Fine structural aspects of phagocytotic activity in inverted follicle cells of cultured porcine thyroids

Inversion of thyroid follicles took place when they were isolated by collagenase and trypsin and cultured in suspension in Eagle's medium supplemented with 10% fetal calf serum without TSH. The apical surface facing the culture medium contained numerous microvilli and a central cilium, while the luminal surface became flattened. Phagocytotic activity by pseudopods was promoted after addition of TSH to the culture medium. When the inverted follicles were incubated in culture medium containing TSH (50 mU/ml) and human red blood cells, or TSH and polystyrene latex beads (2.02 micron in diameter) for 1-3 h, numerous red blood cells or latex beads respectively were observed to be taken up by the epithelial follicle cells by scanning electron microscopy, as well as conventional thin-section electron microscopy. These results show that the apical surface (culture medium side) of the epithelial cell of the cultured thyroid follicle whose polarity is reversed phagocytoses red blood cells and latex beads.     

Release of granule proteins from eosinophils cultured with IL-5.

Eosinophils isolated from normal individuals were cultured in the presence of human rIL-5 (hrIL-5) for up to 14 days, and the effects of this exposure were determined. First, the hrIL-5-cultured eosinophils were activated and degranulated more readily than freshly isolated eosinophils. For example, eosinophils cultured for 7 days with hrIL-5 released 30 and 10% of granule eosinophil-derived neurotoxin (EDN) when exposed to Sepharose 4B beads coupled to secretory IgA and IgG, respectively, whereas freshly isolated eosinophils released only 19 and 4%, respectively, of their EDN in response to the same stimuli. Degranulation of hrIL-5-cultured eosinophils was not augmented by further exposure to hrIL-5, whereas degranulation of freshly isolated cells to secretory IgA and IgG beads was increased by exposure to hrIL-5. Second, eosinophils cultured with hrIL-5 had prolonged viability in vitro. For example, after four days of culture with 50 U/ml of hrIL-5, 86% of eosinophils were viable compared to 12% in medium alone. Third, hrIL-5-cultured eosinophils became hypodense, and electron microscopy showed that they contained granules with core and matrix lucency and with evidence of granule fusion. Fourth, hrIL-5-cultured eosinophils spontaneously lost 30 to 60% of their EDN, eosinophil cationic protein, and eosinophil peroxidase and about 50% of their eosinophil granule major basic protein content compared to freshly isolated eosinophils, and all four of the granule proteins were released into the culture medium. Fifth, detailed studies of eosinophils cultured in hrIL-5 showed that 89 +/- 10% of the starting quantity of EDN could be recovered at 7 days. Whereas 99 +/- 1% of the EDN at day 0 was cell associated, by 7 days 60 +/- 9% was in the cell supernatants. Thus, hrIL-5 activates eosinophils, increases their viability, decreases their density, and their content of granule proteins and causes release of the granule proteins into culture fluids. The striking loss of granule proteins during culture with hrIL-5 may be an important mechanism for deposition of these cationic toxins in various diseases where IL-5 plays a role.     

Uptake of HIV and latex particles by fresh and cultured dendritic cells and monocytes

Blood dendritic cells (DC) efficiently carry HIV-1 and transmit infection to CD4+ T cells in the absence of productive infection of the APC. Fluorescent latex beads were used to define the endocytic pathways that may contribute to this non-infectious pathway of virus carriage. Beads between 14 nm and 2300 nm in diameter were taken up by uncultured blood DC, but uptake of beads larger than 280 nm was much reduced in the DC compared to monocytes. After culture, there was a reduction in bead carriage in DC compared to monocytes. In the DC, beads were found as small aggregates in class II containing compartments or as single beads just below the cell surface. Beads accumulated in monocytes as aggregates in class II negative compartments. Bead recycling occurred in DC, but not in the fresh or cultured monocytes. Electron microscopy of HIV-1-pulsed DC cultured with CD4+ T cells showed accumulation of apoptotic debris and virions within endosomes in the DC. The peripheral location and recycling of endocytosed material in DC provides a pathway for virion transfer from DC to T cells that does not occur in monocytes.     

