Saccharomyces cerevisiae porphobilinogenase: Some physical and kinetic properties

Comp. Biochem. Physiol. Vol.92B,No. 2, pp. 297-301, 1989 Printed in Great Britain

0305-0491/89$3.00+ 0.00 PergamonPress plc

SACCHAROMYCES CEREVISIAE PORPHOBILINOGENASE: SOME PHYSICAL AND KINETIC PROPERTIES* LIDIA SUSANAARAUJO, MARIAELISALOMBARDO,MARIAVICTORIAROSSETTIand ALCIRA M. DEL C. BATLLEt Centro de Investigaciones sobre Porfirinas y Porfirias, CIPYP (CONICET-FCEN, UBA), Ciudad Universitaria, Pabell6n II, 2do. Piso, 1428 Buenos Aires, Argentina (Received 15 March 1988) Abstract--l. Properties of porphobilinogenase (PBGase), the enzyme complex converting porphobilinogen (PBG) into uroporphyrinogens, were studied in a wild strain, D273-10B and a mutant, B231, of Saccharomyces cerevisiae. 2. A well-definedmaximum of enzyme activity was observed for the mutant strain after 20 hr of growth; whilst the activity in the wild strain did not vary significantlyduring growth. 3. Neither PBG consumption nor uroporphyrinogen formation were modified by the presence of air either in the wild or in the mutant strain. 4. In both the wild and mutant strains uroporphyrinogen formation increased linearly with both protein concentration and incubation time. 5. The addition of a mixture of sodium and magnesium salts to the assay system inhibited the enzyme activity of both strains by 50% without modifying the isomer composition. 6. The same optimum pH (7.4) and mol. wt (50,000 + 5000) was found for the enzyme from both strains. 7. The enzyme from both the wild and mutant strains shows Michaelis-Menten kinetics when isolated from cells at either the exponential or the stationary phases of growth. Accumulation of porphyrins and ~-aminolevulinic acid occurring during the exponential phase in the mutant strain, did not modify the kinetics.

INTRODUCTION The biosynthetic pathway leading to protohaem formation plays an important role in cellular metabolism and biogenesis of mitochondria (Sanders et al., 1973). In yeast, the final product is utilized as the prosthetic group of a variety of haemoproteins. The pathway also provides the precursor for the synthesis of sirohaem, a modified uroporphyrin and the prosthetic group of sulphite reductase. On the other hand, protohaem has been shown to directly or indirectly control the synthesis, proteolitic maturation and/or assembly of the protein moieties of many either mitochondrial or cytosolic haemoproteins (LabbeBois et al., 1983). Therefore it appears important to know how protohaem synthesis is regulated in yeast. Moreover, taking into account that this sequence of enzymatic events has been found to be identical in all organisms and tissues so far tested; the study of haem synthesis and its regulation in this simple and well-known eukaryotic organism can contribute to a greater understanding of this pathway. This research would benefit from the utilization of a mutant strain of Saccharomyces cerevisiae blocked or de-regulated in haem production, which can be used as an excellent model system to study both the *Dedicated to Professor Claude Rimington FRS, on occasion of his 85th birthday. "tAll correspondence should be addressed to: Professor Alcira Batlle, Viamonte 1881-10 "A", 1056 Buenos Aires, Argentina.

regulatory events involved in haem synthesis and their mechanisms. Yeast mutants affected in some steps of haem formation have been described, and their study has already provided valuable information (Mattoon et al., 1978; Urban-Grimal and Labbe-Bois, 1981; Bilinski et al., 1981; Arrese et al., 1983; Labbe-Bois et al., 1983; Rytka et al., 1984). Therefore, the study of porphyrin biosynthesis in this class of mutants is one of the areas that has recently attracted our interest. On the other hand, a great deal of experience has been gained in our laboratory in the last 25 years, about the enzymic conversion of porphobilinogen (PBG) into uroporphyrinogens (urogens) (Rossetti et al., 1987 and Refs therein) and although a considerable body of information exists about this step of porphyrin biosynthesis as well as the enzymes concerned (porphobilinogenase (PBGase) and deaminase) in several animal, plant and bacterial sources, very little is known about them in S. cerevisiae. Therefore, to continue our studies of the isolation, purification and properties of these enzymes, to gain more insight into the mechanisms by which haem formation is regulated and also with the aim of elucidating the possible nature of the underlying defects in human porphyrias (Rytka et al., 1984), we have undertaken enzymic studies in a wild-type S, cerevisiae (strain D273-10B) and a mutant (B231) which accumulates substantial amounts of 6-aminolevulinic acid (ALA), octacarboxylic porphyrins and less carboxylated porphyrins to a lesser degree. This

