Characterization of porphobilinogen deaminase from rat liver

Biochimica et Biophysica Acta, 957 (1988) 97-104

97

Elsevier BBA33231

C h a r a c t e r i z a t i o n of porphobilinogen deam~nase from rat liver M a r t a B. M a z z e t t i * a n d J. Ma~'ia Tornio Departamento de Qulmica Biol6gica, Facuitad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Ciudad Universitaria, Buenos Aires (Argentina)

(Received 7 July 1988)

Key words: Porphobilinogen deaminase character!~:io~l; Enzyme purification; Heine synthesis; (Rat liver)

Pollphobilinogen deaminase (porphobilinogen ammonia.lyase, EC 4.3.1.8) was isolated from rat liver. The final preparation was homogeneous according to polyacrylamide gel electrophoresis and immunediffusion criteria. Electropboresis of the native enzyme revealed a single band of activity whT,ch was distributed into three bands after incubation with porphobllinogen. When electrophoresed under denaturing condition it displayed a single pulypeptide band with a molecular weight of 42000 confirmed by exclusion chromatography and by sucrose density gradient centrifugation. The enzyme showed a pH optimum of 7.5 both in 0.! M sodium phosphate and 0.05 M Tris-HC! buffer, when ~ y e d at 37 ° C. An isoe|ectric point of 4.9 for the native purified protein was found. Hepatic porphobfiinc~gen deaminase was remarkably beat-stable showing maximum activity at 5 5 - 6 0 °C with one break in the Arrhenius plot. The kinetic behaviour of the purified enzyme followed the typical Michaelis-Menten kinetics with values of K m--- 17 p M and Vma~ = 29.4 units power mg in 0.1 M phosphate buffer at 37 o C. The amino acid composition was determined, showing that the enzyme had a low content of sulphur-containing amino acids and a considerable number of acidic residues per tool of polypeptide chain. Reagents known to interact with sulphydry| g~'oups have small effect on rat liver enzyme activity.

Intrt~luction

Porphobilinogen de~,minase (or hydroxymethylbilane synthase) (porphobilinogen ammonia-lyase, EC 4.3.1.8) catalyses the polymerization of four porphobilinogen molecules in a head-to-tail fashion to generate hydroxymethylbilane. This linear

* Fellowship of Consejo Nacional de Investigaciones Cientificas y T~cnicas (Argentina). Correspondence: Dr. J. Maria Tomio. Laboratorio de Porfirinas. Departamento de Quimica Biol6gica, F.C.E.y N. Ciudad Universitaria. Pabellbn II (1428). Buenos Aires. Argentina.

tetrapyrrole is converted by uroporphyrinogen IIl synthase into the cyclic uroporphyrinogen III, key intermediate for all the biologically active tetrapyrroles: heroes, coffins and chlorophyls [1]. Biosynthesis of heme is mainly controlled in the liver by the rate of production of 5-aminolevulinate synthase which is present at a limiting level. A negative feed-back control is also exerted by the intracellular concentration of heme. Under certain conditions, another enzyme, the porphobilinogen deaminase becomes rate-limiting due to its very low activity in this tissue [2]. Porphobilinogen deaminase has been purified from different sources [3-11], but little work has been done with hepatic tissues [12,13]. Recent studies using recombinant DNA techniques have

0167-4838/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Bion~edical Division)

98

led to further progress in the elucidation of the porphobilinogen deaminase gene [14-15]. Since this organ is the target of the acute intermittent porphyria and this enzyme is the key to it [16] it would be interesting to study the physicochemical properties of this protein in mammalian livers in order to further investigate the nature of this genetic failure. With this aim, the present report describes the purification to homogeneity and the characterization of porphobilinogen deaminase from rat liver. Materials and Methods