Activated peripheral T lymphocytes undergo apoptosis when cultured with monocytes activated by HLA class II ligation

We treated PBMC with anti-MHC class II mAb known to inhibit T lymphocyte proliferation. Adherent cells from mAb-treated PBMC showed increased metabolic activity by the MTS assay that was not due to cell proliferation. PBMC cultured with solid-phase anti-class II mAb in chamber inserts inhibited, across a membrane, the proliferation of PBMC cultured with soluble anti-CD3 mAb. PBMC treated with both soluble mAb underwent apoptosis as shown by nucleosomal DNA fragmentation. The monocytes formed multinucleated giant cells as shown by fluorescent microscopy, and contained apoptotic bodies as shown by the TUNEL method and by electron microscopy. The apoptotic cells were identified as T cells by double-staining with anti-CD4/CD8-PE and annexin-V-FITC. Thus, MHC class II ligation stimulates monocytes to increase their metabolic activity, induce apoptosis of activated T lymphocytes, and phagocytize the apoptotic cells. TCR-mediated ligation of MHC class II may play a role in the downregulation of immune responses.     

Chlamydia pneumoniae multiply in neutrophil granulocytes and delay their spontaneous apoptosis

The obligate intracellular bacterial pathogen Chlamydia pneumoniae (Cp) is responsible for a range of human diseases, including acute respiratory infection. Although experimental intratracheal infection with Cp results in a massive recruitment of neutrophil granulocytes (polymorphonuclear neutrophils (PMN)), the role of these cells in the defense against Cp is unclear. In this study the interactions of PMN with Cp were investigated. In vitro coincubation experiments showed that human granulocytes were able to internalize Chlamydia in an opsonin-independent manner. Importantly, phagocytosed Cp were not killed; the ingested bacteria survived and multiplied within PMN. Although uninfected granulocytes became apoptotic within 10 h, infected PMN survived up to 90 h. Coincubation with Cp significantly decreased the ratio of apoptotic PMN, as detected by morphological analysis, annexin V, and TUNEL staining. The observed antiapoptotic effect was associated with a markedly lower level of procaspase-3 processing and, consequently, reduced caspase-3 activity in infected PMN. LPS was found as a major, but not exclusive, component responsible for the observed antiapoptotic effect. Chlamydia LPS affected PMN apoptosis both by acting directly on the cells and by inducing the autocrine production of the antiapoptotic cytokine IL-8. These data show that, in contrast to other microbial pathogens that drive phagocytes into apoptosis to escape killing, Cp can extend the life span of neutrophil granulocytes, making them suitable host cells for survival and multiplication within the first hours/days after infection.     

Annexin-1 and peptide derivatives are released by apoptotic cells and stimulate phagocytosis of apoptotic neutrophils by macrophages

The resolution of inflammation is a dynamically regulated process that may be subverted in many pathological conditions. Macrophage (Mphi) phagocytic clearance of apoptotic leukocytes plays an important role in the resolution of inflammation as this process prevents the exposure of tissues at the inflammatory site to the noxious contents of lytic cells. It is increasingly appreciated that endogenously produced mediators, such as lipoxins, act as potent regulators (nanomolar range) of the phagocytic clearance of apoptotic cells. In this study, we have investigated the intriguing possibility that apoptotic cells release signals that promote their clearance by phagocytes. We report that conditioned medium from apoptotic human polymorphonuclear neutrophils (PMN), Jurkat T lymphocytes, and human mesangial cells promote phagocytosis of apoptotic PMN by Mphi and THP-1 cells differentiated to a Mphi-like phenotype. This prophagocytic activity appears to be dose dependent, sensitive to the caspase inhibitor zVAD-fmk, and is associated with actin rearrangement and release of TGF-beta1, but not IL-8. The prophagocytic effect can be blocked by the formyl peptide receptor antagonist Boc2, suggesting that the prophagocytic factor(s) may interact with the lipoxin A(4) receptor, FPRL-1. Using nanoelectrospray liquid chromatography mass spectrometry and immunodepletion and immunoneutralization studies, we have ascertained that annexin-1 and peptide derivatives are putative prophagocytic factors released by apoptotic cells that promote phagocytosis of apoptotic PMN by M[phi] and differentiated THP-1 cells. These data highlight the role of annexin-1 and peptide derivatives in promoting the resolution of inflammation and expand on the therapeutic anti-inflammatory potential of annexin-1.     

