| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
RESEARCH COMMUNICATION |
a Institute of Biochemistry, University of Vienna, A-1030 Vienna, Austria;
b Institute of Applied Genetics, University of Agriculture, Forestry, and Renewable Natural Resources in Vienna, Muthgasse, A-1190 Wien, Austria.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSDISCUSSIONREFERENCES |
|---|
Key Words: human rhinovirus • LDLR • viral inhibition • baculovirus expression
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSDISCUSSIONREFERENCES |
|---|
LDLR is responsible for the cellular homeostasis of cholesterol by taking up low density lipoproteins from the circulation (3). Malfunction due to the presence of mutations or deletions leads to an increased concentration of apolipoprotein-E and -B-associated cholesterol and other lipids in the serum, and ultimately to arteriosclerosis. LDLR is a multidomain protein. Its amino-terminal domain, which is involved in ligand binding, is composed of seven repeats of about 40 amino acids in length, each containing six cysteines (4), all of which are involved in disulfide bonds (5). This domain is followed by a region with similarity to the epidermal growth factor precursor and a short region carrying O-linked oligosaccharide chains. The protein is anchored in the plasma membrane via a stretch of hydrophobic amino acids. The cytoplasmic carboxyl terminus contains a typical internalization signal responsible for clustering in coated pits and for recycling to the cell surface (6).
Members of the LDLR family have similar structural elements. The very low density lipoprotein receptor (VLDLR) contains 8 ligand binding repeats (7); LDLR-related protein contains 31 and gp330 (also termed megalin) contains 36 such repeats, which are arranged in various clusters (8). These members of the LDLR family all serve as attachment proteins for minor group HRVs, although it is not known whether VLDLR and megalin are functional in viral infection (9, 10; unpublished data).
To gain insight into the site of interaction of minor group viruses with their receptors and into the early events in viral infection, we expressed a soluble fragment of LDLR in order to ultimately analyze complexes between virus and receptor by electron microscope imaging techniques. Due to the high content of cysteines, we expected that expression of an active protein would be difficult in the absence of the eukaryotic protein folding machinery. We thus chose insect cells in combination with the baculovirus system for expression. This paper reports on the successful production of large quantities of such a fragment; the protein is expressed in a native conformation and interacts with its ligands, ß-VLDL, and members of the minor group of human rhinoviruses. We demonstrate that this fragment inhibits viral infection in tissue culture by causing aggregation of multiple virions.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSDISCUSSIONREFERENCES |
|---|
Cloning strategies
Starting with plasmid pTZ1, a derivative of pLDLR-2 (4) that contains the whole LDLR cDNA, a fragment encompassing the seven ligand binding repeats was amplified by using two synthetic oligonucleotides (5'-TAATCCCGGGGACTGCAGTGGGCGACA-3' and 5'-TATACCCGGGTTCGTTGGTCCCGCACTC-3'), hybridizing at positions 56 to 76 and 934 to 951, respectively. Nucleotides shown in boldface are not complementary to LDLR cDNA, but were added to create restriction sites for XmaI.
The polymerase chain raction (PCR) fragment was amplified using pfu polymerase; pTZ1 was denatured at 95°C for 3 min, followed by 30 cycles of 95°C for 1 min, 56°C for 45 s, and 72°C for 2 min. The resulting 910 bp fragment was digested with XmaI and purified by agarose gel electrophoresis. It was ligated into the baculovirus cotransfection vector pVT_Bac_His2 (kindly provided by D. Joziasse, Amsterdam), a derivative of pVT_Bac (12) that had been linearized with XmaI, resulting in the plasmid pLDLR1–7h. Cotransfection of pLDLR1–7h with BaculoGold DNA into Sf9 cells was carried out according to the protocol provided by the manufacturer, yielding recombinant baculovirus vLDLR1–7h. Virus was used to infect Sf9 cells for production of rLDLR1–7h (recombinant rLDLR1–7h).