Porphobilinogen and porphyrins were obtained from Porphyrin Products (Utah, U.S.A.). Bovine serum albumin, ovalbumin, cytochrome c, fllactoglobulin, trypsinogen, peroxidase, alkaline phosphatase, dithiothreitol, p-chloromercuribenzoate, N-ethylmaleimide, GSH and DEAE-cellulose were purchased from Sigma Chemical Co. (U.S.A.). Sephadex G-100, Blue dextran 2000, Agarose IEF, Pharmalite ampholine and p l markers were from Pharmacia Fine Chemicals. The materials used for polyacrylamide gel electrophoresis were obtained from Bio-Rad Laboratories (U.S.A.). All other chemicals used were of analytical grade and from commercial sources. Wistar rats, both sexes, (150-300 g) obtained from Farmerit laboratories (Buenos Aires, Argentina), were used. Assay methods, Porphobilinogen deaminase activity was measured fluorometrically by determining uroporphyrin production as previously described [17]. Reaction mixtures were incubated aerobically in the dark at 37°C for up to 1 h in the presence of 20 nmol porphobilinogen, 10/tmol sodium phosphate buffer (pH 7.5) and enzyme solution in a total volume of 100/tl. Coproporphyrin was used as fluorometric standard and porphyrins were measured with ai~ Aminco-Bowman spectrophotofluorometer. Blanks with either enzyme or porphobilinogen alone were made. Analysis of porphyrins and isomer composition were performed by standard methods [18]. One unit of porphobilinogen deaminase activity is defined as the amount of protein required to

catalyse the formation of 1 nmol of uroporphyrinogen I per h under the specified conditions. Protein concentration was estimated by the method of Lowry et al. [19] using bovine serum albumin as standard, or monitored by absorbance at 280 nm. Native molecular weights were determined by gel filtration on Sephadex G-100 columns as described by Andrews [20], using known protein markers in 0.1 M phosphate buffer. The molecular weight of denatured porphobilinogen deaminase was estimated applying the Laemmli method [21] but using a 5-19% polyacrylamide linear gradient slab gel. Electrophoresed gels (12 x 16 × 0.07 cm) were stained with Coomassie brillant blue R-250. Samples were submitted to analytical electrophoresis in polyacrylamide gels as described by Davis [22], using a 4% stacking gel and a 7.5~ separating gel. Electrophoretic runs were at 4 mA per tubes for 3 h and gels were either stained for protein with Coomassie blue or tested for detection of porphobilinogen deaminase activity. The gels were sliced into 2 mm sections which were incubated for 100 min at 37 °C in the presence of 0.03 pmol porphobilinogen and 15 pmoi phosphate buffer pH 7.5 in a final volume of 0.15 ml. Reactions were stopped by chilling the samples. The gel slices were removed and after adding 0.5 ml 10~ (w/v) HCI, porphyrin production was detected fluorometrically. Analytical isoelectric focusing was performed on agarose slab gels (11 x 11 x 0,1 cm) using an Pharmacia FBE apparatus according to the supplier's instructions. The pH range used was 4.-6.5. The pH profile along the gradient was determined using pl markers. Enzyme activity was identified by incubating gel slices (5 x 1 mm) as previously described. The sedimentation coefficient of the porphobilinogen deaminase was obtained according to the Martin and Ames' method [23], using a 5 to 20~ (w/v) sucrose gradient in 0.1 M phosphate buffer and the following standards: cytochrome c, peroxidase and alkaline phosphatase. The Stokes radius was determined by gel filtration according to the method of Laurent and Killander [24]. Purified enzyme samples (50 pC) were hydrolysed under vacuum in 6 M HC1 containing 1

99 m g / m l phenol [25] for 20 and 40 h at l l 0 ° C to perform amino acid analysis. The content of half cystine plus cysteine was estimated as cystei¢ acid after oxidation with performic acid, prior to hydrolysis [26]. The amino acid composition was measure~ with a Beckman 119 CL amino acid analyser. Immunization. New Zealand rabbits were immunized by intradermal injection of purified enzyme (300 #g) suspended in Freund's complete adjuvant (1 : 1, v/v), every 15 days for 4 months. The rabbits were bled at 5, 9 and 17 weeks after the first injection. A partially purified immunoglobulin fraction from pooled rabbit sera was prepared by precipitation with (NH4)2SO 4 [27]. The immunoglobulin fraction was resuspended in 20 mM sodium phosphate buffer (pH 7.4) to 40~ of the initial serum volume and stored at - 2 0 o C. Ouchterlony double immunodiffusion [28] was used to estimate the titer of rabbit anti-rat serum, and to check the homogeneity of the purified enzyme. After incubation of the gels for 24 h, precipitin lines were detected by staining with Coomassie blue. Purification of porphobilinogen deaminase. All the operations were carried out at 4 ° C unless otherwise indicated. Isolation. Since the liver is a profusely irrigated organ and the erythrocyte porphobilinogen deaminase activity is several times higher than the hepatic one, it was necessary to eliminate the occluded blood very carefully. Rats were killed by decapitation and livers were intensively perfused in situ with cold 0.15 M NaCI. After excision, the tissue was homogenized with 4 vol. of 1.15% KCI and centrifuged at 11000 x g for 30 min. Heat treatment. The supematant was divided into fractions and heated with gentle stirring at 55-60 °C for 10 rain before being rapidly chilled in ice water. The heated samples were centrifuged at 8000 x g for 10 min and the pooled supernatants were either stored or used in the following step. Ammonium sulphate fractionation. Solid ammonium sulphate was added to the supematant to give 35~ saturation. Ammonium salt was again added to the supematant obtained by centrifugation (8000 x g, 30 min.) until 60~ saturation was reached. The resulting pellet was resuspended in 5

