High-resolution proton NMR measures mobile lipids associated with Triton-resistant membrane domains in haematopoietic K562 cells lacking or expressing caveolin-1.

High-resolution proton NMR spectra of intact tumour cells generally exhibit intense signals due to isotropically mobile lipids (MLs) of still uncertain nature and origin. NMR studies performed on intact wild-type and caveolin-1-infected haematopoietic K562 cells showed that, under our experimental conditions, part of the ML signals are due to lipid complexes resistant to extraction in Triton X-100 at 4 degrees C. This evidence suggests that a portion of NMR-visible lipid structures are compatible with Triton-resistant membrane rafts and therefore biophysically distinct from NMR-visible Triton-soluble lipid bodies. Similarly to lipid rafts and caveolae, the organization of the Triton-insoluble ML domains could be compromised by treatment with beta-octylglucoside or methyl-beta-cyclodextrin. Exposure to exogenous sphingomyelinase caused an increase in ML NMR visibility, indicating the possible involvement of ceramides in ML formation. The mobility of these lipids was found to be temperature sensitive, suggesting a transition in cells going from 4 degrees C to 25-37 degrees C. These new results are here discussed in the light of possible contributions of plasma membrane microdomains to NMR-visible ML signals.

Detection of polyol accumulation in a new ovarian carcinoma cell line, CABA I: a(1)H NMR study.

Ovarian carcinomas represent a major form of gynaecological malignancies, whose treatment consists mainly of surgery and chemotherapy. Besides the difficulty of prognosis, therapy of ovarian carcinomas has reached scarce improvement, as a consequence of lack of efficacy and development of drug-resistance. The need of different biochemical and functional parameters has grown, in order to obtain a larger view on processes of biological and clinical significance. In this paper we report novel metabolic features detected in a series of different human ovary carcinoma lines, by (1)H NMR spectroscopy of intact cells and their extracts. Most importantly, a new ovarian adenocarcinoma line CABA I, showed strong signals in the spectral region between 3.5 and 4.0 p.p.m., assigned for the first time to the polyol sorbitol (39+/-11 nmol/10(6) cells). (13)C NMR analyses of these cells incubated with [1-(13)C]-D-glucose demonstrated labelled-sorbitol formation. The other ovarian carcinoma cell lines (OVCAR-3, IGROV 1, SK-OV-3 and OVCA432), showed, in the same spectral region, intense resonances from other metabolites: glutathione (up to 30 nmol/10(6) cells) and myo-inositol (up to 50 nmol/10(6) cells). Biochemical and biological functions are suggested for these compounds in human ovarian carcinoma cells, especially in relation to their possible role in cell detoxification mechanisms during tumour progression.

1H NMR-visible mobile lipid domains correlate with cytoplasmic lipid bodies in apoptotic T-lymphoblastoid cells.

The presence of nuclear magnetic resonance (NMR)-visible mobile lipid (ML) domains in apoptotic lymphoblasts suggests alterations in neutral lipid metabolism and compartmentation during programmed cell death. The detection of similar ML signals in activated lymphocytes raises questions about common mechanisms of ML formation during apoptosis and upon lymphoblast stimulation. Structure and subcellular localization of ML domains were therefore investigated by NMR, fluorescence and electron microscopy in Jurkat T-lymphoblasts either induced to apoptosis (by anthracyclines or dexamethasone or by serum deprivation) or activated by phorbol myristate acetate (PMA) plus ionomycin. ML contents in drug-treated cells correlated linearly with apoptosis, irrespective of the specific inducer and cell cycle arrest phase (r = 0.993, P < 0.001). Similar ML levels were measured in drug-induced apoptotic cells (A approximately 30-40%) and in non-apoptotic PMA/ionomycin-treated lymphoblasts (72 h). Lower ML contents were instead formed in serum-deprived apoptotic cells, with respect to controls. Increases in ML signals were associated, in either apoptotic or activated cells, with the accumulation of cytoplasmic, osmophilic lipid bodies (diameter < or = 1.0 microm), surrounded by own membrane, possessing intramembrane particles. The results support the hypothesis that ML are formed in the cytoplasm of drug-induced apoptotic cells during an early, 'biochemically active' phase of programmed cell death.

Double-resonance J-edited (1)H-NMR detection of 6-(13)C-D-2-deoxyglucose uptake in glioma cells.

The C6 methylene protons were selectively detected in (1)H-NMR spectra of intact glioma cells incubated with 6-(13)C-D-2-deoxyglucose (6-(13)C-2dG), a (13)C-enriched glucose analog that is suitable for monitoring glucose utilization in brain tumors. Spectral editing via (1)H-(13)C scalar coupling was performed with twin spin-echo double resonance (T-SEDOR), a pulse sequence which combines chemical specificity and high sensitivity, requires no solvent pre-saturation, and can easily be adapted to imaging protocols. This work demonstrates the suitability of the pulse sequence for monitoring 6-(13)C-2dG uptake in living cells in vitro, in spite of line-broadening and the occurrence of other strong signals in the spectral region of interest (3.5-4.4 ppm).

