Journal Information
Journal ID (publisher-id): chemical
Title: Journal of the Korean Chemical Society
Translated Title (ko): 대한화학회지
ISSN (print): 1017-2548
ISSN (electronic): 2234-8530
Publisher: Korean Chemical Society대한화학회
Bacterial bioluminescence recently has been intrigued as it can be utilized a reporter in single cell imaging system.1-4 Light emission in bioluminescent bacteria, catalyzed by bacterial luciferase, involves the oxidation of FMNH2 and long chain aldehyde resulting in the emission of blue-green light5,6 (Fig. 1). The reduced flavin substrate for this reaction is believed to be regenerated by the lux specific flavin reductase and the reduction of fatty acid to aldehyde is catalyzed by a luminescent-specific fatty acid reductase multienzyme complex which has been purified from the bioluminescent bacterium Photobacterium species.5,6 (Fig. 1). The complex is composed of acyl CoA reductase (EC 1.2.1.50), fatty acyl transferase (EC 2.3.1, -) and fatty acyl synthetase subunit (EC 6.2.1.19) coded by luxC, D and E, respectively5,6 (Fig. 2 and 3).
The natural aldehyde for the bioluminescence reaction is believed to be tetradecanal on the basis of identification of this compound in lipid extracts of luminous bacteria, as well as on the basis of the preference for tetradecanal by luciferase at low substrate concentrations and the specificity of the lux-specific fatty acid reductase system which catalyzes the synthesis of the fatty aldehyde substrate5 (Fig. 1). As shown in Fig. 1, the transferase subunit (LuxD) of this complex diverts fatty acid from fatty acid biosynthetic pathways at the level of acyl-ACP into the luminescent system.7 The fatty acids thus produced are activated with ATP by the synthetase subunit (LuxE) and the activated acyl groups are reduced with NADPH by the reductase subunit (LuxC) to the fatty aldehyde substrate of luciferase5,6 (Figs. 2 and 3). The initial step of conversion from tetradecanoic acid to fatty aldehyde involves activation of the fatty acid with ATP by the synthetase (LuxE) to produce an enzyme-bound fatty acyl-AMP intermediate8 (Fig. 3A). In the following reaction, which is greatly stimulated by the presence of the reductase subunit (r), a nucleophile amino acid residue attacks the activated acyl-group producing in an acyl-synthetase.8 The synthetase subunit (LuxE) catalyzes auto-acylation and acyl-transfer as shown in Fig. 3A.
In this research, to analyze the acylation reaction in fatty acid reductase complex, we generated the aldehyde dark mutants by chemical mutagens from the bioluminescence bacterium of Photobacterium leiognathi, investigated the pattern of acylation peptide in fatty acid complex from wild type and mutants of P. leiognathi, and examined the vicinal thiol effect by checking the acyl-CoA activity with the cell extracts from Escherichia coli 43R. Fatty acid reductase activity has been detected in extracts of a number of Photobacterium species and subunits of the fatty acid reductase complex in the different strains can be specifically identified by acylation with fatty acids or fatty acyl-CoA in crude cell extracts.8,9
In this study, we initially checked the fatty acylation of the native P. leiognathi in vivo with [3H] tetradecanoic acid as well as in vitro acylation with [3H] tetradecanoic acid in the presence of ATP or [3H] tetradecanoyl-CoA resulted in the identification of labelled polypeptides on SDS-PAGE corresponding to the reductase (r), transferase (t), and synthetase (s) polypeptides (Figs. 4A and 4B). As shown in Fig. 4, fatty acid reductase complex contains subunits each of the reductase (58 kDa), synthetase (50 kDa), and transferase (42 kDa), respectively. In Fig. 4A, the synthetase and reductase subunits are shown in vivo acylation with [3H] tetradecanoic acid with uninduced P. leiognathi cells (lane 1) and with induced P. leiognathi cells (lane 3) adding of DTT. Similarly, high degree of [3H] synthetase peptide and relativelely low [3H] reductase peptide are shown in uninduced cells (lane 2) and in induced cells with β-mercaptoethanol (lane 4), respectively.