Sequencing strategies
The entire sequence encoding rLDLR1–7h was verified by PCR-didesoxy sequencing (Autoread, Pharmacia; Piscataway, N.J.) using 
[35S]ATP (Amersham, Arlington Heights, Ill.); the oligonucleotides are listed below. Lowercase letters indicate mispairing nucleotides (the primers were also designed for cloning purposes); the position of complementarity is given in parentheses. Note that nucleotides indicated by italics hybridize to sequences introduced during the cloning procedure. Nucleotides are always given in the 5'–3' direction: taatCCCGGGGACTGCAGTGGGCGACA, (56–76, repeat I); tatacccggGACGTGCTCCCAGGACGAGT, (321–340, repeat III); gacacccgggCACAGCGCAGTTTTCCTCG (681–699 end of repeat V); tataCCCGGGTTCGTTGGTCCCGCACTC, (934–951, repeat VII); CTTAGTCAACGTTGCCCTTG (within the melittin sequence in pVT_Bac_His2); GGTACCAGATCTTTAATG (within pVT_Bac_His2).
DNA sequencing confirmed the correct reading frame and identity of the LDLR fragment. Amino-terminal protein sequencing of Ni-NTA-purified material (100 pmol) was performed on a protein sequenator from Applied Biosystems (Foster City, Calif.).
Time course of expression and purification of LDLR fusion protein
Sf9 cells (1.5x106 cells/ml) being cultured in 1 l Erlenmeyer flasks in 400 ml Insect-Xpress (BioWhittaker, Walkersville, Md.), at 100 rpm and 27°C, were infected with wild-type (w.t.) and recombinant baculovirus, respectively, at a multiplicity of infection (MOI) of 5. At the times specified, 20 ml of the cell suspension was removed and the cells were collected by centrifugation. The supernatant was cleared in a Beckman Ti65 rotor at 50,000 rpm for 50 min at 4°C. Cells were lysed in 3 ml buffer A (20 mM Tris-HCl pH 7.5, 150 mM NaCl, 2 mM CaCl2, 16 mM 3-(3-cholamidopropyl) dimethylammonio-1-propanesulfonate, 1 mM phenylmethyl-sulfonyl-fluoride, 1 mM leupeptin) for 30 min on ice. The lysate was centrifuged in a Beckman TLA 100.3 rotor at 70,000 rpm for 20 min at 4°C. Cleared cell lysate (1.2 ml) or cell culture supernatant (8 ml) were applied onto Ni-NTA spin columns (Qiagen) equilibrated with buffer B (20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 2 mM CaCl2, 20 mM imidazole). Columns were washed five times with 600 µl of buffer B and adsorbed proteins were eluted with 2x 200 µl buffer B containing 250 mM imidazole. Eluants were combined and dialyzed against buffer C (5 mM Tris-HCl pH 7.5, 10 mM NaCl, 0.2 mM CaCl2) overnight and concentrated under vacuum to a final volume of 40 µl. Aliquots (20 µl) were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) under reducing conditions. For large-scale production of rLDLR1–7h, cell culture supernatant was harvested between 80 and 90 h postinfection (p.i.) and the process was upscaled accordingly.
For ligand binding experiments, material was adjusted with 5x Laemmli sample buffer without ß-mercaptoethanol to a final 1x concentration and separated by SDS-PAGE without heating. Proteins were transferred electrophoretically to a polyvinylidene difluoride (PVDF) membrane (Millipore, Bedford, Mass.) in 20 mM Tris-HCl, 150 mM glycine, pH 8.8, at 0.4 A for 2.5 h in the cold. Membranes were blocked with 20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 2 mM CaCl2, and 2% Tween 20 (blocking buffer) for 1 h. About 5 x 107 TCID50 (50% tissue culture infectious dose) of crude HRV2 was then added. This virus was prepared by polyethylene glycol precipitation of infected HeLa cell culture supernatant and suspension of the pellet in 1/50 vol phosphate-buffered saline. After incubation overnight at 4°C, the membranes were washed twice with 20 mM Tris-HCl, pH 7.4, 150 mM NaCl (TBS) containing 2 mM CaCl2 and 0.1% Tween 20 (incubation buffer) for 10 min each and incubated with rabbit anti-HRV2 hyperimmune serum diluted 1:2000, followed by HRP-conjugated goat anti-rabbit antiserum at a 1:5000 dilution. For incubation with [125I]-labeled ß-VLDL (3x106 cpm/µg; kindly provided by M. Huettinger, Vienna), Tween 20 was replaced with 2.5% bovine serum albumin in all solutions. LDLR from HeLa cell membranes was used as the control; cells were homogenized, nuclei were removed by centrifugation, and an S100 was prepared. Membranes in the pellet were solubilized with buffer A and insoluble material was removed by recentrifugation.
