Platform for identifying bacteriophage splitting mode as well as construction method and application thereof

Abstract

The invention relates to the field of bacteriophages, in particular to a platform for identifying a bacteriophage splitting mode as well as a construction method and application of the platform. The method comprises the following steps: respectively constructing a host bacteria library of klebsiella pneumoniae and a bacteriophage library taking klebsiella pneumoniae as host bacteria; detecting the interaction relationship between each bacteriophage in the bacteriophage library and each klebsiella pneumoniae in the host bacterium library to obtain a bacteriophage-host interaction matrix; performing bidirectional clustering analysis on the bacteriophage-host interaction matrix to obtain a plurality of interaction relation modules of bacteriophages and host bacteria; the modules comprise modules with a wide host range and / or modules with a specific host range; wherein the splitting mode refers to the capsular serotype of the phage targeting klebsiella pneumoniae. Depending on the platform, the cracking mode of the unknown bacteriophage on the klebsiella pneumonia can be effectively identified, and the bacteriophage RBP of the klebsiella pneumonia can be specifically identified.

Description

Technical Field

[0001] The present invention relates to the field of bacteriophages, and in particular to a platform for identifying bacteriophage lysis patterns, and a construction method and application thereof. Background Art

[0002] At present, the detection methods of nosocomial pathogens mainly include bacterial culture method, molecular biology method, immunology method, instrument method, etc. The bacterial culture method is to spread the sample to be tested on the culture medium, incubate it at a certain temperature, and then detect the cultivated colonies by physiological and biochemical methods. This method has good accuracy, but it has high requirements on the culture conditions and the number of pathogenic bacteria in the sample, and the detection time is long. With the development of molecular biology technology, polymerase chain reaction has received widespread attention in the detection of pathogens. The PCR method amplifies the specific sequence of bacterial DNA in vitro and analyzes the product by immunofluorescence to achieve the purpose of bacterial species identification. It has the characteristics of high sensitivity and strong specificity. However, conventional PCR detection may result in false detection and missed detection.

[0003] Bacteriophage is a type of virus that uses bacteria as its host. In recent years, it has been widely used in the field of pathogenic bacteria detection. First of all, the structure of bacteriophage is simple, mainly composed of proteins and nucleic acids, and has little effect on the results of bacterial detection. As an active bacterial detection tool, bacteriophage has good stress resistance and can still survive in environments such as strong acids, strong bases, organic solvents, and high temperatures. In addition, bacteriophages reproduce quickly and require a short time for amplification, which effectively simplifies the preparation work before the detection and lays a good foundation for establishing effective bacterial detection methods. In addition, bacteriophages can not only specifically identify bacteria, but also ensure that only live bacteria are infected, which is an important guarantee for the accuracy of bacteriophage detection of bacteria. It is reported that more than 60 kinds of bacteria, including important foodborne pathogens such as Escherichia coli, Salmonella, and Vibrio parahaemolyticus, can enter a viable but non-culturable state (VBNC) under the induction of physical, chemical, and biological factors. Bacteria in the VBNC state still have metabolic activity and pathogenicity, and can be restored to normal bacteria under appropriate conditions. General bacterial detection methods are almost unable to distinguish bacteria in the VBNC state from dead bacteria, which poses a challenge to the accuracy of bacterial detection methods. The characteristic of bacteriophages that they only recognize and infect living bacteria enables them to identify bacteria in the VBNC state.

[0004] The typical life cycle of bacteriophages includes five parts: adsorption, invasion, replication, assembly and release. Among them, adsorption is the first step of phage infection and also the most important part of phage invasion of the host. This process depends on the receptor binding proteins on the surface of the phage and the receptors on the surface of the bacterial cell. The specificity of the adsorption site and the adsorption capacity affect the host range and infection ability of the phage. Depolymerase is an enzyme that plays an important role in the recognition and adsorption process of phages. It can specifically recognize and degrade the capsule of the host bacteria. The tail fibers of some phages even contain more than one depolymerase, which can be bispecific or even multispecific, expanding the host range of the phage. Different depolymerases can recognize different capsule serotypes or multiple capsule serotypes. Therefore, the depolymerase of phages can be used to detect different pathogens in the hospital environment or food.

[0005] Klebsiella pneumoniae (Kp) belongs to the family Enterobacteriaceae. It is commonly found in the skin, respiratory tract, intestines, etc. of humans and animals. It is one of the more common and important pathogens in hospital-acquired and community-acquired infections. In 1986, K. pneumoniae was first found to cause community-acquired liver abscesses with multiple metastatic abscesses. This variant strain is called hypervirulent Klebsiella pneumoniae (hvKP). hvKP is a highly invasive and highly pathogenic pathogen that can cause more severe and disseminated infections. Since the 20th century, there have been more and more reports on infections with KP and hvKP worldwide. Due to the increasingly serious problem of drug resistance of Klebsiella pneumoniae, it has attracted the attention of various health organizations and disease prevention and control centers around the world, and it has been listed as one of the main factors threatening global public safety. Capsular polysaccharides (CPS) are one of the important virulence factors of Klebsiella pneumoniae. CPS can prevent hvKP from being recognized and monitored by the host immune system through immune escape mechanisms such as anti-phagocytosis, inhibition of early inflammatory response, neutralization of antimicrobial peptides to reduce the body's immune response, and inhibition of dendritic cell maturation. This is closely related to the adhesion, growth and proliferation of hvKP in the host. Although the capsule type commonly found in hvKP is KL1 / KL2 / KL57, some classic Klebsiella pneumoniae (cKP) also have the same capsule type, and strains with serum capsule types of KL5, KL16, KL20, KL28, KL54, KL63 and KLN1 also show high virulence. In addition, KL64 is a capsule type that is widely prevalent in my country and is highly resistant to antibiotics.