Intervertebral disc cells as competent phagocytes in vitro: implications for cell death in disc degeneration

Introduction

Apoptosis has been reported to occur in the intervertebral disc. Elsewhere in the body, apoptotic cells are cleared from the system via phagocytosis by committed phagocytes such as macrophages, reducing the chance of subsequent inflammation. These cells, however, are not normally present in the disc. We investigated whether disc cells themselves can be induced to become phagocytic and so have the ability to ingest and remove apoptotic disc cells, minimising the damage to their environment.

Method

Bovine nucleus pulposus cells from caudal intervertebral discs were grown in culture and exposed to both latex particles (which are ingested by committed phagocytes) and apoptotic cells. Their response was monitored via microscopy, including both fluorescent and video microscopy, and compared with that seen by cell lines of monocytes/macrophages (THP-1 and J774 cells), considered to be committed phagocytes, in addition to a nonmacrophage cell line (L929 fibroblasts). Immunostaining for the monocyte/macrophage marker, CD68, was also carried out.

Results

Disc cells were able to ingest latex beads at least as efficiently, if not more so, than phagocytic THP-1 and J774 cells. Disc cells ingested a greater number of beads per cell than the committed phagocytes in a similar time scale. In addition, disc cells were able to ingest apoptotic cells when cocultured in monolayer with a UV-treated population of HeLa cells. Apoptotic disc cells, in turn, were able to stimulate phagocytosis by the committed macrophages. CD68 immunostaining was strong for THP-1 cells but negligible for disc cells, even those that had ingested beads.

Conclusion

In this study, we have shown that intervertebral disc cells are capable of behaving as competent phagocytes (that is, ingesting latex beads) and apoptotic cells. In terms of number of particles, they ingest more than the monocyte/macrophage cells, possibly due to their greater size. The fact that disc cells clearly can undergo phagocytosis has implications for the intervertebral disc in vivo. Here, where cell death is reported to be common yet there is normally no easy access to a macrophage population, the endogenous disc cells may be encouraged to undergo phagocytosis (for example, of neighbouring cells within cell clusters).     

Human and Murine High Endothelial Venule Cells Phagocytose Apoptotic Leukocytes

Apoptotic cell death occurs during normal lymphocyte development and differentiation as well as following lymphocyte exposure to endogenous corticosteroids released during stress, malnutrition, and trauma. Recognition and engulfment of these apoptotic cells is important for the clearance of dying cells before they release potent inflammatory mediators into the vasculature or tissues. Phagocytosis of apoptotic cells is accomplished in part by macrophages. We report for the first time that apoptotic lymphocytes are also phagocytosed by high endothelial venule (HEV) cells. The murine HEV cell line mHEVa rapidly phagocytosed apoptotic lymphoid and myeloid cells with the greatest rate of phagocytosis occurring at 0–6 h. To confirm HEV cell interaction with apoptotic cells, we demonstrated that apoptotic human tonsil lymphocytes were phagocytosed by human tonsil HEV cells in primary cultures. Furthermore, we examined HEV cell phagocytosisin vivo.Mice were treated with a natural corticosterone (4-pregnene-11β,21-diol-3,20-dione) at levels detected during stress or malnutrition (93–180 μg serum cortisol/dl). At 4–12 h posttreatment, apoptotic lymphocytes were present inside vacuoles of HEV cells in axillary lymph node tissue sections, as determined by transmission electron microscopy. These data suggest that, in addition to macrophages, lymph node HEV cells also play a role in the removal of apoptotic lymphocytes. Moreover, since HEV cells are specialized endothelial cells that regulate lymphocyte migration into peripheral lymphoid tissues, they may provide an important checkpoint for clearance of apoptotic lymphocytes within the vasculature, as well as limiting entrance of nonfunctional lymphocytes into the lymph node.     

