WO2018217852A1

WO2018217852A1 - Crispr based tool for characterizing bacterial serovar diversity

Info

Publication number: WO2018217852A1
Application number: PCT/US2018/034064
Authority: WO
Inventors: Nikki SHARIAT
Original assignee: Gettysburg College
Priority date: 2017-05-23
Filing date: 2018-05-23
Publication date: 2018-11-29

Abstract

A process for identifying Salmonella serovars present in an environmental sample includes (i) a first PCR step of amplifying polynucleotides containing one or more of CRISPR1 and CRISPR2 loci using tailed divergent primers directed to the invariant direct repeats in Salmonella CRISPR arrays, (ii) isolating and purifying the products of step (i), (iii) a second PCR amplification step of amplifying the purified products from the second step, (iv) isolating and purifying the products of step (iii), (v) sequencing the purified products of the prior step and using the sequences to identify discrete serovars in the sample.

Description

CRISPR BASED TOOL FOR CHARACTERIZING BACTERIAL SEROVAR DIVERSITY

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This patent application claims benefit of United States Patent Application Serial No. 62/510,129, filed on May 23, 2017, entitled "CRISPR BASED TOOL FOR CHARACTERIZING BACTERIAL SEROVAR DIVERSITY" the disclosure of which is incorporated by reference as if fully rewritten herein.

[0003] STATEMENT AS TO FEDERALLY SPONSORED RESEARCH

[0004] This invention was made with government support under a USDA-NI FA awa rd to Dr. Sha riat: 2016-69003-24615. The government has certain rights in the invention.

[0005] FIELD OF THE INVENTION

[0006] The present invention relates to the detection and identification and characterization of the presence of bacteria of the Salmonella genus. I n particular, the present invention provides methods to characterize the serovars present in a sample by providing rapid a nd inexpensive detection, ideally useful prior to any illness can occur. The methods are particularly useful with respect to Salmonella, and finds application in the poultry industry, other food processing (e.g., nut butter, pepper, tomatoes, etc.), sewage management, waterways and agriculture.

[0007] BACKGROUND OF THE INVENTION

[0008] Salmonella enterica (Salmonella) is a leading cause of bacterial foodborne illness in the

United States and is a tremendously diverse species, comprising six subspecies and >2,500 serovar. Serova rs (aka serotypes) are determined by the presence of cell-surface antigens.

Salmonella enterica subspecies enterica accounts for the majority of clinical cases of salmonellosis and serovar diversity (~1,500 serovars). Different serovars can exhibit varied virulence phenotypes and pathogenicity - that is, all serovars do not cause illness, and some serovars cause worse symptoms than others. Salmonellosis outbreaks are commonly associated with poultry; since 2012 a third of all salmonellosis outbreaks have been poultry associated.

[0009] In routine surveillance for Salmonella, samples need to be enriched in liquid broth to enable detection by plating on selective media. This enrichment step promotes Salmonella growth, while inhibiting growth of other bacteria and is required for detection of Salmonella. Without this, the detection capabilities are tremendously reduced. Enriched cultures are then plated on selective media (XLD), to identify Salmonella colonies prior to subsequent analysis of these presumptive colonies. Importantly, only a small number of colonies (1-5 colonies) are picked and serotyped. Thus, current methodology identifies dominant serovars but does not provide adequate resolution to detect Salmonella serovars present in low numbers (background serovars) during the sampling process.

[0010] As currently known in the art, serovar diversity is addressed in one of two ways. In the first approach, they are evaluated using plating technology. Following plating on selective medium, all colonies can be picked and serotyped. If there are ~200 colonies per plate, and serotyping costs ~$50/sample, this would cost $10,000. Further, if the background serovar is present at 1:1000, several thousand colonies would need to be serotyped to confirm presence. Thus, the conventional plating approach does not provide information about all serovars present and would need to be repeated at great time and expense cost to provide a full understanding of the serovars present in a sample.

[0011] In another approach, a method, termed CRISPOL (US Patent 8673568) relies on probing. CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats) arrays in bacteria. CRISPR arrays are present in ~45 of bacterial genomes and consist of invariant direct repeat sequences, separated by highly variable spacer sequences. CRISPOL involves PCR amplification of all CRISPRs in Salmonella using universal primers, thereafter, discrete identification of a serovar is performed using a known probe. The CRISPOL technology is limited by the number of probes possible to multiplex in a single reaction and by what is known - it is not possible to identify any serovar for which a probe is not used in the CRISPOL process. CRISPOL allows forthe identification of a known and prominent serovar in a sample, but it does not provide an approach for identifying all present serovars.

