`Vol. 93, pp. 250-254, January 1996
`Medical Sciences
`
`Isolation and characterization of a cell surface albumin-binding
`protein from vascular endothelial cells
`(endothelial cell surface/gp60)
`
`CHINNASWAMY TIRUPPATHI*t, ALISON FINNEGANt, AND ASRAR B. MALIK*
`Departments of *Pharmacology and 'Immunology, Rush-Presbyterian-St. Luke's Medical Center/Rush Medical College, Chicago, IL 60)612-3824
`
`Communicated by Ilar Giaever, Rensselear Polytechnic Institute, Troy, NY, August 31, 1995
`
`Albumin-binding proteins identified in vas-
`ABSTRACT
`cular endothelial cells have been postulated to contribute to
`the transport of albumin via a process involving transcytosis.
`In the present study, we have purified and characterized a 57-
`to 60-kDa (gp6O) putative albumin-binding protein from
`bovine pulmonary microvessel endothelial cells. The endothe-
`lial cell membranes were isolated from cultured cells by
`differential centrifugation and solubilized with sodium
`cholate and urea. The solubilized extract was concentrated
`after dialysis by ethanol precipitation and reextracted with
`Triton X-100, and the resulting extract was subjected to
`DEAE-cellulose column chromatography. Proteins eluted
`from this column were further separated using preparative
`sodium dodecyl sulfate/polyacrylamide gel electrophoresis
`and used for immunizing rabbits. Fluorescence-activated cell
`sorter analysis using the anti-gp6O antibodies demonstrated
`the expression of gp6O on the endothelial cell surface. Affinity-
`purified anti-gp6O antibodies inhibited -90% of the specific
`binding of '251-labeled albumin to bovine pulmonary mi-
`crovessel endothelial cell surface. The anti-gp6O antibodies
`reacted with gp6O from bovine pulmonary artery, bovine
`pulmonary microvessel, human umbilical vein, and rat lung
`endothelial cell membranes. Bovine anti-gp6O antibodies also
`reacted with bovine secreted protein, acidic and rich in
`cysteine (SPARC). However, bovine SPARC NH2-terminal
`sequence (1-56 residues) antibodies did not react with gp6O,
`indicating that the endothelial cell-surface-associated albu-
`min-binding protein gp6O was different from the secreted
`albumin-binding protein SPARC. We conclude that the en-
`dothelial cell-surface-associated gp6O mediates the specific
`binding of native albumin to endothelial cells and thus may
`regulate the uptake of albumin and its transcytosis.
`
`The functions of albumin include the delivery of bound ligands
`such as hormones and fatty acids and the maintenance of
`vascular integrity and transvascular oncotic pressure gradient
`(1-4). A number of morphological studies have indicated that
`transcytosis of albumin in vascular endothelial cells can occur
`by a receptor-mediated process (5-8). Some studies have
`suggested that this is a predominant mode of albumin transport
`(6-8). Three major albumin-binding proteins (ABPs; 18, 31,
`and 56-60 kDa) have been identified in vascular endothelial
`cells using the ligand-blotting, photochemical cross-linking,
`and lectin-binding assays (8-10). Recent studies have shown
`that 18- and 31-kDa polypeptides present in vascular endo-
`thelial cells may be similar to the scavenger receptors identified
`on other cell types (11, 12). Schnitzer et al. (10) showed specific
`binding of native albumin with a 56- to 60-kDa (gp6O) polypep-
`tide in rat microvascular endothelial cells using lectin-binding
`analysis. Lectins such as Limax flavus agglutinin and Ricinus
`commutnis agglutinin (RCA) inhibited albumin binding and
`precipitated gp6O from rat microvascular endothelial cells
`
`The publication costs of this article were defrayed in part by page charge
`payment. This articlc must therefore be hereby marked "advertisement" in
`accordance with 18 U.S.C. §1734 solely to indicate this fact.
`
`(10). Studies from our laboratory showed that transendothelial
`albumin transport was inhibited by 40% in the presence of
`RCA but not by control lectins applied to bovine pulmonary
`artery endothelial cell (BPAEC) monolayers (13). RCA also
`precipitated gp6O from BPAECs (13). These results suggest a
`role for gp6O in albumin transport. Studies have shown that
`a-glycophorin (14) and secreted protein, acidic and rich in
`cysteine (SPARC), antibodies (15) also cross-react with gp6O.
`We report the purification and characterization of gp6O from
`cultured vascular endothelial cells. Some of these observations
`have been elsewhere reported in an abstract (16).
`
`MATERIALS AND METHODS
`Materials. Dulbecco's modified essential medium (DMEM)
`and fetal bovine serum (FBS) were from GIBCO. Globulin-
`free bovine serum albumin (BSA), phenylmethylsulfonyl flu-
`oride (PMSF), and protein A-Sepharose were from Sigma.
