OUP user menu

Biochemical and molecular characterization of α-ketoisovalerate decarboxylase, an enzyme involved in the formation of aldehydes from amino acids by Lactococcus lactis

Marta de la Plaza, Pilar Fernández de Palencia, Carmen Peláez, Teresa Requena
DOI: http://dx.doi.org/10.1111/j.1574-6968.2004.tb09778.x 367-374 First published online: 1 September 2004


In this paper, we report for the first time on the identification, purification, and characterization of the α-ketoisovalerate decarboxylase from Lactococcus lactis, a novel enzyme responsible for the decarboxylation into aldehydes of α-keto acids derived from amino acid transamination. The kivd gene consisted of a 1647 bp open reading frame encoding a putative peptide of 61 kDa. Analysis of the deduced amino acid sequence indicated that the enzyme is a non-oxidative thiamin diphosphate (ThDP)-dependent α-keto acid decarboxylase included in the pyruvate decarboxylase group of enzymes. The active enzyme is a homo-tetramer that showed optimum activity at 45 °C and at pH 6.5 and exhibited an inhibition pattern typical for metal-dependant enzymes. In addition to Mg2+, activity was observed in presence of other divalent cations such as Ca2+, Co2+ and Mn2+. The enzyme showed the highest specific activity (80.7 U mg−1) for α-ketoisovalerate, an intermediate metabolite in valine and leucine biosynthesis. On the other side, decarboxylation of indole-3-pyruvate and pyruvate only could be detected by a 100-fold increase in the enzyme concentration present in the reaction.

  • α-Ketoisovalerate decarboxylase
  • Lactococcus lactis
  • Amino acid catabolism

1 Introduction

Aroma development in cheese is both a matter of scientific interest and of economical and consumer concern. Consumer trends demand not only safe products that feature acceptable sensory quality but a range of products that meet high flavor standards. Dairy research efforts are recently addressing the development of technologies designed to control flavor development in cheeses and to develop new cheese flavors [1]. Formation of aroma compounds during cheese ripening progresses along with the development of proteolysis, lipolysis and glycolysis. Amino acid catabolism is the final step of proteolysis. In lactic acid bacteria, it is mainly initiated by transamination that converts amino acids into α-keto acids [24], which are further converted into hydroxyacids or into flavor compounds such as carboxylic acids by an oxidative decarboxylation, or aldehydes by an enzymatic non-oxidative decarboxylation. Conversion of α-keto acids to aldehydes may also be chemically feasible [5,6]. The aldehydes can be further reduced to alcohols or oxidized to carboxylic acids [7].

α-Keto acid decarboxylases are rare in bacteria, being more frequent in plants, yeasts and fungi [810]. Several enzymes involved in the catabolism of α-keto acids derived from aromatic and branched-chain amino acids have been identified in Saccharomyces cerevisiae. The enzyme phenylpyruvate decarboxylase of S. cerevisiae responsible for the decarboxylation of phenyl pyruvic acid to phenylacetaldehyde was genetically characterized [11,12]. Five additional α-keto acid decarboxylases have been identified in this microorganism, namely, pyruvate decarboxylases PDC1, PDC5 and PDC6, α-ketoisocaproate decarboxylase KID1 and the pyruvate decarboxylase-like enzyme YDR380w, which are responsible for the decarboxylation of the branched-chain α-keto acids [1315]. Pyruvate decarboxylase activity has been thoroughly described in Zymomonas mobilis [8,1619]. The enzyme was highly specific with pyruvate, showing no activity against branched-chain or aromatic α-keto acids [9,17]. This high preference for a specific susbstrate was also shown by the indole-3-pyruvate decarboxylase of Enterobacter cloacae that specifically converts indole-3-pyruvate into indole-3-acetaldehyde and to a much lesser extent, pyruvate into acetaldehyde [5,20]. In general, the high specificity of microbial decarboxylases makes their presence a critical step in the amino acid catabolism pathways [9,11].

