Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Nawawi, R. A. Articles by Barnes, A. C. Search for Related Content PubMed PubMed Citation Articles by Nawawi, R. A. Articles by Barnes, A. C. Agricola Articles by Nawawi, R. A. Articles by Barnes, A. C.

 Previous Article  |  Next Article 

Applied and Environmental Microbiology, May 2009, p. 2908-2919, Vol. 75, No. 9
0099-2240/09/$08.00+0     doi:10.1128/AEM.02147-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Host-Directed Evolution of a Novel Lactate Oxidase in Streptococcus iniae Isolates from Barramundi (Lates calcarifer)

Roslina A. Nawawi, Justice C. F. Baiano, E. Charlotte E. Kvennefors, and Andrew C. Barnes^*

The University of Queensland, Aquatic Animal Health Laboratory, Centre for Marine Studies, Brisbane, Queensland 4072, Australia

Received 17 September 2008/ Accepted 2 March 2009

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In Streptococcus iniae, lactate metabolism is dependent upon two proteins, lactate permease that mediates uptake and lactate oxidase, a flavin mononucleotide-dependent enzyme that catalyzes oxidation of -hydroxyacids. A novel variant of the lactate oxidase gene, lctO, in Australian isolates of S. iniae from diseased barramundi was found during a diagnostic screen using LOX-1 and LOX-2 primers, yielding amplicons of 920 bp instead of the expected 869 bp. Sequencing of the novel gene variant (type 2) revealed a 51-nucleotide insertion in lctO, resulting in a 17-amino-acid repeat in the gene product, and three-dimensional modeling indicated formation of an extra loop in the monomeric protein structure. The activities of the lactate oxidase enzyme variants expressed in Escherichia coli were examined, indicating that the higher-molecular-weight type 2 enzyme exhibited higher activity. Growth rates of S. iniae expressing the novel type 2 enzyme were not reduced at lactate concentrations of 0.3% and 0.5%, whereas a strain expressing the type 1 enzyme exhibited reduced growth rates at these lactate concentrations. During a retrospective screen of 105 isolates of S. iniae from Australia, the United States, Canada, Israel, Réunion Island, and Thailand, the type 2 variant arose only in isolates from a single marine farm with unusually high tidal flow in the Northern Territory, Australia. Elevated plasma lactate levels in the fish, resulting from the effort of swimming in tidal flows of up to 3 knots, may exert sufficient selective pressure to maintain the novel, high-molecular-weight enzyme variant.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Streptococcus iniae is a major pathogen of farmed fish, resulting in severe economic losses globally estimated at U.S. $150 million annually (27). S. iniae is essentially a blood pathogen, with infection resulting in a generalized septicemia and meningitis (1). During infection the pathogen avoids phagocytosis by means of an antiopsonic capsule (4, 18, 22) and by binding host serum components including immunoglobulins (3) and fibrinogen (2). Little is known, however, about the metabolism of S. iniae during infection although lactic acid bacteria may produce L-lactate from fermentation of glucose. In S. iniae a lactate oxidase gene, lctO, has been characterized previously (11). The product of the lctO gene in S. iniae is a flavin enzyme (L-lactate 2-monooxygenase, EC 1.13.12.4), which catalyzes the oxidation of lactate to pyruvate, coupled with reduction of O₂ to H₂O₂ (11). Lactate oxidase has been extensively characterized, both structurally and functionally, in the cold-water marine pathogen Aerococcus viridans (9, 30, 35, 36); thus, the catalytic activities of these enzymes are relatively well understood. In S. iniae, lactate can be utilized as an energy source through an aerobic but nonrespiratory mode of metabolism (11), a mechanism that is coupled to hydrogen peroxide production in Streptococcus pyogenes (26).

Since the discovery of the lactate oxidase gene in S. iniae, its presence has been routinely used for PCR-based diagnosis, overcoming the lack of specificity of commercial biochemical diagnostic kits and other molecular methods. Confirmation of isolate identity as S. iniae by commercial bacterial identification kits is problematic because the biochemical profile is absent from databases supplied with the kits or because the databases are unable to identify atypical strains with confidence (25). Identification of isolates by molecular methods such as PCR is more reliable since isolates with atypical biochemical profiles can confidently be identified. PCR has been used to amplify sections of the 16S rRNA gene (37), the chaperonin HSP60 (12), and the 16S-23S rRNA gene intergenic spacer region (5) for identification of S. iniae. The development of the lactate oxidase gene (lctO) PCR assay by Mata et al. (21) reported that the primer pair LOX-1/LOX-2 could be used successfully to aid in the identification of S. iniae via the generation of a specific 870-bp product. Moreover, the LOX-1/LOX-2 primer pair overcame the problem of nonspecific amplification of Streptococcus difficilis that had previously been reported with the 16S rRNA gene primer pair described previously (21, 37).

In Australia, S. iniae causes major economic loss in farmed barramundi (Lates calcarifer, Bloch) (1, 6). Barramundi, also known as Asian sea bass, are perciform euryhaline fish native to Australia and tropical southeast Asia. In Australia, barramundi have both cultural and commercial significance in terms of their iconic status among indigenous populations and the recent rapid growth of commercial farming. The value of intensive barramundi culture in Australia increased from Australian $15.5 million in 2004 to Australian $23.5 million in 2006 (34). There is also increasing farmed output of L. calcarifer in Malaysia, Indonesia, Taiwan, and Vietnam (33) and small to medium recirculating aquaculture ventures in the United States and United Kingdom using imported fingerlings.

During routine diagnostic screening of S. iniae isolated throughout Australia from diseased barramundi, a novel variant of the lctO gene was found that resulted in amplicons of 920 bp following PCR using the LOX-1/LOX-2 primer pair. Isolates expressing the novel lactate oxidase gene were isolated only from a single site in the Northern Territory, Australia. In the present study, the novel lctO variant is investigated genetically and phenotypically in order to better understand how the larger gene product may have arisen from this single site.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Bacterial strains and culturing.
S. iniae isolates from diseased barramundi were obtained from veterinary laboratories throughout Australia; reference isolates from the United States and Canada were obtained from Lynn Shewmaker at the Centers for Disease Control and Prevention, Atlanta, GA; isolates from Thailand, Israel, and Réunion Island were kindly provided by Christian Michel of L'Institut National de la Recherche Agronomique, France (Table 1). Isolates were stored in vegetable peptone broth (Oxoid, Thebarton, Australia) containing 20% glycerol at –80°C and routinely cultured on Columbia agar base containing 5% defibrinated sheep blood at 28°C for 24 h.

