Lymphatic pump-conduit duality: contraction of postnodal lymphatic vessels inhibits passive flow

Am J Physiol Heart Circ Physiol 296: H662 H668, 2009. First published January 2, 2009; doi:10.1152/ajpheart.00322.2008. Lymphatic pump-conduit duality: contraction of postnodal lymphatic vessels inhibits passive flow Christopher M. Quick, Bruce L. Ngo, Arun M. Venugopal, and Randolph H. Stewart Michael E. DeBakey Institute Texas A&M University, College Station, Texas Submitted 27 March 2008; accepted in final form 29 December 2008 Quick CM, Ngo BL, Venugopal AM, Stewart RH. Lymphatic pump-conduit duality: contraction of postnodal lymphatic vessels inhibits passive flow. Am J Physiol Heart Circ Physiol 296: H662 H668, 2009. First published January 2, 2009; doi:10.1152/ajpheart.00322.2008. Lymphangions, the segments of lymphatic vessels between valves, exhibit structural characteristics in common with both ventricles and arteries. Although once viewed as passive conduits like arteries, it has become well established that lymphangions can actively pump lymph against an axial pressure gradient from low-pressure tissues to the great veins of the neck. A recently reported mathematical model, based on fundamental principles, predicted that lymphangions can transition from pump to conduit behavior when outlet pressure falls below inlet pressure. In this case, the axial pressure gradient becomes the major source of energy for the propulsion of lymph, despite the presence of cyclical contraction. In fact, flow is augmented when cyclical contractions are abolished. We therefore used an in vitro preparation to confirm these findings and to test the hypothesis that lymphangion contraction inhibits flow when outlet pressure falls below inlet pressure. Bovine postnodal mesenteric lymphatic vessels harvested from an abattoir were subjected to an inlet pressure of 5.0 cmh 2 O and an outlet pressure that decreased from 6.5 to 3.5 cmh 2 O under control conditions, stimulated with U-46619 (a thromboxane analog) and relaxed with calcium-free solution. Under control conditions, lymphatic flow markedly increased as outlet pressure fell below inlet pressure. In this case, the slopes of the flow versus axial pressure gradient increased with calcium-free conditions (61%, n 8, P 0.016) and decreased with U-46619 stimulation (21%, n 5, P 0.033). Our findings indicate that the stimulation of lymphatic contractility does indeed inhibit lymphatic flow when vessels act like conduits. lymphangion conductance; contractility SINCE LYMPHATIC VESSELS are structured like veins, having a relatively thin wall and valves, and eventually empty into the great veins of the neck, they were originally viewed as conduit passive vessels that dissipate energy as the axial pressure gradient propels lymph (4). It was later shown that, like veins, the external compression caused by the contraction of skeletal muscle, respiration, or intestinal peristalsis can extrinsically pump lymph (1, 4, 7, 17). Although lymphatic vessels were first observed to spontaneously contract in the 19th century, it was not until negative interstitial pressures were reported that lympatic vessels were shown to act as pump chambers that add energy to lymph as it is propelled against a pressure gradient (1). In fact, lymphangions, the segments of vessel between unidirectional valves (12), contract with a frequency of 1 15 min 1 and can expel more than half their luminal volume per contraction (1, 17). The active, intrinsic pumping of lymph is well established for a number of organs where the interstitial pressure is normally lower than central venous pressure (1, 4, 17). Without sources of consistent extrinsic compression, vessels in these organ systems are conventionally recognized as active lymph pumps. Potential for reversal of the normal axial pressure gradient. There are a number of conditions, however, that can cause the inlet pressure of a lymphangion to rise above outlet pressure, reversing the normal axial pressure gradient (15). First, posture, especially in humans, can create a significant hydrostatic pressure gradient. Limb elevation above the level of the great veins of the neck in particular can cause a persistent reversal of the axial pressure gradient. Second, interstitial edema, the accumulation of excess fluid in the interstitial space, can cause a significant increase in interstitial pressure. Venous hypertension has been shown to raise intestinal interstitial pressure as high as 7 mmhg (14). Vena caval occlusion has been shown to increase cysterna chili pressure as high as 6 mmhg (18) and hepatic interstitial