The sphingolipid signaling pathway appears to play an important role in regulating vascular tone. We examined the effect of fumonisin B1, a fungal toxin in corn that blocks ceramide synthase in the sphingolipid signaling pathway, on the ascending aortic impedance spectrum of pigs. Sixteen pigs were fed culture material containing fumonisin B1 (20 mg/kg body wt) (n = 7) or a control diet (n = 9) daily for 3 days and then instrumented under α-chloralose anesthesia for measurement of ascending aortic pressure and flow. Fumonisin ingestion increased serum sphinganine and sphingosine concentrations. Fumonisin ingestion also decreased cardiac output and characteristic impedance and increased the frequency of the first minimum impedance modulus, systemic vascular resistance, and the terminal, first, and second harmonic reflection coefficients, without changing mean arterial pressure. Thus blockade of ceramide synthase is accompanied by decreased vascular tone in systemic conduit arteries and increased vascular tone in systemic resistance vessels. The results indicate that the sphingolipid signaling pathway influences vascular tone in α-chloralose-anesthetized pigs.
- tumor necrosis factor-α
accumulating evidence suggests that the sphingolipid signaling pathway plays an important role in regulating vascular tone in health and disease (41, 42). Two of the major intermediaries in the sphingolipid signaling pathway, ceramide and sphingosine, are suspected to mediate some of the vascular effects of endotoxemia (17, 22; Fig.1), and another component of the pathway, sphingosine-1-phosphate, is a vasoactive molecule released by platelets (36). Acute endotoxemia is accompanied by profound systemic arterial hypotension, which results primarily from vascular smooth muscle relaxation rather than decreased cardiac contractility (4). The endothelium-dependent component of endotoxin-induced hypotension is believed to be due to relaxation of vascular smooth muscle in response to activation of endothelial nitric oxide synthase and generation of nitric oxide, as well as endothelial synthesis of prostanoids (9, 31). The mechanism for the endothelium-independent component of endotoxin-induced hypotension is currently uncertain but is probably mediated by tumor necrosis factor-α (TNF-α) (17), which induces endothelium-independent vasodilation through phospholipase A2-dependent activation of a plasma membrane-associated, neutral pH-operating, magnesium-dependent form of sphingomyelinase that hydrolyzes membrane sphingomyelin to ceramide (17) and ultimately sphingosine or sphingosine-1-phosphate (Fig. 1). Sphingosine relaxes prostaglandin F2α-contracted pig coronary arteries (22) and phenylephrine- contracted pig thoracic aorta (13). Ceramide relaxes phenylephrine-contracted rat mesenteric arteries (18). A membrane-permeable form of ceramide, C2-ceramide, relaxes phenylephrine-contracted rat thoracic aorta and decreases phenylephrine-induced elevations in intracellular Ca2+ concentration ([Ca2+]i) (15-17, 41); similar results were reported after administration of C8-ceramide-1-phosphate and sphingosine (42). The vascular smooth muscle response to sphingosine and ceramide is consistent with blockade of L-type Ca2+channels (41).
Fumonisin provides a valuable pharmacological tool for altering the sphingolipid signaling pathway because it is a potent inhibitor of ceramide synthase (sphinganine/sphingosineN-acyltransferase) (38), a key enzyme in the pathway for de novo sphingolipid biosynthesis (Fig. 1). Fumonisins are a group of mycotoxins produced primarily by Fusarium verticillioides, a fungus that commonly contaminates corn. Ingestion of feed contaminated with fumonisin causes leukoencephalomalacia in horses (39) and pulmonary edema in pigs (10, 11, 28, 34, 35). Fumonisin-induced ceramide synthase inhibition results in increased concentrations of sphinganine and sphingosine in serum and tissues (28, 38; Fig. 1), accompanied by decreased cardiac contractility, heart rate, and cardiac output in pigs (5, 32, 34, 35) and horses (33), and systemic arterial hypotension in pigs (35). Fumonisin mycotoxicosis in pigs and horses provides a naturally occurring example of morbidity and mortality due to abnormal sphingolipid metabolism, thereby providing a valuable model for characterizing the cardiovascular functions of the sphingolipid signaling pathway in health and disease.
