Subcellular localization of annexin V in human foreskin fibroblasts: nuclear localization depends on growth state

Annexin V is a major intracellular calcium-binding protein in human foreskin fibroblasts. Immunocytochemistry revealed that annexin V was localized in the nucleus and throughout the cytoplasm in human foreskin fibroblasts. The presence of annexin V in the nucleus was variable depending on the growth state. Nuclear staining was strongest in proliferating cells immediately after sub-culture, and decreased on prolonged culture without changing the culture medium. The cytoplasmic location of annexin V was not greatly affected by the same conditions. Refeeding cells with fresh serum restored annexin V to the nuclei of all cells within in 24 h indicating that nuclear localization of annexin V is dependent on serum factors.     

Chlamydia (Chlamydophila) pneumoniae-induced cell death in human coronary artery endothelial cells is caspase-independent and accompanied by subcellular translocations of Bax and apoptosis-inducing factor

Atherosclerosis and coronary heart disease are causing high morbidity and mortality worldwide. Different risk factors have been demonstrated, but the exact mechanisms behind these diseases are still not fully understood. Recent studies have suggested Chlamydia pneumoniae to be involved in the pathogenesis, and increased apoptotic indexes in atherosclerotic plaques have been documented. In this study, we show that C. pneumoniae induces apoptosis and necrosis in populations of human coronary artery endothelial cells. Apoptosis was determined by TUNEL and flow cytometry after staining of cells with annexin V and propidium iodide, and defined as TUNEL-reactive or annexin V-positive, propidium iodide-negative cells. The apoptosis was induced within 2 h postinfection and increased with inoculation dose. The general caspase inhibitor z-VAD-fmk did not affect apoptotic frequencies. By immunochemistry and immunoblot, we demonstrated activation and subcellular translocation of the proapoptotic protein Bax, and translocation of apoptosis-inducing factor from the cytosol to the nucleus. These results indicate that C. pneumoniae-induced apoptosis in human coronary artery endothelial cells is caspase-independent and regulated by Bax and apoptosis-inducing factor.     

Interaction of heparin with annexin V

The energetics and kinetics of the interaction of heparin with the Ca2+ and phospholipid binding protein annexin V, was examined and the minimum oligosaccharide sequence within heparin that binds annexin V was identified. Affinity chromatography studies confirmed the Ca2+ dependence of this binding interaction. Analysis of the data obtained from surface plasmon resonance afforded a Kd of approximately 21 nM for the interaction of annexin V with end-chain immobilized heparin and a Kd of approximately 49 nM for the interaction with end-chain immobilized heparan sulfate. Isothermal titration calorimetry showed the minimum annexin V binding oligosaccharide sequence within heparin corresponds to an octasaccharide sequence. The Kd of a heparin octasaccharide binding to annexin V was approximately 1 microM with a binding stoichiometry of 1:1.     

Inhibition of sperm production in mice by annexin V microinjected into seminiferous tubules: possible etiology of phagocytic clearance of apoptotic spermatogenic cells and male infertility

Many differentiating spermatogenic cells die by apoptosis during the process of mammalian spermatogenesis. However, very few apoptotic spermatogenic cells are detected by histological examination of the testis, probably due to the rapid elimination of dying cells by phagocytosis. Previous in vitro studies showed that Sertoli cells selectively phagocytose dying spermatogenic cells by recognizing the membrane phospholipid phosphatidylserine (PS), which is exposed to the surface of spermatogenic cells during apoptosis. We examined here whether PS-mediated phagocytosis of apoptotic spermatogenic cells occurs in vivo. For this purpose, the PS-binding protein annexin V was microinjected into the seminiferous tubules of normal live mice, and their testes were examined. The injection of annexin V caused no histological changes in the testis, but significantly increased the number of apoptotic spermatogenic cells as assessed by the terminal deoxynucleotidyltransferase-mediated dUTP nick end labeling assay. The number of Sertoli cells did not change in the annexin V-injected testes, and annexin V itself did not induce apoptosis in primary cultured spermatogenic cells. These results indicate that annexin V inhibited the phagocytic clearance of apoptotic spermatogenic cells and suggest that PS-mediated phagocytosis of those cells occurs in vivo. Furthermore, the injection of annexin V into the seminiferous tubules brought about a significant reduction in the number of spermatogenic cells and epididymal sperm in anticancer drug-treated mice. This suggests that the elimination of apoptotic spermatogenic cells is required for the production of sperm.     

