- Email: [email protected]
Characterization of β-glucan formation by Lactobacillus brevis TMW 1.2112 isolated from slimy spoiled beer
Accepted Manuscript Title: Characterization of -glucan formation by Lactobacillus brevis TMW 1.2112 isolated from slimy spoiled beer Authors: Marion E. Fraunhofer, Andreas J. Geissler, Daniel Wefers, Mirko Bunzel, Frank Jakob, Rudi F. Vogel PII: DOI: Reference:
S0141-8130(17)32535-7 http://dx.doi.org/10.1016/j.ijbiomac.2017.09.063 BIOMAC 8240
To appear in:
International Journal of Biological Macromolecules
Received date: Revised date: Accepted date:
12-7-2017 1-9-2017 17-9-2017
Please cite this article as: Marion E.Fraunhofer, Andreas J.Geissler, Daniel Wefers, Mirko Bunzel, Frank Jakob, Rudi F.Vogel, Characterization of -glucan formation by Lactobacillus brevis TMW 1.2112 isolated from slimy spoiled beer, International Journal of Biological Macromoleculeshttp://dx.doi.org/10.1016/j.ijbiomac.2017.09.063 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Characterization of β-glucan formation by Lactobacillus brevis TMW 1.2112 isolated from slimy spoiled beer
1
Marion E Fraunhofer, 1Andreas J Geissler, 2Daniel Wefers, 2Mirko Bunzel, 1Frank Jakob*,
1
Rudi F Vogel
1
Lehrstuhl für Technische Mikrobiologie, Technische Universität München, Gregor-Mendel-
Straße 4, 85354 Freising, Germany 2
Karlsruhe Institute of Technology (KIT), Institute of Applied Biosciences, Department of
Food Chemistry and Phytochemistry, 76131 Karlsruhe, Germany
* address for correspondence Dr. Frank Jakob, Lehrstuhl für Technische Mikrobiologie, Technische Universität München, Gregor-Mendel-Straße 4, 85354 Freising, Germany, Email: [email protected]
Highlights:
The beer spoiler L. brevis TMW 1.2112 produces capsular polysaccharides (CPS)
The CPS are composed of β-(1,3-1,2)-linked glucose moieties (β-glucan)
(Putative) glucan-degrading genes suggest CPS as carbon source during starvation
Gtf-2 is an essential enzyme for β-glucan synthesis in L. brevis TMW 1.2112
The gtf-2 gene is located on a plasmid in L. brevis TMW 1.2112
Gtf-2 was proven as suited marker gene for detection of slime-forming beer spoilers
Abstract
1
Despite several hurdles, which hinder bacterial growth in beer, certain bacteria are still able to spoil beer. One type of spoilage is characterized by an increased viscosity and slimy texture caused by exopolysaccharide (EPS) formation of lactic acid bacteria (LAB). In this study, we characterize for the first time EPS production in a beer-spoiling strain (TMW 1.2112) of Lactobacillus brevis, a species commonly involved in beer spoilage. The strain´s growth dynamics were assessed and we found an increased viscosity or ropiness in liquid or on solid media, respectively. Capsular polysaccharides (CPS) and released EPS from cells or supernatant, respectively, were analyzed via NMR spectroscopy and methylation analysis. Both are identical β-(1→3)-glucans, which are ramified with β-glucose residues at position O2. Therefore, we assume that this EPS is mainly produced as CPS and partially released into the surrounding medium, causing viscosity of e.g. beer. CPS formation was confirmed via agglutination test. A plasmid-located glycosyltransferase-2 was found as responsible for excess β-glucan formation, chromosomal glucanases were proposed for its degradation. The glycosyltransferase-2 gene could also be specifically identified in beer-spoiling, slimeproducing Lactobacillus rossiae and Lactobacillus parabuchneri strains, suggesting it as promising marker gene for early detection of β-glucan-producing Lactobacilli in breweries. Highlights:
The beer spoiler L. brevis TMW 1.2112 produces capsular polysaccharides (CPS)
The CPS are composed of β-(1,3-1,2)-linked glucose moieties (β-glucan)
(Putative) glucan-degrading genes suggest CPS as carbon source during starvation
Gtf-2 is an essential enzyme for β-glucan synthesis in L. brevis TMW 1.2112
The gtf-2 gene is located on a plasmid in L. brevis TMW 1.2112
Gtf-2 was proven as suited marker gene for detection of slime-forming beer spoilers
Key words: beer spoilage, Lactobacillus brevis, exopolysaccharide, β-glucan
2
1. Introduction Beer is considered as a stable beverage since it exhibits a harsh environment for bacteria. The low pH and nutrient availability, the oxygen-deficient but carbon dioxide rich environment as well as the presence of ethanol and antimicrobial hop compounds hinder most bacterial growth [1]. Nevertheless, there are bacteria which have developed strategies to overcome these hurdles, namely by picking up respective properties from a trans-species shared plasmid pool [2, 3]. The mechanisms behind bacterial beer spoilage are increasingly understood and were shown to be mainly associated with the ability to counteract hop compounds. Hops act as ionophores, which dissipate the transmembrane proton gradient and consequently abrogate the proton motive force [4-6]. This in turn inhibits nutrient uptake, essential enzyme reactions and finally causes the cell’s death [7-9]. The best-studied mechanisms antagonizing these detrimental effects are the modification of the cell wall to increase the barrier function and consequently to decrease the influx of hops, as well as the expression of multidrug transporters like horA and horC, which extrude hop compounds out of the cell [9, 10]. Utilizing these mechanisms, amongst others, bacteria are able to grow in and spoil beer. Besides strains causing turbidity and off flavors upon growth and metabolite formation, respectively, beer-spoiling bacteria also comprise strains, which produce slime increasing the viscosity of beer. This alteration is mostly attributed to certain lactic acid bacteria (LAB), forming exopolysaccharides (EPS). EPS are long chain, high molecular weight polymers, which are either composed from one type of monosaccharide (mostly glucose or fructose) or comprise a repeating unit of two or more different monosaccharides. They are therefore 3
classified as homopolysaccharides or heteropolysaccharides, respectively. Both can either be associated with the cell surface in form of a capsular polysaccharide (CPS) or be released into the surrounding environment as extracellular polysaccharide (EPS). EPS are thought to confer bacterial cells several advantages like the protection against phagocytosis, desiccation and toxins as well as an facilitated adherence to surfaces and a source of nutrients [11]. Due to their hydrocolloid properties, EPS and/or their producer strains are intentionally used in some food industries to improve e.g. the food texture [12-15]. At the same time, these characteristics make EPS-producing bacteria to an undesired contaminant in beer or other beverages. While the phenomenon of EPS formation by wine-spoiling bacteria is well studied [16, 17], respective beer-spoiling lactobacilli are still sparsely explored. Therefore, the objective of the present study was to get deeper insights into this special type of beer spoilage, by characterizing Lactobacillus (L.) brevis TMW 1.2112, isolated from spoiled, viscous beer regarding its capability to form EPS. Furthermore, the process of EPS formation and degradation was described and linked to the genetic background, which were identified via whole genome sequencing and comparative genomics/genetics. Finally, the structure of the produced polysaccharide was elucidated via NMR spectroscopy. 2. Materials and methods 2.1 Bacterial strains, growth conditions and media L. brevis TMW 1.2112 was isolated from spoiled, viscous beer and identified on species level using Matrix-Assisted-Laser-Desorption/Ionization-Time-Of-Flight Mass (MALDI-TOF MS) Spectrometry as reported by CC Kern, JC Usbeck, RF Vogel and J Behr [18]. For long term preservation the bacterial culture was cryoconserved according to AJ Geissler, J Behr, K von Kamp and RF Vogel [19]. As working culture bacteria were grown at 30°C in modified MRS medium (mMRS) adjusted to pH 4.3. Growth medium composition (quantities per liter): 10 g 4
peptone, 5 g yeast extract, 5 g meat extract, 4 g K2HPO4, 2.6 g KH2PO4·3H2O, 3 g NH4Cl, 1 g Tween80, 0.5 g cysteine-HCl, 20 g maltose, 0.2 g MgSO4·7H2O, 0.038 g MnSO4·H2O. For EPS isolation a variant of the medium without yeast and meat extract, but supplemented with a vitamin mix (quantities per liter: 0.2 mg thiamin, 0.2 mg niacin, 0.2 mg folic acid, 0.2 mg mg pyridoxal, 0.2 mg pantothenic acid, 0.2 mg cobalamine) was used. All other strains used in this study (Table 1) were handled like L. brevis TMW 1.2112. 2.2 Macroscopic characterization of EPS formation To visualize growth dynamics and slime formation a colony of L. brevis TMW 1.2112 was inoculated in 1.8 ml mMRS broth and incubated at 30 °C. Photos were taken over a time span of six days. 2.3 Isolation of capsular polysaccharides and released extracellular polysaccharides To differentiate between cell-bound CPS and released EPS, two methods were used according to R Tallon, P Bressollier and MC Urdaci [20]. The strain was cultured for three days at 30 °C in mMRS broth. Subsequently, the cells were separated from the supernatant by centrifugation (12000 x g, 60 min). The EPS in the supernatant were precipitated with cold ethanol (2:1 v/v) and kept at 4 °C overnight. After centrifugation, the resulting precipitate was dissolved in deionized water, dialyzed against deionized water for three days and lyophilized. To isolate the cell-bound CPS, the cell pellet was washed twice with PBS buffer and dissolved in 1 M NaCl. To detach the polysaccharides from the cells, the pellet was sonicated (3 x 30 sec, power 90 %) (HD-70/ Bandelin electronic, Berlin, Germany). Further processing of CPS was conducted as described for EPS. 2.4 Structural characterization
5
Glycosidic linkages were analyzed by methylation analysis as described previously [21]. Briefly, polysaccharides (3 mg) were dissolved in dimethyl sulfoxide and methylated by the addition of finely ground sodium hydroxide and methyl iodide. The methylated polysaccharides were extracted into dichloromethane, dried, and subsequently hydrolyzed with 2 M TFA for 1.5 h at 121 °C. The solvent was evaporated, and partially methylated monosaccharides were reduced by the addition of sodium borodeuteride. Following acetylation and extraction, the partially methylated alditol acetates (PMAAs) were analyzed by GC-MS and GC-FID (GC2010 Plus, GCMS-QP2010) (Shimadzu, Kyoto, Japan), both equipped with a DB-225 column (30 m x 0.25 mm i.d., 0.25 μm) (Agilent Technologies, Santa Clara, CA). Molar response factors according to DP Sweet, RH Shapiro and P Albersheim [22] were used for semiquantitative estimation of the PMAA ratios. Monosaccharide composition was determined after sulfuric acid hydrolysis (pretreatment in 12 M H2SO4 for 2.5 h, dilution to 1.6 M H2SO4 and hydrolysis for 3 h at 100 °C) according to JF Saeman, JL Bubl and EE Harris [23] and methanolysis (16 h methanolysis with 1.25 M methanolic HCl at 80 °C, 1 h hydrolysis with 2 M TFA at 121 °C) according to GA De Ruiter, HA Schols, AGJ Voragen and FM Rombouts [24]. The hydrolysates were analyzed for their monosaccharide composition by high performance anion exchange chromatography as described previously [21]. To determine the absolute configuration, the exopolysaccharides were hydrolyzed with 2 M TFA for 30 min at 121°C and their silylated (R)-2-octanol derivatives were analyzed by GC-MS as described previously [25]. For NMR spectroscopic characterization, samples (3 mg) were hydrogen-deuterium exchanged and subsequently dissolved in deuterium oxide (500 µL). NMR spectra were acquired at 298 K on a Bruker Ascend 500 MHz spectrometer equipped with a Prodigy cryoprobe (Bruker, Rheinstetten, Germany). Acetone was used as internal reference (1H: 2.22 ppm, 13C: 30.89 ppm) [26]. 2.5 Immunological analysis 6
Agglutination tests were performed with Streptococcus (S.) pneumoniae type 37-specific antisera (Statens Serum Institut, Copenhagen, Denmark) as previously described by E Walling, E Gindreau and A Lonvaud-Funel [27], with slight modifications. Briefly, overnight cultures were centrifuged (10000 x g, 30 min) and the cells were resuspended in PBS buffer. After blending 10 µl of the culture with 10 µl of the antiserum, the mixture was incubated for 30 min at 4 °C and subsequently analyzed microscopically at thousand-fold magnification (Axiostar Plus, Zeiss, Oberkochen, Germany). 2.6 Genomics For genome sequencing, high molecular weight DNA was isolated according to the instructions of the Qiagen Genomic Kit (Qiagen, Hilden, Germany). Single-molecule real-time sequencing (PacBio RS II) was performed at GATC Biotech (Konstanz, Germany) [28]. A library was prepared using one SMRT cell, followed by an assembly with the hierarchical genomeassembly process version 3 (HGAP 3) [29]. The genome was completed by manual curation according to PacBio instructions (https://github.com/PacificBiosciences/BioinformaticsTraining/wiki/Finishing-Bacterial-Genomes) and annotated using the NCBI Prokaryotic Genome Annotation Pipeline and the Rapid Annotations using Subsystems Technology [3032]. For detailed analysis of genes of interest, BLASTP and BLASTN were used [33, 34], for generating genomic images, BLAST Ring Image Generator (BRIG) was applied [35]. 2.7 Glycosyltransferase-2 screening To screen different beer-spoiling LAB for the existence of a glycosyltransferase 2 (gtf-2) gene, primers were designed to amplify a 900 bp region from the respective gene of L. brevis TMW 1.2112.