pressure as high as 20 mmhg (10). Intestinal interstitial pressure has been even estimated to increase by 4 mmhg during lipid absorption (6). Third, an external compression of upstream lymphangions due to peristalsis, respiration, or contraction of skeletal muscle could cause the inlet pressure of a downstream lymphatic vessel to rise transiently (7, 17). In fact, a simple contraction of an upstream lymphangion can cause the pressure gradient across downstream lymphangions to vary. In each case, the reversal of the normal axial pressure gradient has the potential to enhance lymph flow and make active lymphatic pumping unnecessary. Pump-conduit duality of lymphangions. In any lymphangion, two sources of energy are available to propel lymph: 1) lymphangion pumping, when inlet pressure is lower than outlet pressure; and 2) the axial pressure gradient, when inlet pressure is higher than outlet pressure. In the first case, lymphangions act primarily as pumps that provide the energy to propel lymph. In the second case, flow is passive, and lymphangions act primarily as a conduit that dissipates energy. To address the dual nature of lymphangions as both pumps and conduits, a mathematical model was developed (15). Based on fundamental principles, the model predicted that when inlet pressure is lower than outlet pressure, lymphangion contraction was necessary to propel lymph. Acting as pumps, lymph flow only occurred during systole (i.e., when the lymphangions were contracting). Because lymphangion contraction was the only source of lymph propulsion, lymph flow ceased when lym- Address for reprint requests and other correspondence: C. M. Quick, Michael E. DeBakey Inst., Texas A&M Univ., TAMU 4466, College Station, Texas 77843-4466 (e-mail: cquick@tamu.edu). 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. Section 1734 solely to indicate this fact. H662 0363-6135/09 $8.00 Copyright 2009 the American Physiological Society http://www.ajpheart.org

phangions relaxed in diastole. When the inlet pressure was theoretically raised above the outlet pressure, however, the lymphangions were predicted to act as conduits; that is, the lymphangion valves were continuously forced open throughout the contraction cycle. In this case, the only source of energy to propel lymph was provided by the axial pressure gradient. Not only was there passive flow in diastole, but the flow actually decreased transiently during systole. The reason for this could be identified lymphangion contraction increased the resistance to passive flow. Purpose of the present study. Two nonintuitive implications of the pump-conduit duality of lymphangions arose from theoretical study (15), especially when inlet pressure is higher than outlet pressure. First, making lymphatic vessels completely passive by relaxing lymphatic muscle can increase lymph flow. Second, stimulating lymphatic contraction can decrease lymph flow. However, the model predictions of conduit behavior have not yet been validated experimentally. Therefore, the purpose of this study is to use an in vitro lymphatic vessel preparation to confirm the findings of the mathematical model of Quick et al. (15) and to test the hypothesis that, when the inlet pressure is higher than the outlet pressure, lymphatic muscle relaxation increases lymph flow and the stimulation of lymphatic muscle contraction inhibits flow. METHODS Harvesting lymphatic vessels. Bovine postnodal mesenteric lymphatic vessels were isolated from freshly slaughtered cattle at a local abattoir. Physiological saline with Evans blue dye was injected into mesenteric lymph nodes to visualize conducting lymphatic vessels. Postnodal lymphatic vessels ranging from 2 4 mm outer diameter were ligated at the downstream end and then dissected free of the surrounding tissue. Vessels were then transported in an insulated container maintained at 30 C and filled with 1% albumin-supplemented physiological saline solution (APSS), with ph buffered to 7.4, containing 145 mm NaCl, 4.7 mm KCl, 2 mm CaCl 2, 1.17 mm MgSO 4, 1.2 mm NaH 2PO 4, 5 mm dextrose, 2 mm sodium pyruvate, 20 M EDTA, 3 mm MOPS buffer, and 1% wt/vol bovine serum albumin. Vessels were instrumented within 6 h of harvest. In vitro lymphatic vessel preparation. The vessel bath was set up (Fig. 1) as previously reported (19). All lines and tubing were flushed with APSS to remove air pockets and bubbles. Differential pressure transducers (PX26-001GV, Omega, Stamford, CT) were used to measure fluid pressure. Inlet pressure was measured relative to atmospheric pressure, and vessel outlet pressure was measured relative to Fig. 