The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Toh, M.-L.
Right arrow Articles by Miossec, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Toh, M.-L.
Right arrow Articles by Miossec, P.
Right arrowPubmed/NCBI databases
*Compound via MeSH
*Substance via MeSH
Medline Plus Health Information
*Genes and Gene Therapy
*Joint Disorders
*Rheumatoid Arthritis
The Journal of Immunology, 2005, 175: 7687-7698.
Copyright © 2005 by The American Association of Immunologists

Enhancement of Adenovirus-Mediated Gene Delivery to Rheumatoid Arthritis Synoviocytes and Synovium by Fiber Modifications: Role of Arginine-Glycine-Aspartic Acid (RGD)- and Non-RGD-Binding Integrins1

Myew-Ling Toh*, Saw-See Hong{dagger}, Fons van de Loo§, Laure Franqueville{dagger}, Leif Lindholm{ddagger}, Wim van den Berg§, Pierre Boulanger2,{dagger} and Pierre Miossec2,*

* Department of Immunology and Rheumatology, Mixed Unit Civil Hospital of Lyon-BioMérieux, Edouard Herriot Hospital, Lyon, France; {dagger} Laboratoire de Virologie et Pathogenèse Virale, Centre National de la Recherche Scientifique-University of California Biomechanics Laboratory, Faculté de Médecine RTH Laennec, Lyon, France; {ddagger} Department of Medical Microbiology and Immunology, University of Göteborg, and Got-A-Gene, Göteborg, Sweden; and § University Medical Center Nijmegen, Nijmegen Center for Molecular Life Sciences, Rheumatology and Advanced Therapeutics, Nijmegen, The Netherlands


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Rheumatoid arthritis (RA) fibroblast-like synoviocytes (FLS) do not express the coxsackie-adenovirus (Ad) receptor and are poorly permissive to Ad serotype 5 (Ad5). Genetically modified, coxsackie-Ad receptor-independent Ad5 vectors were studied for gene delivery in human RA FLS and synovium explants and murine collagen-induced arthritis. Short-fiber Ad5 vectors with seven fiber shaft repeats Ad5GFP-R7-knob, Ad5GFP-R7-arginine-glycine-aspartic acid (RGD) (RGD-liganded), and Ad5GFP{Delta}knob (knob-deleted) were compared with Ad5GFP-FiWT, a conventional wild-type (WT) Ad5 vector. Gene transfer by Ad5GFP-R7-knob and Ad5GFP-R7-RGD was 40- to 50-fold and 25-fold higher, respectively, than Ad5GFP-FiWT in FLS. Ad5GFP{Delta}knob was more efficacious than its knob-bearing version Ad5GFP-R7-knob in FLS transduction. Virus attachment and entry required RGD- and LDV-binding integrins including {alpha}v, {alpha}v{beta}3, av{beta}5, and {beta}1. Ad5GFP-R7-knob infection of FLS was partially neutralized by synovial fluid (SF), but remained 30- to 40-fold higher than Ad5GFP-FiWT in the presence of SF. Ad5GFP{Delta}knob was partially neutralized by SF at low virus input, but escaped viral neutralization by SF at higher virus input. Gene transfer to human synovium ex vivo explants and murine collagen-induced arthritis in vivo was also more efficient with short fiber-modified vectors (with and without the knob domain) than Ad5GFPFiWT. Gene transfer by short fiber-modified vectors was enhanced by inflammatory cytokines in vitro and in the presence of inflammation in murine synovium in vivo. Our data indicated that the highly efficient gene delivery RA was mediated by RGD- and non-RGD-binding integrins and enhanced by inflammation. Short fiber modifications with knob ablation may be a strategy to enhance gene delivery, reducing vector dose and vector-induced inflammation and toxicity.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Rheumatoid arthritis (RA)3is characterized by chronic synovial inflammation and pannus formation leading to joint destruction. Biological agents such as anti-TNF therapies are efficacious, but may be limited by cost and systemic side effects. Gene therapy has been developed as an alternative approach in targeted delivery of therapeutic agents. Among several viral vectors used, nonreplicative adenovirus (Ad) serotype 5 (Ad5) vectors are relatively safe and well studied in animal models. Although the efficacy of gene therapy in animal models of arthritis has been demonstrated, applicability to human disease is less well documented (1, 2, 3). Key issues include efficient gene delivery to synovial targets.

