Circular and linear DNA molecules of the same size can migrate at very different rates during gel electrophoresis, and which one runs faster depends heavily on the gel conditions you use. In dilute agarose gels, supercoiled circular DNA typically outruns its linear counterpart, but raise the gel concentration or crank up the voltage and that order can flip entirely. The relationship between DNA topology and gel mobility is one of those cases where there is no single answer, only a set of rules that shift with the experimental setup.
Why Shape Changes Everything
Gel electrophoresis separates molecules by forcing them through a porous matrix under an electric field. For DNA, all molecules carry roughly the same charge per unit length, so size and shape become the main factors determining how fast they move. A linear DNA molecule is free to stretch out, snake through gel pores, and reorient itself as it encounters obstacles. A covalently closed circular molecule cannot do this the same way. It has no free ends, which fundamentally changes how it interacts with the gel matrix.
Three common topological forms of the same plasmid show up in the lab. Supercoiled DNA (sometimes called Form I) is a covalently closed circle with extra twists that make it compact. Nicked circular DNA (Form II, also called open circular or relaxed circular) has one or more single-strand breaks, allowing it to relax into a floppy ring. Linear DNA (Form III) has two free ends and behaves more like a rod or worm. Each form has a distinct effective size and flexibility, which is why a single plasmid preparation can produce multiple bands on a gel.
The Migration Order and When It Reverses
Under standard low-percentage agarose conditions, the typical band pattern from fastest to slowest is supercoiled, then linear, then nicked circular. The supercoiled form is tightly wound into a compact shape that slips through gel pores relatively easily. Nicked circular DNA, by contrast, is an open floppy ring with a large effective diameter, so it gets hung up in the matrix and runs slowest.
This ordering is not fixed. Early work on R12-derived plasmids showed that the relative mobilities of supercoiled and linear DNA reverse at a certain molecular weight, and that crossover point shifts downward as you increase the agarose concentration.1Nucleic Acids Research. Physical properties and gel electrophoresis behavior of R12-derived plasmid DNAs In other words, a supercoiled plasmid that outruns its linear counterpart in a 0.7% gel may lag behind in a 1.5% gel. Studies using molecules ranging from roughly 2 to 7 kilobase pairs confirmed the pattern: in dilute gels below about 1% agarose, circular DNA migrates faster than linear DNA of comparable size, while in concentrated gels above about 1.5%, the circular forms migrate much more slowly.2PubMed. Biased sinusoidal field gel electrophoresis for the separation of large DNA
A similar reversal happens between open circular and linear DNA as gel concentration and voltage gradient increase. Detailed fractionation studies showed that at lower voltages, open circular and linear molecules of the same molecular weight migrate at similar rates, but as both voltage and molecular weight increase, the open circular form falls progressively behind.3PubMed. Conformation of double-stranded DNA during agarose gel electrophoresis: fractionation of linear and circular molecules with molecular weights between 3 X 10(6) and 26 X 10(6) This means the band order you see on your gel is genuinely a product of your running conditions, not an intrinsic property of the DNA forms themselves.
How Circular DNA Gets Trapped
The physical reason for these reversals comes down to how each shape navigates the gel. Linear DNA moves through agarose by a snaking motion called reptation: one end threads into a pore, and the rest of the molecule follows. Circular DNA cannot do this because it has no leading end. Instead, research on polyacrylamide gels demonstrated that circles are most likely impaled on gel fibers, threaded like a ring on a post.4PubMed Central. Effects of supercoiling in electrophoretic trapping of circular DNA in polyacrylamide gels Once a gel fiber passes through the ring, the molecule is pinned and can only move by somehow sliding off the fiber, which becomes increasingly difficult as voltage pushes the molecule harder against the obstruction.
This trapping effect becomes dramatic at high electric fields. When open circular bacteriophage lambda DNA (about 48.5 kilobases) was run in a 0.2% agarose gel, it migrated normally at a low voltage gradient of 0.34 V/cm. But when the field was raised to 6.0 V/cm in a perpendicular direction, the circular molecules stopped entirely while linear DNA of the same molecular weight continued to migrate.5ELECTROPHORESIS. A voltage gradient‐induced arrest of circular DNA during agarose gel electrophoresis The explanation is that higher fields push circular DNA more forcefully onto gel fibers, making impalement essentially permanent for large molecules.
