(Received for publication, January 23, 1997, and in revised form, March 23, 1997)
From the Department of Biochemistry, Academic Medical Center, University of Amsterdam, 1105 AZ, Amsterdam, The Netherlands
The low density lipoprotein receptor-related
protein (LRP), a member of the low density lipoprotein receptor gene
family, mediates the cellular uptake of a diversity of ligands. A
folding chaperone, the 39-kDa receptor-associated protein (RAP) that
resides in the early compartments of the secretory pathway inhibits the binding of all ligands to the receptor and may serve to prevent premature binding of ligands to the receptor during the trafficking to
the cell surface. To elucidate the molecular interactions that underlie
the interplay between the receptor, RAP, and the ligands, we have
analyzed and delineated the binding sites of plasminogen activator
inhibitor-1 (PAI-1), tissue-type plasminogen activator (t-PA)·PAI-1
complexes, RAP, and the anti-LRP Fab fragment Fab A8. To that end, we
have generated a series of soluble recombinant fragments spanning the
second cluster of complement-type repeats (C3-C10) and the
amino-terminal flanking epidermal growth factor repeat (E4) of LRP
(E4-C10; amino acids 787-1165). All fragments were expressed by stably
transfected baby hamster kidney cells and purified by affinity
chromatography. A detailed study of ligand binding to the fragments
using surface plasmon resonance revealed the presence of three
distinct, Ca2+-dependent ligand binding
sites in the cluster II domain (Cl-II) of LRP. t-PA·PAI-1 complexes
as well as PAI-1 bind to a domain located in the amino-terminal portion
of Cl-II, spanning repeats E4-C3-C7. Adjacent to this site and
partially overlapping is a high affinity RAP-binding site located on
repeats C5-C7. Fab A8, a pseudo-ligand of the receptor, binds to a
third Ca2+-dependent binding site on repeats
C8-C10 at the carboxyl-terminal end of Cl-II. Next, we studied the
RAP-mediated inhibition of ligand binding to LRP and to Cl-II. As
expected, we observed a strong inhibition of t-PA·PAI-1 complex and
Fab A8 binding to LRP by RAP (IC50 0.3 nM),
whereas in the reverse experiment, competition of t-PA·PAI-1
complexes and Fab A8 for RAP binding to LRP could only be shown at high
concentrations of competitors (
1 µM). Interestingly,
even though the equilibrium dissociation constants for the binding of
RAP to LRP and to Cl-II are similar, the binding of the ligands to
Cl-II is only prevented by RAP at concentrations that are at least 2 orders of magnitude higher than those required for inhibition of ligand
binding to LRP. Our results favor models that propose RAP-induced
allosteric inhibition of ligand binding to LRP that may require LRP
moieties that are located outside Cl-II of the receptor.
The low density lipoprotein receptor-related protein (LRP),1 a 600-kDa glycoprotein, is a multifunctional endocytic receptor (1-3). LRP binds and internalizes a broad spectrum of structurally unrelated ligands including two important components of the fibrinolytic system, tissue-type plasminogen activator (t-PA) and its specific inhibitor plasminogen activator inhibitor type-1 (PAI-1) (4, 5). In rodents, LRP was shown to be one of the in vivo receptor systems involved in the clearance of t-PA (6, 7).
LRP is synthesized as a 600-kDa precursor protein that is subsequently
processed in the trans-Golgi compartment to a 515-kDa -chain and a 85-kDa
-chain. Both chains remain associated in an
unusual noncovalent fashion (8). Ligand binding by the receptor is
thought to be mediated by the so-called complement-type repeats. These
40-residue-long repeats are characterized by a highly conserved spacing
pattern of six cysteine residues, and they are found in all other
members of the low density lipoprotein receptor gene family (1). In the
low density lipoprotein receptor, these complement-type repeats were
shown to mediate ligand binding (9, 10). The
-chain of LRP contains
four potential ligand-binding domains assembled as sequential series of
2, 8, 10, and 11 complement-type repeats (cluster I, II, III, and IV).
