inhibition of human lymphocyte coproporphyrinogen oxidase activity by metals, bilirubin and haemin

262 Biochimica et Biophysica Acta, 1135 (1092) 262-268 © 1992_ Elsev/er Science Publishers B.V. All rights reserved 0167-4889/92/$05.00

Inhibition of human lymphocyte coproporphyrinogen oxidase activity by metals, bilirubin and haemin

E n r i c o R o s s i a, P a u l V. A t t w o o d b a n d P e t e r G a r c i a - W e b b a

= Clinical ,i~o~,em~.- 3- Depurlment. Queen Elizabeth H Medical Centre. Ned.lands (Australia) and ~ Department of Biochemiswy. Unirersio" of Western Auslr~tia, Nedlands Glnstralial

(Received 27 Nm'embcr 1991)

Key words: Coprol~rphyrinogen I!1 oxidase; Melal; Bilirubin; Haemin; Enz~3me inh~ition; Inhibition kinetics; (Human b'rapP.oc31e )

Coprolmrph3nrinogen II1 oxidase activity in human lymphocytes was found to be inh~ited by cadmium and mercury but not by lead. The organometal compounds tributyltin aad methylmercury were effective inh~itors of this haem biosynthetic pathway emyam. Haemin (the ultimate product of the pathway) and bilirubin (a product of haem catabolism) were also shown to be inhibitory. Kinetic studies performed under initial velocity conditions showed that bilirubin was a non-competith'e inhibitor and that one bHirubin molecule was bound to both the enzyme and enzyme substrate complex. The analysis also showed haemin to be a non-competitive inh~itor in which two haemin molecules bind to the enzyme whereas the enzyme substrate complex accepts only one haemin molecule. The poss~le physiological significance of the inh~ition of coprolmrphyrinogen II! oxidase activity by haemin and bfiirubin is discussed.

h t t rodac t ien

Coproporph~,rinogen !!! oxidase (copro oxidase, EC 1.3.3.3) is an enz~ne located in the mitochondrial in ter-membrane space which catalyses the conversion of coproporpbyrinogen I l l (cuprogen) to protopor- pbyrinogen IX by oxidative decarboxylation. The early methods for measur ing its activity were spectrophoto- metric and involved separat ion of the ul t imate product protoporpbyrin IX by solvent parti t ion [1,2]. These methods gave inadequate separat ion of substrate and product and were not sensitive enough to allow mea- surement of copro oxidase activities in lymphocytes or fibroblasts. However, simple and rapid assays for copro oxidase activities in rat liver [3] and human leucocytes [4] are now based on effective HPLC separat ion of substrate and product. The sensitive HPLC methods proved to be applicable to the study of inhibitors of copro oxidase.

Indirect evidence has led to the suggestion that the excretion of excess urine coproporphyrin following in-

Correspondence: E. Rnssi, Clinical Biochemistry Department, Queen Elizabeth I! Medical Centre, Nedlands, WA 6009. Australia.

Abbreviations: Copro oxidase, coproporphyrinogen III oxidase; ow progen, coprel~r0hyrinogen i11; IIPLC, high performance liquid chromatography;, DMSO, dimethy| sulphoxide.

vivo exposure to zome metals results from interference with the action of copro oxidase, in the case of lead, an elevated urine coproporphyrin 1I! has long been recog- nized as a consistent feature [5]. Animal s tudies using lead [6], arsenic [7] and mercury [8] have shown that prolonged exposure to low levels of these metals gives a pronounced increase in urine coproporpbyrin excretion, consistent ~ k h the dose and durat ion of metal exposure. However few studies have a t tempted to demonstrate in-vitro inhibition of copro oxidase.