1. Experimental preparation used to measure the radius (r), inlet pressure (P in), outlet pressure (P out), and flow in an isolated lymphatic segment. The bath depth of 1 cm set the external hydrostatic pressure at 1 cmh 2O. Heights of water columns at the inlet and outlet were adjusted to set inlet and outlet pressures. H663 vessel inlet pressure. Flow was determined with a custom pressure gradient flowmeter used previously (19), consisting of a differential pressure transducer measuring the pressure drop across a short, horizontal length of tubing of fixed resistance. To determine the relationship of the pressure gradient to flow, the luminal fluid flowing through it was collected over a range of flows from 2 to 13 ml/min. The resulting axial pressure gradient was linearly correlated with the measured flow rates (r 2 0.99). The conversion factor of 2.08 was subsequently used to convert the measured axial pressure gradient to flow rate. Lymphatic vessel external diameters were monitored in real time using a monochrome CCD camera (ST-XC50, Sony Electronics, Park Ridge, NJ) and a custom video dimensional analysis program (LabVIEW 7.0, National Instruments, Austin, TX) that measures the distance between high-contrast pixels at the edges of the vessel (15). Image and pressure readings were then digitized at 25 samples/s (NI PCI-6036E and NI PCI-1410, National Instruments). All pressure transducers were calibrated before each experiment. Protocol 1: determining pressure-flow relationship of lymphatic vessels with relaxed smooth muscle. Lymphatic vessels (n 8) were cannulated on both ends and secured to a tubular organ bath. The APSS of the bath and lumen was infused with room air via a membrane oxygenator (OXR, Living Systems Instrumentation, Burlington, VT) and warmed to 37.0 C with a thermoregulator (Lauda E-200, Brinkman Instruments, Westbury, NY). External vessel pressure was held constant at 1 cmh 2O by setting the bath fluid level 1 cm above the vessel. Luminal inlet pressure was maintained at 5 cmh 2O by adjusting the height of the fluid column connected to the inlet port. Luminal outlet pressure was decreased from 6.5 to 3.5 cmh 2Oby adjusting the height of the fluid column connected to the outlet port. Vessels were checked for the presence of at least two valves by confirming a mean positive flow when the outlet pressure was raised above the inlet pressure. Viability was assessed by spontaneous, phasic contractions after the vessel was instrumented. Lymphatic vessels that did not spontaneously contract were subjected to a transient increase in pressure to induce contractions. Vessels that failed to spontaneously pump within 15 min were discarded. Vessels were allowed to acclimate after each change in outlet pressure for at least 1 min. From previous experience (15), this time period was known to be long enough to ensure that vessels reached steady-state. Resulting data were averaged over 1 min. The bath and luminal solutions were then switched from APSS to Ca 2 -free APSS and allowed to equilibrate until the vessels were fully relaxed (5, 9). Ca 2 -free APSS was prepared using the same concentrations as APSS; however, CaCl 2 was not added and the concentration of EDTA was increased to 2 mm. Once the vessels ceased to contract, the data were obtained using the same methods as the control. Protocol 2: determining pressure-flow relationship of lymphatic vessels with stimulated smooth muscle. Lymphatic vessels (n 5) were treated with APSS and APSS containing 10 7 M U-46619 (Cayman Chemical, Ann Arbor, MI), a relatively stable analog of thromboxane. U-46619 is a potent selective mimetic of thromboxane A 2 (3) and has been shown to increase tone (i.e., decrease diameter) in lymphatic vessel segments (8). APSS solutions with 10 7 M U-46619 were prepared by adding 1.34 l of stock solution to 1 liter of APSS. The stock solutions consisted of the drug suspended in 100% ethanol at a concentration of 5 g/ml. The ethanol concentration in the vessel bath and lumen was maintained below 0.003%, which was low enough to have no significant effect on the lymphatic vessels (13). The bath and luminal solutions initially contained APSS for instrumentation, and the lymphatic vessels were tested for viability as in protocol 1. The order of experimentation with APSS or APSS with U-46619 was randomly determined. The axial pressure gradient was then gradually stepped from approximately 1.5 to 1.5 cmh 2O using the same range of outlet pressures described for protocol 1. The bath and luminal solutions were then switched to the alternate solution, and the procedure was repeated.