View this table: [in this window] [in a new window]

TABLE 1. S. iniae isolates used in this study

PCR amplification of the S. iniae lactate oxidase genes.
Initially, the LOX-1/LOX-2 primer pair was employed to amplify the lctO gene as confirmatory diagnostic support for strains that had previously been identified as S. iniae based on biochemical tests and clinical history by the supplying veterinary laboratories. S. iniae genomic DNA was extracted from freshly grown cells using an enzymatic lysis method (24). PCR of the lactate oxidase gene was performed as described previously (21), and PCR amplification of the 16S rRNA gene with primers 27F and 1492R was also performed as described previously (15). PCR amplification of the full-length lctO gene was performed with primers LCTO22F and LCTO1195R designed from the lctO sequence (accession number Y07622) (Table 2). Cycling parameters included an initial denaturation step of 3 min at 94°C followed by 35 cycles at 94°C for 15 s, 58°C for 1 min, and 72°C for 2 min, with a final extension cycle at 72°C for 7 min. The resultant PCR products were visualized in a 1% (wt/vol) agarose gel stained with ethidium bromide. Purified full-length PCR products were ligated into pGEM-T (Promega, Sydney, Australia) and transformed into TOP10 cells (Invitrogen, Melbourne, Australia). Sequencing of the full-length lctO genes from clones was facilitated by primers M13F, M13R, LCTO22F, LOX-1, LOX-2, and LCTO1195R (Table 2). The lactate permease gene was amplified with LCTP F1 and LCTP R primers (Table 2) under similar cycling conditions, except that the annealing temperature was 56.3°C. Partial 16S rRNA gene sequences were obtained by sequencing PCR products amplified using the universal 27F and 1492R bacterial 16S rRNA gene primers with the 530F primer (Table 2).

View this table: [in this window] [in a new window]

TABLE 2. PCR primers used in this study

Construction of recombinant lctO variants from S. iniae and expression in Escherichia coli.
The complete lactate oxidase genes from two strains representative of the two different lctO genes, QMA0165 (type 1) and QMA0177 (type 2), were amplified using primers ELCTO 22F and ELCTO+EK R (Table 2). Genomic DNA (50 ng) from S. iniae strains in 25-µl reaction mixtures containing 5 µl of 5x PrimeSTAR buffer (Takara, Shiga, Japan), 0.2 mM deoxynucleoside triphosphate mixture, 200 ng of forward and reverse primers, and 0.5 units of PrimeSTAR proofreading Taq polymerase (Takara, Shiga, Japan) was amplified with a Mastercycler Gradient EP-S Thermocycler (Eppendorf, Hamburg, Germany). DNA was denatured at 94°C for 10 min, followed by 35 cycles of denaturation at 94°C for 15s, annealing at 56.7°C for 1 min, and elongation at 72°C for 2 min, with a final extension at 72°C for 10 min. To determine the quality of the PCR products, 3 µl of each amplification product was loaded onto 1% agarose gels, stained with 0.5 µg/ml ethidium bromide, and separated by electrophoresis at 80 V for 45 min. PCR products were purified using a commercial kit (iNtRON Biotechnology, Korea) and ligated overnight at 4°C into a Champion pET Directional TOPO expression vector in accordance with the manufacturer's instructions (Invitrogen, Melbourne, Australia).

Vector constructs were transformed into chemically competent E. coli BL21 Star and selected with ampicillin (100 µg/ml) on LB agar. Clones were picked, and the direction and frame of the insert were confirmed by sequencing. Expression of the recombinant protein was induced in BL21 Star cells with isopropyl-β-D-thiogalactopyranoside (IPTG). Cells were harvested by centrifugation at 10,000 x g for 5 min and washed in double-distilled H₂O. To determine when optimal induction occurred, cells harvested at 1, 2, and 3 h postinduction were resuspended in 60 µl of nonreducing sample buffer (0.125 M Tris, 20% glycerol, 20 µg/ml bromophenol blue, 4% sodium dodecyl sulfate [SDS]) and lysed by vortexing with approximately 40 mg of glass beads (150 to 212 µm; Sigma, Castle Hill, Australia). Protein concentrations were determined using a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Rockland, DE).

Assay of lactate oxidase enzyme activity.
Lactate oxidase enzyme activity was determined by a peroxidase-coupled spectrophotometric assay (20). Cell lysates were prepared from recombinant E. coli grown overnight (37°C at 200 rpm) and then induced with IPTG for 3 h. Lysates were prepared by centrifuging the induced cultures at 3,200 x g for 20 min at 4°C. Cells were washed with 10 mM potassium phosphate buffer, pH 6.5, and stored at –20°C until use. Cells were resuspended with the same buffer and treated with lysozyme (0.5 mg/g of cells) (Sigma, Castle Hill, Australia) for 6 h at 4°C, followed by disruption of the cells with 40 mg of glass beads in a MagnaLyser bead mill (Roche Diagnostics, Melbourne, Australia) at 4,500 rpm for 90 s. Cell debris and beads were removed by centrifugation at 10,000 x g for 5 min, and the supernatants were retained for further analysis. Protein concentrations of the lysates were determined using a bicinchoninic acid protein assay kit (Pierce, Rockford, IL) with bovine serum albumin as a standard. Expressed recombinant type 1 and type 2 lactate oxidases were assayed. Lactate oxidase derived from Pediococcus sp. (Sigma, Castle Hill, Australia) with a concentration range of 10, 5, 2, 1, 0.5, and 0.2 U/ml was used as an internal standard for the assays.

Enzyme assays were initiated by adding samples at various dilutions into enzyme diluent comprising 10 mM potassium phosphate (P5379; Sigma, Castle Hill, Australia) with 0.01 mM flavin adenine nucleotide (F6625; Sigma, Castle Hill, Australia) (pH 6.5 at 37°C), 0.02% 2,2'-azinobis (3-ethylbenzthiazolinesulfonic acid) (ABTS Fluka 11557, Sigma, Castle Hill, Australia), 10 mM L-(+)-lactate, and 0.5 U of horseradish peroxidase (Sigma, Castle Hill, Australia). The assay mixture was incubated in a thermomixer (Eppendorf, Hamburg, Germany) at 37°C at 300 rpm for 20 min. The reaction was stopped by adding 0.25% sodium dodecyl benzene sulfonate (Sigma, Castle Hill, Australia). Enzyme activity was determined by reading the absorbance at 405 nm using a BMG Fluostar Optima reader (BMG Labtech, Melbourne, Australia). The absorbance was read versus a reagent blank without enzyme. All the assays were performed in triplicate at each concentration (pseudoreplicates to determine within assay error). One unit of lactate oxidase activity was defined as 1 µmol of H₂O₂ produced per min at 37°C and pH 6.5. All enzyme assay experiments were repeated at least three times, and results were analyzed by one-way analysis of variance (ANOVA) followed by Tukey's multiple comparison posttest to determine whether test results were significantly different from each other and from controls.

