You’ve run your plasmid prep on a gel, and you’re staring at two, three, or five bands where you expected one. Here’s how to identify each common form of uncut plasmid DNA on an agarose gel, assess whether your prep is good enough for downstream work, and fix the ones that aren’t.
Uncut plasmid DNA exists in several conformations, and each migrates differently through agarose. In a good prep from a standard high-copy plasmid, the dominant form should be supercoiled (Form I) — though the exact ratio varies with backbone, host strain, growth conditions, and storage. Nicked circles, linear molecules, denatured single strands, and dimers can all appear alongside it. Which bands you see, and how bright they are relative to each other, tells you how your prep went and what to do next.
Everything below is built around what goes wrong in practice, not just what the textbook says. The band identifier tool suggests forms consistent with your band pattern. The troubleshooting section tells you what caused it and how to fix it.
Plasmid DNA forms: quick reference
The table below covers the five forms you’re most likely to see when running uncut plasmid DNA on a standard agarose gel (0.7–1.0%, TAE or TBE, no ethidium bromide in the gel).
Choose a free resource to help you move forward
PROTOCOL
Chemically Competent Cells Protocol
DOWNLOAD
Nanodrop Results Interpretation Card
| Supercoiled monomer | Form I / CCC | Fastest double-stranded monomer form under standard plasmid QC conditions (runs well below expected size on linear ladder) | Brightest band; should be dominant | Native conformation. Both strands intact, molecule under superhelical tension. |
| Linear monomer | Form III | Runs at true MW (between supercoiled and nicked) | Co-migrates with linearised control | Double-strand break. Nuclease contamination or mechanical shearing. |
| Nicked / open circular | Form II / OC | Slowest monomer form (runs above expected size) | Faint band above linear position | Single-strand nick releases superhelical tension. UV, nucleases, or handling damage. |
| Denatured ssDNA circles | Form IV | Often runs ahead of supercoiled under standard gel conditions — fastest of the common forms in typical miniprep QC gels | Band below the supercoiled band | Overly harsh alkaline lysis permanently denatures the duplex. |
| Dimer / multimer | Plasmid multimer | May appear in the high-molecular-weight region, often at or above where a 2× linear species would migrate — but exact position depends on topology, gel %, buffer, and voltage | Band(s) well above nicked position | Possible plasmid multimer or (if diffuse, near well) genomic DNA contamination — confirm by restriction digest |
Linear and nicked/open circular forms can co-migrate under certain gel conditions. The most practical routine way to distinguish them is to run a linearised control (single restriction enzyme digest) alongside your uncut plasmid.
Identify your bands
You have your gel image and a DNA ladder. Enter your expected plasmid size, your gel conditions (agarose percentage and whether EtBr was in the gel), and then add each band you see — its approximate size from the ladder and its relative brightness (faint, clear, or bright). The tool gives a best-fit interpretation based on the ratios and intensities you enter. It is a diagnostic starting point, not a definitive molecular verdict — gel electrophoresis alone cannot confirm purity, rule out all contaminants, or guarantee performance in your specific application. Use it alongside the troubleshooting sections below.
Plasmid Gel Band Identifier
Describe what you see on your uncut-plasmid gel and find out which structural forms are present — and what to do about the unwanted ones.
Before you start
- Run a linearised control. Digest an aliquot of your plasmid with a single-cutter restriction enzyme and load it alongside the uncut sample. This is the most practical routine way to identify the linear band and distinguish it from nicked circle. Note that supercoiled DNA can also complicate restriction digests — see why supercoiling can derail your cloning if you’re seeing unexpected uncut bands after digestion.
- Note your buffer system. Relative band positions differ between TAE and TBE gels. Do not compare gels run in different buffers.
- Use 0.7–1.0% agarose for plasmid QC. Higher percentages compress the separation between conformational forms and make identification harder. For a broader overview of what to avoid when running agarose gels, see 5 ways to destroy your agarose gel.
- Keep loading amounts moderate. As a general guide, 50–200 ng per lane is a useful starting range for plasmid QC — overloading produces smeared, overlapping bands, while underloading makes faint nicked or linear bands invisible. Adjust to your specific gel system.