`Sodium cholate and Triton X-100 were from Calbiochem.
`DEAE-cellulose was from Whatman. All electrophoretic
`chemicals were from Bio-Rad. Polyclonal antibodies raised
`against BSA were from ICN. LF-BON-1 antiserum (anti-
`bovine SPARC) and LF-56 antiserum (against the NH2-
`terminal amino acid residues 1-56 of bovine SPARC) were
`generous gifts from L. W. Fisher (National Institutes of
`Health, Bethesda, MD). Preparation of LF-56 antiserum has
`been described (17).
`Cell Culture. Bovine pulmonary microvessel endothelial
`cells (BPMVECs), BPAECs, and rat pulmonary artery endo-
`thelial cells were isolated and cultured as described (18, 19).
`Human pulmonary artery endothelial cells (HPAECs) and
`human umbilical vein endothelial cells (HUVECs) were from
`Clonetics (San Diego). HUVECs were grown in RPMI 1640
`medium supplemented with 10% FBS, 90 ,ig of heparin per ml,
`2 mM L-glutamine, and 30 jig of endothelial cell growth factor
`per ml. HPAECs were grown in MDCB 131 medium supple-
`mented with 10% FBS, 10 ng of human epidermal growth
`factor per ml, and 1 ,ug of hydrocortisone per ml. For isolating
`endothelial cell membranes, endothelial cells were cultured in
`850-cm2 roller bottles. To each roller bottle, 75 ml of culture
`medium was added and filled with an air/CO2 mixture. The
`cells were then transferred to a 37°C roller bottle incubator and
`allowed to grow for 10-12 days.
`Endothelial Cell Membrane Isolation. BPMVECs grown in
`roller bottles were washed twice with phosphate-buffered
`saline (PBS). Cells were scraped from roller bottles, suspended
`
`Abbreviations: BPMVEC, bovine pulmonary microvessel endothelial
`cell; BPAEC, bovine pulmonary artery endothelial cell; HPAEC,
`human pulmonary artery endothelial cell; HUVEC, human umbilical
`vein endothelial cell; ABP, albumin-binding protein; BSA, bovine
`serum albumin; SPARC, secreted protein, acidic and rich in cysteine;
`FACS, fluorescence-activated cell sorter; PVDF, poly(vinylidene di-
`fluoride).
`tTo whom reprint requests should be addressed at: Department of
`Pharmacology, Rush-Presbyterian-St. Luke's Medical Center/Rush
`Medical College, 1725 West Harrison, Chicago, IL 60612-3824.
`
`250
`
`Page 1
`
`KASHIV EXHIBIT 1060
`IPR2019-00791
`
`
`
`Medical Sciences: Tiruppathi et al.
`in buffer A (20 mM Hepes/Tris/0.15 M NaCl/0.1 mM PMSF,
`pH 7.4), and washed twice by centrifugation (700 x g, 10 min).
`Cells from six to eight roller bottles were suspended in 75 ml
`of buffer A and homogenized using a Polytron for 2 min at full
`speed. The homogenate was centrifuged (3000 x g, 10 min).
`The supernatant was collected and centrifuged (100,000 x g,
`60 min). The pellet obtained was then suspended in buffer A
`and recentrifuged (100,000 x g, 60 min). The final membrane
`pellet was suspended in a small volume of buffer A containing
`0.2 mM EDTA. The protein concentration of the membrane
`was determined; the preparation was stored at -70°C until
`further use.
`Ligand Blotting. Endothelial cell membranes were preincu-
`bated with 1 mM PMSF/0.5 mM EDTA (20 min, 22°C),
`solubilized by mixing with 1.5 vol of solubilizing buffer (9 M
`urea/2% SDS/2% 2-mercaptoethanol/0.1 M Tris/0.02% bro-
`mophenol blue, pH 6.8), and incubated at 22°C for 30 min. The
`solubilized proteins were separated by SDS/PAGE according
`to Laemmli (20) using the slab-gel electrophoretic system (3%
`acrylamide in the stacking gel, 10% acrylamide in the sepa-
`rating gel). After electrophoresis, the proteins were trans-
`ferred to either a poly(vinylidene difluoride) (PVDF) or a
`nitrocellulose membrane. The nonspecific binding was blocked
`by incubating the membrane with 5 mM CaCl2 in TBS (20 mM
`Tris/0.5 M NaCl, pH 7.5) for 10 min and then with 0.5% Tween
`20 in TBS overnight. After this step, the membrane was washed
`and cut into two strips. One strip was incubated with 0.6 mg of
`globulin-free BSA per ml in TBS containing 1.5% gelatin for
`2 hr, and the other strip was incubated without BSA. The strips
`were washed and incubated with anti-bovine BSA antibodies
`for 60 min in TBS containing 1.5% gelatin. The membranes
`were then washed twice and incubated with a second antibody
`(goat anti-rabbit IgG) conjugated with alkaline phosphatase.
