Tobacco Documents Online

w ( 'b PHY&IOLOGY 01 UIlYGfiN (2ADiCALB ~~wencen P~ye,oln~cal3oc.rry. 1986 1Pcanted an U S A 12 FILE COPY Permeability of Isolated Rat Lungs Perfused With Purine and Xanthine Oxidase Under Constant Perfusing Pressure RICHARD B. FOX. RICHARD B. PARAD, ROBERT H. DF.MLING, -- -- AND MARY JO MERRIGAN Divisions of Pulrnonary and Cell Biology, Children's Hospital; Longwood Area Trau.na Center, Brigham and Women's Hospital; and Departments of Pediatrics and Surgery, Harvard Medical Sehool, Boston, Massachusetts Methods Isolated lung preparation Results Effect of purine and aanthine oxidase exposure on pulmonary vascular resistance Effect of purine arid xanthine oxidase exposure on pulmonary vascular fluid permeability Discussion OXIDANT-1NDUCBD LUNG INJURY can result from 02 radicals produced by 1) environmental sources, such as hyperoxia (2, 19), air pollutants, such as ozone and nitrogen oxides, and cigarette smoke (12); 2) products of inflammatory cells and inflammatory reactions (7, 13, 16, 21, 22); and 3) intracellular metabolites, especially those generated under conditions of ischemia (6, 10). Acute oxidant lung injury, as with most models of acute lung injury, is manifested primarily by edema, which results from an excess of fluid filtration over fluid resorption. The physiological basis of fluid filtration is defined in the Starling equation (5) T J,. = KrJP< - Pt - ~ a, (WC, - tr,.)) (I ) ,at where J„ is the rate of fluid filtration, Kr is the capillary filtration coefficient, P, is the hydrostatic pressure within the fluid-exchanging vessels, P, is the hydrostatic pressure within the interstitial tissues surrounding the fluid ex- changing vessels, a, is the oamotic reflection coefficient of the ith plasma protein (in a group of n total plasma proteins) for the fluid-exchanging vessels under consideration, Rc, is the colloid oncotic press uC vl' the ith plasma protein, and r„ is the colloid oncotic pressure of the ith t.:terstitial protein. This equation demonstrates that fluid filtration depends on two factors. The first important factor in the equation is the Kf, or hydraulic co-lductivity, term_ The Kt is a measure of vascular permeability to the filtering solvent and is an excellent indicator of capillary endothelial permeability in zone III lungs. 163 t PU6L ICRTIONS 012677

164 PHYSIOLOGY OF OXYGE,1t RADICAl.S The second, or hydrostatic, term is within the brackets to tht right of the Kr term in Equation 1. It contains several factors that together define the net hydrostatic force available to move water across the Auid-exchanging vessel. Oxidants could potentially mediate lung edema by increasing either of these important factors in the Starling equation. Oxygen radicals are key intermediates in the production of vasoactive aracbidonic acid met.abolites such as prostaglandins, thromboxanes, and leukotrienes (4), which are known to increasv'e hydrostatic phessures. Furthermore, O2 radicals can also damage cells, such as endothelium (14, 15), and interstitial components, such as glycosaminoglycans (3, 8), which are import.ant components of the barrier to fluid permeability and each tissue factor i n Equ.ati.on l. Tate et aJ. (17, 18) showed that isolated, ventilated rabbit lungs perfused with an enzymatic genersator of 02 radicals (purine and xanthine oaidase) demonst.rated evidence of vasoconstriction due to cyclooxygenase metabolites and evidence of increased fluid permeability. In