(GTF-F:
GAATCCGAACTAGCAATACTCGC,
GTF-R:
ACTAGTGGAATGTGCAACAC TGG). Total genomic DNA was isolated from beer-spoiling LAB strains (Table 1) using E.Z.N.A bacterial DNA Kit (Omega Bio-Tek, Inc., Norcross, 7
USA). PCR amplification was performed with a reaction mixture of 1 µl DNA, 42.25 µl deionized water, 0.25 µl of each primer, 0.25 µl Taq polymerase, 5 µl 10 x MgCl2 buffer (MP Biomedicals, Santa Ana, USA). The PCR program consisted of a step for initial denaturation (94 °C for 2 min), 32 cycles of 94 °C for 45 sec, 60.6 °C for 60 sec, and 72 °C for 54 sec and a final extension at 72 °C for 2 min. The PCR amplicons and a 100-bp plus ladder (Thermo Scientific, Waltham, USA) were separated through gel electrophoresis in 1.3 % (w/v) agarose gels in 0.5 TAE buffer, stained with dimidium bromide and visualized with ultraviolet light. 3. Results 3.1 General growth characteristics of L. brevis TMW 1.2112 L. brevis TMW 1.2112 exhibits a slimy phenotype on solid agar plates and in liquid culture. Growth on agar plates results in ropy colonies forming long filaments when extending the colony surface with an inoculation loop. In contrast, growth in liquid cultures causes highly viscous liquids with macroscopically visible slime formation (Figure 1). The inoculation of cultivation tubes with L. brevis TMW 1.2112 results in diffuse growth and successive slime formation. At the growth maximum, bacterial cells as well as slime occur all over the tube. Subsequently, the effect reverses, starting with the sedimentation of cells and the gradual disappearance of slime. The resulting cell pellet retains the slimy characteristics for several days. In contrast, the supernatant has lost all viscous properties. By shaking the cultivation tube, the cell pellet spirals upwards like a thick mucoid string. Therefore, a sticky network between the cells is indicated, suggesting a CPS more than a released EPS. 3.2 Structural characterization For the structural characterization of the exopolysaccharides formed by L. brevis TMW 1.2112, multiple chromatographic and spectroscopic approaches were applied. Methylation analysis was used to screen the glycosidic linkages of the polysaccharides (Table 2). CPS yielded only 8
glucose-derived PMAAs, whereas the chromatograms of the EPS preparation additionally showed galactose- and mannose-derived PMAAs. Mannose and galactose were also detected by high performance anion exchange chromatography (HPAEC) after acid hydrolysis. Therefore, a control composed of precipitated growth medium was analyzed. Galactose and mannose as well as the detected galactose- and mannose-derived PMAAs were found in the control. In addition, both the control and the samples contained minor amounts of 1,4substituted glucose, demonstrating that galactose, mannose, and (1→4)-linked glucose are medium-derived and most likely not exopolysaccharide constituents. Thus, it was possible to conclude that CPS and EPS contain terminal-, 1,3-, and 1,2,3-substituted D-glucose units (the absolute configuration of glucose was determined by GC-MS after chiral derivatization). However, the high portions of terminal glucose residues suggest an underestimation of the 1,3and/or the 1,2,3-substituted glucose units. Nevertheless, methylation analysis strongly suggests that both EPS and CPS of L. brevis TMW 1.2112 are composed of (1→3)-linked glucans with branches at position O2. The ratios between the 1,3- and 1,2,3-substituted glucose units indicate a ramification at every second backbone unit. To confirm these results and to assess the anomeric configuration of the monomeric units, oneand two-dimensional NMR spectroscopy was applied. Analysis of Correlated spectroscopy (COSY), Total correlated spectroscopy (TOCSY), and Heteronuclear single quantum coherence (HSQC) spectra of the polysaccharides allowed for the assignment of all 1H and 13C chemical shifts of the three structural units present (terminal, 1,3-, and 1,2,3-substituted glucose) (supplementary material Tab. S1). The 1,3- and 1,2,3-substituted glucose units showed a characteristic downfield shift for their C3/H3 or, respectively, their C2/H2 and C3/H3 correlation peaks in the HSQC spectrum (Fig. 2), indicating a substitution at this position. The 13
C chemical shifts of the unsubstituted ring protons/carbons showed chemical shifts
comparable to the β-anomer of monomeric glucose, suggesting the presence of β-glucose 9
residues. In addition, all
13
C chemical shifts of the terminal glucose residues were in good
agreement with the values reported for laminaribiose (β-(1→3)-linked glucobiose) [36]. On the other hand, slightly different chemical shifts were described for a β-(1→3)-linked glucan with branches at position O2 from Pediococcus (P.) damnosus [37]. However, similar trends for the downfield/upfield shifts of the protons/carbons were observed but chemical shifts constantly differ by 0.12 ppm (1H) or 1.5 ppm (13C), respectively. Thus, the chemical shift discrepancies are likely due to varying solvents or different referencing. Taking all results into account, it can be concluded that the L. brevis TMW 1.2112 exopolysaccharides are composed of a backbone of β-(1→3)-linked glucose units, which are ramified with β-glucose residues at position O2. As stated before, the 1,3- and 1,2,3-substituted glucose units were most likely underestimated by methylation analysis. Thus, volume integration was performed for the C2/H2 correlation peaks to get a semiquantitative estimate of the ratios between the structural elements. These signals should have roughly comparable 1JCH coupling constants and consequently an approximately comparable response. The ratio obtained for 1,3- and 1,2,3-substituted glucose units (1/0.8) indicated highly branched polysaccharides with ramifications at about every second backbone residue. In addition, terminal β-glucose residues were detected in amounts comparable to the branched backbone residues, suggesting that the terminal glucose units are mostly derived from ramifications. 3.3 Agglutination test To
confirm
the
possible
capsular
localization
of
the
identified
β-glucan,
an
immunoagglutination test with S. pneumoniae 37-specific antiserum was carried out. The strain agglutinated in presence of the antiserum, thus demonstrating the presence of a β-glucan at the cell surface as CPS (Fig. 3). 3.4 Genome sequencing 10
To gain more insights into the molecular background of β-glucan formation, genomic DNA was isolated from L. brevis TMW 1.2112 and sequenced via single-molecule real-time sequencing technology (PacBio RS II). De novo assembly was carried out by using the HGAP method. Upon sequencing six contigs were generated, which were further assembled to one finished genome. The genome is composed of a chromosome and five plasmids. General genome parameters and accession numbers are listed in Table 3. On pl12112-4, we identified a glycosyltransferase-2 (EC 2.4.1.34) (Fig. 4), which was described as key enzyme for the synthesis of β-(1,3-1,2)-glucans [38] and belongs to the glycosyltransferase family 2 of carbohydrate active enzymes. The gtf-2 gene contains 1704 nucleotides with a predicted protein product of 567 amino acids. Topological analysis indicates that the gene encodes a transmembrane protein with five helices (CBS prediction server; http://www.cbs.dtu.dk/services/TMHMM/). The other eight genes surrounding the gtf-2 are not involved in sugar metabolism, but in the maintenance and transmission of the plasmid. 3.5 Gtf-2 screening in beer-spoiling lactic acid bacteria To verify the relevance of the gtf-2 gene within the slimy spoilage of beer, beer-spoiling LAB (slime forming and non-slime forming ones) were screened for the presence of this gene. Therefore, primers targeting a 900 bp region from the gene from L. brevis TMW 1.2112 were designed and used for PCR. All slime-forming strains generated the expected amplification signal (Fig. 5), whereas amplification with DNA from non-slime forming strains remained negative. Furthermore, random sugar monomer analysis of EPS from different beer-spoiling LAB via HPLC after acidic hydrolysis of the isolated EPS revealed these EPS to be solely composed of glucose (data not shown).
11
These results support the importance of the glycosyltransferase-2 in the formation of glucans and suggest it as diagnostic marker gene for early detection of EPS forming beer-spoiling LAB, which may not be identified along previously described hop tolerance markers. 4. Discussion This study aimed for a better understanding of the EPS-based spoilage of beer by characterization of L. brevis TMW 1.2112, isolated from spoiled, viscous beer. The slime-forming ability of L. brevis TMW 1.2112 was assessed on solid and in liquid media, showing a strong slime formation in both cases. Its phenotype is consistent with the characteristics of slime-forming beer-spoiling lactobacilli from DH Williamson [39]: growth on agar plates resulting in ropy colonies; growth in liquid culture causing increased viscosity. However, the viscosity of liquid cultures disappears upon prolonged fermentation, so we suggest at least a partial degradation of the produced exopolysaccharide, which might serve as carbon source during starvation. Such an enzymatic EPS hydrolysis/degradation has been reported for several EPS-producing LAB strains [40-44]. Since L. brevis TMW 1.2112 encodes different (putative) β-glucan degrading glycosyl hydrolases, e. g. an endoglucanase (glycosyl hydrolase family 8; locus tag AZI09_02135) or a glycosyl hydrolase family 3 protein (locus tag AZI09_02175), this β-glucan could serve as carbohydrate source under certain (possibly limited) growth conditions for L. brevis TMW 1.2112. Both, the appearance on agar and the behavior in liquid culture imply the formation of a CPS, connecting the cells in some kind of sticky network. CPS formation was proved and confirmed in this study via immunological analysis. The structural characterization demonstrated that the capsule is composed of a β-(1→3)-glucan with ramifications at position O2. The PMAA ratios as well as the HSQC peak intensities suggested that about every second 1,3-linked backbone unit is ramified at position O2, which was also described for the β-glucans from P. damnosus 12
[37], Pediococcus sp. [45], and Oenococcus oeni [46]. The polysaccharides isolated from the supernatant are identical to those isolated from the cell pellet. Thus, it can be concluded that the CPS is partially secreted into the surrounding medium. Comparative genome and structural analysis as well as comprehension of known literature revealed a plasmid-located glycosyltransferase-2 as key gene for the formation of this glucan capsule. This type of glycosyltransferase was described to synthesize glucans by catalyzing the polymerization of glycosyl residues from UDP-glucose [38, 47]. Therefore, β-(1,3-1,2)-glucan formation by certain LAB resembles more the mechanisms of heteropolysaccharide biosynthesis than those of homopolysaccharides from energy-rich disaccharides such as sucrose. However, the precise mechanism is still not fully understood [48, 49]. Next to the gtf-2 gene, the plasmid harbors eight other genes, all involved in maintenance and transmission of the plasmid. The maintenance supposedly is ensured by a putative toxinantitoxin system. This system is composed of two closely linked genes encoding a toxin and its cognate antitoxin, protecting the host against the toxic effect. Loss of the plasmid causes cell death since the unstable antitoxin is degraded earlier and the lasting toxin kills the cell [50]. Accordingly, EPS formation poses a stable phenotype of L. brevis TMW 1.2112. The gtf-2 gene is also present on a plasmid of the slimy beer spoiler P. claussenii BAA-344 T. Moreover, all genes encoded on the plasmid of L. brevis are found on the respective one of P. claussenii as well (supplementary material Fig. 1S). Due to these similarities and the fact that all other genes encoded by these plasmids are not involved in sugar metabolism, we assume slimy beer spoilage to result mainly from the action of this plasmid encoded glycosyltransferase-2. This hypothesis is in accordance with the results of D Llull, R Munoz, R Lopez and E Garcia [51] which showed that the highly homologous glycosyltransferase-2 (tts) from S. pneumoniae is the only gene required to drive the biosynthesis and secretion of a
13
capsular β-glucan, while other bacterial beta-glucans such as cellulose are synthesized stepwise from UDP-glucose by the action of multienzyme complexes [52]. Glucan-forming LAB not only affect beer, but also wine and cider. As in beer, viscous wine and cider are attributed to a gtf-mediated β-(1→3)-glucan formation [27, 45-47, 53, 54]. Moreover, the glycosyltransferase is species/genera-independently highly conserved among βglucan-producing LAB. Comparing the gtf gene of the slimy wine/ cider spoilers P. parvulus IOEB8801 (accession no. AF196967), P. damnosus 2.6 (accession no. AY999683) and L. diolivorans G77 (accession no. AY999684) with the slimy beer spoiler P. claussenii ATCC BAA-344 T (locus taq PECL_RS09485) and L. brevis TMW 1.2112 (locus taq AZI09_12770) shows sequence identities of 99 % (Table 4). These similarities in the genetic background and in the structure of the resulting polysaccharide imply a common origin and horizontal gene transfer to be responsible for the dispread of the gtf gene within beverage-spoiling lactobacilli and pediococci. This assumption is strengthened by the finding that the gtf gene is located on a mobilizable plasmid in P. parvulus IOEB8801 [55] and on a conjugative one in P. parvulus 2.6 [53]. In addition, the respective gene of P. parvulus 2.6, L. brevis TMW 1.2112 and P. claussenii ATCC BAA-344 T is neighbored to a transposase (IS 30 family). This enzyme is able to cause a translocation of genes within the genome and an integration into the chromosome. This might explain the finding of chromosomal gtfs in O. oeni (IOEB0205 and I4) and L. suebicus CUPV221 which are in turn highly identical to the plasmid-located ones (Table 4). Indeed, the chromosomal gtf of O. oeni IOEB0205 is found in close proximity to a transposase. The formation of a slimy capsule within these beverage-spoiling bacteria might occur due to three different reasons. One reason for the encapsulation could be an increased resistance against environmental stress factors. Beer and wine represent both a harsh environment for bacteria due to the presence of antimicrobial hop compounds and sulfur dioxide, respectively, 14
of a low pH and a quite high ethanol content. Forming a capsule could serve as protective barrier, while it was demonstrated for P. parvulus that its resistance against these stress factors was increased [16] [17]. Therefore, it is likely that the encapsulation of TMW 1.2112 helps the strain dealing with the hurdles of harsh environments such as beer. The above-discussed role as carbon source during starvation could be another advantage, brought by capsule formation. Finally, capsules might facilitate the adherence to solid surfaces [47, 56], suggesting that L. brevis TMW 1.2112 could be involved in (brewery-associated) biofilm formation. Biofilms in turn are hotspots for gene transfers from one species to another [57]. Consequently, the promotion of the development of gtf-positive lactobacilli or pediococci seems possible. This clarifies the need and the importance for an early detection of such bacteria in breweries. By designing primers targeting the gtf-2 gene, it was possible to detect slimy beer-spoiling LAB species-independently and to distinguish them from non-slimy ones. Thus, this gene represents an appropriate target for the reliable identification of these brewery contaminants with PCR. ACKNOWLEDGEMENT Part of this work was by the German Ministry of Economics and Technology (via AiF) and the Wifoe (Wissenschaftsförderung der Deutschen Brauwirtschaft e.V., Berlin) project AiF 17576 and18194 N. None of the funding sources did have any influence on the study design, the collection, analysis and interpretation of data, the writing of the report and the decision to submit the article for publication. The authors are grateful to the Forschungszentrum Weihenstephan für Brau- und Lebensmittelqualität for providing bacterial strains.