H664 Data analysis. The mean axial pressure gradient (inlet pressure minus outlet pressure), mean flow, maximum diameter, stroke area, and contraction frequency for each measurement period were calculated. The mean flow was plotted as a function of the axial pressure gradient. The flow rates for the positive and negative axial pressure gradients for each the Ca 2 -free group and the U-46619 group were compared with the control data using a custom analysis of covariance model, and the estimated marginal means were analyzed at both negative and positive axial pressure gradients with Bonferroni post hoc analysis. For positive axial pressure gradients, the slopes of flow-axial pressure gradient relationship served as an index of hydraulic conductance. The slopes of the flow-axial pressure gradient relationships and end-diastolic diameters were assessed for each vessel. The control group was compared with the Ca 2 -free group and the U-46619 group using paired t-tests. P values 0.05 were considered significant. RESULTS Fig. 2. Instantaneous flows and diameters of a bovine mesenteric vessel, exhibiting a transition from pump to conduit behavior as the axial pressure gradient reverses, in both control (top) and relaxed (bottom) conditions. Flow (A) and diameter (B) of a vessel actively propelling lymph against an axial pressure gradient (inlet pressure 5 cmh 2O, and outlet pressure 6 cmh 2O) are shown. Flow transiently increases when the vessel contracts, consistent with pump behavior. Because the vessel does not contract uniformly (i.e., the contraction wave takes time to travel the entire length of the lymphangion), part of the vessel started to contract and, thus, initiates increased flow before contraction of the vessel segment at which diameter was being measured. Flow (C) and diameter (D) of same vessel passively resisting flow (inlet pressure 5 cmh 2O, and outlet pressure 4 cmh 2O) are shown. Flow transiently decreases when vessel contracts, and fluid flow rates are positive throughout the contraction cycle (even during diastole), consistent with conduit behavior. Flow (E) and diameter (F) of the same vessel exposed to Ca 2 -free APSS are shown. The vessel does not exhibit contraction, nor does it propel lymph against an axial pressure gradient (inlet pressure 5 cmh 2O, and outlet pressure 6 cmh 2O). Flow (G) and diameter (H) of same vessel passively resisting flow (inlet pressure 5 cmh 2O, and outlet pressure 4 cmh 2O) are shown. Relaxation decreases flow when the vessel is acting as a pump (E vs. A) but increases mean flow when the vessels is acting as a conduit (G vs. C). Lymphatic pump inhibition. In all cases, there was a marked increase in mean lymph flow with a switch from a negative to a positive axial pressure gradient (Figs. 2, 3, 4, and 5). When lymphangions acted as pumps (i.e., when inlet pressure was less than outlet pressure), lymph flow only occurred when the vessels contracted (Fig. 2, A and B). (Lymphangions exhibited nonuniform contraction, causing a small time shift between outlet flow and the diameter measured at a single location.) When lymphangions acted as a conduit (i.e., when inlet pressure was greater than outlet pressure), lymph flow occurred throughout the contraction cycle but decreased transiently during contraction (Fig. 2, C and D). When the vessel was exposed to Ca 2 -free solution, the vessel contractions were abolished (Fig. 2, F and H). End-diastolic diameters were significantly increased compared with spontaneously contracting vessels in APSS (3.13 0.20 vs. 2.63 0.24 mm, n 8, P 0.012). A flow through the vessel with abolished contractions only occurred when the axial pressure gradient was positive (Fig. 2G). Lymphatic vessels in Ca 2 -free APSS had significantly higher rates of fluid flow when the inlet pressure was raised above the outlet pressure (Fig. 2, G vs. C, and Fig. 4A, right). Under positive axial pressure gradients, the slopes of the flow-axial pressure gradient relationships increased 61% compared with control (3.81 0.57 vs. 2.37 0.61 ml cmh 2 O 1 min 1, n 8, P 0.016). Changes in diameters and flows were persistent and not dependent on the duration that the pressure steps were maintained. Lymphatic