297

298

LID1A SUSANA ARAUJO et al.

pattern of porphyrin intermediate accumulation might be due either to a defect in uroporphyrinogen decarboxylase ( U R O - D ) activity, or to an enhancement in PBGase or both. Therefore, preliminary studies on the enzyme complex synthesizing urogens from PBG, were carried out in both yeast strains, and the results obtained are reported here. MATERIALS AND METHODS

PBG was obtained according to Sancovich et al. (1970), from ALA (Sigma Chemical Co., USA) and bovine liver dehydratase (ALA-D) and was estimated by the method of Moore and Labbe (1964). Sephadex gets were obtained from Pharmacia Fine Chemicals Uppsala, Sweden. Tris-HCl and sodium phosphate buffers were used throughout this study, unless indicated. All other reagents employed were of AR grade obtained from several commercial sources. All solutions were made up in ion-free, three-times distilled water. Strains D273-10B and B231 strains of Saccharomyces cerevisiae were kindly supplied by Professor J. Mattoon of the University of Colorado. B231 strain has two key properties: a strong fluorescence and a lysine requirement, which were used to repurify and test it occasionally. Media and growth conditions The growth medium was as follows: 1% Yeast Extract (Difco); 2% Peptone (Difco) and 2% Dextrose (Merck). In agar plates 2% Bacto-agar (Difco) was added. Cells were grown at 3ffC with vigorous magnetic stirring. Preparation of cell~[?ee extracts At the end of the exponential phase, unless otherwise stated, cells were harvested by centrifugation at 12,000 g for 10 min, washed with cold distilled water or 0.05 M phosphate buffer, pH 8.0 and broken by ultrasonic treatment to obtain cell-free extract (H) as shown below. Estimation of enzyme activity The standard incubation system contained, unless otherwise specified, the enzyme preparation (usually 2 ml) together 0.05M sodium phosphate buffer, pH 8.0: 60#g of PBG, in a final volume of 3 ml. Incubation was carried out, unless indicated, aerobically, in the dark, with mechanical shaking at 37°C for 2 hr. Blanks were always run with enzyme without PBG, and with PBG without enzyme; therefore endogenous porphyrins and non-enzymically synthesized porphyrins were taken into account for enzyme activity calculation. After incubation, HC1 (c) was added to a final concentration of 5% (w/v) to inactivate the protein; the mixture was then illuminated with white light for 20 rain to oxidize the porphyrinogens formed, the precipitate was separated by centrifugation and total porphyrins were determined in acid solution (Rimington, 1960). Fractionation, identification and quantitative determination of the porphyrins was performed by the procedures already described (Rossetti et al., 1986). Protein concentration was determined by using the Folin Ciocatteau reagent (Lowry et al., 1951). Enzyme unit The enzyme unit was defined as the amount of enzyme that catalizes the formation of 1 nmol of uroporphyrinogen/2 hr, from PBG under the standard incubation conditions. Specific activity is expressed as the number of units of enzyme/mg of protein.

Determination of mol. wt Molecular wt measurements were performed by gel filtration. A Sephadex G-100 column was prepared and calibrated with protein standards of known mol. wt. Protein content and enzyme activity was determined in all column eluates.

