Volatile Organic Compounds Derived from 2-Keto-Acid Decarboxylase in Microcystis aeruginosa

Volatile organic compounds (VOCs), 2-methyl-1-butanol, 3-methyl-1-butanol and 2-phenylethanol, were detected together with β-cyclocitral from the cyanobacterium Microcystis aeruginosa NIES-843. These alcohols were optimally produced after 35 d of culture, during which nitrate nitrogen in the cultured broth became exhausted. Additionally, these alcohols were definitely produced using the 2-keto-acid decarboxylase (MaKDC) in Microcystis strains. These results suggested that these VOCs from Microcystis are significant for their lifecycle, because these compounds are not produced by any other genus of cyanobacteria. This is the first report of 2-keto-acid decarboxylase producing 3-methyl-1-butanol and 2-phenylethanol by an oxygenic photosynthetic microorganism.

Microcystis can be a major hazard in recreational lakes, drinking water reservoirs, slow-flowing rivers and wetland areas (5,17,27,28). Microcystis can produce a musty odor, and hepatotoxic microcystin poses serious health risks for animals and humans (4,6); therefore, it is highly desirable to regulate the occurrence of Microcystis. We are attempting to develop a biological control system to decrease the outbreak of Microcystis in a lake. With our system we can better understand the lifecycle of Microcystis (19,24), which is probably regulated by several organic compounds derived from the cyanobacterium itself, and then develop a regulation method based on its lifecycle.
In previous studies, we found that Microcystis strains produced β-cyclocitral, 2-methyl-1-butanol and 3-methyl-1butanol as VOCs (11,18). It was determined for the first time that 2-and 3-methyl-1-butanols were excreted into the extracellular fraction of Microcystis aeruginosa NIES-102 (11,18). The functions and biosynthesis of these alcohols in M. aeruginosa have not yet been elucidated. Watson (29) raised a concern as to whether the VOCs are signals or by-products. Furthermore, we observed this finding in the natural environment in 2008 and detected 2-and 3-methyl-1-butanols together with β-cyclocitral (unpublished result). According to our recent results using seven different genera of cyanobacteria, β-cyclocitral and the 2-and 3-methyl-1butanols were produced only by Microcystis strains (12). In this study, we focused on the elucidation of production behavior and the biosynthetic mechanism of the 2-and 3methyl-1-butanols.
While β-cyclocitral is biosynthetically produced from β-carotene, no origin was identified for the 2-and 3-methyl-1butanols from the Microcystis. Branched-chain alcohols, including 3-methyl-1-butanol and 2-methyl-1-butanol, are known to be produced from the corresponding 2-keto acids by nitrogen starvation in yeast (26). The 2-keto acids are intermediates in the amino acid biosynthesis pathways and can be converted into aldehydes by broad-substrate-range 2keto-acid decarboxylases (KDCs) and then to alcohols by alcohol dehydrogenases (ADHs) (1, 10, 21). 1-Propanol, 2methyl-1-propanol, 1-butanol, and 2-phenylethanol are known to be generated from different 2-keto acids using the same steps (8). These branched-chain alcohols accumulate when the nutrients become limiting in Saccharomyces (9). It has been reported that 3-methyl-1-butanol induced the filamentation of S. cerevisiae with a global transcriptional response (2,9,13,15). In the present study, we attempted to elucidate the relationship among the growth of Microcystis aeruginosa NIES-843, nitrate nitrogen in a cultured broth and the production of the 2-and 3-methyl-1-butanols, and to clarify the enzymes involved in the synthesis of these alcohols in order to elucidate their function. Additionally, we report, for the first time, the genetic identification, purification and characterization of the Microcystis enzymatic activity responsible for the conversion of the 2-keto acids derived from amino acids into aldehydes.
M. aeruginosa NIES-843, whose genomic structure has been completed by Kaneko et al. (14), was obtained from the Microbial Culture Collection at the National Institute for Environmental Studies (NIES Collection, Ibaraki, Japan) and used in this study. The cyanobacterial strains were grown under 35 μmol m −2 s −1 continuous illumination by fluorescent (cool white) light and shaking at 80 rpm in 400 mL CB medium at 30°C (23). To determine the growth curve of the cyanobacteria and the amount of nitrate nitrogen as the nitrogen source in the cultured broth, 10 mL cultured broth was collected. Three milliliters were then transferred to a quartz cell (1 cm) and the OD was measured at 660 nm using a V-530 UV/VIS spectrophotometer (JASCO, Tokyo, Japan). The remaining part was filtered using a GF/A filter (Whatman International, Maidstone, UK) and the nitrate nitrogen in the filtrate was measured using a Digital PACKTEST Multi (Kyoritsu Chemical Check Lab., Tokyo, Japan).
