INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

STUDIES FROM OUR LABORATORY have shown that electrical stimulation of the vagal nerve will lower the interstitial fluid pressure (P_if) in the tracheal tissue (24, 25). The normal, slightly subatmospheric P_if becomes negative after the onset of nerve stimulation, and this negativity accompanies the other signs of neurogenic inflammation, namely vasodilation, increased vascular permeability, and plasma protein extravasation followed by edema formation (8, 10). The lowering of P_if is the major hydrostatic pressure driving the rapid formation of edema that develops in this condition.

Transcapillary fluid flux (J_v) is the product of the capillary filtration coefficient (CFC), the hydrostatic pressure (P), and the colloid osmotic pressure (COP) acting across the capillary wall (1). These parameters are interrelated in Eq. 1

(1)

where the subscripts "c" and "if" denote capillary and interstitial fluid pressures, respectively.  is the capillary reflection coefficient for proteins, and P is the net filtration pressure across the capillary wall. P_if is one of the pressures that controls transcapillary fluid transport. Transcapillary fluid flux is normally kept within narrow limits, because of the "autoregulation" by P_if and COP_if. Increased fluid filtration and thereby increased interstitial volume will raise P_if and reduce COP_if. These changes will in turn normalize net filtration pressure and thereby counteract further filtration. This balance is maintained in normal conditions but undergoes dramatic changes in acute inflammatory reactions. The rapid development of edema within minutes after acute tissue injury is a fact of human experience, yet the processes underlying this phenomenon are not well characterized. In earlier studies, the increased vascular permeability and leakage of plasma proteins provoked by agents such as histamine were described, and it was assumed that regulation of gap formation and capillary permeability were key determinants controlling fluid egress from the microcirculation. Another mechanism, identified in our laboratory, is the lowering of P_if. P_if is one of the pressures participating in the control of transcapillary flux according to Starling's hypothesis. P_if normally counteracts edema formation, but in the early phase of inflammation it becomes a driving pressure for edema. Therefore, in our experiments P_if was measured after the induction of circulatory arrest with intravenous KCl to avoid masking the decreased P_if and also to avoid the formation of edema that will occur if the circulation is intact. It was shown that during an inflammatory response, lowering of P_if could withdraw fluid from the capillaries into the interstitial space, which accounts for the rapid development of edema observed, for example, in blisters after a burn injury. By using inflammation of tracheal tissue as an experimental model, it was shown, without induction of cardiac arrest, that increased fluid flux can double interstitial fluid volume within 10 min after initiation of the stimulus (7).

To characterize the factors that influence P_if, we have examined several pharmacological agents that have been shown to reduce inflammatory edema in experimental models of acute tissue injury. These agents include corticotropin-releasing hormone, -trinositol, neurotensin fragments, and mystixins (4, 19, 25, 26). Mystixins are 7- to 11-residue synthetic peptides that inhibit heat-induced swelling of the skin and pulmonary edema at doses of 15-100 µg/kg iv (19, 20). Mystixin-7 (100 µg/kg iv) also reduced substance P (5 µg/kg iv)-induced plasma leakage in rat trachea but did not affect the substance P-enhanced opening of endothelial gaps (2). This action of mystixin was unusual in that protein extravasation is considered to result from the opening of endothelial gaps. The aim of this study was to determine whether the mechanism involved in the anti-inflammatory effect of mystixin-7 is linked to changes in P_if. Here, we tested the effect of mystixin-7 on the lowering of P_if in rat trachea after electrical stimulation of the vagal nerve. The effect of mystixin-7 pretreatment on tissue imbibition of water in vitro was also examined.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Female Wistar rats (200-250 g) were purchased from Møllergaard Breedings (Ry, Denmark). Food was available ad libitum. Animals were anesthetized with pentobarbital sodium (initial dose of 50 mg/kg body wt ip, with supplemental doses administered when required). A branch of the femoral vein was cannulated for injection of mystixin-7 (1, 10, 20, or 100 µg/kg body wt) or saline (0.9% NaCl) or for the injection of radiolabeled tracers. Circulatory arrest was induced with 0.5 ml of saturated KCl injected intravenously. All experimental protocols were approved by, and performed in accordance with the recommendations of, the Norwegian State Commission for Laboratory Animals.