Batt le e t al~ [2] s tudied rat liver copro oxidase using high concentrat ions of meta l ions and showed that Cd 2+ and Pb 2+ at 20 mM and As 3+ at 40 mM inhib- i ted between 30% and 40% of the activity. More recent work has also suggested that significant inhibition of copro oxidase requires high concentrat ions of in-vitro metals. Both rat liver and kidney copro oxidase have been shown to be approx- 50% inhibited by 1 mM Hg 2+ while ! mM As s+ was shown to inhibit the kidney rather than the liver enzyme [9]. The present results show that inh~i t ion of human lymphocyte copro oxidase by Hg 2+ and Cd 2+ can be demonstra ted at much lower concentrat ions (50 /~M). Organometal compounds have not pt-eviously been tested as inhibitors and we show that tn'butyltin and methylmercury also inhibit copro oxidase. These are ,qquatic pollutants; tributyltin is used as a marine antifouling agent and methylmercury is formed by aquatic mi-

croflora (sulphate-reducing bacteria) from inorganic mercury. Bilirubin, the break down prc~act of haem, is a toxic compound known to inhibit dehydrogenases [10] and the enzymes involved in electron transport [11]. Recent studies have shown that bilirubin inhibits a related enzyme, protoporphyrinogen oxidase [12]. The present study shows that bilirnbin is an effective non- competitive inhibitor of lymphocyte copro oxidase. In previous work we showed that significant inhibition of human lymphocyte ferrochelatase occurred at haemin concent~-ations of less than 10 / tM [13]. Here we demonstrate that haemin also inhibits lymphocyte copro oxidase and use enzyme kinetics to examine the mechanisms for bilirnbin and haemin inhibition.

Materials and Methods

Chemicals Haemin (ferriprotoporphyrin IX chloride), mesopor-

phyrin, coproporphyrin and protoporphyrin were from Porphyrin Products (USA). Unconjugated bilirubin was from National Bureau of Standards (USA), tetraethyl lead from Associated Octel (UK), tributyltinchloride from Fiuka (Switzerland) and methylmercuric chloride from K and K Laboratories (USA). Metallic sodium, mercury, ascorbic acid and dimethyl sulphoxi~,: (DMSO) were from BDH Chemicals (UK). Ficoll- Paque was from Pharmacia (Sweden) and t~csferriox- amine mesy!ate from C~a-Geigy (Switzerland).

Lymphocyte enzyme preparation Lymphocytes were isolated from heparinized blood

by centrifugation with FicolI-Paque [14] followed by lysis of contaminating erythrocytes with hypo-osnmtic saline [15]. In previous work we desen'bed optimal sonication conditions and showed that iymphocytes were preferred over mixed leucocytes because of a higher specific activity [16]. Lymphocyte pellets were sonicated at 4°C for 20 s with a microprobe (Model B15, Branson, USA) in 20 mM Tris-HCI buffer (pH 8.0) containing 20% (v/v) glycerol at a protein concentration of between 1.5 and 2.0 g / I . Protein was determined by a trichloroacefic a~d-Ponceau S method using human serum albulnin as standard [17].

Copro oxidase assay Copro oxidase activity was determined using the

HPLC method of Guo et al. [4]. Freshly prepared coprogen I!1 I, fmal concentration 1 / tM) was incubated with the lymphocyte e n z ~ e preparation in the presence of EDTA (final concentration 0.6 raM) to prevent metal chelation of the ultimate product protoporphyrin IX. The reaction was stopped with 20% (w/v) trichloroacetic ac id /DMSO (1 : 1, v /v ) containing meso- porphyrin as internal standard.

263

HPLC conditions A Model 1084 liquid chromatograph (Hewiett-

Packard, USA) was used with an RF-535 fluorescence detector (Shimadzu, Japan) equipped with an R928 red-sensitive photomultiplier tube (Hamamatsu , Japan). The optimal fluorescence wavelengths for protoporphyrin were used: excitation wavelength 400 nm and emission wavelength 625 nm. The separation was performed on a 3.9 m m × 15 cm column packed with 4 p.m Novapak C18 silica (Waters, USA). The flow rate was 1.5 ml /min using a mobile phase prepared by adding 120 ml of ammon:'um acetate buffer (I mol/ l , pH 5.16) to 880 ml of HPLC-grade methanol.

Desferrioxamine as chelator Recent HPLC methods have included EDTA (final

concentration 0.6 raM) to prevent the further conversion of protoporphyrin IX to haem by ferrochelatase [3,4]. However, one of the aims of the present work was to study metal ions as inhibitors of copro oxidase and a general metal chelator such as EDTA could not be used. Desferrioxamine is a specific iron chelator which effectively blocks ferrochelatase action [18].

