192
Biochimica et Biophysica Acta, 1120(1992) 192-198 © 1992ElsevierScience PublishersB.V.All ri~hlsrt:~Jvcd 0005-2760/02/$05.00
BBALIP 53933
Synthesis of platelet activating factor and metabolism of related lipids in embryonic cells K e n n e t h P. C h e p e n i k a a n d R o b e r t L. W y k l e b a Department of Anatomy, le]ferson Medical College of Thomas Jefferson Unieersity. Philadelphia, PA (USA) h Delxartment of Biochemistry, Bowman Gray School of Medicine of Wake Forest Unit'ersity, Winston.Salem, NC (USA)
(Received 12 November1991) (Revised manuscriptreceived 3 March 1992) Key words: PAF: Lipid metabolism;Embryoniccell Primary cultures of mouse embryo palate mesenchyme (MEPM) cells incubated with 1-O-[3H]alkyl-2-1yso-sn-glycero-3-phospho choline ([3H)Iyso-PAF)incorporated radiolabel into l-radyl-2-acyl-sn-glycero-3-phosphocholine(PC) and -phosphoethanolaminc (PE). The radiolabeied FC was insensitive to hydrolysiswith HCI fumes, whereas at least 82% of the 3H found in the PE was hydrolyzcd to 3H-aldehydes by such treatment. Treatment of the PC with Vitride produced ['~H]alkylglycerol; similar treatment of the PE produced [3H]alk-l-enylglycerol. None of the radiolabcled products yielded fatty alcohol upon reduction with Vitride. These findings indicate the radiolabeled PC was 1-O-alkyl-linked whereas the PE contained predominantly l-O-alk-I'-enyl species with smaller amounts of l-O-alkyl species. Homogenates of MEPM cells which had been prelabeled with [~H]Iyso-PAF and [14C]arachidonic acid produced 14C-fatty acid, [3H]lyso-PC, and [3H]alkylglycerol when incubated at selected values of pH and concentrations of calcium. There was no accumulation of [3H]lyso-PE in the various incubation mixtures. Stimulation of MEPM cells with the ionophore A23187 in the presence of calcium and [3H]acetate resulted in the production of 3H-plateletactivating factor (PAF), identified by its migration with authentic PAF and its conversion to l-O-[3H]alkyl-2,3-diaeetylglycerol upon treatment with phospholipase C and acetic anhydride. These studies demonstrate that: (i) MEPM cells are able to incorporate [3H]Iyso-PAFinto l-O-alkyl-2-acyl-sn-glycero-3-phosphocholine,the storage form of PAF, and into l-O-alk-l'-enyl2-acyl-sn-glycero-3-phosphoethanolaminc (PE phtsmalogen); (it) endogenous l-O-[3H]alkyl-2-acyl-sn-glycero-3-phosphocholine can serve as a substrate for phosholipase A: in homogenates; and (iii) MEPM cells have the ability to synthesize PAF, thus raising the possibility t.hnt this compound may play a role in modulating the physiologyof these embryonic cells. Introduction Basic mechanisms underlying differentiation of embryonic tissues are poorly understood. However, there is evidence that mobilization and metabolism of arachidonic acid may be an important event in the differentiation of some tissues. For examole, selected prostaglandins potentiate differentiation of chondrogenic cells [1,2], and inhibitors of eicosanoid synthesis inhibit
Abbreviations: GPC, -su.glycem-3.phosphocholine;PC and PE, Iradyl.2-acyl-sn.glycero-3-phosphocholine, -~hosphoethanolaminc;PS and Pl, 1,2-diacyl-sn-glycero-3-phospboserineand -phosphoinositol; MEPMcells, mouseembryopalate mesenchymecells:TLC, thin-layer chromatography;PAF, platelet activating factor (I-O.alkyl-2-acetylGPC); Vitrid¢, sodium dihydro-bls-(2-mcthoxyethoxy)aluminate, Na(CH3OCH:O),AIH2; lyso-PAF, l-O-alkyl.2-lyso-GPC; Hepes. N-2-hydroxycthylpiperazine-N'-2-ethancsulfonic acid; HPLC, highperformance liquid chromatography;EDTA, cthylenediaminctetraacetic acid. Correspondence: I~P. Chepenik. Department of At.atomy. Jefferson
Medical College of ThomasJeffersonUniversity. 1020 LocustStreet. Philadelphia. PA 19107, USA.
chondrogenesis [1]. The finding that treatment of selected strains of pregnant mice with glucocortieoids produces a 100% incidence of cleft palate in the neonate (reviewed in Ref. 3) is of particular interest since glucocorticoids cause inhibition of phospholipase A2 in some cells [4-7]. Other reports that administration of arachidonic acid to the pregnant dam would alleviate the teratogenic effects of glucocorticoids on the developing palate [8,9] and that arachidonic acid or its metabolites can prevent the inhibitory effects of glococorticoids on the normal process of cell death in the epithelium of the palatal shelves in vivo [9] and in vitro [10,11] further strengthen the notion that mobilization and metabolism of arachidonic acid are critical to the normal differentiation of the palate. More recently, a positive correlation was found between the inhibitory effects of glucocorticoids on phospholipase A activity expressed in primary cultures of MEPM cells from different strains of mice and their differential sensitivity to glucocorticoid induction of cleft palate [12]. Those studles were based on ~n assay wherein the activities of phospholipases were monitored by measuring the release of [3H]arachidonicacid
193
or [14Clstearic acid from endogenous phospholipids in homogenates of MEPM cells which had been radiolabeled in vitro [13]. Although the activities of various phospholipid hydrolases could be documented during the course of the assay, it was not always possible to 'trap' the lysophospholipids produced by such phospholipid hydrolases. Therefore, it was virtually impossible to specify whether (i) sequential hydrolysis by phospholipase A t and lysophospholipase, or (ii) hydrolysis by phospholipase A 2 only, was responsible for mobilization of arachidonic acid in the assay. Tile l-O-alkyl linkage of phospholipids is resistant to hydrolysis by lysophospholipases [14] and could provide a useful approach to the assay for activities of phospholipase A,. We have labeled MEPM cells with 1-O-[ 1',2'- 3H]alkyl-2-1yso-sn-glycero-3-phosphocholine ([3H]Iyso-PAF) and [l-t4C]arachidonic acid ([t4C]AA). The [3H]Iyso-PAF is taken up and acylated to form l-O-[ 3H]alky-2-acyl-sn-glycero-3-phosphocholine whose sl~-I position is resistant to attack by lysophospholipase. Since [t4C]AA is incorporated almost exclusively into (the sn-2 position of) evdogenous phospholipids by MEPM cells [13], (i) its release as free fatty acid would provide evidence of hydrolysis of phospholipids (rather than neutral lipids), and (ii) its accumulation in lysophospholipids produced under certain conditions should give a measure of phospholipase A~ activity. Materials and Methods
Methods Primary cultures of MEPM cells obtained on the 14th day of gestation were prepared as described previ. ously [15]. Cells in the late log phase of growth (4th day in culture; about 7" 105 cells attached per 8 cm 2 culture dish) were incubated in 1 ml of 10% serum which contained 0.5 p.Ci [3H]Iyso-PAF and 0.1 p.Ci [t4C]AA. Radiolabeled compounds were dissolved in ethanol and mixed with the medium. The final concentration of ethanol was 5/zl per ml of medium. At the end of the incubation period the medium was removed and the cells washed three times with warm !abosphate-buffered saline which contained 3 mg defatted bovine serum albumin per ml. Cells were then scraped into ice-cold (2-4°C) 0.25 M sucrose and cen)rifuged 30 rain × 500 g to obtain a pellet which was either homogenized in ice-cold deionized water or extracted according to Bligh and Dyer [16] except the methanol contained 2% acetic acid.
Synthesis of PAF by MEPM cells MEPM cells in the late log phase of growth were washed three times with warm (37°C Hepes-buffered saline (0.9 mM CaCI 2, 0.49 mM MgSO4, 0.373 mM KCi, 0,137 M NaCI, 0.1% dextrose, anti 22 ram Hepes),
and then incubated for 15 rain at 37°C in Hepesbuffered saline which contained A23187 (4/J M), CaCI 2 (2 izM), and 50/zCi [3H]acetate per ml. The incubation was terminated by extraction of the cells and medium according to Bligh and Dyer [16], as above. The lipid extract was chromatographed in the double development system of Emilsson and Sundler [21], along with authentic [3H]PAF an,t [3H]lysoPAF as standards. Flaorographs were prepared by spraying the chromatograph with EnHance (New England Nuclear/ Dupont) and exposing it to Kodak X-OMAT AR film for 12 days at - 70°C to permit localization and subsequent recovery of compounds which chromatographed with authentic PAF.
