Synthesis of platelet-activating factor (PAF) in transformed cell lines of a different origin

Prostaglandins & other Lipid Mediators 70 (2002) 209–226

Synthesis of platelet-activating factor (PAF) in transformed cell lines of a different origin Anna Fallani a,∗ , Barbara Grieco a , Emanuela Barletta a , Gabriele Mugnai a , Gianluca Giorgi b , Laura Salvini b , Salvatore Ruggieri a a

b

Department of Experimental Pathology and Oncology, University of Florence, Viale G.B. Morgagni 50, Florence I-50134, Italy Interdepartmental Center for Structural Analyses and Determinations, University of Siena, Siena, Italy Received 28 January 2002; received in revised form 28 March 2002; accepted 23 July 2002

Abstract Interest in the possible involvement of the platelet-activating factor (PAF) in tumor growth and invasiveness has been stimulated by the recognition that PAF influences various biological responses relevant to metastatic diffusion, such as angiogenesis, adhesiveness to endothelia and cellular motility. In the present study, we investigated the extent to which PAF is synthesized by a series of human and murine transformed cell lines of a different histotype. Synthesis of PAF was studied by combining the 14 C-acetate incorporation into PAF with the quantitative analysis of PAF performed by a procedure based on gas chromatography–mass spectrometry with a negative ion chemical ionization. In the presence of the Ca2+ ionophore A23187, cultures of human melanoma (Hs294T), fibrosarcoma (HT1080) and colon carcinoma (LS180) cell lines synthesized conspicuous amounts of PAF, comparable to those produced by resident peritoneal macrophages. Substantial quantities of PAF were also synthesized by the murine melanoma (F10-M3 cells). PAF synthesis was rather limited in RSV-transformed Balb/c3T3 (B77-3T3) cells and in one of their high metastatic variants (B77-AA6 cells), although it was more abundant in the latter. We also investigated whether certain cytokines, such as TNF␣ and IFN␥ might induce PAF synthesis in our systems of cell lines which we found to express mRNAs encoding receptors for these cytokines. We observed that PAF synthesis was enhanced in human melanoma and colon carcinoma cell lines and in the murine B77-AA6 cells to levels comparable to those obtained with the Ca2+ ionophore. Synthesis of PAF was not inducible by TNF␣ in murine F10-M3 melanoma cells. IFN␥ also stimulated PAF synthesis in human and murine melanoma lines, and in human Abbreviations: PAF, platelet-activating factor; GC–MS/NICI, gas chromatography–mass spectrometry/ negative ion chemical ionization; TNF␣, tumor necrosis factor ␣; IFN␥, interferon ␥; PAFR, PAF receptor; TNF␣R, TNF␣ receptor; IFN␥R, IFN␥ receptor ∗ Corresponding author. Tel.: +39-55-4282301; fax: +39-55-4282333. E-mail address: [email protected] (A. Fallani). 0090-6980/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S 0 0 9 0 - 6 9 8 0 ( 0 2 ) 0 0 1 0 9 - 0

210

A. Fallani et al. / Prostaglandins & other Lipid Mediators 70 (2002) 209–226

LS180 colon carcinoma line, but not in the B77-AA6 cells. PAF synthesis was also inducible by exogenous PAF in the human and murine melanoma lines, and in the human LS180 colon carcinoma line, all of which expressed cell surface PAF receptors. PAF synthesis was not inducible by exogenous PAF in the B77-AA6 cells, which do not express PAF receptors. © 2002 Elsevier Science Inc. All rights reserved. Keywords: PAF; Tumor cell lines; Cytokines; Metastasis; Gas chromatography–mass spectrometry/negative ion chemical ionization

1. Introduction Platelet-activating factor (PAF), 1-O-alkyl-2-acetyl-sn-glycero-3-phosphocholine, is a biological mediator derived from ether-linked lipids [1]. It promotes several pathophysiological processes, e.g. platelet aggregation [2], activation of leukocytes [3] and macrophages [4], cell motility [5–7], cell adhesiveness [8], angiogenesis [6,7,9], and expression of collagenases [10]. PAF is synthesized by a variety of cells such as platelets [2], polymorphonuclear leukocytes [11,12], macrophages [12,13] and endothelial cells [12,14]. The PAF synthesis has also been demonstrated in virally-transformed glomerular endothelial cells [15] and in cell lines established from human leukemia [16], Kaposi’s sarcoma [17], human gastric tumor [18], endometrial adenocarcinoma [19] and breast cancer [20]. In the present study, we investigated whether PAF was synthesized in a series of human and murine transformed cell lines of a different histotype. These lines were: murine Balb/c3T3 cells transformed by the B77 strain of the Rous sarcoma virus (B77-3T3 cells); B77-AA6 cells, a high metastatic variant isolated from B77-3T3 cells [21]; F10-M3 cells, a clone isolated from the murine B16-F10 melanoma line; and human HT1080 fibrosarcoma, Hs294T melanoma and LS180 colon carcinoma cell lines. PAF synthesis was evaluated by combining the determination of 14 C-acetate incorporation into PAF with quantitative analysis of PAF performed by using gas chromatography–mass spectrometry. We further investigated whether PAF synthesis was stimulated by biological mediators such as TNF␣, IFN␥ or PAF itself. The capacity of cytokines to induce PAF synthesis in different normal and transformed cells has been reported by different authors [12,17,22,23]. We also examined the expression of receptors specific for TNF␣, IFN␥ and PAF in most of our systems of transformed cell lines.

