This post is about quantifying DNA samples in the Rieseberg lab, i.e. using the tools we have available to us in our lab in the Biodiversity building as of November 2011.
There are three ways to quantify the amount of DNA in an aqueous solution (e.g. DNA dissolved in water or TE):
- Nanodrop (spectrophotometry).
- Qubit (flourometry).
- Agarose gel with EtBr (also fluorometry -> UV lightbox + your eyes = fluorometer).
- (There is also the BioAnalyzer – I’m going to ignore that here and focus on methods to measure high mw genomic DNA).
These methods have different benefits and limitations but they are all valid and useful.
How they work
Nanodrop: The machine makes a tiny liquid column of your sample between two ends of an optical fibre and shines light through it. It measures the absorbance of light by wavelength. It calculates the concentration of DNA in the sample with an equation that takes the absorbance at 260nm as the key parameter.
Qubit: You mix a small amount of your sample with a dye that intercalates into double-stranded DNA. That mixture goes into the machine where it is exposed to light that excites the dye causing it to fluoresce. The key thing is that the dye only fluoresces if it is bound to DNA. The machine measures the fluorescence and compares it to the fluorescence it has measured from a weak concentration standard and a strong one. As long as the fluorescence of your sample falls between these two standards the machine can calculate the concentration of your sample.
Agarose gel: You run a volume of your DNA solution out on an agarose gel, that contains ethidium bromide (EtBr), alongside the same volume of a range of concentration standards. Your DNA electrophoreses through the gel accumulating EtBr molecules, which intercalate into the double-stranded DNA molecules, You then visualise the DNA by exposing the gel to UV light, which causes the EtBr to fluoresce (whether its bound to your DNA or not). The fluorescence of EtBr stained DNA is linearly related to the amount of DNA in the gel so you can simply compare the fluorescence of your sample to that of the concentration standards. If your sample is twice as bright as a 100ng/ul concentration standard, for example, then the concentration of your standard is 200ng/ul, (only if you loaded the same volume of the standard and your sample).
Range of Measurement
Nanodrop: 2ng/ul – 3700ng/ul.
Qubit: 10pg/ul – 1000ng/ul. (Combining the measurement ranges of two kits – HS and BR).
Agarose+EtBr: this comes down to what you or, more importantly your camera, can detect and resolve. Roughly speaking, its easy to visualize DNA at 2ng/ul and you can distinguish significant differences in concentration up to reasonably high concentrations – in the order of 500-1000ng/ul.
Note that the upper measurement limits on the above are not really important – you can always dilute DNA samples down to concentrations that are accurately measurable.
Also note that the lower detection limit on the agarose gel method is somewhat flexible – you can always load more of your DNA sample.
Nanodrop: we mostly use the Adams lab single-channel machine in the support room where our fridge is in the 3rd floor lab. There is also an 8-channel machine elsewhere in the Biodiv building.
Qubit: The Rieseberg Lab Qubit is on a bench in the 3rd floor lab. The buffer and dye are stored at room temperature in a cupboard below the machine and the concentration standards are in the fridge.
Agarose: There is an obvious gel electrophoresis area in one of the support rooms in the 3rd floor lab.
- Quick – much less than one minute per sample. There is no prep time – you just measure your sample directly.
- Cheap – if you take the purchase price of the machine and the, relatively insignificant, maintenance costs out of the equation you can think of it as almost free.
- Easy – it is.
- Quantity and quality – gives valuable indicators of DNA quality. The absorbances at 230nm and 280nm. See below for more on this.
- You don’t use much of your sample. The assay requires 1-2ul of your DNA sample, which is generally not recovered. I use 1.5ul.
- No pipetting error – the Nanodrop assay is volume independent which means that it doesn’t matter if you load 1ul or 2ul and that pipetting error does not apply.
- Accuracy/Specificity – the estimation of DNA concentration is based on an empirically determined linear curve of fluorescence made with concentration standards and the fluorescence of a DNA specific dye. Its not based on an absorbance that might be influenced by contaminants or an equation that takes average values for key parameters (i.e. the Nanodrop). Note that the assay is specific to double-stranded DNA – this is very important. See below for more on this.
- Sensitivity – the Qubit is capable of measuring very low DNA concentrations, see above.
- You can use as little as 1ul of your DNA sample, you can use as much as 20ul, none of which is recoverable.
- Easy – it is.
- Accuracy/Specificity – see the above, the Qubit is a fancy machine version of this method. If you diluted your samples sufficiently, as per the Qubit, your eye would probably be almost as good as the machine at estimating the relative fluorescence of your samples and the concentration standards. EtBr fluorescence is also specific to double-stranded DNA.
- Standard – running agarose gels is a standard procedure. Its run of the mill and universal. The gear required is probably available in every lab everywhere.