Alteration of the sphingolipid signaling pathway by administration of fumonisin increases systemic vascular resistance (SVR) in pigs (5, 32, 35) and horses (33). The increase in SVR reflected constriction of systemic resistance vessels or was a mathematical result of its method of calculation, where SVR = [mean arterial pressure − central venous pressure]/cardiac output, because affected animals had decreased cardiac output. We hypothesized that the fumonisin-induced increase in SVR was due to contraction of systemic resistance vessels in response to decreased cardiac contractility, heart rate, cardiac output, and sphinganine or sphingosine-induced relaxation of systemic conduit vessels. Therefore, the objectives of this study were to characterize the in vivo effect of altering the sphingolipid signaling pathway by blocking ceramide synthase on systemic conduit and resistance vessels, and to determine whether the vascular responses to standard pharmacological agents (isoproterenol, nitroprusside, phenylephrine, and Ca2+) were similar in fumonisin-fed pigs and control pigs. Isoproterenol, the classic β1- and β2-agonist, was used to examine the effect of altered sphingolipid signaling pathway on β2-mediated vasodilation. Nitroprusside, the direct acting vascular smooth muscle relaxant, was used to examine the effect of altered sphingolipid signaling pathway on systemic arterial vasodilation. Phenylephrine, the classic α1-agonist, was used to examine the effect of altered sphingolipid signaling pathway on systemic arterial vasoconstriction. Ca2+ was administered to examine the effect of increased plasma [Ca2+] on hemodynamic values. To achieve our objectives, we measured the aortic input impedance spectrum in α-chloralose-anesthetized pigs.
MATERIALS AND METHODS
Animals and treatment.
This study was approved by our institutional committee on the care and use of laboratory animals. Sixteen healthy castrated male cross-bred pigs weighing 44 ± 4 kg (means ± SD) were housed individually in stalls 5–7 days before treatment. Pigs had free access to water and were fed an 18% protein grower diet that was free of fumonisin B1 (detection limit <0.1 ppm), fumonisin B2 (detection limit <0.1 ppm), aflatoxin, vomitoxin, T-2 toxin, ochratoxin, and zearalenone. Culture material containing a high concentration of fumonisin was prepared and analyzed as described previously (35). At the end of the acclimation period, a random number table was used to assign pigs to one of two groups. Treated pigs (n = 7) received culture material containing 20 mg · kg−1 · day−1fumonisin B1 mixed into the grower diet, whereas control pigs (n = 9) were fed only the grower diet on the same schedule as treated pigs. Feed consumption was monitored to ensure that the pigs ingested the fumonisin dose. In our laboratory, this fumonisin dose causes pulmonary edema and death within 3–5 days of initial exposure.
Seventy-two hours after the start of fumonisin administration, the pigs were anesthetized by intramuscular injection of 2 mg/kg xylazine and 10 mg/kg ketamine hydrochloride, followed by mask induction with 3–5% halothane in 100% O2. Pigs were then orotracheally intubated, placed in dorsal recumbency on a water-circulating heating blanket, and ventilated with supplemental oxygen (4 l/min) on a volume respirator (Harvard Apparatus) at a tidal volume of 10 ml/kg and respiratory frequency of 20 breaths/min. Anesthesia was subsequently maintained by intravenous administration of α-chloralose (50 mg/kg initially, then 15 mg · kg−1 · h−1; Sigma; St. Louis, MO) and butorphanol (0.5 mg/kg im, every 3 h). This anesthetic protocol produces minimal cardiovascular depression in pigs (5). Arterial pH, Pco 2, and Po 2 were measured periodically during instrumentation and maintained within normal limits (pH 7.40–7.50; PaCO2, 35–45 mmHg; PaO2, >100 mmHg) by adjusting tidal volume. Ventilator settings were not altered once instrumentation was completed. Pulmonary artery blood temperature was monitored during the study, and the temperature of the heating blanket was adjusted to maintain normal blood temperature (39.6 ± 0.6°C). Lactated Ringer solution (4 ml · kg−1 · h−1) was administered intravenously for the duration of the study.
