PLASMIDS101
What is a Plasmid?
Plasmids are circular chains of DNA that contain the replication machinery necessary to be included in the process of binary fission. This circular nature allows the DNA to supercoil into a tight structure that can slip through a cellular membrane much more easily than linear DNA.
Plasmids will often contain genes that convey transient benefits to their host. Plasmid hosts are most often microorganisms, such as prokaryotic bacteria or eukaryotic yeasts - however they’ve been observed in higher order organisms, such as in the mitochondria of corn. Plasmid DNA is often shared between these microorganism hosts via transfer interactions, or co-habitation interactions.
Transfer Interactions -
Co-integration
Facilitation
Mobilisation & Co-transfer
Co-habitation Interactions -
Epistasis
Incompatibility & Interference
Researchers have found several convenient tricks for artificially controlling the utilisation of plasmids in bacteria. The simplicity of these techniques make plasmids an ideal vector for inserting new genetic elements into a host. Including an antibiotic resistance gene on a plasmid, then growing the bacteria on antibiotic-media is a sure-fire way to force a bacteria to keep a plasmid generation after generation. The antibiotic media also acts as a selective agent, killing any other bacteria that might contaminate your experiment. Synthetic Biology on easy mode.
How to mentally picture a Plasmid
3 Degrees of Abstraction
(how to imagine a plasmid into a simple form)
O°: Native Model: Intracellular, actively functioning DNA
Inside the cell, your plasmid could be in any state ranging from a tight supercoil to completely relaxed. If unwound, the plasmid will likely start recruiting transcription proteins to begin the production of mRNA. As such, some regions be unzipped and single stranded. Sitting in the cytosol of a prokaryote, the plasmid may encounter anything from the genome to stored fats. Within a eukaryote, more of the organelles will be membrane bound (including the genome) but there will still be plenty of environmental variables to ruin a good mental model. Internal cytoskeletal structures may provide an anchoring point for the plasmid, or be used in conjunction with transport proteins to move the plasmid to a desired location. When replication occurs, the plasmid will split down the middle, temporarily forming a structure that resembles two circles stuck together. This is an examples of a DNA catenane, a structure that has to be removed at the end of replication by a dedicated protein. Other types of DNA catenanes and knot structures can occur, the purpose of which is not always clear.
1°: 4D Isolated Plasmid DNA in various states of coiling
Plasmid DNA that has been isolated from the cell and purified into a buffer is a more feasible model to imagine, although it still contains more detail than is strictly necessary. The buffer matches the native state pH of the cytosol, allowing the plasmid to fold more naturally. The circular double-stranded DNA molecule may twist into a formidably large number of structures while maintaining the hydrogen bonds between pairs. Coiling can occur via two motions; twist and writhe. Twisting is caused by rotating the plasmid around it’s own vertical or horizontal axis. A loose plasmid that is twisted around an axis will form a helical shape that tightens as you turn. Writhing is caused by the asymmetry around the central axis, this introduces enormous variability as the plasmid can bend in any direction.
A good understanding of how DNA coiling works is especially useful for interpreting the results of Agarose Gel Electrophoresis. It will allow you to understand how the different structures within your purified plasmid prep will affect migration through the gel.
2°: 3D Circular Double-Helix of DNA
Stripping away the 3D geometry we can imagine a double helix of plasmid DNA in its most relaxed state. This allows us to finally look at the sequence of base pairs held within! The benefit of this model over the next one is the inclusion of the backbone structure, so you may wish to use this model when mentally picturing experiments that will introduce single- or double-stranded breaks. It is important to keep track of the phosphates, as at least one of the DNA ends must have an attached phosphate for a ligation experiment to succeed.
3°: 2D circular DNA Sequence
This model will be the one you use for all of your design and plasmid theorycrafting. Genetic design at the sequence level is remarkably simple, especially when working within single-celled organisms. While you should understand how simplified this model is and what it ignores, it is incredibly powerful. A 2D circular DNA sequence is all you really need to succeed while preparing for most experiments. You may occasionally reach for the more complex models for your design, but more often than not you’ll be overthinking it.
What is the structure of a Plasmid?
Almost all of the rules about the structure of DNA from the DNA101 guide apply for plasmid DNA. The main fundamental difference arises in the supercoiling that plasmids are capable of performing.
Image: Mentally picturing supercoiling: Imagine an old-fashioned telephone cord, tangled around itself, or; Take a hairband/rubber-band and twist it. Observe how the structure tightens, fitting more length into less space.
A supercoiled plasmid is a far denser structure than genomic DNA in its native state thanks to significantly more turns in the double helix structure. This dense structure allows it to pass more quickly through an agarose gel than it would in a linear state. This results in a faster moving band relative to size than the DNA ladder - plasmids never match perfectly to DNA ladders!