The first evidence for a role of the cluster II domain (Cl-II) in
ligand binding was presented by Moestrup et al., who
demonstrated the binding of
2-macroglobulin-complexes, urokinase-type plasminogen activator· PAI-1 complexes, and
receptor-associated protein (RAP) to a 75-kDa CNBr-generated LRP
fragment, spanning Cl-II, on a ligand blot (11). Using a recombinant
DNA approach, Willnow et al. expressed LRP minireceptors and
showed t-PA·PAI-1 complex and RAP binding to Cl-II and a low affinity
binding of RAP to a truncated receptor harboring the cluster IV domain
(12). In a previous study, we provided evidence for an additional
ligand binding site in Cl-II by the selection of a high affinity
Cl-II-binding Fab fragment (Fab A8) that bound the receptor in a
Ca2+-dependent fashion and completely inhibited
the binding of pro-urokinase to LRP (13).
The 39-kDa folding chaperone RAP is a molecule with a high affinity for LRP that resides in the endoplasmic reticulum and the cis-Golgi compartments (14). RAP inhibits the binding of all known ligands to LRP, and it has been proposed that this activity may serve to prevent premature binding of ligands to the receptor during the maturation and trafficking of LRP (14, 15). Little is known about the mechanism by which RAP inhibits the binding of the diversity of structurally unrelated ligands. One model assumes a close spatial association of a RAP binding site to each of the independent LRP ligand-binding sites (16, 17). Consequently, RAP would inhibit ligand binding by steric hindrance of each of the LRP ligands. An alternative model for the regulation of ligand binding by RAP proposes a RAP-induced conformational change in the LRP molecule (18). A unique feature of RAP is its repeated triplicate structure, endowing the protein with three potential LRP-binding sites as determined by cross-linking studies (14). Binding studies performed with amino-terminal and carboxyl-terminal fragments of RAP have indicated that at least two high affinity LRP binding sites are present in RAP (19). The presence of RAP-binding sites in the different clusters of LRP (11, 12, 20), as well as the presence of multiple receptor-binding sites in RAP, may support a model in which RAP binds to different clusters synchronically, thereby inducing a conformational change in LRP that would be incompatible with ligand binding. Such a model would imply that competition by RAP for ligand binding is not on the level of the individual clusters but requires the intact LRP molecule. In this study, we have performed a detailed analysis of ligand-binding sites in the cluster II domain of the receptor. Using recombinant fragments spanning Cl-II, we demonstrate that Cl-II contains three distinct binding sites for t-PA·PAI-1 complexes, RAP, and the pseudo-ligand Fab A8. In addition, we provide evidence for an allosteric component in RAP-mediated inhibition of ligand binding to LRP.
Oligonucleotides were from Pharmacia Biotech Inc. Restriction and other DNA-modifying enzymes were from Life Technologies, Inc. All other chemicals used were reagent grade (Sigma or Merck).
ProteinsHuman LRP was kindly provided by Dr. S. K. Moestrup (Institute of Medical Biochemistry, University of Aarhus,
Aarhus, Denmark). Recombinant glutathione
S-transferase-fused RAP was kindly provided by Dr. J. Kuiper
(Sylvius Laboratory, University of Leiden, Leiden, The Netherlands).
Two-chain t-PA was from Biopool (Umea, Sweden) and recombinant t-PA
(Actilyse) was from Boehringer Ingelheim (Ingelheim/Rhein, Germany).
Active PAI-1 purification procedures were essentially as described (21,
22). t-PA·PAI-1 complexes were prepared as follows. Recombinant t-PA
was incubated with an excess of PAI-1 for 2 h at 37 °C.
Complexed t-PA was then separated from the excess of unreacted PAI-1 by
affinity chromatography using lysine-Sepharose. Complexes bound to the
lysine-Sepharose were washed with 20 mM HEPES (pH 7.4), 150 mM NaCl, and 0.001% (v/v) Tween 80 (HBST) followed by
washings with HBST containing 500 mM NaCl and subsequent
elutions with HBST containing 1 M NaCl and 0.2 M -aminocaproic acid. Preparations were concentrated 20-fold by centrifugation in a Sorvall high speed centrifuge at 7,000 rpm for 1 h at 4 °C using Centricon 30 concentrators (Amicon, Beverly, MA) and subsequently dialyzed against filtered (0.45-µm filters, Schleicher & Schuell) and degassed, modified HBST buffer (containing 20 mM HEPES (pH 7.4), 150 mM NaCl,
2 mM CaCl2, 0.001% (v/v) Tween 80).
Preparations revealed a single protein band on a silver-stained
SDS-polyacrylamide gel and were stored at
20 °C until use.