Annexins in the Human Neuroblastoma SH-SY5Y: Demonstration of Relocation of Annexins II and V to Membranes in Response to Elevation of Intracellular Calcium by Membrane Depolarisation and by the Calcium Ionophore A23187

Abstract: The human neuroblastoma SH-SY5Y was found to express annexins I, II, IV, V, and VI by western blot analysis. Calcium-dependent membrane-binding proteins were isolated from SH-SY5Y and analysed by 2-dimensional gel electrophoresis. Proteins with M_r and pl values similar to those of annexins I, II, III, IV, V, and VI were observed. The identity of annexins II and V was confirmed by western blotting. The membrane association of annexins II and V was studied in cells that had been stimulated to release noradrenaline by K⁺ depolarisation or by treatment with the ionophore A23187. Annexins II and V were both found to associate with membranes in a manner that was resistant to elution with EGTA and required Triton X-100 for their solubilisation. Homogenisation of cells in calcium-containing buffers also resulted in the formation of EGTA-resistant membrane-associated annexins II and V. The results demonstrate calcium-dependent relocation of annexins II and V to membranes in intact cells and suggest that these annexins bind in a calcium-dependent manner to non-phospholipid components of SH-SY5Y membranes. Examination of cells by immunofluorescence microscopy demonstrated that annexin II was homogeneously associated with the plasma membrane before treatment with ionophore and relocated to discrete patches of staining after treatment. Annexin V was found by immunofluorescence to be present in the cytoplasm and in the nucleus. Stimulation of the cells produced no change in the cytoplasmic staining pattern but resulted in a partial relocation of nuclear annexin V to the periphery of the nucleus. The results argue for a general role for both annexins in calcium signalling at discrete intracellular locations. The results are not consistent with the specific involvement proposed previously for annexin II in membrane fusion at sites of vesicle exocytosis.     

Multi-modality imaging identifies key times for annexin V imaging as an early predictor of therapeutic outcome

Radiolabeled annexin V may provide an early indication of the success or failure of anticancer therapy on a patient-by-patient basis as an in vivo marker of tumor cell killing. An important question that remains is when, after initiation of treatment, should annexin V imaging be performed. To address this issue, we obtained simultaneous in vivo measurements of tumor burden and uptake of radiolabeled annexin V in the syngeneic orthotopic murine BCL1 lymphoma model using in vivo bioluminescence imaging (BLI) and small animal single-photon emission computed tomography (SPECT). BCL1 cells labeled for fluorescence and bioluminescence assays (BCL1-gfp/luc) were injected into mice at a dose that leads to progressive disease within two to three weeks. Tumor response was followed by BLI and SPECT before and after treatment with a single dose of 10 mg/kg doxorubicin. Biodistribution analyses revealed a biphasic increase of annexin V uptake within the tumor-bearing tissues of mice. An early peak occurring before actual tumor cells loss was observed between 1 and 5 hr after treatment, and a second longer sustained rise from 9 to 24 hr after therapy, which heralds the onset of tumor cell loss as confirmed by BLI. Multimodality imaging revealed the temporal patterns of tumor cell loss and annexin V uptake revealing a better understanding of the timing of radiolabeled annexin V uptake for its development as a marker of therapeutic efficacy.     

Regulation of nucleo-cytoplasmic shuttling of human annexin A2—a proposed mechanism

Studies have long been focused on the functions of annexin A2 in the cytoplasm. However, the involvement of annexin A2 in DNA replication as a part of primer recognition protein complex and the presence of nuclear export signal (NES) suggest that annexin A2 is also functional in the nucleus, and its localization in the nucleus is under regulation by interaction with other nuclear factors through its N-terminus. During the study of the mechanism of annexin A2 sequestering in the nucleus and the regulation of its export from the nucleus, in this study, we show that endogenous annexin A2 is present in both the cytoplasm and the nucleus in HeLa, PC-3 and DU-145 cells. While exogenously expressed annexin A2 is excluded from nuclei of annexin A2-null LNCaP cells in a CRM1 (Chromosome Maintenance Region 1) mediated nuclear export, endogenous annexin A2 in HeLa, PC-3 and DU-145 cell lines does not undergo the CRM1 mediated nuclear export. While investigating the mechanism of the nuclear retention of annexin A2, we found that an anti-annexin A2 antibody that recognizes the C-terminus of annexin A2 (D1/274.5) cannot recognize nuclear annexin A2, suggesting that the domain recognized by this antibody may be masked in the nuclei. In order to find out the role of annexin A2 C-terminus in the nuclear retention of annexin A2, we transiently transfected green fluorescence protein (GFP)-fused N-terminal 29 amino acids of annexin A2 to LNCaP, PC-3 and DU-145 cells, and determined that the C-terminus is not required for the nuclear retention of annexin A2. Based on the finding described above, we propose a model for nuclear retention of annexin A2 where the regulation sites reside in the N-terminus and are adjacent to the NES, and upon modification, the NES is exposed and annexin A2 is exported from the nucleus. Electronic Supplementary Material The online version of this article (doi) contains supplementary material, which is available to authorized users.     

