Regulatory mechanisms in lymphatic vessel contraction under normal and inflammatory conditions

Transcription

1 Pathophysiology 17 (2010) Review Regulatory mechanisms in lymphatic vessel contraction under normal and inflammatory conditions Pierre-Yves von der Weid a,, Mariappan Muthuchamy b, a Department of Physiology & Pharmacology, Inflammation Research Network and Smooth Muscle Research Group, Snyder Institute of Infection, Immunity & Inflammation and Libin Cardiovascular Institute of Alberta, University of Calgary, Calgary, Alberta, Canada T2N 4N1 b Department of Systems Biology and Translational Medicine, Cardiovascular Research Institute Division of Lymphatic Biology, Texas A&M Health Science Center College of Medicine, College Station, TX , United States Received 6 April 2009; received in revised form 10 June 2009; accepted 23 October 2009 Abstract The lymphatic system is composed of a dense network of lymphatic vessels, which are critical components of physiological interstitial fluid transport. These vessels possess intrinsic contractile properties providing the driving force for the fluid to be drained away from the tissues and propelled, as lymph, back into the bloodstream. Lymphatic pumping is also important to carry immune cells, bacteria, macromolecules, viruses and their products to and through lymph nodes, the other component of the lymphatic system, to initiate the adaptive immune response. In addition, among the many circulating mediators known to modulate lymphatic contractile activity and thus lymph flow, mediators of inflammation have potent excitatory or inhibitory actions. The involvement of lymphatic vessels in edema resolution, immune cell trafficking and their sensitivity to inflammatory mediators make them pivotal players of the inflammation process. The ability of lymphatic vessels to generate and regulate lymph flow is provided by the lymphatic muscle present in the vessels wall. Although molecular studies investigating the mechanisms of lymphatic vessel contraction are still very limited, recent findings suggest that lymphatic pumping requires complicated muscle activities that have similarities to those seen in both the heart (striated muscle) and blood vessels (smooth muscle). This review article focuses on presenting and discussing the mechanisms that regulate lymphatic vessel contraction under normal and pathophysiological states, specifically pertaining to inflammatory conditions Published by Elsevier Ireland Ltd. Keywords: Lymphatic pumping; Lymph flow; Lymphatic muscle; Inflammatory mediators; Intestinal inflammation Contents 1. Introduction to the lymphatic system Uniqueness of lymphatic muscle Regulation of smooth muscle contraction Regulation of striated muscle contraction Regulation of lymphatic muscle contraction Lymphatics and inflammation Prostanoids Histamine Abbreviations: SM, smooth muscle; MHC, myosin heavy chain; MLC, myosin light chain; MLCK, myosin light chain kinase; MLCP, myosin light chain phosphatase; TM, tropomyosin; CaD, caldesmon; ERK, extracellular regulated kinase; Tn, troponin; TXA 2, thromboxane A 2 ; PAR2, proteinase-activated receptor 2; NO, nitric oxide; NOS, NO synthase; IBD, inflammatory bowel disease; CD, Crohn s disease; UC, ulcerative colitis; TNF, tumor necrosis factor; IL, interleukin. Corresponding author. Tel.: ; fax: Corresponding author. Tel.: ; fax: addresses: vonderwe@ucalgary.ca (P.-Y. von der Weid), marim@tamu.edu (M. Muthuchamy) /$ see front matter 2009 Published by Elsevier Ireland Ltd. doi: /j.pathophys

2 264 P.-Y. von der Weid, M. Muthuchamy / Pathophysiology 17 (2010) Serotonin Nitric oxide Neuropeptides Impairment of lymphatic contractile functions during inflammation Alteration of the expression and/or phosphorylation of contraction regulatory protein during inflammation General conclusion future perspectives Acknowledgements References Introduction to the lymphatic system The lymphatic system is composed of a network of lymphatic vessels interconnected with lymphoid organs, such as lymph nodes, spleen and Peyer s patches in the small intestine, that are distributed throughout most of the body. The lymphatic system is critical to the maintenance of normal interstitial fluid volume and protein concentration. It is responsible for the daily return of 20 50% of the plasma volume and % of the plasma proteins from the interstitium to the systemic circulation [1,2]. The network of lymphatic vessels or lymphatics, usually parallel to blood vasculature is necessary for the controlled transport of lymph, which contains immune cells, antigens, lipids, macromolecules, fluid and particulate matter. Muscular lymphatic vessels, or collecting lymphatics, are comprised of functional units called lymphangions arranged in series and separated by highly competent valves [3,4]. These vessels transport lymph against a net hydrostatic pressure gradient and a standing protein concentration gradient [5] by means of intrinsic characteristics of phasic contractions of the lymphangions, and through the involvement of extrinsic lymphatic pumps. The extrinsic lymph pump relies on compression and expansion of the lymphatics by external forces applied by surrounding tissues to generate the pressure gradients needed for flow. Tonic contraction/relaxation of the lymphatic vessel via local, neural and humoral factors can modulate the flow in that lymphatic bed by altering the outflow resistance. In the intrinsic lymph pump, flow through a lymphatic bed is generated by coordinated contractions of the lymphatic muscle cells [3,6 9]. The brisk contraction of these cells leads to a rapid reduction of the lymphatic diameter, an increase in the local lymph pressure, a transient reversal of lymph flow with closure of the upstream valve, opening of the downstream valve and ejection of lymph downstream. These contractions produce a complicated pattern of pulsatile lymph flow. The contractions are initiated by pacemaker activity located within the muscle layer of the lymphatic wall, causing depolarization necessary to induce action potential [10 12] and increases in intracellular calcium leading to contraction of the actin/myosin filaments within the muscle cells [13 15]. Thus the intrinsic lymphatic pump acts similar to the heart in its generation of flow. The pumping activity of lymphatics can be investigated using analogies to the cardiac pump with the contractile cycle divided into periods of lymphatic systole and diastole [16] and contraction frequency, ejection fraction, stroke volume and lymph pump flow used to evaluate lymphatic pumping function [17]. These contractions can be modulated in an inotropic (i.e. mediated by changes in the strength of contraction) and/or chronotropic (i.e. mediated by changes in the contraction frequency) fashion by transmural pressure, lymph flow/shear, neural