Genomic Analysis of Oenococcus Oeni PSU-1 and Its Relevance to Winemaking

FEMS Microbiology Reviews 29 (2005) 465–475 www.fems-microbiology.org

Genomic analysis of Oenococcus oeni PSU-1 and its relevance to winemaking

David A. Mills a,*, Helen Rawsthorne a,1, Courtney Parker a,2, Dafna Tamir a,3, Kira Makarova b

a Department of Viticulture and Enology, University of California, Davis, CA 95616, United States b National Center for Biotechnology Information, National Library of Medicine, National Institutes of Health, Bethesda, MD 20894, United States

Received 24 February 2005; accepted 23 April 2005

First published online 28 August 2005

Abstract

Oenococcus oeni is an acidophilic member of the Leuconostoc branch of lactic acid bacteria indigenous to wine and similar environments. O. oeni is commonly responsible for the malolactic fermentation in wine and due to its positive contribution is frequently used as a starter culture to promote malolactic fermentation. In collaboration with the Lactic Acid Bacteria Genome Consortium the genome sequence of O. oeni PSU-1 has been determined. The complete genome is 1,780,517 nt with a GC content of 38%. 1701 ORFs could be predicted from the sequence of which 75% were functionally classified. Consistent with its classification as an obli- gately heterofermentative lactic acid bacterium the PSU-1 genome encodes all the enzymes for the phosphoketolase pathway. More- over, genes related to flavor modification in wine, such as malolactic fermentation capacity and citrate utilization were readily identified. The completion of the O. oeni genome marks a significant new phase for wine-related research on lactic acid bacteria in which the physiology, genetic diversity and performance of O. oeni starter cultures can be more rigorously examined. Ó 2005 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved.

Keywords: Oenococcus oeni; Genome; Physiology; Genetic diversity

Contents

1. Introduction ...... 466 2. Oenococcus oeni...... 466 2.1. Genomic analysis of Oenococcus oeni strain PSU-1 ...... 466 2.2. General genome description ...... 467 3. Metabolism related to growth in wine – selected topics ...... 468 3.1. Carbohydrate metabolism ...... 468 3.2. Organic acids – malate ...... 468

* Corresponding author. Tel.: +1 530 754 7821; fax: +1 530 752 0382. E-mail address: [email protected] (D.A. Mills). 1 Present address: Department of Bioscience and Biotechnology, Drexel University, 32nd and Chestnut Street, Philadelphia, PA 19104-2875, United States. 2 Present address: Osel, Inc., 1800 Wyatt Dr. Suite 14, Santa Clara, CA 95054, United States. 3 Present address: Department of Plant Pathology and Microbiology, Agricultural, Food and Environmental Quality Sciences, The Hebrew University of Jerusalem, P.O. Box 12, Rehovot 76100, Israel.

0168-6445/$22.00 Ó 2005 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.femsre.2005.04.011 466 D.A. Mills et al. / FEMS Microbiology Reviews 29 (2005) 465–475

3.3. Organic acids – citrate ...... 469 3.4. Nitrogen metabolism ...... 470 3.5. Amino acid catabolism ...... 470 3.6. Bacteriophage...... 471 3.7. Stress-related genes ...... 471 3.8. Phylogeny and systematics ...... 472 4. Future work ...... 472 Acknowledgments ...... 472 References...... 473

1. Introduction relatively nutrient-limited, and contains inhibitory compounds such as polyphenolics. To make microbial The production of wine involves an amalgam of growth more challenging, winemakers typically add sul- microbial transformations comprising a complex succes- fur dioxide to grape must to control indigenous yeast sion of various yeast and bacterial species. The initial and bacterial populations. Thus while many LAB are conversion of grape must is an alcoholic fermentation able to perform the MLF, few can thrive sufficiently to carried out by one or more strains of Saccharomyces complete the fermentation in a desired fashion. Oeno- (typically Saccharomyces cerevisiae), either purposely coccus oeni is the agent most commonly associated with inoculated or naturally present in the grape/winery envi- the MLF in the production of wine and for that reason ronment. This primary fermentation produces ethanol has attracted considerable attention. O. oeni is consid- and creates the anaerobic conditions that limit growth ered a desirable bacterium because it can grow in the of other yeast and bacterial species and thus protects hostile environment of wine and, among the wine- the wine from spoilage [1]. The malolactic fermentation related LAB, it is least associated with off flavors or (MLF) is a secondary fermentation that typically occurs other undesirable metabolites. at the end of alcoholic fermentation and is carried out by one or more populations of lactic acid bacteria (LAB). This secondary fermentation can be beneficial 2. Oenococcus oeni or detrimental, depending on the wine style. MLF de- acidifies wine by conversion of malate (L-malic acid) to In 1967, Garvie [8] first proposed a separate species, lactate (L-lactic acid) and is favored in high-acid wines Leuconostoc oenos, for wine-related leuconostoc strains. produced in cool-climate regions. Conversely, this pro- L. oenos was differentiated from other Leuconostoc spe- cess is less desired in warm-climate regions where al- cies primarily on the basis of tolerance to acidic condi- ready low-acid wines are further de-acidified by MLF. tions and a peculiar growth enhancement in media MLF stabilizes wines in a straightforward fashion, by containing tomato juice. Since that time numerous tax- consuming available nutrients and lowering the poten- onomic surveys of L. oenos strains have been under- tial for additional growth of other indigenous microbes. taken, first examining growth characteristics and later Finally MLF contributes to the organoleptic quality of using various molecular methods [9,10]. More recently, wines through direct production of flavor compounds Dicks et al. [11] proposed a reclassification of Leuconos- [2] or through indirect impact on other microorganisms toc oenos into a new genus Oenococcus, of which O. oeni present in the wine [3–5]. From a winemakerÕs perspec- is the only species. tive there is a need to control the MLF, allowing for pre- cise application, or prevention, to enhance the positive 2.1. Genomic analysis of Oenococcus oeni strain PSU-1 attributes or reduce potential negative impacts on the particular wine. Perhaps the greatest measure of control In order to advance the study of O. oeni, we gener- has been achieved through inoculation of starter cul- ated the complete genome sequence of the strain PSU- tures in order to perform the MLF. 1. It was isolated in 1972 by Beelman et al. [12] from a A number of lactic acid bacterial species have been spontaneous malolactic fermentation in an experimental isolated from the wine environment [6,7]. Most belong wine made in Pennsylvania (USA). Initial characteristics to the genera Pediococcus, Lactobacillus, Leuconostoc of the strain demonstrated strong similarity to another and Oenococcus and most possess the capacity to carry well-known oenococcal strain, ML-34, previously iso- out the malolactic fermentation. Wine is a particularly lated in California [13]. Ze-Ze and coworkers [14,15] harsh environment for growth as it possesses a low pH carried out an extensive PFGE analysis of PSU-1. (ranging from 2.5 to 4.0), and following the primary Indeed, we chose PSU-1 for genome sequencing on the fermentation, a high alcohol content. Wine is also basis of this available genetic map. D.A. Mills et al. / FEMS Microbiology Reviews 29 (2005) 465–475 467

