Biochimica et Biophysica Acta 1259 (1995) 203-210
f ic~a A~ta etBiochi Biophysica
Thrombocytes are the predominant source of endogenous sulfidopeptide leukotrienes in the American bullfrog( Rana catesbeiana) Karsten Gronert *, Selene M. Virk, Ceil A. Herman Department of Biology, New Mexico State Unieersity, Box 30001/Dept. 3AF, Las Cruces, NM 88003, USA Received 7 April 1995; accepted 18 July 1995
Nucleated bullfrog erythrocytes have 5-1ipoxygenase(LO) and are the first non-mammalian cell to exhibit endogenous sulfidopeptide leukotriene (LT) synthesis. Non-nucleated mammalian platelets lack 5-LO, but contribute significantly to LTC 4 production by transcellular synthesis. However, nucleated bullfrog thrombocytes have not been examined for 5-LO activity. Endogenous leukotriene synthesis by bullfrog thrombocytes and mixed leukocytes was analyzed by reverse-phase high-performance liquid chromatography (RP-HPLC). Calcium ionophore activated (A23187) leukocytes demonstrated 5-LO, 12-LO, and 15-LO activity. Spectral analysis demonstrated synthesis of LTB4, LTB4 isomers, 15(S)-monohydroxyicosatetraenoic acid (HETE), 5(S),12(S)-diHETE, 5(S),15(S)-diHETE, lipoxin A 4 ( L X A 4) and L X B 4. Thrombocytes synthesized large quantities of sulfidopeptide leukotrienes but no lipoxins. Sulfidopeptide leukotriene and LTB4 radioimmunoassay analysis and the radiological RP-HPLC profile of [3H]AA metabolism further confirmed synthesis. Incubations with [3H]LTC4 demonstrated slow and incomplete conversion to [3H]LTD4. Thrombocyte leukotriene profile changed over time revealing a significant shift from the LTC 4 synthase to LTA 4 hydrolase pathway, corresponding with release of large amounts of LTA 4. Thrombocytes potentially play a pivotal role in inflammatory and cardiovascular responses. 5-LO activity in amphibian homologs to mammalian platelets and erythrocytes compared with the lack of activity in the mammalian counterparts may correspond to the loss of the nucleus in the evolution of these cells. Keywords: Lipoxygenase; Thrombocyte; Leukotriene C4; Lipoxin; Amphibian; (R. catesbeiana)
The importance of leukotrienes as chemotactic/immunoregulatory a g e n t s and p o t e n t b r o n c h o constrictors/modulators of vascular tone has been well established [1-5]. 5-1ipoxygenase (EC 22.214.171.124, 5-LO) and endogenous leukotriene synthesis is restricted to nucleated myeloid cells. Non-nucleated erythrocytes and platelets can synthesize leukotrienes transcellularly in the presence of activated inflammatory cells which release the unstable epoxide LTA 4 (5(S),6(S)-oxido-7,9,11,15-icosatetraenoic acid). By this pathway, erythrocytes synthesize (5(S), 12(R)-dihydroxy-6,8,10,14-icosatetraenoic acid (LTB 4)  and platelets (5(S)-hydroxy-6(R)-glutathionyl7,9,11,14-icosatetraenoic acid (leukotriene (LT) C 4) ,
* Corresponding author. Present address: Harvard Medical SchoolBrigham and Women's Hospital, Department of Anesthesia Research, Center for Experimental Therapeutics and Reperfusion Injury, 75 Francis Street, Boston, MA 02115, USA. Fax: + 1 (617) 2786957. 0005-2760/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0 0 0 5 - 2 7 6 0 ( 9 5 ) 0 0 1 6 0 - 3
respectively. This is especially significant, since platelets are a major source for LTC 4 in the mammalian circulatory system, increasing LTC 4 synthesis approx. 3-fold over endogenous inflammatory cell production [8,9]. In combination with the synthesis of thromboxane  and lipoxins , they modulate vascular tone and inflammatory events. Although studies in non-mammalian vertebrates are less common, lipoxygenase activity and physiological effects have been demonstrated in amphibians [12-20]. In fish blood cells, lipoxygenase activity has been demonstrated [21-26]. However, endogenous sulfidopeptide leukotriene synthesis has only been demonstrated in bullfrog blood. In this system, erythrocytes were the only amphibian cell type characterized for lipoxygenase activity . In erythrocytes, 12-LO activity predominated over 5-LO arachidonic acid metabolism, thus erythrocytes account for only a small percentage of the observed LTD4 synthesis in bullfrog whole blood . It remains to be determined which other amphibian cell types exhibit LTC 4 synthase activity. Endogenous leukotriene synthesis in bullfrog nucleated
K. Gronert et al. / Biochimica et Biophysica Acta 1259 (1995) 203-210
erythrocytes may be explained by the localization of 5lipoxygenase activating protein (FLAP) and 5-LO to the nuclear envelope in mammalian leukocytes [28,29]. In mammalian leukocytes, 5-LO seems to be translocated upon cell activation from either a cytosolic or nuclear pool to the nuclear envelope to initiate leukotriene synthesis . Therefore other nucleated counterparts to nonnucleated mammalian myeloid cells may also exhibit 5-LO activity. Unlike mammals, the hemopoietic tissue in non-mammalian vertebrates is devoid of megakarocytes [30,31]. In 1892, the term thrombocytes was introduced by Dekhuyzen for the spindle shaped cells of amphibian blood that were the cellular constituents of intravascular thrombi . Nucleated thrombocytes appear to be the functional equivalent of mammalian platelets based on ultrastructural similarities (canalicular system) and their ability to promote coagulation in a number of amphibian genera (Xenopus, Bufo, Rana and Batrachoceps). The precise role of thrombocytes in coagulation and inflammatory events remains to be elucidated . The objective of this study was to examine 5-LO activity in thrombocytes and compare it with the lipoxygenase profile of mixed ieukocytes.
