K_ATP channels in lymphatic function

Abstract

K_ATP channels function as negative regulators of active lymphatic pumping and lymph transport. This review summarizes and critiques the evidence for the expression of specific K_ATP channel subunits in lymphatic smooth muscle and endothelium, the roles that they play in normal lymphatic function, and their possible involvement in multiple diseases, including metabolic syndrome, lymphedema, and Cantú syndrome. For each of these topics, suggestions are made for directions for future research.

INTRODUCTION

Widely expressed throughout the body, K_ATP channels are voltage-independent, K⁺-selective channels that are activated by intracellular ADP and inhibited by intracellular ATP, thereby coupling the electrical activity of cells to their metabolic state (1, 2). Mammalian K_ATP channels are heterooctameric complexes of inwardly rectifying K⁺ channel (Kir6) pore-forming subunits and regulatory sulfonylurea receptor (SUR) subunits. Two pairs of adjacent genes (ABCC8, KCNJ11, and ABCC9, KCNJ8), on human chromosomes 11p and 12p, each encode one pair of subunits—SUR1, Kir6.2, and SUR2, Kir6.1, respectively (1). Alternate splicing of ABCC9 produces two major protein variants, SUR2A and SUR2B that differ in their final C-terminal sequences (3, 4). The expression of different combinations of Kir6 and SUR subunit genes in different cell types results in distinct K_ATP channel properties and functional roles in a given tissue (1, 4).

The understood roles of K_ATP channels in arteries/arterioles are of relevance to the present chapter on lymphatic K_ATP channels. Much of the literature suggests that K_ATP channels normally have low activity in arterial smooth muscle (SM) but become activated under ischemic conditions (1), leading to arterial SM hyperpolarization and dilatation, which in turn increases flow and oxygen supply to the ischemic tissue (4). K_ATP channels also contribute to the action of vasodilators that stimulate cAMP-mediated signaling (4). The literature on K_ATP channels in lymphatic vessels is less extensive, but almost all of it suggests that the channels are fundamentally similar to those in vascular smooth muscle and that their activation is a negative regulator of active lymphatic pumping and lymph transport. A discussion and synthesis of that literature, with speculation on the homeostatic significance of K_ATP channels in lymphatic vessels and suggestions for directions for future research, are the aims of this review.

SIMILARITIES AND DIFFERENCES BETWEEN LYMPHATIC AND ARTERIAL SMOOTH MUSCLE

Understanding the differences in the physiology of the lymphatic and blood vascular systems is critical to a discussion of distinct roles of lymphatic K_ATP channels. A primary function of the lymphatic system is the reabsorption of excess interstitial fluid and its return to the central veins. The system consists of networks of initial lymphatic capillaries intertwined among blood capillary networks in almost every tissue and organ. Lymphatic capillaries are composed of overlapping and porous lymphatic endothelial cell (LEC) junctions that facilitate absorption of excess fluid from the interstitium (5). Each lymphatic capillary network has its own specific anatomy apparently adapted to the function of a particular tissue (6). These networks coalesce into lymphatic precollectors containing luminal bileaflet valves that retard backflow. Precollectors in turn form collecting lymphatics, which also contain one-way valves and are surrounded by one or more layers of spontaneously active lymphatic smooth muscle (LSM) cells. The spontaneous contractions of these collecting vessels, in concert with this valve system, allow lymph to be actively transported “uphill” against adverse hydrostatic pressure gradients (7, 8). The contraction frequency of these vessels is exquisitely sensitive to changes in luminal pressure, enabling active lymphatic transport to be matched to the filling state of the lymphatic capillaries (9). Lymph transport can also be facilitated by passive compression of lymphatic vessels by adjacent tissue movement, e.g., skeletal muscle contraction, but the importance of active lymphatic pumping is illustrated by the observation that it accounts for 2/3 of lymphatic transport in the lower legs of humans during quiet standing (10).

Ion Channels

LSM exhibits significant differences from arterial SM in its electrophysiological properties. LSM cells fire spontaneous action potentials (APs) that are required to initiate large-amplitude lymphatic contractions, with the entrainment of APs among LSM cells (11–13) facilitating the effective propulsion of lymph. The membrane potential (V_m) of LSM is not stable but exhibits a gradual, spontaneous depolarization (i.e., a pacemaking potential) during the diastolic phase of the lymphatic contraction cycle (Fig. 1), bringing V_m to a threshold to trigger AP firing (14, 15); the cycle repeats after repolarization. In contrast, most arterial SM cells are quiescent with a stable resting V_m (16, 17), although some can be induced to fire APs under special conditions (18). The complement of ion channels that interacts to generate lymphatic pacemaking potentials has not been fully resolved, but substantial evidence suggests that L-type voltage-gated channels (VGCCs) carry current for the upstroke of the lymphatic AP and mediate much of the Ca²⁺ entry responsible for contraction (19–22), whereas background and voltage-gated K⁺ currents set the resting V_m and facilitate repolarization (14, 23, 24). Two studies have speculated that LSM pacemaker potentials may be driven by interstitial cells, but the evidence to date is only anatomical (25, 26). Other studies have suggested that T-type VGCCs in LSM cells are important for diastolic depolarization (14, 27), but a recent study of lymphatic vessels from Cav3.1^−/−, Cav3.2^−/− and double Cav3.1^−/−; Cav3.2^−/− knockout mice indicates that T-type VGCCs play little or no role in lymphatic pacemaking, at least in that species (28). Roles for other cation channels in LSM diastolic depolarization have also been postulated, including hyperpolarization-activated, cyclic nucleotide-gated (HCN) channels, based on their involvement in pacemaking in other cell systems; however, a role for HCN channels in lymphatic pacemaking under physiological conditions is controversial (29–32). Voltage-gated sodium channels appear to be involved in some species under some conditions (14, 33, 34) but they are not obligatory for the LSM AP under all conditions. Smooth muscle-specific knockout of the calcium-activated Cl⁻ channel anoctamin1 (ANO1) reduces the speed of diastolic depolarization, indicating a contribution of ANO1 to diastolic depolarization. However, ANO1-deficient LSM cells retain a residual level of spontaneous AP firing (35), suggesting the additional involvement of at least one more channel. Neither ANO1 nor most TRP cation channels are intrinsically mechanosensitive (36), so one explanation for spontaneous depolarization is that the channels providing depolarizing current are activated by inositol 1,4,5-trisphosphate receptor (IP₃R)-mediated Ca²⁺ release downstream from mechanosensitive G-protein-coupled receptors (GPCRs) (15, 21, 35, 37). Potentially different types of ion channels may interact to initiate pacemaking in lymphatic vessels of different species (Fig. 1), explaining some of the differences between the studies cited earlier. Whether K_ATP channels contribute to, or modulate, lymphatic pacemaking is an important issue, as considered in depth below.

Figure 1.