A double-stranded nick in the plasmid backbone will result in linearised plasmid. This might be repaired by the cell if it occurs in vivo, but if it occurs in a purified plasmid prep it means you’re very unlikely to succeed with a Plasmid Insertion Protocol. A linear strand of DNA will not be capable of supercoiling into a structure small enough to fit through the outer membrane(s) of a host cell. You may also produce linearised plasmid intentionally via Restriction Enzyme Digest. This linear strand will run slower than the supercoiled structure on an agarose gel, but will more accurately match the size standard. Linearised plasmid is the best method for estimating total plasmid length from a gel. If you linearise your plasmid for an experiment, it will need to be ligated (glued back together) before you can insert it back into a host!
A single nick in the plasmid backbone can relieve the tension in the supercoil, resulting in a relaxed plasmid state with a similar number of turns in the helix as in genomic DNA. A single-stranded nick is easily repaired by DNA ligase without any risk of mutation, so the sequence does not change as the plasmid undergoes uncoiling in vivo. In vitro the nick will not be prepared and the plasmid will remain relaxed from now onwards. As a plasmid prep ages, the relative quantity of relaxed plasmid will increase as more and more nicks arise. The nicked plasmid is in the loosest conformation, and runs the slowest on the gel.
Your plasmid minipreps will most likely contain a mix of supercoiled, nicked and linear plasmid and the relative brightness of these bands will tell you how structurally sound your purified DNA is. Observed on an agarose gel, this looks something like this;
There is a rare, final (cursed) band that might appear on your gels - the circular, single-stranded band. This means that your plasmid DNA has separated into two distinct single-stranded circles of DNA with denatured nucleotides. If you’re seeing this band, you need to review how long you are spending on each step during the plasmid miniprep protocol.
How do Plasmids work?
Our ability to utilise and manipulate plasmids relies upon the use of a standardised parts for each important function the plasmid needs in order to survive. If you too wish to use plasmids, you’ll need to know all of the functions! Fortunately, you can find a very detailed guide on the design principles required for selecting the right parts on the Backbone Selection page. If the genetic design theory is a bit heavy reading for you, here’s a quick summary of the common plasmid features;
Origin of replication - This region dictates the rules for plasmid replication to the host, including how many copies should be made. It is also the binding point for the DNA polymerase during replication. This is the region where the two circles of double-stranded DNA begin splitting apart during replication.
Antibiotic Resistance Gene - This region codes for the genes necessary to resist a specific antibiotic. This region is not necessary in wild-type plasmids, but is extremely useful to the researcher for controlling the uptake and utilisation of a plasmid by a host. These genes can naturally occur in the genome as well.
Multi-Cloning Site - These are standardised junctions, aka. prefix/suffix regions that you can use to insert any gene you wish. Historically these have required a specific set of restriction enzyme binding sites, but this is subject to change as scarless cloning technologies like Gibson Assembly take over. Multicloning Sites (MCSs) occur naturally, but it is unlikely that their natural purpose aligns to ours. The multicloning site may be interior or exterior to the following parts. You will need to take the location of the MCSs into account prior to Gblock Design to ensure that your final construct contains all the necessary machinery for successful protein synthesis.
Gene of Interest - This is the gene that you’ve inserted into your plasmid, hoping to express it in the host. This may be anything from Green Fluorescent Protein to Human Insulin.
The following features are common to all of the expressed genes on your plasmid, from your gene of interest to the Antibiotic Resistance Gene. You will see multiple iterations of these features on each plasmid, flanking the genes. Genes that have constitutive (always on) expression probably won’t have an associated Operator/Repressor system. If they do, it might be broken or only rarely switched off.
Promoters - These regions help control the rates of transcription of each region. The promoter region is recognised by RNA polymerase, which will attempt to bind and initiate transcription. However! A promoter may rely on a nearby operator to be bound by a specific protein (known as a transcription factor) in order to function. If a repressor protein is bound to the operator instead, the promoter will be hidden and the RNA polymerase will look elsewhere.
Operators - Between the promoter and the RBS, the operator binds to transcription factors that will assist the binding of RNA polymerase to the promoter. However! It may instead prefer to bind to a repressor protein that will bend (chelate) the DNA and obscure the promoter from the RNA polymerase.
Ribosome Binding Sites - Region bound by the ribosome for the initiation of translation from RNA strand to protein. Requires a start codon nearby to initiate an Open Reading Frame (ORF).
An Open Reading Frame is a set of specific three base-pair codons that starts with a start codon and ends with a stop codon.
The RBS should be 6-8 base pairs upstream (in the 5’ direction of) the Start Codon.