input and humoral influences. It is well documented that humoral agents such as -adrenergic stimulators, prostaglandins, bradykinin, substance P and others can affect the tone of lymphatics and alter flow resistance. Lymphatic vessels must also act as conduits to carry lymph flow to the nodes and great veins of the neck. Thus lymphatics also act similar to blood vessels in terms of their ability to control resistance and flow within their architecture. It has also been shown that humoral agents can modulate intrinsic pumping by altering the lymph pump pressure relationship [18]. Increases in vessel wall stretch can produce large increases in the lymphatic contraction frequency and strength of contraction [19 21]. Recent work has demonstrated that similar to blood vessels, lymphatics are sensitive to flow/shear [22,23]. Lymph flow produces an endothelial-dependent relaxation and a powerful inhibition of the lymph pump, which is partially mediated by nitric oxide (NO) [22]. Thus lymphatic muscle contraction is central to the ability of the lymphatic system to generate and regulate lymph flow in order to accomplish its functions. This requires complicated muscle activities that have similarities to those seen in both the heart (striated muscle) and blood vessels (smooth muscle). However, molecular studies addressing the lymphatic vessel contraction are very limited. An important question is whether lymphatic muscle exhibits similar or different regulatory mechanisms to modulate its tonic and phasic contractions. This review article will focus on discussing the mechanisms that regulate lymphatic vessel contraction under normal and pathophysiological states, specifically associated with inflammatory conditions. 2. Uniqueness of lymphatic muscle The initial lymphatics or lymphatic capillaries are blindended tubes comprised of single-cell layer of overlapping endothelial cells without continuous basement membrane. They are uniquely tethered to the surrounding interstitial matrix through characteristic anchoring filaments [24 27]. Compression of the initial lymphatic vessels by local tissue

3 P.-Y. von der Weid, M. Muthuchamy / Pathophysiology 17 (2010) forces, closes the inter-endothelial junctions or primary lymphatic valves [28 30] and propels lymph into the collecting lymphatics that have an additional muscular layer and an adventitia containing sparse adrenergic, cholinergic and peptidergic innervation [2,31,32]. The muscle cell layers invested in the outer walls of collecting and transport lymphatic vessels generate and control the movement of lymph along the lymphatic network, even against significant opposing pressure gradients. In the absence of characterization of the contractile and regulatory elements of the lymphatic muscle contractile apparatus, it had been thought, until recently that these muscle cells were smooth muscle (SM) cells. We have recently shown that lymphatic muscle contains both striated and SM contractile elements [33]. Furthermore our data have demonstrated that differences in both contractile function and contractile machinery exist between blood vessels and lymphatics as well as among lymphatics from different body regions. How these different contractile apparatus elements are employed simultaneously within lymphatic muscle to control tonic and phasic contractions is not yet clear. An account of our partial knowledge about lymphatic muscle contractile regulation is provided below, starting with a review of the processes regulating other smooth and striated muscle contraction and how lymphatics may utilize those processes. 3. Regulation of smooth muscle contraction The contractile apparatus of smooth muscle is generally similar to that of striated muscle, although it is not highly organized into a discrete sarcomeric structure. This is principally due to the variation of the content of myosin and actin and other filament organizing proteins in the two muscle types. The molecular components of the contractile apparatus have been characterized in skeletal and cardiac muscle as well as various types of smooth muscle [34 36]. Biochemical and molecular analyses show that smooth muscle expresses at least four muscle-specific and two non-muscle specific myosin heavy chains (MHC), and four types of myosin light chains (MLC) during growth, development, and disease [37]. The smooth muscle MHC variants are SM1A and SM1B, SM2A and SM2B, in which SM1 and SM2 vary in their carboxy-terminal ends and the SM-B isoform has a specific seven amino acid insert that resides in surface loop 1 of the myosin head, adjacent to the ATP-binding domain of MHC. Several in vitro studies have demonstrated that the SM-B (SM1/SM2) isoform expressed in the smooth muscle types with faster contractile properties, such as in the bladder, has nearly two-fold higher ATPase activity. In addition, the rate of MgATP binding to and MgADP release from the active site is increased in the SM-B isoform. Furthermore, in vitro biochemical studies have shown that preparations containing the SM-B isoform have an increase in the velocity of shortening when compared to SM-A preparations [38 40]. The recent mouse model developed by Babu et al. lacking the SM-B MHC isoform exhibits abnormal contractile function of aorta and bladder tissues [41]. Our recent data show the presence of significant amounts of SM-B MHC in mesenteric lymphatic muscle that correlates well with its rapid phasic contractile nature [33]. The thoracic duct and arterioles, on the other hand, express both SMA and SMB isoforms. SM1 and SM2 isoforms of SM-MHC are also detected in mesenteric and thoracic lymphatics, as well as in arterioles. Additionally, a slow-skeletal/fetal cardiac muscle-specific -MHC message is found only in mesenteric lymphatics. All four actin messages, -cardiac, -vascular, -enteric and -skeletal, are present in both mesenteric lymphatics and arterioles; however the thoracic duct, predominantly expresses -cardiac and vascular actins. Furthermore, we have found evidences of expression of some striated muscle regulatory components, such as troponin C and T in lymphatic muscles (M.M., unpublished observations). The smooth muscle MLC variants are smooth musclespecific MLC 20 and non-muscle MLC 20, and MLC 17 a and b isoforms [34]. It is well established that Ca 2+ -dependent MLC phosphorylation plays a major role in the regulation of SM contraction [42]. In general, increases in the intracellular Ca 2+ ([Ca 2+ ] i ) level is the primary mechanism that initiates smooth muscle contraction as is seen in striated muscles. However, in smooth muscle Ca 2+ binds to calmodulin, and the Ca 2+ -calmodulin complex activates the myosin light chain kinase (MLCK) that phosphorylates regulatory myosin light chain 20 (MLC 20 ) [43]. Phosphorylation of MLC 20 permits the myosin ATPase to be activated