Bioactive polar lipids from Chroococcidiopsis sp. (Cyanobacteria)

Comparative Biochemistry and Physiology, Part B 142 (2005) 269 – 282 www.elsevier.com/locate/cbpb

Bioactive polar lipids from Chroococcidiopsis sp. (Cyanobacteria) Smaragdi Antonopoulou a,*, Haralabos C. Karantonis a, Tzortzis Nomikos a, Alexandra Oikonomou a, Elizabeth Fragopoulou a, Adriani Pantazidou b b

a Department of Science of Dietetics – Nutrition, Harokopio University, 70 El. Venizelou Str., 176 71, Athens, Greece Department of Ecology and Systematics, Faculty of Biology, National and Kapodistrian University of Athens, Panepistimioupolis, 157 71, Athens, Greece

Received 7 April 2005; received in revised form 13 July 2005; accepted 14 July 2005

Abstract Many studies indicate that various bioactive metabolites subsist in cyanobacteria. Glycolipids of cyanobacteria are reported as molecules that exert specific bioactivities. In this study, total lipids of Chroococcidiopsissp., a coccoid cyanobacterium isolated from a Greek cave, were separated into neutral and polar-lipids and the latter were further fractionated by high-performance liquid chromatography (HPLC). Each polar lipid fraction was tested in vitro for its ability to inhibit platelet-activating factor (PAF)- and thrombin-induced washed rabbit platelet aggregation and/or to cause platelet aggregation. The structures of the most active fractions were elucidated by biological assays and identified by electrospray mass spectrometry. One fraction was a potent inhibitor of PAF-induced platelet aggregation. Structural studies of this fraction indicated the existence of phospho-glyco analog of ceramide. Another fraction that was a potent inhibitor of PAF- as well as of thrombin-induced platelet aggregation was structurally elucidated as a phospho-acetylated glyco-analog of diglyceride. The fraction that induced platelet aggregation was identified as a phospho-acetylated-glyco analog of ceramide. These novel bioactive polar lipids in cyanobacteria in regard to the structure and biological activity may contribute to the allergic character of cyanobacteria. D 2005 Elsevier Inc. All rights reserved. Keywords: Cyanobacterium; Chroococidiopsis sp.; Lipids; Glycolipids; Ceramides; Platelets; Aggregation; Mass spectrometry; Platelet-activating factor

1. Introduction Cyanobacteria are commonly found in habitats exposed to extreme environmental conditions. Chroococcidiopsis is a very primitive photosynthetic and versatile cyanobacterium, present worldwide, living in a wide range of extreme environments, e.g. arid, hot and cold deserts, hypersaline lagoons and ponds and thermal springs. Members of the genus Chroococcidiopsis are known for their resistance to desiccation and for their ability to survive under nutrient and light deprivation (Anagnostidis and Pantazidou, 1988). According to Komarek and Anagnostidis (1986) both original and several later descriptions of Chroococcidiopsis sp. (order Chr. thermalis) belong without doubt to the Chroococcales family Xenococcaceae. On the other hand,

* Corresponding author. Tel.: +30 210 9549305; fax: +30 210 9577050. E-mail address: [email protected] (S. Antonopoulou). 1096-4959/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.cbpc.2005.07.007

according to Rippka et al. (1979) Chroococcidiopsis belongs to the Pleurocapsalean subgroup. The archaeobacterium of Chroococcidiopsis is an interesting cyanobacterium due to the particular characteristics of its members, e.g. resistance to desiccation, radio-resistance, halotolerance and ability to survive under nutrient and light deprivation, therefore, the biochemical, structural and genetic mechanisms of its strains arouse interest. On the other hand, archaebacteria contain unique ether lipids (neutral dialkyl and tetraalkyl glycerides, dialkyl phospholipids and glycolipids), which are inherent only in these organisms (Kates, 1992). It is well known that ether lipids –cellular structural elements of the earliest living organisms on the earth –are resistant to acidic and alkaline hydrolysis, maintaining intact the bacterial membranes even in the extremely salty, alkaline, acidic, or hot environment (Kates, 1992). Even though structure and composition of ether lipids in general have been significantly changed during evolution, some of them, originating from the early

270

S. Antonopoulou et al. / Comparative Biochemistry and Physiology, Part B 142 (2005) 269 – 282

stages of their evolution, did not change significantly during further evolution while their biological functions increased and became highly specific in certain cells (Kulikov and Muzya, 1997a,b). One such ether lipid is platelet activating factor (PAF) identified as 1-O-alkyl-2-acetyl-sn-glycero-3-phosphocholine (Demopoulos et al., 1979), which is a highly active biological regulator (Braquet et al., 1987; Kulikov and Muzya, 1996) and one of the most potent endogenous mediators in inflammation, allergy, immune disorders and ischemic diseases (Koltai et al., 1991). The existence of PAF has been well established in various protozoans, in yeast, in higher plants, in earthworms, in non-mammalian vertebrates and in mammals (Kulikov and Muzya, 1997a,b; Antonopoulou et al., 1996). PAF is believed to be a member of a large family of lipid mediators since various synthetic as well as naturally occurring compounds have been found to exhibit PAF-like activity (Demopoulos and Antonopoulou, 1996; Avramopoulou et al., 1997). Cyanobacteria comprise a rich source of novel bioactive metabolites, including many cytotoxic, antifungal and antiviral compounds (Patterson et al., 1994). The toxin content of cyanobacteria is not related to the well-recorded

