AJP - Heart Fuel your research with LabChart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Heart Circ Physiol 276: H1012-H1021, 1999;
0363-6135/99 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Frame, M. D. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Frame, M. D. S.
Vol. 276, Issue 3, H1012-H1021, March 1999

Conducted signals within arteriolar networks initiated by bioactive amino acids

Mary D. S. Frame

Department of Anesthesiology, Biomedical Engineering Program, University of Rochester School of Medicine and Dentistry, Rochester, New York 14642


    ABSTRACT
Top
Abstract
Introduction
Methods
Results
Discussion
References

Our purpose was to determine the specificity of L-arginine (L-Arg)-induced conducted signals for intra- vs. extracellular actions of L-Arg. Diameter and red blood cell velocities were measured for arterioles [18 ± 1.6 (SE) µm] in the cremaster muscle of pentobarbital sodium-anesthetized (Nembutal, 70 mg/kg) hamsters (n = 53). Remote (conducted) responses were viewed ~1,000 µm upstream from the local (micropipette) application. Six amino acids were tested: L-arginine, L-cystine, L-leucine, L-lysine, L-histidine, and L-aspartate (100 µM each). Only L-Arg induced a remote dilation; L-lysine and L-aspartate had no effect, and the others each induced a significant remote constriction. There is a second conducted signal initiated by L-arginine that preconditions the arteriolar network and upregulates a direct response of L-arginine to dilate the remote site. This was blocked by inhibition of L-arginine uptake at the local (preconditioning) site (100 µM L-histidine or 1 mM phenformin). Arginine-glycine-aspartate (100 µM)-induced remote dilations (+3.2 ± 0.3 µm) were not mimicked by a peptide control and were prevented by anti- integrin alpha v monoclonal antibody. Remote dilations were greater in animals with a higher wall shear stress for arginine-glycine-aspartate (r2 = 0.92) but not for L-arginine (r2 = 0.12). Thus L-arginine initiates separate conducted signals related to system y+ transport, integrins, and baseline flow.

integrins; flow-dependent dilation; heterogeneity of flow


    INTRODUCTION
Top
Abstract
Introduction
Methods
Results
Discussion
References

THE AMINO ACID ARGININE is involved in a growing list of biologic events from serving as the substrate for nitric oxide (NO) synthase to being a critical component of specific integrin-binding sites [e.g., arginine-glycine-aspartate (RGD)] (42). Previously, we showed that L-arginine (L-Arg) induces two separate conducted signals in the microcirculation (17): 1) a remote dilation that occurs as L-Arg is applied to a local site >= 1,000 µm downstream and 2) a direct dilation of the remote site that can be elicited 10-15 min after the local site stimulation. This upregulated response requires a stimulated conducted signal to precondition the arteriolar network and is observed only while L-Arg is directly applied to the remote site (17). This response implies that any ongoing steady-state conducted processes are not sufficient to precondition the network or they are inhibitory. Because NO donors initiate the remote dilation (10, 13, 18) and induce the "network-preconditioning conducted signal" (13), it may be that remote responses initiated by L-Arg are entirely mediated through NO. However, L-Arg may be affecting several pathways, and other possibilities or components may be involved. These possibilities include nonspecific effects of arginine (i.e., charge) and specific effects, such as binding to the RGD integrin-binding site (42). To address the mechanism(s) of remote responses and expand our understanding of their importance in the coordination of flow distribution within the microcirculation, several of these possibilities were explored.

L-Arg is a basic amino acid [isoelectric point (pI) 10.76]. Thus, presumably at physiological pH, L-Arg will be positively charged. D-Arg (which is not transported) does not mimic the remote response characteristics of L-Arg (17). D-Arg does, of course, have the same pI, suggesting that a nonspecific effect of extracellular pH is not likely to be involved. L-Arg may be acting through an intracellular pH effect. This can be tested by examining amino acids that are transported into cells of the vascular wall and that encompass a range of pI values. Free amino acids were thus chosen to include a range of pI values: 9.47 for L-lysine (L-Lys), 7.64 for L-histidine (L-His), 6.04 for L-leucine (L-Leu), 5.02 for L-cystine (L-Cys), and 2.98 for L-aspartate (L-Asp) (51a).

A factor that compounds the clean evaluation of the pH effect of these amino acids is amino acid transport into the cell. L-Arg is transported into endothelial and vascular smooth muscle cells primarily via the system y+ cationic amino acid transporter (5, 11). This transporter is stimulated by hyperpolarization of the membrane potential (4), as may occur during remote vasodilations (55). A different question is, Does y+ transport activity alone induce a remote response? To answer this question, amino acids were chosen that are, or are not, transported via the system y+ transporter: L-Arg, L-Lys, and L-Cys (system y+), L-Cys (system x-c), L-Leu (system L), L-Asp [system xAG (requires stimulation)], and L-His [system N (not present in endothelial cells)] (5, 25, 27, 31, 32, 38, 48, 51). Thus only L-His is not transported into cells of the vascular wall under normal circumstances. Furthermore, three nonspecific inhibitors of L-Arg uptake were tested: L-His, sulfanilamide, and phenformin (8, 12, 19, 20, 22, 29, 53).

Another possibility is that L-Arg may be acting through a specific extracellular binding site. Several specific extracellular binding sites for arginine-containing peptides are vasoactive. Of importance here are the RGD-containing peptides involved in local dilation (34) and antagonism of flow-dependent dilation (36) in isolated arterioles. The RGD recognition site is common to many forms of the alpha - and beta -integrin subunits (42, 54). Of these, the alpha vbeta 3- and alpha vbeta 5-integrins are involved in vascular remodeling and respond to changes in flow (42, 54). To determine whether integrins are also involved in remote responses, we have used the RGD peptide and a monoclonal antibody to the alpha v-integrin subunit.

The purpose of this study is to examine the L-Arg-induced remote response by specifically determining whether other bioactive amino acids or peptides containing L-Arg induce remote responses. The conducted signal causing the remote dilation and the network-preconditioning conducted signal are investigated.