15
REFERENCES
1.
2.
3.
4. 5. 6. 7. 8. 9.
10.
11.
12.
13.
14.
15.
Suzuki K: 125th Anniversary Review: Microbiological instability of beer caused by spoilage bacteria. Journal of the Institute of Brewing 2011, 117(2):131-155. Vogel RF, Behr J, Geissler AJ: The lifestyle of beer-spoiling lactic acid bacteria is based on digging in the plasmid pool. In: World Brewing Congress: 2016; Denver, USA. American Society of Brewing Chemists. Behr J, Geissler AJ, Schmid J, Zehe A, Vogel RF: The Identification of Novel Diagnostic Marker Genes for the Detection of Beer Spoiling Pediococcus damnosus Strains Using the BlAst Diagnostic Gene findEr. PloS one 2016, 11(3):e0152747. Simpson WJ: Studies on the sensitivity of lactic acid bacteria to hop bitter acids. J Inst Brew 1993, 99:405-412. Simpson WJ: Ionophoric action of trans-isohumulone on Lactobacillus brevis. J Gen Microbiol 1993, 139:1041-1046. Simpson WJ, Fernandez JL: Mechanism of resistance of lactic acid bacteria to trans-isohumulone. J Am Soc brew Chem 1994, 52:9-11. Behr J, Vogel RF: Mechanisms of hop inhibition: hop ionophores. Journal of agricultural and food chemistry 2009, 57(14):6074-6081. Sakamoto K, Konings WN: Beer spoilage bacteria and hop resistance. International journal of food microbiology 2003, 89(2-3):105-124. Sakamoto K, Margolles A, van Veen HW, Konings WN: Hop resistance in the beer spoilage bacterium Lactobacillus brevis is mediated by the ATPbinding cassette multidrug transporter HorA. Journal of bacteriology 2001, 183(18):5371-5375. Suzuki K, Sami M, Kadokura H, Nakajima H, Kitamoto K: Biochemical characterization of horA-independent hop resistance mechanism in Lactobacillus brevis. International journal of food microbiology 2002, 76(3):223-230. Nwodo UU, Green E, Okoh AI: Bacterial exopolysaccharides: functionality and prospects. International journal of molecular sciences 2012, 13(11):14002-14015. Brandt JU, Jakob F, Behr J, Geissler AJ, Vogel RF: Dissection of exopolysaccharide biosynthesis in Kozakia baliensis. Microb Cell Fact 2016, 15(1):170. Ua-Arak T, Jakob F, Vogel RF: Characterization of growth and exopolysaccharide production of selected acetic acid bacteria in buckwheat sourdoughs. International journal of food microbiology 2016, 239:103-112. Jakob F, Pfaff A, Novoa-Carballal R, Rubsam H, Becker T, Vogel RF: Structural analysis of fructans produced by acetic acid bacteria reveals a relation to hydrocolloid function. Carbohydr Polym 2013, 92(2):1234-1242. Ruhmkorf C, Jungkunz S, Wagner M, Vogel RF: Optimization of homoexopolysaccharide formation by lactobacilli in gluten-free sourdoughs. Food microbiology 2012, 32(2):286-294.
16
16.
17.
18.
19. 20.
21.
22.
23.
24.
25. 26.
27.
28.
29.
30.
31.
Stack HM, Kearney N, Stanton C, Fitzgerald GF, Ross RP: Association of beta-glucan endogenous production with increased stress tolerance of intestinal lactobacilli. Appl Environ Microbiol 2010, 76(2):500-507. Coulon J, Houles A, Dimopoulou M, Maupeu J, Dols-Lafargue M: Lysozyme resistance of the ropy strain Pediococcus parvulus IOEB 8801 is correlated with beta-glucan accumulation around the cell. Int J Food Microbiol 2012, 159(1):25-29. Kern CC, Usbeck JC, Vogel RF, Behr J: Optimization of Matrix-AssistedLaser-Desorption-Ionization-Time-Of-Flight Mass Spectrometry for the identification of bacterial contaminants in beverages. Journal of microbiological methods 2013, 93(3):185-191. Geissler AJ, Behr J, von Kamp K, Vogel RF: Metabolic strategies of beer spoilage lactic acid bacteria in beer. Int J Food Microbiol 2016, 216:60-68. Tallon R, Bressollier P, Urdaci MC: Isolation and characterization of two exopolysaccharides produced by Lactobacillus plantarum EP56. Research in microbiology 2003, 154(10):705-712. Wefers D, Bunzel M: Characterization of dietary fiber polysaccharides from dehulled, common buckwheat (Fagopyrum esculentum) seeds. Cereal Chem 2015, 92(6):598-603. Sweet DP, Shapiro RH, Albersheim P: Quantitative-analysis by various GLC response-factor theories for partially methylated and partially ethylated alditol acetates. Carbohydrate Research 1975, 40(2):217-225. Saeman JF, Bubl JL, Harris EE: Quantitative saccharification of wood and cellulose. Industrial and Engineering Chemistry-Analytical Edition 1945, 17(1):35-37. De Ruiter GA, Schols HA, Voragen AGJ, Rombouts FM: Carbohydrate analysis of water-soluble uronic acid-containing polysaccharides with high-performance anion-exchange chromatography using methanolysis combined with TFA hydrolysis is superior to four other methods. Anal Biochem 1992, 207(1):176-185. Wefers D, Tyl CE, Bunzel M: Novel arabinan and galactan oligosaccharides from dicotyledonous plants. Frontiers in Chemistry 2014, 2:100. Gottlieb HE, Kotlyar V, Nudelman A: NMR chemical shifts of common laboratory solvents as trace impurities. J Org Chem 1997, 62(21):75127515. Walling E, Gindreau E, Lonvaud-Funel A: A putative glucan synthase gene dps detected in exopolysaccharide-producing Pediococcus damnosus and Oenococcus oeni strains isolated from wine and cider. Int J Food Microbiol 2005, 98(1):53-62. Eid J, Fehr A, Gray J, Luong K, Lyle J, Otto G, Peluso P, Rank D, Baybayan P, Bettman B et al: Real-time DNA sequencing from single polymerase molecules. Science 2009, 323(5910):133-138. Chin CS, Alexander DH, Marks P, Klammer AA, Drake J, Heiner C, Clum A, Copeland A, Huddleston J, Eichler EE et al: Nonhybrid, finished microbial genome assemblies from long-read SMRT sequencing data. Nature methods 2013, 10(6):563-569. Aziz RK, Bartels D, Best AA, DeJongh M, Disz T, Edwards RA, Formsma K, Gerdes S, Glass EM, Kubal M et al: The RAST Server: rapid annotations using subsystems technology. BMC genomics 2008, 9:75. Overbeek R, Olson R, Pusch GD, Olsen GJ, Davis JJ, Disz T, Edwards RA, Gerdes S, Parrello B, Shukla M et al: The SEED and the Rapid Annotation of 17
32.
33. 34.
35.
36.
37.
38. 39. 40.
41.
42.
43.
44.
45.
46.
microbial genomes using Subsystems Technology (RAST). Nucleic acids research 2014, 42(Database issue):D206-214. Angiuoli SV, Gussman A, Klimke W, Cochrane G, Field D, Garrity G, Kodira CD, Kyrpides N, Madupu R, Markowitz V et al: Toward an online repository of Standard Operating Procedures (SOPs) for (meta)genomic annotation. Omics : a journal of integrative biology 2008, 12(2):137-141. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ: Basic local alignment search tool. Journal of molecular biology 1990, 215(3):403-410. Camacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, Bealer K, Madden TL: BLAST+: architecture and applications. BMC bioinformatics 2009, 10:421. Alikhan NF, Petty NK, Ben Zakour NL, Beatson SA: BLAST Ring Image Generator (BRIG): simple prokaryote genome comparisons. BMC genomics 2011, 12:402. Roslund MU, Tahtinen P, Niemitz M, Sjhohn R: Complete assignments of the 1H and 13C chemical shifts and J(H,H) coupling constants in NMR spectra of D-glucopyranose and all D-glucopyranosyl-D-glucopyranosides. Carbohydr Res 2008, 343(1):101-112. Duenas-Chasco MT, Rodriguez-Carvajal MA, Tejero Mateo P, FrancoRodriguez G, Espartero JL, Irastorza-Iribas A, Gil-Serrano AM: Structural analysis of the exopolysaccharide produced by Pediococcus damnosus 2.6. Carbohydr Res 1997, 303(4):453-458. Karnezis T, McIntosh M, Wardak A, Stanisich VA, Stone BA: The Biosynthesis of ß-Glycans. Trends in Glycoscience and Glycotechnology 2000, 12(66):17. Williamson DH: Studies on lactobacilli causing ropiness in beer. Journal of Applied bacteriology 1959, 22:392-402. Cerning J, Bouillanne C, Landon M, Desmazeaud M: Isolation and Characterization of Exopolysaccharides from Slime-Forming Mesophilic Lactic Acid Bacteria. Journal of Dairy Science 1992, 75(3):692-698. Cerning J, Desmazeaud M, Landon M: Exocellular polysaccharide production by Streptococcus thermophilus. Biotechnology letter 1988, 10(4):255-260. Pham PL, Dupont I, Roy D, Lapointe G, Cerning J: Production of exopolysaccharide by Lactobacillus rhamnosus R and analysis of its enzymatic degradation during prolonged fermentation. Applied and environmental microbiology 2000, 66(6):2302-2310. Gancel F, Novel G: Exopolysaccharide Production by Streptococcus salivarius ssp. thermophilus Cultures. 2. Distinct Modes of Polymer Production and Degradation Among Clonal Variants. Journal of Dairy Science 1994, 77(3):7. Ricciardi A, Parente E, Crudele MA, Zanetti F, Scolari G, Mannazzu I: Exopolysaccharide production by Streptococcus thermophilus SY: production and preliminary characterization of the polymer. Journal of applied microbiology 2002, 92(2):297-306. Llauberes RM, Richard B, Lonvaud A, Dubourdieu D, Fournet B: Structure of an exocellular beta-D-glucan from Pediococcus sp., a wine lactic bacteria. Carbohydr Res 1990, 203(1):103-107. Ibarburu I, Soria-Diaz ME, Rodriguez-Carvajal MA, Velasco SE, Tejero-Mateo P, Gil-Serrano AM, Irastorza A, Duenas MT: Growth and exopolysaccharide (EPS) production by Oenococcus oeni I4 and structural characterization of their EPSs. Journal of applied microbiology 2007, 103(2):477-486. 18
47.