Biophysical and structural characterization of 1H-NMR-detectable mobile lipid domains in NIH-3T3 fibroblasts.

Nature and subcellular localization of 1H-NMR-detectable mobile lipid domains (ML) were investigated by NMR, Nile red fluorescence and electron microscopy, in NIH-3T3 fibroblasts and their H-ras transformants (3T3ras) transfected with a high number of oncogene copies. Substantial ML levels (ratio of (CH2)n/CH3 peak areas R=1. 56+/-0.33) were associated in untransformed fibroblasts with both (a) intramembrane amorphous lipid vesicles, about 60 nm in diameter, distinct from caveolae; and (b) cytoplasmic, osmiophilic lipid bodies surrounded by own membrane, endowed of intramembrane particles. 2D NMR maps demonstrated that ML comprised both mono- and polyunsaturated fatty chains. Lower ML signals were detected in 3T3ras (R=0.76+/-0.37), under various conditions of cell growth. Very few (if any) lipid bodies and vesicles were detected in the cytoplasmic or membrane compartments of 3T3ras cells with R<0.4, while only intramembrane lipid vesicles were associated with moderate R values. Involvement of phosphatidylcholine hydrolysis in ML generation was demonstrated by selective inhibition of endogenous phospholipase C (PC-plc) or by exposure to bacterial PC-plc. This study indicates that: (1) both cytoplasmic lipid bodies and membrane vesicles (possibly in mutual dynamic exchange) may contribute (although to a different extent) to ML signals; and (2) high levels of ras-transfection either inhibit ML formation or facilitate their extrusion from the cell.

Absolute metabolite quantification by in vivo NMR spectroscopy: IV. Multicentre trial on MRSI localisation tests.

The difference between the experimental and theoretical spatial response function (SRF) of a narrow tube with water is used for a localization test for magnetic resonance spectroscopic imaging (MRSI). From this difference a quantitative performance parameter is derived for the relative amount of signal within a limited region in the field of view. The total signal loss by the MRSI experiment and eddy currents is described by a parameter SL derived from the signal intensities of two echoes. Results of a European multi-centre trial show that this approach is suited for assessment of MRSI localization performance.

Absolute metabolite quantification by in vivo NMR spectroscopy: V. Multicentre quantitative data analysis trial on the overlapping background problem.

The goal of this study was to establish the best approach for quantifying nuclear magnetic resonance (NMR) lines, that in the frequency domain are overlapping with broad, unwanted background features. To perform the quantitative data analysis in a controlled way, test signals were designed and utilised, derived from two different real-world in vivo nuclear magnetic resonance signals. One of the main conclusions of the study was that the quantification methods currently available to the biomedical research groups can deliver the correct values of the quantitative parameters, but that great care should be taken in using optimal input parameters for the computer programs concerned.

Lower levels of 1H MRS-visible mobile lipids in H-ras transformed tumorigenic fibroblasts with respect to their untransformed parental cells.

High resolution 1H MRS studies report increased mobile neutral lipid (MNL) signals in transformed and malignant as well as in some in vitro cultured embryonic cells. Nature, subcellular localization and biological function of MNL are still under debate. This work was aimed at assessing alterations induced in MNL signals of NIH-3T3 mouse embryo fibroblasts by transformation with human HJ-ras oncogene. Lower MRS-visible MNL levels were unexpectedly detected in ras-transformed, in vivo tumorigenic fibroblasts, with respect to their untransformed and non-tumorigenic parental cells. MRS, gas chromatography and chemical analysis on cells and their lipid extracts indicated that these spectral differences could hardly be attributed to different triacylglycerol, free fatty acids and total cholesterol levels or to changes in the fatty acyl degree of unsaturation and average chain length. Additional, possibly more relevant mechanisms of regulation of MNL mobility may implicate the extensive morphogenetic changes and reorganization of cytoskeleton components (notably actin) associated with ras-transformation.

Potential role of in vitro 1H magnetic resonance spectroscopy in the definition of malignancy grading of human neuroepithelial brain tumours.