    METHODS
Top
Abstract
Introduction
Methods
Results
Discussion
References

Adult male Golden hamsters [HSD:Syr, 117 ± 11 (SD) g, n = 57] were anesthetized with pentobarbital sodium (70 mg/kg ip), tracheostomized, and maintained on a constant infusion of pentobarbital sodium (10 mg/ml at 0.56 ml/h) via a femoral venous catheter. Systemic hematocrit (Hct) was determined before (55 ± 4%) and after (54 ± 5%) the experiment. Body temperature was maintained between 37 and 38°C by a conductive heat source. Mean arterial pressure was monitored via a left femoral arterial catheter (90 ± 10 mmHg) and was constant (±10 mmHg) for each preparation. Erythrocytes from age-matched donor animals (n = 29) were labeled with a fluorescent fluorophore [substituted tetramethylrhodamine isothiocyanate (XRITC); Molecular Probes, Eugene, OR] by use of an established protocol (44, 46, 47). The fluorescently labeled erythrocytes were injected in tracer quantities via a right jugular catheter and were used to measure cell flow as an index of blood flow changes. The right cremaster muscle was prepared for in vivo microcirculatory observations (16, 52). The preparation was continuously superfused with bicarbonate-buffered saline containing (in mmol/l) 132 NaCl, 4.7 KCl, 2.0 CaCl2, 1.2 MgSO4, and 20 NaHCO3 (equilibrated with gas containing 5% CO2-95% N2, pH 7.4 at 34°C). All chemicals were obtained from Sigma Chemical (St. Louis, MO), unless otherwise noted. The microcirculation was observed with transillumination by using a modified Nikon upright microscope with a ×25 or a ×40 (Nikon) objective at ×1 or ×1.5 magnification. Epi-illumination was used to visualize the XRITC-labeled red blood cells with use of a Chroma G2A filter (Brattleboro, VT). Video images were produced using a CCD 72s video camera and a Gen/Sys II video intensifier (Dage-MTI, Michigan City, IN).

During a 60-min stabilization period, the XRITC-labeled cells were administered (44, 46). Observations were made along a transverse arteriole (feed vessel) located in each preparation, as described previously (14, 15). This feed and its branches are schematized in Fig. 1. This site was chosen for study, because it is functionally an arteriolar network; the branches directly feed adjacent capillary networks (2, 52).


View larger version (21K):
[in this window]
[in a new window]
  		Fig. 1.   Schematic of feed vessel indicating local site of test agent application (local site) and upstream observation site (remote site). Test agent was applied to remote site to determine network-preconditioned responses (direct responses). Typically, 3-5 branches occur between local and remote sites. Stylized micropipettes are shown at local and remote sites to indicate region exposed to pipette contents. Flow out of each micropipette was verified by watching flow stream of fluorescently labeled dextran added to contents of each pipette.

Stable arteriolar diameters (no vasomotion) were verified for 2-5 min before application of test agents. Remote responses were obtained by application of the test agent at the local site (Fig. 1), via micropipette, for 2 min, while the diameter changes upstream at the remote site were observed. Direct responses were obtained by application of the test agent at the remote site for 5 min and simultaneous measurement of diameter changes at the remote site. The micropipette diameter was 20 ± 7 µm. Micropipettes were placed within 50 µm of the vessel wall. All agents were applied using a pneumatic delivery system with -2-cmH2O holding pressure and +30-cmH2O delivery pressure. Each micropipette contained 50 µM fluorescein-labeled dextran (4,000 mol wt) as a flow marker to verify the region of the arteriole that was exposed to pipette contents and was observed using a Chroma B1E filter. Because of the placement of the micropipette, geometry of the site chosen, and continuous flow of suffusate over the tissue, the micropipette contents were washed away from the remote site. Diffusion between the local and remote sites would have taken >2 h in an unstirred area; diffusion was not likely between the two sites. Several protocols were used to address conducted signals causing the remote responses and network-preconditioned conducted signals (Table 1).

                              
View this table:
[in this window]
[in a new window]
  		Table 1.   Summary of responses tested

Protocol 1: inhibition of conducted signals. One amino acid was tested per animal. Direct responses were obtained in response to 100 µM (each) L-Arg, L-Lys, L-Leu, L-Cys, L-His, and L-Asp (see Fig. 3). Remote responses to 2 min of micropipette application of the same amino acid were tested (Fig. 2). After 10-15 min the direct response was again determined (Fig. 3). This sequence was done during inhibition of conducted signals (data in text) and then after a 10-min recovery period. It was mandatory to test for the responses during inhibition first, because once the network-preconditioned signal is activated, it persists for >6 h. Conducted signals were blocked nonspecifically (1, 39) with halothane (~1.7 mM) in bicarbonate-buffered saline or sucrose in bicarbonate-buffered saline (600 mosM total); only one inhibitor was used per animal. Specific inhibitors of gap junction signaling or of conducted responses are not available. Halothane or sucrose was applied continuously to the feed vessel, via micropipette, at a point midway between the local downstream site and the remote observation site. The inhibitors were applied for 10 min before and then continuously with application of the test agent at the local site (remote response) and continuously during application of the test agent at the remote site (direct response). Only the middle region of the feed vessel was exposed to the inhibitors, as verified by the FITC-dextran in the pipette. Neither sucrose nor halothane induced remote diameter changes, as reported previously (17).


View larger version (11K):
[in this window]
[in a new window]
  		Fig. 2.   Peak diameter changes (means ± SE, test - baseline) of remote observation site with local application of (100 µM each) L-histidine (L-His, n = 5), L-cystine (L-Cys, n = 7), L-leucine (L-Leu, n = 7), L-lysine (L-Lys, n = 7), L-aspartate (L-Asp, n = 4), or L-arginine (L-Arg, n = 9). Peak diameter changes within 60 s of exposure were used (protocol 1). x-Axis, isoelectric point (pI) for each amino acid. * Significant change from baseline, P < 0.05.


View larger version (13K):
[in this window]
[in a new window]
  		Fig. 3.   Same amino acid was applied at remote site for 5 min (open bars), then at local site for 2 min (see Fig. 2), and again at remote site for 5 min (filled bars); there was a 10- to 15-min delay between each exposure. Direct responses and diameter changes (means ± SE, test - baseline), with initial direct application of L-Cys (n = 7), L-Leu (n = 7), L-Lys (n = 7), or L-Arg (n = 6) and with direct application of the amino acids after network preconditioning by a local application (protocol 2), are shown. * Significant change from baseline, P < 0.05.

Protocol 2: network-preconditioning conducted signals. L-Arg plus one other agent were tested per animal. Remote responses were tested to micropipette application of 100 µM (each) L-Arg, L-Lys, L-Leu, and L-Cys (Fig. 4) or amino acid sequences RGD or arginine-beta -alanine (Rbeta A; see Fig. 6). After 10-15 min, 100 µM L-Arg was applied to the remote site to test the direct response. For experiments with RGD, responses were obtained with and then without inhibition of the conducted signals with use of halothane, as described for protocol 1 (data in text).


View larger version (14K):
[in this window]
[in a new window]
  		Fig. 4.   Amino acids were applied at local site for 2 min, and 10-15 min later, L-Arg was applied directly at remote site (protocol 2). Peak diameter changes (means ± SE, test - baseline) at remote site during a 5-min direct application of L-Arg to remote site, after local site stimulation with L-Cys (n = 5), L-Leu (n = 3), L-Lys (n = 3), or L-Arg (n = 7; 100 µM each), are shown. * Significant change from baseline; dagger  significant change from L-Arg, P < 0.05.