48. 49.
50. 51.
52.
53.
54.
55.
56.
57. 58.
59.
60.
Dols-Lafargue M, Lee HY, Le Marrec C, Heyraud A, Chambat G, Lonvaud-Funel A: Characterization of gtf, a glucosyltransferase gene in the genomes of Pediococcus parvulus and Oenococcus oeni, two bacterial species commonly found in wine. Appl Environ Microbiol 2008, 74(13):4079-4090. McIntosh M, Stone BA, Stanisich VA: Curdlan and other bacterial (1-->3)beta-D-glucans. Appl Microbiol Biotechnol 2005, 68(2):163-173. Torino MI, Font de Valdez G, Mozzi F: Biopolymers from lactic acid bacteria. Novel applications in foods and beverages. Frontiers in microbiology 2015, 6:834. Yamaguchi Y, Park JH, Inouye M: Toxin-antitoxin systems in bacteria and archaea. Annual review of genetics 2011, 45:61-79. Llull D, Munoz R, Lopez R, Garcia E: A single gene (tts) located outside the cap locus directs the formation of Streptococcus pneumoniae type 37 capsular polysaccharide. Type 37 pneumococci are natural, genetically binary strains. The Journal of experimental medicine 1999, 190(2):241-251. Chawla PR, Bajaj IB, Survase SA, Singhal RS: Microbial Cellulose: Fermentative Production and Applications. Food Technol Biotechnol 2009, 47(2):107-125. Werning ML, Ibarburu I, Duenas MT, Irastorza A, Navas J, Lopez P: Pediococcus parvulus gtf gene encoding the GTF glycosyltransferase and its application for specific PCR detection of beta-D-glucan-producing bacteria in foods and beverages. Journal of food protection 2006, 69(1):161169. Duenas-Chasco MT, Rodriguez-Carvajal MA, Tejero-Mateo P, Espartero JL, Irastorza-Iribas A, Gil-Serrano AM: Structural analysis of the exopolysaccharides produced by Lactobacillus spp. G-77. Carbohydr Res 1998, 307(1-2):125-133. Gindreau E, Walling E, Lonvaud-Funel A: Direct polymerase chain reaction detection of ropy Pediococcus damnosus strains in wine. Journal of applied microbiology 2001, 90(4):535-542. de Palencia PF, Werning ML, Sierra-Filardi E, Duenas MT, Irastorza A, Corbi AL, Lopez P: Probiotic properties of the 2-substituted (1,3)-beta-D-glucanproducing bacterium Pediococcus parvulus 2.6. Applied and environmental microbiology 2009, 75(14):4887-4891. Flemming HC, Wingender J: The biofilm matrix. Nature reviews Microbiology 2010, 8(9):623-633. Pittet V, Abegunde T, Marfleet T, Haakensen M, Morrow K, Jayaprakash T, Schroeder K, Trost B, Byrns S, Bergsveinson J et al: Complete genome sequence of the beer spoilage organism Pediococcus claussenii ATCC BAA-344T. Journal of bacteriology 2012, 194(5):1271-1272. Juvonen R, Honkapaa K, Maina NH, Shi Q, Viljanen K, Maaheimo H, Virkki L, Tenkanen M, Lantto R: The impact of fermentation with exopolysaccharide producing lactic acid bacteria on rheological, chemical and sensory properties of pureed carrots (Daucus carota L.). International journal of food microbiology 2015, 207:109-118. Garai-Ibabe G, Duenas MT, Irastorza A, Sierra-Filardi E, Werning ML, Lopez P, Corbi AL, Fernandez de Palencia P: Naturally occurring 2-substituted (1,3)beta-D-glucan producing Lactobacillus suebicus and Pediococcus parvulus strains with potential utility in the production of functional foods. Bioresource technology 2010, 101(23):9254-9263. 19
FIGURE CAPTIONS
Figure 1: Growth and slime forming dynamics of L. brevis TMW 1.2112 over a period of six days. Figure 2: HSQC spectrum and proposed structure of the L. brevis TMW 1.2112 CPS preparation. The characteristic downfield shifts of the correlation peaks at the substituted positions are encircled. Figure 3: Agglutination induced by S. pneumoniae type 37-specific antiserum. L. brevis TMW 1.2112 without antiserum (A) and after antiserum addition (B). Figure 4: RAST-annotated plasmid pl12112-4 of L. brevis TMW 1.2112. The predicted ORFs are mapped on plasmid 4 using BRIG. Starting from inside: circle 1 shows the general position in kilobases; circle 2 depicts the G+C content; circle 3 presents genes encoded, in red the gtf-2. With exception of gtf-2, the genes are not related to EPS formation, only for plasmid conservation. Figure 5: Gel electrophoresis of PCR amplicons generated with primers Gtf-F/Gtf-R on the following strains: L. brevis (line 2-20), L. parabuchneri (line 21), L. rossiae (line 22-23) – according to table 1. Negative control (line 24), molecular mass standard (line 1). The length of the amplicon is indicated. Figure S1: Similarities between plasmid pl12112-4 of L. brevis TMW 1.2112 and plasmid pPECL7 of P. claussenii ATCC BAA-344 T. Starting from inside: circle 1 shows the general position in kilobases; circle 2 depicts the G+C content of L. brevis TMW 1.2112; circle 3 presents genes encoded on pl12112-4 of L. brevis TMW 1.2112, in red the gtf-2. Circle 4 exhibits the corresponding genes of plasmid pPECL7 of P. claussenii ATCC BAA-344 T.
20
Table 1: Strains and sources of isolation. Species
Strain
Source
L. brevis
TMW 1.2107
Slimy spoiled beer
L. brevis
TMW 1.2108
Slimy spoiled beer
L. brevis
TMW 1.2109
Slimy spoiled beer
L. brevis
TMW 1.2110
Slimy spoiled beer
L. brevis
TMW 1.2111
Slimy spoiled beer
L. brevis
TMW 1.2112
Slimy spoiled beer
L. brevis
TMW 1.2113
Brewery surface
L. brevis
TMW 1.2114
Slimy spoiled beer
L. brevis
TMW 1.2115
Slimy spoiled beer
L. brevis
TMW 1.599
Slimy spoiled beer
L. brevis
TMW 1.240
Slimy spoiled beer
L. brevis
TMW 1.2146
Slimy spoiled beer
L. brevis
TMW 1.2147
Slimy spoiled beer
L. brevis
TMW 1.2148
Slimy spoiled beer
L. brevis
TMW 1.2149
Slimy spoiled beer
L. brevis
TMW 1.2151
Slimy spoiled beer
L. brevis
TMW 1.2153
Slimy spoiled beer
L. brevis
TMW 1.2154
Slimy spoiled beer
L. brevis
TMW 1.2155
Slimy spoiled beer
L. parabuchneri
TMW 1.1141
Beer
L. rossiae
TMW 1.2152
Slimy spoiled beer
L. rossiae
TMW 1.2156
Slimy spoiled beer
TMW = Technische Mikrobiologie Weihenstephan.
21
Table 2: Percentages of the partially methylated alditol acetates from the L. brevis TMW 1.2112 CPS and EPS preparation. Glycosidic linkage
CPS
EPS
t-Glcp
50.7 %
54.6 %
1,3-Glcp
23.8 %
23.5 %
1,2,3-Glcp
25.5 %
21.9 %
t = terminal. Numbers indicate the substituted positions of a sugar unit.
Table 3: Sequencing statistics, genome information and accession numbers of L. brevis TMW 1.2112. The biosample SMNO4517633 is part or the bioproject PRJNA313253. Coverage of HGAP assembly is 396. CDS = Coding sequences based on NCBI PGAP. Contig Accession no.
Internal name
Size (bp)
C+G content (%)
CDS
1
CP016797
chr12112-1
2488939
45.98
2338
2
CP016798
pl12112-1
44555
41.59
60
3
CP016799
pl12112-2
38644
42.58
36
4
CP016800
pl12112-3
33211
45.64
35
5
CP016801
pl12112-4
8521
36.78
9
6
CP016802
pl12112-5
59682
40.35
59
22
Table 4: Overview of EPS-forming beverage-spoiling lactic acid bacteria. Strain
Origin
L. brevis
beer
Gtf
AA
Gtf
Poly
Locus tag*/
location
gtf
identity
saccharide
accession no.
plasmid
567
-
β-(1→3)-
AZI09_12770
TMW 1.2112 P.
glucan
claussenii beer
plasmid
567
99 %
ATCC-BAA-
β-(1→3)-
PECL_RS09485
glucan
344T1,2 P.
