Next Article in Journal
Structural and Kinetic Views of Molecular Chaperones in Multidomain Protein Folding
Next Article in Special Issue
RadA, a Key Gene of the Circadian Rhythm of Escherichia coli
Previous Article in Journal
An Efficient Aequorea victoria Green Fluorescent Protein for Stimulated Emission Depletion Super-Resolution Microscopy
Previous Article in Special Issue
Plasmidome of Listeria spp.—The repA-Family Business
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
IJMS
Volume 23
Issue 5
10.3390/ijms23052487
ijms-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

A Comparative Genomic and Safety Assessment of Six Lactiplantibacillus plantarum subsp. argentoratensis Strains Isolated from Spontaneously Fermented Greek Wheat Sourdoughs for Potential Biotechnological Application

by
Maria K. Syrokou
1,
Spiros Paramithiotis
1,
Eleftherios H. Drosinos
1,
Loulouda Bosnea
2 and
Marios Mataragas
2,*
1
Department of Food Science and Human Nutrition, Agricultural University of Athens, 75 Iera Odos St., 11855 Athens, Greece
2
Department of Dairy Research, Institute of Technology of Agricultural Products, Hellenic Agricultural Organization “DIMITRA”, 3 Ethnikis Antistaseos St., 45221 Ioannina, Greece
*
Author to whom correspondence should be addressed.
Int. J. Mol. Sci. 2022, 23(5), 2487; https://doi.org/10.3390/ijms23052487
Submission received: 6 January 2022 / Revised: 18 February 2022 / Accepted: 21 February 2022 / Published: 24 February 2022
(This article belongs to the Special Issue Microbial Comparative Genomics and Evolutionary Biology)
Browse Figures
Versions Notes

Abstract

:
The comparative genome analysis of six Lactiplantibacillus plantarum subsp. argentoratensis strains previously isolated from spontaneously fermented Greek wheat sourdoughs is presented. Genomic attributes related to food safety have been studied according to the European Food Safety Authority (EFSA) suggestions for the use of lactic acid bacteria (LAB) in the production of foods. Bioinformatic analysis revealed a complete set of genes for maltose, sucrose, glucose, and fructose fermentation; conversion of fructose to mannitol; folate and riboflavin biosynthesis; acetoin production; conversion of citrate to oxaloacetate; and the ability to produce antimicrobial compounds (plantaricins). Pathogenic factors were absent but some antibiotic resistance genes were detected. CRISPR and cas genes were present as well as various mobile genetic elements (MGEs) such as plasmids, prophages, and insertion sequences. The production of biogenic amines by these strains was not possible due to the absence of key genes in their genome except lysine decarboxylase associated with cadaverine; however, potential degradation of these substances was identified due to the presence of a blue copper oxidase precursor and a multicopper oxidase protein family. Finally, comparative genomics and pan-genome analysis showed genetic differences between the strains (e.g., variable pln locus), and it facilitated the identification of various phenotypic and probiotic-related properties.

1. Introduction

Lactic acid bacteria (LAB) are microorganisms that possess important technological characteristics and, therefore, are used in food fermentations as starters or even as adjunct cultures providing the desired quality and organoleptic properties and assuring the safety of the fermented products. Traditionally made fermented foods, i.e., without the addition of a commercial starter culture, constitute a significant source of isolating wild LAB strains comprising the indigenous flora of the products performing the spontaneous fermentation. Although LAB have been acknowledged as safe for human and animal consumption (Generally Recognized as Safe—GRAS by the Food and Drug Administration—FDA and Qualified Presumption of Safety—QPS by the European Food Safety Authority—EFSA) [1,2], infection is still possible—especially for immunocompromised individuals [3,4,5]. Thus, the safety assessment of microbial strains aimed to be used as potential food additives is always timely, and it is highly recommended for investigation [6].
Since they constitute the majority of the respective micro-community, LAB and yeasts play a critical role during sourdough fermentation. There are different types of sourdoughs, based on the applied fermentation and technological processes [7]. Type I sourdough, from which the six Lactiplantibacillus plantarum subsp. argentoratensis strains under study were isolated, is characterized by the dominance of Fructilactobacillus sanfranciscensis, Levilactobacillus brevis, Limosilactobacillus fermentum, L. plantarum, and Companilactobacillus paralimentarius [8]. In the latter work, Greek wheat sourdoughs collected from various geographical regions were analyzed by using culture-dependent and culture-independent molecular techniques. These six strains, initially identified as L. plantarum by 16S rRNA gene sequencing, were further selected for whole-genome sequencing based on their phenotypic and technological properties [9]. Taxonomic classification of the strains based on the sequenced genome revealed that the isolated colonies belong to the subspecies argentoratensis. Sequenced genomes of L. plantarum subsp. argentoratensis are not readily available, especially strains isolated from sourdoughs.
Nowadays, as the sequencing technology has advanced and the cost of genome sequencing has decreased, the use of bacterial whole-genome sequencing as a diagnostic tool is more feasible than a few years ago. When a genome is available completed or as a draft of high quality, its bioinformatic analysis offers an unprecedented way of investigating the biotechnological potential and safety of the isolated sequenced strains [10,11]. Recently, Carpi et al. (2022) [12] performed a detailed pan-genome analysis of L. plantarum using 127 complete genomes. Another three related works have been conducted using the draft genomes of L. plantarum strains [13,14,15]. However, the main limitation of those studies is that acquired antibiotic resistance (AR) genes were annotated using only the Comprehensive Antibiotic Resistance Database (CARD) and/or ResFinder databases, resulting in many cases without any hits or in only one hit. In those works, the authors concluded that AR genes were not detected in the L. plantarum genomes. Chokesajjawatee et al. (2020) [6] mentioned that the limitation of the AR genes search using the CARD and ResFinder databases for non-pathogenic bacteria should be marked because both databases are mainly focused on the AR genes of pathogenic bacteria and those of the non-pathogenic are usually not included, resulting in their inability to identify the AR genes in L. plantarum genomes at the default. On the contrary, the search of the AR genes using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database proved to be efficient for this purpose.
Consequently, in this study the genomes of six L. plantarum subsp. argentoratensis strains, isolated from spontaneously fermented Greek wheat sourdoughs, were bioinformatically analyzed and compared in relation to their biotechnological features and their safety for potential application to foods. In addition, there is no report on the bioinformatic analysis of the L. plantarum subsp. argentoratensis genomes, isolated from spontaneously fermented Greek sourdoughs, and from sourdoughs in general, considering the very limiting number of the available genomes of this subspecies in the NCBI database. The bioinformatic analysis of the genome of the L. plantarum subsp. argentoratensis strains performed in the current work will provide insights regarding their genomic and functional features and their potential use as a food additive.

2. Results and Discussion

2.1. Strain Relatedness

The complete (100%) sequence of the 16S rRNA gene was extracted from the sequenced genome of all strains using the ContEst16S webtool and its length was 1567 bp. It should be noted, however, that although the genomes were of high quality (number of contigs was far below the limit of 200, the sequence length of each contig was ≥500 bp, and the genome coverage was much higher than the threshold of 60–80× and the recommended 100×) they were not closed because it is well known that the genome of lactic acid bacteria contains repeat regions and other mobile elements that make it difficult to finish the genome using only short read sequences [16,17].
The 16S-based ID platform of the EzBioCloud server revealed a high sequence similarity with the following type strains: L. pentosus DSM 20314T (99.87%), L. paraplantarum DSM 10667T (99.80%), L. plantarum subsp. argentoratensis DSM 16365T (99.80%), and L. plantarum subsp. plantarum ATCC 14917T (99.80%). The results from the 16S-based ID showed additional strains with sequence similarity above the threshold of 98.70% (L. fabifermentans DSM 21115, 99.19%; L. daowaiensis 203-3, 99.17%; L. pingfangensis 382-1, 99.16%; L. garii FI11369, 99.10%; L. nangangensis 381-7, 99.07%; L. daoliensis 116-1A, 99.03%; L. herbarum TCF032-E4, 98.99%, L. xiangfangensis LMG 26013, 98.92%; and L. plajomi NB53, 98.79%) but these were not selected further for ANI comparisons because the results obtained from the TYGS webtool indicated very poor pairwise dDDH values (<23%) between these strains and the strains of the current study. The calculated ANI values between the genomes of this work and the closely related type strains (16S rRNA gene sequence similarity ≥99.80%) are displayed in Figure 1. Based on the threshold ANI value of 95–96% and those calculated between the genomes, the six microbial strains isolated from traditional Greek wheat sourdoughs were identified as L. plantarum subsp. argentoratensis. In addition, the results obtained from the TYGS platform on the pairwise comparisons of user genomes and type-strain genomes, restricted to the above-mentioned four type strains, confirmed that these strains belong to the species of L. plantarum subsp. argentoratensis (dDDH value >79%) (Table 1, Figure 2). According to the Qualified Presumption of Safety (QPS) list of EFSA [1] the species L. plantarum has received QPS status.
Some strains presented a high nucleotide sequence identity approximating one hundred percent, although they had been isolated from geographically distinct sites. This could be explained by the fact that orthoANI calculates only the similarity of orthologous fragments. For example, the strains LQC 2422 and LQC 2516 presented genetic differences, such as the length and the organization of the pln loci encoding for plantaricins, which discriminated the two strains (see Section 2.3 “Safety assessment”). A high similarity has been also reported between L. plantarum strains of different origin [12].