Isolation of the monoclonal Fab fragment Fab A8 as well as the
isolation of the anti-factor VIII monoclonal antibody CAg69 were
essentially as described (13, 23).
The BIAcoreTM2000
biosensor system and reagents, including an amine-coupling kit
containing N-hydroxysuccinimide,
N-ethyl-N-(3-diethylamino-propyl)carbodiimide, ethanolamine hydrochloride, and CM5 sensor chips (research grade), were
from Biacore AB (Uppsala, Sweden).
Plasmids encoding recombinant LRP Cl-II fragments
(Cl-II, amino acids 786-1165; Cl-II-1, amino acids 786-914; Cl-II-2,
amino acids 915-1034; and Cl-II-3, amino acids 1041-1165; numbering according to Herz et al. (24)) were obtained by polymerase
chain reaction using a plasmid clone containing the human LRP cDNA
(a gift from Dr. J. Herz, University of Texas Southwestern Medical Center, Dallas) as a template. DNAs encoding recombinant-LRP cluster II
fragments were amplified using primers LRP 3 (5-GGACTCGAGAACAAATGCCGGGTGAAC-3
) and LRP 8 (5
-AACTAGTCTGGTCGCAGAGCTCGCC-3
); LRP3 and LRP4
(5
-GACTAGTGGCTGAACAAGTGGCATT-3
); LRP5
(5
-TTACTCGAGCGCACCTGCCCCCCCAAC-3
) and LRP 6 (5
-AACTAGTGTTGGTGCAGTTGGCGTG-3
); and LRP 7 (5
-TTACTCGAGGGTGGCTGCCACACTGAT-3
) and LRP8 for LRP Cl-II fragments
Cl-II, Cl-II-1, Cl-II-2, and Cl-II-3, respectively. Polymerase chain
reaction products were digested with XhoI and SpeI and subsequently cloned into the
HindIII/XbaI-digested plasmid pRc/CMV
(Invitrogen, La Jolla, California) together with a 192-base pair
HindIII/XhoI fragment, encoding the entire signal
and pro-sequences of t-PA (amino acids
35 to 1) (25) followed by a
16-amino acid tag (KKEDFDIYDEDENQSP) that contains the antigenic
determinant of the mAb CAg69 (26). Subsequently, DNA preparations
encoding Cl-II fragments were obtained from the pRc/CMV constructs
using HindIII and BclII, blunt-ended using T4 DNA
polymerase, and cloned into the SmaI site of vector pZEM229R
(27) to yield expression plasmids encoding LRP Cl-II fragments Cl-II,
Cl-II-1, Cl-II-2, and Cl-II-3. Expression plasmids encoding Cl-II-1/2
and Cl-II-2/3 were obtained by digestion of expression plasmids coding
for Cl-II and Cl-II-2 using either HindIII and
AflIII or AflIII and BclII combinations and ligating the proper DNA fragments. All constructs were
verified by DNA sequence analysis using an ALF express automated DNA
sequencer (Pharmacia).
Baby hamster kidney cells (thymidine kinase-deficient BHK-570 cells, ATCC CRL 10314) were grown to subconfluency (80%) in Dulbecco's modified Eagle's medium (Life Technologies, Inc.), containing 5% (v/v) fetal calf serum, 100 units/ml penicillin, 100 mg/ml streptomycin, and 0.25 mg/ml fungizone. Cells were transfected with 20 µg of DNA using the calcium precipitation method, essentially as described (28). DNA-calcium-phosphate coprecipitates were added, and after 5 h the cells were incubated for 1 min with 15% (v/v) glycerol. Cells were allowed to recover in Dulbecco's modified Eagle's medium for 12 h and subsequently cultured for 12 days on selective medium (Dulbecco's modified Eagle's medium, supplemented with 1 µM methotrexate (amethopterin; Sigma)). Individual colonies were isolated and tested for the expression of recombinant LRP cluster II fragments by ELISAs using mAb CAg69, or a polyclonal antibody directed to LRP (a gift from Dr. S. K. Moestrup). Cl-II producing clones were grown to subconfluency (80%) in 500 cm2 flasks (NUNC, Kamstrup, Denmark). Next, the cells were cultured for several days in Optimem medium (Life Technologies, Inc.), and conditioned media were harvested every 24 h. Cl-II fragments were purified from the conditioned media by affinity chromatography using Sepharose-coupled mAb CAg69. After binding, columns were washed with HEPES-buffered saline (HBS, 20 mM HEPES (pH 7.4), 150 mM NaCl) and eluted with HBS containing 1 M NaCl. If required, another affinity chromatography purification step was included using either RAP- or Fab A8-Sepharose columns. These columns were washed with HBS and eluted with HBS containing 10 mM EDTA. Cl-II preparations were concentrated in a HEPES-buffered 1 M NaCl solution by centrifugation in a Sorvall high speed centrifuge at 7,000 rpm for 1 h at 4 °C, using Centricon 10 or 30 concentrators (Amicon, Beverly, MA) and dialyzed against filtered and degassed, modified HBST buffer (containing 20 mM HEPES (pH 7.4), 150 mM NaCl, 2 mM CaCl2, 0.005% (v/v) Tween 80). All Cl-II fragment preparations were analyzed by ELISA, SDS-polyacrylamide gel electrophoresis, and immunoblotting with mAb CAg69 and anti-LRP antibodies.
