Tissue-specific knockout of the mouse Pig-a gene reveals important roles for GPI-anchored proteins in skin development

Glycosylphosphatidylinositol (GPI)-anchored proteins are widely distributed on plasma membranes of eukaryotes. More than 50 GPI-anchored proteins have been shown to be spatiotemporally expressed in mice with a deficiency of GPI-anchor biosynthesis that causes embryonic lethality. Here, we examine the functional roles of GPI-anchored proteins in mouse skin using the Cre-loxP recombination system. We disrupted the Pig-a gene, an X-linked gene essential for GPI-anchor biosynthesis, in skin. The Cre-mediated Pig-a disruption occurred in skin at almost 100% efficiency in male mice bearing two identically orientated loxP sites within the Pig-a gene. Expression of GPI-anchored proteins was completely absent in the skin of these mice. The skin of such mutants looked wrinkled and more scaly than that of wild-type mice. Furthermore, histological examination of mutant mice showed that the epidermal horny layer was tightly packed and thickened. Electron microscopy showed that the intercellular space was narrow and there were many small vesicles embedded in the intercellular space that were not observed in equivalent wild-type mouse skin preparations. Mutant mice died within a few days after birth, suggesting that Pig-a function is essential for proper skin differentiation and maintenance.

Exogenous Administration of Gangliosides Displaces GPI-anchored Proteins from Lipid Microdomains in Living Cells

Exogenous application of gangliosides to cells affects many cellular functions. We asked whether these effects could be attributed to the influence of gangliosides on the properties of sphingolipid–cholesterol microdomains on the plasma membrane, also termed rafts. The latter are envisaged as lateral assemblies of sphingolipids (including gangliosides), cholesterol, and a specific set of proteins. Rafts have been implicated in processes such as membrane trafficking, signal transduction, and cell adhesion. Recently, using a chemical cross-linking approach with Madin-Darby canine kidney (MDCK) cells permanently expressing a GPI-anchored form of growth hormone decay accelerating factor (GH-DAF) as a model system, we could show that GPI-anchored proteins are clustered in rafts in living cells. Moreover, this clustering was dependent on the level of cholesterol in the cell. Here we show that incubation of MDCK cells with gangliosides abolished subsequent chemical cross-linking of GH-DAF. Furthermore, insertion of gangliosides into the plasma membrane of MDCK GH-DAF cells renders GH-DAF soluble when subjected to extraction with Triton X-114 at 4°C. Our data suggest that exogenous application of gangliosides displaces GPI-anchored proteins from sphingolipid–cholesterol microdomains in living cells.

A knock-out model of paroxysmal nocturnal hemoglobinuria: Pig-a(-) hematopoiesis is reconstituted following intercellular transfer of GPI-anchored proteins.

We created a "knockout" embryonic stem cell via targeted disruption of the phosphatidylinositol glycan class A (Pig-a) gene, resulting in loss of expression of cell surface glycosyl phosphatidylinositol-anchored proteins and reproducing the mutant phenotype of the human disease paroxysmal nocturnal hemoglobinuria. Morphogenesis of Pig-a- embryoid bodies (EB) in vitro was grossly aberrant and, unlike EB derived from normal embryonic stem cells, Pig-A EB produced no secondary hematopoietic colonies. Chimeric EB composed of control plus Pig-A- cells, however, appeared normal, and hematopoiesis from knock-out cells was reconstituted. Transfer in situ of glycosyl phosphatidylinositol-anchored proteins from normal to knock-out cells was demonstrated by two-color fluorescent analysis, suggesting a possible mechanism for these functional effects. Hematopoietic cells with mutated PIG-A genes in humans with paroxysmal nocturnal hemoglobinuria may be subject to comparable pathophysiologic processes and amenable to similar therapeutic protein transfer.

