Patent Grant
7351955
The present invention relates generally to mass spectrometers, and more particularly to an apparatus and method for reducing chemical noise arising from the presence of matrix cluster ions in a MALDI mass spectrometer.
One major problem faced by designers and users of mass spectrometers is the presence of chemical noise in the mass spectra. The source and appearance of chemical noise will vary according to the analyte type, ionization method, and operating conditions. For mass spectrometers utilizing matrix-assisted laser desorption/ionization (MALDI), most of the chemical noise is represented by groups of peaks corresponding to matrix cluster ions with compositions described by (Matrixx−(y+z−1)H)+NayKz(with possible loss of water or ammonia), where x, y, and z may vary from 0 to 8. In addition to these peaks, there is a background noise of complex structure that spans the entire range of mass-to-charge ratios, and which may mask peaks representing the analyte molecule(s) and its fragments. Chemical noise may be particularly problematic in high-throughput operations where careful and frequent cleaning of the MALDI substrate surface (which eliminates some sources of chemical noise) is not practical.
Because the chemical noise imposes a practical lower limit on sensitivity (the minimum analyte quantity that can be reliably detected by a mass spectrometer), there is a strong motivation to develop techniques that eliminate or minimize chemical noise sources. One such technique involves heating the matrix cluster ions to a temperature sufficient to break the relatively weak non-covalent bonds, such as hydrogen bonds or Van der Waals forces, which hold the matrix clusters together. However, applying heat thermally to the matrix cluster ions may have the undesirable effect of promoting decomposition of labile analyte ions. Furthermore, heating the analyte ions increases their kinetic energy and may adversely affect transmission and trapping efficiencies.
In brief, a mass spectrometer according to an embodiment of the present invention includes a MALDI source, ion transport optics, a linear ion trap, and a laser or other radiation source configured to direct a radiation beam along a longitudinal axis of the ion trap. Ions produced by the MALDI source, which include both analyte ions and matrix cluster ions, are trapped within the linear trap and are cooled by collisions with damping gas molecules such that the ion cloud (the volume occupied by most of the ions) shape approximates a narrow cylinder, having a typical diameter of less than 1 mm. The radiation beam is positioned to overlap with and preferably envelop the ion cloud. The radiation source may take the form of a pulsed gas or solid-state laser that emits UV or IR light at a frequency that is strongly absorbed by the matrix cluster ions but is less efficiently absorbed by the analyte ions. The absorption of radiation by the matrix cluster ion results in the breaking of at least one bond and the consequent dissociation of the matrix cluster ion into two or more low molecular weight components having mass-to-charge ratios that lie below the mass-to-charge ratio range of interest (the range in which the analyte and/or its fragments lie). Thus, chemical noise associated with matrix cluster ions is reduced or eliminated. In certain implementations of the invention, the photodissociation process may be applied after or during isolation of an ion of interest in the ion trap.
In accordance with an alternative embodiment, a supplemental oscillating field may be generated in the ion trap to cause analyte ions to travel outside of the region irradiated by the laser beam while still remaining confined within the trap. The frequency (or range of frequencies) of the supplemental field is selected such that the matrix cluster ions are not excited, or excited to a lesser degree relative to the analyte ions, and so the matrix cluster ions stay within the irradiated. In this manner, the analyte ions receive less radiation than the matrix cluster ions. This embodiment may be particularly useful to avoid undesired fragmentation when the analyte ions are absorptive at the laser beam frequency.
In accordance with another aspect of the invention, a MALDI mass spectrometer is provided having a single radiation source that performs the dual functions of ion production and matrix cluster dissociation. This is accomplished by disposing an optical switching device in the radiation beam so that the beam may be selectively switched between a first beam path terminating at the sample, and a second beam path extending along the centerline of the linear ion trap.
In the accompanying drawings:
a) and 3(b) depict mass spectra of melittin from a MALDI mass spectrometer respectively obtained without and with photodissociation of matrix cluster ions; and
a) and 4(b) depict a second embodiment of a mass spectrometer configured for in-trap dissociation of matrix cluster ions, wherein a single laser source is employed for both ion production and cluster dissociation.
The ion plume generated by MALDI source 110 will typically contain matrix cluster ions. The components of the matrix cluster ions are held together by non-covalent or weak covalent bonds. As discussed above in the background section, the matrix cluster ions, if not dissociated or otherwise eliminated prior to mass analysis, may contribute significantly to chemical noise, thereby masking the analyte signal and reducing instrument sensitivity. This problem may be avoided or minimized in accordance with the invention by photodissociation of the matrix cluster ions in the ion trap, in the manner described below.