The right carotid artery, right jugular vein, and right femoral artery were identified by surgical cutdowns. A 7-Fr dual-tipped micromanometer catheter with an electromagnetic fluid-velocity sensor (Millar Instruments; Houston, TX) was advanced through the right carotid artery and positioned across the aortic valve to record left ventricular pressure, ascending aortic pressure, and ascending aortic flow velocity. The aortic and ventricular pressure signals were amplified and displayed (5/6 recorder; Gilson Medical Electronics; Middleton, WI) and the velocity sensor connected to an electromagnetic 500-Hz square wave flowmeter (Carolina Medical Electronics; King, NC) with the low-pass filter set at 100 Hz. The frequency response and sensitivity of this flow-velocity system have been reported previously (19). The catheter position was adjusted to provide optimal pressure and flow-velocity signals, the latter being characterized by a steady diastolic level with a maximal systolic amplitude and minimal late systolic negative flow amplitude. A 7-Fr Swan-Ganz thermodilution catheter (American Edwards Laboratories; Irvine, CA) was placed in the pulmonary artery via the right jugular vein for calibration of the aortic flow-velocity signal, determination of cardiac output, and measurement of mean pulmonary artery pressure, mean central venous pressure, and blood temperature. A polyethylene catheter (3 mm outer diameter) was placed in the right femoral artery for measurement of mean arterial blood pressure and to obtain arterial blood for analysis.
Lead II ECG, ascending aortic pressure, and ascending aortic flow-velocity signals were monitored continuously, calibrated, and digitized at 500 Hz on a 12-bit personal computer. Data were stored on the computer's hard drive. All data were recorded at end expiration with the ventilator turned off.
Aortic pressure signal calibration.
The aortic pressure-flow catheter was presoaked in 0.9% NaCl solution at 37°C for at least 30 min before use and calibrated in vitro with a transducer control unit (Millar Instruments). Zero pressure reference was obtained with the pressure sensor barely submerged in the isotonic saline solution. Pressure transducers had a range of −50–300 mmHg, flat frequency response to 10 kHz, no phase or amplitude error, drift of <6 mmHg in 12 h, and natural frequency of 35 kHz or greater. Because the pressure recording system had no amplitude or phase error at the studied frequencies (<15 Hz), specific adjustments of the pressure signal were not undertaken when characterizing the aortic impedance spectrum and hydraulic power spectrum.
Aortic flow-velocity signal calibration.
The electromagnetic flow-velocity signal (cm/s) was calibrated in terms of volumetric flow (cm3/s) during each hemodynamic recording period by in vivo calibration against simultaneously obtained thermodilution stroke volumes (the mean of three determinations at constant heart rate). The area under the aortic flow-velocity curve was integrated to provide electromagnetic stroke volume, and the gain on the computer was altered to ensure equivalence of the two stroke volumes. This calibration technique was required because the electromagnetic transducer signal was proportional to flow velocity instead of volumetric flow, and assumed the last third of the diastolic flow signal represented zero flow, the velocity profile in the region studied was flat, and that the ascending aortic diameter was constant throughout ejection (25).
Arterial blood was obtained as soon as an artery was catheterized and serum harvested for subsequent sphinganine and sphingosine assays and serum biochemical analyses. After instrumentation had been completed, pigs were monitored for 15 min to ensure hemodynamic stability before obtaining baseline measurements. The cardiovascular response to sequential intravenous administration of the following pharmacological agents was examined: isoproterenol HCl (0.04 μg · kg−1 · min−1; Isuprel, Sanofi Winthrop Pharmaceuticals, New York, NY); sodium nitroprusside (4 μg · kg−1 · min−1; Nitropress, Abbott; N. Chicago, IL); phenylephrine HCl (4 μg · kg−1 · min−1; United Research Laboratories, Philadelphia, PA); and Ca2+(4 mg · kg−1 · min−1; as Ca2+ gluconate). Each pharmacological agent was infused for at least 5 min until stable hemodynamic values were obtained. After each pharmacological agent was infused, hemodynamic parameters were monitored for at least 15 min until they were stable and had returned to preinfusion values. Pigs were euthanized at the end of the study by administration of pentobarbital sodium (60 mg/kg iv), and left ventricular tissue was harvested.