damnosus wine
plasmid
567
99 %
IOEB88013 P. parvulus
cider
plasmid
567
99 %
cider
plasmid
567
99 %
cider
chr
567
99 %
β-(1→3)-
AY999684
β-(1→3)-
GU174474
glucan wine
chr
567
97 %
IOEB02057 O. oeni
AY999683
glucan
CUPV2216 O. oeni
β-(1→3)glucan
G775 L. suebicus
AF196967
glucan
2.64 L. diolivorans
β-(1→3)-
β-(1→3)-
EU556433
glucan cider
chr
567
I48
98 %
β-(1→3)-
AY999685
glucan
AA = amino acids; Gtf identity = sequence identity to gtf-2 of L. brevis TMW 1.2112; chr = chromosome; * locus tag only used when available 1
= V Pittet, T Abegunde, T Marfleet, M Haakensen, K Morrow, T Jayaprakash, K Schroeder,
B Trost, S Byrns, J Bergsveinson, et al. [58], 2 = R Juvonen, K Honkapaa, NH Maina, Q Shi, K Viljanen, H Maaheimo, L Virkki, M Tenkanen and R Lantto [59] 3 = E Walling, E Gindreau and A Lonvaud-Funel [27], 4 = ML Werning, I Ibarburu, MT Duenas, A Irastorza, J Navas and P Lopez [53], 5 = ML Werning, I Ibarburu, MT Duenas, A Irastorza, J Navas and P Lopez [53], 6
= G Garai-Ibabe, MT Duenas, A Irastorza, E Sierra-Filardi, ML Werning, P Lopez, AL Corbi
and P Fernandez de Palencia [60], 7 = M Dols-Lafargue, HY Lee, C Le Marrec, A Heyraud, G Chambat and A Lonvaud-Funel [47], 8 = ML Werning, I Ibarburu, MT Duenas, A Irastorza, J Navas and P Lopez [53]
23
Supplementary material Table S1: 1H and 13C chemical shifts of the three structural elements present in the CPS and EPS of L. brevis TMW 1.2112. 1,3-Glcp = 1,3-substituted β-glucopyranose, 1,2,3-Glcp = 1,2,3substituted β-glucopyranose, t-Glcp = terminal β-glucopyranose bound to position O2 of a 1,3substituted β-glucopyranose. Chemical shifts are given in ppm. Structural element 1,3-Glcp
1
2
3
4
5
6
4.90
3.63
3.79
3.55
3.51
3.74/3.91
13
102.24
73.35
86.30
68.71
76.26
61.30
1
4.88
3.84
3.97
3.55
3.51
3.74/3.91
13
102.26
80.60
84.11
68.71
76.26
61.30
1
4.95
3.31
3.51
3.40
3.51
3.74/3.95
102.95
74.77
77.24
70.42
76.26
61.52
1
H C
1,2,3-Glcp
H C
t-Glcp
H
13
C
24
Figure 5
Figure 1
25
Figure 2
Figure 3
26
Figure 4
27
S0141-8130(17)32535-7 http://dx.doi.org/10.1016/j.ijbiomac.2017.09.063 BIOMAC 8240
To appear in:
International Journal of Biological Macromolecules
Received date: Revised date: Accepted date:
12-7-2017 1-9-2017 17-9-2017
Please cite this article as: Marion E.Fraunhofer, Andreas J.Geissler, Daniel Wefers, Mirko Bunzel, Frank Jakob, Rudi F.Vogel, Characterization of -glucan formation by Lactobacillus brevis TMW 1.2112 isolated from slimy spoiled beer, International Journal of Biological Macromoleculeshttp://dx.doi.org/10.1016/j.ijbiomac.2017.09.063 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Characterization of β-glucan formation by Lactobacillus brevis TMW 1.2112 isolated from slimy spoiled beer
1
Marion E Fraunhofer, 1Andreas J Geissler, 2Daniel Wefers, 2Mirko Bunzel, 1Frank Jakob*,
1
Rudi F Vogel
1
Lehrstuhl für Technische Mikrobiologie, Technische Universität München, Gregor-Mendel-
Straße 4, 85354 Freising, Germany 2
Karlsruhe Institute of Technology (KIT), Institute of Applied Biosciences, Department of
Food Chemistry and Phytochemistry, 76131 Karlsruhe, Germany
* address for correspondence Dr. Frank Jakob, Lehrstuhl für Technische Mikrobiologie, Technische Universität München, Gregor-Mendel-Straße 4, 85354 Freising, Germany, Email: [email protected]
Highlights:
The beer spoiler L. brevis TMW 1.2112 produces capsular polysaccharides (CPS)
The CPS are composed of β-(1,3-1,2)-linked glucose moieties (β-glucan)
(Putative) glucan-degrading genes suggest CPS as carbon source during starvation
Gtf-2 is an essential enzyme for β-glucan synthesis in L. brevis TMW 1.2112
The gtf-2 gene is located on a plasmid in L. brevis TMW 1.2112
Gtf-2 was proven as suited marker gene for detection of slime-forming beer spoilers
Abstract
1
Despite several hurdles, which hinder bacterial growth in beer, certain bacteria are still able to spoil beer. One type of spoilage is characterized by an increased viscosity and slimy texture caused by exopolysaccharide (EPS) formation of lactic acid bacteria (LAB). In this study, we characterize for the first time EPS production in a beer-spoiling strain (TMW 1.2112) of Lactobacillus brevis, a species commonly involved in beer spoilage. The strain´s growth dynamics were assessed and we found an increased viscosity or ropiness in liquid or on solid media, respectively. Capsular polysaccharides (CPS) and released EPS from cells or supernatant, respectively, were analyzed via NMR spectroscopy and methylation analysis. Both are identical β-(1→3)-glucans, which are ramified with β-glucose residues at position O2. Therefore, we assume that this EPS is mainly produced as CPS and partially released into the surrounding medium, causing viscosity of e.g. beer. CPS formation was confirmed via agglutination test. A plasmid-located glycosyltransferase-2 was found as responsible for excess β-glucan formation, chromosomal glucanases were proposed for its degradation. The glycosyltransferase-2 gene could also be specifically identified in beer-spoiling, slimeproducing Lactobacillus rossiae and Lactobacillus parabuchneri strains, suggesting it as promising marker gene for early detection of β-glucan-producing Lactobacilli in breweries. Highlights:
The beer spoiler L. brevis TMW 1.2112 produces capsular polysaccharides (CPS)
The CPS are composed of β-(1,3-1,2)-linked glucose moieties (β-glucan)
(Putative) glucan-degrading genes suggest CPS as carbon source during starvation
Gtf-2 is an essential enzyme for β-glucan synthesis in L. brevis TMW 1.2112
The gtf-2 gene is located on a plasmid in L. brevis TMW 1.2112
Gtf-2 was proven as suited marker gene for detection of slime-forming beer spoilers
Key words: beer spoilage, Lactobacillus brevis, exopolysaccharide, β-glucan
2
1. Introduction Beer is considered as a stable beverage since it exhibits a harsh environment for bacteria. The low pH and nutrient availability, the oxygen-deficient but carbon dioxide rich environment as well as the presence of ethanol and antimicrobial hop compounds hinder most bacterial growth [1]. Nevertheless, there are bacteria which have developed strategies to overcome these hurdles, namely by picking up respective properties from a trans-species shared plasmid pool [2, 3]. The mechanisms behind bacterial beer spoilage are increasingly understood and were shown to be mainly associated with the ability to counteract hop compounds. Hops act as ionophores, which dissipate the transmembrane proton gradient and consequently abrogate the proton motive force [4-6]. This in turn inhibits nutrient uptake, essential enzyme reactions and finally causes the cell’s death [7-9]. The best-studied mechanisms antagonizing these detrimental effects are the modification of the cell wall to increase the barrier function and consequently to decrease the influx of hops, as well as the expression of multidrug transporters like horA and horC, which extrude hop compounds out of the cell [9, 10]. Utilizing these mechanisms, amongst others, bacteria are able to grow in and spoil beer. Besides strains causing turbidity and off flavors upon growth and metabolite formation, respectively, beer-spoiling bacteria also comprise strains, which produce slime increasing the viscosity of beer. This alteration is mostly attributed to certain lactic acid bacteria (LAB), forming exopolysaccharides (EPS). EPS are long chain, high molecular weight polymers, which are either composed from one type of monosaccharide (mostly glucose or fructose) or comprise a repeating unit of two or more different monosaccharides. They are therefore 3
classified as homopolysaccharides or heteropolysaccharides, respectively. Both can either be associated with the cell surface in form of a capsular polysaccharide (CPS) or be released into the surrounding environment as extracellular polysaccharide (EPS). EPS are thought to confer bacterial cells several advantages like the protection against phagocytosis, desiccation and toxins as well as an facilitated adherence to surfaces and a source of nutrients [11]. Due to their hydrocolloid properties, EPS and/or their producer strains are intentionally used in some food industries to improve e.g. the food texture [12-15]. At the same time, these characteristics make EPS-producing bacteria to an undesired contaminant in beer or other beverages. While the phenomenon of EPS formation by wine-spoiling bacteria is well studied [16, 17], respective beer-spoiling lactobacilli are still sparsely explored. Therefore, the objective of the present study was to get deeper insights into this special type of beer spoilage, by characterizing Lactobacillus (L.) brevis TMW 1.2112, isolated from spoiled, viscous beer regarding its capability to form EPS. Furthermore, the process of EPS formation and degradation was described and linked to the genetic background, which were identified via whole genome sequencing and comparative genomics/genetics. Finally, the structure of the produced polysaccharide was elucidated via NMR spectroscopy. 2. Materials and methods 2.1 Bacterial strains, growth conditions and media L. brevis TMW 1.2112 was isolated from spoiled, viscous beer and identified on species level using Matrix-Assisted-Laser-Desorption/Ionization-Time-Of-Flight Mass (MALDI-TOF MS) Spectrometry as reported by CC Kern, JC Usbeck, RF Vogel and J Behr [18]. For long term preservation the bacterial culture was cryoconserved according to AJ Geissler, J Behr, K von Kamp and RF Vogel [19]. As working culture bacteria were grown at 30°C in modified MRS medium (mMRS) adjusted to pH 4.3. Growth medium composition (quantities per liter): 10 g 4
peptone, 5 g yeast extract, 5 g meat extract, 4 g K2HPO4, 2.6 g KH2PO4·3H2O, 3 g NH4Cl, 1 g Tween80, 0.5 g cysteine-HCl, 20 g maltose, 0.2 g MgSO4·7H2O, 0.038 g MnSO4·H2O. For EPS isolation a variant of the medium without yeast and meat extract, but supplemented with a vitamin mix (quantities per liter: 0.2 mg thiamin, 0.2 mg niacin, 0.2 mg folic acid, 0.2 mg mg pyridoxal, 0.2 mg pantothenic acid, 0.2 mg cobalamine) was used. All other strains used in this study (Table 1) were handled like L. brevis TMW 1.2112. 2.2 Macroscopic characterization of EPS formation To visualize growth dynamics and slime formation a colony of L. brevis TMW 1.2112 was inoculated in 1.8 ml mMRS broth and incubated at 30 °C. Photos were taken over a time span of six days. 2.3 Isolation of capsular polysaccharides and released extracellular polysaccharides To differentiate between cell-bound CPS and released EPS, two methods were used according to R Tallon, P Bressollier and MC Urdaci [20]. The strain was cultured for three days at 30 °C in mMRS broth. Subsequently, the cells were separated from the supernatant by centrifugation (12000 x g, 60 min). The EPS in the supernatant were precipitated with cold ethanol (2:1 v/v) and kept at 4 °C overnight. After centrifugation, the resulting precipitate was dissolved in deionized water, dialyzed against deionized water for three days and lyophilized. To isolate the cell-bound CPS, the cell pellet was washed twice with PBS buffer and dissolved in 1 M NaCl. To detach the polysaccharides from the cells, the pellet was sonicated (3 x 30 sec, power 90 %) (HD-70/ Bandelin electronic, Berlin, Germany). Further processing of CPS was conducted as described for EPS. 2.4 Structural characterization
5
Glycosidic linkages were analyzed by methylation analysis as described previously [21]. Briefly, polysaccharides (3 mg) were dissolved in dimethyl sulfoxide and methylated by the addition of finely ground sodium hydroxide and methyl iodide. The methylated polysaccharides were extracted into dichloromethane, dried, and subsequently hydrolyzed with 2 M TFA for 1.5 h at 121 °C. The solvent was evaporated, and partially methylated monosaccharides were reduced by the addition of sodium borodeuteride. Following acetylation and extraction, the partially methylated alditol acetates (PMAAs) were analyzed by GC-MS and GC-FID (GC2010 Plus, GCMS-QP2010) (Shimadzu, Kyoto, Japan), both equipped with a DB-225 column (30 m x 0.25 mm i.d., 0.25 μm) (Agilent Technologies, Santa Clara, CA). Molar response factors according to DP Sweet, RH Shapiro and P Albersheim [22] were used for semiquantitative estimation of the PMAA ratios. Monosaccharide composition was determined after sulfuric acid hydrolysis (pretreatment in 12 M H2SO4 for 2.5 h, dilution to 1.6 M H2SO4 and hydrolysis for 3 h at 100 °C) according to JF Saeman, JL Bubl and EE Harris [23] and methanolysis (16 h methanolysis with 1.25 M methanolic HCl at 80 °C, 1 h hydrolysis with 2 M TFA at 121 °C) according to GA De Ruiter, HA Schols, AGJ Voragen and FM Rombouts [24]. The hydrolysates were analyzed for their monosaccharide composition by high performance anion exchange chromatography as described previously [21]. To determine the absolute configuration, the exopolysaccharides were hydrolyzed with 2 M TFA for 30 min at 121°C and their silylated (R)-2-octanol derivatives were analyzed by GC-MS as described previously [25]. For NMR spectroscopic characterization, samples (3 mg) were hydrogen-deuterium exchanged and subsequently dissolved in deuterium oxide (500 µL). NMR spectra were acquired at 298 K on a Bruker Ascend 500 MHz spectrometer equipped with a Prodigy cryoprobe (Bruker, Rheinstetten, Germany). Acetone was used as internal reference (1H: 2.22 ppm, 13C: 30.89 ppm) [26]. 2.5 Immunological analysis 6
Agglutination tests were performed with Streptococcus (S.) pneumoniae type 37-specific antisera (Statens Serum Institut, Copenhagen, Denmark) as previously described by E Walling, E Gindreau and A Lonvaud-Funel [27], with slight modifications. Briefly, overnight cultures were centrifuged (10000 x g, 30 min) and the cells were resuspended in PBS buffer. After blending 10 µl of the culture with 10 µl of the antiserum, the mixture was incubated for 30 min at 4 °C and subsequently analyzed microscopically at thousand-fold magnification (Axiostar Plus, Zeiss, Oberkochen, Germany). 2.6 Genomics For genome sequencing, high molecular weight DNA was isolated according to the instructions of the Qiagen Genomic Kit (Qiagen, Hilden, Germany). Single-molecule real-time sequencing (PacBio RS II) was performed at GATC Biotech (Konstanz, Germany) [28]. A library was prepared using one SMRT cell, followed by an assembly with the hierarchical genomeassembly process version 3 (HGAP 3) [29]. The genome was completed by manual curation according to PacBio instructions (https://github.com/PacificBiosciences/BioinformaticsTraining/wiki/Finishing-Bacterial-Genomes) and annotated using the NCBI Prokaryotic Genome Annotation Pipeline and the Rapid Annotations using Subsystems Technology [3032]. For detailed analysis of genes of interest, BLASTP and BLASTN were used [33, 34], for generating genomic images, BLAST Ring Image Generator (BRIG) was applied [35]. 2.7 Glycosyltransferase-2 screening To screen different beer-spoiling LAB for the existence of a glycosyltransferase 2 (gtf-2) gene, primers were designed to amplify a 900 bp region from the respective gene of L. brevis TMW 1.2112.