Protocol 3: inhibition of y+ cationic amino acid transport. L-Arg plus one inhibitor were tested per animal. Remote responses and then direct responses to 100 µM L-Arg were tested first with and then without inhibitors (Fig. 5). Specificity for the y+ amino acid transporter was tested by applying L-His (100 µM), phenformin (1 mM), or sulfanilamide (1 mM, separately) to the local site for 5 min before and then continuously with L-Arg (in a 2nd pipette) exposure at the local site. Specific inhibitors of the y+ cationic amino acid transporter are not available. The three agents used are biguanides, sharing common binding features with L-Arg, but they are not transported via the y+ transporter (8, 12, 19, 20, 22, 25, 51).


View larger version (11K):
[in this window]
[in a new window]
 

View larger version (11K):
[in this window]
[in a new window]
  		Fig. 5.   Diameter changes (means ± SE, test - baseline) of remote site with local site stimulation with L-Arg (A, remote response) or direct application of L-Arg to remote site (B, direct response; protocol 3). Responses were obtained in absence (none) or presence of local micropipette application of 1 mM phenformin (n = 4), 1 mM sulfanilamide (n = 5), or 0.1 mM L-His (n = 5). These inhibitor agents were applied continuously at local site for 10 min before local site stimulation with L-Arg and during local and remote application of L-Arg. * Significant change from baseline; dagger  significant change from none, P < 0.05.

Protocol 4: inhibition of alpha v-integrin binding. L-Arg (10-7-10-3 M) or RGD (100 µM) plus the anti-integrin alpha v-monoclonal antibody (1:100; GIBCO BRL, Life Technologies, Grand Island, NY) were tested, with one test agent plus the antibody per animal. All agents were applied to the local site. Remote responses were obtained in sequence to 2 min of exposure to anti-alpha v, 5 min of exposure to anti-alpha v plus RGD or L-Arg (in a 2nd pipette), 2 min of exposure to the test agent alone, and finally 5 min of reexposure to the anti-alpha v plus RGD or L-Arg (in continuous sequence; see Fig. 7). For multiple L-Arg concentrations, remote responses were obtained in sequence to L-Arg (2 min) and then to 2 min of exposure to anti-integrin alpha v-monoclonal antibody (2nd micropipette) plus L-Arg (see Fig. 8).

Baseline red blood cell velocities were obtained for the first (remote site, Fig. 1) and last (local site, Fig. 1) bifurcations of the feed vessel in 23 experiments (n = 12 with RGD and n = 11 with L-Arg; see Fig. 9). Cell flux (F, cells/s) is calculated as follows: F = (mt/p)/t, where mt is the number of fluorescent cells crossing a specified vessel plane in time t and p is the fraction of fluorescent cells in the total red blood cell population [0.62 ± 0.09% (SD)]. Fluorescent cells were taken as representative of the total population (42, 46). Individual velocities (µm/s) were measured as the distance traveled in one video field (1/60th s) for all fluorescent cells crossing the specified sampling plane during a 30-s time interval. The harmonic mean of the individual red blood cell velocities was calculated (mean axial cell velocity, vc) and used to approximate the average velocity of blood in the vessel (6). Hct, the time-averaged volume fraction of cells in the vessel, was calculated as follows (45): Hct = F · Vc/vc · pi r2, where r is the vessel radius and Vc is the mean corpuscular volume [58 × 10-15 liters (46)]. The apparent viscosity (eta app) was calculated from the relationship between vessel Hct, vessel diameter, and the relative viscosity (40). The shear rate (gamma , s-1) was calculated as follows: gamma  = 8 · vc/D, where D is vessel wall diameter. Wall shear stress (Tomega , dyn/cm2) was calculated as follows: Tomega  = eta app · gamma .

All diameter changes were monitored on-line by means of a video caliper system (Microvascular Research Institute, College Station, TX) and a software data acquisition system (Strawberry Tree, Workbench, Sunnyvale, CA) calibrated with a micrometer. Red blood cell velocities (and associated diameters) were determined off-line from the recorded image with use of image analysis software developed for this application (Dept. of Anesthesiology). The calculated values were pooled by test condition to determine the population means and standard errors. Comparisons were made between groups by ANOVA (multiple comparisons). Changes from baseline were evaluated by t-test. Correlation was evaluated using linear regression. For all statistics, differences were considered significant when P < 0.05 (49).


    RESULTS
Top
Abstract
Introduction
Methods
Results
Discussion
References

Remote responses to amino acids. The baseline diameter of the remote site was 18 ± 2 (SE) µm (n = 53). This remote site was 1,240 ± 90 µm upstream from the local site of test agent application. As shown schematically in Fig. 1, the local and remote sites were located at the distal and proximal (respectively) ends of the same feed arteriole (1,465 ± 96 µm). Figure 2 shows the average of the peak remote response within 60 s of local application of L-His, L-Cys, L-Leu, L-Lys, or L-Arg (100 µM each) as a function of the pI of each amino acid. Responses to L-Asp and to L-Lys were not significant diameter changes. For the amino acids that are known to be transported into vascular cells, the magnitude of the remote response is linearly related to the pI of the amino acid (L-Cys, L-Leu, L-Lys, and L-Arg; r2 = 0.92). Of more importance may be the finding that only L-Arg induced a remote dilation.

To determine whether the remote responses likely occurred via gap junction communication, halothane (1.7 mM) or sucrose (600 mosM) was applied to the middle of the feed vessel, halfway between the local and remote sites, for 10 min before and continuously during application of the test agent. L-Arg, L-Cys, L-Leu, or L-Lys was tested in the presence of halothane or sucrose. Neither halothane nor sucrose (diameter change +0.17 ± 1 and -0.06 ± 0.3 µm, respectively) induced a remote response. Each agent completely blocked the remote response to L-Arg (+0.4 ± 0.26 and -0.9 ± 0.7 µm for sucrose and halothane, respectively), as described previously (17). The remote responses to L-Cys were significantly decreased by sucrose (-0.44 ± 0.02 µm, n = 3) or by halothane (-0.95 ± 0.4 µm, n = 3). Neither inhibitor significantly changed the remote responses to L-Leu (-0.8 ± 0.1 and -1.1 ± 0.7 µm for sucrose and halothane, respectively, n = 3 each) or L-Lys (-0.25 ± 0.4 and -0.47 ± 0.3 µm for sucrose and halothane, respectively, n = 3 each). However, because of the magnitude of the responses, we can conclude only that the remote constrictions to L-Cys and remote dilations to L-Arg were blocked by these agents.