2. Materials and methods 2.1. Chemicals Dulbecco’s Modified Essential Medium containing 4500 mg/ml glucose (DMEM 4500) was supplied by GIBCO, Life Technology; fetal calf serum by Boehringer Mannheim, Germany; 3 H-hexadecyl PAF and sodium salt of the 14 C-acetic acid (57 mCi/mmol) by New England Nuclear (Boston, MA). Deutero-PAF (d3 -hexadecyl-PAF) was obtained from Biomol. Res. Lab. Inc. (Plymouth Metting, PA); 2,3,4,5,6-pentafluorobenzoyl chloride and Phospholipase C from Bacillus cereus from Sigma. TLC sheets (Empore 3M) were supplied

A. Fallani et al. / Prostaglandins & other Lipid Mediators 70 (2002) 209–226

211

by Analytichem. Int., Varian Ass. (Harbor City, CA). Standards of hexadecyl-, octadecyland carbamyl-PAF were obtained from Calbiochem (La Jolla, CA). Murine recombinant TNF␣ and human and murine IFN␥ were supplied from Prepo Tech EC Ltd. (London). All glass material was silanized using the Sigmacote mixture provided by Sigma. 2.2. Cells and culture conditions The B77-AA6 and B77-3T3 cells were supplied by Drs. Di Renzo & Bretti, while the F10-M3 cells were donated by Dr. S. Gattoni-Celli, Medical University of South Carolina, USA. The Hs294T, LS180, and HT1080 cells were obtained from ATCC (Rockville, MD). All cell lines were subcultivated by trypsinization (0.25% trypsin–EDTA solution), and grown in DMEM 4500 supplemented with 10% fetal calf serum at 37 ◦ C in a 10% CO2 -humidified atmosphere. Cultures of peritoneal murine macrophages were established according to a procedure described in a previous paper [24], and used as reference material. 2.3. Incorporation of the 14 C-acetate into PAF Confluent cultures were freed from their growth medium and washed twice with phosphate buffered saline (PBS). Cell monolayers were preincubated for 10 min with 2 ml of PBS containing CaCl2 (1.3 mM) and 14 C-acetate (20 ␮Ci/ml), and then stimulated for the indicated periods of time (see Fig. 2), with 6 ␮l of a 7.5 ␮M Ca2+ ionophore A23187 solution in DMSO. The stimulation was stopped by adding 5 ml of an ice-cold mixture of methanol:acetic acid (98:2, v/v). The metabolically-radiolabeled lipids were extracted from the combined supernatant and cell monolayer by using Bligh and Dyer’s method [25]. The upper phase of Bligh and Dyer’s biphasic system was removed, while the lower phase was washed twice with a PBS/chloroform/methanol/sodium acetate 0.1 M (1:3.8:2.5:1, v/v) mixture in order to remove the unincorporated radioactivity. The purified organic phase was evaporated under a flux of nitrogen, dissolved in a small volume of chloroform, and injected into a 5 ␮m Ultrasphere-Si column (Merck, Darmstadt, Germany) mounted in an HPLC apparatus (Perkin-Elmer mod 3B) connected with a radiometric detector (Flo-One/␤ Series a-500, Radiomatic Camberra Packard). The radiolabeled lipids were chromatographed at a flow rate of 1.5 ml/min using a linear gradient from 100% solvent A (n-hexane:isopropanol:water, 6:8:0.4, v/v) to 100% solvent B (n-hexane/isopropanol/water, 6:8:1.7, v/v) during a 20-min period, followed by a 40-min isocratic run in the same solvent. The position of the radiolabeled PAF in the chromatogram was identified by comparison with the retention time of an authentic 3 H-PAF commercial standard. The radiolabeled fraction corresponding to PAF in the radiochromatogram gave a positive reaction when tested with the platelet aggregation assay [26]. PAF biosynthesis in our systems of transformed cells was evaluated by determining the 14 C-acetate incorporation into PAF at different periods of incubation. 2.4. Quantitative analysis of PAF by gas chromatography–mass spectrometry (GC–MS) Confluent cultures were incubated with the Ca2+ ionophore A23187 for a period of time required for a maximal biosynthetic activity as indicated by 14 C-acetate incorporation into

212

A. Fallani et al. / Prostaglandins & other Lipid Mediators 70 (2002) 209–226

PAF. Total lipids were then extracted from the cultures with Bligh and Dyer’s procedure [25] together with a known amount of a d3 -hexadecyl-PAF internal standard (10 ng/106 cells). PAF was isolated from the extracted lipids according to Triolo et al. [27] by chromatography on TLC sheets with chloroform:methanol:water (65:35:6, v/v). After visualization of the chromatogram by exposure to iodine vapors, the area corresponding to PAF was identified by comparison with a standard of hexadecyl-PAF, cut with scissors, and transferred into a silanized vial containing 2 ml of peroxide-free diethyl ether and phospholipase C (20 U/ml) solubilized in 2 ml of Tris 0.05 M at pH 7.35. After an overnight incubation at room temperature, the released alkyl–acetyl glycerols, dissolved in 100 ␮l of cold hexane:pyridine (73:27, v/v) were converted into pentafluorobenzoyl derivatives by reaction with 100 ␮l of a mixture of n-hexane: 2,3,4,5,6-pentafluorobenzoyl chloride (1:1, v/v) for 15 min. The PAF-pentafluorobenzoyl derivatives of PAF were extracted twice with 1.5 ml of n-hexane from the reaction mixture to which 3 ml of water had previously been added. The combined hexane extracts were transferred into vials and kept for 30 min in dry ice in order to obtain the precipitation of the reagent excess. The hexane extracts were recovered from the vials, washed with pyridine, water and Na2 CO3 buffer 0.5 M at pH 9, and evaporated under a flux of nitrogen. The residue was solubilized in 10 ␮l of decane and 1 ␮l of this solution was injected into a capillary SPB-5-HP column (15 m × 0.2 mm i.d.) mounted in a Hewlett-Packard 5890 gas chromatograph interfaced with a double focusing mass spectrometer VG-70-250S. The mass spectrometer was operated in the negative ion chemical ionization (NICI) with isobutane as reagent gas. Analyses were performed at a programmed temperature from 180 to 270 ◦ C at 25 ◦ C/min with helium as a carrier gas. Samples were analyzed by a mass spectra full-scanning and by the use of selected ion monitoring procedure, considering the ions m/z 552, 555 and 580 for hexadecyl-, d3 -hexadecyland octadecyl-PAF, respectively. 2.5. Effects of TNFα and IFNγ on the synthesis of PAF in transformed cell lines Cell cultures were stimulated for 0–8 h with TNF␣ (10 ng/ml) or for 0–36 h with IFN␥ (50 U/ml). Lipids were extracted from these cultures and submitted to the quantitative analysis of PAF by GS–MS/NICI. 2.6. Effect of PAF on the 14 C-acetate incorporation into PAF in transformed cell lines Cell monolayers were preincubated for 10 min with 2 ml of a phosphate buffered saline containing CaCl2 1.3 mM and 14 C-acetate (20 ␮Ci/ml), and then stimulated for 0, 5, 10, 15 or 30 min with carbamyl-PAF (100 nM), a stable analog of PAF. The radiolabeled lipids extracted from cell cultures with Bligh and Dyer’s method [25] were chromatographed by HPLC, and 3 H-PAF was determined radiometrically as described above. 2.7. Detection by PCR of mRNA encoding for PAF and cytokines receptors in human Hs294T and LS180 cell lines, and in murine F10-M3 and B77-AA6 cell lines Cells from stock cultures were rinsed twice with PBS, scraped with a rubber policeman and lysed on ice with 50 mM Tris–HCl pH 8.0, 140 mM NaCl, 1.5 mM MgCl2 , 0.5% NP-40,

