Modulation of lymphatic muscle contractility by the neuropeptide substance P

Abstract

Substance P (SP) is a neuropeptide associated with sensory innervation of lymphoid tissue and a suspected modulator of lymphatic function in inflammation. Only a few studies have examined the effects of SP on lymphatic contraction, and it is not clear to what extent SP acts directly on the lymphatic muscle and/or endothelium or indirectly through changes in intraluminal filling pressure secondary to increases in capillary permeability/filtration. We tested the effects of SP on the spontaneous contractions of rat isolated mesenteric lymphatic vessels under isometric and isobaric conditions, hypothesizing that low concentrations would stimulate lymphatic pumping by enhancing lymphatic muscle contraction in a manner complementary to the effect of increased preload. Under isometric conditions, SP (10 nM) dramatically enhanced lymphatic chronotropy and inotropy. Unlike guinea pig lymphatics, SP actions were not blocked by cyclooxygenase or PLA₂ inhibition. In the absence of SP, ramp increases in isometric preload resulted in ×∼1.6 increases in contraction amplitude (Amp) and ×∼1.7 increases in frequency (Freq). SP increased Freq by ×∼2.4, Amp by ×∼1.9, and the Amp-Freq product (AFP) by ×∼3.5. Under isobaric conditions, the pressure elevation from 0.5 to 10 cmH₂O in the absence of SP decreased Amp by ×∼0.6 and increased Freq by ×∼1.8. SP caused a modest increase in Amp, a robust increase in Freq at all pressures, and shifted the AFP-pressure relationship upward and leftward. Therefore, SP has substantial positive inotropic and chronotropic effects on rat lymphatic muscle, improving pump efficiency independent of the effects of preload and broadening of the working range of the lymphatic pump.

substance p (SP) is an 11-amino acid neuropeptide (36) often associated with cells of lymphoid tissue (15, 19, 21, 22, 46, 47). SP is released from enteric nerves, sensory nerves, and inflammatory cells (27, 41, 45, 52). Correspondingly, SP receptors are expressed on endothelium and muscle of blood and lymphatic vessels, on nerves innervating those vessels, and on associated immune and inflammatory cells (31, 36, 44).

SP appears to have both direct and indirect effects on vascular targets. For example, SP is a powerful stimulator of endothelium-dependent nitric oxide production (28), evoking vasodilation and increasing local blood flow. In addition, SP enhances vascular permeability (18, 25) and stimulates the release of cytokines and chemokines (4, 36) that would indirectly elicit vasodilation and increase vascular permeability. These vascular effects of SP would collectively increase flow/pressure in capillaries and post-capillary venules and lead to increases in transcapillary fluid flux. The resulting increases in interstitial fluid volume and interstitial fluid pressure would enhance lymph formation (42, 43) and stimulate lymphatic pump activity through the well-described relationship of lymphatic filling pressure on lymphatic pumping (11, 34, 51).

SP may also have direct actions on collecting lymphatics. In isolated bovine mesenteric lymphatics, concentrations of SP between 1 and 100 nM increased the spontaneous contraction rate (chronotropy) (10). Rayner and Van Helden (39) found that the positive chronotropic effect of SP on isolated guinea pig mesenteric lymphatics was mediated by PLA₂ activation and the production of thromboxane A₂ (TxA₂) by the lymphatic endothelium. There appeared to be little or no direct effect of SP on guinea pig lymphatic muscle. Amerini et al. (2) described the effects of topical SP administration on the lymphatic pump using an in vivo preparation of the rat mesenteric microcirculation. In that study, SP increased basal tone and chronotropy but reduced both systolic and diastolic diameters, leading to a reduction in spontaneous contraction amplitude (Amp). Because lymphatic vessels have, at best, only weak myogenic responses, the net reduction in both systolic and diastolic diameters in the (probable) face of increased lymphatic filling suggested a direct stimulatory effect of SP on rat lymphatic muscle (2). However, based on in vivo measurements, it was difficult to distinguish between the possible direct effects of SP on lymphatic muscle and/or the endothelium and the indirect effects of SP via enhanced lymph formation and subsequent changes in lymphatic filling.

The purpose of this study was to test the extent to which SP has direct effects on rat mesenteric collecting lymphatic vessels. We used isolated, cannulated lymphatic vessels to distinguish between the possible direct and indirect actions of SP as defined above. We used rat mesenteric vessels because of the extensive amount of available data regarding the contractile behavior of these vessels both in vivo (2, 3, 49) and in vitro (11, 12, 51). A unique aspect of our study was to assess the action of SP on collecting lymphatic vessels under both isometric and isobaric conditions. Our results suggest that SP is a powerful, direct modulator of lymphatic vessel contraction, promoting increases in both chronotropy and inotropy.

METHODS

Vessel isolation.

Male Sprague-Dawley rats (180–250 g) were anesthetized with Nembutal (60 mg/kg ip), and a 3- to 4-cm loop of intestine from each animal was exteriorized through a midline abdominal incision. All animal protocols were approved by the University of Missouri Animal Care and Use Committee and conformed to the Public Health Service (PHS) Policy for the Humane Care and Use of Laboratory Animals (PHS Policy, 1996). Collecting lymphatic vessels (90–150 μm inner diameter by 1 to 2 mm in length) were dissected from the mesentery and cleaned of connective tissue and fat in 3-(N-morpholino) propanesulfonic acid (MOPS)-buffered, albumin-supplemented physiological salt solution (APSS) at room temperature. Each vessel was transferred to a wire or pressure myograph for respective isometric or isobaric studies. After dissection was complete, the rat was euthanized with Nembutal (120 mg/kg ic).

Solutions.