Ionic mechanisms of lymphatic pacemaking. Spontaneous diastolic depolarization triggers the opening of L-type voltage-gated channels (VGCCs) that result in the firing of action potentials (APs). Increased pressure increases pacemaking frequency by increasing the rate of diastolic depolarization through regulation of one or more cation channels and the Cl⁻ channel, anoctamin1 (ANO1), which collectively contribute to depolarizing current. After an AP fires, K⁺ channels repolarize the membrane and the cycle repeats. Calcium entry associated with each AP generates a twitch contraction that propels lymph centrally; lymph flow is unidirectional due to the presence of one-way luminal valves. *Diastolic depolarization phase of the AP; dotted line indicates threshold for AP firing. EDD, end diastolic diameter; ESD, end systolic diameter; LSM, lymphatic smooth muscle.

Contractile Proteins

LSM also exhibits substantial differences from arterial SM in the expression of contractile proteins. Like arterial SM, myosin light chain 20 (MLC₂₀) phosphorylation is the predominant regulator of lymphatic muscle contraction, achieved by the balance of myosin light chain kinase/myosin light chain phosphatase (MLCK/MLCP) activity (38, 39); for this reason, lymphatic muscle is often referred to as a type of smooth muscle. Like arteries, lymphatic collectors exhibit pressure-induced basal vascular tone that can be increased or decreased by appropriate agonists. But unlike arteries, LSM cells express both smooth and striated muscle protein isoforms (40). For example, in rat, the thoracic duct (the central lymphatic trunk) expresses smooth and striated muscle isoforms of actin, tropomyosin, and myosin heavy chain (40–42). The smooth and striated muscle proteins in LSM cells are not organized into defined sarcomeres, but appear to be more aligned than the nonstriated contractile proteins in arterial SM (40). Lymphatic muscle may also express troponin C (43), which in the blood vascular system is only expressed in renal afferent arterioles (44)—vessels with the fastest constrictions of any arteries to pressure elevation (45). It is perhaps not a coincidence that LSM has a substantially higher shortening velocity than arterial SM or phasic venous SM, with maximal unloaded velocities approaching values achieved by cardiac muscle (46). The combination of striated and smooth muscle proteins presumably explains how lymphatic vessels can develop and regulate both sustained tone and phasic contractile activity, although this hypothesis has not been experimentally tested by determining the functional consequences of deleting striated muscle isoforms from lymphatic vessels. Nevertheless, for the reasons stated, some investigators have suggested that LSM may be a hybrid muscle type between cardiac muscle and arterial SM (40, 43, 46, 47).

EXPRESSION OF K_ATP CHANNEL ISOFORMS IN LYMPHATIC SM AND ENDOTHELIUM

Multiple studies provide evidence that LSM cells express Kir6.1 and SUR2B K_ATP channel subunits—the same isoforms predominantly expressed in arterial SM (4). Molecular evidence for Kir6.1 and SUR2B expression in the lymphatic system comes primarily from PCR measurements of collecting lymphatic vessels of various species, including mouse, rat, guinea pig, and human (see Table 1). Published PCR analyses are almost exclusively performed on lysates from whole lymphatic vessels, and thus potentially reflect the expression of K_ATP channel isoforms not only in LSM cells and LECs but in multiple other cell types found in the lymphatic wall, including dendritic cells, macrophages, mast cells, neurons, adipocytes, and adventitial fibroblasts. The evidence for expression of other K_ATP channel subunits, i.e., Kir6.2 and SUR1, in lymphatic vessels is complicated for this same reason. For example, Kir6.2 was not detected in guinea pig mesenteric lymphatics (62), but was detected in whole vessel lysates of human thoracic duct, although at much lower levels than Kir6.1 (24). Kir6.2, Kir6.1, and SUR2B transcripts were all detected in whole popliteal lymphatics from mouse, but only Kir6.1 and SUR2B were detected in FACS-purified popliteal LSM cells (66), suggesting that Kir6.2 message may have come from another cell type. Garner et al. (68) reported evidence for the expression of Kir6.2 in rat mesenteric lymphatics, in addition to Kir6.1 and SUR2B, using both PCR analysis and immunostaining of FACS-purified LSM cells. However, immunopositivity for Kir6.2 was only found in a subset of LSM cells. SUR1 subunits were detected by PCR but only at high cycle thresholds, and no SUR1 immunoreactivity was detected in LSM cells (68). In patch-clamp studies of FACS-purified LSM cells, the same investigators (68) found that the single-channel conductance for cromakalin-activated K⁺ current was 46 pS, compared with values of 29 pS for homotetrameric Kir6.1 channels and 60 pS for homotetrameric Kir6.2 channels measured in expression systems (73). This intermediate conductance level suggests that at least some LSM cells may express heterotetrameric Kir6.1/Kir6.2 channels, which also would be consistent with the postulated hybrid striated/smooth muscle phenotype of LSM (as Kir6.2 is the isoform predominantly expressed in cardiomyocytes). The expression of Kir6.1/Kir6.2 channels in LSM could result in enhanced sensitivity to intracellular ATP (74) and larger whole cell K_ATP currents and hyperpolarization (compared with Kir6.1 homotetrameric channels) when the channels are activated. It remains to be determined the extent to which Kir6.2 expression in LSM cells may be tissue- and species-specific.

Table 1.