It is known that the relative extent of acylation of the fatty acid reductase polypeptide is depend on the procedure, bacterial species, and extent of growth. For example, the small amounts of acylated reductase subunit in vivo are present compared to synthetase subunit by adding the 0.5 mM DTT or 10 mM β-mercaptoethanol which can play roles as acyl group acceptors (Figs. 4A and 4B). It is reflected to the fact that the reduction of acyl group is specific for NADPH with much lower activity with NADH. Additionally, in the absence of NADPH, the reductase was found to have acyl-transferase activity and could transfer the acyl-group to a thiol acceptor such as β-mercaptoetanol or dithiothreitol.8
The reductase was originally identified by its ability to reduce acyl-CoA reductase in vitro and was designated as an acyl-CoA reductase, however, its function appears to be more directly related to the reduction of fatty acid activated by the synthetase (Figs. 3A and 3B). Transfer of acyl-groups between the synthetase and the reductase is readily reversible9 (Figs. 3A and 4A). Consequently, as shown in lane 2 in Fig. 4B, acylation of the synthetase by acyl-CoA can also occur by transfer of the acyl group from the reductase.
Acyl-transferase yields tetradecanoic acid from the pathway leading to fatty acid biosynthesis (Fig. 1), and the transferase can use various fatty acyl esters as substrates including acyl-ACP, acyl-CoA, acyl-glycerol, acyl-p-nitrophenol and acyl-S-mercaptoethanol.9 It is the reason why the transferase peptide is acylated by [3H] tetradecanoyl-CoA on lane 1 in Fig. 4B. Acyl groups on the reductase as well as the synthetase can readily be removed by neutral hydroxylamine; it appears that the acyl-group on the synthetase may be transferred to cysteinyl residue on the reductase as shown in lane 2 in Fig. 4B.
To investigate more precisely the relation on the acylation in the lux specific fatty acid reductase to the light emit reaction in bioluminescence bacteria, we performed mutational studies to generate the fatty acid reductase mutants from bioluminescent bacterium of Photobacterium leiognathi. Chemical mutagens using nitroso-guanidine (NG) were used to obtain mutants that deficient in bioluminescence in P. leiognathi. Out of 80 mutants, the three fatty acid reductase dark mutants recovering of luminescence by addition of tetradecanal or tetradecanoic acid were screened (M23, M56, and M76). We tested the cell growth and light intensity with P. leiognathi wild type and P. leiognathi mutants of 23, 56, 76. Luminescence expression during the cell growth of P. leiognathi wild type and the mutants, the luminescence in cells is increased by exogenous aldehyde or tetradecanoic acid and these cells have 20% the luminescence in vivo relative to P. leiognathi wild type cells.
At various times during bacterial growth, the in vivo luminescence of 1 ml of culture was measured with/without the addition of tetradecanal or tetradecanoic acid (0.1 mM final concentration), where one light unit (LU) equals 6 × 109 quanta per sec.10 During growth in liquid cultures, the luminescence system is induced, i.e. the LU/absorbance was decreased at the early stage of growth and then light emission per cell can be increased up to 100-fold (Figs. 5A and 5B), due to increase of the intracellular concentration of lux proteins.
Although, the interaction between luciferase and fatty acid reductase has never been directly demonstrated, it has been implied and should be expected to occur in vivo.11 Moreover, the fatty aldehyde product of fatty acid reductase is toxic to the cells and should be channeled directly to luciferase. In this regard, it can be explained the luminescence of aldehyde mutants reached up to 20-30% by addition of tetradecanal or tetradecanoic acid comparing to P. leiognathi wild type (Fig. 5B), even though the mutants contain the same levels of extractable luciferase activity (data not shown).
The P. leiognathi mutant 23 was not growing well reached up to A660 value of 3.5 (Fig. 5A). The light intensity was stimulated by adding tetradecaonic acid instead of tetradecanal, indicating that the gene for acyl transferase are mutated on the other hand the genes for acyl-CoA reductase or acyl-protein synthetase are conserved. The subunit interactions were further investigated by fatty acylation of the protein in vitro at peak luminescence (Fig. 6A). In contrast to showing no acylation on the reductase and/or synthetase subunits from P. leiognathi M56 and M76 on lane 3 and 4 in Fig. 6A, the two subunits proteins are very clearly acylated the cell extract from P. leiognathi wild type and M23 (lanes 1 and 2 in Fig. 6A), by [3H] tetradecanoic acid in presence of ATP. These results suggested that the gene for transferase was mutated and the inactive transferase in these cells may not bind to the synthetase and prevent free fatty acid from accessing the latter. It is the reason why the expression of light emission by luciferase genes in the P. leiognathi transferase mutant of M23 produces luminescence was accomplished by adding tetradecanoic acid, as the ability of acylation remain between synthetase and reductase proteins since the fatty acyl groups are transferred via synthtase and reductase subunits in the enzyme complex.