Serum derived from arterial blood and left ventricular tissue were stored at −20°C and thawed immediately before determination of free sphinganine and sphingosine concentrations by using previously described methods (35).
Blood gas analysis.
Blood gas analyses were performed immediately by the platinum and glass electrode technique, blood hemoglobin concentration was determined with the use of spectrophotometry, and plasma ionized [Ca2+] was determined with the use of an ion-specific electrode (model ABL330, Radiometer; Copenhagen, Denmark).
Hemodynamic data were analyzed off-line with the use of a personal computer and custom-designed software. End diastole was defined as the point corresponding to the peak of the R wave of the ECG. This reference point was used for signal averaging techniques. The calibrated ascending aortic pressure and flow-velocity waveforms from 10 consecutive beats were signal averaged to minimize noise in the pressure and flow-velocity recording systems, and submitted to Fourier analysis. The noise level of the flow-velocity and pressure signals were determined for each pig during baseline by performing Fourier analysis on the diastolic portion of the aortic flow and pressure signal. Only flow harmonics with moduli greater than the maximum noise level for the flow-velocity signal (typically 8 cm/s or ∼2% of peak flow) or pressure (typically 0.2 mmHg or ∼0.2% of systolic arterial pressure) were included in subsequent calculations. The phase lag of 1.3/Hz (6 ms) for the flow-velocity signal also was accounted for in subsequent calculations. Postectopic beats and those beats with suboptimal flow-velocity signals were excluded from analysis.
Heart rate, mean aortic pressure, mean aortic flow velocity, peak aortic flow velocity, and time from end diastole to peak aortic flow-velocity were derived from the signal-averaged pressure and flow-velocity waves. The pressure signal was differentiated with respect to time after Fourier analysis to provide peak rate of aortic pressure change (dP/dt max) and time from end diastole to aortic dP/dt max. The flow-velocity signal was differentiated with respect to time after Fourier analysis to provide the maximal flow acceleration of blood in the ascending aorta (dQ/dt max) and time from end diastole to dQ/dt max(30).
Aortic impedance spectrum.
The input impedance modulus and phase angle for each harmonic were calculated as the ratio of the pressure to flow moduli and the difference between the pressure and flow phase angles, respectively, with the convention that a negative-phase angle indicated that flow leads pressure (24). Characteristic impedance (Z c) was calculated from the average of impedance moduli after the first minimum to the noise limit for the flow recording system. A reflectance index, peripheral resistance (Z 0), was calculated from the ratio of mean pressure to mean flow; Z 0 represents the static (nonpulsatile) component of input impedance (Z i) and approximates total peripheral resistance when right atrial pressure is low (24), as in this study. The reflection coefficient at the zero, first, and second harmonics (Γ0, Γ1, and Γ2) were calculated from the frequency-dependent reflection coefficient spectrum by using the ratio (Z i −Z c)/(Z i +Z c) (1). Reflectance characteristics were also assessed by determining the frequency of the first minimum impedance modulus and by estimating the frequency for the first zero crossing of phase angle (phase 0), which was obtained by linear interpolation (24).
Aortic hydraulic power spectrum.
Mean hydraulic pressure power (WM), oscillatory hydraulic pressure power (Wo), and total hydraulic pressure power (WT) were calculated from the hydraulic power spectrum (24). The ratio of oscillatory to total hydraulic pressure power output (Wo/WT) was also determined to assess the efficiency of energy transfer from the left ventricle to the systemic circulation (24, 40).
Data are presented as means ± SD. Two-way analysis of variance (treatment and pharmacological agent) with repeated measures on one factor (pharmacological agent) was used for comparison. Appropriate post tests were conducted whenever the F test was significant (P < 0.05). Multiple pairwise comparisons were conducted between or within treatment groups by using the Bonferroni inequality to keep the experimentwide error rate atP < 0.05 for each family of comparisons. Within-treatment group comparisons were to baseline values. Between-treatment group comparisons for each variable were made for each pharmacological agent. Variables with nonnormal distributions or unequal variances were log transformed or ranked before analysis of variance was performed.
Fumonisin-treated pigs had increased serum and left ventricular sphinganine and sphingosine concentrations, as well as increased serum and left ventricular sphinganine to sphingosine ratios compared with controls (Table 1).