To compare the efficacy of EDTA and desferrioxamine, copro oxidase activity was measured in eight different lymphocy~.e preparations without any chelator, w!th EDTA (final concentration 0.6 mM) a~d with dcsferrioxamine (final concentration 500/zg/r~dL Tak- ing EDTA as representing 100% enzyme activity-, the mean (1 S.D.) activities as a percentage of the EDTA activities for the eight lymphocyte preparations were: desferrioxamine = 95 (5)% and no chelator = 71 (8)%. Desferrioxamine was used as a specific iron chelator in the study with metal ions and organometais. In the work with bilirubin and haemin the assay was conducted with EDTA as originally described [4].

Metals and organomemls Aqueous solutions of Pb 2+, Sn 2+ and Cd 2+ are

unstable due to the precipitation of hydroxides, and solutions were freshly prepared for each experiment. Metal salts were dissolved in water and added in 10-1zl aliquots to give the quoted final concentrations. Reagents were added in the following order: metal, lymphocyte enzyme preparation, assay buffer (Tris-HCI, pH 7.0 buffer containing desferrioxamine) and preincu- bated for 5 min at 37°C. The reaction was started with freshly prepared coprogen substrate :1 /zM) The organometal compounds, tributyltin chloride, methylmercuric chloride and tetraethyl lead, were dissolved in ethanol and added in 10-/tl aliquots to give the quoted final concentration. Reagents were added in the same order as above. The final concentration of ethanol was 3.8% by volume and con*.rols containing ethanol were used in all the experiments with organometals.

264

Non-enzymic formation o f metalloporphyrins The relative ease of nietalloporphyrin formation un-

der certain conditions has been reported by several authors. For example, trace quantities of Cu 2+ and Co 2+ can readily form non-fluorescent complexes with haematoporphyrin r19] and Fe 2+ will form haem non- enzymatically [20]. Spontaneous formation of chelates could be interpreted as inbAbitiou of copro oxidase activity if n o n - e n z ~ controls are not performed. Non-enzymic mcubations of the copro oxidase assay were therefore conducted with each of the metals shown in Table I and organometals shown in Table !! to assess spontaneous metalloporphyrin formation.

Nou-enzymic incubations were conducted under the copro oxidase assay conditions by adding the metal to Tris-glycerol, followed by assay buffer containing desferrioxamine, protoporphyrin to simulate product for- marion and coprogen substrate added last. Following the addition of trichloroacetic acid/DMSO the mix- ture was applied to the I-IPLC and the peak areas of coproporphyrin III (representing snbstrate) and added protoporphyrin IX (representing product) were examined.

Only one metal, Cu 2+ (as CuSOa) gave reduced peak areas. The total coproporphyrin and protoporphyrin peak areas were only 25% of the control, indicating the formation of non-fluorescent copper chelates. Although Cu 2+ may inhibit the en~me, the formation of the copper chelates did not allow this observation to be made.

Pre~rat/o~ o f ~tirt~hb: ~nd haemin solutions Bilirubin (ouconjugated) was dissnlvc6 in dimcthyl-

snlphoxide (DMSO) to give a 6.3-raM stock solution and added in 2-/~i aliquots to the 250-/~1 assay tubes so that the final DMSO concentration was 0.8% (v/v) which did not affect copro oxidase activity. Haemin was initially dissolved in DMSO and 0.2 M Tris-HCI buffer (pH 7.0) added to give a 2.6-raM stock solution containing 20% (v/v) DMSO and then added in 10-1tl aliquots to the 250-/,tl assay tubes so that the final DMSO concentration was 0.8% (v/v).

Results

Effect o f metal ions and organometals Table I shows the results of incubating metal ions at

a final concentration of 50 g M on the activity of lymphocyte copro oxidase. Only two of the metals tested, Cd 2+ and Hg 2+, gave marked inh~ition of enzyme activity. Neither form of Pb 2+ tested was inhibitory and preincubatiou of either form with the enzyme for periods of up to 2 h prior to initiating the enzyme reaction also had no effect.

Table 11 shows the effects of tetraethyl lead, tributyltin chloride and metbyimercuric chloride at final

TABLE i Effect of metals on bmptu~'te copro oxida~ actiLity Results are mean~ of duplicates of a typical experiment. Cu 2+ gave n~-cnz~n~ chelate formation and could not be assessed.