APSS contained (in mM) 145.0 NaCl, 4.7 KCl, 2.0 CaCl₂, 1.17 MgSO₄, 1.2 NaH₂PO4, 0.02 EDTA, 5.0 glucose, 2.0 sodium pyruvate, 3.0 MOPS plus 0.5 g/100 ml purified bovine serum albumin. The composition of Ca²⁺-free APSS was identical except that 3.0 mM EDTA was substituted for CaCl₂. Purified albumin was purchased from US Biochemicals (No. 10856). All other chemicals were from Sigma (St. Louis, MO).

Wire myograph methods.

To precisely control the rate of passive force changes, a servo-controlled piezo stack (Burleigh Inchworm, Fishers, NY) was mounted to a small vessel wire myograph (customized model 310A; Danish Myo Technology, Arhus, Denmark). Stress relaxation of the lymphatic vessel was compensated by the servo-control of passive force, as described previously (7). The direct output option of the myograph allowed the frequency response of the control loop to be increased to 200 Hz. The force signal was amplified and filtered (NPI Electronic, Tamm, Germany) before digitization at 20 Hz with a PCI 6030e analog-digital interface (National Instruments, Austin, TX) in a Pentium 4 computer. The piezo drive was controlled by an EXFO controller (model IW800; EXFO Burleigh, Fishers, NY) through general purpose interface bus and serial commands. Images of the mounted vessel were acquired using a Leica DM IL inverted microscope (at ×100 magnification), Sony X-55 charged-coupled device camera, and video acquisition card (PCI 1409; National Instruments). Control, acquisition, and analysis routines were written in LabView and IMAQ Vision (National Instruments). IGOR (Wavemetrics, Oswego, OR) was used for data display.

Pressure myograph methods.

In parallel experiments, mesenteric lymphatic segments were cannulated with glass micropipettes and pressurized on the stage of an inverted microscope, while diameter changes were tracked by computer (5). To precisely control the rate of intraluminal pressure changes, another servo-controlled system was developed, similar to that used in previous experiments on arterioles (7). Servo-null style shaker pumps (Ling, Hertz, UK) connected to the pipettes were driven by a hardware-based servo controller through unitary-gain power amplifiers (Cardiovascular Research Institute, Texas A&M University). The controller compared the output signal from low-pressure transducers (CyQ model 104; Nicholasville, KY) with reference voltages specified by a computer and adjusted the pump voltages accordingly. For enhanced precision of diameter tracking, the image was digitized with a 1,632 × 1,234 pixel firewire camera (Basler AG model A641FM; Ahrensburg, Germany) and processed as previously described (5). Only the top half of the video field was acquired to increase the video throughput to 30 Hz. All diameters reported in this study represent internal diameters.

Isometric protocols.

The effects of preload on the spontaneous contraction patterns of isometric rat mesenteric lymphatics have been described recently (51). The servo-control system could generate constant speed, ramp increases in passive force to generate essentially continuous relationships between preload and either Amp, frequency (Freq), or their product [Amp-Freq product (AFP)] and then to assess the effects of SP on those relationships. To do this, the vessel was equilibrated for 30 to 60 min at 36° to 37°C, with preload fixed at 0.2 mN, until a stable contraction pattern developed. The force transducer was then rezeroed, and the preload was set to 0.02 mN for 1 to 2 min until a new pattern stabilized. In some cases, the preload had to be increased to 0.05 mN before contractions began. A slow preload ramp up to 1.0 mN was then imposed over a time course of ∼9 min. These parameters permitted the acquisition of a sufficient number of spontaneous contractions (for subsequent data analysis) over a range of preloads equivalent to the physiological range of pressures normally experienced by these vessels in vivo (11, 51). Fresh, preheated APSS was then exchanged for the bath solution, the force transducer was rezeroed, and the preload was again set to 0.02 mN. 1·10⁻⁸ M SP (see results for the rationale) was added to the bath, and an identical preload ramp was imposed. Each vessel was exposed to only a single dose of SP, and only paired data sets (control and SP) were used for analysis. The effects of SP did not completely subside upon washout, so that the order of the control and SP tests could not be reversed; however, in preliminary tests, two consecutive control protocols were performed to ensure that any apparent SP-induced changes were not simply due to the sequence of the tests. After completion of an experiment, custom LabView programs were used to detect the Amps and Freqs of the individual contractions before, during, and following the preload ramps. Amp was taken as the difference between the peak force of the spontaneous contraction and the preload immediately before the contraction. To perform subsequent statistical analyses on the pooled data, the data sets for individual vessels were combined by binning the Amp, Freq, AFP, and rate of change in force development or decline (±dF/dt) data according to the respective preload levels into 40 bins spanning the range from 0.02 to 1.0 mN.

Isobaric protocols.

The effects of pressure on the spontaneous contraction patterns of rat mesenteric lymphatics in the presence and absence of SP were analyzed in a similar manner as described in Isometric protocols. After equilibration of a cannulated lymphatic, for 30–60 min at 36° to 37°C and pressure = 3 cmH₂O, a stable contraction pattern was typically established. Pressure was then lowered to 0.5 or 1 cmH₂O until contractions resumed and a pressure ramp up to 10 cmH₂O, lasting ∼9 min, was imposed, while diameter was tracked continuously. After the ramp was complete, pressure was returned to 0.5 or 1 cmH₂O for 5–10 min until the basal contraction pattern recovered. SP was then added to the bath and, after 1 min, a second identical pressure ramp was imposed. After that ramp was complete, the bath solution was replaced with Ca²⁺-free APSS at a pressure of 5 or 10 cmH₂O. After ∼30 min for equilibration was allowed, pressure was lowered to 0 cmH₂O for 1 min and then set at 1 cmH₂O for 1 min, and a third pressure ramp from 1 to 10 cmH₂O was imposed. Analysis programs were used offline to detect the Amps and Freqs of the individual contractions before, during, and after the pressure ramps. As before, the data sets for individual vessels were combined by binning the Amp and Freq data according to the pressure level into 40 bins spanning the range from 1 to 10 cmH₂O, to perform statistical analyses.

