www.bba-direct.com

Biochimica et Biophysica Acta 1634 (2003) 1–14

Triacsin C inhibits the formation of 1H NMR-visible mobile lipids and

lipid bodies in HuT 78 apoptotic cells

Egidio Iorioa, Massimo Di Vitoa, Francesca Spadaroa, Carlo Ramonia, Emanuela Lococob,Roberto Carnevaleb, Luisa Lentib, Roberto Stromc, Franca Podoa,*

aLaboratory of Cell Biology, Istituto Superiore di Sanita, Viale Regina Elena 299, 00161 Rome, ItalybDepartment of Experimental Medicine and Pathology, University of Rome ‘‘La Sapienza’’, 00185 Rome, Italy

cDepartment of Cellular Biotechnology and Haematology, University of Rome ‘‘La Sapienza’’, 00161 Rome, Italy

Received 7 February 2003; received in revised form 9 June 2003; accepted 24 July 2003

Abstract

Nuclear magnetic resonance-visible mobile lipids (ML) have been reported to accumulate during cell apoptosis in vitro and in vivo. The

biogenesis, biochemical nature and structure of these lipids are still under debate. In this study, a human lymphoblastoid cell line, HuT 78,

was induced to apoptosis by exposure to anti-Fas monoclonal antibodies (a-Fas mAb) followed by incubation for different time intervals (1–

24 h, hypodiploid cell fraction, H, varying from 1% to over 60%) either in the presence or in the absence of 5.0 AM Triacsin C (TRC),

specific inhibitor of long-chain acyl-CoA synthetase (ACS). The increase of ML in apoptotic cells correlated linearly with H and was

associated with: (a) accumulation of intracellular lipid bodies, detected by confocal laser scanning microscopy in lipophilic dye-stained cells;

(b) increases, detected by thin-layer chromatography in total lipid extracts, in the relative abundance of triacylglycerides (TAG) and

cholesteryl esters (CE), with corresponding decreases of phospholipids (PL). TRC completely abolished both ML and lipid body formation in

anti-Fas-treated apoptotic cells, with concomitant reversion of TAG, CE and PL to control levels, but did not alter cell viability nor did it

inhibit apoptosis. ML signals detected during anti-Fas-induced apoptosis therefore appear to originate from neutral lipids assembled in

intracellular lipid bodies, synthesised from cellular acyl-CoA pools.

D 2003 Elsevier B.V. All rights reserved.

Keywords: Apoptosis; Lymphoblast; Mobile lipid; NMR; Triacsin C; Fas

1. Introduction

Apoptosis is a tightly regulated form of physiological cell

death in multicellular organisms, dependent upon the ex-

pression of some cell-intrinsic, genetically controlled sui-

cide machinery [1,2]. This cell death programme not only

plays fundamental physiological roles in embryogenesis,

tissue homeostasis, immune response and aging, but is also

1388-1981/$ - see front matter D 2003 Elsevier B.V. All rights reserved.

doi:10.1016/j.bbalip.2003.07.001

Abbreviations: a-Fas mAb, anti-Fas monoclonal antibodies; ACS, acyl-

CoA synthetase; CE, cholesteryl esters; CHOL, free cholesterol; CLSM,

confocal laser scanning microscopy; CoA, coenzyme A; DAG, diacylgly-

cerides; FFA, free fatty acids; HPTLC, high-performance thin-layer

chromatography; ML, NMR-detectable mobile lipids; NMR, nuclear

magnetic resonance; PC, phosphatidylcholine; PCho, phosphocholine;

PEtn, phosphoethanolamine; PrI, propidium iodide; PL, phospholipids;

TAG, triacylglycerides; TRC, Triacsin C

* Corresponding author. Tel.: +39-6-4990-2686; fax: +39-6-4938-

7144.

E-mail address: fpodo@iss.it (F. Podo).

reputed to be involved in a number of pathological process-

es including neurodegenerative disorders, autoimmune dis-

eases, ischemia, oncogenesis and tumour response to

therapy [3–9].

Although a definite pattern of all the molecular events

leading to apoptosis is not yet available, an increasing level

of knowledge is accumulating on this phenomenon, at the

biological and biochemical level. 1H NMR spectroscopy has

recently added interesting novel information on the bio-

chemistry of programmed cell death, by allowing non-

invasive monitoring of alterations occurring in the energetic

state, as well as in phospholipid and glucose metabolism

[10–17] and by detecting the formation and accumulation

of mobile lipids (ML) in intact apoptotic cells [18–22]. The

production of these lipids, endowed of a sufficiently high

rotational tumbling motion and segmental flexibility of their

fatty acyl chains to be detected in the high-resolution

nuclear magnetic resonance (NMR) time window, is not a

peculiar feature of apoptotic cells, since the characteristic

E. Iorio et al. / Biochimica et Biophysica Acta 1634 (2003) 1–142

ML signals due to saturated (CH2)n segments (d=1.3 ppm)

of the fatty acyl chains, to their olefinic HCCH (5.3 ppm)

and to their terminal methyl groups (0.9 ppm), have been

detected for over two decades in a variety of tumour cells

and tissues ([23] and ref. therein), in activated lymphocytes

[24,25] and in embryo-derived cells [26,27]. The appear-

ance of ML signals in the spectra of intact cells is generally

attributed to the formation of non-bilayer lipid structures,

occurring either at the membrane level [23,26] or within

cytoplasmic compartments [20–22,27–31]. NMR-detect-

able ML have also been reported in tumour cells exposed

to lipophilic cationic compounds that cause mitochondrial

damage, and were found to be associated with the formation

of lipid droplets, which could be visualized by electron

microscopy or by staining with suitable lipophilic dyes [32].

The detection of ML in the NMR spectra of cells undergo-

ing programmed cell death has recently attracted consider-

able interest, in view of the possibility of using ML fatty

acyl signals to quantify, by a non-invasive technique, the

effects of apoptotic agents in cell cultures in vitro, as well as

in tissues in vivo [19–21,31]. An effective use of this

approach requires, however, a better elucidation of the

biogenesis and molecular organization of ML formed during

the apoptotic process.

Previous studies conducted in our laboratory on struc-

ture and subcellular localization of ML in Jurkat lympho-

blastoid cells induced to apoptosis by different drugs [21]

showed that the ML content correlated linearly with the

hypodiploid cell fraction, irrespective of the nature of the

inducing agent and of the phase of cell cycle arrest. The

observed increases in ML signals were associated with the

accumulation of cytoplasmic, osmiophilic lipid bodies

(diameterV1.0 Am) surrounded by its own membrane that

possesses intramembrane particles [21]. In substantial

agreement with these findings, Al-Saffar et al. [22] reported

that ML formation was associated with the synthesis of

triacylglycerides (TAG) and accumulation of cytoplasmic

lipid droplets in Jurkat cells induced to apoptosis by

continuous exposure to anti-Fas monoclonal antibodies

(a-Fas mAb). Furthermore, since 31P NMR analyses had

shown a simultaneous drop in the total content of phos-

phatidylcholine (PC), alterations in PC metabolism have

been indicated as the possible source of TAG accumulation

in apoptosing cells.

The purpose of the present study was to further eluci-

date, in a human T lymphoblastoid cell line (HuT 78)

induced to programmed cell death by stimulation of the

Fas/APO-1 (CD95) receptor [33,34], the biochemical mech-

anisms underlying ML formation and the subcellular com-

partmentation of these components. To this end, we

investigated the changes occurring in ML signal intensities,

in neutral lipid composition and in lipid body formation in

cells undergoing apoptosis, incubated with Triacsin C

(TRC), a selective inhibitor of long-chain acyl-coenzyme

A (acyl-CoA) synthetase (ACS, EC 6.2.1.3) [35,36] that

specifically inhibits the ATP-dependent, ACS-catalysed

activation of free fatty acids (FFA) into the corresponding

acyl-CoAs.