Early apoptosis-related changes triggered by HSV-1 in individual neuronlike cells

Early events of apoptosis following HSV-1 infection were investigated at the single-cell level using intensified fluorescence digital-imaging microscopy. The results provide evidence that infection of differentiated ND7 neuronlike cells by HSV-1 triggers detectable alterations indicative of physiological changes associated with the early stages of apoptosis. Less than 1 h after infection with HSV-1 (KOS strain) or K26GFP (GFP being fused to HSV-1 capsid protein VP26) we observed (i) moderate decrease in mitochondrial membrane potential (about 20%), (ii) exposure of phosphatidyl serine, (iii) morphological change in the mitochondria that became spherical instead of filamentous, and (iv) activation of caspase-8. Within 3 h changes reverted to normal, which indicated that apoptosis was counteracted very early following HSV-1 infection. Similar results were obtained with KOS-TK27GFP, lacking TK and UL24 proteins, suggesting that TK and UL24 play no role in apoptosis. In Vero cells mitochondrial changes characteristic of the apoptotic process were not observed following HSV-1 infection. The UV-inactivated K26GFP had the capacity to induce apoptosis in neuronlike cells. This real-time multiparametric analysis, in combination with relevant viral mutants, could be a useful approach for dissecting the roles of various viral genes in modulating apoptotic pathways during infection.     

Intercellular transfer of apoptotic signals via electrofusion

We determined whether cells that are induced to undergo anoikis by matrix detachment can initiate apoptosis in healthy cells following electroporation-induced fusion. Separate populations of MDCK cells undergoing anoikis and stained with FITC-annexin or viable MDCK cells that were labeled with spectrally discrete fluorescent beads were electroporated. Cells were analyzed by flow cytometry for enumeration of viable cells with beads, apoptotic cells or fused cells. Electroporation promoted a 49-fold increase of the percentage of viable cells that had fused with apoptotic cells. Apoptotic cell-viable cell fusions were 8-fold more likely to not attach to cell culture plastic and 2.3-fold less likely to proliferate after 24hr incubation than viable cell fusion controls. These data demonstrate that apoptotic signals can be transferred between cells by electrofusion, possibly suggesting a novel investigative approach for optimizing targeted cell deletion in cancer treatment.     

Effects of caspase inhibition on camptothecin-induced apoptosis of HL-60 cells

BACKGROUND: During camptothecin (CAM)-induced apoptosis of HL-60 cells, the external exposure of phosphatidylserine (PS) can either precede or follow DNA cleavage. The evidence suggests that cells in S-phase when CAM is added undergo rapid DNA, nuclear, and cellular disintegration before exposing PS on the outside of the plasma membrane, whereas cells moving from G1 into S-phase after CAM is added expose PS before they manifest the other phenomena. This study describes further investigations using the broad spectrum caspase inhibitor Z-VAD-FMK. The cells were cultured for a period long enough to ascertain whether a particular phenomenon was only delayed or was blocked completely. METHODS: Changes in cell light scatter, binding of annexin V-fluorescein isothiocyanate (FITC) to PS, uptake of propidium iodide (PI) as a measure of plasma membrane integrity, and DNA content after membrane fixation/permeabilization were monitored by flow cytometry during 24-h cultures. Fluorescence microscopy was used to examine cell morphology. RESULTS: Caspase inhibition blocked DNA cleavage, breakdown of the nuclear membrane, and formation of apoptotic bodies. It also revealed the existence of a CAM-activated early S-phase checkpoint. Cells arrested in early S-phase preceded the appearance of PS-positive cells. Caspase inhibition delayed both PS exposure and loss of plasma membrane integrity but did not prevent either. CONCLUSIONS: The results support the hypothesis that the sequence of apoptotic phenomena in an individual CAM-treated HL-60 cell depends on the stage of proliferation of that cell when it encounters the CAM. They are also consistent with the hypothesis that caspases are not required for PS exposure or the loss of plasma membrane integrity, but they are involved indirectly in promoting these phenomena.     

Discrimination of early and late apoptotic cells by NBD-phosphatidylserine-labelling and time-lapse observation of phagocytosis of apoptotic cells by macrophages

Since free apoptotic cells are not detected in normal tissues, it is generally believed that apoptotic cells are removed as soon as they appear in vivo. A fluorescent derivative of phosphatidylserine, 1-palmitoyl-2-[6-[(7-nitro-2-1,3-benzoxadiazol-4-yl)amino]hexanoyl]-sn-glycero-3-phospho-L-serine (NBD-PS) is known to be incorporated into living cells, and thereafter gradually absorbed into either fatty acid-free bovine serum albumin or fetal calf serum from the outer leaflet of the cell membrane. When thymocytes were irradiated with X-ray and cultured in the presence of NBD-PS, cells became less fluorescent as apoptosis advanced, but early apoptotic cells were still positive for NBD-PS. We then co-cultured such early apoptotic thymocytes with resident peritoneal macrophages. Upon examination under a time-lapse fluorescence microscope, it was found that the attachment of early apoptotic cells to macrophages does not cause rapid phagocytosis, as compared with late apoptotic cells, suggesting the possibility that, in contrast to the widely held view, early apoptotic cells may not be quickly removed by phagocytes in vivo.     