- If your gel contains ethidium bromide, expect shifted migration. EtBr intercalates into DNA and partially unwinds supercoils. In pre-stained gels, the supercoiled band may run closer to the linear position than expected. Post-stain for the most reliable migration pattern.
The five forms of uncut plasmid DNA
Supercoiled (Form I)
Supercoiled DNA is the native conformation found in vivo. Extra twists are introduced into the double helix, and because the ends of a plasmid are joined into a covalently closed circle, the tension cannot be relieved — think of an over-twisted rubber band coiling back on itself. The compact shape means it migrates faster through agarose than its true size would predict, so under standard gel conditions it runs below its expected position on a linear DNA ladder.
For most standard plasmids, supercoiled is the dominant form and the one you want. Whether a prep with a given supercoiled fraction is sufficient depends on your application — cloning is generally more forgiving than transfection.
Nicked / open circular (Form II)
A single-strand nick releases all superhelical tension and lets the molecule relax into a large, floppy circle. This happens through trace nucleases, UV exposure, or mechanical shearing during purification. Because it is a large, uncompacted structure, it runs slower than any other monomer form on the gel — above the position predicted by the ladder.
A faint nicked band is commonly seen in routine minipreps and often acceptable for most applications. A prominent one means your DNA took damage during purification.
Linear (Form III)
Linear plasmid DNA results from a double-strand break anywhere in the molecule. On a gel, it migrates at its true molecular weight, between the supercoiled and nicked positions. If you see a linear band in an uncut prep, the most likely cause is nuclease contamination in your buffers. Mechanical shearing and trace restriction enzyme contamination are also possible contributors, though routine vortexing is more classically associated with genomic DNA shearing than with plasmid linearisation.
To confirm you’re looking at linear DNA and not a co-migrating nicked circle, compare with a linearised control. The linear band from your uncut prep should run at exactly the same position as the single-enzyme digest.
Denatured single-stranded circles (Form IV)
During alkaline lysis, plasmids are denatured because the hydrogen bonds between strands are disrupted by the alkaline conditions. Normally, when the pH is returned to neutral, the covalently closed circular strands re-anneal and the supercoiled form is restored. But if the lysis step runs too long, the denaturation becomes permanent. The QIAGEN Knowledge Hub explicitly states that lysis should not proceed for longer than 5 minutes, noting that longer exposure causes irreversible plasmid denaturation. The resulting single-stranded closed circles often migrate ahead of the other common plasmid forms under standard miniprep QC gel conditions.
This is not a form you want in your prep. Denatured ssDNA circles often perform poorly in restriction digestion, and transformation and transfection efficiency is typically much reduced. Unless only a faint trace band is present, it is usually safest to re-prep rather than proceed.
Dimers and multimers
Sometimes you’ll see one or more bands well above the nicked/OC position, at approximately 2× (or higher multiples of) the expected plasmid size. These are dimers or multimers: two or more copies of the plasmid joined into a single circular molecule during replication in the bacterial host.
Dimers can be seen with some high-copy plasmids and certain backbones. They can reduce transfection efficiency and produce confusing restriction digest patterns. To confirm a suspected dimer: run a restriction digest with a single-cutter enzyme. Because a tandem dimer contains two copies of the restriction site, the enzyme cuts twice and produces monomer-sized linear fragments that co-migrate at the monomer position, usually with increased band intensity. This means the digest pattern can differ from the uncut lane but may not produce a separate 2× band.
What your gel tells you about your prep
Band identity tells you what forms are present. Band intensity tells you how much. Together they give you a prep quality assessment — but what counts as “good enough” depends on what you’re doing with the DNA. Cloning and sequencing are tolerant of minor nicking or a faint linear band. Transfection is generally more sensitive to non-supercoiled forms, though the degree varies by transfection method, reagent, and cell type. In some systems, open-circular and linear DNA transfect less efficiently than supercoiled, and higher proportions of damaged DNA have been associated with reduced expression and increased cell toxicity — but this is method- and cell-type-dependent. In vitro transcription is generally also sensitive to template integrity, though requirements vary by workflow — circular plasmid templates, linearised templates, and PCR-product templates each have different constraints; consult your kit protocol. For a broader QC workflow after this gel check, see three important things to check after obtaining your plasmid.