by actin and thus, initiates the smooth muscle contraction. A decrease in [Ca 2+ ] i leads to inactivation of MLCK and to the dephosphorylation of MLC 20 by myosin light chain phosphatase (MLCP), which then leads to the deactivation of the myosin ATPase activity and relaxation of the muscle. Thus, in the regulation of smooth muscle contraction, the phosphorylation and dephosphorylation of the MLC 20 plays the central role. Recently, however, several studies have described regulatory mechanisms that modulate the activities of both MLCK and MLCP and thus alter contractile status, which are independent of [Ca 2+ ] i [43,44]. Although myosin ATPase activity is highly correlated with MLC phosphorylation, the Ca 2+ -regulated thin filament system also plays an important modulatory role in SM contraction [45 49]. Smooth muscle contains tropomyosin (TM) and caldesmon (CaD) at a ratio of 1 CaD:2TM:14 actin monomers [50,51], in addition to smaller amounts of filamin, calponin, and calmodulin [52]. CaD acts as an inhibitor of actin activation of myosin ATPase activity, and many characteristics of this inhibition are similar to the role of troponin I in striated muscle [53]. In SM, the phosphorylation of CaD by p21-activated kinase modulates the CaD-inhibition of myosin ATPase activity, thereby controlling the Ca 2+ sensitivity of smooth muscle contraction [54]. The phosphorylation of caldesmon is also regulated by protein kinase C and extracellular regulated kinase 1 (ERK1); however different results have been reported on the inhibitory function of CaD on

4 266 P.-Y. von der Weid, M. Muthuchamy / Pathophysiology 17 (2010) myosin ATPase activity [55]. Similar to the roles of CaD, the role of calponin is also controversial in smooth muscle contraction. There are indications that calponin may directly inhibit the actin-activated myosin ATPase activity [56]. On the other hand, biochemical studies have shown that calponin acts as a signaling molecule that facilitates protein kinase C and ERK redistribution, which then phosphorylates CaD, and activates contraction [57,58]. In lymphatics, distribution of CaD and calponin vary between the different lymphatic beds (M.M. unpublished observation); both semi-quantitative and quantitative RT/PCR assays show that CaD and calponin expression is at a higher level in thoracic duct and cervical lymphatics compared with mesenteric lymphatics. However, CaD protein expression is higher in mesenteric lymphatics compared to thoracic duct and cervical lymphatics indicating the regulation of CaD expression occurs at the translational level. Thoracic duct and cervical lymphatics express higher levels of calponin protein correlating the mrna data. Furthermore both structural and biochemical studies support a role for smooth muscle tropomyosin in the cooperativity of smooth muscle activation [59]. It has also been reported that tropomyosin is necessary for the full inhibition of actin-activated myosin ATPase activity by caldesmon [60]. SM-specific TM isoforms also potentiate actin-activated myosin ATPase activity [61 63]. Recent studies suggest an involvement of TM movement in the on/off switching of SM contraction in an as-yet-undefined manner [64]. 4. Regulation of striated muscle contraction In striated muscle, calcium binding to troponin C (TnC) triggers an allosteric effect on troponin I (TnI) and troponin T (TnT) allowing TM to flex [65]. This causes TM to move away from its steric hindrance position, allowing the attachment of myosin heads to actin. The transition from a weakly bound crossbridge to a strongly bound state locks TM in an open position; this produces a cooperative effect of promoting further crossbridge attachment [66,67]. After the generation of a power stroke, the actomyosin interaction shifts to a rigor state. Subsequent ATP binding then promotes the detachment process. Although the precise mechanism by which the TM Tn complex regulates muscle contraction is still unknown, various models have been proposed (i.e. steric hindrance model, biochemical blocking model [68,69] and a continuum model [70]) for myofilament activation processes. Various experimental evidences including x-ray crystallographic, three-dimensional reconstruction and biochemical data supporting these models demonstrate that Ca 2+ binding to thin filament induces TM movement on the surface of the thin filament, and TM Tn complex may block the actin-myosin interaction. Furthermore, current models for thin filament activation and the crossbridge cycle support a three-state model in which TM exists in blocked, closed (off), and open (on) states [66,67,71,72]. Threedimensional reconstructions of myofilament structure from electron micrographs support the notion that TM requires both calcium and bound myosin to reach the fully activated state [71]. Furthermore the biochemical experiments supported the structural data, demonstrating that in addition to Ca 2+ binding to TnC, full activation of the thin filament requires strongly bound crossbridges [67]. Preliminary evidence from our lab (unpublished observations) indicates that (cardiac) ctnc is present in lymphatic muscle, however what role if any this key determinant of thin filament activation plays in lymphatic muscle contraction is unknown. 5. Regulation of lymphatic muscle contraction Despite the preponderance of information on both smooth and striated muscle regulatory mechanisms, very little is known about lymphatic muscle contraction. In fact, our recent studies provide the first evidence that lymphatic muscle contractile apparatus consists of both striated and smooth muscle contractile elements [33]. In addition, preliminary data demonstrate that lymphatic muscles express striated muscle regulatory components, such as ctnc and (cardiac) ctnt (unpublished observations). Based on the differences in the contractile elements present in lymphatics when compared to blood vasculatures, we proposed that contractile characteristics of lymphatic myofilaments would be different from vascular smooth muscle myofilaments. To test this hypothesis, we have recently determined pca tension relationships for -toxin permeabilized mesenteric lymphatics, arteries and veins [73]. The Ca 2+ sensitivity (pca 50 ) of mesenteric lymphatics was significantly lower compared to arteries (6.16 ± 0.05 vs ± 0.02; p < 0.05), whereas there was no difference in pca 50 between lymphatics and veins (6.16 ± 0.05 vs ± 0.10; n.s.). The Hill coefficient for a-toxin permeabilized lymphatics was not significantly different from arteries, but was significantly greater than that of the veins (1.98 ± 0.19 vs ± 0.18; p < 0.05). These data suggest that differences in myofilament Ca 2+ sensitivity and cooperativity among lymphatic muscle and vascular smooth muscles contribute to the functional differences that exist between these tissues. Zhang et al. [74] have shown that the mechanical comparisons of lymphatic and vascular tissues show dramatic differences; while small arteries and veins exhibit active tension from 80% to 90% of maximal active force (F max ) at 0.8 optimal internal circumference (L 0 ) of the vessel, lymphatics are only about 50 60% of F max, indicating that lymphatic vessels are relatively less compliant, with an increased tissue elastic modulus. Furthermore, Davis et al. [75,76] demonstrate that myogenic responses occur in mesenteric lymphatics in response to changes in intraluminal pressures and lymphatic muscle exhibits rate-sensitive contractile responses to stretch. Hosaka et al. [77] evaluated the participation of the Rho- Rho kinase pathway in the regulation of lymphatic contractile