`The protein bands were localized after adding 5-bromo-4-
`chloro-3-indolyl phosphate and nitroblue tetrazolium salt.
`We also used 125I-labeled monomeric BSA (125I-BSA) in
`ligand-blotting experiments to identify gp6O. In this case,
`nonspecific binding was blocked with bovine y-globulin (BSA
`free, 2 mg/ml in TBS) and then incubated with 125I-BSA (50
`,ug/ml) for 2 hr. The strips were washed with TBS containing
`0.05% Tween 20 and autoradiography was performed.
`Purification of gp6O. BPMVEC membranes (100 mg) were
`preincubated with 1 mM PMSF/0.5 mM EDTA (30 min,
`22°C). The membranes were solubilized using final concen-
`tration of 2.5% sodium cholate and 4 M urea (4°C, 3 hr) with
`gentle stirring. The protein concentration was adjusted to 4
`mg/ml during solubilization. After this treatment, the suspen-
`sion was centrifuged (100,000 x g, 60 min). The supernatant
`was collected and dialyzed against 5 mM Hepes/Tris (pH 7.2).
`More than 80% of membrane proteins were recovered in the
`supernatant. The dialyzed suspension was concentrated by
`60% ethanol precipitation at 4°C. The ethanol precipitate was
`collected by centrifugation (10,000 x g, 30 min, 4°C) and
`suspended in buffer A. This precipitate was solubilized with
`2.5% Triton X-100 (overnight, 4°C) with gentle stirring. The
`suspension was centrifuged (100,000 x g, 60 min). The super-
`natant was collected and dialyzed against 4 liters of buffer B
`(50 mM TrisHCl/0.2 mM EDTA/0.15% Triton X-100/0.1
`mM PMSF, pH 8.0). The dialyzed extract was applied on a
`DEAE-cellulose column (10 x 13 cm), which was previously
`equilibrated with buffer B. The column was washed with 50 ml
`of buffer B and bound proteins were eluted from the column
`with 80 ml of a 0-500 mM linear NaCl gradient in buffer B at
`a flow rate of 15 ml/hr. The fractions from individual peaks
`were pooled separately and concentrated by 50% acetone
`precipitation; the precipitate was used for ligand blotting. Only
`peak 1 showed albumin-binding activity. The proteins present
`in peak 1 were further separated by using preparative SDS/
`PAGE (16 cm x 16 cm, 3-mm thick slab gel), and the gp6O
`eluted from the gel was used for further studies.
`
`Proc. Natl. Acad. Sci. USA 93 (1996)
`
`251
`
`Antibody Production and Purification. The gp6O eluted
`from preparative SDS/PAGE was used to immunize rabbits.
`Approximately 50 jig of protein (per rabbit) was injected i.m.
`after mixing with an equal volume of Freund's complete
`adjuvant (21). A second injection was given after 4 weeks.
`Rabbits were bled at 2 weeks after the second injection, and
`the immune response was checked. The preimmune serum IgG
`and the anti-gp6O-IgG were purified using a protein A-
`Sepharose column (21).
`Fluorescence-Activated Cell Sorter (FACS) Analysis. Con-
`fluent endothelial cell monolayers were washed with serum-
`free medium and incubated with the same medium for 2 hr.
`After this incubation, cells were washed with PBS and de-
`tached by incubating with nonenzymatic cell dissociation
`medium (Sigma) (10-30 min, 37°C). Nonspecific binding was
`blocked by incubating the cells with 10% horse serum in PBS
`(60 min on ice). Cells (106 per tube) were incubated with either
`gp6O antiserum or preimmune serum (1:10 diluted) (60 min,
`4°C), washed, and treated with fluorescein isothiocyanate-
`conjugated goat anti-rabbit IgG for 30 min. After washing, the
`cells were fixed with 1% paraformaldehyde and analyzed with
`an Ortho Cytoron Absolute Flow Cytometer (Ortho Diagnos-
`tic). The mean logarithmic fluorescence intensity for each
`sample was determined and converted into linear relative
`fluorescence units (AFL) by the formula: AFL = 10(E x 0.0137),
`where E is the mean channel fluorescence intensity (22).
`Immunoblotting. Endothelial cell membranes were sub-
`jected to SDS/PAGE and electrophoretically transferred to
`nitrocellulose or PVDF membrane. Nonspecific binding was
`blocked with either 3% gelatin or 5% nonfat dry milk in TBS
`(5 hr, 22°C). The membrane was washed twice with 0.05%
`Tween-20 in TBS and incubated with antibodies diluted in TBS
`containing 1% gelatin. Incubation was carried out for 4-6 hr;
`the mixture was washed and then incubated for 60 min with the
`second antibody (goat anti-rabbit IgG conjugated to alkaline
`phosphatase). After incubation, the membranes were washed
`twice and the protein bands were localized as described above.
