Abstract

PCR-RFLP targeting of the 16S rDNA and rpoB genes, as well as the vdc region, was applied to identify and differentiate between the spoilage and non-spoilage Alicyclobacillus species. Eight reference strains and 75 strains isolated from spoiled juices, juice concentrates, drinks, its intermediates, and fresh apples were subject to study. Hin6I restriction patterns of the 16S rDNA gene enabled distinguishing between all the species analyzed, while the rpoB gene and vdc gene cluster analysis also revealed that there were two major types among the A. acidoterrestris isolates, one similar to the reference strain A. acidoterrestris DSM 2498, and the other similar to the reference strain A. acidoterrestris ATCC 49025. Heterogeneity was also observed among the A. acidocaldarius isolates. RFLP analysis of the 16S rDNA and rpoB genes, as well as vdc region, can be used successfully in the identification and research of intraspecies heterogeneity of the Alicyclobacillus species.

1. Introduction

The contamination of fruit juices by Alicyclobacillus has recently become one of the most important issues in the juice and beverage industry. These acidophilic, thermophilic, and spore-forming bacteria are very hard to eliminate from contaminated drinks.

Alicyclobacillus are Gram positive, aerobic, soil borne bacteria that are able to grow within a range from pH 2.0 to 6.0 and at temperatures from 20 to 70°C [18]. The two main factors which prevent fruit products from spoilage with most other bacteria, which are thermal treatment and low pH values, are insufficient to eliminate Alicyclobacillus. The spores survive under typical pasteurization conditions and are able to germinate and grow in an acidic environment [9, 10]. Thermal treatment may even impel germination of the spores [1113].

Despite being nonpathogenic [14], some Alicyclobacillus species may cause the spoilage of juices and juice-containing products such as nectars and beverages. Alicyclobacillus species have been found all across the world, and their presence has been detected on fruit surfaces [15], in juices produced from several fruits: citrus, apple, banana, berry, and stone fruits [1, 8, 10, 1624], in canned tomatoes [25], and in drinks, for example ice tea, and isotonic drinks [17, 26]. The spoilage mainly manifests itself as the formation of a medical, antiseptic off-odour, from compounds produced by the bacteria. The main compound associated with spoilage is guaiacol, produced from vanillin and vanillic acid [2729], but halophenols, 2,6-dibromophenol and 2,6-dichlorophenol, have also been reported as spoilage agents. Among the 22 currently known Alicyclobacillus species, five have been proven to produce an off-odour: A. acidoterrestris, A. acidiphilus, A. pomorum, A. cycloheptanicus, and A. herbarius. [1, 9, 14, 3037].

The classic method for isolating and characterizing Alicyclobacillus, devised by IFU (Internationale Fruchtsaft Union), which is commonly used in the juice and beverage industry, takes about 15 days. If present in the tested sample, guaiacol can be detected using the peroxidase method [27, 38], by sensory tests ([9, 10, 39]; Siegmund and Pöllinger-Zierler, 2006) or instrumental methods, for example, HPLC or gas chromatography [28, 35, 40, 41].

The identification of the Alicyclobacillus species based on their ability to assimilate erythritol with acid production [19, 42] mainly allows differentiating between two species: A. acidoterrestris and A. acidocaldarius.

Since the classic microbiological Alicyclobacillus detecting methods are time consuming, alternative approaches have been adopted, like flow cytometry ([30], Pieper et al. 2006), Fourier transform infrared spectroscopy [4345], and genetic methods. Except for RAPD-PCR, (Yamazaki et al., 1997; [4649]), most of the genetic methods used in the studies on Alicyclobacillus target the rDNA operon. These include 16S rDNA and ITS region sequencing, 16S rDNA RFLP, Real-Time PCR, and LAMP-PCR of the 16S rDNA fragment ([1, 50]; Durak et al., 2002; [6, 7, 5154]).