Direct responses before and after local site stimulation. The same amino acid was applied at the remote site for 5 min (direct response), at the local site for 2 min (remote response), and then again at the remote site for 5 min (direct response). This was done to determine whether these amino acids could induce the network-preconditioning effect. With the initial application at the remote site, L-Cys dilated the remote site directly, L-Lys constricted the remote site, and neither L-Arg nor L-Leu was vasoactive (Fig. 3, open bars). After local application of L-Cys (i.e., after a conducted signal), L-Cys again caused a dilation when applied at the remote site; however, this dilation was significantly larger than before the local application. Repeated applications of L-Cys to the remote site did not have this effect, as described previously for L-Arg (17). The baseline diameters for L-Cys were not different initially (16.7 ± 1.7 µm) compared with before the remote response (17.9 ± 3 µm) or before the final local application (17.8 ± 2.9 µm). The direct effect of L-Leu or L-Lys at the remote site was not altered after local application of L-Leu or L-Lys (respectively). Likewise, there was no change in the baseline diameters between exposures. As reported previously (17), a direct dilation was obtained with L-Arg only after local application of L-Arg. Thus local site stimulation with L-Cys or L-Arg significantly alters the subsequent direct responses to these two amino acids. Although the remote responses were opposite (constriction with L-Cys and dilation with L-Arg), the network-preconditioned direct response was always dilation.

In a separate series of experiments the direct effect of L-Arg (5 min) at the remote site was determined after a 2-min application of each amino acid to the local site (Fig. 4) to determine whether other amino acids could induce the network-preconditioning response to L-Arg. Before stimulation of the downstream local site, L-Arg is not vasoactive (Fig. 3, open bars), as shown previously (17). After stimulation of the local site with L-Cys, L-Lys, or L-Arg, but not L-Leu, the remote site dilated significantly to directly applied L-Arg (Fig. 4). Thus L-Cys and L-Lys, in addition to L-Arg, each induced the network-preconditioning response to L-Arg. L-Cys had induced a remote constriction, and L-Lys had not induced a measurable remote response. Importantly, neither a specific type of response (constriction vs. dilation) nor the presence of a measurable remote response was essential in causing the network-preconditioned response to L-Arg. Furthermore, L-Leu did not alter the direct effect of L-Arg at the remote site, despite a significant remote constriction, indicating that not all remote responses are capable of inducing this network effect. The common feature of the agents inducing the network-preconditioning effect was the known transport via the system y+ transporter. We therefore tested whether inhibition of the y+ transporter would block the network-preconditioning response to L-Arg.

Inhibition of the y+ cationic amino acid transporter. L-Arg influx via the y+ cationic amino acid transporter was nonspecifically blocked with three biguanides: L-His, phenformin, and sulfanilamide. Each agent is recognizably nonspecific in nature and has known secondary effects. Of these three agents, only L-His induced a statistically significant remote response: -0.6 ± 0.2 µm (n = 5), -0.15 ± 0.4 µm (n = 4), and -0.66 ± 0.4 µm (n = 5) for L-His, phenformin, and sulfanilamide, respectively. The remote dilation to L-Arg applied at the local site was significantly attenuated only by local application of sulfanilamide and persisted during local application of the other two transport blockers (Fig. 5A). Phenformin appeared to accentuate the remote dilation; however, this was not statistically significant; the baseline diameters were 14.5 ± 1.5 µm before the remote response and 15 ± 2.6 µm before the direct response. Because two of these three nonspecific agents had no effect, it is likely that L-Arg transport per se is not required for L-Arg to induce a remote dilation. With continued application of transport inhibitors downstream (at the local site), the direct effect of L-Arg (at the remote site) to directly dilate the remote site was blocked by continued local application of L-His, significantly diminished by phenformin, and not affected by sulfanilamide (Fig. 5B). Thus, although sulfanilamide successfully blocked a remote dilation to L-Arg, the secondary effect of an altered direct response at the remote site was unchanged by sulfanilamide. Furthermore, although neither L-His nor phenformin blocked a remote dilation to L-Arg, the direct effect of L-Arg at the remote site was either significantly decreased or not seen. At least two distinct mechanisms are therefore apparent for the remote vasomotor effects of L-Arg: 1) remote dilation when L-Arg is applied at the local site, which does not require L-Arg transport, and 2) initiation of the network-preconditioning signal, which requires L-Arg transport.

Remote responses and integrins. Evidence presented above suggests that the remote response to L-Arg is dependent on the pI value and not on L-Arg transport. Yet D-Arg (with the same pI) does not mimic the response (17). One manner in which L-Arg could be initiating a remote response is, therefore, through a specific extracellular binding site. To directly test whether specific extracellular binding of L-Arg induced a remote response, the bioactive peptide RGD was applied at the local site. RGD did not induce a diameter change at the local site (0 ± 0.3 µm, n = 12). Application of 100 µM RGD to the local site did, however, induce a significant remote dilation of 1.3-8.1 µm (4.0 ± 0.6 µm) at 1,200 ± 110 µm upstream at the remote site (Fig. 6). The peak dilation occurred 30-120 s after the onset of RGD application. After a remote dilation to RGD, a significant dilation was obtained by applying L-Arg to the remote site (n = 4); the magnitude of this dilation was one-half of the direct response to L-Arg that is normally obtained after a remote dilation to L-Arg (Fig. 4). The baseline diameters in those experiments were 13.2 ± 1.7 µm before the remote response and 12.5 ± 1.6 µm before the direct response. Halothane (1.7 mM) blocked the remote dilation to RGD (-1.2 ± 1.6 µm) and subsequent direct effect of L-Arg at the remote site (-1 ± 0.3 µm, n = 4). As a peptide control, 100 µM Rbeta A induced a significant remote constriction (Fig. 6) but did not significantly alter the direct effect of L-Arg at the remote site (n = 3). The baseline diameters were 14.5 ± 1.7 µm before the remote response and 13.5 ± 2 µm before the direct response. Finally, L-Asp was tested to determine whether the remote response to RGD was induced by aspartate. Figure 2 shows that L-Asp did not induce a remote response.


View larger version (22K):
[in this window]
[in a new window]
  		Fig. 6.   Peak diameter changes (means ± SE, test - baseline) at remote site with local application of arginine-glycine-aspartate (RGD) or arginine-beta -alanine (Rbeta A, filled bars, remote response; protocol 2). At 10-15 min after local exposure to peptide, L-Arg was applied directly to remote site (hatched bars, direct response, n = 4 with RGD, n = 3 with Rbeta A). * Significant change from baseline; dagger  significant change from RGD, P < 0.05.