Genomic sequencing was undertaken as part of a Jazz. Gap filling was carried out in collaboration with larger effort to generate publicly accessible genome a private company, Fidelity Systems, Inc. (Gaithers- sequence for a number of fermentation-associated bacte- burg, MD), using a direct whole genome sequencing ap- ria. This effort was coordinated by a working unit proach as previous described [18]. Annotation was termed the ‘‘Lactic Acid Bacteria Genome Consortium’’ performed at NCBI using the GeneMark program fol- (LABGC), the mission of which is to advance functional lowed by manual curation. genomic studies on the food grade LAB [16,17]. In 2002, the LABGC partnered with the Joint Genome Institute 2.2. General genome description (JGI), a high throughput sequencing facility run by the United States Department of Energy to generate draft At the time of this writing, manual curation of the sequence of the eleven bacterial genomes, five of which O. oeni PSU-1 genome, and comparison to other LAB (Lactobacillus casei, Lactobacillus brevis, Leuconostoc genomes, are ongoing. As such, what is presented is a mesenteroides, Pediococcus pentosaceus and Oenococcus preliminary report on the PSU-1 genome sequence with oeni) can be readily isolated from wines or musts. A high a focus on specific wine-related fermentation properties. quality draft of O. oeni PSU-1 sequences was generated The genome of O. oeni PSU-1 is a single circular chro- by shotgun sequencing of a small-insert library (2–3 kb) mosome of 1,780,517 nt (Fig. 1), which is remarkably to achieve 8X genome coverage based on the estimated similar to the reported genome size of 1,753,879 nt indi- genome size and assembled using the JGI assembler cated for another O. oeni strain, IOEB8413, sequenced

Fig. 1. Genome atlas view of Oenococcus oeni PSU-1 with the predicted origin of replication at the top. The circle was created with GENEWIZ [88] with additional software developed by Eric Alterman. Innermost circle shows COG classification. Predicted ORFs were classified into five major categories: (1) information storage and processing; (2) cellular processes and signaling; (3) metabolism; (4) poorly characterized; and (5) ORFS with uncharacterized COGs or no COG assignment. Circle 2 shows ORF orientation. ORFs in the sense orientation (ORF+) are listed in blue while ORFs in the antisense orientation (ORFÀ) are in red. Circle 3 shows tRNA (green dots) and rRNA genes (blue dots) respectively. Circle 4 shows transposase genes or gene fragments (red dots). Circle 5 shows G + C content deviation. Deviations from the average GC-content are shown in either green (low GC spike) or orange (high GC spike). A box filter was applied to visualize contiguous regions of low or high deviations. Circle 6 shows BLAST similarities. ORFs were compared against the nonredundant (nr) database by using a gapped BLAST [89]. Regions in blue represent unique proteins in PSU-1 while highly conserved features are shown in red. The degree of color intensity indicates the level of similarity. 468 D.A. Mills et al. / FEMS Microbiology Reviews 29 (2005) 465–475

O. oeni PSU-1 COG Distribution has been reported to utilize glucose, fructose, ribose, tre- 30 halose, and cellobiose [12] and various transport systems

25 related to these carbon sources do appear in the PSU-1 genome (PTS of TC#4.A.6, 4.A.1, 4.A.3; several copies 20 of sugar ABS transporters TC#3.A.1.1, 3.A.1.2; several copies of sugar proton-symporter TC#2.A.1.1, glucose/ 15 ribose porter TC#2.A.7.5, etc.). As expected the genes 10 encoding the complete phosphoketolase pathway are Percent of ORFs represented. Like other heterofermentative LAB, O. oeni 5 PSU-1 possesses the ability to convert fructose to mannitol most likely through the action of a mannitol dehy- 0 m n n n n t n d d n s o s n e n r e e i m m d n o o i io p o o t t io l s s m m m m e ti i t m o i o i a t i i s s s s t w t s s i o t p t l a o l l i i i i a p a i i t l a l c b o l l l l ic o drogenase [12]. This activity can be problematic in wine l i c v c e e s c e e o o o o o n s r i i n r n i r r u a b d l u v c - - d t a b b b b b k n c d a a if e t a e a s p h d n e r e e o t a ta a a r n e l s e s t d g c r e t t t p r n r l c / / m e e e e e U T a e e n l n o a n p / m fermentations in that excess production of mannitol can r a l y h e t / m m m m m n C r e it o m y r t / / n T A m l i P u r t t o o C i t l g o r e d r y ti i N e lt t e a q r p o o i r t s a r e e s p m p o a c c D o n s p i p n lead to a ‘‘mannitol taint’’ and also results in high acetic n n c n n s s y d n e m e o a n z L s u f ig i n E n n f u l s t r a n n o e S l / a io t r a a l F e g l t t r e r c a D s r e t o t e r n t d acid levels [20]. Surprisingly, PSU-1 does not contain a C i n e a i e C n S e k a s r c d n c r n d i o i t I a t I e f t y o G f s h o e a n l COG related to this activity (COG0246; mannitol-1- r o o i t P b c r m u r a la c A N phosphate/altronate dehydrogenases), same COG is u ll e c a present in Leuconostoc plantarum [21] a species often iso- tr n I lated from wine. Clearly this activity is linked to a differ- Fig. 2. COG classification of the 1701 predicted O. oeni PSU-1 ent dehydrogenase in PSU-1. proteins. Beelman et al. [12] noted that, when originally isolated, PSU-1 could utilize sucrose, lactose, maltose and by Guzzo and coworkers [16] (note at the time of this raffinose as sole carbon sources. When the same strain writing neither PSU-1 nor IOEB8413 genome sequence was analyzed two years later these abilities were lacking is publicly accessible). The G + C content for the PSU- suggesting that either the strain had lost this genetic 1 genome is 38%. In silico analysis confirmed the previ- capacity or the original tests gave a