RESULTS Optimal assay conditions Jbr PBGase For both strains the shape of the pH curve, as well as the optimum p H (7.4) were the same in both Tris-HC1 and phosphate buffer, however PBGase was 15% less active in the former. It was found that urogen formation was linear with protein concentration over a wide range and with incubation time up to 2 hr. For both strains the activity was assayed at 37~C and the amount of porphyrins formed was the same either in the absence or presence of air. A well defined maximum of enzyme activity was observed for the mutant strain after 20 hr of growing, whilst the activity in the wild strain did not significantly vary during growth; therefore, in both strains enzyme activity was measured in cells harvested at late log phase of growth (20-22 hr). Effect o f p H on enzyme stability For the characterization of an enzyme, knowledge about its pH stability is very important. We have obtained the pH stability profile by pre-incubating the enzyme at the indicated pHs for a time equal to the incubation time used for activity measurements and then activity was determined at the optimum pH. Figure 1 shows that when both strains were preincubated at the optimum pH about 10% of the enzyme units were lost; at higher pHs the loss of enzymic units was greater, while at lower pHs they remained approximately constant. Protein content also significantly decreased at these pHs. In Fig. 2, the upper curve represents the effect of pH on PBGase activity (Araujo et al., 1987); while the lower curve shows the actual effect of pH on the enzyme stability. These results indicate that diminished activity at pHs below or above optimum is largely due to reversible changes in the ionization state of this enzyme. Therefore PBGase from both strains was very stable over a wide range of pH; furthermore, the isomer composition and porphyrin formation was the same independently of the pre-incubation pH. On these grounds, isoelectric precipitation was used as one of the steps to obtain a partially purified enzyme preparation. Effect of sodium and magnesium salts When activity of PBGase from both wild and mutant yeast strains was measured in the presence of sodium and magnesium salts (Table 1), at the same concentrations reported to be activating (Sancovich et al., 1969; Llambias and Batlle, 1971a) or to have not effect (Llambias and Batlle, 1971b; Fumagalli et al., 1982) and expressed on the basis of porphyrin formation, it was found to be 50% lower than the controls without adding any cations. However, isomer composition was not altered.

S. cerevisiae porphobilinogenase

299

(a)

(b)

90

90

.~ 80

80

;

70

"6

u

E

70

o....I/° o~.), ~

60 /~

60

I

i

I

i

6

7

8

9

. ;,I

6

I

I

I

7

8

9

pH

Fig. 1. Effect of pH on enzyme activity. Aliquots of supernatant from cells of wild (a) and mutant (b) strains were pre-incubated at different pHs during 30 min after which inactivated protein was discarded by centrifugation, the pH of the supernatant was adjusted to the optimum pH and incubated under the standard conditions described in the text. Enzyme activity (O) and protein content (O) were determined. The activity of the supernatant from centrifugation at 24,000g, without any treatment, was taken as 100%. Effect o f substrate concentration

When plots of velocity vs PBG concentration were analysed for PBGase, it was found that by measuring the rate of the reaction on the basis of total urogen formation, saturation curves for wild and mutant strains followed classical Michaelis-Menten kinetics, both at the exponential and the stationary phase of growth. Saturation was reached at a PBG concentration of about 70-90/~M. The same profile was obtained when velocity was expressed on the basis of the amount of PBG consumed. Therefore, double reciprocal plots (Figs 3 and 4) were linear and from these plots apparent Km values of 20.01 and 14.26 gM were obtained for the wild and mutant strain respectively, harvested at the late exponential phase. Km values for cells harvested at the stationary phase were of the same order, but Vma~values were about 40% lower for both strains, perhaps as a consequence of the reduced levels of active protein at this phase of growth. We have also found that neither Km values nor the profiles of the saturation curves were modified by endogenous ALA and porphyrins present in the mutant strain (the effect of both compounds was similar). D273 - 10B strain

B231 strain

11)0

i'\

50

/ 0

-//

\

g

i

i

i

I

6

7

8

9

I

Table 1. Effect of sodium and magnesium salts on enzyme activity CINa-CI2 Mg (ml) -0.075 0.15 0.30

Fig. 2. Effect ofpH on activity and stability of PBGase. (O) Activity was measured at the pHs indicated, without any pre-incubation, under the standard incubation conditions. (©) The enzyme was pre-incubated at the indicated pHs for 30 min and then activity was measured at the optimum pH under the standard incubation conditions. The activity of the supernatant pre-incubated at optimum pH was taken as 100%.

0.42 0.37 0.30 0.21

0.83 0.66 0.51 0.39

The supernatant from centrifugation at 24,000g was used as the enzyme preparation. To the standard incubation system a mixture of CINa 0.6 M - C I 2Mg 0.12 M (1:1) was added as indicated, in a final volume of 3 ml. The rest of the experimental conditions were as detailed in the text.

The empirical Hill equation: log(v / Vmax -- V) = n l o g S - logK is very often used to determine the strength of interaction between ligand-binding sites. Hill plots for PBGase of wild and mutant strains resulted in a slope of 1, both at the exponential and stationary phase of growth, indicating the existence of only one class of binding site for PBG per molecule of enzyme (Figs 3 and 4 inset).

