English
Tiếng Việt
(en)Methods are provided for analyses of restriction endonucleases.
1.ApplicationNumber: US-201314144361-A
1.PublishNumber: US-2015184221-A1
2.Date Publish: 20150702
3.Inventor: MEIDLINGER MARC
4.Inventor Harmonized: MEIDLINGER MARC(DE)
5.Country: US
6.Claims:
(en)Methods are provided for analyses of restriction endonucleases.
7.Description:
(en)REFERENCE TO SEQUENCE LISTING
This application contains a sequence listing submitted by EFS-Web, thereby satisfying the requirements of 37 C.F.R. §§1.821-1.825. The Sequence Listing, created on Dec. 15, 2013, is named “Sequence listing P0031.01.US_ST25”.
BACKGROUND
Cyanobacteria, also known as blue-green algae, are photosynthetic bacteria widespread in marine and freshwater environments. Cyanobacteria have simple growth requirements and grow to high densities by using light, carbon dioxide, and other inorganic nutrients for growth. They are useful hosts for the production of a wide range of compounds of interest including biofuels and other commodity chemicals, for example.
A barrier to using cyanobacteria for the production of compounds of interest is that efficient introduction of engineered vectors in cyanobacteria is hampered, or completely prohibited, due to the presence of restriction nucleases, see Thorsten H. et. al., Ch. 24, Synthetic Biology in Cyanobacteria: Engineering and Analyzing Novel Functions, In: Methods in Enzymology, Academic Press, 2011, Vol. 497, pp. 539-579. These nucleases can significantly decrease or completely prevent the uptake of exogenous DNA into cyanobacterial cells, see for example Ruffing A M. et. al., Biotech. Bioeng., 2012, Vol. 109, pp. 2190-2199. Enzymes exhibiting nuclease activity are usually either endonucleolytic (an endonuclease) or exonucleolytic (an exonuclease) and may be either sugar specific or non-sugar specific. A catalog of restriction enzymes and their recognition sites is compiled and maintained by New England Biolabs in Ipswich, Mass. and can be accessed at http://rebase.neb.coil (REBASE).
Various restriction endonucleases exhibit restriction activity to varying degrees in various buffer conditions. Buffer conditions conunonly used for known restriction enzymes are mostly covered across four different buffers available through New England Biolabs, P1, P2, P3 and P4 buffers. Their compositions can be found at REBASE and are well known to those having skill in the art.
Cyanobacteria often contain one or more sequence-specific restriction endonucleases. These nucleases are almost always associated with their cognate methyltransferase. The methyltransferase methylates the host chromosomal and plasmid DNA to provide protection against its associated restriction enzyme. Endogenous cyanobacterial DNA is thus protected against specific nuclease activity by specific methylation, but foreign DNA is not methylated and is therefore targeted for destruction by the cyanobacterial nucleases, see Soper, B W et. al, J. Bacteriology 1994, 176(17):556-5570 and Elhai, J, et. al, J. Bacteriology 1997, 179(6): 1998-2005.
SUMMARY
Disclosed herein are methods that provide for the identification of cyanobacterial restriction endonucleases that are present in crude extracts. The method is based on specific PCR products (ROligos) which are designed with unique restriction endonuclease (“REN”) recognition sequences leading to two easily distinguishable bands upon digestions with cyanobacterial crude extract. In addition, incubation of methylated ROligos in cyanobacterial extracts allows the determination of whether endonucleases present can be rendered ineffective by the provided methylation type. Hence, any methyltransferase that provides the desired type of methylation can be used to methylate and protect DNA in vitro or in vivo prior to transformation.
In an aspect, a method for identifying restriction endonucleases in a cyanobacterial cell is disclosed having the steps of a) incubating oligonucleotides from a ROligo library with a cyanobacterial cell extract from said cyanobacterial cell, wherein each oligonucleotide comprises a recognition sequence of a cyanobacterial restriction endonuclease, and b) analyzing digestion products of said oligonucleotides for digestion at said recognition sequences, and c) identifying restriction endonucleases in said cyanobacterial cell extract by analyzing digestion products of said oligonucleotides. In an embodiment, the method is can be used on cyanobacterial cells including Prochlorococcus, Synechocystis, Synechocococus, Chroococcales, Cyanobium, Oscillatoriales, Cyanobacterium, Pleurocapsales, Geitlerinema, Phormidiumn, Euhalothece, Anabaena, Lyngbya, Spirulina, Nostoc, Pleurocapsa , and Leptoiyngbya . In another embodiment, the method discloses restriction endonucleases of step c) that are analyzed through depicting the results of digestion patterns at recognition sequences of said oligomcleotides from a ROligo library in an infographic table having all incubated oligonucleotides and all restriction endonucleases whose recognition sequences are part of the oligonucleotides, and wherein restriction endonucleases are identified from said cyanobacterial cell extract through the following stepwise manipulations of the digestion patterns depicted in the infographic table including the steps of a) first, restriction endonucleases which do not digest at least one of the oligonucleotides incubated with said cyanobacterial cell extract are removed from the infographic table, and b) second, restriction endonucleases which do not digest oligonucleotides from the ROligo library containing their recognition sequences, but are capable of recognizing other oligonucleotides that have been digested are removed from the infographic table, and c) third, restriction endonucleases in the cyanobacterial extract are identified as the restriction endonucleases not removed from the infographic table. In another embodiment, the uses an infographic table that is a digestion matrix. In an embodiment, the method is disclosed where each oligonucleotide comprises a left arm, a right arm, and a recognition sequence for a cyanobacterial restriction endonuclease wherein the oligonucleotide is produced through using a polymerase chain reaction wherein the left arm is a double stranded polynucleotide and the right arm is a double stranded polynucleotide and wherein the 5′ end of a single stranded reverse primer of said left arm and the 5′ end of a single stranded forward primer of the right arm overlap and make a recognition sequence for a cyanobacterial restriction endonuclease and wherein the left arm and the right arm primers use a template lacking any recognition sites for cyanobacterial restriction endonucleases. In yet another embodiment, the method uses a recognition sequence that is positioned asymmetrically along the length of each oligonucleotide such that digestion by the restriction endonuclease creates two digestion products of different sizes. In an embodiment, the method uses a template that is SEQ ID NO: 1 and wherein said left arm primer and said right arm primer include primers selected from the group consisting of SEQ ID NOs: 5-62.
In an aspect, a method for transfonning cyanobacterial host cells with vectors is disclosed including a) incubating oligonucleotides from a ROligo library with a cyanobacterial cellular extract from the cyanobacterial host cell, wherein each oligonucleotide comprises a recognition sequence of a restriction endonuclease, and b) analyzing digestion products of the oligonucleotides for digestion at the recognition sequences, and c) identifying restriction endonucleases in said cyanobacterial cellular extract by analyzing digestion products of the oligonucleotides, and d) constructing a vector lacking said recognition sequences of the identified restriction endonucleases of step c), and e) transforming said cyanobacterial host cell with said vector of step d). In an embodiment, the method uses restriction endonucleases of step c) that are analyzed through depicting the results of digestion patterns at recognition sequences of said oligonucleotides from a ROligo library in an infographic table comprising all incubated oligonucleotides and all restriction endonucleases whose recognition sequences are part of the oligonucleotides, and wherein restriction endonucleases are identified from the cyanobacterial cell extract through the following stepwise manipulations of the digestion patterns depicted in said infographic table by the following steps: a) first, restriction endonucleases which do not digest at least one of the oligonucleotides incubated with said cyanobacterial cell extract are removed from said infographic table, and b) second, restriction endonucleases which do not digest oligonucleotides containing their recognition sequences, but are capable of recognizing other oligonucleotides that have been digested are removed from said infographic table, and c) third, restriction endonucleases in the cyanobacterial extract are identified as the restriction endonucleases not removed from said infographic table. In another embodiment, the method uses an infographic table that is a digestion matrix. In yet another embodiment, the method uses oligonucleotides wherein each oligonucleotide includes a left arm, a right arm, and a recognition sequence for a cyanobacterial restriction endonuclease, and wherein the oligonucleotide is produced through using a polymerase chain reaction wherein said left arm is a double stranded polynucleotide and the right arm is a double stranded polynucleotide and wherein the 5′ end of a single stranded reverse primer of said left arm and the 5′ end of a single stranded forward primer of said right arm overlap and comprise a recognition sequence for a cyanobacterial restriction endonuclease and wherein the left arm and said right arm primers use a template lacking any recognition sites for cyanobacterial restriction endonucleases. In yet another embodiment, the method uses oligonucleotides having a recognition sequence that is positioned asymmetrically along the length of the each oligonucleotide such that digestion by the restriction endonuclease creates two digestion products of different sizes. In an embodiment, the method uses a template that is SEQ ID NO: 1, and wherein the left arm primer and the right arm primer comprise single stranded polynucleotide primers are selected from the group consisting of SEQ ID NOs: 5-62.
In another aspect a method for transforming cyanobacterial host cells with vectors is disclosed and includes a) incubating oligonucleotides from a ROligo library with a cyanobacterial cellular extract from the cyanobacterial host cell, wherein each oligonucleotide comprises a recognition sequence of a restriction endonuclease, and b) analyzing digestion products of the oligonucleotides for digestion at the recognition sequences, and c) identifying restriction endonucleases in said cyanobacterial cellular extract by analyzing digestion products of the oligonucleotides, and d) methylating recognition sequences of the identified restriction endonucleases in a vector, and e) transforming the cyanobacterial host cell with said vector of step d). In an embodiment, the method uses recognition sequences in a vector that are methylated in step d) and are methylated by methyltransferases. In another embodiment, the method uses methyltransferases that are selected from the group consisting of M.CviPI, M.SssI and M.AvaII. In yet another embodiment, the method of uses restriction endonucleases of step c) that are analyzed through depicting the results of digestion patterns at recognition sequences of the oligonucleotides from a ROligo library in an infographic table comprising all incubated oligonucleotides and all restriction endonucleases whose recognition sequences are part of the oligonucleotides, and wherein restriction endonucleases are identified from the cyanobacterial cell extract through the following stepwise manipulations of the digestion patterns depicted in the infographic table including the steps of a) first, restriction endonucleases which do not digest at least one of the oligonucleotides incubated with the cyanobacterial cell extract are removed from said infographic table, and b) second, restriction endonucleases which do not digest oligonucleotides containing their recognition sequences, but are capable of recognizing other oligonucleotides that have been digested are removed from the infographic table, and c) third, restriction endonucleases in said cyanobacterial extract are identified as the restriction endonucleases not removed from the infographic table. In another embodiment, the method uses an infographic table that is a digestion matrix. In an embodiment, the method uses oligonucleotides wherein each oligonucleotide includes a left arm, a right arm, and a recognition sequence for a cyanobacterial restriction endonuclease, and wherein the oligonucleotide is produced through using a polymerase chain reaction wherein the left arm is a double stranded polynucleotide and the right arm is a double stranded polynucleotide and wherein the 5′ end of a single stranded reverse primer of said left arm and the 5′ end of a single stranded forward primer of the right arm overlap and make a recognition sequence for a cyanobacterial restriction endonuclease and wherein the left arm and the right arm primers use a template lacking any recognition sites for cyanobacterial restriction endonucleases. In yet another embodiment, the method uses oligonucleotides that have recognition sequences that are positioned asymmetrically along the length of each oligonucleotide such that digestion by the restriction endonuclease creates two digestion products of different sizes. In another embodiment, the method uses a template that is SEQ ID NO: 1, and wherein the left arm primer and the right arm primer comprise single stranded polynucleotide primers selected from the group consisting of SEQ ID NOs: 5-62.
In an aspect, a method for identifying restriction endonucleases is disclosed having the steps of a) incubating oligonucleotides including a ROligo library with a cyanobacterial cell extract wherein each oligonucleotide of said ROligo library includes a recognition sequence of a restriction endonuclease, and b) wherein each oligonucleotide of said ROligo library is labeled with a unique marker that is uniquely visualized, and c) analyzing digestion products of said ROligo library of oligonucleotides for digestion at said recognition sequences by visualizing said markers, and d) identifying restriction endonucleases in said cyanobacterial cell extract by correlating said visualized markers of step c) with their corresponding oligonucleotides. In an embodiment, the method uses oligonucleotides wherein each oligonucleotide includes a left arm, a right arm, a marker, and a recognition sequence for a cyanobacterial restriction endonuclease wherein the oligonucleotide is produced through using a polymerase chain reaction wherein the left arm is a double stranded polynucleotide and the right arm is a double stranded polynucleotide and wherein the 5′ end of a single stranded reverse primer of the left arm and the 5′ end of a single stranded forward primer of the right arm overlap and comprise a recognition sequence for a cyanobacterial restriction endonuclease and wherein the left arm and the right arm primers use a template lacking any recognition sites for cyanobacterial restriction endonucleases. In another embodiment, the method uses oligonucleotides wherein the recognition sequence is positioned asymmetrically along the length of said each oligonucleotide such that digestion by the restriction endonuclease creates two digestion products of different sizes. In yet another embodiment, the method uses a template that is SEQ ID NO: 1, and wherein the left arm primer and the right arm primer comprise single stranded polynucleotide primers selected from the group consisting of SEQ ID NOs: 5-62, and wherein the marker is selected from the group consisting of fluorescent dyes, radiolabels, and antigens.
BRIEF DESCRIPTION OF DRAWINGS
FIG. 1 depicts a plasmid map of K230 (left portion of FIG. 1 ) and a schematic depiction of ROligo construction (right portion of FIG. 1 ).
FIG. 2 depicts a list of known cyanobacterial restriction endonucleases, their recognition sequences and the sequences that were used to construct ROligos.
FIG. 3 depicts sequences of primers used to construct the left and right arms of various ROligos.
FIG. 4 depicts an agarose gel analysis of ROligo.AvaI. ROligo.AvaII and ROligo.SphI undigested or digested with commercially available, corresponding restriction nucleases.
FIG. 5 depicts an agarose gel analysis of the restriction enzyme activity of cyanobacterial extracts on plasmids K230 and K244.
FIG. 6 is an infographic chart depicting the theoretical capability of various nucleases to produce a double band pattern when acting upon various ROligos.
FIG. 7 depicts an agarose gel analysis of the restriction enzyme activity of extracts from Triton or pH/urea treated cyanobacterial cells on plasmids K230, K236Cm, and K244 and a PCR product.
FIG. 8 depicts a plasmid map of K236Cm.
FIG. 9 depicts a plasmid map of K244.
FIG. 10 depicts an agarose gel analysis of the restriction enzyme activity in an extract from a first Cyanobacterium strain upon various ROligo pools.
FIG. 11 depicts an agarose gel analysis of the restriction enzyme activity in an extract from a first a first Cyanobacterium strain upon ROligo.AsuI, ROligo.AvaII, ROligo.HaeII, ROligo.SduI, ROligo.AvaIII, ROligo.NspI, ROligo.SphI, and ROligo.BamHI.
FIG. 12 is an infographic chart depicting the results of incubation of various ROligos as a step in the exemplification of the ROligos method for identifying restriction enzymes in a first Cyanobacterium strain extracts.
FIG. 13 is an infographic chart depicting the results of incubation of various ROligos as a step in the exemplification of the ROligos method for identifying restriction enzymes in a first Cyanobacterium strain extract.
FIG. 14 is an infographic chart depicting the results of incubation of various ROligos as a step in the exemplification of the ROligos method for identifying restriction enzymes in a first Cyanobacterium strain extracts.
FIG. 15 is a table depicting cyanobacterial strains analyzed for restriction endonuclease activity using the ROligo method.
FIG. 16 depicts an agarose gel analysis of Gm5C-methylated ROligo.PvuII or TK18 plasmid incubated with a first Lyngbya strain and a Chlorogloeopsis strain extract.
FIG. 17 depicts a plasmid map of TK018.
FIG. 18 depicts an agarose gel analysis of methylated (by a DNA-free Escherichia coli extract harbouring a plasmid expressing M.AvaII methyltransferase) and unmethylated ROligo.AvaII in various cyanobacterial extracts.
FIG. 19 depicts a sequencing readout of a plasmid K230 digested in a Chlorogloeopsis strain extract.
DETAILED DESCRIPTION
Disclosed herein are methods for the identification of restriction endonucleases in cyanobacterial extracts. Identifying restriction endonucleases present within a cyanobacterial strain allows for protection against digestion of introduced vectors or other constructs. In an embodiment, engineered vectors to be used for transformation of a cyanobacterial strain do not contain the recognition site of restriction endonucleases identified in that strain. In another embodiment, engineered vectors to be used for transformation are protected against digestion (restriction) by methylating at extant recognition sites of the identified restriction endonucleases.
DEFINITIONS
The section headings used herein are for organizational purposes only and are not to be construed as limiting the described subject matter in any way. When definitions of terms in incorporated references appear to differ from the definitions provided in the present teachings, the definition provided in the present teachings shall control. Slight and insubstantial deviations from metrics used herein are within the scope of the present teachings herein. In this application, the use of the singular includes the plural unless specifically stated otherwise. Also, the use of “comprise,” “comprises,” “comprising,” “contain,” “contains,” “containing,” “include,” “includes,” and “including” are not intended to be limiting. It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention.
Unless otherwise defined, scientific and technical terms used in connection with the invention described herein shall have the meanings that are commonly understood by those of ordinary skill in the art. Generally, nomenclatures utilized in connection with, and techniques of, cell and tissue culture, molecular biology, protein and nucleotide/polynucleotide chemistry and hybridization described herein are those well-known and commonly used in the art. Standard techniques are used, for example, for nucleic acid purification and preparation, chemical analysis, recombinant nucleic acid, and oligonucleotide synthesis. Enzymatic reactions and purification techniques are performed according to manufacturer's specifications or as commonly accomplished in the art or as described herein. The nomenclatures utilized in connection with, and the laboratory procedures and techniques of described herein are those well-known and commonly used in the art, see, e.g., Sambrook et al., Molecular Cloning: A Laboratory Manual (Third ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. 2000). Standard recombinant DNA procedures were used for gene cloning, plasmid isolation, and electroporation. Manufacturer protocols and standard methods (Sambrook et al., 2000) were followed for DNA purification (e.g., Qiagen, Valencia, Calif.), restriction endonuclease digestion (NEB, Ipswich, Mass.), and DNA amplification using DNA polymerases (NEB, Ipswich, Mass.).
As used in accordance with the embodiments provided herein, the following terms, unless otherwise indicated, shall be understood to have the following meanings:
The term “nucleic acid” and “nucleic acid molecule” refer to natural nucleic acid sequences such as DNA (deoxyribonucleic acid) and RNA (ribonucleic acid), artificial nucleic acids, analogs thereof, or combinations thereof.
The term “oligonucleotide” and “polynucleotide” are synonymous as used herein with each referring to nucleic acids comprising more than one nucleotide. Oligonucleotides and polynucleotides as used herein may be single stranded, double stranded, or a mixture of both single and double strands.
The term “nuclease” applies to enzymes that cleave a phosphodiester bond between nucleotides in a polynucleotide.
The term “recognition sequence” refers to the sequence of polynucleotides comprising the substrate of a nuclease.
The term “endonuclease” applies to nucleases that break nucleic acid chains somewhere in the interior, rather than at the ends, of a polynucleotide.
The term “exonuclease” applies to nucleases that functions by removing nucleotides from the ends of a polynucleotide.
The term “restriction endonuclease” refers to an endonuclease that “restricts”, meaning “digests”, meaning “cuts”, meaning breaks/cleaves phosphodiester bonds between nucleotides on each strand (parallel and antiparallel) in a polynucleotide, thus creating smaller polynucleotides. The terms “restricts”, “digests” and “cleaves” as used herein are synonymous when used in the context of nuclease activity.
The term “plasmid” refers to a circular nucleic acid vector. Generally, plasmids contain an origin of replication that allows many copies of the plasmid to be produced in a bacterial (or sometimes eukaryotic) cell.
A “shuttle vector” refers to a vector which can propagate in two different host species.
The term “construct” as used herein refers to a recombinant nucleic acid molecule that has been generated for the purpose of the expression of a specific nucleotide sequence or sequences, or is to be used in the construction of other recombinant nucleotide sequences. In general, “construct” is used herein to refer to a recombinant nucleic acid molecule. Generally, the term “construct” and the term “vector” are synonymous as used herein.
The term “host cell” as used herein refers to a cell into which DNA can be introduced by any appropriate means (e.g. natural uptake, transformation, transfection, electroporation, and conjugation).
The term “transformation” as used herein refers to a permanent or transient genetic change, e.g., a permanent genetic change induced in a cell following incorporation of non-host nucleic acid sequences.
The term “ROligo” as used herein refers to an oligonucleotide containing arecognition sequence of at least one restriction endonuclease.
The term “ROligo library” refers to multiple ROligos. A ROligo library can consist of a group of ROligos that each individually contain a recognition sequence of a restriction endonuclease from a particular organism and the group of ROligos together make up a ROligo library whose ROligos together contain all of the known recognition sequences of that particular organism.