2. Materials and methods
2.1. Materials Leukotriene (LT) C 4, LTD4, LTB 4, LTE 4, and LTB4/LTC 4 antisera were generous gifts from Merck Frosst, Centre for Therapeutic Research, Pointe Claire, Duval, Quebec, Canada. Arachidonic acid (AA), LTB 4 isomers (6-trans LTB 4, 6-trans,12-epi LTB4), 5(S)-hydroxy icosatetraenoic acid (HETE), 12(S)-HETE, 15(S)HETE, 5(S), 12(S)-diHETE, 5 S, 15S-diHETE, lipoxin (LX) A 4 and L X B 4 w e r e purchased from Cayman, Ann Arbor, MI. [5,6,8,9,11,12,14,15-3H]AA (100 Ci/mmol) and [15,19,20-3H(N)]LTC4 (173 Ci/mmol) were purchased from Du Pont-New England Nuclear, Boston, MA. Highperformance liquid chromatography (HPLC) solvents, Scintiverse E were purchased from Fisher, Pittsburgh, PA. All other chemicals were purchased from Sigma, St. Louis, MO.
2.2. Animals Bullfrogs of both sexes were either collected locally (Las Cruces, NM) or purchased from Charles D. Sullivan, Nashville, TN. Animals were housed in holding tanks with running water and fed beef liver 3-times per week.
2.3. Thrombocyte isolation and preparation Blood was collected from Rana catesbeiana (420 _ 14 g, n = 55) by cardiac puncture and immediately diluted 1:7 with cold heparinized (800 U/ml), Ca 2+ and Mg 2+
free amphibian ringers [ A R ( - ) ] containing NaC1, 118 mM; KCI, 3.5 mM; NaHCO 3, 29.7 mM; glucose 5.5 mM; N%HPO 4, 4.0 mM; NaH2PO 4, 3.8 mM (pH 7.4). Following dilution, blood was placed on ice for 20 min. This solution (20 ml) was carefully loaded onto a 9 ml metrizoate (23.7%)/Ficoll (2.2%) gradient in 50 ml polycarbonate tubes (28.7 X 103 mm) and centrifuged at 250 × g for 27 min (10°C). The thin band immediately below the interphase was carefully collected and diluted with 30 ml cold heparinized [ A R ( - ) ] and pelleted at 163 X g. The resulting thrombocyte pellet was resuspended to desired cell concentrations in appropriate buffers. Thrombocyte purity (89 + 1%, n = 25) was determined by using spindle shape as a morphological characteristic and by Wright stain characteristics according to Humason  with lymphocytes as the major contaminant. All other cell types were present at concentration below 1%. Cell viability was > 99% as determined by Trypan blue exclusion. Cell concentrations were determined by hemocytometer.
2.4. Mixed leukocyte isolation and preparation Blood was collected by cardiac puncture as described above and immediately diluted with heparinized A R ( - ) at a ratio of 1:5. This solution (20 ml) was carefully loaded onto a 9 ml metrizoate (23.2%)/Ficoll (2.4%) gradient in 50 ml polycarbonate tubes (28.7 × 103 mm) and centrifuged at 358 X g for 25 min (18°C). The broad buffy band above the erythrocyte pellet was collected and diluted with 30 ml heparinized A R ( - ) and pelleted at 163 X g. The resulting mixed leukocyte pellet was resuspended to desired cell concentrations in appropriate buffers. Cell composition was determined by morphological and staining characteristics with both Wright and Wright Giemsa according to Humason . Thrombocyte concentration in the neutrophil, eosinophil and monocyte cell population was (38 + 3%). Contamination by erythrocytes was less than 1%. Cell viability was > 99% as determined by Trypan blue exclusion. Cell concentrations were determined by hemocytometer.
2.5. Endogenous AA metabolism The lipoxygenase metabolite profile from purified thrombocyte and mixed leukocytes was compared. Purified thrombocytes or mixed leukocytes were resuspended in A R ( - ) containing i.2 mM Ca 2÷ and 0.5 mM Mg 2÷ (AR + ) at a concentration of 3.75-106-5.00 • 1 0 6 cells/ml. Following a 30 min equilibration, this solution (1 ml or 3 ml) was incubated with 5 / z M calcium ionophore A23187 for 10 min. In some experiments cells were pre-incubated for 10 min with 10 tzM nordihydroguaiaretic acid (NDGA) before activation by A23187. Incubation was terminated by placing tubes on ice and centrifugation at 636 X g (4°C). Supernatants were collected for solid-phase extraction with Sep-Pak C18 cartridges and
K. Gronert et aL / Biochimica et Biophysica Acta 1259 (1995) 203-210
analyzed by RP-HPLC as previously published [35,36], a n d / o r radioimmunoassays (RIA) .