A Multifaceted Study of Scedosporium boydii Cell Wall Changes during Germination and Identification of GPI-Anchored Proteins

International audience

Two Endoplasmic Reticulum (ER) Membrane Proteins That Facilitate ER-to-Golgi Transport of Glycosylphosphatidylinositol-anchored Proteins

Many eukaryotic cell surface proteins are anchored in the lipid bilayer through glycosylphosphatidylinositol (GPI). GPI anchors are covalently attached in the endoplasmic reticulum (ER). The modified proteins are then transported through the secretory pathway to the cell surface. We have identified two genes in Saccharomyces cerevisiae, LAG1 and a novel gene termed DGT1 (for “delayed GPI-anchored protein transport”), encoding structurally related proteins with multiple membrane-spanning domains. Both proteins are localized to the ER, as demonstrated by immunofluorescence microscopy. Deletion of either gene caused no detectable phenotype, whereas lag1Δ dgt1Δ cells displayed growth defects and a significant delay in ER-to-Golgi transport of GPI-anchored proteins, suggesting that LAG1 and DGT1 encode functionally redundant or overlapping proteins. The rate of GPI anchor attachment was not affected, nor was the transport rate of several non–GPI-anchored proteins. Consistent with a role of Lag1p and Dgt1p in GPI-anchored protein transport, lag1Δ dgt1Δ cells deposit abnormal, multilayered cell walls. Both proteins have significant sequence similarity to TRAM, a mammalian membrane protein thought to be involved in protein translocation across the ER membrane. In vivo translocation studies, however, did not detect any defects in protein translocation in lag1Δ dgt1Δ cells, suggesting that neither yeast gene plays a role in this process. Instead, we propose that Lag1p and Dgt1p facilitate efficient ER-to-Golgi transport of GPI-anchored proteins.

Candidate tumor suppressor HYAL2 is a glycosylphosphatidylinositol (GPI)-anchored cell-surface receptor for jaagsiekte sheep retrovirus, the envelope protein of which mediates oncogenic transformation

Jaagsiekte sheep retrovirus (JSRV) can induce rapid, multifocal lung cancer, but JSRV is a simple retrovirus having no known oncogenes. Here we show that the envelope (env) gene of JSRV has the unusual property that it can induce transformation in rat fibroblasts, and thus is likely to be responsible for oncogenesis in animals. Retrovirus entry into cells is mediated by Env interaction with particular cell-surface receptors, and we have used phenotypic screening of radiation hybrid cell lines to identify the candidate lung cancer tumor suppressor HYAL2/LUCA2 as the receptor for JSRV. HYAL2 was previously described as a lysosomal hyaluronidase, but we show that HYAL2 is actually a glycosylphosphatidylinositol (GPI)-anchored cell-surface protein. Furthermore, we could not detect hyaluronidase activity associated with or secreted by cells expressing HYAL2, whereas we could easily detect such activity from cells expressing the related serum hyaluronidase HYAL1. Although the function of HYAL2 is currently unknown, other GPI-anchored proteins are involved in signal transduction, and some mediate mitogenic responses, suggesting a potential role of HYAL2 in JSRV Env-mediated oncogenesis. Lung cancer induced by JSRV closely resembles human bronchiolo-alveolar carcinoma, a disease that is increasing in frequency and now accounts for ≈25% of all lung cancer. The finding that JSRV env is oncogenic and the identification of HYAL2 as the JSRV receptor provide tools for further investigation of the mechanism of JSRV oncogenesis and its relationship to human bronchiolo-alveolar carcinoma.

The association between glycosylphosphatidylinositol-anchored proteins and heterotrimeric G protein alpha subunits in lymphocytes.

Glycosylphosphatidylinositol (GPI)-anchored proteins are nonmembrane spanning cell surface proteins that have been demonstrated to be signal transduction molecules. Because these proteins do not extend into the cytoplasm, the mechanism by which cross-linking of these molecules leads to intracellular signal transduction events is obscure. Previous analysis has indicated that these proteins are associated with src family member tyrosine kinases; however, the role this interaction plays in the generation of intracellular signals is not clear. Here we show that GPI-anchored proteins are associated with alpha subunits of heterotrimeric GTP binding proteins (G proteins) in both human and murine lymphocytes. When the GPI-anchored proteins CD59, CD48, and Thy-1 were immunoprecipitated from various cell lines or freshly isolated lymphocytes, all were found to be associated with a 41-kDa phosphoprotein that we have identified, by using specific antisera, as a mixture of tyrosine phosphorylated G protein alpha subunits: a small amount of Gialpha1, and substantial amounts of Gialpha2 and Gialpha3. GTP binding assays performed with immunoprecipitations of CD59 indicated that there was GTP-binding activity associated with this molecule. Thus, we have shown by both immunochemical and functional criteria that GPI-anchored proteins are physically associated with G proteins. These experiments suggest a potential role of G proteins in the transduction of signals generated by GPI-anchored molecules expressed on lymphocytes of both mouse and human.

