Evaluation of two biosynthetic pathways to d-aminolevulinic acid in Euglena gracilis Katsumi Iida, Ippei Mimura and Masahiro Kajiwara Department of Medicinal Chemistry, Meiji Pharmaceutical University, Kiyose-shi, Tokyo, Japan d-Aminolevulinic acid (ALA), which is an intermediate in the b iosynthesis o f c hlorophyll a, c an be biosynthesized via the C5 pathway and the Shemin pathway in Euglena gracilis. Analysis of the 13 C-NMR spectrum of 13 C-labeled methyl pheophorbide a, derived from 13 C-labeled chlorophyll a biosynthesized from D -[1- 13 C]glucose by E. gracilis,provid- ed evidence suggesting that ALA incorporated in the 13 C-labeled chlorophyll a was synthesized via both the C5 pathway and the Shemin pathway in a ratio of between 1.5 and 1.7 to one. The methoxyl carbon of the methoxycar- bonyl group at C-13 2 of chlorophyll a was labeled with 13 C. The phytyl moiety of chlorophyll a was labeled on C-P2, C-P3 1 ,C-P4,C-P6,C-P7 1 , C-P8, C-P10, C-P11 1 ,C-P12, C-P14, C-P15 1 and C-P16. Keywords: d-aminolevulinic acid; C5 pathway; Shemin pathway; Euglena gracilis; 13 C-NMR. d-Aminolevulinic acid (ALA) (Fig. 1, 3), which is an intermediate in the biosynthesis of tetrapyrrole compounds such as chlorophyll a (1), vitamin B 12 and heme, can be biosynthesized via two pathways, the Shemin pathway (C4 pathway) [1±7] and t he C5 pathway [8±14] (Fig. 1). In the Shemin pathway, ALA (3) is biosynthesized b y the condensation of glycine (4) and succinyl CoA (5). In the C5 pathway, ALA (3) is derived from all the carbons of L -glutamate ( L -glutamic acid; 6). Mayer et al. reported that ALA (3) is biosynthesized via the C5 p athway in Euglena gracilis [12]. B eale et al. reported that E. gracilis contains ALA synthase [15], implying that ALA (3) may also be synthesized via the Shemin pathway. Weinstein et al. [16] reported that the C5 pathway in t he chloroplast and ALA synthase probably in the mitochond- rion of E. gracilis operate simultaneously to biosynthesize ALA. They also showed that [2- 14 C]glycine was incorpo- rated speci®cally into the nontetrapyrrole portion of chlo- rophyll a (1)byE. gracilis . Okazaki et al. [17] found that [2- 13 C]glycine was not incorporated in the tetrapyrrole portion of chlorophyll a (1)viaALA(3), but was incorpo- rated into the methoxyl carbon of the methoxycarbonyl group at C-13 2 of chlorophyll a (1)byE. gracilis. Oh-hama et al. [18] and Porra et al. [19] reported similar results for incorporation o f isotope-labeled glycine into chlorophyll a (1)byScenedesmus obliquus and maize leaves. Thus, the involvement of the Shemin pathway co uld not be assessed in terms of labeling i n the tetrapyrrole portion of chlorophyll a (1) from isotope-labeled glycine fed to the organism. Porra et al. concluded that t he C5 pathway is the predominant biosynthetic pathway to ALA utilized in chlorophyll a (1), as shown from feeding experiments with D , L -[1- 13 C]- and [5- 13 C]glutamic acid in maize leaves [19]. This is in contrast to their previous estimation of approximately equal contri- butions of the C5 pathway and the S hemin pathway, based on feeding experiments with sodium[1- 14 C]- and[5- 14 C] a-ket oglutarate [2 0]. We were interested in investigating the existence of the Shemin pathway for ALA and the ratio of ALA biosyn- thesis from the Shemin pathway to that from the C5 pathway in E. gracilis. Shemin and others reported that ALA (3) is biosynthesized via the Shemin pathway in Propionibacterium shermanii [6,7], but our analysis of the 13 C-NMR s pectrum o f 13 C-labeled vitamin B 12 biosynthe- sized from D -[1- 13 C]glucose by P. shermanii provided evidence that ALA (3) incorporated in the 13 C-labeled vitamin B 12 may have been synthesized via both the Shemin pathway and the C5 p athway [21]. We therefore conducted similar feeding experiments with D -[1- 13 C]glucose in E. grac- ilis, and used 13 C-NMR spectroscopy to examine the 13 C-enrichment ratios of the carbon atoms of 13 C-labeled chlorophyll a or its derivative, 13 C-labeled methyl pheo- phorbide a (Fig. 1 ). Our results indicate that the C5 and Shemin pathways both operate in E. gracilis, an d provide information about the biosynthetic pathways leading to the methoxyl carbon of the methoxycarbonyl group at C-13 2 and the phytyl moiety of chlorophyll a (1). EXPERIMENTAL PROCEDURES Organism and chemicals The strain