| What you see | Cloning / sequencing | Transfection / in vitro applications |
|---|---|---|
| Single dominant supercoiled band, faint or absent nicked band | Proceed | Proceed |
| Supercoiled dominant, visible nicked band (faint) | Proceed | Acceptable — check efficiency; re-prep if low |
| Supercoiled dominant, clear nicked band (similar intensity to SC) | Acceptable with care | Re-prep recommended |
| Supercoiled and linear bands of similar intensity | Investigate first — nuclease contamination will worsen during storage | Re-prep |
| Prominent nicked band, weak supercoiled | Re-prep | Re-prep |
| Fast band below supercoiled (denatured ssDNA) | Redo prep | Redo prep |
| Band in the high-molecular-weight region (possible dimer — confirm by restriction digest) | Acceptable for sequencing; may cause extra bands in digest | Investigate — dimers reduce transfection efficiency |
| Smear, no discrete bands | Redo prep | Redo prep |
My result looks wrong
The quality table above tells you whether to proceed. This table tells you what went wrong and how to fix it. Each row starts from a specific symptom you can see on the gel. If your prep repeatedly fails QC and you’re also seeing low DNA yield, see 11 reasons your plasmid yield is low.
| Symptom | Most likely cause | Fix |
|---|---|---|
| Band running ahead of (faster than) supercoiled | Permanently denatured ssDNA from over-incubation of the alkaline lysis step (P2 buffer > 5 min) | Redo the prep. Keep P2 incubation to 4–4.5 minutes. Mix by gentle inversion only. |
| Prominent linear band in uncut lane | Nuclease contamination in lysis or wash buffers, or mechanical shearing from vortexing | Prepare fresh buffers with nuclease-free water. Stop vortexing. Minimise freeze-thaw cycles. |
| Nicked band brighter than or equal to supercoiled | Nicking from trace nucleases, UV exposure during handling, or aged/contaminated reagents | Avoid UV transilluminator exposure. Use fresh buffers. Handle gently. If using columns, check for degradation. |
| Unexpected band at ~2× plasmid size | Plasmid dimer formed during replication in the host. More common in high-copy backbones. | Confirm by restriction digest with a single-cutter enzyme. A tandem dimer contains two copies of the restriction site, so the enzyme cuts twice and produces monomer-sized linear fragments that co-migrate at the monomer position — usually visible as a brighter monomer-sized band rather than a separate 2× band. Restreak bacterial stock, pick a single colony, re-grow, and re-prep. Consider a recA⁻ host strain if dimers persist. |
| No supercoiled band — only nicked and/or linear forms visible | Complete loss of supercoiling from extensive nuclease damage or prolonged incubation at room temperature | Redo the prep with all fresh reagents. Work on ice where possible. Elute in TE rather than water to chelate trace metals that activate nucleases. |
| Diffuse smear with no discrete bands | Severe nuclease degradation — plasmid fragmented into random-length pieces | Redo with fresh, nuclease-free reagents. Check expiry dates on kit buffers. Use nuclease-free water and clean glassware. |
What the protocol doesn’t tell you
Your gel bands are not pure populations. Even the supercoiled band contains some relaxed and linear molecules. An AFM study of pUC18 found that only about 77% of DNA extracted from the supercoiled gel band was actually supercoiled. The rest was relaxed (16.5%) or linear (6.4%). Gel interpretation gives you an estimate of your prep composition, not an absolute measurement. Treat it as a QC check, not a purity assay.
Ethidium bromide changes the rules. If you cast EtBr into the gel (pre-staining), it intercalates into the DNA and partially unwinds supercoils. Johnson and Grossman (1977) document this effect: the apparent migration of topological forms shifts with intercalating dye concentration. In practice, the supercoiled band may run at a position closer to linear, or even slower than linear at high EtBr concentrations. If you’re trying to assess conformation, post-stain your gel instead.