2.6. [5,6,8, 9,11,12,14,15- 3H]AA experiments Purified thrombocytes were resuspended in AR( + ) at a concentration of 3.75. 1 0 6 - 5 . 0 0 • 106 cells/ml and allowed to equilibrate for 30 min. This solution (1 ml) was incubated with 1.5 /xCi [I-3H]AA (15 nM) and 5 /zM A23187 for 10 min under gentle gyration. In some experiments, cells were pre-incubated for 10 min with 10 /zM NDGA f o l l o w e d by A23187 activation and [5,6,8,9,1 l, 12,15- 3H]AA incubation. Incubations were terminated and supernatants prepared for RP-HPLC. Pellets were digested with 500/zl 1 N NaOH for 30 min, adjusted to pH 7, and [5,6,8,9, I 1,12,15- 3H]AA incorporation by the cells was determined by scintillation counting.
2.7. [14,15,19,20- 3H(N)]LTC4 experiments Purified thrombocytes were resuspended in AR(+ ) at a concentration of 3.75. 106-5.00 • l06 cells/ml. In some experiments, thrombocytes were resuspended in A R ( + ) containing 45 mM serine-borate. The solution (500 ~l) was incubated with 5 /.LM A23187 and 0.02 /.tCi of [l 4,15,19,20-3 H(N)]LTC 4 (0.54 ng) under gentle gyration. At 1, 2.5, 5, 10, 15 min cells were immediately pelleted (636 X g) and supematants collected and prepared for RP-HPLC analysis.
2.8. RP-HPLC and RIA analysis Samples in MeOH were applied to a Waters Nova-Pak CI8 (60 A, 4 /zM,3.9 × 150 mm) HPLC column. Leukotrienes were separated by mobile phase 1 (MPI) containing m e t h a n o l / w a t e r / a c e t o n i t r i l e / a c e t i c acid (23.5:49:27.5:1) (pH 5.6) or MP2 containing methanol/water/acetic acid (68:32:0.1) (pH 5.6). For a complete profile of lipoxygenase products, a step gradient mobile phase (MP3) was developed which changed from m e t h a n o l / w a t e r / a c e t o n i t r i l e / a c e t i c acid (23.5:49:27.5:0.1) (step i) to methanol/water/acetic acid (68:32:0.1) (step 2) to 100% methanol (step 3) at specific intervals according to retention times of authentic standards. All HPLC runs were monitored with a diode array detector (Hewlett-Packard 1040 M Series II, wavelength step, 2 nm), at 280 nm (LTC4, LTD4, LTB 4, LTE4, LTB 4 isomers and 5(S), 12(S)-diHETE), 235 nm (5(S), 15(S)-diHETE, 5(S)-HETE, 12(S)-HETE and 15(S)-HETE), 301 nm (LXA 4 and LXB 4) and 205 nm (arachidonate). The column was extensively washed with MeOH between samples. Fractions (1 min) were collected and analyzed directly by RIA specific for sulfidopeptide leukotrienes or LTB 4 as previously published . LTB 4 antisera demonstrated 100% cross-reactivity with LTB 4 and LTB 5, 25% with 6-trans LTB4, and less than 1% with 6-trans,12-epi
LTB 4, LTC 4, LTD4, LTE 4 or AA. LTC 4 antisera demonstrated 100% cross-reactivity with LTC 4, 80% with LTD 4, 27% with LTE 4 and no cross-reactivity with LTB 4, LTB 4 isomers or AA . No significant cross reactivity was observed with solvents in any HPLC fractions. RIA standard curves were run at the same solvent concentrations as HPLC fractions. Radioactive profiles were determined by counting 1 minute fractions in a Packard Tricarb 4530 scintillation counter or by a /3-Ram flow-through monitor (IN/US Systems, Fairfield, NJ). Specific radioactive peaks and spectra were verified by comparison with authentic standards (corrected for 1.5 min dwell time between detectors). In all incubations the concentration of carrier solvents (EtOH or DMSO) never exceeded one percent. All data are expressed as mean values + S.E.M. Percent LTC 4 to LTD 4 conversion was examined for a completely random design. Effects of time were separated by the least significant difference method. Difference were considered significant at the P < 0.05 level.
3.1. Lipoxygenase profile in mixed leukocytes RP-HPLC separation using a step gradient mobile phase (MP3) and spectral analysis of metabolites from A23187 activated mixed leukocytes (n = 3) revealed four distinct conjugated triene spectra at 280 nm (Fig. IA, Table 1) that matched the retention times of authentic standards: 6-trans 7°i
~zo B o
t5 f ~
Fig. I. Lipoxygenase activity in bullfrog mixed leukocytes. Leukocytes((l.l-l.5). l07 cells) were activated with 5 ~ M A23187, metabolites separated on a RP-HPLC step gradient mobile phase (MP3) and analyzed with a diode array detector. Indicated peaks correspond to the retention time of standards. Data shown (A and B) are representative of the first step in the gradient (step l, flow rate (FR) 0.25 ml/min). 280 nm (A): 6-trans LTB 4 (I), 6-trans, 12-epi LTB 4 (II), LTB 4 (III) and 5(S),12(S)-diHETE (IV); 301 nm (B): LXB 4 (V) and LXA 4 (VI); 235 tim (B): 5(S),15(S)-diHETE (VII). Monohydroxyicosatetraenoic acid compounds eluted in the second step in the gradient (step 2, FR 0.7 m l / m i n ) which was monitored at 235 nm (C): 15(S)-HETE (VIII).
