- Email: [email protected]
Disturbed brain purine metabolism results in a gross opening of the blood-brain barrier in newborn piglets following experimental pneumothorax
Neuroscience Letters, 113 (1990) 163-168
163
Elsevier Scientific Publishers Ireland Ltd.
NSL 06885
Disturbed brain purine metabolism results in a gross opening of the blood-brain barrier in newborn piglets following experimental pneumothorax P6ter Temesvfiri I, Csongor/kbrah~m l, Ferenc Jo6 2, J6zsef Kov~.cs 1, Zsuzsa Baranyai 3 and Katalin R~,cz 1 1Department of Pediatrics, Szent-Gy6rgyi Albert University Medical School, Szeged (Hungary), 2Laboratory of Molecular Neurobiology, Institute of Biophysics, Biological Research Center, Szeged (Hungary) and SDepartment of Pediatrics, University Medical School, Pbes (Hungary) (Received 4 December 1989; Revised version received 5 February 1990; Accepted 5 February 1990)
Key words: Piglet; Pneumothorax; Brain microvessel; Intravital fluorescence microscopy; CSF hypoxanthine Changes in the permeability of pial-arachnoideal microvessels [30-210/~m), of the blood-brain barrier (BBB), were intravitally studied by fluorescent microscopy and compared to the hypoxanthine (HX) level of cerebrospinal fluid (CSF) in newborn piglets (n = 24) using the open cranial window technique. Eight animals served as controls (Group 1), the others were studied in the course of bilateral experimental pneumothorax (BEP) using low (Na+-fluorescein, MW 376, Group 2) and large molecular weight (FITC dextran, MW 40000, Group 3) fluorescent tracer molecules. Cisternal CSF was sampled from the animals: 8 piglets from Group 1, and 4-4 piglets from Groups 2 and 3 at different stages of pathological condition: (i) at the critical (C) stage (severe acidosis, bradycardia, arterial hypotension and hypoxaemia) and also (ii) at the recovery (R) stage (mild metabolic acidosis, tachycardia, arterial hypotension) and the HX concentration was determined with high-pressure liquid chromatography. In Group I neither low (n = 4) nor large (n=4) molecular weight tracers penetrated BBB. In Group 2, however, the fluorescein dye passed BBB as a spotty leakage in animals at C stage (n =8). Diffuse fluorescein penetration was seen at R stage, too (n=4). In Group 3 no change in permeability was found at C stages (n=8), but at R stage (n=4), 2 h after the primary hypoxic insult, when the animals had recovered from cardiovascular and metabolic shock, the tracer passed locally the microvascular wall and appeared as leaky spots (number of leaky sites = 2.3 +0.4/0.10 cm 2, X + SE). The HX content in the CSF was significantly elevated ( P < 0.05) at stage C (26.39+3.2/zM/1, X,+SE), compared to data measured in Group 1 (8.4+ 1.7/LM/I, ~'__+SE). At stage R, it proved to be significantly lower (P<0.05) than in stage C, but still remained elevated (17.7+2.1 /tM/1, ~'+ SE; P < 0.05 vs. Group 1). Our results draw the attention of the size-dependent gross opening of the BBB in different stages of BEP and to the possible importance in the pathogenesis of mediators deriving from the disturbed brain purine metabolism.
Correspondence." P. Temesv~iri, Department of Pediatrics, Szent-Gy6rgyi Albert University, Medical School, P.O. Box 471, H-6701 Szeged, Hungary. 0304-3940/90/$ 03.50 © 1990 Elsevier Scientific Publishers Ireland Ltd.