[0012] What is needed and what Applicant provides here is a method whereby all of the Salmonella serovars present in a sample can be identified, not just the most prevalent or enriched serovar(s).

[0013] DESCRIPTION OF THE INVENTION

[0014] CRISPR and bacterial detection and characterization:

[0015] Clustered Regularly Interspaced Short Palindromic Repeat DNA sequences, or CRISPR, are a repeating sequence of nucleotides found in ~45% bacterial genomes, including Salmonella. The origins of these sequences can be traced to a bacterial adaptive immune system against viruses. Each of the spacers corresponds to a piece of viral DNA. Spacer heterogeneity exists among different Salmonella serotypes, as presumably these serotypes have established different ecological niches. Thus, Salmonella CRISPR arrays can be considered a molecular bar code. As further described herein, the inventors have developed a method, CRISPR-SeroSeq, to interrogate the spacer content of Salmonella that is found in a sample by using PCR and next generation sequencing approach to determine the exact mix of Salmonella serotypes in the sample.

[0016] The ability to evaluate Salmonella serovar diversity, for example, during different stages in poultry processing, enables a more thorough understanding of how environmental and poultry management practices may affect serovar population shifts, especially with respect to those serovars that are multi-drug resistant. According to the methods herein, novel and high- resolution molecular means is provided to detect and identify all Salmonella serovars present in individual poultry samples without the need for isolation of individual colonies, and to investigate the presence of antibiotic resistance in these serovars.

[0017] To effectively mitigate antimicrobial resistance, it is crucial to discern the extent to which elements that confer resistance exist within the Salmonella reservoir. Without a clear picture of serovar diversity, even more beneficially with corresponding profiles of antimicrobial resistance, current mitigation strategies may be futile. The methods described herein enable rapid identification of Salmonella serovar diversity directly from an enriched sample, thus providing information about the complete population of serovars present in a sample.

[0018] CRISPR arrays are present in ~45% of bacterial genomes. These loci consist of invariant direct repeat sequences, separated by highly variable spacer sequences. Referring to FIG 1, the Salmonella CRISPR-Cas system comprises two elements: a CRISPR array and a group of CRISPR- associated (cas) genes (gray arrows). CRISPR arrays consist of tandem invariant direct repeats (black diamonds), which are separated by variable spacer sequences (colored boxes) that are generally derived from bacteriophages or plasmids. In their best-characterized capacity, CRISPR elements function as an elegant nucleic acid-based prokaryotic adaptive immune system. [0019] Referring now to FIG 2, the organization of a CRISPR array is shown. The invariant direct repeats (29 nucleotides in length) are shown as black diamonds. The terminal direct repeat is shown in grey and usually differs by a few nucleotides. The variant spacer sequences (32 nucleotides) are shown as colored boxes, where each unique sequence is depicted as a uniquely colored box. In the bottom panel, the direct repeats are excluded for simplicity.

[0020] Spoligotyping, or spacer-oligonucleotide typing, was first demonstrated in Mycobacterium tuberculosis and has been subsequently applied to other bacteria. The spoligotyping principle involves PCR amplification of the CRISPR array with labeled primers that recognize the invariant direct repeat sequences (FIG 4 C) and hybridization of PCR products to a membrane containing spacer probes. Strain-specific spacer content results in differential hybridization patterns. More recently, Fabre et al. demonstrated that spoligotyping can be successfully utilized for subtyping Salmonella isolates.

[0021] According to the method CRISPOL (for CRISPR polymorphism), fluorescently labeled microbeads provide a high-throughput alternative to traditional spoligotyping. CRISPOL showed 100% concordance between CRISPOL types (equivalent to unique subtypes) and CRISPR array sequence data from 150 Salmonella isolates. While both spoligotyping and CRISPOL are useful techniques, traditional spoligotyping and CRISPOL are both limited by the requirement of individual colonies for analysis and neither provides information about all serovars present in a sample.