`Molecular masses of the proteins were determined using
`known marker proteins.
`1251-Labeled Albumin-Binding Studies. BPMVECs were
`seeded (3 x 105 per well) in six-well Corning tissue culture
`plates and grown to confluence. The monolayers were washed
`twice with DMEM and incubated with DMEM for 20-24 hr in
`a cell culture incubator. After incubation, the monolayers were
`washed twice with binding buffer (10 mM Hepes/DMEM, pH
`7.4) and binding was initiated by adding 1 ml of 1 ,tg of
`125I-BSA in binding buffer. Incubation was carried out at 4°C
`for 60 min. Binding was terminated by washing the monolayer
`three times with the binding buffer. Radioactivity associated
`with the monolayer was determined after lysing the cells with
`1 M NaOH (23). Nonspecific binding was determined by the
`inclusion of unlabeled BSA (40 mg/ml) during the binding
`procedure (10-12). The test components, preimmune-IgG and
`the anti-gp6O-IgG, were preincubated 30 min with the mono-
`layer prior to the addition of 125I-BSA.
`
`RESULTS
`Identification of Native ABPs. We first isolated endothelial
`cell membranes from BPMVECs by differential centrifugation
`and the ABP present in this membrane fraction were identified
`using ligand blotting (see Materials and Methods). The BSA-
`binding regions were identified using polyclonal antibodies
`raised against native BSA. In the absence of exposure of the
`membrane strip to native BSA, the anti-BSA antibodies rec-
`ognized only a 67-kDa polypeptide (Fig. 1A, lane 1), indicating
`that albumin binds to endothelial membranes. However, when
`the strip was treated with BSA, the anti-BSA antibodies
`reacted with three distinct polypeptides: 110, 57-60, and 18
`kDa (Fig. IA, lane 2). Of these, the antibodies reacted most
`
`Page 2
`
`
`
`252
`
`Medical Sciences: Tiruppathi et al.
`
`Proc. Natl. Acad. Sci. USA 93 (1996)
`
`A
`
`1
`
`106-
`80-
`
`49.5-
`
`32.5-
`
`27.5-
`
`2
`
`0
`
`-110
`
`--67
`-57
`
`.
`
`-18
`
`B
`
`1
`
`2
`
`2
`
`2
`
`200 - "
`
`I _ ......
`
`116-
`
`.""
`66- e '.:A
`
`+- 57-60
`
`:.
`
`43 - qm
`
`v 18
`
`29- -
`
`4--
`
`FIG. 3. SDS/PAGE profile of pro-
`teins. Lane 1, BPMVEC membrane;
`lane 2, DEAE peak 1. The arrow in-
`dicates the position of gp6O. BPMVEC
`membranes (100 jtg of protein) and
`DEAE peak 1 (50 jg of protein) were
`separated on SDS/PAGE and stained
`with Coomassie brilliant blue R-250.
`Molecular masses of the proteins were
`determined using known molecular
`mass marker proteins.
`
`(A) Identification of ABP in BPMVEC membranes using
`FIG. 1.
`native albumin. BPMVEC membranes (100 j.g of protein) were
`separated on SDS/PAGE and transferred to PVDF membrane strips.
`Lane 1, control (not treated with albumin); lane 2, incubation with
`albumin. Positions of known molecular mass markers are indicated in
`kDa. Results are representative of three separate experiments. (B)
`Identification of ABP in endothelial cell membranes using 125I-labeled
`albumin. Lane 1, BPMVEC; lane 2, BPAEC.
`
`intensely with the 57- to 60-kDa protein, indicating that gp60
`is the major native ABP.
`We also used 125I-BSA to identify ABP in ligand-blotting
`experiments and found the specific interaction of 125I-BSA
`with the 60- and 18-kDa polypeptides present in endothelial
`cell membranes (Fig. lB).
`Isolation of gp6O. Since native albumin bound primarily to
`gp6O, we next developed a method for the isolation of gp60
`from BPMVEC membranes. Ligand blotting was employed to
`assess the presence of this protein during purification. BPM-
`VEC membranes were initially solubilized with 2.5% sodium
`cholate/4 M urea, and the extract was dialyzed and concen-
`trated by 60% ethanol precipitation. This precipitate was
`reextracted with Triton X-100. The Triton X-100-solubilized
`extract was chromatographed on the DEAE column, and the
`bound proteins were eluted with a linear NaCl gradient (0-500
`mM). The proteins were eluted as three peaks. The fractions
`from each peak were pooled and screened for albumin binding
`using the ligand blotting assay. Only peak 1 showed albumin-
`binding activity (Fig. 2). Fig. 3 shows the SDS electrophoretic
`profile of proteins from native BPMVEC membrane (lane 1)
`and DEAE column peak 1 (lane 2) after staining with Coo-
`
`massie brilliant blue R-250. The presence of 57- to 60-kDa
`protein corresponding to albumin binding was observed with
`ligand blotting in native membranes and DEAE peak I.
