Disparity in Osmolarity-Induced Vascular Reactivity
El Rasheid Zakaria*,,
C. Michelle Hunt*,
Na Li*,
Patrick D. Harris* and
R. Neal Garrison*,,
Departments of * Physiology and Biophysics and Surgery, University of Louisville, and Veterans Administration Medical Center, Louisville, Kentucky
Address correspondence to: Dr. El Rasheid Zakaria, Department of Physiology and Biophysics, Health Sciences Center A-1115, University of Louisville, Louisville, KY 40292. Phone: 502-287-5249; Fax: 502-894-6242; E-mail: erzaka01{at}louisville.edu
Received for publication September 13, 2004.
Accepted for publication June 24, 2005.
Conventional peritoneal dialysis solutions (PDS) are vasoactive.This study was conducted to identify vasoactive components ofPDS and to describe quantitatively such vasoactivity. Anesthetizednonheparinized rats were monitored continuously for hemodynamicswhile the microvasculature of the jejunum was studied with invivo intravital microscopy. In separate experiments, vascularreactivity of rat endothelium-intact and -denuded aortic rings(2 mm) was studied ex vivo in a standard tissue bath. In bothstudies, suffusion of the vessels was performed with filter-sterilizedisotonic and hypertonic solutions that contained glucose ormannitol as osmotic agents. PDS served as a control (Delflex2.25%). Hypertonic glucose and mannitol solutions produced asignificant vascular reactivity in aortic rings and instantaneousand sustained vascular relaxation at all levels of the intestinalmicrovasculature. Similarly, lactate that was dissolved in alow-pH isotonic physiologic salt solution produced significantforce generation in aortic rings. Whereas isotonic glucose andmannitol solutions had no vasoactivity in aortic rings, isotonicglucose produced a selective, insidious, and time-dependentvasodilation in the intestinal premucosal arterioles (18 ±0.2% of baseline), which was not observed in the larger inflowarterioles (100 µm). This isotonic glucose–mediatedvascular relaxation can be attenuated by approximately 50% withcombined adenosine A2a and A2b receptor antagonists and completelyabolished by adenosine A1 receptor inhibition. By using twodifferent experimental techniques, this study demonstrates thathyperosmolality and lactate are the major vasoactive componentsof clinical peritoneal dialysis solutions. The pattern and themagnitude of such reactivity are dependent on vessel size andon the solutes’ metabolic activity. Low pH of conventionalPDS is not a vasoactive component by itself but renders lactatevasoactive. Energy-dependent transport of glucose into cellsmediates vasodilation of small visceral arterioles by an adenosinereceptor–mediated mechanism and constitutes a significantfraction of PDS-mediated vascular reactivity in the visceralmicrovasculature.
Conventional peritoneal dialysis solutions (PDS) dilate visceraland parietal microvasculature by mechanisms possibly relatedto hyperosmolality, low pH, and the buffer anion system of thesesolutions (1,2). Studies of hyperosmolar sodium solutions perfusedinto the intestinal lymph produced vasodilation of submucosalarterioles through a mechanism partially mediated by a hyperosmolality-inducednitric oxide (NO) release (3). Similarly, intravenous infusionof hypertonic galactose or mannitol solutions in pigs increasedthe baseline hepatic blood flow by 37%, presumably by a mechanismattributed to an osmotic stress (4). Massett et al. (5) foundthat in vitro infusion of isolated, cannulated, and pressurizedskeletal muscle arterioles with hypertonic solutions of glucose,sucrose, or mannitol at an osmolality of 330 mOsmol/kgH2O equallydilates these arterioles. This vascular reactivity seems tobe an endothelium-dependent response, which is independent ofthe NO or the cyclooxygenase pathways, but can be nearly abolishedby glibenclamide, an ATP-sensitive potassium channel inhibitor.Similar results were obtained with coronary arterioles perfusedex vivo with either hypertonic glucose or sucrose solutions.In these vessels, only glibenclamide caused attenuation of thehypertonic solution–mediated vascular relaxation. In addition,inhibition of inward rectifier potassium channels or calciumactivated potassium channels had no effect on hyperosmolality-inducedvascular reactivity (6,7). Our recent intravital videomicroscopyof the terminal ileum in rats showed that exposure of the ileumto a conventional PDS produces an instantaneous and sustainedvascular relaxation at all levels of the microvasculature, whichis essentially associated with doubling of blood flow in theinflow arterioles (100 µm in diameter) (8). In the ratmesentery, conventional lactate-buffered PDS preferentiallydilate mesenteric arteries by >20%, passively double mesentericarteriolar blood flow without change in diameter, and increasethe number of perfused mesenteric capillaries by >20%. Thesevasoactive properties were independent of solutions’ pHbut occurred only transiently when the mesentery was exposedto conventional PDS with low-glucose degradation products. Furthermore,bicarbonate-buffered PDS with low-glucose degradation productswere entirely nonvasoactive in the rat mesentery. On the basisof these findings, the authors suggested that the vasoactiveeffects of clinical PDS is due exclusively to their contentsof glucose degradation products and lactate contents and notto hyperosmolality (9).
In a recent study, we attempted to investigate the molecularmechanism of PDS-induced vascular relaxation. In this study,vascular rings from the aorta and superior mesenteric arterieswere exposed to a conventional PDS under controlled conditions.The PDS contracted these arteries by an endothelium-independentmechanism that involved a vascular smooth muscle–derivedprostanoid pathway (10). We therefore hypothesized that thepattern and the magnitude of vascular reactivity produced withconventional peritoneal dialysis and other hypertonic solutionsare determined primarily by the hyperosmolality and the solutes’metabolic activity. We further hypothesized that the role ofother components of the dialysis solution is subordinate. Theaim of this study was to identify the major vasoactive componentsof conventional PDS under controlled experimental conditions.This is required before investigation of the molecular mechanismsand signal transduction pathways involved in such vasoactivity.