allergic skin irritation and allergic asthma cyanobacteria exposure (Falconer, 1996; Torokne et al., 2001), implying a possible role of PAF, since these actions have been well established for this lipid mediator (Jenks et al., 1999; Henig et al., 2000). The major polar lipids isolated from cyanobacteria in general and from genus Chroococcidiopsis are monogalactosyl diacylglycerol (MGDG), digalactosyl diacylglycerol (DGDG), phosphatidyl glycerol (PG), and sulfoquinovosyl diacylglycerol (SQDG) (Murata and Nishida, 1987; Rˇezanka et al., 2003). Rˇezanka et al. (2003) identified also N,N,N-trimethylhomoserin-4-O-yl-diacylglycero (DGTS) in genus Chroococcidiopsis, while an unidentified minor glycolipid was mentioned. Lately, lysoglycerolipids corresponding to their monoacylglycerol types were isolated and identified from the total lipid extract of cyanobacterium Synechocystis sp. PCC 6803 (Kim et al., 1999). The relative proportions of these lipids are different in the envelope, cell and thylakoid membranes (Omata and Murata, 1983). Cyanobacteria have different abilities to synthesize unsaturated fatty acids (Zepke et al., 1978) and have been divided into four groups depending on the acyl group composition of the membrane lipids, however, the

Fig. 1. Representative HPLC separation of glycolipids from Chroococcidiopsissp. on a normal phase column. The solvent system along with the separation conditions is described in the Materials and methods section. PE, phosphatidylethanolamine; PC, phosphatidylcholine; SM, sphingomyelin; LPC, lysophosphatidylcholine; NL, neutral lipids; GALCER, galactosylcerebroside; and DGDG, digalactosyldiglyceride.

S. Antonopoulou et al. / Comparative Biochemistry and Physiology, Part B 142 (2005) 269 – 282

fatty acid groups do not correspond to taxonomic groups (Murata and Wada, 1995). In a recent work concerning the study of glycolipids of Scytonema julianum (Antonopoulou et al., 2005) the existence of three phosphoglycolipids was demonstrated that antagonized PAF and exerted PAF-like activity. These molecules have either sphingosine or phosphatidylglycerol backbone. The existence of the sphingosine backbone is in accordance with previous results demonstrating the existence of acetyl-sphingomyelin in the phospholipid fraction of the S. julianum (Antonopoulou et al., 2002). It is worthy of note that Chroococcidiopsissp. (order Chroococcales,coccoid) differs from S. julianum (order Scytonematales, filamentous) by systematic classification and morphology. Among the biological activities of various glycolipids distributed in photosynthetic eukaryotic and prokaryotic organisms, these are of antiviral, antialgal, antitumorpromoting, hemolytic and anti-inflammatory activities (Reshef et al., 1997). Aims of this study were to establish data on the biological activity of the glycolipids of Chroococcidiopsis sp. and to make an effort to identify these compounds.

2. Materials and methods 2.1. Reagents All reagents and chemicals were of analytical grade and supplied by Merck (Darmstadt, Germany). High-performance liquid chromatography (HPLC) solvents were from Rathburn (Walkerburn, Peebleshire, UK). Lipid standards, bovine serum albumin (BSA), thrombin and BN 52021 were obtained from Sigma (St. Louis, MO, USA). Semisynthetic PAF (80% C-16PAF and 20% C-18PAF) was synthesized in our laboratory as previously described (Demopoulos et al., 1979). PAF-acetylhydrolase was purified from human serum according to the method of Stafforini et al. (1987). Chromatographic material used for thin-layer chromatography (TLC) was silica gel H-60 (Merck, Darmstadt, Germany). 2.2. Instrumentation The separation of lipids was performed at room temperature on a HPLC Series 1100 liquid chromatography model (Hewlett Packard, Waldbronn, Germany) equipped with a 100-AL loop Rheodyne (7725 i) loop valve injector, a degasser G1322A, a quad gradient pump G1311A and a HP UV spectrophotometer G1314A as a detection system. The spectrophotometer was connected to a Hewlett-Packard (Hewlett Packard) model HP-3395 integrator– plotter. The separation of polar lipids was performed on a Partisil 10 Am SCX column (250  4.6 mm i.d.) from MZ Analysentechnik (Mainz, Germany) with an SCX (20  4.0 mm i.d.) precolumn cartridge. The separation of glycolipids was

271

performed on a normal phase column, Sphereclone 5u NH2 250  4.6 mm i.d., from Phenomenex (Hurdsfield Ind. Est., UK). The eluted substances were detected using UV detection at 208 nm. PAF-induced aggregation was measured in a Chrono-Log (Havertown, PA, USA) aggregometer (model 400-VS) coupled to a Chrono-Log recorder. The electrospray ionization (ESI) mass spectrometry experiments were performed on an API 100 Perkin Elmer SCIEX single quadrupole mass spectrometer. Samples were dissolved at a concentration of 10 ng/AL in 1 : 1 v / v aqueous methanol. Electrospray samples are typically introduced into the mass analyzer at a rate of 4.0 AL/min. The positive and negative ions, generated by charged droplet evaporation, enter the analyzer through an interface plate and a 100-mm orifice, while the declustering potential is maintained between 60 and 100 V to control the collisional energy of the ions entering the mass analyzer. The emitter voltage is typically maintained at 4000 V. Tandem mass spectrometric (MS/MS) studies were performed on an API III Perkin Elmer SCIEX triple quadrupole mass spectrometer by selecting a precursor ion with the first (quadrupole) analyzer. These ions were transmitted to a collision cell where they were bombarded with argon gas at energy of 10 – 50 eV. The resulting fragment ions were m / z analyzed by the second (time-of-flight) analyzer. 2.3. Sampling, isolation and culture conditions of Chroococcidiopsis The Chroococcidiopsis strain used in this study was isolated from marble from the ancient theatre of Dionysus, which is situated in Athens of Greece on the south side of the Acropolis. The theatre was built in marble in the 4th