2.2. Phenotypic Characterization and Pan-Genome Analysis

2.2.1. Phenotypic Properties

Based on the phenotype prediction (Figure 3), the six strains were: Gram-positive, catalase-negative, non-motile and non-sporeforming bacilli; unable to grow at 42 °C and produce NH3 from arginine; and negative to indole production, H2S formation, urease activity, and oxidase test. These properties are very common in the genus of Lactobacillus [18]. Regarding the sugar fermentation profile of the isolates, these showed a positive pattern for maltose, melibiose, sorbitol, mannose, glucose, salicin, lactose, raffinose, esculin, sucrose, trehalose, and mannitol, but they were negative for cellobiose, xylose, myo-inositol, and arabinose fermentation (Figure 3). However, genome annotation showed that the strains harbored the respective genes (EC 2.7.1.205 and EC 3.2.1.86) for the utilization of cellobiose. These predictions are in line with the experimental phenotypic tests conducted on the isolates by Syrokou et al. (2020) [8] who found that the L. plantarum subsp. argentoratensis strains can ferment a wide range of carbohydrates, including the cellobiose. Moreover, genome annotation analysis showed that the strains were able to ferment fructose and convert fructose to mannitol, a well-known low-calorie sweetener. Also, the strains LQC 2320 and 2520 were able to catabolize rhamnose and reduce nitrate to nitrite. The latter is mediated by the nitrate reductase which under certain circumstances may lead to nitric oxide (NO) synthesis, a property of great importance during fermented sausage manufacturing. The L. plantarum AJ2 strain isolated from naturally fermented sausage possessed the ability to reduce the residual levels of nitrite and nitrate [19].
Alkaline phosphatase production by all strains and gas (CO2) production from glucose by the LQC 2485 strain was projected by the phypat + PGL predictor only, while the phypat showed a positive result toward acetate utilization (strains LQC 2441, LQC 2422, LQC 2516, and LQC 2485) and growth in 6.5% NaCl (strains LQC 2520 and LQC 2320) (Figure 3). However, positive results predicted only by the phypat should be evaluated experimentally. Weimann et al. (2016) [20] noted that the phypat predictor tends to give more false-positive results when draft genomes are used. Alkaline phosphatase production by L. plantarum has been linked with the increased degradation rate of the organophosphorus pesticides present in wheat dough or skimmed milk [21,22]. Other important phenotypic properties predicted by Traitar were the strains’ ability to utilize the citrate as a carbon source, which may lead to the formation of aroma precursor compounds; the strains’ ability to produce the enzyme beta-galactosidase (ONPG); and the positive reaction to the Voges–Proskauer (VP) test. Beta-galactosidases belong to glycoside hydrolase (GH) groups. Genome annotation revealed that all strains had the GH2 group of beta-galactosidase and especially the LacLM type encoded by two genes, lacL and lacM (EC 3.2.1.23) (the other type is the LacZ, encoded by the lacZ gene). While L. plantarum predominantly possess the LacLM type [23], the LacZ type is found in other LAB [24]. The enzyme hydrolyzes lactose and, therefore, it is used in the dairy industry for health-related applications like the production of lactose-free products or prebiotic galactooligosaccharides (GOS). The addition of beta-galactosidase-producing lactobacilli as probiotics to dairy products can assist lactose-intolerant individuals. The addition of GOS stimulates the growth of probiotics and regulates the gut microflora [25]. The positive VP test, which was negatively correlated with the methyl red test (the MR test was predicted as negative) (Figure 3), indicates the presence of acetoin, which is the precursor of 2,3 butanediol and other aromatic compounds. Thus, the strains do not use the mixed acid fermentation pathway, i.e., production of several organic acids from glucose utilization, but they use the metabolic pathway that leads to the formation of 2,3 butanediol, in addition to lactate production. Thakur et al. (2017) [18] found that a L. plantarum strain isolated from pickle samples displayed the opposite phenotype, i.e., MR+ and VP−. Usually, a bacterial culture is positive to only one pathway, either VP+ or MR+. Indeed, the results on the genome annotation of L. plantarum subsp. argentoratensis confirmed the presence of a complete set of genes for acetoin production and conversion of citrate to oxaloacetate. The latter coincided with the predictions made by Traitar. Finally, 15 GHs (glycoside hydrolases) and 9 GTs (glycosyltransferases) families contained in the CAZymes database were spotted, indicating a high variability in GH- and GT-encoding.