Studies examining the roles of K_ATP channels in lymphatic vessel function

Species	Bed	KATP Isoforms	Preparation(s)*	[Inhibitor]	[Activator]	Effects on Lymphatic Vessel Function	References#
Guinea pig	Mesentery		En face unpressurized vessel-Vm	10 μM GLIB		GLIB had no effect on resting Vm of LECs but depol. LSMCs by 9 mV; CROM had no effect of LEC Vm; GLIB had no effect on ACh-mediated LEC biphasic hyperpol./depol. but blocked hyperpol. in LSMCs	(48)
Guinea pig	Mesentery		Unpressurized vessel-Vm	10 μM GLIB	0.1–10 μM CROM	CROM hyperpol. LSMCs, blocked by GLIB; GLIB inhibited 90% of ACh/NO-induced hyperpol.; GLIB blocked hyperpol. by isoprenaline, forskolin	(49)
Rat	Mesentery		Isolated vessel-contraction	0.1, 1 μM GLIB	0.1–10 μM PIN	PIN inhibited spont. contractions; GLIB blocked PIN effect; GLIB had no effect on spont. contractions	(50)
			Unpressurized vessel-Vm				(51)
Mouse (ddY)	Iliac efferent		Isolated vessel-contraction	1, 10 μM GLIB		B16-BL6 melanoma cells released factor(s) with inhibitory action on lymphatic pump that was partially blocked by GLIB	(52)
Mouse (ddY)	Iliac efferent		Isolated vessel-contraction	1 μM GLIB		GLIB suppressed inhibitory responses to PTH	(53)
Cow	Mesentery			5 μM GLIB		GLIB attenuated acidosis-induced dilation	(54)
Guinea pig	Mesentery		Unpressurized vessel-Vm; in vivo-contraction	0.1, 10 μM GLIB		GLIB caused 4 mV depol. GLIB attenuated 5-HT-induced inhibition of FREQ and hyperpol. (as did PKA inhibitor)	(55)
Rat	Iliac		Isolated vessel-contraction	1 μM GLIB		1 μM GLIB did not affect inhibition of contraction by the RhoK inhibitor Y-27632	(56)
Guinea pig	Mesentery		Unpressurized vessel-Vm	10 μM GLIB		GLIB depolarized LSMC layer; GLIB blocked hyperpol. caused by PAR2 agonists trypsin, SLIGRL-NH2 and hyperpol. by PGE2 and iloprost	(57)
Rat	Iliac afferent		Isolated vessel-contraction	1 μM GLIB		GLIB attenuates ATP- and adenosine-induced inhibition of contractions	(58)
Guinea pig	Mesentery		Ex vivo-perfused	10 μM GLIB		GLIB blocks CGRP-mediated hyperpol.	(59)
Guinea pig	Mesentery		Ex vivo-perfused	1 μM GLIB		GLIB attenuated the actions of PGE2 and iloprost	(60)
Rat	Thoracic duct		Isolated vessel-contraction	1 μM GLIB		GLIB attenuates the effect of l-arginine on FPF	(61)
Guinea pig	Mesentery, iliac	Kir6.1, SUR2B; no Kir6.2	PCR-whole vessel; ex vivo-perfused vessel-contraction;unpressurized vessel-Vm	1, 10 μM GLIB		GLIB blocks VIP-induced hyperpol.; GLIB blocks VIP-induced inhibition of contraction frequency	(62)
Guinea pig	Mesentery	Kir6.1 SUR2B	PCR-whole vessel; ex vivo-perfused vessel	1, 10 μM GLIB	CROM IC50 = 126 nM frequency; CROM IC50 = 322 nM hyperpol.	10 μM GLIB reverses NO-induced inhibition of pumping;CROM blocks pumping GLIB rescues impaired contractions in TNBS-induced ileitis	(63)
Human	Thoracic duct	Kir6.1, Kir6.2	PCR-whole vessel; isometric strips-Vm, contraction	10 μM GLIB	1 μM PIN	GLIB depol. and initiates spont. activity in some quiescent vessels; GLIB increases FREQ of spont. APs, contractions and depolarizes, reduces AMP; PIN silences spont. contractions, hyperpol. by 25 mV	(24)
Rat	Mesentery		Isolated vessel-contraction	1, 10 μM GLIB		10 μM GLIB increases basal FREQ, decreases EF 10 μM GLIB rescues lower FREQ, FPF in metabolic syndrome model	(64)
Rat	Thoracic duct		Isolated vessel-contraction	1 μM GLIB		Hemorrhagic shock impairs lymphatic contraction through NO; GLIB attenuates NO action	(65)
Mouse (WT, Kir6.1−/−, Kir6.1 GOF, LEC, LMC reporters)	Popliteal, inguinal-axillary	Kir6.1, SUR2B in LMCs;+Kir6.2 in whole vessel; Kir6.1 without SUR in LECs	PCR-whole vessel (inguinal-axillary); PCR-sorted LMCs, LECs (inguinal-axillary);isolated vessel-contraction	10 μM GLIB	PIN IC50 = 1.3 × 10−7 M for WT frequency;	GLIB does not significantly alter basal FREQ, AMP, FPF PIN hyperpol. and stops spont. APs; reversed by GLIB; PIN stops contractions PIN IC50 shifted to 8.3 × 10−5 M in Kir6.1−/− vessels;PIN IC50 shifted to 7 × 10−4 M in SUR2[STOP] vessels Expression of Kir6.1 GOF mutant channels in SM inhibits cont. AMP by 70%–90% at various pressures, inhibits FREQ by 40%, inhibits FPF by 40%–70% at various pressures; hyperpol. LSM by 11 mV	(66)
Human	Mesentery		Vm-iso vessel	10 μM GLIB		PIN hyperpolarizes and stops spont. APs, reversed by GLIB	(66)
Cow	Mesentery efferent		Isometric rings	10 μM GLIB		Cocktail of ChTx, Apamin, and GLIB partially blocks H2S-mediated relaxation	(67)
Rat	Mesentery	Kir6.1, Kir6.2, SUR2 in vessels, LMCs	Whole vessel and sorted LMCs-PCR and immunostaining; isolated LMCs-patch-clamp; isolated vessel-contraction; in vivo-diameter, flow	5 μM GLIB	CROM IC50 = 0.3 μM freq.; MINOX IC50 = 0.6 μM freq.; Diazoxide IC50 = 10 μM freq.;	CROM activates K+ current (cell attached patch clamp of LMCs);CROM inhibits lymphatic flow in vivo; Single-channel conductance consistent with Kir6.1/6.2 heteromeric channels in LSM cells	(68)
Mouse(WT, Kir6.1−/−,SUR2[stop]	Popliteal		Isolated vessel-contraction	0.1–10 μM GLIB		10 μM GLIB does not reverse NO-induced inhibition of FREQ.; NO inhibits contractions in Kir6.1−/− vessels; 10 μM GLIB has off-target effects on FREQ, AMP, EDD	(69)

Additional studies on lymph nodes not included in this table show that GLIB partially blocks the inhibitory effects of heparin, acetylcholine (ACh), sodium nitroprusside (SNP) on bovine lymph capsule contraction (70–72). AMP, amplitude; CROM, cromakalim; DIAZ, diazoxide; depol., depolarization; EDD, end diastolic diameter; EF, ejection fraction; FPF, fractional pump flow; FREQ., frequency; GLIB, glibenclamide; GOF, gain-of-function; hyperpol., hyperpolarization; LEC, lymphatic endothelial cell; LMC, lymphatic muscle cell; LSM, lymphatic smooth muscle; MINOX, minoxidil; PIN, pinacidil; SM, smooth muscle; spont., spontaneous; vip, vasoactive intestinal peptide; WT, wild type.

^*All V_m measurements made in LSM cells except Ref. 48;

#listed in chronological order.