FFAþ CoAþ ATP !Mg2þ

ACSacyl-CoAþ AMPþ PPi

The rationale of the proposed experimental design was

that, since acyl-CoA formed in ACS-mediated reaction acts

as acyl donor in the biosynthesis of glycerolipids and of

cholesteryl esters (the two major components of intracellular

lipid bodies), NMR, biochemical and structural studies on

the effects of acyl-CoA synthesis inhibition would be

capable of providing relevant information on biogenesis

and nature of MLs in the apoptotic cells (preliminary report

in Ref. [37]).

2. Materials and methods

2.1. Chemicals and biochemicals

Formaldehyde, Nile Red, paraformaldehyde, poly-L-ly-

sine, propidium iodide (PrI), Triton X-100 and TRC from

Streptomyces were purchased from Sigma Chemicals, St.

Louis, MO, USA. BODIPY 493/503 was supplied by

Molecular Probes, Eugene, OR, USA. Mouse IgM anti-

human Fas mAb, clone CH-11, were from Coulter, Hialeah,

FL, USA.

2.2. Cell culture and induction of apoptosis

The stabilized human T lymphoblastoid cell line HuT 78

was maintained in complete RPMI 1640 medium (Gibco),

supplemented with 10% foetal calf serum (HyClone) and

grown at 37 jC in humidified atmosphere of 5% CO2/95%

air. Cell viability (>99%) was assessed by the trypan blue

exclusion test.

For induction of apoptosis, cells in the exponential

phase of growth were centrifuged at 600�g and the pellet

resuspended in PBS. Cells were then exposed for 2 h to a-

Fas mAb (1 Ag mAb/50�106 cells/ml). After this pulsed

treatment, the cells were 50-fold diluted in complete

medium and incubated at 37 jC for different time inter-

vals. The use of a short exposure to a relatively high

concentration of the inducing agent, diluted away during

the subsequent incubation, should allow a more detailed

study on the progression to apoptosis. As negative controls

for apoptosis (‘‘CTRL’’), other cell samples were exposed

to the same manipulation, but in the absence of a-Fas

mAb.

Apoptosis was detected by flow cytometry analysis of

PrI-stained cells according to a procedure described in detail

elsewhere [21,38,39]. In brief, 1�106 HuT 78 cells were

centrifuged, the pellet gently resuspended in 1.0 ml of a

hypotonic solution containing 50 Ag/ml PrI in 0.1% sodium

citrate plus 0.1% Triton X-100 and the tubes placed at 4 jCin the dark, overnight. The PrI fluorescence of individual

E. Iorio et al. / Biochimica et Biophysica Acta 1634 (2003) 1–14 3

nuclei was then measured using an EPICS XL flow cyto-

meter (Coulter) equipped with a 488 nm Argon laser lamp.

A multiparametric flow-cytometric assay discriminated and

quantified viable, apoptotic and necrotic cells via measure-

ments of forward and side light scatter [40]. The hypodip-

loid (sub-Go) apoptotic nuclei were thus identified and

clearly discriminated from necrotic ones and from the

narrower peak of cells in the Go/G1 phase, that possessed

normal diploid DNA content.

In order to investigate the effects of TRC, the cells

were exposed to this inhibitor (vehicle DMSO 0.1% v/v,

brought to final concentration of 5 AM in complete

medium). The inhibitor was present both during the 2-

h treatment with a-Fas mAb (or with the control medium)

and during the subsequent, variable incubation time inter-

vals, which preceded the actual NMR analyses. Exposure

to TRC did not alter cell viability of either control or

a-Fas-treated cells, did not induce apoptosis in control

cells, nor did it modify the time course of the a-Fas-

induced cell death programme.

2.3. NMR spectroscopy

Intact cells were counted, washed three times in PBS,

centrifuged at 600�g and resuspended in 700 Al of PBS

in D2O, before their transfer to a 5 mm NMR tube (40–

50�106 cells). 1H NMR experiments (25 jC) were

performed at 400 or 700 MHz (Bruker Avance 400 and

700). Analyses at 400 MHz on intact cells were carried

out using 60j flip angle pulses preceded by 1.00 s

presaturation for water signal suppression (interpulse delay

1.00 s, acquisition time 1.86 s, spectral width 11 ppm,

32000 data points, 128 scans). These conditions ensured

that the fatty chain (CH2)n/CH3 ratio was determined at

the magnetization equilibrium (as verified by preliminary

experiments). At 700 MHz, a small correction factor of

1.05 had to be introduced. A series of independent

experiments was also carried out at 200 MHz (Varian

Gemini 200).

Trimethylsilyl-2,2,3,3-D4-propionate (1 mM in D2O) was

used as external chemical shift standard. Signal assignments

were according to previous studies, carried out in our and in

other laboratories [21,27,41,42].

Deconvolution of signals under the one-dimensional 1H

NMR spectral profiles were performed using Varian

software for the spectra at 200 MHz and the Bruker

WIN-NMR software package for the spectra at 400 and

700 MHz.

2.4. Lipid extraction and analysis

Lipid extraction from cells was performed according to

Folch et al. [43]. Total lipid extracts were separated on silica

gel 60 plates (Merck) with a solvent system of toluene/

diethyl ether/ethanol (105:30:3, v/v/v). The zones on silica

gel corresponding to phospholipids (PL) were scraped off,

extracted with chloroform: methanol (2:1, v/v) and further

developed by high-performance thin-layer chromatography

(HPTLC) with a solvent system of chloroform/methanol/

acetic acid/water (50:37.5:3.5:2, v/v/v/v).

The areas containing individual phospholipids were

identified by co-migration of standards and were visualized

by staining with molybdenum blue reagent.

Neutral lipid analyses were performed using a solvent

system of hexane/diethyl ether/acetic acid (70:30:1, v/v/v)

and were detected by staining with 2% copper acetate

solution in 8% phosphoric acid and subsequent heating at

120 jC for 15 min. After about 3 min, free cholesterol

(CHOL) and cholesteryl esters (CE) yielded red spots on a

white background, which converted into pink-brown spots,

like the other lipids, after 10 min.

Relative quantification of individual lipid classes was

performed using the ‘‘Quantity One’’ BioRad software

program.

2.5. Confocal laser scanning microscopy (CLSM) analyses

For CLSM analyses, HuT 78 cells were seeded on poly-

L-lysine (0.01%, 40 min at room temperature), allowed to

attach for 15 min at 37 jC and washed with PBS. For Nile

Red analyses, the cells were fixed by 4% formaldehyde (10

min at room temperature) and then incubated for 15 min at

37 jC, with the lipid probe (100 ng/ml) [44,45]. For

BODIPY 493/503 analyses [46], the cells were fixed with

3% paraformaldehyde (30 min, 4 jC), permeabilised by

Triton X-100 (0.5%, 10 min at room temperature) and then

stained with the lipophilic dye (1:100; 20 min at room

temperature). For double fluorescence experiments, the

cells, after BODIPY staining, were incubated with PrI (20

min, 37 jC) to simultaneously detect morphological mod-

ifications at the nuclear level and cytoplasmic lipid body

formation. After washing in PBS, the samples were resus-

pended in the same buffered medium and immediately

analysed. For CLSM analyses, cells were mounted on the

microscope slide with the ProLong reagent (Molecular

Probes) and observations were performed using a Leica

TCS 4D apparatus, equipped with argon–krypton laser,

510 nm dichroic splitter, 515 nm long-pass filter and

double-dichroic splitters (488/568 nm) for double fluores-

cence. Image acquisition and processing were carried out

by using the SCANware and Multicolor Analysis (Leica

Lasertechnik GmbH, Heidelberg, Germany) and Photoshop

(Adobe Systems Inc., Mountain View, CA, USA) software

programs.