T~

,~ 0.4 n=1.3/ ~ 0 >__ 0.2

/~"///0~/ Km=20.01g.M

og ( PBG )

/>z2.

f //

J -15

Km= 20.23 ~M

.

0~

I

pH

Specific activity D273-10B B231

-lO

-5

ol 0

I

5

I / ( P B G ) M x lO-2

Fig. 3. Double reciprocal plots on velocity against PBG concentration. Cells of wild strain harvested at the exponential (©) and stationary (Q) phase of growth were used. Supernatant from centrifugation at 24,000g was used as the enzyme preparation. Activity was measured as described in the text, with varying concentrations of PBG. The insert shows the Hill plot.

300

LIDIA SUSANA ARAUJO et al.

strains; no aggregates or lower mol. wt species were detected. ~ 9.0 ,-~ 0.6

o.4

i

o=o/

6.0

E

0.2

~ t n=l

0

IO.4, ~ 1.0 1.5 log ( PBG )

DISCUSSION Km = 18.52 ~M

/

Krn= 12.66 ~.M

)./o / Km= 14.26/,~M I

_i~5

_i J0

--5

0

5 1 / (PBG)M x 10-:

Fig. 4. Double reciprocal plots of velocity against PBG concentration. Cells of mutant strain harvested at the exponential (O) and stationary (O) phase of growth were used. Supernatant from cenlrifugation at 24,000g was used as the enzyme preparation. Activity was measured as described in the text, with varying concentrations of PBG, in the presence (O) and absence (D) of endogenous compounds (ALA or porphyrins). ALA was removed by dialysis during 24 hr, against 0.05 M sodium phosphate buffer, pH 8.0. To remove endogenous porphyrins Dowex l-X8 resin was used. The insert shows the Hill plot. Molecular wt determination

A 30-fold purified PBGase preparation obtained after ultracentrifugation (140,000g 45min) and isoelectric precipitation (range of pH: 5.5~4.2) of a 24,000g supernatant, was applied to a Sephadex G-100 column previously calibrated for tool. wt determination. Figure 5 shows the elution diagram for PBGase partially purified from the wild strain of S. cerecMae. The profile obtained for the mutant strain was the same. A sharp peak of tool. wt 50,000 _+ 5000 was found for the enzyme from both

1.6

40

1.2

30

"~ o.8

20 "~

< 0.4

~

/ i - ~ 40

/ 1.0

i 80 120 Elutionvolume( ml )

I~)0

Fig. 5. Elution profile on Sephadex G-100 of a 30-fold purified enzyme. A Sephadex G-100 calibrated column (2.4 x 50 cm) equilibrated and eluted with 0.05 M sodium phosphate buffer, pH 8.0 was used. Protein content (O) and porphyrins formed (O) were determined in column eluates. Methodology was as described in the text.

It is well known that in some organisms, the activity of the enzymes of haem synthesis is significantly influenced by the days of growing (Wider de Xifra et al., 1971; Batlle et al., 1975): when we studied the effect of growth on PBGase activity, we found that in the wild strain it was not dependent on the time of growth, whilst the activity profile in the mutant showed a maximum after about 20 hr growth (late log phase). On the other hand, it has also been reported that PBGase and deaminase were equally active in the absence or presence of oxygen (Fumagalli et al., 1982 and references therein); but it has also been observed that the activity of the soybean callus and Euglena gracilis enzymes (Llambias et al., 1971a; Rossetti and Batlle, 1977) was significantly affected by air and urogens were only formed under anaerobic conditions, although PBG consumption was not seriously modified by air. In the present case, for S. cerevisiae PBGase, the amount of porphyrins formed, their isomer composition and PBG consumption, were the same either in the absence or the presence of air. The pH maximum found for S. cerevisiae PBGase is within the known limits reported for this enzyme in other sources (Rossetti et al., 1987) but when the effect of pH on the ionization state of the enzyme was investigated, it was found that PBGase from both strains seemed to be very stable over a wide range of pH and changes produced by pH on its ionization state were reversible. Llambias and Batlle (1971a), Sancovich et al. (1976) and more recently Clement et al. (1982) have reported some activating effect on activity by the presence of certain concentrations of sodium and magnesium salts, probably due to some sort of association~zlissociation phenomenon. Moreover, Cornford (1964) also observed that changes in salt composition greatly influenced porphyrin synthesis from PBG. However, we found that PBGase activity from both strains of S. cerevisiae was 50% lower in the presence of these salts, but as we had observed that certain ions might or might not stimulate activity depending on whether the concentration of PBG used is below, above or at saturating levels (Juknat, 1983), to actually know what the effect of sodium and magnesium ions might be, further experiments should be performed with different substrate concentrations. Moreover, for both strains, PBGase showed a Michaelis Menten kinetic behaviour, either at an exponential or stationary phase of growth: therefore only one class of binding site for PBG per molecule of enzyme might exist, at variance with the results reported for this enzyme in other sources (Llambias and Batlle, 1970, 1971b; Rossetti et al., 1987; Juknat et al., 1989). Finally, a sharp peak of tool. wt 50,000 _+ 5000 was also found for the enzyme from both strains, in good agreement with the values reported for the bovine liver (Sancovich et al., 1976); pig liver (Fumagalli et

