Atypical mannolipids characterize Thy-1-negative lymphoma mutants.

Essentially all eukaryotic cells, including murine lymphomas, express surface proteins, such as Thy-1, which are anchored by a phosphoinositol mannolipid. Putative mannolipid anchor precursors can be detected in these cells. Six distinct Thy-1-negative lymphoma mutants lack complete mannolipids, and three mutants synthesize atypical mannolipids. The absence of complete mannolipids can account for the lack of expression of multiple mannolipid-anchored proteins and may also account for the lack of lipid anchoring in the human disease paroxysmal nocturnal hemoglobinuria. Structural information on the mannolipids of wild-type and mutant cells indicates that anchor biosynthesis in these cells may involve both transmembrane flip-flop of intermediates and a deacylation step.     

Affected paroxysmal nocturnal hemoglobinuria T lymphocytes harbor a common defect in assembly of N-acetyl-D-glucosamine inositol phospholipid corresponding to that in class A Thy-1- murine lymphoma mutants.

Deficient expression of glycoinositol phospholipid (GPI) anchored proteins in affected paroxysmal nocturnal hemoglobinuria (PNH) cells has been traced to a defect in GPI anchor assembly. In a previous study (Schubert, J., Schmidt, R. E., and Medof, M. E. (1993) J. Biol. Chem., in press) we characterized the biosynthesis of putative Man-containing GPI anchor precursors in normal peripheral blood lymphocytes and investigated assembly of these intracellular GPI intermediates in CD48- affected and CD48+ unaffected T and natural killer cell lines of PNH patients. We found that affected T cells from five patients exhibited a uniform defect in which dolichol-phosphoryl-Man was synthesized but no GPI mannolipids were expressed. In this study, membranes of patients' affected T cells were labeled with UDP-[3H]GlcNAc to evaluate earlier steps in GPI synthesis, and intact cells were fused to Thy-1- murine lymphoma mutants harboring different defects in early GPI assembly to test for the presence of corresponding or complementary lesions. In all cases, affected cell membranes failed to assemble GlcNAc-inositol phospholipid, the initial precursor of GPI anchor structures, and the intact cells failed to complement class A mutants while complementing other classes. Affected polymorphonuclear leukocytes from three additional patients of different origin were then labeled with [3H]Man and the labeling patterns found to correspond to those obtained with the T lymphocytes. Taken together the data indicate that the genetic lesion in PNH cells resides in a DNA element which: 1) encodes a product required for the synthesis of GlcNAc-inositol phospholipid, 2) corresponds to that altered in class A Thy-1- murine lymphoma mutants, and 3) is commonly affected in different patients.     

Extensive labeling with [3H]ethanolamine of a hydrophilic protein of animal cells

Murine T-lymphomas and Thy-1- mutants were labeled overnight with [3H]ethanolamine to detect proteins which possess a glycophospholipid anchor. When labeled cells were treated with 10% trichloroacetic acid and then analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and fluorography, both Thy-1 and a second intensely labeled protein (46 kDa) were observed. The presence of the radiolabeled 46-kDa protein in wild type and class E Thy-1 negative cells (cells in which Thy-1 is synthesized but cannot be labeled with [3H]ethanolamine) suggested incorporation into a distinct moiety. Labeling of the 46-kDa protein with [3H]ethanolamine is rapidly inhibited by cycloheximide. Further characterization of the 46-kDa protein by subcellular fractionation and Triton X-114 partitioning indicated that the protein is located in the cytosol. The protein is basic and does not bind to either concanavalin A or wheat germ agglutinin. Labeling of a 46-kDa protein has also been demonstrated in Chinese hamster ovary, COS, rat myeloma, cloned human T-lymphocytes, and HeLa cells. Pronase digestion of the [3H]ethanolamine-labeled 46-kDa protein of wild type lymphoma cells generated a nonbasic and polar labeled fragment which is labile to strong acid and base ([3H]ethanolamine is liberated), insensitive to periodate oxidation and alkaline phosphatase, and does not bind to concanavalin A or wheat germ agglutinin. Judging from methylation studies, the labeled ethanolamine residue does not contain a free amino group. Based on these results, we report a novel post-translational modification of selected protein(s) by the covalent addition of [3H]ethanolamine.     