Are K_ATP channel subunits expressed in LECs? Functional K_ATP channels are known to be expressed in blood vascular endothelium (75). Hypertension induced by high-salt diet + L-N^G-nitro arginine methyl ester (l-NAME) administration is exacerbated by endothelial cell (EC)-specific deletion of Kir6.1, and mice deficient in both Kir6.1 and apolipoprotein E (ApoE) develop more atherosclerotic lesions than ApoE^−/− mice alone (76). These results suggest that endothelial K_ATP channels aid in protection against hypertension and atherosclerosis. The activation of a K⁺ conductance hyperpolarizes ECs of arteries/arterioles, which is transmitted through myoendothelial gap junctions to the overlying SM layer to cause vasodilatation (77); however, this mechanism does not appear to operate in the lymphatic system. In lymphatic vessels, the LEC and LSM layers do not exhibit strong electrical coupling (11, 48, 78, 79) as in arteries (80); rather, the V_m of LECs is stable at ∼−70 mV (11, 48, 81), while simultaneously the V_m of LSM cells ranges from −35 to −50 mV and oscillates with the firing of APs (11, 22, 82). The K_ATP channel opener (KCO) cromakalim (CROM) did not elicit hyperpolarization of the endothelium of guinea pig mesenteric lymphatics, even though it hyperpolarized the overlying LSM layer, leading von der Weid and van Helden (48) to conclude that lymphatic endothelium does not express functional K_ATP channels. Only a single study has specifically examined the expression of K_ATP channel isoforms in LECs (Table 1). Message for Kir6.1 channels was detected in the absence of message for an SUR subunit in FACS-purified LECs from mouse inguinal-axillary lymphatics (66). This result suggests that only nonfunctional K_ATP channels may be expressed in mouse lymphatic endothelium. Further support for this idea is that overexpression of gain-of-function (GOF) Kir6.1 subunits selectively in mouse LSM caused a profound inhibition of contraction, whereas selective expression in the LEC layer was without significant effect (66). Again, such evidence may apply only to guinea pig mesenteric and mouse popliteal lymphatics and similar experiments need to be carried out on lymphatic vessels from other regions and species before concluding that K_ATP channels are not active in LECs.

EVIDENCE FOR FUNCTIONAL K_ATP CHANNELS IN LYMPHATIC VESSELS

von der Weid and van Helden (48) were first to demonstrate a functional effect of a K_ATP channel activator in lymphatic vessels, where CROM evoked hyperpolarization of LSM in unpressurized strips of guinea pig mesenteric lymphatic vessels; CROM (5 µM) produced 17 mV hyperpolarization, which was almost completely blocked by the K_ATP channel inhibitor glibenclamide (GLIB, 10 µM). Mizuno et al. (50) subsequently showed that K_ATP channel openers (KCOs) inhibit lymphatic contractile function. In pressurized segments of rat mesenteric lymphatic vessels studied ex vivo, 3 × 10⁻⁷ M CROM produced significant slowing of contraction frequency, whereas a 10-fold higher concentration completely stopped spontaneous contractions; these effects of CROM were blocked by GLIB (10 µM). These findings have subsequently been confirmed by a number of investigators (see Table 1) using either CROM or other KCOs, including pinacidil (PIN) or minoxidil. For mouse popliteal lymphatics studied ex vivo, the IC₅₀ for inhibition of contraction frequency by PIN was 1.3 × 10⁻⁷ M. For rat mesenteric lymphatics studied in vivo, the IC₅₀ for inhibition of frequency was 3 × 10⁻⁷ M for CROM and 6 × 10⁻⁷ M for minoxidil (68). In contrast, the IC₅₀ for diazoxide, which has a preferential affinity for Kir6.2/SUR1 channels (83), was shifted some 20-fold to the right of that for CROM. It is interesting that KCOs in these studies appear to act almost exclusively on lymphatic contraction frequency (68, 69), with a lesser or inconsistent effect on contraction amplitude and tone (see examples in Refs. 50, 68, and 69 and discussion in Ref. 69), perhaps pointing to a selective contribution of K_ATP channels to lymphatic pacemaking rather than to modulating the calcium influx associated with contraction.

The activation of K_ATP channels by high concentrations of KCOs can markedly hyperpolarize LSM V_m (49, 66) and completely block lymphatic contractions, but they begin to inhibit AP firing at much lower concentrations. In both mouse popliteal and human mesenteric lymphatics, 1 µM PIN stopped spontaneous AP firing, in many cases with no overt hyperpolarization beyond the peak negative value of V_m in the pacemaking cycle (66). Thresholds of 100–300 nM for PIN, CROM, and minoxidil are reported to inhibit lymphatic contractions in mouse popliteal (66), rat mesenteric (50, 68), and guinea pig mesenteric lymphatics (63), respectively. Although CROM and PIN are considered to be relatively specific KCOs, off-target effects for both compounds on other channels have been reported, but only at concentrations higher than 1 µM (84–86). Accordingly, while the IC₅₀ of PIN for the contraction frequency of mouse popliteal lymphatics was shifted >100-fold to the right in vessels from Kir6.1 or SUR2 knockout mice compared with wild-type (WT) controls, PIN produced significant elevations in lymphatic tone at concentrations ≤1 µM in Kir6.1 and SUR2 knockout vessels (66), consistent with an effect on a target other than K_ATP channels. Collectively, these observations suggest that K_ATP channels contribute some degree of hyperpolarizing current to the balance of currents regulating diastolic depolarization in LSM (Fig. 1), and that even slight K_ATP channel activation can retard AP firing.

K_ATP CHANNELS IN BASAL LYMPHATIC FUNCTION

Although the aforementioned studies demonstrate that functional K_ATP channels are present in lymphatic vessels and that their activation by KCOs leads to LSM hyperpolarization and inhibition of spontaneous lymphatic APs and contractions, the extent to which K_ATP channels are involved in basal (physiologic) lymphatic function is not as clear. Evidence for K_ATP channels normally exerting some inhibitory effects on lymphatic pacemaking and contraction is provided by the effects of GLIB on LSM V_m and on spontaneous contraction frequency. GLIB consistently causes depolarization of LSM, as demonstrated in multiple studies by von der Weid and coworkers (48, 49, 51, 55, 59). For example, 10 µM GLIB depolarized the LSM layer of guinea pig unpressurized mesenteric lymphatics by 5–9 mV (48, 49). Telinius et al. (24) found that 10 µM GLIB increased the frequency of spontaneous contractions in isometric human thoracic duct segments by 2.5-fold and initiated spontaneous contractions in 4 of 7 preparations that were quiescent. GLIB significantly increased the frequency of mouse popliteal lymphatic contractions (and decreased contraction amplitude) at concentrations ≥1 × 10⁻⁷ M (69). However, other studies have reported no significant effect of GLIB on spontaneous lymphatic contractions. For example, Mizuno et al. (50) reported that neither 1 nor 10 µM GLIB altered the frequency of spontaneous contractions of rat mesenteric lymphatics ex vivo. Only two studies have assessed the contractile behavior of lymphatic vessels in K_ATP channel knockout animals. Davis et al. (66) did not detect significant differences in either the frequency or amplitude of spontaneous contractions of popliteal lymphatic vessels in mice with deletion of Kir6.1 or SUR2, compared with WT controls; however, there was a trend for Kir6.1^−/− vessels to have slightly a lower frequency and amplitude at most pressures, resulting in lower calculated pump flows. Likewise, Kim et al. (69) detected very little differences in the properties of spontaneous contractions between WT and Kir6.1- or SUR2-deficient mouse popliteal vessels, except that the maximal passive diameters of mouse popliteal lymphatic vessels in Kir6.1^−/− mice were consistently lower than those in WT mice. In the Kir6.1 and SUR2 global knockout mice, it is possible that compensatory upregulation of other ion channels might have masked the effects of loss of K_ATP channel function.