The collected cyanobacteria were directly analyzed as the total amount of VOC, and the filtrates were analyzed as the amount of released VOC in the medium. The intracellular amount of VOC was obtained by subtracting the VOC amount in the culture filtrate from the total VOC amount. A properly diluted aliquot (10 mL) of each sample was subjected to headspace solid-phase microextraction (HS-SPME) coupled with GC/MS for quantitative determination of the following volatile compounds: 2-methyl-1-butanol, 3-methyl-1butanol, 2-phenylethanol, 1-propanol, 2-methyl-1-propanol, and 1-butanol (11).
M. aeruginosa NIES-843 was cultured well, and 10 mL each was collected at 0, 7, 14, 21, 28, 35, 42 d for VOC analysis. On day 14, the cell numbers of M. aeruginosa NIES-843 peaked in the stationary phase (Fig. 1a). As shown in Fig. 1b, nitrate nitrogen as the nitrogen source in the Microcystis culture was consumed by day 12. These VOCs began to increase around 21 d of culture (Fig. 1c, d and e), and the total maximum concentrations of the 2-and 3-methyl-1-butanols and 2-phenylethanol at 35 d were 47.3 ± 16.1, 107.7 ± 27.2 μg L −1 , and 5.38 ± 2.56 mg L −1 , respectively, and were also present in the culture filtrate. 1-Propanol was not detected, while 1-butanol and 2-methyl-1-propanol were detected at low concentrations (0.76 ± 0.03 and 9.72 ± 1.34 μg L −1 , respectively). As a result, it was found for the first time that 2-phenylethanol was excreted into the extracellular fraction of M. aeruginosa NIES-843, which was produced in a higher concentration than the other branched-chain alcohols. Furthermore, the absolute configuration of 2methyl-1-butanol released from the Microcystis was determined to be (S)-(-)-2-methyl-1-butanol using chiral GC/MS (Fig. S1), indicating that this is a single enantiomer derived from the corresponding L-isoleucine; therefore, 3-methyl-1butanol and 2-phenylethanol were derived from L-leucine and L-phenylalanine, respectively, in the Microcystis (3).
No gene encoding a decarboxylase with activity toward branched-chain 2-keto acids in Microcystis has yet been cloned. Searching the entire genome of M. aeruginosa NIES-843 revealed three proteins (MAE36750, MAE50700, and MAE06010) with homology (34%, 24% and 22% identity, respectively) to the 2-keto-acid decarboxylase, KdcA, from Lactococcus lactis B1157 (25). Since MAE50700 and MAE06010 revealed high homology (77% and 83%, respectively) to IlvG and IlvB of Synechocystis sp. strain PCC6803, it is considered that these two enzymes are not 2-keto-acid decarboxylase. BLASTP homology search indicated that MAE36750 (called MaKDC) was homologous to the indolepyruvate decarboxylases and pyruvate decarboxylases found in various organisms. Highly conserved residues involved in catalysis and cofactor binding were observed in MaKDC, showing that MaKDC is a ThDP-dependent decarboxylase. M. aeruginosa NIES-843 genome DNA was obtained as described by Kaneko et al. (14). The MaKDC gene was amplified by PCR from the M. aeruginosa NIES-843 genome using a forward primer, 5'-TAGCCATGGCA ATCACGATCGGCG AATA-3' (NcoI site is underlined), and a reverse primer, 5'-TAGAGATCTAGCAT GGGGTGA ACGTAAAG-3' (BglII site is underlined). The resultant PCR product was digested and cloned into the NcoI-BglII sites of pQE-60 (Qiagen, Hilden, Germany) to yield pQMDC1, in which MaKDC will be produced with the His6 tag at the Cterminus. The transformed E. coli M15 (Qiagen) was grown in 2×YT medium at 37°C in the presence of 75 μg L −1 ampicillin with shaking at 110 rpm. Production of the recombinant protein was induced by the addition of IPTG to the culture at an OD at 660 nm (OD 660 ) of 0.6 to a final concentration of 2 mM, and the culture was allowed to grow for an additional 7 h at 30°C. Analyzing the VOC compounds in the culture of E. coli M15 expressing MaKDC, 2-and 3methyl-1-butanol, and 2-phenylethanol were detected (Fig.  2) using the same method as above. The most abundantly detected compound was 2-phenylethanol, similar to that from the Microcystis culture.