Neuroprotective effects of genistein in VSC4.1 motoneurons exposed to activated microglial cytokines

Pro-inflammatory cytokines released from activated microglia may be responsible for neuronal damage and resulting motor deficits associated with CNS disorders such as spinal cord injury, Parkinson’s disease, and multiple sclerosis. Estrogen (17β-estradiol) is capable of ameliorating motoneuron death following spinal cord injury, but has a number of deleterious side effects. Genistein (GEN), an estrogen receptor beta agonist and potent antioxidant, may represent an alternative to estrogen in treating neurodegenerative disorders. However, little is known about the neuroprotective effects of GEN. We therefore tested whether GEN would prevent apoptosis in cultured motoneurons following exposure to pro-inflammatory cytokines released from IFN-γ activated microglia. Exposure of ventral spinal cord 4.1 motoneurons to microglial cytokine supernatant in vitro caused significant apoptosis and reduced mitochondrial membrane potential. An increase in reactive oxygen species, intracellular Ca²⁺, calpain, caspases, cytochrome c, and the bax:bcl-2 ratio were also noted. GEN treatment reversed apoptotic death and cellular changes following cytokine exposure and was associated with increased expression of estrogen receptor β suggesting that GEN may promote neuroprotection via receptor-mediated pathways. The addition of ICI 182, 780, an estrogen receptor antagonist following GEN treatment attenuated neuroprotection, suggesting that GEN may act mainly via estrogen receptor β to protect VSC4.1 motoneurons. We conclude that GEN protects cultured ventral spinal cord 4.1 cells from inflammatory insult and thus may represent a potential beneficial therapy in the treatment of neurodegenerative disorders.     

Mast cell chymase induces smooth muscle cell apoptosis by disrupting NF-kappaB-mediated survival signaling

Chymase released from activated mast cells induces apoptosis of vascular smooth muscle cells (SMCs) in vitro by degrading the pericellular matrix component fibronectin, so causing disruption of focal adhesion complexes and Akt dephosphorylation, which are necessary for cell adhesion and survival. However, the molecular mechanisms of chymase-mediated apoptosis downstream of Akt have remained elusive. Here, we show by means of RT-PCR, Western blotting, EMSA, immunocytochemistry and confocal microscopy, that chymase induces SMC apoptosis by disrupting NF-kappaB-mediated survival signaling. Following chymase treatment, the translocation of active NF-kappaB/p65 to the nucleus was partly abolished and the amount of nuclear p65 was reduced. Pretreatment of SMCs with chymase also inhibited LPS- and IL-1beta-induced nuclear translocation of p65. The chymase-induced degradation of p65 was mediated by active caspases. Loss of NF-kappaB-mediated transactivation resulted in downregulation of bcl-2 mRNA and protein expression, leading to mitochondrial swelling and release of cytochrome c. The apoptotic process involved activation of both caspase 9 and caspase 8. The results reveal that, by disrupting the NF-kappaB-mediated survival-signaling pathway, activated chymase-secreting mast cells can mediate apoptosis of cultured arterial SMCs. Since activated mast cells colocalize with apoptotic SMCs in vulnerable areas of human atherosclerotic plaques, they may participate in the weakening and rupture of atherosclerotic plaques.     

Cutting edge: neutrophil granulocyte serves as a vector for Leishmania entry into macrophages

Macrophages (MF) are the final host cells for multiplication of the intracellular parasite Leishmania major (L. major). However, polymorphonuclear neutrophil granulocytes (PMN), not MF, are the first leukocytes that migrate to the site of infection and encounter the parasites. Our previous studies indicated that PMN phagocytose but do not kill L. major. Upon infection with Leishmania, apoptosis of human PMN is delayed and takes 2 days to occur. Infected PMN were found to secrete high levels of the chemokine MIP-1beta, which attracts MF. In this study, we investigated whether MF can ingest parasite-infected PMN. We observed that MF readily phagocytosed infected apoptotic PMN. Leishmania internalized by this indirect way survived and multiplied in MF. Moreover, ingestion of apoptotic infected PMN resulted in release of the anti-inflammatory cytokine TGF-beta by MF. These data indicate that Leishmania can misuse granulocytes as a "Trojan horse" to enter their final host cells "silently" and unrecognized.     