Table 1 Biological activity of the HPLC fractions of Chroococcidiopsis sp. HPLC fraction

Bioactivity

G2 G3 G4 G5 G6 G7 G8 G9 G10 G11 G12 G13 G14 G15

Inhibition of platelet Inhibition of platelet n.d. Inhibition of platelet Inhibition of platelet Inhibition of platelet Inhibition of platelet n.d. n.d. Inhibition of platelet Platelet aggregation Inhibition of platelet Inhibition of platelet n.d.

IC50PAF (AL)

Inhibition of thrombin at the IC50PAF (AL) amount (%)

aggregation aggregation

22.0 22.5

100 100

aggregation aggregation aggregation aggregation

21.0 15.5 31.0 19.0

60 100 97 100

aggregation

35.0

0

aggregation aggregation

5.0 24.0

0 0

Lipids were purified on a normal phase column. The IC50PAF value constitutes the average value of three different experiments performed on different batches of washed rabbit platelets and is expressed as volume of fraction in microliter dissolved in 2.5 mg BSA/mL saline. n.d.: not detected.

272

S. Antonopoulou et al. / Comparative Biochemistry and Physiology, Part B 142 (2005) 269 – 282

century BC with its main features being preserved. The strain was characterized as endolithic as it was isolated underneath of marble exfoliation. Our strain was assigned to genus Chroococcidiopsis as it is characterized by nonpolarized cells dividing consecutively (sometimes in rapid succession) in many planes and finally forming non-motile baeocytes. Further species classification of this strain was not possible, as there is no satisfactory species concept for the genus Chroococcidiopsis. The isolation and the culture conditions of Chroococcidiopsis sp. were the same as the ones described in a previous study (Antonopoulou et al., 2002). 2.4. Extraction and isolation of glycolipids Lipids were extracted from the cell suspensions using the method of Bligh and Dyer (1959) yielding about 70 mg of total lipids. The separation of total lipids into polar and neutral lipids as well as the removal of pigments was achieved by two successive preparative TLC (Antonopoulou et al., 1996). Briefly, the chloroform phase from the extraction procedure, containing the total lipids, was

evaporated to dryness and the extracts were redissolved in a small volume of chloroform/methanol 1 : 1 (v / v) and applied to TLC plates. The plates were developed in petroleum ether (b.p. 40 – 60 -C)/benzene/acetic acid 30 : 70 : 2 (v / v / v). Polar lipids along with the pigments remained in the origin, while neutral lipids migrated along the plates. The band of polar lipids with pigments as well as the band of neutral lipids were scrapped off separately and extracted according to Bligh and Dyer (1959), centrifuged and the organic solvents were phased by adding appropriate volumes of chloroform and water to arrive at a final chloroform/ methanol/water ratio of 1 : 1 : 0.9 (v / v / v). The chloroform extracts that contain polar lipids and pigments were evaporated to dryness and redissolved in a small volume of chloroform/methanol 1 : 1 (v / v). The above extracts were rechromatographed on TLC plates, using acetone/methanol/water 40 : 20 : 1 (v / v / v) as the developing system. Pigments migrated near the solvent front while polar lipids migrated along the plate. The fraction of polar lipids (PL) was recovered. Since glycolipids are partitioned in both fractions of neutral and polar lipids (Antonopoulou et al., 1996), the following steps were necessary in order to

Fig. 2. Positive (A) and negative (B) ESI-MS of fraction G6 from Chroococcidiopsis sp. glycolipids. The m / z ranges from 0 to 1200. The proposed structure along with identified fragmentations is also represented.

S. Antonopoulou et al. / Comparative Biochemistry and Physiology, Part B 142 (2005) 269 – 282

reconstitute the glycolipid extract. The fraction of PL was further fractionated on a cation exchange HPLC column 10 Am using an isocratic elution system consisting of acetonitrile 60% and methanol/water 4 : 1 (v / v) 40% (v / v). The PL fraction eluted from 0 to approximately 6 min under the above HPLC conditions and contained some phospholipids along with polar glycolipids (Antonopoulou et al., 2002) and the fractions of neutral lipids from the TLC separation were pooled together. The above fraction was further fractionated on a normal phase column, Sphereclone 5u NH2. The solvent system consisted of an isocratic elution with 100% solvent A (acetonitrile/methanol: 90 / 10) for 7 min followed by a linear gradient to 100% solvent B (acetonitrile/ methanol: 83/17) in 3 min, a hold for 25 min in 100% B followed by a linear gradient to 100% solvent C (methanol) in 5 min and finally a hold in 100% C. 2.5. Biological assay PAF and the examined samples were dissolved in 2.5 mg BSA/mL saline. Thrombin was dissolved in saline. Various concentrations of the examined samples were added into the aggregometer cuvette and the aggregation