Start codons - First three base-pair codon translated by the Ribosome, always AUG > Methionine. The start codon represents the start of the ORF. All subsequent codons will be in sets of three, starting from the first AUG.
Stop Codons - Three base-pair codon that allows the ribosome to cease translation and detach from the RNA strand. The stop codon must be in frame with the start codon in order to function. The stop codon is not translated, so the length of the protein produced is the distance between the start of the Start Codon and the last base before the Stop Codon.
Terminators - This region allows the RNA polymerase to detach and cease transcription. Without a terminator, the RNA polymerase will just keep going, resulting in a nonsense product.
Repressor proteins - These proteins are expressed separately, with all of the above machinery included. The repressor protein binds to the operator, preventing the production of mRNA and thus downstream protein. Repressor proteins can have their own operators and can be prevented from being transcribed by an additional repressor protein. This allows for signal cascades, whereby multiple genes are switched on in a row thanks to a single change. (Playground of see-saws analogy).
Repressor proteins may preferentially bind a molecule that isn’t the operator, thus freeing up the promoter to bind the RNA polymerase in the gene that was repressed.
e.g. the lac repressor molecule will competitively bind to IPTG or lactose, freeing up the promoter to do it’s work.
Once In Vivo, the systems coded within the plasmid will be interacting with the larger genomic code of the host. This can result in… interesting complications if ignored. For example, insulin synthesised in the cytoplasm of an E. coli will be badly misfolded due to the reductive atmosphere. This necessitates the export of any produced protein to the periplasm of the cell, where it will be folded correctly. This can be coded into the plasmid via another standardised part;
Fusion Protein - A protein that is bonded to the N- or C- terminus of your protein-of-interest (sometimes with a linker). This is coded in DNA by placing the code for the protein immediately upstream or downstream of your gene-of-interest.
Fluorescent fusion proteins (e.g. fuGFP) can be used to quickly and easily identify successful expression of a protein without the need for purification. Keep in mind that this will decrease the efficiency of the production of your actual protein of interest. Works great as an early validation step for an experiment.
Ecotin is a fusion protein used in the Open Insulin experiment to export the produced insulin to the periplasm of the cell.
Alternatively, the split between genomic and plasmid code can be utilised to your advantage. Several regularly used industrial and research systems rely on genetic code split between a plasmid and the genome. Systems designed this way are less likely to be viable if they escape the lab (a feature that pleases regulators) because they will only function properly when combined together. If the host gets out of the lab and attempts to share the plasmid with a wild-type organism, the plasmid will be useless without the genomic code. Vice-versa, if the host cell escapes the lab it is unable to ditch the plasmid without losing a potentially valuable genetic pathway. For example;
T7 operon
Blue-White Screen
When do Plasmids work?
Using Promoters and Repressors in Induction systems
Plasmid Reproduction
Natural sharing of plasmids
How can you (the researcher) make use of these features of Plasmids?
Plasmid insertion is an extremely simple procedure, especially with pre-prepared cells. Heat Shock requires only a short incubation on ice, then 45 seconds at 42°C to coax a plasmid to enter a cell. Electroporation requires a more expensive device but is equally quick to perform. There is also the option of using a gene gun, which can be useful for getting plasmids into difficult cell lines using a small pneumatic cannon. While I’m yet to try this, it sounds like fun.
Once a plasmid is inside your host, simply ensure that you replate the host onto new media every few weeks. Most microorganisms will eventually neutralise the antibiotic nearby the colony, leaving the plasmid as an inefficient genetic element that provides no competitive advantage. The cell will spit out the plasmid and lose the ability to express any genes that were contained within.
Whether you wish to avoid losing plasmids to cell rejection or edit the genes contained within the plasmid - at some stage you’ll have to harvest your plasmid from the host. This is done via a process known as Ethanol Lysis aka. Miniprep. The host is lysed, all the extra gunk is purified away and the plasmid is left bound to a silica column. A special buffer elutes the plasmid from the silica column into a tube for storage at -20°C. Stored like this, plasmids can last decades.
If your plasmid has a Green Fluorescent Protein gene with constitutive expression, simply inserting the plasmid will be enough to change the phenotype of the host. However forcing the host to continuously produce a large protein with no function has a competitive disadvantage. Instead you may consider using an inducible expression system on your plasmid, which will let you grow the cells up to a critical mass (OD600 = 0.3-0.5) before commanding all of them to express the protein at once. This will significantly increase your yield - if that is important to your goals.
You can also design plasmids with logic gates contained within. This can allow for some extremely nifty applications, such as producing a cell that glows green only when both Mercury and Arsenic are present, but not when only one is present.