164 Neonatal pneumothorax and its consequences are a major concern for physicians. Cerebral damage was reported to be more frequent in newborn infants suffering from pneumothorax than in babies who were also under intensive care but did not present this complication [9]. In an experimental model of this disease, we observed [14] severe alterations in the function of the pial-arachnoideal small venules during the acute hypoxic state consisting of leakage of Na +-fluorescein and sludging formation. The objective of the present study was two-fold: (i) to characterize changes in permeability of the pialarachnoideal small vessels in the recovery (R) stage after brain ischemia and (ii) to measure the hypoxanthine(HX) content in the cerebrospinal fluid (CSF) at different stages of pathological condition, in order to elucidate if purine metabolites released from the brain by oxygen deprivation can lead to permanent cerebral damage [5, 12]. In the present study 24 piglets were included. They were managed, monitored and an open cranial window was performed above the right parietal cortex as reported earlier [14]. The animals were divided into 3 experimental groups. Group I piglets. Eight animals without experimental bilateral pneumothorax (BEP). The piglets were ventilated and the pial-arachnoideai microvessels intermittently observed through 4 h with a W I L D Fluorescence Photomacroscope (M 400, Heerbrugg, Switzerland) after having given 2.0 ml/kg b.w. (I.0 vol,%)Na+-fluores cein (Mw 376, Aldrich Chem. Co. Ltd. Gillingham) into the umbilical vein at the beginning and 1 h thereafter in four piglets. The remaining 4 piglets were given 1.0 ml/kg b.w. (5.0 vol.%) FITC-dextran (Mw 40 000, Sigma, St. Louis) at the same time scale. Filtered light for fluorescence excitation was provided by a 50-W mercury lamp and by an I2 filter (Fa. Leitz, Wetzlar, F.R.G.). Microphotographs of the cortex were taken by a WILD MPS 55 photoautomat. CSF samples for the cisterna magna were obtained at the end of the experimental period. Group 2 piglets. Eight animals with BEP through the indwelling intrapleural drains [14]. Circulatory and respiratory failures were produced. In the same dose as in Group 1 Na + -fluorescein was given at the beginning (named basal (B)-stage) and at the critical (C)-stage. Criteria of this condition were: arterial pH < 7.0, HR < 60/ min, MABP < 30 mmHg. At this point, CSF was sampled from 4 animals and the others were resuscitated by giving an intravenous infusion of 8.0 mM/kg b.w. (8.4 vol.%) sodium bicarbonate diluted with an equal volume of glucose (10.0 vol.%) delivered over 15 min. In the meantime, intrapleural gas was gently sucked out, and these 4 animals were further ventilated without BEP. The experiments were terminated at stage R, 2 h after stage C. CSF was sampled. Group 3 piglets. Eight animals with BEP were handled in the same way as piglets in Group 2, except for the fluorescein tracer used. These piglets were given FITCdextran (MW 40 000) in the same doses as in Group I at the beginning and at stage C. As in Group 2, CSF was sampled from 4 animals at stage C, and the others were resuscitated and further ventilated till stage R, when CSF was sampled. Fluorescent photodocumentation was carried out also at B, C and R stages (diameter of vessels: 3(~210/~m) in Group 2 and 3. The CSF samples were stored at -20~'C until analyzed (Du Pont Instruments
165
TABLE I L A B O R A T O R Y D A T A O N N E W B O R N PIGLETS (Group 2 and 3) Values are ~', + SE. Stages of bilateral experimental pneumothorax B (n 16)
C (n 16)
R (n 8)
Time elapsed from B (min)
62.7 + 4.5
189.9+ 13.8
Cardiovascular parameters Mean arterial blood pressure (mmHg) 59.1 + 2.1 H R (beats/min) 162.9 _ 1.0
22.9 + 1.6 ~ 51.8 ___1.81
47.0+3.5 u t89.4+ 1.91,2
Arterial blood gases and acid-base status pH 7.41 +0.02 ( H C O 3 9 mM/l 22.8 __+0.7 PaCO2 (mmHg) 36.5 __+2.0 PaO: (mmHg) 58.0 + 2.5
6.89 + 0.031 9.3 __+0.51 70.6 + 5.51 29.0 + 2.51
7.22+0.031.2 15.6 ___0.8 u 40.9 ___3.02 54.0 ___3.32
Venous bhmd gases and acid-base status (superior vena cava) pH nm 6.81 +0.023 (HCO3-) mM/1 nm 9.2+0.4 PvCO2 (mmHg) nm 84.1 + 2.83 PvO2 (mmHg) nm 17.4+ 1.2~ B = Baseline, C = critical, R Lsignificantly differs (P < 0.05) 2significantly differs (P < 0.05) %ignificantly differs (P < 0.05)
nm nm nm nm
= recovery; n = number of animals studied, nm = not measured. from values measured at B stage. from values measured at C stage. from values measured from arterial samples (Student's unpaired t-test).