273

induced by the sample was studied in a Chrono-Log aggregometer. The aggregatory activity of the sample was expressed as microliters of the fraction dissolved in 2.5 mg BSA/mL saline, which is able to induce 50% of the maximum reversible aggregation of the respective sample. Experiments with the PAF receptor specific antagonist BN 52021 0.1 mM (0.3% DMSO in water) were also performed. BN 52021 was added to the aggregometer cuvette containing washed rabbit platelets 1 min prior to the addition of the examined sample. This experiment was carried out according to Lazanas et al. (1988). In desensitization and cross-desensitization experiments, platelets were desensitized by the addition of the test lipid to the platelet suspension at a concentration that caused reversible aggregation. Second stimulation with PAF or thrombin was performed immediately after complete disaggregation. The platelet aggregation induced by PAF (1.25  10 10 M, final concentration) was measured as PAF-induced aggregation, in washed rabbit platelets, before (considered as 0% inhibition) and after the addition of various concentrations of the examined sample (Lazanas et al., 1988). Consequently, the plot of percentage of inhibition (ranging from 20% to 80%) versus different

Fig. 3. Positive (A) and negative (B) ESI-MS of fraction G6 from Chroococcidiopsis sp. glycolipids. The m / z ranges from 0 to 300.

274

S. Antonopoulou et al. / Comparative Biochemistry and Physiology, Part B 142 (2005) 269 – 282

concentrations of the sample is linear. From this curve, the concentration of the sample that inhibited 50% PAFinduced aggregation was calculated. This value is defined as the IC50. This experiment was also performed with thrombin (0.125 units/cuvette) in order to assess the inhibition to thrombin-induced aggregation. 2.6. Treatment with acetylhydrolase The effect of PAF acetylhydrolase of human serum, an enzyme specific to short or intermediate length sn-2 chains, on the ability of lipid fractions to induce or to inhibit platelet aggregation was examined. Briefly, Tris buffer 50 mM (pH 7.4), human serum acetylhydrolase and the examined sample in BSA at 2.5 mg/mL saline were added to a prewarmed (37 -C) test tube. The enzymatic system was incubated at 37 -C and, at different time intervals, aliquots were taken to test their ability to induce or to inhibit washed rabbit platelet aggregation. 2.7. Mild alkaline hydrolysis and reacetylation An amount of the examined samples was subjected to mild alkaline hydrolysis and reacetylation while another

amount of these lipids was only subjected to acetylation. The biological activities of the lipids derived from the above chemical reactions were then tested on washed rabbit platelets. These procedures were carried out according to Demopoulos et al. (1979). Briefly, the sample, in a quantity as much as twice the quantity inducing platelet aggregation, was dissolved in 1 mL chloroform/methanol (1 : 4 v / v) and then 0.1 mL of 1.2 N NaOH in methanol/water (1 : 1 v / v) was added and allowed to stand for 20 min at 60 -C. The mixture was neutralized with 0.2 mL of 1 N acetic acid and 2 mL of chloroform/methanol (9 : 1 v / v). Afterwards, 1 mL of methanol and 2 mL water were added; two phases resulted which were then separated. The chloroform phase was washed with 1 mL of methanol/water (1 : 2 v / v), while water phase was washed with 1 mL chloroform. Each one of the two phases was separated in two equal fractions. Onehalf was tested for its bioactivity against platelets and the other half was evaporated to dryness under a stream of nitrogen and subjected to reacetylation by the addition of 1 mL acetic anhydrite and incubation at 60 -C for 45 min. After that, the reaction mixture was evaporated and extracted according to the Bligh and Dyer (1959) method and tested for its ability to induce washed rabbit platelet aggregation. Acetylation of the initial compound was also

Fig. 4. Positive (A) and negative (B) ESI-MS of fraction G6 from Chroococcidiopsis sp. glycolipids. The m / z ranges from 300 to 600.

S. Antonopoulou et al. / Comparative Biochemistry and Physiology, Part B 142 (2005) 269 – 282

performed using the same procedure as the one of the above reacetylation.

3. Results and discussion The genus Chroococcidiopsis grows in caves and Chroococcidiopsis kashaii species from cave isolated material (Friedman, 1961, 1962) were confirmed as new. This genus is also reported to have been isolated from the Greek cave of Nympholiptou at Vari of Attica (Pantazidou, 1996) as well as from stalactites of formed inshore cave-like constructions at the area of Aedipsos thermal springs (Anagnostidis and Pantazidou, 1988). The amount of total lipids extracted was about 70 mg. PL fractions eluted from 6 to 35 min under the HPLC conditions described in Section 2 were collected and tested for their biological activity toward washed rabbit platelets. No biological activity was detected in any fraction (data not shown). The fraction of PL eluted from 0 to approximately 6 min under the above HPLC conditions, along with the fraction of neutral lipids from the TLC, were combined and subjected to HPLC purification on a normal phase column,

275

Sphereclone 5u NH2, using the solvent system described in the Materials and methods section. A typical profile of the separation of the above fraction along with the retention times of standard lipids is shown in Fig. 1. This HPLC method was suitable since on one hand it permits purification of glycolipids and phospholipids and on the other hand neutral lipids are separately eluted until the first 7 min. Lipids eluted from the HPLC system, represented by fractions G1 to G15 (Fig. 1), were manually collected and each one was rechromatographed under the same HPLC conditions for its further purification from the adjacent peaks. Consequently, all the purified fractions were tested for their ability to induce washed rabbit platelet aggregation and/or to inhibit PAF-induced as well as thrombin-induced washed rabbit platelet aggregation. The G1 fraction, which contains neutral lipids, was kept for further purification. The biological activity of the purified fractions is represented in Table 1. Only fraction G12 induced rabbit platelet aggregation and thus it was selected for further investigation. Six fractions exerted inhibitory activity against PAF- and thrombin-induced aggregation. Fraction G6 was the most active one of this category. Three fractions were detected to inhibit only PAF-induced washed rabbit platelet aggregation

Fig. 5. Positive (A) and negative (B) ESI-MS of fraction G6 from Chroococcidiopsis sp. glycolipids. The m / z ranges from 600 to 900.

