ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 227, No. 1, November, pp. 2’79-286, 1983

Biosynthesis of Phycocyanobilin from Exogenous Labeled Biliverdin in Cyanidium caldarium

SAMUEL I. BEALE’ AND JUAN CORNEJO

L?iwision of Biology and Medicine, Brown University, Prcrdence, Rhode Island 0.2912

Received June 3, 1983

Phycocyanin is a major light-harvesting pigment in bluegreen, red, and cryptomonad algae. This pigment is composed of phycocyanobilin chromophores covalently attached to protein. Phycocyanobilin is an open-chain tetrapyrrole structurally close to biliverdin. Biliverdin is formed in animals by oxidative ring-opening of protoheme. Recent evidence indicates that protoheme is a precursor of phycocyanobilin in the unicellular rhodophyte, Cyanidium caldarium. To find out if biliverdin is an intermediate in the conversion of protoheme to phycocyanobilin, [14C]biliverdin was administered along with N-meth- ylmesoporphyrin IX (which blocks endogenous protoheme formation) to growing cells of C. caldarium. To avoid phototoxic effects due to the porphyrin, a mutant strain was used that forms large amounts of both chlorophyll and phycocyanin in the dark. After 12 or 24 h in the dark, cells were harvested and exhaustively extracted to remove free pigments. Next, protoheme was extracted. Phycocyanobilin was then cleaved from the apoprotein by methanolysis. Protoheme and phycocyanobilin were purified by solvent partition, DEAE-Sepharose chromatography, and preparative reverse-phase high- pressure liquid chromatography. Absorption was monitored continuously and fractions were collected for radioactivity determination. Negligible amounts of label appeared in the protoheme-containing fractions. A major portion of label in the eluates of the phycocyanobilin-containing samples coincided with the absorption peak at 22 min due to phycocyanobilin. In a control experiment, [14C]biliverdin was added to the cells after incubation and just before the phycocyanobilin-apoprotein cleavage step. The major peak of label then eluted with the absorption peak at 12 min due to biliverdin, indicating that during the isolation biliverdin is not converted to compounds coeluting with phy- cocyanobilin. It thus appears that exogenous biliverdin can serve as a precursor to phycocyanobilin in C. caldarium, and that the route of incorporation is direct rather than by degradation and reincorporation of 14C through protoheme.

Phycocyanin is a major light-harvesting pigment in bluegreen, red, and crypto- monad algae. The pigment is composed of phycocyanobilin chromophores covalently attached to protein. Phycocyanobilin is an open-chain tetrapyrrole ((1); Fig. 1). Structural considerations suggest that phycocyanobilin may be formed by oxi- dative ring-opening of a porphyrin or me- talloporphyrin in analogy with biliverdin

1 To whom correspondence should be addressed.

formation from protoheme (2). Recent ev- idence (3-5) indicates that protoheme is a biosynthetic precursor of phycocyanobilin. Because biliverdin is the direct product of protoheme macrocycle ring-opening in an- imal heme catabolism, it might also be an intermediate in the conversion of proto- heme to phycocyanobilin.

To find out if biliverdin can be converted to phycocyanobilin in Go, 14C-labeled bili- verdin was administered to growing cells of Cyanidium caldarium in the presence of N-methylmesoporphyrin IX, which

279 0003-9861/83 $3.00 Copyright 0 1983 by Academic Press, Inc. All rights of reproduction in any form reserved.

280 BEALE AND CORNEJO

7”’ F”’

FH2

COOH

yH2

COOH

CH=CH2

7”’ 7’2

7”’ 7”’

COOH COOH

Protoporphyrin IX Protoheme

cI&y ( %fi=&& CH2 CH2 7”’

7”’ CHz I

tH2 fH2 y”2

COOH COOH COOH COOH

Phycocyanobilin Biliverdln IXcf.

FIG. 1. Proposed route of phycocyanobilin formation through protoheme and biliverdin.

partially blocks endogenous protoheme formation from protoporphyrin IX (3). [14C]Biliverdin was taken up by the cells and it contributed significant label to phy- cocyanobilin, but not to protoheme, indi- cating a direct conversion of biliverdin to phycocyanobilin.

