Bulgarian Chemical Communications, Volume 40, Number 4 (pp. 464-468) 2008

Pulsed field gradient spin echo NMR spectro-scopy can be used for investigating diffusion phenomena in solution [4, 5]. Mixture analysis continues to be one of the most challenging pursuits in analytical science. Diffusion-ordered spectro-scopy (DOSY) allows the separation of components of different molecular size through differences in their diffusion coefficients and can be a powerful tool for mixture analysis. The diffusion coefficients of the components are dependent on their individual physical properties, such as the size and shape of the molecules as well as the temperature and viscosity of the solutions. In high resolution diffusion-ordered spectroscopy (DOSY), differences in the diffusion coefficient between species are sufficient to allow the spectrum of an intact mixture to be decomposed into the subspectra of individual components. Therefore, the DOSY technique offers a route whereby the components in complex biological systems such as tissue and tissue extracts might be separated and hence identified [5–7].

The aim of the present study has been to apply diffusion-ordered spectroscopy (DOSY) for separa-tion and identification of the metabolites of human lung carcinoma cell extracts, exhibiting different levels of multi-drug resistance (MDR). The human lung carcinoma cell extracts, both hydrophilic and lipophilic, have been studied.

Experimental

The cell lines studied include the parent cell line DLKP (a human squamous non-small cell lung carcinoma) and the resistant daughter cell line DLKP-A5F. DLKP cells express a small amount of the multidrug resistance. The lipophilic (L) and hydrophilic (H) extracts of the parent cell line DLKP (a human squamous cell lung carcinoma) and resistant daughter cell line DLKP-A5F were prepared according to a slightly modified metho-dology published in the literature [8]. The extraction procedure was performed on a number of 1×10⁷ cells per NMR sample on a crushed ice bath at 4ºC. Cells were trypsinised, counted and pelleted into a 1.5 ml eppendorf tube. 500 μl of a 2:1 (v/v) ratio of ice cold methanol-chloroform were added and the pellet resuspended using a vortex. The tube was mixed for 10 min on a blood tube mixer. 500 µl of ice cold 1:1 (v/v) chloroform water mixture were added and again mixed using a vortex. The eppendorf tube was floated in a sonicating bath and sonicated for 10 min. The tube was centrifuged in a microfuge at 13000 rpm for 5 min. The top and bottom layers were removed to separate eppendorf tubes taking care not to disturb the pelleted debris. The solvent was evaporated using a centrifugal con-centrator (MaxiPrep, Holten) and the tubes placed in a darkened box in a freezer for later analysis.

The NMR spectra were recorded on a Bruker AvanceII 600 spectrometer, equipped with a cryo probe and pulse gradient units, capable of producing magnetic field pulsed gradients in the z-direction of 56.0 G/cm.

G. I. Ivanova et al.: Application of diffusion-ordered spectroscopy

1D ¹H NMR spectra of hydrophilic cell extracts (7 mg) in D₂O were recorded at 300K and 0.1% solution of TSP in D₂O as an external chemical shift reference, using a presaturation during the relaxation delay. Two-dimensional ¹H/¹H COSY (correlated spectroscopy) and ¹H/¹H TOCSY (total correlated spectroscopy) have been acquired in a phase sensi-tive mode. In both cases, 64 transients per increment and 256 increments were collected into 2K data points and relaxation delay 2 s. The spectral width in both dimensions was 5000 Hz. The TOCSY NMR spectra were recorded by use of MLEV-17 spin lock scheme for proton-proton transfer with a mixing time of 60 ms.

The DOSY experiments were acquired using the bipolar pulse longitudinal eddy current delay (BPPLED) pulse sequence [9] and presaturation during the relaxation delay for the samples of hydrophilic extracts (in D₂O). Typically, in each experiment a number of 32 BPPLED spectra of 32K data points were collected, with a duration of the magnetic field pulse gradients (δ) of 2.2–3.0 ms, diffusion times (Δ) of 100–200 ms and an eddy current delay set to 5 ms. The pulse gradient (g) has been incremented from 2 to 95% of the maximum gradient strength in a linear ramp. The experiments were acquired at a temperature of 298K and air flow of 535 l/h, in 3 mm NMR tubes. NMR tubes of 3 mm size were used to increase samples concentra-tion and suppress the temperature induced convec-tion currents due the smaller diameter as compared to 5 mm tubes in the diffusion experiments.