false positive [12]. ous identification of two rRNA operons [15] in opposite Examination of the PSU-1 genome reveals some capac- orientation at positions 600 and 1270 kb. Fourty- ity related to these substrates including: a maltosephos- three tRNA genes representing 20 amino acids were phorylase (COG1554; containing a possible frameshift), identified scattered around the genome on either strand three ABC-type maltose permeases (COG3833) and with one specific cluster of 15 tRNA genes at 1136 kb what appears to be a fragment of a sucrose-specific (Fig. 1). Redundant tRNAs were identified for all amino PTS system IIBC component (COG1264). Genes specif- acids except for aspartate, cysteine, histidine, isoleucine, ically linked to sucrose, lactose, maltose or raffinose phenylalanine, tryptophan, tyrosine and valine. The rep- transport were not revealed however several carbohy- lication origin, identified by GC-skew and ORF direc- drate transporters with uncertain substrates are present. tionality (Fig. 1), was found adjacent to the canonical This includes three carbohydrate uptake transporter-1 dnaA gene and a less defined terminus region could be (CUT) family (TC#3.A.1.1) operons (plus one partial localized to around position 1000 kb. Initial analysis operon) and one CUT-2 family (TC#3.A.1.2) operon. predicted 1701 intact ORFs of which 75% were classified Since PSU-1, in its current form, does not contain any into COG (clusters of orthologous groups of proteins) plasmids, it is also possible some of these capabilities categories on the basis of biological or biochemical func- were plasmid-borne in the original isolate and lost with tion (Fig. 2). In addition to the repeated rRNA operon, subsequent passage. 14 different transposase (insertion sequence) genes representing 5 different COGs (COG2826, COG3293 and 3.2. Organic acids – malate COG3464, COG2963, COG2801) are present, as well as additional transposase gene fragments (Fig. 1). O. oeni has the capacity to decarboxylate malic acid to lactic acid and carbon dioxide [22,23]. Malic acid rep- resents a large fraction of the organic acids found in 3. Metabolism related to growth in wine – selected topics wine, and the conversion of the dicarboxylic malic acid to the monocarboxylic lactic acid produces a corre- 3.1. Carbohydrate metabolism sponding increase in the pH, and consequent change in the organoleptic qualities of the wine. This change O. oeni is heterofermentative and utilizes hexoses and has been described as a softening of the mouthfeel [24]. pentoses via the phosphoketolase pathway. Glucose and The genetic locus involved in malolactic conversion fructose, which account for 99% of the sugars in grapes (mle) has been previously identified in O. oeni as well [19] are fermented by all strains of O. oeni [8,9]. PSU-1 as other LAB [25,26]. The PSU-1 mle locus contains D.A. Mills et al. / FEMS Microbiology Reviews 29 (2005) 465–475 469 genes encoding the malate decarboxylase (mleA) and ma- 3.3. Organic acids – citrate late permease (mleP) in an apparent operon. Transcribed in the opposite direction upstream of mleAP is a putative Numerous researchers have examined citrate utiliza- lysR-type regulator mleR. This rather simple genetic lay- tion by O. oeni [7,29]. Like other LAB, O. oeni does out appears conserved across various wine-related LAB not use citrate as a sole carbon source but co-metabo- (Lactobacillus plantarum [21], Pediococcus pentosaceus, lizes citrate with glucose resulting in higher achievable Lactobacillus brevis, Lactobacillus casei, Leuconostoc biomass than growth on glucose alone [30]. Upon trans- mesenteroides; http://www.jgi.doe.gov/). The involve- port into the cell, citrate is converted via the actions of ment of the O. oeni mleR regulator in expression of citrate lyase and oxaloacetate decarboxylase to pyru- mleAP is somewhat unclear [27], however a similar, yet vate, which is then converted to a mixture of lactate, genetically separated, mleR gene has been found essential acetate, diacetyl, acetoin, and 2,3-butanediol using a for expression of a the mle operon in L. lactis [28]. pathway similar to that found in other citrate ferment- While the oenococcal mle operon has been linked to ing dairy lactic acid bacteria [24]. Diacetyl is one of malate conversion it remains to be determined if these the most important aroma compounds generated during three genes are solely responsible for the malate conver- MLF [29], production of which increases the buttery ar- sion. While no other copies of the mleA gene exist on the oma of wine, which may or may not be desirable PSU-1 genome, a gene from the same COG family (mae; depending on the style of wine being produced. The COG0281) is present in the citrate cluster. Similarly, five co-fermentation of citrate with glucose also increases different genes homologous to malate permease can be the production of acetate in the wine, potentially found on PSU-1, all belonging to the same COG impacting sensory attributes [31]. The levels of diacetyl, (COG0679; predicted permease). Indeed, O. oeni PSU- acetate, and other compounds produced during the 1 possesses more genes in this COG family than any MLF vary depending on the substrates present in the other LAB for which genome sequence is available. specific wines and the strain(s) of O. oeni present [32,33]. While this may be a reflection of an adaptation of O. Genomic analysis of PSU-1 revealed a typical citrate oeni to the high malate environment inherent in fruit lyase gene cluster (citR, mae, maeP, citC, citD, citE, citF, juices, it remains to be determined if any these genes citX, citG)(Fig. 3) as well as genes involved in the possess a permease activity related to malate. butanediol pathway pathway (ilvB, alsD, butA [2