MATERIALS AND METHODS

Growth of algal celLa Axenic cultures of C. culdarium were grown in a glucose-based medium at 42°C in complete darkness as previously described (3). Initial

stocks of wild-type cells were a generous gift from

R. F. Troxler (Boston University School of Medicine). For all experiments reported, strain CPD was used. This strain arose in our laboratory as a spontaneous mutant and was isolated from a single cell clone. It forms large amounts of both chlorophyll a and phy-

cocyanin equally well in the light or dark (3). Dark culturing was necessary because the added porphyrin (see below) could render the cells photosensitive. Un- der our growth conditions, cell number doubled in 11

to 12 h and reached a maximum of 3 X lO’/ml. Incubation conditions. Experimental cultures were

grown from dilute suspensions and treatment began at cell densities selected so that at the end of the

PHYCOCYANOBILIN BIOSYNTHESIS FROM BILIVERDIN 281

treatment period, 12 or 24 h later, the final cell density

would be 2 X lO*/ml. Culture volumes were 65 ml.

At the beginning of the treatment period, [14C]biliverdin (8 X 10’ cpm, 100 nmol) was added in

60 ~1 of dimethyl sulfoxide (final concentration O.l%), and N-methylmesoporphyrin IX (final concentration

10 pM) was added in 350 ~1 of 0.1 N HCl. N-Methyl- mesoporphyrin IX was previously shown to partially

block protoheme formation and to cause the accu-

mulation of protoporphyrin IX in these cells (3). Sam- ples were taken periodically to monitor uptake of the

[i4C]biliverdin. Extraction of free pigments. At the end of the in-

cubation period, cells were harvested by centrifugation and washed twice with lo-ml portions of water. Next,

the pigments were extracted six times as follows: The cells were suspended in 3 ml of dimethyl sulfoxide,

and then 10 ml of acetone:l.25 M aqueous ammonia

(99:1, v/v) was added with mixing. The mixture was centrifuged, and the supernatant was removed. The

last two extractions yielded completely colorless su- pernatants.

Extraction of protoheme. After free-pigment ex- traction, cells were suspended in 3 ml of dimethyl sulfoxide, and 0.5 ml of 12 N HCl was added with

mixing, followed immediately by 10 ml of ice-cold acetone. After 15 min, the cells were removed by cen-

trifugation. The above extraction was repeated twice

more using half-volumes of all liquids. The combined supernatants were mixed with 2’7 ml of methylene chloride, then 50 ml of water was added to achieve

phase separation, and the bottom phase (containing protoheme in methylene chloride) was washed with

five 20-ml portions of water. An equal volume of 95% ethanol was added to the methylene chloride and the solution was applied to an g-mm-diameter X lo-mm-

high column of DEAE-Sepharose CL-6B (acetate form), prepared according to (6). After the column

was washed with 20 ml of 95% ethanohmethylene chloride (l:l, v/v), protoheme was eluted with 1 ml

of 95% ethanol:glacial acetic acid:water (92:1:7, V/V/V).

Extraction of phycocyanobilin After protoheme ex-

traction, the cells were washed with three 15-ml por- tions of water, and suspended in 10 ml of 1% tri- chloroacetic acid for 1 h in the dark (7). Next, the

cells were washed with three lo-ml portions of ab- solute methanol. Then the cells were supsended in 10 ml of absolute methanol containing 10 mM HgCl* (8)

and 1.5 mM butylated hydroxytoluene. Cells were kept suspended by shaking for 20 h at 42°C in the dark.

Cells were then separated from the blue phycocy- anobilin-containing solution and washed with 2 ml of methanol. The two methanol solutions were com- bined, 15 ~1 of 2-mercaptoethanol was added to remove Hg, and the white precipitate was removed by cen- trifugation. Ten milliliters of methylene chloride was added with mixing to the methanol, and then 20 ml

of water was added. After phase separation occurred,

the upper phase was discarded and the lower meth-

ylene chloride phase was washed with three 20-ml portions of water. An equal volume of 95% ethanol

was added to the methylene chloride and the solution

was applied to a column of DEAE-Sepharose CL-6B (acetate) as described for protoheme isolation.

The column was washed with 20 ml of 95% ethanohmethylene chloride (l:l, v/v). Finally, the

pigments were eluted with 1 ml of 95% ethanol:glacial acetic acid:water (92:1:7, v/v/v).