Concerning cheese bacteria, production of aldehydes from α-keto acid decarboxylation was first reported in Lactococcus lactis var. maltigenes[21], then in Lactobacillus casei[22] and more recently in some lactococcal wild isolates and one strain of Corynebacterium [2325], [P. Fernández de Palencia, M. de la Plaza, F. Amárita, T. Requena and C. Peláez, unpublished results]. The activity seems not to be present in propionibacteria [26]. We have demonstrated previously that the wild isolate L. lactis IFPL730 converted enzymatically methionine to methional in a process mediated by aminotransferase and α-keto acid decarboxylase activities [24]. In the work by Amárita et al. [24] it was shown that the enzyme involved in the α-keto acid decarboxylase activity from L. lactis IFPL730 shows preference for branched-chain α-keto acids rather than methionine. In later work we demonstrated the ability of this strain to produce methional and the derived 3-methylthiopropanol in cheese slurries added with resting cells and the intracellular fraction of this bacterium [27]. We have also demonstrated enhancement of 2-methylbutanal formation by L. lactis IFPL730 in cheese when used as starter a lactococcal strain producing the bacteriocin lacticin 3147 [28].

In this paper, we report for the first time on the genetic identification, purification and characterization of the Lactococcus enzymatic activity responsible for the conversion into aldehydes of α-keto acids derived from amino acids. The enzyme is a non-oxidative thiamin diphosphate (ThDP)-dependent α-keto acid decarboxylase, which shows highest affinity for α-ketoisovalerate, an intermediate compound in valine and leucine biosynthesis. Therefore, we have named the enzyme as α-ketoisovalerate decarboxylase (Kivd).

2 Material and methods

2.1 Bacterial strains and culture conditions

Lactococcus lactis IFPL730 is a wild isolate from raw goat's milk cheese [29]. L. lactis strains (IFPL730 and IL1403) were cultured in M17 broth (Oxoid) supplemented with 5 g l−1 glucose. Escherichia coli strains NovaBlue GigaSingles™ and BL21(DE3) (Novagen) were employed as cloning and expression hosts, respectively. Cells were routinely grown at 37 °C in Luria-Bertani medium [30] supplemented with 12.5 μg ml−1 tetracycline, 30 μg ml−1 kanamycin and/or 1 mmol l−1 isopropyl β-d-thiogalactopyranoside (IPTG) when appropriate.

2.2 Enzyme purification and amino acid sequencing

Cells from 5 l of a L. lactis IFPL730 culture at late exponential growth phase (OD660= 1.0) were harvested by centrifugation (7500g, 20 min, 4 °C). The pellet was washed twice with 50 mmol l−1 sodium phosphate buffer, pH 6.5. For disruption, the cells were resuspended in 50 ml of phosphate buffer added with 5 mmol l−1 MgCl2, 1 mmol l−1 thiamin diphosphate (ThDP) and passed through a French pressure cell at 12,000 lb in−2. Unbroken cells and cells debris were removed by centrifugation (17,000g, 20 min, 4 °C) and the clear supernatant, which constituted the cell-free extract (CFE), was maintained at −80 °C until use.

CFE was fractionated by ammonium sulfate precipitation at two concentrations (23% and 35% w/v). The enzymatically active protein fraction was obtained in the 35% ammonium sulfate supernatant collected by centrifugation (20,000g, 20 min, 4 °C). The active fraction was desalted by passing through a Sephadex G-25 PD-10 columns (Amersham Biosciences) and then, it was applied twice to a Mono Q HR 5/5 (Amersham Biosciences) column equilibrated with 20 mmol l−1 Bis-Tris propane, pH 6.5. The bound proteins eluted at a flow rate of 0.5 ml min−1 at 4 °C in a linear NaCl gradient (0-1 mol l−1) in the same buffer. Fraction containing α-ketoisovalerate decarboxylase activity, which eluted at 0.1 mol l−1 NaCl, was desalted, liophilized and stored at −80 °C for further studies. The purity and molecular weight of the active fraction was monitored by SDS-PAGE, by using the Phast System cell (Amersham Biosciences) with 12% polyacrylamide gels. SDS molecular weight markers (molecular mass 14–94 kDa; LMW calibration kit, Amersham Biosciences) were used as reference proteins.

Active fraction was transferred to a polyvinylidene difluoride membrane (PVDF) (Bio-Rad) at 200 V for 1 h using a Mini Trans-Blot Electrophoretic Transfer Cell (Bio-Rad). Blotting occurred in 50 mmol l−1 Tris-HCl, pH 8.2, containing 50 mmol l−1 boric acid and the membrane was stained with bromophenol blue. The N-terminal amino acid sequence of the separated proteins was determined by Edman degradation analysis at the Protein Sequence Service in the Centro de Investigaciones Biológicas (CSIC).

