Measuring Masses of Single Bacterial Whole Cells with a Quadrupole Ion Trap

Transcription

1 Measuring Masses of Single Bacterial Whole Cells with a Quadrupole Ion Trap Abstract A novel method is developed to measure precisely the masses of single bacterial whole cells using a quadrupole ion trap as an electrodynamic balance. The bacterial cells were introduced into the ion trap by matrix-assisted laser desorption/ionization, confined in space by audio-frequency ac fields, and detected by elastic light scattering. Mass measurement accuracy approaching 0.1% was achieved for Escherichia coli K-12 with a mass distribution of 3% deduced from 60 repetitive measurements of the particles and their clusters. This is the first high-precision mass measurement reported for any intact microorganisms with masses greater than Da. The method opens new avenues for high-precision mass measurement of single biological whole cells and offers a new approach for rapid identification of microorganisms by mass spectrometry. Introduction Measuring masses of intact microorganisms holds much interest in the scientific and mass spectrometric communities. One of the motivations for such measurement is to provide a means for rapid identification of potentially dangerous viruses and bacteria (1). The conventional ways of quantifying biomass involve both flow cytometry and gravimetric determination, which can be sufficiently accurate but are laborious and time-consuming (2). Recently new technologies, such as quartz crystal microbalances (3), superconducting quantum interference devices (4), and micromechanical oscillators (5,6) have been developed to detect single bacterial and viral particles to achieve highest sensitivity. However, the errors involved in these measurements are relatively large, typically more than Fig 1. Schematic of the experimental setup. The quadrupole ion trap was powered by an audio-frequency driver system comprising a synthesized function generator, a power amplifier, and a bipolar transformer. Six holes were drilled on the ion trap (2 on the end-cap electrodes and 4 on the ring electrode) for sample introduction, entry of the MALDI laser beam, entry of the electron beam, entry/exit of the probe laser beam, and light collection. 25

2 Fig 2. Plot of the star branch number (n) as a function of the trap driving frequency (Ω/2ϖ) and voltage amplitude (Vac). The bacterial particles were trapped at qz 0.5 and the m/ze values were determined by fitting the experimental data to the equation described in text with n = Consecutive changing of the charge state by electron bombardment (from right to left) of the same particle yields 5 distinct m/ze values. Insets: Photos showing the star-like trajectories of the particle s oscillatory motions projected on the radial plane. 10%. Part of the reasons for this inaccuracy is that the mass measurements were all conducted in fairly complicated environments. With the development of soft ionization methods (7,8), both electrospray ionization (ESI) and matrix-assisted laser desorption/ionization (MALDI), it has become possible to directly measure masses of single bioparticles in the gas phase (9). Mass measurement accuracy approaching 1% has been achieved for intact MS2 capsids in a time-of-flight instrument (10). In applying the same technique to mass determination of bacteria (~1 µm in size and ~ Da in mass), the main difficulty faced is the low detection sensitivity. This work illustrates a new approach to detect single bacterial whole cells using elastic light scattering (ELS) and measure their masses with a quadrupole ion trap (QIT) mass spectrometer. The single biological whole cell mass spectrometer The experimental setup consisted of a QIT operating in the audio-frequency region (Fig. 1). Intact bacterial particles were introduced into the ion trap by MALDI from a sample probe inserted in one of the holes on the ring electrode using a frequency-tripled Nd:YAG laser. The evaporated macro-ions (or charged particles) were spatially confined with 300 Hz ac fields and damped to the trap center with 1 mtorr He buffer gas. Only a single macro-ion was needed in each mass measurement. When more than one particle was confined in the QIT, the trap driving frequency (Ω) was decreased slowly to eject particles of the same mass but different charge number axially until there was only one left in the trap for examination. Elastic light scattering from the single isolated particle revealed the trajectory of its oscillatory motion in the radial plane. A stationary star pattern emerged after fine tuning of Ω to resonate with the radial secular frequency ( r), i.e. Ω = n r. From the number (n) of the branch of the stationary star (inset in Fig. 2), the mass-tocharge ratio (m/ze) of the particle was deduced from fitting of the experimental data to theories (11,12). To determine the mass m independently, the value of Z was deduced by using an electron gun to change the charge state of the particle randomly by producing one electron differentials (13). The stationary star (inset in Fig. 2) abruptly lost its stability upon secondary emission of the electrons, an event easily detectable by the CCD camera. A new set of data was then collected after this detection to attain a new m/ze value. In assigning the absolute charge numbers, which are all integers, we iterated the assignment for 5 independently measured m/ze values of the same parti- 26