To further test the specificity of responses induced by RGD for a specific integrin subunit, an anti-alpha v-monoclonal antibody (1:100) was applied simultaneously (2nd pipette) with L-Arg (n = 4) or RGD (n = 4). Figure 7 shows that the anti-alpha v was not vasoactive alone and that the anti-alpha v quickly and reversibly blocked the L-Arg- or the RGD-induced responses. The remote dilation to L-Arg was not completely blocked by the anti-alpha v, once the remote response was elicited, whereas the anti-alpha v completely blocked the remote response to RGD. Thus specific extracellular arginine binding to a site involving the alpha v-integrin subunit activates a conducted signal, which induces a remote dilation and alters the direct effect of L-Arg at the remote site.


View larger version (13K):
[in this window]
[in a new window]
  		Fig. 7.   Peak diameter changes (means ± SE, test - baseline) at remote site with local application of anti-alpha v-monoclonal antibody (1:100), RGD (n = 4), or L-Arg (n = 4, 100 µM; protocol 4). Responses were obtained in sequence: peak response during a 2-min exposure to anti-alpha v (alphaV), peak response to alpha v + RGD or L-Arg (5-min exposure), peak response to RGD or L-Arg within 2 min of stopping exposure to alpha v, and peak response during 5-min reexposure to alpha v + RGD. dagger  Significant change from baseline; * L-Arg differs from RGD, P < 0.05.

There appears to be a component of the L-Arg-induced remote response that is not inhibited by anti-alpha v. To examine this further, remote responses were obtained using a range of L-Arg concentrations and then were obtained during simultaneous exposure to L-Arg plus 1:100 anti-alpha v. Figure 8 shows that remote responses to L-Arg are directly related to the L-Arg concentration, with a remote constriction at 10-7 M and remote dilations at 10-6-10-3 M L-Arg. The maximum remote dilation occurs at 100 µM L-Arg. Vehicle controls yield nonsignificant diameter changes locally and remotely, as reported previously (15, 17). Anti-alpha v has no effect on the remote constriction at low L-Arg but blocks 30-100% of the remote dilation for >10-6 M L-Arg. Thus a portion of the L-Arg-induced remote response is related to binding the alpha v-integrin subunit.


View larger version (15K):
[in this window]
[in a new window]
  		Fig. 8.   Peak diameter changes (means ± SE, test - baseline) at remote site with 2-min local application of L-Arg (10-7-10-3 M) alone and then 2-min exposure to anti-alpha v-monoclonal antibody (1:100, n = 4; protocol 4). Significant remote constriction to L-Arg (no symbol) was not affected by anti-alpha v. Significant remote dilations to L-Arg were partially blocked by anti-alpha v. * L-Arg alone differs from L-Arg + anti-alpha v; dagger  significant change from baseline, P < 0.05.

Wall shear stress and remote responses. There is a strong link between focal adhesion (integrin) binding and vascular responses to flow. In contrast, little is known about the relationship between remote responses and flow. We therefore tested whether the remote responses to L-Arg or RGD were related to the baseline flow state. Baseline red blood cell velocities were obtained immediately before remote responses. In 12 experiments with RGD, the calculated baseline shear rate at the remote site was 70-1,500 s-1 (700 ± 120 s-1). Separately, in 11 experiments with L-Arg, the baseline shear rate was 40-1,300 s-1 (550 ± 120 s-1). The average wall shear stress was 15 ± 3 and 13 ± 2 dyn/cm2 with RGD and L-Arg, respectively. Thus these animals exhibited an equivalent range of hemodynamic conditions for experiments with RGD or L-Arg. Furthermore, the total range in remote responses with RGD (see above) was similar to the range in remote responses with L-Arg (range 7.7-1.0 µm, mean 3.9 ± 0.6 µm). Figure 9 shows the remote dilation initiated by RGD or L-Arg as a function of the baseline wall shear stress at the remote site. The remote dilation to RGD was significantly greater in animals with a higher baseline wall shear stress (r2 = 0.92). With L-Arg, response and wall shear stress were unrelated (r2 = 0.12). For comparison, we examined the relationship between the remote response and the components of wall shear stress (at the remote site) that were directly measured (diameter, red blood cell flux, and velocity) or calculated (Hct, viscosity, and shear rate). The remote response was significantly greater in animals with a higher baseline shear rate at the remote site (r2 = 0.81); other factors were not correlated with this remote response. The flow parameters were determined for the local downstream site of RGD exposure as well. The downstream baseline velocity (local site) was 1,000 ± 160 µm/s, the shear rate was 930 ± 180 s-1, and the wall shear stress was 13 ± 2 dyn/cm2. The wall shear stresses at the remote and local sites were correlated (r2 = 0.68). As expected, the remote dilation to RGD (at the remote site) was also significantly greater in animals with a higher wall shear stress at the local site (r2 = 0.80).


View larger version (14K):
[in this window]
[in a new window]
  		Fig. 9.   Relationship between baseline wall shear stress at remote site and remote dilation to RGD (n = 12) or L-Arg (n = 11) in separate experiments. Solid line, significant correlation between response and wall shear stress with RGD [remote dilation = 0.88 (wall shear stress) + 0.21, r2 = 0.92]. Response and wall shear stress were unrelated with L-Arg [remote dilation = -0.04 (wall shear stress) + 4.3, r2 = -0.12; line not shown].


    DISCUSSION
Top
Abstract
Introduction
Methods
Results
Discussion
References

This study shows that conducted signals do not require a local response, nor do they require a remote response to elicit a network-preconditioning response. The remote response is mechanistically different from the network-preconditioning response. The network response requires L-Arg transport, whereas the remote response does not. The remote response may be related to the charge of the stimulating amino acid. Finally, most of the remote response to L-Arg is directly linked to an RGD binding site involving the alpha v-integrin subunit. Importantly, the integrin-linked remote response is directly related to the baseline flow state of the microcirculation.

The RGD recognition site is associated with four forms of the alpha -integrin subunit: alpha v, alpha 5, alpha 8, and alpha IIIb (42). In the present study we have determined that interstitial application of the RGD peptide sequence induces a significant remote dilation and that the alpha v-subunit is involved in the remote response. Importantly, the magnitude of the remote response induced by RGD is greater in animals with a higher baseline wall shear stress, suggesting that integrin binding increases the network capacity for flow changes in proportion to the baseline flow states. This coupling between flow and the integrin-induced remote response is not due to a nonspecific effect of arginine, because flow and remote responses were unrelated to L-Arg. Furthermore, not all the remote response to L-Arg is blocked by the anti-alpha v-antibody, providing additional evidence for our conclusion that L-Arg induces multiple signals.