pump stimulation. U-46619 was found to induce smooth muscle contraction in lymphatic vessels (n 5), as demonstrated by a 21% decrease in end-diastolic diameters (2.01 0.32 vs. 2.55 0.25 mm, n 5, P 0.033). When compared with control vessels, the lymphatic vessels stimulated with U-46619 did not exhibit significant changes in the flowpressure relationship when actively pumping against a negative axial pressure gradient (Fig. 4B, left). When the axial pressure gradient was positive, however, lymphatic vessels treated with U-46619 exhibited a significant attenuation of fluid flow compared with control vessels (Fig. 3, G vs. C, and Fig. 4B, right). The slopes of the flow-axial pressure gradient relationships under positive gradients were found to be significantly decreased in the presence of U-46619 (1.06 0.40 vs. 3.24 0.27 ml cmh 2 O 1 min 1, n 5, P 0.006). End-diastolic diameters (Fig. 5A) were decreased compared with those of control vessels over all axial pressure gradients when exposed to U-46619. Changes in diameters and flows were persistent and not dependent on the duration that the pressure steps were maintained. The

H665 frequency of contraction was not significantly affected by U-46619 exposure (Fig. 5B). DISCUSSION Transition from pumps to conduits. The present work indicates that spontaneously contracting lymphangions transition from pumps to conduits when the outlet pressure falls below the inlet pressure. In the control case, when outlet pressure is Fig. 3. Instantaneous flows and diameters of a bovine mesenteric vessel, exhibiting a transition from pump to conduit behavior as the axial pressure gradient reverses, both in control (top) and stimulated (bottom) conditions. Flow (A) and diameter (B) of a vessel actively propelling lymph against an axial pressure gradient (inlet pressure 5 cmh 2O, and outlet pressure 6 cmh 2O) are shown. Flow transiently increases when the vessel contracts, consistent with pump behavior. Flow (C) and diameter (D) of same vessel passively resisting flow (inlet pressure 5 cmh 2O, and outlet pressure 4 cmh 2O) are shown. Flow transiently decreases when vessel contracts, and fluid flow rates are positive throughout the contraction cycle (even during diastole), consistent with conduit behavior. Flow (E) and diameter (F) of the same vessel stimulated with U-46619, a thromboxane analog, are shown. Lymph is propelled against an axial pressure gradient (inlet pressure 5 cmh 2O, and outlet pressure 6 cmh 2O) when vessel contracts. Flow (G) and diameter (H) of same vessel passively resisting flow (inlet pressure 5 cmh 2O, and outlet pressure 4 cmh 2O) are shown. Lymphatic muscle stimulation inhibits flow when vessel is acting as a conduit (G vs. C). much higher than the inlet pressure, there was a relatively small amount of flow (Fig. 4A, left). In this case, lymph was propelled only by active lymphangion contraction. Because upstream valves closed during systole and downstream valves closed during diastole, there was a period where lymph flow stopped (Fig. 2A). When the outlet pressure was reduced below the inlet pressure, there was a marked increase in flow (Fig. 2C, and Fig. 4A, right). In fact, the axial pressure gradient became Fig. 4. Lymphatic vessel flow as a function of axial pressure gradient when inlet pressure was 5 cmh 2O and outlet pressure was altered. A: spontaneously contracting lymphangions (n 8) in albumin-supplemental physiological saline solution (APSS) and lymphangions made passive in Ca 2 -free APSS. Vessels under control conditions were able to maintain a positive flow when the inlet pressure was lower than the outlet pressure (left), but vessels exposed to Ca 2 -free conditions were unable to pump fluid. In both cases, fluid flow increases significantly as the pressure gradient reverses (right). Abolishing contraction with Ca 2 -free APSS significantly increases fluid flow. B: lymphatic vessels stimulated with U-46619 APSS (a thromboxane analog) (n 5) did not exhibit a significant difference in fluid flow when inlet pressure was much lower than outlet pressure (left) compared with control. When the axial pressure gradient reversed (right), stimulation with U-46619 APSS significantly attenuated flow.