HPLCof pigments. Pigments were purified and sep-

arated from each other by reverse-phase HPLC on a

Varian Model 5000 liquid chromatograph. A 25-cm- long lO.O-mm-diameter Altex Ultrasphere ODS col-

umn with 5-pm particles was used. The solvent was

95% ethanol:glacial acetic acid:water (92:1:7, v/v/v). The column temperature was set at 30°C and the flow

rate was 4 ml/min. Pigments were dissolved in HPLC solvent before injection. Injection volume was 1.0 ml.

Absorbance was monitored at 373 nm during biliver- din and phycocyanobilin chromatography and at 402

nm during protoheme chromatography. One-minute

(4-ml) fractions were collected for later quantitative spectrophotometry and analysis of radioactivity. After

spectrophotometric measurement in HPLC solvent, fractions eluting between 8 and 38 min were trans- ferred to scintillation vials, solvent was evaporated

at 60°C and 5 ml of scintillation fluid was added. Quantitation of pigments. Purified protoheme and

phycocyanobilin were quantitated spectrophotomet- rically in HPLC solvent. The molar absorption coef-

ficient for protoheme in HPLC solvent was determined to be 170,000 at 399 nm, by measurement of standard

solutions whose concentrations were derived from the reduced minus oxidized pyridine hemochrome spectra,

using a molar difference absorption coefficient of 20,700 (9). The molar absorption coefficient for phy- cocyanobilin in HPLC solvent was found to be 49,940

at 374 nm, based on the value of 47,900 reported by Cole et al. (1) for the pigment in 5% (w/v) HCl in

methanol. Biliverdin concentrations were determined in 5% (w/v) HCl in methanol as reported by McDonagh

and Palma (lo), using an absorption coefficient of 66,200 at 377 nm.

Other procedures. Spectrophotometry was per-

formed on a Cary 219 instrument (Varian) and ra-

dioactivity determination by liquid scintillation on a Beckman Model LS-1OOC spectrometer.

Chemicals and radiochemicals. All reagents, sol-

vents, and culture medium components were pur- chased from Sigma or Fisher. [i4C]Biliverdin was a generous gift from A. F. McDonagh (Department of

Internal Medicine, University of California, San Francisco). It was prepared from [“Clbilirubin as de- scribed in (11). The [Wlbilurubin was in turn formed biosynthetically in rats after administration of 5- amino-[4-“Cllevulinic acid (New England Nuclear).

BEALE AND CORNEJO

TABLE I

CONVERSION OF EXOGENOUS [‘%]BILIVERDIN TO

PHYCOCYANOBILIN IN Cyanidium caldarium

Incubation time (h)

Biliverdin supplied: Amount (nmol)

Final concentration

(PM) Radioactivity (cpm)

Specific radioactivity

(cpm/nmol)

Cells harvested (X10-i’) Culture volume (ml)

Label uptake (%) Purified protoheme:

Amount recovered (nmol)

Radioactivity (cpm) Specific radioactivity

(cpm/nmol) Exogenous biliverdin

Contribution (%) Purified phycocyanobilin:

Amount recovered

(nmol)

Radioactivity (cpm) Specific radioactivity

(cpm/nmol) Exogenous biliverdin

Contribution (%)

12 24

95.6 96.0

1.47 1.51

798,200 801.600

8,350 8,350

1.2 1.5

65.0 63.5

48 44

8.6 11.3

6.0 f 0.4 44.0 f 2.6

0.7

0.008 0.047

13.5 16.5

193.5 + 5.8 200.9 f 8.0

16.4 12.2

0.17

3.91

0.15

Thus, the radioactive precursor was [2,4,6,8,12,14,16,19- i4C]biliverdin IX (A. F. McDonagh, personal com-

munication). N-Methylmesoporphyrin IX was pre- pared by the method of Ortiz de Montellano et al. (12)

as previously described (3).

RESULTS

Uptake of [14C]biliverdin, Percentage up- take of [14C]biliverdin was determined at various times during the incubation pe- riods by monitoring radioactivity in O.l- ml aliquots of cell suspension and in culture medium after removal of the cells by brief centrifugation. A large fraction of the label was apparently taken up by the cells during the first 12 h, and this amount did not change appreciably during the next 12 h (Table I).