A. Fallani et al. / Prostaglandins & other Lipid Mediators 70 (2002) 209–226

213

214

A. Fallani et al. / Prostaglandins & other Lipid Mediators 70 (2002) 209–226

1000 U/ml RNasin, 1 mM dithiothreitol (DTT). Cell lysates were centrifuged at 4 ◦ C for 2 min at 300 × g in order to remove nuclei and cellular debris. Total cytoplasmic RNA was isolated from supernatant according to the RNeasy protocol (Qiagen, GmbH, Germany), and submitted to RNase-free DNase Set protocol (Qiagen, GmbH, Germany) in order to remove DNA. The amount and purity of RNA were determined spectrophotometrically. The integrity of the extracted RNA was ascertained by the analysis of the rRNA fraction on agarose gel. cDNA was synthesized by incubating for 1 h at 37 ◦ C 1 mg of total RNA with 200 U of Moloney leukemia virus reverse transcriptase in 25 ␮l of the reaction buffer (50 mM Tris–HCl, pH 8.3, 75 mM KCl, 3 mM MgCl2 , 10 mM DTT, 2 mM dNTPs, 0.5 mg random hexamer oligonucleotides, 25 U of RNAsin). The reaction was stopped by heating the mixture for 10 min at 95 ◦ C. Aliquots of 3 ␮l of the cDNA mixture were transferred into vials for PCR amplification. The human TNF␣ (hTNF␣Rp60, hTNF␣Rp80) and IFN␥ receptors (hIFN␥R), and the murine PAF (mPAFR), TNF␣ (mTNF␣Rp55, mTNF␣Rp75) and IFN␥ receptors (mIFN␥R-chain1) were identified by the use of antisense and sense primers reported in Table 1. Reaction was carried out in 50 ␮l of a solution containing 50 mM KCl, 10 mM Tris–HCl, pH 9.0, 0.1% Triton X-100, 1.5 mM MgCl2 , 0.2 mM dNTPs, 1 mM of each primer and 1.25 U of Taq DNA polymerase (Promega, Madison, WI, USA). The reaction proceeded through 30 cycles of denaturation at 94 ◦ C for 1 min, followed by annealing for 1 min at the appropriate temperature (see Table 1) and extension at 72 ◦ C for 1 min. The amplified cDNA products were electrophoresed on 1% agarose gels containing ethidium bromide (1 ␮g/ml). The expected sizes of cDNA products reported in Table 1 were evaluated on the basis of a standard PCR marker (PCR Markers, Promega, Madison, WI). 2.8. Characterization of PAF receptors by binding assay Cells were suspended at 1 × 107 /ml in DMEM containing 0.1% BSA, and incubated with at serial concentrations from 0.2 to 16.0 pmol/ml for 1 h at 4 ◦ C. An analogous set 3 of H-PAF dilutions containing a 1000-fold excess of unlabeled PAF was used to estimate the aspecific binding of radiolabeled PAF. At the end of the incubation, cells were washed by centrifugation at 15,000 × g for 1 min, resuspended in a 1% solution of sodium dodecyl sulphate, and quantitatively transferred into scintillation vials for counting. Binding data were analyzed according to the Scatchard method using a data handling software (GraFit version 2.0, Erithacus Software Ltd., Staines, UK). 3 H-PAF

3. Results 3.1.

14 C-acetate

incorporation into PAF

Fig. 1 shows a typical radiochromatogram of metabolically-radiolabeled lipids extracted from human Hs294T melanoma cells stimulated with Ca2+ ionophore after being incubated with 14 C-acetate (panel A). In the radiochromatogram, PAF was positioned immediately

A. Fallani et al. / Prostaglandins & other Lipid Mediators 70 (2002) 209–226

215

Fig. 1. Identification of 14 C-PAF in the radiochromatogram of the metabolically radiolabeled lipids from Hs294T cells. Cells were metabolically radiolabeled for 10 min with 14 C-acetate, and then stimulated for 7.5 min with the Ca2+ ionophore A23187. The radiolabeled lipids were extracted from cell cultures, and chromatographed by HPLC (A). 14 C-PAF was identified by comparison with the retention time of a commercial 3 H-PAF standard chromatographed under the same conditions (B).