H666 Fig. 5. Critical factors affecting lymphangion pumping were measured for vessels stimulated by U-46619 APSS vs. control conditions (n 5). Variables were plotted as a function of axial pressure gradient. A: end-diastolic diameter of the lymphatic vessels decreased over all axial pressure gradients when exposed to U-46619. B: frequency of contraction was not altered significantly during U-46619 exposure. C: stroke area (i.e., stroke volume per length) was decreased with U-46619 exposure, following the trend in end-diastolic diameters (A). the dominant force propelling lymph. Although the valves were completely competent and the lymphangions continued to contract (Fig. 2D), the axial pressure gradient forced the valves open throughout the contraction cycle. The opening of valves allowed for a continuous flow, which peaked in diastole (Fig. 2C). Continuous flow is a defining characteristic of conduit behavior (15). In this case, the energy to propel lymph did not come from the contraction of the lymphangions but from the potential energy stored as inlet hydrostatic pressure. Accordingly, when the lymphangions contracted, the resistance to flow increased, and the flow was transiently inhibited (Fig. 2, C and D). The resulting pressure-flow relationship (Fig. 4) validates the predictions that arose from a previously reported mathematical model (15). Stimulating and inhibiting lymphatic muscle when lymphangions are conduits. The fundamental pump-conduit duality of lymphangions causes the response to altering the lymphangion contractility to have opposite effects depending on whether lymphangion inlet pressure is higher or lower than outlet pressure. When lymphangions pumped against an axial pressure gradient, the relaxation of lymphatic muscle stopped flow (Fig. 2E, and Fig. 4A, left), reflecting the role of the lymphangion as a pump. When inlet pressure was higher than outlet pressure, the amplitude of the contractions can actually increase in the control condition (Fig. 2D). However, the small increase in stroke volume does not account for the increase in flow. In fact, the relaxation of the lymphangion muscle increased the mean flow (Fig. 2, G vs. C, and Fig. 4A, right), reflecting the alternative role of lymphangion as a conduit. In short, when lymphangions are acting as pumps, relaxation makes them worse pumps; when they are acting as conduits, relaxation decreases resistance and makes them better conduits. Similarly, stimulating lymphatic muscle with a thromboxane analog inhibited flow when the lymphangion inlet pressure was higher than outlet pressure (Figs. 3, G vs. C, and Fig. 4B, right); that is, when lymphangions are acting as conduits, stimulation increases resistance and makes them less effective conduits. Reproducing the theoretical mathematical model (15) that motivated the present work would require detailing an extensive set of assumptions and is beyond the scope of the present work. The present experimental results, however, are fully consistent with the fundamental implications of this foundational model and can be approximated by adjusting the indexes of contractility and tone in a simpler algebraic approximation detailed elsewhere (16). Affect of stimulating muscular contraction on lymphangion function. Stimulation of lymphatic muscle contraction by a thromboxane analog did not increase the slope of the flowpressure relationship when the lymphangions were acting as pumps (i.e., when inlet pressure was lower than outlet pressure; Fig. 4B, left). Although the decrease in end-diastolic diameter (Fig. 5A) indicated an increase in lymphangion muscle contraction, the stroke volume actually decreased (Fig. 5C). The expectation to the contrary arises primarily from relying on ventricular analogies to conceptualize lymphangion pump function. Whereas increasing ventricular contractility only increases the slope of the end-systolic pressure volume relationship, increasing lymphangion contractility can also increase the