`SDS/PAGE was also performed under nonreducing condi-
`tions (in absence of 2-mercaptoethanol) and albumin binding
`was observed only with the 57- to 60-kDa region, suggesting
`that gp6O existed as a single polypeptide. We further purified
`gp6O using preparative SDS/PAGE, and the eluted protein
`from the gel was used for antibody preparation.
`Cell Surface Expression of gp6O. We nonenzymatically
`detached endothelial cells and incubated the cells with either
`gp6O antiserum or preimmune serum and carried out FACS
`analysis to investigate the cell surface expression of gp6O on
`endothelial cells. The patterns of preimmune and gp6O anti-
`serum binding to different endothelial cell types are shown in
`Fig. 4. Data are plotted as log fluorescence intensity (in
`arbitrary units) against cell number. The gp6O antiserum-
`specific AFL for each cell type was calculated by subtracting
`preimmune serum AFL. The gp6O-specific AFL values ranged
`from 10 to 12 units for BPMVEC and BPAEC; values ranged
`from 2 to 5.4 units for HUVEC and HPAEC.
`Effect of Anti-gp6O IgG on Binding of '251-BSA to BPMVEC
`Monolayers. Preimmune serum IgG and anti-gp6O IgG were
`affinity purified using a protein A-Sepharose column. We
`investigated the influence of IgG fractions on binding of
`125I-BSA to BPMVEC monolayers at 4°C (Fig. 5). Endothelial
`cell monolayers were extensively washed with serum-free
`medium and then used for 125I-BSA binding. Nonspecific
`
`120
`
`Preimmune
`
`A 10
`
`iPreimmune
`
`B
`
`1
`
`2
`
`t,,q
`
`.M
`
`Identification of gp6O from pro-
`FIG. 2.
`~~teins eluted in the DEAE column. Proteins
`(50 g.g) from peak 1 were subjected to SDS/
`PAGE and transferred to a PVDF membrane.
`Ligand blotting was performed using native
`albumin. Lane 1, control (not treated with
`albumin); lane 2, incubation with albumin.
`The arrow indicates the position of gp6O. The
`wider band above the arrow is BSA eluted
`from the column.
`
`90
`:
`100
`-@150
`CD30
`0120
`
`10'
`
`102
`
`Preimmune
`t
`
`90Anig6
`io3
`102
`100
`10'
`C Pr1m5n0
`120
`30
`
`103
`
`60
`
`--9p60.
`
`2Anti-gp6O
`601
`0
`
`+-
`
`100
`
`101
`
`l
`
`10'
`a131
`o
`log fluorescence intensity
`
`102
`
`103
`
`FIG. 4. FACS analysis of endothelial cells. Endothelial cells were
`detached nonenzymatically and incubated with either preimmune
`antiserum and used for analysis. (A) BPMVECs; (B)
`serum or gp60
`BPAECs; (C) HUVECs; (D) HPAECs.
`
`Page 3
`
`
`
`Medical Sciences: Tiruppathi et al.
`
`Proc. Natl. Acad. Sci. USA 93 (1996)
`
`253
`
`140 -
`
`1
`
`2
`
`4ur
`.....
`
`.
`
`67
`61
`57 -
`43 -
`36 -
`
`Immunoblotting of BPMVEC
`FIG. 7.
`membrane with anti-bovine SPARC and
`LF-56 antibodies. The antibodies, anti-
`bovine SPARC (1:1000) antiserum (lane
`1) and LF-56 antiserum (1:300) (lane 2),
`were diluted and incubated with mem-
`brane strips for 5 hr. Other details are as
`described in the legend to Fig. 6. Results
`are representative of three separate ex-
`periments.
`
`turally different from gp6O. To test the immunological cross-
`reactivity of bovine anti-gp6O antibodies with other species, we
`carried out immunoblotting experiments with human umbilical
`vein and rat lung endothelial cell membranes. The antibodies
`recognized human and rat endothelial cell membrane gp6O
`(Fig. 6, lanes 5 and 6).
`gp6O and SPARC. To study the proposed structural rela-
`tionship between the endothelial membrane-associated and
`secreted ABPs, we carried out immunoblotting of BPMVEC
`membranes with the antibodies raised against purified bovine
`SPARC (Fig. 7). The antibodies raised against purified bovine
`SPARC recognized 67-, 61-, 57- to 60-, 43-, and 36-kDa
`polypeptides in BPMVEC membranes (Fig. 7, lane 1). The
`bovine anti-SPARC NH2-terminal peptide antibodies reacted
`with 36- and 43-kDa polypeptides (Fig. 7, lane 2). We radio-
`labeled the endothelial cell surface with 1251 and immunopre-
`cipitated the endothelial cell lysates using anti-gp6O antibodies
`(13) to study whether SPARC was associated with the endo-
`thelial cell surface. The anti-gp6O antibodies precipitated only
`the 57- to 60-kDa polypeptide (data not shown), suggesting
`that SPARC was not cell surface associated.
