Molecular heterogeneity of PAF in normal human mixed saliva: quantitative mass spectral analysis after direct derivatization of PAF with pentafluorobenzoic anhydride.

Platelet-activating factor (PAF), a family of phospholipid autacoids with potent pro-inflammatory activities, is present in saliva. The current study has quantitated various species of PAF isolated from normal human mixed saliva. Choline-containing, sn-2 acetylated phospholipids with sn-1 ether- or ester-linked fatty alcohol/acid moieties (alkyl-PAF or acyl-PAF, respectively) were evaluated after direct derivatization with pentafluorobenzoic (PFB) anhydride. Individual species of PFB-derivatized PAF were separated by gas chromatography prior to mass spectral analysis; quantitative estimates of six different species of PAF in saliva were made by comparison to corresponding authentic, synthetic PAF standards. In each saliva sample, all six species of PAF were readily detected by this facile procedure. The predominant PAF was 1-O-hexadecyl-2-acetyl-sn-glycero-3-phosphocholine or 16:0-alkyl-PAF (0.75 +/- 0.09 pmol/ml saliva; mean +/- S.E.; n = 5) which represented only 30.4 +/- 1.5% of the total PAF. Substantial amounts of 18:1- and 18:0-alkyl-PAF and 16:0-acyl-PAF were also identified (0.52 +/- 0.07, 0.35 +/- 0.06, and 0.35 +/- 0.02 pmol/ml saliva, respectively). In summary, mass spectrometric analysis of PAF after direct derivatization with PFB anhydride has revealed that at least six different species of PAF are present in normal human mixed saliva. This structural diversity may represent an important aspect of homeostasis in the healthy oral cavity.

Lipid inhibitors of platelet-activating factor (PAF) in normal human plasma.

Endogenous, human plasma-derived lipids that inhibit the platelet stimulating activity of platelet-activating factor (PAF) have been identified. Chromatographic fractionation of neutral lipid PAF inhibitors revealed a majority of PAF inhibitory activity comigrating with cholesterol and a second peak localized with free fatty acids. Plasma phospholipids demonstrated three distinct PAF inhibitory fractions in TLC regions corresponding to those of sphingomyelin, phosphatidylcholine and phosphatidylethanolamine. Three fractions (one neutral lipid and two phospholipid) specifically inhibited PAF-induced platelet activation. Thus, there are both specific and non-specific lipid inhibitors of PAF in normal human plasma. These plasma lipids may be important in the specific regulation of the diverse, potent biological activities of PAF in various physiological states.

The effect of initial periodontal therapy on salivary platelet-activating factor levels in chronic adult periodontitis.

Platelet-activating factor (PAF), a potent phospholipid inflammatory mediator, is increased in the mixed saliva of subjects with periodontal disease and correlates with the extent of oral inflammation. The present study was designed to provide a longitudinal evaluation of the effect of initial periodontal therapy (home care instruction, prophylaxis, and scaling/root planing) on salivary PAF levels in chronic adult periodontitis patients (n = 15). Mixed saliva was collected prior to, during, and after initial therapy and was utilized to assess PAF levels after lipid extraction and fractionation as well as to histologically assess the number of polymorphonuclear leukocytes (PMN). PAF activity was determined in bioassay relative to authentic PAF (1-O-hexadecyl-2-acetyl-sn-glycero-3-phosphocholine; 16:0-alkyl-PAF). Initial salivary PAF levels (12.1 +/- 2.8 pmole equivalents of 16:0-alkyl-PAF/ml saliva; mean +/- SE) decreased following supragingival plaque control (9.6 +/- 2.4) and were further reduced following scaling and root planing (5.7 +/- 1.4). In parallel, salivary PMN levels were significantly reduced and clinical estimates of periodontal disease were significantly improved; i.e., there was a decrease in the percentage of sites with both bleeding on probing (from 46.1 +/- 4.6% of sites at pretreatment to 25.9 +/- 2.6% after scaling and root planing) and probing depths > or = 4 mm (from 16.7 +/- 1.9% of sites to 10.3 +/- 1.2%). Thus, initial periodontal therapy reduced salivary PAF levels in concert with improvements in clinical estimates of marginal and submarginal periodontal inflammation suggesting that PAF may participate in inflammatory events during periodontal tissue injury and disease.