used was E. gracilis IME E-3. Chlorophyll a (1) (from Spirulina) was purchased from Wako Pure Chemical Industries, Ltd. Methyl pheophorbide a (2) was purchased from Tama Biochemical Co., Ltd. D -[1- 13 C]Glucose (90 a tom % 13 C) was purchased from Cambridge Isotope Laboratories. All other chemicals were of analytical grade. Correspondence to K. Iida, Department of Medicinal Chemistry, Meiji Pharmaceutical University, 2-522-1 Noshio, Kiyose-shi, Tokyo 204-8588, Japan. Fa x: + 81 424 95 8612; Tel.: + 81 424 95 8611, E-mail: iida@my-pharm.ac.jp Abbreviations:ALA,d-aminolevulinic acid; DMBI, 5,6-dimethyl- benzimidazole; GSA, glutamate 1-semialdehyde; MPLC, medium- pressure liquid chromatography; ODS, octadecyl silica; TCA, tricarboxylic acid. (Received 31 August 2001, revised 1 November 2001, accepted 2 November 2001) Eur. J. Biochem. 269, 291±297 (2002) Ó FEBS 2002 Instruments All 1 H-NMR (400 MHz) and 13 C-NMR (100 MHz) spec- tra were recorded on a Jeol GSX-400 spectrometer. UV spectra were recorded on a Jasco UVIDEC-610C spectrometer. Examination of optimum amount of D -[1- 13 C]glucose for E. gracilis feeding experiments E. gracilis was cultured as described previously, with some modi®cations [17]. The cultures were grown under illumi- nation (2400 Lx) in seed culture medium (10 mL), which consisted o f L -glutamic a cid (5 gáL )1 ), D , L -malic acid (2 gáL )1 ), L -methionine (50 mgáL )1 ), thiamine hydrochlo- ride (1 mgáL )1 ), cyanocobalamin (5 lgáL )1 ), KH 2 PO 4 (0.4 gáL )1 ), MgSO 4 á7H 2 O(0.5gáL )1 ), CaCO 3 (0.1 gáL )1 ) (NH 4 )PO 4 (0.2 gáL )1 ), EDTA (10 mgáL )1 ), ZnCl 2 (10 mgáL )1 ), FeSO 4 á7H 2 O(4mgáL )1 ), MnCl 2 á4H 2 O (2 mgáL )1 ), CuCl 2 á2H 2 O(0.4mgáL )1 ), CoCl 2 á6H 2 O (2 mgáL )1 )andH 3 BO 4 á7H 2 O(80lgáL )1 ), in a 60-mL test tube at 27 °C. After 7 days, the seed culture medium (10 mL) was added to fermentation culture medium (1 L) in a 3-L conical ¯ask. This fermentation culture medium contained 2.5±20 gáL )1 of D -glucose (9) added in place of L -glutamic acid (5 gáL )1 )and D , L -malic acid (2 gáL )1 )inthe seed culture medium. The cultures of E. gracilis were continuously grown photosynthetically (2400 Lx) at 27 °C with or without bubbling of air. After 7 days, the wet cells, collected by centrifugation of the culture broth for 30 min at 12 300 g, were weighed. Feeding of D -[1- 13 C]glucose to E. gracilis The above seed culture medium (10 mL ´ 2), cultivated for 7 d ays, was added to fermentation culture medium (1 L ´ 2), which consisted of D -[1- 13 C]glucose (2.5 gáL )1 ) added in place of L -glutamic acid (5 gáL )1 )and D , L -malic acid (2 gáL )1 ) in the seed culture medium, in a 3-L conical ¯ask. The cultures of E. gracilis were continuously grown photosynthetically (2400 Lx) at 27 °C for 7 days with bubbling of air. The cells were collected by centrifugation of the culture broth for 30 min at 12 300 g. Isolation of 13 C-labeled chlorophyll a The isolation of chlorophyll a (1) was carried out by modi®cation of the methods described in our previous paper [17]. The growing cultures of E. gracilis were washed with 0.9% NaCl, and this suspension was centrifuged again for 30 min at 12 300 g. The cells were suspended in CH 3 OH (50 m L), disrupted with an ultrasonicator at 0 °Cfor5 min, and centrifuged for 30 min at 12 300 g. The supernatant was evaporated in the dark. Pu ri®cation of the residue by medium-pressure liquid chromatography (MPLC) using a prepacked glass column [2.5 (internal diameter) ´ 30 cm, octadecyl silica (ODS)] with C H 3 OH gave 13 C-labeled chlorophyll a. The amount of 13 C-labeled chlorophyll a Fig. 1. Biosynthetic pathways to chlorophyll a (1) from D -glucose (9) and structure of methyl pheophorbide a (2). Chlorophyll a (1) is biosynthesized through d-aminolevulinic acid (ALA) (3) formed via the C 5 p athway and the Shemin p athway from D -glucose (9), an d methyl pheophorbide a (2)is derived from chlorophyll a (1). 292 K. Iida et al. (Eur. J. Biochem. 