Biochemical Society Transactions

Genome Instability and Cancer

Development of a novel mass spectrometric technique for studying DNA damage

Tony L. Merrigan, C. Adam Hunniford, David J. Timson, Martin Catney, Robert W. McCullough


An experimental system, based upon UV and IR laser desorption, has been constructed to enable the production and characterization of neutral biomolecular targets. These targets are to be used for interaction experiments investigating radiation-induced damage to DNA. The viability of the laser-desorption techniques of MALDI (matrix-assisted laser-desorption ionization), SALDI (surface-assisted laser-desorption ionization) and DIOS (desorption/ionization on silicon), for production of these gas targets is discussed in the present paper. Fluorescent dye tagging and LIF (laser-induced fluorescence) imaging has been used to characterize the biomolecular plumes, revealing their spatial density profiles and temporal evolution.

  • desorption/ionization on silicon (DIOS)
  • laser desorption
  • mass spectrometry
  • matrix-assisted laser-desorption ionization (MALDI)
  • surface-assisted laser-desorption ionization (SALDI)


Biological systems, particularly the human body, can be subject to many forms of radiation from our modern world. Consequently, it is important that we develop a clear understanding of radiation-induced processes and specifically how radiation does damage to DNA which may ultimately lead to cancer. Working environments, such as hospitals or airlines, present an elevated risk to their personnel, as they can experience prolonged exposure to low levels of radiation and may face greater than average doses of radiation in a year. Efforts are made to minimize these risks, for instance the U.K. Health and Safety Executive has a code of practice for working environments dealing with IR (ionizing radiation) [1]. However, the reality is that relatively little is known about how ionizing (and sub-ionizing) radiation causes cancer and what can really be termed a safe level of radiation.

Radiation is also used to target and kill cancerous cells as part of radiotherapy [2]. This most frequently uses high-energy photons which deposit most of their energy near the surface of the tissue. Heavy-ion therapy is an alternative technique making use of a very localized energy deposition, which is ideal for targeting and destroying deep-seated tumours [3]. In contrast with photons, they show an inverse dose-deposition profile where the dose deposition increases with increasing depth up to a maximum at a point known as the Bragg peak. This enables high doses to be delivered to the tumour while maintaining a low dose to the surrounding healthy tissue.

Most investigations of radiation damage concentrate on relatively large systems such as cells. However, so that we can gain a fundamental understanding, it is important to study interactions at the single-molecule level. Many such studies have focused on irradiation of biomolecules in the gas phase where it is possible to use MS to analyse fragment production. This approach is restricted by the limited number of molecules that can be introduced into the gas phase. Thermal evaporation is commonly used, but is unsuitable for complex molecules owing to thermal decomposition. The ability to produce characterized targets of more complex molecules, such as oligonucleotides, would provide an important step in linking fundamental molecular studies to those carried out on larger systems. Several techniques have been used to introduce complex molecules into the gas phase and are often found in MS systems. One approach is the use of laser desorption, and an overview of this is given in the following sections.

Matrix-assisted desorption

One of the earliest approaches was to use a laser to desorb molecules from a dried sample. This technique relied upon inducing rapid heating of the sample, resulting in desorption before the molecules had time to decompose. The disaccharide sucrose was the first compound to be investigated with this laser-desorption method [4], and, unlike glucose, which can be evaporated without extensive thermal fragmentation, sucrose is practically non-volatile. However, as the mass of the analyte molecules is increased, fragmentation during this direct laser-desorption process becomes increasingly problematic. MALDI (matrix-assisted laser-desorption ionization) is a development where a chemical known as a matrix is added to the analyte and is chosen because it has a strong absorption at the desorption laser wavelength [5]. The matrix acts as a receptacle for the incident laser radiation promoting efficient coupling of the energy into the condensed phase, allowing the analyte to be ejected efficiently into the gas phase. Ions are formed by gas-phase collisions in the plume and may be extracted, yielding MS information about the sample. The extended mass range and sensitivity made available by MALDI has made it an established and important tool in the biological sciences. However, it has been developed in a highly empirical manner because of limited understanding of the underlying mechanisms and the technique involves complicated sample preparation procedures, poor shot to shot reproducibility and a large amount of low-mass background interference. As a result of these shortcomings, there is a growing trend towards development of laser-desorption techniques which do not require the presence of a matrix.

Surface-assisted desorption

Surface-assisted desorption techniques [SALDI (surface-assisted laser-desorption ionization) and DIOS (desorption/ionization on silicon)] aim to expand upon the success of MALDI and to make routine sample analysis more straightforward and without the chemical complexity inherent with MALDI.