The cells were harvested by centrifugation for 20 min at 4,000×g and then resuspended in sonication buffer (50 mM Tris [pH 8.0], 300 mM NaCl, 10 mM imidazole). After freezing and thawing, the cells were sonicated on ice (6 times for 10 s each) using a Tomy UD-201 sonicator and then centrifuged for 20 min at 20,000×g. The overexpressed protein was purified by nickel chelate chromatography (Qiagen). The sample was loaded onto the column and washed three times with washing buffer (50 mM Tris-HCl [pH 8.0], 300 mM NaCl, and 30 mM imidazole), and then the recombinant protein was eluted with the same buffer containing 250 mM imidazole. Fractions containing the protein were pooled, dialyzed against dialysis buffer (50 mM Tris [pH 8.0], 0.5 mM DTT, 0.1 mM EDTA, 50% glycerol), and then stored at −80°C. The size of the purified recombinant protein on the SDS-polyacrylamide gel was very consistent with the theoretical molecular mass of 61 kDa (Fig. 3).
To test the substrate specificity of the recombinant enzyme, the decarboxylation reactions of the 2-keto acids were performed. The decarboxylase activity of the purified enzyme was measured using 2,4-dinitrophenylhydrazine (DNPH, Panreac, Barcelona, Spain) as previously described (7,16,20). The substrates used to determine the specificity of the 2-keto-acid decarboxylase were 2-ketoisovalerate, 2ketoisocaproate, 2-ketomethylvalerate, 2-ketomethyl-thiobutyrate, phenylpyruvate, indole-3-pyruvate and pyruvate. Typical 200 μL reactions containing 50 mM sodium phosphate buffer (pH 7.0), 30 mM substrate, 5 mM MgCl 2 , 1.5 mM thiamin diphosphate (ThDP) and MaKDC (10 μg L −1 ) and was incubated at 37°C for 20 min. For pyruvate and indole-3-pyruvate, 100 μg L −1 enzyme was added to the reaction mixture. Indole-3-pyruvate was preincubated at 25°C for 45 min, as recommended by Schutz et al. (22), and added to the reaction mixture at a 2 mM final concentration. All the samples were analyzed by reverse-phase high performance liquid chromatography (ELITE LaChrome System, Hitachi Hitech, Tokyo, Japan) using a 60% (v/v) acetonitrile/water mobile phase. Absorbance was measured at 365 nm. One enzyme activity unit was expressed as μmol of isobutyraldehyde produced per min per mg of protein.
As shown in Table 1, activity was the highest with 2ketoisocaproate. The levels of decarboxylase activity for 2ketomethylvalerate, 2-ketoisovalerate, and 2-phenylpyruvate when compared to 2-ketoisocaproate were 35.2, 74.3, and 64.8%, respectively. Decarboxylation activity with indole-3pyruvate markedly decreased to 0.3%. No activity was detected with pyruvate even if 100-fold enzyme concentration was present in the reaction. The enzyme reactions were performed at different 2-ketoisovalerate concentrations (0.1, 0.5, 1, 1.5, 2, 2.5, 5, 7.5, 10 and 20 mM) under standard conditions. The K m and k cat values were calculated by fitting the data to the Lineweaver-Burk linear transformation of the Michaelis-Menten equation. The Km, kcat, and kcat/Km values for the enzyme were 0.52 mM, 7.22 s −1 , and 13.88 s −1 mM −1 , respectively.
In this paper, we reported the relationship among the growth of M. aeruginosa NIES-843, decrease of nitrate nitrogen in a cultured broth and production of VOCs, along with the genetic identification, purification, and characterization of a MaKDC in M. aeruginosa NIES-843. In a separate study, large amounts of these VOCs were detected together with the outbreak of cyanobacteria under natural conditions (unpublished results). Accordingly, it is considered that the 2-and 3-methyl-1-butanols and 2-phenylethanol are significant for Microcystis, because these alcohols are produced only from Microcystis using MaKDC, which shares homology with KDC involved in the lifecycle of yeast and are correlated with the death phase of the Microcystis   lifecycle. In addition, this is the first report of 2-keto-acid decarboxylase producing the 2-and 3-methyl-1-butanols and 2-phenylethanol by an oxygenic photosynthetic microorganism. A future study should focus on the expressional regulation of the MaKDC gene under nutrition-limiting conditions and the 2-and 3-methyl-1-butanols and 2phenylethanol as a function of the signaling molecule.