Purity of DNA (Ratio of OD = 1.8-2.0) ?

Andre Hamel hamel at ccu.umanitoba.ca
Tue Jan 12 21:26:32 EST 1993

In article <1993Jan12.082445.74 at ntet.nctr.fda.gov> wmelchior at ntet.nctr.fda.gov writes:
>In article <1993Jan12.061103.2973 at kudpc.kyoto-u.ac.jp>, 
>a52041 at sakura.kudpc.kyoto-u.ac.jp (Michael CHENG) writes in part:
>>   Why is a preparation of DNA considered to be pure if OD260/OD280 => 1.8?
>> I heard that there's possibly a theoretical formula that sets the above 
>> value range.
>I don't know about a "theoretical formula".  I suspect it just comes from an 
>empirical study of the spectra of different types of molecules.
>Protein tends to have a higher absorbance at 280 nm (due mainly to 
>tyrosine and tryptophane) than at 260.  This means that the A260/A280 of 
>proteins is LOWER than that of DNA.  RNA, on the other hand, has a higher 
>absorbance at 260 than DNA, leading to LOWER values of A260/A280 and 
>A260/A230.  (So too HIGH a value for A260/A280 also indicates a problem.) 
>Phenol, a common contaminant, absorbs more at 280 than at 260, leading to a
>decreased A260/A280. 
>Note that the spectrum of DNA depends not only on its helical state, but 
>also on its base composition.  The most complete reference I'm aware of is 
>old but worth looking at:  G. Felsenfeld & S.Z. Hirschman, J. Mol. Biol. 
>(1965) 13, 407-427.
>There must be something more recent on RNA (help, anyone?), but I still 
>refer to a figure in an article by E. Chargaff in vol 1 of _The Nucleic 
>Acids_ by Chargaff and Davidson, Academic Press, 1955, from which I 
>estimate A260/A280 = 2.09 and A260/A230 = 2.76; this would, of course, 
>also depend on conformation and base composition.
>(If you add measurements at 230 nm to your studies, be aware that such 
>measurements are much more prone to errors due to contamination from a 
>variety of sources, and to scattered light problems with less expensive 
>spectrophometers.  For double-stranded DNA with 42% GC, Felsenfeld and
>Hirschman give A260/A230 = 2.38.) 
>Bill Melchior                               

Bill's response is, in my opinion (and from my experience in purifiying
nucleic acids from a VERY WIDE variety of biological sources) the most
complete and informative. May I also add that the absorbance measured at
230 nm is also VERY useful in addition to the 260 and 280 values because,
in addition to "other stuff" being detected,  ALL PROTEINS should absorb
"strongly" at 230 nm due to the presence of peptide bonds (amino ester
linkage if my memory offirst year chemistry still holds true). Is this not
why, for instance, many HPLC protein purification systems employ 190- 210 nm
wavelength light to more sensitively detect the outcoming protein? Those
proteins with aromatic prosthetic groups will also exhibit additional
wavelengths of absorbance, around 280 nm. 

	Budget permitting, ideally a scanning spectrophotometer would be used as
opposed to simply taking only 230, 260 and 280 nm measurements. One then
would use the highest absorbance value (as long as it's around 260 nm,
plus/minus about 10 nm). This maximum absorbance (peak) should ideally be
about twice as high in value than the values near 230 and 280 nm. 

Andre Hamel
Manitoba Veterinary Services
Virology Section
Winnipeg, Manitoba
R3T 5S6

email: hamel at ccu.umanitoba.ca
voice: (204) 948-2272

More information about the Methods mailing list