those studies, eonduct,ed with a constant- flow perfusion system, pulmonary vasoconstriction increased pul- monary pressures sufficiently to possibly damage the vasoulature on a purely mechanica] basis- To determine whether 02 radicals can increase permeability in the •absence of increased pressure, we challenged isolated rat lungs with purine and xanthine oxida'se under constant pressure conditions, determining Kr. Pc, and pre- and postcapillary pulmonary vascular reaistances. We found that purine and xanthine oxidase perfusion under constantpressure conditions caused modest increases in both pre- and postcapillary resistances and very substantial increases in solvent permeability. MaJe Sprague-Dawley rats were anesthetized with 75 mg pentobarbital. When thevr were satisfactorily anesthetized but still spontaneously respiring, a tracheostorny was performed an d a plastic catheter was inserted into the trachea, tied in place, and connected to a rodent ventilator. Then the rats were anticoagulated with 100 U heparin injected by femoral cutdown with meticulous care to ensure that no air emboli were injecteci The femoral cutdown approach permitted visualization of the small air emboli that mu- tinely lurked in the very tip of the injection needle and were not appreciated when injecting into the vena cava. The abdomens were then opened a.nd the rats easa.ngumated via the inferior vena cava. The chests were opened from a subdiapbra.'gmatic approach and the pulmonary arteries eannulated witb small- bore intravenous tubing via a right ventriculotorny, again with meticulous care to avoid introduction of air emboli, and secured by a ligature around the pulmonary artery. Then the left ventricles were opened inferiorly and the mitral valves prolapsed and ruptured by retrograde insertion of a blunt 7 Pt in wi th of co Pr sti co• Wa pr< vej coi pre rec ent zer, con coll was fror at a prez that at 3 puto affei left divic the I RESt Effec Pu1rn ll const in pe) with t did a: pre8at resist.; PUBLICATIONS 012678

PtRINE .>ND NANTHINE OXIDASE PERFUSION 165 instrument (clarnpl. This permitted retrograde cannulation of the left atria with a flanged polyethylene catheter secured by a ligaturk around the heart at the le.rel of the mitral valve. After placement of the atrial catheter. perfusion of the lungs was commenced with a solution of Hank's balanced salt solution containing -1`°s bovine serum albumin with a constant pulmonary artery pressure. Purine and xanthine oxidase were added to the perfusate in some studies. ..ith xanthine oxidase added after initial hydrostatic and filtration coefficient determinations were complete. Constant pulmonary artery pressure was achie.e+9 by an overflow chamber maintained at a height sufficient to produce a recorded pressure of --12 crnHzO. After completion of the dis.tection, ventilation of the lungs was discontinued and the tracbeal cathe;er was connected (without allowing the lungs to deflate) to a constant positive pressure of-1 cmH2O. Then the lungs were suspended from a strain gauge to record peight. Pressure transducers were hydraulically connected to the affer- ent and efferent vascular catheters near their insertions into the lungs and zeroed to the level of the le)'t atriAtm. Lung weights and pressures were continuousl.y recorded on a six-channel recorder. Flow rates were measused by collecting efferent fluid in graduated cylinders. The flow of efferent perfusate was arranged