Guaiacol, the main spoilage agent, is produced by nonoxidative decarboxylation of vanillic acid. The ability to produce guaiacol is associated with presence of vdc gene cluster, consisting of three genes, vdcB, vdcC, and vdcD. Chow et al. [55] described vdc gene cluster in Streptomyces sp. Detection of the vdc genes of A. acidoterrestris using RT-PCR was described by Niwa and Kawamoto [38]. The vdc region sequence was published by Matsubara [56]. To date, there are no applications of vdc region analysis for any microorganism. RpoB gene, encoding the β subunit of bacterial RNA polymerase, is one of the single-copy housekeeping genes and is widely used in studies on bacterial taxonomy. These studies include PCR-RFLP analyses of rpoB gene fragments; however, so far, there are no reports on using rpoB gene analysis in research on of Alicyclobacillus.

Our study focuses on the use of rpoB gene, vdc region, and 16SrDNA gene as molecular markers for the identification and differentiation of Alicyclobacillus.

2. Materials and Methods

2.1. Sample Acquisition and Bacterial Strains

Seventy-five strains analyzed in this study were isolated from concentrated apple juice (47), concentrated cherry juice (4), fresh apples (3), concentrated strawberry juice (2), concentrated black currant juice (2), tomato juice (2), orange juice (3), cloudy (1) and clear (1) apple juice, apple beverage (1), concentrated beetroot juice (1), concentrated raspberry juice (1), concentrated orange juice (1), orange beverage (1), banana nectar (1), cherry puree (1), and intermediates used in beverage production (3). All the strains were isolated according to the method described in IFU no. 12 September 2004/March 2007.

Reference strains were obtained from the Leibniz Institute DSMZ – German Collection of Microorganisms and Cell Cultures (Alicyclobacillus acidiphilus DSM 14558; Alicyclobacillus acidocaldarius DSM 446; Alicyclobacillus acidoterrestris DSM 2498; Alicyclobacillus herbarius DSM 13609; Alicyclobacillus hesperidum DSM 12489), and from the American Type Culture Collection (ATCC) (Alicyclobacillus acidoterrestris ATCC 49025; Geobacillus stearothermophilus ATCC 7953; Bacillus subtilis ATCC 6655). Also, Alicyclobacillus acidocaldarius A1 from our own library collection was used as a reference strain, after biochemical and 16S rDNA sequencing confirmation.

2.2. Biochemical Tests

The strains were checked for erythritol utilization and guaiacol production. The ability of erythritol utilization was tested by plating the cultures on agar containing 1% erythritol and bromophenol blue as an indicator [42]. The ability to produce guaiacol was tested using the peroxidase method [27].

2.3. DNA Isolation

Selected Alicyclobacillus strains were cultured in BAT medium (pH 4.0±0.2) at 45°C for 2–3 days. Bacillus subtilis was cultured in TSB medium (pH 7.1±0.2) at 30°C for 2 days, and Geobacillus stearothermophilus was cultured in TSB medium at 45°C for 2 days.

Bacterial chromosomal DNA was purified using a Genomic Mini Kit (A&A Biotechnology), following the manufacturer’s instructions.

2.4. Primer Designing and Amplification

All PCR reactions were performed using a Pequstar 2x Gradient thermocycler (Pequlab).

2.4.1. 16S rDNA Amplification

A fragment of the 16S rDNA gene was amplified using universal primers, similar to that used by Wang et al., 2010 [57]. The primer sequences were 8F (5′– AGAGTTTGATCCTGGCTCAG), E. coli positions 8-27 and 1512R, shortened by 1 nucleotide at 3′ end (5′ – ACGGCTACCTTGTTACGACT), E. coli positions 1512–1493. The size of the amplification product was 1495 bp.

PCR reactions were performed in a total volume of 50 μl, containing 5 ng of the template DNA, 50 pM of each of the primers, and 25 μl of the DreamTaq™ Green PCR Master Mix (Thermo Scientific). PCR was performed under the following conditions: initial denaturation at 94°C for 2 min, 40 cycles of denaturation at 94°C for 30 sec, annealing at 51°C for 35 sec, elongation 72°C for 1 min 40 sec, and the final elongation at 72°C for 2 min.