2.2.2. Pan-Genome Analysis and Annotation

According to Heaps’ law [26], the curve fitting of the pan-genome resulted in an estimated a value of 0.10 (a < 1) which means that when a new genome is added to the pan-genome, new genes are added as well increasing in this way its size (open pan-genome). The increase of pan-genome by 0.10 implies potential evolution-related changes in the L. plantarum subsp. argentoratensis genomes (horizontal gene transfer—HGT, gene gain or gene loss) to cope with the adaptation to various environmental conditions. The pan-genome of the six strains comprised of 3239 genes in total, from which 2454 genes were in the core-genome, 550 genes in the accessory-/unique-genome, and 235 genes were unique (Figure 4). The strain LQC 2485 harbored the highest number of unique genes (202) compared to the other strains. Hence, gene exchange activities associated with potentially different approaches of adaptation and response to environmental conditions are happening at the highest rate in this specific strain. For instance, the genome annotation performed revealed the presence of a multiple-sugar metabolism (msm) gene cluster analogous to the described system found in Streptococcus mutans [27,28]. This multiple-sugar ABC transporter were comprised of two permease proteins (msmG and msmF) and a substrate-binding protein (msmE). Another feature found in the unique-genome of the LQC 2485 strain is the presence of a pyruvate dehydrogenase (quinone) (EC 1.2.5.1), which stimulates the conversion of pyruvate to acetate.
The hierarchical clustering of the gene matrix (presence/absence) (Figure 5), acquired during the pan-genome analysis of the L. plantarum subsp. argentoratensis genomes, was like the clustering obtained based on the WGS (Figure 2). However, the LQC 2485 was quite distinct from the other two groups probably due to the high number of unique genes harbored by this strain. The LQC 2320 and LQC 2520 isolates formed a separate group from the rest of the strains (LQC 2422, LQC 2441, and LQC 2516) attributed to the presence/absence of a group of genes in each microbial cluster (Figure 5). Thus, HGT, gene gain, or gene loss phenomena potentially have occurred between the L. plantarum subsp. argentoratensis strains and other microbial clades. Indeed, the LQC 2320 and LQC 2520 strains possessed a genes cluster related to the reduction of nitrate to nitrite encountered in Staphylococcus carnosus [29] (nitrate/nitrite transporter, NarT; oxygen-sensing two component system sensor histidine kinase, NreBC; and respiratory nitrate reductase alpha, beta, gamma, and delta chain; narGHIJ, EC 1.7.99.4), and the utilization of L-rhamnose (transcriptional regulator of rhamnose utilization, AraC family, L-rhamnose permease, RhaY, major facilitator superfamily—MFS; L-rhamnose mutarotase, RhaM, EC 5.1.3.32; L-rhamnose isomerase, RhaA, EC 5.3.1.14; L-rhamnulose kinase, RhaB, EC 2.7.1.5; and L-rhamnulose-1-phosphate aldolase, RhaD, EC 4.1.2.19); both features were predicted by the Traitar and annotated by the respective web-services. On the other hand, the LQC 2422, LQC 2441, LQC 2485, and LQC 2516 strains harbored the genes cluster associated with the maltodextrin-specific ABC transporter, i.e., the ATP-binding protein MdxE, the membrane-spanning components MdxF and MdxG, and the energizing ATPase MsmX [30], and the utilization of glycerate (phosphotransferase System—PTS, 2-O-alpha-mannosyl-D-glycerate-specific IIA, IIB, and IIC component, EC 2.7.1.195; glycerate kinase, EC 2.7.1.31; and mannosylglycerate hydrolase, EC 3.2.1.170). Annotation showed the presence of 2-dehydro-3-deoxy-D-gluconate-5-dehydrogenase (EC 1.1.1.127) and 2-deoxy-D-gluconate-3-dehydrogenase (EC 1.1.1.125) in the genome of the above four strains where the resulting compound enters the pentose phosphate pathway. Finally, all strains harbored a PTS-system, fructose-specific IIA, IIB, and IIC component (EC 2.7.1.202) and a PTS-system, mannose-specific IIA, IIB, IIC, and IID component (EC 2.7.1.191), but only these four strains had a PTS-system, fructose- and mannose- inducible IIA, IIB, IIC, IID, and IIE component. This phosphoenolpyruvate: fructose phosphotransferase system constitutes an alternative way of entering mannose into Streptococcus salivarius [31].
Gene assignment to the different COG functional groups depicted that the genes of translation, ribosomal structure and biogenesis (J), amino acid transport and metabolism (E), and nucleotide transport and metabolism (F) were enhanced in the core-genome of the L. plantarum subsp. argentoratensis species (Figure 6A), i.e., genes implicated in microbial housekeeping processes [33]. Genes related to cell wall/membrane/envelop biogenesis (M), DNA replication, recombination and repair (L), and carbohydrate transport and metabolism (G) were enhanced in the accessory-/unique-genome, i.e., genes implicated in energy metabolism and DNA repair (Figure 6A) [33]. Moreover, the genes involved in carbohydrate metabolism and membrane transport were the most abundant in the L. plantarum subsp. argentoratensis strains confirming the capacity to metabolize different carbohydrates, a feature which could be strain-dependent (higher abundance of genes in the accessory-/unique-genome compared to the core-genome) (Figure 6B). For example, as discussed above, the Traitar platform predicted that the LQC 2320 and LQC 2520 strains ferment L-rhamnose, but the others do not have this ability. Another observation worth-mentioning is the relatively low gene abundance in the lipid transport and metabolism category (Figure 6), suggesting that the strains’ lipolytic activity is not high. Indeed, experimental results obtained by Syrokou et al. (2020) [8] on the lipolytic activity of the strains under study confirm: strains LQC 2422, LQC 2441 and LQC 2485 exhibited no lipolytic activity, strains LQC 2520 and LQC 2320 showed moderate activity (22.75 AU/mL and 37.25 AU/mL, respectively), and strain LQC 2516 showed low activity (14.50 AU/mL).
All strains displayed the presence of a complete set of genes for riboflavin and folate biosynthesis. Similarly, L. plantarum strains isolated from quinoa sourdough [34] or a traditional maize-based fermented beverage (chicha) [16] had this ability as well. All L. plantarum subsp. argentoratensis strains seem to produce the foldase protein PrsA encoded by the prtM gene, which is responsible for the cleavage of proteins and the subsequent utilization of the large polypeptides by the microbial cells. The Opp (oppA, oppA2, oppBCDF) and Dtp (dtpT) gene clusters identified in their genomes widen the peptide uptake mechanisms. These peptides are subsequently hydrolyzed by various peptidases and proteases such as oligoendopeptidases (pepE, pepF, pepF2, pepO, pepC, and pepN), proline-specific peptidases (pepX, pepI, pepQ, and pip), di- and tri- peptidases (pepV, pepT, and pipD), and other enzymes (map—methionine aminopeptidase; mccF—LD-carboxypeptidase; est—serine aminopeptidase; ydcK—SprT family; ddpX—hydrolysis of the D-alanyl-D-alanine dipeptide; vanY—D-alanyl-D-alanine carboxypeptidase; and ctpA, dacA, amd, gluP, lepB, trpG, ytzB, ymfH, ymfF, glbL, htpX, yvpB, peptidases belonging to different families).
Various probiotic-related genes were identified in the genomes of the six strains suggesting their potential probiotic properties [35,36]. These genes were associated with stress response and immunomodulation (uspA, dltA, dltB, dltC, dltD), salt-stress (nitrate/sulfonate/bicarbonate ABC transporter), acid-tolerance (gadB), DNA protection (ImpB/MucB/SamB and Dps family proteins, clpB, clpC, clpL, msrB, luxS), and adhesion ability (mucin-binding protein—MucBP domain, collagen binding domain protein which encodes an adhesion potentially related to colonization and competition against pathogens, LPxTG-motif cell wall anchor domain protein, fbpA, and groS). Adhesion could also be related to exopolysaccharides (EPS) or surface proteins. However, no cps cluster (cps1, cps2, cps3, cps4) was recognized; only the cps2J gene (accessory-genome) from cluster 2 and the cps4J gene (core-genome) from cluster 4. On the contrary, the surface protein Ef-Tu was identified, which acts as an adhesion factor. Such probiotic-related properties have been found in other L. plantarum strains such as ZJ316 [35] and microbes isolated from Indian fermented foods [36]. Finally, choloylglycine hydrolase family proteins (pva1, pva2, pva3, cbh; EC 3.5.1.24) referred to as bile salt hydrolases (BSH) were detected indicating bile adaptation. This observation confirms the prediction made by the Traitar platform on bile sensitivity of the strains which were negative. Although, bile salt resistance has been used as a criterion, among others, for the selection of a strain as probiotic, this activity could be potentially damaging to the human host and, therefore, the use of the BSH feature as a desirable property of the probiotic microorganism is controversial [6,37].

2.3. Safety Assessment

It is well known that bacteria may produce various biogenic amines (BAs) in foods, leading to health side-effects [38]. Therefore, the candidate microbial cultures should be previously assessed for their ability to produce such compounds before their application to fermented foods as starters because there is the risk of biogenic amines accumulation owing to LAB intensive metabolic activity. To identify the presence of the most important BAs-related genes in the genome of the six L. plantarum subsp. argentoratensis strains, the KEGG database through the egg-NOG mapper results was searched. Genes associated with tyramine, phenylethylamine, putrescine, spermidine, spermine, histamine, agmatine, and tryptamine production were not located in any of the queried genomes. The arcB gene (ornithine carbamoyl transferase) was present in the genome of the strains but the rest of the gene cluster (i.e., arcA, arcC, and arcD) implicated in the ornithine conversion (decarboxylation) to putrescine was lacking. Thus, the six strains can be considered as putrescine-negative producers. On the other hand, the cadA gene (lysine decarboxylase), which is associated with the production of cadaverine (CAD), was found in all strains under study. The cadBC genes involved in the regulation and transportation system of the CAD were not found. Nevertheless, the homologous pot gene cluster (potABCD) was available in all genomes, encoding for the putrescine/ornithine antiporter system, which can be used instead of the cadB antiporter system of CAD/lysine [38]. However, the concentration of CAD should be experimentally determined in order to evaluate the true risk of CAD accumulation. Moreover, the strains harbored some genetic properties related to BA degradation, such as the presence of a blue copper oxidase (CueO) precursor and a multicopper oxidase protein family.
The L. plantarum subsp. argentoratensis strains are able to produce D-lactate because of the presence of D-lactate dehydrogenase and LarA which is implicated in lactate racemization. Hence, the consumption of foods containing LAB with D-lactate producing capabilities may cause problems for people with a high risk of D-lactate acidosis [6], and a precaution is needed. The screening against the CARD and PathogenFinder databases resulted in no identification of antibiotic resistance- or virulence- related genes, respectively. The microorganisms were not predicted as human pathogens since the strains’ probability of being a human pathogen was low, ranging from 4.6 to 8.4%. Chokesajjawatee et al. (2020) [6] noted that one limitation of the CARD database is its focus on the AMR of pathogenic microorganisms, while the AMR-related elements of non-pathogenic microbes are not considered; and, therefore, the search for any antibiotic resistance property against the KEGG database is recommended. Indeed, the results showed seven AMR-related elements (Table 2). However, further experimental investigation on these large families of antibiotics is needed because the possession of the respective genes does not necessarily warrant their resistance as well. For instance, Chokesajjawatee et al. (2020) [6] found that the presence of a macrolide resistance gene msrA and two beta-lactamase genes (penP spotted on two different locations) in the genome of L. plantarum BCC 9546 did not warrant resistance to erythromycin and ampicillin, respectively. According to the authors, this could be attributed to several factors such as the gene expression level and the substrate specificity of the expressed product. Similarly, the identification of hemolysin III protein, the product of the hly gene, is another example [6]. All of these undesirable genes (Table 2) located in the genome of the six strains are commonly encountered in several L. plantarum strains including the reference strain WCFS1 and various probiotics (JDM1, ST-III, and 299V) [6]. Such observations are suggesting the widespread presence of the related genes within the species without jeopardizing, however, their safety level compared to the existing starter/probiotic cultures. The isolated bacteria were bioinformatically examined for the production of antimicrobial drugs, particularly those deemed as clinically important. Based on the performed KEGG database search, no strain displayed the ability to produce any of the concerned antimicrobials.
Another significant factor for the safety assessment of the candidate starters/probiotics is their genome stability. The presence of various mobile genetic elements (MGEs) such as plasmids, insertion sequences (ISs), and prophages may act as vehicles of the horizontal transfer of virulence and/or AMR genes. The conducted analysis returned no virulence or AMR genes within the intact prophage regions (all strains contained two complete phage regions except LQC 2485), plasmids (at least one plasmid was identified; rep28 and repUS73 in LQC 2441, LQC 2485, LQC 2422, and LQC 2516; rep28 in LQC 2320 and LQC 2520), and IS (at least one IS was identified except of strain LQC 2485 in which no IS was located; ISLsa1—IS30 family in LQC 2441 and LQC 2422; ISP1—ISL3 family in LQC 2520; ISP2—IS1182 family and ISLsa1—IS30 family in LQC 2516; ISP1—ISL3 family and ISLpl1—IS30 family in LQC 2320). Similar MGEs have been frequently found in other L. plantarum strains [6,16,39] that pose no safety risk. The L. plantarum subsp. argentoratensis genomes contained two complete phage regions (intact) and some phage remnants (incomplete). Briefly, LQC 2320 and LQC 2520 had two intact regions of 26.2 Kb (total proteins, 29; phage proteins, 27; hypothetical proteins, 2; tRNA, 0; phage-related keywords found in protein names in the region, terminase, portal, head, capsid, tail, plate, and lysin; putative phage attachment site, no; most common phage, PHAGE_Lactob_jlb1; GC, 43.84%) and 52 Kb (total proteins, 71; phage proteins, 47; hypothetical proteins, 24; tRNA, 6; phage-related keywords found in protein names in the region, lysin, tail, head, portal, terminase, and integrase; putative phage attachment site, yes; most common phage, PHAGE_Lactob_Lj965; GC, 41.63%). The LQC 2485 possessed one intact region of 106.7 Kb (total proteins, 120; phage proteins, 91; hypothetical proteins, 29; tRNA, 6; phage-related keywords found in protein names in the region, integrase, lysin, terminase, portal, head, tail, protease, and capsid; putative phage attachment site, yes; most common phage, PHAGE_Lactob_Sha1; GC, 42.18%). Finally, PHASTER identified two intact regions for the strains LCQ 2422, LQC 2441, and LQC 2516 of 50.4 Kb (total proteins, 51; phage proteins, 45; hypothetical proteins, 6; tRNA, 2; phage-related keywords found in protein names in the region, integrase, tail, terminase, head, portal, protease, capsid, and lysin; putative phage attachment site, yes; most common phage, PHAGE_Lactob_Sha1; GC, 42.65%) and 51.3 Kb (total proteins, 65; phage proteins, 46; hypothetical proteins, 19; tRNA, 5; phage-related keywords found in protein names in the region, lysin, tail, head, portal, terminase, and integrase; putative phage attachment site, yes; most common phage, PHAGE_Lactob_Lj965; GC, 41.69%).
The results from the screening of CRISPR sequences are presented in Table 3. Each genome had four Class 1 CRISPR systems, except strain LQC 2485 that had five with different evidence levels (1, 3 or 4). In all cases, only two CRISPR regions of evidence level 4 had one cas gene cluster per CRISPR array in vicinity, which provided immunity against foreign DNA. The TypeIE CRISPRcas system has been associated with the discrimination of self from non-self DNA [40]. This could be a means to avoid acquiring virulence and/or AMR genes through the horizontal gene transfer phenomenon [41]. The remaining CRISPR regions were of low evidence level or without cas genes nearby; and, therefore, these regions were not a real or a functional CRISPR array, respectively [16,42].
The screening of the six genomes with the BAGEL4 platform indicated a variable pln locus encoding for plantaricins and spanning in a region of 8.2 kb (LQC 2485, LQC 2441, and LQC 2422), 20.2 kb (LQC 2516), and 28.9 kb (LQC 2320 and LQC 2520) (Figure 7). The organization of the pln locus for the strains LQC 2485, LQC 2441, LQC 2422, and LQC 2516 resembled that of the L. plantarum 423 strain consisting of four genes, plaABCD [43,44]. Plantaricin 423 is a small pediocin-like plasmid-encoded protein that belongs to the class IIa bacteriocins with a double-glycine leader peptide and strong antilisterial activity. The strains LQC 2320 and LQC 2520 harbored genes encoding for three class IIb bacteriocins, plnJK (two-peptide plantaricin), plnEF (two-peptide plantaricin), and NC8βα (inducible plantaricin) [45]. The structure of the pln operon in these strains was found to be like that of L. plantarum strain NC8 [46,47]. More accurately, the presence of the operons plnJKLR, plnEFI, and plnC8αβc was verified; while, in the case of the transport operon, the presence of only three genes—plnG, plnH, and plnS—was evident, followed by plnY, which was absent in L. plantarum NC8. As far as the regulatory operon was concerned, the presence of three genes—plNC8-IF, plNC8-HK, and plnD—encoding an induction pheromone, a histidine protein kinase, and a response regulator, respectively, was observed. The presence of plnM was also reported. Regarding the plNC8-IF, it was indicated as a small open reading frame (ORF) labelled by BAGEL4 as sORF_12 (Figure 7C). Its sequence was blasted using the BLAST protein tool with the default settings and the results returned a plantaricin precursor peptide, induction factor (98%, L. plantarum).