3. Results

3.1. a-Fas-induced cell apoptosis

Two-hour exposure of HuT 78 cells to a-Fas mAb

induced activation of the cell death programme, whose

E. Iorio et al. / Biochimica et Biophysica Acta 1634 (2003) 1–144

effects were monitored in time by flow cytometry determi-

nation of apoptotic hypodiploid nuclei in cells collected at

different time intervals between 1 and 24 h of post-treatment

incubation (Fig. 1).

While control cells maintained, under these condi-

tions, a very low apoptotic hypodiploid cell fraction

value (H=1.1F0.7%, n=18), cells treated with a-Fas

mAb progressed through increasing levels of apoptosis

(Table 1).

3.2. Formation of 1H NMR-visible ML during a-Fas-induced apoptosis

NMR spectroscopy allowed detection of the progressive

ML formation in HuT 78 lymphoblasts induced to apoptosis

by pulsed exposure to a-Fas mAb, as indicated by the

increase, at different times of incubation (1–24 h) after

treatment, of a major resonance centered at d=1.3 ppm, due

to the saturated fatty acyl chain methylene segments

((CH2)n). Fig. 2 (left traces) shows representative spectra

Fig. 1. Cell cycle analyses of HuT 78 cells exposed to a-Fas mAb and/or to TRC.

to apoptosis by 2-h exposure to a-Fas mAb (1 Ag mAb/50�106 cells/ml) and then

the absence (left) or in the presence (right) of 5.0 AM TRC. The cells were stained w

of events �10�2. (1) Hypodiploid apoptotic peak; (2) diploid peak; (3) S-phase;

of apoptotic cell preparations (‘‘a-Fas’’, at 2, 7 and 24 h after

treatment with a-Fas mAb) and of untreated control cells

(‘‘CTRL’’, bottom and top traces, 2 and 24 h). Deconvolu-

tion of the spectral profiles in the 1.1–1.5 ppm frequency

region (mainly comprising CH2 signals of lipids, amino

acids and/or mobile peptide fragments, with possible con-

tributions from the lactic acid methyl doublet) allowed

determination of the peak area ratio Rchains=a[(CH2)n]lip/

a(CH3)tot, where a[(CH2)n]lip is the integral of the ML

(CH2)n resonance (1.3 ppm) and a(CH3)tot is the integral

of the ‘‘total’’ CH3 resonance at 0.9 ppm, due to amino acids

and lipids. The Rchains value is commonly used as a

parameter for relative quantification of ML in intact cells,

in spite of the fact that the (CH3)tot peak selected as area

reference may also depend (to an a priori unknown extent)

upon ML content. The advantage of using the (CH3)totsignal as internal (intracellular) peak area reference for

ML quantification resides on the independence of the Rchains

ratio from cell density fluctuations within the sensitive

NMR coil region; these fluctuations may in fact lead to a

Representative flow cytometry analyses are shown of HuT 78 cells induced

incubated for different time intervals in complete medium at 37 jC, either inith PrI (see Materials and methods). On the ordinate is reported the number

(4) tetraploid peak.

Rchains* ¼ 1=½1=R� pðCH3Þlip=pðCH2Þnlip

Table 1

Effect of a-Fas mAb and/or TRC on the percentage (H) of apoptotic

hypodiploid in HuT 78 cells

Incubation H (%)

time (h)CTRL a-Fas CTRL/TRC a-Fas/TRC

1 0.9F0.1 (2) 3.2F1.9 (2) N.D. N.D.

2 1.0F0.4 (3) 8.9F3.9 (3)* 1.3F0.9 (2) 7.7F2.9 (2)*

4 0.9F0.2 (3) 27.4F8.2 (3)** 0.8 21.2

7 1.0F0.2 (2) 27.8F5.2 (2) 1.0F0.4 (2) 34.7F7.3 (2)

14 0.6 32.9 N.D. N.D.

24 1.3F1.0 (7) 45.2F17.7

(6)***

2.9F1.8 (6) 49.6F15.0

(6)***

The number of experiments is shown in parentheses. CTRL, control cells.

For the conditions of incubation with a-Fas mAb and/or to TRC, see

Materials and methods. Two-tailed Student’s t-test analyses (applied to pairs

of groups of at least three points each) showed that the difference between

the H values of a-Fas-treated samples and the respective control

preparations were highly statistically significant (a-Fas vs. CTRL and a-

Fas/TRC vs. CTRL/TRC). No statistically significant effects of TRC could

be observed on the hypodiploid cell fraction of control samples (CTRL/

TRC vs. CTRL, P>0.05), nor on that of a-Fas-treated cells (a-Fas/TRC vs.

a-Fas, P>0.5) at 24 h.

* P<0.026.

**P<0.005.

***P<0.00004.

E. Iorio et al. / Biochimica et Bioph

rather large scattering of data when the concentration of

intracellular components is measured with respect to an

external reference (either in the extracellular medium or in

an external capillary).

The Rchains value, measured in apoptotic cells analysed

between 1 and 24 h after treatment with a-Fas mAb (11

experiments) increased linearly with the hypodiploid frac-

tion (Fig. 3). The best fit function was represented by the

equation Rchains=0.020+0.023�H (correlation r2=0.945

from mean values and r2=0.709 from scattered data). The

Rchains value was instead maintained substantially unaltered

in the same time interval (1–24 h) in the corresponding

control samples (Rchains=0.18F0.11; H=1.0F0.3%, n=10).

The Rchains values measured in control and in a-Fas-treated

apoptotic cells, together with the respective ranges of

apoptotic hypodiploid cell fractions, are summarised in

Table 2.

Experiments at 200 MHz (n=9), where the lower spectral

resolution did not allow a satisfactory deconvolution of the

[(CH2)n]lip signal from the basal spectral profile, the peak

area ratio of the overall resonance band in the 1.1–1.5 ppm

region and the (CH3)tot resonance was RVchains=0.95F0.21

in control preparations and increased linearly with the

apoptotic cell fraction with a best fit RVchains=1.00+0.021�H, i.e. with a slope very close to that determined

at higher fields (400 and 700 MHz). In spite of the lower

spectral resolution and consequent limitations in the

deconvolution of spectral profiles at 200 MHz, in vitro

analyses at this lower field (4.7 T) are of some interest,

since they allow a better comparison with the spectral

conditions provided by clinical NMR spectroscopy equip-

ment (1.5–4.0 T).

3.3. Effect of Triacsin C on ML formation during a-Fas-induced apoptosis

Continuous exposure to TRC of the HuT 78 cells

induced to apoptosis by a-Fas treatment severely inhibited

ML formation (Fig. 2, right side and Fig. 3). In fact, the

average Rchains value measured in 1H NMR spectra (400 or

700 MHz) of a-Fas/TRC-treated, apoptotic cells at differ-

ent times of incubation after a-Fas treatment was practi-

cally constant (0.10F0.08, n=5) and not significantly

different from that of the corresponding TRC-treated,

non-apoptotic control cells (Table 2). Similar results were

obtained at 200 MHz, where the RVchains value of a-Fas/

TRC apoptotic cells at 24 h after a-Fas treatment (0.94F0.37, n=3) was not significantly different from that of

CTRL/TRC cells. When, however, cells were exposed to

TRC only during the 2-h a-Fas treatment, and not during

the following 24-h incubation, ML were formed to levels

comparable to those exhibited by cells treated with a-Fas

only (data not shown).

No significant effects were exerted by continuous expo-

sure to TRC on the time course of a-Fas-induced pro-

grammed cell death (Tables 1 and 2). Moreover, exposure

to TRC did not alter to any significant extent cell viability

(>99%) nor the hypodiploid fraction of control cells

(H=2.8F1.7%, n=7).

Analyses at 700 MHz allowed a clearer discrimination

of the partial contributions given by ML to the (CH3)totresonance (shown in shadow in the inset of Fig. 4).