Anchoring and degradation of glycolipid-anchored membrane proteins by L929 versus by LM-TK- mouse fibroblasts: implications for anchor biosynthesis.

Although many cells anchor surface proteins via moieties that are sensitive to phosphatidylinositol-specific phospholipase C (PI-PLC), the anchor moieties of surface proteins of mouse L929 cells resist PI-PLC. By constructing stable hybrids between L929 and lymphoma cells that express glycolipid-anchored proteins in a PI-PLC-sensitive form, we show that PI-PLC resistance behaves as a recessive trait. Since putative mannolipid precursors of the lipid anchors bear alkali-labile substituents which make them resist PI-PLC, these observations are most simply interpreted by postulating that L929 lacks a critical anchor deacylase. Unlike the L929 cell line, two of its descendants, the LM cell line and its thymidine kinase-negative variant (LM-TK-), do not express glycolipid-anchored proteins on their surface. Moreover, unlike L929 cells, LM-TK- cells rapidly inactivate at least one lipid-anchored enzyme in a compartment sensitive to acidotropic amines and leupeptin. By fusion of LM-TK- cells to mouse Thy-1- lymphoma mutants and monitoring of surface expression of lipid-anchored proteins, we assign LM-TK- to lymphoma mutant complementation group H. This genetic assignment is matched by analysis of mannolipids of L929, LM-TK-, wild-type, and class H lymphoma mutant cells: striking similarities are seen between the two wild-type cells by contrast to the mutants. Since the differences pertain to lipids which have properties consistent with their being anchor precursors, we suggest that LM-TK- has a lesion in the synthesis of anchor precursor mannolipids.     

The phenotype of five classes of T lymphoma mutants. Defective glycophospholipid anchoring, rapid degradation, and secretion of Thy-1 glycoprotein

Thy-1 glycoprotein is a member of a class of proteins which are anchored to the plasma membrane via a covalently bound glycophospholipid. The biosynthesis and anchoring of Thy-1 were investigated in a family of wild-type and mutant (complementation groups A, B, C, E, and F) T lymphomas. The mutants all synthesize Thy-1 but fail to express it on the cell surface. Analysis of the size of D-[2-3H]mannose-labeled dolichol-linked oligosaccharides showed that the class E mutant is the only cell line which does not synthesize dolichol-P-P-Glc3Man9GlcNAc2. Turnover and possible secretion of Thy-1 by mutant T lymphoma cells were documented in D-[2-3H]mannose pulse-chase experiments. The turnover of [3H]Thy-1 for all wild-type cells is considerably slower than for the mutant cells. Class B and E cells release appreciably more [3H]Thy-1 than wild-type cells. Additional experiments were performed to determine the electrophoretic mobility and hydrophobicity of cell-associated and released forms of Thy-1 labeled overnight with [3H]mannose. All wild-type and class A, C, E, and F mutant cells contain a major Triton X-114 binding species of cell-associated [3H]Thy-1. All extracellular [3H]Thy-1 was almost exclusively hydrophilic. The presence of two Thy-1 anchor components, ethanolamine and palmitate, was investigated. Biosynthetic labeling with [3H]palmitic acid showed that all of the wild-type cells but none of the mutants incorporated this anchor precursor into Thy-1. In [3H]ethanolamine-labeling experiments, incorporation was detected in the Thy-1 of all wild-type cells and in two mutants, S1A-b and T1M1-c. Based on the above studies, the phenotype of Thy-1 negative T lymphoma mutants can be re-evaluated. In classes A and F, dolichol-linked oligosaccharides appear normal and no anchor is detected. In class B, dolichol-linked oligosaccharides appear normal, a partial anchor may be present, and a substantial amount of Thy-1 is released. In class C, dolichol-linked oligosaccharides appear normal and a partial anchor may be present. In class E, truncated dolichol-linked oligosaccharides are formed, no anchor is detected, but a substantial amount of newly synthesized Thy-1 is released. These observations are discussed with reference to the possibility that the lesions which characterize the mutants pertain to the biosynthesis of the glycophospholipid moiety of Thy-1.     