Densitometric quantification of the lactate oxidase protein in cell lysates.
Proteins from lysates prepared above were separated by electrophoresis in SDS-polyacrylamide gel electrophoresis (PAGE) gels (resolving gel, 10% [wt/vol] acrylamide; stacking gel, 5% [wt/vol] acrylamide), using a Hoeffer SE260 system (GE Healthcare, North Ryde, Australia). The lysate sample was loaded at a concentration of 40 µg per lane. Gels were stained with Brilliant Blue R250 (Sigma, Castle Hill, Australia) for 45 min with gentle agitation before being destained in 40% methanol-10% acetic acid until backgrounds were clear. The percentage of the expressed lactate oxidase gene products in cell lysates was determined by densitometric analysis of four replicate electrophoretic separations of the lysates using an ImageQuant 400 gel documentation system and direct quantification of band intensity with ImageQuant TL, version 2005, analysis software (GE Healthcare, North Ryde, Australia).

MIC.
MICs of lactate for S. iniae isolates expressing the different lactate oxidase genes were determined by broth microdilution or agar dilution. Strains expressing the type 1 enzyme were QMA0076, QMA0140, and QMA0165, while strains expressing the type 2 enzyme were QMA0109, QMA0126, and QMA0177. MICs were also determined using the two recombinant E. coli strains expressing the different enzyme types. The lactate MIC was determined using the following range of dilutions of potassium lactate: 0, 0.1, 0.2, 0.3, 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 4.0, and 5.0%. The MIC using LB broth was determined in 96-well plates. Inocula (10 µl) of an overnight culture of S. iniae or recombinant E. coli with an optical density at 600 nm of 0.2 were inoculated into 200 µl of LB broth. LB broth without lactate was inoculated with each of the strains as a positive control while a negative control was left uninoculated. Absorbance at 600 nm was read after incubation at 37°C overnight using a BMG Fluostar Optima reader.

For determination of the MIC using LB agar (Sigma, Castle Hill, Australia), aliquots (10 µl) of an overnight culture of S. iniae or recombinant E. coli at an optical density at 600 nm of 0.2 were spotted onto surface-dried LB agar plates containing lactate at the concentrations described above. Plates were incubated at 37°C overnight, and the MIC was defined as the lowest concentration (percentage) of lactate that completely inhibited the growth of the bacteria.

Effect of lactate on growth rates of S. iniae.
Growth rate experiments were conducted in 100 µl of LB broth containing potassium lactate concentrations of 0.1%, 0.3%, or 0.5% or without lactate. Standard inocula were prepared by suspending overnight LB agar cultures of S. iniae QMA0165 (type 1 enzyme) or QMA0177 (type 2 enzyme) in sterile phosphate-buffered saline (PBS) to an optical density at 600 nm of 0.400 ± 0.001. These two strains were initially chosen as they were both isolated in the same year (2006) from diseased barramundi farmed in commercial open marine water farms (Table 1). The strains were received immediately by our laboratory from the respective veterinary laboratories and placed into glycerol stock at –80°C with minimal further subculturing. Microbroths were inoculated in 96-well plates with 10 µl of the prepared inoculum and incubated at 28°C with gentle agitation. Growth was recorded hourly by measurement of the optical density at 600 nm using a BMG Fluostar Optima plate reader. Results from three independent experiments, performed simultaneously, were recorded. To determine whether the differences in growth could be attributed to the different lactate oxidase enzymes, the experiment was repeated with six additional strains of S. iniae, three expressing the type 1 enzyme (QMA0088, QMA0218, and QMA0221) and three expressing the type 2 enzyme (QMA0109, QMA0119, and QMA0126). The three strains expressing the type 1 enzyme were selected because they were isolated from geographically disparate locations in different years (Table 1). As strains expressing the type 2 enzyme were found only at one farm, the isolates were selected from separate outbreaks of disease or by isolation on different occasions (Table 1). The experiment was also repeated using the six strains above where either glucose or lactate was used as the sole carbon source. A substrate utilization medium was prepared from PBS containing calcium and magnesium (Thermo Scientific, Sydney, Australia) with 0.1% NH₄Cl added as a nitrogen source. This base diluent was then supplemented with either 0.2% glucose or 0.2% lactate (as potassium lactate) as a carbon source. The experiment was conducted in 100-µl volumes precisely as described above. Data were analyzed with GraphPad Prism, version 4, for Macintosh (GraphPad Software, San Diego, CA; www.graphpad.com). First, growth curves were fitted using the Gompertz equation for bacterial growth, and goodness of fit was determined. Data sets were compared by an F test to determine whether growth curves at various lactate concentrations were different. Then ANOVA was conducted to compare growth at each time point at the different lactate concentrations. Analyses were repeated for each strain.

Sequence analysis and structural modeling of the lactate oxidase proteins.
The protein sequence of lactate oxidase in A. viridans (BAA09172), the predicted protein sequence of the lactate oxidase previously described in S. iniae (accession number Y07622), and the predicted protein sequences of the corrected type 1 lactate oxidase and the novel type 2 variant from S. iniae described in this paper were aligned in Clustal X (28). Enzymatically active sites for A. viridans as described in previous publications (10, 16, 17, 35, 36) that are conserved in S. iniae were marked in the alignment. The Protein Data Bank (PDB) file of 2DU2 (30) on the resolved crystal structure of lactate oxidase in A. viridans was downloaded from the Research Collaboratory for Structural Bioinformatics PDB (http://www.pdb.org/pdb/home/home.do) and subunits 2 and 3 were visualized using PyMOL, version 1.0 (8). Protein models for type 1 and type 2 lactate oxidases in S. iniae were predicted using the CPHmodels, version 2.0, server (www.cbs.dtu.dk/services/) (19), where type 1 was modeled upon PDB entry 2J6X (16) and type 2 was modeled upon PDB entry 2DU2 (30). The protein models were viewed in PyMOL, and two identical subunits were placed adjacent to each other using Adobe Photoshop. Subunit interactions and predicted active sites were as previously described for A. viridans (10, 16, 17, 35, 36).