Moreover, subtraction of control from apoptotic cell

spectra (Fig. 4, bottom) allowed separate detection of

the ML signals, respectively, due to the saturated (CH2)nsegments (1.30 ppm), the a- and h-methylene (2.25 and

1.59 ppm), the terminal CH3 (0.89 ppm), the vinylic

(CHCH, 5.35 ppm), bis-allylic (CHCH2CH, 2.80 ppm),

and allylic (CH2CH2CH, 2.04 ppm) proton groups. The

peak area ratios of these NMR-visible ML signals,

a[HCCH]lip/a[(CH2)n]lip/a[(CH)3]lip, were 4:16.5:3. By

comparison, the corresponding ratios of peaks in oleic

acid (18:1) would be 2:20:3 and those of linoleic acid

(18:2) 4:14:3. It could therefore be estimated that the

‘‘average’’ chain length and unsaturation degree of NMR-

visible ML chains formed in our apoptotic cells were

intermediate between those of oleic and linoleic fatty acyl

chains.

From these data it could be estimated that the contribu-

tion of mobile lipids, (CH3)lip, to the (CH3)tot resonance

increased from less than 2% at the observed Rchains=0.10 to

about 10% at Rchains=0.50 and to about 21% at Rchains=1.20.

This means that if one would like, more correctly, quantify

the [(CH2)n]lip with respect to the presumably constant

[(CH3)tot�(CH3)lip] component, the corrected Rchains* =

[(CH2)n]lip/[(CH3)tot�(CH3)lip] value would be given by

the relationship

ysica Acta 1634 (2003) 1–14 5

Fig. 2. 1H NMR spectra (400 MHz) of intact Hut 78 cells exposed to a-Fas mAb and/or to TRC. Left: representative spectra showing increases in the [CH2)n]lipsignal intensity (1.3 ppm) of ML in cells induced to apoptosis by exposure to a-Fas mAb (see Materials and methods) and analysed at different times (between

1 and 24 h) of subsequent incubation in complete medium at 37 jC. Apoptotic cells (‘‘a-Fas’’): 2 h, H=5.2%, Rchains=0.18; 7 h, H=33.1%, Rchains=0.48; 24 h,

H=56.3%, Rchains=1.55. Control preparations (‘‘CTRL’’): 2 h, H=1.5%, Rchains=0.14; 24 h, H=0.8%, Rchains=0.07. Right: inhibition of the [(CH2)n]lip signal

formation in the cell preparations of left side, continuously exposed to 5.0 AM TRC, both during treatment with a-Fas mAb and during the subsequent

incubation in complete medium at 37 jC, until NMR analyses. Apoptotic cells incubated in the presence of TRC (‘‘a-Fas/TRC’’): 2 h, H=4.7%, Rchains=0.04;

7 h , H=42.0%, Rchains=0.03; 24 h, H=61.4%, Rchains=0.22. Control preparations (‘‘CTRL/TRC’’): 2 h, H=2.2%, Rchains=0.05; 24 h, H=2.3%, Rchains=0.07.

H=percentage of hypodiploid cells (see Fig. 1); Rchains=a[(CH2)n]lip/a(CH3)tot peak area ratio.

E. Iorio et al. / Biochimica et Biophysica Acta 1634 (2003) 1–146

where p(CH3)lip is the number of protons (3) in the

terminal methyl and p[(CH2)n]lip is the average number

of protons (16.5) in the saturated methylene segments of

NMR-visible ML chains. On this basis, the observed

Rchains values should for instance be corrected by a factor

1.10 at Rchains=0.50 (Rchains* =0.55), 1.28 at Rchains=1.2

(Rchains* =1.54) and 1.38 at Rchains=1.5 (Rchains* =2.07) and

the best fit Rchains* (H) function would still be linear

(Rchains* =0.098+0.024�H) within the first 24 h of post-

treatment incubation.

The N+(CH3)3 signal at 3.2 ppm was mainly attributed

to phosphocholine (PCho) on the basis of 1H NMR

analyses of aqueous extracts, the choline and the glycer-

ophosphocholine signals representing about 5–10% of the

total choline containing metabolites (data not shown). The

peak area ratio RPCho=a[N+(CH3)3]/a[(CH3)tot�(CH3)lip]

(measured in intact cells), was practically maintained unal-

tered in apoptotic cells for up to 24 h of incubation after

a-Fas treatment, its average value in 11 assays (2.27F0.47)

being not significantly different from that of an equal

number of controls (2.50F0.51). Cell exposure to TRC

did not modify these values to any significant extent (data

not shown).

These findings were also confirmed by analyses at 200

MHz (not shown).

3.4. Lipid analysis

As shown in Fig. 5A (lanes ‘‘CTRL’’ and ‘‘a-Fas’’) and

in Table 2, the CE, TAG and [diacylglycerol (DAG)+FFA]

bands of total lipid cell extracts increased by factors of

1.7, 3.1 and 2.3, respectively, in cells analysed after 24

h from induction of a-Fas-induced apoptosis, with respect

to untreated cells. CHOL was instead only slightly reduced

(by less than 15%). It is also worth noting that the TAG/

CE and the CE/CHOL ratios increased about 2-fold. These

results were confirmed in two independent series of

experiments.

Analyses of phospholipids (Fig. 5B) showed that the

major PL classes in control cells had relative propor-

tions PE/PC/CL/PI/PS/SM=35:26:13:10:8:8 and that

treatment with a-Fas induced, at 24 h, a reduction of

Fig. 3. Dependence of ML formation upon hypodiploid cell fraction (H) in

HuT 78 cells exposed to a-Fas mAb and/or to TRC. The peak area ratio

Rchains=a[(CH2)n]lip/a(CH3)tot was determined in 1H NMR spectra (400

MHz) of intact cells induced to apoptosis by exposure to a-Fas mAb (see

Materials and methods) and analysed at different times (between 1 and

24 h) of subsequent incubation in complete medium (37 jC) either in the

absence (., ‘‘a-Fas’’) or in the presence of 5.0 AM TRC (E, ‘‘a-Fas/

TRC’’). The fraction of apoptotic cells was measured as the percentage of

the hypodiploid cell fraction (H) in single parameter flow DNA distribution

(see Fig. 1). H=percentage of hypodiploid cells (see Fig. 1); Rchains=

a[(CH2)n]lip/a(CH3)tot peak area ratio.

Fig. 4. Identification of ML fatty acyl chains’ signals in very-high-field

NMR spectra (700 MHz) of intact apoptotic HuT 78 cells. 1H NMR

spectra were performed on cells induced to apoptosis by exposure to

a-Fas (see Materials and methods) and incubated for 24 h in complete

medium either in the absence (‘‘a-Fas’’, H=61.2%; Rchains=0.94) or in the

presence of 5.0 AM Triacsin (‘‘a-Fas/TRC’’, H=65.2%, Rchains=0.08) in

complete medium. The bottom spectrum was obtained as the difference,

in the frequency domain, of the spectrum of apoptotic cells (‘‘a-Fas’’) and

the spectrum of the corresponding control preparation (‘‘CTRL’’,

H=1.3%; R =0.32). Inset: in grey, contribution of ML to the CH

E. Iorio et al. / Biochimica et Biophysica Acta 1634 (2003) 1–14 7

about 30% in the total PL pool, with no significant shift

in the relative levels of major individual phospholipid

classes.