269) Ó FEBS 2002 isolated was calculated from the UV absorption spectrum [22]. Transformation from 13 C-labeled chlorophyll a to 13 C-labeled methyl pheophorbide a Concentrated H 2 SO 4 (0.5 mL) was added dropwise, at 0 °C under argon, to a solution of 13 C-labeled isolated chloro- phyll a in dry CH 3 OH (9.5 mL), and the mixture was stirred for 12 h at room temperature in the dark. The reaction mixture was diluted with CH 2 Cl 2 (200 mL), and quenched with saturated NaHCO 3 . The organic layer was washed with saturated NaHCO 3 , water a nd saturated NaCl, dried over dry MgSO 4 , and then evaporated. Chromatography of the crude product on silica gel with CHCl 3 /CH 3 OH (25 : 1 , v/v) gave 13 C-labeled methyl pheophorbide a. The amount of 13 C-labeled methyl pheophorbide a isolated was calcu- lated from the UV absorption spectrum [22]. 13 C-NMR measurements of chlorophyll a and methyl pheophorbide a The 13 C-NMR spectra were obtained for solutions of 13 C- labeled chlorophyll a (4 .8 m M ) and chlorophyll a (1)in C 2 HCl 3 /C 2 H 3 OH (79 : 6, v/v), an d solutions of 13 C-labeled methyl pheophorbide a (3.8 m M ) and methyl pheophorbide a (2)inC 2 HCl 3 . The signal of C 2 HCl 3 (77.0 p.p.m.) was used as an internal standard. The spectral width w as 24 038.5 Hz with 32 768 data points, which corresponds to a resolution of 0.73 Hz per point. The 10-pulse-width was 4.4 ls, the acquisition time was 0.682 s, the pulse delay time was 2.5 s, and the number of scans was 15 000±18 000. The assignments of 13 C-NMR signals of chlorophyll a (1)and methyl pheophorbide a (2)weremadeonthebasisof reported data [23±28]. Calculation of 13 C-incorporation ratios in 13 C-labeled methyl pheophorbide a The signal of the methoxyl carbon, which was derived from CH 3 OH used in the transesteri®cation reaction to the methyl ester from the phytyl ester of 13 C-labeled chloro- phyll a of the methoxycarbonyl group at C-17 2 of 13 C-labeled m ethyl pheophorbide a shows t he natural abundance of 13 C,andthuscanbeusedasareference signal. T he 13 C-enrichment ratio for each carbon of 13 C-labeled methyl pheophorbide a was calculated from comparison of the signal intensities or half widths in the 13 C-NMR spectrum of 13 C-labeled methyl pheophorbide a, with those of methyl pheophorbide a (2). RESULTS Suitable amount of D -[1- 13 C]glucose for feeding experiment to E. gracilis Cultures of E. gracilis were grown photosynthetically in E. gracilis fermentation culture medium containing various amounts of D -glucose (9)inplaceof L -glutamic acid and D , L -malic acid, which are the carbon sources of chlorophyll a (1), without or with bubbling of air. After 7 days, the culture broth was centrifuged for 30 min at 12 300 g, and the cells were weighed. Without air bubbling, 10, 15 and 20 gáL )1 of D -glucose (9) gave 2.11, 4.01 and 4.72 gáL )1 of E. gracilis, respectively, as shown in Table 1. With air bubbling, 2.5, 5, 10 a nd 15 gáL )1 of D -glucose (9) gave 3.64, 3.64, 4.23 and 5.85 gáL )1 of E. gracilis, respectively. For reasons of economy, we chose to use two cultures, each containing 2.5 gáL )1 of D -[1- 13 C]glucose, with air bubbling f or the feeding experiments. Biosynthesis of 13 C-labeled chlorophyll a and 13 C-incorporation in its phytyl moiety 13 C-Labeled chlorophyll a (2.6 mg) was isolated from growing cultures ( 6.7 g) of E. gracilis cultivated in two 1-L fermentation culture medium in the p resence of D -[1- 13 C]glucose. Its purity was judged to be high by comparison of the 1 H-NMR and UV spectra with those of authentic chlorophyll a (1). The 13 C-enrichments of c arbons (C-P2, C-P3 1 ,C-P4,C-P6,C-P7 1 , C-P8, C-P10, C-P11 1 , C-P12, C-P14, C-P15 1 and C-P16) of the phytyl moiety of 13 C-labeled chlorophyll a were higher than those of carbons of the chlorin ring moiety. Synthesis of 13 C-labeled methyl pheophorbide a and determination of 13 C-incorporation ratios 13 C-Labeled methyl pheophorbide a (1.4 mg) was derived from 13 C-labeled chlorophyll a (2.6 mg). Its purity was judged to be high by comparison of the 1 H-NMR and UV spectra with those o f authentic methyl pheophorbide a (2). The signal of the methoxyl carbon, derived from CH 3 OH used in the transesteri®cation reaction to the methyl ester from the phytyl ester of 13 C-labeled chlorophyll a,ofthe methoxycarbonyl group at C-17 2 of 13 C-labeled methyl pheophorbide a showed the natural abundance of 13 C, and thus was used as a reference signal. Comparison of the signal intensities or half widths in the 13 C-NMR spectrum of 13 C-labeled methyl pheophorbide a with tho se of methyl pheophorbide a (2) (Fig. 2) gave the 13 C-enrichment ratio for each carbon of 13 C-labeled methyl pheophorbide a.The carbons of methyl pheophorbide a (2) are classi®ed