Differential expression of annexin V during spermatogenesis in the newt Cynops pyrrhogaster

 In order to isolate genes whose expression is up-regulated after the initiation of meiosis, we screened a cDNA expression library of newt testes with antiserum against homogenates of testes derived from the spermatogonial and spermatocyte stages. We report the isolation of spermatocyte-specific cDNA clones encoding a newt homologue of the calcium-dependent phospholipid-binding protein, annexin V. Northern blot analysis showed that newt annexin V mRNA was 1.7 kb in length and was expressed strongly in testes, but weakly in other organs. In situ hybridization revealed that the expression of newt annexin mRNA was barely observed in spermatogonia, but increased significantly in leptotene-zygotene primary spermatocytes and reached a maximum level in pachytene spermatocytes and round spermatids. The newt annexin V cDNA predicted a 323-amino acid protein and had a 68% homology to human annexin V. The predicted amino acid sequence contained a conserved 4-fold internal repeat of approximately 70 residues like other annexin proteins. Immunoblot analysis using the monoclonal antibody against newt annexin V showed that the protein was expressed scarcely in spermatogonia but was abundantly expressed in stages from primary spermatocytes to spermatids; this pattern was consistent to that of the mRNA. Immunohistochemical analysis revealed that newt annexin V was localized in the cytoplasm of the spermatogenic cells, but not in somatic cells such as Sertoli cells or pericystic cells. These results indicate that the expression of newt annexin V is up-regulated in the spermatogenic cells after the initiation of meiosis and suggest that newt annexin V plays an important role in spermatogenesis. Received: 8 December 1995 / Accepted: 12 February 1996 Edited by H. Shimada/D. Tautz     

A novel 96-well scintillation proximity assay for the measurement of apoptosis

The translocation of phospholipids across the plasma membrane has been widely documented as one of the earliest measurable biochemical events of apoptosis. Using fluorescently labelled annexin V, which preferentially binds phosphatidylserine (PS) in the presence of Ca²⁺, the externalization of PS can be measured and apoptosis quantified using flow cytometry. Conventional detection methods utilizing annexin V, while faster than in situ DNA end-labelling or DNA laddering, require extensive sample preparation which may compromise samples and makes rapid, high volume screening prohibitive. This paper describes a novel assay for the measurement of apoptosis based upon binding of radiolabelled annexin V to apoptotic cells attached to the growth surface of a 96-well scintillating microplate (Cytostar-T®). We compared measurements of apoptosis made by flow cytometry to those obtained with the scintillating microplate in three model systems, treatment of: mouse connective tissue (L-M) cells with lymphotoxin (LT), human lung carcinoma (H460) cells with Apo-2 ligand and human umbilical vein endothelial (HUVE) cells with staurosporine. In this assay, we compare both direct and indirect labelling methods by utilizing either iodinated annexin V or biotinylated annexin V/[³⁵S] streptavidin to radiolabel apoptotic cells. The signal detected is a direct consequence of the binding of annexin V to externalized PS on apoptotic cells and the proximity of the label to the base of the plate. Using this method, separation of bound and unbound radiolabel signal occurs directly within the well resulting in a sensitive assay that requires minimal manipulation and can accomodate a large number of samples.     