Ad5 cell surface receptors are essential parameters of virus infection. Initial cell attachment is mediated predominantly by binding of the knob domain of the Ad5 fiber to the high-affinity coxsackie-Ad receptor (CAR) (4). Subsequent endocytosis and membrane permeabilization is mediated by binding of {alpha}v-integrins to arginine-glycine-aspartic acid (RGD) motifs in viral penton base (Pb) capsomers, located at each of the 12 apices of the virus icosahedral capsid (5, 6, 7). Attachment and endocytic receptors co-operate for viral entry, which is highly dependent on fiber length and flexibility (8). In human RA and murine fibroblast-like synoviocytes (FLS), CAR is absent, whereas integrins are highly expressed (9, 10, 11, 12, 13). The lack of CAR may account for low permissiveness to Ad5 (6, 9, 14, 15, 16, 17, 18). In animal models, the requirement for increased Ad5 vector dose to obtain efficient gene transduction may induce toxicity including arthritis flares (2). In RA FLS and synovium, Ad5 cellular entry mechanisms have not been previously described. Defining these mechanisms may enable the development of improved vectors to specific synovial targets.

In diseases such as cancer and cystic fibrosis, also hampered by the absence of CAR in cells and tissue, efforts to improve vector efficiency have focused on genetically modifying the knob domain of Ad5 to achieve CAR-independent cell entry (19, 20, 21, 22, 23, 24). In arthritis, this has been less well explored. Approaches have included retargeting of Ad5 vectors to alternative receptors by insertion of the integrin-binding RGD motif in the HI loop or at the C terminus of the fiber, or by replacement of the Ad5 fiber by the fiber of another serotype (9, 13, 25). Recently, an RGD-liganded Ad5 increased efficiency of transduction of the IL-1 receptor antagonist with improved anti-inflammatory effects in murine collagen-induced arthritis (CIA) (13, 26).

An alternative strategy to improve Ad5 vector tropism in CAR-negative cells is to shorten the fiber shaft (8, 9, 19, 27). Members of subgroup B Ad such as Ad3, which have naturally short fibers (e.g., serotypes 11, 16, 35), or chimeric Ad5 vectors carrying subgroup B fibers more efficiently transduced RA FLS compared with Ad5 (9, 14). In the present study, we describe the use of short fiber-modified Ad5 vectors in efficient gene transfer to CAR-negative human RA FLS in vitro, synovium ex vivo explants, and murine CIA in vivo. We also defined the receptor usage and mechanism of viral entry in RA FLS.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Cells, synovium explants, and synovial fluid (SF)

FLS were obtained from synovial tissue from RA patients undergoing joint surgery who fulfilled the American College of Rheumatology criteria for RA (28). FLS were isolated by enzyme digestion and cultured as described elsewhere (29, 30). Cells beyond the third passage were >99% negative for CD45, CD1, CD3, CD19, CD14, and HLA-DR, and positive for the expression of CD10, CD44, and CD54 (BD Biosciences), and were used between passages 4 and 9. An ex vivo synovium explant model was prepared as described (31, 32). In brief, fat and fibrous tissue were removed from freshly harvested synovial tissue and cut into small explants with a volume of ~3 mm3. Explants were cultured in quadruplicate in 6-well plates in DMEM/10% FCS (Invitrogen Life Technologies), 100 U/ml penicillin, 100 µg/ml streptomycin, and 2 mM L-glutamine, at 37°C in a humidified 5% CO2 incubator. After 3–4 days, a cellular outgrowth characterized as FLS are observed migrating out from the edge of synovial explants. FLS were characterized as described above. All data are the results of at least three separate experiments using FLS and synovial tissue from three different human RA donors. A549 cells (lung alveolar carcinoma; CCL 185) were provided by the Clinical Virology Laboratory (Hospices Civils de Lyon, Domaine Rockfeller, Lyon, France). SF samples from six women with RA (mean age, 45 years old; 70% rheumatoid factor positive; mean disease duration, 10 years) were collected and stored at –20°C.