216

A. Fallani et al. / Prostaglandins & other Lipid Mediators 70 (2002) 209–226

Fig. 2. Synthesis of PAF in B77-AA6, Hs294T, F10-M3 and LS180 cells. Cell cultures were metabolically radiolabeled by incubation for 10 min with 14 C-acetate in PBS–CaCl2 (1.3 mM), and then stimulated with Ca2+ ionophore A23187 (7.5 ␮M) for the indicated periods of time. 14 C-PAF was extracted from cell cultures with Bligh and Dyer’s procedure [25], separated from the other radiolabeled lipid components by HPLC, and assayed with a radiometric detector. Values represent the mean ± S.E. of 2–3 experiments performed in duplicate (range of variation ± 10%).

after the phosphatidylcholine-sphingomyelin fraction, as shown by comparison with the retention time of a commercial standard of 3 H-PAF (panel B). As seen in Fig. 2, PAF synthesis stimulated by Ca2+ ionophore A23187 was rather active in F10-M3 murine melanoma, human Hs294T melanoma and LS180 colon carcinoma, but rather limited in murine B77-AA6 cells, and absent in RSV-transformed (B77-3T3) cells. 3.2. Quantitative analysis of PAF by the GC–MS/NICI procedure A typical selected ion monitoring chromatogram of PAF-pentafluorobenzoyl derivatives co-eluted with d3 -PAF is reported in Fig. 3 (panel A), while NICI mass spectra of PAFand d3 -PAF-pentafluorobenzoyl derivatives (m/z 552 and 555, respectively) were illustrated in panel B. Analysis of these spectra showed that hexadecyl-PAF was the only molecular species present in our systems of cell lines. Under basal conditions, production of PAF was quite limited in murine F10-M3 melanoma and human Hs294T melanoma, HT1080 fibrosarcoma and LS180 colon carcinoma cells, and very low in the murine transformed fibroblastic lines (Table 2). In the presence of Ca2+ ionophore A23187, melanoma cell lines, both human and murine, and human colon carcinoma and fibrosarcoma cell lines synthesized amounts of PAF comparable to those

A. Fallani et al. / Prostaglandins & other Lipid Mediators 70 (2002) 209–226

217

Fig. 3. (A) Selected ion monitoring chromatogram of pentafluorobenzoyl derivatives of PAF extracted from Hs294T cells, and of an added d3 -PAF internal standard. (B) NICI mass spectra of the pentafluorobenzoyl derivatives of C16-PAF (m/z 552) and of an added d3 -C16 PAF internal standard (m/z 555).

218

A. Fallani et al. / Prostaglandins & other Lipid Mediators 70 (2002) 209–226

Table 2 Amounts of PAF in cultures of transformed cells stimulated with Ca2+ ionophore A23187a Cell origin

Agent None

Ionophoreb

Melanoma (human) Hs294T

118

3680

Colon carcinoma (human) LS180

121

2766

43

1100

Melanoma (murine) F10-M3

101

3200

Fibroblasts (murine) B77-3T3 B77-AA6

Undetectable 20

Peritoneal macrophages (murine)

100

Fibrosarcoma (human) HT1080

219 509 4500

a PAF was determined by GC–MS using a deuterated PAF internal standard. Values are pg × 106 cells and represent the mean of 2–3 different experiments (range of variation ± 10%). b Each cell line was incubated with Ca2+ ionophore A23187 for the time required for a maximal activity.

generated by murine resident peritoneal macrophages under the same conditions. Among the murine transformed cell lines, the more metastatic B77-AA6 cells stimulated by Ca2+ ionophore produced higher amounts of PAF than the low metastatic B77-3T3 cells. However, synthesis of PAF in both high and low metastatic murine fibroblastic lines did not reach the levels observed in the other systems of transformed cell lines considered in our study. 3.3. Synthesis of PAF in cell lines stimulated by TNFα and INFγ Growth in the presence of TNF␣ stimulated the production of PAF in human Hs294T melanoma and LS180 colon carcinoma cells and in murine B77-AA6 cells, but not in F10-M3 cells (Fig. 4A). In particular, PAF synthesis stimulated by TNF␣ in Hs294T melanoma cells and B77-AA6 cells corresponds, at the maximal rate, to 100% of that found in the same cells exposed to the Ca2+ ionophore. Hs294T melanoma and LS180 colon carcinoma cells as well as F10-M3 cells exposed to IFN␥, synthesized amounts of PAF which correspond, at the maximal rate, to 35, 100, and 80%, respectively, of those found in the same cells stimulated with Ca2+ ionophore. PAF synthesis was not modified in B77-AA6 cells exposed to IFN␥. 3.4. Expression of mRNA transcripts for TNFαR and IFNγ R in transformed cell lines Human Hs294T melanoma and LS180 colon carcinoma cell lines (Fig. 5A) as well as murine F10-M3 and B77-AA6 cell lines (Fig. 5B) were found to express mRNAs specific

A. Fallani et al. / Prostaglandins & other Lipid Mediators 70 (2002) 209–226

219

Fig. 4. Effect of TNF␣ (A) and IFN␥ (B) on the synthesis of PAF in B77-AA6, F10-M3, Hs294T and LS180 cells. PAF was assayed by GC–MS. Values represent the mean ± S.E. of 2–3 experiments performed in duplicate (range of variation ± 10%).