As the expressions of the luminescence in the P. leiognathi mutant 56 and 76 were tested in Fig. 5B, the addition of fatty aldehyde to the complemented mutants raises light emission to levels approaching wild-type cells up to 20-30% compared to P. leiognathi wild type. In addition, the acylation patterns of the mutants were checked with [3H] tetradecanoic acid or with [3H] tetradecanoyl-CoA in vitro (Figs. 6A and 6B). The mutations in subunit interactions of fatty acid reductase in P. leiognathi were further verified by fatty acylation of the protein at peak luminescence (Fig. 6A).
The cell extract from P. leiognathi mutants 56 and 76, The synthetase and reductase proteins are not acylated (lane 3 and 4 in Fig. 6A). Additionally, as shown in Fig. 6B, the synthetase and reductase proteins are very poorly acylated in vitro by acylation with [3H] tetradecanoyl-CoA but there is visible acylation of the transferase peptide (lane 3 in Fig. 6B). It can be explained that the fact of showing no acylation but the recovering of luminesce by addition of tetradecanal in P. leiognathi 56 and 76 mutants as follow. Transfer of acyl-groups between the synthetase and the reductase is readily reversible,9 however the acylation labelling of synthetase by [3H] acyl-CoA cannot occur by transfer of the acyl group from the reductase land 3 and 4 in Fig. 6B. In sharp contrast to the above, as shown on lane 5 in Fig. 6B, the three subunits of fatty acid reductase were clearly acylated from the cell extracts E. coli 43R,3,10 which show high expression of lux gene from P. leiognathi, transformed with PlXba.pT7-3 containing the whole lux genes for light emission in Fig. 2.
The cysteine residue was demonstrated to be acylated on the synthetase and modification of this site blocked acylation without loss of fatty activation indicating two different domains on the synthetase subunit.9 Therefore, it can be speculated that the addition of fatty aldehyde to the mutants raises light emission to levels approaching wild-type cells and shows that the reconstituted fatty acid reductase complexes in these cells cannot fully complement the fatty aldehyde deficiency of the dark mutants. That is the reason why the P. leiognathi dark mutants are unable to produce the fatty aldehyde substrate of luciferase and cannot be stimulated to luminescence as high as P. leiognathi wild type by the exogenous addition of long chain fatty aldehyde (M56, M76) or fatty acid (M23) to the cells.
The interaction between fatty aldehyde substrate and bacterial luciferase may be mediated indirectly between the synthetase and luciferase through the reductase protein. However, the fate of the fatty acid product of luciferase is not known but it also cannot be allowed to diffuse freely inside the cells.11 It is possibly channeled back to the synthetase subunit of fatty acid reductase, through a direct interaction between luciferase and the synthetase, thus establishing a closed fatty acyl recycling loop.
Acyl groups on the reductase as well as the synthetase can readily be removed by neutral hydroxylamine, suggesting that the acyl-group on the synthetase may be transferred to cysteinyl residue on the reductase. This function originally provided the basis for designating this enzyme as an acyl-CoA reductase. It is interesting to compare the active site of acyl-CoA reductase of bioluminescent bacteria and aldehyde dehydrogenase from rat, human, and Vibrio harveyi(Fig. 7). Aldehyde dehydrogenase catalyzes the dehydro-genation of aldehyde as well as the hydrolysis of esters.12 The Cys302 of aldehyde dehydrogenase was shown to be the active site by site specific mutagenesis and chemical modification.13,14
Aside from the same molecular weight (54 kDa), the long chain aldehyde dehydrogenase has a low degree of similarity in sequence with acyl-CoA reductase.5 Moreover, both enzymes can catalyze the reduction of fatty acyl-CoA to aldehyde in the presence of NADPH and have acyl-transferase and acyl-esterase activities in the absence of NADPH.9 Although the percentage of the amino acids identity between acyl-CoA reductase and aldehyde dehydrogenase less than 10% in active site region, the key cysteine amino acid of acylation site of Cys 286 in acyl-CoA reductase15 and Cys302 in aldehyde dehydrogenase are conserved. In addition, the two amino acids of Ala and Phe which are important in structural integrity nearby acylation site are also existed in all enzymes (Fig. 7).