[0022] The Inventive Method: CRISPR SeroSeq

[0023] The inventors have discovered that the presence of serovar-specific spacers allows for discrete identification of all serovars in a sample according to the methods disclosed herein. Within a serovar, spacer composition and organization is highly conserved such that spacers can be considered serovar-specific. The inventors provide here a method, CRISPR-SeroSeq, which uses an amplicon-based sequencing pipeline to identify multiple serovars present in a single environmental sample. A universal PCR primer is used to amplify portions of both CRISPR arrays, and after purification and barcoding, amplicons are sequenced, for example, on the lllumina MiSeq platform. Subsequent bioinformatics analysis is used to determine spacer content.

[0024] In a representative embodiment, the invention provides a process for identifying bacterial serovars present in an environmental sample. The process includes (i) a first PCR step of amplifying polynucleotides containing one or more of CRISPR1 and CRISPR2 loci using tailed divergent primers directed to the invariant direct repeats in Salmonella CRISPR arrays, the process further includes (ii) a PCR product purification step comprising isolating and purifying the products of the first PCR amplification step, the process further includes (iii) a second PCR amplification step comprising amplifying the purified products from the second step, the process further includes (iv) isolating and purifying the products of the second PCR amplification step, the process further includes (v) sequencing the purified products of the prior step and comparing the sequences of the isolated purified products with a database to identify discrete serovars in the sample. In some embodiments, the second amplification step includes use of barcoded primers and tags on the amplified isolated purified products to facilitate multiplexing of samples and next-generation sequencing on an lllumina sequencing platform. In the various embodiments, the amplification steps are carried out by a process selected from the group constituted of: polymerase chain reaction (PCR), ligase chain reaction (LCR), nucleic acid sequence-based amplification (NASBA), cycling probe technology (CPT), nested PCR and multiplex PCR. [0025] In accordance with the veracious embodiments, the method enables detection of serovars present in the environmental sample at a ratio in a range from 1:100,000 to 1:1. It will be understood that a sample may contain from one to any number of possible serovars. It will also be understood that while the described embodiments relate to detection of serovars of Salmonella, that the methods more generally are applicable and are not necessarily limited to the exemplified methods as described herein and shown in the various drawings.

[0026] More particularly, the inventive method involves a two-step process that employs PCR and sequencing technology. In the first step, CRISPR primers are used to amplify heterogeneous CRISPR spacers. The universal primers recognize Salmonella CRISPR regions, specifically the invariant direct repeats. PCR products are prepared and purified, and after purification, these PCR products are re-amplified by PCR, using barcoded primers that also include tags to facilitate multiplexing of samples and next-generation sequencing on the lllumina sequencing platform. A bioinformatics pipeline has been developed to extract spacer sequences and match them against a database to identify all serovars that are present in a sample. In contrast to traditional spoligotyping and CRISPOL, the CRISPR-SeroSeq method described herein sequences all the PCR product of the first step, thus in one reaction it is possible to identify all serovars present in a sample.

[0027] Referring now to FIG 3, two serovars, such as S. Kentucky and S. Enteritidis, can be distinguished based on spacer profile. For simplicity, the invariant direct repeats are not shown in this diagram. According to the inventive methods, differences in the composition of spacers and the presence or absence of spacers within a serovar can be exploited to identify distinct strains of a given serovar. Analysis of the entire CRISPR array can allow subtyping while identification of spacers (differences/absence) allows for discrete serovar identification. [0028] Comparison of the inventive SeroSeq methods with other methods

[0029] Referring now to FIG 4, a schematic shows three different approaches to serovar detection. The far right panel represents the inventive method hereof, CRISPR-SeroSeq. The inventive method uses universal Salmonella CRISPR primers (small, black arrows) that recognize the direct repeat sequences of Salmonella CRISPR arrays. After one round of PCR, small amplicons consisting of 1-3 spacers are generated. These are then purified and subjected to a second round of PCR using barcoded lllumina-adapter primers to facilitate next generation sequence analysis. The dominant serovar is depicted in blue, and the background/less dominant serovar is depicted in pink.

