Obligate fructophilic lactic acid bacteria (FLAB) lack the metabolic pathways used in the metabolism of most carbohydrates and differ from other lactic acid bacteria in that they prefer to ferment d-fructose instead of d-glucose. These characteristics are well conserved at the genus or species level. Leuconostoc citreum F192-5 shows similar growth characteristics. However, the strain is metabolically and genomically different from obligate FLAB. This is an example of a strain that evolved a pseudo-FLAB phenotype to adapt to a fructose-rich environment.
KEYWORDS: Leuconostoc citreum, adhE, comparative genomics, fructophilic lactic acid bacteria, pseudofructophilic
ABSTRACT
Fructophilic lactic acid bacteria (FLAB), composed of Fructobacillus spp., Lactobacillus kunkeei, and Lactobacillus apinorum, are unique in that they prefer d-fructose over d-glucose as a carbon source. Strain F192-5, isolated from the peel of a satsuma mandarin and identified as Leuconostoc citreum, grows well on d-fructose but poorly on d-glucose and produces mainly lactate and acetate, with trace amounts of ethanol, from the metabolism of d-glucose. These characteristics are identical to those of obligate FLAB. However, strain F192-5 ferments a greater variety of carbohydrates than known FLAB. Comparative analyses of the genomes of strain F192-5 and reference strains of L. citreum revealed no signs of specific gene reductions, especially genes involved in carbohydrate transport and metabolism, in the genome of F192-5. The bifunctional alcohol/acetaldehyde dehydrogenase gene (adhE) is conserved in strain F192-5 but is not transcribed. This is most likely due to a deletion in the promoter region upstream of the adhE gene. Strain F192-5 did, however, ferment d-glucose when transformed with a plasmid containing the allochthonous adhE gene. L. citreum F192-5 is an example of a pseudo-FLAB strain with a deficiency in d-glucose metabolism. This unique phenotypic characteristic appears to be strain specific within the species L. citreum. This might be one of the strategies lactic acid bacteria use to adapt to diverse environmental conditions.
IMPORTANCE Obligate fructophilic lactic acid bacteria (FLAB) lack the metabolic pathways used in the metabolism of most carbohydrates and differ from other lactic acid bacteria in that they prefer to ferment d-fructose instead of d-glucose. These characteristics are well conserved at the genus or species level. Leuconostoc citreum F192-5 shows similar growth characteristics. However, the strain is metabolically and genomically different from obligate FLAB. This is an example of a strain that evolved a pseudo-FLAB phenotype to adapt to a fructose-rich environment.
INTRODUCTION
Lactic acid bacteria (LAB) are found in a variety of environments, including dairy products, fermented food, silage, and the gastrointestinal tracts of humans and animals, and are known to adapt to habitats at the genomic level (1). Fructophilic lactic acid bacteria (FLAB) differ from LAB in that they prefer fructose over glucose as a growth substrate (2, 3). External electron acceptors such as pyruvate, oxygen, and fructose dramatically enhance glucose metabolism in FLAB. Species thus far described have been isolated from d-fructose-rich niches such as flowers, fruits, fermented food made from fruits, and the gastrointestinal tracts of insects such as bumblebees, honeybees, tropical fruit flies, and Camponotus ants that feed on d-fructose-rich diets (3). FLAB are divided into two taxonomic groups, i.e., the Fructobacillus group and the Lactobacillus kunkeei-Lactobacillus apinorum group (2). Species from both groups lack the bifunctional alcohol/acetaldehyde dehydrogenase (ADH/ALDH) gene (adhE) (4–6), resulting in an imbalance of NAD/NADH during glucose metabolism (7). No other species have been reported as typical obligate FLAB.
The genus Leuconostoc is the type genus of the family Leuconostocaceae which contains the genus Fructobacillus. Fructobacillus spp. were originally classified into the genus Leuconostoc but later reclassified as members of a novel genus (8). The two genera are phylogenetically related but are clearly separated based on genetic, biochemical, and morphological characteristics (4). Leuconostoc spp. have diverse physiological and biochemical characteristics and have been isolated from different niches, mainly, fermented plant materials and the gastrointestinal tracts of omnivores, including human infants (9–11). Fructophilic characteristics have, however, never been reported for Leuconostoc spp.
During an ecological study of FLAB, a unique strain (F192-5) was isolated from the peel of a satsuma mandarin. The strain shares similar growth characteristics with typical FLAB but is phylogenetically related to Leuconostoc citreum. Comparative genomic analyses and characterization of the adhE gene distinguished strain F192-5 from other strains of L. citreum. Here, we report strain-specific pseudofructophilic evolution.
RESULTS
Isolation, identification, and metabolic characteristics.
Strain F192-5 grew actively in fructose-yeast extract-peptone (FYP) but poorly in glucose-yeast extract-peptone (GYP) broth and was regarded a candidate of FLAB. BLAST analysis of nearly complete 16S rRNA gene sequences (1,542 bp) showed 99.7% similarity to L. citreum ATCC 49370T, and strain F192-5 was identified as L. citreum. Strain F192-5 grew well in the presence of d-fructose (FYP) and d-glucose, but only when d-glucose was supplemented with pyruvate (GYP-P). Good growth of strain F192-5 was recorded in GYP under aerobic conditions but not in GYP without electron acceptors. The type strain of L. citreum (NRIC 1776T), on the other hand, grew well under all conditions tested (Fig. 1a and b). Growth characteristics of L. citreum NRIC 1772, NRIC 1579, and NRIC 1581 were similar to that of strain NRIC 1776T, and the strains reached optical density at 660 nm (OD660) values of 1.5 to 2.0 after 24 h in GYP broth (data not shown). Strain F192-5 produced pinpoint colonies (0.1 to 0.2 mm in diameter) on GYP agar when anaerobically incubated for 48 h (Fig. 1c). Reference strains of L. citreum produced large colonies (2 to 3 mm in diameter) under these conditions. Strain F192-5 produced lactate and acetate, with trace amounts of ethanol, from the fermentation of d-glucose (Table 1). Reference strains of L. citreum produced lactate and ethanol as main end products from the fermentation of d-glucose. Strain F192-5 fermented d-fructose within 1 day but took 2 days to ferment d-glucose. The unique growth and metabolic properties recorded for strain F192-5 are consistent with those of typical FLAB (2, 3). Strain F192-5 does, however, differ from typical FLAB by fermenting more carbohydrates, including l-arabinose, d-xylose, d-mannose, mannitol, methyl-α-d-glucopyranoside, N-acetylglucosamine, arbutin, salicin, cellobiose, maltose, sucrose, trehalose, and turanose, after 1 to 5 days of incubation (3). Reference strains of L. citreum showed similar carbohydrate fermentation patterns to those of strain F192-5 but fermented d-glucose within 1 day.