mM sodium phosphate buffer (pH 7.5) and dialysed overnight against the same buffer. lon-exchange separation. The dialysed fraction was added to a batch of fibrous DEAE-cellulose. The resin was always subject to precycling and treatment with 5 mM phosphate buffer (pH 7.5) before use. After 30 rain, the resin was washed with 2 vol. of equilibration buffer and the supernatants were discarded. The enzyme was eluted batchwise into 0.134 M phosphate buffer (pH 7.5) (three times). Eluates were pooled and ultrafiltrated using an Amicon PM-10 membrane. Sephadex G-IO0 chromatography. The concentrated solution was applied on Sephadex (3-100 columns and protein fractions were collected in 5 mM phosphate buffer (pH 7.5). All fractions with high specific activities were combined, DEAE-ceilulose chromatography. The pooled fractions from the preceding step were immediately loaded onto a column (2 x 22 cm) of the anionic exchanger (preswollen microgranular form) preequilibrated with 5 mM phosphate buffer (pH 7.5). The elution was carried out using a linear salt gradient (0-0.2 M KCI) in the same buffer. Fractions corresponding to the single peak of activity were rapidly pooled, concentrated and stored frozen In order to search for multiple molecular forms of the enzyme, DEAE-cellulose column chromatography was performed after ammonium sulphate precipitation. Results and Discussion

Purification of porphobilinogen deaminase Results of a representative purification of porphobilinogen deaminase are summarized in Table I. In rat liver tissue, this enzyme activity is very low [13,17] when compared with that in human erythrocytes [9], rat spleen [29], or microorganisms and plants such as Euglena gracilis [11], Rhodopseudomonas spheroides [6,7] or spinach leaves [3]. It is worth noting that human hepatic porphobilinogen deaminase activity is still lower (Mazzetti, M. and Tomio, J.M., unpublished data); The specific activity of rat liver homogenate under our standard conditions was 0.030 units per mg of protein and a value of 29.4 units per mg was obtained for pure protein after DEAE-cel-

100 TABLE I PURIFICATION OF PORPHOBILINOGEN DEAMINASE FROM RAT LIVER a Step

Total activity (units) b

Homogenate 11000 x g supernatant Heat treatment (NH4)2SO4 fractionation DEAE.cellulose elution Sephadex G-100 DEAE-cellulose chromatography

Specific activity (units/mg protein)

352 331 275 185 138 70

0.030 0.050 0.181 0.327 1.620 8.770

52

29.400

Purification (n-fold) ] 1.6 6 11 54 292 980

From 80 g of rat liver. b 1 unit of activity is defined as the amount of the enzyme which produce 1 nmol of uroporphyrinogen per hour under standard conditions indicated in Materials and Methods.