No increases in CE, TAG and [DAG+FFA] were instead

detected in extracts of apoptotic cells incubated in the

presence of TRC (Fig. 5A, ‘‘a-Fas’’ vs. ‘‘CTRL/TRC’’

and ‘‘CTRL’’ lanes). Of interest was the increase of the

[DAG+FFA] band in CTRL/TRC cell extracts, an expected

Table 2

Effects of a-Fas mAb and/or TRC on mobile lipid formation, neutral lipid

composition and apoptosis in HuT 78 cells

CTRL a-Fas CTRL/TRC a-Fas/TRC

Incubation

time (h)

1–24 1–24 2–24 2–24

H (%)a 1.1F0.7

(n=18)

1.3–61.2

(range)

2.8F1.7

(n=7)

4.7–65.2

(range)

Rchainsb 0.18F0.11

(n=10)

0.020F0.023�H

(n=11)

0.10F0.08

(n=5)

0.10F0.08

(n=5)

Incubation

time (h)

24 24 24 24

FFA+DAGc 3.8F0.8 8.8F1.8 7.8F1.6 1.8F0.4

TAGc 2.2F0.5 6.9F1.4 1.7F1.7 1.7F1.7

CEc 5.8F1.2 9.6F1.9 6.3F1.3 5.8F1.2

CHOLc 9.0F1.8 7.9F1.6 9.5F1.9 7.8F1.6

TAG/CE 0.38 0.72 0.53 0.60

CE/CHOL 0.64 1.22 0.66 0.74

a H, percentage of apoptotic hypodiploid cells.b Rchains=a[(CH2)n]lip/a(CH3)tot measured at 400 or 700 MHz.c Arbitrary units; optical density (after staining with copper acetate) per

5�106 cells (results obtained by two independent experiments).

chains 3

resonance. An unidentified signal of commercial PBS buffer is indicated

by (*).

effect of acyl-CoA synthetase (ACS) inhibition by TRC

[47]. On the other hand, incubation with the ACS inhibitor

prevented increases in the pools of CE, TAG and [DAG+

FFA] in apoptotic cells and decreases in PL (except phos-

phatidylserine) as shown in Fig. 5 (panels A and B, lanes

‘‘a-Fas/TRC’’).

3.5. Characterization of ML structure

Previous fluorescence and electron microscopy studies

carried out in our laboratory on Jurkat T lymphoblastoid

cells induced to apoptosis by 72 h exposure to dexameth-

asone had shown that the increase of ML signals in 1H

NMR spectra was associated with the formation of cyto-

plasmic lipid bodies [21]. CLSM analyses of Nile Red-

stained apoptotic HuT 78 cells confirmed this finding, by

detecting massive accumulation of cytoplasmic lipid bod-

ies at 24 h of incubation after treatment with a-Fas mAb

Fig. 5. HPTLC analyses of total lipid extracts of HuT 78 cells exposed to a-Fas and/or to TRC. HPTLC analyses were performed on neutral lipids and

phospholipids of cells induced to apoptosis by exposure to a-Fas mAb (see Materials and methods) and then incubated for 24 h in complete medium, either in

the absence (lane 2, ‘‘a-Fas’’) or in the presence of 5.0 AM TRC (lane 4, ‘‘a-Fas/TRC’’). Analyses of the respective control preparations are respectively shown

in lanes 1 and 3 (‘‘CTRL’’ and ‘‘CTRL/TRC’’). The position of pure standard lipids is indicated on the margin of each panel. X=unknown component.

Abbreviations used: CL, cardiolipin; CHOL, free cholesterol; Lys-PC, lysophosphatidylcholine; PA, phosphatidic acid; PE, phosphatidylethanolamine; PI,

phosphatidylinositol; PS, phosphatidylserine; SM, sphingomyelin; for other abbreviations, see list.

Fig. 6. Detection by CLSM of cytoplasmic lipid bodies in Nile Red-stained HuT 78 cells exposed to a-Fas and/or to TRC (24 h incubation after a-Fas

treatment). CLSM analyses were performed on cells stained with the lipid probe Nile Red, after exposure to a-Fas mAb (see Materials and methods) and

subsequent 24-h incubation either in the absence (‘‘a-Fas’’) or in the presence of TRC (‘‘a-Fas/TRC’’). The samples were the same as in Fig. 4. The

corresponding control preparations are reported in panels ‘‘CTRL’’ and ‘‘CTRL/TRC’’. Insets: details of lipid bodies detected in single cell analysis. Bar:

10 Am.

E. Iorio et al. / Biochimica et Biophysica Acta 1634 (2003) 1–148

Fig. 7. Detection by CLSM of cytoplasmic lipid bodies in HuT 78 cells stained with BODIPY 493/503 following exposure to a-Fas and/or to TRC (2–7 h

incubation after a-Fas treatment). CLSM analyses were performed on HuT 78 cells stained with the lipid probe BODIPY 493/503 after exposure to a-Fas

mAb (see Materials and methods) and subsequent incubation for different time intervals (2 h, top; 4 h, middle; 7 h, bottom) either in the absence (‘‘a-Fas’’)

or in the presence of TRC (‘‘a-Fas/TRC’’). The corresponding control preparations are reported in the panels indicated by ‘‘CTRL’’ and ‘‘CTRL/TRC’’.

Bar: 20 Am.

E. Iorio et al. / Biochimica et Biophysica Acta 1634 (2003) 1–14 9

(Fig. 6, panels A,B). Continuous exposure of the same cell

preparations to TRC simultaneously abolished ML signal

formation in NMR spectra (see Figs. 2–4) and the

formation of cytoplasmic lipid bodies (Fig. 6C,D), in

agreement with the depletion and/or strong reduction

induced by this inhibitor in the pools of TAG and CE

(the two major lipid droplet constituents) observed in cell

extracts (see Fig. 5). These results were confirmed by

CLSM analyses of cells stained with another lipophilic

probe, BODIPY 493/503 at earlier times of incubation (2,

4 and 7 h), either in the absence or in the presence of

TRC (Fig. 7). Cell double-staining with BODIPY 493/503

(green) and PrI (red) allowed simultaneous observation of

lipid droplets and apoptotic bodies in cells analysed 4

h after a-Fas treatment. CLSM observations clearly con-

firmed that TRC practically abolished the formation of

lipid bodies, without interfering with that of picnotic

nuclei.

Regarding control cells, the low basal level of Rchains

values measured by NMR spectroscopy (0.18F0.11) was

associated with the presence of a limited number of clearly

detected lipid bodies, which decreased upon cell exposure to

TRC (Figs. 6–8). A qualitative comparison of CLSM

analyses on control and a-Fas-treated apoptotic cells raises

the question whether all lipid droplets present in these two

types of systems are characterized by the same level of

NMR-visibility. In particular, it cannot be excluded that

there are droplets in control cells that are not (or are only

partially) NMR-visible, possibly in relation to the lower

average TAG/CE value (Table 2).

4. Discussion

In agreement with the results of previous NMR studies

on a variety of cell systems induced to programmed cell

death by continuous exposure to drugs [18,19,21], Fas-

stimulation [19,22] or gene therapy [20], a progressive

accumulation of NMR-visible MLs was detected in this

work in HuT 78 cells induced to apoptosis by a-Fas mAb.

The increase in ML signals observed during programmed

cell death has been in the past attributed either to increased

mobility of plasma membrane lipids [18] or to intracellular

accumulation of neutral lipids, such as polyunsaturated fatty

acids [20] or TAG [22], under the form of intracellular lipid

bodies [20–22,31].

Fig. 8. Simultaneous detection by CLSM of lipid bodies and picnotic nuclei in HuT 78 cells double-stained with propidium iodide and BODIPY 493/503,

following exposure to a-Fas mAb and/or to TRC (4 h incubation after a-Fas treatment). CLSM analyses of HuT 78 stained with the lipid probe BODIPY 493/

503 (green) after exposure to a-Fas (see Materials and methods) and subsequent incubation for 4 h either in the absence (‘‘a-Fas’’, H=18.5%) or in the presence

of TRC (‘‘a-Fas/TRC’’, H=21.2%). The corresponding control preparations are reported (‘‘CTRL’’, H=1.0% and ‘‘CTRL/TRC’’, H=0.8%). Bar: 10 Am.

E. Iorio et al. / Biochimica et Biophysica Acta 1634 (2003) 1–1410

The parallel decrease in PC and increases in neutral lipids

(TAG and DAG) and fatty acid synthesis detected by

Engelmann et al. [12] in Miltefosine-treated KB cells pointed

to a possible impairment of the DAG pathway during the cell

death programme induced by this drug. This hypothesis was

in substantial agreement with the inhibition of PC synthesis

detected by Williams et al. [14] in HL-60 cells induced to

apoptosis by a variety of cytotoxic compounds.