The terms “information graphics” or “infographics” are used herein refer to graphic visual representations of information, data or knowledge intended to present complex information quickly and clearly. Infographics utilize graphics to enhance the human visual system's ability to see patterns and trends. As used herein, an example of an infographic is a digestion matrix.
The term “digestion matrix” as used herein refers to an infographic table depicting the theoretical capability of various restriction endonucleases to cleave a ROligo oligonucleotide to the double band pattern. A digestion matrix can be used as an infographic. A digestion matrix depicting the theoretical results of the digestion of all constructed oligonucleotides can be manipulated through the application of a set of rules to afford the identity of active restriction endonuclease in a cyanobacterial extract.
The following figures, description, and examples illustrate certain embodiments of the present disclosure in detail. Those of skill in the art will recognize that there are numerous variations and modifications that are encompassed by its scope. Accordingly, the description of certain embodiments should not be deemed to limit the scope of the present disclosure.
Sequencing Method of Determining Restriction Endonucleases
Incubation of a plasmid in a cyanobacterial extract often results in a large number of bands, especially if enzymes with 4-5 bp recognition sequences like R.HaeIII which has a recognition sequence of GGCC or R.AvaII which has a recognition sequence of GGWCC (where W is A or T) are active. Gel isolation and cloning of every fragment after treatment with mung bean nuclease to degrade both 5′ and 3′ single-stranded extensions from the ends of DNA molecules, to generate blunt ends to enable ligation for sequencing would be very time and labor consuming. In case of multiple cutting enzymes, the same enzyme would be identified repeatedly, rendering most of the work non-informative.
As an example of the sequencing method of identifying recognition sequences, a plasmid, K230 (SEQ ID NO: 1) was incubated with a Chlorogloeopsis strain extract. A plasmid map of K230 is depicted in the left portion of FIG. 1 . The digested plasmidial DNA was subjected to sequencing reactions without gel purification of individual bands. For the sequencing reaction, a set of spanning primers (with around 1 kb distance between two adjacent primers) covering the whole K230 plasmid was used because the cleavage points were not known. FIG. 2 depicts the result of the sequencing reaction for a digested portion of the K230 plasmid. As depicted in FIG. 2 for K230 digested with a Chlorogloeopsis strain crude extract, a cleavage point was identified when the sequencing signal dropped sharply at an R.SphI site.
Through multiple rounds of using the labor-intensive and time-consuming sequencing method, enzymes R.BlpI and R.SphI in a Chlorogloeopsis strain. R.AvaII and R.MstI in a second Nostoc strain and R.AcyI in a second Cyanobacterium strain were identified.
The sequencing method for identifying active restriction endonucleases in a cyanobacterial extract has at least two disadvantages. First, the DNA sent to be sequenced contained a vast amount of cyanobacterial genomic DNA from the extract incubation, so the background often was high and rendered many sequencing reactions non-readable and thus required several repetitions reliable sequencing information. Secondly, the sequencing method relies on complete digestion at least at some cleavage points as even traces of undigested plasmid would allow a read-through instead of a signal drop. Therefore, the sequencing method can only be reliably used if completeness of digestion is assured.
Both gel isolation of bands with subsequent cloning and the direct sequencing method share the same disadvantages as the same enzyme might be identified repeatedly. A different method is disclosed herein that improves identification of restriction endonucleases in cyanobacterial extracts, reduces work load, simplifies and accelerates analyses to provide information about whether all restriction enzymes that are present in a cellular extract have been identified.
ROligo Method for Identification of Restriction Endonucleases
In an embodiment, a method is disclosed (herein referred to as the “ROligo” method “R” from restriction site, “Oligo” from oligonucleotide) to identify restriction endonucleases in cellular extracts. In an embodiment, the ROligo method disclosed herein is an improvement over isolating and sequencing digested polynucleotides using the sequencing method in order to identify restriction endonucleases present in a cell.
In an embodiment, ROligos are an oligonucleotide with one introduced recognition site for a known cyanobacterial restriction endonuclease that are constructed by performing overlapping PCR on two sequences (the “left arm” and “right arm”), whose ends contain the recognition sequence of a given restriction endonuclease to be introduced between the two arms, see the right portion of FIG. 1 . In another embodiment, ROligos contain recognition sequences for restriction endonucleases from any given organism such as, for example, E. coli.
In an embodiment, a collection of multiple different ROligos is referred to as a ROligo library. In an embodiment, a ROligo library contains ROligos designed to test for restriction endonucleases from a particular organism. As an example, a ROligo library could be created containing ROligos that cumulatively contain all of the known recognition sequences of cyanobacterial restriction endonucleases as listed in REBASE or any other catalog listing cyanobacterial restriction endonucleases.
In an embodiment, the ROligo method includes incubating various combinations of different ROligos from a ROligo library in a cellular extract and observing a cleavage pattern of the ROligos by agarose gel electrophoretic analysis. The cleavage pattern (pattern of bands of ethidium bromide stained polynucleotides as visualized in the agarose gel using ultraviolet light) indicates the occurrence of enzymes recognizing the introduced restriction sites. Individual restriction endonucleases can then be identified by subsequent incubation of single ROligos from the digested ROligo pool. Thus, in an embodiment of the ROligo method, specific restriction endonucleases can be rapidly identified without additional isolation and sequencing of digested DNA. Additionally, the ROligo method, in contrast to the sequencing method, does not rely on complete digestion of a polynucleotide sequence, i.e. even if only a fraction of the incubated ROligo molecules are digested at the introduced recognition site, the double band pattern will occur.
FIG. 13 depicts a list of known cyanobacterial restriction endonucleases and their recognition sequences that were used to construct ROligos. The sequence which is introduced between the ROligo arms is in usually identical to the recognition sequence of the corresponding restriction endonuclease. However, in cases of a degenerate recognition sequence, a specific sequence among multiple potential sequences is chosen. Some recognition sequences are preceded (e.g. R.AsuI) or followed (e.g. R.SalI) by additional base(s), here a T base, in order to disrupt otherwise emerging recognition sites of further nucleases in the overlap between the introduced site and the adjacent arm sequence. As depicted in the FIG. 1 in the plasmid of K230, contains a “REfree region” from base pairs 7550 to 7926 of SEQ ID NO: 1 that is devoid of recognition sites for known cyanobacterial enzymes. In an embodiment, this REfree region is used as the template for a two-step PCR protocol to construct the left and right arms of ROligos with specific recognition sites and lacking all other known cyanobacterial recognition sequences elsewhere within the ROligo polynucleotide. Depicted as boxes in the right portion of FIG. 1 , ROligo recognition sequences are introduced at an uncentered, asymmetrical position in the ROligo.
The design of a ROligo library needs to be carefully considered in order to create as many ROligos that are only digested by a single restriction endonuclease to the double band pattern. FIG. 2 depicts known cyanobacterial restriction enzymes, their general recognition sequence and the specific sequence used to generate the recognition sequence in the corresponding ROligo. In an embodiment, a ROligo library can be created using the recognition sequences of the cyanobacterial restriction endonucleases as depicted in FIG. 2 . As an example of decreasing the number of ROligos cut by more than one known cyanobacterial enzyme, the ROligos used in embodiments described herein were designed keeping in mind to following information. As depicted in FIG. 2 , the asterisk (*) denotes that these restriction enzymes, found in a second round of literature search after constructing the first ROligos, cut within the arms of the ROligos and, therefore, do not produce the indicative double band pattern of an embodiment of the ROligo method. The superscripted 1 denotes that a blocking base, T, was inserted before the recognition sequence to disrupt an otherwise emerging ApaI site at the junction to the left arm. The superscripted 2 denotes that a blocking base, T, was inserted before the recognition sequence to disrupt an otherwise emerging AsuI site at the junction to the left arm. The superscripted 3 denotes that a blocking base, T, was inserted before the recognition sequence to disrupt an otherwise emerging GC stretch at the junction to the left arm. As the enzyme's recognition site contains a GC, methylation sensitivity could be checked by M.CviPI. Without the additional T base, a possible blockage might result from the outer C, which could only be methylated in specific nucleotide surroundings. The superscripted 4 denotes a blocking base, T, that was inserted after the recognition sequence to disrupt an otherwise emerging CG stretch at the junction to the right arm. As the enzyme's recognition site contains a CG, methylation sensitivity could be checked by M.SssI. Without the additional T base, a possible blockage might result from the outer C, which could only be methylated in specific nucleotide surroundings.
Use of ROligo Method on Cyanobacteria
In an embodiment, the ROligo method disclosed herein is used to identify restriction endonucleases in cyanobacterial genera, species and/or strains. Non-limiting examples of cyanobacterial genera, species and strains that are members thereof, upon which the ROligo method can be used include, but are not limited to, Prochlorococcus, Synechocystis, Synechococcus, Chroococcales, Cyanobium, Oscillatoriales, Cyanobacterium, Chlorophyta, Pleurocapsales, Geitlerinema, Phormidium, Euhalothece, Anabaena, Lyngbya, Spirulina, Nostoc, Pleurocapsa , and Leptolyngbya . In another embodiment of the ROligo method, cell and cellular extracts to be tested for restriction endonuclease activity include all bacteria and Archaea. In an embodiment, certain eukaryotic cells may be tested by embodiments of the ROligo method.
Development of the ROligo Method
In contrast to other methods used to identify restriction endonucleases, the ROligo method does not rely on complete digestion of the polynucleotides incubated with cellular extracts or other solutions containing restriction endonucleases, i.e. even if only a fraction of the polynucleotide molecules are cleaved at the introduced restriction recognition site, the double band pattern will occur
While the sequencing method uses a plasmid with many restriction sites and aims to identify the cleavage point, the ROligo method uses an oligonucleotide with exactly one recognition site for a known cyanobacterial enzyme. In an embodiment of the ROligo method, about 27 different ROligos were constructed by overlapping PCRs using primers having SEQ ID NOs: 5-62 as depicted in FIG. 3 . The strand backbone for the ROligo originated from the K230 plasmid (see FIG. 1 ) which contains a sequence of 360 bp that shows no recognition site for any of the known cyanobacterial restriction endonucleases, see FIG. 2 . In an embodiment of the ROligo method, this K230 stretch is used and a known cyanobacterial restriction site is inserted at an uncentered, asymmetric position resulting in a PCR-derived DNA fragment of about 380 bp length, see right side of FIG. 1 . In an embodiment, cleavage of the generated ROligos results in a double band of about 216 and about 166 bp that are generated by digestion of a ROligo at the recognition sequence, the approximate length of the individual arms. These bands are readily discernible by using a lane containing size markers as one ROligo cleavage band lies above and one ROligo cleavage band lies below the 200 bp marker band of the size marker standards.
FIG. 4 depicts digested and undigested ROligos analyzed by agarose gel electrophoresis stained with ethidium bromide and visualized under ultraviolet light. Lanes 1-3 contain, respectively, undigested ROligos, ROligo.AvaI, ROligo.AvaII, and ROligo.SphI. Lanes 4-6, contain, respectively, ROligo.AvaI digested with R.AvaI (R. stands for restriction endonuclease, thus R.AvaI is restriction endonuclease AvaI), ROligo.AvaII digested with R.AvaII, and ROligo.SphI digested with R.SphI. Lane M contains a size marker.
The ROligos with introduced recognition sites for R.AvaI (lane 1), R.AvaII (lane 2) and R.SphI (lane 3) were constructed by PCR. The undigested PCR products (lanes 1-3) depict a full length 380 bp band. ROligos were digested with appropriate commercially available enzymes (lanes 4-6) and the desired double band at 216/166 bp was observed.
For some ROligos, digestion with commercially available enzymes leaves a faint undigested band (see FIG. 4 , lane 4). This may be due to the use of three consecutive PCR reactions in the ROligo construction process using a polymerase with relatively low fidelity such as DreamTaq polymerase. To reduce the number of such undigestable PCR molecules, a polymerase with higher fidelity, Pwo available from Roche Applied Science, was used for PCR amplification of the ROligos.
Digestion patterns other than the double band pattern of 216/166 that are created by digestion of a ROligo at the introduced recognition sequence are not investigated. Thus, restriction enzymes that cut at sites other than the introduced recognition sequence will be missed.
Digestions of Plasmids with Cyanobacterial Extracts
FIG. 5 depicts an analysis of the digestion of plasmids by cyanobacterial extracts. The first panel depicts undigested plasmid K230, undigested plasmid K244, and a marker lane. Moving from left to right, the cyanobacterial extracts used were a fourth Cyanobacterium strain, a third Cyanobacterium strain, and a Cyanothece strain (furthest right panel). For each of the cyanobacterial extract panels, lane M is a size marker, lane 1 is a crude cyanobacterial extract without plasmid addition, lane 2 is plasmid K230 incubated in the extract, and lane 3 is plasmid K244 incubated in the extract. Cyanobacterial extracts were incubated with plasmids K230 and K244 to check for overall restriction activity. A fourth Cyanobacterium strain shows a specific digestion banding pattern indicating the presence of one or more restriction endonucleases. A third Cyanobacterium strain shows smearing of both plasmids indicating the presence of an unspecific, possibly extracellular, endonuclease or exonuclease that could obscure specific endonucleases that are active at the same time in the same extract upon the same plasmid. Incubation of the plasmids with a Cyanothece strain extract depicts the plasmids as being undigested.
Potential Digestions of ROligos
In a first embodiment, a restriction enzyme cuts at the introduced recognition site and thus digests the ROligo into a dual 216/166 bp banding pattern as visualized on an agarose gel stained with ethidium bromide. The dual banding pattern indicates a restriction endonuclease that recognizes the complete introduced site, or part of it.
In a second embodiment, a cellular extract contains an active restriction enzyme that cleaves outside the introduced recognition site of the ROligo. If the cleavage point lies in close vicinity to the introduced restriction site, the resulting band pattern will resemble a double band. However, as the sequences outside the introduced site are identical among all ROligos, cleavage of one ROligo outside of its introduced site would mean that all ROligos would likely be digested and therefore a double band would show up on an agarose gel analysis. This universal digestion of all ROligos is unlikely at least because no more than three restriction endonucleases have been found in any cyanobacterial cellular extract. If a cyanobacterial extract contains an active restriction enzyme that cleaves outside the introduced recognition site of the ROligo, this pattern would indicate that there is at least one restriction enzyme present, but because the enzyme does not specifically cut at the introduced recognition sequence, it cannot be identified by the ROligo method. Deducing its recognition sequence would require other more cumbersome and labor intensive methods such as sequencing the ends of gel-isolated ROligo digestion bands.
In a third embodiment of possible cleavages of ROligos, a restriction endonuclease is present that recognizes an overlap between the introduced restriction site and the bases of the flanking sequences. The introduced recognition site of each ROligo is designed such that no known cyanobacterial enzyme would recognize any overlap created by the insertion of the recognition sequence into the ROligo. If the presence of an additional recognition sequence at the junction site between the flanking regions and the introduced recognition sequence is intrinsically unavoidable, a second, discriminatory, ROligo can be constructed which allows exclusion of one of the possible enzymes. For example, restriction enzyme R.NspI recognizes the overlap between the R.AvaI site and the adjacent flanking arm in the ROligo.AvaIII. Thus, if a ROligo was generated with an R.AvaIII recognition sequence and a double band pattern showed up on the gel, it would not be possible to determine whether the active restriction endonuclease is R.NspI or R.AvaIII by using the ROligo method. However when a second ROligo, ROligo.NspI, is generated where only R.NspI cuts, but not R.AvaIII, the identity of the active restriction enzyme or enzymes can be determined, see, for example, the digestion matrix generated in FIG. 6 .
In a fourth and fifth embodiment of possible cleavages of ROligos, a non-double band cleavage pattern or a non-cleavage pattern is generated, respectively. Restriction enzymes that cut an ROligo but do not produce a double band will not be identified by using embodiments of the ROligo method that rely upon the presence of a double band pattern (for example, 216 bp and 166 bp) in the analysis of cellular extract incubated polynucleotides on an agarose gel.
Restriction enzymes that are present in cellular extracts but do not cut any ROligos in the ROligo library will also be missed. This lack of identification of existing, but unknown, restriction endonucleases can be overcome by generating additional ROligos for a ROligo library.
ROligos Containing Degenerate Recognition Sequences
Many cyanobacterial restriction endonucleases have recognition sequences which are degenerate. For example, R.AvaII recognizes GGWCC (where W is A or T) and R.AsuI recognizes GGNCC (where N is A. T, C or G). For the design of a ROligo do distinguish between R.AvaII and R.AsuI using the ROligo method, there are at least two options.
First, degenerate primers can be used to introduce the recognition site into the respective arms. Hence, the ROligo for R.AvaII would contain its two recognized sequences GGACC and GGTCC. Likewise, the ROligo for R.AsuI would contain four sequences, GGACC, GGCCC, GGGCC and GGTCC. The digestion pattern generated using these degenerated ROligos would not be able to distinguish an extract containing R.AvaII from one containing R.AsuI as either enzyme will cut both GGTCC and GGACC containing ROligos to the double band pattern.
In another embodiment, degenerate restriction endonucleases can be distinguished from one another through engineering ROligo recognition sequences of a given ROligo such that fewer of the possible recognition sequences are available to multiple restriction endonucleases. For example, to distinguish between degenerate restriction endonucleases R.AvaII and R.AsuI, GGACC could be used as the sole sequence for the R.AvaII ROligo and GGCCC could be used as the sole sequence for the R.AsuI ROligo. Because GGCCC is not recognized by R.AvaII, extracts with R.AvaII activity can only digest the ROligo.AvaII to the double band pattern but not the ROligo.AsuI, whereas an extract harboring R.AsuI can digest both and therefore the enzymes can be distinguished from one another.
Other examples of identifying degenerate restriction endonucleases include combinations such as R.PvuII having a recognition sequence of CAGCTG and R.NspBII having a recognition sequence of CMGCKG (where M is A or C and K is T or G). In an embodiment, R.PvuII which recognizes CAGCTG and R.NspBII can be distinguished from one another by using CAGCGG as the sole sequence for ROligo.NspBII instead of using all its four sequences recognized, CAGCGG, CAGCTG, CCGCGG, and CCGCTG.
Identification of Restriction Enzymes Using a Digestion Matrix
Infographics can be used to visualize data generated from the analysis of ROligo digestion products in a digestion matrix and to determine the identification of active restriction endonucleases in a cyanobacterial extract through the implementation of various steps of the ROligo method disclosed herein. In an embodiment, a digestion matrix is an infographic table depicting the theoretical capability of various restriction endonucleases as to whether they are capable of digesting the oligonucleotides to a double band pattern. In an embodiment, a digestion matrix can be manipulated through the application of a set of rules to determine the identity of active restriction endonuclease in a cyanobacterial extract. In an embodiment, a digestion matrix is created which depicts the absence or presence of a restriction enzyme in the celhlular extract of a cyanobacterial strain. In an embodiment, the digestion matrix is depicted in the form of an infographic table wherein the rows are labeled with ROligos containing a sequence recognition site and the columns are labeled with known cyanobacterial restriction enzymes. In an embodiment, a digestion matrix depicts, as a filled-in box, which cyanobacterial restriction enzyme(s) would produce a double band digestion pattern (indicative of a digestion of the ROligo at the recognition sequence) for which ROligos, see, for example FIG. 6 .
Restriction enzymes that do not share a recognition sequence, degenerate or overlapping, will create a diagonal pattern in a digestion matrix. In a non-limiting sense, ROligos that are digested by only one restriction enzyme are referred to herein as “Optimal ROligos”.
If all ROligos were Optimal ROligos, identifying the restriction enzyme constituency of a cellular extract would be a matter of having a comprehensive library of Optimal ROligos encoding recognition sequences for all known restriction enzymes. However, at least because many restriction enzymes exhibit overlap in recognition sequences, R.AvaII/R.AsuI, and R.HaeIII/R.ApaI, for example, additional embodiments of the ROligo method are necessary to identify restriction endonucleases present within various cellular extracts.
Washing Cyanobacterial Cells to Remove Extracellular Nucleases
FIG. 7 depicts the results of digestion of plasmids in cyanobacterial extracts that have or have not been washed with either a Triton detergent wash or a pH/urea wash in an attempt to remove extracellular nucleases before disruption of the cyanobacterial cells and testing of their extracts for restriction endonucleases. Plasmids used for digestion include K236Cm (SEQ ID NO: 2) and K244 (SEQ ID NO: 3). Plasmid maps for K236Cm and K244 are depicted in FIG. 8 and FIG. 9 , respectively.
Panel A of FIG. 7 : Lane 1 is a plasmid mixture of K230, K236Cm and K244 incubated in cell-free medium. Lanes 2-4, are plasmid mixtures incubated with supernatants from respective washing steps with triton solutions. Lane 5 is an unincubated plasmid mixture. Lane M is a size marker. Lane 6 is plasmid K236Cm incubated in a third Cyanobacterium strain extract after triton washing at 28° C.; lane 7, plasmid K236Cm incubated in a third Cyanobacterium strain extract after triton washing at 37° C.
