Cancer Research Annual Meeting 2010 Telomeres
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Pérez, Y.
Right arrow Articles by Arús, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Pérez, Y.
Right arrow Articles by Arús, C.
[Cancer Research 62, 5672-5677, October 15, 2002]
© 2002 American Association for Cancer Research

Biochemistry and Biophysics

Measurement by Nuclear Magnetic Resonance Diffusion of the Dimensions of the Mobile Lipid Compartment in C6 Cells1

Yolanda Pérez, Hana Lahrech, Miquel E. Cabañas, Ramon Barnadas, Manel Sabés, Chantal Rémy and Carles Arús2

Unitat de Ciències, Departament de Bioquímica i Biologia Molecular [Y. P., M. E. C., C. A.], Servei de Ressonància Magnètica Nuclear [M. E. C.], and Unitat de Biofísica de Medicina, Departament de Bioquímica i Biologia Molecular [R. B., M. S.], Universitat Autònoma de Barcelona, 08193 Cerdanyola del Vallès, Spain, and Unité Mixte Université Joseph Fourier-INSERM U438, Laboratoire de Recherche Correspondant du Commissariut a l’Energie Atomique, Centre Hospitalier Universitaire de Grenoble, BP 217, 38043 Grenoble Cedex 9, France [H. L., C. R.]


   ABSTRACT
Top
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The 1H spectrum of certain tumor cells, in vivo tumors, and their biopsies in vitro shows a narrow and intense resonance at 1.26 ppm, which has been assigned to the fatty acyl chain of triglycerides [nuclear magnetic resonance (NMR) visible mobile lipids, MLs]. We have used diffusion-weighted NMR spectroscopy to directly address the subcellular origin of MLs in the case of C6 cells in which lactate accumulation had been inhibited by prior iodoacetamide incubation. Borage oil and artificial lipid droplets were used as model systems of free and restricted diffusion, respectively. The characteristic diameter for the ML resonance compartment measured by NMR for the C6 cells was not significantly different from the one obtained with phase contrast microscopy (1.88 ± 0.04 µm from NMR versus 1.37 ± 0.33 µm from microscopy). We herewith provide direct and noninvasive evidence that the lipid signal at 1.26 ppm in C6 cells, which remains visible in long echo time (TE = 136 ms) experiments, mostly originates from subcellular structures with diameters of 1–2 µm, which correspond to the cytosolic lipid droplets that can be detected in optical microscopy preparations of the same cells.


   INTRODUCTION
Top
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
It has been known for some time that the 1H spectra of certain tumor and stimulated/transformed cells, tumor biopsies, experimental tumors, and in vivo/ex vivo human tumors show a narrow and intense resonance at 1.26 ppm that has been usually assigned to MLs3 (reviewed in Ref. 1 ). There is an accepted relationship between the presence of a large ML signal and the existence of necrotic areas in the tumor (2 , 3) . The noninvasive detection of necrosis would immediately suggest a high grade or malignancy in an untreated tumor (4 , 5) or could be used to monitor the response to therapy in a treated tumor (6) . On the other hand, ML resonances have been also detected in tumor biopsies without apparent necrosis (4) and in viable cultured cells (7) , where the presence of MLs is not associated with necrosis. Accordingly, a proper interpretation of the ML detection in a tumor in vivo may require the proper understanding of the locations of these MLs.

Essentially, two different ML locations have been proposed; the ML resonance would arise from fatty acyl chains in triglycerides embedded in the plasma membrane bilayer (PMDs) or from LDs, either cytosolic in intact cells or extracellular in a necrotic core (reviewed in Ref. 1 ). In a recent work (8) , high ML levels were associated in untransformed NIH-3T3 fibroblasts with both intramembrane amorphous lipid vesicles, 60–100 nm in diameter, and cytosolic osmiophilic lipid bodies (0.1–1 µm) surrounded by membrane. Barba et al. (7) showed that MLs were barely visible in log-phase C6 rat glioma cells, but that proliferation arrest caused by saturation density or acid extracellular pH induced the appearance of MLs in the 1H NMR spectra recorded with an echo time (TE) of 136 ms. The intensity of the ML peak height correlated with the calculated increase in the volume of Nile Red-positive cytosolic LDs detected by fluorescence microscopy. In contrast, other authors (9, 10, 11) have proposed that, in their cellular systems, the plasma membrane-associated neutral lipid domains account for most of the ML signal. Studies by ourselves and others (3 , 12) suggest that the LDs in extracellular necrotic areas can be very large, up to 8–10 µm in diameter, whereas intracellular LDs in viable cells (7) or cells undergoing apoptosis (6 , 13) stay within the 0.2–2-µm range. The origin and hence the clinical relevance of the MLs detected in tumors in vivo could be identified if the dimensions of the ML compartment were known or if differential properties were used to devise editing methodologies based on the compartment sizes of MLs.