3 cle (i.e. the same mass) until the standard deviation in the least-squares linear fitting of the data is minimal. Bacterial whole cells Escherichia coli K-12 whole cells were chosen as the target microorganism. Scanning electron microscopy revealed that all the bacteria are nonflagellated and similar in size. They formed aggregates and accumulated on the surfaces of the energy-absorbing materials when the matrix crystallized (Fig. 3). Three matrices have been adopted and tested to ensure the cellular integrity of the gaseous bacterial ions upon laser desorption and ionization: sinapinic acid (SA), 4-hydroxy- -cyanocinnamic acid (4HCCA), and 2-(4-hydroxy-phenylazo)-benzoic acid (HABA). To minimize damage of the bacterial whole cells during MALDI, the laser power density was kept low (~10 6 MW/cm 2 ) and exposure of the same sample spot to repetitive laser irradiation was avoided. In all cases, no scattered laser light was detected with the bare matrix, confirming that the observed signals are indeed derived from the trapped bacterial particles, rather than from clusters composed of the matrix molecules. Further confirmation of the signals came from careful statistical analysis of the masses determined for the bacterial particles, and a control experiment using polystyrene microspheres of known diameters as the sample. Absolute mass determination For the particular measurement shown in Fig. 2, we obtained m/ze = ± prior to charge state change of the E. coli K-12 particle. The reproducibility of the m/ze measurement achieved is 0.1%, meaning that the resolution of this single-particle quadrupole ion trap mass spectrometer is ~1000 at m/ze > An assignment of the charge states with the one electron differentials method Fig 3. Scanning electron microscopy images of E. coli K-12 mixed with the sinapinic acid matrix before laser irradiation [magnifications 650 (left) and 3000 (right)]. 27

4 Fig 4. Plot of the assigned cell number (N) versus the measured mass (m) from 60 independent measurements, showing the linear correlation between these two parameters. Inset: Mass distribution of single E. coli monomers after proper scaling of the masses of the clusters with the cell number (i.e. m/n). leads to m = Da or a weight of fg per E. coli dry cell. In the course of this measurement, clusters of the bacterial whole cells as originally observed on matrix (Fig. 3) were found in the gas phase as well (Fig. 4). These cluster macro-ions, composed of N monomers, can be readily identified from the brightness of their star patterns displayed on the CCD camera. From 60 independent measurements, we determined a statistically averaged mass of m/n = Da for a single E. coli K-12 whole cell in vacuum or a dry weight of fg per particle (inset in Fig. 4). To the best of our knowledge, this is the first high-precision mass measurement for intact microorganisms with masses greater than Da. The mass distribution of E. coli K-12 examined in this work for more than 60 particles is 3%, which may derive from the intrinsic mass variation of the bacteria, fragmentation of the bacterial cells, and/or formation of the bacterium-matrix adducts during MALDI. However, this deviation is unlikely to arise from the mass loss due to electron bombardment or further dehydration of the bacterial particle in vacuum over a long period of time. As a stability test, a mass measurement was conducted for the same particle over 18 h, during which the charge state was changed consecutively 11 times. The measured masses differ by less than 0.1%, well within the limit of our experimental uncertainty. To assess further the degree to which the cells ruptured during the ion formation, we employed 4HCCA and HABA as the matrices for comparison. Masses of Da and Da were determined for 4HCCA and HABA, respectively. Together with the result of SA, the measured masses differ by only 0.2%, strongly suggesting that the bacterial cells remain intact during the laser desorption/ionization. We attribute this remarkable feature to the fact that the bacterial particles are only loosely attached to the surfaces of the energy-absorbing crystals (Fig. 3), rather than being incorporated into the matrix. Conceivably, they evaporate through a process resembling surfaceenhanced neat desorption (14), and multiply charged macro-ions can form in the plume via proton transfer reactions. Conclusion We have demonstrated that by using MALDI-QIT-ELS, it is possible to precisely measure the masses of single E. coli K-12 whole cells with accuracy approaching 28