Incorporation of “C into protoheme. Pro- toheme was purified as described under Materials and Methods. One-minute (4-ml)

fractions of the HPLC eluate were collected and the quantity of protoheme in the frac- tions containing this pigment was deter- mined spectrophotometrically. Then, ra- dioactivity was determined after evapo- ration of the solvent.

In the 12-h incubation (Fig. 2a), no sig- nificant radioactivity was detected in the elution region corresponding to protoheme at 33 min. The only peak of radioactivity occurred at 12 min, the elution time cor- responding to biliverdin. In the 24-h in- cubation sample, a small but significant peak in radioactivity was found to coelute with protoheme (Fig. 2b).

Incorporation of ‘% into ph~cocyanobilin. Phycocyanobilin was purified as described under Materials and Methods. One-minute (4-ml) fractions of the eluate from the HPLC were collected and the quantity of phycocyanobilin in the fractions containing this pigment was determined spectropho- tometrically. Then, radioactivity was de- termined after evaporation of the solvent.

In both the 12-h (Fig. 3a) and 24-h (Fig. 3b) incubations, the majority of the ra- dioactivity coeluted with the peaks con- taining phycocyanobilin. It should be noted that, in addition to the major phycocy- anobilin peak at 22 min, a smaller peak at 17 min also contains radioactivity. This peak probably corresponds to the “phy- cocyanobilin I” peak reported by Fu et al. (13) and may represent a structural isomer of phycocyanobilin. Little or no radioac- tivity was detected in the regions of the elution profiles corresponding to proto- heme or biliverdin.

Incubation requirement for 14C Inca ration into phycocqanobilin. In a control experiment, cells were grown and har- vested as described. After extraction of the free pigments, [14C]biliverdin was added to the cells just before the phycocyanin- methanolysis step. Isolation of phycocy- anobilin was carried out in the normal manner. In the HPLC eluate, radioactivity occurred only in the region corresponding to biliverdin at 12 min (Fig 4).

Quantitative comparism of W incm ration into protohem and phycoc~arwbilin The recovered quantities of protoheme and phycocyanobilin (Table I) include the

PHYCOCYANOBILIN BIOSYNTHESIS FROM BILIVERDIN 283

r I 1 150

il a

! 0

-100 .g .> ; :

0

]

ii

x z

- I 50

J

- rlJiGnJL”r- c 20 30

0 40

Elution Time (min)

Elution Time (min)

FIG. 2. HPLC elution profiles of absorbance at 402 nm (smooth curve) and radioactivity in one- minute eluate fractions (histogram) of the heme-containing extracts from cells grown in the presence

of [%]biliverdin for 12 h (a) and 24 h (b).

amounts which were present at the begin- Both the amount of protoheme recovered ning of the incubation as well as the and the specific radioactivity of this pig- amounts synthesized during the incuba- ment were significantly greater in the 24- tion. h incubation sample than in the 12-h sam-

284 BEALE AND CORNEJO

I

c

5-

o-

I5 -

IO-

Elution Time (mid

b

I - 150

22 4

-100 ': C .?

2 2 P u

Q z

-50

L---d - -I

Elution Time (mid

FIG. 3. HPLC elution profiles of absorbance at 373 nm (smooth curve) and radioactivity in one- minute eluate fractions (histogram) of the phycocyanobilin-containing extracts from cells grown

in the presence of [‘*C]biliverdin for 12 h (a) and 24 h (h).

PHYCOCYANOBILIN BIOSYNTHESIS FROM BILIVERDIN 285

Elution Time (min)

FIG. 4. HPLC elution profile of absorbance at 373 nm (smooth curve) and radioactivity in one- minute eluate fractions (histogram) of the phycocyanobilin-containing extract from cells to which [“Clbiliverdin was added after extraction of free pigments and before cleavage of phycocyanobilin from the phycocyanin by methanolysis.

ple. However, the radioactivity was much greater in phycocyanobilin than in pro- toheme in both incubations. In the 24-h incubation, the specific radioactivity of the phycocyanobilin was over three times greater than that of protoheme, and in the 12-h incubation the specific activity ratio was at least 23.

DISCUSSION

Our results are consistent with and strongly support the hypothesis that the biosynthesis of phycobilin pigments pro- ceeds through biliverdin, the protoheme ring-opening product.