General Animal Care and Surgery
Male Sprague-Dawley rats (Harlan, Inc., Indianapolis, IN) werehoused in Association for Assessment and Accreditation of LaboratoryAnimal Care–approved facilities and were maintained onstandard rat diet and water ad libitum for at least 1 wk beforeuse. All animal care and experimental procedures conformed to"Principles of Laboratory Animal Care" of the National Societyfor Medical Research and the "Guide for the Care and Use ofLaboratory Animals" of the US National Academy of Science aspublished by the National Institutes of Health (NIH publication80-23, revised 1987) and were previously approved by the InstitutionalAnimal Care and Use Committee of the University of Louisvilleand the Louisville Veterans Administration. Experiments wereperformed on rats (200 to 210 g) that had fasted overnight.Anesthesia was induced with intraperitoneal pentobarbital (60mg/kg) and maintained with supplemental subcutaneous injectionas needed. Body temperature was maintained at 37 ± 0.5°Cwith a rectal probe and a servocontrolled heating pad. Surgerywas carried out after loss of the blink and withdrawal reflexes.Tracheostomy was performed to reduce airway resistance, andthe animal was allowed to breathe room air. The right femoralartery was cannulated with a PE-50 catheter to provide continuousmonitoring and online recording of arterial BP.
Bathing Solutions Microvascular Studies.
All chemicals were purchased from Sigma Chemical Co. (St. Louis,MO). The intestinal segment was suffused continuously duringtissue preparation and equilibration with a nonvasoactive modifiedKrebs’ solution that contained 6.92 g/L sodium chloride,0.44 g/L potassium chloride, 0.37 g/L calcium chloride, and2.1 g/L sodium bicarbonate at a pH of 7.4 and osmolality of285 mOsm/L. Isotonic glucose and mannitol solutions were preparedas Tris-buffered physiologic salt solution (PSS). The isotonicglucose solution contained 36.29 mM Tris-HCl, 13.71 mM Tris-base,11.1 mM glucose, 88.98 mM NaCl, 5.87 mM KCl, and 2.55 CaCl2/2H2O.The isotonic mannitol solution had the same components exceptthat mannitol was substituted for glucose. Hypertonic mannitolsolutions were prepared either as a 5% mannitol in Krebs’solution or as a 10% mannitol in deionized water. All solutionswere filter-sterilized and prewarmed to 37°C before use.A conventional 2.25% dextrose-based dialysis solution (Delflex;Fresenius USA, Inc., Ogden, UT) that contained 5.67 g/L sodiumchloride, 3.92 g/L sodium lactate, 0.257 g/L calcium chloride,and 0.152 g/L magnesium chloride at a pH of 5.5 and an initialosmolality of 398 mOsm/L served as control for the hypertonicsolutions.
Macrovascular Studies.
Vascular ring studies are typically conducted in a nonvasoactivePSS. PSS composition in millimolars was 118 NaCl, 4.7 KCl, 2.5CaCl2, 1.2 KHPO3, 1.2 MgSO4, and 11.1 glucose. All aortic ringswere maintained at 37°C and bubbled with a 95% O2 and 5%CO2 gas mixture to yield a pH of 7.4 (gas bubble dispersionsurface; Radnotti Glass Technology, Monrovia, CA). Propertiesof the test solutions are illustrated in Table 1. A conventional2.25% dextrose-based dialysis solution (Delflex) served as control.
Drugs
Phenylephrine hydrochloride [PHE; 1-(5-oxohexyl)-3,7-dimethylxanthine],acetylcholine hydrobromide (ACh), isoproterenol, and sodiumnitroprusside were purchased from Sigma Chemical. PHE and AChwere used to test the integrity of the vascular endotheliumfor each aortic ring at the start and at the end of each experimentaccording to protocol. Isoproterenol and sodium nitroprusside,respectively, were used in the tissue bath to retard peristalsisand to assess the maximum dilation capacity of the intestinalmicrovasculature. All endothelium-intact rings demonstrated>60% of ACh-induced relaxation (81% ± 3.19), and allendothelium-denuded rings demonstrated <5% of the ACh-inducedrelaxation (2.7% ± 1.66). Three selective adenosine receptorantagonists were used: 8-Cyclopentyl-1,3-dipropylxanthine, anadenosine A1 receptor antagonist; 8-(3-chlorostyryl) caffeine,an adenosine A2a receptor antagonist; and alloxazine, an adenosineA2b receptor antagonist. The final concentration in the tissuebath for 8-cyclopentyl-1,3-dipropylxanthine, 8-(3-chlorostyryl)caffeine, and alloxazine were 200 nM, 200 nM, and 600 µM,respectively. These concentrations represent at least threetimes the 50% effective inhibitory concentration for each adenosinereceptor antagonist as determined by information provided bythe manufacturer.
Study I: Microvascular Reactivity Experimental Procedure.
The peritoneal cavity was exposed through a midline abdominalincision of 1.5 cm, and a 2- to 3-cm segment of jejunum waswithdrawn gently from the peritoneal cavity with its neurovascularsupply intact. The segment was opened along the antimesentericborder by electrocautery. The enteric contents and mucus wereremoved gently from the mucosal surface. The animals were positionedon a specially designed polyurethane board. The opened jejunumwas suspended, serosal side up, over a viewing port in a tissuebath with 4-0 silk sutures. The nonvasoactive bathing solutionwas maintained at 37°C and bubbled with nitrogen and carbondioxide to maintain the pH at 7.4. Isoproterenol was added tothe bathing solution in a very dilute concentration (0.01 µg/ml)to retard peristalsis. This dose of isoproterenol is below thethreshold that alters vascular smooth muscle tone (11).
The animal board was positioned on the stage of a trinocularmicroscope for direct in vivo intravital microscopy. Microvascularimages were transmitted through the microscope to a photodiodearray in an optical Doppler velocimeter (Microcirculation ResearchInstitute, Texas A & M University, College Station, TX)to measure center-line red blood cell velocity for the calculationof blood flow in the intestinal A1 inflow arteriole. The microvascularimage then was transmitted to a digital camera (Hitachi Denshi,Models K-P D51/D50), which provided 30 images per second toa computer. The digitized microvascular images were stored asstreamline video in the computer hard drive for later measurementof microvascular diameters with calipers.
