Biotransformation of glyceryl trinitrate to glyceryl dinitrate by human hemoglobin'

Depcrrtment of PRarrnae*ology arzd Toxicology, Queeli's U~ziver.sity. King.~torz, On?., C'annek~ K7L 3NB

Received April 18, I084

BENNETT, B. M . , K . NAKATSL', J . F. BKIEN, and @. S. MARKS. 1984. Bic>transformation of glyceryl trinitrate to glyceryl dinitrate by human heanoglobin. &'an. J . Pkysiol. Pharmacol. 62: 704-706.

The elimination of glyceryl trinitrate (GTN) by man is rapid and its clearance exceedr cardiac output. It is therefore clear that a variety of tissues ~ I I addition to Biver are involved in the biotransformation of GTN. Ii~cubation of G'TN uith the 25 000 x ,q supernatant fraction of lyscd human erythrocytes resulted in a 39.6% -+ 5.5 (SD) elim~inatioa~ of GTN after 40 nun. Atter pretreatment of the lysate supernatant fraction with carbon anonnxidc, G7.N elimination was only 26'k + 3.5. 'I'hese data indicated that hemoglobin might be invoived in Gl'N elimination. When purlfied hemoglobin was inc~tbated uith GTN, a 77.1% f 6.4 elirnination of GTN was observed, accompanied by g%yce~yI dinitrate formation. The biotransformation of GTN was inhibited by pretreatment with carbon monoxide. The results indicate that the biotransforanation of CiTN by human erythrocytes is due, at least in part. to interaction with hernoglobin.

BENWEIT, 5 . M., K. NAKATSU, J . F. BWIEW et G. S. MARKS. 1984. Bic~aransformation of gBycery% trinitrate to glyceryl dinitrate by human hemoglobin. Can. J . Physiol. Phamacol. 62: 704-706.

L'Clirnination d ~ t glyc6ryle trinitrate (GTN) par l'hornme est rapidc et son coefficient d'kpuration dipasse le dkbit cardiaque. II est donc Cvident ququne varikte de tissucs, en plus du foie, partiicipent 2 la biotransforrnation du G'TN. L'incubation du GTN avec une fraction de sumageant de 25 000 Y ,q krythrocytes kurnains lys6s resulk cn une elirnialation de 39.4% a 5.5 (kcart type) du GI'N aprks 44) mian. Aprks pretraitement dc la fraction de sumageant lys6 avec de I'oxyde cBe carbonc, B'6lirnination du GTN n9etait que de 26% k 4.5. Ccs donnkes indiqaaent que IqhCmoglobine peut participcr B l'Cli~-rlination du GTN. Lorsque de I'hk~noglobine purifikc a CtC incuhCe avec du GTN, on a observe une klirnination dc 77.196 + 6.4 du G'I'N, accompagn6e de la formation de glyceryle dinitrate. t e prktraitement avec de l'oxyde de carbone a entravk Ia biotransfor~rnatiora du GTN. Les r6suItats indiquent quc la biotransformation dal CTN par les erythrocytes hunaainf est provoquee, du mnoins en partie, par 19interaction avec 11h6mogIobine.

['I'raduit par le joa~rnal]

I~atroduction TABLE I . Elimination of GTN during incubation with Eysed erythro-

Following the cessation of the intravenous infusion sf gIyc- cyte supernatant faaction

eryl trinitrate (GTN) to man, a very short plasma half-life ( GTN concentration of 1.9 n-min was deterrlrained (Arnlstrong, ~ rms t rong et al. ( x 1w7 M ) 1980). In additicsn, the systemic clearance of GTN exceeded cardiac. output, indicating extrahepatic sites for uptake and (or) 'H'reatrnent 0 rnin 40 min %I elimination biotransformation of GTN. The systemic circulation rnaay be involved in GTW elimination since an arterial-venous gradient YontEated 1 .63+0. 15" 0.98LBB.05 39.6-+5.5

. - L Carbon monoxide- was observed during intravenous infusion to man (Armstrong treated 1.70+0.11 1.25+0.06" 26.054.5"

et al. 1982). Organic nitrates are denitrated in the liver by the glutathlone-dependent glutathione-S-transferase enzymes (EC Noah Samples were rncuhatrd u ~ t h 2 x 10 7 M (JTN at 3:O<' for the tune\ mtinc'tted