lulose chromatography, with a very low yield (-'1"~.: specific activity obtained was much lower than the values reported from other mammalian sources, i.e., 2300 units per mg for human erythrocytes [9] and 1260 units per mg for rat spleen induced with phenylhydrazine [29], suggesting not only species but also tissue variations. It was necessary to use at least 100 g of tissue on a large-scale procedure to obtain milligrams of pure enzyme. The preparation after heat treatment produced only uroporphyrinogen I from porphobilinogen, but the selected assay conditions only gave traces, if any, of uropyrinogen III in the first steps. The chromatographic profile of a Sephadex (3-100 column showed a single synur:trical peak of activity that eluted after the major protein peak. A 50% loss of activity was always observed in this step owing to a decrease in protein concentration. The anionic exchange chromatography ~olved the total enzyme activity in a single peak, eluted between 85 and 100 mM KCI. However, when DEAE-cellulose chromatography was used after (NH4)2SO4 fracttonation, one main peak of activity and three minor ones were observed (Fig. 1). These findings indicate charge heterogeneity of the enzyme preparation, produced either by the presence of multiple forms of the enzyme [8,9] or

as a result or" proteolytic modification [29] during the purification process.

Homogeneity The homogeneity of porphobilinogen deaminase was established according to two different criteria: (a) Under non-denaturant conditions upon polyacrylamide gel electrophoresis at pH 8.3, a single band of protein and activity was observed as illustrated in Fig. 2A. This was confirmed when SDS-polyacrylamide gel electrophoresis was also performed. However, when the native enzyme was preincubated with porphobilinogen, two additional bands coincident with protein-stained bands were observed (Fig. 2B), these new ones being more anodic. These results are in agreement with the findings reported by Anderson and Desnick [9] in support of their hypothesis of enzyme-substrate covalent intermediates; however, not all of the reported intermediates were detected by us. (b) Antibody raised against porphobilinogen deaminase displayed only a single precipitin line on immunodiffusion when directed to heated supernatant and purified enzyme fractions, with a fused single arc of identity between them. These results indicate the absence of other antigenic proteins in the purified enzyme. Moreover, preliminary work showed that the rat liver antibody cross-reacted partially with a fraction of "]1.0 I(

? t

:3.3 I

ut

).3 >.

,1

0.~ I0

2'o 30 FRACTION NUMBER

).1 ""

,,o

Fig. 1. Chromatographic profile of porphobilinogen deaminase on DEAE-cellulose. A desalted ammonium sulfate fraction (70 mg of protein) was applied to a column (2.0 × 22 cm) equifibrated with 5 mM phosphate buffer (pH 7.5). Proteins were eluted with a 0-0.3 M KCI linear gradient in the same buffer. Fractions of 10 ml, at a flow rate of 35 ml/h, were collected.

101

(-} u~

A

B

2.0 > 1£-

_k_

t

45 15 30 SLICE NUMBER Fig. 2. Electrophoresis of native porphobilinogen deaminase. A-purified enzyme was dectrophoresed in polyacrylamide gel, enzyme activity and protein were determined as indicated under Materials and Methods. Bromophenol blue was used as tracking dye ( 1'). B-purified enzyme was preincubated with 0.2 mM porphobilinogen for 10 rain at 37 °C and then subjected to electrophoresis at 4°C. In both runs, the same amount of protein (100 pg) was used.

heat-treated human liver enzyme (Mazzetti, M. and Tomio, J.M., unpublished data). Thus, there appears to.be a partial homology between rat and human liver porphobilinogen deaminase. Further studies to obtain purified human enzyme are in progress.

Molecular weight and subunit composition The M r value for native rat liver porphobilinogen deaminase was estimated to be 41 000 + 4000 by Sephadex G-100 gel filtration when compared with standard marker proteins. Under denaturating conditions using SDS-polyacrylamide gel electrophoresis, a single M r value of 42 000 + 2000 was obtained, indicating that the rat liver enzyme is a monomeric protein. The molecular weight of the isolated enzyme was also estimated by using hydrodynamic parameters. Using the values of the sedimentation coefficient given by sucrose density gradient

centrifugation (S20.w= 3.5 S) and the molecular Stokes radius obtained by gel exclusion (27 A), and considering a partial specific volume of 0.725 cm3/g [23], an .Mr of 42000 was calculated [30] which compares satisfactorily with the value obtained by gel filtration and polyacrylamide gel electrophoresis. The Mr values obtained are very similar to those reported for this enzyme from different sources, i.e., human erythrocytes [9,31], rat spleen [29], spinach leaves [3], R. spheroides [6,7], Chlorella regularis [10] and E. gracilis [11].