276

S. Antonopoulou et al. / Comparative Biochemistry and Physiology, Part B 142 (2005) 269 – 282

without affecting thrombin-induced aggregation. Fraction G13 was the most active of this category. Fractions G6 and G13 were selected for further studies. Fraction G6 co-migrated with standard lysophospatidylcholine on HPLC separation and was detected to inhibit PAF- and thrombin-induced platelet aggregation. Its IC50 value against PAF-induced aggregation was 15.5 AL (expressed as microliters of lipid fraction dissolved in 2.5 mg BSA /mL saline), while at this amount the inhibition of thrombin-induced aggregation was 100%. Structure elucidation of G6 was based on ES/MS data since the limited amount of G6 did not allow the performance of any chemical determination. ES/MS indicated the structure of the phosphoglycolipid shown in Fig. 2 with a relative molecular mass 1154 since the positive ions [M + H]+ and [M + H –CH3CO + H + 3Na – 3H]+ were present at m / z 1155 and 1179, respectively. Furthermore the presence of negative ions [HPO3] and [H2PO4] at m / z 80 and 97, respectively, as well as their correspondent positive ions [H3PO4 + H3]+ at m / z 81 demonstrated a phospholipid structure (Fig. 3A and B). The carboxylate anions from the fatty acyl groups 20 : 0 and

22 : 0 were observed at m / z 311 and 339, respectively, in the negative ion mass spectrum (Fig. 4B). The ESI mass spectrum gave also the ions at m / z 883[M + H – R2CO + H + Na – H]+, at m / z 867[883-CH3COOH – 2H + 2Na]+, at m / z 817[M – R2CO + H –CH3CO + H] , at m / z 815[M + H –R1COOH]+, at m / z 773[815-CH3CO + H]+, at m / z 685[M – 2H + Na – R2COO – sugar A – O] , at m / z 635[M + Na –R1CO + H – sugar A – CH3COO + H]+ (Fig. 5), at m / z 501[M – R1COOH – R2COOH – H] and at m / z 441[501-CH3COOH] (Fig. 4), all supporting the existence of the aforementioned fatty acids as well as of an acetyl group. The existence of the terminal sugar moiety was confirmed by the abundant positive ion at m / z 165 (Fig. 3) and moreover by the other abundant positive ions at m / z 999[M + 3Na – H – CH 3 COOH – sugar] + , m / z 955[M + H + Na – CH 3 COOH – sugar] + , m / z 937[M + Na – H – CH3COOH – sugar – O]+ (Fig. 2) and m / z 617[M + Na – R1CO + H – CH3COOH + H]+ (Fig. 5A) as well as by the negative ion at m / z 953[M + H + Na – CH3COOH – sugar] (Fig. 2B). It is worthy to note that sodium salts of lipids isolated from natural sources are prevalent.

Fig. 6. Positive (A) and negative (B) ESI-MS of fraction G12 from Chroococcidiopsis sp. glycolipids. The m / z ranges from 0 to 300. The proposed structure along with identified fragmentations is also represented.

S. Antonopoulou et al. / Comparative Biochemistry and Physiology, Part B 142 (2005) 269 – 282

The two sugar moieties A and B were also determined by the predominant negative ion at m / z 325[A + B + O – CH3COOH + H] and the positive ion at m / z 365[A + B + O – CH3CO + H + H + Na]+ (Fig. 4). Fraction G12 eluted after DGDG on HPLC separation was detected to induce washed rabbit platelet aggregation in a dose-dependent manner. It seems that this active glycolipid acts through PAF receptor since (1) platelets desensitized to PAF were not aggregated to this lipid and vice versa and (2) the PAF receptor specific antagonist, BN 52021, inhibited the aggregation induced by this active lipid. Treatment with acetylhydrolase resulted in its timedependent total inactivation, with a lower hydrolysis rate than that of PAF, indicating the presence of acetyl group(s), not in a glyceryl backbone position. Structure elucidation of G12 was based on ES/MS data since the limited amount of G12 did not allow the performance of any chemical determination. The positive ions at m / z 81[H2PO3]+ and 99[H3PO4 + + H] along with the negative ions at m / z 80[HPO3] and 97[H2PO4] (Fig. 6) are characteristics of the phosphoric group and indicated the structure of a phospholipid. The presence of the positive ions at m / z 818[M + H]+ and

277

862[M + 2Na –H]+ along with the negative ion at m / z 816[M –H] (Fig. 7) indicated a relative molecular mass of 817 for this lipid molecule. Furthermore, the predominant positive ions at m / z 802[862-CH3COOH]+, m / z 758[818CH3COOH]+ as well as the negative ion at m / z 774[816CH3CO + H] (Fig. 7) indicated the presence of an acetyl group on the sugar moiety, dissociated as 60 units. The presence of this acetyl group is in agreement with the results of the acetylhydrolase treatment. It is worthy to note that the neutral loss of 60 units is common for acetylated glycolipids. The existence of the carboxylate anion from the fatty acyl group 6 : 0 was confirmed by the positive ions at m / z 686[862-62-116] + , m / z 670[862-60-114-H 2 O] + , m / z 642[818-60-116]+ and m / z 626[818-60-114-H2O]+ (Fig. 7). The positive ions at m / z 305, 287, 243 naL 205 as well as the negative ions at m / z 319, 281 naL 227 (Figs. 6 and 8) were attributed to the sugar moiety after the dissociation of either the acetyl group or the fatty acid in the form of RCO – or RCOO – . Finally the phosphorylated sugar gave the predominant negative ions at m / z 339 and 305 and positive ions at m / z 385, 315, 337, and 413 (Fig. 8). The proposed structure of

Fig. 7. Positive (A) and negative (B) ESI-MS of fraction G12 from Chroococcidiopsis sp. glycolipids. The m / z ranges from 600 to 900.