K. Gronert et al. / Biochimica et Biophysica Acta 1259 (1995) 203-210
LTB 4 (I, 25.4 min), 6-trans,12-epi LTB 4 (II, 26.6 min), LTB 4 (III, 32.5 min) and 5(S),12(S) dihydroxy-6,8,10,14icosatetraenoic acid [5(S),12(S)-diHETE] (IV, 35.9 min). Peaks I - I V displayed a typical dihydroxy conjugated triene spectrum with an absorbance maximum at 270 nm and shoulders at 260 nm and 280 nm. No spectra or peaks corresponding to sulfidopeptide leukotrienes were detected. Monitoring at 301 nm (Fig. 1B) revealed four distinct peaks with a conjugated tetraene spectrum. These spectra had a maximum absorbance at 301 nm and shoulders at 289 and 317 nm characteristic for lipoxins (Table 1). Peaks V and VI matched the retention of the authentic standard 5(S), 14(R), 15(S)-trihydroxyicosatetraenoic acid (lipoxin B 4 (LXB4), 9.4 min) and 5 ( S ) , 6 ( R ) , 1 5 ( S ) - t r i h y d r o x y icosatetraenoic acid (lipoxin A 4(LXA 4 ), 13.4 min), respectively, whereas the other peaks (11.1 min, 14.1 min) did not correspond to the retention time of any standards. A peak (VII) overlapping with the retention time of 6-trans,12-epi LTB 4 was detected when the HPLC run was monitored at 235 nm (Fig. I B). This peak had a maximum absorbance at 243 nm (Table 1) and matched the retention time (27.6 min) of 5(S), 15(S)-dihydroxy-6,8,11,13-icosatetraenoic acid [5(S),15(S)-diHETE]. Monitoring the RP-HPLC run for monohydroxy icosatetraenoic compounds at 235 nm revealed one distinct peak (Fig. 1C) that matched the retention time of 15(S)-HETE (51.6 min). This peak (VIII) showed a typical monohydroxy conjugated diene spectrum with an absorbance maximum at 235 nm (Table 1). No peaks corresponding to the retention times of 12(S)-HETE (54.3 min) or 5(S)-HETE (58.6 min) were detected. No peaks or spectra corresponding to icosatetraenoic acid metabolites were detected in leukocytes not activated with A23187 (n = 3).
3.2. Endogenous leukotriene synthesis by purified thrombocytes RP-HPLC separation (MP1) and spectral analysis of dihydroxy icosatetraenoic acid compounds from A23187 activated thrombocytes revealed a lipoxygenase profile (n = 5) distinctly different from activated mixed leuko-
Table 1 RP-HPLC spectrometric data for endogenous metabolites Figure Compound Maximum Hypsochromic Bathochromic absorbance shoulder shoulder (nm) (nm) (nm) I 1 I 1 2 2 2 2
I, II, III, IV V, VI VII VIII I, II III, IV V, VI VII VIII, IX
270 301 243 235 280 270 243 235
260 289 270 260 -
280 317 292 280 -
30 ~ -<
IIl IV i
~ z0 /
20 30 Time (min)
Fig. 2. Lipoxygenase activity in bullfrog thrombocytes. Thrombocytes ((3.8-5.0). 106 cells) were activated with 5 #M A23187, metabolites separated by MPI (FR 0.3 ml/min). Indicated peaks correspond to the retention times of standards. 280 nm (A): LTC4 (I), LTD4 (II), 6-trans LTB4 (III), 6-trans, 12-epi LTB4 (IV), LTB4 (V) and 5(S),I2(S)-diHETE (VI); 235 nm (B): 5(S),I5(S)-diHETE (VII). Monohydroxyicosatetraenoic acid compounds were separated by MP2 (FR 0.7 ml/min). 235 nm (C): 15(S)-HETE (VIII) and 5(S)-HETE (IX).
cytes. Spectral analysis at 280 nm revealed 6 distinct peaks (Fig. 2A) that matched the retention times of authentic standards: LTC 4 (I, 14.6 min), LTD 4 (II, 22.7 min), 6-trans LTB 4 (III, 23.8 min), 6-trans,12-epi LTB 4 (IV, 25.4 min), LTB 4 (V, 30.2 min) and 5(S),I2(S)-diHETE (VI, 34.5 min). Peaks I I I - V I revealed a typical dihydroxy conjugated triene spectrum identical to the observed absorbance profile in peaks l - I V in activated leukocytes (Table 1). The two peaks (I, II) unique to activated thrombocytes exhibited a shifted dihydroxy conjugated triene spectrum characteristic for suifidopeptide leukotriene with a maximum absorbance of 280 nm and shoulders at 270 and 292 nm (Table 1). Analysis of the RP-HPLC run at 235 nm (Fig. 2B) revealed two distinct peaks that showed a maximum absorbance at 243 nm (Table 1). Peak VII overlapped with 6-trans,12-epi LTB 4 and matched the retention time of 5(S),15(S)-diHETE (VII, 26.6 min). The peak at 38.4 min did not correspond to the retention time of any standard. RP-HPLC separation by mobile phase 2 and spectral analysis of monohydroxyicosatetraenoic acid compounds (n = 5) revealed 2 peaks (Fig. 2C) with a maximum absorbance of 235 nm (Table I) that matched the retention times of 15(S)-HETE (VIII, 16.2 min) and 5(S)-HETE (IX, 24.3 min). A peak (17.6 min) eluting post 15(S)-HETE (peak VIII) did not match the retention time of the 12(S)HETE standard (19.0 min). Analysis of 1 min RP-HPLC fractions by LTB 4 and sulfidopeptide specific RIAs (n = 3) revealed cross-reac-
K. Gronert et al. / Biochimica et Biophysica Acta 1259 (1995) 203-210
tivity with specific fractions (Fig. 3) that matched the retention times of LTB 4 and LTCn/LTD 4, respectively. The RIAs measured 26 -+_5 ng (LTC4), 4 ___0.4 ng (LTD4) and 114 ___32 ng (LTB 4) in the combined fractions corresponding to retention times of authentic standards (n = 3).