A defect in glycosylphosphatidylinositol (GPI) transamidase activity in mutant K cells is responsible for their inability to display GPI surface proteins.

The final step in the pathway that provides for glycosylphosphatidylinositol (GPI) anchoring of cell-surface proteins occurs in the lumen of the endoplasmic reticulum and consists of a transamidation reaction in which fully assembled GPI anchor donors are substituted for specific COOH-terminal signal peptide sequences contained in nascent polypeptides. In previous studies we described a human K562 cell mutant line, designated class K, which assembles all the known intermediates of the GPI pathway but fails to display GPI-anchored proteins on its surface membrane. In the present study, we used mRNA encoding miniPLAP, a truncated form of placental alkaline phosphatase (PLAP), in in vitro assays with rough microsomal membranes (RM) of mutant K cells to further characterize the biosynthetic defect in this line. We found that RM from mutant K cells supported NH2-terminal processing of the nascent translational product, preprominiPLAP, but failed to show any detectable COOH-terminal processing of the resulting prominiPLAP to GPI-anchored miniPLAP. Proteinase K protection assays verified that NH2-terminal processed prominiPLAP was appropriately translocated into the endoplasmic reticulum lumen. The addition of hydrazine or hydroxylamine, which can substitute for GPI donors, to RM from wild-type or mutant cells defective in various intermediate biosynthetic steps in the GPI pathway produced large amounts of the hydrazide or hydroxamate of miniPLAP. In contrast, the addition of these nucleophiles to RM of class K cells yielded neither of these products. These data, taken together, lead us to conclude that mutant K cells are defective in part of the GPI transamidase machinery.

Arf6-independent GPI-anchored protein-enriched early endosomal compartments fuse with sorting endosomes via a Rab5/phosphatidylinositol-3 '-kinase-dependent machinery

In the process of internalization of molecules from the extracellular milieu, a cell uses multiple endocytic pathways, consequently generating different endocytic vesicles. These primary endocytic vesicles are targeted to specific destinations inside the cell. Here, we show that GPI-anchored proteins are internalized by an Arf6-independent mechanism into GPI-anchored protein-enriched early endosomal compartments (GEECs). Internalized GPI-anchored proteins and the fluid phase are first visualized in GEECs that are acidic, primary endocytic structures, negative for early endosomal markers, Rab4, Rab5, and early endosome antigen (EEA)1. They subsequently acquire Rab5 and EEA1 before homotypic fusion with other GEECs, and heterotypic fusion with endosomes containing cargo from the clathrin-dependent endocytic pathway. Although, the formation of GEECs is unaffected by inhibition of Rab5 GTPase and phosphatidylinositol-3'-kinase (PI3K) activity, their fusion with sorting endosomes is dependent on both activities. Overexpression of Rab5 reverts PI3K inhibition of fusion, providing evidence that Rab5 effectors play important roles in heterotypic fusion between the dynamin-independent GEECs and clathrin- and dynamin-dependent sorting endosomes.

Characterisation of gp34, a GPI-anchored protein expressed by schizonts of Theileria parva and T. annulata

Using bioinformatics tools, we searched the predicted Theileria annulata and T. parva proteomes for putative schizont surface proteins. This led to the identification of gp34, a GPI-anchored protein that is stage-specifically expressed by schizonts of both Theileria species and is downregulated upon induction of merogony. Transfection experiments in HeLa cells showed that the gp34 signal peptide and GPI anchor signal are also functional in higher eukaryotes. Epitope-tagged Tp-gp34, but not Ta-gp34, expressed in the cytosol of COS-7 cells was found to localise to the central spindle and midbody. Overexpression of Tp-gp34 and Ta-gp34 induced cytokinetic defects and resulted in accumulation of binucleated cells. These findings suggest that gp34 could contribute to important parasite-host interactions during host cell division.