Efficient uptake of relatively large mol- ecules such as biliverdin by unicellular ai- gae is unusual. However, C. caldarium pre- viously has been shown to be capable of taking up protoheme (4) and porphyrins (3, 5). The measured uptake percentages for [‘*C]biliverdin include both true intra-

cellular uptake and adherence onto the cell surface, and therefore the amounts of [‘*C]biliverdin available for metabolic con- version to phycocyanobilin were probably less than the amounts indicated by the up- take results. No attempt was made to as- sess the actual intracellular content of [14C]biliverdin. The slightly lower calcu- lated uptake at 24 h, compared to the value at 12 h, could be due to excretion of 14C- labeled biliverdin-degradation products by the cells.

At 12 h, no significant labeling of pro- toheme was found, although at 24 h there was a small amount of radioactivity as- sociated with this pigment. The most likely explanation for the delayed transfer of the small amount of label to protoheme is in- tracellular degradation of [14C]biliverdin and mixing of the released fragments into general biosynthetic carbon pools.

At both 12 and 24 h, there was much more ‘*C incorporation into phycocyano-

286 BEALE AND CORNEJO

bilin than into protoheme. This result in- dicates that the incorporation route was direct, rather than via degradation of bil- iverdin and reutilization of the 14C in the synthesis of phycocyanobilin via proto- heme.

REFERENCES

1. COLE, W. J., CHAPMAN, D. J., AND SIEGELMAN, H. W. (1967) J. Amer. Chem Sot 89,3643-3645.

2. BROWN, S. B., AND KING, R. F. G. J. (1978) Biochem CT. 170,297-311.

Our results indicate that exogenous bil- iverdin can serve as a direct phycocyano- bilin precursor in viva. It was recently shown (14) that [14C]biliverdin accumulates when C. caldarium cells are grown with exogenous 14C-labeled 5-aminolevulinic acid, but it was not determined whether this biliverdin could serve as a phycocy- anobilin precursor or whether it accu- mulated as a result of degradation of excess protoheme that might be formed as a re- sponse to ALA administration.

3. BEALE, S. I. AND CHEN, N. C. (1983) Plant Physid 71, 263-268.

4. BROWN, S. B., HOLROYD, J. A., TROXLER, R. F., AND OFFNER G. D. (1980) B&hem J. 191,137-147.

5. BROWN, S. B., HOLROYD, J. A., VERNON, D. I., TROXLER, R. F., AND SMITH, K. M. (1982) Biochem J. 208, 487-491.

6. OMATA, T., AND MURATA, W. (1980) Photo&m Photobiol 31, 183-185.

7. TROXLER, R. F., KELLY, P., AND BROWN, S. B. (1978) Biochem J. 172.569-576.

8. CRESPI, H. L., AND SMITH, U. H. (1970) Phgb chemistry 9, 205-212.

ACKNOWLEDGMENTS

9. FUHRHOP, J.-H., AND SMITH, K. M. (1975) in Por- phyrins and Metailoporphyrins (Smith, K. M., ed.), pp. 757-869, Elsevier, Amsterdam.

LO. MCDONAGH, A. F., AND PALMA, L. A. (1980) B&hem .I 189, 193-208.

We are indebted to A. F. McDonagh for supplying the [“‘Cjbiliverdin used in these experiments, and to S. I. Shedlofsky (Veterans Administration Hospital, White River Junction, Ver. for sending an additional sample of [“Clbiiiverdin used in preliminary studies. We thank J. D. Weinstein for assistance with devel- oping the purification procedures and for critically reading the manuscript. This work was supported by NSF Grant PCM-8213948, USDA Grant SEA-59-2442- l-l-679-0, and NIH BRSG Grant 2-SO%RR07085-17.

11. MCDONAGH, A. F. (1979) in The Porphyrins (Dol- phin, D., ed.) Vol. 6A, pp. 293-491, Academic Press, New York.

12. ORTIZDE MONTELLANO, P. R., KUNZE, K. L., COLE, S. P. C., AND MARKS, G. S. (1980) B&hem. Bio- phys. Res. Commun. 97,1436-1442.

13. Fu, E., FRIEDMAN, L., AND SIEGELMAN, H. W. (1979) Biochem. J. 179, l-6.

fwsch. 37C, 105’7-1063. 14. K&T, H.-P., AND BENEDIKT, E. (1982) 2. Natur-