[0006] From the above, we can see that there are many capsular serotypes of Klebsiella pneumoniae, and there are also many types of phages that can identify different capsular serotypes. Therefore, it is necessary to develop a method that can efficiently identify the lysis pattern of unknown phages, so as to provide a basis for infection, treatment, and detection of different KL types of Klebsiella pneumoniae. Summary of the invention

[0007] The first object of the present invention is to provide a method for constructing a platform for identifying phage lysis patterns, thereby obtaining a platform for identifying phage lysis patterns, which can effectively identify the lysis patterns of unknown phages on Klebsiella pneumoniae and the phage receptor binding protein (RBP) that specifically recognizes Klebsiella pneumoniae. Specifically, the method comprises:

[0008] (1) constructing a host bacterial library of Klebsiella pneumoniae (K. pneumoniae) and a phage library with Klebsiella pneumoniae as the host bacterial library;

[0009] (2) detecting the interaction relationship between each phage in the phage library and each Klebsiella pneumoniae in the host bacteria library to obtain a phage-host interaction matrix;

[0010] (3) performing a bidirectional cluster analysis on the phage-host interaction matrix to obtain a plurality of interaction relationship modules between phages and host bacteria; the modules include modules with a broad host range and / or modules with a specific host range;

[0011] The lysis pattern refers to the capsular serotype (KL) of Klebsiella pneumoniae targeted by the phage. For example, if the lysis pattern of the phage is KL1, it means that the capsular serotype of Klebsiella pneumoniae targeted by the phage is KL1.

[0012] According to the method of the present invention, the host bacteria library is composed of a series of known KL Klebsiella pneumoniae, and in some embodiments, the host bacteria library includes at least 50 strains, at least 75 strains, at least 100 strains, at least 125 strains, at least 150 strains, at least 175 strains, at least 200 strains, or at least 225 strains of Klebsiella pneumoniae. In theory, the more strains of the host bacteria library, the higher the accuracy of the constructed platform, but considering the accuracy and labor input, in some preferred embodiments, the host bacteria library includes at least 150 strains, at least 175 strains, at least 200 strains, or at least 225 strains of Klebsiella pneumoniae. In a specific embodiment of the present invention, the host bacteria library includes 238 strains of Klebsiella pneumoniae.

[0013] It is known that the serum capsule type commonly found in hvKP is KL1, KL2, KL57, and is a high toxicity type. In addition, the serum capsule type is KL20, KL47 strains also show high toxicity. There are also some strains that show high toxicity, and the serum capsule type is KL5, KL16, KL28, KL54, KL63 and KN1. In addition, although KL64 is not a high toxicity type, it is more popular in my country and has high antibiotic resistance. In theory, the richer the KL types of strains in the host bacteria library, the higher the accuracy of the constructed platform, but the comprehensive accuracy and labor input, and some KL are uncommon types, therefore, in some preferred embodiments, the KP in the host bacteria library includes at least the most common KLs of the KP with a serum capsule type of KL1, the KP with a serum capsule type of KL2, the KP with a serum capsule type of KL57 and the KP with a serum capsule type of KL64. In some more preferred embodiments, the KP in the host bacterial library includes at least KP with serum capsule type KL1, KP with serum capsule type KL2, KP with serum capsule type KL57, KP with serum capsule type KL64, KP with serum capsule type KL20, and KP with serum capsule type KL47. In a further preferred embodiment, the KL richness of the KP contained in the host bacterial library includes at least 55 species, including KP with serum capsule type KL1, KP with serum capsule type KL2, KP with serum capsule type KL57, and KP with serum capsule type KL64.

[0014] In the method of the present invention, the phage library is composed of a series of phages capable of lysing KP. In some embodiments, the phage library includes at least 20 strains, at least 30 strains, at least 40 strains, at least 50 strains, at least 60 strains, at least 70 strains, at least 80 strains, at least 90 strains, at least 100 strains, or at least 110 strains of phage. In theory, the more strains in the phage library, the higher the accuracy of the constructed platform, but considering the accuracy and labor input, in some preferred embodiments, the phage library includes at least 80 strains, at least 90 strains, at least 100 strains, or at least 110 strains of phage. In a specific embodiment, the phage library includes 80 strains of phage.

[0015] In the method of the present invention, the lysis pattern species richness of the phages contained in the phage library preferably matches the KL species richness of the KP in the host bacteria library. Therefore, in some preferred embodiments, the lysis pattern of the phages in the phage library at least includes phages that lyse KL1 type hvKP, phages that lyse KL2 type KP, phages that lyse KL57 type KP, and phages that lyse KL64 type KP. In some more preferred embodiments, the lysis pattern of the phages in the phage library at least includes phages that lyse KL1 type KP, phages that lyse KL2 type KP, phages that lyse KL57 type KP, phages that lyse KL64 type KP, phages that lyse KL20 type KP, and phages that lyse KL47 type KP. In a further preferred embodiment, the types of lysis patterns of phages in the phage library include at least 55 types of KL, including phages that lyse KL1 type KP, phages that lyse KL2 type KP, phages that lyse KL57 type KP and phages that lyse KL64 type KP.

[0016] In the method of the present invention, the interaction relationship is preferably carried out in an exhaustive manner, that is, the interaction relationship between each phage strain and each KP strain is detected. The detection method can adopt a conventional detection method in the art, for example, whether a plaque can be formed is used for detection. If a phage strain can form a plaque on a plate with KP, it means that the two react with each other. Through this detection, the interaction relationship between each strain and each KP strain can be obtained, and then the phage-host interaction matrix can be obtained, for example, with the phage as the horizontal axis and the KP as the vertical axis, if the interaction relationship is positive, it is marked as "+", and if the interaction relationship is negative, it is marked as "-".

[0017] In the method of the present invention, a clustering software with a conventional underlying algorithm in the art for calculating distance can be used to perform a bidirectional clustering analysis on the phage-host interaction matrix. For example, the heatmap package of R can be used, and the ggplot package can also be used. After the clustering software is selected, the operation can be performed according to the instructions provided in the software, and the present invention will not be repeated here. Through the clustering analysis, a plurality of interaction relationship modules between phages and host bacteria can be obtained, that is, a phage set that can lyse a specific KL type KP. The module can include a phage set with a wide host range, that is, the KL type of the KP that the phage in the module can lyse is different, and at least two KLs are included. The module can also include a phage set with a specific host range, that is, the KL type of the KP that the phage in the module can lyse is the same, and there is only one KL.