Nucleotide sequence accession numbers.
The predicted protein sequences of the corrected type 1 lactate oxidase and the novel type 2 variant from S. iniae described in this paper were deposited in the GenBank database under accession numbers EU086697 and EU086704, respectively.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
PCR amplification and sequencing of the lactate oxidase and lactate permease from S. iniae.
Two different sizes of PCR product were obtained when the LOX-1/LOX-2 primer pair was applied during screening of Australian S. iniae isolates. Most strains gave an amplicon size of 869 bp while some of the strains yielded amplicons of 920 bp (Fig. 1A). Strains producing an 869-bp product were designated as lctO type 1, and those with the 920-bp product designated as lctO type 2 (Table 3). Sequencing and translation of the full-length lctO gene showed that the 51-bp insertion was a repeat sequence (Fig. 2). The first start codon rather than the next one in frame downstream was used as the start of the protein since there is some ambiguity as to which codon is used.

View larger version (72K): [in this window] [in a new window]

FIG. 1. Amplicons of lactate oxidase type 1 and type 2 and lactate permease from S. iniae. (A) Agarose gel (1%) separation of PCR amplicons of the lactate oxidase gene lctO from Australian isolates of S. iniae indicating different amplicon sizes with LOX primers (869 bp and 920 bp, respectively). (B) Agarose gel (0.7%) separation of PCR amplicons of the lactate permease gene lctP from Australian isolates of S. iniae using the LCTP F1 and LCTP R primers. Lane M, Fermentas Generuler molecular weight markers; lane 1, QMA0076; lane 2, QMA0078; lane 3, QMA0080; lane 4, QMA0109; lane 5, QMA0126; lane 6, QMA0140; lane 7, QMA0155; lane 8, QMA165; lane 9, QMA0173; lane 10, QMA0177; lane 11, negative (template free) control.

View this table: [in this window] [in a new window]

TABLE 3. Lactate oxidase enzyme activities of extracts from E. coli expressing recombinant lctO type 1 or type 2

View larger version (26K): [in this window] [in a new window]

FIG. 2. Nucleotide sequence differences in type 1 and type 2 lactate oxidase genes of S. iniae. The shaded gray boxes show differences between the type 1 sequence (EU086702; this study), type 2 sequence (EU086704; this study), and the initial GenBank submission (Y07622) (11) in the LOX-2 primer site and the misread C in the GenBank sequence. The unshaded boxes show the deduced stop codon for the two sequences. The horizontal black bars show the primer sites for LOX-2 and LCTO1195R (which was used to amplify the complete gene).

In light of the size variations obtained for the lactate oxidase gene from the putative S. iniae isolates, partial sequencing of the 16S rRNA gene was performed with the 530F primer (Table 2), and submission to the BLAST server (http://www.ncbi.nlm.nih.gov/BLAST) returned identities of 100% to S. iniae ATCC 29178. While absolute certainty of identity based on high matches in the 16S rRNA sequence may not be guaranteed (13), the results taken in conjunction with the previous biochemical identification by the veterinary laboratories confirm the identity of these isolates as S. iniae. No variation was detected in the lactate permease gene lctP for any of the strains analyzed (Fig. 1B).

A retrospective screen of all confirmed S. iniae isolates in our collection indicated that the type 2 enzyme was restricted to isolates from a single marine farm in the Northern Territory, Australia (Table 1).

Lactate oxidase activity of recombinant lactate oxidases expressed in E. coli.
To determine whether the variation in the lactate oxidase sequence altered the enzyme activity, a series of lactate oxidase enzyme activity assays was conducted on recombinant enzymes expressed in E. coli. Lactate oxidase was assayed at a range of concentrations of cell lysate prepared from the recombinant E. coli strains, and the results presented in Table 3 are a representative set of the results from numerous assays. To determine the specific activities of each enzyme type, lysates from induced recombinant E. coli used in the assays were separated by electrophoresis and stained with Coomassie brilliant blue R250 (Fig. 3). Gels were analyzed densitometrically, and the amount of the induced enzyme band was determined as a percentage of the total protein in 40 µg of lysate (Fig. 3 and Table 3). Lysates from E. coli expressing the larger lctO type 2 had significantly higher (P < 0.001, n = 3) lactate oxidase activity per mg of protein (17.86 units/mg) than lysates from recombinant E. coli expressing lctO type 1 (14.47 units/mg) (Table 3). When specific activity of the two enzymes was calculated from the densitometric analysis, the difference was more substantial as the type 2 enzyme was less efficiently induced in the recombinant system and thus constituted a lower proportion of the total protein in the lysate assayed (Fig. 3 and Table 3).

View larger version (133K): [in this window] [in a new window]

FIG. 3. SDS-PAGE gels of IPTG-induced recombinant E. coli lysates expressing type 1 and type 2 lactate oxidase enzymes. Lysate (40 µg per lane) (lanes 2 to 5) from each of the recombinant isolates was separated in SDS-PAGE gels (10% acrylamide resolving gel) under reducing conditions. Molecular weight markers (Precision Plus; Bio-Rad) were included on each gel (lane 1). Gels were scanned with an Image Quant 400 system (GE Healthcare Biosciences) using transmitted white light.

MIC of lactate against isolates expressing different lactate oxidase enzymes.
In order to ascertain whether the inserted repeat in lctO type 2 and the associated change in enzyme activity had any physiological significance, the MIC of lactate was determined in LB broth and on LB agar. LB was chosen over vegetable peptone or blood agar to minimize potential inhibition of lactate oxidase enzyme expression in S. iniae isolates by glucose in the medium. MICs of lactate using LB broth as the medium were 1.5% for all strains assayed, regardless of the lctO type expressed. When lactate MICs were determined on LB agar, the MIC was higher at 2.5% to 3.0%. There were no consistent differences in the lactate MIC for recombinant strains expressing either the type 1 or type 2 enzyme or between S. iniae isolates QMA0165 and QMA0177.