`
`DISCUSSION
`High-affinity binding sites for modified albumin (e.g., for-
`maldehyde-treated albumin) have been reported in liver sinu-
`soidal and renal plasma membrane preparations (24, 25).
`Receptors for these modified albumin molecules have been
`purified from liver plasma membrane preparations (26) and
`shown to bind specifically to the modified molecules but not to
`the native albumin (27). These ABPs may belong to a family
`of scavenger receptors identified in macrophages or macro-
`phage-derived cells (27-29). This conclusion has recently been
`confirmed by a ligand-blotting experiment showing the exis-
`tence of 18- and 31-kDa scavenger receptors for modified
`albumins on vascular endothelial cell surface (12). The recep-
`tors are also expressed in fibroblasts and smooth muscle cells.
`It has been suggested that the binding of modified albumins
`(i.e., formaldehyde- or maleic anhydride-treated albumin and
`albumin-gold complex) to these receptors may initiate endo-
`cytosis, and the ligand may be subsequently degraded by
`lysosomal proteases (11). The native monomeric albumin can
`also react with these binding proteins in endothelial cells (5, 9).
`Therefore, the ABPs are postulated to play a role in endocy-
`tosis and transcytosis of albumin in various cells. In the case of
`endothelial cells, ABPs may activate transcellular pathways
`and thereby contribute to albumin transcytosis (5, 9, 30).
`In the present study, we isolated membranes from BPM-
`VECs and used these for identification and isolation of gp6O.
`Using the ligand-blotting technique, we demonstrated the
`direct interaction of the native albumin with gp6O in vascular
`
`120 -
`
`"O.
`Co 100-
`
`80-
`
`60-
`
`40-
`
`20 rii
`
`r
`
`m O
`
`0)
`.C
`
`0 3
`
`a)a
`U,
`
`100
`50
`250
`Anti-gp60 IgG, jtg
`
`50
`100
`250
`Preimmune IgG, tig
`Effect of anti-gp6O IgG and preimmune serum IgG on the
`FIG. 5.
`binding of 1251-BSA to BPMVEC monolayer at 4°C. Results are means
`-+- SEM of three separate experiments carried out in a triplicate
`binding assay.
`
`binding ranged from 30% to 40%. Preimmune serum IgG did
`not affect the specific binding of 125I-BSA to the BPMVEC
`monolayers. In contrast, anti-gp6O IgG inhibited the specific
`binding of 125I-BSA to BPMVEC monolayers in a dose-
`dependent manner. The inhibition was -90% at 250 ztg of
`antibody per ml.
`Immunoblotting of Endothelial Cell Membranes with Anti-
`gp6O Antibodies. BPMVEC and BPAEC membrane proteins
`were separated by using SDS/PAGE and transferred to ni-
`trocellulose strips. The strips were immunoblotted with the
`gp6O antiserum (Fig. 6). Preimmune serum (Fig. 6, lanes 1 and
`3) did not recognize any proteins from BPMVEC and BPAEC
`membranes. Anti-gp6O antibodies recognized two major pro-
`teins (57-60 and 36 kDa) and one minor protein (43 kDa) in
`both membrane preparations (Fig. 6, lanes 2 and 4). The
`anti-gp6O antibodies did not react with 18- and 31-kDa
`polypeptides, suggesting that 18- and 31-kDa ABPs are struc-
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`*..:...... ....
`
`Immunoblotting of endothelial cell membrane proteins
`FIG. 6.
`with anti-gp6O antibodies. Endothelial cell membrane proteins (100
`jtg) were separated on SDS/PAGE and transferred to nitrocellulose
`membrane strips. Nonspecific binding was blocked with 5% nonfat dry
`milk in TBS. Antiserum and preimmune serum were diluted in
`blocking solution and incubated 3-4 hr at 40C, washed, and treated
`with goat anti-rabbit IgG conjugated with alkaline phosphatase. Lanes
`1 and 2, BPMVEC membranes; lanes 3 and 4, BPAEC membranes;
`lane 5, HUVEC membranes; lane 6, rat lung endothelial cell mem-
`branes. Lanes 1 and 3 were treated with preimmune serum; lanes 2, 4,
`5, and 6 were incubated with gp6O antiserum. Positions of gp6O and
`36-kDa protein bands are indicated by arrows. The protein band
`between gp6O and 36 kDa is 43-kDa protein. Results are representative
`of three separate experiments.
`
`Page 4
`
`
`
`254
`
`Medical Sciences: Tiruppathi et al.