[0018] In some embodiments of the method of the present invention, the module includes at least one of the following: 1) an M1 module capable of cleaving at least two of KL1, KL2 and KL57; 2) an M2 module that specifically cleaves KL1; 3) an M3 module that specifically cleaves KL2; 4) an M4 module that specifically cleaves KL64; 5) an M5 module that specifically cleaves KL57. In some preferred embodiments, the module includes: an M2 module, an M3 module, an M4 module, an M5 module and an optional M1 module. In some more preferred embodiments, the module includes: an M2 module, an M3 module, an M4 module, an M5 module and an M1 module.

[0019] The second object of the present invention is to provide a platform for identifying phage lysis patterns constructed by the above method. The platform can be used to quickly and high-throughput identify the lysis patterns of unknown phages against Klebsiella pneumoniae.

[0020] The third object of the present invention is to provide a method for identifying a phage receptor binding protein (RBP) that specifically recognizes Klebsiella pneumoniae CPS. It is well known that phages interact with CPS on Kp through the RBP they carry. Therefore, it is crucial to identify RBP that specifically binds to CPS on Kp. The present invention relies on the above platform technology constructed to successfully identify phage RBP that specifically recognizes Klebsiella pneumoniae CPS. Specifically, the method includes:

[0021] (1) classifying the phages in the phage library in the method according to the first object of the present invention according to the modules;

[0022] (2) Annotate the genomes of phages with a genome structure similarity of more than 70% but classified into different modules, draw a phage genome structure map, divide the differential structural genes, and compare the amino acid sequences of all differential genes to construct a phylogenetic tree;

[0023] (3) Analyze the correlation between phage differential protein clusters and their modules to find potential RBPs that recognize specific K. pneumoniae CPS types;

[0024] (4) The RBP identified in step (3) is expressed and purified, and its interaction with Klebsiella pneumoniae of different CPS types as a host is detected to determine the receptor range of the RBP; at the same time, the corresponding RBP gene is replaced on the phage genome, and the receptor specificity of the recombinant phage is detected to obtain a phage RBP that specifically recognizes Klebsiella pneumoniae CPS.

[0025] Based on the above, it can be seen that similar genome structure is not a factor that determines the phage lysis mode, and the method of the present invention can obtain phages with similar genome structures but clustered in different lysis modes through cluster analysis, suggesting that there may be potential RBPs in these phages. With this purpose, the inventors of the present invention further analyzed the correlation between phage differential protein clusters and different interaction relationship modules based on the obtained differential structural gene component phylogenetic tree, and preliminarily obtained potential RBPs.

[0026] In the present invention, genome annotation can be performed using methods known in the art. In a specific embodiment of the present invention, prokka is used to predict and preliminarily annotate ORFs. Further protein annotation integrates the following results: 1. Alignment annotation using blastp and VOGDB; 2. NCBI CDD selects pfam and TIGR database annotation; 3. KEGG database annotation.

[0027] In the present invention, the phage genome structure map can be drawn using methods known in the art, such as using easyfig to draw the structure map.

[0028] In the present invention, the differential structural genes can be divided by methods known in the art. In a specific embodiment of the present invention, the differential structural genes are selected in combination with the following information: 1. After the structural protein genes in the phage genome of the same genus are arranged in sequence, they are compared, and blastn can be used. 2. The similarity value is output, and the sequence with a similarity value lower than 70% is determined to be a differential gene sequence.

[0029] In the present invention, the method for constructing a phylogenetic tree is as described above, for example, using mafft to align the divided differential structural gene sequences, using trimal to trim the sequences, using IQTree to build a tree, and visualizing it in iTOL. The underlying logic is to use the maximum likelihood method to build a phylogenetic tree based on the regions that can be aligned in all RBP sequences, and observe the clustering relationship between RBP sequences. In the data analyzed by the present invention, clustering is directly performed based on the sequence similarity of multiple sequence alignments, and clustering relationships similar to the phylogenetic tree can also be observed. Therefore, using other commonly used maximum likelihood method software to build a tree, similar clustering results can usually be obtained.

[0030] It should be noted that when the candidate RBP sequences used for tree building differ too much, for example, when a protein sequence that is not actually an RBP is added when selecting candidate sequences, the tree building results may be disturbed. You can judge whether the sequence that is not the target of analysis may be included by paying attention to whether the pruned sequence is too short, the similarity matrix between RBP sequences, and whether the clustering results displayed by the evolutionary tree are obviously abnormal, and recheck the genomic location, functional annotation, and sequence similarity of the coding gene of the sequence to evaluate whether it should be excluded.

[0031] Furthermore, we observed the clustering of RBP protein sequences in the phylogenetic tree and whether they were consistent with the pattern of phage infection hosts (M1-M6) from which they were derived, so as to analyze the correlation between phage differential protein clusters and the modules they belonged to, and find RBPs that could potentially identify specific Klebsiella pneumoniae CPS types.

[0032] After obtaining the potential RBP, the inventors further verified whether the RBP was the expected RBP, that is, a phage RBP that specifically recognized Klebsiella pneumoniae. The inventors constructed various Klebsiella pneumoniae CPS receptor deletion mutants (for example, including but not limited to KL1, KL2, KL20, KL47, KL57 and KL64). In addition, the inventors also expressed and purified the potential RBPs initially obtained, and determined the receptor range of the RBPs through a spot plate experiment, that is, whether the potential RBP of the phage classified in a certain module interacts only with the host bacteria of the module. For example, if the potential RBP is a phage RBP that potentially specifically recognizes KL1-type Klebsiella pneumoniae, then whether the potential RBP only interacts with the KL1 host bacteria, but not with its other capsular host bacteria; for example, if the potential RBP is a phage RBP that potentially broadly recognizes KL1 and KL2-type Klebsiella pneumoniae, then whether the potential RBP only interacts with KL1 and / or KL2-type host bacteria, but not with other capsular host bacteria. At the same time, the corresponding RBP gene was replaced on the phage genome to detect the receptor specificity of the recombinant phage, that is, the RBP that potentially recognizes the first KL type was replaced with the RBP that potentially recognizes the second KL type, and the phage will no longer have the ability to recognize the first KL type. For example, the RBP that potentially recognizes the KL1 type was replaced with the RBP that potentially recognizes the KL2 type, and the recombinant phage will no longer have the ability to recognize the KL1 type. In this way, the corresponding relationship between RBP and various receptors of Klebsiella pneumoniae was fully characterized, and an RBP element library with clear surface receptors and clear functions was established.