220

A. Fallani et al. / Prostaglandins & other Lipid Mediators 70 (2002) 209–226

Fig. 5. Detection of IFN␥R and TNF␣R expression in human and murine transformed cell lines. (A) Lane M, PCR marker (range 50–1000 bp); lanes 1 and 4, IFN␥R (110 bp) in Hs294T and LS180 cells, respectively; lanes 2 and 5, TNF␣R 80 (319 bp) in Hs294T and LS180 cells, respectively; lanes 3 and 6, TNF␣R 60 (375 bp) in Hs294T and LS180 cells, respectively; (B) Lane M, PCR marker (range 50–1000 bp); lanes 1 and 4, IFN␥R (660 bp) in F10-M3 and B77-AA6 cells, respectively; lanes 2 and 5, TNF␣R 75 (446 bp) in F10-M3 and B77-AA6 cells, respectively; lanes 3 and 6, TNF␣R 55 (491 bp) in F10-M3 and B77-AA6 cells, respectively. RNAs from Hs294T, LS180, F10-M3 and B77-AA6 cells were reverse transcripted into cDNA and then subjected to cycles of PCR in the presence of the appropriate primers (Table 1), as described in Section 3.

for type 1 (p55–60) and type 2 (p75–80) TNF␣ receptors. mRNA specific for IFN␥R was detected in all transformed cell lines used in our study (Fig. 5A and B). 3.5. Synthesis of PAF in cell lines stimulated by exogenous PAF By measuring the incorporation of 14 C-acetate into PAF, we found that the synthesis of PAF was stimulated by exogenous PAF in murine B16-M3 melanoma cells and human Hs294T melanoma and LS180 colon carcinoma cell lines, but not in B77-AA6 cells (Fig. 6). Table 3 Characteristics of PAF receptors in human melanoma (Hs294T), colon carcinoma (LS180) cells, and in a murine fibroblastic transformed cell line (B77-AA6)a Cell line

Number of receptors (sites × 10−3 /cell)

Dissociation constant (nmol/l)

Hs294T LS180 B77-AA6

70 ± 10 108 ± 24 n.d.b

0.6 ± 0.1 2.1 ± 0.5 n.d.b

a Tumor cells suspended in DMEM were incubated with varying amounts of 3 H-PAF for 1 h at 4 ◦ C, either in the absence or in the presence of an excess of unlabeled PAF in order to determine the specific binding of radiolabeled PAF. Bound radiolabeled PAF was then separated from unbound PAF and the specifically-bound radioactivity was measured. Binding data were analyzed by the Scatchard method. b Not detectable under the analytical conditions used.

A. Fallani et al. / Prostaglandins & other Lipid Mediators 70 (2002) 209–226

221

Fig. 6. Effect of PAF on PAF synthesis in B77-AA6, Hs294T, F10-M3 and LS180 cells. Cell cultures were metabolically radiolabeled by incubation for 10 min with 14 C-acetate in PBS–CaCl2 (1.3 mM), and then stimulated with carbamyl-PAF (100 nM) for the indicated periods of time. 14 C-PAF was separated from the other radiolabeled lipid components by HPLC, and assayed by the use of a radiometric detector. Values represent the mean ± S.E. of 2–3 experiments performed in duplicate (range of variation ± 10%).

Fig. 7. Expression of PAFR in murine transformed cell lines. Lane M, PCR marker (range 50–1000 bp); lane 1, PAFR (399 bp) in F10-M3; lane 2, PAFR in B77-AA6 cells. RNAs from F10-M3 and B77-AA6 cells were reverse transcripted into cDNA, and then subjected to cycles of PCR in the presence of the appropriate primers (Table 1), as described in Section 3.

222

A. Fallani et al. / Prostaglandins & other Lipid Mediators 70 (2002) 209–226

3.6. PAFR expression in human and murine transformed cell lines As shown in Table 3, human Hs294T melanoma and LS180 colon carcinoma cells, but not B77-AA6 cells, expressed functionally active PAF receptors which exhibited a high affinity for the PAF molecule, as determined by ligand binding assay. The analysis of total RNA by PCR revealed the presence of mRNA transcripts for PAFR only in F10-M3 (Fig. 7). The presence of mRNA transcripts for PAFR was previously reported in Hs294T melanoma and LS180 colon carcinoma cell lines [33].

4. Discussion There is a growing body of evidence that tumor growth and invasiveness are promoted by biological mediators generated in the host’s microenvironment. Indeed, cytokines such as TGF␤, TNF␣, IFN␥ and IL-1 were found to potentiate adhesiveness [34], invasiveness [35,36], motility [37] and collagenolytic activity [38,39] of tumor cells. Arachidonic acid metabolites are also involved in the invasive properties of tumor cells. In fact, 12-HETE, a lipoxigenase metabolite released by tumor cells [40] was found to promote the adhesion of tumor cells to the endothelium [41] and to induce endothelial cell retraction [41]. Furthermore, PGE2 enhances tumor cell motility [42] and also inhibits the activity of NK cells [43], an important effector of antimetastatic host defenses [44,45]. Moreover, a role of PAF in the formation of tumor metastasis is indicated by the increased lung colonization of B16-F10 cells in animals post-injected with PAF [46]. In view of the capacity of TNF␣ and IFN␥ to promote PAF synthesis [12,17,22,23], it is possible that the enhancement of metastatic potential in tumor cells treated with TNF␣ [35,47] or IFN␥ [48] be mediated by PAF. The involvement of PAF in metastastatic diffusion might be related to PAF’s capacity to stimulate angiogenesis [17,49], motility [50,51] and adhesiveness to endotelium [33]. In the tumoral microenvironment, PAF could be synthesized by host cells [1] as well as by tumor cells as indicated by the seminal studies by Foa et al. [16], Kester et al. [15], Maggi et al. [19], Bussolino et al. [17], Sobhani et al. [18], Bussolati et al. [20]. We studied PAF synthesis in a series of transformed cells of a different origin by combining 14 C-acetate incorporation into PAF with quantitative analysis of PAF performed by the use of GC–MS. The combined use of the two procedures was motivated by the fact that, although the procedure based on 14 C-acetate incorporation was indicative of an effective PAF synthesis in our systems of transformed lines, the measurement of PAF by GS–MS gave more precise information on the entity of PAF production. The study of PAF synthesis in the presence of Ca2+ ionophore revealed that the various transformed cells considered in our study generated different amounts of PAF according to their histotype and/or biological properties. The highest amounts of PAF were produced by murine and human melanoma lines, followed by human colon carcinoma and human fibrosarcoma lines. A rather limited synthesis of PAF was found in transformed murine fibroblastic lines. However, the high metastatic B77-AA6 cells generated greater amounts of PAF compared to their low metastatic counterpart, a behavior possibly related to the finding that ether-linked lipids, the precursors in PAF [1], are particularly abundant in metastatic cells [52–55].