Flow cytometric estimation of the plasma membrane diversity of transplantable melanomas, using annexin V

Using a recently described flow cytometric assay probing for cell surface exposure of phosphatidylserine with fluoresceine-labeled annexin V, we attempted to establish if there existed any differences in the phospholipid bilayer of the plasma membranes of melanoma cells isolated from two lines of a hamster transplantable melanoma characterized by a common origin but differing in many biological features. In contrast to control nonstaining cells, the cells of both melanoma lines bound annexin V, but at a different rate: 88% of melanotic and 94% of amelanotic melanoma cells were annexin V positive. Among cells of the native melanotic melanoma line we distinguished only one cell population binding annexin but in some experiments with the amelanotic melanoma we observed two annexin V positive cell populations with a different fluorescence intensity. It is possible that these differences in annexin V binding to melanoma cell membranes reflect some changes in the phospholipid bilayer, associated with the progression of these tumors.     

The annexins of Dictyostelium

Annexins are a highly conserved ubiquitous family of Ca2+- and phospholipid-binding proteins present in nearly all eukaryotic cells. Analysis of the Dictyostelium genome revealed the presence of two annexin genes, the annexin C1 gene (nxnA) giving rise to two isoforms of 47 and 51 kDa (previously synexin), and the annexin C2 gene (nxnB) coding for a 56-kDa protein with 33% sequence identity to annexin C1. Annexin C2 is expressed at very low and constant levels throughout development. Quantification by real-time PCR indicated that it is present in about 35-fold lower amounts compared to annexin C1. We have used a GFP-tagged annexin C2 to study its cellular distribution and dynamics. In cell fractionation studies, annexin C2 cofractionates with annexin C1 and is enriched in the 100,000 g pellet. Like annexin C1, GFP-AnxC2 stains the plasma membrane. In addition it is present in the perinuclear region and overlaps to some degree with the Golgi apparatus, whereas annexin C1 is present on intracellular membranes resembling endosomal membranes and in the nucleus. Annexin C2 is not observed in the nucleus. An annexin C1 mutant (SYN-) which shows a defect during multicellular development can be rescued by full-length annexin C1, whereas overexpression of GFP-AnxC2 did not rescue the developmental defect The data support the concept that annexins, although having a highly conserved structure, participate in different functions in a cell.     

Surfactant protein A (SP-A) binds to phosphatidylserine and competes with annexin V binding on late apoptotic cells

The role of surfactant protein A (SP-A) in the recognition and clearance of apoptotic cells is well established, but to date, it is still not clear which surface molecules of apoptotic cells are involved in the process. Here we present evidence that phosphatidylserine (PS) is a relevant binding molecule for human SP-A. The binding is Ca²⁺-dependent and is not inhibited by mannose, suggesting that the sugar-binding site of the carbohydrate recognition domain (CRD) of SP-A is not involved. Flow cytometry studies on apoptotic Jurkat cells revealed apparent inhibition of annexin V binding by increasing concentrations of SP-A in late apoptotic but not early apoptotic cells, and this was consistent for Jurkat cells and neutrophils. Supporting these data, confocal microscopy results show a co-localisation of annexin V and SP-A in late apoptotic but not early apoptotic cells. However, we cannot conclude that this inhibition is exclusively due to the binding of SP-A to PS on the cell surface, as annexin V is not wholly specific for PS and SP-A also interacts with other phospholipids that might become exposed on the apoptotic cell surface.     

Low temperature promotes annexin V expression in newt testis

We examined the effect of low temperatures on annexin V expression in newt testis. When newts were transferred to a low temperature (12 degrees C), up-regulation of annexin V protein was observed in secondary spermatogonia. In primary spermatocytes, high levels of annexin V expression were observed at both 12 degrees C and 22 degrees C, but at 12 degrees C the protein was localized in part of the cytoplasm of primary spermatocytes. These results indicate that in newt testis annexin V is a cold-sensitive protein, suggesting the possibility that annexin V might have a cold stress-related function in newt germ cells.     

Protein kinase C delta (deltaPKC)-annexin V interaction: a required step in deltaPKC translocation and function