Pulsed-field electrophoresis exploits this behavior in a useful way. By alternating the direction of the electric field, linear molecules reorient and keep moving, while large circular molecules remain stuck. This means that at sufficiently high fields, you can selectively trap all circular forms above a certain size threshold, and that threshold drops as you increase the field strength.6PubMed Central. Separations of open-circular DNA using pulsed-field electrophoresis It is an effective method for separating circular from linear DNA when conventional gel electrophoresis produces overlapping bands.
How Supercoiled DNA Behaves Differently from Relaxed Circles
Not all circular DNA is created equal in the gel. Supercoiled plasmids are far more compact than nicked circles, and that compactness explains why they typically run fastest under standard conditions. When both linear and supercoiled molecules are placed in an electric field inside an agarose gel, they align parallel to the field lines and stretch. But the apparent change in shape is much less pronounced for supercoiled molecules, which resist deformation because their tightly wound structure holds them in a smaller volume.7PubMed. Effect of pulsed and reversing electric fields on the orientation of linear and supercoiled DNA molecules in agarose gels
Temperature adds another layer of complexity specifically for supercoiled DNA. A study of a roughly 5,400 base-pair plasmid found that below 37°C, electrophoretic mobility increases steadily as supercoiling increases (that is, as the linking number decreases below the relaxed state). But above 37°C, mobility begins to oscillate in a cyclic pattern as supercoiling changes, and the amplitude of those cycles grows between 37°C and 65°C.8PubMed Central. Temperature dependence of the gel electrophoretic mobility of superhelical DNA Running temperature is one of those variables that many standard protocols gloss over, but it can materially shift where supercoiled bands appear relative to other forms.
Buffer Composition Matters More Than You Might Expect
Most molecular biology labs default to either TBE (Tris-borate-EDTA) or TAE (Tris-acetate-EDTA) buffer for gel electrophoresis without giving it much thought. But EDTA, the chelating agent in both buffers, can influence the relative migration of DNA topological forms in ways that catch people off guard. When EDTA was removed from TBE buffer, supercoiled DNA migrated significantly slower, shifting from running ahead of linear DNA to running behind it. Lowering the ionic strength further made this effect more pronounced. Restoring EDTA brought the normal pattern back, with supercoiled DNA running faster than linear. Interestingly, removing EDTA from TAE buffer did not produce the same reversal.9PubMed Central. Modification of gel architecture and TBE/TAE buffer composition to minimize heating during agarose gel electrophoresis
The ionic strength of the running buffer during electrophoresis also affects whether nicked circular DNA can enter the gel at all. The molecular weight threshold at which nicked circular DNA fails to migrate into the gel depends on ionic strength but, curiously, not on gel concentration.10Nucleic Acids Research. Physical properties and gel electrophoresis behavior of R12-derived plasmid DNAs So if you are working with large open circular molecules and they seem to be stuck in the well, adjusting your buffer’s salt concentration can be more effective than changing your agarose percentage.
The Ethidium Bromide Complication
Ethidium bromide is the most common stain used to visualize DNA on gels, and it also serves as a tool for manipulating DNA topology. But it does not treat all topological forms equally, and this creates both opportunities and pitfalls.
Ethidium bromide intercalates between base pairs, unwinding the helix and introducing positive supercoils. Nicked or linear DNA can absorb the most ethidium bromide because the torsional stress from intercalation is unconstrained: the molecule can simply swivel around its nick or rotate at its free ends. Among circular forms, negatively supercoiled DNA absorbs more than relaxed, and relaxed absorbs more than positively supercoiled. As a result, the same mass of nicked DNA fluoresces more brightly than negatively supercoiled DNA, which in turn fluoresces more brightly than relaxed or positively supercoiled DNA.11PubMed Central. Two-Dimensional Gel Electrophoresis to Resolve DNA Topoisomers
This differential fluorescence is a real problem for quantification. If you are trying to measure what fraction of your plasmid preparation is supercoiled versus nicked versus linear, simply comparing band intensities on an ethidium bromide-stained gel will overestimate the nicked and linear fractions relative to supercoiled DNA. Correcting for this requires either calibrating with known standards or using a stain that binds independently of topology.