so that it could be channeled by means of a stopcock to emerge from the tubing either at a point 6-10 cm below the left atrium or alternatively at a point high enough above the left atrium to give a recorded left atrial pressure of 10 cmH2O. The efferent perfusate then dropped into a reservoir that was maintained at 37°C. The temperature of the lungs was maintained at 37'C by warming the perfusate to 37°C just prior to its entry into the pulmonarv artery. The P,. was determined by simultaneous occlusion of the afferent and efferent catheters (11). The Kr was determined by increasing the left atrial pressure to 10 cmH2O for 3 min and then returning it to zero, dividing the net increase in weight by the net difference in P, due to i.ncreasing . the left atrial pressure (5). The Kr has units of mg-min''-cmH,O'" -lung''- Effect of Purine nnd Xanthine Oxidase Exposure on Pulmono n Vascular Resistance Perfusion of isolated rat lungs with purine and xanthine osio's.,e under constant pulmonary artery pressure conditions caused a significant decrease in perfusate flow (41 %) 45 min after xanthine oxidase addition as compared with the base-line flow rate (Table 1). Effective pulmonary capillary pressure did not change significantly. The reduction in flow in the face of unchanging pressures at all levels resulted from a 66:"o increase in precapillarv •-ascular resistance and a 141% increase in poatcapillary resistance. PUBL ICPTIONS 012679

166 PHYSIOLOGY OF OXYGEN RADICAIS TABLE 1. , E/Jects of purine and xanthine oxidase cluzelenge on t-ascular /luid dynamics of isolated rat lungs 40 Treetmens Parameter ------ - -- Untreated - Parine - - - Punna and saoth.ne osidase Flow rate. ml/min BaseLne 13.8 ± 0.1 12.9=0.3 12.0=0.4 4,imtn 14.5=0.-t 12.7±0.3 7.1±0.9 Pulmonary- anery pressore, cmHrO Baselene 12.3t0.1 12.3t0.2 12.6=0.3 45 mtn 11.8 3 0.3 1'2.0 # 0-3 12.6 :t0.2 Capi)larn pressure, cmHxO Ba_neLne 4.7±0.6 4.9±0.4 4.0=0.3 45men 3.8±0.3 5-3±0-4 4.6~0.4 Left atnum pressure. cmHsO Baseline 0a0 0t0 0:t 0 45mm 0s0 0x0 0 = 0 Precaptllarj. res.stance- cim HtO m1-' -min-' Base Line 0.55 = 0.03 0.58 ± 0.03 0.71 ± 0.02 45 min 0-55 :t 0.03 0.52 :t 0.04' 1.08 :t 0.11 Postcapt111in resistance. rmH aO ml-' • mm-' Base tGne 0.34 t 0.04 0.38 :t 003 0.34 = 0.03 a3men 0.26t0.02 0.41=0.04 0.82:t 0.16 _ Values are means 3 SE for 4 oDservattons. 'Values based on 1 observation. TABLE 2. Effects of purine and santhin.e oxidase challenge on fluid perinetzbility oJ isolated rat lungs _ Punne and zamhine ondase Basel,ne 3.330.6(4) 1.930.3(4) 38= 0.7 (3) 45 min 4.7 s 1.3 (4) 4.7 = 0_2 (4) 121 t 48 (3) e Paratneter - - - _ . Untreated Purine Permeability filtration coefficient' ~ % ah~es are means x SE for number of observations in parentbeses. ' l: mta arn mg- min" cmHtO-' lung''; to convert to ml -mm-' cmHrO-' -ltmg-' multiply by 50 x 10"'. Effect c;- Parine and Xanthin.e Oxidase Exposure on Pulmonar.v Vascular Fluid Perm.eability In contrast to its significant but modest effects on pulmonary vascular resistance, perfusion with purine and xanthine oxidase caused a very larg3 (32-fold) increase in Kt as compared with the base-line permeability (Table 2). This cbange occurred in the absence of any sigriii'cant mechamica] stresses such as zncreased pressures in either the vascular or airway compartments. DISCCSSIO?7 Present studies show that perfusion of isolated rat lungs under constant pressure conditions with an enzymatic generator oC 02 radicals, the purine and PuBL rcArroNS ? J 012680