Inhibition of viral infection
vLDLR1–7h-infected Sf9 cell supernatant (10 µl) was mixed with 90 µl infection medium (minimal essential medium containing 2% fetal calf serum and 30 mM MgCl2) and serial twofold dilutions were made in the same medium. To each dilution (50 µl), 500 TCID50 of the respective HRV serotype in 50 µl infection medium was added and the mixtures were incubated for 1.5 h at 34°C. They were then transferred onto subconfluent monolayers of Rhino HeLa cells (Flow Laboratories, McLean, Va.) in 96-well plates (containing 100 µl infection medium) and incubated for 3 days at 34°C. Cells remaining attached to the plastic were stained with amido black (0.1% in acetic acid/methanol/water 10/40/50 v/v). Noninfected cells and cells infected in the absence of vLDLR1–7h were used as positive and negative controls, respectively.
To determine the neutralization mechanism, two 2.5 µg (3.12x10-13 M) samples of sucrose gradient purified HRV2 (13) at 2.3 x 108 TCID50 were incubated with 15.6 µg (3.12x10-10 M) rLDLR1–7h in 100 µl infection medium for 1.5 h at 34°C. Samples were then centrifuged in an Eppendorf centrifuge at 14,000 rpm for 4 min. Calculations showed that these conditions led to pelleting of complexes, with a sedimentation constant exceeding 1500S. Pellets were resuspended in 100 µl infection medium. To dissociate putative complexes, a final concentration of 100 mM of EDTA was added to one sample, whereas the control sample received the same volume of water. Incubation was for 30 min at 34°C, whereupon the samples were diluted 100-fold with infection medium. Virus was assayed by placing 200 µl of serial twofold dilutions of the 100-fold prediluted samples onto HeLa cell monolayers grown in 96-well plates. Infectious virus in the supernatants was assayed identically. As controls, all experiments were carried out in parallel in the absence of rLDLR1–7h. After incubation of the plates at 34°C for 24 h, wells were stained with amido black. Wells with the most prominent difference in tissue damage were selected for quantification; the stain was dissolved in 100 µl of 1 M NaOH and A560 was determined in a microplate reader (14).
RESULTS
Expression of LDLR fragments
A cDNA fragment encoding the seven LDLR ligand binding repeats (AA 22–317) was PCR amplified from plasmid pTZ1 and subcloned into pVT_Bac_His2, a derivative of pVT_Bac (12), producing pLDLR1–7h. Sf9 cells were cotransfected with the vector and BaculoGold DNA to generate recombinant baculovirus (vLDLR1–7h) encoding the LDLR fragment as a fusion with the melittin signal sequence and a carboxyl-terminal hexa his-tag. The primary structures of LDLR and of the recombinant protein encoded in the vector are shown schematically in 
Fig. 1.
Recombinant baculovirus was recovered from the supernatant of the transfected cells and used to infect suspension cultures.
|
To monitor expression of the desired protein, cells were infected with an MOI of 5. At various times p.i. (as indicated in Fig. 2),
cells were pelleted, pellets were solubilized in detergent buffer, and insoluble material was removed by low-speed centrifugation. Aliquots from the detergent-solubilized cellular material and the cell supernatant (1.2x107 cells at the time of infection) were applied to Ni-NTA spin columns and adsorbed material was eluted and concentrated to 40 µl in a Speedvac concentrator; 20 µl was then run on a 10% polyacrylamide-SDS gel under reducing conditions. The gel after staining with Coomassie blue is shown in Fig. 2A. A protein with an apparent Mr of about 55 kDa, which was retained on the Ni-NTA material, appeared at around 48 h p.i. in the extracts from cells infected with vLDLR1–7h. The amount of this protein further increased up to 96 h p.i., whereupon protein expression decreased. As judged from the Coomassie stained gels, about 30% of the expressed protein was present in the cell pellet ( Fig. 2A;) and the remaining 70% was secreted into the supernatants ( Fig. 2B). As controls, noninfected cells and cells infected with w.t. baculovirus were also analyzed at 72 h p.i., the time of maximum expression of the fusion protein. Although some minor bands of proteins binding unspecifically to the column material were evident, the 55 kDa protein was absent in the Ni-NTA column eluate obtained from mock-infected cells or from cells infected with w.t. baculovirus. Production of the recombinant protein from 1 l of Sf9 suspension cells yielded up to 10 mg/l of electrophoretically pure protein.