Fig. 3. Genetic organization of genes involved in citrate utilization in various LAB. Homologous genes are similarly shaded/patterned. Open arrows denote genes unrelated to citrate utilization. Listed below the species names are GenBank accession numbers. 470 D.A. Mills et al. / FEMS Microbiology Reviews 29 (2005) 465–475 copies]) [34]. Interestingly, the main citrate utilization acid biosynthetic capacity for two amino acids promi- cluster contains a gene related to malate decarboxylase nently present in grape must (proline and serine) while (mae; COG0281) similar to that witnessed in Lb. planta- it retains the capacity to synthesize two amino acids rum (mae), Weissella paramesenteroides (citM) and L. (cysteine and methionine) which are present in grape mesenteroides (mae). As suggested by others [35], this must at low concentrations. gene likely encodes an oxaloacetate decarboxylase activ- To complement the amino acid auxotrophy, PSU-1 ity required to convert citrate to pyruvate. Noteworthy contains a range of amino acid-related transporters is the placement of the putative citrate transporter gene including two ABC-type polar amino acid transporters (maeP) inside the cluster as opposed to flanking the op- (TC#3.A.1.3), an ABC-type spermidine/putrescine eron as observed in other clusters. Elsewhere on the transporter (TC#3.A.1.11), seven amino acid transport- PSU-1 genome are additional genes that, by virtue of ers (TC#2.A.3.6) and a branch chain amino acid perme- their COG classification, may be related to citrate ase (TC#2.A.2.6). In addition two ABC-type oligo/ metabolism; citT (a di/tri carboxylic acid transporter; dipeptide permease transporter systems were identified COG0471) and citB (putative response regulator; as well as a range of peptidases (PepQ, PepM, PepN, COG2197). However these genes contain frameshift PepR1, PepV, PepF, PepF2, PepT, PepXP, PepP, PepC, mutations and the integrity of the expressed products re- PepC2, PepO and PepD/A) suggesting the metabolic main to be determined. capacity to garner amino acids from the proteins and peptides in wine. This is not unexpected since several 3.4. Nitrogen metabolism studies have indicated that wine peptides act as carbon and nitrogen substrates for O. oeni growth [43,44]. Grape juice contains various free amino acids, pep- By comparison to dairy related LAB, relatively little tides and proteins, concentrations of which are altered work has focused on the proteolytic activity of wine-re- during the alcoholic fermentation. While the free amino lated strains. Manca de Nadra and coworkers identified acid level decreases in the early stages of the alcoholic two exoproteases from O. oeni strains isolated in Argen- fermentation their level increases as the fermentation tina [45,46]. These exoproteases were expressed at differ- ceases [36–38] due to excretion of amino acids by yeasts ent points throughout growth [45] and were induced by as well as degradation of grape and yeast proteins nutrient and energy deprivation [47]. Both were able to [39,40]. The practice of leaving the wine in the yeast lees release amino acids from red and white wines and one (sur lee) is known to encourage the malolactic fermenta- activity was more active at the lower pH levels common tion (i.e. oenococcal growth) due, in part, to the increase for wine. In silico analysis of PSU-1 did not reveal any in available nitrogen compounds. Thus the composition Prt-P-like cell-envelope proteases and no peptidase of available nitrogen precursors in wine varies depend- genes with a well-supported signal peptide were identi- ing on the stage of the fermentation, variety of grape, fied. It remains to be determined which, if any, of the and wine processing decisions, in addition to competi- PSU-1 proteases correspond to the activities previously tion by other microbes. examined by Manca de Nadra and coworkers. O. oeni strains appear to vary with regard to amino acid requirements, often presenting different patterns 3.5. Amino acid catabolism of auxotrophy. Garvie [41] examined nine strains and found that all were auxotrophic for arginine, cysteine, In addition to acting as carbon and nitrogen sources, glutamate, tyrosine and valine while alanine, lysine amino acids are exploited by wine-related LAB for other and proline were not required for growth. Other amino purposes. Some oenococcal strains convert arginine, one acid requirements appeared to be strain dependent [41]. of the most abundant amino acids in musts and wines, Genomic analysis of PSU-1 suggests the capacity for to ornithine, ammonia and CO2 via the arginine deimin- biosynthesis of eight amino acids: alanine, aspartate, ase pathway [48,49]. This pathway provides a means for asparagine, cysteine, glutamine, lysine, methionine and direct synthesis of ATP and has been shown to enhance threonine. The ability to synthesize isoleucine, valine cell growth and survival [50]. Metabolism of arginine is and leucine is undefined as key enzymatic steps ilvC of oenological concern in that it can result in excretion (COG0059) and ilvD (COG0129) are absent. Biosyn- of citrulline, which can spontaneously react with alcohol thetic capacity for all other amino acids appears to be to form urethane, a known animal carcinogen [51]. The absent. Previous surveys have indicated the predomi- arginine deiminase pathway (arcABCDE) has been pre- nant amino acids found in grape must (>100 mg/L) are viously characterized in certain oenococcal isolates alanine, arginine, glutamine, glutamate, proline and ser- [52,53]. In silico analysis of PSU-1 did not reveal an in- ine, with arginine and proline are consistently present at tact cluster, however an arginine deiminase gene (arcA) higher levels (>350 mg/L) [42]. Conversely cysteine, was identified. While no significant homologs to the methionine and tryptophan were less well represented arginine/ornithine antiporters (arcDE) were revealed (<10 mg/L). Interestingly, PSU-1 is missing the amino the PSU-1 genome does contain two additional trans- D.A. Mills et al. / FEMS Microbiology Reviews 29 (2005) 465–475 471 porters from the cation amino acid transporter family by as an integration site in several oenococcal hosts [64]. (TC#2.A.3.3) that may provide this function. The genetic landscape surrounding these attachment In acidic media like wine, decarboxylation of amino sites did reveal a few intact phage-related ORFs. Adja- acids to corresponding amines is thought to provide en- cent to tRNAlys_TTT and tRNAarg_CCT are genes are ergy through electrogenic transport as well as assist in conserved hypothetical phage-related proteins while maintaining an optimum internal pH [54]. Conse- downstream of tRNAglu_TTC is a fragmented ORF with quently, various amines can be found in wine, predom- homology to the oenococcal fOg44 phage integrase gene inantly histamine, tyramine, and putrescine, among [63]. Of particular interest is the region downstream of others [55]. Unfortunately some of these amines are con- tRNAlys_TTT that contains a site-specific recombinase sidered ‘‘biogenic’’ and may induce a variety of disorders (COG1961) and an ORF with similarity to a L. planta- when consumed (absorbed) at a high concentration [55]. rum prophage gene. As suggested by San-Jose et al. [63], For the most part, biogenic amine formation in wine has these putative phage-related genes adjacent to attach- been associated with non-oenococcal strains, chiefly ment sites may be a sign of imprecise excision of pro- spoilage lactobacilli and pediococci associated with phage(s) from the PSU-1 genome. Other than these higher pH wine conditions (>3.5). A survey of oenococ- remnants, the lack of phage-related COGs in PSU-1 is cal strains isolated from wine showed that histamine somewhat noteworthy when compared to other se- production is a common phenotype [56]. In silico quenced LAB genomes. analysis of the PSU-1 did not reveal genes similar to the histidine or ornithine decarboxylases previously 3.7. Stress-related genes characterized in other oenococcal isolates [57,58]. How- ever, several transporters in the amino acid-polyamine Various stress-related systems have been proposed to organocation (APC) superfamily are present in PSU-1 assist O. oeni to survive the acidic environment of wine including a glutamine:GABA antiporter (TC#2.A.3.7), [65]. Predominantly among these are genes involved in an L-type amino acid transporter, and two cation amino the malate, citrate and amino acid conversions, dis- acid transporters (TC#2.A.3.3). A direct link between cussed previously, which generate proton motive force the activity of these APC transporters and specific ami- and help maintain internal pH homeostasis. Drici-Ca- no acid catabolism remains to be determined. chon et al. [66] have shown a correlation between acid tolerance in O. oeni and increased ATPase activity. 