Analysis of N-linked Glycosylation of Recombinant LRP Cl-II FragmentsCl-II fragments were analyzed by a nonreducing silver-stained SDS-polyacrylamide gel (12% (w/v) polyacrylamide). To assess the contribution of N-linked glycosylation to the observed migration of the fragments, the Cl-II fragment preparations were treated with glycanase (N-glycosidase F, Boehringer Mannheim). Briefly, the protein samples were incubated for 30 min at 45 °C in 0.4% (w/v) SDS-containing 100 mM potassium phosphate buffer (pH 8). Subsequently, the preparations were diluted 1:1 with 2% Triton X-100 and incubated with 0.1 unit of glycanase for 2 h at 37 °C. Treated samples were reduced and analyzed on a silver-stained SDS-polyacrylamide gel.
Concentration Determinations Using SPRThe tag-binding mAb (CAg69) was immobilized to a CM5 sensor chip, using the amine-coupling kit following the instructions of the supplier at a high density of approximately 50 fmol/mm2 to determine binding of the samples under conditions of mass-transport limitation. A control channel on the sensor chip was activated and blocked using amine-coupling reagents without immobilization of protein. The binding of protein preparations to this noncoated channel was subtracted from specific binding. To calibrate the system, standard Cl-II fragment preparations with known concentrations (BCA kit, Pierce), were passed over the sensor chip at 25 °C at a flow rate of 5 µl/min, using HBST as running buffer. Regeneration of the sensor chip surface was performed with 100 mM H3PO4. For each concentration, the slope of the binding curve was determined by linear fitting using the BIAevaluation software (Biacore AB). Using the BIAconcentration evaluation software, the system was calibrated to determine unknown Cl-II fragment concentrations.
Kinetic Determinations Using SPRPurified human LRP was immobilized on a CM5 sensor chip by amine coupling at a low density of 6.7 fmol/mm2. Using modified HBST as running buffer, RAP and Fab A8 were passed over three separate channels with immobilized LRP and one control (nonimmobilized) channel at 25 °C at a flow rate of 20 µl/min. Each determination was performed at least in triplicate at different concentrations (n = 4) in the appropriate concentration range (around Kd values). RAP/LRP interaction constants were determined by performing fittings according to a double site model (a single site model did not appropriately describe the interaction) according to the following equation:
![]() |
(Eq. 1) |
![]() |
(Eq. 2) |
![]() |
(Eq. 3) |
![]() |
(Eq. 4) |
Recombinant Cl-II fragment-ligand interactions were measured as follows: RAP, purified t-PA·PAI-1 complexes, PAI-1, and Fab A8 were immobilized to the sensor chip at low densities (approximately 15-30 fmol/mm2). Measurements were performed at 25 °C at a flow rate of 20 µl/min. Six different analyses were performed at least at four different concentrations. Kinetic constants were obtained by fitting to single site models (describing the interaction appropriately, as judged from residual plots and statistical parameters) and calculated as indicated above. The data were validated by subjecting them to the tests of selfconsistency (29).