Class F Thy-1-negative murine lymphoma cells are deficient in ether lipid biosynthesis

The glycosyl phosphatidylinositol (PI) membrane anchors of several proteins contain 1-alkyl-2-acyl-glycerophosphoinositol. Although this PI analog has never been found free in cells, the presence of "alkyl-PI" as a component of some membrane anchors suggests its existence. The resistance of ether linkages to cleavage by mild alkali treatment was used to detect possible alkyl chains in the [3H]inositol-labeled phospholipids of several murine lymphoma cell lines which normally express the glycosyl PI-anchored protein Thy-1. One lipid, which arose from alkaline hydrolysis of PI and had mobility on thin layer chromatography similar to lyso-PI, was detected in all wild-type cell lines. Analysis of the base-stable inositol lipids of several lymphoma lines that are deficient in Thy-1 surface expression because of defective biosynthesis of the glycosyl PI membrane anchor revealed that the putative alkyl-PI was missing in the class F mutant. The levels of both the ethanolamine- and choline-containing plasmalogens were also decreased 10-fold in these cells, suggesting a general defect in the production of ether lipids. The activity of the peroxisomal form of dihydroxyacetonephosphate acyltransferase, which catalyzes the first step of ether lipid biosynthesis, was found to be 10-fold decreased relative to the wild-type level. Unlike previously described Chinese hamster ovary cell mutants deficient in ether lipids (Zoeller, R. A., and Raetz, C. R. H. (1986) Proc. Natl. Acad. Sci. U. S. A. 83, 5170-5174), the class F Thy-1- cells contain intact functional peroxisomes. Attempts to restore the putative alkyl-PI to the class F mutants by alkylglycerol supplementation were unsuccessful, despite concomitant restoration of the much larger plasmenylethanolamine pool, suggesting that there are some differences in the biosynthesis of this PI analog and plasmalogens that are presently not understood. Although the deficiencies in ether lipids and surface expression of Thy-1 in the class F mutants could also be due to separate mutations, our findings raise the possibility that alkyl-PI exists in animal cells and may be an obligate precursor for the biosynthesis of the glycosyl-PI membrane anchor of Thy-1.     

Hydrophilic anchor-deficient Thy-1 is secreted by a class E mutant T lymphoma

To investigate the mechanism of glycophospholipid anchoring of the surface antigen Thy-1, we have undertaken a comparative biosynthetic study using a wild-type Thy-1+ murine T lymphoma (BW5147) and a mutant T lymphoma (class E) that synthesizes Thy-1 but fails to express it on the plasma membrane. Labelling experiments with D-[2-3H]mannose demonstrate that, unlike the wild type, the mutant cells are secreting large amounts of Thy-1 and that the secreted molecules are hydrophilic. Moreover, unlike the wild type, they fail to incorporate [3H]palmitic acid into Thy-1. Both wild-type and mutant cells do incorporate labeled galactose and fucose into Thy-1. We conclude that the lack of surface expression of Thy-1 by this mutant results from the failure to add anchor components to Thy-1.     

Derivation and characterization of glycoinositol-phospholipid anchor-defective human K562 cell clones.