HPLC) by high-pressure liquid chromatography [2]. Cardiovascular, blood gas and acid-base parameters. In Group 1, the values remained within the physiological range. Data in both BEP Groups (Groups 2 and 3) were quite similar, therefore are summarized in the same table (Table 1). At stage C severe bradycardia, arterial hypotension, hypoxaemia and combined acidosis developed. At stage R moderate tachycardia, arterial hypotension and metabolic acidosis were observed. Values of blood gases returned almost completely to data measured at stage B. At stage C, venal pH and PvO2 were significantly lower and venal PvCO2 was significantly higher than the arterial ones. BBB permeability changes. In Group 1, both fluorescein dyes remained inside the vessels during the observation (4 h) in each animal. In Group 2, BBB proved to be completely tight to the tracer at stage B (Fig. 1A). At stage C spotty fluorescein leakage occurred around the smaller venules. At stage R, the permeability tracer escaped from the pial circulation and extreme strong fluorescence appeared on the whole
166
Fig. 1. Fluorescent photodocumentation. A and B: Fluorescence microphotographs of the cerebral pialarachnoideal vessels (one field) of a newborn piglet with bilateral experimental pneumothorax (BEP) (blood brain barrier marker: Na ~-fluorescein, MW 376). A: At baseline stage, before the induction of BEP. The indicator remained confined to the intravascular space. B: At recovery stage, extreme brain surlace fluorescence. C: Fluorescence microphotograph of the cerebral pial-arachnoideal vessels of a newborn piglet with BEP (blood brain barrier marker: FITC-dexran, Mw 40000). At recovery stage spotty fluorescein extravasation starts around a small venula (arrow). Bar - 200/ml.
brain surface in all cases (Fig. I B). In Group 3, BBB was also completely tight to the tracer at stage B in all animals. Likewise, no extravasation was seen at stage C, despite the development of severe microvascular alterations, such as vasoconstriction and venous stasis. However, at stage R, spotty fluorescence leakage occurred (number of leaky sites = 2.3 +_0.4/0.10 cm 2, ~'_+ SE) around the smallest veins, which showed progressive sludging (Fig. I C). C S F H X content was significantly higher in animals in stage C (26.39_+ 3.2/tM/1, n = 8 ) than in Group 1 (8.4+ !.7 ~tM/1, X _ S E ; n = 8 , P<0.05). At stage R it decreased (17.7___2.1 /tM/l, X'-+_SE, n = 8 , P<0.05 vs. stage C), but remained elevated compared to Group 1 data (P < 0.05) (unpaired t-test). This is the first study to describe changes in the permeability of pial-arachnoideal microvessels during the early stages of severe cardiovascular and metabolic shock in the neonatal period. As a major finding we showed that progressive BEP resulted
167
in a marked growth of BBB permeability in a size-dependent manner. Simultaneously CSF HX measurements revealed disturbed brain purine metabolism. Our neonatal BEP model has been extensively used and led to a better understanding of brain microcirculation [4, 13, 14]. The open cranial window technique used here was also frequency applied in other animals to study in situ certain features of brain microvessels which have in many respects similar properties to the intraparenchymal ones [4, 10, 16]. In agreement with our previous results, we found that at the C stage BBB had lost its ability to retain Na+-fluorescein within the circulation [14]. From our findings presented here it became clear that the processes leading to enhanced leakage were going on during stage R, resulting in a more intensive parenchymal Na+-fluorescein leakage. Moreover, while FITC-dextran was retained in stage C, it passed BBB and was detected as fluorescent spots at stage R, compromising further the brain parenchyma. These findings are in agreement with our previous result obtained on Evan's blue dye penetration [13]. Changes of the vascular permeability pattern could come from the activation of the same molecular mechanism, which has been hypothesized earlier as regards the pathogenesis of brain edemas [4]. Permeability alterations observed on CSF-covered brain surface could derive, at least in part, from the byproducts (mainly free radicals) of severely disturbed brain purine metabolism caused by oxygen deprivation [12], liberated through posthypoxic reoxygenation between stages C and R [1, 11, 12]. Cerebral microvessels possess enzyme xanthine oxydase [5, 6], which mediates the key step of free radical formation and also results in a gross opening in small pial venules after its topical application on the brain surface [15]. Similarly, capillary endothelial damage with enhanced permeability was achieved intracerebrally when xanthine oxydase together with HX [3] was being given. Moreover, its inhibition resulted in the preservation of BBB for fluorescein-labelled sodium and reduced the secondary infarct size evoked by middle cerebral artery occlusion [6]. The increased intrapleural pressure can also be deleterious on BBB through the elevated intracranial pressure [8] with HX accumulation in CSF [12]. All factors mentioned could open BBB via intercellular or transcellular routes [4, 16]. Moreoever, according to data observed in a similar shock model [7], brain acid-base balance and oxygenization might have been worse during BEP than shown by our arterial blood values. We tend to accept this assumption in the light of our similar observations analysing parallel arterial and venal blood samples. Otherwise, the necessary alkali-glucose infusion at stage C could have serious side-effects on the brain at cellular level. Further studies are warranted to elucidate the molecular mechanism underlying the observed differential leakage at BBB and to ascertain the detailed neuropathological changes in different brain regions following acute phenumothorax. This work was supported by the Alexander von Humboldt Foundation (Bonn, F.R.G.) 1 Armstead, W.M., Mirro, R., Busija, D.W. and Letller, C.W., Postischemic generation of superoxide anion by newborn pig brain, Am. J. Physiol. 255 (Heart Circ. Physiol.), 24 (1988) H401-H403.