DW-NMR is a noninvasive technique that produces information about the microstructure and size of the system (14 , 15) . One of the strengths of this method is its capability of monitoring whether the diffusion is free or restricted. We use the term "restricted diffusion" when the molecules are entrapped in a confining geometry. Further information on the DW-NMR method and its limitations can be obtained from recent reviews (15, 16, 17) .

The aim of the present study was to identify the subcellular origin of the ML resonance in 1H NMR spectra of C6 glioma cells. Because we were interested in mimicking the most common experimental conditions used in in vivo magnetic resonance spectroscopy, all diffusion studies were carried out using an echo time of 136 ms.


   MATERIALS AND METHODS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Materials
ALDs.
ALDs with known diameter distribution were prepared in our laboratory. Pro-Lipo S soybean phospholipids were purchased from Lucas Meyer (Hamburg, Germany). About 20% molar of the phospholipids of this raw material have a net negative charge at physiological pH (18) . The TG was natural borage oil (Gattefossé, Saint Priest, France). To attain a homogenous preparation with 1:1 w/w phospholipid:TG ratio, appropriate amounts of Pro-Lipo S and borage oil were mixed and stirred for 15 min. Deionized water was added to the mixture to obtain a partially homogenous aqueous lipid dispersion with a 2% w/w TG composition. The dispersion was emulsified with a high-pressure homogenizer, Microfluidizer 110S (Microfluidies, Newton, MA) (19) , rendering a macroscopically homogenous mixture containing liposomes and phospholipid:TG droplets (20) . Samples were fractionated by ultracentrifugation for 2 h at 100,000 x g at 4°C in a Sorvall AH-629 swinging bucket rotor (Wilmington, DE). The surface phase, a cake composed of phospholipid:TG droplets, was carefully removed and resuspended in water by gentle vortexing. The aqueous infranatant and the sediment were discarded. No free borage oil was observed.

Culture of C6 Glioma Cells.
C6 glioma cells obtained from European Collection of Animal Cell Cultures repository (Salisbury, United Kingdom) were grown as described previously (7) . For NMR experiments, the cells at saturation density from one 150-cm2 culture flask were washed once with PBS and then treated with 1 ml of trypsin-EDTA for 2 min, and cell viability was assessed by fluorescence microscopy (see below). The cell pellet was centrifuged for 3 min at 425 x g and resuspended twice in [2H]PBS (pH* 7.4; the pH measurement was uncorrected for the 2H isotope effect on glass electrodes). This final cell pellet of about 1 x 108 cells was resuspended in 500 µl of [2H]PBS and transferred to a 5-mm OD NMR tube (Wilmad Glass, Buena, NJ).

For metabolic inhibition of lactate production, C6 cells at saturation density were washed with PBS-IAM (IAM final concentration, 0.6 mM) three times before treatment with trypsin-EDTA solution. After trypsinization, the cell pellet was centrifuged for 3 min at 425 x g and resuspended twice in [2H]PBS (pH* 7.4) with IAM (final concentration, 0.8 mM). Finally, the cell pellet was resuspended in 500 µl of [2H]PBS with IAM (final concentration, 0.8 mM) before transfer to a 5-mm NMR tube.

Cell viability was tested before the NMR experiments and was always >95% after trypsinization. After the diffusion NMR measurements (6 h), at least 50% of the cells were found to be still viable. LD size was tested before and after the NMR experiments by staining cells with Nile Red (7) , and no significant change was detected.

DW-NMR Spectroscopy
All spectra were acquired using a Bruker ARX-400 spectrometer (Bruker Spectrospin, Wissembourg, France) equipped with a 5-mm broadband inverse geometry probe with shielded gradients of up to 0.5 T m-1 in the z direction. Diffusion experiments were carried out using a diffusion-weighted stimulated echo sequence (21) with square gradients with raise and decay times <100 µs. Water suppression was achieved by using a radiofrequency presaturation pulse of 5 mW during 1 s. Elimination of unwanted coherences was performed by the application of a 30-ms spoil gradient (50 mT m-1) during the mixing time (TM) interval and by using an eight-step phase cycling scheme (22) . TE was set to 136 ms, the repetition time (TR) was 2.5 s, and the sweep width (SW) was 5,000 Hz for borage oil and ALDs and 10,000 Hz for C6 cells. The duration ({delta}) of the pulsed gradient was set, depending on the sample, to 5 ms (ALDs) or 10 ms (borage oil and cells), and square gradients used in all cases.

ALD and Borage Oil NMR Experiments.
For diffusion experiments, the gradient factor b was set for each value of diffusion time to at least 10 values. b is calculated as b = {gamma}2G2{delta}2({Delta} - {delta}/3), where G is the gradient strength, {delta} is the length of the gradient pulses, {Delta} is the amount of time elapsing between the onset of the diffusion gradient pulses, and {gamma} is the gyromagnetic constant of the nucleus (21) .

For borage oil, NS = 8, TD = 4096, the diffusion time was varied between 41.67 and 996.67 ms, and the b values were set from 100 to 35,000 s mm-2. For ALDs, NS = 32, TD = 8192, the diffusion time was varied between 36.33 and 798.33 ms, and the b values were set from 100 to 100,000 s mm-2. All measurements were carried out at 308 K, except for borage oil, which was also analyzed at 298 K.