Cells of a third Cyanobacterium strain were washed three times with 0.1% triton in DNase-buffer. A description of the triton washing technique used can be found in Soper, B. W., et. al., J. Bacteriology 1994, 176(17):556-5570. Six hundred nanograms of a plasmid 1:1 mixture containing K230, K236Cm and K244 were incubated over night at 37 OC with the cell's culture media and supernatants. DNA incubated in the cell-free medium (lane 1) showed strong degradation as seen by the smearing towards lower molecular weights and an overall reduction in DNA staining as compared to lanes 3 and 4. DNA incubated in wash supernatant 1 (lane 2) showed weak smearing to lower molecular weights but still a decrease in DNA staining as compared to lanes 3 and 4. DNA incubated in later wash supernatants show only one non-degraded band of similar intensity. The size difference between the unincubated plasmid mixture of lane 5 and the other lanes is likely due to a specific cleavage by an endonucleolytic enzyme which is not removed by triton treatment or a spontaneous reorganization of plasmid DNA during incubation from nicked plasmids to a supercoiled form.
Panel B of FIG. 7 : A 1 kb PCR product was used as a template for incubations of DNA with extract from a Geitlerinema strain. In lane 1, the template was incubated in crude extract of triton treated cells of a Geitlerinema strain incubated overnight at 30° C. In lane 2, the template was incubated overnight at 30° C. in crude extracts of pH/urea washed cells of a Geitlerinema strain. In lane 3, the template was incubated overnight at 37° C. in crude extracts of pH/urea washed cells of a Geitlerinema strain. Lane 4 depicts a crude extract of pH/urea treated cells from lanes 2 and 3. After triton washing of a Geitlerinema strain cells, crude extracts showed unspecific nucleolytic activity (see lane 1) as seen by the blurry band shape after incubating a 1 kb PCR template in a Geitlerinema strain extract over night at 30° C. A fresh culture of a Geitlerinema strain was then washed with a pH change/urea regimen. As depicted in lane 2, this crude extract did not show smearing of the digestion bands as seen by the sharp bands as visualized. Moreover, incubation at 37° C. did not result in unspecific degradation.
Thus, washing cyanobacterial cells with either a triton or a pH/urea washing regimen was useful for decreasing the amount of non-specific nuclease activity in the cellular extract to be tested for the activity of restriction endonucleases using the ROligo method.
Digestion Matrix of a ROligo Library
FIG. 6 depicts a predicted digestion matrix of a library of twenty ROligos. Columns are labeled after cyanobacterial enzymes. Rows are labeled after the ROligo. The introduced recognition sequence is reflected by the name of the ROligo. A black box indicates that the enzyme of that column would digest the ROligo of that row to produce a double band of 216/166 bp. Some ROligos not only get cleaved by their specific enzyme, creating the diagonal black box pattern in the matrix, but are also cut by a second or third enzyme recognizing the same sequence, a part thereof or an overlap to the adjacent arm sequence. The digestion of a given ROligo by multiple restriction enzymes can be reduced through careful engineering of the sequence of the recognition site by the addition of surrounding bases. For example, the ROligo.HaeII used to create the digestion matrix of FIG. 6 was engineered without a T base preceding the recognition sequence in an effort to reduce the number of restriction endonucleases that identify the particular ROligo.HaeIII recognition sequence used.
FIG. 6 depicts the theoretical cleavage capability of a set of known cyanobacterial restriction endonucleases to produce the double band pattern on generated ROligos to identify their occurrence in future cyanobacterial extracts. As depicted in FIG. 6 , columns are labeled after cyanobacterial restriction enzymes. Rows are named after ROligos. The introduced recognition sequence of the respective restriction endonuclease is reflected by the name of the ROligo. For example, AcyI and ROligo.AcyI. In FIG. 6 , a black box indicates that the enzyme of that column can theoretically digest the ROligo of that row to produce a double band of 216 bp and 166 bp as visualized on an agarose gel. Some ROligos not only get cleaved by their respective restriction endonuclease (diagonal black box pattern) but are also cleaved by a second or third enzyme recognizing the same sequence, a part thereof or an overlap to the adjacent arm sequence and thus generating a 216 bp and 166 bp pattern.
After incubation with a cellular extract and analysis by agarose gel, the pattern of digested and undigested ROligos is compared to a predicted digestion matrix to find one enzyme, or a set of enzymes, whose predicted digestion pattern results in exactly the pattern observed. Some incubations result in digestion patterns that predict only one set of enzymes that explain the observed pattern (as the ROligos were designed to be cut by as few other enzymes as possible). However, some combinations of enzymes cannot be distinguished through predicted digestion patterns. For example, the concomitant occurrence of R.HaeIII (recognition sequence of GGCC) and R.AvaII (recognition sequence of GGWCC) would produce the same ROligo digestion pattern as the combination of R.HaeIII and R.AsuI (recognition sequence of GGNCC). This is because R.HaeIII cuts GGCCC and GGGCC, which are the recognition sequences of R.AsuI that are not cleaved by R.AvaII. Thus, using an embodiment of the ROligo method, the concomitant occurrence of R.HaeIII and R.AvaII would be indistinguishable from the concomitant occurrence of R.HaeIII and R.AsuI. Other embodiments of the ROligo method and rules of interpretation of digestion patterns are useful to distinguish concomitant occurrences of restriction endonucleases such as these.
Exemplary Rules Used in Embodiments of the ROligo Method
In an embodiment of the ROligo method, a sequence of rules can be applied to identify a set of restriction endonucleases that explain an observed ROligo digestion pattern as depicted in a digestion matrix.
Rule 1: First, all restriction endonucleases which theoretically do not recognize and cleave at least one of the ROligos that are observed to be cleaved to the double band pattern by a cellular extract are discarded (removed from being included in the digestion matrix). The digestion matrix then only contains enzymes that can theoretically cut at least one of the observably digested ROligos.
Rule 2: Secondly, restriction endonucleases which theoretically recognize and cleave a ROligo that does not show a double band pattern after extract incubation, can be discarded from the digestion matrix.
Rule 3: The occurrence of the remaining restriction endonucleases then explains the observed pattern of digested and non-digested ROligos. Generally, all restriction endonucleases remaining in the digestion matrix after the application of rules 1 and 2 are needed to actually explain the pattern.
However, in some instances, a pattern is present that may be explained by either one enzyme or a combination of restriction endonucleases being responsible for the digestion of multiple ROligos. The digestion pattern presents itself as ambiguous as to whether it is the combination of restriction endonucleases or the single restriction endonuclease is responsible for the observed ROligo digestions. As an example, the ROligo digestion pattern of both R.NspBII (CMGCKG recognition sequence wherein M is A or C, and K is G or T) and R.PvuII having a CAGCTG recognition sequence, is identical to the one produced by R.NspBII alone. The same reasoning applies to the combination of R.AsuI having a GGNCC recognition sequence plus R.AvaII having a GGWCC recognition sequence, and R.AsuI alone. In these ambiguous patterns, the existence of the single restriction enzyme that is responsible for digesting the two ROligos is chosen to exist versus the presence of two restriction endonucleases that perform the identical function. Not being limited by theory, a single restriction endonuclease is more likely to occur in the extract because a cell evolves to conserve metabolic resources and thus would not develop an increase in protection against foreign DNA by expressing two restriction endonucleases where the expression of a single restriction endonuclease would provide the same number and type of recognition sites as the two restriction endonucleases combined. Thus, the restriction endonuclease exhibiting the broader cleavage range is considered more likely to exist alone in the cellular extract and is kept as a restriction endonuclease that exists while the combination of restriction endonucleases are discarded if they cannot be individually eliminated.
Identifying Restriction Endonucleases in a Cyanobacterial Extract Using the ROligo Method
Using a ROligo library of the twenty ROligos depicted in FIG. 6 , active restriction endonucleases from a first Cyanobacterium strain extract were determined. The incubation of the ROligos from the ROligo library with the cellular extract was performed in a two-step process. First, five pools of four ROligos were incubated with a cellular extract and a double band digestion pattern was tested for. Second, the individual ROligos comprising the ROligo pool showing a double band digestion pattern were each individually incubated with the cellular extract and analyzed for the double band digestion pattern.
FIG. 10 depicts the results of an ethidium bromide stained agarose gel analysis of several pools of ROligos incubated in a first Cyanobacterium strain extract. Lanes 1-5, together, comprise the totality of a ROligo library of 20 ROligos. Each lane has four ROligos (a ROligo pool) from the ROligo library. Each ROligo pool is incubated in either a NEB buffer P1 (left part of the gel) and buffer P2 (right part). The most commonly used NEB buffers P1, P2, P3, and P4 are designed such that most known restriction endonucleases will be active in at least one of the four buffers. Lane X is a first Cyanobacterium strain extract without addition of ROligos. Lane M is a size marker.
Some pools of ROligos show a digestion double band at around 216/166 bp (marked by arrows at the top). Pool 1 showed a double band only in buffer P2 but not in buffer P1, pool 3 showed a double band in both P1 and P2 buffers. Not depicted are the incubations performed in NEB buffers P3 and P4. Because pools 2, 4, and 5 showed no double band digestion pattern, their constituent ROligos were eliminated as possibly existing in this first Cyanobacterium strain.
FIG. 11 depicts the results of incubation of individual ROligos in a first Cyanobacterium strain extract. Individual ROligos comprising ROligo pools that resulted in a double band pattern after the first-step incubation, see FIG. 10 , were each individually incubated in a first Cyanobacterium strain extract in NEB buffers P1, P2 or P3. Lanes 1-8 depict the results of ROligos incubated in a first Cyanobacterium strain extract as follows: Lane 1 is ROligo.AsuI having a recognition site of GGNCC constructed as GGCCC. Lane 2 is ROligo.AvaII. Lane 3 is ROligo.HaeIII constructed as GGCC. Lane 4 is ROligo.SduI. Lane 5 is ROligo.AvaII. Lane 6 is ROligo.NspI. Lane 7 is ROligo.SphI constructed as GCATGC. Lane 8 is ROligo.BamHI. Lane M is a marker. ROligos showing a double band were ROligo.AsuI, ROligo.HaeIII and ROligo.SphI and are marked with arrows.
Combining four ROligos in one incubation reaction as a first step in the ROligo method reduces work load significantly as compared to incubating each member of the ROligo library individually. As exemplified above, to identify which of the twenty analyzed cyanobacterial restriction enzymes are present in a first Cyanobacterium , strain, the twenty ROligos in the digestion matrix (see FIG. 6 ) were sorted in a way that ROligos that were digested by the extract, those showing a double banding pattern, were moved to the top of the matrix (see FIG. 12 , top three rows). Rule 1 was then applied to the digestion matrix, and all enzymes which theoretically cannot cut at least one of these three ROligos were removed, leaving four possible enzymes, R.AsuI, R.HaeIII, R.NspI and R.SphI, see FIG. 13 . Applying rule 2 (restriction enzymes which theoretically would also digest at least one other ROligo to the double band pattern than the observed extract-digested double band pattern showing ROligos can be discarded) allowed to discard enzymes R.AsuI (as ROligo.AvaII was undigested) and R.NspI (as neither ROligo.NspI nor ROligo.AvaIII was digested by the extract, see FIG. 14 .
This left two enzymes, R.HaeIII and R.SphI. R.HaeIII theoretically cuts ROligo.AsuI and ROligo.HaeIII. R.SphI only cuts its own ROligo.SphI. The occurrence of both these enzymes in the digestion matrix produces the same pattern (see FIG. 14 ) as was observed in a first Cyanobacterium strain extract, see FIG. 15 . Thus, through using an embodiment of the ROligo method, the active restriction endonucleases in a first Cyanobacterium strain extract were determined and subsequent vectors and constructs used for transformation of a first Cyanobacterium strain extract can be engineered to lack those recognition sequences or can be methylated at those recognition sequences to prevent digestion of the vector by the endogenous restriction endonucleases of a first Cyanobacterium strain.
Analyses of Endonuclease Activity in Cyanobacterial Strains Using the ROligo Method
FIG. 15 depicts various cyanobacterial strains that were analyzed with an embodiment of the ROligo method to identify restriction nuclease activity therein. If a genomic sequence of the cyanobacterial strain was available, RM (“R”estriction “M”odification) systems were predicted. RM systems were identified when a gene encoding a restriction enzyme and a gene encoding its cognate methyltransferase are present in close genomic proximity. Examples of some of the tested methylation of ROligos that blocks the respective cyanobacterial restriction endonucleases from cleavage are given.
As depicted in FIG. 15 , the asterisk (*) denotes that the recognition sequence of a given restriction endonuclease contains two GC stretches. In most cases, both GC stretches were methylated by the methyltransferase M.CviPI. In a prophetic embodiment, restriction enzymes in the respective strain extracts are blocked by methylation of one or both of the GC stretches.
As depicted in FIG. 15 , the pound sign (#) denotes enzymes that were previously not known to exist in cyanobacteria and were identified by using the sequencing method. The identified restriction endonuclease's activities were verified by using the ROligos method to identify restriction endonucleases.
As depicted in FIG. 15 , the double pound sign (##) denotes an enzyme that was initially detected by cloning and sequencing digestion bands.
In an embodiment, and as depicted in FIG. 15 , the restriction endonucleases in extracts of fifteen cyanobacterial strains were analyzed by the ROligo method. Genomic analysis of strains, whose genome had been sequenced, predicted restriction enzymes in a second Lyngbya strain, a Chlorogloeopsis strain, a second Cyanobacterium strain, and a second Nostoc strain, see FIG. 15 . With the exception of a second Nostoc strain, all enzymes whose activity could be demonstrated in crude extracts using the ROligo method, see FIG. 15 , were also predicted by in silico methods. However, not all predicted enzymes were found active in extracts, see a second Nostoc strain in FIG. 15 , for example.
A hitherto unknown cyanobacterial restriction enzyme was detected in a third Cyanobacterium strain. Incubation with a third Cyanobacterium strain extract showed digestion of plasmids, but none of the ROligos in the library were digested to a double band. Direct sequencing of the ends of the digested plasmid bands revealed a region of about 30 bp where the sequences all showed a signal drop, indicating the possible presence of extracellular nucleases slowly and possibly nonspecifically degrading the ends of the digested plasmids. In an effort to remove any extracellular nucleases, the cells were washed with a triton washing protocol as described herein. Analyzing possible restriction enzymes that would have a recognition sequence within that region and further limiting to palindromic enzymes (almost all cyanobacterial restriction endonucleases have recognition sequences that are palindromatic), the recognition sequence of five candidate restriction enzymes were identified and ROligos were constructed accordingly. Of the five candidate restriction enzymes, only the ROligo.EcoRV, having a recognition sequence of R.EcoRV (GATATC), was found to be digested by the extract, see FIG. 15 , and R.EcoRV was the only enzyme that could digest the ROligo to the double band as judged by the appropriate digestion matrix. Because R.EcoRV was not previously known to be a cyanobacterial restriction endonuclease, if a plasmid devoid of EcoRV sites had been incubated with the cellular extract, this restriction enzyme would not have been detected. A Geitlerinema strain was analyzed as well. Again, none of the ROligos in the library at the time of analysis was digested to a double band. However incubating plasmid K230 and several PCR products derived from it, clearly showed digestion. After sequencing the ends of a digested PCR product, a SauI recognition sequence of CCTNAGG was detected and the corresponding ROligo was constructed and was digested by a Geitlerinema strain extract.
In a prophetic embodiment, and according to an embodiment of the ROligo method, the generated ROligo.EcoRV, and ROligo.SauI corresponding to the restriction endonucleases identified could then be methylated by appropriate means. The methylated ROligos could then be incubated with the appropriate cyanobacterial extracts and could be analyzed to determine to what degree digestion was inhibited by their respective restriction endonucleases.
FIGS. 12-14 depict and embodiment of the ROligo method used to determine the restriction enzymes that are active in an extract of a first Cyanobacterium strain.
In an embodiment of the ROligo method, FIG. 12 depicts a digestion matrix that is a useful for the identification of a first Cyanobacterium strain's restriction enzymes. A digestion matrix of a library of twenty ROligos was reorganized to reflect the observed pattern of digested and undigested (i.e., double band digestion of a ROligo detected or not) ROligos in a first Cyanobacterium strain extract. The top most three ROligos (rows) marked yellow, are digested by a first Cyanobacterium strain extract, the unmarked ROligos are not.
FIG. 13 depicts a digestion matrix of a first Cyanobacterium strain extract incubated ROligos after applying an embodiment of rule 1 of the ROligo method (removal of any enzyme which does not cut at least one of the three topmost ROligos). Rule 2 (restriction enzymes which theoretically would recognize other ROligos than the ones having been observed to be extract-digested to the double band pattern, can be discarded) was applied to discriminate the R.AsuI/ROligo.AvaI I digestions (black horizontal arrow) and the R.NspI/ROligo.AvaIII digestions.
FIG. 14 depicts a digestion matrix resulting from the incubation of a ROligo library in a first Cyanobacterium strain extract after applying rules 1 and 2 and then applying rule 3 (the occurrence of the remaining enzymes then explains the observed pattern of digested and non-digested ROligos). The combination of the activity of the two remaining enzymes, R.HaeIII and R.SphI, cut exactly those ROligos which are digested by the extract.
Therefore, in an embodiment of the ROligo method, FIGS. 12-14 depict steps of the ROligo method used to identify the active restriction endonucleases of a cyanobacterial strain. Thus, vectors for transformation of the cyanobacterial strain can be designed that lack the recognition sequences of R.HaeIII and R.SphI and therefore increase the likelihood of transformation of the cyanobacterial strain for which the ROligo method has been performed upon.
Generation of New Digestion Matrices
In an embodiment, after generating new ROligos, the starting digestion matrix will be constructed anew, and would now include any newly identified ROligos and incorporate the predicted digestion pattern of the newly identified ROligos into a new digestion matrix. The restriction enzymes included in the matrix (showing the double band pattern) can be derived from various sets of enzymes, e.g. all cyanobacterial enzymes, all palindromic enzymes, or all restriction enzymes present, for example, in databases such as REBASE.
DNA Methylation
DNA methylation is the process of transferring a methyl group from a donor molecule to either a cytosine or an adenine by DNA methyltransferases. Three types of methylated bases are predominantly found in DNA: 5-methylcytosine (m5C), N4-methylcytosine (m4C), and N6-methyladenine (m6A). The organism-specific pattern of methylation depends on the methyltransferases' specificity.
In prokaryotes, DNA cleavage by a cognate restriction enzyme is prevented by the methylation of DNA by a sequence-specific methyltransferase which is an integral component of every restriction-modification system. E. coli strains used for propagation of plasmid DNA often contain a site-specific DNA adenine methyltransferase, Dam. The methylase encoded by the dam gene methylates the N6-position of an adenine residue within the GATC sequence.
Most restriction enzymes are sensitive to DNA methylation. When a restriction enzyme recognition sequence overlaps a methylation site, no effect, a partial inhibition or complete block of digestion may occur. Furthermore, the ability to cleave methylated DNA is an intrinsic and unpredictable property of each restriction enzyme. Therefore, isoschizomers and neoschizomers which recognize the same DNA sequences can differ in their sensitivity to DNA methylation. To prevent DNA digestion of heterologous vectors, both the type of DNA methylation and the sensitivity of the restriction enzyme to that type of methylation should be considered.
Methylation Sensitivity of Identified Restriction Endonucleases
In an embodiment, the ROligo method is used to improve the efficiency of determining the methylation sensitivity of identified restriction endonucleases. Methylation sensitivity usually has to be checked for each restriction enzyme found and cannot be inferred from a recognition sequence alone. As an example, the non-cyanobacterial restriction enzymes R.DpnI, R.MboI and R.Sau3AI all recognize the four base pair palindrome sequence of GATC. However, they show different sensitivity to methylation of GATC at the adenine or cytosine position (see REBASE). R.DpnI only cuts fully adenine methylated Gm6ATC. MboI is blocked by hemi-methylation at Gm6ATC but is not affected by C5-cytosine methylation. Sau3AI cuts both N6-adenine methylated and unmethylated DNA, but is blocked by C5-methylation GATm5C.
Different types of in vitro methylation can be used to methylate a ROligo. The effect of methylation can subsequently be checked by incubation of the methylated ROligo in an extract that cuts the non-methylated ROligo. If a suitable methylation type that protects from cleavage has been identified, methyltransferase databases like REBASE can be searched for a known methyltransferase that catalyzes that type of methylation. If these methyltransferases can subsequently be expressed in yeast or E. coli , the sequencing of the cyanobacterial genome to find endogenous methyltransferases is not necessary.
