The traditional method for quantification and assessment of purity of DNA samples is spectrophotometric measurement at 260 and 280nm (1). It is generally accepted that a sample of pure, double-stranded DNA of 50 g/mL will have an absorbance at 260nm of 1.0 and that the ratio of absorbance at 260 and 280nm of this sample will be greater than 1.8 (2). If a sample has an A260 / A280 of less than 1.8, it is usually considered to be contaminated by protein. In such measurements, there is frequently little consideration given to the resolution or bandwidth of the spectrophotometer used for the measurements.
Instrument bandwidth is generally defined as the full width at half height of an absorbance band of a reference material that possesses a natural bandwidth less than or equal to the instrument bandwidth. The ideal reference material would be one with many absorbance bands, each having an infinitely small natural bandwidth. For this reason, the atomic lines of elements such as Hg are used as reference materials for determining instrument bandwidth.
Each sample will have a natural or Spectral Bandwidth and, thus, the ability of an instrument to accurately quantify components in a mixture will depend upon several factors, including natural bandwidths of the components and the instrument bandwidth. A widely accepted practice is to use a spectrophotometer with an instrument bandwidth of one tenth of the natural bandwidth of the analyte to be measured, 43 nm in the case of DNA. Thus, following the accepted practice, a spectrophotometer with an instrument bandwidth of = 4.3nm should be used to measure such samples. In the past few years there has been an increasing trend within life sciences toward the use of spectrophotometers with instrument bandwidths of 5nm or greater for quantification of DNA and for estimation of DNA purity. This communication addresses the effects of instrument bandwidth on spectrophotometric determination of DNA concentration and purity.
MATERIALS AND METHODS
Spectrophotometers used in this study were the Shimadzu models UV-2401PC (selectable bandwidth to 0.1nm), UV-1700 (fixed bandwidth, 1nm), and the UV-1240 (fixed bandwidth ,5nm). Calf thymus DNA and Bovine serum albumin were obtained from Sigma Chemical Company, St. Louis, MO. All other reagents were obtained from commercial sources and were of the highest available purity. Samples were prepared in 5mM KH2PO4 containing 145mM NaCl, pH 7.48. Semi-micro, far-UV quartz cells (~1.5mL) were used for all measurements.
RESULTS AND DISCUSSION
An example of the spectrophotometric determination of DNA is illustrated in Figure 1. In this figure, a sample of approximately 73 g/mL calf thymus DNA in physiological buffer was measured using a Shimadzu UV-2401 spectrophotometer with an instrument bandwidth of 1.0nm. As illustrated in Figure 1, this sample of pure DNA has an absorbance maximum (1.47) and minimum (0.62) at approximately 260 and 230nm, respectively. The width at half-height of this spectrum is approximately 43 nm.The spectrum of this same DNA sample was measured using instruments of 1 and 5nm fixed-instrument bandwidth. These data are shown in Figure 2.
The data in Figure 2 indicate that there is very little difference in the spectrum of pure, double-stranded DNA measured using spectrophotometers of 1 and 5nm instrument bandwidth. The absorbance of the sample at 260nm and the A260/A280 ratios are comparable and well within experimental error. A similar comparison of spectra of 1mg/mL Bovine serum albumin is shown in Figure 3.
The spectra in Figure 3 have been offset by 0.1A in order to show the enhanced details in the spectrum of serum albumin measured in the 1nm bandwidth instrument. With enhanced resolution, the contributions from phenylalanine, tyrosine, and tryptophan become more obvious. These residues absorb between 260 and 280nm. However, as was the case for the sample of pure DNA, there was little difference in the measured absorbance (~0.6) and wavelength maximum (~278nm) of the protein sample in the two instruments.
A much different situation applies when DNA purity is assessed. In this situation, the instrument must discriminate between DNA and likely contaminants. These contaminants include protein, chloroform, phenol. Contamination by protein is usually indicated by a decrease in the A260/A280 ratio. Contamination by an extraction solvent such as phenol or chloroform is indicated by increased absorbance at 320nm. If the criterion for acceptable resolution of instrument bandwidth ten-fold less than peak wavelength difference (280-260nm = 20nm) is applied to samples containing DNA and protein, then an instrument with a bandwidth of = 2nm should be used. In this case, there are demonstrative advantages in using a spectrophotometer with a bandwidth of 1nm. The spectra of a sample containing 36g/mL DNA and 500 g/mL serum albumin are shown in Figure 4.
It is clear from these data that the A260/A280 ratios of the sample containing both DNA and protein measured in the 5nm bandpass instrument (1.11) and in the 1nm bandpass instrument (1.02) are not equivalent. While both ratios are significantly less than the value of 1.8 expected for pure DNA, this was a sample heavily weighted with protein (~93% by weight). A more interesting question is the lowest level of protein in a sample that can be determined by this technique.
A series of samples with varying ratios of DNA and BSA were prepared in physiological buffer, and A260/A280 ratios for these samples were measured in spectrophotometers with 1 and 5nm bandwidth. The results are presented in Table 1.
The 1nm bandpass instrument was capable of discriminating a decrease in the A260/A280 ratio of DNA samples containing 10% by weight BSA, or contamination on the order of 10 g. This capability is more readily appreciated by examination of the data in Figure 5.
There appears to be little discernable change in the A260/A280 ratio of DNA-Protein mixtures measured using the 5nm bandpass instrument until approximately 20% by weight BSA or approximately twice the minimum level detected by the 1nm bandpass instrument.
Both 1nm and 5nm bandpass spectrophotometers are appropriate for measurement of pure DNA samples. But for determination of DNA purity, particularly at very low (~10g) amounts of protein, there are significant advantages to the use of a 1nm bandpass spectrophotometer.
1.Glasel, J.A., Validity of Nucleic Acid Purities Monitored by A260/A280 Absorbance Ratios, Biotechniques 18:62-63, 1995.
2. Maniatis T., E.F. Fritsch, and J. Sambrook, Molecular Cloning: A Laboratory Manual , Cold Spring Harbor Laboratory, Cold Springs Harbor, NY, 1982