The region of the active site of acyl-CoA reductase and aldehyde dehydrogenase have some similarity (Fig. 7), suggesting that acyl-CoA reductase and aldehyde dehydrogenase might have evolved from a common thiol esterase precursor. The acylation site of the synthetase has been identified as a conserved cysteine residue (Cys 364) close to the carboxyl terminus.16 For the direct transfer of an acyl-group from the synthetase to the reductase subunit during fatty acid reduction, the acylation site on these two subunits or tetrameric reductase subunit itself must be in close proximity in the enzyme complex or quaternary structure in the reductase subunit.
To check vicinal thiol, we test the effect of phenyl arsene oxide (PhAsO) on the acyl-CoA reductase activity. It was known that PhAsO react with vicinal thiol to form disulfide ring structure,17 resulted in lose enzyme activity. We checked the acyl-CoA reductase activity with the cell extracts from E. coli 43R transformed PlXH.pT7-3 and from PlXba. pT7-33,10 (Fig. 2) by tetradecanoyl-CoA. As you can see in Fig. 8, the acyl-CoA reductase activity was inhibited by increasing the concentration of PhAsO. The enzyme activity was reduced by 50% at the concentration of 0.2 mM PhAsO (Fig. 8). The activity is also restored by following 2,3–dimercapto-1-propane sulfonic acid (DMPS) (Fig. 8). The inhibition of the acyl-CoA reductase activity by PhAsO has suggested that the acylation cysteine site may be covalent disulfide bond by PhAsO. The recovering of the acyl-CoA reductase activity by addition of DMPS, indicated that the disulfide bond is reduced to free –SH at the active cysteine residues.
In order to confirm the correlation between enzyme activity and the acylation pattern of reductase subunit peptide, we checked the acylation with the cell extracts from E. coli 43R transformed with PlXH.pT7-3 and PlXba. pT7-3 by [3H] tetradecanoyl-CoA. The recombinant plasmid PlXba.pT7-3 contains whole genes for luminescence whereas PlXH.pT7-3 does for the luxC and E genes as shown in Fig. 2. In the absence of PhAsO, reductase, synthase, transferase subunits were shown in lanes 2 and 3 from the cell extract of E. coli transformed the above plasmids, however, the acylated peptides were disappeared by addition of DMPS in lanes 4 and 5 in Fig. 9. In parallel with the pattern of recovering of enzyme activities by DMPS in Fig. 8, we obtained the results that the reforming of the acylated [3H] labelled acyl-CoA reductase peptide by addition of DMPS as shown in lanes 6 and 7 in Fig. 9. This result is consistent with the inhibition of the acyl-CoA by PhAsO as shown in Fig. 8, and it is owing to the fact that the reduced cysteine thiols in acyl-CoA reductase subunit are regenerated by DMPS and then to be acylated by [3H] tetradecanoyl-CoA (Fig. 9).
In summary, fatty acid activation, acyl group transfer, and reduction of fatty acids to fatty aldehyde by the fatty acid reductase complex from P. leiognathi species is essential for a bacterial bioluminescence. The presented data including the expression of the light intensities during cell growth and the acylation patterns of fatty acid reductase subunits with mutants and wild type of P. leiognathi as well as inhibition of vicinal effect in acyl-CoA reductase activity by PhAsO, suggested that the direct transfer of an acyl-group from the synthetase to the reductase subunit during fatty acid reduction, the acylation site on these two subunits or tetrameric reductase subunit must be in close proximity in the complex. Studies on the structure and mechanism of fatty acid reductase complex, especially the acylation study of subunit interactions between acyl-CoA reductase subunit and acyl protein transferase as well as fatty acyl synthase, and vicinal thiol effect in acyl-CoA reductase activity will provide valuable information and model on the nature of their function and regulation of multi-enzyme complex.
2,3-dimercapto-1-propane sulfonic acid (DMPS), Phenyl arsene oxide (PhAsO), tetradecanal, and tetradecanoic acid were purchased from Sigma Chemical Company. [3H] tetradecanoic acid (26 Ci/mmol) and [3H] tetradecanoyl-CoA were prepared by New England Nuclear (NEN) Corp.