Protein content was determined by the micromethod of Bradford [31] employing the Bio-Rad (Bio-Rad, Munich, Germany) protein assay and bovine serum albumin (fraction V; Sigma) as standard.

2.3 Cloning of the kivd gene

L. lactis chromosomal DNA was obtained as described by Leenhouts et al. [32]. E. coli plasmid isolation was carried out with the QIAprep Spin Miniprep Kit (Qiagen). DNA manipulations as well as transformation of E. coli cells were carried out by standard methods [30].

The sequenced 20 N-terminal amino acids of the isolated α-ketoisovalerate decarboxylase were identified by sequence similarity with the deduced amino acid sequence encoded by the ipd gene in the L. lactis IL1403 genome (Accession Number NC_002662). The primers kivd1 and kivd2 (Table 1) designed to amplify the ipd gene in L. lactis IL1403 failed to give the expected 1374 bp product using L. lactis IFPL 730 chromosomal DNA. The primer pairs kivd3 and kivd4 (Table 1) located at the starting of the ipd gene and at the ThDP-motif (Prosite PDOC00166) encoding sequence in L. lactis IL1403 were used to amplify a 1249 bp fragment in L. lactis IFPL 730 chromosomal DNA. The remaining part of the kivd gene and the adjacent regions were amplified by inverse PCR with a re-ligated BfaI digest of chromosomal DNA of L. lactis IFPL730 and the primers kivd5 and kivd6 (Table 1) derived from the nucleotide sequence of the 1249 bp fragment from L. lactis IFPL730. The obtained 756 bp PCR fragment was used to design the 3′-terminal primer kivd7 (Table 1). A 1647 bp DNA fragment containing the entire kivd gene was amplified by PCR using L. lactis IFPL 730 chromosomal DNA as a template and the primer pair kivd1 and kivd7. The PCR product was inserted into pET-30 EK/LIC vector (Novagen) according to the manual instructions, resulting in pET-30 EK/LIC::His6-kivd. The recombinant plasmid pET-30 EK/LIC::His6-kivd, encoding the full-length α-ketoisovalerate decarboxylase protein with an additional C-terminal (His)6 tag, was used to transformate by incubation at 42 °C for 30 s E. coli NovaBlue GigaSingles™ (Novagen) for gene cloning and E.coli BL21(DE3) for expression experiments. The correct Kivd encoding gene cloned in E. coli was confirmed by DNA sequencing.

View this table:
Table 1

Primers used in this study

OligonucleotideNucleotide sequence (5′→ 3′)T Annealing (°C)a
  • a DNA strand separation was carried out at 94 °C for 30 s, annealing at the indicated temperature for 30 s and elongation occurred with Taq DNA polymerase (Roche) at 72 °C during 60 s. Thirty cycles were performed.

  • b PCR was carried out with kivd1.

Sequencing of PCR fragments was done at least twice for both strands on all amplified fragments extracted from 0.7% agarose gels with QIAquick Gel Extraction Kit (Qiagen). Sequencing was performed at the DNA Sequence Service (Centro de Investigaciones Biológicas - CSIC).

2.4 Expression of the L. lactis kivd gene in E. coli and purification of the recombinant α-ketoisovalerate decarboxylase

Expression of the L. lactis kivd gene was induced by adding IPTG (final concentration 1 mmol l−1) to an E. coli BL21(DE3)pET-30 EK/LIC::His6-kivd culture grown to an OD600 of approximately 0.4. After 4 h of further incubation, the cells were harvested by centrifugation (10,000g, 10 min). The production of the target protein as well as its cellular localization were verified by the analysis of total cell protein, of the soluble and insoluble cytoplasmic fractions and of the culture supernatant according to the pET System Manual (10th ed., http://novagen.com). The cell pellet was resuspended in 2.5 ml g−1 (wet weight) of 0.02 mol l−1 sodium phosphate, pH 7.4, 0.5 mol l−1 NaCl, 0.04 mol l−1 imidazole, then mixed (1:1, w/v) with glass beads (diameter, 150–212 μm; Sigma) and vortexed. Cell debris and glass beads were collected by centrifugation (12,000g, 5 min). Purification of the recombinant Kivd protein was carried out from the soluble cytoplasmic fraction by immobilized metal affinity chromatography (IMAC), employing a 1-ml HiTrap TM Chelating HP column (Amersham Biosciences). The enzyme was eluted with 0.2 mol l−1 imidazole and the solution was desalted (PD-10 column) and stored in 0.02 mol l−1 sodium phosphate, pH 6.5, at −80 °C. The negative cloning and expression control was done with the EK/LIC β-gal Control Insert.

