cDNA and deduced primary structure of basic phospholipase A2 with neurotoxic activity from the venom secretion of the Crotalus durissus collilineatus rattlesnake.

To illustrate the construction of precursor complementary DNAs, we isolated mRNAs from whole venom samples. After reverse transcription polymerase chain reaction (RT-PCR), we amplified the cDNA coding for a neurotoxic protein, phospholipase A2 D49 (PLA2 D49), from the venom of Crotalus durissus collilineatus (Cdc PLA2). The cDNA encoding Cdc PLA2 from whole venom was sequenced. The deduced amino acid sequence of this cDNA has high overall sequence identity with the group II PLA2 protein family. Cdc PLA2 has 14 cysteine residues capable of forming seven disulfide bonds that characterize this group of PLA2 enzymes. Cdc PLA2 was isolated using conventional Sephadex G75 column chromatography and reverse-phase high performance liquid chromatography (RP-HPLC). The molecular mass was estimated using matrix-assisted laser desorption ionization-time-of-flight (MALDI-TOF) mass spectrometry. We tested the neuromuscular blocking activities on chick biventer cervicis neuromuscular tissue. Phylogenetic analysis of Cdc PLA2 showed the existence of two lines of N6-PLA2, denominated F24 and S24. Apparently, the sequences of the New World's N6-F24-PLA2 are similar to those of the agkistrodotoxin from the Asian genus Gloydius. The sequences of N6-S24-PLA2 are similar to the sequence of trimucrotoxin from the genus Protobothrops, found in the Old World.

cDNA and deduced primary structure of basic phospholipase A2 with neurotoxic activity from the venom secretion of the Crotalus durissus collilineatus rattlesnake

To illustrate the construction of precursor complementary DNAs, we isolated mRNAs from whole venom samples. After reverse transcription polymerase chain reaction (RT-PCR), we amplified the cDNA coding for a neurotoxic protein, phospholipase A2 D49 (PLA2 D49), from the venom of Crotalus durissus collilineatus (Cdc PLA2). The cDNA encoding Cdc PLA2 from whole venom was sequenced. The deduced amino acid sequence of this cDNA has high overall sequence identity with the group II PLA2 protein family. Cdc PLA2 has 14 cysteine residues capable of forming seven disulfide bonds that characterize this group of PLA2 enzymes. Cdc PLA2 was isolated using conventional Sephadex G75 column chromatography and reverse-phase high performance liquid chromatography (RP-HPLC). The molecular mass was estimated using matrix-assisted laser desorption ionization-time-of-flight (MALDI-TOF) mass spectrometry. We tested the neuromuscular blocking activities on chick biventer cervicis neuromuscular tissue. Phylogenetic analysis of Cdc PLA2 showed the existence of two lines of N6-PLA2, denominated F24 and S24. Apparently, the sequences of the New World’s N6-F24-PLA2 are similar to those of the agkistrodotoxin from the Asian genus Gloydius. The sequences of N6-S24-PLA2 are similar to the sequence of trimucrotoxin from the genus Protobothrops, found in the Old World.

L- and T-type voltage-gated Ca2+ currents in adrenal medulla endothelial cells.

We investigated voltage-dependent Ca2+ channels of bovine adrenal medulla endothelial cells with the whole cell version of the patch-clamp technique. Depolarization elicited an inward current that was carried by Ca2+ and was composed of a transient (T) current, present in all the cells tested, and a sustained (L) current, present in 65% of them. We separated these currents and measured their individual kinetic and gating properties. The activation threshold for T current was approximately -50 mV, and its maximum amplitude was -49.8 +/- 4.8 pA (means +/- SE, n = 19) at 0 mV. The time constant was 10.2 +/- 1.5 ms (n = 4) for activation and 18.4 +/- 2.8 ms (n = 4) for inactivation. The L current activated at -40 mV, and it reached a plateau at -20.1 +/- 2.3 pA (n = 6). Its activation time course was a single exponential with an activation time contant of 26.8 +/- 2.3 ms (n = 4). Current-voltage curves, kinetics, gating, response to BAY K 8644, nifedipine, amiloride, and different selectivity for Ba2+ and Ca2+ indicated that the underlying channels for the observed currents are only of the T- and L-types that resemble those of the endocrine secretory cells.

Electrophysiological characteristics of cultured human umbilical vein endothelial cells.