Metal Copro oxidase activity (50 #M) (% of control)

NH 100 MnSO~ 105 SnCI 2 100 NiCi 2 94 Ai2~SO..~ 94 PI~CH3COO) - , 93 ~NO3) 2 93 Cd(CH;COO) z 68 HgCI 2 2 C-aSO 4

concentrations of 100/zM on the activity of lymphocyte copro oxidase. Tributyltin and litethylmercury proved to be effective inhibitors at a concentration of 100 #M.

Inhibition by bilirubin and haemin Wben unconjugated bilirubin was added to the lym-

phocyte copro oxidase assay, inhibition was effective at concentrations of less than 10 /~M, with no further inhibition observed at bilirubin concentrations up to 50 #M (Fig. 1). Haemin gave a concentration-dependent inh~itiou of copro oxidase activity with 50% inhibition occurring at a concentration of 15 ~M (Fig. 1). The inhibition profiles obtained with bilirabin and haemin allowed selection of appropriate inhibitor concentrations for subsequent kinetic analysis.

Substrate ki.~tetics Substrate kinetic analysis using Lineweaver-Burk

plots gave results which failed to give linear double-in- verse plots. Several factors combine to make the K,n determination for copro oxidase difficult to determine accurately. The substrate coprogen is unstable and oxidizes during the course of the reaction, an effect which becomes more important at concentrations around the K m value. The oxidized substrate (coproporphyrin) is also a competitive inhibitor of the enz~ac [.21]. These difficulties are reflected in the range of

TABLE I!

Effect of orgammu'mls on l)wwhocyte copra o~idase actiriO"

Results age means of duplicates of a typical experiment.

Organometal Copro c0gidase activity (100 ~tM) (% of conlrol)

Nil 100 Tetraethyl lead 103 Tributylrin chloride 31 Methylmercuric chlozid:= 8

% of 60 ac t i v i t y

40

20

O0 1 0 2 0 3 0 4 0 5 0

[ InhthRor ] Fig- I. Inhibition of coproporDhyrinogen oxidasc activity by bilirubin (11) and haemin (o). Sonieated lymphocytcs were incubated with each inh~itor and the enzyme acth,'ity assayed. Results are means of

duplicates of a Wpical experiment.

values reported. The K m for rat liver copro oxidase, for example, has been quoted as 48/ tM [22] and 0.16 /~M [23].

Initial attempts to obtain a K m for the substrate revealed that the incubation time was critical. As substrate exhaustion may easily occur in this assay, it was

265

decided to study the time-course with a low substrate concentration (0.04 p.M). The results confirmed that substrate exhaustion had occured within the 60-min incubation time. Therefore the incubation time used for kinetic analyses was reduced to 20 rain.

By using lymphocytes rather than a mixed leucocyte fraction it was possible to determine a K m for copro oxidase from a homogenous ce!lalar population. The K m for the enzyme from human f.vmphocytes was determined by fitting the data to the general velocity equation (Eqn. 1) shown below, using the non-linear regression program of Duggleby [24]. The value was 0.08+0.01 /~M (mean +SD, for six preparations) which is lower than the previously reported values of 0.21 p,M for lymphocytes [23] and 0.12 p.M reported for mixed leucocytes [4].

Analysis of kinetic data The kinetics of bilirubin and haemin inhibition were

examined by assaying copro oxidase activity at different concentrations of inhibitor while varying the concentration ef the coprogen substrate. Each set of data at a particular inhibitor concer, tration was fitted to Eqn. 1 and the slopes and intercepts of the Lineweaver-Burk plots of the data, i.e. Km/Vm~ and I /V,~ were derived from these fits. Initially Eqn. 2 was fitted to the secondary data (slopes and intercepts) to determine if

1 ; v 20

1 0 ~

O0 ' 110 ' 2*0

" [CPg•n] (1 / pM)

7 " / lS, o

0 1 2 3 4 |Bitirubin] (IIM) [Bitirubin] (tIM)

Fig- 2. Inhibition of coproporph~vrinogen oxidase activity by bilirubin. (a) Lineweaver-Burk plots o[ recipFocal ini:ial velocity, I / ( . ( I /nmol h - z mg - t ) with respect to reciprocal copropozphyrinogen concetration, I/[CPgen] ( I /p .M) in the presence; of 0 pM (0) , I t ,M (o). 2 p.M (ElL 3 p M (z~) and 4/~M (e) bilirubin. (b) Secondary plot of slopes (K m / Vm~) with respect to bilirabin concentration (p.M) (c) Secondary plot of

intercepts (l/V,~.~,) with respect to bilirubin concentration (/~M).