Protein kinase C (PKC) plays a critical role in diseases such as cancer, stroke, and cardiac ischemia, and participates in a variety of signal transduction pathways such as apoptosis, cell proliferation, and tumor suppression. Though much is known about PKC downstream signaling events, the mechanisms of regulation of PKC activation and subsequent translocation have not been elucidated. Protein-protein interactions regulate and determine the specificity of many cellular signaling events. Such a specific protein-protein interaction is described here between deltaPKC and annexin V. We demonstrate, at physiologically relevant conditions, that a transient interaction between annexin V and deltaPKC occurs in cells after deltaPKC stimulation, but before deltaPKC translocates to the particulate fraction. Evidence of deltaPKC-annexin V binding is provided also by FRET and by in vitro binding studies. Dissociation of the deltaPKC-annexin V complex requires ATP and microtubule integrity. Furthermore, depletion of endogenous annexin V, but not annexin IV, with siRNA inhibits deltaPKC translocation following PKC stimulation. A rationally designed eight amino acid peptide, corresponding to the interaction site for deltaPKC on annexin V, inhibits deltaPKC translocation and deltaPKC-mediated function as evidenced by its protective effect in a model of myocardial infarction. Our data indicate that translocation of deltaPKC is not simply a diffusion-driven process, but is instead a multi-step event regulated by protein-protein interactions. We show that following cell activation, deltaPKC-annexin V binding is a transient and an essential step in the function of deltaPKC, thus identifying a new role for annexin V in PKC signaling and a new step in PKC activation.     

Magneto/optical annexin V, a multimodal protein

Multimodal proteins, or proteins labeled with both fluorescent and magnetic reporter groups, can be used in a wide range of applications including FACS or fluorescence microscopy, MRI and or near-infrared based optical imaging, or to fractionate cells by magnetic cell sorting. A problem with multimodal proteins, however, is the need to maximize bioactivity, often achieved by minimizing the number of modification points of the protein, while attaching fluorescent and magnetic labels. Here we describe the synthesis of a magneto/optical form of annexin V, achieved by reacting the amino-CLIO nanoparticle with Cy5.5 and SPDP, to produce a fluorescent, sulfhydryl reactive nanoparticle. A single reactive sulfhydryl group was added to annexin V by reaction with SATA that preserved the protein's ability to bind apoptotic Jurkat T cells. Reacting SATAylated annexin V with an SPDP activated nanoparticle yielded Anx-CLIO-Cy5.5, a magneto/optical form of annexin V. The binding of Anx-CLIO-Cy5.5 was specific for apoptotic Jurkat T cells and had an EC(50) of 3.66 nM. This was comparable to the strength of the interaction of unmodified annexin V with apoptotic cells, measured as the displacement of FITC-annexin by annexin V (2.4 nM). Our conjugation strategy preserves the strength of the interaction between annexin V and apoptotic cells, while yielding a probe, Anx-CLIO-Cy5.5, that is readily detectable by standard MR imaging or NIRF optical methods.     

Preapoptotic phenotype of viral epitope-specific CD8 T cells precludes memory development and is an intrinsic property of the epitope

Virus-specific CD8 T cells after clearance of infection reduce their number in lymphoid organs by apoptotic death and by migration into peripheral tissues. During and after infection, many lymphocytic choriomeningitis virus (LCMV)-specific CD8 T cells in lymphoid but not peripheral tissues are in a preapoptotic state, as detected by the early apoptosis marker annexin V. In this report, we investigated the significance of this preapoptotic state and how it may be influenced by viral epitope specificity. Stimulation with anti-CD3 or IL-2 in vitro postponed DNA fragmentation in annexin V+ cells, but adoptive transfer studies in vivo showed that this preapoptotic phenotype precluded the development of functional memory. CD8 T cells specific to LCMV epitopes NP396 and gp33 differed in their preapoptotic state, with NP396-specific T cells binding more annexin V than gp33-specific T cells. These epitope- and tissue-dependent differences were seen in primary, memory, and secondary responses and in mice receiving different displays of Ag by infection with LCMV strains of different tropisms or by infection with vaccinia virus recombinants expressing LCMV proteins. Thus, the epitope-dependent differences in apoptosis were independent of virus tropisms, duration of Ag exposure, and competition within APCs, and were an intrinsic property of the epitope. The tissue-dependent and epitope-dependent preapoptotic state correlated with reduced expression of IL-7Ralpha.     

Lactadherin detects early phosphatidylserine exposure on immortalized leukemia cells undergoing programmed cell death.