|
To investigate whether the fusion protein released into the medium was present as aggregates or as a soluble protein, tissue culture supernatants were subjected to Ni-NTA column chromatography, as described before. Eluted material was dialyzed against TBS/2 mM CaCl2 and aggregates sedimenting at more than 14 S were collected by high-speed centrifugation. The resulting pellet and supernatant were analyzed by SDS-PAGE under reducing conditions. Coomassie blue staining of the gel revealed that all rLDLR1–7h remained in the supernatant, with no material recovered in the pellet fraction (data not shown).
Ligand binding properties of the expressed protein
To confirm the identity of the expressed protein with the LDLR fragment encoded in the recombinant baculovirus, 20 µl of cell culture supernatant 90 h p.i. was subjected to SDS-PAGE under nonreducing conditions and separated proteins were transferred onto PVDF membranes. The membranes were blocked and then incubated with [125I]-labeled rabbit ß-VLDL. As shown in Fig. 3,
lane 2, ß-VLDL bound to two bands migrating with an apparent Mr of about 50 and 95 kDa, respectively. The lower Mr polypeptide thus corresponds to the 55 kDa band found upon gel electrophoresis under reducing conditions (compare to 
Fig. 2). As a control, an extract of HeLa cell membranes was applied in lane 1. Incubation with ß-VLDL revealed a band migrating with an apparent Mr of 120 kDa, which represents the entire human LDLR. Material reactive with [125I]-labeled ß-VLDL was absent from mock-infected Sf9 cells (lane 3) or from cells infected with w.t. baculovirus (lane 4).
|
As further control, an identical blot was incubated with rabbit hyperimmune antibodies raised against bovine LDLR; as seen in lane 6, the antibodies reacted with the expressed protein. The two bands revealed by the antibodies thus correspond to the monomeric and probably to a dimeric form of the recombinant protein, respectively. In HeLa cell membrane extracts, a 120 kDa band was again seen upon incubation with the antiserum (lane 5; compare with lane 1), but no bands were evident in supernatants from mock-infected (lane 7) or w.t. baculovirus-infected Sf9 cells (lane 8). These experiments lend additional proof for the identity of the expressed protein with human LDLR.
We then asked whether minor group HRVs would bind to the recombinant protein. We also wanted to determine the exact time of maximum recovery of binding activity, and so the established virus ligand blotting method (15) was used in a somewhat modified form. Recombinant protein was purified from aliquots of infected cells and cell culture supernatants at different times p.i., as described in Fig. 2. The material was separated on an SDS-polyacrylamide gel run under nonreducing conditions and transferred to a PVDF membrane; virus binding was revealed by incubation with crude HRV2 at 5 x 107 TCID50. Bound virus was then detected with HRV2 antiserum, HRP-conjugated anti-rabbit IgG serum, and chemiluminescent substrate. As expected from the results shown in Fig. 2, maximum binding of HRV2 was seen between 72 and 96 h p.i. ( Fig. 4).
When binding activity recovered from the cells (lanes C) was compared with that present in the cell culture media (lanes S), almost no activity was detectable in cell extracts at any time whereas HRV2 binding activity was strongly enriched in the samples from the supernatant. The strongest binding was to a protein migrating with an apparent Mr of 55 kDa; an additional, weaker band of about 95 kDa was also seen. Molecular weights of the LDLR fragment correspond to those revealed by binding of ß-VLDL (see Fig. 3).
|
In accordance with the known requirement of Ca2+ for binding activity, addition of EDTA during incubation of the blot with HRV2 led to loss of viral attachment (data not shown). Moreover, boiling in SDS in the absence of reducing agent for 45 min did not change the ratio between monomer and dimer. Similar to native LDLR, the recombinant fragment withstands these harsh conditions without any substantial loss in ligand binding activity (16; data not shown).