Tone calculations were made by comparing the control and SP data sets with the Ca²⁺-free APSS data set for the same vessel. At the pressure where each contraction was initiated, the corresponding passive diameter was measured and tone was defined as the difference between the passive diameter and the end-diastolic diameter; tone values were then normalized to percent passive diameter. Vessel tone during the control period or after SP treatment was expressed as a percentage of the passive diameter at each pressure. The tone data from multiple vessels were pooled for subsequent statistical analysis into 40 bins spanning the range from 1 to 10 cmH₂O.

Statistical analyses.

The data sets for each contraction parameter were analyzed using JMP 5.1 (SAS, Cary, NC). For the analyses in Figs. 1, 4, 6, 7, and 8, one-way ANOVAs were performed, with pressure or preload designated as the independent variable. Dunnett's post hoc tests were then used to test for significant within-group variation, e.g., to test for a difference between Amp at any given pressure/preload relative to Amp at a designated pressure/preload. For comparisons between control and SP data sets in Fig. 2, paired or unpaired t-tests were used as appropriate. Significance was defined as P < 0.05.

Fig. 1.

A: dose-response relationship of an isometric lymphatic vessel to substance P (SP). Preload was set at 0.2 mN throughout the experiment. Left side of the recording has an expanded time scale to illustrate the time course and shape of single spontaneous force transients, with analogous systolic and diastolic periods marked. SP was added to the bath and mixed by agitation to achieve the final concentrations as noted. The slight negative deflections upon SP addition are artifacts of the application procedure. When the SP concentration exceeded 3·10⁻⁸ M, individual contractions partially fused. Peak force was achieved at a concentration of 1·10⁻⁶ M. B: summary of the dose-dependent effects of SP on the frequency of spontaneous force transients (•) and on the magnitude of sustained force (○). In this group of vessels (n = 8), sustained force was negligible until the SP concentration reached or exceeded 3·10⁻⁸ M.

Fig. 4.

Analysis of Amp (A), Freq (B), and Amp-Freq product (AFP; C) from 14 isometric vessels subjected to preload ramps in the absence (○) or presence (•) of SP (1·10⁻⁸ M). Amp, Freq, and AFP data were normalized to their respective values at the lowest preload in the presence of SP. To combine the data for statistical analyses, each parameter was binned according to the respective preload level as described in methods. *Significant difference (P < 0.05) between control and SP means at each respective preload level using a paired, 2-tailed t-test; †significant difference (P < 0.05) between the control mean at the indicated preload and the control mean at the lowest preload using an ANOVA followed by Dunnett's post hoc tests; ‡significant difference (P < 0.05) between the SP mean at the indicated preload and the SP mean at the lowest preload using an ANOVA followed by Dunnett's post hoc tests.

Fig. 6.

Analysis of Amp (A), Freq (B), and AFP (C) from 18 isobaric vessels subjected to pressure ramps in the absence (○) or presence (•) of SP (1·10⁻⁸ M). Amp, Freq, and calculated AFP data were normalized to their respective values in the presence of SP at a pressure of 2 cmH₂O. To combine the data for statistical analyses, each parameter was binned according to the respective pressure level as described in methods. *Significant difference (P < 0.05) between the control and SP means at each respective pressure level using a paired, 2-tailed t-test; †significant difference (P < 0.05) between the control mean at the indicated pressure and the control mean at the lowest pressure using an ANOVA followed by Dunnett's post hoc tests; ‡significant difference (P < 0.05) between the SP mean at the indicated pressure and the SP mean at the lowest pressure using an ANOVA followed by Dunnett's post hoc tests.

Fig. 7.

Analysis of rate of change in force development (+dF/dt; A) or decline (−dF/dt; B) and rates of constriction (−dD/dt; C) and dilation (+dD/dt; D) during spontaneous force transients or contractions. ○, data under control conditions; •, data obtained in the presence of 1·10⁻⁸ M SP. A: plot of normalized peak +dF/dt as a function of preload. Peak +dF/dt represents the maximal rate of force development during the systolic phase of a spontaneous isometric contraction. B: plot of normalized peak −dF/dt as a function of preload. Peak −dF/dt represents the maximal rate of force decline during a spontaneous isometric relaxation. C: plot of normalized peak −dD/dt as a function of pressure. Peak −dD/dt represents the maximal rate of constriction during the systolic phase of a spontaneous contraction cycle. D: plot of normalized peak +dD/dt as a function of pressure. Peak +dD/dt represents the maximal rate of dilation during the diastolic phase of a spontaneous contraction cycle. Inset: 2 aligned diameter traces (control and SP) with arrow pointing to the delayed diastolic relaxation in the SP trace. *Significant difference (P < 0.05) between the control and SP means at each respective pressure level, using a paired, 2-tailed t-test; †significant difference (P < 0.05) between the control mean at the indicated preload/pressure and the control mean at the lowest preload/pressure using an ANOVA followed by Dunnett's post hoc tests; ‡significant difference (P < 0.05) between the SP mean at the indicated preload/pressure and the SP mean at the lowest preload/pressure using an ANOVA followed by Dunnett's post hoc tests.

Fig. 8.