Isoelectric focusing Isoelectric focusing of purified enzyme at 4°C revealed the presence of two very close proteins bands, indistinguishable when activity and pH were determined. The pl obtained was 4.9. Acidic predominant forms of porphobilinogen deaminase in mammalian liver were reported by Meisler and Carter [32]. A still more acidic pl was found in other species, i.e., 4.2-4.5 for spinach [3], 4.46 for R. spheroides [7] and 4.2 for C regularis [10]. It is worth noting that the only enzyme showing a more basic p l is the human erythrocyte protein as reported by Anderson and Desnick [6.8-6.2] [9] Meisler and Carter [6.5-5.2] [32] or Miyagi et al. (6.2-5.2) [8]. .4 mino acid composition The results of the amino acid studies of the purified porphobilinogen deaminase after acid hydrolysis are summarized in Table IL Analysis of this composition gave a relatively high proportion (more than 40%) of acidic amino acids, a fact which accounts for the acidic protein pl. The enzyme showed a very low percentage of sulphur-contaitfing aiilino acids (methionine and cysteine or cystine). These results present similarities with those found in the human erythrocyte enzyme but also differences, specially in the contents of tyrosine residues [9]. Spectral analysis The ultraviolet and visible absorption spectra of pure enzyme in 0.1 M phosphate buffer at pH 7.5 showed a typical profile of free protein (maximum 276-278 rim). Thus, there was no evidence of bound cofactors or other compounds, even when searching for stable enzyme-substrate forms.

102 TABLE 1I

Effect of pH

AMINO ACID COMPOSITION OF PORPHOBILINOGEN DEAMINASE FROM RAT LIVER

Dependence of initial velocity with pH showed a classical curve (Fig. 3), wi~h an optimum pH at %5 when both 0.1 M sodium phosphate or 0.05 M Tris-HCi buffers at 37°C were used. Moreover, increased activities of the enzyme were observed in higher ionic strength of TrL,~ buffer (0.1 M) as illustrated in Fig. 3. The liver enzyme was stable in the range pH 7.2-9.0, but no activity was detected below pH 6.2.

Amino acid analysis was carried out as described under Materials and Methods. The results are the average of the values for 20 and 40 h hydrolysates, except as noted below. Amino acid

Residues a moles

Asx Thr b Set h GIx Pro Gly Ala Cys ~ Val Met lie Leu Tyr Phe Trp Lys His Arg

41.7 29.6 33.2 59.2 18.6 41.8 30.6 5.6 20.6 6.7 11.1 30.5 9.7 12.8 n.d. d 18.8 8.4 14.5

Effect of temperature Porphobilinogen deaminase increased with temperature, showing maximum activity in the 55-60 o C range. An 8-fold increase of activity was found over that observed at 37 o C; while at 25 °C it was 50% lower. At temperatures higher than 50°C the yield of porphyrins was significantly increased by chemical polymerization. Applying the Arrhenius equation, a biphasic plot was obtained with a break at 45°C, indicating that a conformational change in the enzyme molecule could take place, favouring the reaction (Fig. 4). Activation energies were estimated to be 62.8

a Residues per 42000 mol. wt. b Values obtained by extrapolation to zero hydrolysis time. Measured as cysteic acid performic oxidation. d n.d,, not determined.

10£

I • Kinetic analyses A hyperbolic dependence between enzyme activity and substrate concentration was always found, giving a typical Michaelis-Menten plot and suggesting a sequential displacement mechanism to bind each of the four porphobilinogen molecules [11]. Under the states conditions the reaction was always linear with respect to time and protein concentration. The K m for its substrate, as measured by uroporphyrin formation, was 17/~M at pH 7.5 in sodium phosphate buffer. This value matches with others from mammalian sources as human erythrocytes. 6 ~,.M [33] and cow fiver, 5 /~M [12], but it is higher than that reported by Piper and van Lier [33] and significantly lower than those for E. gracilis, 70 pM [11], C. regularis, 85/iM [113]or wheat germ, 70 pM [4].

E

!

,=' I

2£ f

&- I ~0 7,0 S

|

pH

80

I

9.0

Fig. 3. The effect of hydrogen ion concentration on porphobilinogen deaminase. Assays were carried out at 3 7 ° C in 0.1 M sodium phosphate buffers (a) or in 0.05 M ( o ) and 0.1 M Tris-HCl buffer (®). Enzyme activity was determined as described under Materials and Methods.