[0033] In the present invention, the construction of Klebsiella pneumoniae mutant strains expressing various Klebsiella pneumoniae CPS types and the replacement of the corresponding RBP gene on the phage genome to construct a recombinant phage are both conventional technical means in the art and will not be described in detail in the present invention.

[0034] The present invention finally determines the phage RBP that specifically recognizes Klebsiella pneumoniae CPS through bidirectional identification.

[0035] In some embodiments, the position corresponding to the bacteriophage RBP that specifically recognizes Klebsiella pneumoniae CPS on the genome starts from the beginning of the terminase large subunit terL gene.

[0036] In some embodiments, the bacteriophage is selected from the genera Webervirus, Jedunavirus, Przondovirus, and Drulisvirus.

[0037] In some embodiments, the bacteriophage belongs to Webervirus, and the position corresponding to its RBP on the genome is based on the beginning of the terminase large subunit terL gene, ranging from about 19600 nt to 23000 nt.

[0038] In some embodiments, the bacteriophage belongs to Jedunavirus, and the position corresponding to its RBP on the genome is based on the beginning of the terminase large subunit terL gene, ranging from about 25500 nt to 28100 nt.

[0039] In some embodiments, the bacteriophage belongs to Przondovirus, and the position corresponding to its RBP on the genome is based on the beginning of the terminase large subunit terL gene, ranging from about 36400 nt to 39500 nt.

[0040] In some embodiments, the bacteriophage belongs to Drulisvirus, and the position corresponding to its RBP on the genome starts from the beginning of the terminase large subunit terL gene and ranges from about 3600 nt to 5600 nt.

[0041] The fourth object of the present invention is to provide a bacteriophage RBP that can specifically recognize Klebsiella pneumoniae CPS.

[0042] A fifth object of the present invention is to provide the use of the bacteriophage RBP as described above in identifying the lysis pattern of bacteriophage against Klebsiella pneumoniae.

[0043] A sixth object of the present invention is to provide a method for high-throughput identification of the lysis pattern of bacteriophage to Klebsiella pneumoniae, the method comprising:

[0044] (1) determining the gene sequence of the phage RBP to be identified, and constructing a phylogenetic tree based on the amino acid sequence translated from the gene sequence, wherein the phylogenetic tree can form different cluster groups;

[0045] (2) The clusters formed by the phylogenetic tree are matched with the modules in the platform according to the second object of the present invention to obtain the corresponding modules of different clusters, thereby determining the lysis pattern of the bacteriophage against Klebsiella pneumoniae.

[0046] The method of the present invention can identify the lysis pattern of phages with high throughput. In some embodiments, the throughput of the method includes at least one phage to be identified.

[0047] According to the method of the present invention, when the genus of the unknown phage is known, for example, when it belongs to Webervirus, Jedunavirus, Przondovirus and Drulisvirus, the RBP sequence can be directly obtained by the position of RBP, and the RBP sequence can be constructed by carrying out a phylogenetic tree with the RBP sequence in the platform library of the present invention, so as to judge the Klebsiella pneumoniae lysis pattern of the phage according to the method of the present invention. When some unknown phages do not belong to Webervirus, Jedunavirus, Przondovirus and Drulisvirus, the determination of RBP can be carried out according to the method of the third aspect of the present invention, wherein the unknown phage can be divided into modules with the phage in the phage library after obtaining genome information. Therefore, even if the method of the present invention has only one phage, its Klebsiella pneumoniae lysis pattern can be efficiently known.

[0048] The method of constructing a phylogenetic tree has been introduced in detail above. In order to avoid unnecessary repetition, it will not be repeated again.

[0049] It can be understood that since the RBP sequences contained in each cluster group and the RBP sequences contained in each module are known, the cluster groups formed by the phylogenetic tree can be matched with the modules in the platform described in the second purpose of the present invention according to the sequences, that is, they can be clustered, and the support degree of the corresponding branch nodes is greater than 90%, which is used to determine the lysis mode of the bacteriophage.

[0050] A seventh object of the present invention is to provide a bacteriophage identified by the above method. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 The phylogenetic relationships of the phages in the phage library of the present invention are shown.

[0052] Figure 2 The diverse morphologies of myo-tailed phages (n=8), long-tailed phages (n=7), and self-assembled short-tailed viruses (n=4) were displayed.

[0053] Figure 3 Graph showing the phylogenetic relationship of K. pneumoniae strains in the host bacterial library of the present invention.

[0054] Figure 4 The interaction matrix between the phages in the phage library of the present invention and the host bacteria in the host bacteria library is shown.

[0055] Figure 5The results of alignment of structural proteins from the genera Webervirus, Jedunavirus, Przondovirus and Drulisvirus by mafft (v.4.787) are shown.

[0056] Fig. 6A A phylogenetic tree was constructed by IQtree (v.2.1.4-beta) of the 55 identified bacteriophage tail protein sequences.

[0057] Figure 6B Results of the interaction of the wild-type strain and the acapsulated mutant binding Klebsiella pneumoniae with phage are shown.

[0058] Figure 6C The putative RBPs were shown to be specific in polysaccharide degradation.

[0059] Figure 7 Expression and purification electropherograms of putative RBPs from clusters I, III, V, and VI are shown.

[0060] Figure 8 The testing of the platform provided by the present invention is shown.