Blood gas analysis.
Fumonisin-treated pigs had an increased blood hemoglobin concentration but similar blood pH, Pco 2, Po 2, bicarbonate concentration, base excess, and plasma ionized [Ca2+] to controls (Table 1).
Fumonisin-treated pigs had a lower cardiac output and stroke volume than control pigs, but similar heart rate, mean arterial pressure, aortic dP/dt max, mean pulmonary artery pressure, and mean central venous pressure, peak aortic flow, and aortic dQ/dt max (Table2). The time from end diastole to aortic dQ/dt max and dP/dt max was longer in fumonisin-treated pigs than in controls. Taken together, these hemodynamic changes indicated decreased left ventricular contractility in fumonisin-treated pigs.
Administration of the four pharmacological agents produced similar hemodynamic changes in both groups, with the only significant group-drug interaction term being the effect on mean aortic pressure (Table 2). Isoproterenol decreased mean aortic pressure in fumonisin-treated pigs but not controls, nitroprusside decreased mean aortic pressure to a greater extent in fumonisin-treated pigs than controls, and phenylephrine increased mean aortic pressure to a greater extent in fumonisin-treated pigs than controls. Therefore, mean aortic pressure in fumonisin-treated pigs was more dependent on SVR tone; this was accompanied by an increased sensitivity of the systemic resistance vessels to isoproterenol, nitroprusside, and phenylephrine.
Representative aortic pressure and flow signals during baseline and administration of isoproterenol, nitroprusside, phenylephrine, and Ca2+ are shown in Figs.2-6. The baseline pressure wave of control pigs (Fig. 2) had a prominant diastolic wave, with the dicrotic notch (foot of the diastolic wave) occurring shortly after the incisura. In comparison, the diastolic wave was not as pronounced in fumonisin-treated pigs because the reflected pressure wave arrived earlier in ejection, leading to augmentation of the pressure wave.
Differences in the shape of the pressure and flow-velocity waves between fumonisin-treated and control pigs were still evident after administration of isoproterenol (Fig. 3). During nitroprusside administration, the pressure wave began to resemble the flow wave, and a decrease in wave reflection caused prolongation of the time at peak flow velocity (Fig.4). Phenylephrine administration decreased the time at peak flow velocity due to the presence of increased wave reflection in early ejection (Fig.5). The administration of Ca2+ increased the pulse pressure and amplified the differences in pressure and flow-velocity signals between control and fumonisin-treated pigs (Fig. 6).
Aortic impedance spectrum.
The impedance spectra of fumonisin-treated pigs differed from that of control pigs (Fig. 2 and Table 3), as indicated by a lower value for Z c, higher values for Z 0, the frequency of the first minimum impedance modulus, and the reflection coefficient at the Γ0, Γ1, and Γ2, and a tendency (P = 0.075) toward an increased value forphase 0. Because mean arterial pressure was similar for both groups (Table 2), these changes in the aortic impedance spectrum indicate relaxation of systemic conduit vessels (decreasedZ c) and contraction of systemic resistance vessels (increased Z 0, Γ0, Γ1, and Γ2, and frequency of first minimum impedance modulus) in fumonisin-treated pigs.
Administration of the four pharmacological agents produced similar changes in the aortic impedance spectrum (Figs. 4 and 5), with the only significant group-drug interaction term being the effect on the frequency of the first minimum impedance modulus (Table 3). Isoproterenol and nitroprusside decreased the value for the first minimum impedance modulus in fumonisin-treated pigs but not controls.
Aortic hydraulic power spectrum.
Fumonisin-treated pigs had similar WM, WT, Wo/WT, and WT/cardiac output (CO), but lower Wo, than control pigs (Table 3). The absence of major changes in the aortic hydraulic power spectrum was indicative that left ventricular aortic coupling was not altered energetically in fumonisin-treated pigs.
Administration of the four pharmacological agents produced disparate changes in the aortic hydraulic power spectrum (Table 3), with significant group-drug interaction terms for WM, WT, and WT/CO. Nitroprusside administration greatly decreased WM, WT, and WT/CO in fumonisin-treated pigs, but not controls, due to marked systemic hypotension.