The Golgi apparatus is a primary site of intracellular damage after photosensitization with Rose Bengal acetate

The aim of the present investigation was to elucidate whether the Golgi apparatus undergoes photodamage following administration of the fluorogenic substrates Rose Bengal acetate (RBAc) and irradiation at the appropriate wavelength. Human HeLa cells were treated in culture and the changes in the organization of the Golgi apparatus were studied using fluorescence confocal microscopy and electron microscopy, after immunocytochemical labeling. To see whether the cytoskeletal components primarily involved in vesicle traffic (i.e., microtubules) might also be affected, experiments of tubulin immunolabeling were performed. After treatment with RBAc and irradiation, cells were allowed to grow in drug-free medium for different times. 24 hr after irradiation, the cisternae of the Golgi apparatus became packed, and after 48-72 hr they appeared more fragmented and scattered throughout the cytoplasm; these changes in the organization of the Golgi cisternae were confirmed at electron microscopy. Interestingly enough, apoptosis was found to occur especially 48-72 h after irradiation, and apoptotic cells exhibited a dramatic fragmentation of the Golgi membranes. The immunolabeling with anti-tubulin antibody showed that microtubules were also affected by irradiation in RBAc-treated cells.     

Phosphatidyl serine exposure during apoptosis precedes release of cytochrome c and decrease in mitochondrial transmembrane potential

Time kinetics of phosphatidyl serine (PS) exposure were compared to other apoptotic parameters following different apoptotic stimuli. Our data indicate that anti-Fas treatment of L929sAhFas cells results in rapid exposure of PS, which precedes decrease in mitochondrial transmembrane potential (DeltaPsi(m)) and release of cytochrome c, indicating that PS exposure occurs independently of these mitochondrial events. Also during TNF-, etoposide- or staurosporine-mediated apoptosis in PC60 RI/RII cells, PS-positive cells were observed before they had a decreased DeltaPsi(m). However, during growth factor depletion-induced death of 32D cells, both phenomena seemed to occur at the same time.     

Endothelial cell-associated platelet-activating factor: a novel mechanism for signaling intercellular adhesion

The binding of neutrophils (polymorphonuclear leukocytes [PMNs]) to endothelial cells (ECs) presents special requirements in the regulation of intercellular adhesion. ECs that are stimulated by certain agonists, including thrombin and cytokines (tumor necrosis factor alpha, interleukin-1), generate molecular signals that induce the adhesion of PMNs (endothelial cell-dependent neutrophil adhesion). Our experiments demonstrate that the mechanism of binding induced by thrombin is distinct from that induced by the cytokines based on the time courses, the requirement for protein synthesis, and differential binding of HL60 promyelocytic leukemia cells to ECs activated by the two classes of agonists. The rapid EC-dependent PMN adhesion (initiated in minutes) that occurs when the ECs are stimulated by thrombin is temporally coupled with the accumulation of platelet-activating factor, a biologically active phosphoglyceride that remains associated with ECs and that activates PMNs by binding to a cell surface receptor. A portion of the newly synthesized platelet-activating factor (PAF) is on the EC surface, as demonstrated by experiments in which the rate of hydrolysis of PAF synthesized by activated ECs was accelerated by extracellular PAF acetylhydrolase. When ECs were treated with exogenous PAF they became adhesive for PMNs; the PMN binding was prevented by incubating the ECs with PAF acetylhydrolase or by treating the PMNs with competitive PAF receptor antagonists. Thus PAF associated with the EC plasma membrane induces PMN binding, an observation supported by experiments in which PAF in model membranes (liposomes) stimulated rapid PMN adhesion to ECs and to cell-free surfaces. In addition, competitive antagonists of the PAF receptor inhibited the binding of PMNs to ECs activated by thrombin and other rapidly acting agonists, but not to ECs activated by tumor necrosis factor alpha, indicating that PAF that is endogenously synthesized by ECs can mediate neutrophil adhesion. These experiments demonstrate a novel mechanism by which a cell-associated phospholipid, PAF, can serve as a signal for an intercellular adhesive event.