164 2 Bouliev, R., Bory, C. and Baltassat, P., High performance liquid chromatographic determination of hypoxanthine in biological fluids, J. Chromatogr., 233 (1982) 131 140. 3 ('han, P.H.. Schmidley, J.W., Fishman, R.A. and Longar, S.M., Brain injury, edema, and vascular permeability changes induced by oxygen-derived free radicals, Neurology, 34 (1984) 315 320. 4 Joo, F. and Klatzo, I., Role ofcerebral endothelium in brain oedema, Neurol. Res., 11 (1989) 67 75. 5 Kjellmar, 1., Andin6, P., Hagberg, H. and Thiringer, K., Extracellular increase of hypoxanthine and xanthine in the cortex and basal ganglia of fetal lambs during hypoxia ischemia, Brain Res., 478 (1989) 241 247. 6 Martz, D., Rayos, G., Schielke, G.P. and Betz, A.L., Allopurinol and dimethylthiourea reduce brain infarction following middle cerebral artery occlusion in rats, Stroke, 20 (1989) 488~,94. 7 Mathias, D.W., Clifford, P.S. and Klopfenstein, H.S., Mixed venous blood gases are superior to arterial blood gases in assessing acid base status and oxygenation during acute cardiac tamponade in dogs, J. Clin. Invest., 82 (1988) 833 838. 8 Mayhan, W.G., Faraci, F.M. and Heistad, D.D., Disruption of the blood-brain barrier in cerebrum and brain stem during acute hypertension, Am. J. Physiol. 251 (Heart Circ. Physiol.), 20 (1986) HII7I Hl175. 9 Nars, P.W., Schubarth, k., Kindler, R., Werthemann, U. and Stalder, G., Systematic tbllow-up of newborns with idiopathic respiratory distress syndrome, Helv. Paediat. Acta, 36 (1981) 389 404. 10 Olesen, S.P., Leakiness of rat brain microvessels to fluorescent probes following craniotomy, Acta Physiol. Scand., 130 (1987) 63 68. 1l Part, A., Harken, A.H., Burton, L.K., Rodell, T.C., Piermattei, D., Schorr, W.J., Parker, N.M, Berger, E.M., Horesh, I.R.. Terada, L.S., Linas, S.L., Cherois, J.C. and Repine, J.E., Xanthine oxidase-derived hydrogen peroxide contributes to ischemia reperfusion-induced edema in gerbil brains, J. Clin. lnvest., 81 (1988) 1556 1562. 12 Saugstad, O.D., Hypoxanthine as an indicator of hypoxia: its role in health and disease through free radical production, Pediatr. Res., 23 (1988) 143 150. 13 Tcmesvfiri, P., Jo6, F., Koltai, M., Eck, E., Ad/tm, G., Sikl6s, L. and Boda, D., Cerebroprotective cffect of dexamethasone by increasing the tolerance to hypoxia and preventing brain oedema in newborn piglets with experimental pneumothorax, Neurosci. Lett., 49 (1984) 87 92. 14 Temesvfiri, P. and Kovacs, J., Selective opening of the blood brain barrier in newborn piglets with experimental pneumothorax, Neurosci. Lett., 93 (1988) 38 43. 15 Underberg, A., Wahl, M. and Baethmann, A., Arachidonic acid induces opening of the blood-brain barrier. In Y. Ianaba, I. Klatzo and M. Spatz (Eds.), Brain Edema, Springer-Verlag, Berlin, 1985, pp. 159 164. 16 Wahl, M., Unterberg, A. and Baethmann, A., lntravital fluorescence microscopy for the study of blood brain barrier function, Int. J. Microcirc. Clin. Exp., 4 (1985) 3 18.