Electrophysiological characteristics of cultured human umbilical vein endothelial cells (HUVEC) were determined using the patch-clamp technique in the whole cell configuration. In isolated cells, membrane potential, capacitance, and input resistance were (Mean +/- SD) - 16.3 +/- 12.7 mV, 53.9 +/- 26 pF, and 2.3 +/- 1.3 G omega, respectively (N = 26); and in confluent cells - 23.6 +/- 5.5 mV, 127 +/- 59 pF, and 0.254 +/- 0.077 G omega, respectively (N = 6). The almost 10 times higher input resistance, and smaller capacitance of isolated versus confluent cells, indicated that the latter were in electrical communication, presumably through open gap junctions, which was confirmed by intercellular diffusion of Lucifer Yellow. Whole-cell currents of isolated cells were made up of at least three components: First, two outward currents, an early transient one with activation-inactivation kinetics and a small delayed sustained component with 6.75 +/- 4.8 and 0.73 +/- 0.089 nS conductance, respectively. Second, an inward component which was rectified and had 1.58 +/-1.2 nS conductance. In contrast to a reported lack of voltage-gated channels in HUVEC, the above currents were voltage dependent. Inhibition of the whole-cell currents by external Ba2, internal Cs, and other K+ blockers indicates that the three observed currents are carried by K+. This was confirmed by changes of outside K+ concentrations shifting the I-V curve intercept in the direction expected for K(+)-selective channels. Voltage-gated Ca2+ currents were not apparent in the whole-cell current records. HUVEC membrane potential was as low as that of microvascular cells, while inward current rectification at normal external K+ was like that in arterial endothelial cells. This mixed phenotypic expression and multipotential behavior suggests that the electrical features of HUVEC may be primarily determined by embryonic origin and the local effect of the microenvironment rather than strictly by vessel size.

Albumin modulation of paracellular permeability of pig vena caval endothelium shows specificity for pig albumin.

We measured endothelial hydraulic conductivity (Lp,e) and endothelial electrical resistance (Re) of pig vena cava in pig serum albumin (PSA) and bovine serum albumin (BSA) to determine whether specificity for the autologous albumin found in dog vena cava is a general property of mammalian endothelium. Pigs were anesthetized with pentobarbital sodium for surgical removal of the thoracic inferior vena cava. A vessel segment was placed as a membrane separating two compartments in a chamber. Vessels kept in 30 mg/ml BSA had Lp,e values of 1.30 +/- 0.81 x 10(-7) cm.s-1.cmH2O-1 (n = 9) and Re values of 13.8 +/- 2.6 omega.cm2 (n = 3). Vessels kept in PSA had Lp,e values of 0.257 +/- 0.125 x 10(-7) cm.s-1.cmH2O-1 (n = 5) and Re values of 21 +/- 6.3 omega.cm2 (n = 4). The differences between Lp,e and Re in BSA and PSA were statistically significant (P < 0.05). In vessels kept in PSA, switching to BSA caused a doubling of Lp,e. Lower Lp,e and higher Re in PSA as compared with those in BSA suggest that specificity for the autologous albumin is a general phenomenon in mammalian endothelium and that albumin binding to specific sites is associated with its permeability effect.

Enzymatic lysis of sulfated glycosaminoglycans reduces the electrophoretic mobility of vascular endothelial cells.

The main purpose of this work was to identify the macromolecules carrying the surface charge of endothelial cells. This was done by measuring changes in cell electrophoretic mobility caused by enzymatic removal of glycocalyx components. Endothelial cells were removed from the bovine pulmonary artery using nonenzymatic procedures, plated, and identified by immunocytochemical methods and electron microscopy. Cultured cells were suspended in saline and placed in the lumen of a capillary in a Rank Brothers electrophoresis instrument. Voltage was applied between the ends of the capillary, and the velocity acquired by the cells was measured with a microscope. Preincubating the cells in protein-free saline for 1 h reduced the mobility by 25%. This reflects the loss of proteoheparan sulfate from the cell surface. Cell mobility was totally suppressed by exposing the entire cell surface to chondroitin sulfate lyase, but it was only slightly diminished when the enzyme was applied only to the cell side facing the culture medium. A partial decrease in mobility was obtained after enzymatic removal of either heparin, heparan sulfate, or collagen. The results indicate that sulfated glycosaminoglycans are the main carriers of the surface change in vascular endothelial cells. The asymmetrical effect of chondroitinase on the two sides of the cell indicates a distribution polarization for glycosaminoglycans in endothelial cells.

Surface charge of endothelial cells estimated from electrophoretic mobility.