Assay for phospholipases Aliquots of cellular homogenates (200 /zg protein't were adjusted to 50 mM glycylglycinc (pH 9.0), Tris-HCI (pH 7.5), or acetate (pH 4.5), as described previously [13]. CaCI 2 (2 raM) was added to the mixtures at pH 7.5 and 9.0, whereas EDTA (2 raM) was added to the mixture at pH 4.5. Mixtures were incubated for 2 h or 4 h at 37°C in an agitating water bath and then extracted according to Bligh and Dyer [16], as above. Control incubation mixtures were prepared by placing homogenates in boiling water for 5 rain. These incubation conditions were chosen because they were earlier found to give optimal expression of phospholipase activities present in MEPM cells [13]. All samples were 'spiked' with 50 p.g phospholipids (extracted from rat brains) during extraction to ensure quantitative recovery of radiolabeled products.
Chemfc,~l procedures To determine whether any of the [3H]alkyl moiety was converted to an alk-l-enyl moiety, cellular pbospholipids were chromatographed along with authentic standards in one dimension in chloroform/methanol/ acetic acid, 65 : 25 : 10 (v/v), the sample lane exposed to HCI fumes, and then chromatographed in a second dimension along with authentic s~andards in chloroform/ methanol/formic acid, 65:25:10 (v/v) to separate mono- and diacyl phospbolipids and aldehydes [17], The chromatograms were sprayed with primulin (Direct Yellow 59, Colour Index 49000; 0.05% in acetone, w/v), and the lipids visualized under ultraviolet light. Addition of 50 p.g each of authentic PC plasmalogen and PE plasmalogen to the cellular extracts prior to chromatography in the first dimension ensured visualization of the various products. To determine the extent to which the [3H]alkyl moiety persisted in the phospholipid molecule, isolated phospholipids were reduced by Vitride [18] and the amount of 3H recovered as 1-O-[3H]alkylglyc.~tol, l-O[~H]alkyl-l-enylglycerol, and 3H-fatty alcohol was determined.
19~ The synthesis of PAF by MEPM cells was examined 0y incubating tire cells in the presence of [3H]acetate and 5 /~M A23187. The radiolabeled products were extracted, separated by TLC, and visualized by fluography. Radiolabel which cochromatographed with authentic PAF wa~ extracted from the silica gel according to Bligh and Dyer [16], and the 1-O-alkyl-2-acctylglycerol acetylated [19] subsequent to digestion of the PAF with phospholipase C [20]. Reference standards of 1,2-diacyl-3-acetylglycerol and 1-acyl-2,3-diacetylglycerol were prepared by acetylation [19] of 1,2-dia~lglycerol and monoacylglycerol, re~peeti'-,ely. 1-O-alkyi2,3-diacetylglycerol and 1-O-alk-l-enyl-2-acyl-3-aeetylglycerol was prepared by treatment of PAF and PE plasmalogen, respectively, with phospholipase C [20] and acetylation [19] of the resultant 1-O-al~l-2acetylglycerol and 1-O-alk-l-enyl-2-acylgly,:erol. Chromatography
The phospbo[ipids were separated on layers of Silica Gel H developed in c h l o r o f o r m / m e t h a n o l / a c e t i c acid:water, 50: 25 : 8: 2 (v/v). Neutral lipids wele separated on layers of Silica Gel G developed in chlorof o r m / methanol/acetic acid, 98 : 2:1 (v/v). PAF and lyso-PAF were separated from each other and from other phospholipids on layers of Silica Gel H by devcioping twice in chloroform/methanol/water, 65 : 35 : 0 (v/v), according to Emilsson and Sundler [211. In some cases phoqpholipids were separated in twG dimensions [17], as described above. Products of Vitride reduction were separated on Silica Gel G lhin-layer plates in chloroform/methanol/acetic acid, 98: 2:1 (v/v), or in water-saturated ether (ethyl ether/water, 98:2, v/v). Acetylated products were separated on Silica Gel G layers developed in hcxane/ethyl ether/acetic acid, 60:40:1 (v/v).
A.H. Thomas (Swedesboro. N.I), or Aldrich Chemicals/Sigrna Chemicals (St. |~uis, MO). Cultu~'e medium and antibiotics were from Grand Island Biological (Grand Island, NY), whereas sera were from Hazelton Research Products (Lenexa, KS). Results
[3H]Lyso-PAF was incorporated into PC and PE of MEPM cells within 1 h after its addition to the culture medium (Table 1). I-Iowever, since 7.5% of the added radiolabe! was incorporated into cellular lipids in 24 h whereas :)nly 1% was incorporated after 1 h, we used the longer incubation period for all s'dbsequent experiments, in contrast to lyso-PAF, 40-44% of the ['4C]arachidonic acid added to the culture, medium was incorporated into cellular iipids duri:,g a 24 h labeling period. No more than 5% of the total 31t and 14C incorporated into cellular lipids cochromatographed with authentic trialyceride following a 24 h labeling period (data not shown). The nature of the 3H-labeled neutral lipid x,:as not identified fltrther, but in all probability, was 1-O-[3H]hexade~-yl-2,3-diacylglycerol. Most of the 3H incorporated into PE was sensitive to ilydrolysis by HCI fumes whereas little was released
3° 1
Materials
[3H]Acetate (92 Ci/mmol), [l-'4C)ataehidonic acid (55 mCi/mmol), were from New England Nuclear P r o d u c t s / D u p o n t (Boston, MA).
1ooo
i.O-[ l',2'.3H]Hexadecyl.2-acetyl-sn-glycero-3-phos -
phocholine (56 Ci/mmol; [3H]PAF) was prepared a~cording to Wyrick et a1.[22], l-O-(l',2'-~H]Hexadecyl 2-1yso-sn-glycero-3-phosphocholine (56 Ci/mmol) was prepared from [~H]PAF (56 C i / m m o l ) [20]. Radiolabeled PAF and lys6-PAF yielded only one radiolabeled product when chromatographed on Silica Gel H developed in c h l o r o f o r m / m e t h a n o l / a c e t i c acid/" water, 50 : 25 : 6:1 (v/v). Cc,mmercially plepared thin-layer plates were from Bodman Chemicals (b~edia, PA), authentic phospholipid and neutral lipid stat, dards were from Serdary Research Laboratories (Ontat'!o, Canada). All solvents were HPLC grade from Fisher Scientific (King of Prussia, PA), and all other chemicals were analytical reagent grade from Fisher Scientific,
o
~
0
20
40 Zorn Number
""
60
-
I~
Fig. I. Zonal profile scan (2 mm) of products obtained by Vitride
reduction of PC and PE extracted from MEPM cells incubated with I-O.13H]alkyl-lyso-GPC.About 2. l06 cells were incubated f~r 24 h at 37°C in 3 rnl mediumwhich contained 0.5 p.Ci [3H]lyso-PAFand 0.1 ttCi [ S4C]arachidonicacid per ml. The cultureswere then washed, extracted according to Bligh and Dyer tiff] and the PC aud PE isolated by T! C and reduced with Vitride, as describe:l in Materials and Methods. The reduction products of the PC (n) and PE (e) were chromatographedon SilicaGel G TLC plates in water-saturated ether (dicthyl ether/water, 98: 2, v/v) alongwith authentic alkTIglycerol and alk-l-f:nylglyeerol.Data are representativeof three separate experiments.
195 TABLE [ Incorporation of ['~lt]lyso-PAF into phospholipid¢ of MEPM cells, in vitro a Length of incubation (h)
Percent distHb.tion ~' of 3H PC ¢ PS/PI
PE
I 24
92.5±0.7 d ~.3±0.9 e
5.2~1 14.7~0.9
1.9±0.1 I ±0.1
a Mouse embuo palate mesenchyme (MEPM) cells (about 7. l0 s) in the late log phase of growth were incubated at 37°C for either 1 h or 24 h in ! ml of medium which contained 0.5 p.Ci l-O-['~H]alkyl2-1yso-GPC (l'~Hllyso-PAF).The medium was then removed and the cells extracted according to a modification of the method of Bligh and Dyer [16]. 1% of the added label was recovered in cellular pbospholipids after incubation for I h whereas 7% was recovered in cellular phospholipids after the 24 h tw:,'iodof incubation. t, Based on total radiolabel recovered in cellular phospholipids. c Phuspholipids were separated by TLC in chloroform/methanol/ acetic acid/water, 50:25:8:2 (v/v). a Data represent the mean 5:1/2 the range of two separate determinations. Data represent the mean±standard deviation of three separate determinations.
by acid hydrolysis of the PC (Table II). Concomitantly, at least one-half the [t4C]AA incorporated into the P E fraction of M E P M cells was associated with the lyso-PE fraction produced on acid hydrolysis (Table 11). These results indicate that the 3H-labeled P E was predominantly 1-O-alk-l'-enyl-2-acyi-sn-glycero-3-phosphoe t h a n o l a m i n e (PE plasmalogen). O n the other hand, little of the [~H]alkyl moiety was metabolized to aik-1enyl groups in the PC. A T least 50% of the [t4C]AA incorporated into the P E of M E P M cells was esterified into P E plasmalogen. Vitride reduction of PC and P E extracted from M E P M cells which had been inct~bated for 24 h with 1-O-[3H]alkyl-2-1yso-PAF and [t4C]AA revealed e s ~ n tially only '~H alkyl moieties in the P(" and [3H]alk-lenyl moieties in the P E (Fig. 1). In no case were SH products found which chromatographed with the t4Calcohols [derived from [t4C]AA), indicating the ['~H]alkyl moieL~ was not metabolized to fatty acid b~ these embryonic ,:ells. W e pre:viously characterized the activities of phospholipase.,; A in homogenates of M E P M cells and founQ
,,.5
pH 9.0
"o° 'O E~'nOUb'ted "h "~ eo / .