Spectra were processed by applying a line broadening of 1 Hz to the free induction decay (FID) and by zero filling the FID up to 16,384 data points before Fourier transformation. After the Fourier transformation, automatic phase and baseline correction of the spectra were performed before measuring the area of the resonances of interest. Calculation of areas by numerical integration was carried out with an ad hoc-written MatlabR program that facilitated the batch processing of files. To obtain the diffusion coefficient (D), the calculated areas were normalized with respect to the area at b = 0 and were then fitted to equation:

to obtain the diffusion coefficient (D). The root mean square displacement ({lambda}) was calculated by applying the Einstein equation:

where tdiff is the diffusion time, equal to {Delta} - {delta}/3 for the case of rectangular gradient pulses. For free diffusion, {lambda} increases linearly with . In the case of motion restriction, the relationship is not linear, with a leveling off where {lambda} reaches the size of the restricting compartment while the ADC decreases. For the calculation of the compartment size, only {lambda} values within the leveling region, corresponding to large tdiff values, are taken into account (23) . Assuming a spherical geometry for the compartment, the diameter of this compartment ({phi}) is calculated as {phi} = {lambda} (16 , 23) . In the case of nonuniform diameter distribution, the diameter measurement corresponds to the characteristic diameter {phi}c, which is related to the diameter distribution as reported by Lahrech et al. (22) .

The average root mean square displacement ({lambda}) was calculated by averaging the root mean square displacement values of those points in the falling part of ADC versus tdiff curve (24) .

Cell NMR Experiments.
C6 cell diffusion experiments with (n = 3) and without (n = 1) IAM treatment were collected as detailed above, except that NS = 64 and TD = 4096 data points. {delta} was set to 10 ms to reach high b values, tdiff was varied between 66.7 and 496.7 ms, and b values were set from 100 to 575,000 s mm-2. For each cell sample, the diffusion experiments lasted 6 h. Spectra were processed as detailed for the model solution. Chemical shift was referenced to total creatine at 3.03 ppm.

Pulse-and-acquire and spin-echo spectra from the cell pellet were recorded before and after diffusion experiments to monitor the samples. In pulse-and-acquire spectra, water suppression was achieved by using a presaturation pulse (1.5 s at 5 mW). Other acquisition parameters were as follows: TD = 4096 data points, NS = 64, SW = 4032 Hz, TR = 4 s. For spin-echo spectra, a spin-echo pulse sequence was used, with water suppression achieved by combining presaturation (2.5 s at 5 mW) and by replacing the excitation pulse with a jump and return scheme: Presat - 90° - {tau} - 90° - TE/2 - 180° - TE/2 - AQ (7 , 25) . TE was set to 136 ms to resolve the lactate and ML resonances by inverting their phases, and {tau} was set to 181.1 µs to place the excitation maximum at 1.26 ppm. The TR between scans was 6.5 s to allow for almost full relaxation of the resonances of interest. A total of 256 scans were recorded, with an acquisition time of ~30 min. All other parameters were as described for the pulse-and-acquire experiments.

OM
Cell Viability.
Cell viability was tested before and after NMR experiments by simultaneous staining with diacetate fluorescein and propidium iodide (26) . Briefly, the cell suspension (2 x 105 cells) in PBS was stained for 10 min with 6 µg ml-1 and 10 µg ml-1 (final concentration) of propidium iodide and diacetate fluorescein, respectively. Stained cells were monitored with a Leica DMRB fluorescence microscope (Leica, Barcelona, Spain). Green fluorescent cells were counted as viable, and red fluorescent cells were counted as nonviable. The viability was expressed as the percentage of viable cells over the total.

Diameter Distribution of LDs.
Nile Red staining was done essentially by following the protocol described by Greenspan et al. (27) . Briefly, ~105 cells were directly stained with 0.1 µg ml-1 (final concentration) of the fluorescent stain Nile Red (Sigma, Madrid, Spain) in PBS for 5 min. Nile Red-stained cells were studied with a Leica DMRB fluorescence microscope equipped with a Hamamatsu C-5310 color chilled CCD camera (Hamamatsu Photonics, Hamamatsu, Japan) (7) .

For each diffusion experiment involving cells at saturation density (postconfluence phase), we measured the LD size in an aliquot of the cell sample. From the diameter values obtained, we calculated the distribution of diameters, the distribution of volumes, and the cumulative distribution of the volumes. No changes in droplet size as measured from fluorescence microscopy happened during the NMR experiments. The size of trypsin-harvested cells was also measured in Nile Red-stained preparations.

The size of the ALDs was determined with the same microscope settings detailed for cells but using phase contrast instead of fluorescence. Briefly, an aliquot of ALDs was placed in a Neubauer chamber 1 mm deep to avoid flattening the droplets when covering the sample with the coverslip. Phase contrast pictures were taken combining an oil immersion x 100 objective (N.A. = 1.3) with a CCD camera as before.