(GTF-F:
GAATCCGAACTAGCAATACTCGC,
GTF-R:
ACTAGTGGAATGTGCAACAC TGG). Total genomic DNA was isolated from beer-spoiling LAB strains (Table 1) using E.Z.N.A bacterial DNA Kit (Omega Bio-Tek, Inc., Norcross, 7
USA). PCR amplification was performed with a reaction mixture of 1 µl DNA, 42.25 µl deionized water, 0.25 µl of each primer, 0.25 µl Taq polymerase, 5 µl 10 x MgCl2 buffer (MP Biomedicals, Santa Ana, USA). The PCR program consisted of a step for initial denaturation (94 °C for 2 min), 32 cycles of 94 °C for 45 sec, 60.6 °C for 60 sec, and 72 °C for 54 sec and a final extension at 72 °C for 2 min. The PCR amplicons and a 100-bp plus ladder (Thermo Scientific, Waltham, USA) were separated through gel electrophoresis in 1.3 % (w/v) agarose gels in 0.5 TAE buffer, stained with dimidium bromide and visualized with ultraviolet light. 3. Results 3.1 General growth characteristics of L. brevis TMW 1.2112 L. brevis TMW 1.2112 exhibits a slimy phenotype on solid agar plates and in liquid culture. Growth on agar plates results in ropy colonies forming long filaments when extending the colony surface with an inoculation loop. In contrast, growth in liquid cultures causes highly viscous liquids with macroscopically visible slime formation (Figure 1). The inoculation of cultivation tubes with L. brevis TMW 1.2112 results in diffuse growth and successive slime formation. At the growth maximum, bacterial cells as well as slime occur all over the tube. Subsequently, the effect reverses, starting with the sedimentation of cells and the gradual disappearance of slime. The resulting cell pellet retains the slimy characteristics for several days. In contrast, the supernatant has lost all viscous properties. By shaking the cultivation tube, the cell pellet spirals upwards like a thick mucoid string. Therefore, a sticky network between the cells is indicated, suggesting a CPS more than a released EPS. 3.2 Structural characterization For the structural characterization of the exopolysaccharides formed by L. brevis TMW 1.2112, multiple chromatographic and spectroscopic approaches were applied. Methylation analysis was used to screen the glycosidic linkages of the polysaccharides (Table 2). CPS yielded only 8
glucose-derived PMAAs, whereas the chromatograms of the EPS preparation additionally showed galactose- and mannose-derived PMAAs. Mannose and galactose were also detected by high performance anion exchange chromatography (HPAEC) after acid hydrolysis. Therefore, a control composed of precipitated growth medium was analyzed. Galactose and mannose as well as the detected galactose- and mannose-derived PMAAs were found in the control. In addition, both the control and the samples contained minor amounts of 1,4substituted glucose, demonstrating that galactose, mannose, and (1→4)-linked glucose are medium-derived and most likely not exopolysaccharide constituents. Thus, it was possible to conclude that CPS and EPS contain terminal-, 1,3-, and 1,2,3-substituted D-glucose units (the absolute configuration of glucose was determined by GC-MS after chiral derivatization). However, the high portions of terminal glucose residues suggest an underestimation of the 1,3and/or the 1,2,3-substituted glucose units. Nevertheless, methylation analysis strongly suggests that both EPS and CPS of L. brevis TMW 1.2112 are composed of (1→3)-linked glucans with branches at position O2. The ratios between the 1,3- and 1,2,3-substituted glucose units indicate a ramification at every second backbone unit. To confirm these results and to assess the anomeric configuration of the monomeric units, oneand two-dimensional NMR spectroscopy was applied. Analysis of Correlated spectroscopy (COSY), Total correlated spectroscopy (TOCSY), and Heteronuclear single quantum coherence (HSQC) spectra of the polysaccharides allowed for the assignment of all 1H and 13C chemical shifts of the three structural units present (terminal, 1,3-, and 1,2,3-substituted glucose) (supplementary material Tab. S1). The 1,3- and 1,2,3-substituted glucose units showed a characteristic downfield shift for their C3/H3 or, respectively, their C2/H2 and C3/H3 correlation peaks in the HSQC spectrum (Fig. 2), indicating a substitution at this position. The 13
C chemical shifts of the unsubstituted ring protons/carbons showed chemical shifts
comparable to the β-anomer of monomeric glucose, suggesting the presence of β-glucose 9
residues. In addition, all
13
C chemical shifts of the terminal glucose residues were in good
agreement with the values reported for laminaribiose (β-(1→3)-linked glucobiose) [36]. On the other hand, slightly different chemical shifts were described for a β-(1→3)-linked glucan with branches at position O2 from Pediococcus (P.) damnosus [37]. However, similar trends for the downfield/upfield shifts of the protons/carbons were observed but chemical shifts constantly differ by 0.12 ppm (1H) or 1.5 ppm (13C), respectively. Thus, the chemical shift discrepancies are likely due to varying solvents or different referencing. Taking all results into account, it can be concluded that the L. brevis TMW 1.2112 exopolysaccharides are composed of a backbone of β-(1→3)-linked glucose units, which are ramified with β-glucose residues at position O2. As stated before, the 1,3- and 1,2,3-substituted glucose units were most likely underestimated by methylation analysis. Thus, volume integration was performed for the C2/H2 correlation peaks to get a semiquantitative estimate of the ratios between the structural elements. These signals should have roughly comparable 1JCH coupling constants and consequently an approximately comparable response. The ratio obtained for 1,3- and 1,2,3-substituted glucose units (1/0.8) indicated highly branched polysaccharides with ramifications at about every second backbone residue. In addition, terminal β-glucose residues were detected in amounts comparable to the branched backbone residues, suggesting that the terminal glucose units are mostly derived from ramifications. 3.3 Agglutination test To
confirm
the
possible
capsular
localization
of
the
identified
β-glucan,
an
immunoagglutination test with S. pneumoniae 37-specific antiserum was carried out. The strain agglutinated in presence of the antiserum, thus demonstrating the presence of a β-glucan at the cell surface as CPS (Fig. 3). 3.4 Genome sequencing 10
To gain more insights into the molecular background of β-glucan formation, genomic DNA was isolated from L. brevis TMW 1.2112 and sequenced via single-molecule real-time sequencing technology (PacBio RS II). De novo assembly was carried out by using the HGAP method. Upon sequencing six contigs were generated, which were further assembled to one finished genome. The genome is composed of a chromosome and five plasmids. General genome parameters and accession numbers are listed in Table 3. On pl12112-4, we identified a glycosyltransferase-2 (EC 2.4.1.34) (Fig. 4), which was described as key enzyme for the synthesis of β-(1,3-1,2)-glucans [38] and belongs to the glycosyltransferase family 2 of carbohydrate active enzymes. The gtf-2 gene contains 1704 nucleotides with a predicted protein product of 567 amino acids. Topological analysis indicates that the gene encodes a transmembrane protein with five helices (CBS prediction server; http://www.cbs.dtu.dk/services/TMHMM/). The other eight genes surrounding the gtf-2 are not involved in sugar metabolism, but in the maintenance and transmission of the plasmid. 3.5 Gtf-2 screening in beer-spoiling lactic acid bacteria To verify the relevance of the gtf-2 gene within the slimy spoilage of beer, beer-spoiling LAB (slime forming and non-slime forming ones) were screened for the presence of this gene. Therefore, primers targeting a 900 bp region from the gene from L. brevis TMW 1.2112 were designed and used for PCR. All slime-forming strains generated the expected amplification signal (Fig. 5), whereas amplification with DNA from non-slime forming strains remained negative. Furthermore, random sugar monomer analysis of EPS from different beer-spoiling LAB via HPLC after acidic hydrolysis of the isolated EPS revealed these EPS to be solely composed of glucose (data not shown).
11
These results support the importance of the glycosyltransferase-2 in the formation of glucans and suggest it as diagnostic marker gene for early detection of EPS forming beer-spoiling LAB, which may not be identified along previously described hop tolerance markers. 4. Discussion This study aimed for a better understanding of the EPS-based spoilage of beer by characterization of L. brevis TMW 1.2112, isolated from spoiled, viscous beer. The slime-forming ability of L. brevis TMW 1.2112 was assessed on solid and in liquid media, showing a strong slime formation in both cases. Its phenotype is consistent with the characteristics of slime-forming beer-spoiling lactobacilli from DH Williamson [39]: growth on agar plates resulting in ropy colonies; growth in liquid culture causing increased viscosity. However, the viscosity of liquid cultures disappears upon prolonged fermentation, so we suggest at least a partial degradation of the produced exopolysaccharide, which might serve as carbon source during starvation. Such an enzymatic EPS hydrolysis/degradation has been reported for several EPS-producing LAB strains [40-44]. Since L. brevis TMW 1.2112 encodes different (putative) β-glucan degrading glycosyl hydrolases, e. g. an endoglucanase (glycosyl hydrolase family 8; locus tag AZI09_02135) or a glycosyl hydrolase family 3 protein (locus tag AZI09_02175), this β-glucan could serve as carbohydrate source under certain (possibly limited) growth conditions for L. brevis TMW 1.2112. Both, the appearance on agar and the behavior in liquid culture imply the formation of a CPS, connecting the cells in some kind of sticky network. CPS formation was proved and confirmed in this study via immunological analysis. The structural characterization demonstrated that the capsule is composed of a β-(1→3)-glucan with ramifications at position O2. The PMAA ratios as well as the HSQC peak intensities suggested that about every second 1,3-linked backbone unit is ramified at position O2, which was also described for the β-glucans from P. damnosus 12
[37], Pediococcus sp. [45], and Oenococcus oeni [46]. The polysaccharides isolated from the supernatant are identical to those isolated from the cell pellet. Thus, it can be concluded that the CPS is partially secreted into the surrounding medium. Comparative genome and structural analysis as well as comprehension of known literature revealed a plasmid-located glycosyltransferase-2 as key gene for the formation of this glucan capsule. This type of glycosyltransferase was described to synthesize glucans by catalyzing the polymerization of glycosyl residues from UDP-glucose [38, 47]. Therefore, β-(1,3-1,2)-glucan formation by certain LAB resembles more the mechanisms of heteropolysaccharide biosynthesis than those of homopolysaccharides from energy-rich disaccharides such as sucrose. However, the precise mechanism is still not fully understood [48, 49]. Next to the gtf-2 gene, the plasmid harbors eight other genes, all involved in maintenance and transmission of the plasmid. The maintenance supposedly is ensured by a putative toxinantitoxin system. This system is composed of two closely linked genes encoding a toxin and its cognate antitoxin, protecting the host against the toxic effect. Loss of the plasmid causes cell death since the unstable antitoxin is degraded earlier and the lasting toxin kills the cell [50]. Accordingly, EPS formation poses a stable phenotype of L. brevis TMW 1.2112. The gtf-2 gene is also present on a plasmid of the slimy beer spoiler P. claussenii BAA-344 T. Moreover, all genes encoded on the plasmid of L. brevis are found on the respective one of P. claussenii as well (supplementary material Fig. 1S). Due to these similarities and the fact that all other genes encoded by these plasmids are not involved in sugar metabolism, we assume slimy beer spoilage to result mainly from the action of this plasmid encoded glycosyltransferase-2. This hypothesis is in accordance with the results of D Llull, R Munoz, R Lopez and E Garcia [51] which showed that the highly homologous glycosyltransferase-2 (tts) from S. pneumoniae is the only gene required to drive the biosynthesis and secretion of a
13
capsular β-glucan, while other bacterial beta-glucans such as cellulose are synthesized stepwise from UDP-glucose by the action of multienzyme complexes [52]. Glucan-forming LAB not only affect beer, but also wine and cider. As in beer, viscous wine and cider are attributed to a gtf-mediated β-(1→3)-glucan formation [27, 45-47, 53, 54]. Moreover, the glycosyltransferase is species/genera-independently highly conserved among βglucan-producing LAB. Comparing the gtf gene of the slimy wine/ cider spoilers P. parvulus IOEB8801 (accession no. AF196967), P. damnosus 2.6 (accession no. AY999683) and L. diolivorans G77 (accession no. AY999684) with the slimy beer spoiler P. claussenii ATCC BAA-344 T (locus taq PECL_RS09485) and L. brevis TMW 1.2112 (locus taq AZI09_12770) shows sequence identities of 99 % (Table 4). These similarities in the genetic background and in the structure of the resulting polysaccharide imply a common origin and horizontal gene transfer to be responsible for the dispread of the gtf gene within beverage-spoiling lactobacilli and pediococci. This assumption is strengthened by the finding that the gtf gene is located on a mobilizable plasmid in P. parvulus IOEB8801 [55] and on a conjugative one in P. parvulus 2.6 [53]. In addition, the respective gene of P. parvulus 2.6, L. brevis TMW 1.2112 and P. claussenii ATCC BAA-344 T is neighbored to a transposase (IS 30 family). This enzyme is able to cause a translocation of genes within the genome and an integration into the chromosome. This might explain the finding of chromosomal gtfs in O. oeni (IOEB0205 and I4) and L. suebicus CUPV221 which are in turn highly identical to the plasmid-located ones (Table 4). Indeed, the chromosomal gtf of O. oeni IOEB0205 is found in close proximity to a transposase. The formation of a slimy capsule within these beverage-spoiling bacteria might occur due to three different reasons. One reason for the encapsulation could be an increased resistance against environmental stress factors. Beer and wine represent both a harsh environment for bacteria due to the presence of antimicrobial hop compounds and sulfur dioxide, respectively, 14
of a low pH and a quite high ethanol content. Forming a capsule could serve as protective barrier, while it was demonstrated for P. parvulus that its resistance against these stress factors was increased [16] [17]. Therefore, it is likely that the encapsulation of TMW 1.2112 helps the strain dealing with the hurdles of harsh environments such as beer. The above-discussed role as carbon source during starvation could be another advantage, brought by capsule formation. Finally, capsules might facilitate the adherence to solid surfaces [47, 56], suggesting that L. brevis TMW 1.2112 could be involved in (brewery-associated) biofilm formation. Biofilms in turn are hotspots for gene transfers from one species to another [57]. Consequently, the promotion of the development of gtf-positive lactobacilli or pediococci seems possible. This clarifies the need and the importance for an early detection of such bacteria in breweries. By designing primers targeting the gtf-2 gene, it was possible to detect slimy beer-spoiling LAB species-independently and to distinguish them from non-slimy ones. Thus, this gene represents an appropriate target for the reliable identification of these brewery contaminants with PCR. ACKNOWLEDGEMENT Part of this work was by the German Ministry of Economics and Technology (via AiF) and the Wifoe (Wissenschaftsförderung der Deutschen Brauwirtschaft e.V., Berlin) project AiF 17576 and18194 N. None of the funding sources did have any influence on the study design, the collection, analysis and interpretation of data, the writing of the report and the decision to submit the article for publication. The authors are grateful to the Forschungszentrum Weihenstephan für Brau- und Lebensmittelqualität for providing bacterial strains.