Criteria for an acceptable microvascular preparation duringintravital microscopy included a baseline mean arterial pressure>90 mmHg, a center-line red blood cell velocity in a first-orderarteriole >20 mm/s, and an active vasomotion in the intestinalpremucosal A3 arterioles. We used a standard nomenclature forintestinal microvessels, as originally described by Bohlen andGore (11). Briefly, first-order arterioles (A1) arise from amesenteric arcade artery to traverse the mesenteric border ofthe bowel wall and then penetrate through the muscle layersto the submucosal layer. In the submucosal layer, second-orderarterioles (A2) arise from the A1 to run along the longitudinalaxis of the bowel. First- and second-order venules parallelthe A1 and A2. A2 give rise to branching second-order arcadevessels as well as to smaller third-order arterioles (A3). TheA3 vessels branch at a right angle from A2 to form distal A3(dA3), which terminates in the mucosa as a central villus arteriole.Along their course, A3 also give rise to smaller proximal A3that supply the seromuscular layers of the bowel wall.
Experimental Protocol and Measurements.
The intestinal segment was allowed to equilibrate for 40 minin the tissue bath. During this time, the segment was suffusedcontinuously with the nonvasoactive Krebs solution. BP, heartrate, rectal and bath temperatures, and bath pH were monitoredcontinuously (Digi-Med Signal Analyzers, Louisville, KY) andrecorded every 5 min. Microvascular data consisted of A1, proximalA3 (pA3), and dA3 arteriolar diameters and center-line red cellvelocity in the inflow A1. Baseline measurements were consideredvalid when the variability in the measurement was <0.5%.After baseline measurements, the nonvasoactive Krebs’solution was aspirated from the tissue bath and a test solution(see bathing solutions) was randomly added into the tissue bath.Microvascular data points were measured initially at 2 min afterthe addition of the test solution and then at 10-min intervalsduring the subsequent 90 min. At the conclusion of the experiment,one dose of ACh (10–4 M) was administered topically inthe tissue bath and microvascular data time points were takenat 1-min intervals over 10 min to assess endothelial cell functionand endothelial-dependent vasodilation. Finally, a single doseof sodium nitroprusside (10–4 M) was administered in thetissue bath to assess the endothelium-independent maximal dilationcapacity.
Study II: Macrovascular Reactivity
The experimental and timeline protocols of these studies aredepicted in Figure 1. Briefly, after induction of anesthesia,the thoracic aorta was excised and submerged in a Petri dishfilled with a PSS. One half of the thoracic aorta was denudedof endothelium by passing a fine-glass rod, approximately thesize of the inner diameter of the aorta, to and fro once throughthe lumen. The other half of the thoracic aorta was regardedas endothelium-intact aorta. The presence/absence of viableendothelium in these rings was verified with an endothelium-dependentACh relaxation according to protocol. The aortic segments closerto the aortic arch and ones closer to the diaphragm behave differentlyto several agonists than the middle segments (unpublished data).Therefore, only the middle 8 mm of the thoracic aorta was usedin these experiments.
Figure 1. Vascular ring preparation and protocol. The aorta was harvested under anesthesia and divided to make two pairs of rings of 2-mm length each. In one pair, the endothelium was removed (denuded rings). Each ring was suspended in physiologic saline solution (PSS) between a transducer and a lower hook under a certain baseline tension level. The change in vessel tension was recorded with a transducer. After equilibration, the PSS was exchanged for the test solution, and the change in tension was recorded over 30 min. A washout of the test solution was performed, and phenylephrine (PHE; -1 adrenergic agonist) was added in the bath to induce contraction, followed by acetylcholine (ACh) to induce endothelium-dependent relaxation to demonstrate presence/absence of functional endothelium.
The endothelium-intact and endothelium-denuded segments of theaorta were divided to produce two 2-mm rings each. Two wires(stainless steel of 0.012 inner diameter) were passed throughthe lumen of each ring and closed on them to form two wire triangles.One triangle was attached to a fixed hook, and the other trianglewas attached via a stainless wire to a force transducer, whichwas connected to a tissue force analyzer. Each vascular ringpreparation was suspended in an individual 20-ml tissue bath,which was filled with PSS. Each vascular ring was stretchedto produce an initial passive tension called "preload" of 1.0g in a bath filled with PSS. Each ring was treated with 1.0µM ACh and 1.0 µM phenylephrine to saturate "nonreceptorbinding sites" for the agonists and then washed with PSS for20 min. After another 40 min of vascular ring equilibration,the preload of each vascular ring was readjusted to the initial1.0-g preload level. Individual rings were contracted in a sequentialmanner with six cumulative phenylephrine doses to give phenylephrinebath concentrations of 0.01 through 3.0 µM (in three steps).Each ring then was relaxed with 3.0 µM ACh for 10 minto demonstrate ACh-induced relaxation and viability of the endothelialcells. Hyperosmolar solutions (see bathing solution) were addedrandomly in the tissue bath to replace the PSS, and the resultantvascular tension was recorded. Both endothelium-intact and -denudedvascular rings were studied in a paired manner design.
Statistical Analyses
All data are presented as mean ± SEM unless stated otherwise.Percentage change of the vessel diameter from baseline was assessedwith one-way ANOVA and Dunnett multiple-range test to evaluatechanges from the baseline within the same animal. Two-way ANOVAwas used to assess the relationship between vascular reactivityand arteriolar type. The maximal force of contraction to theperitoneal dialysis and other hyperosmolar test solutions, themaximal force of contraction to PHE, and the maximal relaxationto ACh was determined for each ring from computer-stored digitizedraw data. Differences between groups were assessed with two-wayANOVA and Bonferroni posttest. ACh-induced endothelium-dependentrelaxation of >60% or <5% was used to determine presence/absence,respectively, of a viable vascular endothelium. A result wasconsidered to be significant when the probability of a type1 error was P < 0.05.