Effect of lactate on growth rates of strains expressing the different enzymes.
As no differences were detected in the MICs of lactate between the strains used, the effect of various lactate concentrations on growth rate in LB broth was determined. A nonlinear regression based on the Gompertz equation was used to fit the curves to the growth data. Goodness of fit was determined, and R² values ranged between 0.9630 and 0.9797 for curves fitted to growth data for QMA0165 at the different lactate concentrations and between 0.9369 and 0.9730 for QMA0177. Having ascertained that curve fit was acceptable, we performed an F test on the data sets for each of the isolates to determine whether there were differences between the growth curves at the different lactate concentrations, with the null hypothesis in each case being that a single curve could fit all the data sets for each strain. In the case of strain QMA0165 expressing the type 1 enzyme, the null hypothesis was rejected (P = 0.0096), indicating a significant effect of lactate concentration on growth (Fig. 4A). In contrast, the F test on the data sets for strain QMA0177, expressing the type 2 enzyme, did not reject the null hypothesis (P = 0.9735), indicating that a single curve could fit all growth data regardless of lactate concentration (P = 0.9735) (Fig. 4B).

View larger version (11K): [in this window] [in a new window]

FIG. 4. Influence of lactate on growth rates of similar marine strains of S. iniae in LB broth. Growth determined by optical density at 600 nm (OD600) of QMA0165 expressing type 1 enzyme and QMA0177 expressing type 2 enzyme in LB broth containing 0, 0.1, 0.3, and 0.5% potassium lactate. Data points indicate means, and error bars represent standard deviations from three independently inoculated cultures. Curves were fitted using the Gompertz equation followed by an F test. The F test supported separate curves for growth data for QMA0165 (P = 0.0096) but a single curve for growth data for QMA0177 (P = 0.9735).

The effect of lactate concentration on growth of each strain over time was also analyzed by ANOVA. In agreement with the nonlinear regression analyses and F test comparisons, no significant effect of lactate concentration on growth over time was detected in strain QMA0177 (P = 0.7628) (Fig. 4B). However, growth of QMA0165 was significantly inhibited at lactate concentrations of 0.3% and 0.5% compared to controls (P = 0.0170) (Fig. 4A).

While assays of additional strains indicated that there was some between-strain variation in the growth kinetics that could not be attributed to the different enzymes (Fig. 5), the strains expressing the type 2 enzyme were not in general adversely affected by high concentrations of lactate in LB medium, and growth seemed to be maximal with 0.3% lactate in these strains (Fig. 5B, D, and F). In contrast, with the exception of strain QMA0218 (which is highly mucoid), the strains expressing the type 1 enzyme were generally inhibited by higher concentrations of lactate (Fig. 5A, C, and E). For all strains except QMA0218, growth curves at each lactate concentration were significantly different from each other, and the null hypothesis (the same curve for each lactate concentration) was rejected by an F test.

View larger version (16K): [in this window] [in a new window]

FIG. 5. Influence of lactate on growth rates of diverse S. iniae strains in LB broth. Growth determined by optical density at 600 nm (OD600) of QMA0088 (type 1), QMA0109 (type 2), QMA0218 (type 1), QMA0119 (type 2), QMA0221 (type 1), and QMA0126 (type 2) in LB broth containing 0, 0.1, 0.3, and 0.5% potassium lactate. Data points indicate means, and error bars represent standard deviations from three independently inoculated cultures. Curves were fitted using the Gompertz equation followed by an F test.

Although LB broth does not contain glucose, it is a complex medium containing many potential sources of carbohydrate. Thus, the experiment was repeated a further time using the same six strains. On this occasion, a basic substrate utilization medium was prepared from commercially available PBS containing calcium and magnesium to which ammonium chloride was added as a source of nitrogen. This basic medium was then supplemented with either lactate or glucose as carbon sources. Growth was very low and slow in this medium for all strains, and there was variation between strains (Fig. 6). Strains did not grow as well in lactate as in glucose (Fig. 6), and this difference was always statistically significant, regardless of the enzyme type expressed. However, strains expressing type 1 enzyme grew considerably less well with lactate instead of glucose as carbon source than strains expressing the type 2 enzyme (Fig. 6). In strains QMA0088, QMA0218, and QMA0221, lactate accounted for 10.30% (P < 0.0001), 5.89% (P = 0.0022), and 8.02% (P < 0.0001) of the total variance, respectively, whereas in strains QMA0109, QMA0119, and QMA0126, lactate accounted for 2.32% (P = 0.0002), 4.09% (P = 0.0029), and 2.49% (P = 0.0055) of the total variance, respectively.

View larger version (13K): [in this window] [in a new window]

FIG. 6. Glucose and lactate utilization by diverse S. iniae strains expressing type 1 or type 2 lactate oxidase. Growth determined by optical density at 600 nm (OD600) of QMA0088 (type 1), QMA0109 (type 2), QMA0218 (type 1), QMA0119 (type 2), QMA0221 (type 1), and QMA0126 (type 2) in PBS containing Ca2+, Mg2+, and NH4Cl supplemented with either 0.2% potassium lactate or 0.2% D-glucose. Data points indicate means, and error bars represent standard deviations from three independently inoculated cultures. Curves were fitted using the Gompertz equation followed by an F test.

Amino acid sequence and structure analysis.
To elucidate the potential structural alterations that may account for the increased enzyme activity, the amino acid sequences derived from the type 1 and type 2 lactate oxidase enzymes from S. iniae were aligned and compared with a thoroughly characterized lactate oxidase from A. viridans (11, 29, 30). The alignment was used to predict a three-dimensional structural model of the lactate oxidase monomer in PyMOL. Amino acid alignment of lactate oxidase in S. iniae from this study with a previously described lactate oxidase (11) indicates some differences in the structure (Fig. 7); the previously described lactate oxidase shows dissimilarities in three positions prior to the first 382 amino acids and major differences in the C-terminal region. Alignment of lactate oxidases from S. iniae and A. viridans shows strongly conserved regions and indicates that the majority of the amino acids involved in the active enzymatic site of A. viridans are conserved in S. iniae (Fig. 7).

View larger version (74K): [in this window] [in a new window]

FIG. 7. Alignments and comparisons of lactate oxidase proteins in S. iniae and A. viridans. Shading indicates homology (black, 100%; gray, 80%; and light gray, 60%). Black dots below alignment denote conserved amino acids within monomers that participate in the enzymatically active site of LOX in A. viridans. The gray dot below the alignment denotes a conserved amino acid suggested to be involved in the enzymatic active site on a neighboring subunit. Accession numbers are as follows: S. iniae lctO type 1, EU086697; S. iniae lctO type 2, EU086704; S. iniae lactate oxidase lctO Si, Y07622; A. viridans lactate oxidase, BAA09172.