2.5 α-Ketoisovalerate decarboxylase activity assay

Standard enzymatic reactions (0.25 ml) were performed in 50 mmol l−1 sodium phosphate buffer, pH 6.5, 30 mmol l−1α-ketoisovalerate, 5 mmol l−1 MgCl2, 1.5 mmol l−1 thiamin diphosphate (ThDP) and appropriately diluted enzyme solution (1 μg ml−1) from purified batches as described above. Samples were incubated at 37 °C for 20 min and the reaction was stopped by adding ethyl acetate (1:1, v/v) prior to chromatographic analysis. To detect isobutyraldehyde production, samples were derivatized with 2,4-dinitrophenylhydrazine (DNPH, Panreac) to form the corresponding dinitrophenylhydrazone derivative as described by Kuntz et al. [33]. The chromatographic conditions employed were those described by Schmidt et al. [34]. Isocratic elution of compounds followed at a flow rate of 1.5 ml min−1, using 60% (v/v) acetonitrile in Milli-Q water as mobile phase. Absorbance was measured at 365 nm. A Jasco RP-HPLC system (Jasco Co., Tokyo, Japan) and a Borwin chromatography software data acquisition system (JBMS Developpments, Grenoble, France) were used. The results were expressed in μmol of isobutyraldehyde produced per min (U) per mg of protein.

2.6 Enzyme characterization

2.6.1 Molecular mass determination

The molecular mass of the enzyme was estimated after gel filtration with the Superdex 200 HR 10/30 column (Amersham Biosciences). Enzyme solution (50 μl) from purified batch was injected onto the column, and the elution was performed at a rate of 0.5 ml min−1 with 50 mmol l−1 sodium phosphate, pH 6.6, containing 0.15 mol l−1 NaCl. The column had been calibrated previously under similar conditions with a mixture of marker proteins that included thyroglobulin (670 kDa), ferritin (440 kDa), catalase (232 kDa) and aldolase (158 kDa) (Amersham Biosciences).

2.6.2 pH and temperature

The effect of pH on the catalytic activity of the recombinant enzyme was investigated in sodium citrate buffer (pH 3–6), sodium phosphate buffer (pH 6.5–7.5), Tris-HCl buffer (pH 8) and glycine-NaOH buffer (pH 9), each at 50 mmol l−1 in standard conditions. α-Ketoisovalerate decarboxylase activity was assayed from 25 to 50 °C in standard conditions.

To estimate pH and temperature stability, the enzyme was preincubated at pH values ranging from 4 to 7.5 (at 37 °C) and between 25 and 45 °C (at pH 6.5), after which the remaining activity was determined in standard conditions. The α-ketoisovalerate decarboxylase activity was also measured at cheese-like conditions (pH 5.4 and 0.7 mol l−1 NaCl) at 12, 25 and 37 °C.

2.7 Effect of inhibitors and bivalent cations

The effect of inhibitors and bivalent cations (Table 2) was studied after incubating the enzyme in the standard reaction at final concentrations of 1 or 10 mmol l−1.

View this table:
Table 2

Effect of inhibitors and metal ions (1 mol l−1) on the activity of recombinant α-ketoisovalerate decarboxylase activity from L. lactis IFPL730

SubstanceRelative activity (%)a
Dithiothreitol (DTT)103.9 ± 0.2
Phenylmethanesulfonyl fluoride (PMSF)71.7 ± 0.1
p-(Hydroxymercuri)benzoic acid (pHMB)59 ± 0.2
Phosphoramidon52 ± 0.3
1,10-Phenanthrolineb4.5 ± 0.0
Ethylenediaminetetraacetate (EDTA)b4.3 ± 0.0
Na+121.5 ± 0.2
Ca2+123.7 ± 0.2
Co2+110.3 ± 0.2
Mg2+107.9 ± 0.1
Mn2+100.2 ± 0.2
Fe2+91.1 ± 0.0
Zn2+65.7 ± 0.0
Cu2+5.2 ± 0.0
  • a Activity on α-ketoisovalerate in absence of metal ion or inhibitor was taken as 100% (80.7 U mg−1).

