Apoptosis specific proteins (ASP) are expressed in the cytoplasm of cultured mammalian cells of various lineages following induction of apoptosis. The cDNA encoding ASP has been cloned from a human expression library and has significant homology to the Saccharomyces cerevisiae APG5 gene which is essential for yeast autophagy. The ASP gene, known as hAPG5, can be transcribed to give mRNAs of 3.3 kbp, 2.5 kbp and 1.8 kbp which are present at comparable levels in viable and apoptotic cells, demonstrating that protein expression must be regulated at the translational level. These data indicate a possible relationship between apoptosis and autophagy and suggest evolutionary conservation in mammalian apoptosis of a degradative process present in yeast.

By analysis with a panel of CD21 MoAbs it is shown that a large part of the soluble CD21 in human blood plasma is of the long isoform (CD21L), as judged by comparison with antigen produced by mouse L cells transfected with CD21L-cDNA and reactivity with the restricted CD21 MoAb R4/23. This is compatible with the hypothesis that soluble CD21 in the blood is mainly derived from follicular dendritic cells (FDC). Cells from a human keratinocyte cell line transfected with cDNA from the Burkitt lymphoma cell line Raji also produced soluble CD21L (sCD21L), whereas the short form of sCD21 (sCD21S) was the major component of sCD21 produced by the B lymphoblastoid cell line LICR-LON-HMy and the T cell line Jurkat. Confocal studies of FDC isolated from human tonsil revealed that CD21 was present in the cytoplasm. On gel filtration sCD21 from untreated serum has an apparent size considerably greater than the 130 kD found by SDS-PAGE analysis. This may be partly accounted for by the non-globular shape of the molecule, but may also indicate, as reported by others, that in its native state sCD21 is complexed with other proteins. However, no evidence of complexing with sCD23 or C3d could be found.

11 beta-Hydroxysteroid dehydrogenase (11 beta HSI) is an enzyme complex responsible for the conversion of hormonally active cortisol to inactive cortisone; two isoforms of the enzyme have been cloned and characterized. Clinical observations from patients with the hypertensive syndrome apparent mineralocorticoid excess, recently explained on the basis of mutations in the human 11 beta HSD2 gene, suggest that it is the 11 beta HSD2 isoform that serves a vital role in dictating specificity upon the mineralocorticoid receptor (MR). We have raised a novel antibody in sheep against human 11 beta HSD2 using synthetic multiantigenic peptides and have examined the localization and subcellular distribution of 11 beta HSD2 in mineralocorticoid target tissues. The immunopurified antibody recognized a single band of approximately 44 kDa in placenta, trophoblast, and distal colon. In kidney tissue, two bands of approximately 44 and 48 kDa were consistently observed. No signal was seen in decidua, adrenal, or liver. Immunoperoxidase studies on the mineralocorticoid target tissues, kidney, colon, and parotid gland indicated positive staining in epithelial cells known to express the MR: respectively, renal collecting ducts, surface and crypt colonic epithelial cells, and parotid duct epithelial cells. No staining was seen in these tissues in other sites. The intracellular localization of 11 beta HSD2 in kidney and colon epithelial cells was addressed using confocal laser microscopy. Parallel measurements of 11 beta HSD2 and nuclear propidium iodide fluorescence on sections scanned through an optical section of approximately 0.1 micron indicated significant 11 beta HSD2 immunofluorescence in the nucleus. In human kidney, colon, and salivary gland, 11 beta HSD2 protects the MR from glucocorticoid excess in an autocrine fashion. Furthermore, within these tissues, 11 beta HSD2, which had been considered to be a microsomal enzyme, is also found in the nucleus, suggesting that the interaction between the MR and aldosterone or cortisol is in part a nuclear event.

Isolated rat hepatocyte couplets were used to study the effect of S-adenosyl-L methionine (SAMe) treatment on disruption of canalicular function caused by cyclosporin A (CyA). Canalicular function was assessed by counting the percentage of couplets that were able to accumulate the fluorescent cholephile choly-lysyl-fluorescein (CLF) into the canalicular vacuole between the two cells, i.e., canalicular vacuole accumulation (CVA). Cotreatment with 1 mmol/L SAMe prevented the inhibition of canalicular vacuole accumulation caused by CyA (75 nmol/L and 100 nmol/L), whereas treatment with it after CyA was unsuccessful. SAMe prevented the dose dependent reduction caused by CyA (5 nmol/L-1 mumol/L) both on CVA and on retention of CLF preaccumulated within the canaliculus, the effect on retention being complete. No difference in intracellular content of reduced glutathione (GSH) between the control and any dose level of the immunosuppressor, with or without SAMe treatment was observed, suggesting that changes in intracellular reduced GSH levels are not involved in the effects of SAMe. F-actin was stained with fluorescein-isothiocyanate phalloidin and fluorescence measurements were performed by confocal microscopy. The ratio of the percanalicular area fluorescence/total couplet fluorescence, indicative of F-actin distribution, significantly decreased with CyA. However, cotreatment of CyA with SAMe protected the integrity of the pericanalicular cytoskeleton, suggesting that this beneficial effect on canalicular function may maintain canalicular contractions and/or preserve tight junction function. Results are discussed in relation to possible involvement of the transmethylation pathway, modifications in membrane fluidity, effects on bile acid transport, and of inhibition of uptake of CyA. They suggest that SAMe could be a good candidate for protecting against CyA-induced membrane dysfunction.