To compare this diameter distribution with the characteristic diameter ({phi}c) obtained by DW-NMR, we had to volume weight the distribution. Briefly, the frequency of each bin in the diameter distribution was multiplied by the volume corresponding to its central diameter. The weighted frequency obtained this way was then renormalized to unit area and then fitted to a Gaussian function. The average diameter of the volume-weighted distribution could be then compared with the one obtained by DW-NMR measurement, {phi}c (22) . Numerical calculations (data not shown) were carried out to assess the largest PMD contribution that would not be resolvable by DW-NMR from the LD compartment under our present experimental conditions.

Statistical Analysis
The sizes determined by DW-NMR and OM were compared by using the unpaired Student’s t test. The diameter distributions obtained by OM before and after NMR experiments were compared using the unpaired z test (28) . Fits were performed by using a nonlinear least squares iterative method (Sigmaplot; SPSS, Inc., Chicago, IL) based on the Marquardt algorithm, with r values >0.9. The uncertainty of the measured values is reported when n = 1; otherwise, values are reported as mean ± SD, when n = 3. Measurement errors were propagated throughout calculations as detailed elsewhere (29) . The significance level was set at P < 0.05 in all cases.


   RESULTS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
DW-NMR of Borage Oil and ALDs
The most intense signals in the spectra of ALDs were the fatty acid -(CH2)n- and -CH3 resonances at 1.26 and 0.91 ppm, respectively. Also, signals from unsaturated fatty acyl chains were observed (data not shown). The sample contained a distribution of LDs as detailed below, made of a phospholipid bilayer spherical container filled with borage oil.

Fig. 1Citation shows the D and ADC, as well as the mean square displacement of fatty acyl CH2 resonances in borage oil (n = 1) and ALDs (n = 3), as a function of tdiff. The data show (Fig. 1C)Citation that the diffusion coefficient of the borage oil is not sensitive to the diffusion time (free diffusion), but the diffusion coefficient of the ALDs depends on the diffusion time (restricted diffusion). The D values obtained from the short tdiff points may be slightly affected by a breakdown of the "narrow pulse" approximation ({delta} << {Delta}), because for ALD {Delta} is 41.67 ms and {delta} is 10 ms for the shortest tdiff. The plateau of ADC for ALDs at long tdiff could be caused by self diffusion of the droplets themselves. The values in this long tdiff region have not been used for the calculation of {lambda}. The D values obtained for borage oil at 298 K, 1.46 ± 0.25 · 10-5 mm2s-1, are similar to those in the literature, e.g., trilinolein, -CH2 1.2 x 10-5 mm2s-1 at 298 K (30) and "oil," 3 x 10-5 mm2s-1 at 298 K (31) .



View larger version (39K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 1. A, phase-contrast image of ALDs. B, left, ALD diameter distribution; right, ALD volume-weighted diameter distribution. The lines show the Gaussian fitting. C, D of borage oil and average ADC (<ADC>) of ALDs versus tdiff; D, average {lambda}2 versus tdiff for the ML resonance in borage oil ({triangledown}, n = 1) and ALDs ({circ}, n = 3). Inset, an expanded view of the first points in D. We show {lambda}2 instead of the root mean square displacement to be able to overlay the plot of D, average ADC, and {lambda}2 versus tdiff. Bars, SD (not shown when smaller than symbol size).

 
Measurement of ALDs with phase contrast microscopy gave an average diameter of 2.32 ± 0.97 µm. The corresponding average volume-weighted diameter was 3.19 ± 0.86 µm. Assuming a spherical geometry, the characteristic diameter for the ML compartment measured by DW-NMR for the ALDs (3.41 ± 0.06) was not significantly different from that obtained with phase contrast microscopy (Table 1)Citation .


View this table:
[in this window]
[in a new window]

  		Table 1 Average characteristic diameter (<{phi}c>) of the compartments calculated for a spherical geometry by DW-NMR and measured by OM

 
Fig. 1BCitation shows the ALD diameter distribution and its corresponding volume-weighted diameter distribution. Images of the ALDs were obtained and measured before and after the diffusion experiments. It was found that the size distribution remained unchanged, possibly because of electrostatic repulsion between vesicle monolayers constituted by phospholipids.

C6 Glioma Cells
Fluorescence Microscopy.
The fluorescence microscopy of the C6 glioma cells demonstrated the presence of LDs in the cytoplasm of the C6 cells in agreement with previous observations (7) . The lipid fluorescent marker Nile Red was used for staining. The average diameter of LDs was 1.09 ± 0.43 µm for untreated cells and 1.12 ± 0.38 µm for cells treated with IAM. The average volume-weighted diameter for C6 LDs was estimated by fitting the volume-weighted diameter distribution with a Gaussian function. A value of 1.38 ± 0.46 µm was found for untreated cells, and a value of 1.37 ± 0.33 µm was found for cells treated with IAM (Table 1)Citation .