'Each \due rzpre\znts the me'in * SIP ( n = 5 ) '$1. Glutathione-S-transferase has been isolated b p 5 0 01 when c.ompnrcd nontredtrd o)nml] by Ssl(jcni3\ r-fril h,r palred ddd

frorn kidney, intestinal mucosa, and erythrocytes. The rapid elinsination of GTN from human whole blood and

provides evidence that the elimination trf GTN by erythrocytes erythrocytes ( t ,/, = 6.2 and 6.6 nnin, respectively; Armstrong, is due, at least in part, to interaction with henloglobin with Slaughter et al. 1980) was attributed to enzy~rnatic bio- resultant formation of glyceryl dianitrate. transformation of the organic nitrate to one or several of its

known metabolites. ~ h e k are. however, conflicting reports concernil~g metabolite formation following incubation of GTN with human erythrocytes. Waa et al. (1981) could not detect metabolite forrnation, whlle Nosnan and Benet (1982), using [%H]GTN, reported the formation of non no nitrate and dinitrate metabolites.

We are earn-ently investigating the meckanism(s) respc~nsible for CTN elimination during incubation with the supernatant derived from lysed human erythrocytes. This communication

'Supported by the Ontario Heart Foundation. grant nutnber 72-14. '~ecipierat of a Medical Research Council of Canada Studentship.

Author to whom correspondence should be addressed.

Rlaterials and methods Drugs U P E ~ solutio~zs

Human hemacjglohin (type IV. Sigana, St. Louis, MO). NADH, and Inethylene blue were dissolved in degassed 61.1 A4 phosphate buffer. pH 7.5. Human erythrocyte methernoglobin reciuctase wa, purchabed froan Calbiochem-Behring Corp. (La Jolla, CA) Glyceryl trinitrate (TridilwY) was diluted in 0.9% saline. All other chemicals wcre at least reagent grade and were obtained from a variety crf' sources.

Preparation qf lysed erythroc.yte S L ~ J ~ T P E ~ E ~ C I P E ~ / k z h . l i o t l

Venous blood obtained from five normal male volunteers rage range, 23-38 years) was centrifuged at 3 10 X ,y and 4°C for 15 min. After rerxaoval of the plasma and buffy coat. the cells were washed three times with 0.90h saline and then lysed by the addition of 2

Can

. J. P

hysi

ol. P

harm

acol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Uni

vers

ity o

f A

uckl

and

on 1

2/05

/14

For

pers

onal

use

onl

y.

TABLE 2. Elli~nination of G'L'N and formation of GDN during incubation with hemoglobin

G'I'N concentration GDN conceratration ( X I 0 7 ~ ) ( X 10

Treatment 0 min 40 rnin elirnination O rnin 40 min

Nontreated 2 .03 iO . lo" tB.47+-0.17 77.126.4 <0.20b 8.09L0.20 Carbon monoxide-

treated 2.04L0.11 1.9420.18' 4 .9+ 10.3' c 0 . 2 0 <0.20

No-n'ti: Hemoglobin (5 x [(:I ' 1W) was incubateti with G'PN (2 x 10 ' M , at 37°C for. the times indicated "Each value represents the mean -C SD ( n - 4). "Valurs below the iirnit of quantitative sensitivity of thc assay. : p 5 0.001 when compared with nontreated control by Student3\ ?-test for paired data.

volumes of distilled water. The lysed cells were then centrifuged at 25 000 X g and 4°C for 20 min, and the supernatant fraction was removed. The pH of supernatant fraction was 7.25-7.35 and de- creased by about 0 .1 pH units after incubation for 40 min at 37°C.

Prepnrcltion of reduc-ed he~noplobirl Hcnr~oglotain was reduced enzymatically by a modification of the

method of Rossi-FaneEli and Antonini ( 1 958). Three-millilitre aliquots of hemoglobin ( 5 x 149Y' M ) were incubated anaerobically with NADH (3 x 10 M B, nmethylene blue (2 x 10 ' ,%a). and methemo- globin reductase (0.5 &I/nnL) for I h at 37°C in a Bubnoff metabolic incubator.

Incubation co~acdifions and ~ s s c q j of' CTR: urld rnrtcibolites After a 5-min preincubation period, 100 ~ 1 , of a G'FN solution were

added to 3-mL aliquots of the lysate supernatant fraction or hemo- globin solution to g~roduce a final concentratic~n of 2 X 10 M GTN, Duplicate samples were incubated at 37°C in a Dubnoff mnaetabolic incubator for 0 or 40 min, after which, 2-mL alicluots were assayed for GTN by the gas chromatographic method described pacviousiy (Amstrong, Marks et al. 1980). Other samples were exposed to a rapid stream of carbon monoxide for I rnin prior to incubation with GTN. The hemc~globin samples were assayed for glycei-yl dinitrate (GDN) by the method described previously for isosorbade rnono- nitrates (Bennett et al. 1983:) with the exception that rn-dinitrc~benzene (6 x lo- ' IIf) was added as the internal standard for the assay. The gas chromatographic conditions used did not distinguish between glyceryl-I ,2-dinitriatc and glyceryl- l,3-dinitrate.