2.4.2. vdc Region Amplification

The vdc gene cluster was amplified using primers vdc fr (5′ – CTGTTGGCTCAATGGCGGCTGAGCGAT), vdc rev (5′ – TTATCAGCGGTTTATCCGCGGTGGAACAGTC), vdc1 fr (5′ – AACGACGCAGGTGTGGAAAC), vdc1 rev (5′ – AGCGTGGGCAAGTTGTCATGTG), vdc K (5′ – TTGGCAACGGAGAAGTGGGAG) and vdc S (5′ – AATCACGCGCTGATGATGGG). The 1586 fragment of vdc region, containing fragments of vdcB and vdcC genes, was used as a template for PCR-RFLP and was amplified using the primers Bur 5 (5′ GCCGACGTGATGCTCAARGAGCGCA) and Bur 6 (5′ GTSGCRTCGAGAATCATCTTGTG). The primers were designed based on a comparison of the raw genome sequences derived from Alicyclobacillus acidoterrestris ATCC 49025 ([58]; GenBank number AURB01000113.1), and Alicyclobacillus herbarius DSM 13609 (GenBank number AUMH01000032.1), the sequence published by Matsubara (GenBank number BD187778.1), and sequences obtained from several Alicyclobacillus acidoterrestris strains analyzed in this study. Sequence alignments were performed using the Serial Cloner program. The positions and directions of the vdc primers are described in Table 1 and shown in Figure 1.

The 2523 bp vdc fr – vdc rev amplification product contained the whole vdc region. The size of the Bur5-Bur6 amplification product was 1586 bp. PCR reactions were performed in a total volume of 50 μl, containing 2.5 ng of the template DNA, 5 pM of each of the primers, and 25 μl of the DreamTaq™ Green PCR Master Mix (Thermo Scientific). PCR was performed under the following conditions: initial denaturation at 94°C for 2 min, 30 cycles of denaturation at 94°C for 30 sec, annealing at 59°C for 50 sec, elongation 72°C for 1 min 40 sec, and the final elongation at 72°C for 5 min.

2.4.3. rpoB Gene Amplification

A fragment of rpoB gene was amplified using Gru3–Gru6 primers. The primers were designed based on a comparison of the rpoB gene sequences of Alicyclobacillus acidocaldarius, Bacillus subtilis, and Geobacillus stearothermophilus, as well as sequences derived from A. acidoterrestris 49025, A. hesperidum URH17-3-68, and A. herbarius DSM 13609 raw genomic sequences. Gru5 and Gru6 primers are nondegenerate versions of the Gru3 and Gru4 primers, respectively. The primer sequences were Gru3 (CGYGACGTDCACTAYTCBCACTA), Gru4 (5′ – GCCCANACYTCCATCTCRCCRAA) Gru5 (5′ – CGCGACGTACACTATTCGCACTA), and Gru6 (5′ – GCCCAAACCTCCATCTCACCAAA). The size of the amplification product was 1735 bp. PCR reactions were performed in a total volume of 50 μl, containing 30 ng of the template DNA, 40 pM of each of the primers, and 25 μl of the DreamTaq™ Green PCR Master Mix (Thermo Scientific). PCR was performed under the following conditions: initial denaturation at 94°C for 2 min, 35 cycles of denaturation at 94°C for 30 sec, annealing at 57°C (for the Bur5 and Bur6 primer pairs) or at 59°C (for the Bur3 and Bur4 primers) for 35 sec, elongation 72°C for 1 min 45 sec, and the final elongation at 72°C for 5 min.

2.5. PCR-RFLP

5-15 μl of the PCR products were digested with BsuRI, Hin6I, and HphI (Thermo Scientific) in 20 μl volumes. The samples were incubated at 37°C for 1-2 hours, and the enzymes were then inactivated with ten-minute incubation at 80°C. The digests were analyzed on 2.5–3% agarose gel.