266

these parameters had a linear or parabolic dependence on the inlu'bitor concentration.

Vm~ v = - - { ! )

I + g . , / I S ]

Slope or intercept = constant (I + [II/K z + [I]2/KtK2 ) (2)

where [i] is the concentration of inhibitor and [S] is the concentration of coprogen substrate. In the case of a slope plot, constant = Km/Vma x, K I = K l, K2 = K ~ (where K I and K~, ate the equih'brium constants de- scribing the dissociation of one inhibitor molecule from EI and EI 2, respectively) and in the case of an intercept plot, constant = 1 / V ~ , K t = K i s , K 2 = K~s (where Kts and K~s, are similar constants for EIS and EI2S, respectively).

,All the fits to data were performed using a non-linear regression program [24] with constant weighting of all data points. The lines shown in the Lineweaver-Burk plots are derived from the final fits to the velocity equations and the points are the means of duplicate experimental determinations.

Kinetic atla~sis o f bilirubin b,hibition The kinetics of bilirubin inhibition were examined

hy determining the inidal velocity of the ~ o o~dase reaction in the presence of five concentrations of bilirubin (0,1, 2, 3 and 4/~M). Eqn. 1 was fitted to the velocities of the reaction at each bilirubin concentration and from these fits values for 1/Vm~ (intercepts) and Km/Vma x (slopes) were obtained.

The secondary plots of these parameters versus bilirubin concentration indicated a linear relationship. This was confirmed by fitting each set of data to Eqn. 2 where in both cases the estimate of K2 tended to infinity. Scheme I was proposed to fit the data in that it depicts non-competitive inh~ition with only one bilirubin molecule binding.

Ks E " E S - - ~ E + P

E1 ~ E I S

Scheme !. where S = coprogen substrate, I = bilirubin inhibitor, E = enzyme and P ~ product. The velocity Eqn. was then derived for this scheme:

k~JE],o,~,l v (3)

l+[i]/Kts + Ks/iS](1 +[ll/K,)

The complete set of velocity data at all five inhibitor concentrations was then fitted to this Eqn. and the values of the inhibition constants were calculated. The

Lineweaver-Burk plot (Fig. 2a) shows reciprocal reaction velocities against reciprocal coprogen substrate concentrations in the presence of bilirubin. The secondary plots (Fig. 2b and c) of slope ( K m / V m ~ ) and intercept (1/Vm,) against bilirubin concentrations were both linear, consistent with the predictions of the velocity equation.

The reaction scheme conforms to a non-competitive inh~ition with respect to coproger, substrate i.e. the bilirubin binds to both the enzyme and en~me coprogen complex with linear secondary plots indicating binding of only one bilirubin molecule The inh~ition constants and their standard errors were K I = 1.2 ('.9.2) /~M and K~s = 1.9 (0.2)/LM, indicating that bilinibin binds only slightly more strongly to the enzyme alone than to the enzyme-coprogen complex. These results demonstrate non-competitive inh~ition of lymphocyte copro oxidase by plr~iological concentrations of bilirubin. The normal range for human plasma hilirubin is less than 20 ~M.

Kinetic analysis o f haemin inhibition Haemin ioln'bition of copro oxidase w-~_s examined

by determining the initial reaction velocities in the presence of five concentrations of haemin (0, 4, 8, 12 and 16/zM). The velocities of the copro oxidase reaction at each haemin concentration were fitted to Eqn. I and, from these fits, values for 1/Vm. ~ (intercept) and K m / V m ~ (slope) were obtained. Fitting of Eqn. 2 to the inte.rcept data indicated that the relationship between the intercepts and haemin concentration could be parabolic, although the standard errors of the estimates of both K I and K 2 were almost as large as the estimates themselves. A fit of Eqn. 4 to the intercept data also gave a reasonable fit (see Fig. 3a) Thus it was possible that either one or two molecules of haemin might bind to the enzyme substrate complex.

