Predicting 260/280 of DNA or RNA from sequence?

Question

I asked this in biology, but I think they don't want to deal with the chemistry behind UV-Vis.

My question has to do with the Poly A tails on mRNA. I've been doing in vitro transcription and tailing for a while now and have noticed that the untailed RNA has a 260/280 of about 2.0, like you'd expect, but successful tailing results in ratios from 2.15 to 2.45, with higher ratios correlating to better band shifts on agarose gels, implying a longer tail. I suspect the reason for this is because the adenine has a much higher 260/280 than the other bases, so if you add a lot of adenines to the 3' end of the RNA you add much more 260 absorbance than 280 absorbance.

After seeing this I tried to come up with a mathematical method to estimate tail length based on 260/280, since gels are annoying, while I can get the ratio directly from the nanodrop.

I started by using the 260/280 for each nucleotide, A: 4.50, G: 1.15, C: 1.51, U: 4.00, T: 1.47 source here: 1. I then took the sequence for the Luciferase gene I'm using for mRNA and counted the nucleotides, 449 A, 566 C, 470 G, 374 U. I tried to get the weighted average of the 260/280 for my RNA.

A: 449 x 4.50 = 2020.50
C: 566 x 1.51 =  854.60
G: 470 x 1.15 =  540.50
U: 374 x 4.00 = 1496.00
Sum:            4911.60

Nucleotides: 449 + 566 + 470 + 374 = 1859
Sum / Nucleotides : 4911.60 / 1859 = 2.64

2.64 is clearly not 2.0. Even accounting for the potential effect of pH and ionic strength on RNA 260/280 ( acidic solutions have lower ratios ) I can't explain why the calculated ratio is so high. Even if we assumed an RNA with 25% of each base, the weighted average is still (1.15 + 4.50 + 1.51 + 4.00)/4 = 2.79, even higher than for my real sequence.

Doing the same calculation for DNA, (1.15 + 4.50 + 1.51 + 1.47)/4 = 2.16.

So why don't my calculated ratios for RNA and DNA 260/280 match the "ideal" ratios of 2.0 and 1.8? It probably has something to do with base stacking interactions, but I don't enough to compensate for it.

Answer 1

You could try a different pH to get rid of secondary structure. It might change the absorbance spectra as well, though. Here is a paper where they measured spectra of DNA and its constituents: https://doi.org/10.1016/S0022-2836(61)80003-X

The abstract says the following (mμ is the same as nm):

The spectra of deoxyadenylic, deoxyguanylic, deoxycytidylic and thymidylic acids are given at pH 1, 2, 3, 5 and 7, from 220 to 310 mμ. The spectrum of thymus DNA denatured in 0·l N-acetic acid at pH3, at low concentration, is almost identical to the calculated spectrum of its constituent nucleotides.

The ratio of the absorbancies of DNA at 260 and 280 mμ reaches a constant value near pH 3, characteristic of the molar proportions of the bases. This appears to form the basis of a rapid method for estimating base composition.

Answer 2

The problem with doing this, especially with RNA, is that secondary structure can have a large impact on the absorption properties of the molecule. Interactions between bases will change the molar absorptivities in ways that may not be easily predictable. The "ideal" ratios are only guidelines and I wouldn't worry too much that a long string of a single base produces a deviation from them. If very accurate measurements are required, hydrolysis into nucleotides before measurement is usually done to avoid these interactions. For your specific case, it may be possible to construct a calibration curve of sorts to get an estimate if you can measure the tail length by some other means. So long as the tail is sort relative to the rest of the molecule, I would expect to see some predictable change in ratios.