2.6. DNA Sequencing

DNA samples were sequenced using 8F and shortened 1512R primers for the 16S rDNA gene; vdc fr, vdc rev, vdc1 fr, vdc1 rev, Bur5, Bur6, and two additional primers to fill the gaps, vdcS (5′ – AATCACGCGCTGATGATGGG) and vdcK (5′ – TTGGCAACGGAGAAGTGGGAG) for the vdc region; and Gru3 – Gru6 for the rpoB gene.

The contig assemblies, sequence alignments, and phylogenetic analysis were performed using Serial Cloner software (SerialBasic) and Clustal Omega. The sequences were compared to the GenBank sequences database using BLAST tools.

3. Results

3.1. RFLP Analysis of 16S rDNA Fragment

The 1495 bp fragment of the 16S rDNA gene, amplified using 8F and shortened 1512R primers, was digested by BsuRI, Hin6I, and HphI.

While BsuRI and HphI digestions did not allow to distinguish between all the species analyzed, the patterns obtained by Hin6I digestion were species specific (Figure 2).

All analyzed isolates with their features and RFLP profiles are described in Table 2.

3.2. RFLP Analysis of the vdc Region Fragment

The fragment of vdc region was amplified using Bur5 and Bur6 primers. The product was a single band of 1586 bp and was observed only for guaiacol producing strains: A. acidoterrestris, A. acidophilus, and A. herbarius. The PCR product was digested by BsuRI (Figure 3), Hin6I (Figure 4), and HphI (Figure 5).

BsuRI and Hin6I RFLP patterns enabled distinguishing between all the species analyzed, and divided the A. acidoterrestris group into two clusters. Type I pattern was identical to the pattern given by A. acidoterrestris DSM 2498, and type II pattern was identical to A. acidoterrestris ATCC 49025. The HphI patterns confirm this rule with the exception of one strain, which represented the type I pattern in the BsuRI and Hin6I analysis.

3.3. RFLP Analysis of the rpoB Gene Fragment

The 1735 bp fragment of the rpoB gene was amplified using Gru3 and Gru4, or Gru5 and Gru6 primers. Gru3 and Gru4 primers were degenerated versions of Gru5 and Gru6 (respectively) and were used for amplification of isolates other than A. acidoterrestris. A acidoterrestris isolates were amplified either with Gru3–Gru4, or Gru5–Gru6 primers, and the Gru5–Gru6 primer pair was chosen due to the better efficiency of the reaction. Considering the individual isolates, the RFLP patterns were identical for both pairs of primers.

The patterns obtained by all of the nucleases used (Figures 68) enabled distinguishing between the species analyzed. With exception of two strains, the BsuRI patterns divided the A. acidoterrestris group into two clusters, analogical to the clusters revealed by the vdc region analysis. The pattern (IIA) is shown by two exceptional strains, most resembling the type II pattern of A. acidoterrestris, and those strains were classified as type II in other analyses. The reference strain ATCC 49025 shows this type of pattern. The A. acidocaldarius group was represented by three types of patterns, two of them resembling each other. A acidocaldarius DSM 446 belongs to the cluster II of A. acidocaldarius group, while A. acidocaldarius A1 was assigned to cluster I.

For Hin6I there were two types of patterns for the A. acidoterrestris and A. acidocaldarius. A. acidoterrestris DSM 2498 belongs to the cluster I, while A. acidoterrestris ATCC 49025 belongs to the cluster II. A. acidocaldarius DSM 446 represents cluster II of the A. acidocaldarius group, while A. acidocaldarius A1 represents cluster I.

HphI endonuclease produces two types of patterns for A. acidoterrestris, analogically as before, and three types of patterns for A. acidocaldarius. One sample of A. acidoterrestris cluster II has a slightly different pattern and has been classified as type II in other analyses. Two of the three patterns of A. acidocaldarius differ with only one band. A. acidocaldarius DSM 446 represents cluster II of the A. acidocaldarius group, while A. acidocaldarius A1 represents cluster I.