Membrane Insertion of C-tail Anchored Proteins

The correct localization of proteins is essential for cell viability. In order to achieve correct protein localization to cellular membranes, conserved membrane targeting and translocation mechanisms have evolved. The focus of this work was membrane targeting and translocation of a group of proteins that circumvent the known targeting and translocation mechanisms, the C-tail anchored protein family. Members of this protein family carry out a wide range of functions, from protein translocation and recognition events preceding membrane fusion, to the regulation of programmed cell death. In this work, the mechanisms of membrane insertion and targeting of two C-tail anchored proteins were studied utilizing in vivo and in vitro methods, in yeast and mammalian cell systems. The proteins studied were cytochrome b(5), a well characterized C-tail anchored model protein, and N-Bak, a novel member of the Bcl-2 family of regulators of programmed cell death. Membrane insertion of cytochrome b(5) into the endoplasmic reticulum membrane was found to occur independently of the known protein conducting channels, through which signal peptide-containing polypeptides are translocated. In fact, the membrane insertion process was independent of any protein components and did not require energy. Instead membrane insertion was observed to be dependent on the lipid composition of the membrane. The targeting of N-Bak was found to depend on the cellular context. Either the mitochondrial or endoplasmic reticulum membranes were targeted, which resulted in morphological changes of the target membranes. These findings indicate the existence of a novel membrane insertion mechanism for C-tail anchored proteins, in which membrane integration of the transmembrane domain, and the translocation of C-terminal fragments, appears to be spontaneous. This mode of membrane insertion is regulated by the target membrane fluidity, which depends on the lipid composition of the bilayer, and the hydrophobicity of the transmembrane domain of the C-tail anchored protein, as well as by the availability of the C-tail for membrane integration. Together these mechanisms enable the cell to achieve spatial and temporal regulation of sub-cellular localization of C-tail anchored proteins.

Transcriptional regulation of the mannosyltransferase-encoding gene pigm in inherited glycosylphosphatidylinositol (GPI) deficiency