There are multiple possible explanations for the discrepancies in the effects of GLIB on basal lymphatic contractions. First is the obvious possibility of tissue and/or species differences. A second explanation is that GLIB may also have off-target effects, particularly at the relatively high concentration of 10 µM most commonly used (Table 1). Patch-clamp studies have demonstrated that GLIB has inhibitory effects on various other ion channels at that concentration, including L-type VGCCs (87–89) and Cl⁻ channels (90, 91); however, inhibitory effects on either of these channels would be predicted to slow lymphatic pacemaking (Fig. 1), opposite to what is observed in studies of lymphatic vessels. Kim et al. (69) tested concentration-response curves for GLIB in mouse popliteal lymphatics and found that concentrations of GLIB ≥ 1 × 10⁻⁶ M decreased contraction amplitude and increased frequency in WT mice, but decreased frequency with little effect on contraction amplitude in Kir6.1^−/− and SUR2^−/− lymphatic vessels; this mixture of effects suggests some off-target actions, at least in the mouse. However, the decrease in frequency was opposite to what was observed in WT vessels, suggesting that if anything, the effect of 10 µM GLIB on contraction frequency might have underestimated the contribution of K_ATP channels to lymphatic contraction frequency in this (69) and other studies. A third explanation for why only some studies find an effect of GLIB relates to the fact that most isolated lymphatic vessel preparations have no imposed pressure gradient for flow (50, 66, 69), whereas in vivo studies (68), or vessels perfused from one end in situ (55, 63), likely have higher basal flow and consequent nitric oxide (NO) production. Several studies (see Nitric Oxide) have suggested that NO production will lead to K_ATP channel activation such that GLIB would be predicted to have a greater effect on contraction in preparations with luminal flow.

A fourth explanation for why only some studies find an effect of GLIB on lymphatic contraction could relate to the oxygenation conditions. K_ATP channels in arteries and myocardium are activated under conditions of hypoxia or ischemia (1, 4), and lymphatic K_ATP channels presumably exhibit the same behavior. Thus, the use of highly oxygenated solutions in many ex vivo experiments on isolated lymphatic vessels (50, 55, 63) might suppress endogenous K_ATP channel activity. Arterial SM cells are in close contact with highly oxygenated arterial blood and therefore likely exposed to higher Po₂ levels (92, 93). In contrast, most lymphatic vessels are in contact with the interstitium, where Po₂ levels are often between 5 and 30 mmHg (94–96), depending on the vascular bed and the proximity of the lymphatic collectors to arteries/arterioles (93). Indeed, many microvascular studies using intravital microscopy have traditionally used deoxygenated solutions to more accurately approximate conditions in the interstitium (97, 98). Thus, inappropriately high oxygenation levels used in some lymphatic studies may have led to an underestimation of the basal activity of K_ATP channels, which otherwise might be predicted to have higher activity than channels in arterial SM. The pH level of the solutions used in lymphatic studies might also be important since interstitial pH may be slightly lower than that of blood and acidosis-induced inhibition lymphatic contractile activity has been shown to be mediated by K_ATP channels (54). These are issues that need further investigation.

ROLES FOR K_ATP CHANNELS IN MEDIATING LYMPHATIC VESSEL RESPONSE TO AGONISTS

The active lymph pump is modulated by numerous mechanisms, including neural, circulating, and locally produced factors that may act either to inhibit or enhance lymphatic contractile properties. The evidence to date suggests that many of these factors act in part or entirely through their effects on K_ATP channels (Table 1).

Nitric Oxide

Nitric oxide (NO) is the most studied of the various factors that modulate lymphatic contractions. In arteries, shear stress enhances the production of endothelium-derived NO, which diffuses to the overlying smooth muscle layer and activates various signaling pathways to inhibit basal (pressure-induced) myogenic tone and cause vasodilation (77); this process is often called flow-induced (or flow-mediated) dilation. Other EC-derived factors can contribute to flow-induced dilation but NO is the most prominent and ubiquitous factor. Gashev et al. (99) were the first to demonstrate that shear stress-induced production of NO inhibits spontaneous lymphatic contractions. In isolated, pressurized lymphatic collectors from rat mesentery, the pressure gradient between the cannulation pipettes was varied to impose different levels of forward flow while maintaining midpoint pressure approximately constant with resistance-matched pipettes. Graded increases in the pressure difference used to drive flow led to progressive inhibition of spontaneous contractions, marked by decreases in frequency, amplitude, and tone in response to flow. The responses were blocked by the endothelial nitric oxide synthase (eNOS) inhibitor l-NAME (99), indicating that they were mediated by NO. These results have been reproduced in a number of subsequent studies (100–103), with some showing that mast cell- or LEC-derived histamine release mediates a small component of the response to flow in certain lymphatic vessels (104, 105).

Inhibition of the active lymph pump is not the only effect of NO on lymphatic contractile properties. The pulsatile production of NO associated with spontaneous lymphatic contractions can enhance active lymph transport. The mechanism is as follows: NO is preferentially produced at the LEC valve regions, where shear stress is highest near the open tips of the valve leaflets (106), and its production oscillates with the lymphatic contraction cycle, as revealed by measurements with NO-sensitive electrodes (107). Under these conditions, NO production is more modest than under imposed flow and leads to a decrease in the time for relaxation following lymphatic systole, with little or no effect on contraction frequency (108). The molecular mechanism is not known, but accelerated relaxation allows more time for filling during lymphatic diastole, which enhances the contraction amplitude and ejection fraction of the subsequent contraction cycle due to a mechanism similar to the Frank–Starling effect in the heart (9). Lymph flow, calculated as the product of frequency and ejection fraction, rises if ejection fraction increases without a fall in frequency (108). Thus, NO appears to have biphasic effects on lymphatic contractions, enhancing the active lymph pump during pulsatile flow and inhibiting it at higher, nonpulsatile, flow.

NO production can also be stimulated by multiple agonists acting on the lymphatic endothelium. Mizuno et al. (109) first demonstrated that acetylcholine (ACh), or the NO donor sodium nitroprusside (SNP), inhibited spontaneous contractions and tone of isolated iliac afferent lymphatics from rat. The effects of both compounds were prevented by denudation of the endothelium. In vessels with an intact endothelium, the effects of ACh, but not SNP, were blocked by the pan-NOS inhibitor N^G-nitro-l-arginine (l-NNA) (109). These findings were consistent with those from eNOS^−/− mice, in which ACh-induced inhibition of spontaneous popliteal lymphatic contractions was lost (110). NO donors do not elicit hyperpolarization of the LEC layer (48, 51), but NO can readily diffuse to the LSM layer. Indeed, ACh, SNP, and another NO donor (DEA-NONOate) all hyperpolarized the LSM layer of guinea pig mesenteric lymphatics and inhibited spontaneous APs (51). Importantly, the hyperpolarizations induced by each of these compounds were substantially attenuated by 10 µM GLIB (51), suggesting that the inhibitory effects were mediated by the activation of K_ATP channels.