FIG 1.
Growth characteristics of L. citreum NRIC 1776T (a) and F192-5 (b) in GYP broth (○), FYP broth (Δ), GYP-P broth (◊), and GYP broth under aerobic conditions (□) and on GYP agar under anaerobic conditions (c). In panels a and b, lines and error bars indicate averages and standard deviations from triplicates, respectively.
TABLE 1.
End products from d-glucose
| Strain | Consumed d-glucose (mM) | End products (mM)a |
Molar ratio of L:E:A | Carbon recovery (%) | ||
|---|---|---|---|---|---|---|
| L | E | A | ||||
| NRIC 1776T | 51.3 | 56.3 | 47.8 | 1.9 | 1:0.85:0.03 | 87.2 |
| NRIC 1772 | 51.4 | 52.4 | 45.5 | 3.2 | 1:0.87:0.06 | 82.6 |
| NRIC 1579 | 51.3 | 51.8 | 44.2 | 1.9 | 1:0.85:0.04 | 80.4 |
| NRIC 1581 | 51.4 | 53.3 | 45.7 | 3.3 | 1:0.86:0.06 | 83.6 |
| F192-5 | 13.9 | 14.6 | 0.2 | 14.1 | 1:0.02:0.97 | 86.7 |
| 31-11 | 53.9 | 58.3 | 71.8 | 1.2 | 1:1.23:0.02 | 99.3 |
| 2-1 | 9.6 | 10.3 | 0.3 | 12.5 | 1:0.03:1.21 | 98.2 |
L, lactate; E, ethanol; A, acetate.
Genomic analysis.
L. citreum F192-5 has a slightly larger genome (2.06 Mbp) and more coding DNA sequences (CDSs; 2,051) than other L. citreum strains (average ± standard deviation [SD], 1.84 ± 0.08 Mbp and 1,835 ± 90 CDSs), as shown in Table 2. Strain F192-5 and all reference strains, including the type strain of L. citreum (NRIC 1776T), had average nucleotide identity (ANI) values of 0.990 to 0.998, indicating that they belong to a single taxon. L. citreum strains have 1,357 core genes, which accounts for 66.2% of the total CDSs in the F192-5 genome. The pan-genome size is 2,893. Functional classification of CDSs predicted in each strain showed large gene repertoires of classes for [J] translation, [G] carbohydrate transport and metabolism, [K] transcription, and [E] amino acid metabolism and transport. Few differences were found between strain F192-5 and other L. citreum strains (see Table S1 in the supplemental material). Contents of genes involved in metabolic/biosynthesis pathways, as determined by ortholog assignment and pathway mapping using the KEGG Automatic Annotation Server (KAAS), showed that strain F192-5 has gene repertoires for metabolism of carbohydrates and amino acids similar to those for other L. citreum strains (data not shown). Strain F192-5 has 14 genes involved in the phosphotransferase system (PTS), encoding an estimated six complete PTS transporters. Similar results were obtained for other L. citreum strains.
TABLE 2.
General genomic characteristics of L. citreum strains used for genomic analysis
| Strain | Source | Genome size (Mbp) | No. of CDSs | Genome statusa | Contamination (%) | Completeness (%) | Accession no. |
|---|---|---|---|---|---|---|---|
| F192-5 | Peel of satsuma mandarin | 2.06 | 2,051 | D | 4.02 | 98.67 | BJJW01000000.1 |
| NRIC 1776T | Honey dew of rye ear | 1.81 | 1,781 | D | 1.49 | 99.65 | BJJX01000000.1 |
| 1301_LGAS | Intensive care units | 1.82 | 1,784 | D | 0.56 | 97.69 | NZ_JVUU00000000.1 |
| NRRL B-1299 | Unknown | 1.75 | 1,717 | D | 0.23 | 98.63 | NZ_CCNH00000000.1 |
| NRRL B-742 | Unknown | 1.75 | 1,749 | D | 0.21 | 97.50 | NZ_CCNG00000000.1 |
| DmW_11 | Wild Drosophila | 1.83 | 1,847 | D | 2.67 | 98.67 | NZ_NDXG00000000.1 |
| KM20 | Kimchi | 1.90 | 1,915 | C | 0.03 | 99.65 | NC_010471.1 |
| LBAE_C10 | Sourdough | 1.93 | 1,939 | D | 1.18 | 99.59 | NZ_CAGE00000000.1 |
| LBAE_C11 | Sourdough | 1.97 | 1,976 | D | 0.52 | 99.65 | NZ_CAGF00000000.1 |
| LBAE_E16 | Sourdough | 1.80 | 1,806 | D | 0.03 | 99.65 | NZ_CAGG00000000.1 |
D, draft genome sequence; C, complete genome sequence.