The RGD recognition site is likewise associated with several forms of the beta -subunit: beta 1, beta 3, beta 5, beta 6, and beta 8 (42). The beta 3-integrin subunit is involved in antagonism of local flow-dependent dilation in isolated coronary vessels when applied intraluminally (i.e., endothelium-mediated response) (36). This response, observed in isolated vessels, would serve to diminish the local ability of a vessel segment to respond to flow changes. In contrast, we found that remote responses to extraluminally applied RGD were amplified in higher flow states. Mogford et al. (34) showed that RGD-containing peptides directly dilate isolated arterioles from skeletal muscle when applied extraluminally, in an endothelium-independent manner. In contrast, we found that localized extraluminal application of RGD in situ was not locally vasoactive. There are several differences among these three studies (34, 36; present study): intraluminal vs. extraluminal application and isolated vs. in situ vessels. These studies illustrate that the local response is not always the network response, as documented elsewhere (6, 15, 16, 28) for adrenergic stimuli. Furthermore, the present study reinforces a previous finding (17) that purely remote responses occur. The physiological question is, of course, What are the relative roles of these responses in regulating blood flow distribution at the microvascular level? The mechanism of a network response could be viewed as the sum of local (or myogenic) responses, as purely remote responses, or, more likely, as combinations. For example, it is interesting to consider that integrins accessible from the extraluminal side of the blood vessels are involved in remote responses in a manner sensitive to the baseline flow state of an entire arteriolar network (present study), whereas intraluminal integrins are sensitive to the flow state locally (36). Importantly, these studies together illustrate the necessity of understanding the local components and the network components of vascular responses.

To determine whether the remote responses occurred via gap junction signaling, two nonspecific inhibitors of gap junction signaling were used (1, 39); the exact mechanism of action of halothane or high-osmolar sucrose on gap junctions is unknown. Although halothane has been shown to block gap junction connexin43 (33), it is also known to block Na channel activity, preventing depolarization (35). In a recent study, local dilation due to hyperosmolarity was linked to ATP-sensitive K channel activity (23). There is typically a ~50% dilation at the site of high-osmolar sucrose exposure (17; present study). Thus these agents may be nonspecifically blocking conducted responses by affecting membrane potential. Importantly, neither agent independently induced a remote response, which means that they did not block endogenous remote responses. In the present study, these gap junction inhibitors significantly diminished remote responses to L-Cys or RGD. We therefore conclude that these remote responses are transmitted via a conducted signaling pathway, as are the responses to L-Arg or NO donors.

Two nonspecific effects of L-Arg were examined in this study: the likely amino acid charge at neutral pH and system y+ transporter activity. The pI of the amino acid side branch is significantly related to the remote response, if we examine only those amino acids that are likely to be transported into cells of the vascular wall. However, this correlation may only reflect a relationship between response and binding characteristics and does not suggest that transport per se has induced these responses. Examination of all the amino acids shows no trend related to the pI values. The data are consistent with a specific effect of L-Arg to induce a remote dilation and of L-Cys to induce a remote constriction. For L-Arg, the remote dilation is dose dependent. We have other evidence that NO mimics the remote and network-preconditioning effects of L-Arg (13, 18). This suggests that exogenously applied L-Arg may enter the cell and be the substrate for NO formation and thus mediate both responses. Although the present study demonstrates that a remote response to L-Arg does not require L-Arg entry into the cells, this may only reflect that a specific effect of extracellular L-Arg is additionally involved. We cannot directly rule out for any of these amino acids that a separate conducted signal is related to amino acid entry.

The network-preconditioning data are clearly not related to the pI of the amino acids. We infer that the network effect is instead related to the y+ amino acid transporter, because two of the three y+ transport inhibitors blocked the response. L-Arg transport into endothelial cells or vascular smooth muscle cells is largely via the y+ amino acid transporter (4, 5). In addition, there is a large transport capacity for L-Lys or L-Cys via the y+ transporter (30, 43). L-Leu enters via the system L transporter, with a maximal rate of transport similar to that of L-Arg (4); therefore, accumulation of L-Leu would follow a time course similar to that for L-Arg or L-Cys. (The Michaelis-Menten constant for each of these amino acids is in the micromolar range.) Only those amino acids transported by the y+ carrier induced the network effect. Importantly, the remote responses initiated by these three amino acids were quite different. These data provide strong evidence that the remote response and the network-preconditioning signal are separate events. We now conclude that y+ transport activity is a significant component of the network-preconditioned conducted signal, which acts to upregulate the direct response. This does not mean that y+ activity alone induces the necessary signal but, instead, that amino acid uptake via system y+ or a subsequent intracellular event plays a significant role.

Three y+ transporter inhibitors were used. Sulfanilamide and phenformin each have other specific and nonspecific effects not shared by L-His. Sulfanilamide (a sulfonylurea) has been used historically as an antibiotic and as a loop diuretic to block Na/H exchange (12). However, it does so through its direct action to inhibit carbonic anhydrase (19). Thus sulfanilamide induces a metabolic acidosis (i.e., decrease of intracellular pH, with a likely concomitant depolarization). Phenformin (also a sulfonylurea) is an oral hypoglycemic agent that induces lactic acidosis over time and, more immediately, prevents Ca release from intracellular stores and binds to ATP-sensitive K channels (8, 29, 53). ATP-sensitive K channels contain a sulfonylurea receptor (24), which requires L-Arg (or L-Lys) for channel activity in small arterioles (26). NO directly activates ATP-sensitive K channels (23) [and Ca-sensitive K channels (37)] in some but not all tissues (50). NO donors can induce a remote dilation (10, 13, 18). Perhaps there is a common underlying mechanism for the responses with L-Arg and NO that are not related to L-Arg serving as the substrate for NO production but, instead, each acting through ATP-sensitive K channels.

The simplest general interpretation for our data is that there are separate signals mediating the remote response and the network-preconditioned response, yet both are clearly mediated by conducted signals. The general conclusion of many investigators regarding remote responses is that there is more than one response pathway. We report that L-Arg transport, although it may induce a remote response, is not required for a remote response. This is not in conflict with previous findings (10, 13, 19, 41) that NO is involved in remote responses under some experimental conditions. Instead, the data illustrate that the specific (and likely multiple) mechanisms involved in the time-dependent (10, 13) and response-specific (18, 41) conducted signals are not entirely understood.

In summary, this study shows that conducted signals are generated in the microcirculation by individual amino acids and by bioactive peptides. Remote constrictions or dilations occur that involve specific extracellular binding sites for L-Cys and L-Arg, respectively. Remote dilations are induced by an RGD binding site involving the alpha v-integrin subunit; the magnitude of this remote response is directly related to the baseline wall shear stress. A second conducted signal initiates a network-preconditioning conducted response. This is triggered by amino acids transported by the system y+ amino acid transporter and is not dependent on the observation of a remote response. Thus bioactive peptides induce two separate types of conducted signals.