RESUMO:O glicosilfosfatidilinositol (GPI) é um complexo glicolipídico utlizado por dezenas de proteínas, o qual medeia a sua ancoragem à superfície da célula. Proteínas de superfície celular ancoradas a GPI apresentam várias funções essenciais para a manutenção celular. A deficiência na síntese de GPI é o que caracteriza principalmente a deficiência hereditária em GPI, um grupo de doenças autossómicas raras que resultam de mutações nos genes PIGA, PIGL, PIGM, PIGV, PIGN, PIGO e PIGT, os quais sao indispensáveis para a biossíntese do GPI. Uma mutação pontual no motivo rico em GC -270 no promotor de PIGM impede a ligação do factor de transcrição (FT) Sp1 à sua sequência de reconhecimento, impondo a compactação da cromatina, associada à hipoacetilação de histonas, e consequentemente, impedindo a transcrição de PIGM. Desta forma, a adição da primeira manose ao GPI é comprometida, a síntese de GPI diminui assim como as proteínas ligadas a GPI à superficie das células. Pacientes com Deficiência Hereditária em GPI-associada a PIGM apresentam trombose e epilesia, e ausência de hemólise intravascular e anemia, sendo que estas duas últimas características definem a Hemoglobinúria Paroxística Nocturna (HPN), uma doença rara causada por mutações no gene PIGA. Embora a mutação que causa IGD seja constitutiva e esteja presente em todos os tecidos, o grau de deficiência em GPI varia entre células do mesmo tecido e entre células de tecidos diferentes. Por exemplo nos granulócitos e linfócitos B a deficiência em GPI é muito acentuada mas nos linfócitos T, fibroblastos, plaquetas e eritrócitos é aproximadamente normal, daí a ausência de hemólise intravascular. Os eventos transcricionais que estão na base da expressão diferencial da âncora GPI nas células hematopoiéticas são desconhecidos e constituem o objectivo geral desta tese. Em primeiro lugar, os resultados demonstraram que os níveis de PIGM mRNA variam entre células primárias hematopoiéticas normais. Adicionalmente, a configuração dos nucleossomas no promotor de PIGM é mais compacta em células B do que em células eritróides e tal está correlacionado com os níveis de expressão de PIGM, isto é, inferior nas células B. A presença de vários motivos de ligação para o FT específico da linhagem megacariocítica-eritróide GATA-1 no promotor de PIGM sugeriu que GATA-1 desempenha um papel regulador na sua transcrição. Os resultados mostraram que muito possivelmente GATA-1 desempenha um papel repressor em vez de activador da expressão de PIGM. Resultados preliminares sugerem que KLF1, um factor de transcrição restritamente eritróide, regula a transcrição de PIGM independentemente do motivo -270GC. Em segundo lugar, a investigação do papel dos FTs Sp demonstrou que Sp1 medeia directamente a transcrição de PIGM em ambas as células B e eritróide. Curiosamente, ao contrário do que acontece nas células B, em que a transcrição de PIGM requer a ligação do FT geral Sp1 ao motivo -270GC, nas células eritróides Sp1 regula a transcrição de PIGM ao ligar-se a montante e não ao motivo -270GC. Para além disso, demonstrou-se que Sp2 não é um regulador directo da transcrição de PIGM quer nas células B quer nas células eritróides. Estes resultados explicam a ausência de hemólise intravascular nos doentes com IGD associada a PIGM, uma das principais características que define a HPN. Por último, resultados preliminares mostraram que a repressão da transcrição de PIGM devida à mutação patogénica -270C>G está associada com a diminuição da frequência de interacções genómicas em cis entre PIGM e os seus genes “vizinhos”, sugerindo adicionalmente que a regulação de PIGM e desses genes é partilhada. No seu conjunto, os resultados apresentados nesta tese contribuem para o conhecimento do controlo transcricional de um gene housekeeping, específico-detecido, por meio de FTs genéricos e específicos de linhagem.-------------ABSTRACTC: Glycosylphosphatidylinositol (GPI) is a complex glycolipid used by dozens of proteins for cell surface anchoring. GPI-anchored proteins have various functions that are essential for the cellular maintenance. Defective GPI biosynthesis is the hallmark of inherited GPI deficiency (IGD), a group of rare autosomal diseases caused by mutations in PIGA, PIGL, PIGM, PIGV, PIGN, PIGO and PIGT, all genes indispensable for GPI biosynthesis. A point mutation in the -270GC-rich box in the core promoter of PIGM disrupts binding of the transcription factor (TF) Sp1 to it, imposing nucleosome compaction associated with histone hypoacetylation, thus abrogating transcription of PIGM. As a consequence of PIGM transcriptional repression, addition of the first mannose residue onto the GPI core and thus GPI production are impaired; and expression of GPI-anchored proteins on the surface of cells is severely impaired. Patients with PIGM-associated IGD suffer from life-threatening thrombosis and epilepsy but not intravascular haemolysis and anaemia, two defining features of paroxysmal nocturnal haemoglobinuria (PNH), a rare disease caused by somatic mutations in PIGA. Although the disease-causing mutation in IGD is constitutional and present in all tissues, the degree of GPI deficiency is variable and differs between cells of the same and of different tissues. Accordingly, GPI deficiency is severe in granulocytes and B cells but mild in T cells, fibroblasts, platelets and erythrocytes, hence the lack of intravascular haemolysis.The transcriptional events underlying differential expression of GPI in the haematopoietic cells of PIG-M-associated IGD are not known and constitute the general aim of this thesis. Firstly, I found that PIGM mRNA levels are variable amongst normal primary haematopoietic cells. In addition, the nucleosome configuration in the promoter of PIGM is more compacted in B cells than in erythroid cells and this correlated with the levels of PIGM mRNA expression, i.e., lower in B cells. The presence of several binding sites for GATA-1, a mega-erythroid lineage-specific transcription factor (TF), at the PIGM promoter suggested that GATA-1 has a role on PIGM transcription. My results showed that GATA-1 in erythroid cells is most likely a repressor rather than an activator of PIGM expression. Preliminary data suggested that KLF1, an erythroid-specific TF, regulates PIGM transcription but independently of the -270GC motif. Secondly, investigation of the role of the Sp TFs showed that Sp1 directly mediates PIGM transcriptional regulation in both B and erythroid cells. However, unlike in B cells in which active PIGM transcription requires binding of the generic TF Sp1 to the -270GC-rich box, in erythroid cells, Sp1 regulates PIGM transcription by binding upstream of but not to the -270GC-rich motif. Additionally, I showed that Sp2 is not a direct regulator of PIGM transcription in B and erythroid cells. These findings explain lack of intravascular haemolysis in PIGM-associated IGD, a defining feature of PNH. Lastly, preliminary work shows that transcriptional repression of PIG-M by the pathogenic -270C>G mutation is associated with reduced frequency of in cis genomic interactions between PIGM and its neighbouring genes, suggesting a shared regulatory link between these genes and PIGM. Altogether, the results presented in this thesis provide novel insights into tissuespecific transcriptional control of a housekeeping gene by lineage-specific and generic TFs.