Results

Typical ¹H NMR spectra of intact cancer cells (for comparison), their lipophilic and hydrophilic extracts are presented in Figure 1. The signal assign-ment of the spectra (Figure 1B and C) is based on the analysis of 1D and 2D (¹H/¹H COSY and ¹H/¹H TOCSY) spectra and is in agreement with the lite-rature data [10–12]. The main resonances observed in the spectra of lipophilic extracts (B) are those from fatty acids (saturated and unsaturated), choline, cholesterol and glycerol backbone moieties. In the ¹H NMR spectra of the hydrophilic extracts, the anticipated resonances of water-soluble cell meta-bolites (amino acids and other organic acids and bases) have been observed. The characteristic reso-nance signals of some of the major metabolites found in the samples are shown in Figure 1.

The analysis of the spectra shows that the lipids extracted from both lung carcinoma cell lines contain at least three lipid components classified according to their self-diffusion coefficients and respectively different molecular size and molecular weight aromatic fractions. The diffusion coefficients of the separated components are presented in Table 1. As depicted from the 2D DOSY plots, the results in table 1 show that on the base of the measured diffu-sion coefficients, differences in the composition of the cellular extracts could be estimated. The presence of cholesteryl esters of fatty acids has been detected in both mixture components of DLKP_L (logD =

2D DOSY (600 MHz) spectra of the hydrophilic extracts in D₂O are presented in Fig. 3. The spectra are dominated by the resonance signals of myo-Inositol (myo-I), choline (Cho), creatine (Cr), gluta-mine (Glu)/glutatione (Glt). The signal overlap in the spectral area between 3.30 and 4.30 ppm as well as the low intensity of some components in the high field spectral area have hampered a more detailed analysis of the DOSY spectra.

G. I. Ivanova et al.: Application of diffusion-ordered spectroscopy

The preliminary results presented here show that Diffusion-Ordered NMR Spectroscopy (DOSY) can be a powerful tool for separation and spectral char-acterisation of the components of complex mixtures, such as tissue extracts. The DOSY techniques can be used as a tool to distinguish between components in a mixture due to the differences in their relative diffusion coefficient values. The diffusion coeffi-cient measurements of cell extracts allow an addi-tional chemical and physical characterization of individual components, in term of their diffusivities, size and shape of the molecules.

G. I. Ivanova et al.: Application of diffusion-ordered spectroscopy

Acknowledgements: G. I. Ivanova acknowledges FCT-Portugal for the grant SFRH/BPD/24930/2005.

References

1. 
   M. Bloom, K. T. Holmes, C. E. Mountford, P. G. Williams, J. Magn. Reson., 69, 73 (1986).
   
2. 
   C. E. Mountford, S. Doran, C. L. Lean, P. Russell, Chem.Rev., 104, 3677 (2004).
   
3. 
   T. L. Whitehead, T. Kieber-Emmons, Prog. NMR Spectr., 47, 165 (2005).
   
4. 
   C. S. Johnson, Progr. NMR Spectr., 34, 203 (1999).
   
5. 
   J. S. Gounarides, A. Chen, M. J. Shapiro, J. Chromatogr. B, 725, 79 (1999).
   
6. 
   Y. Cohen, L. Avram, L. Frish, Angew. Chem., Int. Ed., 44, 520 (2005).
   
7. 
   B. Antalek. Conc. Magn. Reson., 14, 225 (2002).
   
8. 
   J. E. Le Belle, N. G. Harris, S. R. Williams, K. K. Bhakoo, NMR Biomed.,15, 37 (2002).
   
9. 
   D. Wu, A. Chen, C. S. Johnson. J. Magn. Reson. Ser. A, 115, 260 (1995).
   
10. 
    U. Sharma, A. Mehta, V. Seenu, N. R. Jagannathan, Magn. Reson. Imag., 22, 697 (2004).
    
11. 
    V. Govindaraju, K. Young, A. A. Maudsley, NMR Biomed., 13, 129 (2000).
    
12. 
    N. J. Waters, E. Holmes, C. J. Waterfield, R. D. Farrant, J. K. Nicholson, Biochem. Pharmacol., 64, 67 (2002).