Distribution of clustered regularly interspaced palindrome repeats CRISPR2 and CRISPR3 in Lactobacillus delbrueckii ssp. bulgaricus strains

Abstract

The function of several families of clustered regularly interspaced palindrome repeats (CRISPRs) in prokaryotic genomes was recently found to be related to the protection of bacterial cells against the expression of foreign DNA, originating from plasmids or bacteriophages. The present study was the first attempt to screen a broader number of Lactobacillus delbrueckii ssp. bulgaricus strains, widely used in yoghurt and cheese production, for the presence of CRISPRs. Database search of four completely sequenced L. delbrueckii ssp. bulgaricus genomes indicated the presence of CRISPR2 in three of them - ATCC 11842, ATCC BAA-365 and ND02, and the presence of CRISPR3 in strain 2038. In the first three strains, the CRISPR2 was invariably located between a 3′–5′ exonuclease gene and a gene for a ppGpp-synthetase. The location of CRISPR3 in strain 2038 was between a histidine-kinase gene and an acetyl-CoA acetyltransferase gene, 2 kbp downstream of the CRISPR2 locus in ATCC 11842. Specific primers were designed to amplify with polymerase chain reaction the target regions containing the potential CRISPR2 and / or CRISPR3 in a total of 33 L. delbrueckii ssp. bulgaricus strains. Thirteen strains yielded a high molecular mass product corresponding in size and location to CRISPR2 of the type strain ATCC 11842, while another 17 strains indicated the presence of potential CRISPR3, analogous to that of strain 2038. Three strains did not indicate the presence of CRISPRs. Interestingly, none of the tested strains carried both CRISPR2 and CRISPR3 simultaneously in its genome at the investigated region.

Keywords:

• CRISPR
• Lactobacillus delbrueckii ssp. bulgaricus
• strain ATCC 11842
• strain 2038

Introduction

Bacteria from the Lactobacillus delbrueckii ssp. bulgaricus species are a basic component of yoghurt and white-brined cheese starters. Compared to the phage attack on Streptococcus thermophilus, phage infection on L. delbrueckii ssp. bulgaricus cultures occurs much more rarely.[1,2] Nevertheless, bacteriophages that attack the L. delbrueckii group of subspecies are well documented in the literature.[3–6] Recently, the first local isolate of L. delbrueckii ssp. bulgaricus bacteriophage was classified into group ‘b’ L. delbrueckii bacteriophages, based on its partial genome sequencing.[6]

The clustered regularly interspaced palindrome repeats (CRISPRs) were first described in the genome of Escherichia coli K-12 and their presence was confirmed for a multitude of bacterial and archaeal genomes.[7,8] Recently, these palindrome repeats and the cas-genes, associated with them, have been found to function as a new defence mechanism in prokaryotic cells against invading phages and plasmids.[9,10] The regular repeats are interspaced with short sequences named ‘spacers’ which derive from foreign genetic elements.[9,11,12] When, for example, the bacterial host is under a bacteriophage attack, it acquires a new spacer sequence within its CRISPR locus that matches a DNA sequence in the phage genome, referred to as a ‘protospacer’.[13,14] The complete set of repeats and spacers is then transcribed and processed into individual crRNAs that interact with the viral DNA, facilitating its cleavage and rendering the bacterial cell immune to infection by the particular phage.[14–16] Depending on the organization of the cas-genes and the sequence of the palindrome repeats, four classes of CRISPRs are described and numbered from 1 to 4. As an example, in S. thermophilus, all four classes of CRISPR/cas systems are described as CRISPR1 and CRISPR3, which are considered to be the most active ones.[14,16]

As data on CRISPRs in the genome of L. delbrueckii sp. bulgaricus are still lacking, in the present study we reported the results from the database search of potential CRISPRs in this subspecies and from the screening of 33 strains for the presence of potential CRISPR2 and CRISPR3 loci in their genome.