S. cerevisiae porphobilinogenase al., 1982) and Euglena gracilis enzymes (Rossetti et al., 1986). It must be noted that no evidence for PBGase species smaller than a mol. wt of 50,000 was obtained for either the wild or the mutant strain of S. cerevisiae as has been observed for the E. gracilis enzyme (Rossetti et al., 1986). Acknowledgements--M. V. Rossetti and A. M. del C. Batlle hold posts as Scientific Researchers in the Argentine National Research Council (CONICET). M. E. Lombardo is a Research Fellow from the University of Buenos Aires. This work was supported by grants from the CONICET, the SUBCYT, Secretaria de Salud POblica del Ministerio de Bienestar Social, Banco de la Naci6n Argentina and Banco de Galicia y Buenos Aires. This paper forms part of the Doctoral Thesis of L. S. Araujo to be presented to the University of Buenos Aires for her Ph.D. Degree. We wish to express our heartfelt thanks to Professor J. Mattoon, from the Department of Biochemistry, University of Colorado, USA, for generously providing us with all the strains of S. cerevisiae and above all for his continued interest and encouragement. The technical assistance of Mrs B. Riccillo de Aprea is gratefully acknowledged.

REFERENCES

Araujo L. S., Lombardo M. E., Rossetti M. V. and Batlle A. M. del C. (1987) Porphobilinogenase activity in a wild strain and its haem mutant of Saccharomyces cerevisiae. Revta argent. 19, 109-120. Arrese M. R., Carvajal E., Robison S., Sambunaris A., Panek A. and Mattoon J. R. (1983) Cloning of the 6-aminolevulinic acid synthase structural gene in yeast. Curr. Genet. 7, 175 183. Batlle A. M. del C., Llambias E. B. C., Wider de Xifra E. A. and Tigier H. A. (1975) Porphyrin biosynthesis in the soybean callus tissue system--XV. The effect of growth conditions. Int. J. Biochem. 6, 591-606. Bilinski T., Litwinska J., Lukaszkiewicz J., Rytka J., Simon M. and Labbe-Bois R. (1981) Characterization of two mutant strains of Saccharomyces cerevisiae deficient in coproporphyrinogen 3-oxidase activity. J. gen. Microbiol. 122, 79-87. Clement R. P., Kohashi M. and Piper W. (1982) Rat liver purification uroporphyrinogen III cosynthetase. Archs Biochem. Biophys. 214, 657~63. Cornford P. (1964) Transformation of porphobilinogen into porphyrins by preparations from human erythrocytes. Biochem. J. 91, 64-73. Fumagalli S. A., Rossetti M. V., Juknat de Geralnik A. A., Kotler M. L. and Batlle A. M. del C. (1982) Estudios sobre la PBGasa de higado de cerdo. An. Asoc. quire. argent. 70, 375-382. Juknat A. A. (1983) Biosintesis de porfirin6genos. Ph.D. Thesis, University of Buenos Aires. Juknat A. A., Kotler M. L., Koopmann G. E. and Batlle A. M. del C. (1989) Porphobilinogenase from Rhodopseudomonas palustris. Comp. Biochem. Physiol. 92B, 291-295. Labbe-Bois R., Urban-Grimal D., Volland C., Camadro J. M. and Dehoux P. (1983) About the regulation of protoheme synthesis in the yeast Saccharomyces cerevisiae. Mitochondria 523, 534. Llambias E. B. C. and Batlle A. M. del C. (1970) Porphyrin biosynthesis in soybean-callus--V. The porphobilinogen deaminase-uroporphyrinogen cosynthetase system. Kinetic studies. Biochim. biophys. Acta 220, 552-559.