Blocking ceramide synthase by administering fumonisin B1 increased serum and left ventricular sphinganine and sphingosine concentrations and decreased cardiac output. These changes were consistent with those previously reported in pigs ingesting high daily doses of fumonisin B1 (>16 mg/kg; see Ref.11 for review). The major new findings of this study were that altering the sphingolipid signaling system by blocking ceramide synthase caused marked changes in aortic pressure and flow-velocity waves, characterized by decreased vascular tone in systemic conduit vessels and increased vascular tone in systemic resistance vessels. Moreover, ceramide synthase-blocked pigs appeared to have a greater dependence on constriction of systemic resistance vessels for maintaining normal mean arterial pressure, and an increased sensitivity to isoproterenol, nitroprusside, and phenylephrine administration.
The aortic input impedance spectrum provides a quantitative and independent method for characterizing the response of the systemic circulation to an agent that affects the cardiovascular system because it incorporates both steady-state and pulsatile aspects of SVR. The aortic impedance spectrum is determined by the heart rate, mean arterial pressure, the compliance and vascular tone of the proximal aorta, the extent of constriction of systemic resistance vessels, and the distance from the proximal aorta to peripheral reflecting sites (26). The difference in impedance spectra between ceramide synthase-blocked pigs and control pigs (higher values for Z 0, phase 0, and frequency of first minimum impedance modulus, and lower value forZ c for fumonisin-treated pigs) mirrored those in dogs administered the direct acting vasodilator agent nitroprusside, indicating that fumonisin-treated pigs have relaxation of systemic conduit vessels and contraction (presumably reflex) of systemic resistance vessels (37). Vasodilation typically is accompanied by a decrease in Z 0, loss of the discrete minimal modulus value, and a decrease in phase fluctation (26, 27), as observed in the impedance spectra of fumonisin-treated and control pigs administered isoproterenol and nitroprusside. The increase in the frequency of phase 0 and first minimum accompanied the increase in mean aortic pressure after Ca2+ and phenylephrine administration, and the decrease in the frequency of phase 0 and first minimum accompanied the decrease in mean aortic pressure after isoproterenol and nitroprusside administration. These changes were due in part to altered reflection because of a changes in blood pressure and in vascular tone (1,26).
The results of this study clearly demonstrate that alterations in the sphingolipid biosynthetic pathway affects the systemic vasculature of pigs. This response may have been mediated by increased serum sphinganine concentration (0.004 μM in control pigs; 0.165 μM in treated pigs), increased serum sphingosine concentration (0.017 μM in control pigs; 0.032 μM in treated pigs), or vascular smooth muscle sphinganine and sphingosine concentrations (not measured). We believe that both serum and vascular smooth muscle sphingolipid concentrations were increased in the pigs in this study because we (6) recently determined that daily ingestion of 0.22–0.66 mg/kg body wt fumonisin B1 for 6 mo increased aortic sphinganine and sphingosine concentrations in Sinclair minipigs; for comparison, the pigs in the study reported here ingested 20 mg/ kg body wt fumonisin B1 each day. In pigs, relaxation of phenylephrine contracted thoracic aortic rings occurred when sphinganine ≥0.1 μM and sphingosine ≥1 μM (13), and relaxation of prostaglandin F2α-contracted pig coronary arteries occurred when sphingosine ≥3 μM (22). Taken together, these results suggest that sphinganine may be an important vasodilator agent in pigs administered fumonisin; however, a role for sphinganine in vascular reactivity has not been extensively examined. Sphingosine reversibly and potently inhibits L-type Ca2+ channels (EC50 = 0.1 μM) and ryanodine receptors (EC50 = 0.5 μM) of rabbit skeletal muscle fibers (29), and the changes in the aortic input impedance spectrum in pigs administered fumonisin were consistent with sphingosine-mediated L-type Ca2+ channel blockade. Because arterial smooth muscle tone is believed to be regulated primarily by L-type Ca2+ channel activity (23), sphingosine-induced inhibition of systemic arterial smooth muscle L-type Ca2+ channels, if present, would induce vasodilation. However, there may be species-specific differences in the vascular effects of sphingolipids, as sphingosine (10 μM) did not relax the phenylephrine-contracted rat thoracic aorta (15), and ceramide did not relax or contract large bovine coronary arteries but did contract small bovine coronary arteries (20).