15
REFERENCES
1.
2.
3.
4. 5. 6. 7. 8. 9.
10.
11.
12.
13.
14.
15.
Suzuki K: 125th Anniversary Review: Microbiological instability of beer caused by spoilage bacteria. Journal of the Institute of Brewing 2011, 117(2):131-155. Vogel RF, Behr J, Geissler AJ: The lifestyle of beer-spoiling lactic acid bacteria is based on digging in the plasmid pool. In: World Brewing Congress: 2016; Denver, USA. American Society of Brewing Chemists. Behr J, Geissler AJ, Schmid J, Zehe A, Vogel RF: The Identification of Novel Diagnostic Marker Genes for the Detection of Beer Spoiling Pediococcus damnosus Strains Using the BlAst Diagnostic Gene findEr. PloS one 2016, 11(3):e0152747. Simpson WJ: Studies on the sensitivity of lactic acid bacteria to hop bitter acids. J Inst Brew 1993, 99:405-412. Simpson WJ: Ionophoric action of trans-isohumulone on Lactobacillus brevis. J Gen Microbiol 1993, 139:1041-1046. Simpson WJ, Fernandez JL: Mechanism of resistance of lactic acid bacteria to trans-isohumulone. J Am Soc brew Chem 1994, 52:9-11. Behr J, Vogel RF: Mechanisms of hop inhibition: hop ionophores. Journal of agricultural and food chemistry 2009, 57(14):6074-6081. Sakamoto K, Konings WN: Beer spoilage bacteria and hop resistance. International journal of food microbiology 2003, 89(2-3):105-124. Sakamoto K, Margolles A, van Veen HW, Konings WN: Hop resistance in the beer spoilage bacterium Lactobacillus brevis is mediated by the ATPbinding cassette multidrug transporter HorA. Journal of bacteriology 2001, 183(18):5371-5375. Suzuki K, Sami M, Kadokura H, Nakajima H, Kitamoto K: Biochemical characterization of horA-independent hop resistance mechanism in Lactobacillus brevis. International journal of food microbiology 2002, 76(3):223-230. Nwodo UU, Green E, Okoh AI: Bacterial exopolysaccharides: functionality and prospects. International journal of molecular sciences 2012, 13(11):14002-14015. Brandt JU, Jakob F, Behr J, Geissler AJ, Vogel RF: Dissection of exopolysaccharide biosynthesis in Kozakia baliensis. Microb Cell Fact 2016, 15(1):170. Ua-Arak T, Jakob F, Vogel RF: Characterization of growth and exopolysaccharide production of selected acetic acid bacteria in buckwheat sourdoughs. International journal of food microbiology 2016, 239:103-112. Jakob F, Pfaff A, Novoa-Carballal R, Rubsam H, Becker T, Vogel RF: Structural analysis of fructans produced by acetic acid bacteria reveals a relation to hydrocolloid function. Carbohydr Polym 2013, 92(2):1234-1242. Ruhmkorf C, Jungkunz S, Wagner M, Vogel RF: Optimization of homoexopolysaccharide formation by lactobacilli in gluten-free sourdoughs. Food microbiology 2012, 32(2):286-294.
16
16.
17.
18.
19. 20.
21.
22.
23.
24.
25. 26.
27.
28.
29.
30.
31.
Stack HM, Kearney N, Stanton C, Fitzgerald GF, Ross RP: Association of beta-glucan endogenous production with increased stress tolerance of intestinal lactobacilli. Appl Environ Microbiol 2010, 76(2):500-507. Coulon J, Houles A, Dimopoulou M, Maupeu J, Dols-Lafargue M: Lysozyme resistance of the ropy strain Pediococcus parvulus IOEB 8801 is correlated with beta-glucan accumulation around the cell. Int J Food Microbiol 2012, 159(1):25-29. Kern CC, Usbeck JC, Vogel RF, Behr J: Optimization of Matrix-AssistedLaser-Desorption-Ionization-Time-Of-Flight Mass Spectrometry for the identification of bacterial contaminants in beverages. Journal of microbiological methods 2013, 93(3):185-191. Geissler AJ, Behr J, von Kamp K, Vogel RF: Metabolic strategies of beer spoilage lactic acid bacteria in beer. Int J Food Microbiol 2016, 216:60-68. Tallon R, Bressollier P, Urdaci MC: Isolation and characterization of two exopolysaccharides produced by Lactobacillus plantarum EP56. Research in microbiology 2003, 154(10):705-712. Wefers D, Bunzel M: Characterization of dietary fiber polysaccharides from dehulled, common buckwheat (Fagopyrum esculentum) seeds. Cereal Chem 2015, 92(6):598-603. Sweet DP, Shapiro RH, Albersheim P: Quantitative-analysis by various GLC response-factor theories for partially methylated and partially ethylated alditol acetates. Carbohydrate Research 1975, 40(2):217-225. Saeman JF, Bubl JL, Harris EE: Quantitative saccharification of wood and cellulose. Industrial and Engineering Chemistry-Analytical Edition 1945, 17(1):35-37. De Ruiter GA, Schols HA, Voragen AGJ, Rombouts FM: Carbohydrate analysis of water-soluble uronic acid-containing polysaccharides with high-performance anion-exchange chromatography using methanolysis combined with TFA hydrolysis is superior to four other methods. Anal Biochem 1992, 207(1):176-185. Wefers D, Tyl CE, Bunzel M: Novel arabinan and galactan oligosaccharides from dicotyledonous plants. Frontiers in Chemistry 2014, 2:100. Gottlieb HE, Kotlyar V, Nudelman A: NMR chemical shifts of common laboratory solvents as trace impurities. J Org Chem 1997, 62(21):75127515. Walling E, Gindreau E, Lonvaud-Funel A: A putative glucan synthase gene dps detected in exopolysaccharide-producing Pediococcus damnosus and Oenococcus oeni strains isolated from wine and cider. Int J Food Microbiol 2005, 98(1):53-62. Eid J, Fehr A, Gray J, Luong K, Lyle J, Otto G, Peluso P, Rank D, Baybayan P, Bettman B et al: Real-time DNA sequencing from single polymerase molecules. Science 2009, 323(5910):133-138. Chin CS, Alexander DH, Marks P, Klammer AA, Drake J, Heiner C, Clum A, Copeland A, Huddleston J, Eichler EE et al: Nonhybrid, finished microbial genome assemblies from long-read SMRT sequencing data. Nature methods 2013, 10(6):563-569. Aziz RK, Bartels D, Best AA, DeJongh M, Disz T, Edwards RA, Formsma K, Gerdes S, Glass EM, Kubal M et al: The RAST Server: rapid annotations using subsystems technology. BMC genomics 2008, 9:75. Overbeek R, Olson R, Pusch GD, Olsen GJ, Davis JJ, Disz T, Edwards RA, Gerdes S, Parrello B, Shukla M et al: The SEED and the Rapid Annotation of 17
32.
33. 34.
35.
36.
37.
38. 39. 40.
41.
42.
43.
44.
45.
46.
microbial genomes using Subsystems Technology (RAST). Nucleic acids research 2014, 42(Database issue):D206-214. Angiuoli SV, Gussman A, Klimke W, Cochrane G, Field D, Garrity G, Kodira CD, Kyrpides N, Madupu R, Markowitz V et al: Toward an online repository of Standard Operating Procedures (SOPs) for (meta)genomic annotation. Omics : a journal of integrative biology 2008, 12(2):137-141. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ: Basic local alignment search tool. Journal of molecular biology 1990, 215(3):403-410. Camacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, Bealer K, Madden TL: BLAST+: architecture and applications. BMC bioinformatics 2009, 10:421. Alikhan NF, Petty NK, Ben Zakour NL, Beatson SA: BLAST Ring Image Generator (BRIG): simple prokaryote genome comparisons. BMC genomics 2011, 12:402. Roslund MU, Tahtinen P, Niemitz M, Sjhohn R: Complete assignments of the 1H and 13C chemical shifts and J(H,H) coupling constants in NMR spectra of D-glucopyranose and all D-glucopyranosyl-D-glucopyranosides. Carbohydr Res 2008, 343(1):101-112. Duenas-Chasco MT, Rodriguez-Carvajal MA, Tejero Mateo P, FrancoRodriguez G, Espartero JL, Irastorza-Iribas A, Gil-Serrano AM: Structural analysis of the exopolysaccharide produced by Pediococcus damnosus 2.6. Carbohydr Res 1997, 303(4):453-458. Karnezis T, McIntosh M, Wardak A, Stanisich VA, Stone BA: The Biosynthesis of ß-Glycans. Trends in Glycoscience and Glycotechnology 2000, 12(66):17. Williamson DH: Studies on lactobacilli causing ropiness in beer. Journal of Applied bacteriology 1959, 22:392-402. Cerning J, Bouillanne C, Landon M, Desmazeaud M: Isolation and Characterization of Exopolysaccharides from Slime-Forming Mesophilic Lactic Acid Bacteria. Journal of Dairy Science 1992, 75(3):692-698. Cerning J, Desmazeaud M, Landon M: Exocellular polysaccharide production by Streptococcus thermophilus. Biotechnology letter 1988, 10(4):255-260. Pham PL, Dupont I, Roy D, Lapointe G, Cerning J: Production of exopolysaccharide by Lactobacillus rhamnosus R and analysis of its enzymatic degradation during prolonged fermentation. Applied and environmental microbiology 2000, 66(6):2302-2310. Gancel F, Novel G: Exopolysaccharide Production by Streptococcus salivarius ssp. thermophilus Cultures. 2. Distinct Modes of Polymer Production and Degradation Among Clonal Variants. Journal of Dairy Science 1994, 77(3):7. Ricciardi A, Parente E, Crudele MA, Zanetti F, Scolari G, Mannazzu I: Exopolysaccharide production by Streptococcus thermophilus SY: production and preliminary characterization of the polymer. Journal of applied microbiology 2002, 92(2):297-306. Llauberes RM, Richard B, Lonvaud A, Dubourdieu D, Fournet B: Structure of an exocellular beta-D-glucan from Pediococcus sp., a wine lactic bacteria. Carbohydr Res 1990, 203(1):103-107. Ibarburu I, Soria-Diaz ME, Rodriguez-Carvajal MA, Velasco SE, Tejero-Mateo P, Gil-Serrano AM, Irastorza A, Duenas MT: Growth and exopolysaccharide (EPS) production by Oenococcus oeni I4 and structural characterization of their EPSs. Journal of applied microbiology 2007, 103(2):477-486. 18
47.