3.4. DNA Sequencing

DNA sequencing was performed for selected strains. Sequencing of the 16S rDNA gene was performed to confirm the proper identification of the isolates and to identify the non-Alicyclobacillus isolates. The 16S rDNA sequence of strains 31 and 34, classified as type I, showed the greatest similarity to A. acidoterrestris DSM 2498, also classified as type I. The sequences of strains 33, 51, 52, 53, 55, and 56, classified as type II, showed the greatest similarity to A. acidoterrestris ATCC 49025, also classified as type II. The sequence of strain 41, classified as type II, showed the greatest similarity to A. acidoterrestris DSM 3922.

Figure 9 shows the phylogenetic tree constructed from the 16S rDNA sequences of analyzed strains of A. acidoterrestris. The sequence analysis indicates that A. acidoterrestris DSM 2498 and two other type I strains are closely related and it shows greater variation within class II strains. The isolates selected for sequencing represent both groups and different sources which they were isolated from: concentrated apple juice (isolates 33 and 41), fresh apples (isolate 34), concentrated blackcurrant juice (isolate 31), concentrated raspberry juice (isolate 56), concentrated beetroot juice (isolate 51), and concentrated cherry juice (isolates 52, 53, and 55). The sequence analysis shows no apparent correlation between diversity of the 16S rDNA sequence and the sources of the isolates.

Five whole vdc gene clusters were sequenced. Two of them were isolated from the reference strains, A. acidoterrestris DSM 2498, and A. acidoterrestris ATCC 49025. 2416 bp fragments of the vdc regions were aligned. The vdc region sequence of strains 15 and 34, classified as type I, showed 99.9% identity with the vdc sequence of A. acidoterrestris DSM 2498; the sequence of strain 41, classified as type II, showed 99.3% identity with the vdc sequence A. acidoterrestris ATCC 49025. Vdc sequences of A. acidoterrestris DSM 2498 and A. acidoterrestris DSM 49025 showed 94.5% identity. For the last pair, protein sequences of the gene products showed 98.5% positives and 96.5% identities for VdcB, 99.4% and 98.9% for VdcC, and 100% and 97.4% for VdcD.

RpoB gene sequencing was performed to confirm that both primer pairs, Gru3–Gru4 and Gru5–Gru6, enabled to amplify the correct DNA fragment and that degeneration of the primers did not affect the sequence specificity, although the degenerated primers produced significantly lower amount of PCR product. All of the sequences obtained showed greatest similarity to the appropriate rpoB genes.

The GenBank accession numbers are KX371237-KX371249 for 16S rDNA sequences; KX453673-KX453677 for full vdc region sequences; KX453678 and KX453679 for partial vdc sequences of A. herbarius and A. acidiphilus, respectively; and KX453680- KX453682 for partial rpoB sequences.

4. Discussion

Among more than one thousand samples of concentrated apple juice tested in our laboratory between 2004 and 2012, 67% was contaminated with Alicyclobacillus sp., and 31% of the isolates were identified as A. acidoterrestris [59]. The statistics show that Alicyclobacillus spoilage is still a major concern in the fruit processing industry.

In this study, 75 guaiacol producing and non-guaiacol producing strains were isolated from various fruit juices, concentrated fruit juices, and fresh apples. Additionally, 8 reference strains were used. The isolates and reference strains were examined using classic methods, and PCR-RFLP focusing on 16S rDNA and rpoB gene fragments as well as the vdc region fragment. For selected strains, DNA sequencing of the 16S rDNA gene was performed. Sixty of the isolates analyzed were identified as A. acidoterrestris, 12 as A. acidocaldarius, one as Brevibacillus agri, and two as Bacillus ginsengihumi. Four of the isolates, which gave ambiguous results during classic identification (nontypical colour on an erythritol medium or lack of growth at 65°C connected with lack of guaiacol production), were identified by genetic methods as A. acidoterrestris or A. acidocaldarius. A. acidoterrestris isolates were grouped in two major clusters. 27 of the isolates belonged to the cluster I, and 33 to the cluster II. Also, A. acidocaldarius isolates were grouped into two clusters, but showed more intracluster diversity.