Recombinant adenoviral vectors

Ad5 vectors were E1-deleted, with the E1 region replaced by a GFP expression cassette under the control of a CMV promoter. A conventional Ad5 vector, Ad5GFP-FiWT carrying wild-type (WT) fibers with 22 shaft repeats (Fig. 1A), was used as a control and propagated in the E1-trans-complementing cell line HEK-293 (abbreviated 293 cells). Short fiber-modified vectors were propagated in the dual E1+fiber-trans-complementing cell line (referred to as 293-Fiber cells; Ref.22), then in 293 cells for the last amplification step (19). The construction of short fiber-modified vectors has been described elsewhere (19, 22, 33).



View larger version (27K):
[in this window]
[in a new window]
  		FIGURE 1. Schematic representation of adenoviral vectors carrying the WT fiber or recombinant short fibers. The fiber tail domain is shown by {square}, the shaft domain by {cjs2112}, the extrinsic trimerization motif by {blacksquare}, and the knob domain by {circ}. A, Control Ad5 vector (Ad5GFP-FiWT) carried WT fibers that consisted of the following domains from the N terminus to C terminus: a tail domain linked to the Pb, followed by 22 shaft repeats, and an intrinsic trimerization domain including the knob domain. B–D, Ad5 vectors with short-shafted fibers consisted of the tail domain, the shortened shaft with seven repeats, followed by an extrinsic trimerization motif. The fiber shaft was terminated by the native knob domain as in Ad5GFP-R7-knob (B), a linear RGD sequence as in Ad5GFP-R7-RGD (C), or an opale stop codon as in Ad5GFP{Delta}knob (D).

 
In brief, the Ad5 fiber backbone contained the following domains from N terminus to C terminus: 1) the fiber tail; 2) seven shaft repeats; 3) a nonviral trimerization signal (PDVASLRQQVAELQGQVQHLQAAFSQYKKVELFPNG) called the neck region peptide from the human lung surfactant protein D; 4) a tridecapeptide linker sequence (AKKLNDAQAPKSD); and 5) various C-terminal sequences. Ad5GFP-R7-knob carried the native knob domain at the C terminus of the shortened shaft (Fig. 1B), whereas "knobless" vectors Ad5GFP-R7-RGD carried a linear RGD peptide (Fig. 1C), and Ad5GFP{Delta}knob carried no ligand (Fig. 1D). Virus stocks were purified by CsCl ultracentrifugation (34). Infectious titers (concentrations of infectious virions) were determined by plaque titration assay on CAR and integrin-expressing 293 cells, and values were expressed as PFU per milliliter. The number of physical virus particles (physical virus particles = noninfectious + infectious virions) in virus stocks was determined using a Bradford protein assay (Bio-Rad) with BSA (Bio-Rad) as the standard, taking into account the mass for a single virion (2.9 x 10–16 g) and the value of 3.4 x 1012 virions/mg total protein (22). Data was expressed as physical virus particles per milliliter.

In vitro transduction of RA FLS and ex vivo transduction of synovium explants

Cells and tissues were infected with Ad5GFP-FiWT or fiber-modified vectors Ad5GFP-R7-knob, Ad5GFP-R7-RGD and Ad5GFP{Delta}knob at a multiplicity of infection (MOI) of 2, 20, or 200 PFU/cell for 24–72 h, and GFP expression measured by flow cytometry (fluorescence activator cell sorter scan; BD Biosciences) or fluorescence microscopy using an Axiovert 135 microscope (Zeiss) equipped with an AxioCam camera. Briefly, RA FLS were washed with PBS and incubated with 4',6'-diamidino-2-phenylindole (DAPI; Sigma-Aldrich) then analyzed directly by fluorescent microscope. For flow cytometry analysis, RA FLS were washed in PBS, detached using trypsin-EDTA, fixed in 4% paraformaldehyde, and analyzed for GFP expression. Gene transduction efficiency was measured in the presence and absence of TNF-{alpha} (10 ng/ml) or IL-1{beta} (10 ng/ml) (Sigma-Aldrich).