  • b At 10 mmol l−1.

2.7.1 Substrate specificity

The substrates used to determine the specificity of the α-ketoisovalerate decarboxylase were α-ketoisovalerate, α-ketoisocaproate, α-ketomethylvalerate, α-ketomethylthiobutyrate, phenylpyruvate, indole-3-pyruvate and pyruvate. Standard reactions were carried out for each substrate except for pyruvate and indole-3-pyruvate where 100 μg ml−1 enzyme were added in the reaction mixture. Indole-3-pyruvate was preincubated at 25 °C for 45 min as recommended by Schütz et al. [35] and added to the reaction mixture at 2 mmol l−1 final concentration.

2.7.2 Kinetics

Enzyme reactions were performed with different α-ketoisovalerate concentrations (0.1, 0.5, 1, 1.5, 2, 2.5, 5, 7.5, 10 and 20 mmol l−1) and under standard conditions. Km and Vmax values were calculated by fitting data to the Lineweaver-Burk linear transformation of the Michaelis-Menten equation.

3 Results and discussion

3.1 α-Ketoisovalerate decarboxylase isolation

The purification to homogeneity of the enzyme from L. lactis IFPL730 was hindered during the purification process by FPLC due to the loss of the activity after the second chromatographic step, as it also occurred in the work of Amárita et al. [24]. Therefore, the proteins included in the last active fraction were separated by means of SDS-PAGE and transferred onto a PVDF membrane in order to determine their amino-terminal amino acid sequence. The main bands obtained corresponded to proteins of 14, 30 and 60 kDa (results not shown). The Blast search with the 20 amino acids sequenced from the 60 kDa protein indicated a 100% identity with the deduced N-terminal amino acid sequence of the gene ipd identified in L. lactis IL1403. The search indicated that the protein could be included in the pyruvate decarboxylase group of enzymes requiring thiamin diphosphate (ThDP). These enzymes are characterized by the ThDP signature pattern (Prosite PDOC00166). This motif corresponds to positions 412–431 in the L. lactis IL1403 indole-3-pyruvate decarboxylase deduced amino acid sequence (NP_267460). Therefore, the 60 kDa protein was considered to be the likely candidate enzyme for the α-ketoisovalerate decarboxylase activity.

3.2 Genetic characterization of kivd

Primers used for kivd cloning are shown in Table 1. L. lactis IFPL730 total DNA failed to amplify using the PCR primers kivd1 and kivd2 that produced a 1374 bp fragment including the complete gene ipd in L. lactis IL1403 (results not shown). A new primer pair, kivd3 and kivd4, designed at the starting of the ipd gene and at the ThDP-motif sequence in L. lactis IL1403, succeeded in amplifying a 1249 bp fragment in L. lactis IFPL730. The DNA sequence of this fragment was determined and used to design the primers kivd5 and kivd6 (Table 1), which allowed amplification of a 756 bp fragment by inverse PCR (see Section 2) that contained the remaining part of the gene, as well as part of the adjacent regions. The sequence of the overall region (1954 bp) contained one 1647-bp open reading frame (ORF) designated kivd (Accession Number AJ746364). The ORF encoded a predicted 548-amino acid protein with a deduced molecular mass of 61 kDa, in agreement with the molecular mass of the 60 kDa protein included in the MonoQ active fraction from L. lactis IFPL730. The N-terminal deduced amino acid sequence of Kivd (20 amino acids) showed 100% identity to the N-terminal amino acid sequence of the 60 kDa protein included in the MonoQ active fraction. BlastP analysis of the deduced amino acid sequence of the kivd gene from L. lactis IFL730 revealed significant sequence identities to yeast and bacterial pyruvate and indolepyruvate decarboxylases. The highest homologies were 40% identity to indolepyruvate decarboxylase from E. cloacae, 38% to pyruvate decarboxylase (PDC) from Schizosaccharomyces pombe and 37% to PDC1 from S. cerevisiae and PDC from Kluyveromyces lactis. A comparison of Kivd with the deduced 457-amino acid sequence of the ipd gene of L. lactis IL1403 (Accession Number NC_002662) showed 98.6% identity within the starting 438 amino acids. At position L439 in L. lactis IL1403, the gene ipd is interrupted by the insertion of an IS983 element. A Blast search of the additional 330 bp of the gene kivd from L. lactis IFPL 730 revealed that this part of the gene is not present in the L. lactis IL1403 genome sequence, indicating that in this microorganism the enzyme C-terminal region would have been lost due to genomic recombination between transposable elements.