Early work used a fine graphite powder suspended in a solution of analyte and glycerol [6], resulting in intense analyte signals with minimal background signal. An evolution of this technique used porous activated carbon (AC) in the sample suspension as it was found this helped give a more reproducible signal across the substrate surface. The principal difficulty results from evaporation of glycerol, liberating carbon particles which could coat the system and lead to electrical discharges. Sucrose may be added to the samples, restricting the movement of carbon while causing little suppression of analyte ions, with sodiated and potassiated sucrose peaks being the only addition to the spectra. SALDI subsequently made use of thin AC substrates instead of suspended carbon particles [7]. AC powder was sprinkled on to an aluminium surface which was covered with an adhesive layer. This immobilized the carbon particles and, because of the increased surface area exposed to the analyte, increased the sensitivity of the system. Glycerol was still used as an additive as it enabled ion signals to be detected for extended periods. It should be noted that, for the purpose of producing neutral gaseous molecules, glycerol would not be necessary, allowing removal of associated contaminants. In spite of the interest in this AC SALDI technique, an upper mass limit of only a few thousand Daltons restricts its applicability.

DIOS [8] is a matrix-free technique using porous silicon to mimic the effects produced by a matrix. Both analyte and solvent are adsorbed on to the large surface area presented by the porous framework of the silicon. The silicon absorbs UV light and then can transfer energy to the bound molecules. The primary advantages of this technique are that the sample preparation is greatly simplified and, critically when trying to produce a ‘clean’ target, it is free from matrix contamination. The mass spectra from resultant ions demonstrate the lack of low-mass species, indicating little fragmentation results from the process.

Porous silicon is produced from crystalline silicon by a chemical etching process, resulting in oxidative degradation, forming a network of pores. Pore size and porosity are important because they dictate the size of the surface area which should be sufficiently large for effective energy transfer to the analyte [9]. Freshly etched DIOS surfaces are hydrophobic, leading to small and concentrated sample spots when using aqueous samples. The lack of low-mass background signal allows study of low-molecular-mass analytes which are not easily studied by MALDI (which typically observes molecules of >700 Da). This simplicity and flexibility has led to interest in the possibility of developing a chip-based system for high-throughput analysis [10]. However, there is again a relatively low upper mass limit of ∼3000 Da, beyond which analyte fragmentation becomes problematic. Ongoing study is focusing on further surface modification to enable DIOS to be useful with larger molecules. Recent work in this area has resulted in the development of the DIOSD (desorption/ionization on silicon dioxide) technique, whereby a DIOS-like chip is made by making a silicon substrate selectively porous [11]. The result is a silicon chip with an array of hydrophilic porous silicon dioxide spots surrounded by a dense hydrophobic silicon dioxide layer. This approach produces signals five times greater than those found with the standard DIOS chip. It has also been discovered that alternative silicon substrates may be used, but require ‘activation’ [12]. This may be achieved using UV irradiation and is particularly effective in the presence of water vapour, suggesting that the silicon surface is modified to contain hydroxy groups, i.e. Si-OH.

From the point of view of a gas-phase biomolecular target, ease of sample preparation, dense localized samples, high sensitivity and low fragmentation make SALDI/DIOS a promising technique. The relatively low mass limit available with these techniques means that there is still a great restriction imposed upon which molecules may be studied.

IR laser desorption

All of the techniques so far mentioned have used UV radiation for desorption. However, a number of studies have also used IR wavelengths to try to ‘softly’ desorb the analyte molecules, i.e. removal from the surface with minimal fragmentation. IR desorption has again been studied for matrix- and surface-assisted techniques, but nowhere near as extensively as UV desorption.

IR MALDI suffers from the same inherent matrix complication as UV MALDI and thus provides no better an option for neutral target production. There have been a limited number of studies so far which have looked at direct IR laser desorption from substrates such as stainless steel, aluminium, copper, polyethylene and silicon. The best performance in terms of signal level and low mass interference was obtained with silicon surfaces. In contrast with the UV case, it was found that no enhancement in the spectra was observed using a porous substrate [13]. Peptides and proteins (up to 17 kDa) could be desorbed with samples with masses below 6000 Da producing spectra of similar quality to MALDI, but without the associated low-mass background signal. A later study explored the effect of IR wavelength on ionization efficiency and found that the threshold fluence for desorption of insulin molecules was lowest when operating at 2.94 μm, coincident with the stretch mode of hydroxy groups within the molecule [14]. Another local maxima was found at 3.4 μm coinciding with the C-H stretch. This seems to be evidence that it is in fact the analyte itself that is doing the absorption and not the presence of any residual water that may be present in the sample. The analyte is therefore considered to act as its own intrinsic matrix, resulting in an inherent level of sample fragmentation.

Overall, this IR surface-assisted technique shows much promise as a method for producing a gas-phase target of biomolecules. The mass range is considerably higher than the ranges observed for UV SALDI and DIOS, and there is no need to include glycerol, matrix or any other solution in the sample mixture as was required for the other techniques.