Measurements

P_if. After the induction of circulatory arrest, an extrathoracic portion of the trachea was exposed, and the muscle lying over the trachea was split longitudinally. The left vagal nerve was dissected free from the carotid artery, at the same level as the exposed trachea, and placed on a stimulating electrode. The tracheal surface and the vagal nerve were covered with mineral oil to avoid desiccation. The time required for the experimental preparation was between 3 and 5 min. Measurements of P_if were initiated thereafter and continued until 60 min after cardiac arrest. P_if was measured with a sharpened glass pipette (4-10 µm) penetrating the frontal portion of the tracheal tissue between the cartilage rings. The pipette tip was inserted into the tissue from the outermost layer of the adventitia, via the submucosa, and may have gone as deep as the mucosa layer, but the registration of pressures was most likely measured in the submucosa. Measurements were performed without opening the trachea. The glass capillary was connected to a servo-controlled counterpressure system (22, 23), and the counterpressure created by the servo-controlled pump (model 201; Ling Dynamic Systems, Royston, UK) was recorded with a pressure transducer (1280C, Hewlett-Packard) connected to an amplifier and recorder (Gould Instruments, Ballainvilliers, France). The glass pipettes were filled with 0.5 M NaCl colored with Evans blue to visualize the tip of the pipette. Micropuncture was performed under visual guidance with a microscope (Wild M3C, Heerbrugg, Switzerland), and care was taken to minimize stretch or compression at the site of puncture (23). The measurements were accepted when the following criteria were met: 1) no change occurred in recorded pressure when feedback gain was increased; 2) suction applied by the servo-controlled pump gave an increased electrical resistance in the pipette, verifying an open communication to interstitial fluid because of the lower tonicity of the fluid entering the pipette; and 3) baseline measurements before and after P_if registration were unchanged. This last measurement was performed in a plastic cup filled with saline placed at the level of the site of puncture. All surgical manipulations on or near the trachea were performed postmortem to prevent local inflammation that might increase interstitial fluid volume and confound accurate estimates of P_if.

Electrical stimulation. Electrical stimulation (ES) was performed with a Grass stimulator (S60, Grass, Quincy, MA) with the parameters 20 V, 20 Hz, and 0.5 ms for a total of 15 min.

Interstitial fluid volume, total tissue water, and albumin extravasation. The animals in these experiments were bilaterally nephrectomized to prevent filtration of the tracer ⁵¹Cr-EDTA, which gives the measure for extracellular volume (ECV) and interstitial fluid volume (IFV). The radiolabeled tracer ⁵¹Cr-EDTA (1 MBq in 0.3 ml; Institute for Energiteknikk, Kjeller, Norway) was injected intravenously 65 min before induction of cardiac arrest. Fifteen minutes after the injection of ⁵¹CR-EDTA, 0.2 ml of ¹²⁵I-labeled human serum albumin (¹²⁵I-HSA; 0.05 MBq; Institute for Energiteknikk) was injected intravenously, followed by intravenous mystixin (1, 10, 20, or 100 µg/kg body wt) or saline. The injected volume was 0.1 ml/100 g body wt plus 0.2 ml of saline for flushing the cannula. Five minutes before cardiac arrest and forty-five minutes after mystixin injection, the rats received intravenous injection of 0.2 ml of ¹³¹I-HSA (0.10 MBq; Institute for Energiteknikk). Blood samples (0.5-1.0 ml) were taken by cardiac puncture before cardiac arrest. Samples of trachea were placed in preweighed vials, which were sealed immediately and reweighed. The radioactivity in the tissue samples and plasma was determined in an LKB Wallac 1282 gamma counter (LKB, Turku, Finland) with automatic spillover and background subtraction. The samples were then dried at 65°C until a constant weight was attained and total tissue water content (TTW) was calculated as fresh weight (fw) minus dry weight (dw) divided by dry weight:TTW=(fwdw)/dw. Distribution volumes and extravasation of albumin (E_alb) were calculated as plasma equivalent space, i.e., counts per minute (cpm) of tracer per gram of dry tissue weight divided by cpm per milliliter of plasma. IFV was estimated as ⁵¹Cr-EDTA space (i.e., the ECV of tissue) minus the 5-min plasma equivalent space for ¹³¹I-HSA (i.e., the plasma volume of the tissue, or PV): IFV=ECVPV. Albumin extravasation (E_alb) was taken as the difference between the 50-min plasma equivalent space for ¹²⁵I-HSA and PV: E_alb = ¹²⁵I-HSA  PV. All calculations were performed per gram of dry tissue weight.

