Ultraviolet Photodissociation Spectroscopy of [dAMP-H]- at Low Temperature

Abstract

Nucleotide fragmentation after photoexcitation in the ultraviolet is a potential cause for damage to DNA strands. Consequently, the fragmentation process needs to be explored to understand the stability of nucleotides on a molecular level. Here, we present wavelength-dependent relative photoabsorption cross section measurements of [dAMP-H]^- below the photodetachment threshold, which lead to fragmentation along several different channels. Several spectral features are observed in the broad absorption peak in the range of 240 to 270 nm, the resolution of which we attribute to the low temperature of 3 K achieved in our cryogenic 16-pole radiofrequency wire trap. These features likely originate from different Franck-Condon-active vibrational bands in only one or two different conformers. Quantum chemical calculations predict that the spectrum originates from a strong ππ* excitation located at the adenine moiety. Furthermore, the wavelength-dependent yield of the five observed photofragments was studied. This revealed no preferred single photofragment, but showed different trends for different fragments as a function of photon energy. Finally, an absolute photofragmentation cross section of [dAMP-H]^- was obtained by comparison with the photodetachment cross section of I^-.

1

16-pole wire trap setup used for this experiment. A custom-built nano-ESI sprays into a double skimmer source. It guides the ions into an octupole ion guide, which serves as a pretrap. The quadrupole ion filter, with its segmented end-cap, guides the ions into the 16-pole wire trap. After the trap, a lens stack guides the ions into the Wiley–McLaren time-of-flight mass spectrometer. The mirror of the reflectron was not used in the present experiments, instead the ions were detected on a microchannel plate (MCP) behind the mirror.

2

Exemplary set of timings of our experiment. Note that the x-axis of this plot is cut in multiple positions. Each cycle starts with a pulse of helium buffer gas being let into the trap. The ions, which were pretrapped in the octupole are then let into the main trap. These two timings are matched to each other so that the ions and the buffer gas arrive in the trap at the same time. Shortly before the exit of the octupole is opened, its entrance gets closed to prevent any nonthermalized ions from flying into the trap. When the entrance opens again, the pretrapping for the next experimental cycle begins. This usually overlaps with the current experimental cycle to speed up the experiment. Once the ions are trapped in the main trap, they thermalize via interaction with the buffer gas. After this, the laser shutter opens and the ions are irradiated for a selected time, unloaded from the trap, and guided toward the Wiley–McLaren time-of-flight mass spectrometer. It is triggered once the cold ions from the trap have reached it. Afterward the experiment restarts. The laser shutter reopens for a short time after the ions have left the trap to record the laser power.

3

Measured relative photoabsorption cross section, depicted with 1 σ error bars. The red line displays a fit of seven Lorentzian peaks, one for each observed features A to G in the structure and a final one for the shoulder around 5.0 eV. Alongside our data we also plot the data recorded by Marcum, Halevi and Weber. We chose to plot this in a different representation than in their original paper, in order to present a less convoluted figure. For this purpose a running average was calculated over the data points of their work. The gray solid line shows the average of this data, the shaded area around it the 1 σ confidence interval.

4

Scheme of [dAMP–H]⁻ photochemistry, showing the bright ππ* transition and the lowest-lying nπ* transition as calculated for conformer A. The transition energies are provided in eV as calculated at the ωB97XD/aug-cc-pVDZ level, f refers to the oscillator strength. For comparison, the analogous scheme for conformer G is shown in Figure S7.

5

Combined mass spectrum of the detected [dAMP–H]⁻ fragments. This is taken using two different time-of-flight measurements between the trap and Wiley–McLaren plates (light and dark blue) since we are unable to couple all ion masses simultaneously into the mass spectrometer (refer to Supporting Information for details). Differences in the coupling efficiencies for the masses prevent the peak intensities from being directly comparable to each other.

6

Wavelength-dependent fragment yields for the five fragments that were observed for the photofragmentation of [dAMP–H]⁻. The yield of each fragment has been normalized to its own average over all wavelengths, to show wavelength dependent changes of each fragment. We additionally include linear fits of the fragment data with the shaded region depicting the fit variation as we vary the fit parameters by up to one σ of their error.