Materials and methods

Bacterial strains, culture conditions and DNA isolation

Thirty-three L. delbrueckii ssp. bulgaricus strains maintained in the LBB culture collection (LB Bulgaricum PLC, Sofia, Bulgaria) were included in the study. All cultures were grown in MRS medium (peptone – 10 g/L; meat extract – 8 g/L; yeast extract – 4 g/L; glucose – 20 g/L; sodium acetate trihydrate – 5 g/L; Tween-80 – 1 g/L; dipotassium hydrogen phosphate – 2 g/L; triammonium citrate – 2 g/L; magnesium sulfate heptahydrate – 0.2 g/L; manganese sulfate – 0.05 g/L; pH was adjusted to 6.2) [17] for 24 hours at 37 °C. We used 5 mL of the cultures for DNA isolation with the GenElute Bacterial Genomic DNA Kit (Sigma-Aldrich, St. Louis, MO, USA), according to the manufacturer's instructions. A list of the tested L. delbrueckii ssp. bulgaricus strains is presented in Table 1.

Table 1. List of the tested bacterial strains with the presence (+) or absence (−) of the respective products from a CRISPR2-specific and CRISPR3-specific PCR amplification.

Strain	pr21/pr22*	pr35/pr32**	pr31/pr32***	Strain	pr21/pr22*	pr35/pr32**	pr31/pr32***
B5/1S	+	−	+	B2/15J	−	+	−
B5/5J	+	−	+	B31/2S	−	+	−
B8/1S	+	−	+	B37/5S	−	+	−
B43/9S	+	−	+	B48/8J	−	+	−
B51/7J	+	−	+	B54/6J	−	+	−
B55/8J	+	−	+	B56/6J	−	+	−
B60/3J	+	−	+	B61/7J	−	+	−
B60/8S	+	−	+	B127/8S	−	+	−
B140/3S	+	−	+	B128/1S	−	+	−
B208/5S	+	−	+	B215/6S	−	+	−
B226/2S	+	−	+	B313/7S	−	+	−
B278/2S	+	−	+	B340/10S	−	+	−
B314/5S	+	−	+	B344/8S	−	+	−
B55/7J	−	−	+	B346/2S	−	+	−
B57/4J	−	−	+	B350/9S	−	+	−
B66/1J	−	−	+	B389/10S	−	+	−
				B69/3J	−	+	−

Note: * CRISPR2-specific primers.

** CRISPR3-specific primers.

*** Primers amplifying a short sequence between the histidine-kinase gene and acetyl-CoA acetyltransferase gene, proving the absence of CRISPR3.

Database search for CRISPRs and primer design

Four publicly available genomes of L. delbrueckii ssp. bulgaricus' strains ATCC 11842, ATCC BAA-365, ND02 and 2038 (GenBank Acc. Nos. NC_008054; NC_008529; NC_014727 and NC_017469) were searched for CRISPRs with the CRISPRFinder software.[18] Conservative regions upstream and downstream of CRISPR2 and CRISPR3 were identified by alignment with the CLC Sequence Viewer (www.clcbio.com) and suitable primers were designed with the PrimerBLAST tool.[19] A complete list of the primers, their sequence and location is presented in Table 2.

Table 2. List of the designed primers and their location in the analysed genomic region.

Primer	Direction	Sequence	Location
Pr21	Forward	3′-AGGGATTATCGGAAATGCG-5′	a 3′-5′ exonuclease gene upstreme of CRISPR2
Pr22	Reverse	3′-ATATAGCTTTTGCCCGTGTT-5′	a ppGpp-synthetase gene downstream of CRISPR2
Pr31	Forward	3′-CAGAAGTATTGCGATCCAGA-5′	a histidine-kinase gene upstream of CRISPR3
Pr35	Forward	3′-GCGATTGATGATAATGTTTGA-5′	CRISPR3-associated cas1 gene
Pr32	Reverse	3′-TAGCAGAAAAGTCAGCCAAT-5′	an acetyl-CoA acetyltransferase gene downstream of CRISPR3