3. Materials and Methods

3.1. Microbial Strains

The bacterial strains LQC 2320, LQC 2422, LQC 2441, LQC 2485, LQC 2516, and LQC 2520 of L. plantarum subsp. argentoratensis have been isolated from spontaneously fermented Greek wheat sourdoughs [8]. Information on the whole-genome sequencing of the strains and the subsequent bioinformatics processing of the raw reads (fastq files), up to genome assembly and annotation, can be found in the work of Syrokou et al. (2021) [9] as well as in the NCBI database (GenBank) under the accession numbers JAEQMR01, JAEQMQ01, JAEQMP01, JAEQMO01, JAEQMN01, and JAEQMM01, respectively.

3.2. Strains Relatedness

Overall genome-related index (OGRI) analysis [48] was performed according to Mataragas (2020) [49]. Species relatedness was examined by comparing the average nucleotide identity (ANI) and digital DNA-DNA hybridization (dDDH) values between the current strains and the type strains of L. paraplantarum DSM 10667T, L. plantarum subsp. plantarum ATCC 14917T, L. plantarum subsp. argentoratensis DSM 16365T, and L. pentosus DSM 20314T, downloaded from the GenBank database. The above phylogenetic neighbors were chosen following the workflow of genome-based classification at the species level proposed by Chun et al. (2018) [50]: first, the full length of the 16S rRNA gene sequence was extracted from the whole-genome sequences of each studied strain using the ContEst16S tool [51], and then the extracted sequences were loaded on the 16S-based ID platform of the EzBioCloud server [52] for identifying a bacterial isolate based on the 16S rRNA gene database of type strains. Bacterial identifications that displayed a high 16S sequence similarity value (≥98.7%) were further selected for OGRI analysis. The ANI values (threshold for species identification 95–96%) were calculated with the Orthologous Average Nucleotide Identity Tool (OAT) [53] while dDDH values (threshold for species and subspecies identification 70% and 79%, respectively) with the Type (Strain) Genome Server (TYGS) [54]. The TYGS was also used for establishing a bootstrapped genome-based phylogenetic relationship of the strains.

3.3. Phenotypic Characterization and Pan-Genome Analysis

Phenotype assignments to bacterial genomes of L. plantarum subsp. argentoratensis strains were achieved through the Traitar web service, accounting for 67 diverse traits [20]. Pan-genome analysis was conducted using the BPGA v1.3.0 software [55]. For the core-genome extraction, the default values of the tool were used; USEARCH software v10.0.240 [56] with a threshold of sequence identity equal to 0.5 (50%). The functional genes from the core-, accessory- and unique-genome of the L. plantarum subsp. argentoratensis strains were clustered to each COG and KEGG category with the USEARCH software, against the respective databases, embedded in the BPGA pipeline, applying the default parameters. Finally, the clusters of functional genes (pan-matrix), presented as a binary matrix (presence or absence), were introduced in the GENE-E software (https://software.broadinstitute.org/GENE-E/) (accessed on 6 September 2021) for visualizing the genes against the genomes. A hierarchical clustering (HC) of genes and genomes was performed using the option one minus the Pearson correlation distances available in the GENE-E. The web services PATRIC v3.6.8 [57], KofamKOALA [58], and eggNOG-mapper [59] were used to annotate the predicted proteins obtained during the pan-genome analysis of L. plantarum subsp. argentoratensis strains. The results of eggNOG-mapper also offered CAZymes analysis towards the CAZymes database [60].

3.4. Genomic Aspects Related to Food Safety

A safety assessment of the L. plantarum subsp. argentoratensis strains was performed by using whole-genome analysis and following the guidelines and recommendations of Chokesajjawatee et al. (2020) [6] and the EFSA FEEDAP Panel (2018) [10], respectively.
Antibiotic resistance and potential pathogenicity of the microbes were checked with the Resistance Gene Identifier (RGI) tool of the CARD [61] and PathogenFinder [62], respectively, following the analytical approaches of Rodrigo-Torres et al. (2019) [16]. Moreover, the egg-NOG mapper results based on the KEGG database was searched for undesirable genes as suggested by Chokesajjawatee et al. (2020) [6]. Bacteriocin-encoding genes were inspected using the BAGEL4 webtool [63]. Annotation results from PATRIC v3.6.8, KofamKOALA and eggNOG-mapper were screened for the ability of the strains to produce biogenic amines. Finally, the stability of the genomes was evaluated by investigating the following genetic features: (a) plasmids and insertion sequences (IS) were analyzed by PlasmidFinder [64] and MobileElementFinder [65], respectively. MobileElementFinder was also able to annotate acquired antimicrobial resistance genes using ResFinder [66,67] and virulence genes using VirulenceFinder [68]; (b) regions of Clustered Palindromic Interspaced Palindromic Repeats (CRISPR) with cas genes were detected with the CRISPRCasFinder webserver [42] using the default parameters; and (c) the presence of prophages was searched by means of the PHASTER webtool [69]. According to Chokesajjawatee et al. (2020) [6], the WHO CIA list (list of critically important antimicrobials of World Health Organization) served as a reference to evaluate the ability of the microbial strains to produce antimicrobial drugs of clinical importance via the KofamKOALA and egg-NOG-mapper results of the KEGG database.