In vitro swelling of tracheal tissues. The animals were treated as before, with cannulation of the femoral vein and intravenous injection of mystixin-7 (20 or 100 µg/kg body wt) or saline. After cardiac arrest, the tracheas were dissected out, split longitudinally, placed in preweighed vials, sealed, and weighed. The vials containing one trachea were then filled with 5 ml of phosphate-buffered saline (PBS, 0.15 M adjusted to pH 7.4). After 2, 4, 8, and 24 h of soaking, the tracheas were removed from the vials, the surface fluid was wiped gently with a paper towel, and the tracheas were weighed. After being weighed, the sample was placed back into the vial. After the last weighing (24 h), samples were dried at 65°C to obtain tissue dry weight. TTW was then estimated for each time interval by assuming a constant dry weight during the 24 h of immersion.

Solutions. Mystixin-7 [molecular weight (MW) 1,041] was synthesized using the solid phase method, as described previously (19). The stock solution of 100 µg/ml was prepared by dissolving mystixin-7 in 100 µl of 0.1 M acetic acid and diluting to the required concentration with 0.9% NaCl. This solution or further dilutions with 0.9% NaCl were used for intravenous injections at 0.1 ml/100 g.

Experimental Protocol

P_if. Mystixin (1, 10, or 20 µg/kg body wt, n = 8, 8, and 4, respectively) was administered intravenously. Forty-five minutes postinjection, cardiac arrest was induced with intravenous injection of 0.5-1 ml of saturated KCl. Measurement of P_if was made shortly after the left vagal nerve had been surgically isolated and placed in the electric stimulator and after the trachea had been exposed. One or two recordings of P_if were obtained in the controls before sensory nerve stimulation. After nerve stimulation, repeated measurements were performed until 60 min after circulatory arrest. Animals in the control group that had electrical stimulation without pretreatment of mystixin (n = 8) were given 0.1 ml/100 g body wt 0.9% saline intravenously, and circulatory arrest was induced with intravenous injection of saturated KCl.

IFV, TTW, and E_alb. The series of measurements of IFV, TTW, and E_alb included four groups that were pretreated with intravenous mystixin (1, 10, 20, or 100 µg/kg body wt, n = 8 in each group) and one control group (n = 8) that received saline intravenously instead of mystixin. The injection volume was 0.1 ml/100 g body wt plus 0.20 ml of saline to flush the catheter. The timing for the injection of the tracers was according to the following schedule: 0 min, ⁵¹Cr-EDTA; 15 min, ¹²⁵I-HSA and mystixin/vehicle; 60 min, ¹³¹I-HSA; and 65 min, blood sample and KCl.

In vitro swelling of tracheal tissues. The experiments, with the two doses of mystixin-7, were not carried out simultaneously (see DISCUSSION). The first series included mystixin 100 µg/kg body wt (n = 5) and a control group (n = 6). The second series included mystixin 20 µg/kg body wt (n = 8) and a control group (n = 8). The anesthetized animals were cannulated and pretreated with mystixin or saline at an injection volume of 0.1 ml/100 g body wt plus 0.20 ml of saline to flush the catheter.

Statistical Analysis

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

P_if. P_if before ES (n = 8) was 1.2 ± 0.7 mmHg and decreased to 4.7 ± 1.0 mmHg (P < 0.01) after stimulation (Fig. 1), which is in agreement with results from previous studies (24). Mystixin-7 suppressed the lowering of P_if produced by ES. The attenuation was significant (P < 0.01) at 10 and 20 µg/kg body wt iv but not at 1 µg/kg body wt (Fig. 1). The measurements of P_if, with mystixin 10 and 20 µg/kg body wt iv were stable during the 60-min observation period.