The strains of Photobacterium leiognathi used in this study were the strain of ATCC 25521 and the isogenic dark mutants M23, M56, and M76. Briefly, the three mutants obtained by nitrosoguanidine mutagenesis are dark mutants with defective fatty acid reductase level and recovering of light intensity by addition of tetradecanal or tetradecanoic acid. P. leiognathi (ATCC25521) was obtained from American Type Culture Collection. The mutant of E. coli RR1, designated E. coli 43R, was used for propagation of the recombinant plasmids. The recombinant plasmid pT7-3, PlXH.pT7-3, and PlXba.pT7-3 are described in author’s previous paper.3,10
P. leiognathi cells were grown at 30 ℃ in complex medium containing 5 g of yeast extract, 2 ml of glycerol, 10 g of NaCl, 3.7 g of Na2HPO4, 1g KH2PO4, 0.5 g of (NH4)2HPO4, and 0.1 g MgSO4 per liter. Whereas, the E. coli 43R transformed with PlXba in pT7-3 and PlXH in pT7-3 plasmids for the expression of lux C, D and E genes were grown LB media with adding 100 μg ampicillin/ml.
For in vivo fatty acylation, 2 ml 3.7 mM [3H] tetrade-canoic acid (23 Ci/mmol) was added to 400 ml of cells at park luminescence (OD660 = 2.5). After 10 minutes the cells were isolated by centrifugation and washed with 500 ml cold 1% complex containing 1 mg/ml bovine serum albumin (Boehringer). For in vitro fatty acylation 25 μl of cell extracts was added to a total volume of 100 μl containing 18μM [3H] tetradecanoic acid (23 Ci/mmol) plus 2.5 mM ATP or 22 μM [3H] tetradecanoyl-CoA. The final cell pellet was resuspended in 50 ml 1 mM phosphate pH 7 containing 10 mM N-ethylmaleimide. The cells were lysed by sonication in a water bath sonicator for 60 sec. The cell debris was removed by centrifugation and the protein concentration in the supernatant was adjusted to 6 mg/ml.
Acyl-CoA reductase activity was conducted at room temperature. Aliquots of the cell extract were added to 0.002% tetradecanal in 1 ml of 50 mM phosphate (pH 7) containing 0.2% bovine serum albumin. Acyl-CoA reductase activity was measured by a luciferase coupled assay in above solution. Extracts were mixed with 4 μg of P. phosphoreum luciferase, 5 μM tetradecanoyl-CoA and 0.1 mM NADPH. Light emission was then measured on injection of FMNH2 into 1.0 ml of the assay mixture as described above.
The samples were finally diluted with one volume of SDS-PAGE sample buffer (0.1 M Tris pH 6.8, 2.5% SDS, 25% (v/v) glycerol, 0.1% Bromophenol blue) containing 10 mM N-ethylmaleimide. The protein was separated by SDS-PAGE using the Bio-Rad Mini-Protean apparatus with 4% stacking and 12% separating gels. The gels were stained with Coomassie brilliant blue R, soaked in En3Hance (DuPont), dried, and exposed overnight with Kodak X-OMAT AR films.
C. Gregor K. C. Gwosch S. J. Sahl S. W. Hell Proc. Natl. Acad. Sci. U S A.2018115962 [CrossRef]
C. Gregor J. K. Pape K. C. Gwosch T. Gilat S. J. Sahl S. W. Hell Proc. Natl. Acad. Sci. U S A201911626491 [CrossRef]
A. C. Love J. A. Prescher Cell Chem. Biol.202027904 [CrossRef]
E. A. Meighen Microbiol. Rev.199155123 [CrossRef]
E. Brodl A. Winkler P. Macheroux Comput. Struct. Biotechnol. J.201816551 [CrossRef]
A. Rodriguez L. Wall S. Raptis C. G. Zarkadas E. Meighen Biochimica. Et. Biophysica. Acta.1988964266 [CrossRef]
L. Wall E. A. Meighen Biochemistry1986254315 [CrossRef]
C. Y. Lee R. B. Szittner E. A. Meighen Eur. J. Biochem.1991201161 [CrossRef]
S. Ulitzur J. W. Hastings Proc. Natl. Acad. Sci. U.S.A.U.S.A.197976265 [CrossRef]
S. Ulitzur J. W. Hastings Proc. Natl. Acad. Sci. U.S.A.U.S.A.197875266 [CrossRef]
J. Hempel H. Nicholas H. Jornvall Proteins199111176 [CrossRef]
E. E. Blater D. P. Abriola R. Pietruszko Biochem. J.1992282353 [CrossRef]
R. R. Soly E. A. Meighen J. Mol. Biol.199121969 [CrossRef]
M Jauiainen K. T. Stevenson P. J. Dolphin J. Biol. Chem.19882636525 [CrossRef]