Polymerase chain reaction (PCR) amplifications

All amplifications were performed on a 9600 GeneAmp PCR System (Perkin-Elmer, Norwalk, Connecticut) in a 25 μL reaction mixture consisting of diluted VWR Taq DNA Polymerase Master Mix (VWR International, Haasrode, Belgium), 50 ng template DNA and 10 pmol of each primer. The PCR programme was as follows: one cycle of 3 min at 95 °C, 30 cycles of 30 s at 93 °C, 30 s at 60 °C and 1 min at 72 °C, and one cycle of 7 min at 72 °C. The potential CRISPR2 and CRISPR3 regions were amplified with primer pairs pr21 / pr22 and pr35 / pr32, respectively. Additionally, primers pr31 / pr32 were used to amplify a short region to confirm the absence of CRISPR3 in the searched location. The position of the primers within the analysed region is illustrated in Figure 1. The amplified products were separated by electrophoresis in a 2% agarose gel in Tris-Acetate-EDTA buffer (40 mmol/L Tris-acetate and 1 mmol/L EDTA, pH 8.3) at 100 V and visualized after staining with ethidium bromide. The size of the obtained product was determined using a suitable size marker (Gene Ruler 100 bp Plus DNA Ladder, Thermo Scientific, Pittsburgh, PA, USA).

Figure 1. Schematic representation of the location of CRISPR2 and CRISPR3 in the genomes of Lactobacillus delbrueckii ssp. bulgaricus strains ATCC 11842, ATCC BAA-365, ND02 and 2038 and the position of the designed primers.

Note: Acetyl-CoA acetyltransferase gene (A); 3′–5′ exonuclease gene (E); histidine-kinase gene (H); ppGpp-synthetase gene (S); CRISPR3-associated gene (cas1); specific primers (pr21 to pr35).

Results and discussion

The database search of the four completely sequenced L. delbrueckii ssp. bulgaricus genomes indicated the presence of CRISPR2 in three of them – ATCC 11842, ATCC BAA-365 and ND02 and the presence of CRISPR3 in strain 2038.

In the first three strains the CRISPR2 was invariably located between a 3′–5′ exonuclease gene and a gene for a ppGpp-synthetase (Figure 1). In strain 2038, the region preceding the analogue of the ppGpp-synthetase gene contained completely different genes instead of CRISPR2. The location of CRISPR3 of strain 2038 was between a histidine-kinase gene and an acetyl-CoA acetyltransferase gene, 2 kbp downstream of the CRISPR2 locus in ATCC 11842 (Figure 1). In strains ATCC 11842, ATCC BAA-365 and ND02, which did not have the CRISPR3 region, the acetyl-CoA acetyltransferase gene followed immediately after the histidine-kinase gene. The CRISPRFinder software listed 40, 20 and 64 spacers within CRISPR2 of strains ATCC 11842, ATCC BAA-365 and ND02, respectively, and 19 spacers within CRISPR3 of strain 2038. A list of the spacers within CRISPR2 and CRISPR3 of strains ATCC 11842 and 2038 is presented in Table 3.

Table 3. List of spacers within CRISPR2 and CRISPR3 of Lactobacillus delbrueckii ssp. bulgaricus strains ATCC 11842 and 2038, respectively, retrieved by CRISPRFinder software.[18]

Using the specific primers pr21/pr22 for the CRISPR2 region, 13 strains yielded a high molecular mass product, corresponding in size (>3 kbp) and location to CRISPR2 of the type strain ATCC 11842 (Table 1). The separation of the CRISPR2-specific amplified product for strain ATCC 11842 and five representative ‘CRISPR2-positive’ strains is presented in Figure 2(A). Neither of the ‘CRISPR2-positive’ strains yielded an amplification product with the CRISPR3-specific primer pair pr35/pr32 (Table 1, Figure 2(A)). Additionally, for all ‘CRISPR2-positive’ strains a short amplification product was obtained with primers pr31 / pr32 that proved that the histidine-kinase gene was followed immediately by the acetyl-CoA acetyltransferase gene with no insertion of CRISPR3 between them. A visualization of this short amplification product for strain ATCC 11842 and five representative ‘CRISPR2-positive’ strains is included in Figure 2(A).