A. Fallani et al. / Prostaglandins & other Lipid Mediators 70 (2002) 209–226

223

Our studies also revealed that PAF receptors were expressed by certain transformed cell lines such as human colon carcinoma and melanoma lines, but not by B77-AA6 cells. The expression of PAF receptors has previously been shown in other types of malignant cells [17,19,49,56–58]. We found that stimulation with exogenous PAF enhanced PAF synthesis only in those lines which expressed PAF receptors. Stimulation of PAF synthesis by exogenous PAF has already been demonstrated in endothelial cells [59]. Given the contribution by cytokines to the tumoral inflammatory microenvironment, we examined to what extent our systems of transformed cells synthesized PAF in the presence of TNF␣ and IFN␥. We found that human colon carcinoma and melanoma cells exposed to TNF␣ and IFN␥ synthesized PAF to levels comparable to those found in the same cells examined in the presence of Ca2+ ionophore. IFN␥, but not TNF␣, stimulated PAF synthesis in murine F10-M3 cells, while. TNF␣, but not IFN␥, induced PAF synthesis in murine B77-AA6 cells. Bussolino et al. [17] reported an increased PAF synthesis in Kaposi’s sarcoma lines following stimulation with TNF␣. A functional inactivity or structural deficiency of TNF␣ or IFN␥ receptors might account for the lack of response to specific cytokines observed in murine transformed cells which, however, express mRNAs specific for the corresponding receptors. To conclude, the results of the present study indicate that transformed cells, regardless of their origin, can synthesize PAF, although to different degrees depending on their histotype. This capacity appears to be inducible by biological agents present in the host microenvironment, such as inflammatory cytokines. This characteristic may contribute to the autonomy of tumor cells by providing them with a potent biological agent, such as PAF, capable of recruiting cell functions which are relevant to metastatic diffusion, e.g. stimulation of cell motility [5,6] and adhesiveness [8], induction of angiogenesis [6] and expression of collagenases [10]. Moreover, tumor cells that express PAF receptors are equipped with an additional mechanism of autonomy in view of their capacity for being stimulated by PAF produced by the host cells or the tumor cells themselves.

Acknowledgements This work was supported by grants from MURST 40%-Cofin 1999 and Murst ex 60%. References [1] Snyder F. Platelet-activating factor and related acetylated lipids as potent biologically active cellular mediators. Am J Physiol 1990;259:c697–708. [2] Chignard M, Le Couedic JP, Tence M, Vargftig BB, Benveniste J. The role of platelet-activating factor in platelet aggregation. Nature 1979;279:799–800. [3] O’Flaherty JT, Miller CH, Lewis JC, Wykle RL, Bass DA, McCall CE, et al. Neutrophil responses to platelet-activating factor. Inflammation 1981;5:193–201. [4] Prpic V, Uhing RJ, Weiel JE, Jakoi L, Gawdi G, Herman B, et al. Biochemical and functional responses stimulated by platelet-activating factor in murine peritoneal macrophages. J Cell Biol 1988;107:363–72. [5] Bussolino F, Camussi G, Aglietta M, Braquet P, Bosia A, Pescarmona G, et al. Human endothelial cells are targets for platelet-activating factor. I. Platelet-activating factor induces changes in cytoskeletal structure. J Immunol 1987;139:2439–46.