I~
'~' "--'
0
"
PC
pH 4 : 5 . - - - o L ='~ neUad 4 ~ Incubated 2h ~0 I ~-~ |nco~,&ted 4h
'
IJPC
OE I,PE PE
PC
LPC
Cg
LPE Pg
°1 PC
pH '~,0
pH ?.5
~o
~.
o
r~', uPc uPz
pr. pc
FF& LPC UPE PI
PE PC
~ & LPC UPE P!
BE
PC
h d i o l l b e l u d Up|dl Fig. 2. Effect of the activities of endogenous phnspholipaseson the distribution of [ZH]lyso-PAF and [z4C]arachidonic acid amongst endogenous lipids of MEPM cells, Primak~, cultures of ,noose embryo palai,c mesencltyme ~MEPM) cells were radiolabelod for 24 h, homogenized, and incubated at 37°C for the indicated times and values of pH, as described in Materials and Methods. The reaction mixture contained
154042± 6489 DPM as 3H and 124135+ 5068 DPM as 14C. Lipids in the incubation mixtures were extracted into chloroform and separated b~ TLC, as described in Materials and Methods. Percent distribution = (DPM '.vh:ch chromatographod with designated authentic standard ± tot~:l DPM on the thin layer chromatograph)x lflO. Data arc the means of at least three separate determinations+one standard deviation. Values obtaiued from boiled controls incubated for either 2 or 4 h were not different and have been summed. GE, l-O-alkylglycerol;FFA, free fatty acid. Remaining abbreviations ate as in "Fables! an¢. H.
196 TABLE !1
I~corporation of [ ~HIlyswPAFa and [ ~C]arachidonic acid into p/asmalogen species of I'C apd PE of MEPM cells Parent compound
Products of actd hydrolysis~
Percent distribution b 3H ~4C
PC
aldehyde ly~,o-PC nc t hydrolysed
3+2 d 2+_ I 95 + I
4+_ I 4=!:3 92 +_4
PE
aldehyde lyso.PE
82 + 2 I =1:I 17 ± 3
10_+4 53 + 3 37 + l
not hy~rolvsed
" Primary cultures ot mouse embryo palate mesenchyme (MEPM) cells were incubated at 37°C in I ml of medium which contained 0.5 p.Ci I-O-[I',2'-3H]alkyl-2-1yso-GPC([3H]LPAF) and 0.1 /zCi [IJ4Clarachidonic acid, as described in Materials and Methods. The medium was removed after 24 h and the cells washed, fixed with a solution of methanol/water (2:1, v/v). scraped into test tubes and their lipids extracted as de~ribed in Materials and Methods. Approx. 7% of the added 3H and 42% of the added l l4C wece incorporated into cellular lipids during the incubation periot~. b Percent Distribution = (DPM in unhydrolyscd component, aldehyde, or lyso clcrivative/sum of the DPM in unhydrolyscd component, aldehyde, and lyso derivative)X 100. c Phospholipids were separated on EM 60 Silica Gel H TLC plates and the distribution of radiolabel into alkyl and alk-l-enyl moieties determined as described in Materials and Methods. '~ Data are the mean±one standard deviation obtained on analysis of four separate samples.
peaks of hydrolytic activity at p H 4.5, 7.5, and 9.0 [13]. In those studies hydrolysis of e n d o g e n o u s phospholipids was found to bc almost linear over a 6 h incubation period [13]. Therefore, M P E M cells radiolabeled with [3H]Iyso-PAF and [t4C]AA were homogenized and incubated 2 h or 4 h at e i t h e r pH 4.5, 7.5, or 9.0 to obtain a measure of hydrolysis of e n d o g e n o u s e t h e r lipids. T,~cre was an almost linear accumulation of 14Cfatty acid during ute 4 h incubation period at all values of pH (Fig. 2). Maximum accumulation of 14Cfatty acid c,:curred at pl-! 9.0, and m i n i m u m accumulation was at pH 4.5 (Fig. 2). l u c r e was a relatively greater decrease m [14C]PC than in [t4C]pE at all values of pH, and in all cases [14C]P! decreased to the least extent (Fig. 2). 14C-Labeled lyso-PC and lyso-PE accumulated at p H 4.5 only (Fig. 2). In contrast, l-O[3H]alkyl-2-acyl-sn-glycero-3-phosphocholine was hydrolyzed to yield [3H]lyso-PC at all values of pH (Fig. 2). T h e r e was also some accumulation of 3H products which c h r o m a t o g r a p h e d with authentic l-O-octadec9'-enyl-sn-gfycerol (glycer(d ether, GE; Fig. 2). The label, however, would be the 1 6 : 0 alkylglycerol (citimyl alcohol). The percent of the total ' H found as iyso-FC and alkylgly¢crol was i n w r s e l y related to the percentage of 3H lost from PC , t pH 4.5 (Fig. 2). A t pH 7.5 and 9.0 the accumulation of ['~H]lyso-PC and alkylglycerol was about o~.e-half that found at pH 4.5. This may
reflect the presence of transacylase activity, which could reacylate lyso-PAF as it is formed, since 15 + 3% ( ± S . D . , n = 4 ) of tl,e total [3H]Iyso-PAF (4 pmol) a d d e d to homogenates of unlabeled M E P M cells was incorporated into PC at p H 7.5 and 9.0; no radiolabel was incorporated into PC at pH 4.5. Radiolabeled a/achidonic acid was iiot incorporated into phospholipids when incubated with homogenates of M E P M cells u n d e r similar conditions [13]. Since M E P M cells actively incorporated lyso-PAF into l-O-all~,l-2-acyI-PC, the storage form of P A F [23], we explored the i~ossibility these embryonic cells might be able to synthesize P A F itself. Indeed, incubation of M E P M cells with the Ca 2+ ionophore A23187 and [3H]acetate resulted in production of a radiolabeled product which c h r o m a t o g r a p h e d w:.th authentic P A F and was resistant to hydroly*is with HCI fumes (data not shown). Digestion of thi~ putative [3H]PAF with phospholipase C and acetylation of the resultant diradylglycerol produced a radiolabeled derivative that e h r o m a t o g r a p h e d with a u t h e n t i c l-O-alkyl-2,3-diacetylglycerol (Fig. 3). Digestion of the [3H]PAF with the phospholipa~e A 2 of Crotalus adementeous venom released over 70% of the radiolabel into the aqueous phase, indicating the label was present as a 3H acetyl moiety.
~,o. Zone number: compound 15-16:acyldlace~)lyc~ ol 17-21:alkyldt~catylolyc~ol 23-2S:dtacylacely~yceml 26-30:aJk-I-enykcy~et l~ycerol
,¢00.
3ooo
2000
t0oo 0 ...... .......
j -I~
20 30 ~ 7onql l~rnl~h Fig. 3. Identification of radlolabelod l-O-a~kyl-2,3.diacely|glycerol
derived fro~ I -O.alkyl-2.l'~Hjacety\.GPC synthesized by MEFM ccl!:.
Cells (2. l0 ~') were incubaled wit;, [3HJacetate a.ld stimulated with A~3t87 (O) or vehicle (*), as d..'.scribed in MateGals and Methods. Chromatography of extraels of cells stimulated with A23187 resulted in 20P~ DPM recovered in tht; fraction which cochromatographed with authc.tic PAF, Ti~',s radiolabeled product w~s r~:oa':ered, digested, acetylated and analyzed by zonal TLC (5 mm s~ans) after scparatior~ by development in hexane/ethyl ether~acetic acid, 60:40: I, v/v), as described in Materials ~nd Methods. Separation ot V~l'iousacetylatod products was achieved with thi~ solvem system, as indict:ted in the figure.