BACKGROUND: Phosphatidylserine (PS) appears on the outer membrane leaflet of cells undergoing programmed cell death and marks those cells for clearance by macrophages. Macrophages secrete lactadherin, a PS-binding protein, which tethers apoptotic cells to macrophage integrins. METHODS: We utilized fluorescein-labeled lactadherin together with the benchmark PS Probe, annexin V, to detect PS exposure by flow cytometry and confocal microscopy. Immortalized leukemia cells were treated with etoposide, and the kinetics and topology of PS exposure were followed over the course of apoptosis. RESULTS: Costaining etoposide-treated leukemoid cells with lactadherin and annexin V indicated progressive PS exposure with dim, intermediate, and bright staining. Confocal microscopy revealed localized plasma membrane staining, then diffuse dim staining by lactadherin prior to bright generalized staining with both proteins. Annexin V was primarily localized to internal cell bodies at early stages but stained the plasma membrane at the late stage. Calibration studies suggested a PS content less, less than or approximately equal to 2.5%-8% for the membrane domains that stained with lactadherin but not annexin V. CONCLUSIONS: Macrophages may utilize lactadherin to detect PS exposure prior to exposure of sufficient PS to bind annexin V. The methodology enables detection of PS exposure at earlier stages than established methodology.     

Inhibition of protein kinase C by annexin V.

Annexin V is a protein of unknown biological function that undergoes Ca(2+)-dependent binding to phospholipids located on the cytosolic face of the plasma membrane. Preliminary results presented herein suggest that a biological function of annexin V is the inhibition of protein kinase C (PKC). In vitro assays showed that annexin V was a specific high-affinity inhibitor of PKC-mediated phosphorylation of annexin I and myosin light chain kinase substrates, with half-maximal inhibition occurring at approximately 0.4 microM. Annexin V did not inhibit epidermal growth factor receptor/kinase phosphorylation of annexin I or cAMP-dependent protein kinase phosphorylation of the Kemptide peptide substrate. Since annexin V purified from both human placenta and recombinant bacteria inhibited protein kinase C activity, it is not likely that the inhibitor activity was associated with a minor contaminant of the preparations. The following results indicated that the mechanism of inhibition did not involve annexin V sequestration of phospholipid that was required for protein kinase C activation: similar inhibition curves were observed as phospholipid concentration was varied from 0 to 800 micrograms/mL; the extent of inhibition was not significantly affected by the order of addition of phospholipid, substrate, or PKC, and the core domain of annexin I was not a high-affinity inhibitor of PKC even though it had similar Ca2+ and phospholipid binding properties as annexin V. These data indirectly indicate that inhibition occurred by direct interaction between annexin V and PKC. Since the concentration of annexin V in many cell types exceeds the amounts required to achieve PKC inhibition in vitro, it is possible that annexin V inhibits PKC in a biologically significant manner in intact cells.     

Bisecting GlcNAc mediates the binding of annexin V to Hsp47

The bisecting N-acetylglucosamine (GlcNAc) structure, formed through catalysis by UDP-N-acetylglucosamine : beta-D-mannoside beta-1,4-N-acetylglucosaminyltansferase III (GnT-III), is responsible for a variety of biological functions. We have previously shown that annexin V, a member of the calcium/phospholipid-binding annexin family of proteins, has binding activity toward the bisecting GlcNAc structure. In this study, we reported on a search for potential target glycoproteins for annexin V in a rat hepatoma cell line, M31. Using a glutathione S-transferase (GST)-annexin V immobilized sepharose 4B affinity column to trap interacting proteins produced by the GnT-III-transfected M31 cells, we isolated a 47 kDa protein. It was identified as Hsp47 by an N-terminal sequence analysis. Immunoprecipitation experiments showed that annexin V interacted with Hsp47. The association of annexin V and Hsp47 was abolished by treatment with N-glycosidase F or preincubation with sugar chains containing bisecting GlcNAc, suggesting that the bisecting GlcNAc plays an important role in the interaction. An oligosaccharide analysis of Hsp47 purified from GnT-III-transfected M31 cells was shown to have the bisecting GlcNAc structure, as detected by erythroagglutinating phytohemagglutinin (E4-PHA) and matrix assisted laser desorption/ionization-time of flight (MALDI-TOF) mass spectrometry (MS) analysis. Surface plasmon resonance analysis showed that annexin V was bound to Hsp47, bearing a bisecting GlcNAc with a Kd of 5.5 microM, whereas no significant binding was observed in the case of Hsp47 without a bisecting GlcNAc. In addition, immunofluorescence microscopy revealed the colocalization of annexin V, Hsp47, and a bisecting GlcNAc sugar chain around the Golgi apparatus. Collectively, these results suggest that the binding of annexin V to Hsp47 is mediated by a bisecting GlcNAc oligosaccharide structure and that Hsp47 is an intracellular ligand glycoprotein for annexin V.