[0061] Fig. 9 The identification of an unknown phage lysis pattern using the platform of the present invention is shown.

[0062] Fig.10 The lytic range of engineered phages with expanded host capsule preference is shown. DETAILED DESCRIPTION

[0063] Example 1 Construction of a platform for identifying phage lysis patterns

[0064] 1.1 Construction of phage library

[0065] For constructing a Klebsiella pneumoniae (K.pneumoniae) phage library representing a wide host range, the present invention uses a group of 39 K.pneumoniae strains comprising 15 capsule serotypes (KLs) and 23 sequence types (STs) as hosts for phage separation, including most of the epidemic clones. After a large number of separations were carried out from hospital wastewater and domestic sewage, the present embodiment obtained more than 200 phages. Genome sequencing identified 114 dsDNA Caudovirales phages with non-redundant sequences, and the genome size range was 27,814 to 350,174bp. Using multiple databases to perform genome annotation shows that these phages carry 48 to 634 predicted coding sequences, without antibiotic resistance genes or virulence factors. Among them, three phages (RCIP0057, RCIP0098 and RCIP0114) encode integrase genes, indicating that they are mild phages. To evaluate the phylogeny of phages, this example retrieved 121 K. pneumoniae phage genomes from the National Center for Biotechnology Information (NCBI) Refseq database, merged them with the genome of the phage library of this example, and constructed a protein distance tree ( Figure 1 The phages in the phage library of this example represent seven families (Ackermannviridae, Autographiviridae, Demerecviridae, Drexlerviridae, Peduoviridae, Siphoviridae, and Straboviridae) and ten genera (Taipeivirus, Drulisvirus, Przondovirus, Sugarlandvirus, Webervirus, Eganvirus, Jiaodavirus, Slopekvirus, Alcyoneusvirus, and Jedunavirus), covering most of the reported Klebsiella phage families and genera. Morphological observations using transmission electron microscopy (TEM) supported this phylogenetic classification, Figure 2 The diverse morphologies of myo-tailed phages (n=8), long-tailed phages (n=7), and self-assembled short-tailed viruses (n=4) are shown.

[0066] 1.2 Construction of host bacterial library

[0067] This example constructs a large host set, including 238 K. pneumoniae strains sequenced on the Illumina platform. Based on the core genes (Roary) of these 238 strains, IQTree constructs a phylogenetic tree ( Figure 3). Six branches with greater than 90% bootstrap support and dozens of deeply branched phylogenetic lineages indicated the phylogenetic diversity of the strain set. Multilocus sequence typing of seven genes (gapA, infB, mdh, pgi, phoE, rpoB, and tonB) and capsular polysaccharide (K) antigen typing at the K locus were performed using Kleborate version v.2.2.0. The strains belonged to 105 sequence types, 94 of which were known and 11 were unknown, including most pandemic clones such as ST258, ST11, ST101, ST23, ST65, and ST25, and 55 prevalent KLs, including KL1, KL2, KL20, KL47, KL57, and KL64. Antibiotic resistance genes (ARGs) were searched in each genome by ResFinder, and a total of 53 ARGs associated with nine classes of antibiotics were found. Among them, the most prevalent ARGs, present in more than 30% of the genomes, confer resistance genes to β-lactams, fluoroquinolones, macrolide-lincomycin-streptomycin B (MLSB), chloramphenicol, sulfonamides, tetracycline, and trimethoprim.

[0068] 1.3 Phage-host interactions

[0069] The host range of each phage in the library was tested on 238 K. pneumoniae strains by dot plaque assay, resulting in a large phage-host interaction matrix ( Figure 4 ), which included 3,021 lytic phenotypes among 27,132 possible interactions. Overall, 223 of the 238 strains (93.70%) were lysed by at least one of the 114 phages, demonstrating the large capacity of the phage library of the present invention.

[0070] Hierarchical clustering of the host-phage bidirectional matrix using the R heatmap package ( Figure 4 ) Further inspection revealed that the lytic interactions were divided into five distinct modules, named M1 to M5. Notably, there was a strong correlation between these modules and the capsule types (KLs) of Klebsiella pneumoniae strains ( Figure 4). These five typical modules can be divided into two groups according to their coverage of host KLs. One group of modules with a broad host range contains only M1, which contains 15 phages that can lyse multiple KLs, mainly KL1, KL2, and KL57. The other group of modules with high specificity for KLs includes M2, M3, M4, and M5. M2 contains 14 phages that mainly lyse KL1, M3 consists of 20 phages that lyse KL2, M4 contains 8 phages that lyse KL64, and M5 contains 17 phages that lyse KL57. It is worth noting that 40 phages did not correspond to the KLs of their hosts, and these phages were defined as M6.

[0071] Example 2 Determination of RBP

[0072] 2.1 Determination of the correlation between phage lysis pattern and phage phylogeny

[0073] In order to clarify the correlation between the phylogeny and lysis modules of phages, the present invention constructed a whole genome phylogenetic tree of 114 phages and calculated the average nucleotide homology (ANI) of each phage. It was observed that five clusters containing more than five strains corresponded to five genera (Webervirus, Jedunavirus, Przondovirus, Drulisvirus and Slopekvirus), each of which contained strains with close phylogenetic relationships (ANI>70%). 97 phages were selected from these five genera for further study. Interestingly, phages in the same genus exhibited a diverse host range for different KLs. Specifically, phages of the Webervirus and Drulisvirus genera showed lysis modules of M1, M2, M3, M5 and M6; Jedunavirus contained lysis modules of M1, M3, M5 and M6; Przondovirus had all confirmed lysis modules. It is worth noting that Slopekvirus is an exception, and its seven phage strains are not specific to host serotypes.