197 Discussion
We have explored the synthesis of PAF and the metabolisms of related compounds by primary cultures of mouse embryo palate mesenchyme cells. Unlike differentiated polymorphonuclear leukocytes [24], these embryonic cells incorporated radiolabel into alkylacylPC and PE plasm.~logen. Incorporation of the radiolabeled alkyl moiety as the alk-l-enyl into PE plasmalogen suggests an active pathway is present that can transfer the alkylglycerol component from PC to PE where it can then be desatvrated to the plasmalogen (reviewed in Ref. 25). The synthesis of PE plasmaiogens by embryonic cells is consistent with previous findings that embryonic tissues contain significant amounts of PE plasmalogen [26]. Incubation of homogenates of prelabeled cells at acid pH resulted in *.he accumulation of 3H- and t4Clabeled lys3-PC. Accumulation of those products of hydrolysis indicates the presence of phospholipases A 2 and A t, r¢spectively, and is consistent with previous studies of these cells [13]. Most important, we were able to obtain a minimal measure of the activities of ~hospholipase A 2 in homogenates incubated at neutral and alkaline pH since there was accumulation of [3H]Iy,,~o-PC which was parallel to a loss of [~4C]AA from the PC fraction. This contrasts with our failure to find any radiolabeled lysophospholipids under similar conditlons when endogenous phospholipids of MEPM cells wcae radioiabeled in the sn-1 position with acyl nloietles [13]. These contrasting results caa he explained by the presence of a iysophosphelipase that hydrolyzes l-acyl-linked lysophospholipids, hut cannot hydrolyze the ether-linked species. It would appear that PE plasutalogens are either not hydroly:,ed by the cellular phospholipase A 2 or there is rapid r¢ acyl~tion of l-O-alk-l'-enyl-2-1yso-PE under the conditions of our assay since (i) 50% of the [14C]araehidonate and essentially all of the 3H alkyl moiety incorporated into PE was in PE plasmalogen, (i0 ~:C-tatty acid was lost from PE during the course of the assay but there was neither accumulation of [3H]lyso-PE nor loss of [3H]PE during; I.he course of the assay. During the course of the assay there was accumulation of a 3H-labeled compound which migrated with l-O-alk3,1glycerol. This finding may reflect sequential hydrolysis of choline phospholipids by phospholipase C and diglyceride lipase, or phospholipase D activity in concert with a phospbatidic acid phosphatase and a lipase. Phospholipase~ C and D which hydrolyze 1-Oalkyl-2-acyl-sn-glycero-3-phosphocholine have been reported in various types of cells recently [27,28]. We found MEPM ceils have the ability to synthesize PAF and its immediate precursor, l-O-alkyl-2-acyi-PC (by acylation nf lyso.PAF). In turn, O'Neill [29-31] has reported recent'y on the ability of the very early era-
bryo (fertilized ooc~,te) to produce PAF, and raised the possibility that it :nay serve as an early embryonic message. Our finding that cells obtained from a discrete embryonic structure (the palate) are able to synthesize PAF raises the possibility this biologically 13otent compound might serve as an autocrine in regulating certain aspects of differentiation. Many of the responses of differentiated cells to PAF, such as migration, adhesion, secretion, and uptake of Ca 2÷ (reviewed in ReL 32) are processes which are also associated with differentiation of embryonic cells and tissues. In summary, we found that: (i) embryonic cells are able to incorporate lyso-PAF into l-O-alkyl-2-acs,I-PC (the storage form of PAF) and ethanolamine p!asmalogen, (it) endogenous, double-labeled 1-O-aikyl-2-acylPC serves as a substrate for intracellular phospholipase A2, with subsequent accumulation of radiolabeled lyso-PAF, and (iii) embryonic palatal cells are able to synthesize PAF, thus prompting the notion this effector molecule may play a role in regulating growth a n d / o r differentiation of embryonic tissues. Acknowledgement This work was supported by N.I.H. grant Nos. DE 06553, HL 26818, and AI 17287. ~eferences 1 Chepenik, K.P., Ho, W.C., Waite, ~.M, and Parker, C.L. {1984) Calcif. Tissue. tat. 36, 175-181. 2 Gay, S.W. and Kosher, R.A. (1984) J. Exp. Zool. 232, 317-326. 3 Pratt, R.M., Kim, C.S. and Grove, R.L (1984) in Current Topics in DevelopmentalBiology(Zimmerman,LF., ed.), Vol. 19, pp. 81-10t, AcademicPress, New York. 4 Crutchley,D.J., Ryan, U,S, and Ryan, J.W. (1985) J. PharmacoL Exp. Thor, 233(3), 650-655. 5 Maridonneau-Parini,i., Errasfa, M. and Russo-Marie,F. (1989) J. Clin. Invest,83, 1936-1940, 6 Rothhut, B., Russo-Marie,F., Wood, J., DiRosa, M. and Flower, RJ. (1983) Biochem.Biophys.Res. Commun. 117,878-884. 7 Flower, RJ. and Blackwen,G.J. (1979) Nature 278. 456-459. 8 Tzortzatou, G.G., Goldman, A.S. and BoutwelL W.C, (1¢81) Prec. Soc. Exp. Biol, Med. 166,321-324. 9 Piddington,R., Herold, R. and Goldman, A.S. (1983}Prec., Soc, Exp. BioL Med, 174,336-342, 10 Goldman, A,S,, Baker, M.K,, Piddington, R. and Herold, R, (1983) Proc. :~.oc.Exp, Biol. Med. 174, 239-243. I1 Montenegro,M,A. and Paz de La Vega, Y. (1982) Arch. Oral. Biol. 27, 771-775. 12 Chepenik,K.P, George, M. and Greene, R.M. (1985)Teratology 32,119-123. 13 Geroge, M. and Chepeaik, K.P. (1985) Biochlm. Biophys. Acts .~36.45- 55, 14 Wykle,R.L. and Snyder, F. (1976) in The Enzymesof Biological Membranes (Martonosi, A., ed.), Vol. 2, pp. 87-117, Plenum PpJblishing,New York. 15 Chabot, M.C. and Chepenik K.P. (1986) J. Craniofac. Genet. Dev. Biol. 6, 223-234. 16 Bligh. E.G, and Dyer, W.J, (1959) Can. J, Biochem Physiol.37. 911-917.
1.98 17 Esko, LD. and Raetz, C.R.H. (19~0} J. Biol, Chore, ~ 5 , ~ 7 4 ~80. 18 S-~'a.~-~. r:., Blank, M. and Wykle, R.L. (1971) J. Biol. Chem. 246, 3b.; ~ -3645. 19 Waku, K., !1o, H., Bito, T. and Nakazawa, v . (!o7a) 1. ~inrhe::t. 75, 1307-1312. 20 Mueller, H.W., O'Flahaerty, J.T. and W~/k'.e.R.L. (1983) J. Biol. Chem. 258, 6213-6218. 21 Emilsson, A. and Sundlcr, R. (1985) Biochim. Biophys. Acta 816. 265-274. 22 Wyrick, S.D,, McClanaha,, J.S., Wykle, R.L. and O'Flaherty, J.T. (1£85) J. Labeled. Comod. Radiopharm. 22, 1169-1174. 23 Snyder, F. (1988) in Biological Membrane~: Aberrations in Membrane Structure and Function, pp. 57-.72. Alan R. li~s. New York.
24 Chilton, F.H., O'Flahcr,"j, 3iF., Ellis, 3.M,, Swei*ds~a, C.L, and Wykle, R,L. (1983)J. BioL Chem. 258, 6357-6361. 25 Snyder, F., Lee. T-C. and Wykle, R.L. (1985) in The Enzymes of Biological Membranes (Martone3i, ,~..N,, cd.), Voi. 2, pp. 1-58, Plenum Publishing, New York. 26 Chepenik, K.P., Blank, M., Snyder, F. and Waite, B.M. (1980) int. J. Biochem. !1,605-607. 27 Exton, J.H. (1990) 3. Biol. Chem. 265, I-4. 28 Exton, J.H. (198~) FA~;EB J. 2, 2670-2676. 29 O'Neill, C. (198:~) .L Reprod. Fort. 75, 3'75-~t~U. 30 O'Neill, C. (1985)J. Reprod. Ferl. 73, 559-566. 31 O'Neill, C. (1985; J. Reprod. Fen. 73, 567-677. 32 Lee, T-C. and Snyder, F, (1985) in Phospholipids and Cellular Activation (Kuo, J.F., ed.), Vol. 2, pp, l-39, CRC Press, Boca Raton.