3.3. Time-course study of [14,15,19,20-3H(N)]LTC4 metabolism and leukotriene profile in purified thrombocytes
Time (rain) Fig. 3. Radioimmunoassay analysis of metabolites from bullfrog thrombocytes. Metabolites were separated by MPI and minute fractions analyzed by LTB~ and sulfidopeptide leukotriene specific antisera (n = 3). Retention times of authentic standards are indicated.
For characterization of gamma-glutamyl transferase activity, thrombocytes were incubated with [14,15,19,203H(N)]LTC4. Metabolism to LTD4 was 16 _+ 3% (1 rain), 19-+ 6% (2.5 min), 28_+5% (5 min) and 31 -+4% (10 min) post LTC 4 addition. Percent LTC 4 metabolism at 1 min was statistically different from 5 and 10 rain ( P < 0.05). Extending incubations to 15 min did not increase the percent of L T C 4 to L T D 4 metabolism (data not shown). In the presence of the gamma-glutamyl transferase specific inhibitor serine-borate, LTC 4 metabolism was inhibited 100% (n = 3). RP-HPLC spectral analysis at 1, 2.5, 5 and 10 min revealed a pronounced change in the leukotriene profile over time (Fig. 4). Leukotriene quantities were approximated by computer integration of peaks that correspond to the retention times of authentic standards. At 1 min post A23187 activation, LTC 4 and LTB 4 were present at approximately equal concentrations (1:1). With time the ratio of LTC 4 to LTB 4 decreased 1:2 (2.5 min), 1:3 (5 rain) and > 1:3 (10 min). The same pattern was also observed with the ratio of LTC 4 to both LTB 4 isomers; 3:1 (1 rain), 2:1 (2.5 rain), 1:1 (5 min), 1:1 (10 min). The [14,15,19,20-3H(N)]LTC4 added to all incubations
v,~, I" 24
Time (rain) Fig. 4. Changes in thrombocyte leukotriene profile over time. Representative lipoxygenase profile at 1 min (A), 2.5 min (B), 5 min (C) and 10 min (D). Metabolites were separated by MP1 (FR 0.4 ml/min) and RP-HPLC run monitored at 280 nm. Retention times of authentic standards are indicated.
(0.54 ng) is below the detection limit of the RP-HPLC and therefore did not contribute significantly to the observed endogenous LTC 4 synthesis by thrombocytes.
3.4. [5,6,8,9,11,12,14,15-3H]AA thrombocytes
metabolism by purified
After a 10 min incubation 56 ___3% (A23187 activated, n = 1 0 ) and 5 2 + 2 % (NDGA inhibited, n = 8 ) of the [5,6,8,9,11,12,14,15- 3H]AA was incorporated by the cells. Analysis of the complete radiological profile of thrombocytes revealed that only a small percent (6 + 1%, n = 4) of labeled AA was located in the incubation supernatant, with the remainder accounted for as metabolites. The radiological profile for dihydroxy icosatetraenoic compounds was determined by analyzing radioactivity in 1 min RP-HPLC (MP1) fractions. This revealed two distinct peaks. Fractions 22-27 correspond to the retention times of LTD4 and the LTB 4 isomers and fractions 28-32 correspond to the retention time of LTB 4. Metabolites were inhibited 79 -+_9% by NDGA (n = 3) (Fig. 5A). No defined radiological peaks were detected for LTC 4 or 5(S),12(S)-diHETE. A peak was also detected in fractions 33-35. All standards eluted prior to this peak. NDGA did not inhibit the synthesis of this compound, indicating that it is not a lipoxygenase metabolite. RP-HPLC analysis (MP2) of the radiological profile for monohydroxy icosatetraenoic compounds revealed 2 distinct peaks (Fig. 5B). Fractions 14-17 correspond to the retention time of 15(S)-HETE and fractions 21-24 correspond to the retention time of 5(S)-HETE. Metabolites were inhibited 6 2 + 11% [15(S)-HETE] and 85-+9% [5(S)-HETE] with NDGA. No radiological peak corre-
K. Gronert et al. / Biochimica et Biophysica Acta 1259 (1995) 203-210
Time (rain) Fig. 5. Radiological lipoxygenase profile of activated and NDGA inhibited thrombocytes. Thrombocytes were incubated with 1.5 /zCi [5,6,8,9,11,12,14,15- 3H]AA and activated with 5 /xM A23187 (black bars) or pre-incubated with 10 /xM NDGA (white bars) (n = 3). Leukotrienes were separated by MPI (FR 0.3) and the RP-HPLC run monitored at 280 nm (A). Monohydroxyicosatetraenoic acid compounds were separated by MP2 (FR 0.7) and the RP-HPLC run monitored at 235 nm (B). Retention times of authentic standards are indicated.
sponded to the retention time of 12(S)-HETE. No radiological peaks corresponding to retention times of standards were detected in cells not activated with A23187 (n = 3).