TAE and TBE gels give different band positions. The relative migration of supercoiled, linear, and nicked forms shifts between buffer systems. A band that sits clearly between supercoiled and nicked in TAE may sit in a different position in TBE. Never compare plasmid QC gels run in different buffers, and note which buffer you used so you can reproduce the result.
Common mistakes
| Mistake | How to spot it | How to prevent it |
|---|---|---|
| Estimating plasmid size from the supercoiled band | Size estimate doesn’t match the known sequence | Supercoiled DNA migrates faster than its true size. Use the linearised control or a supercoiled DNA ladder for size estimation. |
| Confusing linear with nicked/open circular | Band assignment changes between preps or doesn’t match restriction digest | Always run a linearised control alongside uncut plasmid. |
| Vortexing during the lysis step | Excess linear and nicked bands appear; possible genomic DNA contamination (smear at top of gel) | Mix by gentle inversion (6–8 times). Never vortex after adding P2 or neutralisation buffer. |
| Letting alkaline lysis run too long | Fast-migrating band below supercoiled (denatured ssDNA) | Time the P2 step. Keep it at 4–5 minutes maximum. Set a timer. |
| Overloading the gel | Bands are broad, smeared, or overlapping | Load 50–200 ng per lane for a clean QC gel. Less is better for conformation analysis. |
| Comparing gels run in different buffer systems | Band positions seem inconsistent between experiments | Standardise on one buffer (TAE or TBE) and gel percentage for all plasmid QC runs. |
A gel tells you a lot, but not everything
Band pattern and intensity give you a useful indicator of prep quality for the most common problems: denaturation, nicking, nuclease contamination, and dimerisation. What a gel cannot tell you is whether the prep is free of RNA, endotoxin, residual salt, or protein — all of which matter for sensitive applications. Treat gel QC as the first checkpoint, not the last. A prep with a clean, dominant supercoiled band and no obvious damage is likely good. Whether it is good enough for your specific experiment depends on the application, and sometimes on a spectrophotometer reading alongside it — see determining DNA concentration and purity for what A260/A280 and A260/A230 ratios can and can’t tell you.
If your prep fails the gel check, the troubleshooting table above covers the most common causes. For persistent problems with low supercoiled yield, see 6 ways to get more supercoiled plasmid from your preps.
References
- Johnson PH, Grossman LI. Electrophoresis of DNA in agarose gels. Optimizing separations of conformational isomers of double- and single-stranded DNAs. Biochemistry. 1977;16(19):4217–4225. doi: 10.1021/bi00638a014. PMID: 332225.
- Jiang Y, Rabbi M, Mieczkowski PA, Marszalek PE. Separating DNA with different topologies by atomic force microscopy in comparison with gel electrophoresis. J Phys Chem B. 2010;114(37):12162–12165. doi: 10.1021/jp105603k.
- Schleef M, ed. Plasmids for Therapy and Vaccination. Wiley; 2001. ISBN 978-3-527-30269-3.
- Remaut K, Sanders NN, Fayazpour F, Demeester J, De Smedt SC. Influence of plasmid DNA topology on the transfection properties of DOTAP/DOPE lipoplexes. J Control Release. 2006;115(3):335–343. doi: 10.1016/j.jconrel.2006.08.009. PMID: 17010468.
- Azzoni AR, Ribeiro SC, Monteiro GA, Prazeres DMF. The impact of polyadenylation signals on plasmid nuclease-resistance and transgene expression. J Gene Med. 2007;9(5):392–402. doi: 10.1002/jgm.1031.
- Ledley FD. Pharmaceutical approach to somatic gene therapy. Pharm Res. 1996;13(11):1595–1614. doi: 10.1023/A:1016420102549.
- QIAGEN. Lysis of bacterial cells for plasmid purification. QIAGEN Knowledge Hub. Accessed June 2026.
This blog was originally written on the 21st of April 2026, and was updated on the 29th of June 2026 for accuracy and clarity.
You made it to the end—nice work! If you’re the kind of scientist who likes figuring things out without wasting half a day on trial and error, you’ll love our newsletter. Get 3 quick reads a week, packed with hard-won lab wisdom. Join FREE here.