5 P.-Y. von der Weid, M. Muthuchamy / Pathophysiology 17 (2010) activity. This study employed the Rho-kinase inhibitor, Y and the phosphatase inhibitor, okadaic acid on lymph pump activity and myogenic, pressure- and agonist-induced tone in isolated rat lymphatics. Y produced a reduction of basal and activated tonic contraction with a cessation of the lymph pump activity at high doses. Okadaic acid produced an increase in lymphatic tonic constriction but reduced the frequency of the phasic lymphatic contractions. Additionally, the effects of Y were in part reversed by the pretreatment with okadaic acid. These findings indicate that Rho kinase and myosin phosphatase contribute to the regulation of lymphatic tonic contractile activity and may be involved in the development of activity associated with the phasic contractions. Our recent data show that the MLCK specific inhibitor, ML-7 decreases MLC 20 phosphorylation in lymphatics; while the tonic contraction of lymphatic vessel is reduced in the presence of ML-7, phasic contraction amplitude is not altered (M.M., unpublished observation). These data indicate that tonic contraction strength and phasic contraction amplitude of the lymphatics can be differentially regulated, whereby the activation of MLC 20 phosphorylation produces an increase in tonic contraction without a change in phasic contraction amplitude. Taken together, we propose (Fig. 1) that the differences in contractile characteristics, calcium sensitivity and cooperativity, that exist in lymphatic myofilaments contribute to the functional differences existing between vascular and lymphatic tissues. Furthermore, the presence of striated muscle regulatory and contractile elements in lymphatic muscle regulates its phasic contraction; whereas, the smooth muscle regulatory elements govern the tonic contraction of lymphatic muscle. Thus, the unique mechanical and contractile characteristics of lymphatic muscle would comprise of both smooth and striated muscle regulatory mechanisms to accomplish its essential function in maintenance of the normal lymph transport containing nutrients and metabolites as well as in protection against the formation of gross edema and other pathological conditions. 6. Lymphatics and inflammation Through its function to propel lymph, the lymphatic system is strongly implicated in the adaptive immune response. It is responsible for transporting antigens to lymphoid tissues to allow initiation of an immune response during disease and in response to infection. Lymphoid structures such as lymph nodes are distributed along the lymphatic vessel network. They are composed of lymphatic and blood vessels spread out inside a parenchyma, subdivided into B-cell follicles and a T-cell area that together form the cortex and the medulla. Cellular and molecular traffic between these compartments is an essential aspect of lymph node physiology. Immune cells, fluid, bacteria, macromolecules, viruses and their products enter lymph nodes from tissues via afferent lymphatic vessels or from the cardiovascular network via Fig. 1. Stylized lymphatic vessel diameter and calcium tracing, and proposed regulatory mechanisms for lymphatic muscle. Changes in vessel diameter due to an increased pressure or flow are shown as an example for functional and contractile behavior of the lymphatic vessel. An example of spontaneous contractions and calcium transients of rat mesenteric lymphatic vessel is shown. As discussed in the text, presence of both smooth and striated muscle-specific contractile and regulatory proteins in lymphatic muscle are listed. Possible combinations of smooth and striated muscle contractile and regulatory proteins that may involve in modulating tonic and phasic contractile mechanisms are proposed. SM smooth muscle, MHC myosin heavy chain, MLC myosin light chain, MLCK myosin light chain kinase, MLCP myosin light chain phosphatase, TM tropomyosin. high endothelial venules. Antigens present in fluid entering the lymph node can effectively elicit an immune response by activating resident naïve T- and B-cells. Activated lymphocytes exit the lymph nodes via blood vessels or via efferent lymphatics, eventually returning to the bloodstream, where they are transported to tissues throughout the body and act as patrols on the lookout for foreign antigens. Hence the lymphatic vasculature provides an exclusive environment where immune cells can respond to foreign antigens, and the means for circulating lymphocytes to return to the bloodstream. The lymphatic system also functions as a one-way communication system for molecular messages, such as inflammatory mediators or chemokines, which can be transmitted to cellular constituents in lymph nodes. Although no investigations

6 268 P.-Y. von der Weid, M. Muthuchamy / Pathophysiology 17 (2010) have been performed yet to formally address the possibility that chemokines could play a role in controlling lymph flow, effect of more classical inflammatory mediators, such as prostanoids, histamine or nitric oxide have been correlated with modulation of lymphatic pumping and drainage. Similarly, neuromediators important in immune and inflammatory responses, such as substance P, calcitonin gene related peptide (CGRP), neuropeptide Y or vasoactive intestinal polypeptide (VIP), have also been reported to strongly affect lymphatic vessel contractility. In vivo however, it is difficult to determine whether a lymphatic response is due to a direct stimulation by inflammatory mediators, or is a secondary consequence of other aspects of the inflammatory response, such as edema and subsequent vessel filling. Indeed, the same inflammatory mediators induce vascular leakage, causing net flow of fluid into the interstitial space and this, in itself, enhances lymphatic contractility. These inflammatory mediator s effects on lymphatic pumping is discussed in the next few sections Prostanoids Arachidonic acid metabolites are among the most important regulators of lymphatic vessel contractility, and because they are also important mediators