103 light inactivation were found, even after 45 rain of pre-incubation at 55 or 60 ° C. Higher increases in temperature (80°C) led to a progressive loss of activity with time; however, even after 30 rain at 80 ° C, the enzyme still retained 15~g of its normal activity (Fig. 5). These findings are in accordance with those reported by other researchers about the inherent thermal stability of this enzyme from different sources [4,9,10,29].

E U~

;O'C/

O'C

v

> Ot O ,--I

Stability

0 0

,

2.8

I

I

3.0

3.2

1/T ( K'I)(=IO3)

~25"C 3./,

Fig. 4. Arrhenius plot for determination of the activation energy for the enzyme polymerization of porphobilinogen. Activi:ies were measured fluorometricallyafter incubation at the temperatures.Initial velocitieswere alwaysevaluated. k J / m o l (25-45 o C) and 108.8 k J / m o l (45-55 ° C). Thereafter inactivation prevailed. Studies of thermal stability were carried out by pre-incubating porphobilinogen deaminase in the absence of porphobilinogen for various periods of t~me and at different temperatures, as i~dicated in Fig. 5. The results revealed that the rat liver enzyme was very stable, ~ince no loss of activity or

50

o------c-

~

= 4.0

E

~3,C w

1.(

15 30 45 PREINCUBATIONTIME(rain) Fig. 5. Thermal inactivation of rat liver porphobilinogen deaminase. Aliquots of partially purified enzyme were preincubated at 37°C (O), 45°C (A), 65°C (n) and 80°C (o), for different periodsof time. Residual activitieswere then assayed at 37 °C as indicatedin Materials and Methods. Initial activity withoutpreincubationwas 5.0 units/ml.

Partially purified porphobilinogen deaminase (after ammonium sulphate precipitation or heat treatment) can be stored at - 2 0 ° C for at least 6 month with little or no loss of activity. However, over 60~ of the initial activity is lost when left at room temperature for 24 h o Diluted proteh| fractions or very low ionic strength solutions made the enzyme very labile to both, temperature and time.

Effect of sulphydryl reagents The effects of several compounds on porphobilinogen deaminase activity were studied. The activity of partially purified enzyme was measured under standard conditions in the presence of known interacting SH group reagents, such as cysteine, dithiothreitol and glutathione. These reagents show little influence on enzyme liver activity at 1 raM. On the other hand, 0.1 mM 5,5'-dithiobis(2-nitrobenzoate), N-ethylmaleimide and iodoacetamide incubated at 3 7 ° C had slight or no effect on porphobilinogen deaminase activity either, whereas p-chloromercuribenzoate provoked an inhibition of about 40% only at 0.1 mM that could not be overcome by the subsequent addition of 0.1 or 1 mM dithiothreitol. The results reported here indicate that rat liver porphobilinogen deaminase is a monomeric enzyme, sharing many physicochemical properties with porphobilinogen deaminase from other sources. Thus, similar M r catalytic properties such as hyperbolic dependence of activity on substrate concentration, and thermal stability were found. An exception was the more acidic isoelectric point for the rat liver enzyme. Moreover, as with the other enzymes, the rat liver porphobilinogen deaminase does not contain a chromophoric prosthetic group nor is there evidence of cofactor or

104

metal requirement for maximal activity, although stable-substrate complexes can be obtained when it reacts with porphobilinogen [8].

Acknowledgements We wish to thank Dr. C.E.M. Wolfenstein for the amino acid analysis. This research was supported by grants from Consejo Nacional Investigaciones Cientificas y T~nicas and Secretaria Ciencia y T&nica, Universidad de Buenos Aires.