A: examples of the data obtained using the protocol to evaluate spontaneous, basal lymphatic tone. The diameter data are plotted as a function of pressure under control conditions (darker trace) and in the presence of 1·10⁻⁸ M SP (lighter trace) during slow pressure ramps from 0.5 to 10 cmH₂O (lasting ∼10 min). SP induced a slight basal constriction and slight decrease in both systolic and diastolic diameters. After the data sets were collected, the same vessel was equilibrated for 30 min in Ca²⁺-free albumin-supplemented physiological salt solution and subjected to a third, identical pressure ramp (top trace). Basal tone was defined as the difference between diastolic diameter and the maximal passive diameter in Ca²⁺-free solution at each pressure. After tone was determined at each pressure level for each of 12 lymphatics, the data were binned according to pressure level for subsequent statistical analysis. B: plots of tone are shown as a function of pressure for control (absence of SP; ○) and SP-treated (1·10⁻⁸ M SP; •) vessels. *Significant difference (P < 0.05) between the control and SP means at each respective pressure level using a paired, 2-tailed t-test.

Fig. 2.

A: summary of isometric amplitude (Amp) and frequency (Freq) responses (for spontaneous force transients) to 10 nM SP in the absence (control) or presence of the cyclooxygenase inhibitor indomethacin (Indo; 10 μM; 30 min treatment; n = 9). Inset: a recording from 1 vessel in the continuous presence of Indo; arrows denote successive exposure to 10, 20, and 30 nM SP. Calibration bar = 0.5 mN, 1 min. B: summary of isometric Amp and Freq responses to 10 nM SP in the absence or presence of the PLA₂ inhibitor mepacrine (Mepa; 10 μM; 30 min treatment; n = 8). Inset: a recording from 1 vessel in the continuous presence of Mepa; arrows denote successive exposure to 10, 20, and 30 nM SP. Calibration bar = 0.5 mN, 1 min. C: isometric Amp and Freq responses to 30 nM SP in the absence or presence of the nonspecific neurokinin (NK) receptor blocker spantide (50 nM; 15–20 min treatment; n = 4). D: isometric Amp and Freq responses to 30 nM SP in the absence or presence of the NK₁ antagonist WIN57108 (10 μM; 15–20 min treatment; n = 7). For A–D, preload was 0.3 mN. #Significant difference (P < 0.05) in the Amp or Freq response to SP in the absence or presence of inhibitor (i.e., comparisons between bars of the same color); *significant difference (P < 0.05) in the absence vs. presence of inhibitor (i.e., comparisons between the white and black bars).

RESULTS

Preliminary isometric experiments indicated that low doses of SP (≤3·10⁻⁸ M) enhanced the spontaneous activity of lymphatic vessels while preserving distinct individual contractions. Higher doses typically produced a partial fusion of individual contraction spikes (0.1–0.5 mN in Amp), leading to sustained tetanic contractions (peak = 1.2–2.5 mN). Subsequent protocols therefore used concentrations of SP (≤3·10⁻⁸ M) as subtetanic doses that consistently produced a robust modulation of individual contractile events, yet allowed both isometric and isobaric indexes of contraction to be studied and compared in parallel.

Isometric experiments.

The dose-response relationship of rat mesenteric lymphatic vessels to SP under isometric conditions is shown in Fig. 1. The threshold dose of SP appeared to be ≤1·10⁻⁹ M. 1·10⁻⁹ M SP produced a slightly higher sustained contraction frequency than what occurred spontaneously. The immediate reaction to the addition of SP and to the subsequent mixing of the bath (note the associated force artifact) was a burst of contractions that declined slightly to a value that was still substantially higher than the control frequency. Similar effects were evident at SP doses of 3·10⁻⁹ M and 1·10⁻⁸ M, as shown in the example in Fig. 1A. The waning of the initial contraction bursts may have been caused by a partial desensitization of neurokinin (NK) receptors during the first 1 to 2 min following SP exposure (39). A dose of 3·10⁻⁸ M SP typically produced a partial summation of individual contractions, whereas 1·10⁻⁷ M SP always produced a sustained contraction and 1·10⁻⁶ M SP always produced a nearly fused contraction characterized by a peak lasting 1–5 min, followed by a plateau of variable Amp (50). Doses higher than 1·10⁻⁶ M did not typically produce additional force (not shown).

Figure 1B illustrates the effect of SP on spontaneous contraction Freq. Consistent with a previous report (39), the lowest dose, 1·10⁻⁹ M SP, produced a slight, but significant and sustained, increase in contraction Freq. Higher doses produced successively larger increases in contraction Freq, up to a dose of 1·10⁻⁷ M, which increased the average Freq to nearly 50 contractions/min (from an average control rate of ∼9 min⁻¹ in this group). The effect of SP on Freq at doses exceeding 1·10⁻⁷ M was difficult to measure because of the very low contraction Amp in most vessels (see example in Fig. 1A). The fivefold increase in Freq was considerably larger than reported previously (2, 10, 39). SP evoked substantial increases in the spontaneous contraction Amp in the same vessels (for the lower SP doses), even though the effect on Amp was not apparent in the sustained force analysis shown in Fig. 1B.

The large effect of SP on spontaneous contraction Freq, the SP-induced positive inotropy, and the fact that these changes occurred in wire-mounted vessels that may have had compromised endothelial function suggested that SP acted by a different mechanism than reported in guinea pig lymphatics. Rayner and Van Helden (39) found that SP stimulated the production of TxA₂ from guinea pig lymphatic endothelium, which subsequently produced a positive chronotropic response of the underlying lymphatic muscle. In light of that study, we specifically tested whether TxA₂ mediated the effects of SP on rat mesenteric lymphatics.

