Recruitment of null cells during graft versus host reactions in the chicken

Untreated SC (B2/B2) chicken spleen or thymus cells (2 × 10⁷) caused significantly increased [³H]thymidine incorporation in spleens of heavily irradiated FP (B15/B21) recipient chicks on Day 4 after iv injection. Mitomycin-treated SC spleen cells or spleen cells treated with rabbit anti-T-cell serum and complement failed to raise the [³H]thymidine incorporation over that in uninjected, bursa cell-injected or FP spleen cell-injected controls. However, the combination of mitomycin-treated spleen or thymus cells and anti-T-treated spleen cells caused an increased [³H]thymidine uptake, suggesting the recruitment of non-T cells into proliferation by alloreactive mitomycin-treated T cells. Bursa cells did not proliferate during GVH reactions even though they could be shown to undergo proliferation in vivo upon mitogen (lipopolysaccharide and dextran sulfate) stimulation. In contrast, anti-T-treated spleen cells from agammaglobulinemic chickens were recruited into proliferation, suggesting that the recruited cell was not only not a T cell, but also no pre-B or B cell and most likely represented a cell of the monocyte-macrophage series.     

Inhibition of phorbol ester-receptor binding by a factor from human serum.

The inhibition of receptor binding of [3H]phorbol-12,13-dibutyrate (PDBu) by a factor from human serum was characterized. The serum factor inhibited [3H]PDBu binding in intact monolayer cultures of the rat embryo cell line CREF N and in a subcellular system containing membranes from these cells. Inhibition occurred at both 37 and 4 degrees C and was rapid and reversible. An analysis of [3H]PDBu binding in the presence of the serum factor indicated that inhibition of [3H]PDBu binding by the serum factor was noncompetitive. Using gel filtration to separate the serum factor from free [3H]PDBu, we obtained evidence that the serum factor does not act by binding or trapping the [3H]PDBu. Unlike the phorbol ester tumor promoters, the serum factor alone did not stimulate the release of choline or arachidonic acid from cellular phospholipids, nor did it inhibit the binding of 125I-labeled epidermal growth factor to cellular receptors. The factor did, however, antagonize the inhibition of epidermal growth factor binding induced by PDBu. Sera from pregnant women were, in general, more inhibitory of [3H]PDBu binding than were those from nonpregnant women, which were more inhibitory than those from men. During these studies we found that CREF N cells responded to being grown in the presence of PDBu by partial down regulation of the phorboid receptor. The 50% effective dose for down regulation was 8 nM PDBu, and the maximum effect occurred after 6 h. Taken together, our results indicate that the serum factor inhibits [3H]PDBu binding by a direct physical effect at the level of the phorboid receptors or their associated membranes. It would appear that if this factor acts in vivo, then it might antagonize certain effects of this class of tumor promoters.     

Distribution of 32P-labelled endotoxin in the frog.

Distribution and elimination of 32P-labelled endotoxin were studied in normal frogs and in frogs pretreated with a high dose of unlabelled endotoxin 24 hr previously. Like in other animals, the spleen and the liver of the frogs were found the most active R.E.S. organs. The alimentary tract too exhibited considerable R.E.S. activity. Pretreatment with unlabelled endotoxin resulted in a decrease of 32P-endotoxin uptake by the liver, the spleen and the alimentary tract. Evidence was obtained that the well-known effects of endotoxin on the R.E.S. of the mammals can be demonstrated in the frog as well.     

Stimulation of chicken lymphocytes by T- and B-cell mitogens.