Biochimica et Biophysica Acta, 1120(1992) 192-198 © 1992ElsevierScience PublishersB.V.All ri~hlsrt:~Jvcd 0005-2760/02/$05.00
BBALIP 53933
Synthesis of platelet activating factor and metabolism of related lipids in embryonic cells K e n n e t h P. C h e p e n i k a a n d R o b e r t L. W y k l e b a Department of Anatomy, le]ferson Medical College of Thomas Jefferson Unieersity. Philadelphia, PA (USA) h Delxartment of Biochemistry, Bowman Gray School of Medicine of Wake Forest Unit'ersity, Winston.Salem, NC (USA)
(Received 12 November1991) (Revised manuscriptreceived 3 March 1992) Key words: PAF: Lipid metabolism;Embryoniccell Primary cultures of mouse embryo palate mesenchyme (MEPM) cells incubated with 1-O-[3H]alkyl-2-1yso-sn-glycero-3-phospho choline ([3H)Iyso-PAF)incorporated radiolabel into l-radyl-2-acyl-sn-glycero-3-phosphocholine(PC) and -phosphoethanolaminc (PE). The radiolabeied FC was insensitive to hydrolysiswith HCI fumes, whereas at least 82% of the 3H found in the PE was hydrolyzcd to 3H-aldehydes by such treatment. Treatment of the PC with Vitride produced ['~H]alkylglycerol; similar treatment of the PE produced [3H]alk-l-enylglycerol. None of the radiolabcled products yielded fatty alcohol upon reduction with Vitride. These findings indicate the radiolabeled PC was 1-O-alkyl-linked whereas the PE contained predominantly l-O-alk-I'-enyl species with smaller amounts of l-O-alkyl species. Homogenates of MEPM cells which had been prelabeled with [~H]Iyso-PAF and [14C]arachidonic acid produced 14C-fatty acid, [3H]lyso-PC, and [3H]alkylglycerol when incubated at selected values of pH and concentrations of calcium. There was no accumulation of [3H]lyso-PE in the various incubation mixtures. Stimulation of MEPM cells with the ionophore A23187 in the presence of calcium and [3H]acetate resulted in the production of 3H-plateletactivating factor (PAF), identified by its migration with authentic PAF and its conversion to l-O-[3H]alkyl-2,3-diaeetylglycerol upon treatment with phospholipase C and acetic anhydride. These studies demonstrate that: (i) MEPM cells are able to incorporate [3H]Iyso-PAFinto l-O-alkyl-2-acyl-sn-glycero-3-phosphocholine,the storage form of PAF, and into l-O-alk-l'-enyl2-acyl-sn-glycero-3-phosphoethanolaminc (PE phtsmalogen); (it) endogenous l-O-[3H]alkyl-2-acyl-sn-glycero-3-phosphocholine can serve as a substrate for phosholipase A: in homogenates; and (iii) MEPM cells have the ability to synthesize PAF, thus raising the possibility t.hnt this compound may play a role in modulating the physiologyof these embryonic cells. Introduction Basic mechanisms underlying differentiation of embryonic tissues are poorly understood. However, there is evidence that mobilization and metabolism of arachidonic acid may be an important event in the differentiation of some tissues. For examole, selected prostaglandins potentiate differentiation of chondrogenic cells [1,2], and inhibitors of eicosanoid synthesis inhibit
Abbreviations: GPC, -su.glycem-3.phosphocholine;PC and PE, Iradyl.2-acyl-sn.glycero-3-phosphocholine, -~hosphoethanolaminc;PS and Pl, 1,2-diacyl-sn-glycero-3-phospboserineand -phosphoinositol; MEPMcells, mouseembryopalate mesenchymecells:TLC, thin-layer chromatography;PAF, platelet activating factor (I-O.alkyl-2-acetylGPC); Vitrid¢, sodium dihydro-bls-(2-mcthoxyethoxy)aluminate, Na(CH3OCH:O),AIH2; lyso-PAF, l-O-alkyl.2-lyso-GPC; Hepes. N-2-hydroxycthylpiperazine-N'-2-ethancsulfonic acid; HPLC, highperformance liquid chromatography;EDTA, cthylenediaminctetraacetic acid. Correspondence: I~P. Chepenik. Department of At.atomy. Jefferson
Medical College of ThomasJeffersonUniversity. 1020 LocustStreet. Philadelphia. PA 19107, USA.
chondrogenesis [1]. The finding that treatment of selected strains of pregnant mice with glucocortieoids produces a 100% incidence of cleft palate in the neonate (reviewed in Ref. 3) is of particular interest since glucocorticoids cause inhibition of phospholipase A2 in some cells [4-7]. Other reports that administration of arachidonic acid to the pregnant dam would alleviate the teratogenic effects of glucocorticoids on the developing palate [8,9] and that arachidonic acid or its metabolites can prevent the inhibitory effects of glococorticoids on the normal process of cell death in the epithelium of the palatal shelves in vivo [9] and in vitro [10,11] further strengthen the notion that mobilization and metabolism of arachidonic acid are critical to the normal differentiation of the palate. More recently, a positive correlation was found between the inhibitory effects of glucocorticoids on phospholipase A activity expressed in primary cultures of MEPM cells from different strains of mice and their differential sensitivity to glucocorticoid induction of cleft palate [12]. Those studles were based on ~n assay wherein the activities of phospholipases were monitored by measuring the release of [3H]arachidonicacid
193
or [14Clstearic acid from endogenous phospholipids in homogenates of MEPM cells which had been radiolabeled in vitro [13]. Although the activities of various phospholipid hydrolases could be documented during the course of the assay, it was not always possible to 'trap' the lysophospholipids produced by such phospholipid hydrolases. Therefore, it was virtually impossible to specify whether (i) sequential hydrolysis by phospholipase A t and lysophospholipase, or (ii) hydrolysis by phospholipase A 2 only, was responsible for mobilization of arachidonic acid in the assay. Tile l-O-alkyl linkage of phospholipids is resistant to hydrolysis by lysophospholipases [14] and could provide a useful approach to the assay for activities of phospholipase A,. We have labeled MEPM cells with 1-O-[ 1',2'- 3H]alkyl-2-1yso-sn-glycero-3-phosphocholine ([3H]Iyso-PAF) and [l-t4C]arachidonic acid ([t4C]AA). The [3H]Iyso-PAF is taken up and acylated to form l-O-[ 3H]alky-2-acyl-sn-glycero-3-phosphocholine whose sl~-I position is resistant to attack by lysophospholipase. Since [t4C]AA is incorporated almost exclusively into (the sn-2 position of) evdogenous phospholipids by MEPM cells [13], (i) its release as free fatty acid would provide evidence of hydrolysis of phospholipids (rather than neutral lipids), and (ii) its accumulation in lysophospholipids produced under certain conditions should give a measure of phospholipase A~ activity. Materials and Methods
Methods Primary cultures of MEPM cells obtained on the 14th day of gestation were prepared as described previ. ously [15]. Cells in the late log phase of growth (4th day in culture; about 7" 105 cells attached per 8 cm 2 culture dish) were incubated in 1 ml of 10% serum which contained 0.5 p.Ci [3H]Iyso-PAF and 0.1 p.Ci [t4C]AA. Radiolabeled compounds were dissolved in ethanol and mixed with the medium. The final concentration of ethanol was 5/zl per ml of medium. At the end of the incubation period the medium was removed and the cells washed three times with warm !abosphate-buffered saline which contained 3 mg defatted bovine serum albumin per ml. Cells were then scraped into ice-cold (2-4°C) 0.25 M sucrose and cen)rifuged 30 rain × 500 g to obtain a pellet which was either homogenized in ice-cold deionized water or extracted according to Bligh and Dyer [16] except the methanol contained 2% acetic acid.
Synthesis of PAF by MEPM cells MEPM cells in the late log phase of growth were washed three times with warm (37°C Hepes-buffered saline (0.9 mM CaCI 2, 0.49 mM MgSO4, 0.373 mM KCi, 0,137 M NaCI, 0.1% dextrose, anti 22 ram Hepes),
and then incubated for 15 rain at 37°C in Hepesbuffered saline which contained A23187 (4/J M), CaCI 2 (2 izM), and 50/zCi [3H]acetate per ml. The incubation was terminated by extraction of the cells and medium according to Bligh and Dyer [16], as above. The lipid extract was chromatographed in the double development system of Emilsson and Sundler [21], along with authentic [3H]PAF an,t [3H]lysoPAF as standards. Flaorographs were prepared by spraying the chromatograph with EnHance (New England Nuclear/ Dupont) and exposing it to Kodak X-OMAT AR film for 12 days at - 70°C to permit localization and subsequent recovery of compounds which chromatographed with authentic PAF.