Analysis of RAP Inhibition of Ligand Binding by Competitive ELISAELISAs were performed at half-maximal saturating concentrations of ligands versus immobilized LRP or Cl-II. To determine half-maximal saturating concentrations, 50 ng of LRP or 160 ng of Cl-II was immobilized to 96-well plates for 30 min at 37 °C. Wells were blocked with 3% (w/v) bovine serum albumin for 30 min at 37 °C, washed, and incubated with a range of ligand concentrations in 20 mM Tris-HCl (pH 7.4), 150 mM NaCl, 0.005% (v/v) Tween 20 (TBST). Bound proteins were detected using either biotinylated anti-PAI-1 mAb CLB 2C8, biotinylated rat anti-mouse kappa light chain mAb (CLB products, Amsterdam, The Netherlands), or goat anti-glutathione S-transferase polyclonal antibodies (Pharmacia Biotech Inc.) for t-PA·PAI-1 complexes, Fab A8, and glutathione S-transferase-RAP, respectively. Staining was performed as described (13). Competitive ELISAs were subsequently performed under identical conditions, and experiments were performed in triplicate.
In a
previous study, we confirmed and extended the evidence for the
importance of Cl-II for ligand binding by the isolation and analysis of
a "phage display" derived Fab fragment (Fab A8) (13). This Fab
fragment behaves like a pseudo-ligand because it binds LRP in a
Ca2+-dependent fashion and its binding is
completely inhibited by RAP. Using this Fab fragment, we provided
evidence for a binding site for pro-urokinase in Cl-II, distinct from
the binding sites for 2-macroglobulin and t-PA·PAI-1
complexes. To investigate ligand binding by Cl-II in more detail, we
have expressed a series of soluble recombinant cluster II fragments
(Fig. 1) in stably transfected baby hamster kidney cell
lines. Each of the fragments is preceded by a t-PA-derived signal
pro-sequence for targeting the fragments into the secretory pathway and
a tag sequence to facilitate detection and purification. The epidermal
growth factor repeat (E4) that precedes Cl-II is included in all
fragments corresponding to the amino-terminal end of Cl-II, because
this repeat has been postulated to be important for the binding of
methylamine activated
2-macroglobulin to LRP (11, 12).
Purified recombinant fragments were analyzed by SDS-polyacrylamide gel
electrophoresis, and gels were stained with silver (Fig.
2A). The mobility of the fragments was
somewhat slower than would be expected on the basis of the calculated
molecular masses (approximately 15.9, 14.9, 15.4, 28.9, 29.3, and 43.5 kDa for Cl-II-1, Cl-II-2, Cl-II-3, Cl-II-1/2, Cl-II-2/3, and Cl-II,
respectively). This observation might be readily explained by
N-linked glycosylation of several of the complement-type
repeats of Cl-II (20). To show that N-linked glycosylation
is indeed responsible for the altered mobility of the fragments, we
treated the fragments with glycanase (N-glycosidase F) and
compared the migration pattern to that of the untreated fragments. As
shown in Fig. 2B, after glycanase treatment Cl-II-2/3
migrates with the mobility of a protein of the expected molecular mass
(29.3 kDa). Similar results were obtained for the other recombinant fragments (data not shown), and we conclude that the mobility of the
fragments during SDS-polyacrylamide gel electrophoresis is reduced as a
result of carbohydrate moieties attached to the fragments by
N-linked glycosylation. These results can also explain the
observed minor bands (Fig. 2A), which may represent
unglycosylated forms of the respective fragments.
Analysis of Ligand Binding to LRP Using SPR
In this study we have focused on the binding of the cluster II ligands RAP, t-PA·PAI-1 complexes, PAI-1, t-PA, and Fab A8. First, we determined the Kd values for interactions of these ligands with the intact receptor. To that end, LRP was immobilized on a sensor chip and the association and dissociation rate constants were determined by SPR measurements. As shown in Table I, RAP binding to LRP could only be described using a double exponential decay model, indicating the presence of at least two RAP-binding sites on LRP, as reported by Williams and co-workers (30). The two sites have Kd values of 39.4 and 275.0 nM, respectively, which are in good agreement with similar SPR measurements described by Moestrup et al. (high and low affinity sites of 41.7 and 240 nM, respectively) (31). However, Kd values for the RAP-LRP interaction determined by conventional binding studies turn out to be lower than the ones calculated here. This apparent discrepancy may be due to the different experimental approach. The interactions of t-PA·PAI-1 complexes, PAI-1, t-PA, and Fab A8 could accurately be described by a single site model, yielding Kd values of 6.1, 35, 158, and 3.3 nM, respectively.