- Quantity and quality – the great thing about running your DNA out on a gel is that you get to see it assayed for size. Size is a really useful indicator of DNA quality. Usually you know what bp size you are expecting for your DNA – so seeing the actual size can be very educational. For example, if you are running out whole genomic DNA you don’t want to see anything but a tight high molecular weight band on the gel. If you see lower mw DNA there is something wrong – you have mechanically, chemically or (worst case) enzymatically damaged some or all of your DNA. If you have a lot of RNA in your DNA sample that will also be visible.
- It doesn’t use up much of your DNA – I routinely run out just 2ul of a 200ul DNA sample.
- Contaminants can mislead you. Anything in your sample that is not DNA but that absorbs at 260nm will inflate your estimated DNA concentration.
- The above is, generally speaking, not a serious issue. However there is a “contaminant” that is commonly present in DNA samples and that absorbs at 260nm – single stranded DNA and/or nucleotides. It is this absorbance that is most likely to inflate your estimates – if you consider the amount of double-stranded DNA to be the thing you are estimating.
- Note that “contamination” by single stranded DNA and nucleotides is not a problem for the two fluorescence based methods.
- Relatively expensive – you have to use the manufacturer’s reagents and “special” tubes.
- Subject to pipetting error. This may be important! There is a dilution step involved where you aliquot 2-20ul of your sample into the buffer-dye mix this step, and the dilution of the dye in the buffer both introduce pipetting error into the process. This can matter due to the sensitivity of the Qubit assay – if you are trying to measure precisely and accurately, especially at low concentrations, pipetting error might really matter.
- A bit more hassle than the Nanodrop – you dilute your samples then measure that dilution then convert up to your actual DNA sample concentration (the machine can do the arithmetic).
- You have to have some idea of your DNA sample’s concentration before you start – you have to choose between “high sensitivity” and “broad range” reagents.
EtBr Agarose Gel
- Time and hassle – you have to make a gel, prepare your samples and the standards then load, run and visualise the gel.
- EtBr – EtBr is a mutagen and suspected carcinogen. Its easy to handle EtBr and gels stained with it safely.
- Subjectivity – you are relying on your ability to compare fluorescence. This can be done quite accurately but its still subjective and beginners might be inaccurate.
What I do
For genomic DNA extractions, I assay them on the Nanodrop (1.5ul), and record the ng/ul estimate, the 260/230 and 260/280nm absorbance rations, and I run 2ul out on an EtBr-Agarose gel against a selection of lambda DNA concentration standards. This is a really good, and cheap, combination of quantification methods that also gives a lot of quality information.
I use the Qubit for small numbers of samples where I need high precision and/or estimates for very low DNA concentrations.
Usually, when you quantify a DNA sample you also want to know if its good. The three quantification methods provide different information about the quality of your sample.
The Nanodrop measures the absorbance at 260nm to estimate the amount of DNA in the sample but it also measure absorbances at wavelengths that indicate common contaminants – at 230nm and 280nm. It gives these absorbances in the form of a ratio of the 260nm absorbance. The higher these ratios are the better. A DNA sample that has a 260nm/230nm ratio of 2.0 and a 260nm/280nm ratio of 1.8 can be considered pure – its a good DNA sample. Ratios that are higher than 2.0 and 1.8 are good, lower is bad.
The most important absorbance ratio is the 260nm/280nm. This one is an estimate of protein (or phenol) contamination and proteins can cause problems – enzymes are proteins. A sample with a 260nm/280nm ratio of less than 1.7 is a worry.
Take a look at the Nanodrop specs for two batches of DNA that I linked to in my DNA Extraction – CTAB post to get some idea of the usual sorts of values.
The Qubit does not offer much in the way of DNA quality specs but the one thing it can tell you is sometimes vital information. The Qubit assay is specific for double-stranded DNA. Sometimes you really want to know how much double-stranded DNA you have, or simply whether you have any at all. In such cases the Qubit is very handy. (But note that an EtBr stained agarose gel also tells you this – and more). In a nutshell, if you ever want to know if a DNA solution contains double-stranded DNA and, especially if you want to know this about a low concentration DNA solution that you can’t afford to waste much of, the Qubit is the way to go.
You can learn three possibly important things about the quality of your DNA sample by running it out on agarose: the size, or range of sizes, of your DNA molecules, that at least some of it is double-stranded, and if there is a significant amount of RNA in it. For a genomic DNA sample, for example, you expect only high mw DNA – so a smear or band of low mw DNA would be bad; you expect to see something – you will only see double-stranded DNA; and you probably don’t expect to see a lot of RNA – RNA will appear on a gel as two fuzzy “bands” down at very low mw DNA levels. If you are running concentration standards, for example commercial lambda standards, you can compare the low mw fluorescence of those with that of your DNA standards.
Here is the single-channel Nanodrop manual: Nanodrop1000_UserManual.
Here is the Qubit manual: Qubit-2-Fluorometer-User-Manual.
Dan E. November 2011.