4.Dicussion This is the first study to demonstrate that thrombocytes are a predominant source for endogenous LTC 4 in bullfrogs (Rana catesbeiana) and that leukotriene synthesis (LTB 4 and LTC 4) by thrombocytes is greater than reported for any mammalian cell type. Bullfrogs are the only non-mammalian vertebrate that synthesizes sulfidopeptide leukotrienes [14,27]. Endogenous LTB 4 is synthesized by amphibian peritoneal leukocytes [14-16]. The precise cell composition in these septic peritonitis studies was undetermined but was mainly composed of granulocytes. Endogenous synthesis of 5(S)-HETE a n d / o r 15(S)-HETE but no LTC4, 12(S)-HETE, 5(S),12(S)/5(S),15(S) diHETE or LXA 4 were detected in Rana catesbeiana, Rana temporia, Rana arvalis, and Bufo americanus. LTC 4 was detected in activated peritoneal leukocytes of bullfrogs only in the presence of 50 /zM exogenous AA . Activated mixed bullfrog leukocytes synthesized both leukotrienes and lipoxins and demonstrated the activity of 3 lipoxygenases (Fig. 1). This the first study to report lipoxin synthesis in amphibians. Both 5(S),I5(S)-diHETE (Fig. 1B)  and lipoxins  are synthesized by the interaction of 5-LO and 15-LO, a potential metabolic route further confirmed by large amounts of 15(S)-HETE (Fig. IC). Small quantities of 5(S),12(S)-diHETE (Fig. 1A)
suggest 12-LO activity in mixed leukocytes since this compound is synthesized by the interaction of both 5-LO and 12-LO . While lipoxins are also synthesized via the interaction of these two lipoxygenases , it is not the dominant pathway of synthesis in bullfrog leukocytes. It is not clear if lipoxins in mixed leukocytes are synthesized transcellularly as has been reported for mammals , a n d / o r endogenously by one cell type as has been reported for trout macrophages . Unexpectedly, purified thrombocytes synthesized large quantities (ng/ml) of sulfidopeptide leukotrienes. Similar to mixed leukocytes, the activity of three lipoxygenases was detected. Large quantities of LTB 4, LTB 4 isomers, LTC4, LTD4 and 5(S)-HETE suggest a highly active 5-LO (Fig. 2). The ratio of 15(S)-HETE (Fig. 1) and 5(S),15(S)-diHETE (Fig. 2) to other lipoxygenase metabolites suggests lower 15-LO activity in thrombocytes compared with mixed leukocytes. The presence of small quantities of 5(S),I2(S)-diHETE (Fig. 2A) also suggests low 12-LO activity. No lipoxins were detected. Thrombocyte concentration ranged from 85% to 100% with a similar leukotriene profile in all experiments (n = 25). RP-HPLC fractions were analyzed by RIA to confirm LTC 4, LTD 4 and LTB 4 synthesis (Fig. 3). Quantification of leukotriene specific RIA cross-reactivity demonstrated that 3750-5000 thrombocytes synthesized 30__+5 pg of sulfidopeptide leukotrienes and 114+ 32 pg of LTB 4. These values correlate with absorbance values obtained using authentic standards. Direct comparison of leukotriene synthesizing capabilities must be considered cautiously due to differences in cell size and protein content. However, mammalian reports for endogenous a n d / o r transcellular synthesis of L T C 4 / D 4 / E 4 or LTB 4 range widely from picograms to nanograms and are based on 1-600 million cells [8,38-41]. Bullfrog thrombocytes clearly have the capability to synthesize both LTB 4 and LTC 4 endogenously in quantities greater than reported for any mammalian cell type. Endogenous sulfidopeptide leukotriene synthesis and rapid conversion of LTC 4 to LTD4 in both purified erythrocytes and whole blood of bullfrog differs from the profile obtained for thrombocytes . To confirm the apparent slow and incomplete conversion of LTC 4 to LTD 4 in thrombocytes, a time-course study with [[ 14,15,19,20]- 3H(N)]LTC 4 was designed. LTC 4 to LTD4 metabolism increased from 16% at 1 rain to 31% at 10 rain with no further increase at 15 min. The metabolism rate in thrombocytes is similar to LTC 4 to LTDn/LTE 4 metabolism by human platelets (22% at 10 min) . These results confirm the observed profile from both RP-HPLC spectral and RIA analysis, demonstrating large quantities of LTC 4 (26 + 5 ng) and only small amounts of LTD4 (4 + 0.4 ng) at the end of 10 rain incubations. A dramatic change in the metabolic profile was observed over time (Fig. 4). At I min post activation, LTC 4 was the dominant leukotriene synthesized by thrombo-
K. Gronert et al. / Biochimica et Biophysica Acta 1259 (1995) 203-210