4. Conclusions

This work deals with the comparative genomic analysis of six L. plantarum subsp. argentoratensis strains previously isolated from Greek wheat sourdoughs. The analysis showed that the lactic acid bacteria possessed properties with biotechnological/probiotic potential (e.g., as a starter culture in fermented foods). Their safety based on the EFSA recommendations was evaluated. Regarding biogenic amines production, the six strains encoded a lysine carboxylase gene, cadA, associated with cadaverine production but the real risk of such production and accumulation should be confirmed phenotypically. No virulence factors or production of antimicrobial drugs were detected in any of the queried genomes. Genes related to antibiotic resistance were found but the experimental validation of this resistance was not performed. Our aim was the bioinformatic characterization of the L. plantarum subsp. argentoratensis strains and the evaluation of their safety to have a reference study for any future validation experiment involving the current bacteria. Similar resistance antibiotic genes, however, are frequently encountered in strains of the same species including probiotics (WCFS1, JDM1, ST-III, and 299V) [6]. All microorganisms were able to produce plantaricins belonging to different classes. Finally, plasmids, prophage regions, insertion sequences, and CRISPRcas systems were identified in the six genomes. Consequently, the in-silico analysis performed and the genomic results obtained guarantee the safety of the microbial strains for food applications.

Author Contributions

Conceptualization, M.K.S., S.P. and M.M.; methodology, M.M.; software, M.M.; formal analysis, M.M.; investigation, M.K.S. and S.P.; resources, M.M.; data curation, M.M.; writing—original draft preparation, M.M.; writing—review and editing, M.K.S., S.P., E.H.D., L.B. and M.M.; supervision, S.P., E.H.D., L.B. and M.M.; project administration, S.P. and M.M.; funding acquisition, S.P. and M.M. All authors have read and agreed to the published version of the manuscript.

Funding

This research has been co-financed by the European Union and Greek national funds through the Operational Program Competitiveness, Entrepreneurship and Innovation, under the call RESEARCH—CREATE—INNOVATE (T1EDK-05339).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The genome sequences are available in the NCBI Bioproject ID PRJNA689714 [9].