3.6. Bacteriophage Moreover ATPase mutants were shown to be impaired in malolactic activity [67], suggesting a linkage between The impact of bacteriophage on the malolactic fer- MLF and H+-ATPases systems. In silico analysis of the mentation in wine is somewhat ambiguous. Numerous PSU-1 genome identified a complete atp operon encod- bacteriophages have been identified in fermenting ing a putative F0–F1 ATPase sytem (atpBEFHAGDC). must/wine [59–61]. However, unlike the situation with Also identified were two cation-translocating ATPases dairy fermentations, direct evidence suggesting wide- (COG2217) which encode activities that have been also spread occurrence of incomplete malolactic fermenta- implicated in oenococcal pH homeostasis [68]. tions due to bacteriophage-induced cell lysis is lacking. Other genes previously shown to be involved in the This may be due to the multiplicity of indigenous species oenococcal stress response(s) were readily identified in and strains present in wine fermentations that carry out PSU-1. These include clpX [69], clpLP [70], trxA [71], the malolactic conversion. Thus bacteriophage lysis of hsp18 ([72]; COG0071), ftsH [73] and omrA [74].In any one population may not impact the overall malolac- addition, the class I heat shock genes (groESL and dnaK tic conversion as other LAB populations remain to operons) were identified. PSU-1 contains three copies accomplish this task. However, with the repeated use the ATP-binding subunit of Clp protease (clpA) how- of specific oenococcal starter cultures bacteriophage at- ever no clpB or clpC genes were identified. Immediately tack could certainly become problematic, delaying over- upstream of one clpA allele is a class III stress gene reg- all conversion and/or opening the way for less desired, ulator ctrS (COG4463). PSU-1 does not appear to con- indigenous populations to dominate. tain other heat shock transcriptional regulators such as The PSU-1 sequence does not contain any intact tem- hrcA (COG1420) or alternative sigma factors. Like perate bacteriophages, or larger tracts of obvious bacte- other LAB, O. oeni is microaerophilic, lacks catalases riophage origin, often found in other LAB [21,62]. and possesses systems to deal with oxidative stress. Regardless, strain PSU-1 can be a host for bacterio- Genes involved in the disulfide-reducing pathway (trxA phage. Sao-Jose et al. [63] identified several prophage and trxB homologs), and systems that remove reactive integration sites in the draft PSU-1 sequence, mostly oxygen (NADH-oxidase [COG1902] and NADH-perox- adjacent to tRNA genes (tRNAlys_TTT, tRNAarg_CCT, idase [COG0778]) were found in PSU-1, however no tRNAleu_ CCA, tRNAglu_TTC, tRNAglu_CTC). The site homolog to superoxide dismutase was identified. In addi- adjacent to tRNAleu_CAG had been previously identified tion to omrA, a bacterial multidrug resistance transporter 472 D.A. Mills et al. / FEMS Microbiology Reviews 29 (2005) 465–475 similar to the lmrA from L. lactis [75] and horA of addition, comparison of the PSU-1 genome to the other L. brevis [76], PSU-1 contains two other transporters sequenced O. oeni strain (IOEB8413) [16], isolated in of the same drug-exporter 2 family (TC#3.A.1.117). France, will shed light on the geographic diversity of Given that the horA gene in L. brevis has been shown these strains as well as the debate on mode of evolution to provide resistance to plant hops [76], perhaps these of the species [79,82]. Thus one benefit emerging from transporters in O. oeni provide protection from similar the sequence data will be a new, and comprehensive, fruit-derived inhibitors. ability to discriminate oenococcal strains, allowing for better definition of genotypes that have arisen in various winemaking regions. Future use of microarrays to probe 3.8. Phylogeny and systematics genomic differences among isolates will expand this analysis to better reveal how this species evolved. Exact differentiation of oenococcal strains is an Use of microarrays to probe O. oeni gene expression important aim both for commercial and scientific ratio- in wine fermentations will also advance our understand- nales. Studies using a variety of molecular approaches ing of the MLF. Given the similarity in oenococcal iso- have suggested a homogenous nature to the species lates [15,79], it may be possible to use the same array for [11,77–80]. Using PFGE methods, Tenreiro and co- multiple strains. This would give an unprecedented view workers proposed two major lineages for O. oeni [81]. of gene expression within a single species and help define However, a subsequent analysis suggested the two the diversity of stress and MLF-related responses, groupings were less divergent than originally believed among others. An applied outcome to this will be a bet- [15]. Recently De Rivas et al. [82] used multilocus se- ter discrimination of starter cultures with regard to quence typing (MLST) of five genes (gyrB, ddl, mleA, MLF performance and/or production of off-flavors. pgm and recP) to examine allelic diversity and popula- Moreover, as other LAB sequencing projects come to tion structure of various oenococcal isolates. Interest- fruition (particular for pediococci, leuconostoc and lac- ingly, MLST was able to differentiate 18 strains that tobacilli) the involvement of these other LAB genera in could only be differentiated into two groups by ribotyp- wine-related spoilage, as well as their interaction with O. ing. This allelic diversity suggests a higher level of genetic oeni, can be examined in more detail. heterogeneity among oenococccal isolates than had been Future work will inevitably require viable shuttle previously suggested by other molecular typing meth- plasmids and other genetic tools for chromosomal gene ods. De Rivas et al. [82] also concluded that recombi- analysis. Unfortunately the lack of such tools has slo- nation has played a major role in generating genetic wed analysis of O. oeni by comparison to other, genet- diversity in O. oeni. ically pliable, LAB. However the recent development of Yang and Woese [83] first noted that the phylogenetic conjugative methods for transposon [85] or plasmid position of O. oeni as determined by 16S rRNA sug- delivery [86,87] into O. oeni suggest that these barriers gested a tachytelic, or fast-evolving, organism. Subse- are being resolved. The combination of PSU-1 genome quent analysis of the b subunit of the DNA dependent sequence and viable genetic tools sets the stage for a RNA polymerase did not support the hypothesis that tremendous expansion in our understanding of oeno- O. oeni is tachytelic [84]. Preliminary phylogenetic com- coccal biology as well as their critical role in wine parisons based on concatenated ribosomal proteins production. from available LAB genome sequences indicate that the Leuconostoc group has evolved rapidly relative to the Lactobacillus–Pediococcus grouping with O. oeni Acknowledgments being the most divergent member of the Leuconostoc group (data not shown). Adaptation to growth in acidic D.A.M. acknowledges the tremendous contributions fruit juices environments is one possible, albeit specula- of Fidelity Systems Inc. (in particular Serg Kozyavkin tive, explanation for the divergence of O. oeni away and Alexi Slesarev) and the Joint Genome Institute (in from the other leuconostocs. particular Paul Richardson, Dan