Figure 2 shows the effects of the cyclooxygenase (COX) inhibitor indomethacin and the PLA₂ inhibitor mepacrine on the inotropic and chronotropic actions of SP. Surprisingly, indomethacin, at a concentration of 10 μM [3-fold higher than the inhibitory dose for guinea pig lymphatics (39)], did not block the effects of SP (Fig. 2A). On average, 10 nM SP produced ∼50% increase in Amp and ∼100% increase in Freq in the presence of indomethacin (Fig. 2A, black bars). Likewise, mepacrine (10 μM), at a three- to tenfold higher concentration than used in other studies to effectively block PLA₂ (1, 35), produced a substantial increase in basal contraction Freq but failed to block either the positive inotropic or chronotropic effect of 10 nM SP (Fig. 2B). The effects of SP were also not prevented by the TxA₂ synthase inhibitor imidazole (50 μM; n = 4; not shown). Collectively, these results suggest that the actions of SP on rat mesenteric lymphatics were not mediated by a prostanoid.

In guinea pig mesenteric lymphatics, the nonspecific NK receptor blocker spantide (50 μM) and the more selective NK₁ receptor blocker WIN51708 blocked SP-induced chronotropy (39), consistent with the action of SP on NK₁ receptors on the lymphatic endothelium. However, we were unable to find a concentration of spantide that effectively blocked the action of 30 nM SP on isometric contractions of rat mesenteric lymphatics. As shown in Fig. 2C, 50 nM spantide (39) appeared to act as a partial agonist because its application (15–20 min before SP) increased the spontaneous concentration Freq by approximately twofold. However, in the presence of spantide, 30 nM SP increased Freq by about the same amount (∼2-fold) as in the absence of spantide. Similar effects of spantide were observed on contraction Amp (Fig. 2C). We also tested the effects of WIN51708 at the same concentration used previously on guinea pig lymphatics (pretreatment with 10 μM for 15–20 min). WIN51708 had little effect on basal contraction Amp or Freq (Fig. 2D). Curiously, WIN51708 blocked about one-half of the SP-induced increase in Amp but potentiated the effect of SP (30 nM) on Freq, with Freq increasing approximately fourfold in response to 30 nM SP, compared with an approximately threefold increase in the absence of WIN51708 (n = 7).

We then tested the interaction between SP and pressure on spontaneous contractions. Figure 3 shows the response of a representative isometric lymphatic vessel to two successive ramp increases in preload from 0.02 to 1.0 mN, spanning the range of stretch (pressure) normally experienced by similar vessels in vivo under normal conditions and during edemagenic stress (3). The advantage of using a continuous preload ramp over a series of preload steps is that this procedure allowed the measurement of the essentially continuous relationships between preload and the various lymphatic contraction parameters over a relatively short time interval (51). The raw data for two successive preload ramps are shown in Fig. 3A (control) and 3B (1·10⁻⁸ M SP), whereas Fig. 3, C and D, shows the subsequent analyses of Amp and Freq for the individual contractions as a function of preload. Amp was determined by finding the peak of each force transient relative to the passive force immediately before contraction. Freq was determined from the time interval between each successive contraction. Under control conditions, contraction Amp increased from 0.18 to 0.3 mN as preload was slowly elevated from 0.02 to 1.0 mN (Fig. 3, A and C). Contraction Freq increased from ∼3 min⁻¹ to ∼13 min⁻¹ over the same preload range (Fig. 3D). A second identical control ramp produced very similar effects on Amp and Freq (not shown). The addition of SP (1·10⁻⁸ M) at a preload = 0.05 mN increased Amp from ∼0.18 to 0.34 mN and Freq from ∼3 to 9 min⁻¹. As preload was subsequently raised to 0.3 mN, there was a further small increase in Amp, followed by a slight decline at higher preloads. In the presence of SP, Freq increased with preload, from ∼9 to 18 min⁻¹, with a slight decline at higher preloads. This record is fairly representative in showing the extent to which Amp and Freq were modulated by preload and in showing the enhancement of both parameters by SP under isometric conditions.

Fig. 3.

A and B: effect of a preload (passive force) ramp on spontaneous force transients of isometric lymphatics in the absence (A) or presence (B) of 1·10⁻⁸ M SP. Preload was initially set to a minimal value (0.02–0.05 mN) at which spontaneous contractile activity was observed. Preload was then increased to 1.0 mN at a constant rate over the course of ∼10 min. After completion of the ramp, preload was returned to its initial level until the original contraction pattern resumed. SP was then added to the bath and, after 1 to 2 min, a second, identical preload ramp was initiated. C and D: after the data were collected, a computer algorithm was used to detect the individual contraction peaks associated with the control and SP preload ramps. The detected Amps (C) and peak-to-peak Freqs (D) for the traces in A and B were then plotted as a function of preload.

The results of 14 experiments on isometric lymphatic vessels before and after the application of 1·10⁻⁸ M SP are summarized in Fig. 4. The data for each vessel were combined according to preload level, as described in methods. To minimize vessel-to-vessel variation, the Amp and Freq data for each vessel were normalized to their respective minimum values in the presence of SP. The relationship between preload and AFP, an index of lymph pump function analogous to cardiac output, was also determined (Fig. 4C). The calculation of AFP (49), rather than another index such as fractional pump flow (11), permitted the subsequent comparison of isometric and isobaric data (51). AFP peaked at a preload of ∼0.4 mN for control vessels. SP (1·10⁻⁸ M) caused an increase in AFP by ×∼4 at nearly every level of preload, with the increase being greatest at ∼0.4 mN.

Isobaric experiments.