Analysis of the DNA sequence upstream of kivd identified a putative ribosome binding site and −10 and −35 promoter regions. Downstream of the ORF a putative ρ-independent transcriptional terminator with a ΔG of −14.5 kcal mol−1 was identified. These results indicate that the gene is transcribed monocistronically. No ORFs were identified in the 154 bp sequence upstream from kivd or the 153 bp downstream from the ORF.

3.3 Characterization of the recombinant Kivd

Analysis by SDS-PAGE of the protein fractions of an IPTG induced culture of E. coli BL21(DE3) harbouring pET-30 EK/LIC::His6-kivd (see Section 2) showed a strong signal in the soluble cytoplasmic fraction at a molecular weight of about 66 kDa (Fig. 1). This size is in agreement with the calculated molecular mass (64.5 kDa) deduced from the kivd DNA sequence, extended by the vector encoded amino acids. Analysis of induced E. coli cells bearing pET EK/LIC β-gal Control Insert showed no remarkable signal at the mentioned position in any cellular fraction, confirming the strong band to originate from the cloned gene. Besides, only E. coli cells bearing pET-30 EK/LIC::His6-kivd produced isobutyraldehyde from α-ketoisovalerate, providing genetic evidence for the gene encoding the enzyme. The C-terminal (His)6 tag facilitated the purification to homogeneity of the recombinant α-ketoisovalerate decarboxylase from the protein extract of induced cells by affinity chromatography (IMAC). The pure fraction, which had a specific activity of 80.7 U mg−1 of protein with α-ketoisovalerate as substrate was used for characterization.

Figure 1

SDS-PAGE of purified protein fraction of IPTG induced Escherichia coli BL21(DE3) harbouring the recombinant α-ketoisovalerate decarboxylase. Lanes: 1, protein standard; 2, soluble cytoplasmic fraction eluted with 0.02 mol l−1 sodium phosphate, pH 7.4, 0.5 mol l−1 NaCl, 0.2 mol l−1 imidazole. Numbers indicated the molecular mass (kDa) of the standard proteins. The 12% polyacrylamide gel was stained with Coomassie blue.

The apparent molecular mass of the recombinant enzyme was estimated to be 240 kDa at pH 6.6 by gel filtration, indicating that Kivd from L. lactis IFPL730 associates as a homo-tetramer with a deduced subunit protein of 548 amino acids. Both characteristics are common in bacterial and yeast pyruvate decarboxylases and indolepyruvate decarboxylases [8,10,20,36]. Crystal structures of the indolepyruvate and pyruvate decarboxylases have demonstrated that the active tetramers are assembled by binding the cofactors ThDP and Mg2+ [10,16,37], whereas monomers and dimers are inactive [5].

The recombinant Kivd was active over a broad range of temperature values with a maximum of activity (determined at pH 6.5) at 45 °C (Fig. 2), although the enzyme showed very low thermostability since a complete inactivation was registered after heating the enzyme at 45 °C during 30 min. Therefore, activity was measured at 37 °C for enzyme characterization. Optimal pH of activity (determined at 37 °C) was at 6.5 (Fig. 2) with reduced activity at pH values over 7.5 and below pH 5.5. Loss of activity at both basic and acidic pH values could be related to a pH-dependent subunit association equilibrium of the enzyme as it has been described for the E. cloacae indolepyruvate decarboxylase tetramer structure, which is stabilized in the range pH 5.6–7.5 [35].

Figure 2

Effect of temperature (▪) and pH (▲) on the activity of the recombinant α-ketoisovalerate decarboxylase. Activities are expressed relative to the maximum activity.

Activity of the enzyme was also analyzed at conditions of pH and NaCl concentration that simulated those present in cheese (pH 5.4 and 0.7 mol l−1 NaCl). Incubation of the enzyme in the presence of NaCl increased by 3-fold the enzyme activity registered at pH 5.4.