|
To delineate the binding sites for RAP, t-PA·PAI-1 complexes, PAI-1, and Fab A8 on Cl-II, we immobilized these LRP ligands at a low density on sensor chips and determined the rate constants for the interactions with the Cl-II fragments. As presented in Table II, RAP binds to a subset of Cl-II fragments according to a single site model with Kd values of 20.9 nM for the whole Cl-II and 57.9 nM for the smallest fragment (Cl-II-2). The fragments Cl-II-1/2 and Cl-II-2/3 bind to RAP with similar Kd values. These results indicate the presence of a single RAP-binding site in Cl-II, localized on a fragment spanning complement-type repeats C5, C6, and C7.
|
t-PA·PAI-1 complexes bind to recombinant fragments Cl-II and Cl-II-1/2. The respective Kd values are 44.4 (Cl-II) and 54.0 nM (Cl-II-1/2), indicating an interaction site on Cl-II localized on a fragment spanning epidermal growth factor repeat E4 and complement-type repeats C3-C7. Cl-II and Cl-II-1/2 interact with PAI-1 as well, although with higher Kd values (196.3 and 219.1 nM for Cl-II and Cl-II-1/2, respectively). We could not detect any significant binding of Cl-II fragments to immobilized t-PA. We conclude that t-PA·PAI-1 complexes as well as PAI-1 bind to a site that is distinct from but partially overlapping the RAP binding site in Cl-II.
The anti-LRP Fab fragment Fab A8, which inhibits the binding of pro-urokinase-type plasminogen activator to LRP, binds to another subset of Cl-II fragments including Cl-II, Cl-II-2/3, and Cl-II-3. The Kd values are 29.3, 36.6, and 114.6 nM for Cl-II, Cl-II-2/3, and Cl-II-3, respectively, indicating a unique binding site for Fab A8 localized on complement-type repeats C8, C9, and C10.
The results of these studies are schematically represented in Fig.
3. We conclude that the recombinant fragments of Cl-II harbor functional ligand binding sites and Cl-II contains a linear array of at least three distinct binding sites for t-PA·PAI-1 complexes (PAI-1), RAP, and Fab A8.
Determination of IC50 Values by Competitive ELISA
We previously demonstrated that RAP can completely prevent
the binding of Fab A8 to LRP, whereas Fab A8 is unable to compete for
the RAP/LRP interaction (13). Furthermore, it has been shown that RAP
also inhibits t-PA·PAI-1 complex binding to LRP (4), whereas the
reverse has not been observed. One possible mechanism to explain RAP
inhibition of ligand binding to LRP is the direct sterical hindrance
due to the close spatial association of the RAP- and ligand-binding
sites. To test this hypothesis, we performed a series of competition
experiments to assess the cross-competition of RAP-, t-PA·PAI-1
complex-, and Fab A8-binding to LRP and to Cl-II. First, we determined
the concentration required for half-maximal binding to the immobilized
LRP and to Cl-II. As can be seen in Fig. 4A,
half-maximal saturation of RAP binding to either LRP or Cl-II is
obtained at nearly identical concentrations, being 0.3 or 0.5 nM, respectively. From panels B and C
in Fig. 4, it can be deduced that the same conclusion is valid for
t-PA·PAI-1 complexes and Fab A8, with half-maximal saturation at
concentrations of 0.5 and 0.8 nM for t-PA·PAI-1 complex
and 0.4 and 0.8 nM for Fab A8 binding to LRP and Cl-II,
respectively. These results again demonstrate that the functional
integrity of the ligand-binding sites of Cl-II is fully maintained.
Next, we studied the cross-competition of RAP-, t-PA·PAI-1 complex-,
and Fab A8-binding to LRP and to Cl-II. As shown in Fig. 4D,
t-PA·PAI-1 complex binding to LRP is totally inhibited by RAP at an
IC50 value of 0.3 nM. Surprisingly, an
approximately 250-fold higher concentration of RAP is required to
inhibit t-PA·PAI-1 complex binding to Cl-II (IC50 = 76 nM). The latter residual inhibition is likely to be the
result of sterical hindrance of RAP for t-PA·PAI-1 complex binding as
a consequence of the partial overlap of the binding sites. This is
illustrated by the reverse experiment shown in Fig. 4F,
where similarly high concentrations are required to compete for RAP
binding to Cl-II and to LRP. Even more pronounced is the difference in
the RAP inhibition of Fab A8 binding to LRP and Cl-II (Fig.