Assay for phospholipases Aliquots of cellular homogenates (200 /zg protein't were adjusted to 50 mM glycylglycinc (pH 9.0), Tris-HCI (pH 7.5), or acetate (pH 4.5), as described previously [13]. CaCI 2 (2 raM) was added to the mixtures at pH 7.5 and 9.0, whereas EDTA (2 raM) was added to the mixture at pH 4.5. Mixtures were incubated for 2 h or 4 h at 37°C in an agitating water bath and then extracted according to Bligh and Dyer [16], as above. Control incubation mixtures were prepared by placing homogenates in boiling water for 5 rain. These incubation conditions were chosen because they were earlier found to give optimal expression of phospholipase activities present in MEPM cells [13]. All samples were 'spiked' with 50 p.g phospholipids (extracted from rat brains) during extraction to ensure quantitative recovery of radiolabeled products.
Chemfc,~l procedures To determine whether any of the [3H]alkyl moiety was converted to an alk-l-enyl moiety, cellular pbospholipids were chromatographed along with authentic standards in one dimension in chloroform/methanol/ acetic acid, 65 : 25 : 10 (v/v), the sample lane exposed to HCI fumes, and then chromatographed in a second dimension along with authentic s~andards in chloroform/ methanol/formic acid, 65:25:10 (v/v) to separate mono- and diacyl phospbolipids and aldehydes [17], The chromatograms were sprayed with primulin (Direct Yellow 59, Colour Index 49000; 0.05% in acetone, w/v), and the lipids visualized under ultraviolet light. Addition of 50 p.g each of authentic PC plasmalogen and PE plasmalogen to the cellular extracts prior to chromatography in the first dimension ensured visualization of the various products. To determine the extent to which the [3H]alkyl moiety persisted in the phospholipid molecule, isolated phospholipids were reduced by Vitride [18] and the amount of 3H recovered as 1-O-[3H]alkylglyc.~tol, l-O[~H]alkyl-l-enylglycerol, and 3H-fatty alcohol was determined.
19~ The synthesis of PAF by MEPM cells was examined 0y incubating tire cells in the presence of [3H]acetate and 5 /~M A23187. The radiolabeled products were extracted, separated by TLC, and visualized by fluography. Radiolabel which cochromatographed with authentic PAF wa~ extracted from the silica gel according to Bligh and Dyer [16], and the 1-O-alkyl-2-acctylglycerol acetylated [19] subsequent to digestion of the PAF with phospholipase C [20]. Reference standards of 1,2-diacyl-3-acetylglycerol and 1-acyl-2,3-diacetylglycerol were prepared by acetylation [19] of 1,2-dia~lglycerol and monoacylglycerol, re~peeti'-,ely. 1-O-alkyi2,3-diacetylglycerol and 1-O-alk-l-enyl-2-acyl-3-aeetylglycerol was prepared by treatment of PAF and PE plasmalogen, respectively, with phospholipase C [20] and acetylation [19] of the resultant 1-O-al~l-2acetylglycerol and 1-O-alk-l-enyl-2-acylgly,:erol. Chromatography
The phospbo[ipids were separated on layers of Silica Gel H developed in c h l o r o f o r m / m e t h a n o l / a c e t i c acid:water, 50: 25 : 8: 2 (v/v). Neutral lipids wele separated on layers of Silica Gel G developed in chlorof o r m / methanol/acetic acid, 98 : 2:1 (v/v). PAF and lyso-PAF were separated from each other and from other phospholipids on layers of Silica Gel H by devcioping twice in chloroform/methanol/water, 65 : 35 : 0 (v/v), according to Emilsson and Sundler [211. In some cases phoqpholipids were separated in twG dimensions [17], as described above. Products of Vitride reduction were separated on Silica Gel G lhin-layer plates in chloroform/methanol/acetic acid, 98: 2:1 (v/v), or in water-saturated ether (ethyl ether/water, 98:2, v/v). Acetylated products were separated on Silica Gel G layers developed in hcxane/ethyl ether/acetic acid, 60:40:1 (v/v).
A.H. Thomas (Swedesboro. N.I), or Aldrich Chemicals/Sigrna Chemicals (St. |~uis, MO). Cultu~'e medium and antibiotics were from Grand Island Biological (Grand Island, NY), whereas sera were from Hazelton Research Products (Lenexa, KS). Results
[3H]Lyso-PAF was incorporated into PC and PE of MEPM cells within 1 h after its addition to the culture medium (Table 1). I-Iowever, since 7.5% of the added radiolabe! was incorporated into cellular lipids in 24 h whereas :)nly 1% was incorporated after 1 h, we used the longer incubation period for all s'dbsequent experiments, in contrast to lyso-PAF, 40-44% of the ['4C]arachidonic acid added to the culture, medium was incorporated into cellular iipids duri:,g a 24 h labeling period. No more than 5% of the total 31t and 14C incorporated into cellular lipids cochromatographed with authentic trialyceride following a 24 h labeling period (data not shown). The nature of the 3H-labeled neutral lipid x,:as not identified fltrther, but in all probability, was 1-O-[3H]hexade~-yl-2,3-diacylglycerol. Most of the 3H incorporated into PE was sensitive to ilydrolysis by HCI fumes whereas little was released
3° 1
Materials
[3H]Acetate (92 Ci/mmol), [l-'4C)ataehidonic acid (55 mCi/mmol), were from New England Nuclear P r o d u c t s / D u p o n t (Boston, MA).
1ooo
i.O-[ l',2'.3H]Hexadecyl.2-acetyl-sn-glycero-3-phos -
phocholine (56 Ci/mmol; [3H]PAF) was prepared a~cording to Wyrick et a1.[22], l-O-(l',2'-~H]Hexadecyl 2-1yso-sn-glycero-3-phosphocholine (56 Ci/mmol) was prepared from [~H]PAF (56 C i / m m o l ) [20]. Radiolabeled PAF and lys6-PAF yielded only one radiolabeled product when chromatographed on Silica Gel H developed in c h l o r o f o r m / m e t h a n o l / a c e t i c acid/" water, 50 : 25 : 6:1 (v/v). Cc,mmercially plepared thin-layer plates were from Bodman Chemicals (b~edia, PA), authentic phospholipid and neutral lipid stat, dards were from Serdary Research Laboratories (Ontat'!o, Canada). All solvents were HPLC grade from Fisher Scientific (King of Prussia, PA), and all other chemicals were analytical reagent grade from Fisher Scientific,
o
~
0
20
40 Zorn Number
""
60
-
I~
Fig. I. Zonal profile scan (2 mm) of products obtained by Vitride
reduction of PC and PE extracted from MEPM cells incubated with I-O.13H]alkyl-lyso-GPC.About 2. l06 cells were incubated f~r 24 h at 37°C in 3 rnl mediumwhich contained 0.5 p.Ci [3H]lyso-PAFand 0.1 ttCi [ S4C]arachidonicacid per ml. The cultureswere then washed, extracted according to Bligh and Dyer tiff] and the PC aud PE isolated by T! C and reduced with Vitride, as describe:l in Materials and Methods. The reduction products of the PC (n) and PE (e) were chromatographedon SilicaGel G TLC plates in water-saturated ether (dicthyl ether/water, 98: 2, v/v) alongwith authentic alkTIglycerol and alk-l-f:nylglyeerol.Data are representativeof three separate experiments.
195 TABLE [ Incorporation of ['~lt]lyso-PAF into phospholipid¢ of MEPM cells, in vitro a Length of incubation (h)
Percent distHb.tion ~' of 3H PC ¢ PS/PI
PE
I 24
92.5±0.7 d ~.3±0.9 e
5.2~1 14.7~0.9
1.9±0.1 I ±0.1
a Mouse embuo palate mesenchyme (MEPM) cells (about 7. l0 s) in the late log phase of growth were incubated at 37°C for either 1 h or 24 h in ! ml of medium which contained 0.5 p.Ci l-O-['~H]alkyl2-1yso-GPC (l'~Hllyso-PAF).The medium was then removed and the cells extracted according to a modification of the method of Bligh and Dyer [16]. 1% of the added label was recovered in cellular pbospholipids after incubation for I h whereas 7% was recovered in cellular phospholipids after the 24 h tw:,'iodof incubation. t, Based on total radiolabel recovered in cellular phospholipids. c Phuspholipids were separated by TLC in chloroform/methanol/ acetic acid/water, 50:25:8:2 (v/v). a Data represent the mean 5:1/2 the range of two separate determinations. Data represent the mean±standard deviation of three separate determinations.
by acid hydrolysis of the PC (Table II). Concomitantly, at least one-half the [t4C]AA incorporated into the P E fraction of M E P M cells was associated with the lyso-PE fraction produced on acid hydrolysis (Table 11). These results indicate that the 3H-labeled P E was predominantly 1-O-alk-l'-enyl-2-acyi-sn-glycero-3-phosphoe t h a n o l a m i n e (PE plasmalogen). O n the other hand, little of the [~H]alkyl moiety was metabolized to aik-1enyl groups in the PC. A T least 50% of the [t4C]AA incorporated into the P E of M E P M cells was esterified into P E plasmalogen. Vitride reduction of PC and P E extracted from M E P M cells which had been inct~bated for 24 h with 1-O-[3H]alkyl-2-1yso-PAF and [t4C]AA revealed e s ~ n tially only '~H alkyl moieties in the P(" and [3H]alk-lenyl moieties in the P E (Fig. 1). In no case were SH products found which chromatographed with the t4Calcohols [derived from [t4C]AA), indicating the ['~H]alkyl moieL~ was not metabolized to fatty acid b~ these embryonic ,:ells. W e pre:viously characterized the activities of phospholipase.,; A in homogenates of M E P M cells and founQ
,,.5
pH 9.0
"o° 'O E~'nOUb'ted "h "~ eo / .