[0030] The inventive method combines the power of next-generation sequencing and a PCR reaction that amplifies CRISPR loci, for example, in Salmonella. Referring again to FIG 4, panel C, sequence identity in the direct repeats (black diamonds) allows a single set of primers to be used (gray arrows) to amplify CRISPRs from all Salmonella. Sequencing reads will correspond to spacers from different serovars, in this case, Serovars A (dominant) and B (background). The inset shows expected PCR products and their sizes. In some examples, (eg, lllumina MiSeq 150 cycles) the expected reads are about 150 nucleotides in length (dashed line) which the first two spacers and nine nucleotides of the third spacer to be identified. It will be appreciated by those of ordinary skill in the art that in some other examples that employ longer reads, for example, 300bp reads, the sequence identity of all three spacers could be established.

[0031] In contrast to the inventive method, which provides information about the populations of bacteria that are present in a sample, the species of bacteria and identifies all serovars, conventional methods provide much more limited information. Referring again to FIG 4, the left pa nel depicts the approach for identification the presence of a particula r bacteria in a sam ple. The method includes 16S ribosomal sequencing wherein hyper variable regions in the 16S rRNA gene provide a species-specific signature sequence which is useful for bacterial identification. Identical primers are used to PCR the 16S gene from all microbes in a sample and the sequences are aligned. Sequence differences are used to identify a species. This method is limited in that it does not provide any information about the various serovars that are present in the sample.

[0032] Referring again to FIG 4, the middle panel depicts conventional serotyping, a method wherein 1-5 colonies that are agar plated a re selected for subtyping. According to this method, only one population at a time can be evaluated, and typically, only the dominant microbial population is selected for serotyping, and the less dominant population(s), a representative example depicted in the figure in red (i.e., a more rare serovar), would not be routinely detected by serotyping. This method is limited in that it does not provide any information about other populations of microbes in the sample, or about any other serovars in a sample. That information can be gained by serotyping, but only by repeating the serotyping for each population in the sample.

[0033] As shown, each of the two conventional methods as compared with the inventive method provides the following information:

[0034] The above table highlights that 16S sequencing and CRISPR-SeroSeq can be applied to samples containing diverse populations of bacteria (16s) or Salmonella (CRISPR-SeroSeq), rather than single isolates (as for conventional analysis) but that CRISPR-SeroSeq can also provide serovar information for all serovars present at ratios of 1:10,000. CRISPR-SeroSeq is the first high- throughput method to detect multiple Salmonella serovars in a single sample, particularly where the serovars are not equally present. As further reported herein, the inventors used the inventive methods to identify serovars present at 1:10,000, a ratio that would not be possible with conventional approaches.

[0035] In accordance with the various methods, CRISPR-SeroSeq involves determining the presence of serovar-specific spacers, and does not provide information to quantify serovars within a sample based on spacer content and number of sequencing reads. Thus, while the method is semi-quantitative (for example using CRISPR-SeroSeq in the presence/absence of different antibiotics), given that it is well established that amplicon length and GC content can bias a particular PCR sample, accurate quantification based on read number alone is unexpected for Salmonella.

[0036] EXAMPLE 1

[0037] CRISPR-SeroSeq PCR

[0038] The divergent primers used for the PCR step of CRISPR-SeroSeq are those primers that are as shown below.

[0039] Letters shown in underline and bold are the adapters/tails for the 2nd step PCR (in this example, from lllumina Nextera kit) which are appended to the sequencesthat bind to the CRISPR regions. [0040] Forward CRISPR SeroSeq primer:

[0041] TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGCGCGCCAGCGGGGATAAACC [0042] Reverse CRISPR SeroSeq primer:

[0043] GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGCTGGCGCGGGGAACAC

[0044] In some alternative embodiments, the primers may be selected from those used for CRISPOL (refer to the below included APPENDIX) and shown to work in Salmonella and may include 5' adaptors to facilitate library construction, for example lllumina Nextera library construction. FIG 4C demonstrates the heterogeneity of expected PCR products and their sizes. PCR products are purified using established methods. In some examples, purification may be achieved using a resin bead column. In some specific examples, purification is achieved using AMpure beads (Beckman Coulter, Inc.). The purified PCR products are then re-amplified with indexed (barcoded) primers to a llow sample multiplexing. Samples a re 'sequence-ready' with just two PCR reactions.