Results and discussion Incubation of GTN with the lysed erythrocyte supernatant

fraction resulted in a 39.6% k 5.5 elimination of GTN after 40 mln (Table 1). When GTN was incubated in 0.9% saline. adjusted to pH values between 6.5 and 9.0 with 0.1 LW phos- phate buffer, no loss of GTN was observed. Whcn the lysate supernatant fraction was pretreated with carbon naonoxide. GTN eiimination was decreased to 26.0% 2 4.5 ( p < 0.0 1). Since carbon monoxide binds to the ferrous atom of heme in hemoglobin, it appeared that the heme moiety of hemc3globin was in some way involved in tl1e elinaination of GTN. This interpretation was supported by the observation that oxidation of hemoglobin in the lysate supernatant fraction to methe- moglobin by ferricyanide resulted in inhibition of G'I'N elimination similar to that seen after carbon ~nonoxide treat- ment (data not shown). Since the elimination of GTN was only partially inhibited after carbon monoxide treatment, it would appear that at least one other mechanism is involved in the overall elimination of GTN by erythrocytes.

Further evidence for the possible invslvel-nent of hemoglobin in the elirnination of GTN was sought in experiments in which hemoglobin was used instead of the 1ysed erythrocyte super- natant fraction. Since commercially available hemoglc~bin con-

tains up to 75% methernoglobin, it was necessary ~ C B reduce the methernogobin prior tc~ incubation with GTN. This was achieved enzymatically with methemoglobin reductase. Incu- bation of GTN with 5 x 10-' 1C;I hemoglobin resulted in a 77.1 % * 6.4 elimination of GTN (Table 2), and this dim- ination was almost completely inhibited (4.9% -+ 10.3 elimination) by pretreatment of the hemoglobin samplec with carbon monoxide ( p 5 0.001). In a control experinnent. no elimination of GTN was observed when hemoglobin was omitted from the methemoglobin reductase reaction mixture.

To determine whether the elimination of GTN involved bio- transformation. the hemoglobin incubation sararples were as- sayed for glyceryl diiaitrate metabolite(s). The mean concen- tration of GDN formed after 40 min incubation with 2 X 10 ' 1W GTN was 1.09 x 10 rW * 0.20 (Table 2). Metabolite formation at 0 min in untreated samlaples and at 0 and 40 nlin in carbon monoxide-treated samples was below the lower limit of quantitative sensitivity of the assay for GDN (2 x IBt ' Ad). We. therefore. conclude that hemoglobin in the ferrous forn~ is capable of metabolizing GTN to dinitrate metabolites.

The interaction of GTN with hemoglobin may have impli- cations regarding the mechanisrna of action of GTN as a vaso- dilator. Organic nitrate-induced vasodilation is preceded by cyclic GNIP elevation in vascular smooth muscle (Diamc3nd and Blisard 1976). The cyclic GMP elevation is produced by activation of guanylate cyclase (EC 4.6.1 .%), an enzyme that has been purified to apparent homogeneity in a form that contains heme (Gerzer, Bkih~nae et al. 1981; Ignarro, Wood et al. 1982). Activation of purified guanylate cyclase by nitric oxide (NO), nitrc~prusside, and related agents requires heme (Gcrzer, Hofmann et al. 198 I ; Ignarro, Degnan et al. 1982; Craven and DcRubertis 1983), and it is thought that formation of an NO-heme complex is an obligate step in this process.

Since GTN biotransformation by hemoglobin requires that the iron in hemoglobin be in the ferrous state and since carbon monoxide inhibits GTN biotransfor~natiora, it is likely that GTN Interacts wlth the ferrous atom during biotransfcjrnaaatic3n. It is, therefore, pc~ssible that the activation of guanylate cyclasc by GTN may occur by the direct interaction of one of the nitrate groups of GTW with the iron of guanylate cyclase-bound heme, rather than by prior biotransformation of' GTN to NO and S-nitrosothiols which. in turn, activate guanylate cyclase, as has been suggested previously (Ignarro et al. 198 1).

Elucidation of the mechanism of GTN biotransformation by hemoglobin may shed light on the mechanism by which GTN induces vasodilation.