4E). Whereas Fab A8 binding to LRP is efficiently prevented
by RAP at an IC50 of 0.2 nM, RAP inhibition of
Fab A8 binding to Cl-II was only partial and required high RAP
concentrations (approximately 40% inhibition at 1 µM
RAP). Results from the reverse competition experiment, depicted in Fig.
4G, show a complete lack of inhibition of RAP binding to
either LRP or Cl-II by Fab A8. From these experiments we conclude that
although RAP is fully able to bind to Cl-II, it is not able to display
its ligand inhibitory role.
To further analyze ligand binding by Cl-II of LRP, we have generated a subset of Cl-II fragments. We show that expression of soluble Cl-II by baby hamster kidney cell lines yields preparations with fully functional Ca2+-dependent ligand binding sites. Using these fragments we revealed the presence of an array of three distinct binding sites for t-PA·PAI and PAI-1, for RAP, and for the pseudo-ligand Fab A8. In some cases, the Kd values for the ligand interactions with the fragments are somewhat higher than those for the intact LRP molecule. This may be explained by the fact that for the Kd determinations of the cluster II fragments, the ligands were immobilized in contrast to the analysis of ligand binding to LRP, where the receptor was immobilized to the sensor chip. Attempts to immobilize the Cl-II fragments to the sensor chips failed, possibly due to the acidic nature of the proteins that is incompatible with the immobilization protocol. The lack of t-PA binding to the fragments and the observation that t-PA·PAI-1 complexes and PAI-1 both bind to the same domain support our previous observation that the PAI-1 moiety of the complex is the major determinant responsible for binding to Cl-II of LRP.2 Because no binding of free t-PA to Cl-II could be detected, at present we cannot exclude that a t-PA binding site outside Cl-II is responsible for the binding to LRP (5, 33, 34).
The arrangement of the RAP binding site in Cl-II, in between a partially overlapping t-PA·PAI-1 complex binding site and an independent binding site for Fab A8, renders the recombinant Cl-II an interesting tool to study the mechanism of inhibition of ligand binding by RAP. Surprisingly, even though the binding of RAP to Cl-II and LRP are quite similar, RAP is a poor inhibitor of ligand binding to the isolated Cl-II compared with its inhibition of ligand binding to the intact molecule. This observation makes it unlikely that RAP regulates ligand binding by a mechanism that involves mutually exclusive binding to overlapping binding sites. Our data are compatible with a model of ligand inhibition by RAP that assumes RAP-induced allosteric changes in the receptor, as suggested by Willnow et al. (18). First, different studies, including this one, confirm the presence of at least two RAP binding sites on LRP (12, 16, 20, 30, 31). One site is localized in Cl-II, the second in the cluster IV domain. A third, low affinity site is present on cluster III, as reported recently by Bu and Rennke (20). Second, the RAP molecule may be multivalent for LRP binding based on the internal triplication of LRP binding domains (14). At least two of these domains, present at the amino- and carboxyl-terminal ends, have been shown to bind to LRP with high affinity (19). The presence of multiple receptor binding sites may endow a single RAP molecule with the property to interact with different RAP binding sites on LRP, thereby inducing a conformational change in the receptor that prevents ligand binding. Cl-II may lack the structural requirements to maintain the proposed allosteric model needed for efficient RAP inhibition of ligand binding. The observed residual cross-competition of RAP and t-PA·PAI-1 complexes for binding to Cl-II may be explained by steric hindrance, consistent with the findings that sites are partially overlapping.
In a study by Saito et al., the importance of Cl-II in the homologous glycoprotein 330/megalin was demonstrated (32). This region contains a major epitope involved in the pathogenesis of nephritis, suggesting a highly exposed localization that may be well suited for protein-protein interactions. Our studies focusing on ligand binding to Cl-II of LRP support this view. Therefore, the data reported here may serve as a paradigm for ligand binding to other members of the low density lipoprotein receptor gene family as well.
Recently it was demonstrated that RAP functions as a novel type of specialized chaperone, inhibiting premature ligand binding, which may result in subsequent degradation of the receptor in the endoplasmic reticulum/proteasome pathway (14, 15, 20). The model described here may apply to the intracellular function of RAP as well.