Strain-specific genes in each strain were initially screened by orthologous clustering. The number of strain-specific genes ranged from 16 to 175 (average ± SD, 81.6 ± 51.7) (Table 3). Specificity of the genes was further studied based on standard protein BLAST (BLASTP) analysis against other L. citreum strains using the NCBI BLAST web server. The analysis dramatically reduced the number of strain-specific genes, ranging from 1 to 68 (average ± SD, 27.5 ± 24.4) (Table 3). Strain F192-5 possessed the largest number of specific genes (68 genes). The number of genes missing specifically in F192-5 but conserved in the other nine strains was one, which was the smallest in this analysis (ranging from 1 to 39 in the tested strains). BLASTP analysis has shown that 7 of the 68 specific genes in strain F192-5 were conserved in certain strains of Fructobacillus spp. The seven genes encode (i) abortive infection gene product AbiN, (ii) DNA replication initiation protein, (iii) type II toxin-antitoxin system antitoxin, (iv) RelB/DinJ family, (v) zinc-binding alcohol dehydrogenase family protein, (vi) phage portal protein, and (vii) hypothetical protein. The zinc-binding alcohol dehydrogenase gene was sandwiched between genes encoding the MarR family transcriptional regulator and a hypothetical protein in the F192-5 genome. This gene organization is identical to that observed in Fructobacillus fructosus NRIC 1058T (GCA_001047095.2). Similarities of the three proteins between strains F192-5 and F. fructosus NRIC 1058T ranged from 66% to 75%.
TABLE 3.
Strain-specific genes in L. citreum strains
| Strain | No. of strain-specific genes |
No. of genes shared with Fructobacillus spp.a | |
|---|---|---|---|
| Orthologous clustering | BLASTP analysis | ||
| F192-5 | 123 | 68 | 7 |
| NRIC 1776T | 16 | 5 | 1 |
| 1301_LGAS | 70 | 23 | 0 |
| NRRL B-1299 | 43 | 10 | 1 |
| NRRL B-742 | 45 | 1 | 1 |
| DmW_11 | 58 | 37 | 4 |
| KM20 | 47 | 7 | 0 |
| LBAE_C10 | 149 | 68 | 0 |
| LBAE_C11 | 175 | 24 | 0 |
| LBAE_E16 | 90 | 32 | 4 |
As determined by BLASTP analysis.
The bifunctional alcohol/acetaldehyde dehydrogenase gene (adhE) consisting of 2,703 bp was conserved in the genomes of strain F192-5 and the other strains of L. citreum. The adhE/AdhE of F192-5 and NRIC 1776T shared sequence similarities of 99.7% at the DNA level and 99.8% at the amino acid level. The adhE gene in all strains studied had no stop codon.
Characterization of adhE gene and relevant enzyme activities.
As described above, the adhE gene was found in the genome of strain F192-5. The gene is sandwiched between genes encoding glycerophosphoryl diester phosphodiesterase and an uncharacterized protein (Fig. 2). This gene organization was identical with two of the nine reference strains (NRIC 1776T and DmW_11). In seven of the reference strains, the gene encoding the uncharacterized protein was replaced with a gene encoding a membrane protein. In the genome of strain F192-5, only 62 bp separated the adhE gene and the gene encoding glycerophosphoryl diester phosphodiesterase (Fig. 2). This area in the genomes of the nine reference strains was much longer (229 or 230 bp) (Fig. 2). Alignment of the sequences has shown a 167- to 168-bp gap in the genome of F192-5 (see Fig. S1). A putative promoter was found between the adhE and glycerophosphoryl diester phosphodiesterase genes of the reference strains but not for F192-5 (Fig. 2).
FIG 2.
Organization of adhE and its upstream and downstream genes in L. citreum NRIC 1776T and F192-5. White, black, and slashed arrows indicate genes encoding glycerophosphoryl diester phosphodiesterase, AdhE, and uncharacterized protein, respectively. Single-elbow arrows indicate promoters predicted.
Reverse-transcriptase PCR revealed that L. citreum NRIC 1776T expressed its adhE (adhE-citreum) gene, but F192-5 did not (Fig. 3). The expression of reference genes, i.e., glk and the 16S rRNA gene, was recovered in both strains. Strong ADH activities and relatively weak ALDH activities were recorded in NRIC 1776T but not in strain F192-5 (Table 4).
FIG 3.
Reverse transcriptase PCR (RT-PCR) detection of expression of adhE-citreum, glk, and 16S rRNA gene in L. citreum NRIC 1776T and F192-5. Lanes: M, Hyper ladder 1 kb (Bioline, London, UK); 1, genomic DNA of NRIC 1776T; 2, cDNA of NRIC 1776T; 3, genomic DNA of F192-5; 4, cDNA of F192-5; N, negative control.
TABLE 4.
Enzyme activities of ADH, ALDH, and NADH oxidase
| Strain | Activity (mU/mg protein) |
||
|---|---|---|---|
| ADH | ALDH | NADH oxidase | |
| NRIC 1776T | 214 | 4 | 60 |
| F192-5 | 0 | 0 | 8,137 |
| 31-11 | 468 | 3 | 496 |
| 2-1 | 0 | 0 | 1,285 |
Transformation of strain F192-5 and characterization of the derivatives.
Leuconostoc citreum F192-5 was successfully transformed with pSJE-adhEpro, and a recombinant strain (strain 31-11) was obtained. Sequencing analysis of the inserted promoter region (PslpA) in pSJE-adhEpro of the recombinant strain revealed single nucleotide mutation compared with the original PslpA sequence in pIGM1. This mutation is located in the upstream −35 region of the dual tandem −35 and −10 regions (Fig. 4b; see also Fig. S2). The mutated region was not recognized as a promoter by promoter prediction. Sequences of adhE-mesenteroides in strain 31-11 were identical to those of adhE-mesenteroides in Leuconostoc mesenteroides (data not shown). Strain 2-1 was also obtained by transformation of L. citreum F192-5 with pSJE and used as a control strain.
FIG 4.
Construction of a vector (a) and features of PslpA (b). The arrow in panel b indicates a mutation point in pSJE-adhEpro.