278

S. Antonopoulou et al. / Comparative Biochemistry and Physiology, Part B 142 (2005) 269 – 282

G12 is presented in Fig. 6 along with its fragmentation pathways that lead to the identified fragments. G13, eluted between DGDG and PE on HPLC separation, was chosen for its structure elucidation since it inhibited only the PAF-induced washed rabbit aggregation without affecting thrombin-induced aggregation. Its IC50 value against PAF-induced aggregation was 5.0 AL (expressed as microliters of lipid fraction dissolved in 2.5 mg BSA /mL saline). Acetylation of the fraction resulted in a molecule with a different biological activity; more specifically, it induced washed rabbit platelet aggregation. This result indicated that at least one free hydroxyl group is present. After alkaline hydrolysis, the chloroform soluble fraction maintained in full the biological activity of the initial fraction

(G13). Moreover, acetylation of the chloroform soluble fraction from the hydrolysis resulted in a molecule that induced platelet aggregation in an identical way compared to the one derived from the acetylation of the initial fraction. These data suggest that no ester bond is present on G13. The relatively wide elution area of G13 along with the data from ES/MS revealed the existence of at least three molecular species that differ in carbon chain size. The first analog with relative molecular mass 743 was confirmed by the positive ions at m / z 744[M + H]+ and 788[M + 2Na –H]+ along with the negative ion at m / z 742[M – H] (Fig. 9). In addition the positive ions at m / z 708[744 - 2H2O] +, 730[708 + Na – H]+ and 774[708 + 3Na – 3H]+ were also detected.

Fig. 8. Positive (A) and negative (B) ESI-MS of fraction G12 from Chroococcidiopsis sp. glycolipids. The m / z ranges from 300 to 430.

S. Antonopoulou et al. / Comparative Biochemistry and Physiology, Part B 142 (2005) 269 – 282

279

Fig. 9. Positive (A) and negative (B) ESI-MS of fraction G13 from Chroococcidiopsis sp. glycolipids. The m / z ranges from 600 to 900. The proposed structure along with identified fragmentations is also represented.

The second analog with relative molecular mass 771 was detected by the negative ion at m / z 770[M –H] while the third analog with relative molecular mass 757 was determined by the positive ions at m / z 758[M + H]+ and 802[M + 2Na – H]+ (Fig. 9). The ES/MS data for G13, which are in agreement with the data from mild alkaline hydrolysis and acetylation treatment, confirmed the structure of a phospho-glyco analog of ceramide with a short ether-bonded carbon chain moiety on the sugar moiety. The three aforementioned analogs differ in length of this short carbon chain moiety. The proposed structure of G13 is presented in Fig. 11 along with its identified fragments. The existence of the sugar moiety for the first analog was determined by the positive ions at m / z 229, 295[229 3Na – 3H] + , 193[229 – 2H 2 O] + and 145[229 – CH 3 (CH_ CH)2OH]+ as well as by the negative ions at m / z 267 (Fig. 10) and at m / z 325 (Fig. 11). The positive ions at m / z 259[257 + 2H]+ and 279[257 + Na – H]+ and the negative one at m / z 353 (Fig. 11) confirmed the existence of the sugar moiety for the second analog. The sugar moiety for the third analog was determined by the positive ion at m / z 243 (Fig. 10) and by the negative ions at m / z 281 (Fig. 10) and 339

(Fig. 11). Fragmentation pathways resulting to the above fragments are presented in Fig. 11. The presence of phosphoric group was confirmed by the negative ions at m / z 80[HPO3] and 97[H2PO4] as well as by their corresponding positive ions at m / z 81[H2PO3]+ naL 99[H3PO4 + H]+ (Fig. 10). The proposed structure of the phospho-ceramide was also based on the presence of the positive ions at m / z 538[phospho-ceramide + H + Na]+, 582[538 + 2Na – 2H]+, 522[phosphono-ceramide + H + Na] + , 504[522 – H 2 O] + , 480[ceramide + 2Na – H]+ and 502[480 + Na –H]+ (Fig. 11). The corresponding fragmentation pathways are presented in Fig. 11. The predominant positive fragment at m / z 320 was attributed to the fragment Na 2 PO 4 CH 2 CH_+ NHCO (CH2)3(CH_CH)2CH3. This fragment is common at the positive ES/MS spectrums (Olsson et al., 1999). Chroococcidiopsissp. glycolipids exert mainly inhibitory action against washed rabbit platelet aggregation. The most active fractions were G6 and G13. The structure of G6 as confirmed by ES/MS is referred to phospho-acetylated glyco analog of diglyceride. It is well established (Avramopoulou et al., 1997) that phospho-acetylated glyco-glycero lipids