Microvascular Reactivity Effect of Solute’s Metabolic Activity.
d-Glucose, unlike mannitol, is a metabolically active solutethat can be transported readily and actively into cells. Asshown in Figure 2A, isotonic d-glucose causes a differentialreactivity in the intestinal microcirculation (n = 12). Thisreactivity is characterized by dilation of the smaller premucosalA3 (8 to 15 µm, pA3, dA3, respectively). This is in contrastto the largely transient initial constriction observed in thelarger inflow A1 (100 µm). The maximum vascular responseduring a 90-min exposure of the intestine to the isotonic glucosesolution, expressed as percentage from baseline, was A1 (–10.58± 1.06% recorded at 10 min), pA3 (17.17 ± 1.66%recorded at 80 min), and dA3 (19.09 ± 2.41% at 80 min).There was no significant vascular reactivity when the intestinewas exposed to the isotonic mannitol solution (P > 0.05;n = 12). The averaged 90-min vascular reactivity to mannitolwas –0.13 ± 1.26%, –2.90 ± 1.18%,and –4.45 ± 1.07% from baseline diameter in theA1, pA3, and dA3, respectively. Isotonic glucose-mediated premucosalintestinal arteriolar dilation is an insidious and time-dependentresponse (Figure 2B, top). This vascular reactivity was partiallyattenuated by –50% in pA3 and –45% in dA3, withcombined adenosine A2a and A2b receptor antagonists (P <0.05; n = 12; Figure 2B, middle), and completely abolished whenthe adenosine A1 receptor was inhibited (P < 0.01; n = 12;Figure 2B, bottom). There was no effect of the adenosine receptorsubtype inhibition on the isotonic glucose-elicited selectiveconstriction of the A1 inflow arterioles.
Figure 2. (A) Effect of solute’s metabolic activity on intestinal microvascular reactivity. Isotonic glucose but not mannitol produced a differential vascular reactivity in intestinal microcirculation characterized by a significant selective vasodilation of the smaller premucosal arterioles (8 to 15 µm) and a vasoconstriction of the larger inflow A1 arterioles (100 µm). *P < 0.01 by ANOVA and Bonferroni post test versus isotonic mannitol; P < 0.01 by ANOVA and Bonferroni post test versus A3 premucosal arterioles. (B) Mechanism of isotonic glucose–induced microvascular reactivity. BL, baseline arteriolar diameter; A1, intestinal inflow arteriole; pA3 and dA3, intestinal proximal (p) and distal (d) A3 premucosal arterioles. *P < 0.01 by ANOVA and Bonferroni post test versus BL. P < 0.05 by ANOVA and Bonferroni post test versus BL.
Effects of Osmolarity Perturbation.
The effect of osmolarity perturbation on the intestinal microvasculatureis depicted in Figure 3. Osmolarity was enhanced with eitherthe addition of a 5% mannitol to a nonvasoactive Krebs’solution to obtain a final osmolality of 560 mOsmol/L (n = 6)or a 10% mannitol to deionized water to obtain a final osmolalityof 575 mOsmol/L (n = 6). A conventional 2.5% dextrose-basedPDS (Delflex) with osmolality of 398 mOsmol/L served as control(n = 6). Enhancement of osmolarity caused a generalized vasodilationat all levels of the intestinal microvasculature. Although theosmolarity of the hypertonic mannitol solutions was significantlyhigher than that of the dialysis solution, the dialysis solutioninduced a significantly greater vasodilation in the premucosalarterioles (36.85 ± 3.84 and 44.38 ± 6.63%), comparedwith the 5% mannitol (30.6 ± 4.7 and 22.4 ± 3.5%)and the 10% mannitol (33.7 ± 3.1 and 27.5 ± 2.7%)in the premucosal pA3 and dA3, respectively. In addition, allhyperosmolar solutions tested in this series equally doubledthe blood flow of intestinal inflow A1.
Figure 3. Effect of mannitol-enhanced osmolality on intestinal microvascular reactivity. PDS, conventional peritoneal dialysis solution; Mn, mannitol, 5% in Krebs or 10% in deionized water. *P < 0.05 by ANOVA and Bonferroni post test versus PDS-induced vascular reactivity.
Macrovascular Reactivity Effects of Osmolarity Perturbation.
Seven solutions were tested in this series (n = 12 each; Table 1).Of these solutions, only the solutions made hypertonic witheither d-glucose or d-mannitol produced a significant vascularreactivity in the aorta as measured by force generation (Figure 4).The magnitude of this aortic vascular reactivity was largestwith the conventional PDS and quantitatively higher in hyperosmolarsolutions that contained ions, compared with identical hyperosmolarsolutions that lacked ionic contents (Figure 4), suggestinga significant role of the ionic contents of the solution indetermining the magnitude of the prevailing vascular response.There was no vascular effect of isotonic glucose or mannitolsolutions on the aorta (data not shown).
Figure 4. Box and whisker plot of altered osmolality-induced macrovascular reactivity in endothelium (En) intact (+) and denuded (–) aortic rings. Glu, glucose. *P < 0.05 versus other hyperosmotic solutions by ANOVA and Bonferroni post test.
Effects of Vascular Endothelium.
In all of the hypertonic solutions tested in our study, endotheliumremoval significantly attenuated the magnitude of the solution-mediatedaortic contraction by –32% (range –20 to –54%;P < 0.05). Such attenuation was maximally seen in endothelium-denudedaortic rings that were exposed to the 5% glucose (–37%)and the 5% mannitol (–54%) in PSS (Figure 5).
Figure 5. Effect of vascular endothelium (En) on hyperosmolality-induced macrovascular reactivity. *P < 0.05 versus endothelium-intact aortic rings by ANOVA and Bonferroni post test.
Effects of Solution pH.