Cultures of chicken spleen, peripheral blood, thymus, and bursal lymphocytes were tested for mitogenic stimulation by phytohaemagglutinin (PHA), concanavalin A (ConA), pokeweed mitogen (PWM), bacterial lipopolysaccharide (LPS), trypsin, and insulin. Spleen and blood leukocytes were stimulated by both the lectins and LPS, and also to some degree by trypsin and insulin as judged by increased incorporation of [³H]thymidine into acid-insoluble material. This was observed in cultures incubated in serum-free medium as well as in the presence of foetal bovine serum or autologous plasma. Thymus cells were reproducibly stimulated by high concentrations of PHA. No significant responses were obtained in bursal cell cultures with any of the compounds tested. Removal of cotton wool-adherent cells from the spleen cell suspensions resulted in a subpopulation of cells which were stimulated by PHA but showed little response to ConA, PWM, or LPS. This procedure did not remove surface immunoglobulin-bearing cells from the original suspension. Both these enriched spleen lymphocytes and the unfractionated spleen, blood and thymus leukocyte cultures were effectively stimulated by a partially purified PHA but with a highly purified PHA preparation only at very high concentrations. These and other results suggest that the mitogenic components in crude PHA preparations are different for chicken and human or mouse cells.     

The effects of mastoparan on the carrot cell plasma membrane polyphosphoinositide phospholipase C.

When [3H]inositol-labeled carrot (Daucus carota L.) cells were treated with 10 or 25 microM wasp venom peptide mastoparan or the active analog Mas-7 there was a rapid loss of more than 70% of [3H]phosphatidylinositol-4-monophosphate (PIP) and [3H]phosphatidylinositol-4,5-bisphosphate (PIP2) and a 3- and 4-fold increase in [3H]inositol-1,4-P2 and [3H]inositol-1,4,5-P3, respectively. The identity of [3H]inositol-1,4,5-P3 was confirmed by phosphorylation with inositol-1,4,5-P3 3-kinase and co-migration with inositol-1,3,4,5-P4. The changes in phosphoinositides were evident within 1 min. The loss of [3H]PIP was evident only when cells were treated with the higher concentrations (10 and 25 microM) of mastoparan or Mas-7. At 1 microM Mas-7, [3H]PIP increased. The inactive mastoparan analog Mas-17 had little or no effect on [3H]PIP or [3H]PIP2 hydrolysis in vivo. Neomycin (100 microM) inhibited the uptake of Mas-7 and thereby inhibited the Mas-7-stimulated hydrolysis of [3H]PIP and [3H]PIP2. Plasma membranes isolated from mastoparan-treated cells had increased PIP-phospholipase C (PLC) activity. However, when Mas-7 was added to isolated plasma membranes from control cells, it had no effect on PIP-PLC activity at low concentrations and inhibited PIP-PLC at concentrations greater than 10 microM. In addition, guanosine-5'-O-(3-thiotriphosphate) had no effect on the PIP-PLC activity when added to plasma membranes isolated from either the Mas-7-treated or control cells. The fact that Mas-7 did not stimulate PIP-PLC activity in vitro indicated that the Mas-7-induced increase in PIP-PLC in vivo required a factor that was lost from the membrane during isolation.     

ATP/Mg2+-dependent binding of vincristine to the plasma membrane of multidrug-resistant K562 cells

To study the mechanism of active drug efflux in multidrug-resistant cells, the interaction between [3H] vincristine (VCR) and plasma membrane prepared from an adriamycin (ADM)-resistant variant (K562/ADM) of human myelogenous leukemia K562 cells was examined by filtration method. [3H]VCR bound to the plasma membrane prepared from K562/ADM cells, but not from parental K562 cells, depending on the concentrations of ATP and Mg2+. Adenosine 5'-O-(3-thio)triphosphate was not effective in the binding of [3H]VCR, indicating that ATP hydrolysis is required for this binding. Dissociation constant (Kd) of VCR binding was 0.24 +/- 0.04 microM in the presence of 3 mM ATP. In the absence of ATP, specific binding of VCR to K562/ADM membrane was also observed; however, the affinity (Kd = 9.7 +/- 3.1 microM) was 40 times lower than that observed in the presence of ATP. The high affinity VCR binding to K562/ADM membrane was dependent on temperature. The bound [3H]VCR molecules were rapidly released by unlabeled VCR added to the reaction mixture at 25 degrees C. The high affinity binding of [3H]VCR to K562/ADM membrane was inhibited by VCR, vinblastine, actinomycin D, and ADM, to which K562/ADM cells exhibit cross-resistance, whereas 5-fluorouracil and camptothecin, to which K562/ADM cells are equally sensitive as K562 cells, did not inhibit the [3H]VCR binding. Furthermore, verapamil and other agents, which are known to circumvent drug resistance by inhibiting the active efflux of antitumor agents from resistant cells, could also inhibit the high affinity [3H]VCR binding. These results indicate that ATP/Mg2+-dependent high affinity VCR binding to the membrane of resistant cells closely correlates with the active drug efflux of this resistant cell line.     