280

S. Antonopoulou et al. / Comparative Biochemistry and Physiology, Part B 142 (2005) 269 – 282

Fig. 10. Positive (A) and negative (B) ESI-MS of fraction G13 from Chroococcidiopsis sp. glycolipids. The m / z ranges from 0 to 300.

exhibit a double action, that is, in low concentrations they act as inhibitors of washed rabbit platelets aggregation while in high concentrations they induce washed rabbit platelet aggregation. In addition, the presence of an acetylated phosphoglycoglycerolipid that exhibits a biological activity very similar to that of PAF has been well established in S. julianum (Antonopoulou et al., 2002). It is worth mentioning that the acetyl group in both previous studies existed also at the sn-2 position of the glycerol backbone. The results from the analysis of glycolipids of Chroococcidiopsissp. show that the acetyl group exists only on the sugar moiety, which leadsto a molecule with inhibitory activity. The structure of G13 that is based on the results from mild alkaline hydrolysis and reacetylation as well as on ES/ MS data is referred to a phospho-glyco analog of ceramide. This fraction seems to be a non-acetylated analog of G12 since its acetylation resulted in a molecule that induced platelet aggregation. Finally G12 was the only fraction that caused platelet aggregation. The results from the acetylhydrolase treatment along with the ES/MS data confirm a structure of the phospho-acetylated glyco analog of ceramide. It is important to notice that the ceramide backbone seems to be particularly prevalent at the cyanobacteria

lipids. Concerning the biological activity of these lipid analogs, it seems that the acetylated phospho-glyco analogs of ceramides (Antonopoulou et al., 2005) as well as the acetylated sphingomyelin (Antonopoulou et al., 2002) induce platelet aggregation, while the non-acetylated analogs and more specific phospho-glyco ether analogs of ceramides (like G13) are specific inhibitors of PAF activity. In conclusion, we demonstrated that the polar lipid fraction of Chroococcidiopsissp. contains three potent biologically active molecules concerning their ability to induce washed rabbit platelet aggregation or to antagonize the actions of PAF. All three molecules were phosphoglycolipids, two of them having a sphingosine backbone, while the other one had a phosphoglycerol backbone. The existence of acetyl groups esterified to the sugar moiety is a common structural feature of these molecules. The comparison between fractions G12 and G13, which have similar structures, leads to the conclusion that acetylation of sugar moiety on ceramide analogs can change the bioactivity from inhibition to aggregation of platelets, something which may be of physiological importance. The existence of the sphingosine backbone is in accordance with previous results demonstrating the existence of acetyl-

S. Antonopoulou et al. / Comparative Biochemistry and Physiology, Part B 142 (2005) 269 – 282

281

Fig. 11. Positive (A) and negative (B) ESI-MS of fraction G13 from Chroococcidiopsis sp. glycolipids. The m / z ranges from 300 to 600.

sphingomyelin in the phospholipid fraction of S. julianum (Antonopoulou et al., 2002). The presence of a phosphoglycerol derivative (fraction G6) along with the aforementioned data demonstrates the existence of novel bioactive compounds in cyanobacteria. The identified PAF-like activity of some of the isolated glycolipids may contribute to the allergic skin irritation and allergic asthma appearing after cyanobacterial exposure. References Anagnostidis, K., Pantazidou, A., 1988. Endolithic cyanophytes from the saline thermal springs of Aedipsos, Hellas (Greece). Arch. Hydrobiol. Algolog. Stud. 50 – 53 (suppl. 80), 555 – 559. Antonopoulou, S., Demopoulos, C.A., Andrikopoulos, N.K., 1996. Lipid separation from Urtica dioica: existence of platelet-activating factor (PAF). J. Agric. Food Chem. 44, 3052 – 3056. Antonopoulou, S., Oikonomou, A., Karantonis, C., Fragopoulou, E., Pantazidou, A., 2002. Isolation and structure elucidation of biologically active phospholipids from Scytonema julianum (Cyanobacteria). Biochem. J. 367, 287 – 293. Antonopoulou, S., Nomikos, T., Oikonomou, A., Kyriacou, A., Andriotis, M., Fragopoulou, E., Pantazidou, A., 2005. Characterization of

bioactive glycolipids from Scytonema julianum (Cyanobacteria). Comp. Biochem. Physiol. B 140, 219 – 231. Avramopoulou, V., Antonopoulou, S., Frousios, K., Argyropoulos, D., Demopoulos, C.A., 1997. Synthesis of a new phosphoglycolipid with biological activity towards platelets. Int. J. Biochem. Cell Biol. 29, 767 – 774. Bligh, E.G., Dyer, W.J., 1959. A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37, 911 – 917. Braquet, P., Touqui, L., Shen, T.Y., Vargaftig, B.B., 1987. Pharmacol. Rev. 39, 97 – 145. Demopoulos, C.A., Antonopoulou, S., 1996. A discovery trip to compounds with PAF-like activity. Adv. Exp. Med. Biol. 416, 59 – 63. Demopoulos, C.A., Pinckard, R.N., Hanahan, D.J., 1979. Platelet-activating factor. Evidence for 1-O-alkyl-2-acetyl-sn-glyceryl-3-phosphoryl-cholineas the active component. (A new class of lipid chemical mediators). J. Biol. Chem. 254, 9355 – 9358. Falconer, I.R., 1996. Potential impact on human health of toxic cyanobacteria. Phycologia. 35, 6 – 11. Friedman, E.I., 1961. Chroococcidiopsis kashaii sp. nov. and the genus ¨ sterr. Bot. Chroococcidiopsis (studies on cave algae from Israel III). O Z. 108, 354 – 367. Friedman, E.I., 1962. The ecology of theatomphytic nitrate algae Chroococcidiopsis kashaii Friedman. Studies on cave algae from Israel: IV. Archiv. Mikrobiol. 42, 42 – 45. Henig, N.R., Aitken, M.L., Liu, M.C., Yu, A.S., Henderson Jr., W.R., 2000. Effect of recombinant human platelet-activating factor-acetylhydrolase