I~
'~' "--'
0
"
PC
pH 4 : 5 . - - - o L ='~ neUad 4 ~ Incubated 2h ~0 I ~-~ |nco~,&ted 4h
'
IJPC
OE I,PE PE
PC
LPC
Cg
LPE Pg
°1 PC
pH '~,0
pH ?.5
~o
~.
o
r~', uPc uPz
pr. pc
FF& LPC UPE PI
PE PC
~ & LPC UPE P!
BE
PC
h d i o l l b e l u d Up|dl Fig. 2. Effect of the activities of endogenous phnspholipaseson the distribution of [ZH]lyso-PAF and [z4C]arachidonic acid amongst endogenous lipids of MEPM cells, Primak~, cultures of ,noose embryo palai,c mesencltyme ~MEPM) cells were radiolabelod for 24 h, homogenized, and incubated at 37°C for the indicated times and values of pH, as described in Materials and Methods. The reaction mixture contained
154042± 6489 DPM as 3H and 124135+ 5068 DPM as 14C. Lipids in the incubation mixtures were extracted into chloroform and separated b~ TLC, as described in Materials and Methods. Percent distribution = (DPM '.vh:ch chromatographod with designated authentic standard ± tot~:l DPM on the thin layer chromatograph)x lflO. Data arc the means of at least three separate determinations+one standard deviation. Values obtaiued from boiled controls incubated for either 2 or 4 h were not different and have been summed. GE, l-O-alkylglycerol;FFA, free fatty acid. Remaining abbreviations ate as in "Fables! an¢. H.
196 TABLE !1
I~corporation of [ ~HIlyswPAFa and [ ~C]arachidonic acid into p/asmalogen species of I'C apd PE of MEPM cells Parent compound
Products of actd hydrolysis~
Percent distribution b 3H ~4C
PC
aldehyde ly~,o-PC nc t hydrolysed
3+2 d 2+_ I 95 + I
4+_ I 4=!:3 92 +_4
PE
aldehyde lyso.PE
82 + 2 I =1:I 17 ± 3
10_+4 53 + 3 37 + l
not hy~rolvsed
" Primary cultures ot mouse embryo palate mesenchyme (MEPM) cells were incubated at 37°C in I ml of medium which contained 0.5 p.Ci I-O-[I',2'-3H]alkyl-2-1yso-GPC([3H]LPAF) and 0.1 /zCi [IJ4Clarachidonic acid, as described in Materials and Methods. The medium was removed after 24 h and the cells washed, fixed with a solution of methanol/water (2:1, v/v). scraped into test tubes and their lipids extracted as de~ribed in Materials and Methods. Approx. 7% of the added 3H and 42% of the added l l4C wece incorporated into cellular lipids during the incubation periot~. b Percent Distribution = (DPM in unhydrolyscd component, aldehyde, or lyso clcrivative/sum of the DPM in unhydrolyscd component, aldehyde, and lyso derivative)X 100. c Phospholipids were separated on EM 60 Silica Gel H TLC plates and the distribution of radiolabel into alkyl and alk-l-enyl moieties determined as described in Materials and Methods. '~ Data are the mean±one standard deviation obtained on analysis of four separate samples.
peaks of hydrolytic activity at p H 4.5, 7.5, and 9.0 [13]. In those studies hydrolysis of e n d o g e n o u s phospholipids was found to bc almost linear over a 6 h incubation period [13]. Therefore, M P E M cells radiolabeled with [3H]Iyso-PAF and [t4C]AA were homogenized and incubated 2 h or 4 h at e i t h e r pH 4.5, 7.5, or 9.0 to obtain a measure of hydrolysis of e n d o g e n o u s e t h e r lipids. T,~cre was an almost linear accumulation of 14Cfatty acid during ute 4 h incubation period at all values of pH (Fig. 2). Maximum accumulation of 14Cfatty acid c,:curred at pl-! 9.0, and m i n i m u m accumulation was at pH 4.5 (Fig. 2). l u c r e was a relatively greater decrease m [14C]PC than in [t4C]pE at all values of pH, and in all cases [14C]P! decreased to the least extent (Fig. 2). 14C-Labeled lyso-PC and lyso-PE accumulated at p H 4.5 only (Fig. 2). In contrast, l-O[3H]alkyl-2-acyl-sn-glycero-3-phosphocholine was hydrolyzed to yield [3H]lyso-PC at all values of pH (Fig. 2). T h e r e was also some accumulation of 3H products which c h r o m a t o g r a p h e d with authentic l-O-octadec9'-enyl-sn-gfycerol (glycer(d ether, GE; Fig. 2). The label, however, would be the 1 6 : 0 alkylglycerol (citimyl alcohol). The percent of the total ' H found as iyso-FC and alkylgly¢crol was i n w r s e l y related to the percentage of 3H lost from PC , t pH 4.5 (Fig. 2). A t pH 7.5 and 9.0 the accumulation of ['~H]lyso-PC and alkylglycerol was about o~.e-half that found at pH 4.5. This may
reflect the presence of transacylase activity, which could reacylate lyso-PAF as it is formed, since 15 + 3% ( ± S . D . , n = 4 ) of tl,e total [3H]Iyso-PAF (4 pmol) a d d e d to homogenates of unlabeled M E P M cells was incorporated into PC at p H 7.5 and 9.0; no radiolabel was incorporated into PC at pH 4.5. Radiolabeled a/achidonic acid was iiot incorporated into phospholipids when incubated with homogenates of M E P M cells u n d e r similar conditions [13]. Since M E P M cells actively incorporated lyso-PAF into l-O-all~,l-2-acyI-PC, the storage form of P A F [23], we explored the i~ossibility these embryonic cells might be able to synthesize P A F itself. Indeed, incubation of M E P M cells with the Ca 2+ ionophore A23187 and [3H]acetate resulted in production of a radiolabeled product which c h r o m a t o g r a p h e d w:.th authentic P A F and was resistant to hydroly*is with HCI fumes (data not shown). Digestion of thi~ putative [3H]PAF with phospholipase C and acetylation of the resultant diradylglycerol produced a radiolabeled derivative that e h r o m a t o g r a p h e d with a u t h e n t i c l-O-alkyl-2,3-diacetylglycerol (Fig. 3). Digestion of the [3H]PAF with the phospholipa~e A 2 of Crotalus adementeous venom released over 70% of the radiolabel into the aqueous phase, indicating the label was present as a 3H acetyl moiety.
~,o. Zone number: compound 15-16:acyldlace~)lyc~ ol 17-21:alkyldt~catylolyc~ol 23-2S:dtacylacely~yceml 26-30:aJk-I-enykcy~et l~ycerol
,¢00.
3ooo
2000
t0oo 0 ...... .......
j -I~
20 30 ~ 7onql l~rnl~h Fig. 3. Identification of radlolabelod l-O-a~kyl-2,3.diacely|glycerol
derived fro~ I -O.alkyl-2.l'~Hjacety\.GPC synthesized by MEFM ccl!:.
Cells (2. l0 ~') were incubaled wit;, [3HJacetate a.ld stimulated with A~3t87 (O) or vehicle (*), as d..'.scribed in MateGals and Methods. Chromatography of extraels of cells stimulated with A23187 resulted in 20P~ DPM recovered in tht; fraction which cochromatographed with authc.tic PAF, Ti~',s radiolabeled product w~s r~:oa':ered, digested, acetylated and analyzed by zonal TLC (5 mm s~ans) after scparatior~ by development in hexane/ethyl ether~acetic acid, 60:40: I, v/v), as described in Materials ~nd Methods. Separation ot V~l'iousacetylatod products was achieved with thi~ solvem system, as indict:ted in the figure.