`
`Proc. Natl. Acad. Sci. USA 93 (1996)
`
`endothelial cell membranes (Fig. 1) and with the purified gp6O
`(Fig. 2). This finding is in close agreement with previous work
`using lectin-binding analysis, which showed that native albumin
`interacts with gp6O on rat endothelial cell surface (10, 31).
`We prepared antibodies against the bovine gp6O to investi-
`gate characteristics of this protein in endothelial cells. We
`performed FACS analysis of endothelial cells using anti-gp6O
`antibodies. gp6O was expressed in vascular endothelial cells
`(Fig. 4) and the surface expression was greater in BPMVECs
`and BPAECs than in HUVECs and HPAECs. Affinity-
`purified anti-gp60 antibodies significantly reduced the specific
`binding of 125I-BSA to BPMVEC monolayer, whereas preim-
`mune serum IgG had no effect on the binding of 125I-BSA to
`the cell monolayers (Fig. 5). These results demonstrate that the
`antibodies developed against gp6O specifically recognize the
`native albumin-binding sites on the endothelial cell surface.
`We performed immunoblotting experiments with endothe-
`lial cell membranes to study the specificity of the anti-gp6O
`antibodies. The antibodies recognized two major polypeptides,
`57-60 and 36 kDa, and one minor polypeptide, 43 kDa (Fig.
`6), in endothelial cell membrane proteins. That the antibodies
`recognized only these proteins suggests that gp6O was purified
`to an apparent homogeneity. To test immunological cross-
`reactivity of bovine anti-gp6O antibodies with other species, we
`carried out immunoblotting experiments using human and rat
`endothelial cell membranes (Fig. 6). These results indicated
`that the bovine anti-gp6O antibodies reacted with human and
`rat endothelial cell membranes.
`Studies have shown immunological cross-reactivity of anti-
`SPARC antibodies and SPARC peptide-based antibodies with
`the endothelial cell membrane-associated gp6O (15, 31).
`SPARC is secreted into the culture medium by endothelial
`cells and other cell types (32). SPARC is identical to the bone
`noncollagenous protein osteonectin (33, 34). The molecular
`size of this protein ranges from 36 to 43 kDa and it binds to
`albumin and extracellular matrix proteins (32-34). Recent
`studies have shown specific interaction of SPARC with endo-
`thelial cells (35). We carried out immunoblotting of bovine
`endothelial cell membranes using the antibodies raised against
`purified bovine SPARC and the NH2-terminal sequence (1-56
`residues) of bovine SPARC. The bovine anti-SPARC antibod-
`ies recognized many polypeptides, including gp6O in BPMVEC
`membranes (Fig. 7). The bovine NH2-terminal SPARC pep-
`tide antibodies recognized only the 36- and 43-kDa polypep-
`It did not react with other endothelial membrane
`tides.
`proteins, indicating that the membrane-associated gp6O was
`different from SPARC and that SPARC did not derive from
`gp60. The 36- and 43-kDa polypeptides identified (Fig. 6) in
`BPMVEC membranes using anti-gp60 antibodies may be due
`to the presence of SPARC associated with endothelial cell
`matrix proteins (35). The anti-gp6O antibodies also did not
`recognize the 18- and 31-kDa polypeptides [proposed as the
`scavenger receptors in endothelial cells (11, 12)], suggesting
`that scavenger receptors are structurally different from native
`albumin receptor, gp6O.
`
`We thank Dr. Peter J. Del Vecchio for providing us the BPMVECs
`and Asma Naqvi for her excellent technical assistance. This work was
`supported by National Institutes of Health Grants HL-46350 and
`HL-45638 and by Andaris Limited (England).
`
`1.
`
`Pardridge, W. M. (1979) Endocrinology 105, 605-612.
`
`2.
`3.
`
`4.
`5.
`
`6.
`
`7.
`
`8.
`
`9.
`
`10.
`
`11.
`12.
`
`13.
`
`14.
`
`15.
`
`16.
`
`17.
`
`18.
`
`19.
`
`20.
`21.
`
`22.
`
`23.
`
`24.
`
`25.
`
`26.
`
`27.
`
`28.
`
`29.
`
`30.
`
`31.
`32.
`
`33.
`
`34.
`
`35.
`
`Peters, T., Jr. (1975) Adv. Protein Chem. 37, 161-245.
`Peters, T., Jr. (1975) The Plasma Proteins: Structure, Function, and
`Genetic Control, ed. Putman, F. W. (Academic, New York), Vol.
`1, pp. 133-181.
`Forker, E. L. & Luxon, B. (1983) J. Clin. Invest. 72, 1764-1771.
`Ghinea, N., Fixman, A., Alexandru, D., Popou, D., Hasu, M.,
`Ghitssen, L., Eskenasy, M., Simionescu, M. & Simionescu, N.
