Reactivity of aromatic σ,σ-biradicals toward riboses

Anthony Adeuya 1 2 3 Linan Yang 1 2 F. Sedinam Amegayibor 0 1 2 John J. Nash 1 2 Hilkka I. Kenttmaa 1 2 0 Current address: Kos Pharmaceuticals , Inc., 1 Cedar Brook Drive, Cran- bury, NJ 08512, USA 1 Published online August 28, 2006 Address reprint requests to Dr. H. Kenttmaa, Department of Chemistry, Purdue University , 1393 Brown Building, West Lafayette, IN 47907-1393, USA 2 Department of Chemistry, Purdue University , West Lafayette, Indiana, USA 3 Current address: USDA-ARS-NCAUR, Bioproducts and Biocatalysis Re- search Unit, 1815 N. University St. , Peoria, IL 61604, USA ion. The overall reaction efficiencies of the biradicals towards a given substrate were found to be directly related to the magnitude of their EA values, and inversely related to their S-T gaps. The EA of a biradical appears to be a very important rate-controlling factor, and it may even counterbalance the reduced radical reactivity characteristic of singlet biradicals that have large S-T gaps. (J Am Soc Mass Spectrom 2006, 17, 1325-1334) 2006 American Society for Mass Spectrometry - Aradicals) and , -biradicals (e.g., benzynes) romatic carbon-centered -radicals (e.g., phenyl have attracted a great deal of recent interest because of their role in nonhydrolytic DNA damage [1]. For example, some , -biradicals have been identified as the biologically active intermediates of the enediyne class of antitumor antibiotics [2]. These intermediates are believed to irreversibly damage double-stranded DNA via hydrogen atom abstraction from a sugar moiety in each strand [2]. Therefore, a better understanding of the factors controlling the reactivity of these biradicals toward sugars is important. Solution [3] and gas-phase [4] studies on the reactivity of neutral and charged phenyl radicals have confirmed that these monoradicals can abstract hydrogen atoms from sugars as well as from the sugar moiety in nucleosides and dinucleoside phosphates. Polar effects (i.e., polarization of the transition state) play a major role in controlling these reactions [57]. However, no such studies have been reported for the analogous biradicals. The magnitude of the singlet-triplet (S-T) gap has been proposed earlier [8] as the major reaction rate controlling factor for aromatic , -biradicals with singlet ground states. As the magnitude of the S-T gap increases, the reaction efficiency for hydrogen atom abstraction from simple substrates has been observed to decrease, presumably because of the energetically high cost of uncoupling the biradicals electrons in the transition state [8, 9]. Biradicals with large S-T gaps appear to avoid this penalty by undergoing nucleophilic or electrophilic (nonradical) addition reactions [10]. Recent gas-phase studies have shown that in addition to S-T gap effects [9], reactions of biradicals with simple organic substrates are also sensitive to polar effects (which is reflected by the biradicals calculated vertical electron affinity, EA) [11]. Here, we report an examination of the reactivity of several , -biradicals (Scheme 1) toward various sugars, and show that these reactions are also affected by the S-T gap and the EA of the biradical. All experiments were carried out in a Finnigan model FTMS 2001, 3-tesla dual-cell Fourier-transform ion cyclotron resonance mass spectrometer (FT-ICR). The instrument has been described in detail previously [12]. The (bi)radical precursors, 4,6-dinitroisoquinoline, 5,8-dinitroisoquinoline, 5,7-dinitroquinoline, and 4iodoisoquinoline, were synthesized using known methods [13]. They were introduced into the dual cell via a heated solid probe at a nominal pressure of about 1.0 2.0 108 torr. Generation of (bi)radicals 1a 4a (Scheme 1) involved N-protonation of the precursors, 4,6-dinitroisoquinoline, 5,8-dinitroisoquinoline, 5,7-dinitroquinoline, or 4-iodoisoquinoline, respectively, via methanol chemical ionization, whereas the generation of 1b 4b involved N-methylation of the same precursors via methyl iodide chemical ionization. The N-protonated and N-methylated precursors, generated in one side of the dual cell, were transferred into the other cell by grounding the conductance limit plate for about 154 s. The transferred ions were allowed to cool for 1 s via IR emission and collisions with the neutral sugar molecules (e.g., D-ribose at about 2.7 10 8 torr). The technique of sustained off-resonance irradiated collision-activated dissociation [14] (SORI-CAD) was used to homolytically cleave the carbonOiodine and/or carbonOnitrogen bonds to generate the corresponding (bi)radicals. This was accomplished by introducing argon (at a nominal pressure of about 10 6 torr) into the cell via a pulsed valve assembly. The ions were accelerated by continuously exciting them for one second at a frequency 1 kHz higher than their cyclotron frequency, and activated via collisions with argon. The (bi)radicals were then isolated by applying a series of stored-waveform inverse Fourier-transform [15] (SWIFT) excitation pulses to the plates of the cell to eject unwanted ions from the cell. The isolated (bi)radical ions of interest were allowed to react with D-ribose, 2-deoxy-D-ribose, and 1-O-methyl-2deoxy-D-ribose (at a nominal pressure of 1.0 10 8 to 1.2 10 7 torr) for variable periods of time (typically 1 to 900 seconds). Ion excitation for detection was achieved using a chirp of bandwidth 2.56 MHz and a sweep rate of 3200 Hz/ s. Each spectrum was collected as 64 k data points with one zero fill before Fourier transformation. All measured reaction spectra were background-corrected as reported earlier [5]. A background reaction spectrum was collected in the absence of the isolated ion of interest. This spectrum was then subtracted from the reaction spectrum to remove peaks that are not indicative of the isolated ions reaction products. The product branching ratios were derived from the constant abundance ratios of the product ions at short reaction times. The sugars, D-ribose, 2-deoxy-D-ribose, and 1-Omethyl-2-deoxy-D-ribose, were introduced into the instrument via a solids probe. The nominal pressure for each sugar was measured by an ion gauge and the pressure reading was corrected for the sensitivity [16] of the ion gauge towards each sugar and the pressure gradient between the cell and the ion gauge. The correction factor for each sugar was obtained by measuring the rate of the highly exothermic electron-transfer to ionized carbon disulfide (i.e., every collision is assumed to lead to a reaction). The concentration of the sugars is much greater than that of the radical ions in the cell. Hence, the reactions studied here follow pseudofirst-order kinetics. A semilogarithmic plot of the relative abundance of the reactant ion as a function of time, and knowledge of the concentration of the neutral reagent, yield the second-order reaction rate constants. The reaction efficiency is given as kreaction/kcollision kcollision was obtained via a parameterized trajectory theory (...truncated)