197 Discussion
We have explored the synthesis of PAF and the metabolisms of related compounds by primary cultures of mouse embryo palate mesenchyme cells. Unlike differentiated polymorphonuclear leukocytes [24], these embryonic cells incorporated radiolabel into alkylacylPC and PE plasm.~logen. Incorporation of the radiolabeled alkyl moiety as the alk-l-enyl into PE plasmalogen suggests an active pathway is present that can transfer the alkylglycerol component from PC to PE where it can then be desatvrated to the plasmalogen (reviewed in Ref. 25). The synthesis of PE plasmaiogens by embryonic cells is consistent with previous findings that embryonic tissues contain significant amounts of PE plasmalogen [26]. Incubation of homogenates of prelabeled cells at acid pH resulted in *.he accumulation of 3H- and t4Clabeled lys3-PC. Accumulation of those products of hydrolysis indicates the presence of phospholipases A 2 and A t, r¢spectively, and is consistent with previous studies of these cells [13]. Most important, we were able to obtain a minimal measure of the activities of ~hospholipase A 2 in homogenates incubated at neutral and alkaline pH since there was accumulation of [3H]Iy,,~o-PC which was parallel to a loss of [~4C]AA from the PC fraction. This contrasts with our failure to find any radiolabeled lysophospholipids under similar conditlons when endogenous phospholipids of MEPM cells wcae radioiabeled in the sn-1 position with acyl nloietles [13]. These contrasting results caa he explained by the presence of a iysophosphelipase that hydrolyzes l-acyl-linked lysophospholipids, hut cannot hydrolyze the ether-linked species. It would appear that PE plasutalogens are either not hydroly:,ed by the cellular phospholipase A 2 or there is rapid r¢ acyl~tion of l-O-alk-l'-enyl-2-1yso-PE under the conditions of our assay since (i) 50% of the [14C]araehidonate and essentially all of the 3H alkyl moiety incorporated into PE was in PE plasmalogen, (i0 ~:C-tatty acid was lost from PE during the course of the assay but there was neither accumulation of [3H]lyso-PE nor loss of [3H]PE during; I.he course of the assay. During the course of the assay there was accumulation of a 3H-labeled compound which migrated with l-O-alk3,1glycerol. This finding may reflect sequential hydrolysis of choline phospholipids by phospholipase C and diglyceride lipase, or phospholipase D activity in concert with a phospbatidic acid phosphatase and a lipase. Phospholipase~ C and D which hydrolyze 1-Oalkyl-2-acyl-sn-glycero-3-phosphocholine have been reported in various types of cells recently [27,28]. We found MEPM ceils have the ability to synthesize PAF and its immediate precursor, l-O-alkyl-2-acyi-PC (by acylation nf lyso.PAF). In turn, O'Neill [29-31] has reported recent'y on the ability of the very early era-
bryo (fertilized ooc~,te) to produce PAF, and raised the possibility that it :nay serve as an early embryonic message. Our finding that cells obtained from a discrete embryonic structure (the palate) are able to synthesize PAF raises the possibility this biologically 13otent compound might serve as an autocrine in regulating certain aspects of differentiation. Many of the responses of differentiated cells to PAF, such as migration, adhesion, secretion, and uptake of Ca 2÷ (reviewed in ReL 32) are processes which are also associated with differentiation of embryonic cells and tissues. In summary, we found that: (i) embryonic cells are able to incorporate lyso-PAF into l-O-alkyl-2-acs,I-PC (the storage form of PAF) and ethanolamine p!asmalogen, (it) endogenous, double-labeled 1-O-aikyl-2-acylPC serves as a substrate for intracellular phospholipase A2, with subsequent accumulation of radiolabeled lyso-PAF, and (iii) embryonic palatal cells are able to synthesize PAF, thus prompting the notion this effector molecule may play a role in regulating growth a n d / o r differentiation of embryonic tissues. Acknowledgement This work was supported by N.I.H. grant Nos. DE 06553, HL 26818, and AI 17287. ~eferences 1 Chepenik, K.P., Ho, W.C., Waite, ~.M, and Parker, C.L. {1984) Calcif. Tissue. tat. 36, 175-181. 2 Gay, S.W. and Kosher, R.A. (1984) J. Exp. Zool. 232, 317-326. 3 Pratt, R.M., Kim, C.S. and Grove, R.L (1984) in Current Topics in DevelopmentalBiology(Zimmerman,LF., ed.), Vol. 19, pp. 81-10t, AcademicPress, New York. 4 Crutchley,D.J., Ryan, U,S, and Ryan, J.W. (1985) J. PharmacoL Exp. Thor, 233(3), 650-655. 5 Maridonneau-Parini,i., Errasfa, M. and Russo-Marie,F. (1989) J. Clin. Invest,83, 1936-1940, 6 Rothhut, B., Russo-Marie,F., Wood, J., DiRosa, M. and Flower, RJ. (1983) Biochem.Biophys.Res. Commun. 117,878-884. 7 Flower, RJ. and Blackwen,G.J. (1979) Nature 278. 456-459. 8 Tzortzatou, G.G., Goldman, A.S. and BoutwelL W.C, (1¢81) Prec. Soc. Exp. Biol, Med. 166,321-324. 9 Piddington,R., Herold, R. and Goldman, A.S. (1983}Prec., Soc, Exp. BioL Med, 174,336-342, 10 Goldman, A,S,, Baker, M.K,, Piddington, R. and Herold, R, (1983) Proc. :~.oc.Exp, Biol. Med. 174, 239-243. I1 Montenegro,M,A. and Paz de La Vega, Y. (1982) Arch. Oral. Biol. 27, 771-775. 12 Chepenik,K.P, George, M. and Greene, R.M. (1985)Teratology 32,119-123. 13 Geroge, M. and Chepeaik, K.P. (1985) Biochlm. Biophys. Acts .~36.45- 55, 14 Wykle,R.L. and Snyder, F. (1976) in The Enzymesof Biological Membranes (Martonosi, A., ed.), Vol. 2, pp. 87-117, Plenum PpJblishing,New York. 15 Chabot, M.C. and Chepenik K.P. (1986) J. Craniofac. Genet. Dev. Biol. 6, 223-234. 16 Bligh. E.G, and Dyer, W.J, (1959) Can. J, Biochem Physiol.37. 911-917.
1.98 17 Esko, LD. and Raetz, C.R.H. (19~0} J. Biol, Chore, ~ 5 , ~ 7 4 ~80. 18 S-~'a.~-~. r:., Blank, M. and Wykle, R.L. (1971) J. Biol. Chem. 246, 3b.; ~ -3645. 19 Waku, K., !1o, H., Bito, T. and Nakazawa, v . (!o7a) 1. ~inrhe::t. 75, 1307-1312. 20 Mueller, H.W., O'Flahaerty, J.T. and W~/k'.e.R.L. (1983) J. Biol. Chem. 258, 6213-6218. 21 Emilsson, A. and Sundlcr, R. (1985) Biochim. Biophys. Acta 816. 265-274. 22 Wyrick, S.D,, McClanaha,, J.S., Wykle, R.L. and O'Flaherty, J.T. (1£85) J. Labeled. Comod. Radiopharm. 22, 1169-1174. 23 Snyder, F. (1988) in Biological Membrane~: Aberrations in Membrane Structure and Function, pp. 57-.72. Alan R. li~s. New York.
24 Chilton, F.H., O'Flahcr,"j, 3iF., Ellis, 3.M,, Swei*ds~a, C.L, and Wykle, R,L. (1983)J. BioL Chem. 258, 6357-6361. 25 Snyder, F., Lee. T-C. and Wykle, R.L. (1985) in The Enzymes of Biological Membranes (Martone3i, ,~..N,, cd.), Voi. 2, pp. 1-58, Plenum Publishing, New York. 26 Chepenik, K.P., Blank, M., Snyder, F. and Waite, B.M. (1980) int. J. Biochem. !1,605-607. 27 Exton, J.H. (1990) 3. Biol. Chem. 265, I-4. 28 Exton, J.H. (198~) FA~;EB J. 2, 2670-2676. 29 O'Neill, C. (198:~) .L Reprod. Fort. 75, 3'75-~t~U. 30 O'Neill, C. (1985)J. Reprod. Ferl. 73, 559-566. 31 O'Neill, C. (1985; J. Reprod. Fen. 73, 567-677. 32 Lee, T-C. and Snyder, F, (1985) in Phospholipids and Cellular Activation (Kuo, J.F., ed.), Vol. 2, pp, l-39, CRC Press, Boca Raton.