To aid in studies of human glycoinositol-phospholipid (GPI) anchor pathway biochemistry in normal and affected paroxysmal nocturnal hemoglobinuria cells, GPI anchor-defective human K562 cell lines were derived by negative fluorescent sorting of anti-decay-accelerating factor (DAF) monoclonal antibody-stained cells either following or in the absence of ethylmethylsulfonate pretreatment. The resulting cloned cells showed deficiencies of both DAF and GPI-anchored CD59, some (designated group A) exhibiting total absence and some (designated group B) exhibiting approximately 10% levels of surface expression of the two proteins. In heterologous cell fusions, group A clones complemented defective Thy-1 expression by class A, B, C, E, and I Thy-1-negative lymphoma lines, but not H or D lines, the latter of which is defective in the Thy-1 structural gene. In contrast, group B clones complemented all previously described GPI anchor pathway-defective lymphoma classes. Immunoradiomatic assays of cells and supernatants and 35S biosynthetic labeling showed that group A cells degraded DAF protein while group B cells secreted it but failed to attach a GPI anchor structure. [3H]Man labeling of intact cells and UDP-[3H]GlcNAc and GDP-[3H]Man labeling of broken cell preparations demonstrated that group A cells failed to synthesize GlcNAc- and GlcN-PI (GPI-A and -B) as well as more polar mannolipids, whereas group B cells showed accumulation of GlcNAc-PI with approximately 10-fold diminished levels of GlcN-PI and more polar mannolipids. The failed assembly of GlcNAc-PI in group A cells and the reduced conversion of this intermediate to GlcN-PI in group B cells indicates that the former harbors a defect in UDP-GlcNAc transferase or in assembly of its PI acceptor, while the latter harbors a defect in GlcN-PI deacetylase activity.     

High expression of the Thy-1 antigen on natural killer cells recently derived from bone marrow

The subset of murine natural killer (NK) cells that kills lymphoma targets contains about 50% cells expressing the Thy-1 antigen and this has been one of the reasons for assigning NK cells to the T-cell differentiation lineage. It has now been shown that the proportion of the Thy-1+ NK cells is not constant: ca. 90% of the NK cells appearing in the spleens of irradiated mice injected 10-14 days previously with bone marrow cells (anti-Thy-1 plus complement treated) express this antigen. The donor origin of these Thy-1+ NK cells was demonstrated by using semisyngeneic bone marrow cells in transfers but this same phenomenon could also be observed after entirely syngeneic transfers, excluding the possibilities that this Thy-1+ NK activity is due to activated T cells or to the effect of T-cell activation products on NK cells. Additionally, these early NK cells expressed the asialo-GM1 antigen, which is found on murine NK cells but not on cytotoxic T cells. These data suggest that the precursors for NK cells in the bone marrow are Thy-1-, and that the first splenic NK cells derived from these progenitors express this antigen.     

No glycolipid anchors are added to Thy-1 glycoprotein in Thy-1-negative mutant thymoma cells of four different complementation classes.

Recent evidence shows that the mature Thy-1 surface glycoprotein lacks the C-terminal amino acids 113 to 143 predicted from the cDNA sequence and is anchored in the plasma membrane by a complex, phosphatidylinositol-containing glycolipid attached to the alpha-carboxyl group of amino acid 112. Here we studied the biosynthesis of Thy-1 in two previously described and two newly isolated Thy-1-deficient mutant cell lines. Somatic cell hybridization indicated that their mutations affected some processing step rather than the Thy-1 structural gene. The Thy-1 made by mutants of classes C, F, and H bound detergent but, in contrast to wild-type Thy-1, their detergent-binding moieties could not be removed by phospholipase C. In addition, tryptophan, which only occurs in position 124, was incorporated into Thy-1 of these mutants but not of wild-type cells. Last, the Thy-1 of wild-type but not mutant cells could be radiolabeled with [3H]palmitic acid. Together, these findings strongly suggest that mutants of classes C, F, and H accumulate a biosynthetic intermediate of Thy-1 which retains at least part of the hydrophobic C-terminal peptide. The Thy-1 of these mutants remained endoglycosidase H sensitive, suggesting that it accumulated in the rough endoplasmic reticulum or the Cis-Golgi. A different Thy-1 intermediate was found in a class B mutant cell line: the Thy-1 of this mutant was 2 kilodaltons smaller than the Thy-1 of other cell lines, did not bind detergent, and was rapidly secreted via a normal secretory pathway.     