Predicted models of lactate oxidase in S. iniae both showed very good fit to template models (E = 1e–104 for type 1 and E = 1e–100 for type 2, respectively), indicating that our models are relatively robust. Comparisons of A. viridans lactate oxidase and the two S. iniae lactate oxidases suggest that the inserted sequence for type 2 lactate oxidase in S. iniae results in an additional loop structure in close proximity to the active site in the neighboring subunit of the lactate oxidase protein complex (Fig. 8).

View larger version (51K): [in this window] [in a new window]

FIG. 8. Protein structure of lactate oxidase in A. viridans and predicted protein structure models of type 1 and type 2 lactate oxidases in S. iniae visualized in PyMOL. (A) Partial crystal structure of A. viridans (PDB entry 2DU2) showing interactions between subunits 2 (yellow) and 3 (green). (B) Predicted model of type 1 lactate oxidase in S. iniae (based on PDB entry 2J6X). (C) Predicted model of type 2 lactate oxidase in S. iniae (based on PDB entry 2DU2). Green and yellow denote two separate subunits. Blue sticks show determined (A) or predicted (B and C) active sites for substrate. Light blue sticks show suggested (A) and predicted (B and C) amino acids from a neighboring subunit participating in the enzymatic reaction at the active site. Red sticks show predicted positions of inserted amino acids in S. iniae type 2 lactate oxidase variants.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Cloning and sequencing the full-length lactate oxidase gene from representative strains of S. iniae allowed two lctO types to be identified on the basis of size. Apart from the insertion sequence found in the 3' end of the gene in some isolates, three nucleotides in positions 211 to 213, not previously detected when the gene was described by Gibello et al. (11), resulted in an inserted valine residue in the translated product from all isolates. We note an apparent error in the primary sequence and translation of the GenBank sequence (Y07622). This is likely to be due to an inserted C nucleotide at position 1148 of the lctO gene at the far 3' end of the gene sequence (inside the LOX-1/LOX-2 priming region) that has altered the reading frame (11). This means that the expected PCR product size of 870 bp is incorrect and is actually 869 bp. A factor that has greater implications for the specificity of the diagnostic PCR is the misinterpretation of the position of the stop codon since the 3' end of the LOX-2 primer is located downstream of the stop codon in all of the isolates that we sequenced. Variations in intergenic spacer regions are likely to produce results that are slightly different from those expected. We also detected a mismatch in the LOX-2 primer sequence with all of the isolates presented here and suggest that the basis of the design of the LOX-2 primer is flawed since the data used to design the primer contained errors. The redesign of the LOX-2 primer or its replacement with a primer that is located in the lctO gene itself should be considered although this may reduce the specificity of the assay due to the conserved 3' end of the lactate oxidase gene.

In Australian isolates the production of an amplicon of 869 bp is not the only result that can be obtained using the LOX-1/LOX-2 primer set with S. iniae. While the method has been tested for specificity (21), there needs to be awareness among diagnosticians that amplicons that are of higher molecular weight can be produced by atypical strains. These larger amplicons should not be disregarded and may be serendipitous indicators that can be used in epidemiological studies of S. iniae.

Type 1 lactate oxidase appears to be the dominant form of LOX protein in S. iniae in the strain collection used in this study. In total, 105 strains mainly from Australia, but also from Israel, Réunion Island, Thailand, the United States, and Canada were investigated. The type 1 lactate oxidase protein structure of S. iniae shares similarities with lactate oxidase proteins described in other organisms such as that of A. viridans, and it is interesting that the type 2 lactate oxidase in S. iniae was detected only in strains derived from a single location in the Northern Territory, Australia. Type 2 variants contain a repeat insert of 17 additional amino acids that are predicted to form an additional loop in close proximity to the active binding site of a neighboring subunit in the enzyme complex. The lactate oxidase activity in type 2 is enzymatically more efficient than that of type 1, leading us to speculate on the function of the additional insert. The predicted structure suggests that the additional loop may function in enhancing enzyme activity at the binding site by improving capture of the substrate. The existence of the extra loop changes conformation of the multimeric lactate oxidase structure where, in addition to increasing the enzyme efficiency, it may act to stabilize the enzyme. The flip-flop action of the His265 in the lactate oxidase enzyme-D-lactate complex in A. viridans increases the space within the active site of the lactate oxidase enzyme. Changes in the conformational structure have also led to the formation of a hydrogen-bonding network between the residues His265-Asp174-Lys221. All these residues are very well conserved in the -hydroxyoxidase family members (29). It has also been suggested that the hydrogen bond network is important for the enzymatic activity (10). Changes in the lactate oxidase structure have also been reported to broaden the active site of the lactate oxidase enzyme (17). Binding of the active site to the substrate may induce a conformational change, as reported during substrate binding in the A. viridans lactate oxidase enzyme (10), which may involve the loop in efficient substrate channeling to the active site. However, future crystallographic studies on the interaction of the enzyme with the substrate will be required to reveal the true function of the insert.

The questions of why and how this variant of the gene arose at only one farm site and how a larger gene product has been maintained remain to be answered. The increased activity of the larger enzyme is indicative of a potential requirement for improved lactate metabolism in that particular environment. This is supported by the preliminary experiment investigating the effect of lactate on the growth rate of the two strains isolated in the same year under similar conditions (both from barramundi in large commercial open seawater farms) but expressing two different enzymes. Lactate concentrations of 0.3% and 0.5% significantly reduced the growth rate of the strain containing the type 1 enzyme (QMA0165) while there was no significant effect on growth rate of the strain containing the type 2 enzyme (QMA0177). Further investigation using multiple strains indicated there was significant strain-to-strain variability among growth rates in the presence or absence of lactate that could not be accounted for by enzyme type, but there was a trend for strains expressing type 2 enzyme to be less adversely affected by high concentrations of lactate than those expressing the type 1 variant. Clearly, carbohydrate metabolism in bacteria is highly complex and will be affected by multiple factors including nutrient availability, exopolysaccharide production, oxygen availability, and other factors; thus, this area requires substantial further expert investigation using highly controlled culture environments. Moreover, S. iniae is a highly diverse species (14, 23), and variability in growth characteristics should be expected. However, our preliminary experiments suggest that the type 2 enzyme may provide a competitive advantage under conditions where the lactate concentration is high. Intriguingly, the farm from which the variant was isolated is unusual as it is a marine farm with extremely high tidal flow (8 to 9 meters). This results in currents in excess of 3 knots through the anchored stainless steel cages, forcing the fish to swim periodically in bursts to prevent collision with the cage. During exercise, lactate is produced in the white muscle tissue, which comprises around 65% of the total muscle, of teleost fish (7, 31). The fate of muscle lactate and the rate at which it is shuttled into plasma in teleosts are species dependent (31, 32). While lactate metabolism has not been studied in barramundi, it is a strike predator that will undergo periods of burst swimming likely to result in plasma lactate accumulation similar to that reported for the silver trevally (Pseudocarnax dentex) (32), in which plasma lactate levels in excess of 21 mM were reported; this level is reported to be inhibitory to S. iniae growth in brain heart infusion broth (11) and reduced the growth rate of the type 1 enzyme-containing strain in this study. The strains isolated from this site are the only ones to date that have a type 2 lctO gene from an international collection of more than 100 strains from geographically disparate locations and hosts. Moreover, our earliest isolate from this site (QMA0109) isolated in April 2005 expresses the type 2 enzyme, and the type 2 enzyme has been maintained throughout subsequent outbreaks of disease. Intriguingly, unpublished work conducted in our laboratory has indicated that later strains isolated from this site following vaccination in July 2005 (QMA0126 and all other 2006 isolates from the farm) have substantial coding mutations in the capsular operon, resulting in a serotype switch. In spite of these changes, perhaps selected by vaccination, the type 2 lactate oxidase gene has been maintained. Since these isolates from the Northern Territory are able to process lactate at a faster rate than normal lctO type 1 strains, it is tantalizing to hypothesize that environmental influences of large tidal flows linked with increased swimming activity have led to the evolution and retention of the larger lctO gene encoding a more efficient enzyme in these isolates of S. iniae.