PURINE AND XANTHINE OXIDASE PERFUSION 167 xanthine oxidase system, causes moderate increases in both pre- and postcap- illarv resistances and a more marked increase in Kt. These findings are in substantial agreement with those of Tate et all. (17, 18), who used a constant-flow circuit and reported pressure increase$ that could have caused mechanical lung injury. Also we used modest constant distending airway pressure instead of volvme-cycled ventilation because our early experiments showed that peak inspiratory pressures in edematous ven- tilated lungs can attain levels in excess of 50 cmHzO, causing further edema from overdistension (1). The 02 radical generator employed in these studies, purine and xanthine oxidase, produces the superoxide anion radical (02•) as its prir+nary product (9). The O?• is a relatively weak oxidant, which then undergoes a complex of reactions with itself to yield 171202 and the hydroxyl radical (HO•). necent studies have suggested that HO• is the species primarily responsible for actual lung damage (2, 6, 7). The isolated perfused lung is an especially powerful model for the study of oxidant lung injury because I) the fluid perfusing the lungs can be consti- tuted to include or exclude any and all blood components or other special enz.•mes or factors to be studied; 2) perfusing pressures can be controlled and measured at the pulmonary artery, left atrial, and capillary levels, and flow rates can be measured to yield direct determinations of pre- and postcapillary resi>tances: and 3) changes in permeability can be measured as changes in lung tkeight in response to known changes in capillary pressure. With these powerful tools and the application of the Starling equation it is now possible to determine quantitatively whether oxidant-induced lung edema is due to changes in hydrostatic forces, changes in permeability, or both. This distinc- tion is usually not possible using methods that measure only pulmonary artery pressure, lung water, or extravasation of reference molecules such as albumin, which are moderately permeable across the pulmonary endothelium with bulk fluid transport (20). %Ve thank Aubrey E. Taylor and Mary Towesley for teaching us tbe use of the isolatied lung preparatroa for the determination of Xr and P,. Thr_v work was supported by grants from the Council For Tobacco Research (1666) t+nd the Nat:onal Heart, Lung. and Blood Institute (HL-30068). Dr- Fox is a Clinical Investigator Awardee oi the Dras on of Lung Diseasea, National Heart. Lung, and Blood Institute rK08 HL-01943). Present address of R. B. Fox: Pulmonary Critical Care Division. Children's Haspital, MLnLeapohs, MN. RFFE RE SCES 1. Ec.aN, E. A. Lung inflation, lung solute permeability, and alveolar edema. J Appl Physwt 63 121-125, 1982. 2 Fox. R. B. Prevention of granulocyte•mediat.ed oxidant lung injury in rats by a bydroxyl radical scavenger, dimethylthiourea J. C4n Incest 74: 1456-1464. 198.4. 3. Fox. R. B.. AND W. K. Fox. Dimetbyl sulfoznde preventa hydroxyl tadical-medrated depolym• ertzation of hyaluronie acid. Ann. NY Acad. Sri. 411:14-18, ]983. 4. FR1DOVICH, S. E.. AND N. A. PosTert. Oxidation of arachidonie acid in micellee by superozide and hydrogen peroxide. J Biot Chem 25fi= 260- 265, t981. PUBLIL"4 BTIONS 012681