163
Elsevier Scientific Publishers Ireland Ltd.
NSL 06885
Disturbed brain purine metabolism results in a gross opening of the blood-brain barrier in newborn piglets following experimental pneumothorax P6ter Temesvfiri I, Csongor/kbrah~m l, Ferenc Jo6 2, J6zsef Kov~.cs 1, Zsuzsa Baranyai 3 and Katalin R~,cz 1 1Department of Pediatrics, Szent-Gy6rgyi Albert University Medical School, Szeged (Hungary), 2Laboratory of Molecular Neurobiology, Institute of Biophysics, Biological Research Center, Szeged (Hungary) and SDepartment of Pediatrics, University Medical School, Pbes (Hungary) (Received 4 December 1989; Revised version received 5 February 1990; Accepted 5 February 1990)
Key words: Piglet; Pneumothorax; Brain microvessel; Intravital fluorescence microscopy; CSF hypoxanthine Changes in the permeability of pial-arachnoideal microvessels [30-210/~m), of the blood-brain barrier (BBB), were intravitally studied by fluorescent microscopy and compared to the hypoxanthine (HX) level of cerebrospinal fluid (CSF) in newborn piglets (n = 24) using the open cranial window technique. Eight animals served as controls (Group 1), the others were studied in the course of bilateral experimental pneumothorax (BEP) using low (Na+-fluorescein, MW 376, Group 2) and large molecular weight (FITC dextran, MW 40000, Group 3) fluorescent tracer molecules. Cisternal CSF was sampled from the animals: 8 piglets from Group 1, and 4-4 piglets from Groups 2 and 3 at different stages of pathological condition: (i) at the critical (C) stage (severe acidosis, bradycardia, arterial hypotension and hypoxaemia) and also (ii) at the recovery (R) stage (mild metabolic acidosis, tachycardia, arterial hypotension) and the HX concentration was determined with high-pressure liquid chromatography. In Group I neither low (n = 4) nor large (n=4) molecular weight tracers penetrated BBB. In Group 2, however, the fluorescein dye passed BBB as a spotty leakage in animals at C stage (n =8). Diffuse fluorescein penetration was seen at R stage, too (n=4). In Group 3 no change in permeability was found at C stages (n=8), but at R stage (n=4), 2 h after the primary hypoxic insult, when the animals had recovered from cardiovascular and metabolic shock, the tracer passed locally the microvascular wall and appeared as leaky spots (number of leaky sites = 2.3 +0.4/0.10 cm 2, X + SE). The HX content in the CSF was significantly elevated ( P < 0.05) at stage C (26.39+3.2/zM/1, X,+SE), compared to data measured in Group 1 (8.4+ 1.7/LM/I, ~'__+SE). At stage R, it proved to be significantly lower (P<0.05) than in stage C, but still remained elevated (17.7+2.1 /tM/1, ~'+ SE; P < 0.05 vs. Group 1). Our results draw the attention of the size-dependent gross opening of the BBB in different stages of BEP and to the possible importance in the pathogenesis of mediators deriving from the disturbed brain purine metabolism.
Correspondence." P. Temesv~iri, Department of Pediatrics, Szent-Gy6rgyi Albert University, Medical School, P.O. Box 471, H-6701 Szeged, Hungary. 0304-3940/90/$ 03.50 © 1990 Elsevier Scientific Publishers Ireland Ltd.
