanism of local control of vascular tone in those vessels. That finding contrasts with many in vivo and ex vivo studies that show suppression of NO production results in substantial vasoconstriction in arteriolar segments and that almost one half of the basal smooth muscle contractile state (as judged by vasodilation) can be suppressed by increased shear forces within the physiologic range (11-13). It is well established that, in vascular beds such as the intestine, small arteries and large arterioles—rather than the terminal microvasculature—are the major contributors to intestinal vascular regulation and organ perfusion. Thus, relaxation of those small vessels in response to exposure to conventional PDS is detrimental in the number of perfused capillaries and in the modulation of the effective capillary surface area available for exchange during peritoneal dialysis.
Conclusions
Conventional PDS induces contraction in large vessels, which contrasts with the relaxation seen in microvessels. This PDS-induced large-vessel contraction involves production of a constrictor prostanoid in vascular smooth muscle. Peritoneal dialysis solutions do not induce NO in aortic endothelium. The PDS-induced, prostanoid-mediated contraction of smooth muscle may contribute to worsening of hypertension and to the premature uterine contraction observed in the rare cohort of pregnant uremic patients on peritoneal dialysis.
References
- 1.Miller FN, Nolph KD, Joshua IG, Wiegman DL, Harris PD, Andersen DB. Hyperosmolality, acetate, and lactate: dilatory factors during peritoneal dialysis. Kidney Int. 1981;20:397–402. doi: 10.1038/ki.1981.152. [DOI] [PubMed] [Google Scholar]
- 2.Miller FN, Joshua IG, Harris PD, Wiegman DL, Jauchem JR. Peritoneal dialysis solutions and the microcirculation. Contrib Nephrol. 1979;17:51–8. doi: 10.1159/000402980. [DOI] [PubMed] [Google Scholar]
- 3.Zakaria E, Spain DA, Harris PD, Garrison RN. Generalized dilation of the visceral microvasculature by peritoneal dialysis solutions. Perit Dial Int. 2002;22:593–601. [PubMed] [Google Scholar]
- 4.Mortier S, De Vriese AS, Van de Voorde J, Schaub TP, Passlick–Deetjen J, Lameire NH. Hemodynamic effects of peritoneal dialysis solutions on the rat peri-toneal membrane: role of acidity, buffer choice, glucose concentration, and glucose degradation products. J Am Soc Nephrol. 2002;13:480–9. doi: 10.1681/ASN.V132480. [DOI] [PubMed] [Google Scholar]
- 5.Wang R, Liu Y, Sauvé R, Anand–Srivastava MB. Hyperosmolality-induced abnormal patterns of calcium mobilization in smooth muscle cells from non-diabetic and diabetic rats. Mol Cell Biochem. 1998;183:79–85. doi: 10.1023/a:1006813223216. [DOI] [PubMed] [Google Scholar]
- 6.Clark SG, Fuchs LC. Role of nitric oxide and Ca++-dependent K+ channels in mediating heterogeneous microvascular responses to acetylcholine in different vascular beds. J Pharmacol Exp Ther. 1997;282:1473–9. [PubMed] [Google Scholar]
- 7.McCulloch AI, Randall MD. Sex differences in the relative contributions of nitric oxide and EDHF to agonist-stimulated endothelium-dependent relaxations in the rat isolated mesenteric arterial bed. Br J Pharmacol. 1998;123:1700–6. doi: 10.1038/sj.bjp.0701781. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Nishikawa Y, Stepp DW, Chilian WM. In vivo location and mechanism of EDHF-mediated vasodilation in canine coronary microcirculation. Am J Physiol. 1999;277:1252–9. doi: 10.1152/ajpheart.1999.277.3.H1252. [DOI] [PubMed] [Google Scholar]
- 9.Petersson J, Zygmunt PM, Hogestatt ED. Characterization of the potassium channels involved in EDHF-mediated relaxation in cerebral arteries. Br J Pharmacol. 1997;120:1344–50. doi: 10.1038/sj.bjp.0701032. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Tare M, Parkington HC, Coleman HA. EDHF, NO and a prostanoid: hyperpolarization-dependent and -independent relaxation in guinea-pig arteries. Br J Pharmacol. 2000;130:605–18. doi: 10.1038/sj.bjp.0703332. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Koller A, Kaley G. Endothelial regulation of wall shear stress and blood flow in skeletal muscle microcirculation. Am J Physiol. 1991;260:H862–8. doi: 10.1152/ajpheart.1991.260.3.H862. [DOI] [PubMed] [Google Scholar]
- 12.Koller A, Sun D, Huang A, Kaley G. Corelease of nitric oxide and prostaglandins mediates flow-dependent dilation of rat gracilis muscle arterioles. Am J Physiol. 1994;267:H326–32. doi: 10.1152/ajpheart.1994.267.1.H326. [DOI] [PubMed] [Google Scholar]
- 13.Bohlen HG, Nase GP. Dependence of intestinal arteriolar regulation on flow-mediated nitric oxide formation. Am J Physiol Heart Circ Physiol. 2000;279:H2249–58. doi: 10.1152/ajpheart.2000.279.5.H2249. [DOI] [PubMed] [Google Scholar]