Application of the isotope-dilution principle to the analysis of factors affecting the incorporation of [3H]uridine and [3H]cytidine into cultured lymphocytes. Evaluation of pools in serum and culture media

1. Rat lymph-node cells were incubated in serum and medium 199 with [5-(3)H]uridine or [5-(3)H]cytidine and acid-precipitable radioactivity was measured. Results were interpreted in terms of an isotope-dilution model. 2. Both serum and medium 199 contained pools that inhibited radioactive labelling in a competitive manner. The serum activity was diffusible and inhibited labelling with [(3)H]cytidine more than with [(3)H]uridine; in these respects the activity resembled cytidine (14mum). 3. The pools in serum and plasma were the same size; however, the rate of labelling was greater in plasma, owing to a diffusible factor. 4. Paradoxically, relatively simple media (Earle's salts and Eagle's minimum essential) appeared to have a larger pool than the more complex pyrimidine-containing medium 199; this suggests a contribution to the pool by cells in the simple media. 5. In the absence of pools the average cell was capable of incorporating 2000 radioactive nucleoside molecules/s.     

Immunological and molecular characterization of the cAMP-dependent protein kinases in AtT20 cells

The properties of the cAMP-dependent protein kinases in AtT20 mouse pituitary tumor cells were characterized by a combination of immunological and biochemical techniques. Ninety per cent of the total cAMP-dependent protein kinase was in the 40,000 X g supernatant fraction. Protein kinases I and II were immunoprecipitated with specific antisera directed against their regulatory subunits. The immunoprecipitated kinases bound [3H]cAMP and were catalytically active when incubated with [gamma-32P]ATP-Mg and protamine or histone H2B. Immunoprecipitated protein kinases I and II bound [3H]cAMP with apparent Kb values of 1.5 and 15 nM, respectively. Regulatory subunit concentrations in AtT20 cells were measured by immunoprecipitation of [3H]cAMP-R complexes. R-I and R-II levels were 2.7 and 3.0 pmol of [3H]cAMP binding activity per mg of cytosolic protein, respectively, however, the ratio of protein kinase II to protein kinase I was 2.5 indicating the presence of a significant amount of free R-I. This was confirmed by DEAE-cellulose chromatography and the isolation of immunoreactive R-I devoid of protein kinase activity. A significant amount of R-I also coeluted with protein kinase II when AtT20 cell extracts were subjected to DEAE-cellulose chromatography. In quantitative immunoprecipitation experiments, 0.1 microliter of anti-brain R-II serum complexed up to 0.5 pmol of the [3H]cAMP-binding activity of protein kinase II prepared from bovine and rat brain, and AtT20 cells while 2 microliter of anti-brain R-II serum was required to precipitate an equal amount of protein kinase II from bovine skeletal muscle showing that the protein kinase II in AtT20 cells contained the neural-specific R-II subunit.     