Amino-terminal amino acid sequencing of rLDLR1–7h
The amino-terminal amino acids of rLDLR1–7h were determined with an automatic sequencer as DPSPGTAV. . . . (see Fig. 1). Therefore, signal peptidase cleavage occurs carboxyl-terminally to the last amino acid residue of the melittin-derived signal sequence. The first ligand binding repeat of the mature recombinant protein thus contains four additional amino acids resulting from the cloning procedure. The calculated Mr of rLDLR1–7h is therefore 35,467 Da. The large difference between calculated and observed Mr on polyacrylamide gels (see Figs. 2–4) apparently reflects the well-known aberrant behavior of LDLR upon gel electrophoresis, which is related to the presence of three disulfide bonds per ligand binding repeat. It might also be due to modification at three potential N-glycosylation sites present in the truncated receptor (17, 18).
Soluble LDLR fragments inhibit minor group HRV infection
We have previously shown that detergent-solubilized OVR, the avian homologue of mammalian VLDLR, inhibits infection of HeLa cells by HRV2 (9). However, since this receptor is not involved in viral infection (rhinoviruses do not replicate in any species except humans and other primates), it was of interest whether soluble LDLR fragments would also inhibit infection. Various HRV serotypes at 500 TCID50 were preincubated in a final volume of 100 µl of infection medium containing different amounts of vLDLR1–7h-infected tissue culture supernatant, as indicated in Fig. 5.
HeLa cells in 96-well plates were then challenged with the incubation mixtures and incubated for 3 days at 34°C. Addition of rLDLR1–7h inhibited infection by the minor group viruses HRV1A, HRV2, HRV29, HRV30, HRV47, and HRV49 in a concentration-dependent manner, whereas the major group virus HRV14 was not inhibited. Slightly different amounts of receptor were necessary to prevent growth of the various serotypes (compare HRV1A and HRV49).
|
Affinity chromatography yielded maximally 10 mg of rLDLR1–7h per liter of infected tissue culture supernatant (see above). This allows us to estimate the minimal amount of rLDLR1–7h necessary to completely inhibit viral growth (i.e., 6.3 ng contained in 0.63 µl, see Fig. 5). Based on a final volume of 200 µl medium per well, this corresponds to a concentration of 31.5 ng/ml at most. For a better quantification, the protein concentration and inhibitory activity were determined before (crude tissue culture supernatant) and after purification (Ni-NTA eluate). As seen in Table 1,
specific activity of the receptor fragment increased about 400-fold upon purification on the Ni-NTA column. In this experiment, 50% cell protection was achieved with about 25 ng of pure protein per well, indicating a slight underestimation of the receptor content in the tissue culture supernatant in the previous experiment.
|
As controls, supernatants of Sf9 cells either mock infected or infected with w.t. baculovirus as well as flow-through from the Ni-NTA column were used and resulted in no inhibition of infection (data not shown). This proves that inhibition of minor receptor group HRVs by rLDLR1–7h is specific.
Mechanism of inhibition of infection
Soluble viral receptor can inhibit infection by blocking attachment of the virus to the cellular receptor, by uncoating of the viral genomic RNA, or by aggregation (19). To investigate whether rLDLR1–7h mediates viral aggregation, mixtures of HRV2 and receptor were subjected to centrifugation. First, the highest amount of purified virus being neutralized by a given amount of purified rLDLR1–7h was determined. The soluble receptor (160 ng) was incubated with increasing quantities of HRV2 in 100 µl infection medium for 90 min at 34°C. Viral neutralization was then monitored on HeLa cell monolayers in 96-well plates as protection from a cytopathic effect after 24 h upon challenge with the incubation mixture. In this experiment, 160 ng of purified rLDLR1–7h protected the cells from lysis upon challenge with 25 ng of virus whereas 50 ng virus resulted in destruction of the monolayer (data not shown). In the absence of rLDLR1–7h, the cells were lysed in all wells containing from 2.5 to 250 ng of purified virus. Chelating of Ca2+ ions leads to inactivation of LDLR; therefore, EDTA was used to dissociate putative complexes between HRV2 and rLDLR1–7h. To keep its concentration low in the subsequent infectivity assay, total amounts of virus and receptor were increased 100-fold with respect to the previous experiment to allow for sufficient dilution. The mixture was incubated as before and subjected to centrifugation under conditions that led to pelleting of particles sedimenting at 1500S (i.e., aggregates of about 10 virions) (20). The pellet was resuspended in infection medium and infectivity was determined directly or after dissociation of putative virus/receptor complexes with EDTA. Infectivity in the supernatant was also assayed after incubation with or without EDTA. Samples were prediluted 100-fold and HeLa cells were challenged with serial twofold dilutions. The cells were incubated for 24 h and stained with amido black. The cytopathic effect was quantified in those wells exhibiting the strongest difference between control incubations and incubations in the presence of rLDLR1–7h ( Fig. 6).