of inflammation, they likely also play an important role in lymphatic dysfunction in the setting of inflammation. Numerous metabolites produced from arachidonic acid are processed in inflammatory cells and surrounding tissues. Importantly, lymphatic vessels themselves are capable of producing them. Depending on which one, it will lead to variable lymphatic responses. Early studies using inhibitors of cyclo-oxygenase (COX) and of other arachidonate metabolism pathways demonstrated that spontaneous lymphatic pumping is abolished when these mediators are repressed [78]. The same study also showed that application of leukotrienes, as well as a PGH 2 /thromboxane A 2 (TXA 2 ) mimetic, induces rhythmic constriction in non-contracting lymphatic vessels. In a study of rat iliac lymphatics, arachidonic acid caused a significant decrease in vessel diameter. These effects were blocked by indomethacin, and significantly reduced by removal of the endothelial layer, suggesting that these metabolites were produced through cyclo-oxygenase, at least in part, in the lymphatic endothelium [79]. Constrictions caused by arachidonic acid were converted to dilations when PGH 2 /TXA 2 receptors were blocked by antagonists, and dilation was mimicked by exogenous PGE 2 [79]. A role for prostaglandins is also illustrated in studies looking at the effect of substance P and ATP, where both enhanced the rate of constriction in guinea pig mesenteric lymphatics, with the effect attenuated by indomethacin [80,81]. These findings indicate that prostanoid PGH 2 /TXA 2 release from the endothelium is activated by both substance P and ATP. More recently, we showed that the decrease in lymphatic pumping induced by the proteinase-activated receptor 2 (PAR2), another potent inflammatory mediator [82], also involves prostaglandins as the effects were blocked by indomethacin and mimicked by PGE 2 and prostacyclin [83] Histamine Histamine release during the inflammatory response causes an increase in microvascular permeability, leading to an accumulation of fluid in the surrounding interstitial space. As a consequence, lymph formation and vessel filling must increase in order to maintain tissue fluid homeostasis. Systemic administration of histamine increases lymph flow [84 88], an effect partly due to increased microvascular permeability [85,88]. As histamine is also present in the interstitium during inflammation, and is detected in lymph draining injured tissues it could well directly stimulate lymphatic vessels [89]. Indeed, histamine potently modulates lymphatic contractile activity. Depending on species and vessel location, it increases the frequency and decreases the amplitude of lymphatic pumping via activation of H 1 receptors located on smooth muscle cells [90 92], decreases contraction frequency via H 2 receptor stimulation [90,92] or via an indirect effect mediated by the endothelium [91]. Histamine is also a predominant constituent of granules in mast cells, predominant players in inflammatory and allergenic reactions. We have demonstrated, using the food allergy model of milk-sensitized guinea pig, that histamine is the main mediator of increased lymphatic pumping caused by mast cell activation [93] Serotonin Serotonin, or 5-hydroxytryptamine (5-HT) is involved in changes in blood flow, vascular permeability and microcirculatory adjustments [94]. As with other modulators of vascular function, 5-HT also has an impact on lymphatic circulation in normal as well as in pathophysiological situations (see [95]). In most quiescent lymphatic vessels, 5-HT produces an increase in tone [96 100] and an increase in spontaneous contractions in bovine mesenteric vessels [101]. However, 5- HT causes a relaxation when these vessels are pre-constricted [101] or an inhibition of spontaneous contractions in both guinea pig and sheep mesenteric vessels [102,103]. Pharmacological characterization of the receptors indicates that increase in tone and contraction frequency are mediated by 5- HT 2 receptors present in the lymphatic vessel wall [94,101], whereas relaxation and decreases in contraction frequency are due to the activation of 5-HT 4 receptors in guinea pig and 5-HT 7 receptors in sheep mesenteric vessels [ ] Nitric oxide Previously identified as endothelium-derived relaxing factor (EDRF), the gaseous molecule nitric oxide is involved in many different physiological processes. NO is continuously released from the endothelium in blood vessels to affect the smooth muscle tone of their walls. More recently,

7 P.-Y. von der Weid, M. Muthuchamy / Pathophysiology 17 (2010) it has been shown that not differently, NO also affects lymphatic muscle function. Studies conducted using several different animal species have consistently demonstrated that NO abolishes spontaneous constriction dilation cycles that are seen in lymphatic vessels [79]. In addition, nitric oxide-synthase (NOS) inhibitors increase the rhythm and amplitude of vessel contraction [104]. NO inhibits vasomotion through arresting intracellular calcium release associated with spontaneous transient depolarizations and contractions. Intracellular microelectrode recordings from guinea pig mesenteric lymphatic vessels showed that endotheliumderived or exogenously applied nitric oxide decreased the frequency and amplitude of spontaneous transient depolarizations, an effect abolished in the presence of ODQ, a guanylate cyclase blocker. These findings suggest that inhibition of pumping by NO involves activation of guanylate cyclase and cgmp-dependent protein kinase [105]. NO is also released in response to flow and this pathway has been implicated in the flow-induced decrease in active pumping in rat mesenteric and thoracic vessels, as well as in the decreased vessel tone caused by the flow generated by the vessel contractions in rat thoracic duct [22,106]. Taken together, these findings indicate that lymphatic muscle tone is continuously modulated by endothelium-derived NO. In inflammatory situations, large quantities of NO additionally synthesized by immune cells, such as macrophages, or by inflamed tissues, such as smooth muscle could certainly further alter lymphatic contractility [107] Neuropeptides Neuromodulators with vasoactive properties such as substance P, VIP, neuropeptide Y and CGRP are present in enteric and sensory nerves and are present in peptidergic innervation of mesenteric lymphatics [32,108]. They are often associated with cells of lymphoid tissue [32, ] and are also released from inflammatory cells potentially present in and around lymphatic vessels [ ]. In vitro effect of these neuropeptides on lymphatic contractile activity has been reported. VIP potently inhibits mesenteric lymphatic pumping [108], an action suggested to be independent of the endothelium and mediated at least in part via stimulation of the