into six groups according t o their biosynthetic origin [12,16,17], i.e. from each carbon of ALA (3) and the methyl carbon of L -methionine, as summarized in Table 2. The average 13 C-enrichment ratio of carbons (C-13 3 and C-17 3 ) derived from C-1 of ALA (3) was 2.4-fold, that of carbons (C-2 1 , Table 1. Determination of suitable amount of D -glucose (9) for feeding experiment. The cultures of E. gracilis were grown photosynthetically in the fermentation cu lture medium, wh ich contained of 2 .5±20 gáL )1 of D -glucose (9) ad ded in p lace of L -glutamic acid and D , L -malic acid, without or with bubbling of air. After 7 days, the cells were collected by centrifugat ion of the culture broth, an d the wet weight was mea- sured. See Experimental procedures for details. Amount of D -glucose (gáL )1 ) Yield (gáL )1 )ofE. gracilis cells No bubbling of air Bubbling of air 2.5 ± 3.64 5 ± 3.64 10 2.11 4.23 15 4.01 5.85 20 4.72 Ó FEBS 2002 Evaluation of two ALA biosynthetic pathways (Eur. J. Biochem. 269) 293 C-3 2 ,C-7 1 ,C-8 2 ,C-12 1 ,C-13 2 ,C-17 2 and C-18 1 ) derived from C-2 of ALA (3) was 8.8-fold, that of carbons (C-2, C-3 1 ,C-7,C-8 1 , C-12, C-13 1 ,C-17 1 and C-18) derived from C-3 of ALA (3) was 4.1-fold, that of carbons (C-1, C-3, C-6, C-8, C-11, C-13, C-17 and C-19) derived from C-4 of ALA (3) was 4.1-fold, and that of carbons (C-4, C-5, C-9, C-10, C-14, C-15, C-16 and C-20) derived from C-5 of ALA (3) was 3.7-fold. The 13 C-enrichment ratio o f the methoxyl carbon, which is derived from the methyl carbon of L -methionine, of the meth oxycarbonyl group at C-13 2 was 1.8-fold. T he C-1 to C-5 carbons of ALA (3) and the methyl carbon of L -methionine were thus labeled with 13 Cfrom D -[1- 13 C] glucose. DISCUSSION Biosynthetic pathways leading to ALA and L -methionine in E. gracilis The chlorin ring moiety of methyl pheophorbide a (2), in addition to the methyl carbon of L -methionine, is derived from the carbons of ALA (3), which may in principle be formed via the C5 pathway or the Shemin p athway (Fig. 1) [12,16,17]. As shown in Table 2, the average 13 C-enrichment ratios of carbons derived from C-1 to C-5 of ALA (3)are 2.4-, 8.8-, 4.1-, 4.1- and 3.7-fold, respectively. The 13 C-enrichment ratio of the methoxyl carbon, which is derived from the methyl carbon of L -methionine, of the methoxycarbonyl group at C-13 2 is 1.8-fold. These results demonstrate that the C-1 to C-5 carbons of ALA (3)andthe methyl carbon of L -methionine were labeled with 13 Cfrom D -[1- 13 C]glucose. Figure 3 shows the positions that are predicted to be labeled in ALA (3ii-5ii to 3vii-5vii and 3i-7i to 3v-7v) biosynthesized from 13 C-labeled succinyl CoA (5ii to 5vii) and 13 C-labeled a-ketoglutaric acid (7i to 7v)viatheC5 Fig. 2. 13 C-NMR spectra of 13 C-labeled methyl pheophorbide a and methyl pheophorbide a (2). Upper: spectrum of 13 C-labeled methyl pheophorbide a derived from 13 C-labeled chlorophyll a,whichwas biosynthesized from D -[1- 13 C]glucose in E. gracilis.Lower:spectrum of methyl pheophorbide a (2). Table 2. 13 C-Enrichment ratios for c arbon atoms in 13 C-labeled methyl pheophorbide a derived from 13 C-labeled chlorophyll a bio synthes ized from D -[1- 13 C]glucose in E. gracilis. The cultures of E. gracilis were grown photosynthetically in fermentation culture medium containing D -[1- 13 C]glucose with bubbling of air. The E. gracilis cells collected gave rise to 13 C-labeled chlorophyll a after puri®cation. The 13 C-enrichment ratios for each c arbon of 13 C-labeled methyl pheophorbide a were obtained by comparison of the 13 C-NMR spectrum of 13 C-labeled methyl pheophorbide a, which was derived from th e 13 C-labeled chlorophyll a, with those of methyl pheophorbide a (2). For each group shown in t he table, the ®rst line indicates the carbon positions, the second line gives 13 C-NMR ch emical shift values in p.p.m., and t he third line shows the 13 C-enrichment ratio. For de tails of calculation o f 13 C-incorporation ratio in 13 C-labeled m ethyl pheophorbide a, see Experimental pro ce dures. The reference carb on (reference signal) was the methoxyl carbon of the metho xycarbonyl group at C-17 2 (51.66 p.p.m., 13 C-Enrichment ratio of 1.0). Carbons of m ethy l p heophorbi de a are c lassi®ed in to six groups accordin g t o their biosynthetic origin: C-1 to C -5 ind icate carb ons o f A LA (3), and