Expression of adhE-mesenteroides was recorded from strain 31-11 but not from strains F192-5 and 2-1 (Fig. 5). All strains studied expressed reference genes glk and 16S rRNA. None of the strains expressed the adhE-citreum gene. Strong ADH activities and relatively weak ALDH activities were recorded in strain 31-11 but not in strains F192-5 and 2-1 (Table 4).
FIG 5.
RT-PCR detection of expression of adhE-mesenteroides, adhE-citreum, glk, and 16S rRNA gene in L. citreum F192-5, 31-11, and 2-1. Lanes: M, Hyper ladder 1 kb; 1, genomic DNA of NRIC 1776T; 2, cDNA of F192-5; 3, cDNA of 31-11; 4, cDNA of 2-1; N, negative control.
The adhE-mesenteroides-expressing strain 31-11 grew rapidly on d-glucose and reached stationary phase after 1 day (Fig. 6a), whereas strain 2-1 grew poorly on d-glucose. Strain 31-11 produced large colonies (1 to 2 mm in diameter) after 48 h on GYP agar under anaerobic conditions, but strain 2-1 produced pinpoint colonies (approximately 0.1 to 0.2 mm in diameter) (Fig. 6b). Strain 31-11 produced lactate and ethanol at almost equimolar ratios, but strain 2-1 mainly produced lactate and acetate with trace ethanol (Table 1).
FIG 6.
Growth characteristics of L. citreum 31-11 (○) and 2-1 (Δ) in GYP broth (a) and on GYP agar under anaerobic conditions (b). In panel a, lines and error bars indicate averages and standard deviations from triplicates, respectively.
DISCUSSION
LAB are known to adapt to their habitats. Adaptation on a genomic level usually occurs at a species level or a niche-associated group level within species (12–14). Previous studies have shown that niche-specific fructophilic evolutions are common at the genus level, e.g., the genus Fructobacillus, or at species level, e.g., Lactobacillus kunkeei and Lactobacillus apinorum (4–6).
Strain F192-5 grows poorly on d-glucose but well on d-fructose. Acetate is mainly produced with lactate from d-glucose metabolism, but not ethanol. Supplementation of pyruvate or aerobic culturing on d-glucose dramatically enhanced the growth of strain F192-5. These characteristics differentiated the strain from the reference strains of L. citreum but were consistent with the characteristics of obligate FLAB (2, 8, 15, 16). On the other hand, strain F192-5 metabolized 15 carbohydrates, separating it from obligate FLAB that usually metabolize two to four carbohydrates (2, 16). Obligate FLAB species are known to have poor gene repertoires for class of carbohydrate transport and metabolism (ranging from 61 to 79 genes) compared to their phylogenetic relatives (4, 6). The gene class of FLAB ranks 8th or 9th in the 21 Cluster of Orthologous Groups (COG) functional classes (3). Strain F192-5 has 171 genes in the class, and the class ranks 3rd in the 21 classes (see Table S1 in the supplemental material). FLAB have poor carbohydrate transport systems, and they especially lack a complete PTS (3). Strain F192-5, on the other hand, has six complete PTS transporters encoded by its genome. Specific reduction of genomic size and total CDSs, previously reported in FLAB (4, 6), was not seen in the genome of F192-5. These characteristics indicate that strain F192-5 is a phenotypically FLAB-like strain but is genetically distinct from other FLAB.
Orthologous clustering is frequently used for the identification of specific genes in specific bacterial species or strains (17, 18). We found 16 to 175 strain-specific genes in L. citreum strains, which accounted for 1% to 9% of total CDSs in each strain. The ratio of specific genes seemed to be too large within a single species. In fact, the numbers of strain-specific genes dramatically reduced after manual BLASTP analyses (Table 3). This discrepancy may be due to more stringent criteria of get-homologues, which depends on bidirectional BLAST hits as evidence for orthologues. Strain F192-5 has the highest number of strain-specific genes of all strains included in the present study. Seven of the strain-specific genes are conserved in Fructobacillus spp. L. citreum LBAE_C10, originating from sourdough, has the same number of specific genes as strain F192-5 (Table 3), but none of the genes were shared with Fructobacillus spp. Concluded from these findings, the same set of genes was stored for survival in fructose-rich environments in Fructobacillus spp. and strain F192-5.
The gene encoding bifunctional ADH/ALDH (adhE) plays a key role in the phosphoketolase pathway of heterofermentative LAB. This gene is either absent in obligate FLAB or is partially present (4–6, 19). Transformation of the adhE gene into F. fructosus NRIC 1058T resulted in a remarkable change of metabolic properties, i.e., there was no requirement for the external electron acceptors during the metabolism of d-glucose (7). Furthermore, ethanol instead of acetate was produced from d-glucose (7). The key gene, i.e., adhE, was conserved in all L. citreum strains tested, including strain F192-5. However, strain F192-5 did not grow on d-glucose without electron acceptors. A detailed study of the genome sequence revealed a 167- to 168-bp gap upstream of the adhE gene on the genome of F192-5 compared with reference strains of L. citreum. A single promoter was predicted upstream of the adhE gene of the reference strains but not in the region of F192-5. This would explain why strain F192-5 did not express its adhE. Introduction of the adhE-mesenteroides gene led to significant phenotypic changes, and strain 31-11, transformed with the adhE gene, did not require the electron acceptors for d-glucose metabolism. These results clearly indicated that deletion of the promoter region of adhE is the reason for the fructophile-like metabolic characteristics of strain F192-5.
Toxicity of AdhE was reported in Escherichia coli (20, 21). In the former study, the terminator region of pSJE-adhEprot2 was not practically removed in F. fructosus for transformation of adhE, and a leaky promoter activity assisted the cells in producing acceptable levels of AdhE (7). Interestingly, the recombinant strain 31-11 contained a single nucleotide mutation in one of the dual tandem −35 and −10 regions of the inserted promoter sequences (PslpA), and the mutated region was not predicted as a promoter. This would attenuate promoter activity. This attenuation would repress excess production of AdhE, resulting in a successful production of the recombinant.