The analyte ions and matrix cluster ions are gated into ion trap 140 by applying or removing a DC voltage to or from an entrance section of the ion trap or a lens or other electrostatic device. Various automatic gain control (AGC) methods known in the art may be utilized to control and optimize the ion population within the ion trap. Ion trap 140 is preferably implemented as a conventional two-dimensional (also known as linear) ion trap of the type discussed in U.S. Pat. No. 5,420,425 to Bier et al. and incorporated into the Finnigan LTQ mass spectrometer (Thermo Electron, San Jose, Calif.). The trap is constructed from four parallel spaced-apart elongated electrodes 145, each preferably having a hyperbolic surface. Each electrode 145 may be segmented into entrance 150, exit 155 and medial 160 sections, which are electrically isolated from one another so that they may be held at different DC potentials to create a potential well that axially confines ions to the medial portion of the trap interior. Alternatively, axial confinement may be effected by providing opposed end electrodes and applying a supplemental oscillating voltage to the end caps to create an axial pseudopotential. Radial confinement of ions in the trap interior is achieved by applying RF voltages to opposed electrode pairs to generate a substantially quadrupolar electric field within the trap. One or more of electrodes 145 may be adapted with a slot to allow ejection of ions therethrough. Supplemental oscillating voltages may be applied to the electrodes to resonantly excite ions for precursor selection and/or collisionally induced dissociation of ions of interest for MSn analysis.
The kinetic energy of ions injected into ion trap 140 is preferably reduced by collisional cooling prior to initiating photodissociation of the matrix cluster ions. The interior of trap 140 is preferably filled with an inert damping gas, such as helium, that provides cooling of ions within the trap so that the ions are focused to a narrow cylindrical volume centered about the trap axial centerline. A cooling time of about 1-2 milliseconds (ms) is usually sufficient to achieve the desired degree of kinetic energy damping. Typical damping gas pressures for ion trap operation will be around 1 millitorr. The ion-occupied volume 165 (the smallest volume occupied by about 95 percent of the collisionally cooled ions), also referred to herein as the ion cloud, will depend on trap and ion parameters. Based on computer modeling of ion motion within the Finnigan LTQ trap at typical operating conditions, the diameter of ion cloud 165 is believed to be approximately 1 mm. The ion cloud 165 has an axial dimension that is roughly coextensive with the longitudinal extent of the medial section of the electrodes.
Signal-to-noise ratios and instrument sensitivity may be improved by isolating in trap 140 ions having mass-to-charge ratios within a range of interest, thereby removing components of chemical noise arising from those matrix cluster ions having mass-to-charge ratios lying outside of the range of interest. As is known in the art, isolation may be achieved by application of a notched broadband isolation voltage to the trap electrodes 145. The isolation voltage is composed of a plurality of frequency components that correspond to the secular frequencies of the ions having the undesired mass-to-charge ratios, while being substantially devoid of frequencies that match the secular frequencies of the ions within the range of interest. Thus, the undesired ions become kinetically excited and are removed from the trap via ejection or collision with electrode surfaces, while the ions in the range of interest are not kinetically excited, or are excited to a much lesser degree, and remain confined within the trap.
Following completion of the ion cooling and (optionally) the ion isolation operations, the matrix ion clusters are dissociated by passing a beam of radiation at a wavelength that is strongly absorbed by the matrix cluster ions along the centerline of the trap. The beam 167 may be generated by a laser 170, such as a gas or solid-state laser, that is capable of generating radiation having suitable wavelength and power. One or more beam turning elements, such as mirror 175, may be disposed in the beam path to direct the beam along the trap 140 centerline. The beam 167 is oriented and sized to substantially overlap ion cloud 165 such that a large fraction or all of the ions in ion cloud 165 are positioned within the beam and are exposed to the radiation. In a typical implementation, beam 167 has a diameter of about 1 mm. The absorption of the beam energy by the matrix ion clusters will cause a substantial fraction of the weakly-bonded clusters to dissociate into a plurality of fragments. Generally, a matrix ion cluster composed of a greater number of matrix molecules will exhibit more efficient absorption of photons relative to a matrix ion cluster composed of a lesser number of matrix molecules, so the time required for dissociation to occur will increase as the number of matrix molecules (and mass) decreases. The radiation is preferably delivered as a train of pulses. The number of pulses required to achieve a desired degree of fragmentation may depend on (inter alia) the pulse energy and width, ion cloud and laser beam geometries, cluster bond strength and absorption efficiencies. In tests done using a sample consisting of 1 femtomole of angiotensin in 5 mg/ml α-CHA matrix solution in a modified LTQ ion trap using a Nd:YLF laser having a pulse energy of 150 μJ, a wavelength of 349 nm, a pulse width of 9 ns, and a repetition rate of 1 kHz, it was found that a total of 90 pulses produced an appreciable improvement in signal-to-noise ratios and reduction of chemical noise attributable to matrix cluster ions.