methyl indicates the methyl carbon of L -methionine. C-1 a 13 3 169.56 2.4 17 3 173.34 2.4 C-2 b 2 1 3 2 7 1 8 2 12 1 13 2 17 2 18 1 12.05 122.83 11.26 17.42 12.10 64.70 29.84 23.06 9.2 9.5 8.1 8.2 9.2 8.6 8.4 8.8 C-3 c 23 1 78 1 12 13 1 17 1 18 131.88 128.93 136.21 19.48 129.06 189.62 31.01 50.09 4.4 4.6 3.7 3.9 4.6 4.0 3.4 4.2 C-4 d 1 3 6 8 11 13 17 19 142.07 136.32 155.70 145.26 137.94 129.00 51.08 172.19 4.4 4.0 4.6 3.2 3.4 4.6 4.4 4.2 C-5 e 4 5 91014151620 136.53 97.58 151.01 104.47 149.67 105.18 161.19 93.13 3.4 4.1 3.8 3.8 3.2 3.5 3.7 4.4 Methyl Methoxyl carbon of the methoxycarbonyl group at C-13 2 52.85 1.8 a Average 13 C-enrichment ratio for C-1 of ALA (3) is 2.4. b Average 13 C-enrichment ratio for C-2 of ALA (3) is 8.8. c Average 13 C- enrichment ratio for C-3 of ALA (3) is 4.1. d Average 13 C-enrichment ratio for C-4 of ALA (3) is 4.1. e Average 13 C-enrichment ratio for C-5 of ALA (3) is 3.7. 294 K. Iida et al. (Eur. J. Biochem. 269) Ó FEBS 2002 pathway and the Shemin pathway. As shown in Figs 1 and 3, the C-2 to C-5 carbons of ALA (3) generated via the C5 pathway are labeled with 13 Cfrom D -[1- 13 C]glucose. The C-1 carbon of ALA (3) formed via the C 5 pathway is not labeled with 13 Cfrom D -[1- 13 C]glucose, as this carbon is derived from C-1, whose carbon is not labeled with 13 Cfrom D -[1- 13 C]glucose, of acetyl CoA (8). On the other hand, the C-1 to C-4 carbons of ALA (3) produced via the Shemin pathway are labeled with 13 Cfrom D -[1- 13 C]glucose. The C-5 carbon of ALA (3) formed via the Shemin pathway is not labeled with 13 Cfrom D -[1- 13 C]glucose, as this carbon is derived from C-2, whose carbon is not labeled with 13 Cfrom D -[1- 13 C]glucose, of glycine (4) derived from L -[3- 13 C]serine, which is generated from D -[1- 13 C]glucose via [2- 13 C]acetyl CoA and [3- 13 C]pyruvic acid. Thus, ALA (3) labeled with 13 C on C-1 appears via the Shemin pathway, never via the C5 pathway, and ALA (3) labeled with 13 C on C-5 appears via the C5 pathway, never via the Shemin pathway. Therefore, the observed 13 C-enrichment at carbons of 13 C- labeled methyl pheophorbide a derived from C-1 and C-5 of ALA (3) suggests that both pathways to ALA (3)operatein E. gracilis . As shown in Fig. 3 and discussed in our previous report [21], the biosynthesis of ALA molecules (3iv-5iv and 3v-7v) labeled with 13 ConC-1andC-5canberationalizedas follows. Succinyl CoA, which is formed in the s econd cycle of the tricarboxylic acid (TCA) cycle, is labeled with 13 Con C-1atthe®rstentryof[2- 13 C]acetyl CoA (8i) into the TCA cycle and transformed to succinic acid. At this time, succinic acid molecules labeled with 13 C on C-4 and C-1 appear in equal quantity. Succinic acid labeled with 13 ConC-4and C-1 can revert to succinyl CoA (5iv and 5v), giving rise to succinyl CoA (5iv and 5v)labeledwith 13 ConC-4andC-1 in equal quantity. Part of succinyl CoA (5iv and 5v) labeled with 13 C on C-4 and C-1 goes into the Shemin pathway, and condenses with glycine ( 4). ALA (3iv-5iv) labeled with 13 C on C-1 is biosynthesized from succinyl CoA (5iv) labeled with 13 C on C-4, and gives rise to a 2.4-fold 13 C-enrichment in 13 C-labeled methyl pheophorbide a.ALA(3v-5v) labeled with 13 C on C-4 is concomitantly biosynthesized from succinyl CoA (5v) labeled with 13 C on C-1. The rest of succinyl CoA (5iv and 5v)labeledwith 13 ConC-4andC-1 re-enters the TCA cycle, and generates 13 C-labeled a-ket oglutaric acid ( 7iv and 7v)via 13 C-labeled succinic acid, 13 C-labeled oxaloacetic acid, 13 C-labeled citric acid and other 13 C-labeled intermediates. 13 C-Labeled L -glutamic acid, which is formed from 13 C-labeled a-ketoglutaric acid (7iv and 7v), goes into the C5 pathway, and generates 13 C- labeled ALA (3iv-7iv and 3v-7v). Namely, succinyl C oA (5v) labeled with 13 C on C-1 generates ALA (3v-7v) labeled w ith 13 ConC-5viaa-ketoglutaric acid (7v) labeled with 13 Con C-1, and 13 C on C-4 of succinyl CoA (5iv) labeled at the ®rst entry of [2- 13 C]acetyl CoA (8i) into the TCA cycle disappears from 13 C-labeled ALA (3iv-7iv). The 13 C- enrichment ratio of C-5 of 13 C-labeled ALA (3v-7v)is decreased in c omparison with that of C-1 of 13 C-labeled succinyl CoA (5v) that re-entered the TCA cycle owing to the many pathways leaving from the pathway between succinyl CoA (5)and L -glutamic acid (6), and ALA (3v-7v) labeled with 13 