Beyond staining, ethidium bromide actually changes DNA topology during electrophoresis if it is present in the running buffer. As it intercalates, it progressively removes negative supercoils, relaxes the molecule, and eventually introduces positive supercoils. This progressive unwinding is the basis of two-dimensional gel electrophoresis protocols designed specifically to resolve topoisomers. In the first dimension, DNA runs without ethidium bromide, separating largely by mass. In the second dimension, ethidium bromide is added to the gel and buffer, changing the effective supercoiling of each topoisomer and shifting their mobilities relative to one another.12Journal of Raman Spectroscopy. DNA tertiary structure and changes in DNA supercoiling upon interaction with ethidium bromide and gyrase monitored by UV resonance Raman spectroscopy
Using Topology Differences to Detect DNA Damage
One of the most practical applications of the different gel behaviors of circular and linear DNA is in radiation biology and toxicology. Covalently closed circular plasmids serve as sensitive reporters of DNA damage because each type of lesion converts the plasmid into a different topological form with a distinct gel mobility.
A single-strand break converts a supercoiled plasmid into a nicked circle, which runs slower on a standard gel. A double-strand break, whether direct or from the chance alignment of two single-strand breaks on opposite strands, converts the molecule into a linear form.13PubMed. Determination of G-values for single and double strand break induction in plasmid DNA using agarose gel electrophoresis and a curve-fitting procedure By running irradiated plasmid samples on an agarose gel and quantifying the fraction of DNA in each band, researchers can calculate the yields of single- and double-strand breaks from a given dose of radiation. This approach has been used with X-rays and also with low-energy ion beams, where nitrogen ion irradiation was shown to induce both types of breaks, visible as increased nicked circular and linear bands on the gel.14Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms. Electrophoresis examination of strand breaks in plasmid DNA induced by low-energy nitrogen ion irradiation
The simplicity of this assay is its main appeal. You do not need antibodies, fluorescent probes, or expensive sequencing. A plasmid, an agarose gel, and a UV transilluminator give you a direct readout of damage types. The main limitation is sensitivity at very low doses, where only a small fraction of molecules are converted, and the quantification issues with ethidium bromide staining described above.
Two-Dimensional Gels for Resolving Complex Mixtures
Standard one-dimensional gel electrophoresis can separate supercoiled, nicked, and linear forms reasonably well for small plasmids under the right conditions. But real biological samples often contain many more topological species than those three: catenanes (interlocked rings), knotted circles, replication intermediates, and molecules with varying degrees of supercoiling. One-dimensional gels collapse many of these into overlapping bands.
Two-dimensional agarose gel electrophoresis solves this by running a sample under one set of conditions in the first dimension and then under different conditions in the second. For studying DNA replication, the neutral/neutral 2D technique separates restriction-digested replication intermediates first by mass, then by shape. A gel slice from the first dimension is turned 90 degrees and run in a second gel at higher agarose concentration, which preferentially retards branched and circular intermediates relative to linear fragments. The result is an arc pattern where different replication structures, like simple forks, bubbles, and double forks, each trace a characteristic curve on the 2D gel.15PubMed. Two-dimensional agarose gel analysis of DNA replication intermediates
This technique has been adapted broadly. Plasmid replication studies have used it to visualize replication fork blockage, fork reversal, knotting that accumulates during replication, and the interplay between supercoiling and catenation as a circular molecule is copied.16PubMed. Plasmid DNA replication and topology as visualized by two-dimensional agarose gel electrophoresis A variant called 2D-IMAGE, developed for intact mitochondrial DNA, can resolve more than 25 distinct topoisomers from a single sample, including various catenated species and molecules containing single-stranded regions associated with replication or damage.17PubMed. Using Two-Dimensional Intact Mitochondrial DNA (mtDNA) Agarose Gel Electrophoresis (2D-IMAGE) to Detect Changes in Topology Associated with Mitochondrial Replication, Transcription, and Damage
Alkaline Gels and Single-Stranded DNA
Everything discussed so far concerns double-stranded DNA under native conditions. The picture changes again when you denature the DNA. Under alkaline conditions, hydrogen bonds between strands break, collapsing the double helix into single strands. For linear DNA, this gives you two free single strands. For a covalently closed circular molecule, the two strands remain topologically linked and cannot separate, forming a tangled single-stranded circle.