    ACKNOWLEDGEMENTS

The author thanks Yoshia Makino for developing the velocity measurement software (development supported by The Whitaker Foundation); Eric Higashi, Pieter Van Horn, Lisa French, Keith Rivers, and Lisa Pudaserri for excellent technical assistance; and Drs. William M. Chilian and Richard J. Rivers for critical reading of the manuscript.


    FOOTNOTES

This study is supported by National Heart, Lung, and Blood Institute Grant HL-55492.

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Address for reprint requests: M. D. S. Frame, Dept. of Anesthesiology, Box 604, University of Rochester School of Medicine and Dentistry, 601 Elmwood Ave., Rochester, NY 14642.

Received 4 May 1998; accepted in final form 4 November 1998.


    REFERENCES
Top
Abstract
Introduction
Methods
Results
Discussion
References

1.   Barr, L., W. Berger, and M. M. Dewey. Electrical transmission at the nexus between smooth muscle cells. J. Gen. Physiol. 51: 347-369, 1968[Abstract/Free Full Text].

2.   Berg, B. R., and I. H. Sarelius. Functional capillary organization in striated muscle. Am. J. Physiol. 268 (Heart Circ. Physiol. 37): H1215-H1222, 1995[Abstract/Free Full Text].

4.   Bogle, R. G., A. R. Baydoun, J. D. Pearson, and G. E. Mann. Regulation of L-Arg transport and nitric oxide release in superfused porcine aortic endothelial cells. J. Physiol. (Lond.) 490: 229-241, 1996[Medline].

5.   Bussolati, O., R. Sala, A. Astorri, B. M. Rotoli, V. Dall'asta, and G. C. Gazzola. Characterization of amino acid transport in human endothelial cells. Am. J. Physiol. 265 (Cell Physiol. 34): C1006-C1014, 1993[Abstract/Free Full Text].

6.   Chilian, W. M., S. M. Layne, C. L. Eastham, and M. L. Marcus. Heterogeneous microvascular coronary alpha -adrenergic vasoconstriction. Circ. Res. 64: 376-388, 1989[Abstract/Free Full Text].

7.   Cokelet, G. R. Experimental determination of the average hematocrit of blood flowing in a vessel. Microvasc. Res. 7: 382-384, 1974[Medline].

8.   Corpus, V. M., S. M. Bressie, L. I. Stillwell, and G. M. Olins. Interaction of guanidinium compounds and K+ channel modulators with imidazoline binding sites in rabbit kidney. Eur. J. Pharmacol. 266: 197-200, 1994[Medline].

9.   D'Angelo, G., J. E. Mogford, G. E. Davis, M. J. Davis, and G. A. Meininger. Integrin-mediated reduction in vascular smooth muscle [Ca2+]i induced by RGD-containing peptide. Am. J. Physiol. 272 (Heart Circ. Physiol. 41): H2065-H2070, 1997[Abstract/Free Full Text].

10.   Doyle, M. P., and B. R. Duling. Acetylcholine induces conducted vasodilation by nitric oxide-dependent and -independent mechanisms. Am. J. Physiol. 272 (Heart Circ. Physiol. 41): H1364-H1371, 1997[Abstract/Free Full Text].

11.   Durante, W., L. Liao, and A. I. Schafer. Differential regulation of L-arginine transport and inducible NOS in cultured vascular smooth muscle cells. Am. J. Physiol. 268 (Heart Circ. Physiol. 37): H1158-H1164, 1995[Abstract/Free Full Text].

12.   Feit, P. W. Bumetanide---the way to its chemical structure. J. Clin. Pharmacol. 21: 531-536, 1981[Abstract].

13.   Frame, M. D. S. Nitric oxide donors induce a biphasic conducted dilation, altering subsequent upstream responsivity in small arterioles (Abstract). Circulation 96: I-148, 1997.

14.   Frame, M. D. S., and I. H. Sarelius. Arteriolar bifurcation angles vary with position and when flow is changed. Microvasc. Res. 46: 190-205, 1993[Medline].

15.   Frame, M. D. S., and I. H. Sarelius. Endothelial cell dilatory pathways link flow and wall shear stress in an intact arteriolar network. J. Appl. Physiol. 81: 2105-2114, 1996[Abstract/Free Full Text].

16.   Frame, M. D. S., and I. H. Sarelius. Regulation of capillary perfusion by small arterioles is spatially organized. Circ. Res. 73: 155-163, 1993[Abstract].

17.   Frame, M. D. S., and I. H. Sarelius. L-Arginine induced conducted signals alter upstream arteriolar responsivity to L-arginine. Circ. Res. 77: 695-701, 1995[Abstract/Free Full Text].

18.   Frame, M. D. S., and I. H. Sarelius. Vascular communication and endothelial cell function in the control of arteriolar flow distribution. Microcirculation 3: 233-235, 1996[Medline].

19.   Furesz, T. C., A. J. Moe, and C. H. Smith. Two cationic amino acid transport systems in human placental basal plasma membranes. Am. J. Physiol. 261 (Cell Physiol. 30): C246-C252, 1991[Abstract/Free Full Text].

20.   Henry, R. P., Y. Wang, and C. M. Wood. Carbonic anhydrase facilitates CO2 and NH3 transport across the sarcolemma of trout white muscle. Am. J. Physiol. 272 (Regulatory Integrative Comp. Physiol. 41): R1754-R1761, 1997[Abstract/Free Full Text].

21.   Hoshi, T., K. Sudo, and Y. Suzuki. Characteristics of changes in the intracellular potential associated with transport of neutral, dibasic and acidic amino acids in Triturus proximal tubule. Biochim. Biophys. Acta 448: 492-504, 1976[Medline].

22.   Inagaki, N., Y. Tsuura, N. Namba, K. Masuda, T. Gonoi, M. Horie, Y. Seino, M. Mizuta, and S. Seino. Cloning and functional characterization of a novel ATP-sensitive potassium channel ubiquitously expressed in rat tissues, including pancreatic islets, pituitary, skeletal muscle, and heart. J. Biol. Chem. 270: 5691-5694, 1995[Abstract/Free Full Text].

23.   Ishizaka, H., and L. Kuo. Endothelial ATP-sensitive potassium channels mediate coronary microvascular dilation to hyperosmolarity. Am. J. Physiol. 273 (Heart Circ. Physiol. 42): H104-H112, 1997[Abstract/Free Full Text].