C on C-5 gives rise to at least a 3.7-fold 13 C- enrichment in 13 C-labeled methyl pheophorbide a.Further, the 13 C-enrichment ratio of C-5 of 13 C-labeled ALA (3v-7v ) generated from [2- 13 C]acetyl CoA (8i) v ia only t he C5 pathway in the third cycle of the TCA cycle can not be larger Fig. 3. Positions of 13 C in products derived from D -[1- 13 C]glucose. Changes of 13 C-label position d uring the biosynthesis of ALA (3ii-5ii to 3vii-5vii and 3i-7i to 3v-7v), through the C5 pathway or the Shemin pathway via the TCA cycle f rom [2- 13 C]acetyl CoA (8i to 8iii) derived from D - [1- 13 C]glucose. (ccc cc) r epresents a-ket oglutaric acid (7i to 7v ), (cccc ) r epresen ts suc cinyl C oA ( 5ii to 5vii), and (cc) represents acetyl CoA (8i to 8iii ). (c) is unlabeled carbon, (C)is 13 C-carbon from ®rst entry of [2- 13 C]acetyl CoA (8i)intotheTCAcycle,( )is 13 C-carbon from the second entry of [2- 13 C]acetyl CoA (8ii) into the TCA cycle, and (C) is 13 C-carbon from the third entry of [2- 13 C]acetyl CoA (8iii)intotheTCAcycle. 13 C-Labeled positions of succinyl CoA (cccc) (5ii to 5vii) are those of the product formed by reversion from succinic acid. Numbers shown under (ccccc) (cccc) and (cc) a re the carbon numbers of th e com pounds. 13 C-Labeled positions of ALA ( 3ii-5ii to 3vii-5vii and 3i-7i to 3v-7v) formed via the C 5 p athway from each 13 C-labeled ( cccc c) a nd via the Shemin pathway from ea ch 13 C-labeled (cccc) a re s hown a t the s ide. (a ) an d (b) o n arro ws ( ® ) show the C5 pathway and the Shemin p athway, respe ctively. Ó FEBS 2002 Evaluation of two ALA biosynthetic pathways (Eur. J. Biochem. 269) 295 than the average 13 C-enrichment ratio (4.1-fold), which is mainly due to ALA (3ii-7ii and 3v-5v)labeledwith 13 Con C-4 generated from [2- 13 C]acetyl CoA (8i) via both the C5 pathway and the Shemin pathway in the second cycle of the TCA cycle, of carbons of 13 C-labeled methyl pheop horbide a derived from C-4 of ALA (3). Thus, the 13 C-enrichment ratio of C-5 of 13 C-labeled ALA (3v -7v) takes the value of between 3.7- and 4.1-fold. On the basis of relation of the biosynthetic pathways of ALA (3iv-5iv and 3v-7v) labeled with 13 ConC-1andC-5, the 13 C-enrichment ratio (2.4-fold) of carbons of 13 C-labeled methyl pheophorbide a derived from C-1 of ALA (3) should re¯ect the ratio of ALA biosynthesis from t he Shemin pathway, and the 13 C-enrichment ratio ( between 3.7-fold and 4.1-fold) of carbons of 13 C-labeled methyl pheophor- bide a derived from C-5 of ALA (3) should re¯ect the ratio of ALA biosynthesis from the C5 pathway. Thus, on the assumption that substantial scrambling o f the label does not occur, we can estimate the relative contributions of the C5 pathway and the Shemin path way to ALA biosynthesis in a ratio of between 1.5 (i.e. 3.7/2.4 ) and 1.7 (i.e. 4.1/2.4) to one. E. gracilis also biosyn thesizes ALA from the conden- sation of glycine (4) and succinic acid [15,29,30]. However, simultaneous biosynthesis of ALA from succinyl CoA and succinic acid would not in¯uence the estimation of the ratio of ALA biosynthesis via the C5 p athway to that via t he Shemin pathway. It is crucial to evaluate the extent of scrambling of the label due to possible alternative or competing biosynthetic pathways or degradative reactions, particularly as a culture period of 7 days was employed. Although we cannot assess the importance of a ll the possible reactions, we can assess the contribution of the second passage through the TCA cycle, which is likely to be one of the major contributors to label scrambling. That is, there is a contribution to the biosynthesis of ALA, which would be labeled with 13 Con C-1 and C-5, from [2- 13 C]acetyl CoA (8ii) generated in the second cycle of the TCA cycle (shown as c ). As this results in the synthesis of AL A (3iii-7iii, 3vi-5vi and 3vii-5vii )with adjacent labeled carbons at C-2 and C-3 ( Fig. 3), we can estimate the contribution of [2- 13 C]acetyl CoA (8ii)fromthe second turn of the TCA cycle from the ratio of doublet and singlet signals in the 13 C-NMR spectrum; the average was  10 %. This suggests that extensive scrambling of the label does not occur, and that this approach to evaluate the contributions of the two pathways is reasonable. It is worth noting that the contributions of more complex scrambling pathways would tend to be diluted out. A