 
* Corresponding author. Mailing address: Aquatic Animal Health Laboratory, Centre for Marine Studies, The University of Queensland, Brisbane, Queensland 4072, Australia. Phone: 61 7 3346 9416. Fax: 61 7 3365 4755. E-mail: a.barnes{at}uq.edu.au

Published ahead of print on 6 March 2009.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Agnew, W., and A. C. Barnes. 2007. Streptococcus iniae: an aquatic pathogen of global veterinary significance and a challenging candidate for reliable vaccination. Vet. Microbiol. 122:1-15.[CrossRef][Medline]
2
2. Baiano, J. C., R. A. Tumbol, A. Umapathy, and A. C. Barnes. 2008. Identification and molecular characterisation of a fibrinogen binding protein from Streptococcus iniae. BMC Microbiol. 8:67.[CrossRef][Medline]
3
3. Barnes, A. C., M. T. Horne, and A. E. Ellis. 2003. Streptococcus iniae expresses a cell surface non-immune trout immunoglobulin-binding factor when grown in normal trout serum. Fish Shellfish Immunol. 15:425-431.[CrossRef][Medline]
4
4. Barnes, A. C., F. M. Young, M. T. Horne, and A. E. Ellis. 2003. Streptococcus iniae: serological differences, presence of capsule and resistance to immune serum killing. Dis. Aquat. Org. 53:241-247.[Medline]
5
5. Berridge, B. R., J. D. Fuller, J. de Azavedo, D. E. Low, H. Bercovier, and P. F. Frelier. 1998. Development of specific nested oligonucleotide PCR primers for the Streptococcus iniae 16S-23S ribosomal DNA intergenic spacer. J. Clin. Microbiol. 36:2778-2781.[Abstract/Free Full Text]
6
6. Creeper, J. H., and N. B. Buller. 2006. An outbreak of Streptococcus iniae in barramundi (Lates calcarifera) in freshwater cage culture. Aust. Vet. J. 84:408-411.[CrossRef][Medline]
7
7. Dalla Via, J., G. vanden Thillart, O. Cattani, and P. Cortesi. 1997. Environmental versus functional hypoxia/anoxia in sole Solea solea: the lactate paradox revisited. Mar. Ecol. Prog. Ser. 154:79-90.[CrossRef]
8
8. DeLano, W. L. 2002. The PyMOL molecular graphics system. DeLano Scientific, Palo Alto, CA.
9
9. Duncan, J. D., J. O. Wallis, and M. R. Azari. 1989. Purification and properties of Aerococcus viridans lactate oxidase. Biochem. Biophys. Res. Commun. 164:919-926.[CrossRef][Medline]
10
10. Furuichi, M., N. Suzuki, B. Dhakshnamoorhty, H. Minagawa, R. Yamagishi, Y. Watanabe, Y. Goto, H. Kaneko, Y. Yoshida, H. Yagi, I. Waga, P. K. Kumar, and H. Mizuno. 2008. X-ray structures of Aerococcus viridans lactate oxidase and its complex with D-lactate at pH 4.5 show an alpha-hydroxyacid oxidation mechanism. J. Mol. Biol. 378:436-446.[CrossRef][Medline]
11
11. Gibello, A., M. D. Collins, L. Dominguez, J. F. Fernandez-Garayzabal, and P. T. Richardson. 1999. Cloning and analysis of the L-lactate utilization genes from Streptococcus iniae. Appl. Environ. Microbiol. 65:4346-4350.[Abstract/Free Full Text]
12
12. Goh, S. H., D. Driedger, S. Gillett, D. E. Low, S. M. Hemmingsen, M. Amos, D. Chan, M. Lovgren, B. M. Willey, C. Shaw, and J. A. Smith. 1998. Streptococcus iniae, a human and animal pathogen: specific identification by the chaperonin 60 gene identification method. J. Clin. Microbiol. 36:2164-2166.[Abstract/Free Full Text]
13
13. Janda, J. M., and S. L. Abbott. 2007. 16S rRNA gene sequencing for bacterial identification in the diagnostic laboratory: pluses, perils, and pitfalls. J. Clin. Microbiol. 45:2761-2764.[Free Full Text]
14
14. Lahav, D., M. Eyngor, A. Hurvitz, C. Ghittino, A. Lublin, and A. Eldar. 2004. Streptococcus iniae type II infections in rainbow trout Oncorhynchus mykiss. Dis. Aquat. Org. 62:177-180.[CrossRef][Medline]
15
15. Lane, D. J. 1991. 16S/23S rRNA sequencing, p. 115-175. In E. Stackebrandt and M. Goodfellow (ed.), Nucleic acid techniques in bacterial systematics. John Wiley, New York, NY.
16
16. Leiros, I., E. Wang, T. Rasmussen, E. Oksanen, H. Repo, S. B. Petersen, P. Heikinheimo, and E. Hough. 2006. The 2.1 Å structure of Aerococcus viridans L-lactate oxidase (LOX). Acta Crystallogr. F 62:1185-1190.[CrossRef]
17
17. Li, S. J., Y. Umena, K. Yorita, T. Matsuoka, A. Kita, K. Fukui, and Y. Morimoto. 2007. Crystallographic study on the interaction of L-lactate oxidase with pyruvate at 1.9 angstrom resolution. Biochem. Biophys. Res. Commun. 358:1002-1007.[CrossRef][Medline]
18
18. Lowe, B. A., J. D. Miller, and M. N. Neely. 2007. Analysis of the polysaccharide capsule of the systemic pathogen Streptococcus iniae and its implications in virulence. Infect. Immun. 75:1255-1264.[Abstract/Free Full Text]
19
19. Lund, O., M. Nielsen, C. Lundegaard, and P. Worning. 2002. CPHmodels 2.0: X3M a computer program to extract 3D models. Center for Biological Sequence Analysis, Technical University of Denmark, Lyngby.