Rokshar, and Trevor Hawkins) for completion of the PSU-1 genome sequence. D.A.M. is grateful to Eric Alterman, Bart Wei- 4. Future work mer, Todd Klaenhammer, Kalliopi Rantsiou, Jae Han Kim, Milton Saier, and the LABGC for various assis- Public access to genome sequence for any industrially tance, comment and encouragement with the LABGC useful bacteria is only a starting point for future genome project as well as the PSU-1 genome analysis. ‘‘omics’’ research. The most immediate utility of the K. Boundy-Mills is thanked for critical review of the O. oeni PSU-1 genome access is the in silico examination manuscript. D.A.M. gratefully acknowledges the Amer- of cell metabolism, comparative analyses with other ican Vineyard Foundation, the California Competitive LAB and prediction of nutritional requirements. In Grants Program for Research in Viticulture and Enol- D.A. Mills et al. / FEMS Microbiology Reviews 29 (2005) 465–475 473 ogy, the US Department of Energy for financial support [18] Slesarev, A.I., Mezhevaya, K.V., Makarova, K.S., Polushin, of the O. oeni PSU-1 genome project. N.N., Shcherbinina, O.V., Shakhova, V.V., Belova, G.I., Arav- ind, L., Natale, D.A., Rogozin, I.B., Tatusov, R.L., Wolf, Y.I., Stetter, K.O., Malykh, A.G., Koonin, E.V. and Kozyavkin, S.A. (2002) The complete genome of hyperthermophile Methanopyrus References kandleri AV19 and monophyly of archaeal methanogens. Proc. Natl. Acad. Sci. 99, 4644–4649. [1] Fleet, G.H. (1998) The microbiology of alcoholic beverages. [19] Coombe, B.G. and Dry, P.R. (1992) Viticulture. Practices, vol. 2. Microbiology of Fermented Foods, vol. 1, pp. 217–262. Blackie Hyde Park Press, Underdale, Australia, 376 pp. Academic & Professional, London; New York. [20] Boulton, R.B., Singleton, V.L., Bisson, L.F., Kunkee, R.E. [2] Matthews, A., Grimaldi, A., Walker, M., Bartowsky, E., Grbin, (1996). Principles and practices of winemaking, Chapman & P. and Jiranek, V. (2004) Lactic acid bacteria as a potential source Hall, London, 604 pp. of enzymes for use in vinification. Appl. Environ. Microbiol. 70, [21] Kleerebezem, M., Boekhorst, J., vanKranenburg, R., Molenaar, 5715–5731. D., Kuipers, O.P., Leer, R., Tarchini, R., Peters, S.A., Sandbrink, [3] Huang, Y.C., Edwards, C.G., Peterson, J.C. and Haag, K.M. H.M., Fiers, M.W., Stiekema, W., Lankhorst, R.M., Bron, P.A., (1996) Relationship between sluggish fermentations and the Hoffer, S.M., Groot, M.N., Kerkhoven, R., deVries, M., Ursing, antagonism of yeast by lactic acid bacteria. Am. J. Enol. Vitic. B., deVos, W.M. and Siezen, R.J. (2003) Complete genome 47, 1–10. sequence of Lactobacillus plantarum WCFS1. Proc. Natl. Acad. [4] Edwards, C.G., Haag, K.M., Collins, M.D., Hutson, R.A. and Sci. 100, 1990–1995. Huang, Y.C. (1998) Lactobacillus kunkeei sp. nov.: a spoilage [22] Salema, M., Lolkema, J., SanRamao, M.V. and Dias, M.C.L. organism associated with grape juice fermentations. J. Appl. (1996) The proton motive force generated in Leuconostoc oenos by Microbiol. 84, 698–702. L-malate fermentation. J. Bacteriol. 178, 3127–3132. [5] Alexandre, H., Costello, P.J., Remize, F., Guzzo, J. and Guilloux- [23] Salema, M., Capucho, I., Poolman, B., San Romaõ, M.V. and Benatier, M. (2004) Saccharomyces cerevisiae–Oenococcus oeni Dias, M.C. (1996) In vitro reassembly of the malolactic fermen- interactions in wine: current knowledge and perspectives. Int. J. tation pathway of Leuconostoc oenos (Oenococcus oeni). J. Food Microbiol. 93, 141–154. Bacteriol. 178, 5537–5539. [6] Wibowo, D., Eschenbruch, R., Davis, C.R., Fleet, G.H. and Lee, [24] Nielsen, J.C. and Richelieu, M. (1999) Control of flavor devel- T.H. (1985) Occurrence and growth of lactic acid bacteria in wine: opment in wine during and after malolactic fermentation by a review. Am. J. Enol. Vitic. 36, 302–313. Oenococcus oeni. Appl. Environ. Microbiol. 65, 740–745. [7] Lonvaud-Funel, A. (1999) Lactic acid bacteria in the quality [25] Denayrolles, M., Aigle, M. and Lonvaud, F.A. (1994) Cloning improvement and depreciation of wine. Ant. van Leeuwenhoek and sequence analysis of the gene encoding Lactococcus lactis 76, 317–331. malolactic enzyme: relationships with malic enzymes. FEMS [8] Garvie, E.I. (1967) Leuconostoc oenos sp.nov. J. Gen. Microbiol. Microbiol. Lett. 116, 79–86. 48, 431–438. [26] Labarre, C., Guzzo, J., Cavin, J.F. and Divie`s, C. (1996) Cloning [9] Van Vuuren, H.J.J. and Dicks, L.M.T. (1993) Leuconostoc oenos: and characterization of the genes encoding the malolactic enzyme A review. Am. J. Enol. Vitic. 44, 99–112. and the malate permease of Leuconostoc oenos. Appl. Environ. [10] Versari, A., Parpinello, G.P. and Cattaneo, M. (1999) Leuconos- Microbiol. 62, 1274–1282. toc oenos and malolactic fermentation in wine: a review. J. Ind. [27] Labarre, C., Divie`s, C. and Guzzo, J. (1996) Genetic organization Microbiol. Biotechnol. 23, 447–455. of the mle locus and identification of a mleR-like gene from [11] Dicks, L.M.T., Dellaglio, F. and Collins, M.D. (1995) Proposal to Leuconostoc oenos. Appl. Environ. Microbiol. 62, 4493–4498. reclassify Leuconostoc oenos as Oenococcus oeni (corrig ) gen nov, [28] Renault, P., Gaillardin, C. and Heslot, H. (1989) Product of the comb nov. Int. J. Syst. Bacteriol. 45, 395–397. Lactococcus lactis gene required for malolactic fermentation is [12] Beelman, R.B., Gavin III, A. and Keen, R.M. (1977) A new strain homologous to a family of positive regulators. J. Bacteriol. 171, of Leuconostoc oenos for induced malolactic fermentation in 3108–3114. eastern wines. Am. J. Enol. Vitic. 28, 159–165. [29] Bartowsky, E.J. and Henschke, P.A. (2004) The ÔbutteryÕ attribute [13] Ingraham, J.L., Vaughn, R.H. and Cooke, G.M. (1960) Studies of wine–diacetyl–desirability, spoilage and beyond. Int. J. Food on the malolactic organisms isolated from California wines. Am. Microbiol. 96, 235–252. J. Enol. Vitic. 11, 1–4. [30] Salou, P., Loubiere, P. and Pareilleux, A. (1994) Growth and [14] Ze-Ze, L., Tenreiro, R., Brito, L., Santos, M.A. and Paveia, H. energetics of Leuconostoc oenos during cometabolism of glucose (1998) Physical map of the genome of Oenococcus oeni PSU-1 with citrate or fructose. Appl. Environ. Microbiol. 60, 1459– and localization of genetic markers. Microbiology 144, 1145– 1466. 1156. [31] Liu, S.Q. (2002) Malolactic fermentation in wine – beyond [15] Ze-Ze, L., Tenreiro, R. and Paveia, H. (2000) The Oenococcus deacidification. J. Appl. Microbiol. 92, 589–601. oeni genome: Physical and genetic mapping of strain GM and [32] Maicas, S., Gil, J.V., Pardo, I. and Ferrer, S. (1999) Improvement comparison with the genome of a ÔdivergentÕ strain, PSU-1. of volatile composition of wines by controlled addition of Microbiology 146, 3195–3204. malolactic bacteria. Food Res. Int. 32, 491–496. [16] Klaenhammer, T., Altermann, E., Arigoni, F., Bolotin, A., Breidt, [33] Davis, C.R., Wibowo, D., Eschenbruch, R., Lee, T.H. and Fleet, F., Broadbent, J., Cano, R., Chaillou, S., Deutscher, J., Gasson, G.H. (1985) Practical implications of malolactic fermentation: A M., van de Guchte, M., Guzzo, J., Hartke, A., Hawkins, T., Hols, review. Am. J. Enol. Vitic. 36, 290–301. P., Hutkins, R., Kleerebezem, M., Kok, J., Kuipers, O., Lubbers, [34] Garmyn, D., Monnet, C., Martineau, B., Guzzo, J., Cavin, J.F. M., Maguin, E., McKay, L., Mills, D., Nauta, A., Overbeek, R., and Divies, C. (1996) Cloning and sequencing of the gene Pel, H., Pridmore, D., Saier, M., van Sinderen, D., Sorokin, A., encoding alpha-acetolactate decarboxylase from Leuconostoc Steele, J., OÕSullivan, D., de Vos, W., Weimer, B., Zagorec, M. oenos. FEMS Microbiol. Lett. 145, 445–450. and Siezen, R. (2002) Discovering lactic acid bacteria by [35] Martin, M., Corrales, M.A., de Mendoza, D., Lopez, P. and genomics. Ant. Van Leeuwenhoek 82, 29–58. Magni, C. (1999) Cloning and molecular characterization of the [17] Mills, D.A. (2004) The Lactic Acid Bacteria Genome Project. J. citrate utilization citMCDEFGRP cluster of Leuconostoc para- Food Sci. 69, M28–M30. mesenteroides. FEMS Microbiol. Lett. 174, 231–238. 474 D.A. Mills et al. / FEMS Microbiology Reviews 29 (2005) 465–475

