Previous Article | Next Article ![]()
Applied and Environmental Microbiology, December 2006, p. 7954-7958, Vol. 72, No. 12
0099-2240/06/$08.00+0 doi:10.1128/AEM.01213-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
| SHORT REPORT |
Departamento de Microbiología,1 Departamento de Caracterización de Alimentos, Instituto de Fermentaciones Industriales, CSIC, Juan de la Cierva 3, 28006 Madrid, Spain2
Received 25 May 2006/ Accepted 9 October 2006
|
|
|---|
|
|
|---|
In acidic media like wine, decarboxylation of amino acids to their corresponding amines is thought to provide energy through electrogenic transport as well as assist in maintaining an optimal internal pH (11). Some of these amines are considered "biogenic" and may cause intoxication when consumed. The biogenic amine putrescine, which can potentiate the action of histamine, is the most prevalent amine in wine and is found in almost all analyzed wines (12, 15).
Biogenic amines are formed primarily by decarboxylation of the corresponding amino acids by microorganisms through substrate-specific decarboxylases. The capability of biogenic amine production appears to be strain dependent rather than species specific. Previously, we reported the identification of the odc gene in the putrescine producer O. oeni RM83 (formerly O. oeni BIFI-83) for the first time (14). The odc gene encodes a deduced 745-amino-acid putative ornithine decarboxylase (ODC) (EC 4.1.1.17) which catalyzes the conversion of ornithine to putrescine. The odc gene is seldom present in the O. oeni genome, as it has not been detected in a screen of 42 O. oeni strains tested to date (14). Moreover, in silico analysis of the draft O. oeni PSU-1 genome did not reveal the presence of any odc homologs (16).
Recently, Lucas et al. described that the potential for producing histamine in Lactobacillus hilgardii 0006 is encoded on an unstable 80-kb plasmid (13); the authors further suggested that it is very likely that the histamine producer Tetragenococcus muriaticus and O. oeni 9204 harbor the same plasmid (13). However, the localization of the odc gene in O. oeni RM83 remains unknown.
This study was undertaken to gain deeper insight into the origin of putrescine production in O. oeni RM83. Additionally, O. oeni RM83 ODC was expressed in Escherichia coli and biochemically characterized.
|
|
|---|
|
|
|---|
To construct the O. oeni RM83 DNA library, chromosomal DNA was partially digested with Sau3AI restriction enzyme and ligated to the ZAP Express vector (Stratagene, La Jolla, CA) digested with BamHI. The screening of the library using the 1.4-kb internal odc DNA fragment as a probe yielded five positive clones. Since the inserts of three of them were included in pAM4 and pAM8, only these plasmids were sequenced (Fig. 1). A 10,891-bp O. oeni RM83 DNA fragment was sequenced from the pAM4 and pAM8 plasmids. Two successive reverse PCR experiments, utilizing SnaBI and EcoRV, allowed us to sequence the 5' end of the fragment (Fig. 1). Similarly, a reverse PCR experiment using SpeI allowed for the sequencing of the total 17,270-bp EcoRV-SpeI chromosomal DNA fragment surrounding the O. oeni RM83 odc gene (Fig. 1).
![]() View larger version (14K): [in a new window] |
FIG. 1. Genetic organization of the O. oeni RM83 17.2-kb DNA region containing the odc gene. Thick and thin arrows represent complete and interrupted ORFs, respectively. The locations of putative promoters (vertical bent arrow) and predicted transcriptional terminator regions (ball and stick) are indicated. Some of the plasmids used in this study are indicated, as are relevant restriction sites: E, EcoRV; S, Sau3AI; Sp, SpeI; Sn, SnaBI. Only some of the corresponding restriction sites present in this fragment are represented.
|
|
View this table: [in a new window] |
TABLE 1. odc region-encoded proteins: properties and similarities to proteins in the databases
|
Upstream of orf4, there is an 876-bp region that might correspond to an insertion sequence-like element on the basis of sequence similarity (58.5% nucleotide identity to a Lactobacillus plantarum transposase). Almost-perfect 17-bp inverted repeats were found at positions 6124 to 6140 and 6984 to 7000. Another IS, a copy of ISLpl4, is found 722 nucleotides further upstream (4).
The odc gene is located downstream of ISLpl4. The ODC protein is predicted to possess 745 amino acid residues, including conserved residues involved in enzymatic activity as well as the consensus sequence containing the pyridoxal-5-phosphate binding domain (14). The highest sequence identity (67%) was found between O. oeni and Lactobacillus sp. strain 30a (Table 1). Surprisingly, O. oeni RM83 ODC showed lower identity with similar proteins found in other members of the lactic acid bacteria (49% with Lactobacillus johnsonii and Lactobacillus acidophilus) than with enzymes from unrelated microorganisms, such as Haemophilus influenzae (64%), Pasteurella multocida (64%), and E. coli (58%) (14).