Growth-inhibitory activity of lymphoid cell plasma membranes. I. Inhibition of lymphocyte and lymphoid tumor cell growth

Membranes isolated from normal spleen cells or lymphoid tumor cells were found to inhibit in vitro growth of several murine tumor cell lines including a B cell hybridoma, a thymoma, and a mastocytoma. 50% inhibition occurred at membrane protein concentrations of 60-100 micrograms/ml. A similar concentration dependence was found for inhibition of [3H]-thymidine incorporation by tumor cells and for the lipopolysaccharide-induced mitogenic response of normal spleen cells. The inhibitory activity co-purified with the plasma membrane upon fractionation of crude membranes. Membrane solubilization with deoxycholate followed by dialysis to remove the detergent gave good recovery of inhibitory activity in the resulting reconstituted membranes. Membrane-mediated growth inhibition resulted from a decreased rate of proliferation and not from increased cell death. A toxic effect of the membranes was further ruled out by the finding that increasing the fetal calf serum content of the medium could substantially reverse the growth inhibition. Thus, the plasma membrane of lymphoid cells contains a component that can slow or stop the growth of cells in culture. This membrane component may have a role in cell contact-mediated regulation of growth.     

Antibody in the sera of tumor-bearing mice that mediates spleen cell cytotoxicity toward the autologous tumor.

Pretreatment of MTV-induced BALB/cfC3H mammary tumor cells with autologous serum results in increased spleen cell cytotoxic activity and the recruitment of previously inactive spleen cells to cytotoxic activity against the target cells. These recruiting antibodies are tumor-specific for individual tumors; pretreatment with such serum of target cells of an MTV-induced mammary tumor obtained from a different BALB/cfC3H female results in blocking of spleen cell activity. The autologous recruiting factors are active at dilutions of 1000 or more of whole serum are found in the 19S fraction after gel filtration.     

Effect of Anti-Ia Serum In Vitro on Lipopolysaccharide-Induced Proliferative Responses of Murine Bone Marrow and Spleen Cells

The in vitro proliferative response of murine bone marrow cells and spleen cells to bacterial lipopolysaccharide (LPS) and the effect of anti-Ia serum on the response were studied. The incorporation of [³H]thymidine into cells prepared from bone marrow increased in the presence of LPS, but the addition of anti-Ia serum to the cultures reduced the incorporation. Pretreatment of bone marrow cells with anti-Ia serum and complement did not abolish the ability of the cells to respond to LPS, while the same pretreatment destroyed this ability in spleen cells. These results suggest that cultures of Ia-negative bone marrow cells generate Ia-positive cells during the culture period, and the Ia-positive cells are responsive cells to LPS. The proliferative response of 1- or 2-week-old spleen cells was easily suppressed by anti-Ia serum when compared with that of 4-week-old spleen cells. Furthermore, the responses of spleen cells obtained from γ-irradiated and syngeneic bone marrow cell-reconstituted mice were prominently suppressed by anti-Ia serum in comparison with that of normal adult spleen cells. These findings suggest that LPS-responsive lymphocytes in the developmental stage are quite sensitive to anti-Ia serum. The effect of anti-Ia serum on the maturation of bone marrow-derived lymphocytes was discussed.     

The relative importance of the bone marrow and spleen in the production and dissemination of B lymphocytes.

The relative importance of the bone marrow and spleen in the production of B lymphocytes was investigated in guinea pigs by the combined use of [³H]TdR radio-autography and fluorescent microscopy after the staining of B cells by FITC-F(ab′)₂-goat-anti-guinea pig Ig. Large and small lymphoid cells possess sIg in the marrow and spleen but B cell turnover in the marrow exceeds that in the spleen. That newly generated bone marrow B cells are not derived from an extramyeloid bursa equivalent was demonstrated by the absence of [³H]TdR labeled B cells in tibial marrow 72 hr after [³H]TdR was administered systemically, while the circulation to the hind limbs was occluded. Pulse and chase studies with [³H]TdR showed that large marrow B cells are derived from sIg-negative, proliferating precursors resident in the bone marrow and not from the enlargement of activated small B lymphocytes. The acquisition of [³H]TdR by splenic B cells lagged behind that observed in the marrow. Three days after topical labeling of tibial and femoral bone marrow with [³H]TdR, a substantial proportion of splenic B cells were replaced by cells that had seeded there from the labeled marrow. The studies unequivocally identify the bone marrow as the organ of primary importance in B cell generation and indicate that in the guinea pig rapidly renewed B lymphocytes of the spleen are replaced by lymphocytes recently generated in bone marrow. The rate of replacement of B lymphocytes in the lymph node by cells newly generated in the bone marrow takes place at a slower tempo than in the spleen.     