Combined NMR, HPTLC and CLSM analyses per-

formed in this study showed that (a) the amount of ML

formed in HuT 78 cells within 24 h from a 2-h treatment

with a-Fas mAb exhibited positive linear correlation with

the apoptotic cell fraction; (b) the build-up of ML signals in

NMR spectra was associated with increases in the levels of

the two major constituents of lipid droplets, TAG (over 3�)

and CE (ca. 1.6�), with a resulting 2-fold increase in the

TAG/CE ratio; (c) the increased neutral lipid contents were

associated with progressive accumulation of cytoplasmic

lipid bodies, as detected by CLSM analysis of intact

apoptotic cells stained with lipophilic probes; (d) the

average length and unsaturation degree of the NMR-visible

ML chains was intermediate between those of oleic and

linoleic acid.

The increases in neutral lipid abundance during a-Fas-

induced apoptosis (accompanied by substantial decreases—

by about 30%—in the total phospholipid content) required

unrestricted acyl-CoA availability. In fact, no increases in

TAG or CE masses (nor decreases in PL) were observed

when apoptotic cells were incubated in the presence of

TRC. This potent competitive inhibitor of long-chain

ACS, blocks activation and reuse of fatty acids [35] and is

known to prevent synthesis of TAG and PL from glycerol,

without inhibiting PL re-acylation from lysophospholipids

[36]. Furthermore, since ACS-mediated acyl-CoA synthesis

from FFA is an ATP-dependent reaction, it is reasonable to

conclude that the observed increased production of TAG

and CE and the concomitant inhibition of PL synthesis

occur under conditions in which the cells, although defi-

nitely progressing through the apoptotic process, are still

bioenergetically competent. At variance from the results

obtained by Shimabukuro et al. [48] in pancreatic h cells,

the effects of TRC did not modify the progression towards

E. Iorio et al. / Biochimica et Biophysica Acta 1634 (2003) 1–14 11

apoptosis, possibly because of a different expression of

enzymes involved in lipid metabolism and to the different

method of induction of programmed cell death.

The results of this study can be interpreted in the light of

the well-established concept of coordinated regulation of

the PL biosynthetic reactions with the pathways of TAG

synthesis and turnover [49]. These pathways are schemat-

ically represented in Fig. 9. Under conditions of unrestricted

ATP and acyl-CoA levels, TAG synthesis occurs from DAG

and acyl-CoA (via acyltransferase) and therefore subtracts

these substrates from PL re-synthesis; at the same time,

lipase-mediated TAG degradation products, FFA and glyc-

erol, can actively re-enter their respective utilisation cycles

(e.g. FFA activation to acyl-CoA; FFA h-oxidation in

mitochondria; the cascade of glycerol phosphorylation/ac-

ylation, resulting into re-synthesis of phosphatidate and new

DAG production). The detailed mechanisms responsible for

TAG accumulation (at the expense of the PL pools) in

apoptotic cells require further investigations. In general

terms, it could be envisaged that a reduction in the activi-

Fig. 9. Schematic representation of the coordinated regulation of phospholipid

coenzyme A; CDP-Cho, cytidine diphospho-choline; CDP-DAG, cytidine dip

choline; CL, cardiolipin; CoA, coenzyme A; CTP, cytidine triphosphate; DAG, dia

3P, glycerol 3-phosphate; Ins, inositol; PA, phosphatidic acid; PC, phosphatid

phosphoethanolamine; PGro, phosphatidylglycerol; P-GroP, phosphatidylglycerolp

sphingomyelin; TAG, triacylglycerides. Enzymes: ACS, acyl-CoA synthetase;

phospholipase C; pld, phospholipase D.

ty(ies) of some enzymes responsible for the de novo route

of PL biosynthesis and/or for the activation of specific

phospholipases is likely to lead to increased levels of

DAG, which would be toxic to the cell if not converted

into TAG or degraded to FFAs, whose subsequent activa-

tion to acyl-CoA would in turn allow further synthesis of

TAG (and CE from FC, via acyl-CoA cholesterol acyl-

transferase (ACAT). Such a possible impairment of

enzymes responsible for PL synthesis would be in agree-

ment with the reported increase of CDP-choline in cells

induced to apoptosis by a variety of cytotoxic drugs [14]

and with the interpretation of altered lipid components

observed in Miltefosine-treated cells [12]. These and similar

findings actually suggested the interest of measuring the

levels of other intermediates of the Kennedy’s pathway in

PC and phosphatidylethanolamine biosynthesis, with par-

ticular attention to PCho and phosphoethanolamine (PEtn)

[17]. The results of these studies appeared rather contradic-

tory, however, or at least strictly dependent upon the

particular cell system or inducing agent. In particular,

biosynthesis and turnover of glycerolipids. Metabolites: acyl-CoA, acyl-

hospho-diacylglycerol; CDP-Etn, cytidine diphospho-ethanolamine; Cho,

cylglycerides; Etn, ethanolamine; FFA, free fatty acids; Gro, glycerol; Gro-

ylcholine; PCho, phosphocholine; PE, phosphatidylethanolamine; PEtn,

hosphate; PI, phosphatidylinositol; PS, phosphatidylserine; Ser, serine; SM,

dgk, diacylglycerol kinase; pap, phosphatidate phosphohydrolase; plc,

E. Iorio et al. / Biochimica et Biophysica Acta 1634 (2003) 1–1412

significant decreases in PCho levels have been reported in

the human promyelocytic leukemia cell line HL-60 induced

to apoptosis by chelerythrine (a protein kinase C inhibitor)

or by ceramide [14] under conditions of lowered ATP

content, or in T lymphoblastoid Jurkat cells after prolonged

exposure to doxorubicin [18], anti-Fas [19] or dexametha-

sone [21]. An increase in PCho has been instead measured

in apoptotic neutrophils [11], suggesting the attractive

hypothesis that NMR could allow monitoring of the acti-

vation of a PC-specific phospholipase, following stimula-

tion of Fas receptors [50]. However, no substantial increase

in PCho levels has been found in a previous study on a-

Fas-stimulated apoptotic Jurkat T cells [22], nor in the

present study on a-Fas-stimulated HuT 78 T cells. Similar-

ly, apparently conflicting results have been reported on

PEtn levels during apoptosis of different cell systems—this

compound increasing, e.g. in farnesol- or chelerythrine-

treated HL-60 cells, but remaining at unaltered levels in

the same cells induced to apoptosis by other agents such as

camptothecin, etoposide or ceramides [14], and decreasing

in dexamethasone-treated human leukemia cells [51]. Pos-

sible explanations for the different results reported in the

literature on PCho and PEtn levels could refer either to

decreased levels of ATP present under different experimen-

tal conditions (which would clearly affect the outcome of

choline and ethanolamine kinase activities) or to different

time courses of increase and decrease of these metabolites

under the action of transiently activated phospholipases. As

a consequence, NMR cannot always give an unambiguous

information on phospholipase-dependent processes occur-

ring during apoptosis, unless experimental conditions are

kept under strict control. In our case, combined NMR

analyses on intact cells and their aqueous extracts showed

that there was a substantial lack of modification in the PCho

levels in a-Fas-treated with respect to control cells, making

it difficult to obtain clear information on the underlying

biochemical mechanisms in the explored time window (1–

24 h of cell incubation after end of Fas stimulation).