[36] Monteiro, F.F. and Bisson, L.F. (1992) Nitrogen supplementa- [55] Lonvaud-Funel, A. (2001) Biogenic amines in wines: role of lactic tion of grape juice II effect on amino acid and urea release acid bacteria. FEMS Microbiol. Lett. 199, 9–13. following fermentation. Am. J. Enol. Vitic. 43, 11–17. [56] Coton, E., Rollan, G., Bertrand, A. and Lonvaud-Funel, A. [37] Monteiro, F.F. and Bisson, L.F. (1991) Amino acid utilization (1998) Histamine-producing lactic acid bacteria in wines: Early and urea formation during vinification fermentations. Am. J. detection, frequency, and distribution. Am. J. Enol. Vitic. 49, Enol. Vitic. 42, 199–208. 199–204. [38] Monteiro, F.F. and Bisson, L.F. (1992) Nitrogen supplementa- [57] Coton, E., Rollan, G.C. and Lonvaud-Funel, A. (1998) Histidine tion of grape juice I effect on amino acid utilization during carboxylase of Leuconostoc oenos 9204: purification, kinetic fermentation. Am. J. Enol. Vitic. 43, 1–10. properties, cloning and nucleotide sequence of the hdc gene. J. [39] Perrot, L., Charpentier, M., Charpentier, C., Feuillat, M. and Appl. Microbiol. 84, 143–151. Chassagne, D. (2002) Yeast adapted to wine: nitrogen compounds [58] Marcobal, A., de Las Rivas, B., Moreno-Arribas, M.V. and released during induced autolysis in a model wine. J. Ind. Munoz, R. (2004) Identification of the ornithine decarboxylase Microbiol. Biotechnol. 29, 134–139. gene in the putrescine-producer Oenococcus oeni BIFI-83. FEMS [40] Charpentier, C., Dos Santos, A.M. and Feuillat, M. (2004) Microbiol. Lett. 239, 213–220. Release of macromolecules by Saccharomyces cerevisiae during [59] Davis, C., Silveira, N.F. and Fleet, G.H. (1985) Occurrence and ageing of French flor sherry wine Vin jaune. Int. J. Food properties of bacteriophages of Leuconostoc oenos in Australian Microbiol. 96, 253–262. wines. Appl. Environ. Microbiol. 50, 872–876. [41] Garvie, E.I. (1967) The growth factor and amino acid require- [60] Henick-Kling, T., Lee, T.H. and Nicholas, D.J.D. (1986) Char- ments of species of the genus Leuconostoc, including Leuconostoc acterization of the lytic activity of bacteriophages of Leuconostoc paramesenteroides (sp.nov.) and Leuconostoc oenos. J. Gen. oenos isolated from wine. J. Appl. Bacteriol. 61, 525–534. Microbiol. 48, 439–447. [61] Sozzi, T., Hose, H., DÕAmico and Gnaegi, F. (1982) Bacterio- [42] Henschke, P. and Jiranek, V. (1993) Yeast-metabolism of nitrogen phage of Leuconostoc oenos isolated from wine. Rev. Suisse compounds. In: Wine Microbiology and Biotechnology (Fleet, Arboric. Hortic. 14, 2–8. G.H., Ed.), pp. 77–164. Harwood Academic, Chur, SW. [62] Bolotin, A., Wincker, P., Mauger, S., Jaillon, O., Malarme, K., [43] Aredes Fernandez, P.A., Saguir, F.M. and Manca de Nadra, Weissenbach, J., Ehrlich, S.D. and Sorokin, A. (2001) The M.C. (2004) Effect of dipeptides on the growth of Oenococcus oeni complete genome sequence of the lactic acid bacterium Lactococ- in synthetic medium deprived of amino acids. Curr. Microbiol. 49, cus lactis ssp. lactis IL1403. Genome Res. 11, 731–753. 361–365. [63] Sao-Jose, C., Santos, S., Nascimento, J., Brito-Madurro, A.G., [44] Feuillat, M., Brillant, G. and Rochard, J. (1980) Mise en evidence Parreira, R. and Santos, M.A. (2004) Diversity in the lysis- dÕune production de proteases exocellulaires par les levures au integration region of oenophage genomes and evidence for cours del la fermentation alcoholique du mout de raisin. Conn. multiple tRNA loci, as targets for prophage integration in Vinge. Vin. 14, 37–52. Oenococcus oeni. Virology 325, 82–95. [45] Rollan, G.C., Farias, M.E. and Manca de Nadra, M.C. (1993) [64] Gindreau, E., Torlois, S. and Lonvaud Funel, A. (1997) Identi- Protease production by Leuconostoc oenos strains isolated from fication and sequence analysis of the region encoding the site- wine. World J. Microbiol. Biotechnol. 9, 587–589. specific integration system from Leuconostoc oenos (Oenococcus [46] Rollan, G.C., Farias, M.E. and Manca de Nadra, M.C. (1995) oeni) temperate bacteriophage phi10MC. FEMS Microbiol. Lett. Characterization of two extracellular proteases from Leuconostoc 147, 279–285. oenos. World J. Microbiol. Biotechnol. 11, 153–155. [65] Guzzo, J., Coucheney, F., Pierre, F., Fortier, L.-C., Delmas, F., [47] Rollan, G.C., Farias, M.E., Strasser De Saad, A.M. and Manca Divies, C. and Tourdot-Marechal, R. (2002) Acidophilic behav- De Nadra, M.C. (1998) Exoprotease activity of Leuconostoc oenos iour of the malolactic bacterium Oenococcus oeni. Sciences des in stress conditions. J. Appl. Microbiol. 85, 219–223. Aliments 22, 107–111. [48] Mira De Orduna, R., Patchett, M.L., Liu, S.Q. and Pilone, G.J. [66] Drici-Cachon, Z., Guzzo, J., Cavin, J.F. and Divies, C. (1996) (2001) Growth and arginine metabolism of the wine lactic acid Acid tolerance in Leuconostoc oenos: Isolation and characteriza- bacteria Lactobacillus buchneri and Oenococcus oeni at different tion of an acid-resistant mutant. Appl. Microbiol. Biotechnol. 44, pH values and arginine concentrations. Appl. Environ. Microbiol. 785–789. 67, 1657–1662. [67] Galland, D., Tourdot-Marechal, R., Abraham, M., Chu, K.S. [49] Mira de Orduna, R., Liu, S.Q., Patchett, M.L. and Pilone, G.J. and Guzzo, J. (2003) Absence of malolactic activity is a (2000) Kinetics of the arginine metabolism of malolactic wine characteristic of H+-ATPase-deficient mutants of the lactic acid lactic acid bacteria Lactobacillus buchneri CUC-3 and Oenococcus bacterium Oenococcus oeni. Appl. Environ. Microbiol. 69, 1973– oeni Lo111. J. Appl. Microbiol. 89, 547–552. 1979. [50] Tonon, T. and Lonvaud-Funel, A. (2000) Metabolism of arginine [68] Fortier, L.C., Tourdot-Marechal, R., Divies, C., Lee, B.H. and and its positive effect on growth and revival of Oenococcus oeni.J. Guzzo, J. (2003) Induction of Oenococcus oeni H+-ATPase Appl. Microbiol. 89, 526–531. activity and mRNA transcription under acidic conditions. FEMS [51] Ough, C.S. (1976) Ethyl carbamate in fermented beverages and Microbiol. Lett. 222, 165–169. foods Part 2 possible formation of ethyl carbamate from [69] Jobin, M.P., Garmyn, D., Divies, C. and Guzzo, J. (1999) The diethyl dicarbonate addition to wine. J. Agric. Food Chem. Oenococcus oeni clpX homologue is a heat shock gene preferen- 24, 328–331. tially expressed in exponential growth phase. J. Bacteriol. 181, [52] Tonon, T., Bourdineaud, J.-P. and Lonvaud-Funel, A. (2001) The 6634–6641. arcABC gene cluster encoding the arginine deiminase pathway of [70] Beltramo, C., Grandvalet, C., Pierre, F. and Guzzo, J. (2004) Oenococcus oeni, and arginine induction of a CRP-like gene. Res. Evidence for multiple levels of regulation of Oenococcus oeni clpP- Microbiol. 152, 653–661. clpL locus expression in response to stress. J. Bacteriol. 186, 2200– [53] Divol, B., Tonon, T., Morichon, S., Gindreau, E. and Lonvaud- 2205. Funel, A. (2003) Molecular characterization of Oenococcus oeni [71] Jobin, M.P., Garmyn, D., Divies, C. and Guzzo, J. (1999) genes encoding proteins involved in arginine transport. J. Appl. Expression of the Oenococcus oeni trxA gene is induced by Microbiol. 94, 738–746. hydrogen peroxide and heat shock. Microbiology 145, 1245–1251. [54] Konings, W.N. (2002) The cell membrane and the struggle for life [72] Guzzo, J., Delmas, F., Pierre, F., Jobin, M.P., Samyn, B., Van of lactic acid bacteria. Ant. Van Leeuwenhoek 82, 3–27. Beeumen, J., Cavin, J.F. and Divies, C. (1997) A small heat shock D.A. Mills et al. / FEMS Microbiology Reviews 29 (2005) 465–475 475