Changes in fibroblast-derived trypomastigotes of Trypanosoma cruzi during long-term culture.

Fibroblast-derived trypomastigotes (FDTs) of Trypanosoma cruzi that had been in culture for extended periods of time were found to differ in their ability to proliferate in culture when compared to blood-form trypomastigotes (BFTs) and FDTs that had been recently established from blood-forms. "Old" FDTs transform into amastigotes/spheromastigotes and epimastigotes and readily incorporate [3H]thymidine in medium alone or in the presence of mouse spleen cells, whereas "new" FDTs and BFTs did not incorporate [3H]thymidine although they did transform in culture. These differences should be considered when FDTs are used for physiologic and immunologic studies of T. cruzi.     

Cytoplasmic location of undermethylated messenger RNA in Novikoff cells.

Novikoff cells in culture were labeled with L-[methyl-3H]methionine and [U-14C]uridine in the presence of (a) TubHcy2, (b) AdoHcy, (c) homocysteine, (d) tubercidin, or (e) without any additions. Only in cultures labeled in the presence of TubHcy were undermethylated cap structures observed to represent a significant portion of [3H]methyl radioactivity. Novikoff cells in culture were then simultaneously labeled with L-[methyl-3H]methionine and [32P]orthophosphate in the presence or absence of TubHcy. Total cytoplasmic, polysomal and monosomal poly(A)-containing RNAs were analyzed. Both monosomal and polysomal mRNA fractions from TubHcy-treated cells contain partially methylated cap structures, suggesting that 2'-O-methylation of the nucleoside adjacent to the pyrophosphate linkage in caps is not required for transport, ribosomal binding or translation. Comparison of nuclear and cytoplasmic cap structures from normal and inhibited cultures indicate that an altered mRNA population is generated in the presence of TubHcy.     

Differential distribution of liposome-entrapped [H]methotrexate and labelled lipids after intravenous injection in a primate

Positive liposomes consisting of phosphatidylcholine, cholesterol and stearylamine and negatively charged liposomes consisting of phosphatidylcholine, cholesterol and phosphatidylserine, were double labelled with either ³H-labelled dipalmitoyl phosphatidylcholine and [¹⁴C]cholesterol or with [¹⁴C]cholesterol and [³H]methotrexate entrapped in the aqueous phase. The plasma levels and urinary excretion of radioactivity from sonicated and non-sonicated liposomes were then compared with the levels of radioactivity from free [³H]methotrexate during a 4 h experimental period after an initial intravenous injection in cynomolgous monkeys. Tissue uptake at the completion of the 4 h experimental period was also measured.It was found that plasma radioactivity from [³H]methotrexate and [¹⁴C]cholesterol in sonicated positive liposomes was cleared more slowly than from comparable non-sonicated liposomes, and considerably slower than from free [³H]methotrexate. Radioactivity from sonicated negative liposomes was cleared more rapidly than from positive sonicated liposomes. Positive liposomes captured considerably more [³H]methotrexate than negative liposomes and showed very low permeability to [³H]methotrexate in in vitro studies, even in the presence of high concentrations of serum.[¹⁴C]Cholesterol radioactivity was cleared more rapidly from plasma than ³H-radioactivity from liposome-entrapped [³H]methotrexate for double-labelled sonicated liposomes and generally showed greater uptake into tissues and red blood cells. ³H-labelled dipalmitoyl phosphatidylcholine in sonicated positive liposomes was cleared faster than [¹⁴C]cholesterol during the first 3 h. The more rapid disappearance of [¹⁴C]cholesterol from the plasma was complemented by greater uptake into a number of tissues, and positive non-sonicated liposomes were taken up to a greater extent by the spleen than equivalent sonicated liposomes.Renal excretion of ³H from liposome-entrapped [³H]methotrexate was considerably less than that of ³H from free [³H]methotrexate. There was insignificant excretion, however, of ¹⁴C from cholesterol in the urine.Entrapment in liposomes completely prevented the otherwise considerable breakdown of free methotrexate to ³H-containing products in plasma and partially prevented its breakdown in tissues.These studies indicate marked differences in the distribution of liposomes in vivo due to surface charge and size, and some degree of exchange of the lipid components of the liposome bilayer independent of the distribution of the entrapped species. They also show that entrapment in liposomes can reduce metabolic degradation of a drug, maintain high plasma levels and reduce its renal excretion.     