Alkaline agarose gel electrophoresis exploits this difference. Once secondary and tertiary structures are disrupted, linear and circular single-stranded DNA forms of the same molecule resolve clearly on the gel. The ratio of linear to circular mobility depends strongly on gel concentration and somewhat on DNA size.18PubMed. Separation and size determination of circular and linear single-stranded DNAs by alkaline agarose gel electrophoresis This makes alkaline gels useful for confirming whether a DNA molecule is truly covalently closed or just appears circular under native conditions because of base pairing that holds the strands together at a nick.
Capillary Electrophoresis as an Alternative
Traditional slab gels are fine for qualitative work, but they struggle with reproducibility and throughput when you need precise quantification of different plasmid forms, as in pharmaceutical manufacturing of DNA vaccines or gene therapy vectors. Capillary electrophoresis offers a faster, more quantitative alternative.
Using dilute polymer solutions like hydroxyethyl cellulose inside narrow capillaries, researchers have achieved separation of linear and supercoiled DNA in as little as three minutes for molecules in the range of 1 to 16 kilobase pairs. The key insight is that supercoiled plasmids, being stiffer, respond less to changes in electric field strength and buffer concentration than linear DNA does. That difference in sensitivity to running conditions can be exploited to maximize the resolution between the two forms.19Analytical Chemistry. High-Speed Separation of Linear and Supercoiled DNA by Capillary Electrophoresis. Buffer, Entangling Polymer, and Electric Field Effects Capillary methods have been validated for plasmids across a wide size range and produce quantitative results that agree well with conventional agarose gels.20PubMed. Electrophoretic separation of linear and supercoiled DNA in uncoated capillaries
Atomic force microscopy has also been tested as a complementary approach for quantifying the ratio of topological forms in a sample. When corrections are applied carefully, AFM imaging and gel electrophoresis give consistent results for the relative abundance of circular and linear species.21PubMed Central. Separating DNA with different topologies by atomic force microscopy in comparison with gel electrophoresis The main advantage of AFM is that you directly visualize individual molecules rather than inferring topology from mobility, but it is far slower and more labor-intensive than either gel or capillary methods.
Practical Troubleshooting for Common Gel Confusions
A few recurring problems trip people up when running plasmid preps on gels. The most common is misidentifying bands. If you see three bands from a supposedly pure plasmid prep, the fastest band under standard conditions is usually supercoiled, the middle band is linear (from double-strand breaks during purification), and the slowest is nicked circular. But if your gel concentration is high or your voltage is unusually high, supercoiled DNA may actually run slower than linear, flipping the expected pattern. Running a linearized control alongside your sample, using a restriction enzyme that cuts once, eliminates the guesswork.
Another common issue is the appearance of extra bands above the main supercoiled band. These are often multimeric forms: dimers or trimers of the plasmid that arise during replication. Catenanes, where two or more circular molecules are physically interlinked, can also appear. These run more slowly because of their larger effective size and can be confused with nicked circles or large linear contaminants.22PubMed. Quantitative analysis of plasmid forms by agarose and capillary gel electrophoresis Digesting with a single-cut restriction enzyme converts all forms to linear, so if an extra band collapses into the main linear band after digestion, you know it was a topological variant rather than a contaminating DNA of different sequence.
Finally, be cautious about quantifying band intensities directly from ethidium bromide-stained gels. Because different topological forms absorb different amounts of dye, the brightness of a band does not linearly correspond to the amount of DNA. Supercoiled DNA will appear dimmer per microgram than nicked DNA. For rough estimates this may not matter, but for quality control of therapeutic plasmid batches, the error can be significant enough to misclassify a preparation.