The surface charge density of endothelial cells was estimated from cell electrophoresis. Cultured endothelial cells from the bovine pulmonary artery were suspended in saline and placed in the lumen of a glass capillary. A voltage was applied across the capillary ends and the velocity imparted to the cells was measured with a microscope. Erythrocyte mobility was also measured. The mobility in (micron/s)/(V/cm) was 0.74 +/- 0.08 for endothelial cells and 1.03 +/- 0.15 for erythrocytes. Charge density in esu/cm2 was calculated as 2.62 x 10(4) and 0.91 x 10(4) for endothelial and red cells, respectively. Removal of sialic acid did not affect the mobility of endothelial cells, but it reduced that of red cells to near zero. Endothelial cell mobility decreased either with ionic strength or calcium concentration. Our results strongly suggest that the surface charge of endothelial cells is dependent on sulfated glycosaminoglycans.

Effect of albumin on the width of water channels in venous endothelium.

The effect of bovine and dog albumin and plasma was measured on the hydraulic conductivity (Lp) and electrical resistance (R) across the endothelium of the dog vena cava. An estimate of changes in size of transendothelial water channels was then obtained by combining Lp and R values. Dogs were anesthetized with pentobarbital sodium to surgically remove their thoracic inferior vena cava. Lp was measured by a gravimetric method at constant vessel volume. R was obtained from the transvascular voltage changes caused by current pulses. Lp measured at the beginning of the experiments was 0.97 +/- 0.17 X 10(-7) cm.s-1.cmH2O-1 in plasma and 2.75 +/- 0.48 X 10(-7) cm.s-1.cmH2O-1 in Ringer solution. Bovine albumin produced an 18% reduction in Lp relative to its value in Ringer solution. Dog albumin and plasma lowered Lp 50 and 54%, respectively. The differences were statistically significant with P less than 0.05. R increased from 15.17 +/- 7.0 to 26.5 +/- 16.0 omega.cm2 in dog albumin and 27.4 +/- 9.6 omega.cm2 in dog plasma. Calculations using a model for the transendothelial channel and equations for Lp and R showed a decrease in channel width of 172 A due to the protein. This could be accounted for by the thickness of a monolayer of albumin absorbed to the channel walls. Results suggest that the permeability response to plasma proteins is a general property of vascular endothelium.

Hydraulic conductivity of venous endothelium as measured with a volume clamp method.

Hydraulic conductivity (Lp) of cava vein endothelium was obtained with a new method that measures fluid flow at constant volume. Dogs were anesthetized with pentobarbital sodium. A vessel segment was removed, cannulated, and hung free from a force transducer calibrated to measure weight. Hydraulic pressure difference drove fluid across the vessel wall. This fluid was washed out from the vessel surface by external perfusion, causing a weight loss that was transmitted to a Grass polygraph through the force transducer. Shifting the polygraph balance caused a voltage output which was used to activate a controller operating a microsyringe pump. By injecting fluid into the vessel lumen, this servomechanism counteracted weight changes and kept volume and pressure at a fixed level. Recording the volume injected allowed continuous monitoring of fluid flow. This avoided the problems caused by frequent pressure and volume resetting in previous methods. Fluid flow was a linear function of the applied pressure (r = 0.87). Lp was 0.91 +/- 0.05 (SD) X 10-7 cm X s-1 X cmH2O-1. This value was very similar to those in continuous capillaries and arterial endothelium.

Electrical conductivity and its use in estimating an equivalent pore size for arterial endothelium.

Transport characteristics of capillary endothelia have been studied in numerous preparations. Little is known, however, regarding transport across the ultrastructurally similar arterial endothelium. The present studies describe a method for determining endothelial electrical resistance in the isolated, Ringer-perfused rabbit aorta. By use of silver-silver chloride electrodes, total resistance (RT) of the vessel wall was found to be 18.4 +/- 6.5 omega . cm2 (mean +/- SD; n = 10) in response to transmural passage of constant current pulses. After detergent removal of the endothelium, outer layer resistance (Rol) was 8.7 +/- 5.1 omega . cm2. Endothelial resistance (Re) was calculated to be 9.7 +/- 4.1 omega . cm2, since RT = Rol + Re. Combination of Re with previously measured hydraulic conductivity of this endothelium yielded an equivalent endothelial pore radius of 85 A, whereas a fractional pore area of 6.5 X 10(-4) was estimated from conductivity data. An alternative analysis demonstrated an endothelial slit width of 102 A. Re is similar to electrical resistances of leaky epithelia, frog mesenteric capillary, and frog muscle capillary, suggesting similar permeability characteristics in both arterial and capillary endothelia.

Vascular resistance and transit time of sickle red blood cells.