As a general conclusion of this study, it can be proposed

that the increases in TAG (and CE) and the consequent

formation of ‘‘NMR-visible’’ lipid bodies in HuT 78 cells

induced to apoptosis by Fas-stimulation were the result of

alterations occurring under these conditions, during the

apoptotic process, in some enzyme activities of the coordi-

nated PL/neutral lipid pathways. Since the content of NMR-

visible ML correlated linearly with the hypodiploid cell

fraction, the ML signals can actually be used for the

quantification of apoptosis, at least under conditions of

unrestricted availability of ATP and acyl-CoA. However,

the production of these ‘‘mobile lipids’’ in apoptotic cells is

not an ‘‘indispensable’’ process for the accomplishment of

the cell death programme, since inhibition of acyl-CoA

synthesis, with the consequent block of acyltransferase

reactions, abolished the formation of the ‘‘NMR-visible’’

lipid bodies, without perturbing the time course of apoptosis

to any significant extent. Even PL synthesis returned, under

these conditions, to the basal levels measured in untreated

control cells.

By affecting reactions which are crucial to ML formation

in apoptotic cells, TRC acted as a particularly effective tool

to investigate biogenesis, structure and nature of ML formed

in an apoptotic cell system and might provide interesting

approaches to the biochemical/biological characterization of

ML in other cell systems, such as cancer cells and activated

lymphocytes.

5. Conclusions

(1) A progressive increase in 1H NMR-visible MLs was

measured in intact apoptotic lymphoblastoid cells in the

first 24 h after 2-h exposure to a-Fas mAb. This

increase is attributed to changes in neutral lipid contents

(mainly TAG) and accumulation of cytoplasmic lipid

bodies in cells undergoing programmed cells death.

(2) By inhibiting long-chain ACS and therefore reducing

the pool of cellular acyl-CoAs, TRC induced TAG

depletion, decrease of CE and disappearance of ML

signals in the NMR spectra of apoptotic cells, without

altering cell viability nor the fraction of apoptotic cells.

These effects were associated with the disappearance of

cytoplasmic lipid bodies.

(3) The results supported the view that ML signals formed

in HuT 78 cells during a-Fas-induced apoptosis derive

from partially unsaturated neutral lipids synthesised

during a biochemically active (ATP-dependent) phase

of apoptosis and assembled in intracellular lipid bodies.

Acknowledgements

This work was partially supported by grants of the

Istituto Superiore di Sanita, Rome, ISS-2045/RI and ISS-

0D/C. We thank Mr. Massimo Giannini and Mr.

Massimiliano Ferrara, for technical assistance in maintain-

ing the NMR equipment at an excellent performance level

and Dr. Carla Raggi for expert collaboration in lipid

analysis.

References

[1] J.F. Kerr, A.H. Wyllie, A.R. Currie, Apoptosis: a basic biological

phenomenon with wide-ranging implications in tissue kinetics, Br.

J. Cancer 26 (1972) 239–257.

[2] M.-T. Heemels, R. Dhand, R. Allen, Apoptosis, Nature 407 (2000)

769–916.

[3] A.H. Wyllie, J.F.R. Kerr, A.R. Currie, Cell death: the significance of

apoptosis, Int. Rev. Cyt. 68 (1980) 251–306.

[4] M.C. Raff, Social controls on cell survival and death, Nature 356

(1997) 397–399.

[5] M.J. Arends, A.H. Willie, Apoptosis: mechanisms and roles in path-

ology, Int. Rev. Exp. Pathol. 32 (1991) 223–254.

[6] G.T. Williams, Programmed cell death: apoptosis and oncogenesis,

Cell 65 (1991) 1097–1098.

E. Iorio et al. / Biochimica et Biophysica Acta 1634 (2003) 1–14 13

[7] C.B. Thompson, Apoptosis in the pathogenesis and treatment of dis-

ease, Science 267 (1995) 1456–1462.

[8] D.S. Ucker, Death and dying in the immune system, in: S.H. Kauf-

mann (Ed.), Advances in Pharmacology, vol. 41, Academic Press,

New York, 1997, pp. 179–218.

[9] P. Mesner Jr., I.I. Budihardio, S.H. Kaufmann, Chemotherapy-in-

duced apoptosis, Adv. Pharmacol. 41 (1997) 461–499.

[10] K.K. Bhakoo, J.D. Bell, The application of NMR spectroscopy to the

study of apoptosis, Cell. Mol. Biol. 43 (1997) 621–629.

[11] A.V.W. Nunn, M.L. Barnard, K. Bhakoo, J. Murray, E.J. Chilvers,

J.D. Bell, Characterisation of secondary metabolites associated with

neutrophil apoptosis, FEBS Lett. 392 (1996) 295–298.

[12] J. Engelmann, J. Henke, W. Willker, B. Kutscher, G. Nossner,

J. Engel, D. Leibfritz, Early stage monitoring of miltefosine induced

apoptosis in KB cells by multinuclear NMR spectroscopy, Anti-

cancer Res. 16 (1996) 1429–1440.

[13] F. Adebodun, Phospholipid metabolism and resistance to glucocorti-

coid-induced apoptosis in a human leukemic cell line: a 31P-NMR

study using a phosphonium analog of choline, Cancer Lett. 140

(1999) 189–194.

[14] S.N.O. Williams, M.L. Anthony, K.M. Brindle, Induction of apoptosis

in two mammalian cell lines results in increased levels of fructose-1,6-

bisphosphate and CDP-choline as determined by 31P MRS, Magn.

Reson. Med. 40 (1998) 411–420.

[15] S.M. Ronen, F. Distefano, C.L. McCoy, D. Robertson, T.A.D. Smith,

N.M. Al-Saffar, J. Titley, D.C. Cunningham, J.R. Griffiths, M.O.

Leah, P.A. Clarke, Magnetic resonance detects metabolic changes

associated with chemotherapy-induced apoptosis, Br. J. Cancer 80

(1999) 1035–1041.

[16] M.L. Anthony, M. Zhao, K.M. Brindle, Inhibition of phosphatidyl-

choline biosynthesis following induction of apoptosis in HL-60 cells,

J. Biol. Chem. 274 (1999) 16686–16692.

[17] K.M. Brindle, Detection of apoptosis in tumors using magnetic reso-

nance imaging and spectroscopy, Adv. Enzyme Regul. 42 (2002)

101–112.

[18] F.G. Blankenberg, R.W. Storrs, L. Naumovski, T. Goralski, D. Spiel-

man, Detection of apoptotic cell death by proton magnetic resonance

spectroscopy, Blood 87 (1996) 1951–1956.

[19] F.G. Blankenberg, P.D. Katsikis, R.W. Storrs, C. Beaulieu, D. Spiel-

man, J.Y. Chen, L. Naumovski, J.F. Tait, Quantitative analysis of

apoptotic cell death using proton nuclear magnetic resonance spectro-

scopy, Blood 89 (1997) 3778–3786.

[20] J.M. Hakumaki, H. Poptani, A.-M. Sandmair, S. Yla-Herttuala, R.A.

Kauppinen, 1H MRS detects polyunsaturated fatty acid accumulation

during gene therapy of glioma: implications for the in vivo detection

of apoptosis, Nat. Med. 5 (1999) 1323–1327.

[21] M. Di Vito, L. Lenti, A. Knijn, E. Iorio, F. D’Agostino, A. Molinari,

A. Calcabrini, A. Stringaro, S. Meschini, G. Arancia, A. Bozzi, R.

Strom, F. Podo, 1H NMR-visible lipid domains correlate with cyto-

plasmic lipid bodies in apoptotic T-lymphoblastoid cells, Biochim.

Biophys. Acta 1530 (2001) 47–66.

[22] N.M.S. Al-Saffar, J.T. Titley, D. Robertson, P.A. Clarke, L.E. Jackson,

M.O. Leach, S.M. Ronen, Apoptosis is associated with triacylglycerol

accumulation in Jurkat T-cells, Br. J. Cancer 86 (2002) 963–970.

[23] C.E. Mountford, W.B. Mackinnon, P. Russell, A. Rutter, E.J. Deli-

katny, Human cancers detected by proton MRS and chemical shift

imaging ex vivo, Anticancer Res. 16 (1996) 1521–1532.