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. EFSA. Introduction of a Qualified Presumption of Safety (QPS) approach for assessment of selected microorganisms referred to EFSA–Opinion of the Scientific Committee. EFSA J. 2007, 5, 587. [Google Scholar] [CrossRef]
  2. Leuschner, R.G.; Robinson, T.P.; Hugas, M.; Cocconcelli, P.S.; Richard-Forget, F.; Klein, G.; Licht, T.R.; Nguyen-The, C.; Querol, A.; Richardson, M. Qualified presumption of safety (QPS): A generic risk assessment approach for biological agents notified to the European Food Safety Authority (EFSA). Trends Food Sci. Technol. 2010, 21, 425–435. [Google Scholar] [CrossRef]
  3. Cannon, J.P.; Lee, T.A.; Bolanos, J.T.; Danziger, L.H. Pathogenic relevance of Lactobacillus: A retrospective review of over 200 cases. Eur. J. Clin. Microbiol. Infect. Dis. 2005, 24, 31–40. [Google Scholar] [CrossRef] [PubMed]
  4. Kayser, F.H. Safety aspects of enterococci from the medical point of view. Int. J. Food Microbiol. 2003, 88, 255–262. [Google Scholar] [CrossRef]
  5. Salminen, M.K.; Rautelin, H.; Tynkkynen, S.; Poussa, T.; Saxelin, M.; Valtonen, V.; Järvinen, A. Lactobacillus bacteremia, species identification, and antimicrobial susceptibility of 85 blood isolates. Clin. Infect. Dis. 2006, 42, e35–e44. [Google Scholar] [CrossRef]
  6. Chokesajjawatee, N.; Santiyanont, P.; Chantarasakha, K.; Kocharin, K.; Thammarongtham, C.; Lertampaiporn, S.; Vorapreeda, T.; Srisuk, T.; Wongsurawat, T.; Jenjaroenpun, P.; et al. Safety assessment of Nham starter culture Lactobacillus plantarum BCC9546 via whole-genome analysis. Sci. Rep. 2020, 10, 10241. [Google Scholar] [CrossRef] [PubMed]
  7. Siepmann, F.B.; Ripari, V.; Waszczynskyj, N.; Spier, M.R. Overview of sourdough technology: From production to marketing. Food Bioprocess Technol. 2018, 11, 242–270. [Google Scholar] [CrossRef]
  8. Syrokou, M.K.; Themeli, C.; Paramithiotis, S.; Mataragas, M.; Bosnea, L.; Argyri, A.A.; Chorianopoulos, N.G.; Skandamis, P.N.; Drosinos, E.H. Microbial ecology of Greek wheat sourdoughs identified by a culture-dependent and a culture-independent approach. Foods 2020, 9, 1603. [Google Scholar] [CrossRef]
  9. Syrokou, M.K.; Paramithiotis, S.; Skandamis, P.N.; Drosinos, E.H.; Bosnea, L.; Mataragas, M. High-quality draft genome sequence data of six Lactiplantibacillus plantarum subsp. argentoratensis strains isolated from various Greek wheat sourdoughs. Data Br. 2021, 37, 107172. [Google Scholar]
  10. EFSA. Guidance on the characterization of microorganisms used as feed additives or as production microorganisms. EFSA J. 2018, 16, e05206. [Google Scholar]
  11. Sun, Z.; Harris, H.M.B.; McCann, A.; Guo, C.; Argimón, S.; Zhang, W.; Yang, X.; Jeffrey, I.B.; Cooney, J.C.; Kagawa, T.F.; et al. Expanding the biotechnology potential of lactobacilli through comparative genomics of 213 strains and associated genera. Nat. Comm. 2015, 6, 8322. [Google Scholar] [CrossRef] [PubMed]
  12. Carpi, F.M.; Coman, M.M.; Silvi, S.; Picciolini, M.; Verdenelli, M.C.; Napolioni, V. Comprehensive pan- genome analysis of Lactiplantibacillus plantarum complete genomes. J. Appl. Microbiol. 2022, 132, 592–604. [Google Scholar] [CrossRef] [PubMed]
  13. Choi, S.; Jin, G.-D.; Park, J.; You, I.; Kim, E.B. Pan-genomics of Lactobacillus plantarum revealed group- specific genomic profiles without habitat association. J. Microbiol. Biotechnol. 2018, 28, 1352–1359. [Google Scholar] [CrossRef] [PubMed]
  14. Inglin, R.C.; Meile, L.; Stevens, M.J.A. Clustering of pan- and core- genome of Lactobacillus provides novel evolutionary insights for differentiation. BMC Genom. 2018, 19, 284. [Google Scholar] [CrossRef]
  15. Evanovich, E.; de Souza Mendonça Mattos, P.J.; Guerreiro, J.F. Comparative genomic analysis of Lactobacillus plantarum: An overview. Int. J. Genom. 2019, 2019, 4973214. [Google Scholar]
  16. Rodrigo-Torres, L.; Yépez Aznar, A.R.; Arahal, D.R. Genomic insights into five strains of Lactobacillus plantarum with biotechnological potential isolated from chicha, a traditional maize-based fermented beverage from northwestern Argentina. Front. Microbiol. 2019, 10, 2232. [Google Scholar] [CrossRef]
  17. Schmid, M.; Frei, D.; Patrignani, A.; Schlapbach, R.; Frey, J.E.; Remus-Emsermann, M.N.P.; Ahrens, C.H. Pushing the limits of de novo genome assembly for complex prokaryotic genomes harboring very long, near identical repeats. Nucleic Acids Res. 2018, 46, 8953–8965. [Google Scholar] [CrossRef]
  18. Thakur, M.; Deshpande, H.W.; Bhate, M.A. Isolation and identification of lactic acid bacteria and their exploration in non-dairy probiotic drink. Int. J. Curr. Microbiol. Appl. Sci. 2017, 6, 1023–1030. [Google Scholar]
  19. Sawitzki, M.C.; Fiorentini, A.M.; Cunha Junior, A.; Bertol, T.M.; Sant’anna, E.S. Lactobacillus plantarum AJ2 isolated from naturally fermented sausage and its effects on the technological properties of Milano-type salami. Food Sci Technol. 2008, 28, 709–717. [Google Scholar] [CrossRef] [Green Version]
  20. Weimann, A.; Mooren, K.; Frank, J.; Pope, P.B.; Bremges, A.; McHardy, A.C. From genomes to phenotypes: Traitar, the microbial trait analyzer. mSystems 2016, 1, e00101–e00116. [Google Scholar] [CrossRef] [Green Version]
  21. Dorđević, T.M.; Siler-Marinković, S.S.; Durović-Pejčev, R.D.; Dimitrijević-Branković, S.I.; Gajić Umiljendić, J.S. Dissipation of pirimiphos-methyl during wheat fermentation by Lactobacillus plantarum. Lett. Appl. Microbiol. 2013, 57, 412–419. [Google Scholar] [CrossRef] [PubMed]
  22. Zhang, Y.-H.; Xu, D.; Liu, J.-Q.; Zhao, X.-H. Enhanced degradation of five organophosphorus pesticides in skimmed milk by lactic acid bacteria and its potential relationship with phosphatase production. Food Chem. 2014, 164, 173–178. [Google Scholar] [CrossRef] [PubMed]
  23. Iqbal, S.; Nguyen, T.H.; Nguyen, T.T.; Maischberger, T.; Haltrich, D. β-galactosidase from Lactobacillus plantarum WCFS1: Biochemical characterization and formation of prebiotic galacto-oligosaccharides. Carbohydr. Res. 2010, 345, 1408–1416. [Google Scholar] [CrossRef] [PubMed]
  24. Kittibunchakul, S.; Pham, M.-L.; Tran, A.-M.; Nguyen, T.-H. β-galactosidase from Lactobacillus helveticus DSM 20075: Biochemical characterization and recombinant expression for applications in dairy industry. Int. J. Mol. Sci. 2019, 20, 947. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  25. Roberfroid, M.; Gibson, G.R.; Hoyles, L.; McCartney, A.L.; Rastall, R.; Rowland, I.; Wolvers, D.; Watzl, B.; Szajewska, H.; Stahl, B.; et al. Prebiotic effects: Metabolic and health benefits. Br. J. Nutr. 2010, 104, S1–S63. [Google Scholar] [CrossRef] [Green Version]
  26. Tettelin, H.; Riley, D.; Cattuto, C.; Medina, D. Comparative genomics: The bacterial pan-genome. Curr. Opin. Microbiol. 2008, 11, 472–477. [Google Scholar] [CrossRef]
  27. Russell, R.R.; Aduse-Opoku, J.; Sutcliffe, I.C.; Tao, L.; Ferretti, J.J. A binding protein-dependent transport system in Streptococcus mutans responsible for multiple sugar metabolism. J. Biol. Chem. 1992, 267, 4631–4637. [Google Scholar] [CrossRef]
  28. McLaughlin, R.E.; Ferretti, J.J. The multiple-sugar metabolism (msm) gene cluster of Streptococcus mutans is transcribed as a single operon. FEMS Microbiol. Lett. 1996, 140, 261–264. [Google Scholar] [CrossRef]
  29. Fedtke, I.; Kamps, A.; Krismer, B.; Götz, F. The nitrate reductase and nitrite reductase operons and the narT gene of Staphylococcus carnosus are positively controlled by the novel two-component system NreBC. J. Bacteriol. 2002, 184, 6624–6634. [Google Scholar] [CrossRef] [Green Version]
  30. Schönert, S.; Seitz, S.; Krafft, H.; Feuerbaum, E.-A.; Andernach, I.; Witz, G.; Dahl, M.K. Maltose and maltodextrin utilization by Bacillus subtilis. J. Bacteriol. 2006, 188, 3911–3922. [Google Scholar] [CrossRef] [Green Version]
  31. Pelletier, G.; Frenette, M.; Vadeboncoeur, C. Transport of mannose by an inducible phosphoenolpyruvate: Fructose phosphotransferase system in Streptococcus salivarius. Microbiology 1994, 140, 2433–2438. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  32. Bardou, P.; Mariette, J.; Escudié, F.; Djemiel, C.; Klopp, C. jvenn: An interactive Venn diagram viewer. BMC Bioinform. 2014, 15, 293. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  33. Chun, B.H.; Kim, K.H.; Jeon, H.H.; Lee, S.H.; Jeon, C.O. Pan-genomic and transcriptomic analyses of Leuconostoc mesenteroides provide insights into its genomic and metabolic features and roles in kimchi fermentation. Sci. Rep. 2017, 7, 11504. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  34. Carrizo, S.L.; Montes de Oca, C.E.; Laiño, J.E.; Suarez, N.E.; Vignolo, G.; LeBlanc, J.L.; Rollán, G. Ancestral Andean grain quinoa as source of lactic acid bacteria capable to degrade phytate and produce B-group vitamins. Food Res. Int. 2016, 89, 488–494. [Google Scholar] [CrossRef] [PubMed]
  35. Li, P.; Li, X.; Gu, Q.; Lou, X.-Y.; Zhang, X.-M.; Song, D.-F.; Zhng, C. Comparative genomic analysis of Lactobacillus plantarum ZJ316 reveals its genetic adaptation and potential probiotic profiles. J. Zhejiang Univ. Sci. B 2016, 17, 569–579. [Google Scholar] [CrossRef]
  36. Goel, A.; Halami, P.M.; Tamang, J.P. Genome analysis of Lactobacillus plantarum isolated from some Indian fermented foods for bacteriocin production and probiotic marker genes. Front. Microbiol. 2020, 11, 40. [Google Scholar] [CrossRef]