The next identified ORF is a putative potE gene. It encodes a 441-amino-acid, 47.6-kDa protein showing 67% identity to the putrescine-ornithine antiporter (PotE) from Haemophilus influenzae and 66% to PotE proteins from some Enterobacteriaceae. Unexpectedly, O. oeni PotE only shows a 14% identity to amino acid transporters from lactic acid bacteria (data not shown). PotE can catalyze both the uptake and the excretion of putrescine (8).
|
|
|---|
Cell extracts for the ODC enzymatic assay were obtained from E. coli HT414 cells harboring the control plasmid pIN-III(lppp-5)A3 or the recombinant plasmid pAM11 as previously described (18). The ODC assay was performed in 50 mM sodium phosphate buffer (pH 6.5) in the presence of 3.6 mM ornithine and 0.4 mM pyridoxal-5-phosphate. The reaction mixture was incubated at 37°C for 1 h. Subsequently, the putrescine formed in the reaction mixture was derivatized and detected by thin-layer chromatography (6) and by reverse-phase high-pressure liquid chromatography (14). Extracts from the strains harboring pAM11 were able to decarboxylate the supplied ornithine to putrescine, whereas extracts prepared from control cells containing the vector plasmid alone did not (Fig. 2). Therefore, we have provided experimental evidence that the odc gene encodes a functional ODC.
![]() View larger version (9K): [in a new window] |
FIG. 2. Putrescine production by soluble cell extracts of IPTG (isopropyl-ß-D-thiogalactopyranoside)-induced cultures of E. coli HT414 harboring pAM11. The putrescine produced during the enzymatic reaction was subjected to an automatic precolumn derivatization with o-ophthaldialdehyde prior to injection. Putrescine was determined by reverse-phase high-pressure liquid chromatography as previously described (14). (A) Results from reaction with E. coli HT414 bearing the control pIN-III(lppp-5)A3 plasmid. (B) Results from reaction with E. coli HT414 bearing the recombinant pAM11 plasmid.
|
|
|
|---|
![]() View larger version (27K): [in a new window] |
FIG. 3. Schematic overview of the sequence conservation between O. oeni PSU-1 and O. oeni RM83 chromosomal regions containing the proposed recombination site. Genes are represented by arrows. The rectangle corresponds to the interrupted gene. The genes present in these regions are indicated: orf6 and orf7 in O. oeni RM83 and Ooen02001058 (encoding a putative transcriptional terminator), Ooen02001059 (encoding the carbamoylphosphate synthase large subunit), and Ooen02001060 (coding for a hypothetical protein, ZP_00319317) in O. oeni PSU-1. The complete ORF coding for the hypothetical protein in O. oeni PSU-1 is also represented. Open squares and hexagons represent GW domains and MucBP domains, respectively. ORF regions with identical shading correspond to regions having the same degree of sequence identity. The degrees of amino acid identity between the protein fragments encoded by these ORFs are also shown. The colors of the upper and lower bars indicate the degrees of nucleotide identity between the 2-kb DNA regions: black, 47% identity; gray, 61.5%; and white, 100%. Two black arrows indicate the recombination site. The nucleotide positions corresponding to both sequences are also indicated. The O. oeni PSU-1 nucleotide sequence appears in the GenBank database under accession number NZ_AABJ03000005.
|
80-amino-acid-long domains contain the dipeptide Gly-Trp (GW modules). The MucBP domains consist of sequences of around 50 residues in length found in bacterial peptidoglycan bound proteins. It is noteworthy that orf7 in O. oeni RM83 appears to be a chimeric protein originating from the fusion of a gene encoding a protein 50% identical to the O. oeni PSU-1 putative carbamoyl phosphate synthase (Ooen02002059) and a gene encoding a protein 55% identical on its MucBP domain to the PSU-1 hypothetical protein (Ooen02002060). Taking into account that GW and MucBP domains are found in a variety of bacterial proteins, it is possible that the unknown donor protein could have domains encoded by DNA regions showing high nucleotide similarity with the corresponding O. oeni regions. This similarity could facilitate the crossover between this unknown donor DNA and O. oeni chromosomal DNA. Upon examination of regions of maximal identity, the crossover point appears to reside at nucleotide position 16747 of the O. oeni RM83 sequence described in this work and position 62120 of O. oeni PSU-1 (GenBank accession number NZ_AABJ03000005). This recombination site is located in the junction of the MucBP domain and the first GW domain. Interestingly, in O. oeni PSU-1 the gene coding for the hypothetical protein Ooen02001060 is found less than 5 kb downstream of the gene recP, coding for a transketolase. Recently, de las Rivas et al. described for the recP locus a possible example of a recombinatorial event from an unknown source (3). The description of two recombinatorial events in the same DNA region indicates a region of great flexibility in the O. oeni chromosome, as described recently for L. plantarum (17).
It is now understood that horizontal gene transfer provides an important mechanism for generating genotypic and phenotypic diversity in bacteria. This phenomenon has been studied extensively in relation to bacterial adaptability or fitness under certain growth conditions. Accordingly, it has been reported widely that adaptability traits can be encoded by mobile genetic elements. Genomic islands (GI) are clusters of chromosomal genes that have been described as horizontally acquired DNA regions (5). They often possess genes (or pseudogenes) coding for mobility-related elements, such as phage genes, insertion sequences, transposases, and origins of replication. A typical GI carries genes encoding traits that may increase bacterial adaptability under certain growth conditions. All of these observations taken together suggest that the 16.7-kb O. oeni RM83-specific DNA may be a fragment of a GI transferred by horizontal gene transfer.
|
|
|---|
We thank E. García and D. Llull for their help with the pulsed-field gel electrophoresis experiments. We thank the E. coli Genetic Stock Center (http://cgsc.biology.yale.edu) for generously providing the E. coli HT414 strain. We also thank D. Sela for correcting the English version of the manuscript. The technical assistance of M. V. Santamaría and A. Gómez is greatly appreciated.
Published ahead of print on 20 October 2006. ![]()
|
|
|---|
(speA-speB)
speC]. J. Bacteriol. 147:702-704.This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»