slope of the end-diastolic pressure-volume relationship (11). In this case, increasing diastolic stiffness inhibits lymphangion filing, a phenomenon analogous to chronic diastolic dysfunction in ventricles. Complementary views of lymphangion pump-conduit duality. It is common practice for in vitro preparations either to set pressures and measure flow or to set flow and measure pressures. In reality, neither axial pressure gradient nor flow has set values in vivo, since lymphangion contraction will affect both variables. In the present work, the pressure gradient was set and the resulting mean flow was measured. Although this strict pump-conduit duality, where inlet pressure is above or below outlet pressure throughout the contraction cycle, is likely rare in whole animal preparations, our findings can provide insight into in vivo studies. For instance, Benoit (2) reported that in a preparation that increased transmicrovascular flux enough to yield a 25-fold increase in thoracic duct flow, the terminal lymphangion pump flow calculated from stroke volume times contraction frequency only increased by 3-fold. Although this would seem to violate the principle of the conservation of mass, it was previously shown that when lymphangions act like conduits, a calculated pump flow underestimates true flow (15). We can thus infer that transmicrovascular flux outstripped the ability of these small lymphangions to pump, and they thus were acting as conduits. Possibility for a lymphatic network to exhibit both pump and conduit behavior. The behavior of lymphangions in vivo is complicated by the possibility that the axial pressure gradient can vary with time. For instance, external compression of upstream lymphangions due to peristalsis, respiration, or contraction of skeletal muscle could cause the inlet pressure of a downstream lymphatic vessel to rise transiently (7, 17). In fact, a simple contraction of an upstream lymphangion can cause the pressure gradient across downstream lymphangions to transiently reverse. Such variations in local pressure gradients can ensure that a lymphatic network can contain some lymphangions that act as pumps and others that act as conduits. The mathematical model (15) suggested there was a range of axial pressure gradients between 0 0.25 mmhg in which the lymphangion valves are open during some portion of diastole, indicating partial conduit behavior. The present experimental work complements previously reported theoretical work (15) to provide the first step in understanding how lymphangion pump-conduit duality can affect the function of an entire lymphatic network. Implications of a lymphangion pump-conduit switch. Although lymphangions perform a number of critical functions, two implications to lymph flow in particular bear inspection. First, when transmicrovascular flux is great enough to cause transition to a conduit, spontaneous lymphangion contraction inhibits flow (Fig. 2, G vs. C, and Fig. 4A). In high-flow edemagenic conditions where inlet pressure is higher than outlet pressure (6, 10, 14, 18), lymphangion pumping would likely decrease lymph flow and exacerbate edema. The present work suggests that if vessels were to relax in high-flow conditions, edema would be ameliorated. Although the reported shear-mediated relaxation (5, 9) was not evident in the present study, the present work suggests it may be a normal physiological response that acts as a novel antiedema mechanism. Second, in response to sustained increases in transmicrovascular flux, one might expect changes resulting from lymphatic remodeling that would promote passive flow, such as increased vessel diameter, decreased contractility, and decreased contraction frequency. Two implications to clinical research can also derive from the possibility that lymphangions transition from a pump into a conduit in edematous conditions. First, a strategy for resolving edema based on increasing lymphangion contractility can be flawed. When edema is great enough to raise interstitial pressure above central venous pressure, or if a limb is raised above the great veins of the neck, increasing lymphatic contractility could potentially decrease lymph flow. Second, the possibility exists that the transition from a pump into a