DW-NMR Spectroscopy.
The pulse-and-acquire and spin-echo spectra of C6 cells at saturation density with and without treatment with IAM are shown in Fig. 2Citation . C6 glioma cells at saturation density show clear ML peaks at 1.26 ppm (methylene groups of fatty acyl chains) and 0.91 ppm (terminal methyl of fatty acyl chains). As shown in reference (7) , these signals are not observable for cells in the log phase of growth. We can also see the creatine and TMA peaks. As expected, the spin-echo spectrum from IAM-treated cells does not exhibit the inverted lactate signal.



View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 2. 1H NMR spectra of C6 cells at saturation density, measured as described in "Materials and Methods." A and B, cells without treatment with IAM. C and D, cells treated with IAM. Spectra were acquired using a pulse-acquire sequence (A and C) and a spin-echo sequence with water presaturation (B and D). The range from 0.5 to 5.5 ppm is shown. Lac, lactate; FA, methyl and methylene groups of fatty acyl chains (see Refs. 8 and 7 for assignments); Cr, total creatine.

 
The ADCs and root mean square displacements for creatine, TMAs, and ML signals for IAM-treated C6 cells are plotted in Fig. 3Citation , where Nile Red staining of C6 cells and the diameter distribution of cytosolic LDs are also shown. The ADC for creatine and TMAs in treated and untreated C6 cells exhibit the same behavior; thus, for simplicity, only data from IAM-treated cells are shown. Fig. 3Citation clearly shows that diffusion is restricted because the ADC depends on tdiff. Results for C6 cells are summarized in Table 1Citation .



View larger version (37K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 3. A, fluorescence micrograph of C6 cells stained with Nile Red at yellow-gold fluorescence. Shown in black and white, droplets appear as white spots (bar, 10 µm); B, left, C6 ML droplets diameter distribution; right, C6 ML droplets volume weighted diameter distribution; the lines show the Gaussian fit. C, average ADC versus tdiff for the ML ({circ}), creatine ({square}), and TMA ({triangledown}) resonances in IAM-treated C6 glioma cells (n = 3). C, inset, an expanded view of average ADC versus tdiff for the ML ({circ}) resonance. D, average mean square displacement for the ML ({circ}), creatine ({square}), and TMA ({triangledown}) resonances in IAM-treated C6 glioma cells (n = 3). The dotted lines indicate the average mean square displacement used to calculate the compartment size. Bars, SD (not shown when smaller than symbol size). Note that <{lambda}2> (D) is displayed in log-scale.

 
The measurement of the size of trypsin-harvested C6 cells with microscopy yielded a cellular size of ~14 µm, and no significant difference in cell diameter was found between cells treated and untreated with IAM (Table 1)Citation . The compartment sizes calculated from diffusion experiments for creatine and TMAs are the same as the cell sizes measured by light microscopy. This strongly suggests that the diffusion of these metabolites is restricted by the cell plasma membrane. Finally, using this approach, the characteristic diameter of the ML resonance compartment calculated from the mean square displacement of the -CH2-signal in IAM-treated cells was 1.88 ± 0.04 µm, which is not statistically different from the value obtained by fluorescence microscopy for cytosolic LDs, 1.37 ± 0.33 µm.

Additionally (results not shown), the attenuation expected for the ML signal for a compartment of a characteristic diameter of 1.88 µm (LDs) or 80 nm (PMDs; Refs. 8 , 32 ) was calculated according to Tanner and Stejskal (23) for the longest diffusion time used by us (496.7 ms). It was found that the contribution of 80-nm domains should produce a significant change in the calculated {phi}c, when its contribution to the total ML volume in the cell exceeds 40%. This suggests a bottom limit to the detection of PMD contribution to the ML signal under our present experimental conditions.


   DISCUSSION
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
ALDs were used as a control to set up the experimental conditions and assess the methodology. Because the NMR signal is proportional to the amount of each contributing species, which in turn is proportional to the volume, the ALD diameter distribution obtained by OM was volume weighted. This weighting allowed us to calculate the characteristic diameter (22) . The characteristic diameter for ALDs obtained by DW-NMR (3.41 ± 0.06 µm) was not significantly different from the value obtained by OM (3.19 ± 0.86 µm). The extremely good agreement between these results validates the use of DW-NMR to evaluate the restriction to the diffusional motion and to calculate the size of metabolite compartments under our experimental conditions.

In our work on C6 cells, we found that the presence of the overlapping methyl signal arising from lactate (see Fig. 2Citation ) hampered the accurate measurement of the ML compartment size when using DW-NMR (33) . Thus, it was mandatory for us to remove the lactate signal contribution from the spectra if we wanted to have an accurate measure.