A monoclonal antibody detecting unusual Thy-1 determinants

20-10-5S is a monoclonal antibody produced by the fusion of C3H anti-C3H.SW splenocytes with the SP2/0 cell line. The antibody appears to react with Thy-1 determinants by several criteria including cytotoxicity patterns, functional assays, genetic analyses, and competitive binding experiments. However, the antibody and the determinants it detects are unusual in that: 1) 20-10-5S is autoreactive; 2) the antibody shows allospecificity for Thy-1.2 vs Thy-1.1 antigens only on peripheral lymphocytes and not on thymocytes; and 3) the antibody reacts only with determinants on murine T cells and not with antigens on brain tissue or on rat thymocytes. It therefore seems that 20-10-5S reacts with murine T cell-specific Thy-1 determinants that are lost or modified during maturation of the cells on which they are expressed.     

Studies on the antigenic determinants of the Thy-1.2 alloantigen as expressed by the murine lymphoblastoid line S-49.1 TB-2-3.

Sialic acid has been shown to be part of the antigenic determinant of the Thy-1.2 alloantigen as expressed by the murine cell line S-49.1 TB-2-3 (S-49). This conclusion is based on the loss of cytotoxic inhibitory activity by the action of neuraminidase on the Thy-1.2 alloantigen. Also, sialic acid has the ability to inhibit the cytotoxic assay of AKR anti-C3H Thy-1.2 serum for S-49 cells. Further evidence for the protein nature of the Thy-1.2 alloantigen is apparent by trypsin digestion. Possibly the Thy-1.2 alloantigen as expressed on S-49 cells is a glycoprotein.     

Somatic genetic analysis of the expression of cell surface molecules

Many mutations that affect the biosynthesis and expression of cell surface molecules are potentially lethal in vivo. Somatic cell genetics provides a means of isolating novel mutants and studying their effects. This approach has been applied to study the genes important in mediating the cell surface expression of the murine Thy-1 glycoprotein, a molecule present on subsets of cells within the hematopoietic system. Nine classes of mutants with no known in vivo counterpart have been identified and studied.     

DNA methylation alters chromatin structure and regulates Thy-1 expression in EL-4 T cells

Thy-1 exhibits marked differences in expression in various tissues in many species; therefore, it is of interest to define possible mechanisms that may regulate Thy-1 expression. We produced Thy-1 negative variants of the murine T cell lymphoma EL-4 by mutagenesis with ethylmethanesulfonate (EMS), negative selection with anti-Thy-1 monoclonal antibodies (mAb) plus complement, and fluorescence-activated cell sorting (FACS). Thy-1 surface negative (Thy-1-) mutants produced in this manner were shown to produce no detectable Thy-1 mRNA, but contained an intact Thy-1 gene as determined by Southern blotting. 5' CG sequences, which had been demethylated in the parent EL-4 clone, were completely methylated in the EMS-induced Thy-1-variant. In addition, a DNase I hypersensitive site that mapped to the 5' end of the Thy-1 gene in EL-4 was absent in the Thy-1- variant. Treatment of this Thy-1- clone with 5-azadeoxycytidine (5-dAZA) resulted in re-expression of surface Thy-1, demethylation of the 5' CG sequences, and regeneration of the DNase I hypersensitive site. These studies indicate that methylation of certain critical DNA sequences in the 5' region of the Thy-1 gene can alter local chromatin structure and regulate expression of this gene.     

Induction and long-term maintenance of Thy-1 positive T lymphocytes: Derivation from continuous bone marrow cultures

In the present study we investigated the presence of T-lymphocyte progenitors in the long-term murine bone marrow culture system described by Dexter: mature Thy-1 antigen-bearing T lymphocytes are lost in these cultures after a few days. By culturing nonadherent cells from such cultures in the presence of a supernatant of concanavalin A-stimulated spleen cells, a source of T-cell growth factor, we found that Thy-1 positive blast cells proliferated together with a second population of Thy-1 negative cells. These two populations of cells have been maintained in long-term in vitro cultures by passaging the cells in fresh conditioned medium at regular intervals. Moreover, we have been able to establish pure cultures of the Thy-1-bearing blast cells after separating them from the non-T cells using their adherence property to plastic surfaces. Long-term cultures of T lymphocytes can thus be established from long-term marrow cultures as well as from the spleen, thymus or fresh bone marrow.     