Intercept = (1 +[I]/KI) (4)

The secondary plot for the slopes was paraboP.c (Fig. 3b), indicating that two molecules of haemin bound to the enzyme. However, on examining the parabolic fit for the slopes, the concentration of El was found to be negligible, as the K t value tended to infinity and therefore E12 was formed immediately. Due to the uncertainty of the relationship between the intercepts and concentration of haemin thc~-e were two possible reaction schemes, as shown below.

E_ gs ~ E S ~ E + P

E1 s EIS

Sch~.me I!

E Ks ,ES t . . . .~ E+p

El, EIS

EI2S

Scheme Ill where S = coprogen substrate, 1 = haemin inhibitor, E = enzyme and P = product.

The rate equations for the two schemes are:

~ [ E ] ~ , I+[I]/K,s+Ks/[S](I+[I]2/K,K,,) (5, Scheme I!)

k,~lEl,o~a L

I + [I]/K~ + [I]2/KIsKIs, + K s/[:5;]( I + [l]2/KI K 1, )

(6, Scheme !!I)

Eqns. 5 and 6 were fitted to the complete set of velocity data. Eqn. 5 gave the best fit to the data, showing that Scheme B is the best model, whilst in the

8

-!- o 4 | 12 16 -20

2O

lS

1 Iv 10

5

e .4

. . . . . . . 0 4 8 12 16 20

[Hum~] (~u)

C

lO 2 0

1 / [CPgen] (1 1 IJL~)

Fig. 3. Inhibition of COl~mimr~ogen oxidase act~ity by hae ,~n~ (a) Seconda~ plot of intercepts (l/Veto) with respect to haemin concentration (uM). (b) Secondary plot of slopes (Km/V ~ ) with respect to hacmin concentration t/zM). (c) Lineweaver-Burk plots of reciprocal initial velocity, l / r (l/nmol h -t mg -t ) with respect to reciprocal coptotmrpbyfitmgen concentration, l/[CPgen] (l//zM) in the presence ~f0/zM(ra), 4/zM (o), 8 FM(III), 12 laM(A)and 16

~M (O) haemin.

267

fit of Eqn. 6 KisKls, tended to infinity, indicating no formation of EI2S.

The complete set of velocity data at all five inhibitor concentrations was then fitted to this equation and the values of the inhibitor constants were calculated. The Lineweaver-Burk plot (Fig. 3¢) shows reciprocal reaction velocities against reciprocal coprogen substrate concentrations in the presence of haemin. The inhibition constants and their standard errors were K I K I, = 120 05) /zM 2 and Kts = 27 (2.9) FM.

The results show that haemin is a non-competitive inhibitor of lymphocyte copro oxidase. Furthermore, when the enzyme is not bound to the substrate it has two binding sites for haemin such that the binding of the first haemin molecule is immediately followed by the binding of a second, so that the enzyme with only one haemin bound exists at negligible concentrations over the range of inhibitor concentrations used (4-16 p.M). On the other hand, the enzyme substrate complex can only accept one haemin molecule.

Discussion and Conclusion

Metal and organometal inhibition In the present work, the assay method was modified

by substituting a specific iron chelator (desferrioxamine) for EDTA, allowing study of a range of metals. Of the seven metals tested in this study only C_.d 2+ and Hg 2+ gave inhibition at 50-/zM concentrations. No effect was observed with Mn 2+, Sn 2+, Ni 2+, AI 3+ or Pb 2+. Virtually complete inhibition of lymphocyte copro oxidase was obtained with 50 ~M Hg 2+. This finding does not agree with a recent study examining the effect of Hg 2. on rat liver and kidney copro oxidase where 1000/zM Hg 2+ inhibited approx. 50% of the activity [9]. The reasons for the discrepant find- ings are probably the use of EDTA in their copro oxidase method and preparation of the enzyme in 3 mM dithiothreitoL Both EDTA and dithiothreitol would bind Hg 2+, accounting for the relatively high Hg 2+ concentrations required to demonstrate inhibition. This conclusion is supported by another study which demonstrated inhibition of copro oxidase in anaerobic bacteria with 100/zM Hg 2+ when the assay was carried out in the absence of any metal chelator [25]. "llte latter study also reported that Cu 2+ gave non-enzymic chelation which prevented observation of enzyme inhibition.