164 Neonatal pneumothorax and its consequences are a major concern for physicians. Cerebral damage was reported to be more frequent in newborn infants suffering from pneumothorax than in babies who were also under intensive care but did not present this complication [9]. In an experimental model of this disease, we observed [14] severe alterations in the function of the pial-arachnoideal small venules during the acute hypoxic state consisting of leakage of Na +-fluorescein and sludging formation. The objective of the present study was two-fold: (i) to characterize changes in permeability of the pialarachnoideal small vessels in the recovery (R) stage after brain ischemia and (ii) to measure the hypoxanthine(HX) content in the cerebrospinal fluid (CSF) at different stages of pathological condition, in order to elucidate if purine metabolites released from the brain by oxygen deprivation can lead to permanent cerebral damage [5, 12]. In the present study 24 piglets were included. They were managed, monitored and an open cranial window was performed above the right parietal cortex as reported earlier [14]. The animals were divided into 3 experimental groups. Group I piglets. Eight animals without experimental bilateral pneumothorax (BEP). The piglets were ventilated and the pial-arachnoideai microvessels intermittently observed through 4 h with a W I L D Fluorescence Photomacroscope (M 400, Heerbrugg, Switzerland) after having given 2.0 ml/kg b.w. (I.0 vol,%)Na+-fluores cein (Mw 376, Aldrich Chem. Co. Ltd. Gillingham) into the umbilical vein at the beginning and 1 h thereafter in four piglets. The remaining 4 piglets were given 1.0 ml/kg b.w. (5.0 vol.%) FITC-dextran (Mw 40 000, Sigma, St. Louis) at the same time scale. Filtered light for fluorescence excitation was provided by a 50-W mercury lamp and by an I2 filter (Fa. Leitz, Wetzlar, F.R.G.). Microphotographs of the cortex were taken by a WILD MPS 55 photoautomat. CSF samples for the cisterna magna were obtained at the end of the experimental period. Group 2 piglets. Eight animals with BEP through the indwelling intrapleural drains [14]. Circulatory and respiratory failures were produced. In the same dose as in Group 1 Na + -fluorescein was given at the beginning (named basal (B)-stage) and at the critical (C)-stage. Criteria of this condition were: arterial pH < 7.0, HR < 60/ min, MABP < 30 mmHg. At this point, CSF was sampled from 4 animals and the others were resuscitated by giving an intravenous infusion of 8.0 mM/kg b.w. (8.4 vol.%) sodium bicarbonate diluted with an equal volume of glucose (10.0 vol.%) delivered over 15 min. In the meantime, intrapleural gas was gently sucked out, and these 4 animals were further ventilated without BEP. The experiments were terminated at stage R, 2 h after stage C. CSF was sampled. Group 3 piglets. Eight animals with BEP were handled in the same way as piglets in Group 2, except for the fluorescein tracer used. These piglets were given FITCdextran (MW 40 000) in the same doses as in Group I at the beginning and at stage C. As in Group 2, CSF was sampled from 4 animals at stage C, and the others were resuscitated and further ventilated till stage R, when CSF was sampled. Fluorescent photodocumentation was carried out also at B, C and R stages (diameter of vessels: 3(~210/~m) in Group 2 and 3. The CSF samples were stored at -20~'C until analyzed (Du Pont Instruments
165
TABLE I L A B O R A T O R Y D A T A O N N E W B O R N PIGLETS (Group 2 and 3) Values are ~', + SE. Stages of bilateral experimental pneumothorax B (n 16)
C (n 16)
R (n 8)
Time elapsed from B (min)
62.7 + 4.5
189.9+ 13.8
Cardiovascular parameters Mean arterial blood pressure (mmHg) 59.1 + 2.1 H R (beats/min) 162.9 _ 1.0
22.9 + 1.6 ~ 51.8 ___1.81
47.0+3.5 u t89.4+ 1.91,2
Arterial blood gases and acid-base status pH 7.41 +0.02 ( H C O 3 9 mM/l 22.8 __+0.7 PaCO2 (mmHg) 36.5 __+2.0 PaO: (mmHg) 58.0 + 2.5
6.89 + 0.031 9.3 __+0.51 70.6 + 5.51 29.0 + 2.51
7.22+0.031.2 15.6 ___0.8 u 40.9 ___3.02 54.0 ___3.32
Venous bhmd gases and acid-base status (superior vena cava) pH nm 6.81 +0.023 (HCO3-) mM/1 nm 9.2+0.4 PvCO2 (mmHg) nm 84.1 + 2.83 PvO2 (mmHg) nm 17.4+ 1.2~ B = Baseline, C = critical, R Lsignificantly differs (P < 0.05) 2significantly differs (P < 0.05) %ignificantly differs (P < 0.05)
nm nm nm nm
= recovery; n = number of animals studied, nm = not measured. from values measured at B stage. from values measured at C stage. from values measured from arterial samples (Student's unpaired t-test).