[0074] 2.2 Identification of potential RBPs

[0075] Bacteriophages recognize their hosts through RBPs on the fibers or tail fibers at the distal end of their tails. RBPs bind to specific surface receptors on bacteria, which may be components of bacterial outer membrane proteins, lipopolysaccharides, or bacterial capsules, cilia, or flagella. To understand the factors that determine the RBPs of the lytic module, this example performed a comparative genomic analysis of phages M2 to M5 belonging to the same genus to determine their preferences. This example used mafft (v.4.787) to align the structural proteins from the Webervirus, Jedunavirus, Przondovirus, and Drulisvirus genera ( Figure 5 ). Genes encoding morphologically related proteins in the same genus shared more than 70% nucleotide homology within a given genus, except for one gene, the predicted phage tail protein gene. In the same genus, tail proteins of phages belonging to the same lytic module shared high homology within the same module. A phylogenetic tree was constructed by IQtree (v.2.1.4-beta) based on the 55 phage tail protein sequences identified in this study. These proteins formed six monophyletic groups, named I, II, III, IV, V, and VI ( Fig. 6A ). Cluster I of Drulisvirus and Webervirus and Cluster II of Przondovirus correspond to the M2 lysis module, which specifically lyses the KL1 strain. Cluster III contains phages of Drulisvirus, Webervirus, and Jedunavirus, and Cluster IV contains the M3 lysis module of Przondovirus, which specifically lyses the KL2 strain. Cluster V contains Przondovirus, represents the M4 module, and lyses the KL64 strain. Finally, Cluster VI contains Drulisvirus, Przondovirus, Webervirus, and Jedunavirus, represents the M5 module, and lyses the KL57 strain. The strong correlation between these putative phage tail protein clusters and the lysis modules suggests that these phage tail proteins are RBPs responsible for host recognition. It is noteworthy that other putative phage tail proteins within the same genus, which are found in conserved regions within a given genus, are not associated with the lysis module.

[0076] 2.3 Identification of RBPs that determine host capsule preference of Klebsiella pneumoniae phages

[0077] To test the hypothesis that putative phage tail proteins acting as RBPs are capable of binding the capsular polysaccharide (CPS) of K. pneumoniae, this example compares binding of the wild-type strain and acapsular mutant.

[0078] First, the present invention generated CPS-deficient mutants of KL1, KL2, KL57, and KL64 by serial passage on Luria–Bertani (LB) plates. Phages with possible RBPs from the six clusters were selected and spotted on agar with wild-type strains or acapsular mutants. Phage RCIP0039, which has a putative phage tail protein in cluster I, and phage RCIP0024 in cluster II formed visible spots on the KL1 wild-type strain lawn, but no spots on the acapsular mutant plates ( Figure 6B ). Similarly, phage RCIP0035 in cluster III and phage RCIP0001 in cluster IV formed visible plaques on agar of wild-type KL2 strain, but no plaques on plates of acapsular mutants. Phage RCIP0028 in cluster V and phage RCIP0079 in cluster VI formed visible plaques on agar of wild-type KL64 and KL57 strains, but no plaques on plates of acapsular mutants. These results suggest that some components of the bacterial capsule are binding receptors for these phages.

[0079] To characterize the polysaccharide degradation function of RBPs in different clusters, candidate genes were expressed and purified from putative RBPs in clusters I, III, V, and VI. The molecular weights of RBPI_RCIP0039, RBPIII_RCIP0035, RBPV_RCIP0028, and RBPVI_RCIP0079 were 69.9, 60.8, 110, and 82.3 kDa, respectively. Figure 7 ), and spot tests were performed using KL1, KL2, KL64, and KL57 strains. RBPI_RCIP0039 produced a translucent spot on plates with KL1 strain 8-105, but not on plates with strains from other KLs. Similarly, RBPIII_RCIP0035, RBPV_RCIP0028, and RBPVI_RCIP0079 showed translucent spots on the corresponding KL plates, but not on plates with other KLs.

[0080] Taken together, these results suggest that these RBPs are specific in glycan degradation ( Figure 6C ).

[0081] Example 3 Validation of potential RBPs in Example 2 in the M1 module

[0082] Most of the phages of the four genera of Webervirus, Jedunavirus, Przondovirus and Drulisvirus in the library of the present invention lyse specific KL strains. However, some phages were identified as belonging to the M1 and M6 lysis modules, showing a preference for multi-capsular serotypes. This example identified 35 potential RBPs, and a phylogenetic tree was constructed using MEGA software based on the above 55 RBPs and the protein sequences of these 35 potential RBPs, and seven different clusters were observed ( Figure 8 ). Among them, a new cluster was observed, which was named Cluster VII. This cluster contains 13 potential RBPs from Webervirus phages with the M6 ​​cleavage module, as well as the six clusters mentioned above. In addition, 14 potential RBPs from M1 phages, which can cleave Kp of KL1, KL2 and KL57, are clustered with specialized phages in clusters I, III, IV and VI. Specifically, the potential RBPs of M1 phage have very high amino acid homology (from 90% to 99.47%) with the RBPs of specialized phages in the same genus. Similarly, seven potential RBPs with the M6 ​​module are clustered with specific RBPs, belonging to clusters II, III, IV, V and VI, respectively, and have high amino acid homology (from 97.35% to 99.31%).

[0083] Two potential RBPs were expressed from M1 lytic module phage and M6 lytic module phage. The RBP from phage RCIP0104 belongs to cluster I, which includes RBPs that specifically target the KL1 capsule type, while the RBP from phage RCIP0046 belongs to cluster III, which includes RBPs that are specific for the KL2 capsule type. Experiments performed on these two RBPs demonstrated that both formed translucent spots on bacterial plates of KL1 and KL2 capsule type strains. In contrast, on plates of KL57 strain, no spots were observed when these RBPs were applied, indicating that the RBPs have specific binding to KL1 and KL2 capsular polysaccharides and are unable to degrade KL57 capsular polysaccharides. Extensive analysis showed that there is high amino acid sequence identity between RBPs of broad-host range phages and narrow-host range phages, such as 96.93% identity between RBP-RCIP0104 and RBP-RCIP0039, despite 19 amino acid substitutions. Similarly, the identity between RBP-RCIP0046 and RBP-RCIP0035 is 97.23% with 16 amino acid differences. These sequence variations may explain the specificity and functional differences observed between broad-host range and narrow-host range RBPs.