224

A. Fallani et al. / Prostaglandins & other Lipid Mediators 70 (2002) 209–226

[6] Camussi G, Montrucchio G, Lupia E, De Martino A, Perona L, Arese M, et al. Platelet-activating factor directly stimulates in vitro migration of endothelial cells and promotes in vivo angiogenesis by a heparin-dependent mechanism. J Immunol 1995;154:6492–501. [7] Montrucchio G, Lupia E, Battaglia E, Del Sorbo L, Boccellino M, Biancone L, et al. Platelet-activating factor enhances vascular endothelial growth factor-induced endothelial cell motility and neoangiogenesis in a murine matrigel model. Arterioscler Thromb Vasc Biol 2000;20:80–8. [8] McIntyre TM, Zimmerman GA, Prescott SM. Leukotrienes C4 and D4 stimulate human endothelial cells to synthesize platelet-activating factor and bind neutrophils. Proc Natl Acad Sci USA 1986;83:2204–8. [9] Jackson RJ, Bolognese B, Mangar CA, Hubbard WC, Marshall LA, Winkler JD. The role of platelet-activating factor and other lipid mediators in inflammatory angiogenesis. Biochim Biophys Acta 1998;1392:145–52. [10] Bazan HEP, Tao Y, Bazan NG. Platelet-activating factor induces collagenase expression in corneal epithelial cells. Proc Natl Acad Sci USA 1993;90:8678–82. [11] Camussi G, Aglietta M, Coda R, Bussolino F, Pacibello W, Tetta C. Release of platelet-activating factor (PAF) and istamine. II. The cellular origin of human PAF: monocytes, polymorphonuclear neutrophils and basophils. Immunology 1981;42:191–9. [12] Camussi G, Bussolino F, Salvidio G, Baglioni C. Tumor necrosis factor/cachectin stimulates peritoneal macrophages, polymorphonuclear neutrophils, and vascular endothelial cells to synthesize and release platelet-activating factor. J Exp Med 1987;166:1390–404. [13] Mencia-Huerta JM, Benveniste J. Platelet-activating factor and macrophage. 1. Evidence for the release from rat and mouse peritoneal macrophages and not from mastocytes. Eur J Immunol 1979;9:409–15. [14] McIntyre TM, Zimmerman GA, Satoh K, Prescott SM. Cultured endothelial cells synthesize both platelet-activating factor and prostacyclin in response to histamine, bradykinin and adenosine triphosphate. J Clin Invest 1985;76:271–80. [15] Kester M, Nowinski RJ, Holthofer H, Marsden P, Dunn MJ. Characterization of platelet-activating factor synthesis in glomerular endothelial cell lines. Kidney Int 1994;46:1404–12. [16] Foa R, Bussolino F, Ferrando ML, Guarini A, Tetta C, Mazzone R, et al. Release of platelet-activating factor in human leukemia. Cancer Res 1985;45:4483–5. [17] Bussolino F, Arese M, Montrucchio G, Barra L, Primo L, Benelli R, et al. Platelet-activating factor produced in vitro by Kaposi’s sarcoma cells induces and sustains in vivo angiogenesis. J Clin Invest 1995;96:940–52. [18] Sobhani I, Denizot Y, Moizo L, Laigneau JP, Bado A, Laboisse C, et al. Regulation of platelet activating factor production in gastric epithelial cells. Eur J Clin Inv 1996;26:53–8. [19] Maggi M, Bonaccorsi L, Finetti G, Carloni V, Muratori M, Laffi G, et al. Platelet-activating factor mediates an autocrine proliferative loop in the endometrial adenocarcinoma cell line HEC-1A. Cancer Res 1994;54: 4777–84. [20] Bussolati B, Biancone L, Russo S, Rola-Pleszczynsky M, Montrucchio G, Camussi G, et al. PAF produced by human breast cancer cells promotes migration and proliferation of tumor cells and neo-angiogenesis. Am J Pathol 2000;157:1713–25. [21] Di Renzo MF, Bretti S. Characterization of stable spontaneous metastatic variant lines of RSV-transformed mouse fibroblasts. Int J Cancer 1982;30:751–7. [22] Valone FH, Epstein LB. Biphasic platelet-activating factor synthesis by human monocytes stimulated with IL-1-␤, tumor necrosis factor or IFN-␥. J Immunol 1988;141:3945–50. [23] Montrucchio G, Lupia E, Battaglia E, Passerini G, Bussolino F, Emanuelli G, et al. Tumor necrosis factor ␣-induced angiogenesis depends on in situ platelet-activating factor biosynthesis. J Exp Med 1994;180: 377–82. [24] Cecconi O, Calorini L, Mannini A, Mugnai M, Ruggieri S. Enhancement of lung-colonizing potential murine tumor cell lines co-cultivated with activated macrophages. Clin Exp Metastasis 1997;15:94–101. [25] Bligh EG, Dyer WJ. A rapid method of total lipid extraction and purification. Can J Biochem Phys 1959;37:911–7. [26] Bossant MJ, Ninio E, Delautier D, Benveniste J. Biossay of PAF-acether by rabbit platelet aggregation. Meth Enzymol 1990;187:125–30. [27] Triolo A, Bertini J, Mannucci C, Perico A, Pestellini V. Analysis of platelet-activating factor by gas chromatography-mass spectrometry: low-energy electron impact of the corresponding 3-acetyl-2-tertbutyldimethylsilyl derivative. J Chromatogr 1991;568:281–9.