PAF molecular heterogeneity: pathobiological implications.

The existence and potential (patho)physiologic significance of PAF molecular heterogeneity can no longer be summarily dismissed or ignored. While significant advances in the chemistry of PAF have been made, the (patho)physiologic behaviors of most of the PAF molecular species of biologic origin await further study. This is because to date, investigators have studied the biologic activities of what was previously thought to be PAF, i.e., only 16:0- and 18:0-AGEPC. In view of the evidence presented in this review, a comprehensive investigation of the possible biologic relevance and significance of PAF molecular heterogeneity is warranted. Hopefully, such studies will be designed to elucidate the extent to which the various molecular species of PAF differ in their intrinsic in vitro and in vivo (patho)physiologic behaviors (agonistic, synergistic, and possibly antagonistic) and modes of action. Once this additional information has been derived, the pathobiologic relevance of this class of phospholipid autacoid may be better understood.

Differential responsiveness of human neutrophils to the autocrine actions of 1-O-alkyl-homologs and 1-acyl analogs of platelet-activating factor.

The phlogistic actions of six molecular species of platelet-activating factor (PAF) (1-O-alkyl-PAF homologs, 16:0-, 18:0- and 18:1-alkyl-PAF, 1-O-alkyl-2-acetyl-sn-glycero-3-phosphocholine (AGEPC) and their respective 1-acyl-PAF analog counterparts, 16:0-, 18:0- and 18:1-acyl-PAF, 1-acyl-2-acetyl-sn-glycero-3-phosphocholine (AGPC)) were assessed relative to five human neutrophilic polymorphonuclear leukocyte (PMN) functional responses: 1) lysosomal enzyme secretion; 2) specific desensitization to 16:0-AGEPC-induced lysosomal enzyme secretion; 3) O2- production; 4) chemotaxis; and 5) priming for enhanced O2- production. With respect to inducing lysozyme secretion, 18:0-AGEPC was 30- and 75-fold less potent than 16:0-AGEPC and 18:1-AGEPC, respectively, and was 25- and 40-fold less potent for inducing beta-glucuronidase secretion. 18:0-AGEPC was also 10-fold less active than 18:1- and 16:0-AGEPC for inducing O2- production. Thus, the rank order of potency of the alkyl-PAF homologs for inducing both lysosomal enzyme secretion and O2- production was 18:1- greater than or equal to 16:0- much greater than 18:0-AGEPC. In contrast, these three alkyl-PAF homologs had the same potency for desensitizing PMN to subsequent 16:0-AGEPC-induced lysosomal enzyme secretion and for priming PMN for augmented O2- production in response to FMLP or human recombinant C5a. Paradoxically, however, the rank order of potency of the alkyl-PAF homologs for effecting PMN chemotaxis was 18:0- greater than 18:1- much greater than 16:0-AGEPC. At concentrations as high as 1.0 microM, the acyl-PAF analogs did not initiate PMN lysosomal enzyme secretion, O2- production, or chemotaxis. However, the acyl-PAF analogs induced partial PMN desensitization to 16:0-AGEPC. A novel finding of potential (patho)-physiologic significance was the ability of acyl-PAF at nM concentrations to prime PMN for significantly enhanced O2- production after stimulation with FMLP or human recombinant C5a. The priming action of acyl-PAF was due to an increase in the rate as opposed to a prolongation of O2- production. The differing rank orders of potency of the alkyl-PAF homologs and acyl-PAF analogs for stimulating several physiologic responses of the same target cell, the human PMN, support the premise that there may be more than one PAF receptor subtype on the PMN and/or that differences in the biophysical properties of the various molecular species of PAF modulate their interaction with PAF receptor(s) linked to stimulus-response coupling.

A rabbit model for bacteria-induced preterm pregnancy loss.