Pluripotential hematopoietic stem cells in post-5-fluorouracil murine bone marrow express the Thy-1 antigen

Expression of the Thy-1 alloantigen by hematopoietic stem and progenitor cells in post-5-fluorouracil (5-FU) murine bone marrow was investigated. FACS analysis of BDF1 bone marrow stained for Thy-1.2 with a triple-layer amplified labeling technique demonstrated that 35% of the total bone marrow population expressed Thy-1.2 (Thy-1.2+). Two distinct size subpopulations were observed in post-5-FU BDF1 marrow. Thy-1.2+ cells were present in both the large and the small subpopulations. FACS-separated bone marrow cells were also plated in methylcellulose cultures. Ninety percent of all colony-forming cells surviving in vivo administration of 5-FU were Thy-1.2+. Replating of primary hemopoietic colonies and morphologic examination of primary and secondary colonies demonstrated that the most primitive stem cells including "stem" (S) cells were Thy-1.2+. These cells (Thy-1.2+) were capable of self-renewal in vitro and exhibited multiple differentiation potentials in comparison to Thy-1.2-cells, which lacked significant self-renewal capability and were mono- or bipotent progenitor cells. Separation of Thy-1.2+ cells into large or small Thy-1.2+ subpopulations showed that only the large Thy-1.2+ colony-forming cells possessed significant self-renewal capacity. Treatment of BDF1 bone marrow with anti-Thy-1.2 plus complement reduced primary colony formation by 67% and eliminated those colony-forming cells which had extensive self-renewal properties. In the presence of PWMSCM, depletion and reconstitution of T lymphocytes had no effect on primary or secondary colony formation. These data demonstrate that Thy-1 is present on primitive hematopoietic stem cells in post-5-FU bone marrow. In addition, they show that the murine S cell is Thy-1+.     

Identification of defects in glycosylphosphatidylinositol anchor biosynthesis in the Thy-1 expression mutants.

A number of eukaryotic proteins are anchored to the membrane by glycosylphosphatidylinositol (GPI), of which the core structure is conserved from protozoan to mammalian cells. Here, we used a panel of thymoma mutants, which synthesize Thy-1 but cannot express it on the cell surface, to study the GPI biosynthetic pathway in mammalian cells. These mutants have been assigned into six complementation classes (A, B, C, E, F, H) by the technique of somatic cell hybridization. Using a combination of metabolic labeling and chemical/enzymatic tests, the biosynthetic defects were mapped to four different steps. Class A, C, and H mutants cannot transfer N-acetylglucosamine (GlcNAc) to a phosphatidylinositol acceptor, suggesting that the first step of GPI synthesis is regulated by at least three genes. The Class E mutant does not synthesize dolichol-phosphate-mannose, the donor for the first mannose residue transferred to the GPI core, and thus cannot form any mannose-containing GPI precursors. Class B and F mutants are defective in the addition of the third mannose residue or ethanolamine phosphate, respectively, to the elongating GPI core. Our findings have implications for the biosynthesis and attachment of the mammalian GPI anchor.     

A role for the thymic epithelium in the selection of pre-T cells from murine bone marrow