[0045] To demonstrate the sensitivity of CRISPR-SeroSeq, serovars with known CRISPR sequences are mixed in different ratios (e.g. 1:1, 1:10, 1:100 and 1:1000 as well as template mixtures with >2 serovars). As PCR favors smaller am plicons, it is expected that the majority of products contain one, two or three spacers (see inset, FIG 4C,). Samples including 16 barcoded samples are run, representing different ratios of different serovars, on the lllumina MiSeq platform (yielding approx. 0.8 x 10⁵ reads/sample) to establish the depth of sequencing sufficient to identify spacers corresponding to a serovar present at 1/lOOOth the levels of a dominant serovar. PCR conditions are employed to enable the fewest cycles sufficient for detection of spacers from less prevalent serova rs (10-20 reads for a ra re spacer). [0046] EXAMPLE 2: Development of the CRISPR-SeroSeq bioinformatics pipeline

[0047] For analysis of sequencing data bioinformatic scripts are used to accomplish three steps: (1) removal of any reads that do not contain GCCAGCGGGGATAAACCG (corresponding to the 3' end of the direct repeat). (2) trimming of the ends of the reads to remove the sequencing adaptors, and, (3) collapsing of redundant reads to enable more streamlined downstream analysis. BLAST with custom spacer datasets from Salmonella spacer database are used to query sequencing reads, the database consisting of spacers collected from both Salmonella CRISPR loci from over 150 serovars. The database is non-redundant such that only spacers unique to a single serovar are included.

[0048] EXAMPLE: 3

[0049] CRISPR-SeroSeq has the potential to be an extremely powerful tool to investigate the diversity of Salmonella populations in many different reservoirs that could include sewage, waterways, agriculture, and food processing. The method

[0050] 1) Is independent of culturing individual Salmonella clones; and,

[0051] 2) Can determine the complexity of serotypes from a diverse population of Salmonella.

[0052] FIG 5 shows that the dominant Salmonella serovars in the poultry industry over time has fluctuated in response to interventions by the National Poultry Improvement plan that target specific serotypes (arrows). The serovars shown in bold are known human pathogens and all are in the top five serotypes that cause illness.

[0053] Referring now to FIG 6, it is posited that exposure of bacterium to different viral pathogens has led to the divergence of CRISPR spacer sequences in different Salmonella serotype. Importantly, within a serotype, spacer composition and organization is highly conserved. Salmonella CRISPR arrays exhibit distinct spacer composition within serovars. CRISPR1 arrays, from the top four chicken-associated serovars are shown. The colored boxes represent unique spacer sequences (for clarity, the direct repeat sequences are removed). Spacers that are found in multiple serovars shown in the example, are shown by a gray bar and are excluded from our database. It will be appreciated that these are merely representative, and other serovars have different spacers. In this example, serovar-specific spacers are indicated by a black bar below and these are included in our Salmonella serovar spacer database. Different strains of a serovar exhibit inclusion of different spacers - this is the basis for using CRISPR as a subtyping tool. The arrow spacer in Kentucky represents a spacer >50 nucleotides. CRISPR2 arrays also exhibit similar serovar differences and will also be included in the CRISPR-SeroSeq analysis.

[0054] The inventive method is aimed at enabling the discrete identification of even low numbers of various serovars in a sample taking advantage of the variations seen in the spacers across serovars. Referring again to FIG 4, CRISPR-SeroSeq employs similar methodology to 16s sequencing and will provide serovar resolution, which is not possible with 16s sequencing and is time and resource intensive using conventional serotyping approaches.

[0055] Referring now to FIG 7, a schematic overview of the CRISPR-SeroSeq protocol is shown.

[0056] Samples were provided having a mixture of Salmonella serovars. As shown in the top panel of FIG 7, tailed divergent primers recognizing the direct repeats present in all Salmonella CRISPR arrays were used to amplify the spacer sequences (1) and produced tagged, heterogeneous amplicons consisting of 1 or more spacers (2). After purification, these products were used as templates in a second PCR reaction that adds barcoded lllumina sequencing tags (3). Products were purified and sequenced (4). [0057] Purified and sequenced products from the two PCR steps were evaluated, as shown in the lower panel of FIG 7. The resultant 150 nucleotide sequencing reads were first trimmed at the 3'end (1), the spacer sequences were extracted (2), and then analyzed using local BLAST analysis against a database of Salmonella spacer sequences

[0058] Referring now to FIG 8, the pilot CRISPR-SeroSeq experiment shows serovar detection with genomic DNA from Salmonella Kentucky and Salmonella Enteritidis mixed in different ratios. Using the CRISPR-SeroSeq pipeline, the inventors successfully detected reads from Salmonella Kentucky, even when it was present at 1:100,000 as compared with Salmonella Enteritidis DNA. Salmonella Kentucky is shown in blue and Salmonella in red. Referring again to FIG 8, the upper panel shows the proportion of the reads that were detected from both Salmonella Kentucky and Enteritidis. The lower panel expands the dotted region in the upper panel to clearly show the presence of the background serovar. These results show that the CRISPR-SeroSeq Protocol is able to detect Salmonella serovars in low concentrations.