Methylation of ROligos
Identification of restriction endonucleases that are present in a given strain of cyanobacteria allows for site specific methylation of heterologous DNA prior to transformation of that given strain of cyanobacteria. In an embodiment, methylation of ROligos can be achieved by heterologous overexpression of methyltransferases. For example, plasmids that overexpress methyltransferases, such as M.AvaI and M.AvaII methyltransferases, may be constructed and used in an E. coli donor strain and then transferred into a cyanobacterial strain via conjugation (see Elhai J., et. al., J. Bacteriology 1997, 179(6):1998-2005) or other transformation means.
In another embodiment, commercially available methyltransferases can be used to methylate ROligos. Methylation success is checked by restriction digestion with a corresponding commercially available restriction enzyme that is blocked by the type of methylation used (which can be found for example in charts depicting methylation sites for known restriction endonucleases, e.g., REBASE). Any methylations that result in protection of the ROligos from digestion can then be applied to methylate plasmids or other exogenous DNA prior to transformation of cyanobacterial strains having the identified restriction endonuclease activity. For example, M.CviPI methylates cytosine at the C5 position in its GC recognition sequence. M.SssI methylates cytosine at the C5 position in its CG recognition sequence. M.CviPI or M.SssI, for example, can therefore be used to methylate DNA at cytosines. Many of the cyanobacterial enzymes contain a CG or GC stretch and can, therefore, be checked for their sensitivity to m5CG and/or Gm5C methylation, see FIG. 2 . As an example, protection from digestion of ethanologenic plasmids and ROligos methylated by M.CviPI and M.SssI from restriction endonucleases is depicted in FIG. 16 .
FIG. 16 depicts analysis of a possible protective effect of Gm5C-methylation against a first Lyngbya strain extract and a Chlorogloeopsis strain extract. The left panel of FIG. 16 depicts lane M which is a reference marker; lane 1 which is an unmethylated ROligo.PvuII incubated in a first Lyngbya strain extract; lane 2 which is Gm5C-methylated ROligo.PvuII incubated in a first Lyngbya strain extract; and lane 3 which is methylated ROligo.PvuII that has not been digested. Thus, as depicted in the left panel of FIG. 16 , ROligo.PvuII, having a recognition sequence of CAGCTG was methylated using commercially available M.CviPI which methylates the cytosine at the C5 position in its recognition sequence GC resulting in CAGm5CTG. Methylated and unmethylated ROligo.PvuII were incubated in a first Lyngbya strain extract and only the umnethylated ROligo.PvuII was digested by the extract. Thus, R.PvuII activity in a first Lyngbya strain is blocked by M.CviPI methylation. Therefore, the left panel of FIG. 8 depicts the protective effect of methylation of ROligo.PvuII against digestion from a restriction endonuclease in the extract of a first Lyngbya strain.
The right panel of FIG. 16 depicts analysis of a possible protective effect of Gm5C-methylation against a Chlorogloeopsis strain extract. TK018 (SEQ ID NO: 4), an exemplary plasmid containing an ethanologenic cassette, was methylated with M.CviPI and then column purified. A plasmid map of TK018 is depicted in FIG. 17 . Methylated and unmethylated TK018 were then incubated in a Chlorogloeopsis strain extract. The right panel of FIG. 16 depicts lane M which is a size marker; lane 4 which is a Chlorogloeopsis strain extract without DNA; lane 5 which is TK18 incubated in a Chlorogloeopsis strain extract; lane 6 which is TK18 methylated with M.CviPI and then incubated in a Chlorogloeopsis strain extract; lane 7 which is unincubated TK018; and lane 8 which is methylated TK18 that has not been incubated. As depicted in the right panel of FIG. 16 , the methylated plasmid in lane 6 does not show the digestion bands of the unmethylated plasmid of lane 5. Therefore, the right panel of FIG. 16 depicts the protective effect of methylation of an exemplary plasmid, TK018, against digestion from a restriction endonuclease in the extract of a Chlorogloeopsis strain.
In yet another embodiment, the extract from E. coli strains containing methyltransferases expressed from helper plasmids can be used to methylate ROligos. FIG. 18 depicts the protective effect of M.AvaII methylation of ROligo.AvaII against digestion from restriction endonucleases within the extracts of a first Lyngbya strain, a first Nostoc strain, and a second Nostoc strain. As depicted in FIG. 18 , lane 1 is an unmethylated ROligo.AvaII digested with commercially available R.AvaII; lane 2 is a methylated ROligo.AvaII subjected to digestion with R.AvaII; lane 3 is a methylated ROligo.AvaII that is undigested; lanes 4, 6, and 8 are unmethylated ROligo.AvaII incubated in extract of a first Lyngbya strain, a first Nostoc strain, and a second Nostoc strain, respectively; lanes 5, 7, and 9 are methylated ROligo.AvaII incubated in extract of a first Lyngbya strain, a first Nostoc strain, and a second Nostoc strain, respectively; and lane M is a reference marker.
ROligo.AvaII as used in FIG. 18 incubations was, as applicable, methylated using an extract from E. coli HB101 containing a M.AvaII expressing plasmid pRL528. Success of methylation of ROligo.AvaI was verified by digesting the methylated ROligo.AvaII with commercially available and methylation sensitive R.AvaII, see lanes 1 and 2 of FIG. 18 . The methylated and unmethylated ROligo.AvaII were incubated in extracts of a first Lyngbya strain, a first Nostoc strain and a second Nostoc strain which all demonstrated R.AvaII activity, see lanes 4, 6 and 8 of FIG. 18 . R.AvaII activity in all strains depicted in FIG. 18 is blocked by M.AvaII methylation; compare lanes 4 to 5, 6 to 7, and 8 to 9 of FIG. 18 .
In another embodiment, endogenous E. coli Dam methyltransferase from a DNA free extract of E. coli can be used to methylate ROligos and plasmids. Effects of methylation by an endogenous E. coli Dam methyltransferase can be analyzed by using an E. coli extract to methylate ROligos. ROligos complimentary to restriction enzymes that contain a GATC site, such as R.BglII with a recognition site of AGATCT or R.BamHI with a recognition site of GGATCC, can be methylated and tested for protection against digestion.
In another embodiment, methylated nucleotides could be incorporated into ROligos by PCR. For recognition sites which cannot be methylated by in vitro methyltransferases, PCR could be used to incorporate methylated nucleotides. This method could be especially useful when there is no commercial enzyme available which methylates any given position in a restriction enzyme recognition sequence.
Additional Means of Identification of Unknown Restriction Endonucleases
Restriction endonucleases that do not recognize a ROligo from an existing ROligo library will not be identified. Thus, increasing the number of ROligos in a ROligo library will increase the number of potentially identifiable restriction enzymes.
In an embodiment, a method of identifying unidentified, cyanobacterial, restriction endonucleases is to generate a sequence that contains recognition sequences for restriction enzymes which are currently unknown in cyanobacteria. Thus, in a prophetic example, a sequence could be constructed in silico that contains recognition sequences of all known restriction enzymes in a database, e.g., REBASE, except for cyanobacterial recognition sequences or recognition sequences that are cleaved by cyanobacterial enzymes. For example, the cyanobacterial R.HaeIII has a recognition sequence of GGCC and also cuts the recognition sequence of the non-cyanobacterial restriction endonuclease R.NotI having a recognition sequence of GCGGCCGC. Therefore, the recognition sequence of R.NotI would be omitted.
After construction of the polynucleotide sequence containing all of the appropriate recognition sequences, the sequence would be incubated with cyanobacterial extracts. If the sequence was digested, the resulting band(s) would be isolated from a gel and directly sequenced with primers spanning the artificial sequence. This sequencing information would provide the recognition sequence and also the identity of the restriction enzyme active in the cyanobacterial extract. Thus, this prophetic embodiment would allow for the detection and identification of new enzymes currently not described in cyanobacteria.
In a prophetic embodiment, restriction endonucleases that cleave outside of the recognition sequence of a given ROligo could be detected through construction of a “megaROligo”. Detecting a double band cleavage product of a ROligo digested at the recognition site can be obfuscated when two restriction enzymes occur in one cyanobacterial extract, each digesting the ROligo. The restriction activity of both of the restriction endonucleases acting upon the ROligo at, for example, the recognition sequence and at a site in one of the flanking regions would result in a digestion pattern different from the predicted double band digestion product indicative of a single digestion at the recognition sequence by the correlating restriction endonuclease.
To account for restriction endonucleases that cut within a flanking arm of a given ROligo, a megaROligo could be constructed. The megaROligo would contain a recognition sequence that contains all recognition sequences that occur within the flanking arms that are used in the construction of the ROligos in a ROligo library. Consistent with the design principles of ROligos, this multiple recognition sequence would be flanked by an artificial sequence that is devoid of any recognition sequences from any known restriction enzymes.
In a prophetic embodiment, pre-incubation of this megaROligo in cyanobacterial extracts would allow testing for the presence of restriction endonucleases that recognize sequences in the flanking arms of all ROligos in the library. Thus, by incubating the megaROligo in the extract of cyanobacterial cells to be analyzed by the ROligo method, multiple restriction enzymes that digest a given ROligo and obscure the double band pattern expected by only digestion at the recognition sequence could be detected. This detection of multiple restriction enzymes digesting a ROligo in parallel would provide an additional quantum of information useful for determining the full complement of restriction endonucleases in a cyanobacterial extract.
In a prophetic example, to avoid digestion of ROligos in flanking regions ROligos could be engineered to be devoid of any recognition sequences in the flanking regions for any known restriction endonucleases.
All currently identified cyanobacterial enzymes except R.MboII are palindromic. Therefore, in a prophetic example, a “polypalindromic” sequence could be constructed that contains all possible palindromic sequences from 4-8 bp in length that are not recognized by anly known restriction enzymes. In a prophetic embodiment, incubating this polypalindromic sequence in a cyanobacterial extract could identify novel restriction enzymes from cyanobacteria.
Labeled ROligos
In a prophetic embodiment, ROligos could be labeled with fluorescent dyes, radiolabels, antigens or other labels. The labels could be incorporated through post-labeling techniques or could be incorporated using pre-labeled nucleotides through polynucleotide synthetic techniques well known in the art, see for example oligonucleotide synthetic services available through Iba life sciences in Olivette, Mo. at http://www.iba-lifesciences.com/Service_Custom_oligos.html and Shibata A., Molecules 2012 Feb. 29; 17(3):2446-63 which discloses various techniques and references to techniques well known in the art of oligonucleotide labeling.
In an embodiment, the labels used on each individual ROligo in a given ROligo pool would be uniquely identifiable. As an example, if a ROligo pool contained four different ROligos, each could be labeled with a dye that fluoresces at a wavelength that is different from the other three labels. Thus, even if multiple ROligos were digested and they all created digestion bands that were indistinguishable from one another on an ethidium bromide stained agarose gel, the identity of the double bands present could be visualized by measuring fluorescence at the four different wavelengths at which the four dyes fluoresce. For example, four different ROligos, ROligo.MstI, ROligo.SduI, ROligo.BlpI, and ROligo.BsteII could each be respectively labeled at a flanking region with a blue, green, yellow and red fluorescing dye. If a dual band pattern was visualized, one could not a priori determine which of the four ROligos had been digested. However, if the bands were then measured for fluorescence and a blue and red fluorescence were detected, the restriction enzymes of R.MstI and R.BsteII would have been identified. This multi-dimensional measurement would eliminate the additional step of runnming each of the ROligos individually from a ROligo pool that demonstrated a dual band pattern via ethidium bromide visualization on an agarose gel.
In another prophetic embodiment, as many different labels as there are ROligos in a library could individually be incorporated into an individual ROligo. Thus, if there were twenty ROligos in a ROligo library, there could be twenty different labels that are visualized by twenty distinct wavelengths, antibody-antigen bindings or other visualization means. In this prophetic embodiment, all twenty ROligos could be incubated in a cellular extract. The entire extract would then be run in a single lane on an agarose gel and the digestion bands present would be measured according to the twenty different visualization means.
Universal Methyltransferases
In a prophetic embodiment, E. coli strains are used to express cyanobacterial methyltransferases, e.g., M.SphI, M.BlpI, and M.AsuI from a Chlorogloeopsis strain, M.AcyI from a second Cyanobacterium strain and M.PvuI from a second Lyngbya strain. Activity of the expressed methyltransferases will then be analyzed in crude E. coli extracts similar to those depicted in FIG. 16 . The expressed methyltransferases will be analyzed for methylation and transformation efficiency. The effects that the expressed methyltransferases have on the resistance of ROligos to digestion and to transformation efficiency of methylated vectors in a cyanobacterial strain will also be compared to 2 bp-methyltransferases such as M.CviPI and M.SssI.
M.CviPI and M.SssI lead to hypermethylation of DNA while cyanobacterial methyltransferases methylate DNA at specific recognition sites. Hypermethylation could result in difficulty of the initial replication of plasmids in cyanobacteria and/or for the expression of a resistance marker. However, hypermethylation would not be maintained during replication as newly synthesized strands of DNA would be subject to only methylation by the endogenous methyltransferases which may or may not be able to act upon hemi-hypermethylated double stranded DNA.
Transformation success in a Chlorogloeopsis strain was increased by M.CviPI hypermethylation of a vector used for transformation. The cloned cyanobacterial methyltransferases that were used to methylate the vector used for transformation of a Chlorogloeopsis strain could also be used to check for their effect on digestion of vectors in different strains with the corresponding restriction enzyme.
Digestion by most of the identified cyanobacterial restriction endonucleases can be protected against by a combination of M.CviPI, M.SssI and M.AvaII. Thus, in a prophetic embodiment, a “universal” helper plasmid would be constructed wherein M.CviPI, M.SssI and M.AvaII methyltransferases are expressed. This universal helper plasmid could be used to help increase the transformation efficiency of many different vectors into various cyanobacterial strains.
Examples
Triton Washing of Cyanobacterial Cells
Cyanobacterial cells were washed three times with DNase-buffer (40 mM TRIS pH 7.4, 6 mM magnesium chloride, 2 mM calcium chloride) supplemented with 0.1% (v/v) triton X-100. After a final wash step with lysis buffer, cells were disrupted with glass beads. Cell-free medium (supplemented with DNase buffer) and cell-free supernatants after each wash step were incubated with about 300 ng to about 900 ng plasmid overnight at about 28° C. to account for the activity of unspecific nucleases.
Washing of Cyanobacterial Cells Using Buffers with pH Changes and Urea Supplementation
Cyanobacterial cells were washed two to three times with alternating buffers of acidic and basic pH (acidic buffer: 20 mM potassium phosphate pH 3 supplemented with 1 M urea; basic buffer: 20 mM potassium phosphate pH 11 supplemented with 1 M urea). Urea was used to destabilize proteins and expose amino acids. By rapidly changing the pH, a permanent denaturation of proteins might result and render unspecific nucleases ineffective at digestion. Finally, cells were washed twice with lysis buffer and disrupted with glass beads. The cell-free supernatant was used as the crude extract.
Preparation of Cyanobacterial Extracts
From a liquid preculture, 50 mL of a given cyanobacterial strain was inoculated to an OD 750 of between 0.5 to 1.0. After 10 days, 30 mL of that culture was pelleted (5 min 3000×g at room temperature), washed once with lysis buffer (40 mM sodium hydrogenphosphate pH 7.4, 1 mM EDTA, 5% (v/v) glycerol) and resuspended in 1 mL lysis buffer with 1 U RNaseI. Cells were disrupted with glass beads in a tissue lyser. The cell debris was removed by two consecutive centrifugation steps for each 5 min at 14000×g at room temperature. The final supernatant was used as crude extract.
Construction of ROligos
Examples of PCR primers used to generate various left arms and right arms of various ROligos can be found in FIG. 3 . The left arm, approximately a 170 bp fragment, was amplified using K230 ( FIG. 1 ) as a template and DreamTaq polymerase (35 cycles of (1 min at 95° C., 1 min at 59° C., 1 min at 72° C.) with forward primer #474 (see FIG. 3 ) and a reverse primer containing the site to be introduced. Similarly, the right arm, approximately a 210 bp fragment, was amplified with a forward primer containing the site to be introduced and thus was overlapping with the reverse primer of the respective left arm and a reverse primer #475. Both arms were gel isolated and used in a 1:20 dilution in an overlapping PCR with booster primers #472 and #473 (3′-truncated versions of primers #474 and #475) that resulted in a approximately 380 bp fragment, which was gel isolated and quantitatively amplified using primers #474 and #475 again and a proofreading polymerase, Pwo, with 250 μM dNTP each, 250 nM each primer, 2 U Pwo/100 μL reaction, 1.5 mM MgSO 4 and 35 cycles of (94° C. 15 sec, 63° C. 30 sec, 45 sec at 68° C.). ROligos were then checked by digesting with commercially available restriction enzymes to verify correct insertion.
Incubation of Plasmids with Crude Extract
Plasmid DNA was incubated in cyanobacterial extracts over night at 28° C. with up to 30 μg protein (with the extract volume comprising up to about 25% (v/v) of the total reaction volume) in NEB buffer P4. To test for restriction endonuclease activity, a 40 μL reaction with 600 ng plasmid was loaded directly onto a 0.8% (w/v) agarose gel. For direct sequencing, 2 μg of plasmidial DNA were digested and then column purified, for example.
Determination of Restriction Endonucleases in a Cyanobacterial Crude Extract
A cyanobacterial extract was incubated with a mixture of four different ROligos, a “ROligo pool”. Each constituent ROligo was added to the pool at 150 ng overnight at 28° C. with 25% (v/v) extract, in NEB buffers P1, P2, P3 and P4 (compositions thereof are well known in the art and can be found at REBASE or ordered from New England Biolabs) and analyzed on a 2% (w/v) agarose gel. Pool/buffer combinations, where a double band at 216/166 bp was observed were further analyzed by incubating the individual ROligos separately in the respective buffer overnight. The resulting pattern of digested, i.e., showing a double band of 216/166 bp, and non-digested ROligos in a given extract was evaluated using a digestion matrix and restriction endonucleases were identified accordingly.
Methylation of ROligos
Identification of restriction endonucleases that are present in a given strain of cyanobacteria allows for site specific methylation of heterologous DNA prior to transformation. Methylation of ROligos was achieved through using the following protocol: 1 μg ROligo was methylated in a buffer containing 40 mM potassium phosphate buffer at pH 8.0 (or TRIS), 10 mM DTT, 25 mM EDTA supplemented with 640 μM S-adenosylmethionine (SAM) and a suitable methyltransferase. Methyltransferases were used at either 50 μL of an E. coli extract containing suitable methyltransferases or commercially available methyltransferases such as M.CviPI or M.SssI at about 0.5 to 2 U per μg DNA to be methylated. The final volume for the methylation reaction was 200 μL. The methylation reaction proceeded overnight at 37° C. Afterwards, the ROligo was column purified. Methylation success was checked by restriction digestion with an appropriate, commercially available restriction enzyme that is blocked by the type of methylation used (which can be found in charts depicting methylation sites for known restriction endonucleases, see REBASE). Methylation events that prevented the methylated ROligos from digestion by cyanobacterial extracts can then be applied to methylate heterologous plasmids prior to transformation into a given cyanobacterial strain.
Preparation of an E. coli Extract with Methyltransferase Activity
To methylate using the endogenous E. coli Dam methyltransferase, an extract of E. coli alphaSelect (available from Bioline USA Inc. of Taunton Mass.) was grown for 2 days at room temperature in 1% (w/v) soy bean peptone/0.5% (w/v) yeast extract was used (the final extract volume was 1/100 of the initial culture volume). Cells were disrupted in lysis buffer as above for cyanobacterial extracts. To methylate using the M.AvaII methyltransferase, E. coli HB101 harboring the pRL528 plasmid (Elhai, J, et. al, J. Bacteriology 1997, 179(6):1998-2005) grown for 2 days was used. E. coli extracts were DNaseI-treated (2 U per ml extract volume) for 30 min at 37° C. after supplementation with DNaseI buffer (NEB). The DNA-free extract was then used to methylate ROligos.
Sequencing of Extract-Digested Plasmids
A cyanobacterial-extract digested plasmid was sequenced after column purification with a variety of sequencing primers that span the whole plasmid in about 1 kb distances between adjacent primers. A cleavage point was identified when the sequencing signal drops sharply, see FIG. 19 , for example. A recognition site around a given cleavage point was then identified and thus the corresponding restriction endonuclease was identified.
Other Embodiments
A number of embodiments of the disclosure have been described. Nevertheless, it will be understood that various modifications may be made without departing from the spirit and scope of the disclosure. Such modifications will become apparent to those skilled in the art from a consideration of this disclosure or practice of the disclosure disclosed herein. Consequently, it is not intended that this disclosure be limited to the specific embodiments disclosed herein, but that it cover all modifications and alternatives coming within the true scope and spirit of the disclosure as embodied in the attached claims.
1.PublishNumber: US-2015184221-A1
2.Date Publish: 20150702
3.Inventor: MEIDLINGER MARC
4.Inventor Harmonized: MEIDLINGER MARC(DE)
5.Country: US
6.Claims:
(en)Methods are provided for analyses of restriction endonucleases.
7.Description:
(en)REFERENCE TO SEQUENCE LISTING
This application contains a sequence listing submitted by EFS-Web, thereby satisfying the requirements of 37 C.F.R. §§1.821-1.825. The Sequence Listing, created on Dec. 15, 2013, is named “Sequence listing P0031.01.US_ST25”.
BACKGROUND
Cyanobacteria, also known as blue-green algae, are photosynthetic bacteria widespread in marine and freshwater environments. Cyanobacteria have simple growth requirements and grow to high densities by using light, carbon dioxide, and other inorganic nutrients for growth. They are useful hosts for the production of a wide range of compounds of interest including biofuels and other commodity chemicals, for example.
A barrier to using cyanobacteria for the production of compounds of interest is that efficient introduction of engineered vectors in cyanobacteria is hampered, or completely prohibited, due to the presence of restriction nucleases, see Thorsten H. et. al., Ch. 24, Synthetic Biology in Cyanobacteria: Engineering and Analyzing Novel Functions, In: Methods in Enzymology, Academic Press, 2011, Vol. 497, pp. 539-579. These nucleases can significantly decrease or completely prevent the uptake of exogenous DNA into cyanobacterial cells, see for example Ruffing A M. et. al., Biotech. Bioeng., 2012, Vol. 109, pp. 2190-2199. Enzymes exhibiting nuclease activity are usually either endonucleolytic (an endonuclease) or exonucleolytic (an exonuclease) and may be either sugar specific or non-sugar specific. A catalog of restriction enzymes and their recognition sites is compiled and maintained by New England Biolabs in Ipswich, Mass. and can be accessed at http://rebase.neb.coil (REBASE).
Various restriction endonucleases exhibit restriction activity to varying degrees in various buffer conditions. Buffer conditions conunonly used for known restriction enzymes are mostly covered across four different buffers available through New England Biolabs, P1, P2, P3 and P4 buffers. Their compositions can be found at REBASE and are well known to those having skill in the art.
Cyanobacteria often contain one or more sequence-specific restriction endonucleases. These nucleases are almost always associated with their cognate methyltransferase. The methyltransferase methylates the host chromosomal and plasmid DNA to provide protection against its associated restriction enzyme. Endogenous cyanobacterial DNA is thus protected against specific nuclease activity by specific methylation, but foreign DNA is not methylated and is therefore targeted for destruction by the cyanobacterial nucleases, see Soper, B W et. al, J. Bacteriology 1994, 176(17):556-5570 and Elhai, J, et. al, J. Bacteriology 1997, 179(6): 1998-2005.
SUMMARY
Disclosed herein are methods that provide for the identification of cyanobacterial restriction endonucleases that are present in crude extracts. The method is based on specific PCR products (ROligos) which are designed with unique restriction endonuclease (“REN”) recognition sequences leading to two easily distinguishable bands upon digestions with cyanobacterial crude extract. In addition, incubation of methylated ROligos in cyanobacterial extracts allows the determination of whether endonucleases present can be rendered ineffective by the provided methylation type. Hence, any methyltransferase that provides the desired type of methylation can be used to methylate and protect DNA in vitro or in vivo prior to transformation.
In an aspect, a method for identifying restriction endonucleases in a cyanobacterial cell is disclosed having the steps of a) incubating oligonucleotides from a ROligo library with a cyanobacterial cell extract from said cyanobacterial cell, wherein each oligonucleotide comprises a recognition sequence of a cyanobacterial restriction endonuclease, and b) analyzing digestion products of said oligonucleotides for digestion at said recognition sequences, and c) identifying restriction endonucleases in said cyanobacterial cell extract by analyzing digestion products of said oligonucleotides. In an embodiment, the method is can be used on cyanobacterial cells including Prochlorococcus, Synechocystis, Synechocococus, Chroococcales, Cyanobium, Oscillatoriales, Cyanobacterium, Pleurocapsales, Geitlerinema, Phormidiumn, Euhalothece, Anabaena, Lyngbya, Spirulina, Nostoc, Pleurocapsa , and Leptoiyngbya . In another embodiment, the method discloses restriction endonucleases of step c) that are analyzed through depicting the results of digestion patterns at recognition sequences of said oligomcleotides from a ROligo library in an infographic table having all incubated oligonucleotides and all restriction endonucleases whose recognition sequences are part of the oligonucleotides, and wherein restriction endonucleases are identified from said cyanobacterial cell extract through the following stepwise manipulations of the digestion patterns depicted in the infographic table including the steps of a) first, restriction endonucleases which do not digest at least one of the oligonucleotides incubated with said cyanobacterial cell extract are removed from the infographic table, and b) second, restriction endonucleases which do not digest oligonucleotides from the ROligo library containing their recognition sequences, but are capable of recognizing other oligonucleotides that have been digested are removed from the infographic table, and c) third, restriction endonucleases in the cyanobacterial extract are identified as the restriction endonucleases not removed from the infographic table. In another embodiment, the uses an infographic table that is a digestion matrix. In an embodiment, the method is disclosed where each oligonucleotide comprises a left arm, a right arm, and a recognition sequence for a cyanobacterial restriction endonuclease wherein the oligonucleotide is produced through using a polymerase chain reaction wherein the left arm is a double stranded polynucleotide and the right arm is a double stranded polynucleotide and wherein the 5′ end of a single stranded reverse primer of said left arm and the 5′ end of a single stranded forward primer of the right arm overlap and make a recognition sequence for a cyanobacterial restriction endonuclease and wherein the left arm and the right arm primers use a template lacking any recognition sites for cyanobacterial restriction endonucleases. In yet another embodiment, the method uses a recognition sequence that is positioned asymmetrically along the length of each oligonucleotide such that digestion by the restriction endonuclease creates two digestion products of different sizes. In an embodiment, the method uses a template that is SEQ ID NO: 1 and wherein said left arm primer and said right arm primer include primers selected from the group consisting of SEQ ID NOs: 5-62.
In an aspect, a method for transfonning cyanobacterial host cells with vectors is disclosed including a) incubating oligonucleotides from a ROligo library with a cyanobacterial cellular extract from the cyanobacterial host cell, wherein each oligonucleotide comprises a recognition sequence of a restriction endonuclease, and b) analyzing digestion products of the oligonucleotides for digestion at the recognition sequences, and c) identifying restriction endonucleases in said cyanobacterial cellular extract by analyzing digestion products of the oligonucleotides, and d) constructing a vector lacking said recognition sequences of the identified restriction endonucleases of step c), and e) transforming said cyanobacterial host cell with said vector of step d). In an embodiment, the method uses restriction endonucleases of step c) that are analyzed through depicting the results of digestion patterns at recognition sequences of said oligonucleotides from a ROligo library in an infographic table comprising all incubated oligonucleotides and all restriction endonucleases whose recognition sequences are part of the oligonucleotides, and wherein restriction endonucleases are identified from the cyanobacterial cell extract through the following stepwise manipulations of the digestion patterns depicted in said infographic table by the following steps: a) first, restriction endonucleases which do not digest at least one of the oligonucleotides incubated with said cyanobacterial cell extract are removed from said infographic table, and b) second, restriction endonucleases which do not digest oligonucleotides containing their recognition sequences, but are capable of recognizing other oligonucleotides that have been digested are removed from said infographic table, and c) third, restriction endonucleases in the cyanobacterial extract are identified as the restriction endonucleases not removed from said infographic table. In another embodiment, the method uses an infographic table that is a digestion matrix. In yet another embodiment, the method uses oligonucleotides wherein each oligonucleotide includes a left arm, a right arm, and a recognition sequence for a cyanobacterial restriction endonuclease, and wherein the oligonucleotide is produced through using a polymerase chain reaction wherein said left arm is a double stranded polynucleotide and the right arm is a double stranded polynucleotide and wherein the 5′ end of a single stranded reverse primer of said left arm and the 5′ end of a single stranded forward primer of said right arm overlap and comprise a recognition sequence for a cyanobacterial restriction endonuclease and wherein the left arm and said right arm primers use a template lacking any recognition sites for cyanobacterial restriction endonucleases. In yet another embodiment, the method uses oligonucleotides having a recognition sequence that is positioned asymmetrically along the length of the each oligonucleotide such that digestion by the restriction endonuclease creates two digestion products of different sizes. In an embodiment, the method uses a template that is SEQ ID NO: 1, and wherein the left arm primer and the right arm primer comprise single stranded polynucleotide primers are selected from the group consisting of SEQ ID NOs: 5-62.
In another aspect a method for transforming cyanobacterial host cells with vectors is disclosed and includes a) incubating oligonucleotides from a ROligo library with a cyanobacterial cellular extract from the cyanobacterial host cell, wherein each oligonucleotide comprises a recognition sequence of a restriction endonuclease, and b) analyzing digestion products of the oligonucleotides for digestion at the recognition sequences, and c) identifying restriction endonucleases in said cyanobacterial cellular extract by analyzing digestion products of the oligonucleotides, and d) methylating recognition sequences of the identified restriction endonucleases in a vector, and e) transforming the cyanobacterial host cell with said vector of step d). In an embodiment, the method uses recognition sequences in a vector that are methylated in step d) and are methylated by methyltransferases. In another embodiment, the method uses methyltransferases that are selected from the group consisting of M.CviPI, M.SssI and M.AvaII. In yet another embodiment, the method of uses restriction endonucleases of step c) that are analyzed through depicting the results of digestion patterns at recognition sequences of the oligonucleotides from a ROligo library in an infographic table comprising all incubated oligonucleotides and all restriction endonucleases whose recognition sequences are part of the oligonucleotides, and wherein restriction endonucleases are identified from the cyanobacterial cell extract through the following stepwise manipulations of the digestion patterns depicted in the infographic table including the steps of a) first, restriction endonucleases which do not digest at least one of the oligonucleotides incubated with the cyanobacterial cell extract are removed from said infographic table, and b) second, restriction endonucleases which do not digest oligonucleotides containing their recognition sequences, but are capable of recognizing other oligonucleotides that have been digested are removed from the infographic table, and c) third, restriction endonucleases in said cyanobacterial extract are identified as the restriction endonucleases not removed from the infographic table. In another embodiment, the method uses an infographic table that is a digestion matrix. In an embodiment, the method uses oligonucleotides wherein each oligonucleotide includes a left arm, a right arm, and a recognition sequence for a cyanobacterial restriction endonuclease, and wherein the oligonucleotide is produced through using a polymerase chain reaction wherein the left arm is a double stranded polynucleotide and the right arm is a double stranded polynucleotide and wherein the 5′ end of a single stranded reverse primer of said left arm and the 5′ end of a single stranded forward primer of the right arm overlap and make a recognition sequence for a cyanobacterial restriction endonuclease and wherein the left arm and the right arm primers use a template lacking any recognition sites for cyanobacterial restriction endonucleases. In yet another embodiment, the method uses oligonucleotides that have recognition sequences that are positioned asymmetrically along the length of each oligonucleotide such that digestion by the restriction endonuclease creates two digestion products of different sizes. In another embodiment, the method uses a template that is SEQ ID NO: 1, and wherein the left arm primer and the right arm primer comprise single stranded polynucleotide primers selected from the group consisting of SEQ ID NOs: 5-62.
In an aspect, a method for identifying restriction endonucleases is disclosed having the steps of a) incubating oligonucleotides including a ROligo library with a cyanobacterial cell extract wherein each oligonucleotide of said ROligo library includes a recognition sequence of a restriction endonuclease, and b) wherein each oligonucleotide of said ROligo library is labeled with a unique marker that is uniquely visualized, and c) analyzing digestion products of said ROligo library of oligonucleotides for digestion at said recognition sequences by visualizing said markers, and d) identifying restriction endonucleases in said cyanobacterial cell extract by correlating said visualized markers of step c) with their corresponding oligonucleotides. In an embodiment, the method uses oligonucleotides wherein each oligonucleotide includes a left arm, a right arm, a marker, and a recognition sequence for a cyanobacterial restriction endonuclease wherein the oligonucleotide is produced through using a polymerase chain reaction wherein the left arm is a double stranded polynucleotide and the right arm is a double stranded polynucleotide and wherein the 5′ end of a single stranded reverse primer of the left arm and the 5′ end of a single stranded forward primer of the right arm overlap and comprise a recognition sequence for a cyanobacterial restriction endonuclease and wherein the left arm and the right arm primers use a template lacking any recognition sites for cyanobacterial restriction endonucleases. In another embodiment, the method uses oligonucleotides wherein the recognition sequence is positioned asymmetrically along the length of said each oligonucleotide such that digestion by the restriction endonuclease creates two digestion products of different sizes. In yet another embodiment, the method uses a template that is SEQ ID NO: 1, and wherein the left arm primer and the right arm primer comprise single stranded polynucleotide primers selected from the group consisting of SEQ ID NOs: 5-62, and wherein the marker is selected from the group consisting of fluorescent dyes, radiolabels, and antigens.
BRIEF DESCRIPTION OF DRAWINGS
FIG. 1 depicts a plasmid map of K230 (left portion of FIG. 1 ) and a schematic depiction of ROligo construction (right portion of FIG. 1 ).
FIG. 2 depicts a list of known cyanobacterial restriction endonucleases, their recognition sequences and the sequences that were used to construct ROligos.
FIG. 3 depicts sequences of primers used to construct the left and right arms of various ROligos.
FIG. 4 depicts an agarose gel analysis of ROligo.AvaI. ROligo.AvaII and ROligo.SphI undigested or digested with commercially available, corresponding restriction nucleases.
FIG. 5 depicts an agarose gel analysis of the restriction enzyme activity of cyanobacterial extracts on plasmids K230 and K244.
FIG. 6 is an infographic chart depicting the theoretical capability of various nucleases to produce a double band pattern when acting upon various ROligos.
FIG. 7 depicts an agarose gel analysis of the restriction enzyme activity of extracts from Triton or pH/urea treated cyanobacterial cells on plasmids K230, K236Cm, and K244 and a PCR product.
FIG. 8 depicts a plasmid map of K236Cm.
FIG. 9 depicts a plasmid map of K244.
FIG. 10 depicts an agarose gel analysis of the restriction enzyme activity in an extract from a first Cyanobacterium strain upon various ROligo pools.
FIG. 11 depicts an agarose gel analysis of the restriction enzyme activity in an extract from a first a first Cyanobacterium strain upon ROligo.AsuI, ROligo.AvaII, ROligo.HaeII, ROligo.SduI, ROligo.AvaIII, ROligo.NspI, ROligo.SphI, and ROligo.BamHI.
FIG. 12 is an infographic chart depicting the results of incubation of various ROligos as a step in the exemplification of the ROligos method for identifying restriction enzymes in a first Cyanobacterium strain extracts.
FIG. 13 is an infographic chart depicting the results of incubation of various ROligos as a step in the exemplification of the ROligos method for identifying restriction enzymes in a first Cyanobacterium strain extract.
FIG. 14 is an infographic chart depicting the results of incubation of various ROligos as a step in the exemplification of the ROligos method for identifying restriction enzymes in a first Cyanobacterium strain extracts.
FIG. 15 is a table depicting cyanobacterial strains analyzed for restriction endonuclease activity using the ROligo method.
FIG. 16 depicts an agarose gel analysis of Gm5C-methylated ROligo.PvuII or TK18 plasmid incubated with a first Lyngbya strain and a Chlorogloeopsis strain extract.
FIG. 17 depicts a plasmid map of TK018.
FIG. 18 depicts an agarose gel analysis of methylated (by a DNA-free Escherichia coli extract harbouring a plasmid expressing M.AvaII methyltransferase) and unmethylated ROligo.AvaII in various cyanobacterial extracts.
FIG. 19 depicts a sequencing readout of a plasmid K230 digested in a Chlorogloeopsis strain extract.
DETAILED DESCRIPTION
Disclosed herein are methods for the identification of restriction endonucleases in cyanobacterial extracts. Identifying restriction endonucleases present within a cyanobacterial strain allows for protection against digestion of introduced vectors or other constructs. In an embodiment, engineered vectors to be used for transformation of a cyanobacterial strain do not contain the recognition site of restriction endonucleases identified in that strain. In another embodiment, engineered vectors to be used for transformation are protected against digestion (restriction) by methylating at extant recognition sites of the identified restriction endonucleases.
DEFINITIONS
The section headings used herein are for organizational purposes only and are not to be construed as limiting the described subject matter in any way. When definitions of terms in incorporated references appear to differ from the definitions provided in the present teachings, the definition provided in the present teachings shall control. Slight and insubstantial deviations from metrics used herein are within the scope of the present teachings herein. In this application, the use of the singular includes the plural unless specifically stated otherwise. Also, the use of “comprise,” “comprises,” “comprising,” “contain,” “contains,” “containing,” “include,” “includes,” and “including” are not intended to be limiting. It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention.
Unless otherwise defined, scientific and technical terms used in connection with the invention described herein shall have the meanings that are commonly understood by those of ordinary skill in the art. Generally, nomenclatures utilized in connection with, and techniques of, cell and tissue culture, molecular biology, protein and nucleotide/polynucleotide chemistry and hybridization described herein are those well-known and commonly used in the art. Standard techniques are used, for example, for nucleic acid purification and preparation, chemical analysis, recombinant nucleic acid, and oligonucleotide synthesis. Enzymatic reactions and purification techniques are performed according to manufacturer's specifications or as commonly accomplished in the art or as described herein. The nomenclatures utilized in connection with, and the laboratory procedures and techniques of described herein are those well-known and commonly used in the art, see, e.g., Sambrook et al., Molecular Cloning: A Laboratory Manual (Third ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. 2000). Standard recombinant DNA procedures were used for gene cloning, plasmid isolation, and electroporation. Manufacturer protocols and standard methods (Sambrook et al., 2000) were followed for DNA purification (e.g., Qiagen, Valencia, Calif.), restriction endonuclease digestion (NEB, Ipswich, Mass.), and DNA amplification using DNA polymerases (NEB, Ipswich, Mass.).
As used in accordance with the embodiments provided herein, the following terms, unless otherwise indicated, shall be understood to have the following meanings:
The term “nucleic acid” and “nucleic acid molecule” refer to natural nucleic acid sequences such as DNA (deoxyribonucleic acid) and RNA (ribonucleic acid), artificial nucleic acids, analogs thereof, or combinations thereof.
The term “oligonucleotide” and “polynucleotide” are synonymous as used herein with each referring to nucleic acids comprising more than one nucleotide. Oligonucleotides and polynucleotides as used herein may be single stranded, double stranded, or a mixture of both single and double strands.
The term “nuclease” applies to enzymes that cleave a phosphodiester bond between nucleotides in a polynucleotide.
The term “recognition sequence” refers to the sequence of polynucleotides comprising the substrate of a nuclease.
The term “endonuclease” applies to nucleases that break nucleic acid chains somewhere in the interior, rather than at the ends, of a polynucleotide.
The term “exonuclease” applies to nucleases that functions by removing nucleotides from the ends of a polynucleotide.
The term “restriction endonuclease” refers to an endonuclease that “restricts”, meaning “digests”, meaning “cuts”, meaning breaks/cleaves phosphodiester bonds between nucleotides on each strand (parallel and antiparallel) in a polynucleotide, thus creating smaller polynucleotides. The terms “restricts”, “digests” and “cleaves” as used herein are synonymous when used in the context of nuclease activity.
The term “plasmid” refers to a circular nucleic acid vector. Generally, plasmids contain an origin of replication that allows many copies of the plasmid to be produced in a bacterial (or sometimes eukaryotic) cell.
A “shuttle vector” refers to a vector which can propagate in two different host species.
The term “construct” as used herein refers to a recombinant nucleic acid molecule that has been generated for the purpose of the expression of a specific nucleotide sequence or sequences, or is to be used in the construction of other recombinant nucleotide sequences. In general, “construct” is used herein to refer to a recombinant nucleic acid molecule. Generally, the term “construct” and the term “vector” are synonymous as used herein.
The term “host cell” as used herein refers to a cell into which DNA can be introduced by any appropriate means (e.g. natural uptake, transformation, transfection, electroporation, and conjugation).
The term “transformation” as used herein refers to a permanent or transient genetic change, e.g., a permanent genetic change induced in a cell following incorporation of non-host nucleic acid sequences.
The term “ROligo” as used herein refers to an oligonucleotide containing arecognition sequence of at least one restriction endonuclease.
The term “ROligo library” refers to multiple ROligos. A ROligo library can consist of a group of ROligos that each individually contain a recognition sequence of a restriction endonuclease from a particular organism and the group of ROligos together make up a ROligo library whose ROligos together contain all of the known recognition sequences of that particular organism.
The terms “information graphics” or “infographics” are used herein refer to graphic visual representations of information, data or knowledge intended to present complex information quickly and clearly. Infographics utilize graphics to enhance the human visual system's ability to see patterns and trends. As used herein, an example of an infographic is a digestion matrix.
The term “digestion matrix” as used herein refers to an infographic table depicting the theoretical capability of various restriction endonucleases to cleave a ROligo oligonucleotide to the double band pattern. A digestion matrix can be used as an infographic. A digestion matrix depicting the theoretical results of the digestion of all constructed oligonucleotides can be manipulated through the application of a set of rules to afford the identity of active restriction endonuclease in a cyanobacterial extract.
The following figures, description, and examples illustrate certain embodiments of the present disclosure in detail. Those of skill in the art will recognize that there are numerous variations and modifications that are encompassed by its scope. Accordingly, the description of certain embodiments should not be deemed to limit the scope of the present disclosure.
Sequencing Method of Determining Restriction Endonucleases
Incubation of a plasmid in a cyanobacterial extract often results in a large number of bands, especially if enzymes with 4-5 bp recognition sequences like R.HaeIII which has a recognition sequence of GGCC or R.AvaII which has a recognition sequence of GGWCC (where W is A or T) are active. Gel isolation and cloning of every fragment after treatment with mung bean nuclease to degrade both 5′ and 3′ single-stranded extensions from the ends of DNA molecules, to generate blunt ends to enable ligation for sequencing would be very time and labor consuming. In case of multiple cutting enzymes, the same enzyme would be identified repeatedly, rendering most of the work non-informative.
As an example of the sequencing method of identifying recognition sequences, a plasmid, K230 (SEQ ID NO: 1) was incubated with a Chlorogloeopsis strain extract. A plasmid map of K230 is depicted in the left portion of FIG. 1 . The digested plasmidial DNA was subjected to sequencing reactions without gel purification of individual bands. For the sequencing reaction, a set of spanning primers (with around 1 kb distance between two adjacent primers) covering the whole K230 plasmid was used because the cleavage points were not known. FIG. 2 depicts the result of the sequencing reaction for a digested portion of the K230 plasmid. As depicted in FIG. 2 for K230 digested with a Chlorogloeopsis strain crude extract, a cleavage point was identified when the sequencing signal dropped sharply at an R.SphI site.
Through multiple rounds of using the labor-intensive and time-consuming sequencing method, enzymes R.BlpI and R.SphI in a Chlorogloeopsis strain. R.AvaII and R.MstI in a second Nostoc strain and R.AcyI in a second Cyanobacterium strain were identified.
The sequencing method for identifying active restriction endonucleases in a cyanobacterial extract has at least two disadvantages. First, the DNA sent to be sequenced contained a vast amount of cyanobacterial genomic DNA from the extract incubation, so the background often was high and rendered many sequencing reactions non-readable and thus required several repetitions reliable sequencing information. Secondly, the sequencing method relies on complete digestion at least at some cleavage points as even traces of undigested plasmid would allow a read-through instead of a signal drop. Therefore, the sequencing method can only be reliably used if completeness of digestion is assured.
Both gel isolation of bands with subsequent cloning and the direct sequencing method share the same disadvantages as the same enzyme might be identified repeatedly. A different method is disclosed herein that improves identification of restriction endonucleases in cyanobacterial extracts, reduces work load, simplifies and accelerates analyses to provide information about whether all restriction enzymes that are present in a cellular extract have been identified.
ROligo Method for Identification of Restriction Endonucleases
In an embodiment, a method is disclosed (herein referred to as the “ROligo” method “R” from restriction site, “Oligo” from oligonucleotide) to identify restriction endonucleases in cellular extracts. In an embodiment, the ROligo method disclosed herein is an improvement over isolating and sequencing digested polynucleotides using the sequencing method in order to identify restriction endonucleases present in a cell.
In an embodiment, ROligos are an oligonucleotide with one introduced recognition site for a known cyanobacterial restriction endonuclease that are constructed by performing overlapping PCR on two sequences (the “left arm” and “right arm”), whose ends contain the recognition sequence of a given restriction endonuclease to be introduced between the two arms, see the right portion of FIG. 1 . In another embodiment, ROligos contain recognition sequences for restriction endonucleases from any given organism such as, for example, E. coli.
In an embodiment, a collection of multiple different ROligos is referred to as a ROligo library. In an embodiment, a ROligo library contains ROligos designed to test for restriction endonucleases from a particular organism. As an example, a ROligo library could be created containing ROligos that cumulatively contain all of the known recognition sequences of cyanobacterial restriction endonucleases as listed in REBASE or any other catalog listing cyanobacterial restriction endonucleases.
In an embodiment, the ROligo method includes incubating various combinations of different ROligos from a ROligo library in a cellular extract and observing a cleavage pattern of the ROligos by agarose gel electrophoretic analysis. The cleavage pattern (pattern of bands of ethidium bromide stained polynucleotides as visualized in the agarose gel using ultraviolet light) indicates the occurrence of enzymes recognizing the introduced restriction sites. Individual restriction endonucleases can then be identified by subsequent incubation of single ROligos from the digested ROligo pool. Thus, in an embodiment of the ROligo method, specific restriction endonucleases can be rapidly identified without additional isolation and sequencing of digested DNA. Additionally, the ROligo method, in contrast to the sequencing method, does not rely on complete digestion of a polynucleotide sequence, i.e. even if only a fraction of the incubated ROligo molecules are digested at the introduced recognition site, the double band pattern will occur.
FIG. 13 depicts a list of known cyanobacterial restriction endonucleases and their recognition sequences that were used to construct ROligos. The sequence which is introduced between the ROligo arms is in usually identical to the recognition sequence of the corresponding restriction endonuclease. However, in cases of a degenerate recognition sequence, a specific sequence among multiple potential sequences is chosen. Some recognition sequences are preceded (e.g. R.AsuI) or followed (e.g. R.SalI) by additional base(s), here a T base, in order to disrupt otherwise emerging recognition sites of further nucleases in the overlap between the introduced site and the adjacent arm sequence. As depicted in the FIG. 1 in the plasmid of K230, contains a “REfree region” from base pairs 7550 to 7926 of SEQ ID NO: 1 that is devoid of recognition sites for known cyanobacterial enzymes. In an embodiment, this REfree region is used as the template for a two-step PCR protocol to construct the left and right arms of ROligos with specific recognition sites and lacking all other known cyanobacterial recognition sequences elsewhere within the ROligo polynucleotide. Depicted as boxes in the right portion of FIG. 1 , ROligo recognition sequences are introduced at an uncentered, asymmetrical position in the ROligo.
The design of a ROligo library needs to be carefully considered in order to create as many ROligos that are only digested by a single restriction endonuclease to the double band pattern. FIG. 2 depicts known cyanobacterial restriction enzymes, their general recognition sequence and the specific sequence used to generate the recognition sequence in the corresponding ROligo. In an embodiment, a ROligo library can be created using the recognition sequences of the cyanobacterial restriction endonucleases as depicted in FIG. 2 . As an example of decreasing the number of ROligos cut by more than one known cyanobacterial enzyme, the ROligos used in embodiments described herein were designed keeping in mind to following information. As depicted in FIG. 2 , the asterisk (*) denotes that these restriction enzymes, found in a second round of literature search after constructing the first ROligos, cut within the arms of the ROligos and, therefore, do not produce the indicative double band pattern of an embodiment of the ROligo method. The superscripted 1 denotes that a blocking base, T, was inserted before the recognition sequence to disrupt an otherwise emerging ApaI site at the junction to the left arm. The superscripted 2 denotes that a blocking base, T, was inserted before the recognition sequence to disrupt an otherwise emerging AsuI site at the junction to the left arm. The superscripted 3 denotes that a blocking base, T, was inserted before the recognition sequence to disrupt an otherwise emerging GC stretch at the junction to the left arm. As the enzyme's recognition site contains a GC, methylation sensitivity could be checked by M.CviPI. Without the additional T base, a possible blockage might result from the outer C, which could only be methylated in specific nucleotide surroundings. The superscripted 4 denotes a blocking base, T, that was inserted after the recognition sequence to disrupt an otherwise emerging CG stretch at the junction to the right arm. As the enzyme's recognition site contains a CG, methylation sensitivity could be checked by M.SssI. Without the additional T base, a possible blockage might result from the outer C, which could only be methylated in specific nucleotide surroundings.
Use of ROligo Method on Cyanobacteria
In an embodiment, the ROligo method disclosed herein is used to identify restriction endonucleases in cyanobacterial genera, species and/or strains. Non-limiting examples of cyanobacterial genera, species and strains that are members thereof, upon which the ROligo method can be used include, but are not limited to, Prochlorococcus, Synechocystis, Synechococcus, Chroococcales, Cyanobium, Oscillatoriales, Cyanobacterium, Chlorophyta, Pleurocapsales, Geitlerinema, Phormidium, Euhalothece, Anabaena, Lyngbya, Spirulina, Nostoc, Pleurocapsa , and Leptolyngbya . In another embodiment of the ROligo method, cell and cellular extracts to be tested for restriction endonuclease activity include all bacteria and Archaea. In an embodiment, certain eukaryotic cells may be tested by embodiments of the ROligo method.
Development of the ROligo Method
In contrast to other methods used to identify restriction endonucleases, the ROligo method does not rely on complete digestion of the polynucleotides incubated with cellular extracts or other solutions containing restriction endonucleases, i.e. even if only a fraction of the polynucleotide molecules are cleaved at the introduced restriction recognition site, the double band pattern will occur
While the sequencing method uses a plasmid with many restriction sites and aims to identify the cleavage point, the ROligo method uses an oligonucleotide with exactly one recognition site for a known cyanobacterial enzyme. In an embodiment of the ROligo method, about 27 different ROligos were constructed by overlapping PCRs using primers having SEQ ID NOs: 5-62 as depicted in FIG. 3 . The strand backbone for the ROligo originated from the K230 plasmid (see FIG. 1 ) which contains a sequence of 360 bp that shows no recognition site for any of the known cyanobacterial restriction endonucleases, see FIG. 2 . In an embodiment of the ROligo method, this K230 stretch is used and a known cyanobacterial restriction site is inserted at an uncentered, asymmetric position resulting in a PCR-derived DNA fragment of about 380 bp length, see right side of FIG. 1 . In an embodiment, cleavage of the generated ROligos results in a double band of about 216 and about 166 bp that are generated by digestion of a ROligo at the recognition sequence, the approximate length of the individual arms. These bands are readily discernible by using a lane containing size markers as one ROligo cleavage band lies above and one ROligo cleavage band lies below the 200 bp marker band of the size marker standards.
FIG. 4 depicts digested and undigested ROligos analyzed by agarose gel electrophoresis stained with ethidium bromide and visualized under ultraviolet light. Lanes 1-3 contain, respectively, undigested ROligos, ROligo.AvaI, ROligo.AvaII, and ROligo.SphI. Lanes 4-6, contain, respectively, ROligo.AvaI digested with R.AvaI (R. stands for restriction endonuclease, thus R.AvaI is restriction endonuclease AvaI), ROligo.AvaII digested with R.AvaII, and ROligo.SphI digested with R.SphI. Lane M contains a size marker.
The ROligos with introduced recognition sites for R.AvaI (lane 1), R.AvaII (lane 2) and R.SphI (lane 3) were constructed by PCR. The undigested PCR products (lanes 1-3) depict a full length 380 bp band. ROligos were digested with appropriate commercially available enzymes (lanes 4-6) and the desired double band at 216/166 bp was observed.
For some ROligos, digestion with commercially available enzymes leaves a faint undigested band (see FIG. 4 , lane 4). This may be due to the use of three consecutive PCR reactions in the ROligo construction process using a polymerase with relatively low fidelity such as DreamTaq polymerase. To reduce the number of such undigestable PCR molecules, a polymerase with higher fidelity, Pwo available from Roche Applied Science, was used for PCR amplification of the ROligos.
Digestion patterns other than the double band pattern of 216/166 that are created by digestion of a ROligo at the introduced recognition sequence are not investigated. Thus, restriction enzymes that cut at sites other than the introduced recognition sequence will be missed.
Digestions of Plasmids with Cyanobacterial Extracts
FIG. 5 depicts an analysis of the digestion of plasmids by cyanobacterial extracts. The first panel depicts undigested plasmid K230, undigested plasmid K244, and a marker lane. Moving from left to right, the cyanobacterial extracts used were a fourth Cyanobacterium strain, a third Cyanobacterium strain, and a Cyanothece strain (furthest right panel). For each of the cyanobacterial extract panels, lane M is a size marker, lane 1 is a crude cyanobacterial extract without plasmid addition, lane 2 is plasmid K230 incubated in the extract, and lane 3 is plasmid K244 incubated in the extract. Cyanobacterial extracts were incubated with plasmids K230 and K244 to check for overall restriction activity. A fourth Cyanobacterium strain shows a specific digestion banding pattern indicating the presence of one or more restriction endonucleases. A third Cyanobacterium strain shows smearing of both plasmids indicating the presence of an unspecific, possibly extracellular, endonuclease or exonuclease that could obscure specific endonucleases that are active at the same time in the same extract upon the same plasmid. Incubation of the plasmids with a Cyanothece strain extract depicts the plasmids as being undigested.
Potential Digestions of ROligos
In a first embodiment, a restriction enzyme cuts at the introduced recognition site and thus digests the ROligo into a dual 216/166 bp banding pattern as visualized on an agarose gel stained with ethidium bromide. The dual banding pattern indicates a restriction endonuclease that recognizes the complete introduced site, or part of it.
In a second embodiment, a cellular extract contains an active restriction enzyme that cleaves outside the introduced recognition site of the ROligo. If the cleavage point lies in close vicinity to the introduced restriction site, the resulting band pattern will resemble a double band. However, as the sequences outside the introduced site are identical among all ROligos, cleavage of one ROligo outside of its introduced site would mean that all ROligos would likely be digested and therefore a double band would show up on an agarose gel analysis. This universal digestion of all ROligos is unlikely at least because no more than three restriction endonucleases have been found in any cyanobacterial cellular extract. If a cyanobacterial extract contains an active restriction enzyme that cleaves outside the introduced recognition site of the ROligo, this pattern would indicate that there is at least one restriction enzyme present, but because the enzyme does not specifically cut at the introduced recognition sequence, it cannot be identified by the ROligo method. Deducing its recognition sequence would require other more cumbersome and labor intensive methods such as sequencing the ends of gel-isolated ROligo digestion bands.
In a third embodiment of possible cleavages of ROligos, a restriction endonuclease is present that recognizes an overlap between the introduced restriction site and the bases of the flanking sequences. The introduced recognition site of each ROligo is designed such that no known cyanobacterial enzyme would recognize any overlap created by the insertion of the recognition sequence into the ROligo. If the presence of an additional recognition sequence at the junction site between the flanking regions and the introduced recognition sequence is intrinsically unavoidable, a second, discriminatory, ROligo can be constructed which allows exclusion of one of the possible enzymes. For example, restriction enzyme R.NspI recognizes the overlap between the R.AvaI site and the adjacent flanking arm in the ROligo.AvaIII. Thus, if a ROligo was generated with an R.AvaIII recognition sequence and a double band pattern showed up on the gel, it would not be possible to determine whether the active restriction endonuclease is R.NspI or R.AvaIII by using the ROligo method. However when a second ROligo, ROligo.NspI, is generated where only R.NspI cuts, but not R.AvaIII, the identity of the active restriction enzyme or enzymes can be determined, see, for example, the digestion matrix generated in FIG. 6 .
In a fourth and fifth embodiment of possible cleavages of ROligos, a non-double band cleavage pattern or a non-cleavage pattern is generated, respectively. Restriction enzymes that cut an ROligo but do not produce a double band will not be identified by using embodiments of the ROligo method that rely upon the presence of a double band pattern (for example, 216 bp and 166 bp) in the analysis of cellular extract incubated polynucleotides on an agarose gel.
Restriction enzymes that are present in cellular extracts but do not cut any ROligos in the ROligo library will also be missed. This lack of identification of existing, but unknown, restriction endonucleases can be overcome by generating additional ROligos for a ROligo library.
ROligos Containing Degenerate Recognition Sequences
Many cyanobacterial restriction endonucleases have recognition sequences which are degenerate. For example, R.AvaII recognizes GGWCC (where W is A or T) and R.AsuI recognizes GGNCC (where N is A. T, C or G). For the design of a ROligo do distinguish between R.AvaII and R.AsuI using the ROligo method, there are at least two options.
First, degenerate primers can be used to introduce the recognition site into the respective arms. Hence, the ROligo for R.AvaII would contain its two recognized sequences GGACC and GGTCC. Likewise, the ROligo for R.AsuI would contain four sequences, GGACC, GGCCC, GGGCC and GGTCC. The digestion pattern generated using these degenerated ROligos would not be able to distinguish an extract containing R.AvaII from one containing R.AsuI as either enzyme will cut both GGTCC and GGACC containing ROligos to the double band pattern.
In another embodiment, degenerate restriction endonucleases can be distinguished from one another through engineering ROligo recognition sequences of a given ROligo such that fewer of the possible recognition sequences are available to multiple restriction endonucleases. For example, to distinguish between degenerate restriction endonucleases R.AvaII and R.AsuI, GGACC could be used as the sole sequence for the R.AvaII ROligo and GGCCC could be used as the sole sequence for the R.AsuI ROligo. Because GGCCC is not recognized by R.AvaII, extracts with R.AvaII activity can only digest the ROligo.AvaII to the double band pattern but not the ROligo.AsuI, whereas an extract harboring R.AsuI can digest both and therefore the enzymes can be distinguished from one another.
Other examples of identifying degenerate restriction endonucleases include combinations such as R.PvuII having a recognition sequence of CAGCTG and R.NspBII having a recognition sequence of CMGCKG (where M is A or C and K is T or G). In an embodiment, R.PvuII which recognizes CAGCTG and R.NspBII can be distinguished from one another by using CAGCGG as the sole sequence for ROligo.NspBII instead of using all its four sequences recognized, CAGCGG, CAGCTG, CCGCGG, and CCGCTG.
Identification of Restriction Enzymes Using a Digestion Matrix
Infographics can be used to visualize data generated from the analysis of ROligo digestion products in a digestion matrix and to determine the identification of active restriction endonucleases in a cyanobacterial extract through the implementation of various steps of the ROligo method disclosed herein. In an embodiment, a digestion matrix is an infographic table depicting the theoretical capability of various restriction endonucleases as to whether they are capable of digesting the oligonucleotides to a double band pattern. In an embodiment, a digestion matrix can be manipulated through the application of a set of rules to determine the identity of active restriction endonuclease in a cyanobacterial extract. In an embodiment, a digestion matrix is created which depicts the absence or presence of a restriction enzyme in the celhlular extract of a cyanobacterial strain. In an embodiment, the digestion matrix is depicted in the form of an infographic table wherein the rows are labeled with ROligos containing a sequence recognition site and the columns are labeled with known cyanobacterial restriction enzymes. In an embodiment, a digestion matrix depicts, as a filled-in box, which cyanobacterial restriction enzyme(s) would produce a double band digestion pattern (indicative of a digestion of the ROligo at the recognition sequence) for which ROligos, see, for example FIG. 6 .
Restriction enzymes that do not share a recognition sequence, degenerate or overlapping, will create a diagonal pattern in a digestion matrix. In a non-limiting sense, ROligos that are digested by only one restriction enzyme are referred to herein as “Optimal ROligos”.
If all ROligos were Optimal ROligos, identifying the restriction enzyme constituency of a cellular extract would be a matter of having a comprehensive library of Optimal ROligos encoding recognition sequences for all known restriction enzymes. However, at least because many restriction enzymes exhibit overlap in recognition sequences, R.AvaII/R.AsuI, and R.HaeIII/R.ApaI, for example, additional embodiments of the ROligo method are necessary to identify restriction endonucleases present within various cellular extracts.
Washing Cyanobacterial Cells to Remove Extracellular Nucleases
FIG. 7 depicts the results of digestion of plasmids in cyanobacterial extracts that have or have not been washed with either a Triton detergent wash or a pH/urea wash in an attempt to remove extracellular nucleases before disruption of the cyanobacterial cells and testing of their extracts for restriction endonucleases. Plasmids used for digestion include K236Cm (SEQ ID NO: 2) and K244 (SEQ ID NO: 3). Plasmid maps for K236Cm and K244 are depicted in FIG. 8 and FIG. 9 , respectively.
Panel A of FIG. 7 : Lane 1 is a plasmid mixture of K230, K236Cm and K244 incubated in cell-free medium. Lanes 2-4, are plasmid mixtures incubated with supernatants from respective washing steps with triton solutions. Lane 5 is an unincubated plasmid mixture. Lane M is a size marker. Lane 6 is plasmid K236Cm incubated in a third Cyanobacterium strain extract after triton washing at 28° C.; lane 7, plasmid K236Cm incubated in a third Cyanobacterium strain extract after triton washing at 37° C.
Cells of a third Cyanobacterium strain were washed three times with 0.1% triton in DNase-buffer. A description of the triton washing technique used can be found in Soper, B. W., et. al., J. Bacteriology 1994, 176(17):556-5570. Six hundred nanograms of a plasmid 1:1 mixture containing K230, K236Cm and K244 were incubated over night at 37 OC with the cell's culture media and supernatants. DNA incubated in the cell-free medium (lane 1) showed strong degradation as seen by the smearing towards lower molecular weights and an overall reduction in DNA staining as compared to lanes 3 and 4. DNA incubated in wash supernatant 1 (lane 2) showed weak smearing to lower molecular weights but still a decrease in DNA staining as compared to lanes 3 and 4. DNA incubated in later wash supernatants show only one non-degraded band of similar intensity. The size difference between the unincubated plasmid mixture of lane 5 and the other lanes is likely due to a specific cleavage by an endonucleolytic enzyme which is not removed by triton treatment or a spontaneous reorganization of plasmid DNA during incubation from nicked plasmids to a supercoiled form.
Panel B of FIG. 7 : A 1 kb PCR product was used as a template for incubations of DNA with extract from a Geitlerinema strain. In lane 1, the template was incubated in crude extract of triton treated cells of a Geitlerinema strain incubated overnight at 30° C. In lane 2, the template was incubated overnight at 30° C. in crude extracts of pH/urea washed cells of a Geitlerinema strain. In lane 3, the template was incubated overnight at 37° C. in crude extracts of pH/urea washed cells of a Geitlerinema strain. Lane 4 depicts a crude extract of pH/urea treated cells from lanes 2 and 3. After triton washing of a Geitlerinema strain cells, crude extracts showed unspecific nucleolytic activity (see lane 1) as seen by the blurry band shape after incubating a 1 kb PCR template in a Geitlerinema strain extract over night at 30° C. A fresh culture of a Geitlerinema strain was then washed with a pH change/urea regimen. As depicted in lane 2, this crude extract did not show smearing of the digestion bands as seen by the sharp bands as visualized. Moreover, incubation at 37° C. did not result in unspecific degradation.
Thus, washing cyanobacterial cells with either a triton or a pH/urea washing regimen was useful for decreasing the amount of non-specific nuclease activity in the cellular extract to be tested for the activity of restriction endonucleases using the ROligo method.
Digestion Matrix of a ROligo Library
FIG. 6 depicts a predicted digestion matrix of a library of twenty ROligos. Columns are labeled after cyanobacterial enzymes. Rows are labeled after the ROligo. The introduced recognition sequence is reflected by the name of the ROligo. A black box indicates that the enzyme of that column would digest the ROligo of that row to produce a double band of 216/166 bp. Some ROligos not only get cleaved by their specific enzyme, creating the diagonal black box pattern in the matrix, but are also cut by a second or third enzyme recognizing the same sequence, a part thereof or an overlap to the adjacent arm sequence. The digestion of a given ROligo by multiple restriction enzymes can be reduced through careful engineering of the sequence of the recognition site by the addition of surrounding bases. For example, the ROligo.HaeII used to create the digestion matrix of FIG. 6 was engineered without a T base preceding the recognition sequence in an effort to reduce the number of restriction endonucleases that identify the particular ROligo.HaeIII recognition sequence used.
FIG. 6 depicts the theoretical cleavage capability of a set of known cyanobacterial restriction endonucleases to produce the double band pattern on generated ROligos to identify their occurrence in future cyanobacterial extracts. As depicted in FIG. 6 , columns are labeled after cyanobacterial restriction enzymes. Rows are named after ROligos. The introduced recognition sequence of the respective restriction endonuclease is reflected by the name of the ROligo. For example, AcyI and ROligo.AcyI. In FIG. 6 , a black box indicates that the enzyme of that column can theoretically digest the ROligo of that row to produce a double band of 216 bp and 166 bp as visualized on an agarose gel. Some ROligos not only get cleaved by their respective restriction endonuclease (diagonal black box pattern) but are also cleaved by a second or third enzyme recognizing the same sequence, a part thereof or an overlap to the adjacent arm sequence and thus generating a 216 bp and 166 bp pattern.
After incubation with a cellular extract and analysis by agarose gel, the pattern of digested and undigested ROligos is compared to a predicted digestion matrix to find one enzyme, or a set of enzymes, whose predicted digestion pattern results in exactly the pattern observed. Some incubations result in digestion patterns that predict only one set of enzymes that explain the observed pattern (as the ROligos were designed to be cut by as few other enzymes as possible). However, some combinations of enzymes cannot be distinguished through predicted digestion patterns. For example, the concomitant occurrence of R.HaeIII (recognition sequence of GGCC) and R.AvaII (recognition sequence of GGWCC) would produce the same ROligo digestion pattern as the combination of R.HaeIII and R.AsuI (recognition sequence of GGNCC). This is because R.HaeIII cuts GGCCC and GGGCC, which are the recognition sequences of R.AsuI that are not cleaved by R.AvaII. Thus, using an embodiment of the ROligo method, the concomitant occurrence of R.HaeIII and R.AvaII would be indistinguishable from the concomitant occurrence of R.HaeIII and R.AsuI. Other embodiments of the ROligo method and rules of interpretation of digestion patterns are useful to distinguish concomitant occurrences of restriction endonucleases such as these.
Exemplary Rules Used in Embodiments of the ROligo Method
In an embodiment of the ROligo method, a sequence of rules can be applied to identify a set of restriction endonucleases that explain an observed ROligo digestion pattern as depicted in a digestion matrix.
Rule 1: First, all restriction endonucleases which theoretically do not recognize and cleave at least one of the ROligos that are observed to be cleaved to the double band pattern by a cellular extract are discarded (removed from being included in the digestion matrix). The digestion matrix then only contains enzymes that can theoretically cut at least one of the observably digested ROligos.
Rule 2: Secondly, restriction endonucleases which theoretically recognize and cleave a ROligo that does not show a double band pattern after extract incubation, can be discarded from the digestion matrix.
Rule 3: The occurrence of the remaining restriction endonucleases then explains the observed pattern of digested and non-digested ROligos. Generally, all restriction endonucleases remaining in the digestion matrix after the application of rules 1 and 2 are needed to actually explain the pattern.
However, in some instances, a pattern is present that may be explained by either one enzyme or a combination of restriction endonucleases being responsible for the digestion of multiple ROligos. The digestion pattern presents itself as ambiguous as to whether it is the combination of restriction endonucleases or the single restriction endonuclease is responsible for the observed ROligo digestions. As an example, the ROligo digestion pattern of both R.NspBII (CMGCKG recognition sequence wherein M is A or C, and K is G or T) and R.PvuII having a CAGCTG recognition sequence, is identical to the one produced by R.NspBII alone. The same reasoning applies to the combination of R.AsuI having a GGNCC recognition sequence plus R.AvaII having a GGWCC recognition sequence, and R.AsuI alone. In these ambiguous patterns, the existence of the single restriction enzyme that is responsible for digesting the two ROligos is chosen to exist versus the presence of two restriction endonucleases that perform the identical function. Not being limited by theory, a single restriction endonuclease is more likely to occur in the extract because a cell evolves to conserve metabolic resources and thus would not develop an increase in protection against foreign DNA by expressing two restriction endonucleases where the expression of a single restriction endonuclease would provide the same number and type of recognition sites as the two restriction endonucleases combined. Thus, the restriction endonuclease exhibiting the broader cleavage range is considered more likely to exist alone in the cellular extract and is kept as a restriction endonuclease that exists while the combination of restriction endonucleases are discarded if they cannot be individually eliminated.
Identifying Restriction Endonucleases in a Cyanobacterial Extract Using the ROligo Method
Using a ROligo library of the twenty ROligos depicted in FIG. 6 , active restriction endonucleases from a first Cyanobacterium strain extract were determined. The incubation of the ROligos from the ROligo library with the cellular extract was performed in a two-step process. First, five pools of four ROligos were incubated with a cellular extract and a double band digestion pattern was tested for. Second, the individual ROligos comprising the ROligo pool showing a double band digestion pattern were each individually incubated with the cellular extract and analyzed for the double band digestion pattern.
FIG. 10 depicts the results of an ethidium bromide stained agarose gel analysis of several pools of ROligos incubated in a first Cyanobacterium strain extract. Lanes 1-5, together, comprise the totality of a ROligo library of 20 ROligos. Each lane has four ROligos (a ROligo pool) from the ROligo library. Each ROligo pool is incubated in either a NEB buffer P1 (left part of the gel) and buffer P2 (right part). The most commonly used NEB buffers P1, P2, P3, and P4 are designed such that most known restriction endonucleases will be active in at least one of the four buffers. Lane X is a first Cyanobacterium strain extract without addition of ROligos. Lane M is a size marker.
Some pools of ROligos show a digestion double band at around 216/166 bp (marked by arrows at the top). Pool 1 showed a double band only in buffer P2 but not in buffer P1, pool 3 showed a double band in both P1 and P2 buffers. Not depicted are the incubations performed in NEB buffers P3 and P4. Because pools 2, 4, and 5 showed no double band digestion pattern, their constituent ROligos were eliminated as possibly existing in this first Cyanobacterium strain.
FIG. 11 depicts the results of incubation of individual ROligos in a first Cyanobacterium strain extract. Individual ROligos comprising ROligo pools that resulted in a double band pattern after the first-step incubation, see FIG. 10 , were each individually incubated in a first Cyanobacterium strain extract in NEB buffers P1, P2 or P3. Lanes 1-8 depict the results of ROligos incubated in a first Cyanobacterium strain extract as follows: Lane 1 is ROligo.AsuI having a recognition site of GGNCC constructed as GGCCC. Lane 2 is ROligo.AvaII. Lane 3 is ROligo.HaeIII constructed as GGCC. Lane 4 is ROligo.SduI. Lane 5 is ROligo.AvaII. Lane 6 is ROligo.NspI. Lane 7 is ROligo.SphI constructed as GCATGC. Lane 8 is ROligo.BamHI. Lane M is a marker. ROligos showing a double band were ROligo.AsuI, ROligo.HaeIII and ROligo.SphI and are marked with arrows.
Combining four ROligos in one incubation reaction as a first step in the ROligo method reduces work load significantly as compared to incubating each member of the ROligo library individually. As exemplified above, to identify which of the twenty analyzed cyanobacterial restriction enzymes are present in a first Cyanobacterium , strain, the twenty ROligos in the digestion matrix (see FIG. 6 ) were sorted in a way that ROligos that were digested by the extract, those showing a double banding pattern, were moved to the top of the matrix (see FIG. 12 , top three rows). Rule 1 was then applied to the digestion matrix, and all enzymes which theoretically cannot cut at least one of these three ROligos were removed, leaving four possible enzymes, R.AsuI, R.HaeIII, R.NspI and R.SphI, see FIG. 13 . Applying rule 2 (restriction enzymes which theoretically would also digest at least one other ROligo to the double band pattern than the observed extract-digested double band pattern showing ROligos can be discarded) allowed to discard enzymes R.AsuI (as ROligo.AvaII was undigested) and R.NspI (as neither ROligo.NspI nor ROligo.AvaIII was digested by the extract, see FIG. 14 .
This left two enzymes, R.HaeIII and R.SphI. R.HaeIII theoretically cuts ROligo.AsuI and ROligo.HaeIII. R.SphI only cuts its own ROligo.SphI. The occurrence of both these enzymes in the digestion matrix produces the same pattern (see FIG. 14 ) as was observed in a first Cyanobacterium strain extract, see FIG. 15 . Thus, through using an embodiment of the ROligo method, the active restriction endonucleases in a first Cyanobacterium strain extract were determined and subsequent vectors and constructs used for transformation of a first Cyanobacterium strain extract can be engineered to lack those recognition sequences or can be methylated at those recognition sequences to prevent digestion of the vector by the endogenous restriction endonucleases of a first Cyanobacterium strain.
Analyses of Endonuclease Activity in Cyanobacterial Strains Using the ROligo Method
FIG. 15 depicts various cyanobacterial strains that were analyzed with an embodiment of the ROligo method to identify restriction nuclease activity therein. If a genomic sequence of the cyanobacterial strain was available, RM (“R”estriction “M”odification) systems were predicted. RM systems were identified when a gene encoding a restriction enzyme and a gene encoding its cognate methyltransferase are present in close genomic proximity. Examples of some of the tested methylation of ROligos that blocks the respective cyanobacterial restriction endonucleases from cleavage are given.
As depicted in FIG. 15 , the asterisk (*) denotes that the recognition sequence of a given restriction endonuclease contains two GC stretches. In most cases, both GC stretches were methylated by the methyltransferase M.CviPI. In a prophetic embodiment, restriction enzymes in the respective strain extracts are blocked by methylation of one or both of the GC stretches.
As depicted in FIG. 15 , the pound sign (#) denotes enzymes that were previously not known to exist in cyanobacteria and were identified by using the sequencing method. The identified restriction endonuclease's activities were verified by using the ROligos method to identify restriction endonucleases.
As depicted in FIG. 15 , the double pound sign (##) denotes an enzyme that was initially detected by cloning and sequencing digestion bands.
In an embodiment, and as depicted in FIG. 15 , the restriction endonucleases in extracts of fifteen cyanobacterial strains were analyzed by the ROligo method. Genomic analysis of strains, whose genome had been sequenced, predicted restriction enzymes in a second Lyngbya strain, a Chlorogloeopsis strain, a second Cyanobacterium strain, and a second Nostoc strain, see FIG. 15 . With the exception of a second Nostoc strain, all enzymes whose activity could be demonstrated in crude extracts using the ROligo method, see FIG. 15 , were also predicted by in silico methods. However, not all predicted enzymes were found active in extracts, see a second Nostoc strain in FIG. 15 , for example.
A hitherto unknown cyanobacterial restriction enzyme was detected in a third Cyanobacterium strain. Incubation with a third Cyanobacterium strain extract showed digestion of plasmids, but none of the ROligos in the library were digested to a double band. Direct sequencing of the ends of the digested plasmid bands revealed a region of about 30 bp where the sequences all showed a signal drop, indicating the possible presence of extracellular nucleases slowly and possibly nonspecifically degrading the ends of the digested plasmids. In an effort to remove any extracellular nucleases, the cells were washed with a triton washing protocol as described herein. Analyzing possible restriction enzymes that would have a recognition sequence within that region and further limiting to palindromic enzymes (almost all cyanobacterial restriction endonucleases have recognition sequences that are palindromatic), the recognition sequence of five candidate restriction enzymes were identified and ROligos were constructed accordingly. Of the five candidate restriction enzymes, only the ROligo.EcoRV, having a recognition sequence of R.EcoRV (GATATC), was found to be digested by the extract, see FIG. 15 , and R.EcoRV was the only enzyme that could digest the ROligo to the double band as judged by the appropriate digestion matrix. Because R.EcoRV was not previously known to be a cyanobacterial restriction endonuclease, if a plasmid devoid of EcoRV sites had been incubated with the cellular extract, this restriction enzyme would not have been detected. A Geitlerinema strain was analyzed as well. Again, none of the ROligos in the library at the time of analysis was digested to a double band. However incubating plasmid K230 and several PCR products derived from it, clearly showed digestion. After sequencing the ends of a digested PCR product, a SauI recognition sequence of CCTNAGG was detected and the corresponding ROligo was constructed and was digested by a Geitlerinema strain extract.
In a prophetic embodiment, and according to an embodiment of the ROligo method, the generated ROligo.EcoRV, and ROligo.SauI corresponding to the restriction endonucleases identified could then be methylated by appropriate means. The methylated ROligos could then be incubated with the appropriate cyanobacterial extracts and could be analyzed to determine to what degree digestion was inhibited by their respective restriction endonucleases.
FIGS. 12-14 depict and embodiment of the ROligo method used to determine the restriction enzymes that are active in an extract of a first Cyanobacterium strain.
In an embodiment of the ROligo method, FIG. 12 depicts a digestion matrix that is a useful for the identification of a first Cyanobacterium strain's restriction enzymes. A digestion matrix of a library of twenty ROligos was reorganized to reflect the observed pattern of digested and undigested (i.e., double band digestion of a ROligo detected or not) ROligos in a first Cyanobacterium strain extract. The top most three ROligos (rows) marked yellow, are digested by a first Cyanobacterium strain extract, the unmarked ROligos are not.
FIG. 13 depicts a digestion matrix of a first Cyanobacterium strain extract incubated ROligos after applying an embodiment of rule 1 of the ROligo method (removal of any enzyme which does not cut at least one of the three topmost ROligos). Rule 2 (restriction enzymes which theoretically would recognize other ROligos than the ones having been observed to be extract-digested to the double band pattern, can be discarded) was applied to discriminate the R.AsuI/ROligo.AvaI I digestions (black horizontal arrow) and the R.NspI/ROligo.AvaIII digestions.
FIG. 14 depicts a digestion matrix resulting from the incubation of a ROligo library in a first Cyanobacterium strain extract after applying rules 1 and 2 and then applying rule 3 (the occurrence of the remaining enzymes then explains the observed pattern of digested and non-digested ROligos). The combination of the activity of the two remaining enzymes, R.HaeIII and R.SphI, cut exactly those ROligos which are digested by the extract.
Therefore, in an embodiment of the ROligo method, FIGS. 12-14 depict steps of the ROligo method used to identify the active restriction endonucleases of a cyanobacterial strain. Thus, vectors for transformation of the cyanobacterial strain can be designed that lack the recognition sequences of R.HaeIII and R.SphI and therefore increase the likelihood of transformation of the cyanobacterial strain for which the ROligo method has been performed upon.
Generation of New Digestion Matrices
In an embodiment, after generating new ROligos, the starting digestion matrix will be constructed anew, and would now include any newly identified ROligos and incorporate the predicted digestion pattern of the newly identified ROligos into a new digestion matrix. The restriction enzymes included in the matrix (showing the double band pattern) can be derived from various sets of enzymes, e.g. all cyanobacterial enzymes, all palindromic enzymes, or all restriction enzymes present, for example, in databases such as REBASE.
DNA Methylation
DNA methylation is the process of transferring a methyl group from a donor molecule to either a cytosine or an adenine by DNA methyltransferases. Three types of methylated bases are predominantly found in DNA: 5-methylcytosine (m5C), N4-methylcytosine (m4C), and N6-methyladenine (m6A). The organism-specific pattern of methylation depends on the methyltransferases' specificity.
In prokaryotes, DNA cleavage by a cognate restriction enzyme is prevented by the methylation of DNA by a sequence-specific methyltransferase which is an integral component of every restriction-modification system. E. coli strains used for propagation of plasmid DNA often contain a site-specific DNA adenine methyltransferase, Dam. The methylase encoded by the dam gene methylates the N6-position of an adenine residue within the GATC sequence.
Most restriction enzymes are sensitive to DNA methylation. When a restriction enzyme recognition sequence overlaps a methylation site, no effect, a partial inhibition or complete block of digestion may occur. Furthermore, the ability to cleave methylated DNA is an intrinsic and unpredictable property of each restriction enzyme. Therefore, isoschizomers and neoschizomers which recognize the same DNA sequences can differ in their sensitivity to DNA methylation. To prevent DNA digestion of heterologous vectors, both the type of DNA methylation and the sensitivity of the restriction enzyme to that type of methylation should be considered.
Methylation Sensitivity of Identified Restriction Endonucleases
In an embodiment, the ROligo method is used to improve the efficiency of determining the methylation sensitivity of identified restriction endonucleases. Methylation sensitivity usually has to be checked for each restriction enzyme found and cannot be inferred from a recognition sequence alone. As an example, the non-cyanobacterial restriction enzymes R.DpnI, R.MboI and R.Sau3AI all recognize the four base pair palindrome sequence of GATC. However, they show different sensitivity to methylation of GATC at the adenine or cytosine position (see REBASE). R.DpnI only cuts fully adenine methylated Gm6ATC. MboI is blocked by hemi-methylation at Gm6ATC but is not affected by C5-cytosine methylation. Sau3AI cuts both N6-adenine methylated and unmethylated DNA, but is blocked by C5-methylation GATm5C.
Different types of in vitro methylation can be used to methylate a ROligo. The effect of methylation can subsequently be checked by incubation of the methylated ROligo in an extract that cuts the non-methylated ROligo. If a suitable methylation type that protects from cleavage has been identified, methyltransferase databases like REBASE can be searched for a known methyltransferase that catalyzes that type of methylation. If these methyltransferases can subsequently be expressed in yeast or E. coli , the sequencing of the cyanobacterial genome to find endogenous methyltransferases is not necessary.
Methylation of ROligos
Identification of restriction endonucleases that are present in a given strain of cyanobacteria allows for site specific methylation of heterologous DNA prior to transformation of that given strain of cyanobacteria. In an embodiment, methylation of ROligos can be achieved by heterologous overexpression of methyltransferases. For example, plasmids that overexpress methyltransferases, such as M.AvaI and M.AvaII methyltransferases, may be constructed and used in an E. coli donor strain and then transferred into a cyanobacterial strain via conjugation (see Elhai J., et. al., J. Bacteriology 1997, 179(6):1998-2005) or other transformation means.
In another embodiment, commercially available methyltransferases can be used to methylate ROligos. Methylation success is checked by restriction digestion with a corresponding commercially available restriction enzyme that is blocked by the type of methylation used (which can be found for example in charts depicting methylation sites for known restriction endonucleases, e.g., REBASE). Any methylations that result in protection of the ROligos from digestion can then be applied to methylate plasmids or other exogenous DNA prior to transformation of cyanobacterial strains having the identified restriction endonuclease activity. For example, M.CviPI methylates cytosine at the C5 position in its GC recognition sequence. M.SssI methylates cytosine at the C5 position in its CG recognition sequence. M.CviPI or M.SssI, for example, can therefore be used to methylate DNA at cytosines. Many of the cyanobacterial enzymes contain a CG or GC stretch and can, therefore, be checked for their sensitivity to m5CG and/or Gm5C methylation, see FIG. 2 . As an example, protection from digestion of ethanologenic plasmids and ROligos methylated by M.CviPI and M.SssI from restriction endonucleases is depicted in FIG. 16 .
FIG. 16 depicts analysis of a possible protective effect of Gm5C-methylation against a first Lyngbya strain extract and a Chlorogloeopsis strain extract. The left panel of FIG. 16 depicts lane M which is a reference marker; lane 1 which is an unmethylated ROligo.PvuII incubated in a first Lyngbya strain extract; lane 2 which is Gm5C-methylated ROligo.PvuII incubated in a first Lyngbya strain extract; and lane 3 which is methylated ROligo.PvuII that has not been digested. Thus, as depicted in the left panel of FIG. 16 , ROligo.PvuII, having a recognition sequence of CAGCTG was methylated using commercially available M.CviPI which methylates the cytosine at the C5 position in its recognition sequence GC resulting in CAGm5CTG. Methylated and unmethylated ROligo.PvuII were incubated in a first Lyngbya strain extract and only the umnethylated ROligo.PvuII was digested by the extract. Thus, R.PvuII activity in a first Lyngbya strain is blocked by M.CviPI methylation. Therefore, the left panel of FIG. 8 depicts the protective effect of methylation of ROligo.PvuII against digestion from a restriction endonuclease in the extract of a first Lyngbya strain.
The right panel of FIG. 16 depicts analysis of a possible protective effect of Gm5C-methylation against a Chlorogloeopsis strain extract. TK018 (SEQ ID NO: 4), an exemplary plasmid containing an ethanologenic cassette, was methylated with M.CviPI and then column purified. A plasmid map of TK018 is depicted in FIG. 17 . Methylated and unmethylated TK018 were then incubated in a Chlorogloeopsis strain extract. The right panel of FIG. 16 depicts lane M which is a size marker; lane 4 which is a Chlorogloeopsis strain extract without DNA; lane 5 which is TK18 incubated in a Chlorogloeopsis strain extract; lane 6 which is TK18 methylated with M.CviPI and then incubated in a Chlorogloeopsis strain extract; lane 7 which is unincubated TK018; and lane 8 which is methylated TK18 that has not been incubated. As depicted in the right panel of FIG. 16 , the methylated plasmid in lane 6 does not show the digestion bands of the unmethylated plasmid of lane 5. Therefore, the right panel of FIG. 16 depicts the protective effect of methylation of an exemplary plasmid, TK018, against digestion from a restriction endonuclease in the extract of a Chlorogloeopsis strain.
In yet another embodiment, the extract from E. coli strains containing methyltransferases expressed from helper plasmids can be used to methylate ROligos. FIG. 18 depicts the protective effect of M.AvaII methylation of ROligo.AvaII against digestion from restriction endonucleases within the extracts of a first Lyngbya strain, a first Nostoc strain, and a second Nostoc strain. As depicted in FIG. 18 , lane 1 is an unmethylated ROligo.AvaII digested with commercially available R.AvaII; lane 2 is a methylated ROligo.AvaII subjected to digestion with R.AvaII; lane 3 is a methylated ROligo.AvaII that is undigested; lanes 4, 6, and 8 are unmethylated ROligo.AvaII incubated in extract of a first Lyngbya strain, a first Nostoc strain, and a second Nostoc strain, respectively; lanes 5, 7, and 9 are methylated ROligo.AvaII incubated in extract of a first Lyngbya strain, a first Nostoc strain, and a second Nostoc strain, respectively; and lane M is a reference marker.
ROligo.AvaII as used in FIG. 18 incubations was, as applicable, methylated using an extract from E. coli HB101 containing a M.AvaII expressing plasmid pRL528. Success of methylation of ROligo.AvaI was verified by digesting the methylated ROligo.AvaII with commercially available and methylation sensitive R.AvaII, see lanes 1 and 2 of FIG. 18 . The methylated and unmethylated ROligo.AvaII were incubated in extracts of a first Lyngbya strain, a first Nostoc strain and a second Nostoc strain which all demonstrated R.AvaII activity, see lanes 4, 6 and 8 of FIG. 18 . R.AvaII activity in all strains depicted in FIG. 18 is blocked by M.AvaII methylation; compare lanes 4 to 5, 6 to 7, and 8 to 9 of FIG. 18 .
In another embodiment, endogenous E. coli Dam methyltransferase from a DNA free extract of E. coli can be used to methylate ROligos and plasmids. Effects of methylation by an endogenous E. coli Dam methyltransferase can be analyzed by using an E. coli extract to methylate ROligos. ROligos complimentary to restriction enzymes that contain a GATC site, such as R.BglII with a recognition site of AGATCT or R.BamHI with a recognition site of GGATCC, can be methylated and tested for protection against digestion.
In another embodiment, methylated nucleotides could be incorporated into ROligos by PCR. For recognition sites which cannot be methylated by in vitro methyltransferases, PCR could be used to incorporate methylated nucleotides. This method could be especially useful when there is no commercial enzyme available which methylates any given position in a restriction enzyme recognition sequence.
Additional Means of Identification of Unknown Restriction Endonucleases
Restriction endonucleases that do not recognize a ROligo from an existing ROligo library will not be identified. Thus, increasing the number of ROligos in a ROligo library will increase the number of potentially identifiable restriction enzymes.
In an embodiment, a method of identifying unidentified, cyanobacterial, restriction endonucleases is to generate a sequence that contains recognition sequences for restriction enzymes which are currently unknown in cyanobacteria. Thus, in a prophetic example, a sequence could be constructed in silico that contains recognition sequences of all known restriction enzymes in a database, e.g., REBASE, except for cyanobacterial recognition sequences or recognition sequences that are cleaved by cyanobacterial enzymes. For example, the cyanobacterial R.HaeIII has a recognition sequence of GGCC and also cuts the recognition sequence of the non-cyanobacterial restriction endonuclease R.NotI having a recognition sequence of GCGGCCGC. Therefore, the recognition sequence of R.NotI would be omitted.
After construction of the polynucleotide sequence containing all of the appropriate recognition sequences, the sequence would be incubated with cyanobacterial extracts. If the sequence was digested, the resulting band(s) would be isolated from a gel and directly sequenced with primers spanning the artificial sequence. This sequencing information would provide the recognition sequence and also the identity of the restriction enzyme active in the cyanobacterial extract. Thus, this prophetic embodiment would allow for the detection and identification of new enzymes currently not described in cyanobacteria.
In a prophetic embodiment, restriction endonucleases that cleave outside of the recognition sequence of a given ROligo could be detected through construction of a “megaROligo”. Detecting a double band cleavage product of a ROligo digested at the recognition site can be obfuscated when two restriction enzymes occur in one cyanobacterial extract, each digesting the ROligo. The restriction activity of both of the restriction endonucleases acting upon the ROligo at, for example, the recognition sequence and at a site in one of the flanking regions would result in a digestion pattern different from the predicted double band digestion product indicative of a single digestion at the recognition sequence by the correlating restriction endonuclease.
To account for restriction endonucleases that cut within a flanking arm of a given ROligo, a megaROligo could be constructed. The megaROligo would contain a recognition sequence that contains all recognition sequences that occur within the flanking arms that are used in the construction of the ROligos in a ROligo library. Consistent with the design principles of ROligos, this multiple recognition sequence would be flanked by an artificial sequence that is devoid of any recognition sequences from any known restriction enzymes.
In a prophetic embodiment, pre-incubation of this megaROligo in cyanobacterial extracts would allow testing for the presence of restriction endonucleases that recognize sequences in the flanking arms of all ROligos in the library. Thus, by incubating the megaROligo in the extract of cyanobacterial cells to be analyzed by the ROligo method, multiple restriction enzymes that digest a given ROligo and obscure the double band pattern expected by only digestion at the recognition sequence could be detected. This detection of multiple restriction enzymes digesting a ROligo in parallel would provide an additional quantum of information useful for determining the full complement of restriction endonucleases in a cyanobacterial extract.
In a prophetic example, to avoid digestion of ROligos in flanking regions ROligos could be engineered to be devoid of any recognition sequences in the flanking regions for any known restriction endonucleases.
All currently identified cyanobacterial enzymes except R.MboII are palindromic. Therefore, in a prophetic example, a “polypalindromic” sequence could be constructed that contains all possible palindromic sequences from 4-8 bp in length that are not recognized by anly known restriction enzymes. In a prophetic embodiment, incubating this polypalindromic sequence in a cyanobacterial extract could identify novel restriction enzymes from cyanobacteria.
Labeled ROligos
In a prophetic embodiment, ROligos could be labeled with fluorescent dyes, radiolabels, antigens or other labels. The labels could be incorporated through post-labeling techniques or could be incorporated using pre-labeled nucleotides through polynucleotide synthetic techniques well known in the art, see for example oligonucleotide synthetic services available through Iba life sciences in Olivette, Mo. at http://www.iba-lifesciences.com/Service_Custom_oligos.html and Shibata A., Molecules 2012 Feb. 29; 17(3):2446-63 which discloses various techniques and references to techniques well known in the art of oligonucleotide labeling.
In an embodiment, the labels used on each individual ROligo in a given ROligo pool would be uniquely identifiable. As an example, if a ROligo pool contained four different ROligos, each could be labeled with a dye that fluoresces at a wavelength that is different from the other three labels. Thus, even if multiple ROligos were digested and they all created digestion bands that were indistinguishable from one another on an ethidium bromide stained agarose gel, the identity of the double bands present could be visualized by measuring fluorescence at the four different wavelengths at which the four dyes fluoresce. For example, four different ROligos, ROligo.MstI, ROligo.SduI, ROligo.BlpI, and ROligo.BsteII could each be respectively labeled at a flanking region with a blue, green, yellow and red fluorescing dye. If a dual band pattern was visualized, one could not a priori determine which of the four ROligos had been digested. However, if the bands were then measured for fluorescence and a blue and red fluorescence were detected, the restriction enzymes of R.MstI and R.BsteII would have been identified. This multi-dimensional measurement would eliminate the additional step of runnming each of the ROligos individually from a ROligo pool that demonstrated a dual band pattern via ethidium bromide visualization on an agarose gel.
In another prophetic embodiment, as many different labels as there are ROligos in a library could individually be incorporated into an individual ROligo. Thus, if there were twenty ROligos in a ROligo library, there could be twenty different labels that are visualized by twenty distinct wavelengths, antibody-antigen bindings or other visualization means. In this prophetic embodiment, all twenty ROligos could be incubated in a cellular extract. The entire extract would then be run in a single lane on an agarose gel and the digestion bands present would be measured according to the twenty different visualization means.
Universal Methyltransferases
In a prophetic embodiment, E. coli strains are used to express cyanobacterial methyltransferases, e.g., M.SphI, M.BlpI, and M.AsuI from a Chlorogloeopsis strain, M.AcyI from a second Cyanobacterium strain and M.PvuI from a second Lyngbya strain. Activity of the expressed methyltransferases will then be analyzed in crude E. coli extracts similar to those depicted in FIG. 16 . The expressed methyltransferases will be analyzed for methylation and transformation efficiency. The effects that the expressed methyltransferases have on the resistance of ROligos to digestion and to transformation efficiency of methylated vectors in a cyanobacterial strain will also be compared to 2 bp-methyltransferases such as M.CviPI and M.SssI.
M.CviPI and M.SssI lead to hypermethylation of DNA while cyanobacterial methyltransferases methylate DNA at specific recognition sites. Hypermethylation could result in difficulty of the initial replication of plasmids in cyanobacteria and/or for the expression of a resistance marker. However, hypermethylation would not be maintained during replication as newly synthesized strands of DNA would be subject to only methylation by the endogenous methyltransferases which may or may not be able to act upon hemi-hypermethylated double stranded DNA.
Transformation success in a Chlorogloeopsis strain was increased by M.CviPI hypermethylation of a vector used for transformation. The cloned cyanobacterial methyltransferases that were used to methylate the vector used for transformation of a Chlorogloeopsis strain could also be used to check for their effect on digestion of vectors in different strains with the corresponding restriction enzyme.
Digestion by most of the identified cyanobacterial restriction endonucleases can be protected against by a combination of M.CviPI, M.SssI and M.AvaII. Thus, in a prophetic embodiment, a “universal” helper plasmid would be constructed wherein M.CviPI, M.SssI and M.AvaII methyltransferases are expressed. This universal helper plasmid could be used to help increase the transformation efficiency of many different vectors into various cyanobacterial strains.
Examples
Triton Washing of Cyanobacterial Cells
Cyanobacterial cells were washed three times with DNase-buffer (40 mM TRIS pH 7.4, 6 mM magnesium chloride, 2 mM calcium chloride) supplemented with 0.1% (v/v) triton X-100. After a final wash step with lysis buffer, cells were disrupted with glass beads. Cell-free medium (supplemented with DNase buffer) and cell-free supernatants after each wash step were incubated with about 300 ng to about 900 ng plasmid overnight at about 28° C. to account for the activity of unspecific nucleases.
Washing of Cyanobacterial Cells Using Buffers with pH Changes and Urea Supplementation
Cyanobacterial cells were washed two to three times with alternating buffers of acidic and basic pH (acidic buffer: 20 mM potassium phosphate pH 3 supplemented with 1 M urea; basic buffer: 20 mM potassium phosphate pH 11 supplemented with 1 M urea). Urea was used to destabilize proteins and expose amino acids. By rapidly changing the pH, a permanent denaturation of proteins might result and render unspecific nucleases ineffective at digestion. Finally, cells were washed twice with lysis buffer and disrupted with glass beads. The cell-free supernatant was used as the crude extract.
Preparation of Cyanobacterial Extracts
From a liquid preculture, 50 mL of a given cyanobacterial strain was inoculated to an OD 750 of between 0.5 to 1.0. After 10 days, 30 mL of that culture was pelleted (5 min 3000×g at room temperature), washed once with lysis buffer (40 mM sodium hydrogenphosphate pH 7.4, 1 mM EDTA, 5% (v/v) glycerol) and resuspended in 1 mL lysis buffer with 1 U RNaseI. Cells were disrupted with glass beads in a tissue lyser. The cell debris was removed by two consecutive centrifugation steps for each 5 min at 14000×g at room temperature. The final supernatant was used as crude extract.
Construction of ROligos
Examples of PCR primers used to generate various left arms and right arms of various ROligos can be found in FIG. 3 . The left arm, approximately a 170 bp fragment, was amplified using K230 ( FIG. 1 ) as a template and DreamTaq polymerase (35 cycles of (1 min at 95° C., 1 min at 59° C., 1 min at 72° C.) with forward primer #474 (see FIG. 3 ) and a reverse primer containing the site to be introduced. Similarly, the right arm, approximately a 210 bp fragment, was amplified with a forward primer containing the site to be introduced and thus was overlapping with the reverse primer of the respective left arm and a reverse primer #475. Both arms were gel isolated and used in a 1:20 dilution in an overlapping PCR with booster primers #472 and #473 (3′-truncated versions of primers #474 and #475) that resulted in a approximately 380 bp fragment, which was gel isolated and quantitatively amplified using primers #474 and #475 again and a proofreading polymerase, Pwo, with 250 μM dNTP each, 250 nM each primer, 2 U Pwo/100 μL reaction, 1.5 mM MgSO 4 and 35 cycles of (94° C. 15 sec, 63° C. 30 sec, 45 sec at 68° C.). ROligos were then checked by digesting with commercially available restriction enzymes to verify correct insertion.
Incubation of Plasmids with Crude Extract
Plasmid DNA was incubated in cyanobacterial extracts over night at 28° C. with up to 30 μg protein (with the extract volume comprising up to about 25% (v/v) of the total reaction volume) in NEB buffer P4. To test for restriction endonuclease activity, a 40 μL reaction with 600 ng plasmid was loaded directly onto a 0.8% (w/v) agarose gel. For direct sequencing, 2 μg of plasmidial DNA were digested and then column purified, for example.
Determination of Restriction Endonucleases in a Cyanobacterial Crude Extract
A cyanobacterial extract was incubated with a mixture of four different ROligos, a “ROligo pool”. Each constituent ROligo was added to the pool at 150 ng overnight at 28° C. with 25% (v/v) extract, in NEB buffers P1, P2, P3 and P4 (compositions thereof are well known in the art and can be found at REBASE or ordered from New England Biolabs) and analyzed on a 2% (w/v) agarose gel. Pool/buffer combinations, where a double band at 216/166 bp was observed were further analyzed by incubating the individual ROligos separately in the respective buffer overnight. The resulting pattern of digested, i.e., showing a double band of 216/166 bp, and non-digested ROligos in a given extract was evaluated using a digestion matrix and restriction endonucleases were identified accordingly.
Methylation of ROligos
Identification of restriction endonucleases that are present in a given strain of cyanobacteria allows for site specific methylation of heterologous DNA prior to transformation. Methylation of ROligos was achieved through using the following protocol: 1 μg ROligo was methylated in a buffer containing 40 mM potassium phosphate buffer at pH 8.0 (or TRIS), 10 mM DTT, 25 mM EDTA supplemented with 640 μM S-adenosylmethionine (SAM) and a suitable methyltransferase. Methyltransferases were used at either 50 μL of an E. coli extract containing suitable methyltransferases or commercially available methyltransferases such as M.CviPI or M.SssI at about 0.5 to 2 U per μg DNA to be methylated. The final volume for the methylation reaction was 200 μL. The methylation reaction proceeded overnight at 37° C. Afterwards, the ROligo was column purified. Methylation success was checked by restriction digestion with an appropriate, commercially available restriction enzyme that is blocked by the type of methylation used (which can be found in charts depicting methylation sites for known restriction endonucleases, see REBASE). Methylation events that prevented the methylated ROligos from digestion by cyanobacterial extracts can then be applied to methylate heterologous plasmids prior to transformation into a given cyanobacterial strain.
Preparation of an E. coli Extract with Methyltransferase Activity
To methylate using the endogenous E. coli Dam methyltransferase, an extract of E. coli alphaSelect (available from Bioline USA Inc. of Taunton Mass.) was grown for 2 days at room temperature in 1% (w/v) soy bean peptone/0.5% (w/v) yeast extract was used (the final extract volume was 1/100 of the initial culture volume). Cells were disrupted in lysis buffer as above for cyanobacterial extracts. To methylate using the M.AvaII methyltransferase, E. coli HB101 harboring the pRL528 plasmid (Elhai, J, et. al, J. Bacteriology 1997, 179(6):1998-2005) grown for 2 days was used. E. coli extracts were DNaseI-treated (2 U per ml extract volume) for 30 min at 37° C. after supplementation with DNaseI buffer (NEB). The DNA-free extract was then used to methylate ROligos.
Sequencing of Extract-Digested Plasmids
A cyanobacterial-extract digested plasmid was sequenced after column purification with a variety of sequencing primers that span the whole plasmid in about 1 kb distances between adjacent primers. A cleavage point was identified when the sequencing signal drops sharply, see FIG. 19 , for example. A recognition site around a given cleavage point was then identified and thus the corresponding restriction endonuclease was identified.
Other Embodiments
A number of embodiments of the disclosure have been described. Nevertheless, it will be understood that various modifications may be made without departing from the spirit and scope of the disclosure. Such modifications will become apparent to those skilled in the art from a consideration of this disclosure or practice of the disclosure disclosed herein. Consequently, it is not intended that this disclosure be limited to the specific embodiments disclosed herein, but that it cover all modifications and alternatives coming within the true scope and spirit of the disclosure as embodied in the attached claims.
You are contracting for Methods for analyses of cyanobacterial restriction endonucleases
Expert Methods for analyses of cyanobacterial restriction endonucleases
You are commenting for Methods for analyses of cyanobacterial restriction endonucleases