In this respect, in a recent work Lahrech et al. (22) have minimized the contribution of lactate to the 1.26 ppm signal in DW-NMR studies of in vivo C6 rat brain tumors. This was carried out by fitting their experimental data to Eq. A only for data points acquired at b > 5000 s mm-2. It can be assumed that lactate diffusion was unrestricted, because the metabolite would be mostly extracellular in necrotic cores. Then, a lactate-free diffusion coefficient in the range of 0.38–1.0·10-3 mm2s-1 was assumed. The combination of the lactate attenuation attributable to stimulated echo sequence (TE = 68 ms) and diffusion losses yields an attenuation factor in the range of 4.6–6.7. They found that the mean characteristic diameter for the ML resonance compartment of C6 glioma in rats was 4.27 ± 0.71 µm. Unfortunately, no insight could be produced from microscopy studies on the diameter distribution of the LDs or their average diameter in the tumor volume sampled for NMR; thus, it is difficult to assess the level of agreement between the DW-NMR values and the values found by using other techniques such as electron microscopy or OM. But we have to take into account that the amount of lactate present in highly necrotic C6 tumors may be lower than in more viable ones (34) and accordingly would perturb less the estimated characteristic diameter. Therefore, when attempting a similar approach to the one described by Lahrech et al. (22) with C6 cells at saturation density, we found a mean spherical diameter of 4.28 ± 0.12 µm, in clear disagreement with the volume-weighted diameter obtained from fluorescence microscopy, 1.36 ± 0.02 µm (33) . Our data from intact C6 cells would then suggest that we may have had yet some contribution from lactate, probably because lactate diffusion is restricted similarly to other cytosolic metabolites, and it is still visible at b values >=5000 s mm-2 in our experimental conditions, or because of higher lactate concentration being accumulated in the C6 cell pellet.

In view of the above-mentioned difficulties, we preferred the much simpler approach of inhibiting lactate production in the C6 cells. IAM is known to impair lactate production by inhibiting the enzymes glucose-6-phosphate isomerase, glyceraldehyde-3-phosphate dehydrogenase, and enolase (35) . Treating cells with IAM reduces to undetectable levels the lactate content in its NMR spectra (Fig. 2)Citation . Using this approach, the characteristic diameter of the ML resonance compartment, calculated from the mean square displacement of the -CH2-signal in IAM-treated C6 cells, was 1.88 ± 0.04 µm, which was not statistically different from the value obtained by fluorescence microscopy, 1.37 ± 0.33 µm. It is worth noting the large difference (one order of magnitude) in the characteristic displacement, and thus, in the restriction size, shown by creatine and TMA on one side and the -CH2-MLs on the other (see Table 1Citation ). The former correlates well with the cell size; the latter indicates a much smaller compartment and correlates with the size of the cytosolic LDs.

To date, there have been no reports on the existence of PMDs in C6 cells; nonetheless, we may try to approximate the value of the fractional volumes of MLs in LDs and PMDs of C6 cells from the values reported by Ferretti et al. (8) for NIH-3T3 fibroblasts. They report a total volume of <=0.2 µm3/cell for PMDs, whereas the LD total volume amounts to 4–8 µm3/cell. Using their PMD total volume and our previously reported LD total volume in C6 cells (~6 µm3/cell; Ref. 7 ), we can estimate that the maximal signal contributed by the PMD compartment would be 3% of the total ML signal. Taking into account the restriction size measured by DW-NMR in this work and the limits mentioned in the results section for the maximal PMD contribution, ~40%, there would still be a gap (between 3 and 40%) for the maximal contribution of PMDs to the ML resonance detected in C6 cells. Further experiments at longer diffusion times will be needed to address this point in C6 cells and other cells and tissues for which MLs are detected.

In summary, in the presence of cytosolic LDs (1–2 µm diameter), as in the case of the saturation density C6 cells, the ML resonance at 1.26 ppm in 1H NMR spectra arises mainly from these large cytosolic droplets. Although the presence of intramembrane lipid particles cannot be discarded, our present results suggest that their contribution to the NMR pattern of tumoral C6 cells recorded at an echo time of 136 ms would be <40% of the ML resonance.

In conclusion, we provide direct noninvasive evidence that the lipid signal at 1.26 ppm that remains visible in long echo time (TE = 136 ms) NMR experiments recorded from C6 cells mainly originates from structures with an approximate characteristic diameter of 1.88 ± 0.04 µm, which correspond to cytosolic LDs.


   ACKNOWLEDGMENTS
 
We thank Dr. Michel Décorps for useful advice, guidance, and discussion; Dr. Salvador Bartolomé (LAF-Universitat Autònoma de Barcelona) for help in the acquisition and processing of the microscope images; and Dr. Teodor Parella (Servei de Ressonància Magnètica Nuclear, Universitat Autònoma de Barcelona) for helpful advice in setting up the NMR experiments. We also thank Dr. Jonathan Coles for assistance with the English language and helpful suggestions.


   FOOTNOTES
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 This work has been funded by projects CICYT SAF99-0101 and Generalitat de Catalunya (ACI98-21 and SGR2001-194) and grants from La Ligue contre le Cancer and l’Association pour la Recherche sur le Cancer. Back

2 To whom requests for reprints should be addressed, at Departament de Bioquímica i Biologia Molecular, Universitat Autònoma de Barcelona, Edifici Cs, 08193 Cerdanyola del Vallès, Spain. Phone: 34-93-581-1257; Fax: 34-93-581-1264; E-mail: Carles.Arus{at}uab.es Back

3 The abbreviations used are: ML, nuclear magnetic resonance visible mobile lipid; ADC, apparent translational diffusion coefficient; ALD, artificial lipid droplet; DW-NMR, diffusion-weighted NMR; IAM, iodoacetamide; LD, lipid droplet; NMR, nuclear magnetic resonance; OM, optical microscopy; PMD, plasma membrane domain; TMA, trimethylamine-containing compound, TG, triacylglycerol. Back

Received 12/17/01. Accepted 8/15/02.