Conclusion.
A novel unique L. citreum strain, F192-5, was isolated from the peel of a satsuma mandarin. The strain shows fructophile-like growth characteristics. Unlike other FLAB, strain F192-5 possesses phenotypically and genetically rich carbohydrate metabolic systems, and the genome of strain F192-5 remained large; the size is comparable to the genomes of nonfructophilic L. citreum strains. Furthermore, unlike obligate FLAB, the adhE gene of strain F192-5 is intact. However, a large gap upstream of the adhE gene with no promoter was observed in the genome of strain F192-5, explaining the inability to ferment d-glucose. These results do not classify the strain as obligate FLAB, and the strain could be described as pseudo-FLAB. d-Glucose metabolism is a key for energy production in most organisms, including LAB, and a strain-specific deficiency of glucose metabolism is unique in LAB. This strain-specific unique phenotype might be one of the strategies to survive in diverse environments of LAB. All heterofermentative LAB have a possibility to have a pseudofructophilic phenotype by inactivating the adhE during adaptation to fructose-rich niches. This adaptation seems to be strain specific.
MATERIALS AND METHODS
Isolation.
A satsuma mandarin was purchased at a supermarket in Stellenbosch, South Africa. Fruit peels were collected in sterilized plastic bags and crushed. Five milliliters of FYP broth was added to the crushed samples and incubated at 30°C for 24 h. FYP broth was composed of (per liter) 10 g d-fructose, 10 g yeast extract, 5 g polypeptone, 2 g sodium acetate, 0.5 g Tween 80, 0.2 g MgSO4·7H2O, 0.01 g MnSO4·4H2O, 0.01 g FeSO4·7H2O, 0.01 g NaCl, 0.05 g cycloheximide, and 0.05 g sodium azide (pH 6.8). After incubation, a loop full of sample was inoculated in 30% FYP broth and incubated aerobically on an orbital shaker (120 rpm) at 30°C until growth was visible. The 30% FYP broth was FYP broth supplemented with 300 g liter−1 d-fructose and was used for selective isolation of FLAB (2). The culture was streaked onto FYP agar containing (per liter) 5 g CaCO3 and 12 g agar. Plates were incubated at 30°C under aerobic conditions. After 48 h of incubation, ten colonies were selected from the plates based on the sizes of the colony and clear zone surrounding the colonies and transferred to GYP and FYP broth, respectively. Incubation was conducted for 24 h at 30°C without aeration. GYP broth differed from FYP broth by containing 10 g liter−1 of d-glucose instead of d-fructose. The isolate was stocked in MRS broth containing 20% (vol/vol) glycerol at −80°C.
Reference strains.
Leuconostoc citreum NRIC 1776T, NRIC 1772, NRIC 1579, and NRIC 1581 were obtained from NODAI Culture Collection Center, Tokyo University of Agriculture, and used as reference strains for biochemical characterization (Table 5). The strains were cultured in MRS broth at 30°C for 24 h.
TABLE 5.
Bacterial strains and plasmids used in the present study
| Bacterial strain or plasmid | Description | Source or reference |
|---|---|---|
| Strains | ||
| Escherichia coli DH5α-pSJE-adhEprot2 | Donor of pSJE-adhEprot2 | 7 |
| Leuconostoc citreum NRIC 1776T | Reference strain | NRICa |
| Leuconostoc citreum NRIC 1772 | Reference strain | NRIC |
| Leuconostoc citreum NRIC 1579 | Reference strain | NRIC |
| Leuconostoc citreum NRIC 1581 | Reference strain | NRIC |
| Leuconostoc citreum F192-5 | Wild type (fructophile-like isolate) | This study |
| Leuconostoc citreum 31-11 | Derivative of F192-5 possessing adhE-mesenteroides | This study |
| Leuconostoc citreum 2-1 | Derivative of F192-5 containing pSJE | This study |
| Plasmids | ||
| pSJE | Emrb; E. coli-Leuconostoc shuttle vector | 29 |
| pSJE-adhEprot2 | pSJE harboring adhE-mesenteroides and promoter region of pIGMt (removed pSJE-originated NotI site) | 7 |
| pSJE-adhEpro | pSJE-adhEprot2 derivative with terminator removed | This study |
NODAI Research Institute Culture Collection, Tokyo, Japan.
Emr, erythromycin resistant.
Identification and biochemical characterization.
Identification of the isolate was performed by phylogenetic analysis based on the 16S rRNA gene sequence. Almost full-length 16S rRNA gene sequences (1,542 bp) were determined by a method described previously (22). BLAST analysis was performed for species identification.
Carbohydrate fermentation reactions were recorded using API CHL galleries (bioMérieux, Marcy-l'Etoile, France), according to the manufacturer’s instructions. Readings were taken for 7 days at 30°C. Growth characteristics on d-glucose and d-fructose and the requirement of external electron acceptors for d-glucose dissimilation were determined in GYP broth, FYP broth, and GYP broth supplemented with 10 g liter−1 pyruvate (GYP-P). Oxygen utilization as an electron acceptor was determined in GYP broth under aerobic conditions on a shaker (170 rpm). Incubation was at 30°C for 5 days. Growth was monitored at 660 nm with a spectrophotometer (model U-2800A; Hitachi, Tokyo, Japan). Anaerobic growth on d-glucose was determined by streaking cultures onto GYP agar incubated in an anaerobic workstation (10% H2, 10% CO2, and 80% N2, Bugbox chamber; Baker Ruskinn, Sanford, ME, USA). Production of lactate, acetate, and ethanol from d-glucose metabolism was determined after 2 days of incubation at 30°C in GYP broth using the F-kit (Roche Diagnostics, Mannheim, Germany). All experiments were performed in triplicates, unless indicated otherwise.