24.   Islam, S. A., C. R. Wolf, M. S. Lennard, and M. J. E. Sternberg. A three-dimensional molecular template for substrates of human cytochrome P450 involved in debrisoquine 4-hydroxylation. Carcinogenesis 12: 2211-2219, 1991[Abstract/Free Full Text].

25.   Kilberg, M. S., and D. Haussinger (Editors). Mammalian Amino Acid Transport. New York: Plenum, 1992Kilberg, M. S., and D. Haussinger (Editors). Mammalian Amino Acid Transport. New York: Plenum, 1992.

26.   Kontos, H. A., and E. P. Wei. Cerebral arteriolar dilations by KATP channel activators need L-lysine or L-arginine. Am. J. Physiol. 274 (Heart Circ. Physiol. 43): H974-H981, 1998[Abstract/Free Full Text].

27.   Malandro, M. S., and M. S. Kilberg. Molecular biology of mammalian amino acid transporters. Annu. Rev. Biochem. 65: 305-336, 1996[Medline].

28.   Marshall, J. M. The influence of the sympathetic nervous system on individual vessels of the microcirculation of skeletal muscle of the rat. J. Physiol. (Lond.) 332: 169-186, 1982[Abstract/Free Full Text].

29.   McGuinness, M. E., and R. L. Talbert. Phenformin-induced lactic acidosis: a forgotten adverse drug reaction. Ann. Pharmacother. 27: 1183-1186, 1993[Abstract].

30.   McNamara, P. D., C. T. Rea, and S. Segal. Ion dependence of cystine and lysine uptake by rat renal brush-border membrane vesicles. Biochim. Biophys. Acta 1103: 101-108, 1992[Medline].

31.   McNamara, P. D., C. T. Rea, and S. Segal. Lysine uptake by rat renal brush-border membrane vesicles. Am. J. Physiol. 251 (Renal Fluid Electrolyte Physiol. 20): F734-F742, 1986.

32.   Miura, K., T. Ishii, Y. Sugita, and S. Bannai. Cystine uptake and glutathione level in endothelial cells exposed to oxidative stress. Am. J. Physiol. 262 (Cell Physiol. 31): C50-C58, 1992[Abstract/Free Full Text].

33.   Miyoshi, H., M. B. Boyle, L. B. Mackay, and R. E. Garfield. Gap junction currents in cultured muscle cells from human myometrium. Am. J. Obstet. Gynecol. 178: 588-593, 1998[Medline].

34.   Mogford, J. E., G. E. Davis, S. H. Platts, and G. A. Meininger. Vascular smooth muscle alpha vbeta 3 integrin mediates arteriolar vasodilation in response to RGD peptides. Circ. Res. 79: 821-826, 1996[Abstract/Free Full Text].

35.   Molliex, S., B. Dureuil, M. Aubier, G. Friedlander, J. M. Desmonts, and C. Clerici. Halothane decreases Na,K-ATPase, and Na channel activity in alveolar type II cells. Anesthesiology 88: 1606-1613, 1998[Medline].

36.   Muller, J. M., W. M. Chilian, and M. J. Davis. Integrin signaling transduces shear stress-dependent vasodilation of coronary arterioles. Circ. Res. 80: 320-326, 1997[Abstract/Free Full Text].

37.   Onoue, H., and Z. S. Katusic. Role of potassium channels in relaxations of isolated canine basilar arteries to acidosis. Stroke 28: 433-437, 1997[Abstract/Free Full Text].

38.   Pan, M., M. Wasa, and W. W. Souba. Tumor necrosis factor stimulates system XAG-transport activity in human endothelium. J. Surg. Res. 58: 659-664, 1995[Medline].

39.   Peracchia, C. Effects of the anesthetics heptanol, halothane and isoflurane on gap junction conductance in crayfish septate axons. J. Membr. Biol. 121: 67-78, 1991[Medline].

40.   Pries, A. R., T. W. Secomb, P. Gaehtgens, and J. F. Gross. Blood flow in microvascular networks: experiments and simulations. Circ. Res. 67: 826-834, 1990[Abstract/Free Full Text].

41.   Rivers, R. Conducted arteriolar dilations persist in the presence of nitroarginine. J. Cardiovasc. Pharmacol. 30: 309-312, 1997[Medline].

42.   Ruoslahti, E. RGD and other recognition sequences for integrins. Ann. Rev. Cell Dev. Biol. 12: 697-715, 1996[Medline].

43.   Samarzija, I., and E. Fromter. Electrophysiological analysis of rat renal sugar and amino acid transport. IV. Basic amino acids. Pflügers Arch. 393: 210-214, 1982[Medline].

44.   Sarelius, I. H. Microcirculation in striated muscle after acute reduction in systemic hematocrit. Respir. Physiol. 78: 7-17, 1989[Medline].

45.   Sarelius, I. H. Cell flow path influences transit time through striated muscle capillaries. Am. J. Physiol. 250 (Heart Circ. Physiol. 19): H899-H907, 1986.

46.   Sarelius, I. H., D. N. Damon, and B. R. Duling. Microvascular adaptations during maturation of striated muscle. Am. J. Physiol. 241 (Heart Circ. Physiol. 10): H317-H324, 1981.

47.   Sarelius, I. H., and B. R. Duling. Direct measurement of microvessel hematocrit, red cell flux, velocity, and transit time. Am. J. Physiol. 243 (Heart Circ. Physiol. 12): H1018-H1026, 1982[Abstract/Free Full Text].

48.   Segal, S., P. D. McNamara, and L. M. Pepe. Transport of cystine and dibasic amino acids in renal brush border vesicles. Science 197: 169-171, 1977[Abstract/Free Full Text].

49.   Snedecor, G. W., and W. G. Cochran. Statistical Methods (6th ed.). Ames: Iowa State University Press, 1967.

50.   Sobey, C. G., and F. M. Faraci. Effect of nitric oxide and potassium channel agonists and inhibitors on basilar artery diameter. Am. J. Physiol. 272 (Heart Circ. Physiol. 41): H256-H262, 1997[Abstract/Free Full Text].

51.   Sobrevia, L., P. Cesare, D. L. Yudilevich, and G. E. Mann. Diabetes-induced activation of system y+ and nitric oxide synthase in human endothelial cells: association with membrane hyperpolarization. J. Physiol. (Lond.) 489: 183-192, 1995[Medline].

51a.   Stryer, L. (Editor). Biochemistry (2nd ed.). San Francisco, CA: Freeman, 1975Stryer, L. (Editor). Biochemistry (2nd ed.). San Francisco, CA: Freeman, 1975.

52.   Sweeney, T. E., and I. H. Sarelius. Arteriolar control of capillary cell flow in striated muscle. Circ. Res. 64: 112-120, 1989