As expected from studies on arteries (111), the inhibitory effects of NO are mediated by stimulation of the cGMP/soluble guanylate cyclase (sGC) signaling pathway (49, 51, 63, 112). A number of studies have subsequently showed that K_ATP channels mediate, in part or in whole, the inhibitory effects of NO on lymphatic contractions (49, 58, 63, 65). It should be noted that these studies all used lymphatic vessels from rats or guinea pigs and relatively high concentrations of NO donors [e.g., 100 µM NONOate (63); 50–100 µM SNP (49); 100 µM SNP (51)]. In contrast, Kim et al. (69) found that K_ATP channels did not mediate the inhibitory effects of NO on the spontaneous contractions of popliteal lymphatic vessels from mice. ACh and NONOate each inhibited spontaneous contractions of WT mouse popliteal vessels, with IC₅₀ values of 3 ×·10⁻⁸ M for ACh and 9·× 10⁻⁸ M for NONOate, respectively, on contraction frequency. However, the sensitivities to both ACh and NONOate were comparable in Kir6.1^−/− vessels, with IC₅₀ values of 2·× 10⁻⁸ M for ACh and 9·× 10⁻⁸ M for NONOate, respectively, on contraction frequency. Subsequent protocols showed that NONOate was working through stimulation of cGMP and soluble guanylate cyclase as expected, but that calcium-activated K⁺ channels, rather than K_ATP channels, mediated most (but not all) of the inhibitory effects of NO (69). This study raises questions about whether the role for K_ATP channels in NO-mediated inhibition of lymphatic pacemaking is unique to the rat and guinea pig, and if mouse lymphatics are the exception, or whether low and high concentrations of NO work through different mechanisms and/or on different molecular targets to modulate lymphatic contractions. Some potential mechanisms by which NO may act on K_ATP channels to inhibit lymphatic pacemaking are illustrated in Fig. 2. The effects of NO on human lymphatics do not appear to have been tested to date and it will be important to determine whether they are mediated by K_ATP channel activation—for reasons related to potential therapies for lymphatic contractile dysfunction as discussed in the K_ATP channels and Lymphatic Dysfunction.

Figure 2.

Hypothesis to explain how nitric oxide (NO) and prostanoids act through K_ATP and/or large-conductance Ca²⁺-activated K⁺ channel (BKCa) channels to regulate the lymphatic pacemaking cycle. K_ATP channel activation slows the diastolic depolarization rate of the ionic pacemaker in lymphatic smooth muscle (LSM). In rat, K_ATP channel activity is stimulated by the NO-sGC-cGMP-PKG signaling axis; in mouse, the same pathway activates BKCa rather than K_ATP channels, but produces similar slowing of diastolic depolarization. Although cGMP production is associated with NO-mediated activation of K_ATP channels, the increased channel activity may be mediated either by channel phosphorylation by PKG or through cross-activation of PKA. Prostanoid production leads to increased cAMP levels that also activate PKA. AC, adenylate cyclase; ANO1, anoctamin1; AP, action potential; eNOS, endothelial nitric oxide synthase; iNOS, inducible NO synthase; sGC, soluble guanylate cyclase; VGCC, voltage-gated channel; GPCR, G-protein-coupled receptor.

Other Agonists

Although NO stimulates sGC-cGMP-PKG signaling, K_ATP channel activation appears to be largely, or additionally, through downstream activation of protein kinase A (PKA) (62). NO-sGC-cGMP-PKG signaling might activate PKA through cross-activation of PKA by cGMP (113, 114). Because cellular cGMP levels are normally much lower than cAMP levels, cGMP will compete with cAMP to activate PKA when cGMP levels rise (e.g., when NO is produced). The relative levels of the two cyclic nucleotides are also affected by the activities of phosphodiesterases (e.g., PDE3), which can degrade one nucleotide while being inhibited by the other (115). In LSM, hyperpolarizations evoked by ACh or SNP were inhibited by glibenclamide, by the sGC blocker ODQ and by the PKA inhibitor H-89 (49). SNP also inhibited spontaneous transient depolarizations, which are thought to contribute to the pacemaking potential in LSM (51), and this inhibition was attenuated by H-89 or the structurally unrelated PKA inhibitor KT-5720. The effects of several other agonists that inhibit lymphatic contractions also appear to be mediated by K_ATP channels, including vasoactive intestinal peptide (VIP) (62), calcitonin gene-related peptide (CGRP) (59), ATP (58), parathyroid hormone (53), and β2-adrenoreceptor signaling (49), with each study providing evidence for PKA involvement (Table 1). For example, VIP-induced hyperpolarization of LSM was almost completely blocked by GLIB or by H-89, and H-89 shifted the IC₅₀ for VIP on contraction frequency to the right by ∼10-fold (62). The inhibitory effects of CGRP on spontaneous lymphatic contractions were also prevented by H-89 or GLIB (59). As with KCOs, the effects of these agonists were primarily on the frequency of lymphatic APs or contractions. Thus, a common link between NO or other agonists and inhibition of lymphatic contractions may be PKA-mediated activation of K_ATP channels, as depicted in Fig. 2. However. this conclusion relies heavily on the specificity of H-89, which inhibits at least eight other kinases, and off-target effects are likely as the typical concentration used in the earlier studies (10 µM) well exceeds the IC₅₀ (135 nM) for PKA (116).

Prostanoids

Inflammation is known to compromise lymphatic contractile function, in part through cytokine-induced activation of inducible NO synthase (iNOS), resulting in the excessive generation of NO, and in part through production of multiple vasoactive prostanoids resulting from the metabolism of arachidonic acid by increased cyclooxygenase activity (63, 117–122). Wu et al. (123) observed dilated lymphatics with blunted spontaneous contractions occurring in a guinea pig model of ileitis; the contractile dysfunction was partially reversed by cyclooxygenase-1, -2 (COX-1 and COX-2) inhibition. Subsequent studies by the same laboratory identified PGE₂ and PGI₂ as the major prostanoids mediating lymphatic contractile dysfunction in both rat and guinea pig models of ileitis (60, 121). These prostanoids bind to their respective GPCRs (e.g., EP4 for PGE₂, IP for PGI₂), which couple to adenylate cyclase (AC) through Gs proteins (60), resulting in enhanced production of cAMP. The inhibitory effects of exogenous PGE₂ and PGI₂ were blocked by H-89 or GLIB (60), consistent with roles for both PKA and K_ATP channels. In the guinea pig ileitis model, Mathias et al. (63) also found that the impaired lymphatic contractions were partially rescued by inhibition of either K_ATP channels, iNOS, or sGC. Collectively, these findings suggest that NO and prostanoids generated in these models of ileitis inhibit spontaneous lymphatic pumping through the activation of LSM K_ATP channels.