The role of pH in hyperosmolality-induced aortic reactivityseems to be influenced by the aortic endothelium and the ioniccomposition of the solution (Figure 6). Of all of the hyperosmolarsolutions tested in this series, the dialysis solution producedthe highest aortic ring contraction, regardless of the endothelium(Figure 6). In the aortic denuded rings, the conventional PDSproduced a –23% less contraction force than rings withintact endothelium. This pattern was also preserved in denudedaortic rings suffused with 5% mannitol (–19%), 5% glucose(–33%), 10% mannitol (–27%), and 10% glucose (–15%),when the pH of these hyperosmolar solutions was adjusted to<6 pH units (n = 7 for each solution). Low pH accounted for4.5% (P < 0.05) of the total variation in the vascular reactivityproduced by the 5% glucose or mannitol, which contained otherions (Figure 6, top). In comparison, in the 10% osmotic solute(glucose or mannitol) in deionized water (Figure 6, bottom),low pH accounted for only 0.3% (P > 0.5) of the total variation,which is accounted for exclusively by the vascular reactivity(15.3%; P < 0.001). In the presence of ions, low pH seemsto attenuate vascular reactivity by 30% in denuded rings andby 10% in endothelium-intact rings regardless of the main osmoticsolute (Figure 6, top). In contrast, in low pH deionized solutions,hyperosmolality as a result of glucose enhanced aortic reactivityin endothelium intact rings by 16% and attenuated that of endothelium-denudedrings by 14%, whereas in hyperosmolality as a result of mannitol,aortic reactivity in endothelium-intact rings was enhanced by8% and attenuated by 37% in endothelium-denuded rings (Figure 6,bottom).
Figure 6. Effect of pH perturbation on hyperosmolality-induced macrovascular reactivity. *P < 0.01 versus endothelium-intact aortic rings; P < 0.05 versus controlled pH by ANOVA and Bonferroni post test.
Effects of the Buffer Anion.
In this series, sodium lactate was dissolved in an isotonicPSS to match the lactate concentration of PDS. Addition of sodiumlactate (0.392 g/L) to the nonvasoactive PSS yielded a pH >8pH units. At this high pH, which was maintained in the tissuebath during the experiment, the maximum contraction force inaortic rings with intact endothelium was 0.01 ± 0.02g (n = 12; P > 0.1) versus –0.02 ± 0.02 g (n= 12; P > 0.1) in endothelium-denuded rings. When tissuebath pH was adjusted to 5.03, there was a significant contractionforce in both endothelium-intact aortic rings (0.33 ±0.11 g; n = 12; P < 0.05) and endothelium-denuded aorticrings (0.45 ± 0.16 g; n = 12; P < 0.01). This pH-dependentlactate-induced aortic contraction was significantly greaterin the endothelium-denuded rings (P < 0.05; Figure 7).
Figure 7. Lactate-induced macrovascular reactivity. En (+), endothelium intact; En (–), endothelium denuded. *P < 0.01 versus high pH, P < 0.01 versus endothelium-intact aortic rings by ANOVA and Bonferroni post test.
The salient findings of these studies are that (1) conventionalPDS produce an instantaneous and sustained vasodilation at alllevels of the intestinal (visceral) microvasculature; (2) thepattern and magnitude of such dilation is dependent on vesselsize and on the osmotic solute’s metabolic activity; (3)energy-dependent transport of glucose into cells mediates aninsidious vasodilation preferentially on small visceral arteriolesby an adenosine receptor–mediated mechanism, which constitutesa significant fraction of a glucose-based PDS-mediated reactivityin the visceral microvasculature; (4) hyperosmolality is themajor vasoactive component of the conventional PDS, whereasthe lactate buffer anion system of this solution is vasoactiveonly at low pH; and (5) low pH and other ionic contents of conventionalPDS modifies the magnitude of vascular reactivity instigatedby these solutions.
Technique
Our studies were performed with a standard tissue bath procedureunder well-controlled experimental conditions. The tissue underinvestigation, a small segment of the intestine (2 to 4 cm)or aortic rings (2 mm), was positioned in a relatively largetissue bath in which temperature, pH, Po2, Pco2, and osmolalitywere monitored and controlled while we simultaneously made directobservations of the intestinal microcirculation or continuouslyrecord the change in force of individual aortic rings. Duringthe equilibration period, the small intestinal segment and theaortic rings were suffused continuously with a nonvasoactivePSS, which results in variability of <0.5% in vessel diameterand force generation measurements. This small variability inthe measurements indicates that changes in vessel reactivityin our studies can be attributed only to a specific experimentalintervention rather than to a baseline change in vascular reactivity.The experimental design of our studies does not allow for thesimultaneous determination of solute and water transport acrossthe blood-peritoneal barrier. Therefore, other issues such asthe effect of bioincompatibility and dialysis solution compositionon transperitoneal exchange are not addressed in this study.
Vascular Reactivity
In our studies, two standard techniques were used to identifythe vasoactive components of conventional PDS and quantitativelydescribe their vasoactivity. All hyperosmolar solutions, includingour control PDS, produced significant vascular reactivity inaortic rings and intestinal microvasculature. In addition, sodiumlactate dissolved in isotonic PSS produced a significant vascularreactivity in aortic rings only when H+ concentration in thesolution was increased. Results of our studies were consistentwith literature data, which supported the concept that hyperosmolalityand lactate are the major vasoactive components of PDS and otherhyperosmolar solutions (1,2,8,12–15).
Mortier et al. (9) found that in the rat mesentery, conventionaland new bicarbonate-buffered re-sterilized PDS that containhigh-glucose degradation products (GDP) dilate mesenteric arteries(250 to 350 µm), whereas new bicarbonate-buffered solutionsthat contain low GDP were nearly nonvasoactive. Although noneof the measured GDP was singled out as a possible potent vasoactiveagent, the authors postulated that a possible combination ofthe measured GDP is likely the cause of the dilation response.In their study, multiple solutions with or without nitroglycerin(10–4 M) were randomly tested in the same animal. In addition,the vascular reactivity to the test solution was observed inmesenteric arteries, whereas a change in blood flow in a smallernetwork arteriole was interpreted to mirror a change in thediameter of the mesenteric artery. Our experience with intravitalmicroscopy of the intestinal microcirculation suggests thattesting multiple solutions or pharmacologic manipulations ofthe tissue bath drastically affects baseline microvascular hemodynamics.In particular, endothelium-independent NO donors such as sodiumnitroprusside and nitroglycerin cause a transient maximum dilationfollowed by a rapid drop in the arterial BP and the local bloodflow as a result of systemic absorption of the drug. The vascularreactivity (approximately 20% of baseline) of mesenteric arteriesto conventional PDS seems to be of the same magnitude seen inthe intestinal A1 in this (Figure 3) and previous studies (8).However, the interpretation of the results as to what componentof the dialysis solution is vasoactive differs between our studyand that by Mortier et al. (9). There are several potentialexplanations that could account for the difference in interpretationof the results between the two studies. It is well establishedthat vessels of similar size use different mechanisms for endothelium-dependentregulation of vascular tone depending on vascular bed (16).Similarly, the relative contribution of agonist-stimulated NOand endothelium-dependent hyperpolarizing factor to endothelium-dependentrelaxation seems to differ between genders (17), arteriolarsize within the same vascular bed (18), and arterioles fromdifferent vascular beds (16,19,20). Consistent with these experimentallyvalidated observations is the differential microvascular responsewithin the same vascular bed to specific events such as hemorrhageand sepsis (21,22) or exposure to agonists such as serotonin,angiotensin, and activated complement (23–25). In addition,our experiments were performed in a well-defined vascular bedthat has a unique microvascular architect, which is identicalin all rats (11). In contrast, the mesentery is relatively voidof cells and possesses a poorly defined microvascular network,which accounts for much of the variations in local blood flowin different segments and for the variety of effects of vasoactiveagents on the dynamics of the mesenteric microcirculation (26,27).