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Mystixin-7 (1, 10, and 20 µg/kg body wt iv) inhibits development of lowering interstitial fluid pressure (Pif) in the trachea of anesthetized rats after antidromic stimulation of the vagus nerve. Values are means ± SD. *P < 0.01, using 1-way ANOVA with subsequent Bonferroni's comparison and t-tests of the changes in Pif from before and after stimulation [change in Pif = (Pif before stimulation)  (Pif after stimulation)]. **Significant inhibition of fall in Pif (P < 0.01).

IFV, TTW, and E_alb. Forty-five minutes after mystixin-7 (100 µg/kg body wt iv) was applied, E_alb and plasma equivalent space for ¹²⁵I-HSA were significantly elevated (Table 1). The lower doses of mystixin-7, namely 1, 10, and 20 µg/kg body wt iv, did not change IFV, TTW, E_alb, or plasma equivalent space for ¹²⁵I-HSA, compared with values for saline-injected animals.

In vitro swelling of tracheal tissues. The samples of tracheal tissues removed from the animals and placed in PBS did accumulate fluid. As expected from the higher E_alb, tracheal samples from animals treated with mystixin-7 at 100 µg/kg body wt iv (n = 5) also had greater fresh weight TTW (TTW0h) values (2.5 ± 0.2 ml/g dry wt) than the corresponding control (2.0 ± 0.1 ml/g dry wt) (P < 0.01; Table 2). At this dose, the rate of fluid accumulation in the tissue samples at various time intervals did not differ from the controls. In tissues from animals treated with 20 µg/kg body wt iv, TTW0h was similar to that of controls, but there was a significant inhibition of fluid accumulation as measured by TTW at 2 and 4 h after placement of the tissues in phosphate buffer (Table 2 and Fig. 2).

View larger version (13K): [in this window] [in a new window]   Fig. 2.   In vitro swelling of tracheal tissues was significantly reduced with mystixin-7 (20 µg/kg body wt iv). The inhibitory effect was observed at the 2- and 4-h intervals. Values are means ± SE; n = no. of tissues. **P < 0.05; *P = 0.054, using 1-way ANOVA with subsequent Bonferroni and t-tests. P < 0.0000 using 2-way ANOVA with repeated measures.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The results from this study show that the lowering of P_if, which developed during neurogenic inflammation in the trachea of rats, is inhibited by the peptide mystixin-7. It was also shown that pretreatment of animals with a low dose of mystixin-7 (20 µg/kg body wt iv) reduced fluid accumulation in the tissues placed in PBS for 2-4 h. These results may provide new insights into how a peptide such as mystixin-7 inhibits edema formation in vivo.

A change in mean arterial pressure (P_a) may reduce transmural driving forces for fluid and protein egress from the microcirculation via an effect on capillary pressure (P_c), as described in Eq. 1, and thereby could also possibly affect P_if. Both ES of the vagus nerve and intravenous injection of mystixin-7 change P_a. The effect of vagal stimulation on P_a is not, however, applicable in the current experiments because ES was imposed after cardiac arrest. In a previous report (2), mystixin-7 was shown to reduce protein extravasation in rat trachea without a reduction of endothelial gap formation. The fall in P_a after mystixin injection was considered to be a possible contributor in the reduction of protein extravasation. However, several other observations suggest that the effect of mystixin-7 on P_if is not determined by changes in P_a. First, the P_if measurement after mystixin injection and before nerve stimulation is the same as that in the vehicle-treated animals, i.e., slightly subatmospheric. Second, it has been shown that agents such as substance P and calcitonin gene-related peptide lower P_a but have independent effects on P_if (5). We therefore conclude that the effects produced by mystixin-7 on P_if are not due to hemodynamic changes.

The observation that mystixin-7 (at 100 µg/kg body wt iv) increased TTW, E_alb, and ¹²⁵I-HSA plasma equivalent space was unexpected. The in vitro swelling experiments of tracheal tissue were first carried out using mystixin-7 at 100 µg/kg body wt. Compared with the controls, there was a significant increase in accumulated fluid in the mystixin pretreated rats before the swelling occurred, indicating a proinflammatory effect of mystixin-7 at this dose. Therefore, we performed a study using the radiolabeled tracer technique to determine the IFV, TTW, and E_alb with animals pretreated with mystixin-7 at doses from 1 to 100 µg/kg body wt. We have also noted in unpublished studies (Strukova H and Wei ET) that mystixin-7 at concentrations exceeding 0.1 µM (corresponding to 0.1 µg/ml mystixin, MW 1,041) will cause a release of histamine from isolated rat peritoneal mast cells in vitro, and this action may explain the local tissue inflammation. Furthermore, it is also known that some substances can induce both pro- or anti-inflammatory reactions depending on the route of administration (13), tissue, or species (3, 14). With the radiolabeled tracer technique, the proinflammatory effects were not observed at 20 µg/kg body wt iv, a dose that inhibits lowering of P_if after nerve stimulation. At this lower dose, the ability of mystixin-7 to inhibit tissue imbibition of water, as measured by tissue weights, was demonstrated.