Draft genome sequencing and de novo assembly.
Whole-genome sequencing was conducted by Illumina MiSeq, with insert lengths of approximately 500 bp. A total of 5,502,480 and 6,743,618 reads with average lengths of 294 and 275 bp were obtained from L. citreum F192-5 and NRIC 1776T, respectively. De novo assembly, using the Velvet Assembler for short reads with parameters optimized by the VelvetOptimiser (version 1.2.10), resulted in 42 contigs (length, 2,061,777 bp; N50, 161,012 bp) and 8 contigs (length, 1,806,779; N50, 1,253,651 bp) for strains F192-5 and NRIC 1776T, respectively. Sequences shorter than 300 bp were eliminated. The genome was annotated using the DDBJ Fast Annotation and Submission Tool (DFAST; https://dfast.nig.ac.jp) (23), which is based on the annotation pipeline Prokka (24) and the curated reference database tailored for lactic acid bacteria. The sequence coverages of L. citreum F192-5 and NRIC 1776T were 785- and 1,028-fold, respectively.
Genomic data of other L. citreum strains and genome analysis.
The draft genome and complete genome sequences of eight strains of L. citreum, which were all genomic data of L. citreum strains available in the GenBank or RefSeq database at the time of analysis (May 2017), were obtained and were included in the comparative genomic study (Table 2). The completeness and contamination of the genomic data were assessed by CheckM (version 1.0.4) (25), by which the existence of gene markers specific to Leuconostoc spp. was inspected. Genomic data of L. citreum 1300_LCIT (NZ_JVUV00000000.1), which were available at the GenBank database, were excluded in the analysis, since the average nucleotide identity (ANI) analysis clearly separated this strain from L. citreum and classified it as Leuconostoc lactis.
To estimate the size of conserved genes in L. citreum, all protein sequences were grouped into orthologous clusters by GET_HOMOLOGUES software (version 1.3) based on the all-against-all bidirectional BLAST alignment and the MCL graph-based algorithm. Conserved genes were defined as gene clusters present in all analyzed genomes (26). This analysis was also used to identify strain-specific genes (nonorthologs). For functional comparison of the gene contents among L. citreum strains, coding sequences (CDSs) predicted in each strain were assigned to Cluster of Orthologous Groups (COG) functional classification using the COGNITOR software (27). Prediction of promoter sequences was conducted by using Berkeley Drosophila Genome Project (http://www.fruitfly.org/seq_tools/promoter.html) (28).
DNA manipulation, recombination, and transformation.
Escherichia coli strain DH5α possessing the pSJE-adhEprot2 plasmid was cultured at 37°C in brain heart infusion broth (Oxoid, Thermo Fisher Scientific, Tokyo, Japan), supplemented with 200 μg/ml erythromycin. The pSJE-adhEprot2 (Table 5) was produced from the promoterless E. coli-Leuconostoc shuttle vector pSJE (29) by insertion of adhE of Leuconostoc mesenteroides (here, adhE-mesenteroides) and the modified slpA promoter of Lactobacillus brevis (Fig. 4) used in a previous study (7). The modified promoter contains dual tandem −35 and −10 regions. The pSJE-adhEprot2 plasmid was digested with NotI and ligated to itself to restore the promoter activity by removing the terminator (pSJE-adhEpro). L. citreum F192-5 was transformed with the plasmid. A Qiagen Plasmid MAXI kit (Qiagen, Hilden, Germany), restriction endonucleases (TaKaRa-bio), and a Ligation-Convenience kit (Nippon Gene, Tokyo, Japan) were used as specified by the manufacturers.
Preparation of competent cells and electrotransformation of L. citreum F192-5 were conducted based on the method used for Fructobacillus fructosus (7), with slight modifications. Concentrations of sucrose, glycine, and penicillin G in culture broth were changed to 100 mM, 1% (wt/vol), and 0.375 μg/ml, respectively. After electrotransformation, cells were immediately resuspended in 1 ml freshly prepared MRS broth containing 0.5 M sucrose, 1% (wt/vol) sodium pyruvate, and 0.1 M magnesium chloride and incubated statically for 1 h at 30°C. The transformed cells were plated onto MRS agar supplemented with 2 μg/ml erythromycin and incubated aerobically at 30°C until colonies developed.
Gene expression of adhE and enzyme assays.
Gene expression of autochthonous adhE (adhE-citreum) and/or adhE-mesenteroides was confirmed from 20-h-old cultured cells of Leuconostoc citreum NRIC 1776T, F192-5, and derivatives (strains 31-11 and 2-1) by reverse transcription of total RNA combined with a PCR with gene-specific primers, as described previously (6). Genes encoding 16S rRNA and glucokinase (glk) were used as references. Primers used are listed in Table 6.
TABLE 6.
Primers used in the present study
| Primer | Sequence (5′→3′) | Purpose |
|---|---|---|
| pSJE-adhE-BamHI-F2 | AAAGGATCCGTGCACTAATCTAAATATGTG | Amplification of adhE-mesenteroides without promoter |
| pSJE-XbaI-adhESD-R | AAATCTAGATTAAAACTGATCTTTCAACAATTGACG | |
| adhE-R100 | ATCAATACCTCAGTCATCTTC | Sequencing of PslpA |
| adhE-F350 | CAAGGTAAGGTAGAATTAGC | Sequencing of adhE-mesenteroides |
| adhE-R500 | ATAAACTGATTCTGAGGCGA | |
| adhE-F1100 | TGCATACCTTGTTCCTAAGA | |
| adhE-R1250 | TAATTCGAACAAAAGTACAT | |
| adhE-F1700 | CTTACTTACAAGACTTGCCT | |
| adhE-R1800 | AATCGCCTCGTCATCTTCTA | |
| adhE-F2000 | GTTGCTATTGTTGATCCAGA | |
| adhE-F2440 | GCACGCACAAAGATGCACTA | |
| pLPD4-1t | ATTACAAAGGGTTTAAGCAG | Sequencing of PslpA |
| pLPD4-2t | GAACAACCGGGCCTCGAC | Confirmation for deletion of terminator region |
| F192-5-adhE-EXP-F | TCACAGAACGCCAAGAAGCA | Amplification of adhE-citreum |
| F192-5-adhE-EXP-R | CATGATCCGGTACCAAGCGT | |
| Lcit_glk_F | TTAGACAACGATGCGAATGC | Amplification of glk |
| Lcit_glk_R | TCTTGTGCCAAGCGTACAAC | |
| 8F | AGAGTTTGATCATGGCTCAG | Amplification of 16S rRNA gene |
| 15R | AAGGAGGTGATCCAACCGCA |
ADH, ALDH, and NADH oxidase activities were determined in cells of L. citreum NRIC 1776T, F192-5, and the derivatives (31-11 and 2-1). Culturing of the cells, preparation of cell extracts, and enzyme assays were performed as described previously (7).