Data of this study indicate that an isotonic glucose solutioncauses an insidious and preferential vasodilation in the smallerpremucosal intestinal arterioles and a slight but significantconstriction of the larger inflow arterioles. This unique vasoactiveproperty of the isotonic glucose is mediated by an adenosinereceptor mechanism. In contrast, the intestinal microvesselsremained at baseline during exposure to an isotonic mannitolsolution. However, hypertonic mannitol solutions either as 5%in Krebs’ or as 10% in deionized water caused an instantaneousand sustained dilation similar in magnitude to the Delflex-induceddilation at all levels of the intestinal microvasculature. Althoughmannitol is an efficient osmotic agent, it is a metabolicallyinert solute that is totally excluded from cells. These dataclearly demonstrate that hyperosmolality is a major vasoactivecomponent by itself but that the mechanisms of this dilationeffect may differ depending on the specific metabolic activityof the osmotic solute. Isolated cannulated and pressurized skeletalmuscle arterioles elicit a dilation response proportional toincreasing concentrations of glucose added to a superfusionsolution (5). Similar changes in arteriolar diameter were obtainedin response to superfusion with sucrose or mannitol (5). Thesedata suggest that in in vitro models, arteriolar vasodilationoccurs in proportion to the degree of hyperosmolality. Thiscontrasts with the aortic vascular reactivity in this studyin which the least aortic vascular reactivity was obtained withsolution’s osmolality >550 mOsmol/L. Therefore, translationof the in vitro data to the in vivo situation should be approachedtentatively. This does not refute that such in vitro vascularmodels are the gold standard for assessing vascular controlmechanisms and the signal transduction pathways of these mechanisms.These and previous data do not support a linear relationshipbetween hyperosmolality and vascular reactivity. However, itseems that the magnitude of such reactivity is modified by theH+ concentration, ionic contents, and other vasoactive componentsof the solution such as lactate, as well as by the specificsolute metabolic activity. Indeed, earlier intravital videomicroscopystudies of the rat’s cremaster muscle have shown thatthe magnitude of arteriolar dilation evoked by hyperosmolarsolutions of dextrose, sucrose, or sodium chloride was similarbut that the dilation rate constant differs among the threehyperosmolar solutions (2). Other perfusion studies of the dog’sforelimb (28) and cat’s ileum (29) have found that boththe magnitude and the time course of the dilatory effects ofhyperosmolar dextrose and sodium chloride solutions differedsignificantly.
These data on the intestinal vasoactivity suggest that hypertonicglucose–based solutions dilate the intestinal microvasculatureby at least two mechanisms: One is an instantaneous microvascularvasodilation related to the osmotic stress, and the second isa more insidious, time-dependent vasodilation, stimulated byan energy-dependent transport of glucose into cells. This pathwayis accounted for exclusively by an adenosine receptor–mediatedmechanism as demonstrated in our study. For the vasodilationrelated to the osmotic stress, we suggest that during crystalloid-inducedosmosis, the osmotic water flux through the transendothelialwater-exclusive channels (aquaporin-1) is the primary mechanismwhereby the endothelium is being stimulated to instigate vasodilationeffects. Initial cell shrinkage caused by osmotic-driven waterflow results in a relative increase in cellular ionic contents,especially, Ca2+ and K+, which are known to stimulate endothelium-dependentdilation pathways. Simultaneously, such initial cell volumedecrease triggers a regulatory volume increase characterizedby net water and ionic uptakes as well as stimulation of organicosmolyte transporters to restore the original cell volume (30,31).It has been shown that during conditions of osmotic stress,there is activation of Ca2+-activated K+ channels and ATP-sensitiveK+ channels. Such activation results in a dilation responsethat can be nearly completely abolished by specific inhibitionof these channels (5). Large arteries and other inflow arteriolesminimally express aquaporin-1 (32) and possess insufficientadenosine receptor subtypes (33). This explains the subordinatemagnitude of dilation seen in theses macrovessels compared withthe marked dilation observed in the smaller premucosal intestinalarterioles in these studies.