3.3.1 Effect of ions and potentially inhibitory agents on the recombinant enzyme

Table 2 shows the effect of some inhibitors and metal ions (at 1 mmol l−1) on the activity of the recombinant Kivd. The inhibition pattern suggested that it is a metal-dependent enzyme. Among the inhibitors, phophoramidon was more effective for inhibition of the enzyme than EDTA and 1,10-phenanthroline, that did not exert any inhibitory effect at 1 mmol l−1. Enzyme inactivation was observed by raising the inhibitor concentration to 10 mmol l−1, that could suggest that the mineral Mg2+ is rather tightly bound to the enzyme. In addition to Mg2+, activity was observed in presence of other divalent cations such as Ca2+, Co2+ and Mn2+, whereas activity was inhibited by Cu2+ and reduced by Zn2+. These ions probably bind to the enzyme but produce an inactive holoenzyme. Activity was also stimulated by Na+. Dithiothreitol (DTT) was the only agent that acted as an activator, which could be due to its reducing character, whereas pHMB, which has an oxidizing effect, reduced the activity.

3.3.2 Substrate specificity and enzyme kinetics

Table 3 shows results of substrate specificity of the recombinant enzyme. The activity was highest with α-ketoisovalerate and was 4- and 6-fold less active with the other two branched-chain α-keto acids. Low activity was also determined with methionine and phenylalanine derived α-keto acids. Decarboxylation of indole-3-pyruvate and pyruvate only could be detected by increasing 100-fold the enzyme concentration present in the reaction. The enzyme showed specific activities of 0.07 and 0.46 U mg−1 with indole-3-pyruvate and pyruvate, respectively. The Km and Vmax for the enzyme, using different concentrations of α-ketoisovalerate, were 1.9 mmol l−1 and 117.9 μmol min−1 per mg protein, respectively. High substrate specificity for pyruvate has been described for the pyruvate decarboxylase of Z. mobilis [9,17] and for indole-3-pyruvate for the E. cloacae indolepyruvate decarboxylase [5]. In S. cerevisiae there is a phenylpyruvate decarboxylase specific for the decarboxylation of phenylpyruvate [11,12]. Therefore, the highest specificity of the enzyme for α-ketoisovalerate, an intermediate metabolite in both valine and leucine biosynthetic pathways, would indicate that it is the native substrate of the enzyme and the decarboxylase from L. lactis IFPL730 should be named α-ketoisovalerate decarboxylase.

View this table:
Table 3

Substrate specificity of recombinant α-ketoisovalerate decarboxylase activity from L. lactis IFPL730

SubstrateSpecific activity (U mg−1)Relative activity (%)a
α-Ketoisovalerate80.7 ± 5.7100 ± 7.1
α-Ketoisocaproate18.3 ± 3.122.7 ± 3.9
α-Ketomethylvalerate13.5 ± 1.316.7 ± 1.6
α-Phenylpyruvate7.1 ± 0.38.8 ± 0.4
α-Ketomethylthiobutyrate5.8 ± 1.27.2 ± 1.5
Pyruvateb0.46 ± 0.00.6 ± 0.0
Indole-3-pyruvateb0.07 ± 0.00.1 ± 0.0
  • a Expressed as a percentage of maximum activity measured towards α-ketoisovalerate.

  • b Reaction with 100 μg ml−1 of the recombinant enzyme.

We report here for the first time on the genetic identification, purification, and characterization of an α-ketoisovalerate decarboxylase in L. lactis, a novel enzyme responsible for the decarboxylation of α-keto acids, which are intermediate compounds in the biosynthesis of amino acids. The high specificity of the α-ketoisovalerate decarboxylase would entitle it as key enzyme in the enzymatic conversion of amino acids, therefore being the rate-controlling step in the formation of branched-chain aldehydes by L. lactis. The potential role of the α-ketoisovalerate decarboxylase in the formation of aldehydes during cheese ripening could be favored by the activating effect demonstrated by NaCl even at the average concentration present in cheese. Additional studies on the enzyme activity during cheese ripening and on the regulation mechanisms for activity are in progress.


This work was performed under the auspices of the Consejo Superior de Investigaciones Científicas and was supported by Research Project AGL2002-03277.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
  30. [30].
  31. [31].
  32. [32].
  33. [33].
  34. [34].
  35. [35].
  36. [36].
  37. [37].
View Abstract