The authors wish to thank Glen Tarn and Chris Slack for

Can

. J. P

hysi

ol. P

harm

acol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Uni

vers

ity o

f A

uckl

and

on 1

2/05

/14

For

pers

onal

use

onl

y.

706 CAN. J . PHYSIOL. PHARMACOL. VOL. 62. 1984

their technical assistance and Mrs. Debbie Browne for her assistance in the preparation of this manuscript.

ARMSTRONG, 9. A., 6. S. MARKS, and P. W. ARMSTRONC;. 1980. Absence of metabolite formation during nitroglycerin-induced re- laxation of isolated blood vessels. Mol. Pharmacol. 18: 1 12- 1 16.

ARMSTRONG, J. A., S. E. SLAUGHTER, G. S. MARKS, and P. W. ARMSTRONG. 1980. Rapid disappearance of nitroglycerin following incubation with human blood. Can. 9 . Physiol. Pharmacol. 58: 459-462.

ARMSTRONG, P. W., J . A. ARMSTRONG, and G. S. MARKS. 1980. Pharmacskinetic-henlodynamic studies of intravenous nitroglycerin in congestive cardiac failure. Circulation. 62: 160- 166.

ARMSTRONG, P. W., J . A. MOFFAT, and G. S. MARKS. 1982. Arterial- venous nitroglycerin gradient during intravenous infusion in man. Circulation, 66: 1273 - L 276.

BENNETT, B. M . , D. A. S. TWIDDY, J . A. MOFFAT. P. W. ARM- STRONG, and G. S. MARKS. 1983. Sex-related difference in the metabolism of isosorbide dinitrate following incubation in human blood. Biochem. Phamacol. 32: 3729 - 3734.

CRAVEN, P. A., and F. R. BERUBERTIS. 1983. Requirement for heme in the activation of purified guanylate cyclase by nitric oxide. Biochim. Biophys. Acta, 745: 3 10-321.

DIAMOND. J. , and K. S. BLISARD. 1976. Effects of stimulant and relaxant drugs on tension and cyclic nucleotide levels in canine femoral artery. Mol. Pharmacol. 12: 688-692.

GERZER. R., E. B ~ H M E , F. WOPMANN, and G. SCHULTZ. 1981. Solu- ble guanylate cyclase purified from bovine lung contains heme and

copper. FEBS Lett. 132: 71 -74. GERZER, R., F. HOFMANN, and G. SCHULTZ. 1981. Purification of a

soluble, sodium nitropnrsside-stimulated guanylate cyclase from bovine lung. Eur. 9. Biochem. 116: 479-486.

IGNARRO, L. J . , J. N. DEGNAN, W. H. BARICOS, P. J. M r % ~ o w ~ ~ z , and M. S. WOLIN. 1982. Activation of purified guanylate cyclase by nitric oxide requires heme. Comparison of heme-deficient, heme- reconstituted and heme-containing forms sf soluble enzyme from bovine lung. Biochim. Biophys. Acta, 718: 49-59.

IGNARRO, L. J . , H. LIPPTON, J. C. EDWARDS. W. H. BARICOS, A. L. WYMAN, P. J . KADOWITZ, and C. A. GRUETTER. 198 1. Mechanism of vascular smooth muscle relaxation by organic nitrates, nitrites. nitropnmsside and nitric oxide: evidence for the involvement of S-nitrosothiols as active intermediates. 9. PharmacoB. Exp. Ther. 218: 739-749.

IGNARRO, L. J., K. S. WOOD, and M. S. WOLIN. 1982. Activation of purified soluble guanylate cyclase by protoporphyrin IX. Proc. Natl. Acad. Sci. U.S.A. 79: 2870-2873.

NOONAN, P. M., and L. Z. BENET. 1982. Formation of mono- and dinitrate metabolites of nitroglycerin following incubation with hu- man blood. Int. J. Pharm. 12: 331 -340.

ROSSI-FANELLI, A., and E. ANTONINI. 1958. Studies on the oxygen and carbon monoxide equilibria of human myoglobin. Arch. Bio- chern. Biophys. 77: 478-492.

Wu, C. C. , T. SOKOLOSKI. M. F. BLANFORD, and A. M. BIJRKMAN. 198 1. Absence of metabolite in the disappearance of nitroglycerin following incubation with red blood cells. Int. J . Pharm. 8: 323 -329.

Can

. J. P

hysi

ol. P

harm

acol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Uni

vers

ity o

f A

uckl

and

on 1

2/05

/14

For

pers

onal

use

onl

y.