The interactions of SP and pressure were next tested using isobaric lymphatic preparations. The responses of a pressurized lymphatic vessel to pressure ramps under control conditions and in the presence of 1·10⁻⁸ M SP are shown in Fig. 5. At a baseline pressure of 1 cmH₂O, Freq was 3 min⁻¹ and increased gradually to 15 min⁻¹ as pressure was raised to 10 cmH₂O. Under control conditions, Amp was nearly constant until pressure exceeded 3 cmH₂O but then declined progressively with further pressure elevation. About one-half of the vessels showed this pattern of response, whereas in the other half, Amp was maximal at 0.5 or 1 cmH₂O and declined at all higher pressures. The effects of pressure and SP are clearer in the Amp-pressure and Freq-pressure plots shown in Fig. 5, C and D. In the presence of 1·10⁻⁸ M SP, Freq increased from 3 to 7 min⁻¹ at pressure = 1 cmH₂O and increased further to 20 min⁻¹ at pressure = 10 cmH₂O. The net effect of SP was an upward, parallel shift in the Freq-pressure curve. SP caused a decrease in both diastolic and systolic diameter so that the Amp-pressure relationship remained approximately the same as under control conditions.

Fig. 5.

A and B: effect of a pressure ramp on the spontaneous contraction pattern of isobaric lymphatics in the absence (A) or presence (B) of 1·10⁻⁸ M SP. Pressure was initially set to the minimal value (0.5–1.0 cmH₂O) associated with the development of spontaneous contractile activity. After a stable contraction pattern was established, pressure was increased to 10 cmH₂O at a constant rate over the course of ∼10 min. After completion of the ramp, pressure was returned to its initial value until the original contraction pattern resumed. SP was then added to the bath and, after 1 to 2 min, a second, identical pressure ramp was initiated. C and D: plots of individual contraction parameters [Amp (C) and peak-to-peak Freq (D)] as a function of pressure.

Figure 6 summarizes the responses of 18 isobaric lymphatic vessels to pressure under control conditions and in the presence of 1·10⁻⁸ M SP. The overall patterns of the responses are similar to those shown for the single vessel in Fig. 5. Under control conditions, Amp remained constant until pressure exceeded 3 cmH₂O, after which it declined progressively with further pressure elevation. SP caused a slight increase in Amp at all pressure levels, but the increase was only significant at about one-half of the pressure values (indicated by asterisks in Fig. 6A). Under control conditions, Freq increased progressively with pressure over the range of 1–10 cmH₂O, but most of the response (about a 75% increase) occurred between pressures of 1 and 5 cmH₂O. At the lowest pressure, SP increased Freq by about 60%. In the presence of SP, increases in pressure produced large increases in Freq, especially over the lower pressure range, <5 cmH₂O (Fig. 6B). Under control conditions, the relationship between AFP and pressure was somewhat flat, with a broad peak near 4 cmH₂O. SP shifted this relationship upward (significant at about one-half of the pressures) and slightly to the left, so that peak AFP occurred at ∼3 cmH₂O (Fig. 6C).

Measurement of +dF/dt is often used as an index of contractility (9, 23, 24). To estimate possible contractility changes in isometric lymphatic vessels induced by SP, +dF/dt and the rate of change in force decline (−dF/dt) were computed for each acquired force point in the spontaneous contraction/relaxation cycle, after which the peak +dF/dt and −dF/dt values during each cycle were determined. Likewise, the rates of constriction (−dD/dt) and dilation (+dD/dt) for isobaric vessels were computed for each data point during each spontaneous contraction cycle, and the respective peak values were determined for each contraction. The +dF/dt, −dF/dt, +dD/dt, and −dD/dt data sets were pooled separately for a subsequent statistical analysis into 40 bins spanning the range from 1 to 10 cmH₂O. Figure 7 shows the results of this analysis. Under isometric conditions in the absence of SP, the relationship between +dF/dt and preload was relatively flat (Fig. 7A). 1·10⁻⁸ M SP shifted the entire +dF/dt-preload relationship upward so that there was a ∼40% increase in +dF/dt at all preload levels, with about one-half of the SP values being significantly different than their paired controls. Figure 7B shows a similar analysis for the relaxation phase. The effects of 1·10⁻⁸ M SP were less dramatic on dF/dt than on Amp and Freq, but at some of the lower preloads, SP produced a significant increase in the speed of isometric relaxation.

Under isobaric conditions, the −dD/dt-preload relationship (Fig. 7C) resembled the Amp-preload relationship (Fig. 6A) in that −dD/dt was maximal at 0.5 cmH₂O and declined at all higher pressures. Recall that −dD/dt reflects the maximal rate of constriction during a spontaneous contraction cycle, so that Fig. 7C indicates the speed of constriction declined as pressure increased. 1·10⁻⁸ M SP shifted the entire relationship slightly downward, i.e., producing a reduced rate of constriction at any given pressure; however, only the −dD/dt values at low pressures were significantly different from control. An explanation for this effect is suggested in the discussion. Figure 7D illustrates the relationship between pressure and the rate of dilation (+dD/dt) during the spontaneous contraction cycle; this graph resembled the AFP-pressure relationship (Fig. 6D), except that 1·10⁻⁸ M SP shifted the curve slightly downward, suggesting that it slowed the peak rate of dilation, especially at low pressures. In addition, 1·10⁻⁸ M SP substantially prolonged the length of diastole, an effect not evident in the overall +dD/dt analysis but shown in the inset in Fig. 7D.

The isobaric method permitted an evaluation of basal tone, which has been defined previously as the difference between diastolic diameter (i.e., peak dilation during a contraction cycle) and the maximal passive diameter in Ca²⁺-free solution at the same pressure (11, 12). Analyses of data obtained from pressure-ramp protocols generated an essentially continuous relationship between tone and pressure, and examples of data obtained with that protocol are shown in Fig. 8A. The diameter data are plotted as a function of pressure under control conditions (darker trace) and in the presence of 1·10⁻⁸ M SP (lighter trace) during slow pressure ramps from 0.5 to 10 cmH₂O. This particular vessel was fairly representative of most vessels subjected to this protocol; SP produced a slight increase in Amp that was more prominent at higher pressures so that, after activation with SP, the vessel was able to sustain larger amplitude contractions at high pressure. For this vessel, SP increased Freq (from 7 to 12 min⁻¹ at 1 cmH₂O), although the chronotropic effect is not apparent on the diameter-pressure plot. The diameter data after equilibration in Ca²⁺-free APSS are also shown (top trace). It is evident that tone (the difference between diastolic diameter and maximal passive diameter) was lowest at 0.5 cmH₂O and progressively increased with pressure. Notably, SP increased the amount of tone at each pressure.