The fragments resulting from photodissociation will have mass-to-charge ratios that are significantly reduced relative to the matrix cluster ions. Certain of these fragments may have mass-to-charge ratios that are below the low-mass cutoff (LMCO) limit and will develop unstable trajectories that will remove them from the trap. Fragments that remain confined within the trap may be removed by a second isolation step, or may alternatively be scanned out of the trap during a mass-selective analysis scan. Since the mass-to-charge ratios of the fragments will typically be substantially below those of the analyte molecules, the presence of the fragments in the trap do not cause noise at the mass-to-charge ratios of interest and hence do not adversely affect instrument sensitivity.
While the irradiation and consequent photodissociation of matrix cluster ions may be performed in other regions of the mass spectrometer system, e.g., within sections of the ion optics or within an associated collision cell (not depicted), in-trap irradiation offers several advantages. Cooling of the ions within trap 140 by collision with damping gas atoms or molecules produces an ion cloud 165 having a narrow diameter that can be closely matched to the radiation beam diameter. In contrast, the ion beam diameter within low-pressure regions (having pressures well below 1 millitorr) of the ion optics will typically be significantly wider than the ion cloud 165 diameter within trap 140. If photodissociation of matrix cluster ions were to be performed within the low-pressure ion optics, the radiation beam would need to be expanded using appropriate optical elements to match its diameter to that of the ion beam; furthermore, the reduced beam fluence associated with the expanded beam size may reduce the efficiency of cluster fragmentation. We note that even if the ion optics (or a region thereof) were to be operated at a pressure similar to that of ion trap 140 (close to 1 millitorr), the residence time of ions within the ion optics is typically inadequate for the delivery of a sufficient number of radiation pulses for dissociation of matrix cluster ions to occur to an appreciable degree. Within the intermediate—pressure regions of the ion optics (having pressures on the order of 10-100 millitorr), the ion beam diameter may be comparable in size to that of ion cloud 165 (facilitating matching to the radiation beam diameter), but the loss of energy through collisions occurs at a significantly elevated rate, thereby requiring a larger radiation energy input to initiate declusterization.
The above-discussed cluster fragmentation technique assumes that the ions of interest are non-absorptive or weakly absorptive at the wavelength of laser 170. If, however, the ions of interest exhibit moderate or high absorption at the laser wavelength, they may undergo photo-induced dissociation as well. This undesirable result may be reduced by employing the technique depicted in
The benefit realized by employing the cluster dissociation technique of the invention is illustrated by
a) and 4(b) depict an alternative embodiment of a mass spectrometer 400, which differs from the
In the implementation shown in
It should be appreciated that the rotatable mirror structure described above represents but one example of a beam switching element. As used herein, the term “beam switching element” is intended to include any structure or combination of structures that enable the beam to be controllably switched between at least two paths. One or both of the paths may include a light-guiding structure such as an optical fiber or planar waveguide. In other implementations, the beam switching element may take the form of (without limitation) a different configuration of electromechanically actuated device (such as a rotating or sliding mirror or prism), an electro-optical or acousto-optical switch, or an all-optical switch.
The foregoing description, for purpose of explanation, has been described with reference to specific embodiments. However, the illustrative discussions above are not intended to be exhaustive or to limit the invention to the precise forms disclosed. Many modifications and variations are possible in view of the above teachings. The embodiments were chosen and described in order to best explain the principles of the invention and its practical applications, to thereby enable others skilled in the art to best utilize the invention and various embodiments with various modifications as are suited to the particular use contemplated.
| Number | Name | Date | Kind |
|---|---|---|---|
| 5073713 | Smith et al. | Dec 1991 | A |
| 5180914 | Davis et al. | Jan 1993 | A |
| 5381006 | Wells et al. | Jan 1995 | A |
| 6075244 | Baba et al. | Jun 2000 | A |
| 6342393 | Hofstadler et al. | Jan 2002 | B1 |
| 6610976 | Chait et al. | Aug 2003 | B2 |
| 6642516 | Hansen et al. | Nov 2003 | B1 |
| 6797949 | Hashimoto et al. | Sep 2004 | B2 |
| 6919562 | Whitehouse et al. | Jul 2005 | B1 |
| 7119331 | Chang et al. | Oct 2006 | B2 |
| 7157698 | Makarov et al. | Jan 2007 | B2 |
| 20020148972 | Krutchinsky et al. | Oct 2002 | A1 |
| 20040119015 | Hashimoto et al. | Jun 2004 | A1 |
| 20040217272 | Horning et al. | Nov 2004 | A1 |
| 20050151073 | Kato | Jul 2005 | A1 |
| 20050178955 | Baba et al. | Aug 2005 | A1 |
| 20050258353 | Berkout et al. | Nov 2005 | A1 |
| 20060261267 | Sze et al. | Nov 2006 | A1 |
| Number | Date | Country |
|---|---|---|
| 2390937 | Jan 2004 | GB |
| Number | Date | Country | |
|---|---|---|---|
| 20070057173 A1 | Mar 2007 | US |