Several explanations for the mystixin-7 inhibition of fluid uptake in vitro can be proposed on the basis of observations by Meyer et al. (11, 12) on the swelling properties of loose connective tissue, which they determined using umbilical cord as their model system. First, it is known that the tissue content of hyaluronan and other glycosaminoglycans facilitate uptake of water and are responsible for tissue swelling. Recently, mystixins were tested and found to inhibit transforming growth factor--stimulated hyaluronan production in cultures of human HS27 dermal fibroblasts at an EC₅₀ of 0.01-0.10 µg/ml (21). Furthermore, because hyaluronidase inhibits swelling in the umbilical cord by enzymatic degradation of hyaluronan (11, 12), mystixin-7 could enhance the production of hyaluronidase, which would both inhibit swelling of the tracheal tissue and inhibit lowering of P_if. Swelling occurs at a different rate for mystixin-7 (20 µg/kg body wt) than for the controls, but at the end of the experiment the swelling had reached the same level, making it unlikely that mystixin-7 could affect the content of hyaluronan or the other glycosaminoglycans as well as hyaluronidase activity.

A second possibility is that there is a direct interaction between collagen and mystixin-7 in the swelling process. This mechanism appears unlikely because the concentration of collagen (MW 95,000, -chain, type I collagen), 8.4 nmol/g (0.8 mg/g) wet wt skin interstitium (15) relative to the injected mystixin dose of 0.2 nmol/kg body wt (20 µg/kg body wt), would be distributed in the whole rat interstitium at 5.7 fmol/g (1.43 × 10⁴ mg/g), and it is therefore not likely that mystixin-7 is directly targeting the collagen network and, hence, strengthening the structure to avoid swelling. The solution in which the swelling took place will also provide a supplementary dilution of mystixin-7 with a reduction of concentration by another 1,000 times, further minimizing the likelihood of a direct interaction.

A third possibility is that mystixin-7 may act on cellular receptors that control the tension of the extracellular matrix. A possible pathway would be the stimulation of the ₁-integrin receptor system, which appears to be a common and final step involved in generating a lowering of P_if, because polyclonal and monoclonal antibodies to ₁- and ₂₁-integrin, respectively, decrease P_if (16-18). Thus we postulate that mystixin-7 acts via a cellular receptor, which in turn causes strengthening of the cytoskeleton, stabilizing the ₁-integrin function. This will subsequently vanish in our system because we make no attempt to keep the cells viable, and swelling stops at the same end point regardless of the treatment regime used. It is difficult to see how the effects of mystixin-7, shown here as attenuation of swelling and inhibition of changes in P_if, would not involve the components of the extracellular matrix. However, the exact process by which mystixin-7 may affect P_if and swelling via the integrin system and the intracellular signaling pathway is unclear. In this context, it is of interest to note the structural similarity of mystixin-7 to a fragment derived from the COOH-terminal G-domain of the laminin-1 chain. This peptide, designated AG73 and having the sequence Arg-Lys-Arg-Leu-Gln-Val-Gln-Leu-Ser-Ile-Arg-Thr, has profound effects on epithelial morphogenesis and endothelial cell adhesion (6, 9). Together, these results suggest specific mystixin receptor mechanisms in epithelial and connective tissues that may influence P_if during tissue injury.

In summary, we have described the actions of a peptide that is able to alter the physical forces that generate edema in the acute phase of inflammation. Further studies of the receptive mechanisms of this peptide in the extracellular matrix may allow a more accurate description of molecular targets and cellular receptors that influence P_if and thereby control edema formation.