protein from Leuconostoc oenos induced by multiple stresses and [81] Tenreiro, R., Santos, M.A., Paveia, H. and Vieira, G. (1994) during stationary growth phase. Lett. Appl. Microbiol. 24, 393– Inter-strain relationships among wine leuconostocs and their 396. divergence from other Leuconostoc species, as revealed by low [73] Bourdineaud, J.P., Nehme, B., Tesse, S. and Lonvaud-Funel, A. frequency restriction fragment analysis of genomic DNA. J. Appl. (2003) The ftsH gene of the wine bacterium Oenococcus oeni is Bacteriol. 77, 271–280. involved in protection against environmental stress. Appl. Envi- [82] de Las Rivas, B., Marcobal, A. and Munoz, R. (2004) Allelic ron. Microbiol. 69, 2512–2520. diversity and population structure in Oenococcus oeni as deter- [74] Bourdineaud, J.P., Nehme, B., Tesse, S. and Lonvaud-Funel, A. mined from sequence analysis of housekeeping genes. Appl. (2004) A bacterial gene homologous to ABC transporters protect Environ. Microbiol. 70, 7210–7219. Oenococcus oeni from ethanol and other stress factors in wine. Int. [83] Yang, D. and Woese, C.R. (1989) Phylogenetic structure of the J. Food Microbiol. 92, 1–14. leuconostocs an interesting case of a rapidly evolving organism. [75] van Veen, H.W., Venema, K., Bolhuis, H., Oussenko, I., Kok, J., Syst. Appl. Microbiol. 12, 145–149. Poolman, B., Driessen, A.J. and Konings, W.N. (1996) Multidrug [84] Morse, R., Collins, M.D., OÕHanlon, K., Wallbanks, S. and resistance mediated by a bacterial homolog of the human multi- Richardson, P.T. (1996) Analysis of the betaÕ subunit of DNA- drug transporter MDR1. Proc. Natl. Acad. Sci. 93, 10668–10672. dependent RNA polymerase does not support the hypothesis [76] Sakamoto, K., Van Veen, H.W., Saito, H., Kobayashi, H. and inferred from 16S rRNA analysis that Oenococcus oeni (formerly Konings, W.N. (2002) Membrane-bound ATPase contributes to Leuconostoc oenos) is a tachytelic (fast-evolving) bacterium. Int. J. hop resistance of Lactobacillus brevis. Appl. Environ. Microbiol. Syst. Bacteriol. 46, 1004–1009. 68, 5374–5378. [85] Zuniga, M., Pardo, I. and Ferrer, S. (1996) Transposons Tn916 [77] Kelly, W.J., Huang, C.M. and Asmundson, R.V. (1993) Com- and Tn925 can transfer from Enterococcus faecalis to Leuconostoc parison of Leuconostoc oenos strains by pulsed-field gel electro- oenos. FEMS Microbiol. Lett. 135, 179–185. phoresis. Appl. Environ. Microbiol. 59, 3969–3972. [86] Zuniga, M., Pardo, I. and Ferrer, S. (2003) Conjugative [78] Sato, H., Yanagida, F., Shinohara, T., Suzuki, M., Suzuki, K. and plasmid pIP501 undergoes specific deletions after transfer from Yokotsuka, K. (2001) Intraspecific diversity of Oenococcus oeni Lactococcus lactis toOenococcus oeni. Arch. Microbiol. 180, isolated during red wine-making in Japan. FEMS Microbiol. Lett. 367–373. 202, 109–114. [87] Beltramo, C., Oraby, M., Bourel, G., Garmyn, D. and Guzzo, J. [79] Zavaleta, A.I., Martinez-Murcia, A.J. and Rodriguez-Valera, F. (2004) A new vector, pGID052, for genetic transfer in Oenococcus (1997) Intraspecific genetic diversity of Oenococcus oeni as derived oeni. FEMS Microbiol. Lett. 236, 53–60. from DNA fingerprinting and sequence analyses. Appl. Environ. [88] Jensen, L.J., Friis, C. and Ussery, D.W. (1999) Three views of Microbiol. 63, 1261–1267. microbial genomes. Res. Microbiol. 150, 773–777. [80] Viti, C., Giovannetti, L., Granchi, L. and Ventura, S. (1996) [89] Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Species attribution and strain typing of Oenococcus oeni (formerly Z., Miller, W. and Lipman, D.J. (1997) Gapped BLAST and PSI- Leuconostoc oenos) with restriction endonuclease fingerprints. BLAST: a new generation of protein database search programs. Res. Microbiol. 147, 651–660. Nucl. Acids Res. 25, 3389–3402.