In the absence of rLDLR1–7h, almost all HRV2 remained in the supernatant, with only little virus being recovered from the pellet regardless of EDTA incubation. In contrast, when rLDLR1–7h was present, only low viral infectivity was found in the pellet or in the supernatant without EDTA treatment. Incubation with EDTA led to recovery of infectivity from the pellet; comparison with the control sample (supernatant without rLDLR1–7h) shows that EDTA treatment of the pellet restored viral infectivity almost entirely. Virions thus formed large aggregates exceeding sedimentation constants of 1500S in the presence of soluble receptor.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSDISCUSSIONREFERENCES |
|---|
Fig. 1), members of the LDLR family are supposed to be difficult to express in a native conformation in bacteria. Thus we used the baculovirus system and produced large quantities of a soluble fragment of LDLR encompassing the seven ligand binding repeats. A monomer and a form of the protein with roughly twice the apparent molecular weight were obtained. This latter form probably constitutes a dimer, since it is stable in detergent and reverts to the lower molecular form upon reduction (compare Fig. 2and Fig. 3). The existence of LDLR mutants or recombinant LDLR fragments with an abnormal tendency to form dimers has been reported previously (17, 21). Both forms of the expressed LDLR fragment are active in binding the natural ligand ß-VLDL; they also bind HRV2, a member of the human rhinovirus minor receptor group, indicating that the ligand binding repeats are sufficient for virus recognition. HRV binding to rLDLR1–7h transferred to PVDF membranes is very sensitive. The soluble receptor was easily detectable with a crude HRV2 virus preparation, using virus-specific antiserum in conjunction with a chemiluminiscence detection method.
Provided that soluble rLDLR1–7h attaches to HRVs in solution, it should inhibit virus binding to cellular receptors and thus block infection. Indeed, HeLa cells were protected from infection by all minor group viruses tested whereas no inhibition was seen for the major group virus HRV14. Neutralization was very efficient. Only 0.08 µl of vLDLR1–7h-infected Sf9 cell supernatant (corresponding to about 150 Sf9 cells) was sufficient to inhibit cell damage by 500 TCID50 of HRV1A, HRV29, HRV30, and HRV47 ( Fig. 5). All minor group viruses were completely inhibited with 0.63 µl of supernatant. However, a comparison of the inhibitory potency of various batches revealed substantial variations that were related to the density and status of the Sf9 cells at the time of infection. Inhibition was tested using the same number of infectious particles for each serotype as determined by TCID50. For rhinoviruses, however, the ratio between infectious and noninfectious particles might vary from 1:24 to 1:240 (22). The concentration of rLDLR1–7h necessary to neutralize a given HRV serotype certainly depends on the total number of virus particles competing for the receptor but escaping detection by our assay. Considering all these points, it is possible to roughly estimate the minimal inhibitory concentration as less than 30 ng/ml for all serotypes examined. This is much less than that reported for the inhibition of major group HRVs by soluble ICAM-1, which is between 0.1 and 7.9 µg/ml (23, 14).
Taking into account the calculated Mr of soluble ICAM-1 (82 kDa) and of the LDLR fragment used in our study (35 kDa) as well as the slightly different assay conditions, these numbers indeed indicate at least the same if not a higher inhibitory potency of the LDLR fragment as compared to ICAM-1. Ni-NTA does not select for active material, but retains only those molecules that contain the hexa-his tag. In an attempt to enrich for receptors that are correctly folded and are thus active in virus binding, we used chromatography on ß-VLDL columns for further purification. Preliminary experiments indicate that this results in a substantial increase of the specific inhibitory activity of rLDLR1–7h yielding receptors suitable for the determination of exact minimal inhibitory concentrations.