HPLC) by high-pressure liquid chromatography [2]. Cardiovascular, blood gas and acid-base parameters. In Group 1, the values remained within the physiological range. Data in both BEP Groups (Groups 2 and 3) were quite similar, therefore are summarized in the same table (Table 1). At stage C severe bradycardia, arterial hypotension, hypoxaemia and combined acidosis developed. At stage R moderate tachycardia, arterial hypotension and metabolic acidosis were observed. Values of blood gases returned almost completely to data measured at stage B. At stage C, venal pH and PvO2 were significantly lower and venal PvCO2 was significantly higher than the arterial ones. BBB permeability changes. In Group 1, both fluorescein dyes remained inside the vessels during the observation (4 h) in each animal. In Group 2, BBB proved to be completely tight to the tracer at stage B (Fig. 1A). At stage C spotty fluorescein leakage occurred around the smaller venules. At stage R, the permeability tracer escaped from the pial circulation and extreme strong fluorescence appeared on the whole
166
Fig. 1. Fluorescent photodocumentation. A and B: Fluorescence microphotographs of the cerebral pialarachnoideal vessels (one field) of a newborn piglet with bilateral experimental pneumothorax (BEP) (blood brain barrier marker: Na ~-fluorescein, MW 376). A: At baseline stage, before the induction of BEP. The indicator remained confined to the intravascular space. B: At recovery stage, extreme brain surlace fluorescence. C: Fluorescence microphotograph of the cerebral pial-arachnoideal vessels of a newborn piglet with BEP (blood brain barrier marker: FITC-dexran, Mw 40000). At recovery stage spotty fluorescein extravasation starts around a small venula (arrow). Bar - 200/ml.
brain surface in all cases (Fig. I B). In Group 3, BBB was also completely tight to the tracer at stage B in all animals. Likewise, no extravasation was seen at stage C, despite the development of severe microvascular alterations, such as vasoconstriction and venous stasis. However, at stage R, spotty fluorescence leakage occurred (number of leaky sites = 2.3 +_0.4/0.10 cm 2, ~'_+ SE) around the smallest veins, which showed progressive sludging (Fig. I C). C S F H X content was significantly higher in animals in stage C (26.39_+ 3.2/tM/1, n = 8 ) than in Group 1 (8.4+ !.7 ~tM/1, X _ S E ; n = 8 , P<0.05). At stage R it decreased (17.7___2.1 /tM/l, X'-+_SE, n = 8 , P<0.05 vs. stage C), but remained elevated compared to Group 1 data (P < 0.05) (unpaired t-test). This is the first study to describe changes in the permeability of pial-arachnoideal microvessels during the early stages of severe cardiovascular and metabolic shock in the neonatal period. As a major finding we showed that progressive BEP resulted
167
in a marked growth of BBB permeability in a size-dependent manner. Simultaneously CSF HX measurements revealed disturbed brain purine metabolism. Our neonatal BEP model has been extensively used and led to a better understanding of brain microcirculation [4, 13, 14]. The open cranial window technique used here was also frequency applied in other animals to study in situ certain features of brain microvessels which have in many respects similar properties to the intraparenchymal ones [4, 10, 16]. In agreement with our previous results, we found that at the C stage BBB had lost its ability to retain Na+-fluorescein within the circulation [14]. From our findings presented here it became clear that the processes leading to enhanced leakage were going on during stage R, resulting in a more intensive parenchymal Na+-fluorescein leakage. Moreover, while FITC-dextran was retained in stage C, it passed BBB and was detected as fluorescent spots at stage R, compromising further the brain parenchyma. These findings are in agreement with our previous result obtained on Evan's blue dye penetration [13]. Changes of the vascular permeability pattern could come from the activation of the same molecular mechanism, which has been hypothesized earlier as regards the pathogenesis of brain edemas [4]. Permeability alterations observed on CSF-covered brain surface could derive, at least in part, from the byproducts (mainly free radicals) of severely disturbed brain purine metabolism caused by oxygen deprivation [12], liberated through posthypoxic reoxygenation between stages C and R [1, 11, 12]. Cerebral microvessels possess enzyme xanthine oxydase [5, 6], which mediates the key step of free radical formation and also results in a gross opening in small pial venules after its topical application on the brain surface [15]. Similarly, capillary endothelial damage with enhanced permeability was achieved intracerebrally when xanthine oxydase together with HX [3] was being given. Moreover, its inhibition resulted in the preservation of BBB for fluorescein-labelled sodium and reduced the secondary infarct size evoked by middle cerebral artery occlusion [6]. The increased intrapleural pressure can also be deleterious on BBB through the elevated intracranial pressure [8] with HX accumulation in CSF [12]. All factors mentioned could open BBB via intercellular or transcellular routes [4, 16]. Moreoever, according to data observed in a similar shock model [7], brain acid-base balance and oxygenization might have been worse during BEP than shown by our arterial blood values. We tend to accept this assumption in the light of our similar observations analysing parallel arterial and venal blood samples. Otherwise, the necessary alkali-glucose infusion at stage C could have serious side-effects on the brain at cellular level. Further studies are warranted to elucidate the molecular mechanism underlying the observed differential leakage at BBB and to ascertain the detailed neuropathological changes in different brain regions following acute phenumothorax. This work was supported by the Alexander von Humboldt Foundation (Bonn, F.R.G.) 1 Armstead, W.M., Mirro, R., Busija, D.W. and Letller, C.W., Postischemic generation of superoxide anion by newborn pig brain, Am. J. Physiol. 255 (Heart Circ. Physiol.), 24 (1988) H401-H403.