282

S. Antonopoulou et al. / Comparative Biochemistry and Physiology, Part B 142 (2005) 269 – 282

on allergen-induced asthmatic responses. Am. J. Respir. Crit. Care Med. 162, 523 – 527. Jenks, P.J., Kavanagh, G.M., Brooks, J., Bradfield, J.W., Archer, C.B., 1999. Comparison of weal and flare responses to platelet-activating factor (PAF) and histamine and the ultra structural effects of PAF in the skin of atopic and normal subjects. Clin. Exp. Dermatol. 24, 112 – 117. Kates, M., 1992. Archaebacterial lipids: structure, biosynthesis and function. Biochem. Soc. Symp. 58, 51 – 72. Kim, Y.H., Choi, J.-S., Yoo, J.S., Park, Y.-M., Kim, M.S., 1999. Structural identification of glycerolipid molecular species isolated from cyanobacterium Synechocystis sp. PCC 6803 using fast atom bombardment tandem mass spectrometry. Anal. Biochem. 267, 260 – 270. Koltai, M., Hosford, D., Guinot, P., Esanu, A., Braquet, P., 1991. Plateletactivating factor (PAF): a review of its effects, antagonists and possible future clinical implications. Drugs (Part I) 42, 9 – 29. Komarek, J., Anagnostidis, K., 1986. Modern approach to the classification system of cyanophytes: 2—Chroococcales. Algological Studies 43, 157 – 226 (Arch. Hydrobiol. Suppl. 72, 3). Kulikov, V.I., Muzya, G.I., 1996. Biochemistry (Moscow) 61, 387 – 403. Kulikov, V.I., Muzya, G.I., 1997a. Ether lipids and platelet-activating factor: evolution and cellular function. Biochemistry (Moscow) 62, 1103 – 1108. Kulikov, V.I., Muzya, G.I., 1997b. Ether lipids and platelet-activating factor: evolution and cellular function. Biochemistry (Moscow) 62, 1103 – 1108. Lazanas, M., Demopoulos, C.A., Tournis, S., Koussissis, S., LabrakisLazanas, K., Tsarouhas, X., 1988. PAF of biological fluids in disease: IV. Levels in blood in allergic reactions induced by drugs. Arch. Dermatol. Res. 280, 124 – 126. Murata, N., Nishida, I., 1987. Lipids of blue-green algae (Cyanobacteria). In: Stumpf, P.K. (Ed.), The Biochemistry of Plants, Lipids: Structure and Function, vol. 4. Academic Press, New York, pp. 315 – 347. Murata, N., Wada, H., 1995. Acyl-lipid desaturases and their importance in the tolerance and acclimatization to cold cyanobacteria. Biochem. J. 308, 1 – 8.

Olsson, B.M., Karlsson, H., Larsson, T., Lanne, B., 1999. Mass spectrometric analysis of ceramide composition in mono-, di-, tri-, and tetraglycosylceramides from mouse kidney: an experimental model for uropathogenic Escherichia coli. J. Mass Spectrom. 34, 942 – 951. Omata, T., Murata, N., 1983. Isolation and characterization of cytoplasmic membranes from the blue green algae (cyanobacterium) Anacystis nidulans. Plant. Cell Physiol. 24, 1101 – 1112. Pantazidou, A., 1996. Cyanophytes (Cyanobacteria) in lighted parts of various Greek Caves—problems associated with their growth. Algological Studies 83, 455 – 456. Patterson, G.M.L., Larsen, L.K., Moore, R.E., 1994. Bioactive natural products from blue-green algae. J. Appl. Phycol. 6, 151 – 157. Reshef, V., Mizrachi, E., Maretzki, T., Silberstein, C., Loya, S., Hizi, A., Carmeli, S., 1997. New acylated sulfoglycolipids and digalactolipids and related known glycolipids from cyanobacteria with a potential to inhibit the reverse transcriptase of HIV-1. J. Nat. Prod. 60, 1251 – 1260. Rˇezanka, T., Viden, I., Go, J.V., Dor, I., Dembitsky, V.M., 2003. Polar lipids and fatty acids of three wild cyanobacterial strains of genus Chroococcidiopsis. Folia Microbiol. 48 (6), 781 – 786. Rippka, R., Deruells, J., Waterbury, J.B., Herdman, M., Stanier, R.Y., 1979. Generic assignments, strain histories and properties of pure cultures of cyanobacteria. J. Gen. Microbiol. 111, 1 – 61. Stafforini, D.M., McIntyne, T.M., Carter, M.E., Prescott, S.M., 1987. Human plasma platelet-activating factor acetylhydrolase. J. Biol. Chem. 262, 4223 – 4230. Torokne, A., Palovics, A., Bankine, M., 2001. Allergenic (sensitisation, skin and eye irritation) effects of freshwater cyanobacteria—experimental evidence. Environ. Toxicol. 16, 512 – 516. Zepke, H.D., Heinz, E., Radunz, A., Linscheid, M., Pesch, R., 1978. Combination and positional distribution of fatty acids in lipids from blue-green algae. Arch. Microbiol. 119, 157 – 162.