Until now, attempts to detect stoichiometric complexes assembled between rLDLR1–7h and HRV2 by rate zonal centrifugation have failed. Because of the unequivocal demonstration of infection inhibition by rLDLR1–7h (see Fig. 5), it is assumed that complexes between virus and receptor must exist, but are not stable enough to persist under conditions of sucrose gradient separation. This might be due to a low affinity of the monomeric receptor. The situation could be similar to that of Fab fragments prepared from the monoclonal antibody 8F5, which fail to attach to HRV2, whereas bidentate binding of the entire 8F5 IgG molecule is stable enough for a complex to be isolated (24). Recent work on echovirus 7 also demonstrates that despite recombinant soluble CD55-inhibited virus infection, no complexes between virus and receptor could be isolated by sucrose density gradient centrifugation (25). The avidity of rLDLR1–7h is obviously increased upon immobilization to PVDF membranes, leading to high local concentrations of the protein. In vivo, this local concentration is certainly further potentiated upon invagination of the cell membrane, leading to coalescence of many LDLR molecules, which all interact with one single virion and thus facilitate viral entry. Nevertheless, this presumed low affinity is in contradiction to the strong inhibitory activity of the soluble receptor fragment.
To investigate the mechanism of infection inhibition, we incubated purified HRV2 with a 1000-fold molar excess of purified rLDLR1–7h and wondered whether infectivity was being reduced as a consequence of aggregation. As demonstrated in Fig. 6, HRV2 indeed formed aggregates with rLDLR1–7h of a size corresponding to assemblies that contain at least 10 virions. These could be dissociated and infectivity could be restored by addition of EDTA, which complexes Ca2+ ions that are required for the function of LDLR. This is in line with earlier experiments indicating that LDLR serves only as a vehicle for virus internalization (26, 27), whereas soluble ICAM-1 also uncoats major group viruses upon incubation in vitro (19, 28). How aggregation occurs is not clear at this time, since rLDLR1–7h is not aggregated in the absence of virus and does not sediment under conditions that lead to the pelleting of particles >14S. Assuming 60 identical receptor binding sites on the viral capsid (29), it is likely that aggregation is either induced as a consequence of the high local concentration of rLDLR1–7h attained upon multiple binding to the viral shell or by the presence of multimeric forms of the receptor present in the preparation. Therefore, soluble LDLR inhibits infection by competition with the viral receptor present on the cell surface and/or by reducing the number of infectious units due to aggregation.
In summary, we succeeded in expressing a fragment of LDLR in an active, soluble form in great quantity, with strong in vitro inhibitory activity for minor group human rhinoviruses. We are currently investigating whether shorter fragments of LDLR also inhibit infection; such fragments, which might lend themselves to bacterial expression, could be seen as a starting point for the development of therapeutically useful compounds.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Abbreviations: AP, alkaline phosphatase; HRV, human rhinovirus; ICAM-1, intercellular adhesion molecule; LDLR, low density lipoprotein receptor; MOI, multiplicity of infection; p.i., postinfection; TCID50, 50% tissue culture infectious dose; NBT, nitroblue tetrazolium salt; BCIP, 5-bromo-4-chloro-3-indolylphosphate; HRP, horseradish peroxidase; w.t., wild-type; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis; PCR, polymerase chain reaction; PVDF, polyvinylidene difluoride; VLDLR, very low density lipoprotein receptor; vLDLR1–7h, recombinant baculovirus; rLDLR1–7h, recombinant rLDLR1–7h.