164 2 Bouliev, R., Bory, C. and Baltassat, P., High performance liquid chromatographic determination of hypoxanthine in biological fluids, J. Chromatogr., 233 (1982) 131 140. 3 ('han, P.H.. Schmidley, J.W., Fishman, R.A. and Longar, S.M., Brain injury, edema, and vascular permeability changes induced by oxygen-derived free radicals, Neurology, 34 (1984) 315 320. 4 Joo, F. and Klatzo, I., Role ofcerebral endothelium in brain oedema, Neurol. Res., 11 (1989) 67 75. 5 Kjellmar, 1., Andin6, P., Hagberg, H. and Thiringer, K., Extracellular increase of hypoxanthine and xanthine in the cortex and basal ganglia of fetal lambs during hypoxia ischemia, Brain Res., 478 (1989) 241 247. 6 Martz, D., Rayos, G., Schielke, G.P. and Betz, A.L., Allopurinol and dimethylthiourea reduce brain infarction following middle cerebral artery occlusion in rats, Stroke, 20 (1989) 488~,94. 7 Mathias, D.W., Clifford, P.S. and Klopfenstein, H.S., Mixed venous blood gases are superior to arterial blood gases in assessing acid base status and oxygenation during acute cardiac tamponade in dogs, J. Clin. Invest., 82 (1988) 833 838. 8 Mayhan, W.G., Faraci, F.M. and Heistad, D.D., Disruption of the blood-brain barrier in cerebrum and brain stem during acute hypertension, Am. J. Physiol. 251 (Heart Circ. Physiol.), 20 (1986) HII7I Hl175. 9 Nars, P.W., Schubarth, k., Kindler, R., Werthemann, U. and Stalder, G., Systematic tbllow-up of newborns with idiopathic respiratory distress syndrome, Helv. Paediat. Acta, 36 (1981) 389 404. 10 Olesen, S.P., Leakiness of rat brain microvessels to fluorescent probes following craniotomy, Acta Physiol. Scand., 130 (1987) 63 68. 1l Part, A., Harken, A.H., Burton, L.K., Rodell, T.C., Piermattei, D., Schorr, W.J., Parker, N.M, Berger, E.M., Horesh, I.R.. Terada, L.S., Linas, S.L., Cherois, J.C. and Repine, J.E., Xanthine oxidase-derived hydrogen peroxide contributes to ischemia reperfusion-induced edema in gerbil brains, J. Clin. lnvest., 81 (1988) 1556 1562. 12 Saugstad, O.D., Hypoxanthine as an indicator of hypoxia: its role in health and disease through free radical production, Pediatr. Res., 23 (1988) 143 150. 13 Tcmesvfiri, P., Jo6, F., Koltai, M., Eck, E., Ad/tm, G., Sikl6s, L. and Boda, D., Cerebroprotective cffect of dexamethasone by increasing the tolerance to hypoxia and preventing brain oedema in newborn piglets with experimental pneumothorax, Neurosci. Lett., 49 (1984) 87 92. 14 Temesvfiri, P. and Kovacs, J., Selective opening of the blood brain barrier in newborn piglets with experimental pneumothorax, Neurosci. Lett., 93 (1988) 38 43. 15 Underberg, A., Wahl, M. and Baethmann, A., Arachidonic acid induces opening of the blood-brain barrier. In Y. Ianaba, I. Klatzo and M. Spatz (Eds.), Brain Edema, Springer-Verlag, Berlin, 1985, pp. 159 164. 16 Wahl, M., Unterberg, A. and Baethmann, A., lntravital fluorescence microscopy for the study of blood brain barrier function, Int. J. Microcirc. Clin. Exp., 4 (1985) 3 18.