Fluorescence imaging

In order to decide which technique best suits the purpose of producing a gas-phase biomolecular target, it is also necessary to characterize the resulting ‘plume’ of material. Spatial and temporal information of large molecules has been gained previously by monitoring laser-generated plumes using laser-based optical spectroscopic probes. These operate by measuring absorption in the plume or by observing reflection from the surface. This allows monitoring of both the neutral and the ionic components of the desorbed species. However, because of the substantially lower concentrations involved in MALDI and SALDI/DIOS, the above probes are not suitable. In an attempt to better understand the evolution of gas-phase molecules produced by MALDI, a laser may be swept across a plume of dye-tagged molecules and the fluorescent signal used to produce a spatial and temporal map [15]. This provides the extra sensitivity required for MALDI events and reveals similar information to absorption probes. Puretzky and Geohegan [16] improved upon this technique when they performed the first gated LIF (laser-induced fluorescence) imaging of laser-desorbed plumes from a MALDI matrix, 3-HPA (3-hydroxypicolinic acid). Figure 1 illustrates this method. This technique, along with the use of ion-probe measurements, enabled them to investigate desorbed products, densities, fractional ionization and velocity distributions of the 3-HPA plumes. They were able to infer the number of unexcited 3-HPA molecules and derived an absolute concentration of 3-HPA to be between 2.5×1015 and 6×1015 cm−3. It was also found that ions propagated within a relatively narrow distribution located at the leading edge of the neutral 3-HPA plume. An estimated 3.5×1011 ions·cm−3 was found at a distance of 0.3 cm for an ArF laser-generated plume. This indicated that the fraction of ions generated is very small, ni/n0=10−5. This group then reported results of the first gated LIF imaging of the expansion of both matrix and analyte molecules using MALDI [17]. LIF imaging of dye-tagged DNase I proteins shows that these heavy molecules (∼30 kDa) propagate within a very narrow angular distribution compared with that of the 3-HPA matrix (139 Da). This imaging technique has been applied successfully to our own experiment to aid in characterization of a variety of laser-desorbed biomolecular plumes.

Figure 1 Schematic diagram of the LIF method of the biomolecular plume

Experimental approach

A schematic diagram of the experimental arrangement is shown in Figure 2. A tuneable laser system comprising a Nd:YAG (neodymium-doped yttrium aluminium garnet) pumped OPO (optical parametric oscillator) is used to provide a UV output of 355 nm or a tuneable infrared output of 2.6–3.1 μm. The desorption beam is incident at 45° with respect to the normal to the target surface: a multiple-well stainless steel substrate on which is deposited the molecules of interest. A probe sheet beam from the 532 nm output of a Nd:YAG laser is directed parallel to the sample plate surface at variable time delays. The resulting fluorescence from the desorbed dye-tagged biomolecules is focused and imaged by an Andor Technology IStar™ ICCD (intensified charge-coupled device) camera via a viewing port on the top of the interaction chamber, providing a series of ‘snapshots’ of the expanding plume. A linear time-of-flight mass spectrometer is located on the axis of the expanding plume and provides the ability to monitor fragmentation within the gas target.

Figure 2 Schematic diagram of the apparatus used for the production and characterization of gas-phase biomolecular targets

(a) A probe beam excites fluorescence from the laser-desorbed dye-tagged biomolecules, and (b) the fluorescence emission is focused on to and detected by an Andor Technology IStar™ ICCD camera.


A system has been constructed to enable production of gas-phase targets of biomolecules to be used for radiation-interaction experiments. A variety of laser-desorption methods have been examined, including MALDI, SALDI and DIOS. LIF imaging has been used to monitor the propagation of the plumes and provide information on plume evolution and density. Time-of-flight MS has been used to monitor the ionic species present in the plume to determine the extent of fragmentation.


The support received by the European Union Project ITS-LEIF (Ion Technology and Spectroscopy at Low Energy Ion Beam Facilities) [grant number RII3/026015] is gratefully acknowledged. T.L.M. is grateful for the award of an ITS-LEIF studentship.


This experiment has been performed at the QU-LEIF facility, part of the distributed LEIF (Low Energy Ion Beam Facilities) infrastructure.


  • Genome Instability and Cancer: Biochemical Society Irish Area Section Focused Meeting held at National University of Ireland, Galway, Ireland, 4 December 2008. Organized and Edited by Michael Carty (National University of Ireland, Galway, Ireland).

Abbreviations: AC, activated carbon; DIOS, desorption/ionization on silicon; 3-HPA, 3-hydroxypicolinic acid; ICCD, intensified charge-coupled device; IR, ionizing radiation; LIF, laser-induced fluorescence; MALDI, matrix-assisted laser-desorption ionization; Nd:YAG, neodymium-doped yttrium aluminium garnet; SALDI, surface-assisted laser-desorption ionization


View Abstract