   REFERENCES
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Hakumäki J. M., Kauppinen R. A. 1H NMR visible lipids in the life and death of cells. Trends Biochem. Sci., 25: 357-362, 2000.[Medline]
  2. Kuesel A. C., Sutherland G. R., Halliday W., Smith I. C. P. 1H MRS of high grade astrocytoma: mobile lipids accumulation in necrotic tissue. NMR Biomed., 7: 149-155, 1994.[Medline]
  3. Rémy C., Foulhié N., Barba I., Sam-Laï E., Lahrech H., Cucurella M. G., Izquierdo M., Moreno A., Ziegler A., Massarelli R., Décorps M., Arús C. Evidence that mobile lipids detected in rat brain glioma by 1H nuclear magnetic resonance correspond to lipid droplets. Cancer Res., 57: 407-417, 1997.[Abstract/Free Full Text]
  4. Kuesel A. C., Donnelly S. M., Halliday W., Sutherland G. S., Smith I. C. P. Mobile lipids and metabolic heterogeneity of brain tumours as detectable by ex vivo 1H MR spectroscopy. NMR Biomed., 7: 172-180, 1994.[Medline]
  5. Lin A., Blum S., Mamelak A. N. Efficacy of proton magnetic resonance spectroscopy in clinical decision making for patients with suspected malignant brain tumors. J. Neuro-Oncol., 45: 69-81, 1999.[Medline]
  6. Hakumäki J. M., Poptani H., Sandmair A. M., Ylä-Herttuala S., Kauppinen R. A. 1H MRS detects polyunsaturated fatty acid accumulation during gene therapy of glioma: implications for the in vivo detection of apoptosis. Nat. Med., 5: 1323-1327, 1999.[Medline]
  7. Barba I., Cabañas M. E., Arús C. The relationship between NMR-visible lipids, lipid droplets and cell proliferation in cultured C6 cells. Cancer Res., 59: 1861-1868, 1999.[Abstract/Free Full Text]
  8. Ferretti A., Knijn A., Iorio E., Pulciani S., Giambenedetti M., Molinari A., Meschini S., Stringaro A., Calcabrini A., Freitas I., Strom R., Arancia G., Podo F. Biophysical and structural characterization of 1H-NMR-detectable mobile lipid domains in NIH-3T3 fibroblasts. Biochim. Biophys. Acta, 1438: 329-348, 1999.[Medline]
  9. Rosi A., Luciani A. M., Matarrese P., Arancia G., Viti V., Guidoni L. 1H-MRS lipid signal modulation and morphological and ultraestructural changes related to tumor cell proliferation. Magn. Res. Med., 42: 248-257, 1999.[Medline]
  10. Lamanna R., Motta A., Romano R., Rainaldi G., Flamma F., Pentimalli M., Tancredi T., Indovina P. L., Santini M. T. Forced adhesive growth of K562 leukemic cells that normally grow in suspension induces variations in membrane lipids and energy metabolism: a proton NMR study. J. Biomed. Materials Res., 46: 171-178, 1999.
  11. Foley L. M., Towner R. A., Painter D. M. In vivo image-guided 1H-magnetic resonance spectroscopy of the serial development of hepatocarcinogenesis in an experimental animal model. Biochim. Biophys. Acta, 1526: 230-236, 2001.[Medline]
  12. Fouilhé N. . Caractérisation des Lipides Mobiles Détectés par Spectroscopie RMN du Proton dans un Modèle de Gliome Intracérebral chez le Rat, PhD thesis, Université Joseph Fourier 1997.
  13. Finstad H. S., Dyrendal H., Myhrstad M. C., Heimli H., Drevon C. A. Uptake and activation of eicosapentaneoic acid are related to accumulation of triacylglycerol in Ramos cells dying from apoptosis. J. Lipid Res., 41: 554-564, 2000.[Abstract/Free Full Text]
  14. Posse S., Cuenod C. A., Le Bihan D. Human brain: proton diffusion MR spectroscopy. Radiology, 188: 719-725, 1993.[Abstract/Free Full Text]
  15. Nicolay K., Braun K. P. J., de Graaf R. A., Dijkhuizen R. M., Kruiskamp M. J. Diffusion NMR spectroscopy. NMR Biomed., 14: 94-111, 2001.[Medline]
  16. Price W. S. Pulsed-field gradient nuclear magnetic resonance as a tool for studying translational diffusion. Part I. Basic theory. Concepts Magn. Reson., 9: 299-336, 1997.
  17. Price W. S. Pulsed-field gradient nuclear magnetic resonance as a tool for studying translational diffusion. Part II. Experimental aspects. Concepts Magn. Reson., 10: 197-237, 1997.
  18. Barnadas R. . Optimació de Formulacions amb Fosfolípids per a Ús Dermatològic. PhD thesis, Departament de Bioquímica i Biologia Molecular, Unitat de Biofisica, Universitat Autònoma de Barcelona 1999.