[24] M.F. Veale, A.J. Dingley, G.F. King, N.J.C. King, 1H NMR visible

neutral lipids in activated T lymphocytes: relationship to phosphati-

dylcholine cycle, Biochim. Biophys. Acta 1303 (1996) 215–221.

[25] M.F. Veale, N.J. Roberts, G.F. King, N.J.C. King, The generation of1H-NMR-detectable mobile lipid in stimulated lymphocytes: relation-

ship to cellular activation, the cell cycle, and phosphatidylcholine-

specific phospholipase C, Biochem. Biophys. Res. Commun. 239

(1997) 868–874.

[26] G.L. May, L.C. Wright, K.T. Holmes, P.G. Williams, I.C.P. Smith, P.E.

Fox, R.M. Fox, C.E. Mountford, Assignment of methylene proton

resonances in NMR spectra of embryonic and transformed cells to

plasma membrane triglyceride, J. Biol. Chem. 261 (1986) 3048–3053.

[27] A. Ferretti, A. Knijn, E. Iorio, S. Pulciani, M. Giambenedetti, A.

Molinari, S. Meschini, A. Stringaro, A. Calcabrini, I. Freitas, R.

Strom, G. Arancia, F. Podo, Biophysical and structural characteriza-

tion of 1H-NMR-detectable mobile lipid domains in NIH-3T3 fibro-

blasts, Biochim. Biophys. Acta 1438 (1999) 329–348.

[28] R. Callies, R.M. Sri-Pathmanathan, D.Y.P. Ferguson, K.M. Brindle,

The appearance of neutral lipid signals in the 1H NMR spectra of a

myeloma cell line correlates with the induced formation of cytoplas-

mic lipid droplets, Magn. Reson. Med. 29 (1993) 546–550.

[29] C. Remy, N. Fouilhe, I. Barba, E. Sam-Laı, H. Lahrech, M.-G.

Cucurella, M. Izquierdo, A. Moreno, A. Ziegler, R. Massarelli,

M. Decorps, C. Arus, Evidence that mobile lipids detected in rat

brain glioma by 1H nuclear magnetic resonance correspond to lipid

droplets, Cancer Res. 57 (1997) 407–414.

[30] M. Barba, M.E. Cabanas, C. Arus, The relationship between nuclear

magnetic resonance-visible lipids, lipid droplets, and cell proliferation

in cultured C6 cells, Cancer Res. 59 (1999) 1861–1868.

[31] J.M. Hakumaki, R.A. Kauppinen, 1H NMR visible lipids in the life

and death of cells, Trends Biochem. Sci. 25 (2000) 9–14.

[32] E.J. Delikatny, W.A. Cooper, S. Brammah, N. Nathavisan, D.C. Ride-

out, Lipid accumulation induced by cationic lipophilic chemothera-

peutic agents are accompanied by increased lipid droplet formation

and damaged mitochondria, Cancer Res. 62 (2002) 1394–1400.

[33] P.H. Krammer, et al., The role of APO-1-mediated apoptosis in the

immune system, Immunol. Rev. 142 (1994) 175–191.

[34] S. Nagata, P. Golstein, The Fas death factor, Science 267 (1993)

1449–1456.

[35] I. Namatame, H. Tomoda, H. Arai, K. Inoue, S. Omura, Complete

inhibition of mouse macrophage-derived foam cell formation by tri-

acsin C, J. Biochem. 125 (1999) 319–327.

[36] R.A. Igal, P. Wang, R.A. Coleman, Triacsin C blocks de novo syn-

thesis of glycerolipids and cholesterol esters but not recycling of fatty

acid into phospholipid: evidence for functionally separate pools of

acyl-CoA, Biochem. J. 324 (1997) 529–534.

[37] E. Iorio, M. Di Vito, F. Spadaro, C. Ramoni, E. Lococo, R. Carnevale,

L. Lenti, R. Strom, F. Podo, Triacsin C inhibits the formation of 1H

NMR-visible mobile lipids and lipid bodies in HuT 78 apoptotic cells,

Magn. Reson. Mater. Phys. Biol. Med. 15 (S1) (2002) 292–293

(MAG*MA).

[38] I. Nicoletti, G. Migliorati, M.C. Pagliacci, F. Grignani, C. Riccardi,

A rapid and simple method for measuring thymocyte apoptosis by

propidium iodide staining and flow cytometry, J. Immunol. Methods

139 (1991) 271–279.

[39] A. De Maria, L. Lenti, F. Malisan, F. D’Agostino, B. Tomassini, M.R.

Rippo, R. Testi, Requirement for GD3 ganglioside in CD95- and

ceramide-induced apoptosis, Science 277 (1997) 1652–1655.

[40] C. Dive, C.D. Gregory, D.J. Phipps, D.L. Evans, A.E. Milner, A.H.

Wyllie, Analysis and discrimination of necrosis and apoptosis (pro-

grammed cell death) by multiparameter flow cytometry, Biochim.

Biophys. Acta 1133 (1992) 275–285.

[41] A. Knijn, A. Ferretti, P.J. Zhang, M. Giambenedetti, A. Molinari, S.

Meschini, S. Pulciani, F. Podo, Lower levels of 1H MRS-visible

mobile lipids in H-ras transformed tumorigenic fibroblasts with re-

spect to their untransformed parental cells, Cell. Mol. Biol. 43 (1997)

691–701.

[42] D.Y. Sze, O. Jardetzky, Characterization of lipid composition in

stimulated human lymphocytes by 1H-NMR, Biochim. Biophys. Acta

1054 (1990) 198–206.

[43] J. Folch, M. Lees, G.H. Sloane-Stanley, A simple method for the

isolation and purification of total lipids from animal tissue, J. Biol.

Chem. 226 (1957) 497–509.

[44] I. Freitas, P. Pontiggia, S. Barni, V. Bertone, M. Parente, A. Novarina,

G. Roveta, G. Gerzeli, P. Stoward, Histochemical probes for the de-

tection of hypoxic tumour cells, Anticancer Res. 10 (1990) 613–622.

[45] P. Greenspan, E.P. Mayer, S.D. Fowler, Nile Red: a selective fluores-

E. Iorio et al. / Biochimica et Biophysica Acta 1634 (2003) 1–1414

cence stain for intracellular lipid droplets, J. Cell Biol. 100 (1985)

965–973.

[46] P.M. Gocze, D.A. Freeman, Factors underlying the variability of lipid

droplet fluorescence in MA-10 Leydig tumor cells, Cytometry 17

(1994) 151–158.

[47] T. Arai, Y. Kawakami, T. Matsushima, Y. Okuda, K. Yamashita, Intra-

cellular fatty acid downregulates ob gene expression in 3T3-L1 adi-

pocytes, Biochem. Biophys. Res. Commun. 297 (2002) 1291–1296.

[48] M. Shimabukuro, Y.T. Zhou, M. Levi, R.H. Unger, Fatty acid-induced

beta cell apoptosis: a link between obesity and diabetes, Proc. Natl.

Acad. Sci. U. S. A. 95 (1998) 2498–2502.

[49] D.E. Vance (Ed.), Phosphatidylcholine Metabolism, CRC Press, Boca

Raton, FL, 2000.

[50] M.G. Cifone, P. Roncaioli, R. De Maria, G. Camarda, A. Santoni, G.

Ruberti, R. Testi, Multiple pathways originate at the Fas/APO-1

(CD95) receptor: sequential involvement of phosphatidylcholine-spe-

cific phospholipase C and acidic sphingomyelinase in the propagation

of the apoptotic signal, EMBO J. 23 (1995) 5859–5868.

[51] F. Adebodun, J.F. Post, 31P NMR characterization of cellular metab-

olism during dexamethasone induced apoptosis in human leukemic

cell lines, J. Cell. Physiol. 1 (1994) 180–186.