Changes in vascular resistance and mean transit time caused by the introduction either normal or sickled red blood cells (SSRBC) into the circulation were measured in the coronary and mesenteric vascular beds. The cat mesentery was perfused at constant flow through the superior mesenteric artery with a Ringer solution containing albumin. A rise in input pressure was observed after introducing RBC's. (10% hematocrit) at a constant rate, into the perfusate. The pressure rise caused by SSRBC's relative to that caused by normal RBC's represents the relative change in vascular resistance caused by the former cells as the perfusion flow was the same for both infusions. Similar measurements were performed in Ringer's perfused rabbit hearts in vitro. SSRBC's produced a 2 +/- 0.12 (M +/- SE) times relative increase in resistance in the mesentery and a 4 +/- 0.5 times increase in the coronary vasculature. Mean transit time, t was measured by monitoring the light transmission change caused by the injection of a small amount of RBC's into the arterial side of a Ringer's perfused rabbit heart. For normal RBC's t was 3.67 +/- 0.08 s and it rose to 6.1 +/- 0.6 s when SSRBC's were used. The simultaneous measurement of both mean transit time and pressure rise may help to identify the cell changes that cause circulatory impairment is SS cell anemia patients and it may also provide a convenient index to relate with the severity of the crisis.

Secondary driving forces affecting transcapillary osmotic flows in perfused heart.

We evaluated secondary hydrostatic and osmotic pressure gradients caused by an osmotic flow across the wall of rabbit heart capillaries in vitro. Tissue fluid loss after addition of 80 mM sucrose to the Ringer perfusate depressed interstitial fluid pressure (IFP) by 13.2 +/- 1.1 cmH2O, as measured with a fine intramyocardial needle. This, plus a calculated rise in capillary pressure, caused a difference of 4.3 cmH2O in transcapillary pressures which represents 1.4% of the simultaneous and opposite osmotic pressure difference. The effect of transient-induced experimental changes in interstitial effect of transient-induced experimental changes in interstitial concentration of NaCl (main resident solute) during an osmotic transient and osmotic balance experiments using opposing transcapillary flows (Jv) and a sigma sucrose = 2.9 sigma NaCl. Long-term perfusion with solutions having abnormal concentrations of NaCl did not influence Jv, whereas inclusions of sucrose as resident solute depressed Jv by 10%. Results indicate that secondary hydrostatic and osmotic forces are small relative to the main osmotic driving force. Compliance measurements and calculations suggest that cells contribute three-fourths organ water loss during an osmotic transient, thus buffering volume, pressure, and concentration changes in the interstitial spaces.

Transcapillary osmotic flows in the in vitro perfused heart.

Hearts were removed from rabbits anesthetized with pentobarbitol sodium, heparinized, and then perfused through the aorta with Ringer solution. Addition of sucrose to the perfusate caused an osmotic transcapillary flow (Jv). Measurements of Jv with two independent methods, one using the rate of organ weight change and the other the variation of effluent concentration of an impermeant dye (Blue dextran), agreed very closely, giving an initial Jv per unit concentration and wet heart weight of 0.306 (microliters/s) (mmol/l).g (corrected for viscosity, 25 degrees C). Because dye dilution was free of vascular volume interference, its agreement with weight measurements suggests that vascular volume changes were relatively small. Measurements using 51Cr-labeled erythrocytes supported the above conclusion. The effect of temperature and concentration on Jv was ascribed to viscosity changes. Organ condition remained stable for several hours, based on maximum ventricular pressure (107 +/- 6.4 cmH2O) and dP/dt (1,145 +/- 98 cmH2O/s) values close to those in blood-perfused rabbit hearts. Repeat weight response to osmotic testing showed approximately 5% variation during an experiment.

Permeability and model testing of heart capillaries by osmotic and optical methods.

Rabbits were anesthetized with Nembutal; their hearts were removed and perfused with Ringers solution. Osmotic weight transients were then produced by test solute infusion into the perfusate. The rate constant for organ weight change was used to predict test solute venous concentration (CV) and the distribution volume (V) during an osmotic transient by use of the Johnson and Wilson (6) model for capillary tissue exchange. By use of Cr EDTA, strongly purple in solution, we compared the above predictions with a value of CV obtained directly by optical measurement of outflow venous concentration and values of V calculated from our measurements of organ weight and the known sucrose distribution volume. The close agreement between both sets of values with a permeability coefficient of 5 x 10(-5) cm/s and a volume of distribution of 3.2 ml/10 g heart lead us to conclude that the model used closly represents the conditions of the isolated perfused heart, and that both the osmotic transient and the extraction measurements provide good estimates of organ capillary permeability.

Hydraulic conductivity of the endothelial and outer layers of the rabbit aorta.