[0059] The results further showed that the inventive method provided information about serovar complexity in samples collected from poultry houses through the first steps of processing. Of 96 samples (boot socks, fecal and rinses), approximately one third resulted in Salmonella-pos^'ti^'we colonies when plated on Xylose lysine deoxycholate (XLD) agar. The inventive CRISPR-SeroSeq method identified the presence of multiple serovars in several individual samples, as shown in FIG 9, where the upper panel shows the proportion of reads from multiple Salmonella serovars, and the lower panel expands the dotted region in the upper panel to clearly show the presence of the background serovars. [0060] EXAMPLE 4

[0061] CRISPR-SeroSeq is used to determine antibiotic resistance in background Salmonella serovars in chickens.

[0062] Determining fluctuations in serovar diversity in response to antibiotic exposure involves evaluating whether the background Salmonella serovars identified according to the process above harbor antibiotic resistance elements not found in the dominant serovar. Each enriched sample (frozen stocks from the poultry house collections in Fig 9) is screened for antibiotic susceptibility against antibiotics routinely analyzed in Salmonella by the National Antimicrobial Resistance Monitoring System (NARMS). Experimentally, enriched samples are sub-cultured in 3 ml of tryptic soy broth (TSB) with and without antibiotics. Antibiotics are used at minimal inhibitory concentrations (MIC), as defined by the Clinical and Laboratory Standards Institute (CLSI). Using these cultures, we the following are evaluated with each sample:

[0063] First, 0.6ml is used to assess for presence of antibiotic-resistant bacteria, as determined by increased OD600 values after a 12-hour incubation. Then, 1 ml is used to prepare DNA template for a serovar-specific qPCR to identify the serovars present, and specifically identify spacer content.

[0064] With the results from the above, multiplex qPCR with spacer-specific probes is completed. Referring now to FIG 10, when compared to 'no-antibiotic' controls, a significant decrease in the number of cycles required to reach the cycle number threshold (Ct value) is observed when a resistant serovar is present. This data provides information about the antibiotic resistance of the serovars in the sample, which will be confirmed as follows: 1 ml of the culture is plated onto xylose lysine deoxycholate (XLD) media (plus appropriate antibiotic) to select individual resistant Salmonella isolates. Single colonies are serotyped to confirm the serovar identified and pure cultures are stored at -80 to enable additional analysis.

[0065] These experiments are designed to detect differential antibiotic resistance profiles between background and dominant Salmonella serovars, and provide novel insight into antimicrobial resistance in background serovars. Experiments such as this can provide information regarding the presence of antibiotic resistance that would not otherwise be identified.

[0066] While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes may be made and equivalents may be substituted for elements thereof without departing from the scope of the invention. In addition, many modifications may be made to adapt a particular situation or material to the teachings of the invention without departing from the essential scope thereof. Therefore, it is intended that the invention not be limited to the particular embodiment disclosed as the best mode contemplated for carrying out this invention, but that the invention will include all embodiments falling within the scope of the appended claims.

APPENDIX

[0067] The terms "nucleic acid" and "DNA" are equivalent and comprise single-stranded or double-stranded nucleic acids.

[0068] The term "primer" denotes a single-stranded or double-stranded oligonucleotide, preferably a single-stranded oligonucleotide for optimal efficiency. A primer (an oligonucleotide primer), once hybridized to a single-stranded nucleic acid sequence, termed "template", is a substrate for at least one DNA polymerase (i.e. a primer hybridized to a nucleic acid sequence has the property of binding, at its 3'OH end, at least one DNA polymerase). In the presence of appropriate nucleotides (A, C, G, T), of a DNA polymerase (for example, the Taq polymerase), of an appropriate buffer (comprising cofactors or compounds affecting the pH or the ionic strength of the reaction medium, for example) and at an appropriate temperature, the 3ΌΗ end of a primer can be extended, thus resulting in the synthesis of a strand complementary to the template sequence to which said primer is hybridized. The primers, according to the present invention, preferably have a length of less than or equal to 50 nucleotides, i.e. less than or equal to 40, 30, 20, 15 or 10 nucleotides. More preferably, the primers according to the present invention have a length of between 15 and 30 nucleotides, preferably between 18 and 25 nucleotides. The primers according to the present invention can be advantageously labeled.