168 PHYSIOLOGY OF OXYGEN RADICAIS 5. GAAR. K. A., Jlt, A. E- TAVi-ort, L J. OweNS, AND A- C. GttYA'ox- Pulmanary capillary pressure and filtrauon coefficient in the isolated perfused lung. Am. J PhysioL 213: 910- 914, 1967. 1 6. GRI.NGER. D. N.,,G. RLTT1Lt, AND J. M. McCoRD. Superoxide rasdicala in feiine inteatinal isc6emia. ~Gastroenterology 8l: 22-29, 1981. 7. JoHNSON. K. .1., J; C. FAtvTONE III, J. KAPLAN, AND P. A- WARD. Ih vivo damage of rat lungs by oxygen metsGohtea. J. Chn. lntre.+t. 67: 983-993. 1981. 8. N1cCoR.D, J. M_ Ftea radicals and inflammation: protection of aynovtal fhiid by supero:ide dismutase' Sci;en,ce Wash. DC 185: 529-531, 1974. 9. %iCCORfl, J. M.. ANb I. F1uDDY1CH. Superoxide dismut.aae. J. BwL Cher-a 244: 6049-fA55- 1969. 10. Pa LsR, M. S.. J. HbtoAL. AND R F. Fenfus. Ozygen free radicals in ischernic acute renal failure in thp rat J Clin_ /nuest. 74: 1156-1164, 1984. 11 P+RxER. J. C., P. ft KvtF-rvs, K_ P. RvAN, AND A. E. TAYLOR. Comparison of isogravimetric and venoua occlusion capiUary pressures :o isolated dog lungs. J AppL PhysioL 55: 964- 968.1983. 12. PRVOR, W. A_. B. J. HALES. P. I. PREMOVIC, A>iD D. F. CHURCH. Tbe radicals in cigarette smoke. tbeir nature and suggested physiologic implications. Science Wash. DC 220: 425- 42 . 1983. 13. RsvtNE, J. E., J. W. EATON, M. W. ANDERS, J. R. HOIDAL. AND R. B. Fox. Generation of h.dso:yl r¢diciils by enzymea, chemicals and human phagocytee in vitro. Detection using tbe anti-inflnmmat.ory agent, dimetbyl sulfo=ide. J C4n, lnueat: 64: 16Q2- 1651, 1979. 14. SAcx3. T.. C. F: 1pIoLDOw, P_ R CRADDOCK, T_ K. BowERS, wniD H. S. JACOe. Oxygen radicals mkdiate endothehal cell damage by colmplement-atimulated granulncytrs. An in vitro model of immune vascular damage. J. Ctin. lnuest. 61: 1161-1167, 19'8. 15. Sxwssv. D. M.; t. S. Stfwsnv, AND M_ J. PEActe. Granulocytea and phorbol myristate acetate increase peYmeaability to albumin of cultured endothelial monoleyers and isolated perfueed lungs. Am, Rev; Respir. Du 12: 72-76. 1983. 16. SRAsev, D_ M., K. M. VANeENt'HUSEN, IL M. TA're, S. S. SHASITY, I. Mcb1uR'rttv, AND J. E_ RBPiNE. GTantalocyte9 mediate acute edematous lung injury in rabbits and in isolated rabbit lungb perfused with phorbot myriatate acetate: role of oxygen radicale. Ant. Rev Respis. Dia. 12& 443-447, 1982. 17. TATw- R M_, H- G~.'. MoRRts, W. R. SCHROEDER,,AND J. E. REPINE. Oxy-geo metabolites atamulate thnomboxane production and vaaocotnltrictioo in isolated saline-perfiteed rabbit lumgs- J. Clin'. lnuest. 74: 60&-613, 1984. 18. TA98, R. M.. K. M. VANBENTHU9EN, D. M. SHwsev, L F_ McMuRt7er, AxD J. E. REptNE. Otygen.radieal-mediated permeability edema and vaaoconatriction in iaolated perfuaed rabbit lungs. Ar~ Reu Respir. Dis. 126: 802,80f, 1982. 9 19. 'h.'RREWs, J. F., J.,D. CRAPO, AND B. A. FREEBI.aN_ Protection against oxygen toxicity by intravenous injQction of lipoaome-entrapped catelaae and superozide dismtttase. J. Clin I ru.est 73: 87-95, 1984. 20. RA.vcENS'rEEta, b.; Ii~ YANROVtcH, J- HomAL, A1vD D. NtEwoEHNES. Bleomycin-induced changes in piulYnonary microvascular albumin permeability and eatravasculas nlbumin spsae. Ara. Reu., Resp.r_ Dis 127: 204-208, 1983- 21. WwxD, P. A., G. Q. "1`a.L, J. R. HATHER1Lt., T. M. ANNeSI-Ett. AND R G. -%Ct,-NR81-. Systemic complement acuvauon, lung inlury, and products of lipid peroaJdation. J. Clin Intxst :6: 517--527, ;9Ei5. 2_^ WAaD, P. A.. G. 0.. TtLL. R. KuNxEL, ANn C. BEAUCHAa,tP. Evidence for role of bydltozyl radical in complement and neutrophil -dependent tibsue injury. J Ctin Invest 72: 789- 8A1, 1983. PuBL ICHrIONs .1