301

Llambias E. B. C. and Batlle A. M. del C. (1971a) Studies on the porphobilinogen deaminase-uroporphyrinogen cosynthetase system of cultured soybean cells. Biochem. J. 121, 327-340. Llambias E. B. C. and Batlle A. M. del C. (1971b) Porphyrin biosynthesis--VIII. Avian erythrocyte porphibilinogen deaminase-uroporphyrinogen III cosynthetase. Its purification, properties and the separation of its components. Biochim. biophys. Acta 227, 180-191. Lowry O. H., Rosebrough N. J., Farr A. L. and Randall R. J. (1951) Protein measurement with the Folin phenol reagent. J. biol. Chem. 193, 265-275. Mattoon J. R., Malamud D. R., Brunner A., Braz C., Carvajal E., Lancashire W. E. and Panek A. D. (1978) Regulation of heme formation and cytochrome biosynthesis in normal and mutant yeast. In Biochemistry and Genetics of Yeast, Pure and Applied Aspects (Edited by Bacila M., Horecker B. and Stoppani A. O. M.), pp. 317-337. Academic Press, New York. Moore D. and Labbe R. (1964) Assays for ALA and PBG determination. Clin. Chem. 10, 1105-1109. Rimington C. (1960) Spectral-absorption coefficients of some porphyrins in the Soret-band region. Biochem. J. 75, 62tY623. Rossetti M. V. and Batlle A. M. del C. (1977) Polypirrol intermediates in porphyrin biosynthesis. Studies with Euglena gracilis. Int. J. Biochem. 8, 277-283. Rossetti M. V., Lombardo M. E., Juknat de Geralnik A. A., Araujo L. S. and Batlle A. M. del C. (1986) Porphyrin biosynthesis in Euglena gracilis--V. Soluble and particulate PBG-ase. Comp. Biochem. Physiol. 85B, 451-458. Rossetti M. V., Araujo L. S., Lombardo M. E., Correa Garcia S. and Batlle A. M. del C. (1987) Porphyrin biosynthesis in Euglena gracilis--VI. The effect of growth conditions on porphobilinogenase activity and further properties. Comp. Biochem. Physiol. g/B, 593~00. Rykta J., Bilinski T. and Labbe-Bois R. (1984) Modified uroporphyrinogen decarboxylase activity in yeast mutant which mimics porphyria cutanea tarda. Biochem. J. 218, 405-413. Sancovich H. A., Batlle A. M. del C. and Grinstein M. (1969) Porphyrin biosynthesis--VI. Separation and purification of porphobilinogen deaminase and uroporphyrinogen isomerase from cow liver. Porphobilinogenase as an allosteric enzyme. Biochim. biophys. Acta 191, 130-143. Sancovich H. A., Ferramola A. M., Batlle A. M. del C. and Grinstein M. (1970) Preparation of porphobilinogen. In Methods in Enzymology (Edited by Tabor H. and Tabor C. W.), Vol. XVII A, pp. 220-222. Academic Press, New York. Sancovich H. A., Ferramola A. M., Batlle A. M. del C., Kivilevich A. and Grinstein M. (1976) Studies on cow liver porphobilinogen deaminase. Acta physiol, latinoam. 26, 376-386. Sanders H. K., Mied P. A , Briquet M., Hern/mdezRodriguez A., Gottal R. F. and Mattoon J. R. (1973) Regulation of mitochondrial biogenesis: yeast mutants deficient in synthesis of ~-aminolevulinic acid. J. molec. Biol. 80, 17-39. Urban-Grimal D. and Labbe-Bois R. (1981) Genetic and biochemical characterization of mutants of Saccharomyces eerevisiae blocked in six different steps of heme biosynthesis. Molec. Gen. Genet. 183, 85-92. Wider de Xifra E. A., Batlle, A. M. del C. and Tigier H. A. (1971) 6-Aminolaevulinate synthetase in extracts of cultured soybean cells. Biochim. biophys. Acta 235, 511-517.