The affected gene underlying the class K glycosylphosphatidylinositol (GPI) surface protein defect codes for the GPI transamidase

The final step in glycosylphosphatidylinositol (GPI) anchoring of cell surface proteins consists of a transamidation reaction in which preassembled GPI donors are substituted for C-terminal signal sequences in nascent polypeptides. In previous studies we described a human K562 cell mutant, termed class K, that accumulates fully assembled GPI units but is unable to transfer them to N-terminally processed proproteins. In further work we showed that, unlike wild-type microsomes, microsomes from these cells are unable to support C-terminal interaction of proproteins with the small nucleophiles hydrazine or hydroxylamine, and that the cells thus are defective in transamidation. In this study, using a modified recombinant vaccinia transient transfection system in conjunction with a composite cDNA prepared by 5′ extension of an existing GenBank sequence, we found that the genetic element affected in these cells corresponds to the human homolog of yGPI8, a gene affected in a yeast mutant strain exhibiting similar accumulation of GPI donors without transfer. hGPI8 gives rise to mRNAs of 1.6 and 1.9 kb, both encoding a protein of 395 amino acids that varies in cells with their ability to couple GPIs to proteins. The gene spans ≈25 kb of DNA on chromosome 1. Reconstitution of class K cells with hGPI8 abolishes their accumulation of GPI precursors and restores C-terminal processing of GPI-anchored proteins. Also, hGPI8 restores the ability of microsomes from the mutant cells to yield an active carbonyl in the presence of a proprotein which is considered to be an intermediate in catalysis by a transamidase.

Segregation of Heterotrimeric G Proteins in Cell Surface Microdomains: Gq Binds Caveolin to Concentrate in Caveolae, whereas Gi and Gs Target Lipid Rafts by Default

Select lipid-anchored proteins such as glycosylphosphatidylinositol (GPI)-anchored proteins and nonreceptor tyrosine kinases may preferentially partition into sphingomyelin-rich and cholesterol-rich plasmalemmal microdomains, thereby acquiring resistance to detergent extraction. Two such domains, caveolae and lipid rafts, are morphologically and biochemically distinct, contain many signaling molecules, and may function in compartmentalizing cell surface signaling. Subfractionation and confocal immunofluorescence microscopy reveal that, in lung tissue and in cultured endothelial and epithelial cells, heterotrimeric G proteins (Gi, Gq, Gs, and Gβγ) target discrete cell surface microdomains. Gq specifically concentrates in caveolae, whereas Gi and Gs concentrate much more in lipid rafts marked by GPI-anchored proteins (5′ nucleotidase and folate receptor). Gq, apparently without Gβγ subunits, stably associates with plasmalemmal and cytosolic caveolin. Gi and Gs interact with Gβγ subunits but not caveolin. Gi and Gs, unlike Gq, readily move out of caveolae. Thus, caveolin may function as a scaffold to trap, concentrate, and stabilize Gq preferentially within caveolae over lipid rafts. In N2a cells lacking caveolae and caveolin, Gq, Gi, and Gs all concentrate in lipid rafts as a complex with Gβγ. Without effective physiological interaction with caveolin, G proteins tend by default to segregate in lipid rafts. The ramifications of the segregated microdomain distribution and the Gq-caveolin complex without Gβγ for trafficking, signaling, and mechanotransduction are discussed.

Intercellular transfer of a glycosylphosphatidylinositol (GPI)-linked protein: release and uptake of CD4-GPI from recombinant adeno-associated virus-transduced HeLa cells.

A diverse group of GPI-anchored protein structures are ubiquitously expressed on the external cell membranes of eukaryotes. Whereas the physiological role for these structures is usually defined by their protein component, the precise biological significance of the glycolipid anchors remains vague. In the course of producing a HeLa cell line (JM88) that contained a recombinant adeno-associated virus genome expressing a GPI-anchored CD4-GPI fusion protein on the surface of the cells, we noted the transfer of CD4-GPI to native HeLa cells. Transfer occurred after direct cell contact or exposure to JM88 cell supernatants. The magnitude of contact-mediated CD4-GPI transfer correlated with temperature. Supernatant CD4-GPI also attached to human red blood cells and could be cleaved with phosphatidylinositol-specific phospholipase C. The attached CD4-GPI remained biologically active after transfer and permitted the formation of syncytium when coated HeLa cells were incubated with glycoprotein 160 expressing H9 cells. JM88 cells provide a model for the production, release, and reattachment of CD4-GPI and may furnish insight into a physiologic role of naturally occurring GPI-anchored proteins. This approach may also allow the production of other recombinant GPI-anchored proteins for laboratory and clinical investigation.