Bacterial infection has been implicated in premature labor in humans. To elucidate mechanisms and potential intervention strategies, we sought to develop a model of infection-induced pregnancy loss in rabbits. On day 21 (70% of gestation), each uterine horn was inoculated hysteroscopically with 0.2 ml containing saline solution of 10(6) cfu Escherichia coli or Bacteroides bivius or Fusobacterium necrophorum. Fetal viability was assessed. Animals were sacrificed at various times or as delivery occurred. Serum progesterone and amniotic fluid prostaglandins were measured. Cultures and histologic sections were prepared. Compared with the saline solution group, E coli and F. necrophorum-inoculated rabbits were significantly more likely to deliver (16 of 16 and six of seven with mean times of 31.9 +/- 10.7 and 28.3 +/- 11.5 hours, respectively for E. coli and F. necrophorum). Positive amniotic fluid cultures for the E. coli group were found in 11 of 12 (92%) and for the F. necrophorum group in three of three cases (100%). Histologic inflammation was seen heavily in both the E. coli and F. necrophorum groups, whereas it was absent in the saline solution group. Inoculation with B. bivius led to a much lower pregnancy loss rate (eight of 32) and less histologic inflammation despite positive uterine cultures in most animals. This model may provide an opportunity to determine mechanisms of clinical or subclinical intraamniotic infection and to test intervention strategies.

Decrease in ovarian platelet-activating factor during ovulation in the gonadotropin-primed immature rat.

Platelet-activating factor (PAF) is a biologically active phospholipid that is released locally during acute inflammatory reactions and tissue injury. Since there is evidence that the biochemical events of mammalian ovulation resemble an inflammatory reaction, the objective of this study was to determine whether ovarian levels of PAF change during ovulation. At 2-h intervals during the ovulatory process in gonadotropin-primed 25-day-old Wistar rats, the ovaries were extirpated, homogenized, and extracted for lipids. The extracts were subjected to thin-layer chromatography (TLC), and the portion of the silica gel that comigrated with PAF was re-extracted and assayed for PAF activity. The PAF was measured (in fmole equivalents of synthetic PAF) by a bioassay based on the capacity of aliquots of the extracts to release [3H]-serotonin from platelets isolated from whole blood of rabbits and prelabeled with [3H]-serotonin. The ovarian level of PAF decreased (p less than 0.01) by 36% from 6.67 +/- 0.77 to 4.27 +/- 0.45 fmoles/mg ovary by 2 h after treatment with human chorionic gonadotropin (hCG), and it declined another 14% by 4 h after hCG. The ovarian PAF remained at this reduced level for up to 24 h after hCG. The administration of indomethacin (5 mg/rat, s.c.) or epostane (5 mg/rat, s.c.) at 1 h after hCG prevented ovulation, but neither drug affected the decline in ovarian PAF. Preliminary tests showed that the lipid extracts from the ovaries also contained PAF inhibitor(s) that comigrated with PAF on the TLC plates. Similar to PAF, the lipid-soluble inhibitor(s) decreased (p less than 0.05) in the ovaries within 4 h after hCG treatment.(ABSTRACT TRUNCATED AT 250 WORDS)

Prostaglandin synthesis by day-6 rabbit blastocysts in vitro.

Day-6 rabbit blastocysts were recovered from superovulated donor animals, washed in ice-cold Krebs-Ringer-bicarbonate (KRB) buffer, pooled and randomly allocated to polypropylene incubation tubes, usually 10 blastocysts in 1 ml KRB. The blastocysts were ruptured with a dissecting needle and incubated at 37 degrees C for periods of 1-3 h with 10 microCi [3H]arachidonic acid/tube. A control tube without blastocysts was run in each experiment. At the end of the incubation, the samples were acidified, extracted with ethyl acetate, dried down and resuspended in h.p.l.c., using a solvent system for prostaglandins (PGs), was subtracted from each experimental run in the same experiment. The remaining radioactivity constituted 0.14% of the original [3H]arachidonic acid added to each incubation tube. This was considered to have been the result of conversion of the radiolabelled arachidonic acid to prostanoids. In the absence of 10 mM-EDTA no conversion occurred, whereas in its presence peaks of radioactivity co-eluting with [3H]PGF-2 alpha and [3H]PGE-2 were seen. A third peak that eluted was either 15-keto metabolites of these PGs or PGD-2. These 3 peaks were always significantly above background, and usually did not differ from each other. No differences in amount of conversion could be related to incubation time. Addition of indomethacin (100 micrograms/ml) or radioinert arachidonic acid (10 micrograms/ml) inhibited production of [3H]PG, even in the presence of EDTA. Removal of calcium from the incubation medium was per se without effect. Addition of atropine (0.15 mM) or carbachol (0.15 mM) in the presence or absence of EDTA did not change the pattern of conversion of [3H]arachidonic acid to [3H]PGs. These experiments demonstrate that rabbit blastocysts have the capacity for de-novo synthesis of PGs from exogenous substrate, when utilization of endogenous substrate is inhibited. The extent of conversion observed may not be a true reflection of the capacity for conversion of endogenous substrate.