  37. Begley, M.; Hill, C.; Gahan, C.G.M. Bile salt hydrolase activity in probiotics. Appl. Environ. Microbiol. 2006, 72, 1729–1738. [Google Scholar] [CrossRef] [Green Version]
  38. Benkerroum, N. Biogenic amines in dairy products: Origin, incidence, and control means. Compr. Rev. Food Sci. Food Saf. 2016, 15, 801–826. [Google Scholar] [CrossRef] [Green Version]
  39. Liu, C.-J.; Wang, R.; Gong, F.-M.; Liu, X.-F.; Zheng, H.-J.; Luo, Y.-Y.; Li, X.-R. Complete genome sequences and comparative genome analysis of Lactobacillus plantarum strain 5-2 isolated from fermented soybean. Genomics 2015, 106, 404–411. [Google Scholar] [CrossRef]
  40. Westra, E.R.; Semenova, E.; Datsenko, K.A.; Jackson, R.N.; Wiedenheft, B.; Severinov, K.; Brouns, S.J.J. Type I-E CRISPR-cas systems discriminate target from non-target DNA through base pairing-independent PAM recognition. PLoS Genet. 2013, 9, e1003742. [Google Scholar] [CrossRef] [Green Version]
  41. Marraffini, L.A.; Sontheimer, E.J. CRISPR interference limits horizontal gene transfer in staphylococci by targeting DNA. Science 2008, 322, 1843–1845. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  42. Couvin, D.; Bernheim, A.; Toffano-Nioche, C.; Touchon, M.; Michalik, J.; Néron, B.; Rocha, E.P.C.; Vergnaud, G.; Gautheret, D.; Pourcel, C. CRISPRCasFinder, an update of CRISRFinder, includes a portable version, enhanced performance and integrates search for Cas proteins. Nucleic Acids Res. 2018, 46, 246–251. [Google Scholar] [CrossRef] [Green Version]
  43. Van Reenen, C.A.; Chikindas, M.L.; Van Zyl, W.H.; Dicks, L.M.T. Characterization and heterologous expression of a class IIa bacteriocin, plantaricin 423 from Lactobacillus plantarum 423, in Saccharomyces cerevisiae. Int. J. Food Microbiol. 2003, 81, 29–40. [Google Scholar] [CrossRef]
  44. Van Reenen, C.A.; Van Zyl, W.H.; Dicks, L.M.T. Expression of the immunity protein of plantaricin 423, produced by Lactobacillus plantarum 423, and analysis of the plasmid encoding the bacteriocin. Appl. Environ. Microbiol. 2006, 72, 7644–7651. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  45. Diep, D.B.; Straume, D.; Kjos, M.; Torres, C.; Nes, I.F. An overview of the mosaic bacteriocin pln loci from Lactobacillus plantarum. Peptides 2009, 30, 1562–1574. [Google Scholar] [CrossRef]
  46. Maldonado, A.; Ruiz-Barba, J.L.; Jimenez-Diaz, R. Purification and genetic characterization of plantaricin NC8, a novel coculture-inducible two-peptide bacteriocin from Lactobacillus plantarum NC8. Appl. Environ. Microbiol. 2003, 69, 383–389. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  47. Maldonado, A.; Ruiz-Barba, J.L.; Jimenez-Diaz, R. Production of plantaricin NC8 by Lactobacillus plantarum NC8 is induced in the presence of different types of Gram-positive bacteria. Arch. Microbiol. 2004, 181, 8–16. [Google Scholar] [CrossRef] [Green Version]
  48. Chun, J.; Rainey, F.A. Integrating genomics into the taxonomy and systematics of the Bacteria and Archaea. Int. J. Syst. Evol. Microbiol. 2014, 64, 316–324. [Google Scholar] [CrossRef]
  49. Mataragas, M. Investigation of genomic characteristics and carbohydrates’ metabolic activity of Lactococcus lactis subsp. lactis during ripening of a Swiss-type cheese. Food Microbiol. 2020, 87, 103392. [Google Scholar]
  50. Chun, J.; Oren, A.; Ventosa, A.; Christensen, H.; Arahal, D.R.; da Costa, M.S.; Rooney, A.P.; Yi, H.; Xu, X.-W.; De Meyer, S.; et al. Proposed minimal standards for the use of genome data for the taxonomy of prokaryotes. Int. J. Syst. Evol. Microbiol. 2018, 68, 461–466. [Google Scholar] [CrossRef]
  51. Lee, I.; Chalita, M.; Ha, S.M.; Na, S.I.; Yoon, S.H.; Chun, J. ContEst16S: An algorithm that identifies contaminated prokaryotic genomes using 16S RNA gene sequences. Int. J. Syst. Evol. 2017, 67, 2053–2057. [Google Scholar] [CrossRef]
  52. Yoon, S.H.; Ha, S.M.; Kwon, S.; Lim, J.; Kim, Y.; Seo, H.; Chun, J. Introducing EzBioCloud: A taxonomically united database of 16S rRNA and whole genome assemblies. Int. J. Syst. Evol. Microbiol. 2017, 67, 1613–1617. [Google Scholar] [CrossRef] [PubMed]
  53. Lee, I.; Kim, Y.O.; Park, S.C.; Chun, J. OrthoANI: An improved algorithm and software for calculating average nucleotide identity. Int. J. Syst. Evol. Microbiol. 2015, 66, 1100–1103. [Google Scholar] [CrossRef]
  54. Meier-Kolthoff, J.P.; Göker, M. TYGS is an automated high-throughput platform for state-of-the-art genome-based taxonomy. Nat. Commun. 2019, 10, 2182. [Google Scholar] [CrossRef] [PubMed]
  55. Chaudhari, N.M.; Gupta, V.K.; Sutra, C. BPGA-an ultra-fast pan-genome analysis pipeline. Sci. Rep. 2016, 6, 24373. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  56. Edgar, R.C. Search and clustering orders of magnitude faster than BLAST. Bioinformatics 2010, 26, 2460–2461. [Google Scholar] [CrossRef] [Green Version]
  57. Davis, J.J.; Wattam, A.R.; Aziz, R.K.; Brettin, T.; Butler, R.; Butler, R.M.; Chlenski, P.; Conrad, N.; Dickerman, A.; Dietrich, E.M.; et al. The PATRIC Bioinformatics Resource Center: Expanding data and analysis capabilities. Nucleic Acids Res. 2020, 48, D606–D612. [Google Scholar] [CrossRef] [Green Version]
  58. Aramaki, T.; Blanc-Mathieu, R.; Endo, H.; Ohkubo, K.; Kanehisa, M.; Goto, S.; Ogata, H. KofamKOALA: KEGG Ortholog assignment based on profile HMM and adaptive score threshold. Bioinformatics 2020, 36, 2251–2252. [Google Scholar] [CrossRef] [Green Version]
  59. Huerta-Cepas, J.; Forslund, K.; Coelho, L.P.; Szklarczyk, D.; Jensen, L.J.; von Mering, C.; Bork, P. Fast genome-wide functional annotation through orthology assignment by eggNOG-mapper. Mol. Biol. Evol. 2017, 34, 2115–2122. [Google Scholar] [CrossRef] [Green Version]
  60. Lombard, V.; Golaconda Ramulu, H.; Drula, E.; Coutinho, P.M.; Henrissat, B. The carbohydrate active enzymes database (CAZy) in 2013. Nucleic Acids Res. 2021, 42, D490–D495. [Google Scholar] [CrossRef] [Green Version]
  61. Jia, B.; Raphenya, A.R.; Alcock, B.; Waglechner, N.; Guo, P.; Tsang, K.K.; Lago, B.A.; Dave, B.M.; Pereira, S.; Sharma, A.N.; et al. CARD 2017: Expansion and model-centric curation of the comprehensive antibiotic resistance database. Nucleic Acids Res. 2017, 45, 566–573. [Google Scholar] [CrossRef] [PubMed]
  62. Cosentino, S.; Larsen, V.M.; Aarestrup, F.M.; Lund, O. PathogenFinder–distinguishing friend from foe using bacterial whole genome sequence data. PLoS ONE 2013, 8, e77302. [Google Scholar] [CrossRef]
  63. Van Heel, A.J.; De Jong, A.; Montalbán-López, M.; Kok, J.; Kuipers, O.P. BAGEL3: Automated identification of genes encoding bacteriocins and (non-) bactericidal post-translationally modified peptides. Nucleic Acids Res. 2013, 41, 448–453. [Google Scholar] [CrossRef] [PubMed]
  64. Carattoli, A.; Zankari, E.; Garcia-Fernandez, A.; Larsen, M.V.; Lund, O.; Villa, L.; Aarestrup, F.M.; Hasman, H. In Silico detection and typing of plasmids using PlasmidFinder and Plasmid Multilocus Sequence Typing. Antimicrob. Agents Chemother. 2014, 58, 3895–3903. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  65. Johansson, M.H.K.; Bortolaia, V.; Tansirichaiya, S.; Aarestrup, F.M.; Roberts, A.P.; Petersen, T.N. Detection of mobile genetic elements associated with antibiotic resistance in Salmonella enterica using a newly developed web tool: MobileElementFinder. J. Antimicrob. Chemother. 2021, 76, 101–109. [Google Scholar] [CrossRef] [PubMed]
  66. Bortolaia, V.; Kaas, R.S.; Ruppe, E.; Roberts, M.C.; Schwarz, S.; Cattoir, V.; Philippon, A.; Allesoe, R.L.; Rebelo, A.R.; Florensa, A.F.; et al. ResFinder 4.0 for predictions of phenotypes from genotypes. J. Antimicrob. Chemother. 2020, 75, 3491–3500. [Google Scholar] [CrossRef] [PubMed]
  67. Zankari, E.; Allesøe, R.; Joensen, K.G.; Cavaco, L.M.; Lund, O.; Aarestrup, F.M. PointFinder: A novel web tool for WGS-based detection of antimicrobial resistance associated with chromosomal point mutations in bacterial pathogens. J. Antimicrob. Chemother. 2017, 72, 2764–2768. [Google Scholar] [CrossRef] [Green Version]
  68. Joensen, K.G.; Scheutz, F.; Lund, O.; Hasman, H.; Kaas, R.S.; Nielsen, E.M.; Aarestrup, F.M. Real-time whole-genome sequencing for routine typing, surveillance, and outbreak detection of verotoxigenic Escherichia coli. J. Clin. Microbiol. 2014, 52, 1501–1510. [Google Scholar] [CrossRef] [Green Version]
  69. Arndt, D.; Grant, J.R.; Marcu, A.; Sajed, T.; Pon, A.; Liang, Y.; Wishart, D.S. PHASTER: A better, faster version of the PHAST phage search tool. Nucleic Acids Res. 2016, 44, 16–21. [Google Scholar] [CrossRef] [Green Version]
Ijms 23 02487 g001 550
Figure 1. Calculated ANI values showing the relatedness between the sequenced genomes of the strains isolated from traditional Greek wheat sourdoughs (LQC 2422, LQC 2441, LQC 2516, LQC 2485, LQC 2320, and LQC 2520) and the type strains restricted to those displayed a very high 16S rRNA gene similarity (≥99.80%).
Figure 1. Calculated ANI values showing the relatedness between the sequenced genomes of the strains isolated from traditional Greek wheat sourdoughs (LQC 2422, LQC 2441, LQC 2516, LQC 2485, LQC 2320, and LQC 2520) and the type strains restricted to those displayed a very high 16S rRNA gene similarity (≥99.80%).
Ijms 23 02487 g001
Ijms 23 02487 g002 550