cytes. It decreased significantly when compared to LTB 4 and LTB 4 isomer concentrations at 2.5, 5 and 10 min. Several factors could be responsible for thrombocytes shifting leukotriene synthesis from the LTC 4 synthase to the LTA 4 hydrolase pathway. Availability of reduced glutathione could be a limiting factor in the reaction causing a shift to the LTA 4 hydrolase pathway. No data are available for glutathione concentrations in frog thrombocytes, but mammalian platelets have a high (mM) concentration of reduced glutathione . If glutathione is not a limiting factor, the accumulation of LTA 4 may directly or indirectly shift pathways towards LTB 4 synthesis. The limiting step for transcellular platelet LTC 4 synthesis is the availability of LTA 4 . Both LTC 4 synthase and LTA 4 hydrolase are inefficient in converting LTA 4 in mammals. Yields are usually below 15%, and activated cells can release more than 45% of the LTA 4 extracellularly . In thrombocytes, large amounts of LTA 4 are released and coincide with the shift towards the LTA 4 hydrolase pathway at 2.5 min. High amounts of LTA 4 in combination with large quantities of 15(S)-HETE (Fig. 1), a known inhibitor of 5-LO  and lipoxins (Fig. IB) in mixed leukocytes containing 38% thrombocytes may explain the absence of any HPLC detectable sulfidopeptide leukotrienes. Two defined radiological peaks corresponded to L T D 4 / L T B 4 isomers and LTB 4 respectively in [5,6,8,9,11,12,14,15-3H]AA experiments (Fig. 5A). These peaks were inhibited 79 + % by NDGA, a potent lipoxygenase inhibitor , further confirming these products as lipoxygenase metabolites. RP-HPLC analysis of endogenous leukotriene synthesis in both mixed leukocytes and thrombocytes confirmed inhibition by NDGA, as spectral analysis demonstrated 100% inhibition. Metabolites at the beginning and end of the RP-HPLC run, which were not inhibited by NDGA, prevented detection of radiological peaks for both 5(S), 12(S) diHETE and LTC ~. 15(S)-HETE and 5(S)-HETE (Fig. 5B) were inhibited 62% and 85%, respectively, by NDGA. The observed difference is likely due to the disparate sensitivity of 5-LO and 12-LO to this inhibitor . Arachidonic acid was rapidly metabolized and/or incorporated into cells. After 10 min incubations, only 6% of [5,6,8,9,11,12,14,15-3H]AA remained in the supematant. The balance of the labeled AA was incorporated, 56% (A23187 activated) and 52% (NDGA inhibited), or accounted for in metabolites. The combination of large a m o u n t s of e n d o g e n o u s AA metabolites and [5,6,8,9,11,12,14,15- 3H]AA metabolism/cell incorporation suggests that deacylation and reacylation of A A occurs in A23187 activated thrombocytes simultaneously. It has been suggested that reacylation of AA contributes to the regulation of icosanoid synthesis in mammalian platelets . In view of the apparent AA sparing by thrombocytes, it is unclear what effect large amounts of exogenous leukocyte LTA 4 have on AA metabolism or if
thrombocytes transcellularly synthesize LTA 4, a metabolically 'cheaper' pathway for leukotriene synthesis. In the circulatory system of mammals and amphibians, leukotriene production is completely dependent on the production of LTA 4 by nucleated inflammatory cells. Further endogenous or transceilular metabolism of this unstable epoxide by blood a n d / o r endothelial cells determines the leukotriene profile at the site of an inflammatory stimulus which is highly dependent on the local cell composition. In view of the high endogenous sulfidopeptide leukotriene synthesis by pure thrombocytes, the absence of detectable L T C n / L T D 4 in mixed leukocytes and the reported LTB 4 synthesis by peritoneal granulocytes make it reasonable to assume that thrombocytes are the predominant source of LTC~ in bullfrog blood. The absence of 5-LO in mammalian platelets and erythrocytes strongly suggests that the loss of the nucleus in the evolution of these mammalian myeloid cells correlates with the lack of endogenous leukotriene synthesis. It is unclear what differences between mammalian and amphibian leukocyte regulatory mechanisms account for the apparent higher rate of leukotriene synthesis in bullfrogs. The novel finding that bullfrog thrombocytes synthesize large quantities of LTB 4 and sulfidopeptide leukotrienes endogenously, suggests a function other than coagulation for these cells. Thrombocytes potentially play a pivotal role in the regulation of inflammatory events and modulation of vascular tone or smooth muscle contraction.
Acknowledgements This study was supported in part by National Science Foundation Grant IBN-9318047 to C.A,H., and a graduate student research award from the Graduate School, New Mexico State University to K.G.S.M.V. was supported by the Minority Access to Research Careers Programs. The authors thank Dr. R. Peter Herman, Department of Biology, New Mexico State University for helpful discussion of the data and critical review of the manuscript.
References [I] Samuelsson. B., Hammarstrom,S., Murphy, R.C. and Borgeat, P. (1980) Allergy (Copenhagen) 35, 375-381.  Samuelsson, B. (1983) Science 220, 568-575.  Rouzer, C.A., Matsumoto,T. and Samuelsson, B. (1986) Proc. Natl. Acad. Sci. USA 83, 857-861.  Borgeat, P. (1988) Can. J. Physiol. Pharmcol. 67, 936-942.  Ford-Hutchinson,A.W., Gresser, M. and Young, R.N. (1994) Annu. Rev. Biochem. 63, 383-417.  Fitzpatrick, F., Liggett, W., McGee, J., Bunting, S., Morton, D. and Samuelsson, B. (1984)J. Biol. Chem. 259, 11403-11407.  Pace-Asciak, C.R., Klein, J. and Spielberg, S.P. (1986) Biochim. Biophys. Acta 877, 68-74.  Maclouf. J., Murphy, R.C. and Henson, P.M. (1989) Blood 74, 703-707.