PKA presumably activates K_ATP channels via phosphorylation of specific residues on one or both of the channel subunits. An initial report suggested that PKA phosphorylates one site on Kir6.1 and two sites on SUR2B (124). However, subsequent mutation of each predicted site on each subunit revealed that Kir6.1 was not phosphorylated by PKA while two residues on SUR2B were critical for mediating the effects of isoproterenol and forskolin, and only one of these (Ser 1381) was phosphorylated (125). K_ATP channels can also be phosphorylated by PKC (126) and PKG (127, 128), which further complicates the interpretation of many lymphatic studies, and thus, the extent to which regulation of K_ATP channels in LSM may be mediated or modified by any or all of these protein kinases remains to be definitively determined. Protein kinase-mediated phosphorylation of Kir6.1 in LSM would presumably require the close proximity of the kinase and Kir6.1, which might be facilitated by scaffolding proteins, as has been established for other ion channel phosphorylation mechanisms (125, 129–132). Such studies typically use a combination of patch clamp protocols, super-resolution microscopy, proximity ligation assays, site-directed mutagenesis, and/or transgenic mice to establish these relationships. As yet, none of these techniques has been applied to LSM in the context of protein kinase-mediated regulation of K_ATP channels.

K_ATP CHANNELS AND LYMPHATIC DYSFUNCTION

Cantú Syndrome

Chronic activation of K_ATP channels in LSM would lead to suppression of spontaneous lymphatic contractions and active lymph transport, and a clear example of SM dysfunction mediated by K_ATP channels is now provided by Cantú syndrome (CS). First described in 1982 by J.M. Cantú (133), and later recognized to have similarity of symptoms to those of chronic, systemic minoxidil use (134), CS is characterized by obvious physical features of congenital hypertrichosis and facial dysmorphia (133, 135). In addition, subjects with CS exhibit multiple cardiovascular abnormalities, including hypotension, cardiomegaly, patent ductus arteriosis, and dilated and tortuous cerebral arteries (135–138), as well as gastrointestinal SM problems that disrupt normal gastrointestinal (GI) motility (135). Over 50% of patients with CS develop lymphedema at some stage, especially in the lower extremities (135, 139), presumably related to hyperactive K_ATP channels in LSM. Other aspects of lymphatic dysfunction in patients with CS, such as localized head and neck edema (140) may be underdiagnosed. It is possible that the pericardial effusion observed in some patients with CS may be associated with depressed lymphatic contractility of the coronary lymphatics, and that CS-related neurological disorders may be related to impaired cerebral lymphatic/glymphatic drainage, as reported for other neurological diseases (141–144).

The cause of CS is now known to be gain-of-function (GOF) mutations in either the Kir6.1 or SUR2 K_ATP channel subunits (145, 146), with over 30 mutations identified thus far (135). Mouse models of CS engineered by knock-in CS mutations into the endogenous KCNJ8 (Kir6.1[V65M]) or ABCC9 (SUR2[A478V]) loci reproduce important features of the syndrome (147). The electrophysiological characteristics of GOF K_ATP channels in arterial SM are as predicted: whole cell patch-clamp recordings reveal an enhanced basal K_ATP conductance in vascular smooth muscle (147–149). In both Kir6.1[V65M] and SUR2[A478V] animals, this causes arterial dilation, lower diastolic and systolic blood pressures, and increased arterial compliance (147), with the severity of cardiovascular symptoms increasing with severity of the molecular GOF (147). The chronic reduction in arterial pressure results in increased sympathetic neural output to the heart (138) and activation of the renin-angiotensin system (150), resulting in high output cardiac hypertrophy. Importantly, most of the cardiovascular abnormalities in mice can be reversed with chronic, systemic GLIB treatment (151).

Lymphatic function has not yet been studied in these knock-in models. However, SM-specific expression of a transgenic GOF Kir6.1 subunit (149) resulted in severe contractile dysfunction in mouse popliteal lymphatics studied ex vivo, where contraction amplitude was reduced by 70%–90% and contraction frequency by 50% compared with those of vessels from control animals (66). The amplitude and frequency of spontaneous contractions were similarly reduced in popliteal lymphatics studied in vivo. V_m recordings from LSM cells in pressurized popliteal vessels revealed that V_m was hyperpolarized by 11 mV on average in Kir6.1 GOF vessels (66). Similar hyperpolarization and impairment of lymphatic contractile function are predicted to be observed in mice with CS-related GOF mutations in Kir6.1 or SUR2B.

Chronic K_ATP channel GOF in arterial SM also leads to more permanent vascular remodeling. GOF mutations in either Kir6.1 or SUR2B resulted in increases in the maximum passive diameters of multiple arteries and increased arterial compliance (147), an effect that does not appear to be reversed by GLIB treatment (151). Whether any similar remodeling of lymphatic vessel diameter occurs in CS mice remains to be tested. Interestingly, the maximal passive diameters of mouse popliteal lymphatic vessels in Kir6.1^−/− mice were consistently lower than those in WT mice (69), i.e., the reverse of what happens in CS mouse blood vessels, indicating that K_ATP channel dysfunction may indeed have novel consequences for lymphatic vessel development.

Lymphedema

K_ATP channels might also contribute to lymphatic contractile dysfunction in patients with more generalized lymphedema—defined as edema with a component caused by lymphatic dysfunction and classified as either primary or secondary. Typically resulting from hereditary mutations in genes controlling lymphatic vessel and valve development, primary lymphedema in CS may be the first example caused by impaired lymphatic contractile function. Secondary lymphedema occurs as a consequence of another pathological process, e.g., congestive heart failure, chronic venous disease, or obesity. The incidence of secondary lymphedema correlates highly with basal metabolic index (BMI). Patients with BMI over 40 have a progressive risk for lymphedema (152), with morbid obesity almost always leading to the development of secondary lymphedema (153). The most common cause of lymphedema in the United States is surgical intervention and/or radiotherapy for breast cancer (154) and 20%–40% of breast cancer survivors develop lymphedema in the affected arm(s) at some point after cancer treatment due, in part, to impairment of active lymphatic transport (155–157). If left untreated, secondary lymphedema can progress through several stages, culminating in extreme interstitial fluid and protein retention, inflammation, and cytokine production, which eventually result in fibrosis and adipose deposition. All these factors can further contribute to deteriorating lymphatic function, which includes both valve incompetency and contractile defects (158–161). The only approved treatment for lymphedema is massage therapy combined with compression stockings or devices, but recent clinical studies have explored the use of anti-inflammatory treatments, including the T cell inhibitor tacrolimus (162), regulatory T cell transfer (163), and the leukotriene B4 inhibitor ketoprofen (164). Although no studies have yet linked K_ATP channel activation to lymphatic contractile dysfunction in secondary lymphedema, part of the rationale for anti-inflammatory therapy is to prevent excessive production of NO and/or eicosanoids through cytokine-induced activation of iNOS and/or COX2. K_ATP involvement in this context hinges on whether NO and prostanoids activate K_ATP channels in human lymphatics.