A. Fallani et al. / Prostaglandins & other Lipid Mediators 70 (2002) 209–226

225

[28] Hube F, Birgel M, Lee Y-M, Hauner H. Expression pattern of tumor necrosis factor receptors in subcutaneous and omental human adipose tissue: role of obesity and non-insulin-dependent diabetes mellitus. Eur J Clin Invest 1999;29:672–8. [29] Nagao M, Nakajima Y, Kanehiro H, Hisanaga M, Aomatsu Y, Ko S, et al. The impact of interferon gamma receptor expression on the mechanism of escape from host immune surveillance in hepatocellular carcinoma. Hepatology 2000;32:491–500. [30] Chau LI, Peck K, Yen HH, Wang JY. Agonist-induced down-regulation of platelet activating factor receptor gene expression in U937 cells. Biochem J 1994;301:911–6. [31] Zhuang L, Wang B, Shinder GA, Shivji GM, Mak TW, Sauder DN. TNF receptor p55 plays a pivotal role in murine keratinocyte apoptosis induced by ultraviolet B irradiation. J Immunol 1999;162:1440–7. [32] Wei YP, Kita M, Shinmura K, Yan XQ, Fukuyama R, Fushiki S, et al. Expression of IFN-␥ in cerebrovascular endothlial cells from aged mice. J Interferon Cytokine Res 2000;20:403–9. [33] Mannori G, Barletta E, Mugnai G, Ruggieri S. Interaction of tumor cells with endothelia: role of plateletactivating factor. Clin Exp Metastasis 2000;18:1–8. [34] Okahara H, Yagita H, Miyake K, Okumura K. Involvement of very late activation antigen 4 (VLA-4) and vascular cell adhesion molecule 1 (VCAM-1) in tumor necrosis factor ␣ enhancement of experimental metastasis. Cancer Res 1994;54:3233–6. [35] Lollini PL, De Giovanni C, Nicoletti G, Bontadini A, Tazzari PL, Landuzzi L, et al. Enhancement of experimental metastatic ability to tumor necrosis factor-alpha alone or in combination with interferon-gamma. Clin Exp Metastasis 1990;8:215–24. [36] Welch DR, Fabra A, Nakajima M. Transforming growth factor ␤ stimulates mammary adenocarcinoma cell invasion and metastatic potential. Proc Natl Acad Sci USA 1990;87:7678–82. [37] Rosen EM, Goldberg ID, Liu D, Setter E, Donovan MA, Bhargava M, et al. Tumor necrosis factor stimulates epithelial tumor cell motility. Cancer Res 1991;51:5315–21. [38] Lauricella-Lefebvre MA, Castronovo V, Sato H, Seiki M, French DL, Merville MP. Stimulation of 92kDa type IV collagenase promoter and enzyme expression in human melanoma cells. Invasion Metastasis 1993;13:289–300. [39] Dabbous MK, North SM, Haney L, Nicolson GL. Effects of mast cells-macrophage interaction on the production of collagenolytic enzymes by metastatic tumor cells and tumor-derived and stromal fibroblasts. Clin Exp Metastasis 1995;13:33–41. [40] Chen YQ, Duniec ZM, Liu B, Hagmann W, Gao X, Shimoji K, et al. Endogenous 12(S)-HETE production by tumor cells and its role in metastasis. Cancer Res 1994;54:1574–9. [41] Honn KV, Tang DG, Gao X, Butovich IA, Liu B, Timar J, et al. 12-lipoxygenases and 12(S)-HETE: role in cancer metastasis. Cancer Metastasis Rev 1994;13:365–96. [42] Young MR, Okada F, Tada M, Hosokawa M, Kobayashi H. Association of increased tumor cell responsiveness to prostaglandin E2 with more aggressive tumor behavior. Invasion Metastasis 1991;11:48–57. [43] Fulton AM, Heppner G. Relationships of prostaglandin E and natural killer sensitivity to metastatic potential in murine mammary adenocarcinomas. Cancer Res 1985;45:4779–84. [44] Hanna N, Burton RC. Definitive evidence that natural killer (NK) cells inhibit experimental tumor metastasis in vivo. J Immunol 1981;127:1754–8. [45] Gorelik E, Wiltrout RH, Okumura K, Habu S, Herberman R. Role of NK cells in the control of metastatic spread and growth of tumor cells in mice. Int J Cancer 1982;30:107–12. [46] Im SY, Ko HM, Kim JW, Lee HK, Ha TJ, Lee HB, et al. Augmentation of tumor metastasis by plateletactivating factor. Cancer Res 1996;56:2662–5. [47] Orosz P, Echtenacher BE, Falk W, Ruschoff J, Weber D, Mannel DN. Enhancement of experimental metastasis by tumor necrosis factor. J Exp Med 1993;177:1391–8. [48] Ramani P, Balkwill FR. Enhanced metastases of a mouse carcinoma after in vitro treatment with murine interferon gamma. Int J Cancer 1987;40:830–4. [49] Montrucchio G, Sapino A, Bussolati B, Ghisolfi G, Rizea-Savu S, Silvestro L, et al. Potential angiogenic role of platelet-activating factor in human breast cancer. Am J Pathol 1998;153:1589–96. [50] Biancone L, Cantaluppi V, Boccellino M, Bussolati B, Del Sorbo L, Conaldi PG, et al. Motility induced by human immunodeficiency virus-1 Tat on Kaposi’s sarcoma cells requires platelet-activating factor synthesis. Am J Pathol 1999;155:1731–9.

226

A. Fallani et al. / Prostaglandins & other Lipid Mediators 70 (2002) 209–226

[51] Boccellino MR, Biancone L, Cantaluppi V, Ye RD, Camussi G. Effect of platelet-activating factor receptor expression on CHO cell motility. J Cell Physiol 2000;183:254–64. [52] Friedberg SJ, Smaidek J, Anderson K. Surface membrane O-alkyl lipid concentration and metastasizing behavior in transplantable rat mammary carcinomas. 1986;46:845–9. [53] Fallani A, Mannori G, Ruggieri S. Composition of ether-linked sub-classes of glycerophospholipids in clones with a different metastatic potential isolated from a murine fibrosarcoma line (T3 cells). Int J Cancer 1995;62:230–2. [54] Calorini L, Fallani A, Tombaccini D, Mugnai G, Ruggieri S. Lipid composition of cultured B16 melanoma cell variants with different lung-colonizing potential. Lipids 1987;22:651–6. [55] Calorini L, Fallani A, Barletta E, Mugnai G, Di Renzo MF, Comoglio P, et al. Lipid characteristics of RSVtransformed Balb/c3T3 cell lines with different spontaneous metastatic potentials. Lipids 1989;24:685–90. [56] Mutoh H, Ishii S, Izumi T, Kato S, Shimizu T. Platelet-activating factor (PAF) positively regulates the expression of human PAF-receptor transcript 1 (leukocyte type) through NF-kappa B. Biochem Biophys Res Commun 1994;205:1137–42. [57] Dupuis F, Levasseur S, Jean-Louis F, Dulery C, Praloran V, Denizot Y, et al. Production, metabolism and effect of platelet-activating factor on the growth of the human K562 erythroid cell line. Biochim Biophys Acta 1997;1359:241–9. [58] Merendino N, Dwinell MB, Varki N, Eckmann L, Kagnoff MF. Human intestinal epithelial cells express receptors for platelet-activating factor. Am J Physiol 1999;277:810–8. [59] Heller R, Bussolino F, Ghigo D, Garbarino G, Pescarmona G, Till U, et al. Human endothelial cells are target for platelet-activating factor. II. Platelet-activating factor induces platelet-activating factor synthesis in human umbilical vein endothelial cells. J Immunol 1992;149:3682–8.