PAC1 receptor [120]. CGRP also caused a decrease in lymphatic contraction frequency in the guinea pig mesentery. It has been shown to be mediated at concentrations between 1 and 100 nm by the stimulation of endothelial CGRP-1 receptors and the enhancement of the release or the actions of endogenous NO, and at high concentrations (500 nm), by direct activation of the lymphatic muscle in a NO-independent manner [121]. In the rat mesentery however, CGRP enhanced phasic contractions of the lymphatic vessels [122]. The effect of neuropeptide Y on lymphatic pumping has not yet been fully assessed. One study reported an increase in the pacemaking activity correlating with an increase in pumping in guinea pig mesenteric lymphatics [123]. Substance P also has direct actions on collecting lymphatics and is probably the most studied of these neuropeptides. In isolated bovine mesenteric lymphatics, substance P increased spontaneous contraction rate [124]. A positive chronotropic effect mediated by PLA 2 activation and the production of thromboxane A 2 (TxA 2 ) by the lymphatic endothelium has also been shown in isolated guinea pig mesenteric lymphatics. Little or no direct effect on the lymphatic muscle could be found [125]. In rat mesenteric lymphatics, substance P also has substantial positive inotropic and chronotropic effects that improve pump efficiency, however in this case, it acts directly on the lymphatic muscle [126,127]. 7. Impairment of lymphatic contractile functions during inflammation Despite the cardinal function of lymphatic vessels to adapt their contractile activity to changes in fluid load, they are still not often considered as active players in inflammatory pathologies where edema is a hallmark. In these situations, mediators of inflammation, released in the vicinity of lymphatic vessels, either directly or through changes in vessel load, could certainly affect lymphatic muscle and alter lymphatic pumping. Keeping in mind that dysfunction of lymphatic pumping and impairment of lymph flow may occur in most inflammatory pathologies, such as heart and renal failures or pulmonary edema, we will discuss in the following section the role of lymphatics in chronic inflammatory bowel diseases (IBD). IBD is an ensemble of complex disorders including Crohn s Disease (CD) and Ulcerative Colitis (UC), which involve chronic inflammation of the gastrointestinal tract with symptoms such as diarrhea, abdominal pain and cramping. CD can affect any part of the gastrointestinal tract from the mouth to the anus and is characterized by patches of inflammation with intermittent areas of healthy tissue and with the inflammation involving all layers of the bowel wall. UC on the other hand is confined to the rectum, with uniform inflammation of the mucosal and submucosal layers only. Current theories about the pathogenesis of IBD point to an impaired mucosal immune response in a host susceptible to the microbes within the intestinal flora. However, the exact mechanisms of immune, environmental and genetic involvement are still not well understood. Despite the uncertainty of IBD s etiology, some of the most consistent pathological features observed in patients suffering from CD and UC are mucosal exudation, interstitial (submucosal) edema and extensive dilation of lacteals, the intestinal initial lymphatics [ ]. Lymphatic vessel dilation and edema could be consequent to either lymphatic obstruction, as reported in IBD patients undergoing surgery [131] or impaired contractile function [129,131]. The presence of edema suggests poor lymph drainage, which could then cause mucosal hypoxia and fibrosis, two features that have been intimately associated with chronic inflammation in CD [132,133]. The poor

8 270 P.-Y. von der Weid, M. Muthuchamy / Pathophysiology 17 (2010) drainage of interstitial fluid could also lead to impaired transport of large molecules, particles, dead cells, and bacteria away from the intestine, which may promote infection and delay the immune response. This concept is supported by earlier experimental studies where injection of sclerotic agents into canine and porcine mesenteric lymphatic vessels or lymph nodes to obstruct them led to lesions similar to those seen in human IBD [134,135]. Due to increases in vascular permeability and resultant interstitial fluid, lymph flow is thought to increase during inflammatory reactions. Indeed, mesenteric lymphatic pumping was increased during edemagenic stress caused by dilution of plasma in rats in vivo [17]. This effect was due to an increase in distension of the lymphatic wall, which may be the situation occurring when lymphatics are overloaded in cases of edema. Although the involvement of lymphatic vessels was demonstrated in earlier investigations, the study of lymphatics in human IBD was not pursued until recently and new information is now available. Initial lymphatic vessels have been shown to be associated with a proportion of granulomas seen in CD patients, leading to the suggestion that granulomatous lymphangitis is a primary lesion of CD [136,137]. Proliferation of initial lymphatics has been demonstrated in the colonic mucosa of patients with UC, but not in healthy controls [138,139]. In the colon, lymphatics, which are normally distributed beneath the muscularis mucosa, proliferate into the lamina propria and submucosa in patients with UC in proportion to the severity of the disease [138]. In addition, the integrity of the lamina propria, with regards to lymphatic distribution, is restored with disease resolution. Like in UC, proliferation of lymphatic initials is also observed in CD, where it can occur in each layer of the inflamed small and large bowel [140]. Lymphatic capillaries are prominent in lymphoid aggregates and also observed in fibrotic areas. These findings suggested that lymphangiogenesis in IBD may be triggered by chronic inflammation and is maintained in fibrotic end-stage disease [140]. Demonstration of lymphangiogenesis has also been made in other inflammatory situations [ ]. Lymphatic circulation is thus likely to play a crucial role during IBD, but the involvement of lymphatic contractile function in inflammatory diseases and its role during IBD has only been recently addressed. Findings from our laboratory reveal that the contractile function of mesenteric collecting lymphatics is impaired in the guinea pigs model of 2,4,6- trinitrobenzenesulfonic acid (TNBS)-induced ileitis with a strong correlation between the contractile dysfunction and the degree of mucosal inflammation. Specifically, experiments performed on isolated mesenteric lymphatic vessels showed that in TNBS-treated animals vessel diastolic diameters increased with the severity of inflammation, whereas their constriction frequency was significantly lower than in those from sham-treated animals, with