The three organometals tested in this study, tetraethyl lead, tributyltin and metbyimerc~-y are all known to be toxic and the latter two were shown to be potent inhibitors of lymphocyte copro oxidase. We have previously shown that lymphocyte ferrochelatase is also inhibited by methylmercury [26] which is lipid soluble and readily taken up by brain and neural tissue leading to the development of cerebellar ataxia and

268

cort ical bl indness . T h e observed symptoms may be par t ly due to inhibit ion o f these two enzymes o f haem biosynthesis resul t ing in insuffu.-ient haem for neura l funct ion.

B// /mb/n/rib/b/t /on T h e s tudy o f bi l irobin inhibit ion showed tha t cop ro

oxidase inh~i t~an could be demons t r a t ed bo th a t maxi- mal velocity, condi t ions (subst ra te concen t ra t ions 12 x K m) a n d initial velocity condi t ions (substrate concen- t ra t ions a r o u n d the K ~ ) by bil irubin concen t ra t ions o f less t han 5 # M . T h e kinet ic s tudy revealed tha t the mechan i sm was non-compet i t ive in type, with bi l irubin b ind ing to bo th the enzyme a n d enzyme subs t ra te oDml~ex.

Litt le is known o f the mechan i sm o f c o p r a oxidase; however , it does not conta in o r utilize the ch romo- p lmr ic cofac tors t3pical o f o t h e r enzymes which catal- yse e lec t ron t rans fe r react ions, bu t does make use o f p ro te in tyrosine res idues as e lec t ron accep tors [27]. The next enzyme in the pa thway, p ro toporphyr inogen oxidase, a lso lacks these cofactors , catalyses a n elec- t ron t r ans fe r reac t ion a n d has been s h o ~ to be i n h ~ Red by bil irubin. However , in this case bil irubin was a compet i t ive inh~ i to r , sugges t ing tha t it compe tes with the subs t ra te for the same ca ta 'y t ic site o n the enzyme: [12.]. A l though the mechan i sm o f act ion o f this enzyme is unknown, it has been specu la ted tha t it may be s imilar to c o p r o oxidase [12]. T h e presen t f inding o f a d i f ferent mechan i sm for bi l irubin inhibit ion o f c o p r o oxidase suggests tha t the reac t ion mechan i sms o f the two enzymes a re also different .

Haemin inhibition U n d e r initial velocity condi t ions (substrate concen-

t ra t ions a r o u n d the Km o f 0.08 /zM) haemin i nh~ i t s l )mphocy te c o p r o oxidase a t concen t ra t ions be tween 4 a n d 16 /zM. H a e m i n does no t a p p e a r to have been tes ted to see if it would inh.;bit this enzyme, a l though we have previously shown tha t it i nh~ i t s lymphocyte fe r roche la tase [13]. T h e pos s~ l e physiological siguifi- canee o , ¢ this f inding is unknown becaase little is known o f the precise regu la tory role o f f ree haem. Free haem is very recent ly ~ t h e s i z e d a n d not yet bound , o r pos s~ ly has jus t been re leased f rom haemopro te ins , a n d the existence o f free h a e m pools in mi tochondr ia , cytosol a n d endoplasmic re t iculum has been p roposed [28].

I t is known that haem c~nnot freely penetrate mito- chondr ia l m e m b r a n e s [29]. Fo r the synthesis o f haemog lob in a n d some cy tochromes the haem must the re fo re be t r anspo r t ed f rom its site o f synthesis ( the mi tochondr ia l matr ix) to the cytoplasm. Binding pro- teins, poss~ ly l igandin, a re t hough t to be impor tan t in the efflux o f h a e m f rom mi tochondr ia [30] The assess- m e n t o f a n y pos s~ l e physiological significance for the

haemin inh~ i t ion o f c o p r o oxidase must awai t fu r the r knowledge o f bo th the modes o f t ranspor t o f haem a n d the funct ioning o f the regu la tory pool o f free haem.

W e t h a n k the Sir Char les G a i r d i n e r Hospi ta l Re- search F u n d for a g r a n t to suppor t this work, a n d Dr . E. Leake for useful discussions a n d the provision o f a comprehens ive da t a base o f the publ ished l i terature.