Figure 8B summarizes the tone-pressure relationships for 12 vessels before (control) and after the addition of 1·10⁻⁸ M SP. At this concentration, SP produced ∼50% increase in tone at most pressures, with a slightly greater effect at low pressures. Even greater effects on tone were observed at higher SP concentrations (not shown). There were actually three patterns of the tone-pressure relationship that were somewhat obscured by the average data analysis shown in Fig. 8B. Ten out of 17 vessels exhibited a pattern similar to that shown, where tone was lowest at 0.5 cmH₂O and increased with increasing pressure. However, two vessels showed essentially no change in tone with pressure, and five vessels showed a pattern similar to that previously reported using steady-state analyses (11, 12), where tone was highest at 1 cmH₂O and progressively declined with increasing pressure. Some possible reasons for these differences are discussed below. In each group, however, SP produced an upward, parallel shift in the tone-pressure curve.

DISCUSSION

Our results suggest that SP exerts substantial positive inotropic and chronotropic effects on rat mesenteric lymphatic muscle. At low concentrations, SP enhances lymphatic contractility and pump efficiency independent of the effects of preload and thus broadens the working range of the lymphatic pump. These actions of SP on lymphatic muscle would act in concert with its effects on permeability to promote fluid filtration from exchange vessels (18, 25), stimulating lymphatic pump activity through enhanced lymphatic filling pressure. Because SP release from numerous cell types is associated with various inflammatory conditions (27, 41, 45, 52), its stimulatory actions, at low concentrations, on the lymphatic pump may play an important role in the homeostatic control of interstitial fluid balance in both normal and inflammatory states.

Effects of SP on lymphatic vessels.

Our findings both confirm and contradict various aspects of the reported effects of SP on lymphatic vessels of other species. In bovine mesenteric lymphatic vessels studied under isometric conditions, 100 nM SP increased the spontaneous contraction rate by approximately threefold (10). The concomitant effects of SP on inotropy were not quantified, but from the single recording shown (10), doses of 1 and 10 nM SP appeared to produce modest increases (∼5%) in the Amp of spontaneous contractions. In isolated, perfused guinea pig mesenteric lymphatics, the threshold for a significant positive chronotropic effect of SP was 1·10⁻⁹ M, a value that agrees exactly with our findings (Fig. 1). In the guinea pig, 1 μM SP produced a 153% increase in chronotropy that was mediated by PLA₂ activation and the production of TxA₂ from the lymphatic endothelium (39). TxA₂ subsequently initiated a chronotropic effect on an unidentified pacemaker within the vessel wall. No significant chronotropic effect was observed in denuded guinea pig vessels, suggesting that SP exerted this effect entirely through its action on the lymphatic endothelium (39).

Our results confirm the positive chronotropic effect of SP on isolated rat mesenteric lymphatic vessels, with approximately threefold enhancement in contraction frequency in response to 1·10⁻⁸ M SP under isometric conditions and approximately twofold enhancement under isobaric conditions (Figs. 4 and 6). Much greater (∼5-fold) positive chronotropic effects were observed at higher SP concentrations (Fig. 1B), but we comprehensively studied only the effects of 1·10⁻⁸ M SP to compare the results of isometric and isobaric tests. Importantly, SP also produced a substantial positive inotropic effect that was particularly evident in isometric preparations (Figs. 4A and 7A). A significant inotropic effect of 1·10⁻⁸ M SP under isobaric conditions (Figs. 6A and 7C) may have been masked by the substantial SP effect on basal tone (Fig. 8B).

In contrast with the guinea pig, the chronotropic responses of rat mesenteric lymphatics to SP were fully preserved in the presence of the COX inhibitor indomethacin, the PLA₂ inhibitor mepacrine (Fig. 2B), and the TxA₂ synthase inhibitor imidazole. Although we do not have independent confirmation of the effectiveness of these inhibitors, we used a threefold higher dose of indomethacin that completely blocked the SP action in guinea pig lymphatics (39), a three- to tenfold higher dose of mepacrine than required to block PLA₂ in other studies (1, 35), and the same dose of imidazole that blocked SP-induced lymphatic chronotropy in guinea pig (39). It should be noted that flow-dependent diastolic relaxation in the rat thoracic duct was independent of COX activity as well (13). Collectively, our results show that SP produces both positive chronotropy and inotropy in rat mesenteric lymphatic vessels through a mechanism that is different than in the guinea pig (39). We were unable to acutely denude or functionally disrupt the lymphatic endothelium in rat mesenteric lymphatics without compromising contractile function, despite trying almost all of the methods that we and others have shown to be successful for arterioles (6). However, in another study (manuscript in preparation), we found that rat mesenteric lymphatics denuded of endothelium by an air bubble recovered spontaneous contractions after 1 to 2 days in organ culture and subsequently showed positive inotropic and chronotropic responses to SP.

SP receptors on lymphatic vessels.