most vessels not contracting at all (Fig. 2). When constrictions occurred, their amplitude was significantly lower than that of vessels from sham-treated animals. Consistent with these in vitro findings, intravital measurements, performed in anesthetized animals demonstrated that the diameters of lymphatic vessels proximal to the TNBS injection site were significantly larger than those from their sham counterpart (Fig. 2) and that the number of spontaneously contracting lymphatics were reduced by half. Importantly, the contractile function was partially, but significantly, restored in the presence of the non-selective COX inhibitor, indomethacin or a combination of selective COX-1 and COX-2 inhibitors (Fig. 2). These findings suggest the involvement of arachidonic acid metabolites in the inflammation-induced impairment of lymphatic pumping [145]. Subsequent investigation looking at the lymphatic expression of enzymes involved in the synthesis of inhibitory prostaglandins, such as PGE 2 and prostacyclin, known to be abundantly produced during intestinal inflammation, revealed that although most of them seem to be down regulated in mesenteric lymphatic vessels from TNBS-treated animals, COX-2 was overexpressed (Rehal and vdw, unpublished observation). Moreover and consistent with PGE 2 and prostacyclin receptors being coupled with Gs proteins [146], activation of EP 4 and IP receptors in these vessels has been shown to involve camp-dependent protein kinase dependent phosphorylations. Neuropeptides are other potential candidates responsible for the lymphatic dysfunction observed during intestinal inflammation. In addition to their presence or release in the vicinity of lymphatic vessels, these neuropeptides are suggested to play key roles in IBD, where either their content and/or expression of their receptors is altered (see [147]). For example, colonic muscle strips from patients with IBD exhibited impaired CGRP and substance P-mediated contractility. Also of interest is the marked reduction of substance P measured in the intestinal mucosa and submucosa of the guinea pig during TNBS-induced ileitis [148]. 8. Alteration of the expression and/or phosphorylation of contraction regulatory protein during inflammation Altered expression of contractile proteins during inflammation has been shown to occur in cardiac and various smooth muscles, but this has yet to be reported in lymphatic muscles. It has been clearly demonstrated that myocardial function is altered in patients, as well as in animal models during sepsis [ ]. Wu et al.[152] have shown that bi-phasic changes in phosphorylation of TnI, myosin binding protein-c and myosin light chain 2 and calcium sensitivity of the cardiac myofilaments in the cecal ligation and puncture-induced sepsis rat model; thus the authors have suggested modulation of -adrenergic pathway is involved in the sepsis-induced myocardial dysfunction. Tumor necrosis factor (TNF)- and interleukin-1 (IL-1) family cytokines are the primary inflammatory mediators in sepsis-induced cardiac

9 P.-Y. von der Weid, M. Muthuchamy / Pathophysiology 17 (2010) Fig. 2. Effect of intestinal inflammation on mesenteric lymphatic tonic and phasic contractile activity in the guinea pig model of TNBS-induced ileitis. Collecting lymphatic vessels in the guinea pig mesentery (arrows and dotted lines in c) have diameters markedly larger in TNBS-treated animals in vivo (c) as well as in isolated vessels (d), than those from sham-treated animals (a and b). The contraction frequency of isolated vessels is also strongly impaired by the treatment (compare b and d). Scales in a and b apply to c and d. dysfunction. Several clinical and experimental studies suggest that inflammatory mediators play a significant role in the progression of heart failure; in experimental animal models it has been shown that these inflammatory mediators cause left ventricular dysfunction, myocyte hypertrophy, chamber dilation, activation of fetal gene expression and myocyte apoptosis [153]. Li et al.[154] have recently demonstrated that gp91(phox)-nadph oxidase-mediated calpain-1 activation induces caspase-3 activation and TNF- expression in lipopolysaccharide (LPS) stimulated cardiomyocytes. Thus, recent studies on inflammatory-mediated cardiac dysfunction have provided information on the signaling mechanisms and altered myofilament proteins phosphorylation during inflammation conditions. CPI-17, the endogenous myosin phosphatase inhibitor, is downregulated in intestinal smooth muscles of UC patients and of mice models of intestinal inflammation [155,156]. In colonic segments from TNBS-treated rats, carbacholinduced phosphorylation of CPI-17 was also reduced [157]. In this study, the reduced colonic smooth muscle contractility observed in IBD was further associated with inflammation-induced changes in Ca 2+ -sensitization as colon strips from TNBS-treated rats showed greater sensitivity to Ca 2+ and had a higher basal tone as compared to controls. The Rho-kinase inhibitor Y induced significantly greater relaxation in these tissues, consistent with increased expression of Rho-kinase in the inflamed colon. Moreover, total protein kinase C expression was reduced in inflamed colonic muscle, a finding that could be correlated with an attenuation of carbachol-induced colonic contraction. These effects, leading to motility disorders, are likely to be induced by the chronic production of TNF- and/or IL-1 [ ]. Importantly, data reported by Ohama et al. [160] also implicated a decrease in the phosphorylation level of the regulatory subunit of myosin phosphatase (MYPT-1) during IL-1 treatment, which could contribute, together with CPI-17 downregulation, in an increase in myosin phosphatase activity to reduce force generation, a possible cause of gastrointestinal motility disruption associated with IBD. IL-1 has also been shown to cause a ten-fold increase in -MHC mrna expression in myocytes [161]. IL-1 and TNF- are key immunoregulatory cytokines that amplify the inflammatory response by activating a cascade of immune cells [162]. They play important roles in IBD and their production is increased in inflamed mucosa. They have indeed been detected in rat lymph following intestinal ischemia/reperfusion [163]. They are also secreted by immune cells that are known to traffic through lymph between inflamed sites and lymph nodes. It is thus very likely that cytokine-induced dysregulation of expression or function of contractile and/or regulatory proteins, similar to that reported in intestinal muscle during inflammation, occurs in lymphatic vessels, and could account at least partly for the contractile inhibition observed in mesenteric vessels from TNBS-treated guinea pigs. To our knowledge, no study addressing this idea has been performed in lymphatic tissues yet. However, and in agreement with this hypothesis, Hanley et al. [164] reported an inhibition of fluid pumping in isolated bovine mesenteric lymphatics by IL-1 and. There is substantial evidence that the lymphatic system is intimately involved in and highly altered during inflammatory diseases particularly of the gut. Release of inflammatory mediators, in addition to increasing vascular permeability during inflammation, is thought to play a pivotal role in modulating lymphatic vessel function. Although the exact role of lymphatics is not yet known and failure in the lymphatic system is probably not the direct cause of the ailment, inflammation has a significant