[0084] Example 4 Prediction of phage lysis host range

[0085] The purpose of this example is to use the platform constructed by the present invention to verify the lysis pattern of phages other than the present invention. After a large amount of separation work from hospital wastewater, this example obtained a phage RCIP0146. Genome sequencing identified it as a dsDNA Caudovirales phage, Autographiviridae family, Drulisvirus genus, with a genome size of 43,270bp. According to Example 2, its RBP was determined to be ORF00051. The results of protein cluster analysis showed that the RBP was clustered in Cluster II ( Fig. 9 ). According to the method established by the present invention, it is determined that the phage has the ability to lyse the Klebsiella pneumoniae KL1 strain.

[0086] Thirty Klebsiella pneumoniae KL1 strains were randomly selected to verify the lysis ability of RCIP0146. RCIP0146 was able to lyse 28 strains, proving that the method of this platform can accurately determine the lysis host range of bacteriophages.

[0087] Example 5 Engineered phages with variable host capsule tropicity

[0088] The newly identified receptor binding protein provides a toolbox for customizing phages for targeting different types of Klebsiella capsules. This example assumes that exchanging RBPs may result in phages with altered host capsule preferences. To verify this hypothesis, this example selected phage RCIP0035 in the library of the present invention. The genome length of RCIP0035 is 43,252bp, the GC content is 54.24%, and it does not contain tRNA, antibiotic resistance, toxins, virulence factors or lysogeny gene clusters. Genomic analysis showed that phage RCIP0035 belongs to the genus Drulisvirus in the family Autographiviridae. Subsequently, the present invention successfully constructed a recombinant phage by replacing the RBPIII-RCIP0035 gene of phage RCIP0035 with the RBPIII-RCIP0082 gene of phage RCIP0082, which is named RCIP0035ΔrbpIII::rbp0082 in this example. This novel phage exhibited a shift in host specificity; it lost the ability to lyse its original host with the KL2 capsule but acquired new lytic activity against strains with the KL57 capsule.

[0089] Specifically,

[0090] In order to obtain RBP-replacement recombinant phage, the plasmid pColadRED carrying kanamycin resistance and L-arabinose-induced λ-red recombination system was constructed. Phage RCIP0035 belongs to the genus Drulisvirus in the family Autographiviridae. The number of Drulisvirus genus in the phage library is relatively large. RCIP0035 was selected as the chassis phage, with a genome length of 43,252bp, a GC content of 54.24%, and no tRNA, antibiotic resistance, toxin, virulence factor or lysogeny gene clusters. The genomic DNA of RCIP0035 was used as a template to amplify the upstream and downstream regions of the gene (about 500bp), and the genomic DNA of RCIP0082 was used as a template to amplify the replaced tail fiber gene, and these genes were combined to form pColadRED using Gibson assembly to obtain a recombinant plasmid. The plasmid was introduced into the host strain Kp7-50 of RCIP0035 by electroporation, and then screened on LB agar supplemented with 50μg / ml kanamycin. The host carrying the recombinant plasmid was cultured in LB medium containing 1mM L-arabinose to induce the expression of the recombinase system. The induced bacterial culture in the logarithmic growth phase (optical density at 600nm = 0.5) was incubated with various titers of RCIP0035 at 37°C for 30 minutes. The mixture of phage and bacteria was inoculated on the top agar and then covered on the LB agar plate. Using the primer set RCIP0035tailfiber2-up-F and RCIP035 tail fiber-2-up-R, a single plaque was selected to screen the recombinant phage with tail fiber replacement. The obtained recombinant phage was further purified and separated on the host strain RCIP0082. In order to eliminate the interference of RCIP0035, the process was repeated two to three times using the agar overlay method. Recombinant phage RCIP0035ΔrbpIII::rbp0046 with the RCIP0046 tail fiber and reconstructed phage RCIP035ΔrbpIII::rbp0104 with the RCIP1004 tail fiber were constructed by a similar method.

[0091] The lytic capacity of wild-type and recombinant phages in different hosts was evaluated using a double agar overlay plaque assay. Briefly, wild-type and recombinant phages were serially diluted in triplicate, and 100 μL of each dilution was inoculated with the desired test strain (10 8 CFU / mL, 100 μL) and then poured onto the surface of a dried and labeled pad. After incubation at 37°C overnight, the phage titer was determined by counting the number of plaques at the appropriate dilution. The initial titer of phage was standardized to 8×10 9 [PFU] / mL.

[0092] By replacing the RBPIII-RCIP0035 gene of phage RCIP0035 with the RBPIII-RCIP0082 gene of phage RCIP0082, a recombinant phage was successfully constructed, which was named RCIP0035ΔrbpIII::rbp0082 in this example. This new phage exhibited a shift in host specificity; it lost the ability to lyse the initial host with the KL2 capsule, but gained a new lytic activity to lyse strains with the KL57 capsule.

[0093] Example 6 Engineered phage with expanded host capsule preference

[0094] The present embodiment investigates whether the replacement of RBPIII-RCIP0035 gene with multivalent RBPs with cross-capsule lysis potential can produce phage with extended host capsule preference. The present embodiment replaces the RBPIII-RCIP0035 gene of phage RCIP0035 with the RBP gene of RCIP0046 and RCIP0104 phage. These donor phages have a wide range of hosts, can crack the bacterial strain with KL1, KL2 and KL57 capsule types, and RBPIII-RCIP0046 and RBPI-RCIP0104 show the ability of degrading KL1 and KL2 capsule polysaccharides. It is worth noting that RBPIII-RCIP0046 and RBPIII-RCIP0035 belong to cluster III, belong to cluster I with RBPI-RCIP0104, and the similarity with RBPIII-RCIP0035 is less than 30%. The obtained recombinant phages were named RCIP0035ΔrbpIII::rbp0046 and RCIP0035ΔrbpIII::rbp0104, respectively.