A rat thymic epithelial cell line IT45-R1 has been previously described as secreting soluble molecules that in vitro chemoattract rat hemopoietic precursor cells. The development of such an in vitro migration assay was based on the ability of cells to migrate across polycarbonate filters in Boyden chambers. In the present paper, by using the same strategy, we studied murine bone marrow cells capable of migrating in vitro toward IT45-R1 conditioned medium. The responding cells were shown to represent a minor bone marrow subpopulation characterized by a low capacity to incorporate tritiated thymidine in vitro (less than 10% of control). Moreover, this cell subset was considerably impoverished with respect to granulocyte-macrophage CFU (less than 7% of control) and pluripotent hemopoietic stem cells (less than 12% of control). Potential generation of T cells of donor-type in the lymphoid organs of irradiated recipients was measured by using C57BL/Ka Thy-1.1 and Thy-1.2 congenic mice. Thy-1.1 irradiated mice were injected intrathymically or intravenously with the selectively migrated cell subset of Thy-1.2 donor-type bone marrow cells. The use of an i.v. transfer route allowed us to show that these cells possess thymus-homing and colonization abilities. In a time-course study after intrathymic cell transfer, these migrated cells were able to generate Thy-1.2+ donor-type thymocytes represented by all cortical and medullary cell subsets in a single wave of repopulation from day 20 to day 30 after transfer, with a peak around days 23 to 25. The degree of repopulation closely resembled that seen with unfractionated bone marrow cells in terms of absolute numbers of donor cells per thymus (82% of control, 22 x 10(6) Thy-1.2+ cells) as well as in percent donor cells per thymus (105% of control). Thy-1.2+ cells were also detected in the lymph nodes and the spleens of reconstituted recipient mice. Taken together, these results support the idea that the supernatant of the established thymic epithelium IT45-R1 induces the migration of a murine bone marrow subset that contains hemopoietic stem cells already committed to the lymphoid lineage (i.e., pre-T cells).     

A method for removal of fibroblasts from human tissue culture systems

The phenomenon of fibroblast overgrowth is one of the major problems encountered during long-term culture of more slowly growing specialized cell types. A cell surface glycoprotein, Thy-1, which was originally found to be present on murine T-lymphocytes and brain cells, is also found to be present on only a few human cell types, mainly fibroblasts and neuronal cells. We have taken advantage of this fact, using a solid-phase immunoadsorption technique termed "panning", to rid our culture system (normal human keratinocytes) of contaminating dermal fibroblasts. A mouse monoclonal antibody raised against human brain Thy-1 was used to attach dermal fibroblasts to a goat anti-mouse immunoglobulin-coated plastic surface. By this method we were able to separate a 1:1 mixture of human dermal fibroblasts and keratinocytes with greater than 97.5% efficiency. Furthermore we have successfully removed dermal fibroblasts from naturally arising contaminated keratinocyte cultures, where the proportion of fibroblasts (less than 10%) was considerably less than that of the artificially mixed populations. These results compare favorably with those expected of the fluorescence-activated cell sorter (FACS) method of cell separation. In addition this technique is comparatively simple and inexpensive and is thought to be of use to other primary tissue culture systems (especially human) where contamination and subsequent overgrowth with fibroblasts remains a problem.     

Spontaneous in vitro evolution of lytic specificity of cytotoxic T lymphocyte clones isolated from murine intestinal epithelium

Three long-term clonally derived cytotoxic lines have been established from isolates of murine intraepithelial lymphocytes (IEL). All three lines were selected for with antigen and represent two allospecific cytotoxic T lymphocyte (CTL) clones and a major histocompatibility complex (MHC)-restricted clone specific for a murine minor histocompatibility antigen. On long-term in vitro culture, IEL clones gradually lost antigen-specific lytic activity and simultaneously acquired the capacity to lyse natural killer (NK)-sensitive target cells which, in some cases, required high-level lymphokine activation. Of interest was the finding that, despite changes in lytic specificity, IEL clones remained strictly antigen-dependent for proliferation. A murine CTL clone of splenic origin, which was propagated under culture conditions identical to those used for IEL, did not exhibit changes in lytic specificity, suggesting that acquired changes in IEL function cannot be attributed solely to the influence of in vitro culture. Phenotypic analyses of IEL clones with altered lytic specificity revealed that all lines remained Thy-1+, Lyt-2+, L3T4-, with or without lytic activation by lymphokines. The expression of CT-1, a murine CTL activation antigen, and asialo GM1, a murine NK cell marker, were variable on IEL clones, and their presence did not correlate with the changes in lytic behavior. Collectively, these findings provide evidence, at the clonal level, that at least some NK activity present in isolates of murine IEL may originate from antigen-specific CTL. The data also indicate that, on binding antigen, different signals are conveyed to T cells, resulting in proliferation or target cell lysis.