Refenmces

I Sano. S. and GrankL $.(1961) J. BioL Chem. 236. 1173-1180. 2 BattY. A-M.C~ Benson. A. and Rimlngtor~ C. (1965) Biochem. J.

97. 731-740. 3 LL F~ lain. C.K. and Peters. TJ. (19861 Biochem. J. 239, 481-484. 4 Guo. R-, L/m, C.K. and Peters, TJ. (19881 Gin. Chem. Acta 177,

245-25Z 5 Watsoa, C..J. (1936) .L C]in. InvesL 15. 327-343. 6 Fo~er, B.A_ K/mmeL CA_ Woads, J.S~ McConnell, ILE. and

CnaaL LD. (19S01Tmg'col. Appl. P h a ~ 56. 59-77. 7 Woods. $.S. and Foamier. B.A. (1978} ToxicoL Appl. PharmacoL

43. 361-371. 8 Woods. J.S. and Forker. B.A. (1977) $. Lab. Clin. Mnd. 90,

~6-27Z 9 Woods. J.S. and Saml~m. M.R. (1989} Tm~ol. AppL Pharma-

col. 97.183-190. 10 F]itman. R. a,"td ~orth. N.G. (1966) J. BioL Chem. 241. 669-674. I! Nob', B.A_ Boceri~ A_. Ga.razoo P.A.M. and StoppanL A.O.M.

(19T2} FEBS LelL 27, 270-274. 1.2 Ferreh'a, O.C. -,rod Dm~', H.A. (1988) Bioc~m. J. ?_sO, 597-603. 13 Rossi, E. Airax~L~J, P.V~ Ga.rcia-Webb. P_ Costin, K.A. (19901

Biodfim. Bmplcys. Acta ItO8. 375-38!_ 14 Bgg~t. A. (1968) Scand. J. CErt. Lab. l~esL 2l {Sg.opl} 97,

77-89. 15 PateL A.K., HalleU. M.B. and CampbelL A.K. {1987) Biochem. J.

248, 273-180. 16 Ros~ E-. ~ - W e b b , P. and Costin, ICA. (1989) Gin. Chim

Acta 181.115-118. 17 Pesce. M.A. and Strande. C.5. (l~'yr.rj) Clin. Chem. 19,1265-1267. 18 Srivaslava. G~ Brooker. J.D. May, BJC. EllioU. W. (1980)

B/ochem. J. IS& 781-788. 19 Sommer. S_ Riminglon. C. and Moan. J. {1984) FEBS LelL 179.

267-271. 20 Takelan~, S. and Tokunaga. IL (1984) B;ochim. Biogh~. Acre

798, 226-230. 21 Elder, GH. and E,-ans, J.O. (1978) Biochem. J. 169, 205-214. 22 Yoshinaga. T. and Sam3, S. (1980) J. B/ol. Chem. 255. 4722--4726. 23 Gran~hamp. B. and No,mann. Y. (1982) En~me 28. 196-205. 24 Duggleby, R.G. (1981) Anal. Biochem. II0. 9-1S. 25 Moil. M. and Smm. S. (1972_) Biochilm Bioph~..~.-'la 264. 252-

262. 26 RossL E.. An~nd . P.V_ Garcia-Webb. P. and C_.ostin. K.A.

(19W3) Cfin Ch/m. A~ta 188,1-14. 27 Yo~hinaga. T. and Sano, S. (1980) J. B/ol. Chem. 255, 4727-4731. 28 Kappas, A., Sassa, S, Galbraith, R.A. and N0rdmann, Y. (i989)

in The Me:abohe Basis of Inherited Disease, 6th ndn. (Stanbu:y, J.B. Wyngaarden, J.B. and Fredcrickson, D.S. e~.). pp. 1312- 1317. McGraw Hall tqew York.

29 Kollet. M.F_ and Romslo. !. (1977) Biochim. Bh3phys. Acla 461. 283-296.

30 Keuerar. B~ SraL KS. and Christodoulides. L. (1976) Biochim. Bioph~x Acta 428, 683-689.

inhibition of human lymphocyte coproporphyrinogen oxidase activity by metals, bilirubin and haemin

Documents