Figure 2. Analysis of the presence/absence of CRISPR2 (primer pair pr21/pr22; lanes 1–6 and 19–24), CRISPR3 (primer pair pr35/pr32; lanes 7–12 and 25–30) and histidine-kinase / acetyl-CoA acetyltransferase gene (primer pair pr31/pr32; lanes 13–18 and 31–36) specific amplification products in a selection of ‘CRISPR2-positive’ (A) or ‘CRISPR3-positive’ (B) L. delbrueckii ssp. bulgaricus strains.

Note: representative ‘CRISPR2-positive’ strains: ATCC 11842 (lanes 1, 7, 13), B5/1S (2, 8, 14), B60/8S (3, 9, 15), B208/5S (4, 10, 16), B226/2S (5, 11, 17) and B278/2S (6, 12, 18) (A); representative ‘CRISPR3-positive’ strains: B37/5S (19, 25, 31), B54/6J (20, 26, 32), B56/6J (21, 27, 33), B69/3J (22, 28, 34), B127/8S (23, 29, 35) and B215/6S (24, 30, 36) (B). 100 bp Plus DNA ladder (M).

Another 17 strains gave positive amplification with primers pr35/pr32, specific for the CRISPR3 region of strain 2038, indicating the presence of potential CRISPR3 in these cultures (Table 1). The separation of the CRISPR3-specific amplified product for strain 2038 and five representative ‘CRISPR3-positive’ strains is presented in Figure 2(B). All of the ‘CRISPR3-positive’ strains failed to produce amplification product with the CRISPR2-specific primer pair pr21/pr22 suggesting the absence of CRISPR2 in these strains (Table 1). The ‘CRISPR3-positive’ strains also did not yield the short amplification product of primer pairs pr31/pr32, possibly due to the large size of the targeted region, resulting from the insertion of CRISPR3 between the histidine-kinase and acetyl-CoA acetyltransferase genes. The absence of CRISPR2-specific products and the short amplification product of primers pr31/pr32 for strain 2038 and five representative ‘CRISPR3-positive’ strains is demonstrated in Figure 2(B).

The indirect evidence that the amplified region contains CRISPRs was the formation of amplification sub-products with regularly increasing size, which formed a typical ladder-like electrophoretic pattern (Figure 2(A) and 2(B)). This phenomenon can be explained by the formation of loops in some of the template molecules due to internal annealing of the palindromic repeats that prematurely terminate the elongation process. Interestingly, none of the tested strains carried both CRISPR2 and CRISPR3 simultaneously at the investigated region of their genome. Three strains did not indicate the presence of CRISPRs, due to their absence in the genome of the strains, or a completely different location of the potential CRISPRs. A complete list of all strains with the respective amplification products is given in Table 1.

Conclusions

Thirty-three strains of L. delbrueckii ssp. bulgaricus were analysed for the presence of CRISPR2 and CRISPR3 in their genomes. For 13 and 17 of them it was found that CRISPR2 and CRISPR3, respectively, were present and had the same location as that in strains ATCC 11842 and 2038. Neither strain contained the two CRISPRs simultaneously. Although the tested strains may still have CRISPRs in different locations that were not analysed in this study, it seems an interesting finding that with respect to the presence and location of CRISPR2 and CRISPR3 L. delbrueckii ssp. bulgaricus strains could be divided into two lineages - that of strain ATCC 11842 and that of strain 2038. The actual work on determining the sequences of spacers within the CRISPRs of L. delbrueckii ssp. bulgaricus strains is still ahead and may reveal important evolutionary relations between different cultures and highlight the history of particular strains with respect to encounters with foreign DNA.

Disclosure statement

No potential conflict of interest was reported by the authors.