Accession number(s).
Newly determined sequence data have been deposited in the GenBank database under accession numbers BJJW01000000.1 and BJJX01000000.1.
Supplementary Material
ACKNOWLEDGMENTS
We thank Yuna Kanatani (Department of Food and Cosmetic Science, Tokyo University of Agriculture) for technical assistance in the study of enzyme activities.
The present study was supported by JSPS KAKENHI (grant number 26850054), the MEXT Program for the Strategic Research Foundation at Private Universities 2013 to 2017 (S1311017), and NIG-JOINT (2016 to 2019).
Computational analysis was performed on the NIG supercomputer at ROIS.
Footnotes
Supplemental material for this article may be found at https://doi.org/10.1128/AEM.01077-19.
REFERENCES
- 1.Ventura M, Turroni F, van Sinderen D. 2012. Probiogenomics as a tool to obtain genetic insights into adaptation of probiotic bacteria to the human gut. Bioeng Bugs 3:73–79. doi: 10.4161/bbug.18540. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Endo A, Futagawa-Endo Y, Dicks LM. 2009. Isolation and characterization of fructophilic lactic acid bacteria from fructose-rich niches. Syst Appl Microbiol 32:593–600. doi: 10.1016/j.syapm.2009.08.002. [DOI] [PubMed] [Google Scholar]
- 3.Endo A, Maeno S, Tanizawa Y, Kneifel W, Arita M, Dicks L, Salminen S. 2018. Fructophilic lactic acid bacteria, a unique group of fructose-fermenting microbes. Appl Environ Microbiol 84:e01290-18. doi: 10.1128/AEM.01290-18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Endo A, Tanizawa Y, Tanaka N, Maeno S, Kumar H, Shiwa Y, Okada S, Yoshikawa H, Dicks L, Nakagawa J, Arita M. 2015. Comparative genomics of Fructobacillus spp. and Leuconostoc spp. reveals niche-specific evolution of Fructobacillus spp. BMC Genomics 16:1117. doi: 10.1186/s12864-015-2339-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Maeno S, Dicks L, Nakagawa J, Endo A. 2017. Lactobacillus apinorum belongs to the fructophilic lactic acid bacteria. Biosci Microbiota Food Health 36:147–149. doi: 10.12938/bmfh.17-008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Maeno S, Tanizawa Y, Kanesaki Y, Kubota E, Kumar H, Dicks L, Salminen S, Nakagawa J, Arita M, Endo A. 2016. Genomic characterization of a fructophilic bee symbiont Lactobacillus kunkeei reveals its niche-specific adaptation. Syst Appl Microbiol 39:516–526. doi: 10.1016/j.syapm.2016.09.006. [DOI] [PubMed] [Google Scholar]
- 7.Maeno S, Kajikawa A, Dicks L, Endo A. 2019. Introduction of bifunctional alcohol/acetaldehyde dehydrogenase gene (adhE) in Fructobacillus fructosus settled its fructophilic characteristics. Res Microbiol 170:35–42. doi: 10.1016/j.resmic.2018.09.004. [DOI] [PubMed] [Google Scholar]
- 8.Endo A, Okada S. 2008. Reclassification of the genus Leuconostoc and proposals of Fructobacillus fructosus gen. nov., comb. nov., Fructobacillus durionis comb. nov., Fructobacillus ficulneus comb. nov. and Fructobacillus pseudoficulneus comb. nov. Int J Syst Evol Microbiol 58:2195–2205. doi: 10.1099/ijs.0.65609-0. [DOI] [PubMed] [Google Scholar]
- 9.Nieminen TT, Sade E, Endo A, Johansson P, Bjorkroth J. 2014. The family Leuconostocaceae, p 215–240. In Rosenberg E, DeLong EF, Lory S, Stackebrandt E, Thompson F (ed), The Prokaryotes. Firmicutes and Tenericutes. Springer-Verlag, Berlin, Germany. [Google Scholar]
- 10.Aakko J, Endo A, Mangani C, Maleta K, Ashorn P, Isolauri E, Salminen S. 2015. Distinctive intestinal Lactobacillus communities in 6-month-old infants from rural Malawi and southwestern Finland. J Pediatr Gastroenterol Nutr 61:641–648. doi: 10.1097/MPG.0000000000000878. [DOI] [PubMed] [Google Scholar]
- 11.Endo A, Futagawa-Endo Y, Dicks LM. 2010. Diversity of Lactobacillus and Bifidobacterium in feces of herbivores, omnivores and carnivores. Anaerobe 16:590–596. doi: 10.1016/j.anaerobe.2010.10.005. [DOI] [PubMed] [Google Scholar]
- 12.Hols P, Hancy F, Fontaine L, Grossiord B, Prozzi D, Leblond-Bourget N, Decaris B, Bolotin A, Delorme C, Dusko Ehrlich S, Guedon E, Monnet V, Renault P, Kleerebezem M. 2005. New insights in the molecular biology and physiology of Streptococcus thermophilus revealed by comparative genomics. FEMS Microbiol Rev 29:435–463. doi: 10.1016/j.fmrre.2005.04.008. [DOI] [PubMed] [Google Scholar]