Recent reports have claimed that activation of protein kinase C (PKC)-β is sufficient for both differentiation and apoptosis in promyeloid HL60 cells. Phorbol esters which differentially activate PKC isoenzymes in vitro were used to induce differentiation and apoptosis in U937 cells; TPA and Dopp activate all U937 PKC isoenzymes, except PKC-ζ and Doppa activate only PKC-βl. At concentrations of Doppa below 50 nM, only PKC-βl was activated by 2 min and apoptosis was induced, but there was no differentiation of cells towards monocytes. TPA (1–25 nM) and Dopp (5–100 nM) activated PKC-α,-βl and-δ within 2 min and induced differentiation, but only increased apoptosis at the highest concentrations used. Thus, initial activation of PKC-βl is insufficient for differentiation of U937 cells, but may lead to the induction of apoptosis.

The capacity to be recognized and engulfed by phagocytes is an important characteristic of cells dying by apoptosis. Phagocytosis of apoptotic cells occurs rapidly in vivo, probably prior to plasma membrane breakdown. While the molecular mechanisms mediating phagocytosis of apoptotic cells are beginning to be defined, little is yet known of the relationship between the cell‐death program itself and the surface changes on the dying cells that signal for engulfment. Here, we investigate to what extent the apoptosis repressor Bcl‐2 can modulate the recognition and phagocytosis of human B cells exposed to triggers of apoptosis. Burkitt lymphoma (BL)‐derived, Bcl‐2− B cells were induced into apoptosis either by the Ca2+‐ionophore ionomycin or by the inhibitor of protein synthesis cycloheximide. Apoptotic BL cells, but not viable BL cells, were recognized and phagocytosed by monocyte‐derived macrophages. bcl‐2‐transfected BL populations showed a reduced capacity both to undergo apoptosis in response to these inducing agents and to interact with macrophages. Like their Bcl‐2− counterparts, Bcl‐2+ BL cells interacted with macrophages only after activation of their apoptotic program as assessed by changes in nuclear morphology. These results demonstrate not only that continued protein synthesis in B cells undergoing apoptosis is not essential for their recognition by macrophages, but also that macrophage recognition of apoptotic B cells cannot be uncoupled from the cell‐death program that is controlled by Bcl‐2. In this respect, the behavior of B cells contrasts markedly with that of neutrophils in which Bcl‐2 has been reported to inhibit apoptosis without affecting phagocytic clearance (Lagasse and Weissman, J. Exp. Med. 1994. 179: 1047).

The subcellular localization of protein kinase C (PKC)-delta was determined in HL60 cells differentiated toward monocytes/macrophages by treatment with TPA. PKC-delta was detected in the nucleus and cytoplasm of differentiated HL60 cells and, more specifically, associated with structures resembling intermediate filaments. Indirect immunostaining revealed that PKC-delta colocalized with vimentin in the cytosol and perinuclear region of these cells. Immunoprecipitation studies showed that PKC-delta was in an active (autophosphorylated) state in differentiated HL60 cells and that vimentin immunoprecipitated from these cells was also phosphorylated. Treatment of HL60 cells with the PKC-specific inhibitor chelerythrine decreased the phosphorylation of vimentin. These data suggest that vimentin is a substrate for PKC-delta and that this PKC isoenzyme may play a specific role in the regulation of shape change and cell adhesion during HL60 differentiation.