20
20. Maeda-Yorita, K., K. Aki, H. Sagai, H. Misaki, and V. Massey. 1995. L-Lactate oxidase and L-lactate monooxygenase: mechanistic variations on a common structural theme. Biochimie 77:631-642.[CrossRef][Medline]
21
21. Mata, A. I., M. M. Blanco, L. Dominguez, J. F. Fernandez-Garayzabal, and A. Gibello. 2004. Development of a PCR assay for Streptococcus iniae based on the lactate oxidase (lctO) gene with potential diagnostic value. Vet. Microbiol. 101:109-116.[CrossRef][Medline]
22
22. Miller, J. D., and M. N. Neely. 2005. Large-scale screen highlights the importance of capsule for virulence in the zoonotic pathogen Streptococcus iniae. Infect. Immun. 73:921-934.[Abstract/Free Full Text]
23
23. Nawawi, R. A., J. C. F. Baiano, and A. C. Barnes. 2008. Genetic variability amongst Streptococcus iniae isolates from Australia. J. Fish Dis. 31:305-309.[CrossRef][Medline]
24
24. Pruksakorn, S., N. Sittisombut, C. Phornphutkul, C. Pruksachatkunakorn, M. F. Good, and E. Brandt. 2000. Epidemiological analysis of non-M-typeable group A streptococcus isolates from a Thai population in northern Thailand. J. Clin. Microbiol. 38:1250-1254.[Abstract/Free Full Text]
25
25. Roach, J. C. M., P. N. Levett, and M. C. Lavoie. 2006. Identification of Streptococcus iniae by commercial bacterial identification systems. J. Microbiol. Methods 67:20-26.[CrossRef][Medline]
26
26. Seki, M., K. Iida, M. Saito, H. Nakayama, and S. Yoshida. 2004. Hydrogen peroxide production in Streptococcus pyogenes: involvement of lactate oxidase and coupling with aerobic utilization of lactate. J. Bacteriol. 186:2046-2051.[Abstract/Free Full Text]
27
27. Shoemaker, C. A., P. H. Klesius, and J. J. Evans. 2001. Prevalence of Streptococcus iniae in tilapia, hybrid striped bass, and channel catfish on commercial fish farms in the United States. Am. J. Vet. Res. 62:174-177.[CrossRef][Medline]
28
28. Thompson, J. D., T. J. Gibson, F. Plewniak, F. Jeanmougin, and D. G. Higgins. 1997. The CLUSTAL_X Windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25:4876-4882.[Abstract/Free Full Text]
29
29. Umena, Y., K. Yorita, T. Matsuoka, M. Abe, A. Kita, K. Fukui, T. Tsukihara, and Y. Morimoto. 2005. Crystallization and preliminary X-ray diffraction study of L-lactate oxidase (LOX), R181M mutant, from Aerococcus viridans. Acta Crystallogr. F 61:439-441.[CrossRef]
30
30. Umena, Y., K. Yorita, T. Matsuoka, A. Kita, K. Fukui, and Y. Morimoto. 2006. The crystal structure of L-lactate oxidase from Aerococcus viridans at 2.1 Å resolution reveals the mechanism of strict substrate recognition. Biochem. Biophys. Res. Commun. 350:249-256.[CrossRef][Medline]
31
31. van Ginneken, V. R. Boot, T. Murk, G. van den Thillart, and P. Balm. 2004. Blood plasma substrates and muscle lactic-acid response after exhaustive exercise in common carp and trout: indications for a limited lactate-shuttle. Anim. Biol. Leiden Neth. 54:119-130.[CrossRef]
32
32. Wells, R. M. G., and J. Baldwin. 2006. Plasma lactate and glucose flushes following burst swimming in silver trevally (Pseudocaranx dentex: Carangidae) support the "releaser" hypothesis. Comp. Biochem. Physiol. A 143:347-352.[CrossRef][Medline]
33
33. Wijkström, U., A. Gumy, and R. Grainger. 2004. The state of world fisheries and aquaculture. Food and Agriculture Organization of the United Nations, Rome, Italy.
34
34. Wood, R., A. Peat, and D. F. Galeano. 2007. Australian fisheries statistics, 2006. Report 1037-6897. Australian Bureau of Agricultural and Resource Economics, Canberra.
35
35. Yorita, K., T. Matsuoka, H. Misaki, and V. Massey. 2000. Interaction of two arginine residues in lactate oxidase with the enzyme flavin: conversion of FMN to 8-formyl-FMN. Proc. Natl. Acad. Sci. USA 97:13039-13044.[Abstract/Free Full Text]
36
36. Yorita, K., H. Misaki, B. A. Palfey, and V. Massey. 2000. On the interpretation of quantitative structure-function activity relationship data for lactate oxidase. Proc. Natl. Acad. Sci. USA 97:2480-2485.[Abstract/Free Full Text]
37
37. Zlotkin, A., A. Eldar, C. Ghittino, and H. Bercovier. 1998. Identification of Lactococcus garvieae by PCR. J. Clin. Microbiol. 36:983-985.[Abstract/Free Full Text]

---

Applied and Environmental Microbiology, May 2009, p. 2908-2919, Vol. 75, No. 9
0099-2240/09/$08.00+0     doi:10.1128/AEM.02147-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Nawawi, R. A. Articles by Barnes, A. C. Search for Related Content PubMed PubMed Citation Articles by Nawawi, R. A. Articles by Barnes, A. C. Agricola Articles by Nawawi, R. A. Articles by Barnes, A. C.