  19. Sjostrom B., Westesen K., Bergenstahl B. Preparation of submicron drug particles in lecitin-stabilized O/W emulsions. II. Characterization of cholesteryl acetate particles. Int. J. Pharmaceut., 94: 89-101, 1993.
  20. Hamilton J. A. Interaction of triglycerides with phospholipids: incorporation into the bilayer structure and formation of emulsions. Biochemistry, 28: 2514-2520, 1989.[Medline]
  21. Tanner J. E. Use of the stimulated echo in NMR diffusion studies. J. Chem. Phys., 52: 2523-2526, 1970.
  22. Lahrech H., Zoula S., Farion R., Rémy C., Décorps M. In vivo measurement of size of lipid droplets in intracerebral gliomas in rat brain. Magn. Reson. Med., 45: 409-414, 2001.[Medline]
  23. Tanner J. E., Stejskal E. O. Restricted self-diffusion of protons in colloidal systems by the pulsed-gradient, spin-echo method. J. Chem. Phys., 49: 1768-1777, 1968.
  24. Van Zijl P. C. M., Davis D., Moonen C. T. W. Diffusion spectroscopy in living systems Gillies Robert J. eds. . NMR in Physiology and Biomedicine, 185-198, Academic Press San Diego 1994.
  25. Plateau P., Guéron M. Exchangeable proton NMR without base-line distortion using new strong-pulse sequences. J. Am. Chem. Soc., 104: 7310-7311, 1982.
  26. Jones K. H., Senft J. A. An improved method to determine cell viability by simultaneous staining with fluorescein diacetate-propidium iodide. J. Histochem. Cytochem., 33: 77-79, 1985.[Abstract]
  27. Greenspan P., Mayer E. P., Fowler S. D. Nile Red: a selective fluorescent stain for intracellular lipid droplets. J. Cell Biol., 100: 965-973, 1985.[Abstract/Free Full Text]
  28. Rice J. . Mathematics Statistics and Data Analysis, Waldsworth & Brooks/Cole Advanced Books & Software Pacific Grove, CA 1988.
  29. Bevington P. R. . Data Reduction and Error Analysis for the Physical Sciences, McGraw-Hill Book Company New York 1969.
  30. Weybright P., Millis K., Campbell N., Cory D. G., Singer S. Gradient, high-resolution, magic angle spinning 1H magnetic resonance spectroscopy of intact cells. Magn. Reson. Med., 39: 337-344, 1998.[Medline]
  31. Assaf Y., Cohen Y. In vivo and in vitro bi-exponential diffusion of N-acetyl aspartate (NAA) in rat brain: a potential structural probe?. NMR Biomed., 11: 67-74, 1998.[Medline]
  32. Di Vito M., Lenti L., Knijn A., Iorio E., D’Agostino F., Molinari A., Calcabrini A., Stringaro A., Meschini S., Arancia G., Bozzi A., Strom R., Podo F. 1H NMR-visible mobile lipid domains correlate with cytoplasmic lipid bodies in apoptotic T-lymphoblastoid cells. Biochim. Biophys. Acta, 1530: 47-66, 2001.[Medline]
  33. Pérez Y., Lahrech H., Cabañas M. E., Sabés M. E., Rémy C., Décorps M., Arús C. NMR diffusion measurement of the size of the intracellular compartment containing the 1.26 ppm mobile lipid resonance in C6 cells. Magn. Reson. Materials Phys. Biol. Med., 8: 228 1999.
  34. García-Martín M. L., Herigault G., Rémy C., Farion R., Ballesteros P., Coles J. A., Cerdán S., Ziegler A. Mapping extracellular pH in rat brain gliomas in vivo by 1H magnetic resonance spectroscopic imaging: comparison with maps of metabolites. Cancer Res., 61: 6524-6531, 2001.[Abstract/Free Full Text]
  35. Lust D. W., Schwartz J. P., Passoneau J. V. Glycolytic metabolism in cultured cells of the nervous system. 1. Glucose transport and metabolism in the C6-glioma cell line. Mol. Cell. Biochem., 28: 169-176, 1975.



This article has been cited by other articles:


Home page
 J. Physiol.Home page
I. Barba, L. Chavarria, M. Ruiz-Meana, M. Mirabet, E. Agullo, and D. Garcia-Dorado
Effect of intracellular lipid droplets on cytosolic Ca2+ and cell death during ischaemia-reperfusion injury in cardiomyocytes
J. Physiol., March 15, 2009; 587(6): 1331 - 1341.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Pérez, Y.
Right arrow Articles by Arús, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Pérez, Y.
Right arrow Articles by Arús, C.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online