Granulocyte macrophage colony-forming cells (GM-CFC) are bipotential progenitor cells that can proliferate and develop into macrophages in response to macrophage CSF or into neutrophils in response to stem cell factor or granulocyte CSF. These cytokines promoted growth and development in highly enriched GM-CFC. In [3H]thymidine suicide assays, IL-4 was shown to stimulate proliferation of GM-CFC to the same degree as IL-3 and other potent mitogens for GM-CFC. IL-4 also maintained the clonogenic potential of enriched GM-CFC over a 2-day period. However, after several days in the presence of IL-4, the GM-CFC began to die and retained blast cell morphology characteristic of the isolated GM-CFC. When a high concentration of IL-4 was added to GM-CFC with neutrophilic stimuli, the response of these cells was altered because macrophages were formed. This effect was achieved by a 4-h preincubation with IL-4, suggesting that an early signal produced by IL-4 promotes lineage restriction, although IL-4 itself cannot promote development. IL-4, like macrophage CSF, translocates PKC-alpha to the nucleus in GM-CFC, this redistribution of protein kinase C alpha (PKC-alpha) being inhibited by calphostin C (a PKC inhibitor). Calphostin C also blocked IL-4-mediated development of macrophages in stem cell factor- and granulocyte-CSF-treated cells. This is further evidence that PKC-alpha translocation is involved in the commitment of GM-CFC to macrophage development. This data also suggests that agonist-stimulated lineage commitment can be uncoupled from development in normal hematopoietic cells.

Most tonsil B cells have high levels of surface CD44 but this molecule is either expressed at low levels or is absent from germinal centre B cells (GCB). On average 62% of isolated GCB were found to be CD44- and the remainder CD44low. Most CD44- GCB were in cell cycle, indicating that they were centroblasts, while centrocytes, non-dividing GCB, were mainly CD44low. Immunohistological analysis confirms that centrocytes, which are located in the light zone of germinal centres, express low levels of CD44, while centroblasts, cells of the dark zone, are CD44-. While most CD77high GCB are centroblasts and CD77low GCB centrocytes, many centroblasts and centrocytes express intermediate levels of CD77, making this less reliable than CD44 for discriminating between these cells. Most CD44low and CD44- GCB were shown to have undergone Ig switch recombination in vivo. This indicates that switch recombination is independent of the maturation of centroblasts to centrocytes and precedes the signals that induce GCB to differentiate to plasma cells or memory B cells. The average rate of entry of the CD44- GCB fraction to apoptosis on culture at 37 degrees C was faster than that of the total GCB preparation. It is suggested that this may reflect strict stromal-dependence of centroblasts while centrocytes have to survive for long enough to have the chance of receiving antigen-specific selection signals. Inhibition of apoptosis by CD40 mAb with IL-4 or phorbol myristate acetate with ionomycin was similar in the CD44- and CD44low preparations.

Cell surface isoforms of meprin A (EC 3.4.24.18) from mice and rats contain β subunits that are type I integral membrane proteins and α subunits that are disulfide-linked to or noncovalently associated with membrane-anchored meprin subunits. Both α and β subunits are synthesized with COOH-terminal domains predicted to be cytoplasmic, transmembrane, and epidermal growth factor-like; these domains are retained in β subunits but are removed from α during maturation. The present studies establish that an inserted 56-amino acid domain (the “I” domain), present in α but not in β, is necessary and sufficient for COOH-terminal proteolytic processing of the α subunit. This was demonstrated by expression of mutant meprin subunits (deletion mutants, chimeric αβ subunits, and β mutants containing the I domain) in COS-1 cells. Mutations of two common processing sites present in the I domain (a dibasic site and a furin site) did not prevent COOH-terminal proteolytic processing, indicating that the proteases responsible for cleavage are distinct from those having these specificities. Deletion of the I domain from the α subunit resulted in accumulation of unprocessed subunits in a preGolgi compartment. Furthermore, COOH-terminal proteolytic processing of wild-type α subunits occurred before acquisition of endoglycosidase H resistance. Pulse-chase experiments and expression of an α subunit transcript containing a c-myc epitope tag, confirmed that proteolytic processing at the COOH terminus occurs in the endoplasmic reticulum. This work identifies the region of the α subunit that is essential for COOH-terminal processing and demonstrates that the differential processing of the evolutionarily-related subunits of meprin A that results in a structurally unique tetrameric protease begins in the endoplasmic reticulum.

In previous histological studies, biliary epithelial cells (BEC) in the liver of patients with primary biliary cirrhosis (PBC), but not controls, reacted strongly with antibodies specific for the major autoantigen associated with PBC, the E2 component of pyruvate dehydrogenase complex (PDC‐E2). In this study we have used transmission electron microscopy (TEM) to document the precise subscellular localization of PDC‐E2 in BEC. Two antibodies which recognize PDC‐E2 were used: affinity‐purified anti‐PDC‐E2 raised in rabbits; and human antibody from the serum of patients with PBC, affinity‐purified against human heart PDC. The intracellular localization of antibody binding was determined by laser scanning confocal microscopy and TEM. Both antibodies bound to the inner membrane of mitochondria in BEC isolated from both patients with PBC and controls, but binding to the external aspect of the plasma membrane was observed only in BEC from patients with PBC. Surface antigen expression in PBC may make BEC immunological targets.