Obesity and Metabolic Syndrome

In addition to its role in the reabsorption of excess interstitial fluid, the lymphatic system also plays critical roles in dietary fat absorption and immune function. Metabolic syndrome is defined by a constellation of clinical findings that includes obesity, insulin resistance, dyslipidemia, and hypertension. The strong association between these pathologies and lymphatic dysfunction has been highlighted in a recent series of review articles (165–171). At least three lines of evidence support a causal link between lymphatic dysfunction and the development of obesity and/or metabolic syndrome.

First, several key genes and signaling pathways that control lymphatic development are linked to obesity and insulin resistance. Prox1 is the key transcription factor that determines lymphatic identity (172). Certain Prox1 gene variants are associated with increased incidences of hyperlipidemia, obesity, and type 2 diabetes in humans (173–176). Prox1^−/− mice die at embryonic day 14.5 (E14.5) when the lymphatic system is developing (177). Most haplodeficient Prox1 mice die shortly after birth but those that survive to adulthood develop adult-onset obesity, with mispatterning of the lymphatic vasculature, defective barrier function, and impaired lymphatic transport (178, 179). Importantly, restoration of lymphatic function by Prox1 re-expression reverses this adult-onset obesity (179). Similarly, Vegfr3, the critical growth factor controlling lymphatic development (180, 181), and Foxc2, a transcription factor controlling lymphatic valve development (182–184), are also linked to the development of obesity and insulin resistance (185–189).

A second link between the lymphatic system and obesity/metabolic syndrome relates to the role of the lymphatic system in fat absorption and fat deposition (190, 191). Dietary fat absorption occurs through the intestinal villi, each of which contain a central lymphatic lacteal capillary with highly permeable, discontinuous, button-like LEC-LEC junctions. Lacteals drain into networks of lymphatic precollectors and collectors with less permeable zipper-like junctions (5) that transport chylomicrons through the mesentery to the intestinal lymph nodes (190). These intestinal lymphatics play a key role in the deposition of visceral adipose tissue (VAT), which, when accumulated in excess, is associated with the production of adipokines and fatty acids, leading to impaired arterial SM function, NO bioavailability, and insulin resistance (192). Adipose tissue is known to accumulate around sites of lymph leakage (178, 193, 194) and in mice, a high-fat diet stimulates the growth of mesenteric lymphatics with abnormal permeability, promoting leakage of chylomicron-rich lymph that stimulates deposition of VAT (193). Conversely, elevated levels of VEGF-A can convert button junctions in the lacteal network to zipper junctions, leading to impairment of chylomicron uptake and resistance to diet-induced obesity (195). Thus, fat deposition is closely correlated with lymph leakage and is enhanced if lymphatic permeability is abnormally elevated. Enhancing lymphatic function under these conditions may thus protect against diet-induced obesity and insulin resistance (196).

Third, there is accumulating evidence that lymphatic dysfunction may be both a cause and a consequence of the factors that contribute to obesity and metabolic syndrome (197–199). Dysfunction in this context may include several components, such as the enhanced leakiness of lymphatic collectors observed in db/db mice, a genetic model of diabetes/obesity (115), valve back-leak and closure defects in mice fed a Western diet (200), and lymphatic contractile dysfunction in animals fed high-fat or high-fructose diets as models of metabolic syndrome (199, 201–203). Lymphatic contractile dysfunction also occurs in animals with advanced age, with the major effect being a reduction in spontaneous contraction frequency (101, 102, 204). A common theme linking these conditions is metabolic stress and inflammation, and it is therefore likely that K_ATP channel activation may be involved. Metabolic stress may result in increased intracellular ADP/ATP ratio that would directly activate K_ATP channels in LSM (Fig. 3) and NO conversion to reactive oxygen species (ROS) might also facilitate K_ATP channel activation (205–207). Inflammation-related production of NO and prostanoids would inhibit lymphatic pumping through K_ATP channels, as discussed in an earlier section. To date, however, there are only two indications of K_ATP channel involvement in these pathologies: the profound lymphatic contractile dysfunction in the 2,4,6-trinitrobenzene sulfonic acid (TNBS) model of ileitis, due to excessive PGE₂ production, and the impaired lymphatic flow and contractile function in rats fed a high-fructose diet, both of which were rescued by GLIB (63, 64). The extent to which lymphatic contractile function is compromised in other models of metabolic stress, and whether the involvement of K_ATP channels is a common theme, are potentially important issues that require further investigation.

Figure 3.

Summary of mechanisms through which K_ATP channels regulate lymphatic pacemaking. Activation of K_ATP channels through the production of cAMP and cGMP results in a slower diastolic depolarization rate of the ionic pacemaker in lymphatic smooth muscle (LSM), leading to reduced lymphatic contraction frequency and reduced pumping. Gain-of-function mutations in Cantu syndrome cause constitutive activation of K_ATP channels and lymph pump dysfunction, eventually resulting in peripheral lymphedema. K_ATP channels may also be activated to varying degrees by metabolic stress in animal models of metabolic syndrome. AC, adenylate cyclase; ANO1, anoctamin1; AP, action potential; eNOS, endothelial nitric oxide synthase; GOF, gain-of-function; iNOS, inducible NO synthase; NO, nitric oxide; ROS, reactive oxygen species; sGC, soluble guanylate cyclase; GPCR, G-protein-coupled receptor.

An important related issue is whether inhibitors of mitochondrial respiration cause the activation of lymphatic K_ATP channels. This topic has been extensively studied in arterial SM and myocardium (1, 4), but there have been no comparable studies in lymphatic vessels. Multiple conditions related to metabolic stress could lead to K_ATP channel activation, both as an initiating factor and as a feed-forward mechanism to potentiate and prolong lymphatic contractile dysfunction. For example, the activation of K_ATP channels by metabolic stress would result in compromised active lymphatic pumping and transport, potentially leading to a buildup of metabolic products in the interstitium, exacerbating the already elevated ADP/ATP ratio in LSM cells and promoting further activation of K_ATP channels. Once lymphedema begins to develop, increased fibrosis and adipose deposition would result in an increased diffusion distance for O₂ (and CO₂) from arteries to lymphatic collectors, which could further trigger or reinforce K_ATP channel activation. Surprisingly, these topics have not been studied experimentally but are areas ripe for further investigation.

In summary, it is likely that a spectrum of K_ATP channel activation levels, ranging from modest (metabolic syndrome) to severe (Cantú syndrome), leads to the suppression of lymphatic contractile function and impaired lymph transport that contributes to multiple diseases, including obesity, metabolic syndrome, and secondary lymphedema. A better understanding of the contribution of K_ATP channels to these processes could potentially lead to their treatment using sulfonylurea receptor inhibitors such as glibenclamide.

GRANTS

This research was supported by National Institutes of Health Grants R01 HL-141107 and HL-122578 (to M. J. Davis) and R35 HL-140024 (to C. G. Nichols).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

M.J.D. and C.G.N. conceived and designed research; M.J.D., H.J.K., and C.G.N. drafted manuscript; M.J.D., H.J.K., and C.G.N. edited and revised manuscript; M.J.D., H.J.K., and C.G.N. approved final version of manuscript.