comment is necessary re garding the enrichment ratio (1.8-fold) of the methoxyl carbon of the methoxycarbonyl group at C-13 2 of 13 C-labeled methyl pheophorbide a. During the exchange of the phytyl ester to the m ethyl ester i n the transformation of 13 C-labeled chlorophyll a to 13 C-labeled methyl pheophorbide a in CH 3 OH and con- centrated H 2 SO 4 , it is possible that some exchange of the methoxyl carbon of the methoxycarbonyl group at C-13 2 with the carbon of CH 3 OH also occurs, though the reactivities of the phytyl and methyl esters are likely to b e different. Thus, all we can say about the 13 C-enrichment of the methoxyl car bon of the methoxycarbonyl group at C-13 2 of chlorophyll a, is that the observed value of 1.8-fold in m ethyl pheophorbide a represents a minimum value. With regard to the source of the methoxyl carbon of the methoxycarbonyl group at C-13 2 of 13 C-labeled methyl pheophorbide a,[2- 13 C]acetyl CoA, which would be formed from D -[1- 13 C]glucose by glycolysis, is transformed to L -[3- 13 C]serin e via [3- 13 C]pyruvic acid. The L -[3- 13 C]serine is transformed to glycine (4) in the presence of tetrahydrof- olic acid , and N 5 ,N 10 -[ 13 C]methylenetetrahydrofolic acid is derived f rom t he 13 C-carbon of L -[3- 13 C]serine and tetra- hydrofolic acid. N 5 ,N 10 -[ 13 C]Methylenetetrahydrofolic acid gives rise to L -[methyl- 13 C]methionine. Therefore, the methoxyl carbon of the methoxycarbonyl group at C-13 2 of 13 C-labeled methyl pheophorbide a is labeled with 13 Cfrom D -[1- 13 C]glucose, as this carbon is derived from the methyl carbon of L -methionine [17±19]. CONCLUSION Our results suggest that ALA (3) is synthesized via both the C5 pathway and the Shemin pathway from the TCA cycle in E. gracilis, with the relative contributions being in a ratio of between 1.5 and 1.7 to one. The extent of label scrambling could not be quantitatively determined, but the effect of second passage through the TCA cycle (likely to be a major contributor) was estimated to be only 10%. We also found that the phytyl moiety of chlorophyll a (1) is synthesized via the condensation of 13 C-labeled isoprene ([1,2-methyl, 3- 13 C 3 ]2-methyl-1,3-butadiene) generated from D -[1- 13 C]- glucose via [2- 13 C]acetyl CoA. The methoxyl carbon of the methoxycarbonyl group at C-13 2 of chlorophyll a (1) was derived from the 13 C-labeled methyl carbon of L -[methyl- 13 C]methionine generated from D -[1- 13 C]glucose via [2- 13 C]acetyl CoA and L -[3- 13 C]serine. ACKNOWLEDGEMENT We thank Prof. R. Timkovich (University of Alabama, AL, USA) for advice on ALA biosynthetic pathways. REFERENCES 1. Shemin, D. & Rittenberg, D. (1945) The utilization of glycine for the synthesis of a porphyrin. J. Biol. Chem. 159, 567±568. 2. Shemin, D. & Rittenberg, D. (1946) The biological utilization of glycine for the syn thesis of the protoporphyrin of hemoglo bin. J. Biol. Chem. 166, 621±625. 3. Shemin, D. & Rittenberg, D. (1946) The life span of the human red blood cell. J. Biol. Chem. 166, 627±636. 4. Radin, N.S., Rittenberg, D. & Shemin, D. (1950) The role of glycine in the biosynthesis of heme. J. Biol. Chem. 184, 745±753. 5. Radin, N.S., Rittenberg, D. & S hemin, D. (1950) The rol e of acetic acid in the biosynthesis of heme. J. Biol. Chem. 184, 755±767. 6. Menon, I.A. & Shemin, D. (1967) Concurrent decrease of e nzymic activities concerne d with t he syn thesis of coenzyme B 12 and of propionic acid in Propionibacteria. Arch. Biochem. Biophys. 121, 304±310. 7. Mauzerall, D. & Granick, S. (1956) The occurrence and deter- mination of d-aminolevulinic acid and porphobilinogen in urine. J. Biol. Chem. 219, 435±446. 8. Beale, S.I. & Castelfranco, P.A. (1973) 14 C incorporation from exogenous compounds into d-aminolevulinic acid by greening cucumber cotyledons. Biochem. Biophys. Res. Commun. 52, 143± 149. 9. Beale, S.I. & Castelfranco, P.A. (1974) Biosynthesis of d-amino- levulinic acid in higher plants. Plant Physi ol. 53, 291±296. 10. Beale, S.I. & Castelfranco, P.A. ( 1974) B iosynthesis o f d-amino- levulinic acid in higher plants II. Formation of d-aminolevulinic 296 K. Iida et al. (Eur. J. Biochem. 269) Ó FEBS 2002 acid from labeled precursors in greening plant tissues. Plant Physiol. 53, 297±303. 11. Weinstein, J.D. & Beale, S.I. (1985) RNA is required for enzy- matic con versio n of glutamate t o d-aminolevulinic acid by extracts of Chlorella vulgaris. Arch. Biochem. Biophys. 