Differential distribution of liposome-entrapped [3H]methotrexate and labelled lipids after intravenous injection in a primate

Positive liposomes consisting of phosphatidylcholine, cholesterol and stearylamine and negatively charged liposomes consisting of phosphatidylcholine, cholesterol and phosphatidylserine, were double labelled with either ³H-labelled dipalmitoyl phosphatidylcholine and [¹⁴C]cholesterol or with [¹⁴C]cholesterol and [³H]methotrexate entrapped in the aqueous phase. The plasma levels and urinary excretion of radioactivity from sonicated and non-sonicated liposomes were then compared with the levels of radioactivity from free [³H]methotrexate during a 4 h experimental period after an initial intravenous injection in cynomolgous monkeys. Tissue uptake at the completion of the 4 h experimental period was also measured.It was found that plasma radioactivity from [³H]methotrexate and [¹⁴C]cholesterol in sonicated positive liposomes was cleared more slowly than from comparable non-sonicated liposomes, and considerably slower than from free [³H]methotrexate. Radioactivity from sonicated negative liposomes was cleared more rapidly than from positive sonicated liposomes. Positive liposomes captured considerably more [³H]methotrexate than negative liposomes and showed very low permeability to [³H]methotrexate in in vitro studies, even in the presence of high concentrations of serum.[¹⁴C]Cholesterol radioactivity was cleared more rapidly from plasma than ³H-radioactivity from liposome-entrapped [³H]methotrexate for double-labelled sonicated liposomes and generally showed greater uptake into tissues and red blood cells. ³H-labelled dipalmitoyl phosphatidylcholine in sonicated positive liposomes was cleared faster than [¹⁴C]cholesterol during the first 3 h. The more rapid disappearance of [¹⁴C]cholesterol from the plasma was complemented by greater uptake into a number of tissues, and positive non-sonicated liposomes were taken up to a greater extent by the spleen than equivalent sonicated liposomes.Renal excretion of ³H from liposome-entrapped [³H]methotrexate was considerably less than that of ³H from free [³H]methotrexate. There was insignificant excretion, however, of ¹⁴C from cholesterol in the urine.Entrapment in liposomes completely prevented the otherwise considerable breakdown of free methotrexate to ³H-containing products in plasma and partially prevented its breakdown in tissues.These studies indicate marked differences in the distribution of liposomes in vivo due to surface charge and size, and some degree of exchange of the lipid components of the liposome bilayer independent of the distribution of the entrapped species. They also show that entrapment in liposomes can reduce metabolic degradation of a drug, maintain high plasma levels and reduce its renal excretion.     