Spermicidal effect of antagonists of platelet-activating factor.

A variety of antagonists of platelet-activating factor (PAF) have been examined for their ability to prevent pregnancy, by admixture with spermatozoa in vitro followed by insemination, or in vivo by administration to female rabbits. When the antagonists were added to the ejaculate at a concentration of 10(-4) M 30 minutes before insemination of the females, a significant failure of fertilization was seen only with CV-3988, U66985, and SRI 63-441--all structural analogs of PAF. Antagonists that were not structural analogs were not effective. All compounds to a lesser or greater extent caused some qualitative agglutination and loss of sperm motility, but these effects were not correlated with inhibition of fertilization. SRI 63-441 was the most effective compound and was subjected to further study. When given intravenously prior to ovulation (5 mg/kg at 1, 5, and 9 hours after the ovulating injection), no effect on fertilization was seen. If SRI 63-441 (40 mg in 0.5 ml aqueous solution) was instilled into the vagina 2 minutes before insemination, a highly significant reduction in the fertilization rate was achieved. It is concluded that these compounds act by an action on the sperm membrane rather than by direct PAF antagonism on spermatozoa.

The final step in the de novo biosynthesis of platelet-activating factor. Properties of a unique CDP-choline:1-alkyl-2-acetyl-sn-glycerol choline-phosphotransferase in microsomes from the renal inner medulla of rats.

Final steps in the synthesis of platelet activating factor (PAF) occur via two enzymatic reactions: the acetylation of 1-alkyl-2-lyso-sn-glycero-3-phosphocholine by a specific acetyltransferase or the transfer of the phosphocholine base group from CDP-choline to 1-alkyl-2-acetyl-sn-glycerol by a dithiothreitol (DTT)-insensitive cholinephosphotransferase. Our studies demonstrate that rat kidney inner medulla microsomes synthesize PAF primarily via the DTT-insensitive cholinephosphotransferase since the specific activity of this enzyme is greater than 100-fold higher than the acetyltransferase. The two cholinephosphotransferases that catalyze the biosynthesis of phosphatidylcholine and PAF have similar Mg2+ or Mn2+ requirements and are inhibited by Ca2+. Also topographic experiments indicated that both activities are located on the cytoplasmic face of microsomal vesicles. PAF synthesis was slightly stimulated by 10 mM DTT, whereas the enzymatic synthesis of phosphatidylcholine was inhibited greater than 95% under the same conditions. The concept of two separate enzymes for PAF and phosphatidylcholine synthesis is further substantiated by the differences in the two microsomal cholinephosphotransferase activities with respect to pH optima, substrate specificities, and their sensitivities to temperature, deoxycholate, or ethanol. Study of the substrate specificities of the DTT-insensitive cholinephosphotransferase showed that the enzyme prefers a lipid substrate with 16:0 or 18:1 sn-1-alkyl chains. Short chain esters at the sn-2 position (acetate or propionate) are utilized by the DTT-insensitive cholinephosphotransferase, but analogs with acetamide or methoxy substituents at the sn-2 position are not substrates. Also, CDP-choline is the preferred water-soluble substrate when compared to CDP-ethanolamine. Utilization of endogenous neutral lipids as a substrate by the DTT-insensitive cholinephosphotransferase demonstrated that sufficient levels of alkylacetylglycerols are normally present in rat kidney microsomes to permit the synthesis of physiological quantities of PAF. These data suggest the renal DTT-insensitive cholinephosphotransferase could be a potentially important enzyme in the regulation of systemic blood pressure.