Received for publication October 2, 1997. Accepted for publication January 19, 1998.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSDISCUSSIONREFERENCES |
|---|
Lys mutation. Biochim. Biophys. Acta 1255, 285–292[Medline]
This article has been cited by other articles:
|
|
 |
|
  B. M. Krenn, B. Holzer, E. Gaudernak, A. Triendl, F. J. van Kuppeveld, and J. Seipelt Inhibition of Polyprotein Processing and RNA Replication of Human Rhinovirus by Pyrrolidine Dithiocarbamate Involves Metal Ions J. Virol., November 15, 2005; 79(22): 13892 - 13899. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Vlasak, I. Goesler, and D. Blaas Human Rhinovirus Type 89 Variants Use Heparan Sulfate Proteoglycan for Cell Attachment J. Virol., May 15, 2005; 79(10): 5963 - 5970. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Duque, M. LaRocco, W. T. Golde, and B. Baxt Interactions of Foot-and-Mouth Disease Virus with Soluble Bovine {alpha}V{beta}3 and {alpha}V{beta}6 Integrins J. Virol., September 15, 2004; 78(18): 9773 - 9781. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Herdy, L. Snyers, M. Reithmayer, P. Hinterdorfer, and D. Blaas Identification of the Human Rhinovirus Serotype 1A Binding Site on the Murine Low-Density Lipoprotein Receptor by Using Human-Mouse Receptor Chimeras J. Virol., July 1, 2004; 78(13): 6766 - 6774. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Khovidhunkit, M.-S. Kim, R. A. Memon, J. K. Shigenaga, A. H. Moser, K. R. Feingold, and C. Grunfeld Thematic review series: The Pathogenesis of Atherosclerosis. Effects of infection and inflammation on lipid and lipoprotein metabolism mechanisms and consequences to the host J. Lipid Res., July 1, 2004; 45(7): 1169 - 1196. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Reithmayer, A. Reischl, L. Snyers, and D. Blaas Species-Specific Receptor Recognition by a Minor-Group Human Rhinovirus (HRV): HRV Serotype 1A Distinguishes between the Murine and the Human Low-Density Lipoprotein Receptor J. Virol., June 14, 2002; 76(14): 6957 - 6965. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Reischl, M. Reithmayer, G. Winsauer, R. Moser, I. Gosler, and D. Blaas Viral Evolution toward Change in Receptor Usage: Adaptation of a Major Group Human Rhinovirus To Grow in ICAM-1-Negative Cells J. Virol., October 1, 2001; 75(19): 9312 - 9319. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Suzuki, M. Yamaya, M. Kamanaka, Y. X. Jia, K. Nakayama, M. Hosoda, N. Yamada, H. Nishimura, K. Sekizawa, and H. Sasaki Type 2 rhinovirus infection of cultured human tracheal epithelial cells: role of LDL receptor Am J Physiol Lung Cell Mol Physiol, March 1, 2001; 280(3): L409 - L420. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. C. Marlovits, C. Abrahamsberg, and D. Blaas Soluble LDL Minireceptors. MINIMAL STRUCTURE REQUIREMENTS FOR RECOGNITION OF MINOR GROUP HUMAN RHINOVIRUS J. Biol. Chem., December 11, 1998; 273(50): 33835 - 33840. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Bayer, D. Schober, E. Prchla, R. F. Murphy, D. Blaas, and R. Fuchs Effect of Bafilomycin A1 and Nocodazole on Endocytic Transport in HeLa Cells: Implications for Viral Uncoating and Infection J. Virol., December 1, 1998; 72(12): 9645 - 9655. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. C. Marlovits, C. Abrahamsberg, and D. Blaas Very-Low-Density Lipoprotein Receptor Fragment Shed from HeLa Cells Inhibits Human Rhinovirus Infection J. Virol., December 1, 1998; 72(12): 10246 - 10250. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Bayer, D. Schober, M. Huttinger, D. Blaas, and R. Fuchs Inhibition of Clathrin-dependent Endocytosis Has Multiple Effects on Human Rhinovirus Serotype 2 Cell Entry J. Biol. Chem., February 2, 2001; 276(6): 3952 - 3962. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. M. Okun, R. Moser, B. Ronacher, E. Kenndler, and D. Blaas VLDL Receptor Fragments of Different Lengths Bind to Human Rhinovirus HRV2 with Different Stoichiometry. AN ANALYSIS OF VIRUS-RECEPTOR COMPLEXES BY CAPILLARY ELECTROPHORESIS J. Biol. Chem., January 5, 2001; 276(2): 1057 - 1062. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