K. Gronert et al. / Biochimica et Biophysica Acta 1259 (1995) 203-210
 Grimminger, F., Menger, M., Becker, G. and Seeger, W. (1988) Blood 72, 1687-1692.  Hamberg, M., Svensson, J. and Samuelsson, B. (1975) Proc. Natl. Acad. Sci. USA 72(8), 2994-2998.  Serhan, C.N. (1994) Biochim. Biophys. Acta 1212, 1-25.  Andazola, J.J., Underwood, J.A., Chiono, M., Torres, O.A. and Herman, C.A. (1992)J. Pharm. Exp. Therap. 263, 1117-1123.  Chiono, M., Heller, R.S., Andazola, J.J. and Herman, C.A. (1991) J. Pharm. Exp. Therap. 265, 1042-1048.  Green, F.A. (1987)Biochem. Biophys. Res. Commun. 148, 15331539.  Green, F.A., Herman, C.A., Herman, R.P., Claesson, H. and Hamberg, M. (1987)J. Exp. Zool. 243, 211-215.  Green, F.A., Herman, C.A., Herman, R.P., Claesson, H. and Hamberg, M. (1987) Biochim. Biophys. Res. Commun. 142, 309-314.  Heller, R.S., Herman, R.P. and Herman, C.A. (1987) Can. J. Physiol. Pharmacol. 67, 309-314.  Herman, R.P., Heller, R.S., Canavan, C.M. and Herman, C.A. (1988) Can. J. Physiol. Pharmacol. 66, 980-984.  Herman, C.A., Charlton, G.A. and Cranfill, R.L. (1991) Am. J. Physiol. 260, R834-R838.  Sun, J. and Herman, C.A. (1995) Gen. Comp. Endocrinol. 97, 199-208.  German, J.B., Bruckner, G.G. and Kinsella, J.E. (1986) Biochim. Biophys. Acta 875, 12-20.  German, J.B. and Kinsella, J.E. (1986) Biochim. Biophys. Acta 877, 290-298.  Piomelli, D. (1985) Naturwissenschaften 72, 5276-5277.  Pettitt, T.R., Rowley, A.F. and Secombes, C.J. (1989) FEBS Lett. 259, 168-170.  Pettitt, T.R. and Rowley, A.F. (1991) Comp. Biochem. Physiol. 99B, 647-652.  Rowley, A.F., Lloyd-Evans, P., Barrow, S.E. and Serhan, C.N. (1994) Biochemistry 33, 856-863.  Gronert, K., Virk, S.M. and Herman, C.A. (1995) Biochim. Biophys. Acta 1255, 3, 311-319.
 Woods, J.W., Evans, J.F., Ethier, D., Scott, S. and Vickers, P.J. (1993) J. Exp. Med. 178, 1935-1946.  Brock, T.G., Paine, R. and Peters-Golden, M (1994) J. Biol. Chem. 269, 22059-22066.  Andrews, W. (1965) Comparative hematology, Grune and Stratch, New York/London  Daimon, T., Mizuhira, V. and Uchida, K. (1979) Cell. Tiss. Res. 201, 431-439.  Dekhuyzen, M.C. (1892)Verh. Anat. Ges. (Jena)(6. Versammlung Wien) 6, 90-103.  Turner, R.J. (1988) Vertebrate Blood Cells (Rowley, A.F. and Ratcliff, N.A., eds.), Cambridge University Press, Cambridge, England.  Humason, G.L. (1979) Animal Tissue Techniques 4th Ed., W.H. Freeman, San Francisco, CA.  Pfeifer, C.A., Furilla, R.A., Gronert, K., Goss, D.D. and Herman, C.A. (1992)Can. J. Physiol. Pharmacol. 70, 1442-1449.  Pfeifer, C.A.. Furilla, R.A., Gronert, K., Goss, D.D., Romig, K.E. and Herman, C.A. (1993) Prostaglandins 45, 203-219.  Martinez, J.M., Chapunoff, D., Romero, M.A. and Herman C.A. (1994) J. Exp. Zool. 269, 298-307.  Surette, M.E., Odeimat, A., Palmantier, R., Marleau, S., Poubelle, P.E. and Borgeat, P. (1994) Anal. Biochem. 218, 392-400.  Antoine, C., Murphy, R.C., Hensen, P.M. and Maclouf, J. (1992) Biochim. Biophys. Acta 1128, 139-146.  Maclouf, J. and Murphy, R. C. (1988) J. Biol. Chem. 263, 1, 174-181.  Shindo, K., Baker, J.R., Munafo, D.A. and Bigby, T.D. (1994) J. Immun. 153, 5750-5759.  Salari, H., Braquet, P. and Borgeat, P. (1984) Prostaglandins Leukot. Med. 13, 53-60.  Preuss, A. and Patscheke, H. (1992) Prostaglandins Cardiovasc. Syst. 37, 34-40.