conduit may lead to instability. Once a lymphangion becomes a conduit, it cannot actively pump lymph unless the inlet pressure falls below the outlet pressure. Lymphangion pumping, however, is necessary for the inlet pressure to fall below the outlet pressure. The transition to a conduit, therefore, has the potential to delay rather than promote edema resolution. This potential becomes even more likely with acute flow-induced relaxation and chronic vascular remodeling. Although this potential is beyond the scope of the present work, lymphatic pump-conduit duality provides a new conceptual tool to address interstitial edema. ACKNOWLEDGMENTS H667 We thank our research assistant Melinda Webster for procuring viable lymphatic vessels and ensuring proper setup of the vessel bath. GRANTS This work was supported by National Heart, Lung, and Blood Institute Grant K25-HL-070608 (to C. M. Quick) and American Heart Association Grants 0565116Y (to C. M. Quick) and 0365127Y (to R. H. Stewart). REFERENCES 1. Aukland K, Reed RK. Interstitial-lymphatic mechanisms in the control of extracellular fluid volume. Physiol Rev 73: 1 78, 1993. 2. Benoit JN. Relationships between lymphatic pump flow and total lymph flow in the small intestine. Am J Physiol Heart Circ Physiol 261: H1970 H1978, 1991. 3. Coleman RA, Humphrey PP, Kennedy I, Levy GP, Lumley P. Comparison of the actions of U-46619, a prostaglandin H 2-analogue, with those of prostaglandin H 2 and thromboxane A 2 on some isolated smooth muscle preparations. Br J Pharmacol 73: 773 778, 1981. 4. Gashev AA, Zawieja DC. Physiology of human lymphatic contractility: a historical perspective. Lymphology 34: 124 134, 2001. 5. Gashev AA, Davis MJ, Zawieja DC. Inhibition of the active lymph pump by flow in rat mesenteric lymphatics and thoracic duct. J Physiol 540: 1023 1037, 2002. 6. Granger DN, Korthuis RJ, Kvietys PR, Tso P. Intestinal microvascular exchange during lipid absorption. Am J Physiol Gastrointest Liver Physiol 255: G690 G695, 1988. 7. Ikomi F, Schmid-Schonbein GW. Lymph pump mechanics in the rabbit hind leg. Am J Physiol Heart Circ Physiol 271: H173 H183, 1996. 8. Johnston MG, Gordon JL. Regulation of lymphatic contractility by arachidonate metabolites. Nature 293: 294 297, 1981. 9. Koller A, Mizuno R, Kaley G. Flow reduces the amplitude and increases the frequency of lymphatic vasomotion: role of endothelial prostanoids. Am J Physiol Regul Integr Comp Physiol 277: R1683 R1689, 1999. 10. Laine GA, Hall JT, Laine SH, Granger J. Transsinusoidal fluid dynamics in canine liver during venous hypertension. Circ Res 45: 317 323, 1979. 11. Meisner JK, Stewart RH, Laine GA, Quick CM. Lymphatic vessels transition to state of summation above a critical contraction frequency. Am J Physiol Regul Integr Comp Physiol 293: R200 R208, 2007. 12. Mislin H. Active contractility of the lymphangion and coordination of lymphangion chains. Experientia 32: 820 822, 1976. 13. Mizuno R, Koller A, Kaley G. Regulation of the vasomotor activity of lymph microvessels by nitric oxide and prostaglandins. Am J Physiol Regul Integr Comp Physiol 274: R790 R796, 1998. 14. Mortillaro NA, Taylor AE. Interstitial fluid pressure of ileum measured from chronically implanted polyethylene capsules. Am J Physiol Heart Circ Physiol 257: H62 H69, 1989.

H668 15. Quick CM, Venugopal AM, Gashev AA, Zawieja DC, Stewart RH. Intrinsic pump-conduit behavior of lymphangions. Am J Physiol Regul Integr Comp Physiol 292: R1510 R1518, 2007. 16. Quick CM, Venugopal AM, Dongaonkar RM, Laine GA, Stewart RH. First-order approximation for the pressure-flow relationship of spontaneously contracting lymphangions. Am J Physiol Heart Circ Physiol 294: H2144 H2149, 2008. 17. Schmid-Schönbein GW. Microlymphatics and lymph flow. Physiol Rev 70: 987 1028, 1990. 18. Stewart RH, Laine GA. Flow in lymphatic networks: interaction between hepatic and intestinal lymph vessels. Microcirculation 8: 221 227, 2001. 19. Venugopal AM, Stewart RH, Laine GA, Dongaonkar RM, Quick CM. Lymphangion coordination minimally affects mean flow in lymphatic vessels. Am J Physiol Heart Circ Physiol 293: H1183 H1189, 2007.