10 272 P.-Y. von der Weid, M. Muthuchamy / Pathophysiology 17 (2010) effect in impairing normal contractile function of the vessels. Whether the dysfunction is correlated with a modification of the expression or function of contractile or regulatory proteins in the course of the inflammatory process has not been addressed. 9. General conclusion future perspectives A small number of research groups have recently begun to unravel the regulatory mechanisms of lymphatic muscle contraction. Addressing the molecular mechanisms that modulate force generation in lymphatic muscle during tonic and phasic contractions under normal and diseased states will provide valuable insight into the targeting molecules that regulate lymphatic muscle contraction, and could provide the basis for better diagnosis and treatment of lymphatic dysfunction. Perturbation in lymph flow can result in a wide range of clinical disorders, including lymphedema, various inflammatory conditions, lymphocyte circulation, and impaired lipid metabolism, among others. Thus understanding the regulatory mechanisms that modulate the lymphatic functions and the generation of lymph flow under normal and diseased states would lead us to develop therapeutic agents targeted to improve lymphatic function in these various pathophysiological conditions. Acknowledgements This study was supported by Grants HL and KO2 HL from the NIH (to M.M.) and by Grants from the Canadian Institutes of Health Research (CIHR) the Crohn s & Colitis Foundation of Canada (CCFC) and the National Sciences and Engineering Research Council of Canada (NSERC) (to P.-Y. vdw). We would like to thank the following people for their help in the studies that support this work Drs. Michael J. Davis, Anatoliy A. Gashev and David C. Zawieja. References [1] A.E. Taylor, The lymphatic edema safety factor: the role of edema dependent lymphatic factors (EDLF), Lymphology 23 (1990) [2] J.M. Yoffey, F.C. Courtice, Lymphatics, Lymph and the Lymphomyeloid Complex, Academic Press, London, [3] H. Mislin, Active contractility of the lymphangion and coordination of lymphangion chains, Experientia 32 (1976) [4] H. Mislin, Structural and functional relations of the mesenteric lymph vessels, New Trends in Basic Lymphology, Proceedings of a Symposium held at Charleroi (Belgium), July 11 13, 1966, Experientia Suppl. 14 (1966) [5] G.W. Schmid-Schonbein, Microlymphatics and lymph flow, Physiol. Rev. 70 (1990) [6] J.M. Allen, N.G. McHale, Neuromuscular transmission in bovine mesenteric lymphatics, Microvasc. Res. 31 (1986) [7] D.C. Zawieja, K.L. Davis, R. Schuster, W.M. Hinds, H.J. Granger, Distribution, propagation, and coordination of contractile activity in lymphatics, Am. J. Physiol. 264 (1993) H1283 H1291. [8] N.G. McHale, M.K. Meharg, Co-ordination of pumping in isolated bovine lymphatic vessels, J. Physiol. 450 (1992) [9] M.J. Crowe, P.Y. von der Weid, J.A. Brock, D.F. Van Helden, Coordination of contractile activity in guinea-pig mesenteric lymphatics, J. Physiol. 500 (1997) [10] H.M. Toland, K.D. McCloskey, K.D. Thornbury, N.G. McHale, M.A. Hollywood, Ca(2+)-activated Cl(-) current in sheep lymphatic smooth muscle, Am. J. Physiol. Cell Physiol. 279 (2000) C1327 C1335. [11] D.F. Van Helden, Pacemaker potentials in lymphatic smooth muscle of the guinea-pig mesentery, J. Physiol. 471 (1993) [12] P.Y. von der Weid, M.J. Crowe, D.F. Van Helden, Endotheliumdependent modulation of pacemaking in lymphatic vessels of the guinea-pig mesentery, J. Physiol. 493 (1996) [13] Y. Shirasawa, J.N. Benoit, Stretch-induced calcium sensitization of rat lymphatic smooth muscle, Am. J. Physiol. Heart Circ. Physiol. 285 (2003) H2573 H2577. [14] D.C. Zawieja, E. Kossman, J. Pullin, Dynamics of the microlymphatic system, J. Prog. Appl. Microcirc. 23 (1999) [15] N.G. McHale, J.M. Allen, The effect of external Ca2+ concentration on the contractility of bovine mesenteric lymphatics, Microvasc. Res. 26 (1983) [16] H.J. Granger, S. Kovalcheck, B.W. Zweifach, G.E. Barnes, Quantitative analysis of active lymphatic pumping, in: Proceedings of the VII Summer Computer Simulation Conference, Simulation Council, La Jolla, CA, [17] J.N. Benoit, D.C. Zawieja, A.H. Goodman, H.J. Granger, Characterization of intact mesenteric lymphatic pump and its responsiveness to acute edemagenic stress, Am. J. Physiol. 257 (1989) H2059 H2069. [18] R.M. Elias, M.G. Johnston, A. Hayashi, W. Nelson, Decreased lymphatic pumping after intravenous endotoxin administration in sheep, Am. J. Physiol. 253 (1987) H1349 H1357. [19] A.R. Hargens, B.W. Zweifach, Contractile stimuli in collecting lymph vessels, Am. J. Physiol. 233 (1977) H57 H65. [20] N.G. McHale, I.C. Roddie, The effect of transmural pressure on pumping activity in isolated bovine lymphatic vessels, J. Physiol. (Lond.) 261 (1976) [21] H. Mislin, R. Schipp, Structural and functional relations of the lymph vessels. Progress in lymphology, in: Proceedings of the International Symposium on Lymphology, Zurich, Switzerland, July 19 23, 1966, pp [22] A.A. Gashev, M.J. Davis, D.C. Zawieja, Inhibition of the active lymph pump by flow in rat mesenteric lymphatics and thoracic duct, J. Physiol. 540 (2002) [23] A. Koller, R. Mizuno, G. Kaley, Flow reduces the amplitude and increases the frequency of lymphatic vasomotion: role of endothelial prostanoids, Am. J. Physiol. 277 (1999) R1683 R1689. [24] L. Leak, J. Burke, Ultrastructural studies on the lymphatic anchoring filaments, J. Cell Biol. 36 (1968) [25] H. Collin, The ultrastructure of conjunctival lymphatic anchoring filaments, Exp. Eye Res. 8 (1969) [26] L.V. Leak, The structure of lymphatic capillaries in lymph formation, Fed. Proc. 35 (1976) [27] P. Bock, Histochemical staining of lymphatic anchoring filaments, Histochemistry 58 (1978) [28] P.M. Lynch, F.A. Delano, G.W. Schmid-Schonbein, The primary valves in the initial lymphatics during inflammation, Lymph. Res. Biol. 5 (2007) [29] E. Mendoza, G.W. Schmid-Schonbein, A model for mechanics of primary lymphatic valves, J. Biomech. Eng. 125 (2003) [30] J. Trzewik, S.K. Mallipattu, G.M. Artmann, F.A. Delano, G.W. Schmid-Schonbein, Evidence for a second valve system in lymphatics: endothelial microvalves, FASEB J. 15 (2001)