5 0.1%. The result signifies the potential application of this MALDI-QIT-ELS mass spectrometer to detection and, possibly, identification of single viral particles (such as SARS viruses). The biological whole cell mass spectrometry constitutes an extension of the top-down approach toward rapid identification of intact microorganisms. Huan-Cheng Chang Institute of Atomic and Molecular Sciences, Academia Sinica Journal of the American Chemical Society 126, (2004) References 1. Fenselau, C. & Demirev, P. A. Characterization of intact microorganisms by MALDI mass spectrometry. Mass Spectrom. Rev. 20, (2001). 2. Sonnleitner, B., Locher, G. & Fiechter, A. Biomass determination. J. Biotechnol. 25, 5-22 (1992). 3. Cooper, M. A., Dultsev, F. N., Minson, T., Ostanin, V. P., Abell, C. & Klenerman, D. Direct and sensitive detection of a human virus by rupture event scanning. Nature Biotechnol. 19, (2001). 4. Grossman, H. L., Myers, W. R., Vreeland, V. J., Bruehl, R., Alper, M. D., Bertozzi, C. R. & Clarke, J. Detection of bacteria in suspension by using a superconducting quantum interference device. Proc. Natl. Acad. Sci. USA 101, (2004). 5. Ilic, B., Yang, Y. & Craighead, H. G. Virus detection using nanoelectromechanical devices. Appl. Phys. Lett. 85, (2004). 6. Gupta, A., Akin, D. & Bashir, R. Single virus particle mass detection using microresonators with nanoscale thickness. Appl. Phys. Lett. 84, (2004). 7. Wong, S. F., Meng, C. K. & Fenn, J. B. Multiple charging in electrospray ionization of poly(ethylene glycols). J. Phys. Chem. 92, (1988). 8. Karas, M. & Hillenkamp, F. Laser desorption ionization of proteins with molecular masses exceeding 10,000 daltons. Anal. Chem. 60, (1988). 9. Fuerstenau, S. D., Benner, W. H., Thomas, J. J., Brugidou, C., Bothner, B. & Siuzdak, G. Mass spectrometry of an intact virus. Angew. Chem. Int. Ed. 40, (2001). 10. Tito, M. A., Tars, K., Valegard, K., Hajdu, J. & Robinson, C. V. Electrospray time-of-flight mass spectrometry of the intact MS2 virus capsid. J. Am. Chem. Soc. 122, (2001). 11. Hars, G. & Tass, Z. Application of quadrupole ion trap for the accurate mass determination of submicron size charged particles. J. Appl. Phys. 77, (1995). 12. March, R. E. & Hughes, R. J. Quadrupole Storage Mass Spectrometry, Wiley: New York, 1989, p Schlemmer, S., Illemann, J., Wellert, S. & Gerlich, D. Nondestructive high-resolution and absolute mass determination of single charged particles in a three-dimensional quadrupole trap. J. Appl. Phys. 90, (2001). 14. Hutchens, T. W. & Yip, T.-T. New desorption strategies for the mass spectrometric analysis of macromolecules. Rapid Commun. Mass Spectrom. 7, (1993). 29