The receptor type mediating the actions of SP on rat mesenteric lymphatics was not completely resolved in the present study. Three tachykinin/NK receptors have been identified, NK₁, NK₂, and NK₃, each coupled through various G proteins to PLC and thereby linked to the control of phosphoinositide metabolism (26). All three receptors are widely expressed in the central and peripheral nervous systems, in the gastrointestinal tract, and in blood vessels (26, 30). SP interacts with all three receptors but has the highest affinity for the NK₁ receptor (26, 30). Only one study to date has determined the SP receptor subtype on lymphatic vessels. In guinea pig mesenteric lymphatics, the nonspecific NK receptor blocker spantide (50 nM) and the more selective NK₁ receptor blocker WIN51708 (10 μM) blocked SP-induced chronotropy (39), suggesting that SP actions were mediated by NK₁ receptors (no inotropic effect of SP was reported in that preparation). Because the SP effect on chronotropy was eliminated by endothelial cell disruption in guinea pig lymphatics, SP probably acted via NK₁ receptors on the endothelium of those vessels. The results of our tests using spantide and WIN51708 on rat mesenteric lymphatic vessels suggest that the SP-induced positive inotropy is at least partially mediated by NK₁ receptors but that the positive chronotropic effect may not be. It is highly likely that SP exerts both its positive inotropic and chronotropic effects in rat mesenteric lymphatics by acting on the lymphatic muscle and not the endothelium. Furthermore, the effects of SP on rat mesenteric lymphatics are probably mediated by a combination of NK receptors (NK₁/NK₂), as has been recently shown for human visceral smooth muscle where a strong inotropic response is also observed (30). Alternatively, if the lymphatic pacemaker is controlled by interstitial cells of Cajal-like neurons (32, 33), it is possible that NK₃ receptors may be involved, since this receptor is known to be associated with neurons (8, 40). Thus the specific SP receptors mediating positive inotropic and chronotropy in rat mesenteric lymphatics remain uncertain, and systematic experiments using more selective antagonists for specific NK receptor subtypes will be required to resolve this issue.

Interactions of pressure and SP.

A primary goal of the present studies was to assess possible interactions between pressure and SP. Our results showed that the actions of SP and pressure were additive but not synergistic, as is evident in Figs. 4 and 6. With the use of AFP as an index of pump function, stretch from 0.05 to 0.2 mN produced ∼2.5-fold increase in AFP. In a pressurized vessel, this range of preload would correspond to a pressure change from 1.5 to 6 cmH₂O (51). After exposure to 1·10⁻⁸ M SP, AFP increased approximately fourfold at the lowest preload and then increased twofold further as preload was raised to the optimal level. The effect of the same dose of SP on AFP in isobaric lymphatics was more modest (∼50% increase) and was due largely to an effect on Freq rather than Amp. Interestingly, under isobaric conditions, lower concentrations of SP (<10⁻⁸ M) usually led to an increased constriction Amp, whereas higher SP doses produced even more reduction in Amp in a greater percentage of vessels (not shown); the latter observation is consistent with the in vivo findings of Amerini et al. (2).

In this regard, isometric indexes of contraction may be better indicators of the potential chronotropic and inotropic effects of SP because the inotropic action of SP was much more consistent and graded according to the SP dose under isometric conditions (Figs. 1 and 4A). +dF/dt in isometric contractions was also consistently elevated by SP across all levels of preload, whereas the effect of SP on −dD/dt in isobaric vessels was somewhat inconsistent (Fig. 7), for reasons that are not completely clear but are probably related to the concomitant increase in basal tone (Fig. 8B). Regardless, in vivo lymphatics, which contract neither isometrically nor isobarically (due to internal pressure spikes), would be stimulated by the additive effects of pressure and low doses of SP to promote vigorous changes in lymphatic pumping. Under conditions of edemagenic stress, when both pressure and local SP levels are most likely to be elevated, the inotropic and chronotropic actions of SP would allow the pumping ability of the collecting lymphatics to be preserved or even enhanced in the face of chronically elevated intraluminal pressure (3, 14, 20, 37, 48).

Effect of SP on basal tone.

The relationship between pressure and lymphatic basal tone (Fig. 7B) was different than reported in our previous steady-state pressurized vessel studies. Earlier (11, 12), our laboratory found that the basal tone of rat mesenteric lymphatic vessels was highest at low pressures and declined as pressure increased (between 1 and 7 cmH₂O). Although that same relationship was observed in a subset of vessels in this study (5 of 17), in the majority of the present experiments (10 of 17), tone was minimal at pressures ≤1 cmH₂O and rose with pressure to a value of ∼8% (a maximum level that agrees with Refs. 11 and 12). In the latter vessels, the increase in tone represents a mild myogenic constriction that is consistent with the stimulatory effect of pressure on spontaneous contraction Amp. The myogenic constriction appeared to be time dependent in response to comparatively fast, step changes in pressure (29) and thus may be more likely to be detected with the ramp pressure protocols used here than by averaging over a 5-min equilibration period as used previously (11, 12).

The effect of SP on tone may be important pathologically because, at very high intraluminal pressures associated with chronic edema (3, 14, 20, 37, 48), the normal uphill pressure gradient (16, 17) may be reversed and collecting lymphatics may switch, over a prolonged period of time, from functioning as pumps to serving as conduits (12, 38). Although this effect might be advantageous to interstitial fluid balance at moderately elevated interstitial and lymphatic pressures, excessive basal tone development promoted by SP could be disadvantageous by increasing the resistance of the collecting lymphatic conduits and potentially exacerbating the edema. Based on Figs. 1 and 8, tone could potentially be quite pronounced at high SP doses, if such concentrations are ever achieved in vivo, e.g., during edema. With the assumption that SP release is graded in proportion to the severity of edema (which has not yet been tested), a logical conclusion is that SP receptor blockers would exacerbate mild edema (by interfering with SP-induced pump enhancement) but would partially alleviate severe edema [by blocking the SP-induced constriction of passive lymphatic conduits (38)]. Whether this prediction is correct remains to be determined.

GRANTS

This work was supported by National Institutes of Health Grants HL-075199, HL-89784, and RR-017353.