- 13.Vogel RF, Pavlovic M, Ehrmann MA, Wiezer A, Liesegang H, Offschanka S, Voget S, Angelov A, Bocker G, Liebl W. 2011. Genomic analysis reveals Lactobacillus sanfranciscensis as stable element in traditional sourdoughs. Microb Cell Fact 10(Suppl 1):S6. doi: 10.1186/1475-2859-10-S1-S6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Zhao X, Ganzle MG. 2018. Genetic and phenotypic analysis of carbohydrate metabolism and transport in Lactobacillus reuteri. Int J Food Microbiol 272:12–21. doi: 10.1016/j.ijfoodmicro.2018.02.021. [DOI] [PubMed] [Google Scholar]
- 15.Endo A, Irisawa T, Futagawa-Endo Y, Sonomoto K, Itoh K, Takano K, Okada S, Dicks LM. 2011. Fructobacillus tropaeoli sp. nov., a fructophilic lactic acid bacterium isolated from a flower. Int J Syst Evol Microbiol 61:898–902. doi: 10.1099/ijs.0.023838-0. [DOI] [PubMed] [Google Scholar]
- 16.Endo A, Irisawa T, Futagawa-Endo Y, Takano K, Du Toit M, Okada S, Dicks LM. 2012. Characterization and emended description of Lactobacillus kunkeei as a fructophilic lactic acid bacterium. Int J Syst Evol Microbiol 62:500–504. doi: 10.1099/ijs.0.031054-0. [DOI] [PubMed] [Google Scholar]
- 17.Zhang W, Ji H, Zhang D, Liu H, Wang S, Wang J, Wang Y. 2018. Complete genome sequencing of Lactobacillus plantarum ZLP001, a potential probiotic that enhances intestinal epithelial barrier function and defense against pathogens in pigs. Front Physiol 9:1689. doi: 10.3389/fphys.2018.01689. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.El Kafsi H, Binesse J, Loux V, Buratti J, Boudebbouze S, Dervyn R, Kennedy S, Galleron N, Quinquis B, Batto JM, Moumen B, Maguin E, van de Guchte M. 2014. Lactobacillus delbrueckii ssp. lactis and ssp. bulgaricus: a chronicle of evolution in action. BMC Genomics 15:407. doi: 10.1186/1471-2164-15-407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Endo A, Tanaka N, Oikawa Y, Okada S, Dicks L. 2014. Fructophilic characteristics of Fructobacillus spp. may be due to the absence of an alcohol/acetaldehyde dehydrogenase gene (adhE). Curr Microbiol 68:531–535. doi: 10.1007/s00284-013-0506-3. [DOI] [PubMed] [Google Scholar]
- 20.Membrillo-Hernandez J, Echave P, Cabiscol E, Tamarit J, Ros J, Lin EC. 2000. Evolution of the adhE gene product of Escherichia coli from a functional reductase to a dehydrogenase. Genetic and biochemical studies of the mutant proteins. J Biol Chem 275:33869–33875. doi: 10.1074/jbc.M005464200. [DOI] [PubMed] [Google Scholar]
- 21.Koo OK, Jeong DW, Lee JM, Kim MJ, Lee JH, Chang HC, Kim JH, Lee HJ. 2005. Cloning and characterization of the bifunctional alcohol/acetaldehyde dehydrogenase gene (adhE) in Leuconostoc mesenteroides isolated from kimchi. Biotechnol Lett 27:505–510. doi: 10.1007/s10529-005-2541-z. [DOI] [PubMed] [Google Scholar]
- 22.Endo A, Okada S. 2005. Lactobacillus satsumensis sp. nov., isolated from mashes of shochu, a traditional Japanese distilled spirit made from fermented rice and other starchy materials. Int J Syst Evol Microbiol 55:83–85. doi: 10.1099/ijs.0.63248-0. [DOI] [PubMed] [Google Scholar]
- 23.Tanizawa Y, Fujisawa T, Kaminuma E, Nakamura Y, Arita M. 2016. DFAST and DAGA: web-based integrated genome annotation tools and resources. Biosci Microbiota Food Health 35:173–184. doi: 10.12938/bmfh.16-003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Seemann T. 2014. Prokka: rapid prokaryotic genome annotation. Bioinformatics 30:2068–2069. doi: 10.1093/bioinformatics/btu153. [DOI] [PubMed] [Google Scholar]
- 25.Parks DH, Imelfort M, Skennerton CT, Hugenholtz P, Tyson GW. 2015. CheckM: assessing the quality of microbial genomes recovered from isolates, single cells, and metagenomes. Genome Res 25:1043–1055. doi: 10.1101/gr.186072.114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Contreras-Moreira B, Vinuesa P. 2013. GET_HOMOLOGUES, a versatile software package for scalable and robust microbial pangenome analysis. Appl Environ Microbiol 79:7696–7701. doi: 10.1128/AEM.02411-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Tatusov RL, Galperin MY, Natale DA, Koonin EV. 2000. The COG database: a tool for genome-scale analysis of protein functions and evolution. Nucleic Acids Res 28:33–36. doi: 10.1093/nar/28.1.33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Reese MG. 2001. Application of a time-delay neural network to promoter annotation in the Drosophila melanogaster genome. Comput Chem 26:51–56. doi: 10.1016/S0097-8485(01)00099-7. [DOI] [PubMed] [Google Scholar]
- 29.Park JY, Jeong SJ, Lee AR, Park JY, Jeong WJ, Kim JH. 2007. Expression of alpha-galactosidase gene from Leuconostoc mesenteroides SY1 in Leuconostoc citreum. J Microbiol Biotechnol 17:2081–2084. [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.