[0095] RCIP0035ΔrbpIII::rbp0046 and RCIP0035ΔrbpIII::rbp0104 were tested for their lysis efficacy against a diverse range of bacterial strains, including 14 KL1 strains, 18 KL2 strains, and 9 KL57 strains. The recombinant phage RCIP0035ΔrbpIII::rbp0046 not only maintained its lysis efficacy against KL2 capsulated strains, but also expanded its lysis range against KL1 and KL57 capsulated strains, such as Fig.10Similarly, RCIP0035ΔrbpIII::rbp0104 exhibited an extended host range, maintaining activity against KL2-capsulated strains and expanding its lytic capacity to include strains of KL1 and KL57 capsular types. Interestingly, the lytic spectrum of the recombinant phages was not completely consistent with that of their donor phages. For example, compared to its donor phage RCIP0046, RCIP0035ΔrbpIII::rbp0046 was able to lyse three additional KL2-type strains ( Fig. 9 ). This suggests that these hybrid phages not only acquire the broad host range of their donors but may also develop unique lytic properties, highlighting the complexity of phage-host interactions and the novel and enhanced capabilities that engineered phages may exhibit in bacterial lysis.

[0096] The preferred embodiments of the present invention are described in detail above. However, the present invention is not limited to the specific details in the above embodiments. Within the technical concept of the present invention, a variety of simple modifications can be made to the technical solution of the present invention, and these simple modifications all belong to the protection scope of the present invention.

[0097] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present invention will not further describe various possible combinations.

[0098] In addition, various embodiments of the present invention may be arbitrarily combined, and as long as they do not violate the concept of the present invention, they should also be regarded as the contents disclosed by the present invention.

Claims

1. A method for constructing a platform for identifying phage lysis patterns, the method comprising: (1) constructing a host bacterial library of Klebsiella pneumoniae (K. pneumoniae) and a phage library with Klebsiella pneumoniae as the host bacterial library; (2) detecting the interaction relationship between each phage in the phage library and each Klebsiella pneumoniae in the host bacteria library to obtain a phage-host interaction matrix; (3) performing a bidirectional cluster analysis on the phage-host interaction matrix to obtain a plurality of interaction relationship modules between phages and host bacteria; the modules include modules with a broad host range and / or modules with a specific host range; Wherein, the lysis mode refers to that the phage targets the capsular serotype (KL) of Klebsiella pneumoniae.

2. The method according to claim 1, wherein: The host bacterial library includes at least 50 strains, at least 75 strains, at least 100 strains, at least 125 strains, at least 150 strains, at least 175 strains, at least 200 strains, or at least 225 strains of Klebsiella pneumoniae; and / or The KLs covered by the host bacterial library include at least KL1, KL2, KL57 and KL64; Preferably, the KLs covered by the host bacterial library include at least Klebsiella pneumoniae of KL1, KL2, KL20, KL47, KL57 and KL64; More preferably, the host bacterial library contains at least 55 KLs, including KL1, KL2, KL57 and KL64.

3. The method according to claim 1 or 2, wherein: The phage library comprises at least 20 strains, at least 30 strains, at least 40 strains, at least 50 strains, at least 60 strains, at least 70 strains, at least 80 strains, at least 90 strains, at least 100 strains, or at least 150 strains of phage, at least 200 strains of phage, or at least 250 strains of phage; and / or The lysis patterns of the phage library include at least KL1, KL2, KL57 and KL64; Preferably, at least KL1, KL2, KL20, KL47, KL57 and KL64 are included; More preferably, at least 55 kinds of KL are included, including KL1, KL2, KL57 and KL64.

4. The method according to any one of claims 1 to 3, wherein: The module includes at least one of the following: 1) M1 modules capable of cleaving at least two of KL1, KL2, and KL57; 2) Specific cleavage of the M2 module of KL1; 3) Specific cleavage of the M3 module of KL2; 4) Specific cleavage of the M4 module of KL64; 5) Specific cleavage of the M5 module of KL57; Preferably, the modules include: an M2 module, an M3 module, an M4 module, an M5 module and an optional M1 module; More preferably, the modules include: an M2 module, an M3 module, an M4 module, an M5 module and an M1 module.

5. A platform for identifying phage lysis patterns constructed by the method according to any one of claims 1 to 4.

6. A method for identifying a bacteriophage receptor binding protein (RBP) that specifically recognizes Klebsiella pneumoniae capsular polysaccharides (CPS), the method comprising: (1) classifying the phages in the phage library according to any one of claims 1 to 4 according to the modules; (2) Annotate the genomes of phages with a genome structure similarity of more than 70% but classified into different modules, draw a phage genome structure map, divide the differential structural genes, and compare the amino acid sequences of all differential genes to construct a phylogenetic tree; (3) Analyze the correlation between phage differential protein clusters and their modules to find RBPs that can potentially recognize specific K. pneumoniae CPS types; (4) The RBP identified in step (3) is expressed and purified, and its interaction with Klebsiella pneumoniae of different CPS types is detected to determine the receptor range of the RBP; at the same time, the corresponding RBP gene is replaced on the phage genome, and the receptor specificity of the recombinant phage is detected to obtain a phage RBP that specifically recognizes the CPS of Klebsiella pneumoniae.

7. The phage RBP identified by the method of claim 6, and its use in identifying the lysis pattern of phage against Klebsiella pneumoniae.

8. A method for high-throughput identification of the lysis pattern of bacteriophage against Klebsiella pneumoniae, the method comprising: (1) determining the gene sequence of the RBP expressed on the phage to be identified, and constructing a phylogenetic tree based on the amino acid sequence translated from the gene sequence, wherein the phylogenetic tree can form different cluster groups; (2) The clusters formed by the phylogenetic tree are matched with the platform modules described in claim 5 to obtain the corresponding modules of different clusters, thereby determining the lysis pattern of the bacteriophage against Klebsiella pneumoniae.

9. The method according to claim 8, wherein: The throughput of the method includes at least one phage to be identified.

10. A bacteriophage identified by the method of claim 8 or 9.

Images