Immunocytochemical and ultrastructural characterization of endocrine cells in the larval stomach of the frog Rana temporaria tadpoles: a comparison with adult specimens

According to immunostaining and ultrastructural patterns, Rana temporaria tadpole stomach displays a well-differentiated endocrine population comprising, at least, six cellular types: ECL, EC [serotonin], D [somatostatin] - all three of them abundant -, P [bombesin] - less numerous -, CCK-8 [cholecystokinin/gastrin] and A [glucagon/glicentin] - both very scarce. Larval endocrine cells are mainly located in the surface epithelium and show open or closed morphologies. Cellular diversity is similar in tadpoles and frogs, with the exception of immunoreactivity for gastrin-17, found in adults in numerous cells. Larval cells display mature ultrastructural traits, although with smaller secretory granules. The different distribution of endocrine cells, which in adults are preferentially located in the glands, probably refers to different functional requirements. However, the rich vascular plexus present in larval mucosa may be an efficient transport medium of surface hormones to-gastric targets. The enhancement in adults of endocrine population and correlative increase in hormonal secretion indicates a more active functional role, probably related to the shift from herbivorous to carnivorous habits. In summary, the tadpole gastric endocrine population, although not as numerous as that of adult frogs, displays histological traits that indicate a relevant (immunoreactive and ultrastructural properties, cellular diversity) and specific (surface location, relative abundance of open-type cells) role of local regulatory factors in amphibian larval gastric function.     

Identification and characterization of a soluble suppressor factor(s) in the serum of AKR mice bearing lymphocytic leukemia

Leukemia in AKR mice was found to be associated with the presence of a serum factor(s) termed AKR leukemic suppressor factor (AKR-LSF). Suppression was quantitated by measuring the inhibition of PHA-stimulated [³H]thymidine incorporation by normal AKR spleen cells at various dilutions of leukemic mouse serum (LMS). AKR-LSF activity was expressed as units per milliliter, which is the reciprocal of the LMS dilution that inhibited [³H]thymidine uptake by 50% with respect to fetal calf serum control cultures. The amount of activity in the serum directly correlated to the rate of tumor cell growth. Mice receiving 10⁷ BW5147 transplanted leukemia cells had 130 ± 12 units of AKR-LSF activity/ml of serum compared to 40 ± 8 units/ ml for mice with spontaneous leukemia. Normal mouse serum contained 33 ± 11 units/ml. The leukemic serum exhibited no strain specificity in either phytohemagglutinin or lipopolysaccharide assays, but was found to be twofold more inhibitory against mouse spleen cells than that against rat spleen cells. Human lymphocyte blastogenesis was not inhibited by the leukemic serum. LMS did not inhibit the growth of L929 fibroblasts or murine tumor cells in vitro. Further work is necessary to determine what role the suppressor factor may play in the regulation of antitumor cell immunity.     

Metabolic labeling of glycosylphosphatidylinositol-anchor of heparan sulfate proteoglycans in rat ovarian granulosa cells.

The glycosylphosphatidylinositol (GPI)-anchor of the plasma membrane-associated heparan sulfate (HS) proteoglycan was metabolically radiolabeled with [3H]myristic acid, [3H]palmitic acid, [3H]inositol, [3H]ethanolamine, or [32P]phosphate in rat ovarian granulosa cell culture. Cell cultures labeled with [3H]myristic acid or [3H]palmitic acid were extracted with 4 M guanidine HCl buffer containing 2% Triton X-100 and the proteoglycans were purified by ion exchange chromatography after extensive delipidation. Specific incorporation of 3H into GPI-anchor was demonstrated by removing the label with a phosphatidylinositol-specific phospholipase C (PI-PLC). Incorporation of 3H activity into glycosaminoglycans and core glycoproteins was also demonstrated. However, the specific activity of 3H in these structures was approximately 2 orders of magnitude lower than that in the GPI-anchor, suggesting that 3H label was the result of the metabolic utilization of catabolic products of the 3H-labeled fatty acids. PI-PLC treatment of cell cultures metabolically labeled with [3H]inositol, [3H]ethanolamine, or [32P]phosphate specifically released radiolabeled cell surface-associated HS proteoglycans indicating the presence of GPI-anchor in these proteoglycans. GPI-anchored HS proteoglycans accounted for 20-30% of the total cell surface-associated HS proteoglycans and virtually all of them were removed by PI-PLC. These results further substantiate the presence of GPI-anchored heparan sulfate proteoglycan in ovarian granulosa cells and its cell surface localization.