These results support the observations made by Osopale at al. [60] and Durak et al. [23] who also reported two genetic clusters among analyzed A. acidoterrestris samples, by RAPD analysis,and by 16S rDNA sequencing, respectively.

Most of the strains analyzed in this study were isolated from concentrated apple juice, as this is the main subject of the Alicyclobacillus focused screening made or outsourced by polish fruit industry; however, it is one of the main subjects of similar screenings performed by fruit industry worldwide. A. acidoterrestris representing both clusters have been found in concentrated apple juice. For other sources, three A. acidoterrestris cluster II isolates have been found in orange juices, two cluster I in concentrated black currant juice, and three cluster II in concentrated cherry juice; however, these are only single observations and further research is needed to establish if A. acidoterrestris strains representing both clusters are found in other juices.

Genetic methods are broadly used in the detection, characterization, and differentiation of microorganisms. The application of PCR-RFLP in the characterization of Alicyclobacillus and other thermoacidophilic bacteria isolated from the apple juice processing environment has been described by Chen et al. [52], although only the 16S rDNA gene was the subject of this study and the Hin6I enzyme was not used. RpoB gene has never been subjected to PCR-RFLP or any other genetic analysis of Alicyclobacillus so far, although the value of this gene in taxonomy has been confirmed by many studies on other microorganisms. To date, there are no reports on the use of the vdc gene cluster in taxonomic studies of microorganisms, although its sequence may be proven valuable for research both on the vectors and diversity and evolution of the element itself. Hin6I restriction patterns for 16S rDNA were sufficient to differentiate between all the species analyzed, but provided no closer data on intraspecies diversity. The diversity was revealed both by the rpoB gene and vdc region RFLP analyses.

Vdc gene cluster of Streptomyces sp., described by Chow et al. [55], consists of three genes: vdcB (0.6kb), vdcC (1.4 kb), and vdcD (0.2 kb), transcribed as single, policistronic mRNA molecule. All three genes were essential to produce guaiacol from vanillic acid. Niwa and Kawamoto [38] described analogical gene cluster in A. acidoterrestris, consisting of ORF1 (597 bp), ORF2 (1425 bp), and ORF3 (321 bp). Three respective open reading frames were found in all five analyzed vdc sequences.

Alicyclobacillus acidoterrestris RFLP patterns of both the rpoB gene and vdc region showed consistently that there were two major types among the isolates, one similar to the reference strain A. acidoterrestris DSM 2498, and the other similar to the reference strain A. acidoterrestris ATCC 49025. Obtaining such consistent data concerning two genetic elements of distant function reveals that there is deeper intraspecies genetic diversity in the A. acidoterrestris species. This division may be considered when examining other features of Alicyclobacillus such as susceptibility to temperature or high hydrostatic pressure, in order to establish if there are any differences between the groups.

Both the rpoB gene and vdc region seem to be single-copy genetic elements and showed no signs of intragenomic heterogeneity, which often makes the RFLP comparison of rDNA genes harder to perform.

The sequence analysis of the 16S rDNA fragment of selected isolates shows no apparent correlation between diversity of the 16S rDNA sequence and the sources which the strains were isolated from.

In conclusion, the application of PCR-RFLP has been proven to be a fast and reliable method for Alicyclobacillus identification and differentiation. The method is also technically less difficult than most of other molecular techniques. Two major groups of A. acidoterrestris have been identified. The primers designed for this study could be useful in further research on Alicyclobacillus.

Data Availability

The DNA sequences obtained during this study have been deposited in GenBank, and the accession numbers are included within the article.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

Our research project was fully sponsored by Polish Ministry of Science and Higher Education, with funds for statutory activity.