Ceramide synthase blockade increased the sensitivity of systemic resistance vessels to isoproterenol, nitroprusside, and phenylephrine. The increased sensitivity to the vasodilator agents isoproterenol and nitroprusside was due, in part, to decreased cardiovascular reserve, which we have documented previously in pigs ingesting similar doses of fumonisin (5, 32, 34, 35). The mechanism for the increased sensitivity to the vasoconstrictor agent phenylephrine is probably sphingosine enhancement of α1-receptor signaling because cytosolic sphingosine increases phospholipase C activity, thereby resulting in G protein-coupled enhancement of phosphatidylinositol turnover, increased release of Ca2+ from the sarcoplasmic reticulum (3), and vasoconstriction. Another potential pathway by which ceramide synthase blockade increases the sensitivity of systemic resistance vessels to phenylephrine is rapid metabolism of cytosolic sphingosine to sphingosine-1-phosphate by endoplasmic reticulum-bound sphingosine kinase (Fig. 1); sphingosine-1-phosphate then directly induces Ca2+ release from intracellular Ca2+stores (7, 8), leading to vasoconstriction.
We did not examine the effect of fumonisin ingestion on the pulmonary circulation. It is possible that ceramide synthase blockade exerts a different physiological response on pulmonary vasculature, as fumonisin B1 induces pulmonary hypertension in pigs (34,35) that is accompanied by accumulation of membranous material in the cytocavitary network of the pulmonary capillary endothelium. In contrast, fumonisin B1 does not appear to cause morphologic changes in the systemic vasculature (10, 11).
The results of this study provide additional information on a potential mechanism for altered vascular responsiveness in endotoxemic shock. Acute endotoxemia causes the release of TNF-α (21), and TNF-α induces endothelium-dependent and endothelium-independent vasodilation (31). The endothelium-independent TNF-α signaling pathway is shown in Fig. 1, which involves an acute signal-transduction mechanism involving phospholipase A2activation of neutral sphingomyelinase, hydrolysis of sphingomyelin to ceramide, and conversion of ceramide to sphingosine. Activation of the TNF-α signaling pathway will therefore increase ceramide and sphingosine concentrations in vascular smooth muscle; in contrast, ceramide synthase blockade by fumonisin increases sphinganine and sphingosine concentrations in vascular smooth muscle (6) and is widely believed to decrease tissue ceramide concentration. Therefore, the results of this and other studies (13, 22,41) suggest that in vascular smooth muscle, TNF-α induced increased sphingosine concentrations will lead to vasodilation through blockade of L-type Ca2+ channels, while increasing the sensitivity to α1-mediated vasoconstriction. Interestingly, carotid and cranial mesenteric arteries from pigs with acute (3 h) endotoxemia have increased sensitivity to the vasodilator effects of nitroglycerin and the vasoconstrictive effects of norepinephrine (2); these responses are consistent with those expected from TNF-α induced increases in vascular smooth muscle sphingosine concentration.
In conclusion, the results of this study in pigs indicate that ceramide synthase blockade caused relaxation of systemic conduit vessels, which was accompanied by contraction of systemic resistance vessels. The vascular effects associated with ceramide synthase blockade were probably due to an increase in plasma and tissue sphinganine or sphingosine concentrations. Because sphingosine is an important physiological mediator in numerous cellular pathways and disease processes, including endotoxic shock, additional studies are warranted to improve our understanding of the mechanism for the vascular effects of altered sphingolipid metabolism.
We thank Catherine Simutis for technical assistance.
This study was supported in part by American Heart Association Grant-in-Aid 9706561A and American Heart Association Fellowship Award 9804717X (to G. W. Smith). Part of this study was presented in: PD Constable, GW Smith, and WM Haschek. Effect of fumonisin on left ventricular afterload in swine (Abstract). Exper Biol 123: A93, 1999.
Address for reprint requests and other correspondence: P. D. Constable, Dept. of Veterinary Clinical Medicine, Univ. of Illinois at Urbana-Champaign, 1008 W. Hazelwood Dr., Urbana, IL 61802 (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2003 the American Physiological Society