Pressure-driven fluid flow across the arterial wall was measured to determine wall hydraulic conductivity (Lp) before and after removal of the endothelium. The thoracic aortas of rabbits, anesthetized with Nembutal, were cannulated, perfused with oxygenated Ringer solution, and removed. With one cannula connected to a capillary manometer and the other closed, the manometer meniscus shift could be used as an indication of fluid loss through the wall plus vessel volume increase (creep). The latter effect, when measured, accounted for about one-fourth of the total volume displacement. The Lp given in cm/(s.cmH2O) +/- SD, was 3.30 +/- 0.96 x 10(-8). Another method employed continuous weighing of a closed aortic segment to obtain fluid loss, and yielded an Lp of 4.07 +/- 1.3 x 10(-8), and after mechanically removing the endothelium, the Lp became 7.73 +/- 2.8 x 10(-8). Using the above data, an Lp could be calculated for aortic endothelium of 8.6 x 10(-8). This suggests that about half the total transmural pressure drop occurs across the endothelium. Scanning electronmicrographs were used to check the condition of the endothelium.

Effect of hyperosmolarity on resting and developed tension in heart muscle.

Simultaneous changes in weight, tension, and electrical activity were studied in the isolated perfused rabbit heart when the Ringer solution perfusion fluid was made hypertonic by the addition of sucrose, urea, glycerol, ethylene glycol, or formamide. The typical responses to each of the molecules was an initial drop in weight and tension followed by a return toward the base-line level. A 0.4 M concentration of sucrose, urea, or glycerol reduced the weight to 40 +/- 4.6, 48 +/- 3, and 52.2 +/- 3% of the initial value, respectively. The tension was simultaneously reduced to 23 +/- 3.5, 31 +/- 2, and 41 +/- 4% of its initial value. The tension drop produced by the solutes tested was linearly related to the amount of water lost by the heart. The falling phases of both tension and weight loss were closely correlated phases of both tension and weight loss were closely correlated in time a magnitude and were both related to the effectiveness of a particular molecule to move water out of the cells. In contrast, the subsequent rising phases of tension and weight were not as well correlated in time and magnitude. Whereas for most of the molecules used, the tension recovery was incomplete; urea, in contrast, caused an overshoot of the control tension level, thus pointing toward a unique inotropic effect of this compound. Resting tension rose for both urea and sucrose but not for the other compounds.

Permeability of barnacle muscle fibers to water and nonelectrolytes.

The permeability of isolated muscle fibers of the giant barnacle Megabalanus psittacus to water and nonelectrolytes was studied by determining their reflection and permeability coefficients. Reflection coefficients were obtained by comparing the osmotic fluxes produced by a test molecule to that produced by the impermeant sucrose molecule. Permeability coefficients were determined for measurements of tracer uptake. The results indicate that, in these fibers, nonelectrolyte permeability is closely related to lipid solubility and molecular size. A value of 2.16 X 10(-12) cm3/sec dyne for the hydraulic conductivity and a value of 10.45 X 10(-4) cm/sec for 3HHO permeability coefficient were obtained. The effect of membrane surface invaginations and clefts on the determination of permeability coefficients is discussed.

Filtration coefficient of the axon membrane as measured with hydrostatic and osmotic methods.

The hydraulic conductivity of the membranes surrounding the giant axon of the squid, Dosidicus gigas, was measured. In some axons the axoplasm was partially removed by suction. Perfusion was then established by insertion of a second pipette. In other axons the axoplasm was left intact and only one pipette was inserted. In both groups hydrostatic pressure was applied by means of a water column in a capillary manometer. Displacement of the meniscus in time gave the rate of fluid flowing across the axon sheath. In both groups osmotic differences across the membrane were established by the addition of a test molecule to the external medium which was seawater. The hydraulic conductivity determined by application of hydrostatic pressure was 10.6 +/- 0.8.10(-8) cm/sec cm H(2)O in perfused axons and 3.2 +/- 0.6.10(-8) cm/sec cm H(2)O in intact axons. When the driving force was an osmotic pressure gradient the conductivity was 4.5 +/- 0.6 x 10(-10) cm/sec cm H(2)O and 4.8 +/- 0.9 x 10(-10) cm/sec cm H(2)O in perfused and intact axons, respectively. A comparable result was found when the internal solution was made hyperosmotic. The fluid flow was a linear function of the hydrostatic pressure up to 70 cm of water. Glycerol outflux and membrane conductance were increased 1.6 and 1.1 times by the application of hydrostatic pressure. These increments do not give an explanation of the difference between the filtration coefficients. Other possible explanations are suggested and discussed.