References 1 Battersby, A.R., Fookes, C.J.R., Matcham, G.W.J, McDonald, E. and Gustafson-Potter, K.E. (1979) J. Chem. Soc. Chem. Commun. 316-319. 2 Kappas, A., Sassa~ S. and Anderson, K.E. (1983) in The Metabolic Basis of Inherited Disease (Stanbury, J.B., Wyngaarden, J.B., Fredrickson, D.S., Goldstein, J.L. and Brown, M.S., eds.), pp. 1301-1384, McGraw-Hill. New York. 3 I-liguchi, M. and Bogorard, L. (1975) Ann. NY Acad. Sci. 244, 401-408. 4 Frydman, R.B. and Frydman, B. (1970) Arch. Biochem. Biophys. 136, 193-202. 5 Llamblas, E.B.C. and Badle, A.M. dei C. (1971) Biochim. Biophys. Acta 227, 180-191. 6 Jordan, P.M. and Shemin, D. (1973) J. Biol. Chem. 248, 1019-1024. 7 Davies, R.C. and Neuberger, A. (1973) Biochem. J. 133, 471-492. 8 Miyagi, K., Kaneshima, M., Kawakami, J., Nakada, F., Petrika, Z.J. and Watson, C.J. (1979) Proc. Natl. Acad. Sci. USA 76, 6172-6276. 9 Anderson, P.M. and Desnik, R.J. (1980) £ Biol. Chem. 255, 1993-1999. I0 Shioi, Y, Nagamine, M., Kuroki, M. and Sasa, T. (1980) Biochim. Biophys. Acta 616, 300-309. 11 Williams, D.C., Morgan, G.S, McDonald, E. and Battersby, A.R. (1981) Biochem. J. 193, 301-310.

12 Sancovich, H.A., Battle, A.M.C. and Grinstein, M. (1969) Biocitim. Biophys. Acta 191, 130-143. 13 Piper, W.N. and van Lier, R.B.L. (1977) Mol. Pharmacol. 13, 1126-1135. 14 Grandchamp, B., Romeo, P.H, Dubart~ A., Raich, N., Rosa, £, Nordmann, Y. and Goossens, M. (1984) Proc. Natl. Acad. Sci. USA 81, 5036-5040. 15 Romeo. P.H., Raich, N., Dubart, A., Beaupair, D., Mattei, M.G. and Goosen~, M. (1986) in Porphyrins and Porphyrias (Nordmann, Y., ed.), pp. 25-34, Jolm Libbey, Paris. 16 Strand~ L.J., Meyer, Y.A., Felsher, B.F., Redeker, A.G. and Marver, H.S. (1972) J. Clin. Invest. 51, 2530-2536. 17 Maines, M.D., Janousek, V., Tomio, J.M. and Kappas, A. (1976) Proc. Natl. Acad. Sci. USA 73, 1499-1503. 18 Schwartz, S., Berg, M.H., Bossenmaier, I. and Dismore, J. (1960) Methods Biochem. Anal. 8, 221-293. 19 Lowry, O.H, Rosebrough, N., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275. 20 Andrews, P. (1964) Biochem. J. 91, 222-233. 21 Laemmli, U.K. (1970) Nature (Lood.) 222, 680-685. 22 Davis, B.J. (1964) Ann. NY Acad. Sci. 121,404-427. 23 Martin, R.G. and Ames, B.N. (1961) J. Biol. Chem. 236, 1372-1379. 24 Laurent, T.C. and Killander, J. (1964) J. Chromatogr. 14, 304-316. 25 Wolfenstein, C.E.M., Santomd, J.A. and Paladini, A.C. (1966) Acta Physiol. Lat. Am. 16, 194-202. 26 Hits, C.H.W. (1956) J. Biol. Chem. 219, 611-621. 27 Masters, B.S.S., Baron, J., Taylor, W.E., Isaacson, E.L. and LoSpalluto, J. (1971) J. Biol. Chem. 246, 4143-4150. 28 Ouchterlony, O. (1964) Coll. Ges. Physiol. Chem. 15, 13-35. 29 Williams, D.C. (1984) Biochem. J. 217, 675-683. 30 Siegel, L.M. and Monty, K.J. (1966) Biochim. Biophys. Acta 112, 346-362. 31 Brown, R.C, Elder, G.H. and Urquhart, A.J. (1985) Biochem. Soc. Trans. 13, 1227-1228. 32 Meisler, M.H. and Carter, M.L.C. (1980) Proc. Natl. Acad. Sci. USA 77, 2848-2852. 33 Kreimer-Bimbaum, M. and Tomio, J.M. (1975) in Porphyrin in Human Diseases (Doss, M., ed.), pp. 182-188, S. Karger, A.G., Basel. 34 Piper, W.N. and Van Lier, R.B.L. (1976) Life Sci. 19, 1225-1234.