It is generally conceived that peritoneal dialysis–inducedvasodilation occurs only initially during a hypertonic dwell.Carlsson and Rippe (34) attributed an inflation of the permeabilitysurface area product (PS) of small solutes to an initial vasodilationduring a hypertonic dwell. However, our data suggest that asignificant fraction of the dilation response is sustained preferentiallyin the small precapillary arterioles for as long as these vesselsare exposed to an isotonic solution that contains glucose. Theinfluence of this precapillary vasodilation on the mass transferarea coefficient (MTAC; or PS) for small solutes during thelate phase of the dwell, where the osmotic gradient has dissipated,remains to be determined. It is likely that relaxation of thesesmall vessels in response to exposure to conventional PDS isdetrimental in the number of perfused capillaries and in themodulation of the effective capillary surface area availablefor exchange during peritoneal dialysis. Thus, the instantaneoussubmaximal vasodilation at all levels of the intestinal microvasculatureincluding the inflow feed A1, which doubles its blood flow inthe initial phase of the dwell when osmolality is high, couldprovide a plausible explanation for the high PS for small solutesduring the early phase of the dwell. With dissipation of theosmotic gradient with time as a result of glucose absorption,a subordinate vasodilation is preferentially maintained in thesmaller intestinal premucosal A3, whereas the feed A1 at mostrestore their baseline diameter and blood flow, which explainsthe relatively lower PS for small solutes during the late phaseof the dwell. In contrast, in peritoneal transport rate studiesin humans, the addition of a clinical dose of sodium nitroprussideto a dialysis solution produced no effect on peritoneal fluidkinetics, a slight increase in MTAC for small solutes, but agreat increase in macromolecular clearances (35). Similarly,in rabbits, a clinical dose of sodium nitroprusside exclusivelyenhanced peritoneal macromolecular clearance (36). These clearancedata might reflect a change in peritoneal microvascular permeabilityrather than a vasodilation-mediated modulation of the functionalperitoneal surface area in terms of capillary recruitment. Forrecruitment of functional capillary surface area by dialysissolution–induced dilation, a contact between the dialysateand peritoneal tissue must be established. We have shown thatless than half of the mouse anatomic peritoneum is in contactwith a large volume of solution in the peritoneal cavity (37)and that agitation or use of surfactant-supplemented dialysissolution increases the fraction of contact area, resulting inenhanced transperitoneal exchange (37,38). Although the visceralperitoneum accounts for approximately 60% of the anatomic peritoneum,its fractional contribution to the overall PS for small solutesis only of the order of 30%, whereas the major fraction of PSfor small solutes is accounted for by the much smaller parietalperitoneum (38). This is attributed to the complex geometryof the visceral peritoneum that encompasses "macro-unstirred"pockets of fluid, which equilibrates faster than the rest ofthe dialysate, limiting small-solute diffusion (38). Furthermore,in single-membrane models, permeability and surface area aremultiplicatively linked to form the lumped parameter MTAC, whichis difficult to separate. Therefore, pharmacologic targetingof this lumped parameter in an attempt to improve adequacy ismore likely to change permeability and surface area simultaneously.Thus, efforts to improve dialysis adequacy should be directedtoward improving the wetted peritoneal surface area, which isthe anatomic peritoneum in contact with the dialysate, and tocreate favorable mixed conditions at the visceral peritoneum.
In conclusion, using two different experimental techniques,we demonstrated that hyperosmolality and lactate are the majorvasoactive components of clinical PDS. The pattern and the magnitudeof such reactivity are dependent on vessel size and on the solutes’metabolic activity. Low pH of conventional PDS is not a vasoactivecomponent by itself but renders lactate vasoactive. Energy-dependenttransport of glucose into cells mediates vasodilation of smallvisceral arterioles by an adenosine receptor–mediatedmechanism and constitutes a significant fraction of PDS-mediatedvascular reactivity in the visceral microvasculature. Furtherinvestigation is required to define the signal transductionpathway and the molecular mechanisms of hyperosmolality-inducedvascular reactivity.
Acknowledgments
This project was supported by a VA Merit Review grant and byNational Institutes of Health Research Grant R01 HL76163-01,funded by the National Heart, Lung, and Blood Institute andthe US Army Medical Resources and Material Command.
Part of this study was presented at the first joint InternationalSociety for Peritoneal Dialysis/European Peritoneal Dialysis(ISPD/EUROPD) Congress held in Amsterdam, The Netherlands, August28 to 31, 2004, and partially published in abstract form (39).
Footnotes
Published online ahead of print. Publication date availableat www.jasn.org.
Miller FN, Joshua IG, Harris PD, Wiegman DL, Jauchem JR: Peritoneal dialysis solutions and the microcirculation.
Contrib Nephrol 17
: 51
–58, 1979[Medline]
Miller FN, Nolph KD, Joshua IG, Wiegman DL, Harris PD, Andersen DB: Hyperosmolality, acetate, and lactate: Dilatory factors during peritoneal dialysis.
Kidney Int 20
: 397
–402, 1981[Medline]
Steenbergen JM, Bolen HG: Sodium hyperosmolarity of intestinal lymph causes arteriolar vasodilation in part mediated by EDRF.
Am J Physiol 265
: H323
–H328, 2000
Winkler K, Henriksen JH, Tygstrup N: The effect of hyperosmotic solutions on the hepatic blood flow.
Clin Physiol 13
: 645
–652, 1993[Medline]
Massett MP, Koller A, Kaley G: Hyperosmolality dilates rat skeletal muscle arterioles: Role of endothelial K(ATP) channels and daily exercise.
J Appl Physiol 89
: 2227
–2234, 2000[Abstract/Free Full Text]
Ishizaka H, Kuo L: Endothelial ATP-sensitive potassium channels mediate coronary microvascular dilation to hyperosmolarity.
Am J Physiol 273
: H104
–H112, 1997
Ishizaka H, Gudi SR, Frangos JA, Kuo L: Coronary arteriolar dilation to acidosis: Role of ATP-sensitive potassium channels and pertussis toxin-sensitive G proteins.
Circulation 99
: 558
–563, 1999[Abstract/Free Full Text]
Zakaria ER, Spain DA, Harris PD, Garrison RN: Generalized dilation of the visceral microvasculature by peritoneal dialysis solutions.
Perit Dial Int 22
: 593
–601, 2002[Abstract/Free Full Text]
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 peritoneal membrane: Role of acidity, buffer choice, glucose concentration, and glucose degradation products.
J Am Soc Nephrol 13
: 480
–489, 2002[Abstract/Free Full Text]
Bohlen HG, Gore RW: Preparation of rat intestinal muscle and mucosa for quantitative microcirculatory studies.
Microvasc Res 11
: 103
–110, 1976[CrossRef][Medline]
Duling BR, Staples E: Microvascular effects of hypertonic solutions in the hamster.
Microvasc Res 11
: 51
–56, 1976[CrossRef][Medline]
Steenbergen JM, Bohlen HG: Sodium hyperosmolarity of intestinal lymph causes arteriolar vasodilation in part mediated by EDRF.