The increasing sensitivity of neuro-imaging in the diagnosis of brain expanding lesions is not directly related to biopathological specificity and new technological approaches are under study. In particular Magnetic Resonance Spectroscopy (MRS) allows evaluation of some biochemical pathways whose metabolic alterations may be correlated with the nature and malignancy grading of primary brain tumours. In the present study the author performed an in vitro high field 1H MRS (9.4 and 14.1 T) analysis of specimens obtained from stereotactic biopsy or microsurgical removal of primary brain tumours. Different samples derived from heterogeneous areas and/or infiltrated perilesional regions were examined. This study was principally focused on malignancy grading of gliomas and its correlation with the ratio (R) between the resonance band arising from choline containing compounds (between 3.14 and 3.35 ppm) and the total creatine signal (3.0 ppm). Analyses allowed significant discrimination between astrocytomas (R = 2.4 +/- 0.6) and glioblastoma (GBM) (R = 4.4 +/- 1.3) [p < 0.002]; however the results did not allow discrimination between differentiated and anaplastic astrocytomas. The GBM showed the largest spread of values corresponding to their higher level of tissue heterogeneity and de-differentiation. Studies on non astrocytic brain tumours indicated that even higher R values were exhibited by oligodendrogliomas, even in well differentiated forms (p < 0.02 with respect to GBM). Moreover, preliminary observations indicated that signals arising from other metabolites may also contribute to a differential diagnosis of different oncotypes. Among these glycine appears particularly relevant, since higher levels were measured for this amino acid in GBM with respect to both astrocytomas and oligodendrogliomas.

Evaluation of signal processing methods for the quantification of a multi-exponential signal: the glycogen 13C-1 NMR signal.

The 13C-1 NMR peak in proton-decoupled spectra of liver glycogen solution was quantitatively analyzed by three types of model-function fitting algorithms: iterative line-fitting in the frequency domain (MDCON); iterative least-squares fitting (VARPRO) in the time domain; and noniterative singular value decomposition-based analysis (HTLS), also in the time domain. Quantification results were compared with manual integration values. Performance of the algorithms was tested at different signal-to-noise ratios (S/N) of the glycogen C-1 peak. This was achieved by varying the number of scans summed prior to analysis. Since T2 relaxation in glycogen has been shown to be multiexponential [Overloop, K. et al. Magn. Reson. Med. 36, 45-51 (1996], the exact quantification of the C-1 glycogen signal requires a model function comprising a sum of Lorentzian components, each with a different broadening at the glycogen frequency. This paper focuses on the performances of the above methods to fit such a multicomponent resonance line. In the frequency domain, line fitting with two Lorentz lines gives good results at sufficiently high S/N. In the time domain, VARPRO performs better than HTLS because fixed values can be imposed to the linewidth of the components at the common C-1 frequency, thereby reducing convergence problems at low S/N.

Detection of phosphatidylcholine-specific phospholipase C in NIH-3T3 fibroblasts and their H-ras transformants: NMR and immunochemical studies.

Although evidence supports constitutive activation of phosphatidylcholine specific phospholipase C (PC-plc) in rastransformed fibroblasts, no studies have been devoted to measure the basal activity levels of this enzyme, its molecular characteristics and subcellular localization. This paper reports for the first time measurements of the activity of different enzymes responsible for PC hydrolysis (PC-plc; phospholipases A2 (pla2) and A1 (pla1)) in homogenates of murine NIH-3T3 fibroblasts (3T3) and their transformants obtained by human H-ras transfection (3T3ras). To this end, 31P NMR analyses were carried out on total cell homogenates, incubated in the presence of mixed diheptanoylphosphatidylcholine: sphingomyelin (DHPC:SM) unilamellar vesicles (SLUV), in which DHPC acts as a suitable substrate for water-soluble lipolytic enzymes. The basal PC-plc activity levels (0.66 +/- 0.14 and 0.38 +/- 0.10 nmol/10(6) cells.hour in 3T3 and 3T3ras fibroblasts, respectively),were substantially higher (over 30-50x) than those reported in the literature for normal mammalian cells (dog heart myocytes). Moreover the PC-plc activity was about 15-30 times lower than the overall PC deacylation activity in both clones. The use of high titer polyclonal antibodies, raised in a rabbit against bacterial PC-plc, allowed identification of one cross-reactive mammalian PC-plc component (M(r) 66 kD) in cell lysates of both 3T3 and 3T3ras fibroblasts, and detection, by indirect immunofluorescence, of its subcellular localization. In control 3T3 fibroblasts (in the late log-phase of growth) the enzyme was exclusively located in the cytosol, while in H-ras transformed cells it was massively exposed on the external side of the membrane. This new finding strongly suggests that the oncogenic product p2Iras is able to induce (or mediate) translocation of PC-plc across the plasma membrane of ras transformed cells, with possible implications not only on cell biochemistry (enhancement of PC-plc activity, and consequent production of intra- and extracellular PCho and accumulation of neutral lipids) but also on cell-cell interaction mechanisms which facilitate tumour invasion and metastasis of oncogene-transformed cells.