239, 87±93. 12. Mayer, S.M., Beale, S.I. & W einstein, J.D. (1987) Enzymatic conversion of glutamate to d-aminolevulinic acid in soluble extracts of Euglena gracilis. J. Biol. Chem. 262, 12541±12549. 13. Smith, A.G., Marsh, O. & Elder, G.H. (1993) Investigation of the subcellular location of the tetrapyrrole-biosynthesis enzyme coproporphyrinogen oxidase in higher plants. Biochem. J. 292, 503±508. 14. Oh-hama, T., Stolowich, N.J. & Scott, A.I. (1988) 5-Aminolevu- linic acid formation from glutamate via the C 5 pathway in Clos- tridium thermoaceticum. FEBS Lett. 228, 89±93. 15. Beale, S.I., Foley, T. & Dzelzkalns, V. (1981) d-Aminolevulinic acid synthase from Euglena gracilis. Pr oc. Natl Acad. Sci. USA 78 , 1666±1669. 16. Weinstein, J.D. & Beale, S.I. (1983) Separate physiological roles and subcellular comparttments for two tetrapyrrole biosynthetic pathways in Euglena gracilis. J. Biol. Chem. 258, 6799±6807. 17. Okazaki, T., Kurumaya, K., Sagae, Y. & Kajiwara, M. (1990) Studies on the biosynthesis of corrinoids and porphyrinoids. IV. Biosynthesis o f chlorophyll in Euglena gracilis. Chem. Pharm. Bull. 38, 3303±3307. 18. Oh-hama, T., Seto, H., Otake, N. & Miyachi, S. (1982) 13 C-NMR evidence for the pathway of chlorophyll biosynthesis in green algae. Biochem. Biophys. Res. Commun. 105, 647±652. 19. Porra, R.J., Klein, O. & Wright, P.E. (1983) The proof by 13 C- NMR spectroscopy of the predominance o f the C 5 pathway over the Shemin in chlorophyll biosynthesis in higher plants and of the formation of the methyl ester group of chlorophyll from glycine. Eur. J. Biochem. 130, 509±516. 20. Klein, O. & Porra, R.J. (1982) The participation of the Shemin and C5 pathways in 5-aminolevulinate and chlorophyll formation in higher plan ts and facultative and photosynthetic bacteria. Hoppe- Seyler's Z. Physiol. Chem. 353, 1365±1368. 21. Iida, K. & Kajiwara, M. (2000) Evaluation of biosynthetic path- ways to d- aminolevulinic acid in Propionibacterium shermanii based on bio synthesis of vitamin B 12 from D -[1- 13 C]glucose. Biochemistry 39, 3666±3670. 22. Smith, K.M. (1975) Electronic absorption spectra of chlorins in or ga nic solvents . In Porphyrins and Metalloporphyrins pp. 8 80±881. Else vier Sc ienti®c P ublishing, Amsterdam, T he Netherlands. 23. Katz, J.J. & Janson, T.R. (1973) Chlorophyll±chlorophyll inte r- actions from proton and carbon-13 nuclear magnetic resonance spectroscopy. Ann. NY Acad. Sci. 206, 579±603. 24. Boxer, S.G., Closs, G.L. & Katz, J.J. (1974) The eect of mag- nesium c oordination on the 13 Cand 15 N magn etic resonance spectra of chlorophyll a. The relative energies of nitrogen np* states as deduced from a complete assignment of chemical shifts. J. Am. Chem. So c. 96, 7058±7066. 25. Smith, K.M. & Unsworth, J.F. (1975) The nuclear magnetic res- onance spectra of porphyrins-IX. Carbon-13 NMR spectra o f some chlorins and other chloro phyll degradation products. Tetrahedron 31, 367±375. 26. Wray, V., Ju È rgenns, U. & Brockmann, H. Jr (1979) Electrophilic reactions of chlorin derivatives and a comprehensive collection of 13 C data of these products and closely related compounds. Tetrahedron 35, 2275±2283. 27. Smith, K.M., Bushell, M.J., Rimmer, J. & Unsworth, J.F. (1980) Bacterioch lorophy lls c from Chloropseudomonas ethylicum. Composition and NMR studies of the pheophorbides and deriv- atives. J. Am. Chem. Soc. 102, 2437±2448. 28. Lo È tjo È nen, S . & Hynninen, P.H. (1981) Com plete assignment of the carbon-13 NMR spectrum of chlorophyll a. Org. Magn. Reson. 16, 304±308. 29. Dzelzkalns, V., Foley, T. & Beale, S.I. (1982) d-Aminolevulinic acid synthase of Euglena gracilis: ph ysical and kinetic properties. Arch. Biochem. Biophys. 216, 196±203. 30. Corriveau, J.L. & Beale, S.I. (1986) In¯uence of gabaculine on growth, chlorophyll synthesis, and d-aminolevulinic acid synthase activity in Euglena gracilis. Plant Science 45, 9±17. Ó FEBS 2002 Evaluation of two ALA biosynthetic pathways (Eur. J. Biochem. 269) 297 . Evaluation of two biosynthetic pathways to d-aminolevulinic acid in Euglena gracilis Katsumi Iida, Ippei Mimura and Masahiro Kajiwara Department of. and Shemin pathways both operate in E. gracilis, an d provide information about the biosynthetic pathways leading to the methoxyl carbon of the methoxycarbonyl