Am J Physiol 265
: H323
–H328, 1993
Miller FN, Nolph KD, Harris PD, Rubin J, Wiegman DL, Joshua IG, Twardowski ZJ, Ghods AJ: Microvascular and clinical effects of altered peritoneal dialysis solutions.
Kidney Int 15
: 630
–639, 1979[Medline]
Miller FN, Wiegman DL, Joshua IG, Nolph KD, Rubin J: Effects of vasodilators and peritoneal dialysis solution on the microcirculation of the rat cecum.
Proc Soc Exp Biol Med 161
: 605
–608, 1979[Medline]
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 282
: 1473
–1479, 1997[Abstract/Free Full Text]
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 123
: 1700
–1706, 1998[CrossRef][Medline]
Nishikawa Y, Stepp DW, Chilian WM: In vivo location and mechanism of EDHF-mediated vasodilation in canine coronary microcirculation.
Am J Physiol 277
: 1252
–1259, 1999
Petersson J, Zygmunt PM, Hogestatt ED: Characterization of the potassium channels involved in EDHF-mediated relaxation in cerebral arteries.
Br J Pharmacol 120
: 1344
–1350, 1997[CrossRef][Medline]
Tare M, Parkington HC, Coleman HA: EDHF, NO and a prostanoid: Hyperpolarization-dependent and -independent relaxation in guinea-pig arteries.
Br J Pharmacol 130
: 605
–618, 2000[CrossRef][Medline]
Zakaria ER, Spain DA, Harris PD, Garrison RN: Resuscitation regimens for hemorrhagic shock must contain blood.
Shock 18
: 567
–573, 2002[CrossRef][Medline]
Zhao H, Spain DA, Matheson PJ, Vaughn C, Harris PD, Garrison RN: Sustained infection induces 2 distinct microvascular mechanisms in the splanchnic circulation.
Surgery 128
: 513
–519, 2000[CrossRef][Medline]
Alsip NL, Harris PD: Receptor mediation of microvascular responses to serotonin in striated muscle.
Am J Physiol 261
: H1525
–H1533, 1991
Fleming JT, Harris PD, Joshua IG: Endogenous prostaglandins selectively mask large arteriole constriction to angiotensin II.
Am J Physiol 253
: H1573
–H1580, 1987
Lubbe AS, Garrison RN, Harris PD: Endothelium-dependent microvascular responses to activated complement.
J Surg Res 57
: 654
–660, 1994[CrossRef][Medline]
Banks RO, Gallavan RH Jr, Zinner MH, Bulkley GB, Harper SL, Granger DN, Jacobson ED: Vasoactive agents in control of the mesenteric circulation.
Fed Proc 44
: 2743
–2749, 1985[Medline]
Harper SL, Bohlen HG, Granger DN: Vasoactive agents and the mesenteric microcirculation.
Am J Physiol 249
: G309
–G315, 1985
Gazitua S, Scott JB, Swindall B, Haddy FJ: Resistance responses to local changes in plasma osmolality in three vascular beds.
Am J Physiol 220
: 384
–391, 1971[Free Full Text]
Levine SE, Granger DN, Brace RA, Taylor AW: Effects of hyperosmolality on vascular resistance and lymph flow in the cat ileum.
Am J Physiol 234
: H14
–H20, 1978
O’Neill WC, Steinberg DF: Functional coupling of Na(+)-K(+)-2Cl- cotransport and Ca(2+)-dependent K+ channels in vascular endothelial cells.
Am J Physiol 269
: C267
–C274, 1995
O’Neill WC: Physiological significance of volume-regulatory transporters.
Am J Physiol 276
: C995
–C1011, 1999
Fang W, Qian J, Yu Z, Chen S: Expression of aquaporin-1 in the human peritoneum and the effect of peritoneal dialysis on its expression.
Chin Med J (Engl) 116
: 1370
–1373, 2003
Li N, Harris PD, Zakaria ER, Matheson PJ, Garrison RN: Microvascular responses to adenosine help explain functional and pathologic differences between intestinal segments.
Am J Surg 188
: 526
–531, 2004[CrossRef][Medline]
Carlsson O, Rippe B: Enhanced peritoneal diffusion capacity of 51Cr-EDTA during the initial phase of peritoneal dialysis dwells: Role of vasodilatation, dialysate 'stirring’, and of interstitial factors.
Blood Purif 16
: 162
–170, 1998[CrossRef][Medline]
Douma CE, de Waart DR, Struijk DG, Krediet RT: The nitric oxide donor nitroprusside intraperitoneally affects peritoneal permeability in CAPD.
Kidney Int 51
: 1885
–1892, 1997[Medline]
Douma CE, Zweers MM, de Waart DR, van der Wardt AB, Krediet RT, Struijk DG: Substrate and inhibitor for nitric oxide synthase during peritoneal dialysis in rabbits.
Perit Dial Int 19
: 358
–364, 1999
Flessner MF, Lofthouse J, Zakaria ER: Improving contact area between the peritoneum and intraperitoneal therapeutic solutions.
J Am Soc Nephrol 12
: 807
–813, 2001[Abstract/Free Full Text]
Zakaria ER, Carlsson O, Rippe B: Limitation of small-solute exchange across the visceral peritoneum: Effects of vibration.
Perit Dial Int 17
: 72
–79, 1997[Abstract/Free Full Text]
Zakaria ER, Hunt MC, Na L, Harris PD, Garrison RN: On the disparity in osmolarity-induced vascular reactivity [Abstract].
Perit Dial Int 24
: S11
, 2004
Received for publication September 13, 2004.
Accepted for publication June 24, 2005.
Top
Abstract
Introduction
-1 adrenergic agonist) was added in the bath to induce contraction, followed by acetylcholine (ACh) to induce endothelium-dependent relaxation to demonstrate presence/absence of functional endothelium.
P < 0.01 by ANOVA and Bonferroni post test versus A3 premucosal arterioles. (B) Mechanism of isotonic glucose–induced microvascular reactivity. BL, baseline arteriolar diameter; A1, intestinal inflow arteriole; pA3 and dA3, intestinal proximal (p) and distal (d) A3 premucosal arterioles. *P < 0.01 by ANOVA and Bonferroni post test versus BL. 