[0069] For the purpose of the present invention, two nucleic acid molecules are "complementary" when each of the bases in successive positions of the 5' end of the first nucleic acid molecule is paired with the corresponding residue in the second molecule, starting from the 3' end, according to the rules of base-pair pairing (i.e. A and T, C and G). Under suitable conditions known to those skilled in the art, two complementary single strands of DNA reassociate to form a double-stranded DNA molecule.

[0070] The term "CRISPR locus" is intended to mean a genomic DNA sequence composed of a series of repeat nucleotide sequences (called DR) having a size of approximately 21 to 37 base pairs, spaced out by variable nucleotide sequences (spacers) having a size of approximately 20 to 40 base pairs. In the context of the disclosure of the present invention, the CRISPRl locus is defined as being located in a position 5' with respect to the CRISPR2 locus.

[0071] The bacteria of the Salmonella genus have one or two CRISPR loci. The following table 1 shows the location of the CRISPRl and CRISPR2 loci of some strains of bacteria of the Salmonella genus:

[0072]

Claims

What is claimed is:

1. A process for identifying Salmonella serovars present in an environmental sample, the process comprising (i) a first PCR step of amplifying polynucleotides containing one or more of CRISPRl and CRISPR2 loci using tailed divergent primers directed to the invariant direct repeats in Salmonella CRISPR arrays, the process further comprising (ii) a PCR product purification step comprising isolating and purifying the products of the first PCR amplification step, the process further comprising (iii) a second PCR amplification step comprising amplifying the purified products from the second step, the process further comprising (iv) isolating and purifying the products of the second PCR amplification step, the process further comprising (v) sequencing the purified products of the prior step and comparing the sequences of the isolated purified products with a database to identify discrete serovars in the sample.

2. The process according to claim 1, wherein the second amplification step includes use of barcoded primers and tags on the amplified isolated purified products to facilitate multiplexing of samples and next-generation sequencing on an lllumina sequencing platform.

3. The process according to claim 1, characterized in that amplification steps are carried out by a method selected from the group constituted of: polymerase chain reaction (PCR), ligase chain reaction (LCR), nucleic acid sequence-based amplification (NASBA), cycling probe technology (CPT), nested PCR and multiplex PCR.

5. The process according to claim 1, wherein the serovars present in the environmental sample may be present at a ratio in a range from 1:10,000 to 1:1.

6. The process according to claim 1, characterized in that the sample comprises at least one isolated bacterial strain of the Salmonella genus.

7. The process according to claim 1, wherein the sample is obtained from a food supply.

8. The process according to claim 7, wherein the sample is from one of a chicken farm and a chicken processing facility.

9. An in vitro method for detecting and characterizing serovar diversity of bacterium of the Salmonella genus in a sample, the method comprising at least steps:

(a) amplifying at least one nucleic acid fragment from a bacterium of the Salmonella genus, the fragment comprising one or more of the CRISPR1 locus and the CRISPR2 locus, using at least one set of tailed divergent primers directed to the direct repeats present in Salmonella CRISPR arrays,

(b) isolating and purifying tagged heterogeneous amplicons obtained in step (a), the isolated amplicons each comprising at least one CRISPR1 or CRISPR2 spacers,

(c) amplifying at least one nucleic acid fragment obtained from the purification step (b), the amplification including incorporation of barcoded illumine sequencing tags;

(d) isolating and purifying the barcoded illumine tagged products from step (c),

(e) sequencing the products from step (d),

(f) analyzing the sequenced products by comparing the sequences obtained for the CRISPR1 and/or CRISPR2 spacers using BLAST analysis to identify discrete serovars of the bacteria by matching the sequenced products to a database.

10. The method according to claim 3, wherein step (a) is preceded by a step of extraction of nucleic acids present in said sample.