`(1988) J. Cell Biol. 107, 231-239.
`Ghitescu, L., Fixman, A., Simionescu, M. & Simionescu, N.
`(1986) J. Cell Biol. 102, 1304 -1311.
`Milici, A., Watrous, N. E., Stukenbrok, H. & Palade, G. E. (1987)
`J. Cell Biol. 105, 2603-2612.
`Predescu, D. M., Simionescu, N., Simionescu, N. & Palade, G. E.
`(1988) J. Cell Biol. 107, 1729-1738.
`Ghinea, N., Eskenasy, M., Simionescu, M. & Simionescu, N.
`(1989) J. Biol. Chem. 264, 4755-4758.
`Schnitzer, J. E., Carley, W. W. & Palade, G. E. (1988) Proc. Natl.
`Acad. Sci. USA 85, 6773-6777.
`Schnitzer, J. E. & Bravo, J. (1993) J. Biol. Chem. 268, 7562-7570.
`Schnitzer, J. E., Sung, A., Horvat, R. & Bravo, J. (1992) J. Biol.
`Chem. 267, 24544-24553.
`Siflinger-Birnboim, A., Schnitzer, J. E., Lum, H., Blumenstock,
`F. A., Shen, C. P. J., Del Vecchio, P. J. & Malik, A. B. (1991) J.
`Cell. Physiol. 149, 575-584.
`Schnitzer, J. E., Ulmer, J. B. & Palade, G. E. (1990) Proc. Natl.
`Acad. Sci. USA 87, 6843-6847.
`Schnitzer, J. E. & Oh, P. (1992) Am. J. Physiol. 263, H1872-
`H1879.
`Tiruppathi, C., Siflinger-Birnboim, A., Del Vecchio, P. J. &
`Malik, A. B. (1993) Mol. Biol. Cell 4, 338a (abstr.).
`Fisher, L. W., Lindner, W., Young, M. F. & Termine, J. D. (1989)
`Connect. Tissue Res. 21, 43-50.
`Del Vecchio, P. J. & Lincoln, D. W., 11 (1983) in The Endothelial
`Cell-a Pluripotent Control Cell of the Vessel Wall, eds. Thio-
`Kroner, D. D. S., & Freshney, R. I. (Karger, Basel), pp. 67-83.
`Del Vecchio, P. J., Siflinger-Birnboim, A., Belloni, P. N., Holle-
`ran, L. A., Lum, H. & Malik, A. B. (1992) In Vitro Cell. Dev. Biol.
`28, 711-715.
`Laemmli, U. K. (1970) Nature (London) 227, 680-685.
`Harlow, E. & Lane, E. (1988) Antibodies: A Laboratory Manual
`(Cold Spring Harbor Lab. Press, Plainview, NY).
`Makni, H., Malter, J. S., Reed, J. C., Nobuhiko, S., Lang, G.,
`Kioussis, D., Trinchieri, G. & Kamoun, M. (1991) J. Immunol.
`146, 2522-2529.
`Tiruppathi, C., Lum, H., Anderson, T. T., Fenton, J. W., II, &
`Malik, A. B. (1992) Am. J. Physiol. 263, L595-L601.
`Horiuchi, S., Takata, K. & Morino, Y. (1985) J. Biol. Chem. 260,
`475-481.
`Ranganathan, P. N. & Mego, J. L. (1986) Biochem. J. 239,
`537-543.
`Horiuchi, S., Takata, K. & Morino, Y. (1985) J. Biol. Chem. 260,
`482-488.
`Knowles, S. E., Ballard, F. J., Livesey, G. & Williams, K. E.
`(1981) Biochem. J. 196, 41-48.
`Brown, M. S. & Goldstein, J. L. (1983) Annu. Rev. Biochem. 52,
`223-261.
`Pratten, M. K., Williams, K. E. & Lloyd, J. B. (1977) Biochem. J.
`168, 365-372.
`Malik, A. B. & Siflinger-Birnboim, A. (1993) in Biological Bar-
`riers to Protein Delivery, eds. Audus, K. L. & Raub, T. J. (Plenum,
`New York), pp. 231-267.
`Schnitzer, J. E. & Oh, P. (1994) J. Biol. Chem. 269, 6072-6082.
`Sage, H., Johnson, C. & Bornstein, P. (1984) J. Biol. Chem. 259,
`3993-4007.
`Sage, H., Vernon, R. B., Funk, S. E., Everitt, E. A. & Angello, J.
`(1989) J. Cell Biol. 109, 341-356.
`Termine, J. D., Kleinman, H. K., Whitson, S. W., Conn, K. M.,
`McGarvey, M. L. & Martin, G. R. (1981) Cell 26, 99-105.
`Yost, J. C. & Sage, E. H. (1993)J. Biol. Chem. 268, 25790-25796.
`
`Page 5
`
`