Figure 2. A bootstrapped genome-based phylogenetic relationship of the strains isolated from traditional Greek wheat sourdoughs (LQC 2422, LQC 2441, LQC 2516, LQC 2485, LQC 2320, and LQC 2520) and the type strains restricted to those displayed a very high 16S rRNA gene similarity (≥99.80%). The numbers above branches are pseudo-bootstrap support values > 60% from 100 replications. Matrix on the right (columns from left to right), dDDH species (>70%) and subspecies (>79%) cluster, strains with the same color belong to the same species or subspecies group; GC content (%), min 43.9 (L. paraplantarum DSM 10667T) and max 46.3 (L. pentosus DSM 20314T), strains with the same color have similar GC content (e.g., strains of this study and L. plantarum subsp. argentoratensis DSM 16365T have a GC content of 45.0–45.1); delta (δ) values (<1) showing the tree-likeness of the data set, if a strain has an exceptionally high value this indicates that this genome should be removed because it negatively affects phylogenetic inference; Genome size in Mb, min 3.13 (LQC 2422) and max 3.67 (L. pentosus DSM 20314T); Protein content, min 2958 (LQC 2422) and max 3310 (L. pentosus DSM 20314T); Strains provided by the user (cross symbol); and Type species (tick symbol).
Figure 2. A bootstrapped genome-based phylogenetic relationship of the strains isolated from traditional Greek wheat sourdoughs (LQC 2422, LQC 2441, LQC 2516, LQC 2485, LQC 2320, and LQC 2520) and the type strains restricted to those displayed a very high 16S rRNA gene similarity (≥99.80%). The numbers above branches are pseudo-bootstrap support values > 60% from 100 replications. Matrix on the right (columns from left to right), dDDH species (>70%) and subspecies (>79%) cluster, strains with the same color belong to the same species or subspecies group; GC content (%), min 43.9 (L. paraplantarum DSM 10667T) and max 46.3 (L. pentosus DSM 20314T), strains with the same color have similar GC content (e.g., strains of this study and L. plantarum subsp. argentoratensis DSM 16365T have a GC content of 45.0–45.1); delta (δ) values (<1) showing the tree-likeness of the data set, if a strain has an exceptionally high value this indicates that this genome should be removed because it negatively affects phylogenetic inference; Genome size in Mb, min 3.13 (LQC 2422) and max 3.67 (L. pentosus DSM 20314T); Protein content, min 2958 (LQC 2422) and max 3310 (L. pentosus DSM 20314T); Strains provided by the user (cross symbol); and Type species (tick symbol).
Ijms 23 02487 g002
Ijms 23 02487 g003 550
Figure 3. Heatmap of the phenotypic features of the strains isolated from traditional Greek wheat sourdoughs (LQC 2422, LQC 2441, LQC 2516, LQC 2485, LQC 2320, and LQC 2520) as predicted by the Traitar webtool.
Figure 3. Heatmap of the phenotypic features of the strains isolated from traditional Greek wheat sourdoughs (LQC 2422, LQC 2441, LQC 2516, LQC 2485, LQC 2320, and LQC 2520) as predicted by the Traitar webtool.
Ijms 23 02487 g003
Ijms 23 02487 g004 550
Figure 4. Venn diagram (A) showing the number of genes located in the core-, accessory- and unique genome of each L. plantarum subsp. argentoratensis strain. The plots below the Venn diagram show the total number of genes in each genome of the strains (B) and the total number of genes shared by the respective number of genomes (C). Venn diagram was constructed using the jvenn webtool [32].
Figure 4. Venn diagram (A) showing the number of genes located in the core-, accessory- and unique genome of each L. plantarum subsp. argentoratensis strain. The plots below the Venn diagram show the total number of genes in each genome of the strains (B) and the total number of genes shared by the respective number of genomes (C). Venn diagram was constructed using the jvenn webtool [32].
Ijms 23 02487 g004
Ijms 23 02487 g005 550
Figure 5. Heat-map of the genes presence (red) and absence (blue) of the six L. plantarum subsp. argentoratensis genomes. Yellow box marks the difference in the genes presence/absence between the two microbial clusters.
Figure 5. Heat-map of the genes presence (red) and absence (blue) of the six L. plantarum subsp. argentoratensis genomes. Yellow box marks the difference in the genes presence/absence between the two microbial clusters.
Ijms 23 02487 g005
Ijms 23 02487 g006 550
Figure 6. Plots displaying the distribution of the COG (A) and KEGG (B) functional categories in the core- and accessory-/unique-genome of the L. plantarum subsp. argentoratensis strains.
Figure 6. Plots displaying the distribution of the COG (A) and KEGG (B) functional categories in the core- and accessory-/unique-genome of the L. plantarum subsp. argentoratensis strains.
Ijms 23 02487 g006
Ijms 23 02487 g007 550
Figure 7. Organization of the pln loci of the six L. plantarum subsp. argentoratensis strains. (A) LQC 2485, LQC2441, and LQC 2422, (B) LQC 2516, and (C) LQC 2320 and LQC 2520.
Figure 7. Organization of the pln loci of the six L. plantarum subsp. argentoratensis strains. (A) LQC 2485, LQC2441, and LQC 2422, (B) LQC 2516, and (C) LQC 2320 and LQC 2520.
Ijms 23 02487 g007
Table
Table 1. Pairwise comparisons (dDDH values) between the genomes of the strains isolated from traditional Greek wheat sourdoughs (LQC 2422, LQC 2441, LQC 2516, LQC 2485, LQC 2320, and LQC 2520) and the genomes of the type strains restricted to those displayed a very high 16S rRNA gene similarity (≥99.80%).
Table 1. Pairwise comparisons (dDDH values) between the genomes of the strains isolated from traditional Greek wheat sourdoughs (LQC 2422, LQC 2441, LQC 2516, LQC 2485, LQC 2320, and LQC 2520) and the genomes of the type strains restricted to those displayed a very high 16S rRNA gene similarity (≥99.80%).
a/aStrains12345678910
1LQC 2422-
2LQC 244199.9-
3LQC 2516100100-
4LQC 248599.599.599.5-
5LQC 232089.289.289.288.9-
6LQC 252089.289.289.288.9100-
7L. paraplantarum DSM 10667T31.131.231.231.531.131.2-
8L. plantarum subsp. argentoratensis DSM 16365T95.195.195.29489.990.531.5-
9L. plantarum subsp. plantarum ATCC 14917T62.562.462.462.363.16331.162.8-
10L. pentosus DSM 20314T24.424.424.424.5242424.424.923.8-
Values in bold indicate that the threshold value for subspecies (79%) or species (70%) delineation is exceeded.
Table
Table 2. AMR (antimicrobial resistance) genes identified in the L. plantarum subsp. argentoratensis genomes.
Table 2. AMR (antimicrobial resistance) genes identified in the L. plantarum subsp. argentoratensis genomes.
a/aAMR ElementKEGG KOGene (EC No)Description
1BacitracinK06153uppP, EC 3.6.1.27Catalyzes the dephosphorylation of the undecaprenyl diphosphate (UPP). Confers resistance to bacitracin
ABC transporter, permease protein, probably the 2 or 3 component bacitracin resistance efflux pump, BcrAB or BcrABC
Chloramphenicol acetyltransferase
2BacitracinK01992-D-alanyl-D-alanine-carboxypeptidase
Catalyzes hydrolysis of the D-alanyl-D-alanine dipeptide
Beta-lactamase enzyme family
3PhenicolK19271catA, EC 2.3.1.28Mediates bacterial resistance to the antibiotics streptomycin and spectomycin
4VancomycinK07260vanY, EC 3.4.17.14
5VancomycinK08641ddpX, EC 3.4.13.22
6Beta-LactamK17836bla1, bla2, EC 3.5.2.6
7AminoglycosideK00984aadA, EC 2.7.7.47
Table
Table 3. CRISPRcas system identified in the L. plantarum subsp. argentoratensis genomes.
Table 3. CRISPRcas system identified in the L. plantarum subsp. argentoratensis genomes.
StrainTotal CRISPR
Regions		CRISPR
Regions with EL 1 or 2 1		CRISPR
Regions with EL 3 or 4		cas Gene Clusters Nearby CRISPR
Regions 2 		Type of cas Gene ClusterTotal cas Genes 3
LQC 232042 of EL12 of EL42CAS-typeI and CAS-typeIE11
LQC 242241 of EL11 of EL3 & 2 of EL42CAS-typeI and CAS-typeIE11
LQC 244141 of EL11 of EL3 & 2 of EL42CAS-typeI and CAS-typeIE11
LQC 248551 of EL14 of EL42CAS-typeI and CAS-typeIE11
LQC 251641 of EL11 of EL3 & 2 of EL42CAS-typeI and CAS-typeIE11
LQC 252042 of EL12 of EL42CAS-typeI and CAS-typeIE11
1 When the evidence level (EL) is 1 or 2 then the CRISPR region is not likely to be a real CRISPR array. 2 Only two CRISPR regions of evidence level (EL) 4 had in vicinity cas gene clusters (one gene cluster per CRISPR array). 3 CAS-typeI gene cluster, four cas genes; CAS-typeIE, seven cas genes.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Syrokou, M.K.; Paramithiotis, S.; Drosinos, E.H.; Bosnea, L.; Mataragas, M. A Comparative Genomic and Safety Assessment of Six Lactiplantibacillus plantarum subsp. argentoratensis Strains Isolated from Spontaneously Fermented Greek Wheat Sourdoughs for Potential Biotechnological Application. Int. J. Mol. Sci. 2022, 23, 2487. https://doi.org/10.3390/ijms23052487

AMA Style

Syrokou MK, Paramithiotis S, Drosinos EH, Bosnea L, Mataragas M. A Comparative Genomic and Safety Assessment of Six Lactiplantibacillus plantarum subsp. argentoratensis Strains Isolated from Spontaneously Fermented Greek Wheat Sourdoughs for Potential Biotechnological Application. International Journal of Molecular Sciences. 2022; 23(5):2487. https://doi.org/10.3390/ijms23052487

Chicago/Turabian Style

Syrokou, Maria K., Spiros Paramithiotis, Eleftherios H. Drosinos, Loulouda Bosnea, and Marios Mataragas. 2022. "A Comparative Genomic and Safety Assessment of Six Lactiplantibacillus plantarum subsp. argentoratensis Strains Isolated from Spontaneously Fermented Greek Wheat Sourdoughs for Potential Biotechnological Application" International Journal of Molecular Sciences 23, no. 5: 2487. https://doi.org/10.3390/ijms23052487

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop