Introduction

Here will we look briefly at some enzymes that can be used to synthesize DNA, polymerases, and to cut DNA, nucleases. They are important in genetic engineering either because we can use them in our cloning process, or because we need to know how to avoid them in our isolation of nucleic acids (as you have already seen in Chapter 3). We will finish by briefly describing DNA ligases; these can be used to join DNA molecules together.

Before you begin reading this chapter, review basic DNA structure if you don’t have this information accessible in your head. In particular you need to understand the components of DNA – base, sugar, phosphate – which carbons are the 1′, 2′ 3′ and 5′ carbons, the 5′ – 3′ polarity of a DNA strand and how this leads to the antiparallel structure of the double helix. It is important to have a sense of DNA structure while working with molecular techniques to help you understand what is going on in the reactions you are running.

A. DNA polymerases:

DNA polymerases synthesize DNA molecules by forming new phosphodiester bonds. They add new nucleotides to the 3’OH of a pre-existing DNA or RNA molecule. DNA polymerases cannot synthesize DNA de novo; that means they must always add to a pre-existing molecule. Therefore, they require a primer, a short stretch of nucleotides to which they can attach new nucleotides. When DNA replicates in a cell (in vivo), these primers are made of RNA (because RNA can be synthesized de novo) by an enzyme called primase. When we synthesize DNA in the lab (in vitro), we use primers that have been synthesized to our specifications and these are made of DNA.

Polymerases also require a template – a piece of DNA or RNA that is used to determine which nucleotides are added to the new molecule and in which order. The nucleotides added are complementary to those of the template and the new strand is antiparallel to the template strand.

There are many DNA polymerases we could cover, but we will confine ourselves here to those that are most relevant to the methods we will cover in the course. The polymerases we will discuss can be categorizes as either DNA-directed or RNA-directed. This term identifies the template; DNA-directed DNA polymerases use DNA as a template while RNA-directed DNA polymerases make DNA but use RNA as a template. However, note that both are DNA polymerases which means that they synthesize DNA, regardless of the template.

Some DNA polymerases are thermostable, meaning that they do not lose activity when subjected to very high temperatures. If you consider that a reason why high fevers are so dangerous is that proteins denature just a few degrees above body temperature, then the existence of enzymes that can remain intact at close to 100^oC is quite remarkable. These enzymes are found in thermophilic organisms that live in hot springs or near hydrothermal vents in the ocean.

When we are making DNA in vitro, whatever polymerase we use, we will require the appropriate buffer, and template, dNTPs (deoxynucleotides) and primers. We use DNA primers for in vitro work.

This video describes the basics of DNA structure and replication and some terminology. Some of you may not need this review (Click here to download the powerpoint slides presented in the recording):

A-1. DNA polymerase I

A1-i. What does DNA polymerase I do?

DNA polymerase I was the first polymerase isolated from E. coli. Its role in DNA replication is not what we might think of as the major role; it does not replicate the majority of the DNA. Instead it is responsible for the removal of the RNA primers in all the newly replicated pieces of DNA. The RNA primers are then replaced by complementary DNA. Because of this role in DNA synthesis, this enzyme has several different activities. The first is the DNA polymerase activity; the enzyme can find the 3′ OH of a piece of DNA and add a new nucleotide to it, complementary to the template. The polymerase cannot move forward though, while the RNA primer of the adjacent (beside it) DNA piece is in the way. So, the second activity is a 5′ to 3′ exonuclease activity. The enzyme removes the nucleotides of the primer ahead of it one at a time, from the 5′ end. As the RNA primer nucleotides are removed, they are replaced by deoxyribonucleotides. DNA nucleotides can also be removed by this activity when we do in vitro (literally “in glass”; it means outside of the cell) work.

The third type of activity of this enzyme is 3′ to 5′ exonuclease activity. This is a proofreading activity. If the incorporated nucleotide is not the correct one, the shape of the double helix will be incorrect. This causes the enzyme to back up, cleaving the recently added incorrect nucleotide, replacing it with the correct one, and proceeding. For some types of applications we may require proofreading activity but for many we do not. Proofreading makes DNA replication quite accurate. The error rate varies with the enzyme but in bacteria is something like one error per hundred million bases or even 1 per billion. Some organisms have more error-prone polymerases with much higher rates of incorrect nucleotide incorporation. If DNA polymerases were perfect, there would be no genetic variation.

A1-ii. Modifications of DNA polymerase I

When we are making DNA in vitro we usually don’t require or even want the exonuclease activities of DNA polymerase I. There are some modified forms that are frequently used for particular applications in which, for instance, proofreading is not needed.

The Klenow fragment is DNA polymerase I without the 5′ to 3′ exonuclease activity. It was found (by Hans Klenow, hence the name Klenow fragment) that a protease called subtilisin from the bacterium, Bacillus subtilis, is able to cleave the polymerase to release the part responsible for that activity. The remainder of the enzyme is then purified. This type of cleavage is possible for proteins that have a modular nature – domains for a particular activity are found in a particular region of the protein, thus it is possible to cleave that region of the protein to remove that activity, leaving the rest of the protein intact.

There is also a Klenow called exo ^– that is available; it has no exonuclease activities at all. It is often used in cases where no proofreading is required. Sometimes we want to make a large amount of DNA quickly, but the odd single nucleotide sequence error is not important. In such a case, Klenow exo^– is a good choice.

DNA sequencing by the Sanger method (see Chapter 9) uses the Klenow fragment or the Klenow exo^– enzyme. These are also used when making labeled probes for for a variety of purposes.

One purpose is Southern blot hybridization. In this procedure, gDNA is cut with enzymes and run out on a gel. There will be many pieces of DNA on the gel, of different sizes, that appear as a smear in the lane. The DNA is then transferred to a membrane. To detect a particular sequence, a probe is made that is complementary to the sequence of interest and the probe is labeled with radioactivity or fluorescence or another type of label. The probe is hybridized to the membrane under conditions in which only the complementary sequence – usually a gene we are interested in studying – will be detected. Further details of the procedure are not needed here, but we will come back to the labeling of probes and hybridization in a later chapter. Northern blots are similar to Southerns, except that RNA is run on the gel (which is a denaturing gel, see Chapter 3) and transferred to the membrane which is then probed by essentially the same type of process as for a Southern.

DNA polymerases that lack 5′ to 3′ exonuclease activity are important for applications where “strand displacement” synthesis is taking place. In this case the same template can be used to make many copies of complementary DNA. If DNA polymerase I was used to make probe for instance, and it was on a DNA template that already had been copied, the intact polymerase I would degrade the previously made strand as it made the new one. This would limit the amount of DNA produced. However, Klenow, or Klenow exo^– are not able to degrade the existing replicated strand. Instead, they synthesize the new DNA strand while displacing (pushing aside) the previous strand. In this way many copies of a sequence can be made with a small amount of template. When trying to visualize this process, I use the image of a snowplow, pushing away the snow in front of it.

The video below explains the basics of DNA polymerase I and its derivatives (Click here to download the powerpoint slides presented in the recording):

A-2. Terminal transferase

Terminal transferase (TdT) is a template independent polymerase. That means it can work in situations where there is a template but also when there is no template. If there is a template it will act like any other polymerase. But it can also add nucleotides to 3’OHs that are not double-stranded and in this case random nucleotides are added because there is no template strand to direct which complementary nucleotides are to be added. It is used for a variety of purposes, such as filling in a recessed end in DNA to make it blunt, and in making the second strand of a cDNA (see below). These applications are hard to explain here without some needed background. Therefore, watch for descriptions in subsequent sections where terminal transferase is mentioned.

A-3. Reverse transcriptase

Reverse transcriptase is RNA-directed, using RNA as a template and producing a complementary DNA. It is still a DNA polymerase, and so still requires primers. This often misunderstood by students because of the use of RNA as a template.

Reverse transcriptases are found in retroviruses, viruses with RNA genomes. These viruses infect cells and try to turn the host cell into a factory for producing more viruses- so many in fact, that the cell will eventually burst open, spewing the newly made viruses out to infect more cells. But the infected cells are not accustomed to using RNA as a genome. So the virus provides reverse transcriptase enzyme, that makes a DNA copy of the viral genome. This involves the enzyme also using the newly made DNA copy as a template to make the “second strand” of the cDNA. Some reverse transcriptases also have RNase activity, degrading the original RNA genome as they make the second strand of cDNA. Once the genome has been transformed into double stranded DNA, it can integrate into the host genome. The integrated viral genome can then be replicated and passed on to new daughter cells along with the rest of the genes of the host. Eventually some kind of stress will trigger the viral genome to be excised from the host genome and the host cell will start produces massive amounts of viral components – all the materials needed to assemble many new viruses and eventually lyse the host cell.

We use reverse transcriptase in the lab for making cDNA.

A3-i. Why cDNA?

cDNA is a copy of mRNA, and thus it contains an open reading frame (ORF) that codes for the protein product of the gene. If we are studying a gene and we want to know the protein product, it is much easier to discern from the cDNA than the genomic sequence, in which the ORF is interrupted by introns- sequences that function like “spacers” and have no coding information. We can also determine the sequence of the UTRs on the message. This may provide information as well; we can find the ribosome binding site on the leader sequence. Sometimes there is regulatory sequence on a UTR or one that is needed for sequestration of the transcript before it is translated. Those of you who have taken developmental biology may remember a little about sequestration of the bcd (bicoid) transcript during the anterior-posterior specification of fly embryos. If you haven’t yet taken development, watch for this. Making connections between the classes you take will help you understand both courses better.

If we want to express a protein product from the gene of interest and we plan to use bacteria, we can’t use the genomic sequence as it most likely contains introns. Bacteria have no mechanism for splicing introns. Using the genomic sequence of a gene would certainly result in an incorrect protein being made. The genomic DNA would be transcribed but when translated, the intron sequences would be included, leading to incorrect amino acids at the least and most likely premature termination. In any case the wrong protein would be produced. We must use the cDNA.

And cDNAs are shorter than the corresponding genomic DNA, and sometimes are very much shorter. It is easier to clone shorter inserts into a plasmid than longer inserts. I once worked on a gene that was ~ 250,000 bp long but the cDNA of that gene was only ~2.5 kb. Imagine trying to find the open reading frame in the genomic sequence. And it was much too big to be cloned into a plasmid.

We can work with DNA more easily than RNA – it is more stable. And we cannot sequence RNA directly – we need to first make a DNA copy and sequence that. Once we have the cDNA sequence we can align it against the genomic sequence and determine the number and position of introns in the gene. We can also look at multiple cDNAs for a particular gene and see if they are all the same. If there are differences we can determine that alternative splicing may be taking place. Many genes can produce several variant forms of a protein, most often in different tissues. This is efficient because the different proteins can be produced from a single gene. One of the most spectacular examples of alternative splicing is the Dscam (Down syndrome cell adhesion molecule) gene, which in fruit flies (Drosophila melanogaster) can produce 38,016 different mRNAs (!) through the alternative splicing of its 95 exons¹.

A3-ii. The first strand

When the first strand of the cDNA is made, RNA is used as a template to produce DNA. Then, to produce a double stranded DNA, that first strand of DNA is used as a template to make the complementary second strand, which requires a different enzyme, as is described below.

Most often we want to make the first strand of all the mRNAs in the sample. This is the most efficient approach, simply because you can make a first strand for most of the mRNAs in your prep and then you will have material from which you can amplify many genes. Even if you are only studying one or two genes, having this resource can be useful. You may discover that there are other genes important to the process you are studying, and you will be able to use your first strand material to amplify those genes too.

In this case we use a primer- remember ALL DNA polymerases require a primer– that will recognize something common to all or most mRNAs. This is the poly A tail. So a primer made of a series of T nucleotides should bind to the poly A tail, and then we can make a strand of DNA that is called the “first strand”. This will be a sample of many first strands, of all the mRNAs that were in the sample. The primer is called oligo (dT) meaning it is an oligonucleotide (that is the oligo part) a short stretch of nucleotides, and the oligo is made of just deoxy-T nucleotides (the (dT) part).

If we really want the mRNA for only the gene we are studying and we know the sequence of the gene, we can order a primer complementary to part of the 3′ UTR of the mRNA and we can make a first strand of just the gene of interest. This would not be very efficient if we were studying multiple genes, though, as it would require a separate primer for each gene. If we make a first strand for all the mRNAs it only requires the one primer.

When we reverse transcribe the first strand from an RNA prep, we might first do a DNase I treatment to get rid of any contaminating DNA in the prep. We want the cDNA we make to be from the RNA we isolated, not a DNA contaminant! Then we have to carefully heat kill the DNase I to ensure it is no longer active. Otherwise, it will degrade whatever DNA is produced in the RT reaction. We heat the sample at high temperature – say 70^oC – for about 15 minutes to ensure it is inactivated. I never trust the protocols for inactivating enzymes and usually heat kill for longer, just to be sure.

Sometimes we might want to make complementary copies of all RNAs, including those without a poly (A) tail. In this case we can use random priming. We can purchase pre-made primers that are random hexamers. This means the primers are 6 nucleotides long (hexamers) and are a mixture of many combinations of sequences that are 6 nucleotides long. These are not unique sequences in a genome and so they can initiate the first strand from all the RNAs in the sample. We often don’t get full length cDNA from this approach because the primers can bind anywhere along the RNA. So each first strand that gets made is only part of the sequence.

A3-iii. The second strand

There are several ways of making the second strand of DNA though this isn’t absolutely needed. In our lab we normally use the first strand reaction in a PCR and use primers that recognize both ends of the gene we’re studying because in the lab exercise we’re only amplifying one gene. In that case the second strand is automatically made in the PCR reaction and these days, most of the time we are amplifying a gene with a known sequence.

However there are situations where you may want to make a cDNA library of all the mRNAs in particular tissues or stages of an organism, or even from a sample of multiple organisms. In such cases you would not know the sequences of the all genes from the organisms and you would need to make the second strand of the cDNA in a different way. In this case we could use several approaches. They are more complicated than described here, but this is a good introduction to a few approaches. There are also many different approaches than these which you can read about online if you want to find out more. I selected just a couple of different approaches here as examples.

AMV RT enzyme can randomly initiate the formation of a hairpin at the 3′ end of the newly synthesized DNA strand. If this happens, there will be a 3’OH available for a polymerase to add nucleotides to, as the DNA molecule folds back on itself. DNA Polymerase I can be used to add complementary nucleotides, forming the second strand. After the reaction is completed, S1 nuclease (see below) can be used to degrade the single stranded part of the structure, i.e. the loop. This leaves double stranded cDNAs of many genes. Whether or not the hairpin forms at the 3′ end of the new strand is variable and exactly where it happens is also not controllable. Also, the S1 nuclease removes the loop part of the structure, which is part of the cDNA sequence, so this procedure does not result in a full length cDNA. We get only a partial cDNA.

Some RTs also have intrinsic RNase activity. These leave nicks in the RNA as it is being reverse transcribed. This results in a DNA:RNA hybrid molecule in which the RNA part is in fragments. This looks a bit like DNA replication on the lagging strand before the Okazaki fragments have been dealt with. DNA polymerase I can be used in this case as well. It will find the 3’OHs of the RNA fragments, and add deoxynucleotides complementary to the DNA part of the strand. As the polymerase moves forward along the DNA it will degrade any RNA (or even DNA) ahead of it. Given sufficient time, the result should be a nearly complete double-stranded cDNA molecule.

After first strand synthesis is complete and the RNA has been degraded, it is also possible to use terminal transferase (TdT, briefly described above) to add nucleotides to the 3′ OH of the DNA molecules. In this case there is no template so the TdT will add nucleotides at random to the first strand DNAs. You can set up the reaction: first strand DNA, appropriate buffers and sufficient water and then provide only one nucleotide in the reaction – dCTP. You perform the reaction for a short time, enough for the enzyme to have added maybe 15-20 Cs to the 3′ end of the first strand. Then you used DNA polymerase I and an oligo (dG) primer (made only of G nucleotides) to manufacture the second strand of the DNA. This is the most popular method used for second strand synthesis.

This recording covers the main aspects of RT use in the lab (Click here to download the powerpoint slides presented in the recording):

A-4. Thermostable DNA polymerases

Thermostable DNA polymerases are very unusual because they can be incubated at high temperatures (95^oC for instance) for some time without losing activity. They were originally isolated from thermophiles, organisms that live in hot springs and around hydrothermal vents. These enzymes are useful for situations where DNA is made through repeated cycles that include a denaturation step (where DNA strands are made single stranded via high temperature). The production of DNA this way is called PCR and we will devote all of Chapter 5 to the basics of PCR, one of the key most important techniques in biology.

B. Nucleases:

Nucleases are enzymes that break phosphodiester bonds and thus cut DNA or RNA into small pieces, or down to single nucleotides. Some nucleases are endonucleases – these cut inside a molecule. Others are exonucleases, which begin at the end of a molecule and remove nucleotides one at a time. Whether the entire molecule is degraded or not depends on the type of enzyme. Some are site-specific; these restriction endonucleases are very useful in genetic engineering. Others are not sequence specific but might affect only (or mainly) single stranded parts of a molecule. These each have their purposes or roles in genetic engineering. In general we like to use the enzymes we can control and you’ll see many such uses of nucleases and/or polymerases throughout the course. And we deal with enzymes we cannot control, either through heat killing them after we have used them or by avoiding them altogether (for example, ensuring there are no nucleases in our DNA or RNA preps)

B-1. Restriction endonucleases – essential components of molecular biology

Restriction enzymes are an example of a great idea co-opted from other organisms, in this case from bacteria. We will briefly consider what restriction enzymes are used for in nature and then how one class, the type IIP enzymes, work. We’ll touch on one other class of enzymes that is used for alternative cloning approaches with details to follow in a later chapter. We will finish with a short overview of the role of restriction enzymes in many cloning procedures.

B1-i. origins of restriction enzymes

Restriction enzymes are found in many types of bacteria and archaea and are a first line of defence against viral infection. These enzymes recognize a particular sequence of DNA for instance: 5′-GAATCC-3′, or 5′-GGCC-3′ and cleave both strands of DNA somewhere within that sequence. Broken or cleaved pieces of DNA can be degraded by exonucleases, thus preventing the phage (virus that attacks bacteria) from infecting the cell. Bacterial DNA may also contain the target sequences, but a sequence-specific methylase adds methyl (CH₃) groups to one or more nucleotides in the recognition sequence which prevents the restriction enzyme from recognizing and cleaving the bacterium’s own DNA. This system is called the restriction- modification system. The “restriction” part refers to the cleavage of the invading viral DNA by the restriction enzyme and the “modification” part refers to the bacterium modifying its own DNA to protect it from the restriction enzyme. Bacteria can further protect themselves by somewhat compartmentalizing the enzymes. The restriction enzymes are localized to the periplasmic space (in gram negative bacteria) between the two membranes surrounding the cell. This is where the phage DNA is inserted into the cell, therefore the restriction enzymes are poised to cut this DNA as soon as it enters the cell. The corresponding methylases are cytoplasmic, in the region where they are most likely to encounter the bacterium’s own DNA and less so the invading DNA.

Many different restriction enzymes have been isolated from bacteria and these are named according to the species from which they are derived and the order of isolation. For example, EcoRI was isolated from E. coli (that is the Eco part and why it is italicized; it is derived from the species name which is by convention italicized). The R in this name refers to the strain of E. coli from which the enzyme was isolated. It is not italicized. And I means it was the first such enzyme described in this species and strain.

Click here to download the powerpoint slides presented in the video below.

Restriction enzymes have been instrumental in genetic engineering, as described below.

B1-ii. type IIP enzymes

Type I restriction enzymes are fairly common, and these cut randomly, at variable distances from their recognition sequences. Although they are very useful to the bacteria, and interesting to study, they are not useful in genetic engineering because they don’t produce discrete and predictable DNA fragments. Type II restriction enzymes are used in genetic engineering. These restriction enzymes cut either in or close to their recognition sequences and generate discrete fragments of DNA as a result – if you know the sequence of your DNA, you can be sure of where type II enzymes will cut, what pieces of DNA will be produced and what type of overhang, if any, will be generated.

We will first look at type IIP enzymes because these are the most commonly used restriction enzymes. The P in the name means palindromic. A palindrome can be read backwards and forwards and give the same meaning: bob, aha, eve, rotor, a man a plan a canal Panama (!) etc. The recognition sequence for EcoRI is 5′-G/AATTC-3′. Consider the double stranded nature of DNA and see if you can see why this sequence is called “palindromic”. The  /  in the sequence indicates the position at which the enzyme cleaves the phosphodiester bond. Remember that DNA is double stranded and so the cut on the other strand is: 3′-CTTAA/G-5′. Draw this out to see the symmetry and notice that the cut is between the same two nucleotides on each strand. If the cuts take place on both strands and you separate the two resulting pieces of DNA, you will see a 5′ overhang – a place where the first 4 nucleotides of the DNA molecule are single stranded. The 3′ end is recessed and the 5′ end sticks out. The exact same overhang is generated on both strands of the DNA. It is really important to be drawing this out in order to see how it works. Being able to visualize what is going on with the DNA you’re working with is essential to understanding the techniques you are learning. These single-stranded ends of the molecule are able to base pair with complementary nucleotides on other pieces of DNA cut with the same enzyme. Therefore, the overhangs are also referred to as sticky ends. If we cut two different pieces of DNA with the same enzyme, the resulting fragments will have complementary sticky ends. This allows us to combine these fragments to make a construct: a vector that contains DNA we have engineered (combined from different sources) for a particular purpose.

EcoRV is the fifth restriction enzyme isolated from E. coli, strain R, and its recognition sequence is: 5′-GAT/ATC-3′. If you draw both strands and look at the cut site you will see that they line up. If you separate the two cut pieces of DNA in this case you will have no overhang, and therefore, this enzyme is called a blunt-cutter because the ends are even, or blunt. Blunt-ended DNA molecules can be combined but it is more challenging than using sticky-ends because there is no complementary base-pairing holding the pieces of DNA together. This will be discussed briefly in section C.

(Click here to download the powerpoint slides presented in the recording):

B1-iii. type IIS enzymes

Type II S enzymes do not have palindromic recognition sequences. For example the enzyme BsaI has the sequence: 5′-GGTCTC-3′. Hopefully you can see that this sequence is not palindromic. These enzymes also have two separate activities; the DNA binding and cleavage activities of the enzymes are distinct domains of the protein. The result of this is that the enzymes cut to one side of the recognition sequence, because the parts of the enzyme that bind to the DNA and do the actual cutting are physically contacting different DNA sequences. Their activity is still predictable, so if you know the sequence of a particular piece of DNA you can predict exactly where the cut site will be, but multiple pieces of DNA cut with the same type IIS enzyme will not necessarily have compatible sticky ends. That is because different pieces of DNA will have different nucleotides beside the recognition sequence.

The way the cut sites are indicated is like this: 5′-GGTCTC-3′ (1,5). This gives the recognition sequence: GGTCTC, followed by the cut sites, which in the case of BsaI is 1 nucleotide 3′ of the C in the recognition sequence on the same strand, and 5 nucleotides 5′ of the corresponding G on the other strand. This will yield a 4 bp 5′ overhang that will differ depending on the sequence of the DNA cut by the enzyme. (Draw it, using Ns for the nucleotides to either side of the recognition sequence)

The enzyme MlyI cuts this sequence: 5′-GAGTC-3′ (5/5) and thus would leave a blunt cut. and the enzyme BspCNI cuts the sequence 5′-CTCAG-3′ (9/7), which would give a 2 bp 3′ overhang. Draw out all three of these recognition sequences, remembering to include both strands and this will help you see how the sticky ends are generated.

We use these types of enzymes in some alternative approaches to cloning, so they will be discussed more extensively later in the semester.

B1-iv. Use in cloning

The video below is free access material that is to be used along with a wet lab (doing the actual procedure in the class). But this introductory part is a good brief explanation of some of what we’ve covered in this and previous chapters- a VERY SIMPLE DNA isolation, restriction digest and running a gel – no protocols or details are given, just the basics of how it works. You must unfortunately disregard the last few seconds that explain what the wet lab will involve 🙁

https://www.jove.com/science-education/10628/dna-isolation-and-restriction-enzyme-analysis

Below is some information on the general use of restriction enzymes (just the IIP enzymes for now), including a sample recipe, some considerations regarding the temperature and amount of enzyme to use and some examples of useful resources online that can help you plan an experiment.

(Click here to download the powerpoint slides for the video below)

B-2. S1 nuclease

S1 nuclease is a single strand specific nuclease and can be used as described above, to remove the loops that are sometimes generated during cDNA synthesis. There are other uses of the enzyme in techniques that are fairly out of date now. Watch for mention of SI nuclease in procedures described in other sections of the text.

B-3. DNase I

DNAase I can cut either single-stranded (ssDNA) or double-stranded DNA (dsDNA) into fragments. Because it lacks specificity, it has limited uses. As mentioned in Chapter 3, it is too expensive to use to remove large amounts of DNA from RNA preparations but is sometimes used prior to a reverse transcription reaction (described above), to ensure that cDNA is made from the RNA, and not from contaminating genomic DNA (we are only concerned about the second strand part in this case).

It is also sometimes used to identify protein binding sites in DNA, a technique known as DNA foot printing. In this technique, amplified DNA in a gene or region of interest is incubated with a protein that is being studied. The protein should recognize and bind to only its recognition sequence, if any, on the DNA. A sample of the same DNA not incubated with the protein serves as a control. Both samples are subjected to a short DNase I treatment and then the DNA is run on a gel. The DNA that wasn’t bound by the protein is cut in random positions that are characteristic of that DNA sequence, called DNase I sensitive sites. When this DNA is run on a gel, there will be a ladder of bands reflecting those sites. The DNA that was bound by the protein will be cut in the same positions except where the protein is bound to the DNA at a DNase I sensitive site. Here, the DNase I does not cut the DNA and thus there will be a “gap” in the DNA fragments seen in the gel. When we compare the two samples side by side we can identify which pieces were bound by protein.

Why do we do this type of experiment? There are still many proteins that regulate gene expression that are not well understood- we don’t know what sequences they bind to either activate or repress the transcription of a gene. This is one approach to trying to identify or map the binding sites of these proteins for each gene. It helps in defining the exact regulatory regions of those genes.

Here is a short video that briefly describes the technique:

B-4. Exonuclease activities of DNA polymerases

The exonuclease activity of some DNA polymerases are used in alternative cloning approaches. This is described in detail in a later chapter, but here just note that the exonuclease activity of these enzymes can be somewhat controlled; the exonuclease activity is activated in the absence of nucleotides. DNA incubated without dNTPs and with the polymerase will lead to the inactivation of the polymerase activity (because there are no nucleotides to add to a DNA strand) and the activation of the exonuclease activity. Some enzymes like T4 DNA polymerase (isolated from T4 bacteriophage), remove nucleotides in a 3′ to 5′ direction, essentially “chewing back” the 3′ end of a DNA molecule and generating a 5′ overhang. The length of the overhang depends on how long the incubation takes place. T5 polymerase (isolated from T5 bacteriophage), on the other hand, removes nucleotides in a 5′ to 3′ direction, meaning that is “chews back” the 5′ end of the molecule, leaving a 3′ overhang. Careful timing is important to ensure that the overhang you are producing is an appropriate length for the procedure you are performing. The reaction is stopped abruptly by the addition of a huge excess of just one nucleotide, dCTP for example, to stimulate the polymerase activity of the enzyme and suppress the exonuclease activity. Consider that we are trying to generate an overhang of a certain length for some purpose and ask yourself why we add just one nucleotide rather than a mixture of all the dNTPs. What would happen if we added all four nucleotides?

T4 DNA polymerase can also be used when we have DNA molecules with 5′ and 3′ overhangs and want to generate a blunt ended molecule (no overhangs) for cloning or some other purpose. In this case the DNA polymerase activity adds nucleotides to the recessed 3′ end of a 5′ overhang and the exonuclease activity will remove the 3′ overhang. This process occurs in the presence of dNTPs in the reaction, so the exonuclease activity ceases when the molecule end is blunt.

B-5. Exonuclease I

Exonuclease I  (ExoI) is specific for single stranded DNA and is sometimes used to degrade the single stranded primers left over in PCR reactions. There are large excesses of primers in the reactions and so once the PCR reaction is complete there may still be many primers left. Exo I can degrade these so they do not interfere with the next steps of the cloning procedure. There are also certain column prep methods that exclude small pieces of DNA, so we can eliminate primers that way also.  Column preps in my experience have poor yields so the ExoI method might be preferable so long as it is thoroughly heat killed after use.

C. DNA ligase:

DNA ligase is used to join DNA fragments together. Although we think of it as joining the pieces together, ligase acts differently than you may think. It “activates” the free 5′ ends of the DNA molecules by adding AMP  to them. This creates a situation in which the actual joining reaction- the formation of a phosphodiester bond, can occur spontaneously.  So the ligase enzyme does not actually catalyze the ligation; it does however, generate the conditions for the reaction to occur on its own. The most commonly used ligase is T4 DNA ligase. You will notice a lot of enzymes with T5, T4, T3, and T7 in the name. These are isolated from bacteriophages with those names. There are DNA and RNA polymerases, and ligases  that come from these phages.

A key component of a ligation reaction is thus the ATP. If ligation has not worked correctly, the first thing to check is whether the buffer contained sufficient ATP. In our lab (a long time ago now, 🙁 ) we used to routinely add some ATP to our reactions if we’d had the ligase buffer for a while, to ensure that there was sufficient ATP.

If your enzyme is not expired and the amount of ATP in the buffer is correct, another possible reason for failure of a ligation reaction is having the wrong insert to vector ratio. Successful ligation based cloning depends on the right ratio of the DNA fragments to be combined. We will discuss the ratio of insert to vector when we discuss plasmids. In general we want an excess of insert molecules over vector molecules. And here we are specifically talking about the molar ratio: the numbers of molecules rather than just the weight. The same weight of different sized molecules necessarily contains different numbers of molecules. Think about this and we’ll come back to it in a later chapter, when we discuss making a construct. If we have molecules with sticky ends to join a 3:1 insert:vector ratio works well. If we are joining molecules with blunt ends, ligation is not as efficient, since there is no complementary base pairing to hold the molecules together. In this case a much larger excess of insert is needed, between 5:1 and 10:1 insert to vector. In addition, the vector that is blunt cut could simply recircularize in a ligation – there is nothing stopping the enzyme from closing the vector. This can make the ligation of the insert into the vector very inefficient. There are two ways to deal with this. One is to treat the vector with phosphatase enzyme (calf intestinal phosphatase, CIP, or shrimp alkaline phosphatase, SAP) to remove free 5′ phosphates. Without a 5′ phosphate, ligase cannot close the vector. This is not 100% effective – some vector molecules will escape this -but can reduce the amount of “empty” vector produced in our experiments. And in the lab in the past when we’ve done blunt end ligation we have included the restriction enzyme we used to cut open the plasmid in the ligation reaction. The idea is that if the plasmid recircularizes, the enzyme will cut it open again.  Once an insert is in the vector, there is no more restriction site for the enzyme to cut and so this also enhances our ability to clone the desired insert into the vector. There will be more information on this – and other methods to select for the construct we want – in the chapter on plasmids, transformation and selection.

The reactions can be done at various temperatures- this is not a fussy enzyme. The lower temperatures take longer. Some people prefer to do slow reactions at lower temperatures.

Ligation is not always necessary in genetic engineering; it depends on the cloning method that is used. For instance in some methods, there is enough overlap in the compatible ends of the molecules you are joining to hold them together until they enter the bacteria (more on this in later chapters on alternative cloning methods). Once inside the bacterium, any gaps can be filled with DNA polymerase, and the molecules ligated together with the bacterial ligase.

References

Previous (Chapter 3)Next (Chapter 5)