48. 49.
50. 51.
52.
53.
54.
55.
56.
57. 58.
59.
60.
Dols-Lafargue M, Lee HY, Le Marrec C, Heyraud A, Chambat G, Lonvaud-Funel A: Characterization of gtf, a glucosyltransferase gene in the genomes of Pediococcus parvulus and Oenococcus oeni, two bacterial species commonly found in wine. Appl Environ Microbiol 2008, 74(13):4079-4090. McIntosh M, Stone BA, Stanisich VA: Curdlan and other bacterial (1-->3)beta-D-glucans. Appl Microbiol Biotechnol 2005, 68(2):163-173. Torino MI, Font de Valdez G, Mozzi F: Biopolymers from lactic acid bacteria. Novel applications in foods and beverages. Frontiers in microbiology 2015, 6:834. Yamaguchi Y, Park JH, Inouye M: Toxin-antitoxin systems in bacteria and archaea. Annual review of genetics 2011, 45:61-79. Llull D, Munoz R, Lopez R, Garcia E: A single gene (tts) located outside the cap locus directs the formation of Streptococcus pneumoniae type 37 capsular polysaccharide. Type 37 pneumococci are natural, genetically binary strains. The Journal of experimental medicine 1999, 190(2):241-251. Chawla PR, Bajaj IB, Survase SA, Singhal RS: Microbial Cellulose: Fermentative Production and Applications. Food Technol Biotechnol 2009, 47(2):107-125. Werning ML, Ibarburu I, Duenas MT, Irastorza A, Navas J, Lopez P: Pediococcus parvulus gtf gene encoding the GTF glycosyltransferase and its application for specific PCR detection of beta-D-glucan-producing bacteria in foods and beverages. Journal of food protection 2006, 69(1):161169. Duenas-Chasco MT, Rodriguez-Carvajal MA, Tejero-Mateo P, Espartero JL, Irastorza-Iribas A, Gil-Serrano AM: Structural analysis of the exopolysaccharides produced by Lactobacillus spp. G-77. Carbohydr Res 1998, 307(1-2):125-133. Gindreau E, Walling E, Lonvaud-Funel A: Direct polymerase chain reaction detection of ropy Pediococcus damnosus strains in wine. Journal of applied microbiology 2001, 90(4):535-542. de Palencia PF, Werning ML, Sierra-Filardi E, Duenas MT, Irastorza A, Corbi AL, Lopez P: Probiotic properties of the 2-substituted (1,3)-beta-D-glucanproducing bacterium Pediococcus parvulus 2.6. Applied and environmental microbiology 2009, 75(14):4887-4891. Flemming HC, Wingender J: The biofilm matrix. Nature reviews Microbiology 2010, 8(9):623-633. Pittet V, Abegunde T, Marfleet T, Haakensen M, Morrow K, Jayaprakash T, Schroeder K, Trost B, Byrns S, Bergsveinson J et al: Complete genome sequence of the beer spoilage organism Pediococcus claussenii ATCC BAA-344T. Journal of bacteriology 2012, 194(5):1271-1272. Juvonen R, Honkapaa K, Maina NH, Shi Q, Viljanen K, Maaheimo H, Virkki L, Tenkanen M, Lantto R: The impact of fermentation with exopolysaccharide producing lactic acid bacteria on rheological, chemical and sensory properties of pureed carrots (Daucus carota L.). International journal of food microbiology 2015, 207:109-118. Garai-Ibabe G, Duenas MT, Irastorza A, Sierra-Filardi E, Werning ML, Lopez P, Corbi AL, Fernandez de Palencia P: Naturally occurring 2-substituted (1,3)beta-D-glucan producing Lactobacillus suebicus and Pediococcus parvulus strains with potential utility in the production of functional foods. Bioresource technology 2010, 101(23):9254-9263. 19
FIGURE CAPTIONS
Figure 1: Growth and slime forming dynamics of L. brevis TMW 1.2112 over a period of six days. Figure 2: HSQC spectrum and proposed structure of the L. brevis TMW 1.2112 CPS preparation. The characteristic downfield shifts of the correlation peaks at the substituted positions are encircled. Figure 3: Agglutination induced by S. pneumoniae type 37-specific antiserum. L. brevis TMW 1.2112 without antiserum (A) and after antiserum addition (B). Figure 4: RAST-annotated plasmid pl12112-4 of L. brevis TMW 1.2112. The predicted ORFs are mapped on plasmid 4 using BRIG. Starting from inside: circle 1 shows the general position in kilobases; circle 2 depicts the G+C content; circle 3 presents genes encoded, in red the gtf-2. With exception of gtf-2, the genes are not related to EPS formation, only for plasmid conservation. Figure 5: Gel electrophoresis of PCR amplicons generated with primers Gtf-F/Gtf-R on the following strains: L. brevis (line 2-20), L. parabuchneri (line 21), L. rossiae (line 22-23) – according to table 1. Negative control (line 24), molecular mass standard (line 1). The length of the amplicon is indicated. Figure S1: Similarities between plasmid pl12112-4 of L. brevis TMW 1.2112 and plasmid pPECL7 of P. claussenii ATCC BAA-344 T. Starting from inside: circle 1 shows the general position in kilobases; circle 2 depicts the G+C content of L. brevis TMW 1.2112; circle 3 presents genes encoded on pl12112-4 of L. brevis TMW 1.2112, in red the gtf-2. Circle 4 exhibits the corresponding genes of plasmid pPECL7 of P. claussenii ATCC BAA-344 T.
20
Table 1: Strains and sources of isolation. Species
Strain
Source
L. brevis
TMW 1.2107
Slimy spoiled beer
L. brevis
TMW 1.2108
Slimy spoiled beer
L. brevis
TMW 1.2109
Slimy spoiled beer
L. brevis
TMW 1.2110
Slimy spoiled beer
L. brevis
TMW 1.2111
Slimy spoiled beer
L. brevis
TMW 1.2112
Slimy spoiled beer
L. brevis
TMW 1.2113
Brewery surface
L. brevis
TMW 1.2114
Slimy spoiled beer
L. brevis
TMW 1.2115
Slimy spoiled beer
L. brevis
TMW 1.599
Slimy spoiled beer
L. brevis
TMW 1.240
Slimy spoiled beer
L. brevis
TMW 1.2146
Slimy spoiled beer
L. brevis
TMW 1.2147
Slimy spoiled beer
L. brevis
TMW 1.2148
Slimy spoiled beer
L. brevis
TMW 1.2149
Slimy spoiled beer
L. brevis
TMW 1.2151
Slimy spoiled beer
L. brevis
TMW 1.2153
Slimy spoiled beer
L. brevis
TMW 1.2154
Slimy spoiled beer
L. brevis
TMW 1.2155
Slimy spoiled beer
L. parabuchneri
TMW 1.1141
Beer
L. rossiae
TMW 1.2152
Slimy spoiled beer
L. rossiae
TMW 1.2156
Slimy spoiled beer
TMW = Technische Mikrobiologie Weihenstephan.
21
Table 2: Percentages of the partially methylated alditol acetates from the L. brevis TMW 1.2112 CPS and EPS preparation. Glycosidic linkage
CPS
EPS
t-Glcp
50.7 %
54.6 %
1,3-Glcp
23.8 %
23.5 %
1,2,3-Glcp
25.5 %
21.9 %
t = terminal. Numbers indicate the substituted positions of a sugar unit.
Table 3: Sequencing statistics, genome information and accession numbers of L. brevis TMW 1.2112. The biosample SMNO4517633 is part or the bioproject PRJNA313253. Coverage of HGAP assembly is 396. CDS = Coding sequences based on NCBI PGAP. Contig Accession no.
Internal name
Size (bp)
C+G content (%)
CDS
1
CP016797
chr12112-1
2488939
45.98
2338
2
CP016798
pl12112-1
44555
41.59
60
3
CP016799
pl12112-2
38644
42.58
36
4
CP016800
pl12112-3
33211
45.64
35
5
CP016801
pl12112-4
8521
36.78
9
6
CP016802
pl12112-5
59682
40.35
59
22
Table 4: Overview of EPS-forming beverage-spoiling lactic acid bacteria. Strain
Origin
L. brevis
beer
Gtf
AA
Gtf
Poly
Locus tag*/
location
gtf
identity
saccharide
accession no.
plasmid
567
-
β-(1→3)-
AZI09_12770
TMW 1.2112 P.
glucan
claussenii beer
plasmid
567
99 %
ATCC-BAA-
β-(1→3)-
PECL_RS09485
glucan
344T1,2 P.
damnosus wine
plasmid
567
99 %
IOEB88013 P. parvulus
cider
plasmid
567
99 %
cider
plasmid
567
99 %
cider
chr
567
99 %
β-(1→3)-
AY999684
β-(1→3)-
GU174474
glucan wine
chr
567
97 %
IOEB02057 O. oeni
AY999683
glucan
CUPV2216 O. oeni
β-(1→3)glucan
G775 L. suebicus
AF196967
glucan
2.64 L. diolivorans
β-(1→3)-
β-(1→3)-
EU556433
glucan cider
chr
567
I48
98 %
β-(1→3)-
AY999685
glucan
AA = amino acids; Gtf identity = sequence identity to gtf-2 of L. brevis TMW 1.2112; chr = chromosome; * locus tag only used when available 1
= V Pittet, T Abegunde, T Marfleet, M Haakensen, K Morrow, T Jayaprakash, K Schroeder,
B Trost, S Byrns, J Bergsveinson, et al. [58], 2 = R Juvonen, K Honkapaa, NH Maina, Q Shi, K Viljanen, H Maaheimo, L Virkki, M Tenkanen and R Lantto [59] 3 = E Walling, E Gindreau and A Lonvaud-Funel [27], 4 = ML Werning, I Ibarburu, MT Duenas, A Irastorza, J Navas and P Lopez [53], 5 = ML Werning, I Ibarburu, MT Duenas, A Irastorza, J Navas and P Lopez [53], 6
= G Garai-Ibabe, MT Duenas, A Irastorza, E Sierra-Filardi, ML Werning, P Lopez, AL Corbi
and P Fernandez de Palencia [60], 7 = M Dols-Lafargue, HY Lee, C Le Marrec, A Heyraud, G Chambat and A Lonvaud-Funel [47], 8 = ML Werning, I Ibarburu, MT Duenas, A Irastorza, J Navas and P Lopez [53]
23
Supplementary material Table S1: 1H and 13C chemical shifts of the three structural elements present in the CPS and EPS of L. brevis TMW 1.2112. 1,3-Glcp = 1,3-substituted β-glucopyranose, 1,2,3-Glcp = 1,2,3substituted β-glucopyranose, t-Glcp = terminal β-glucopyranose bound to position O2 of a 1,3substituted β-glucopyranose. Chemical shifts are given in ppm. Structural element 1,3-Glcp
1
2
3
4
5
6
4.90
3.63
3.79
3.55
3.51
3.74/3.91
13
102.24
73.35
86.30
68.71
76.26
61.30
1
4.88
3.84
3.97
3.55
3.51
3.74/3.91
13
102.26
80.60
84.11
68.71
76.26
61.30
1
4.95
3.31
3.51
3.40
3.51
3.74/3.95
102.95
74.77
77.24
70.42
76.26
61.52
1
H C
1,2,3-Glcp
H C
t-Glcp
H
13
C
24
Figure 5
Figure 1
25
Figure 2
Figure 3
26
Figure 4
27