Identification of Ad5 cell surface receptors and receptor usage in RA FLS

Construction and isolation of recombinant Ad5 proteins hexon, knob, WT Pb, and its two substitution mutants R340E and D288K have been described in detail elsewhere (35, 36, 37). R340E carries an Arg-to-Glu substitution at position 340 in the RGD motif, and D288K an Asp-to-Lys substitution at position 288 in the LDV motif. Receptor usage was determined by cell binding competition assays using viral proteins as competitors (WT Pb, R340E and D288K mutant, hexon or fiber knob), as described previously (22, 23). Briefly, 1 x 105 RA FLS were incubated with hexon (3, 30, or 300 ng total protein), Pb protein, WT or mutant (0.5, 5, or 50 ng), or knob protein (4, 40 or 400 ng) for 1 h at 4°C, in 200 µl/well of a 24-well plate. These amounts were estimated to be equivalent to 10-, 102-, and 103-fold excess over the hexon capsid proteins present in the infectious viral inoculum, 3-, 30-, and 300-fold excess over the Pb capsomers, and 102-, 103-, and 104-fold over the knob content, respectively. This was based on the occurrence of 240 copies of hexon, 12 copies of Pb, and 12 copies of fiber knob per virion. Cells were washed in cold PBS then incubated with Ad5GFP-R7-knob at a MOI of 200 for a further 30 min at 4°C. Cells were washed and transferred to 37°C overnight, and GFP expression was measured by flow cytometry.

Rabbit polyclonal anti-Pb Ab was used to inhibit viral Pb-integrin recognition in gene transfer assays (35, 36). Ad5GFP-R7-knob at a MOI of 200 was preincubated with anti-Pb Abs or control preimmune rabbit serum at 3 and 30 µg/ml for 1 h at 4°C, then complexes were added to FLS for 30 min at 4°C. Cells were washed in PBS and transferred to 37°C overnight, and GFP expression was measured by flow cytometry.

Integrin expression was determined by flow cytometry using specific mAbs against {alpha}v{beta}3 (CD51/CD61; BD Pharmingen), {alpha}v{beta}5 (CD49E; Chemicon International), {alpha}v (I3C2), {alpha}3 (P1B5; Chemicon International), and {beta}1 integrins (JBS5; Chemicon International). RA FLS were incubated with 40 µg/ml mAbs against integrins for 30 min at 4°C, washed with PBS, and incubated with FITC-conjugated goat anti-mouse Ab (BD Pharmingen) for 30 min at 4°C. Control samples were incubated with secondary Ab conjugate alone. For function-blocking integrin assays, RA FLS were incubated with 40 µg/ml mAbs against integrins, {alpha}v{beta}3 (LM609), {alpha}v{beta}5 (P3G2), anti-{alpha}v (I3C2), anti-{alpha}3 (P1B5), anti-{beta}1 (JBS5; Chemicon International) for 1 h at 4°C, and cells were washed in PBS then incubated with Ad5GFP-R7-knob at a MOI of 200 for 30 min at 4°C (38). Cells were washed and returned to 37°C overnight, and GFP expression was measured by flow cytometry.

Virus neutralization by SF

SF from six RA patients was centrifuged at 12,000 rpm for 15 min at 4°C. The supernatant was collected and incubated for 30 min at 55°C to inactivate complement. To determine the neutralizing activity, SF was diluted to 1/10, 1/100, and 1/1,000. Ad5GFP-FiWT, Ad5GFP-R7-knob, and Ad5GFP{Delta}knob were incubated with SF dilutions for 1 h, then added to RA FLS for 1 h at 37°C. Supernatants were removed, RA FLS were incubated with fresh medium, and GFP expression was measured by flow cytometry after 24 h. IgG was depleted from SF as described previously (39). Briefly, SF was incubated overnight at 4°C with protein G-Sepharose affinity gel (Amersham Biosciences), IgG-protein G-Sepharose complex was removed by centrifugation at 12,000 rpm for 5 min, then 10-fold dilutions of IgG-depleted SF were incubated with Ad5GFP-FiWT and added to RA FLS as described above.

CIA model

CIA was induced in male (8–10 wk old) DBA1 mice (Bomholtgard Breeding and Research Center) with bovine type II collagen (bCII) as described previously (40). All in vivo studies complied with national legislation and were approved by local authorities of the Care of Use of Animals with related codes of practice. Briefly, CIA was induced by intradermal injection at the tail base with 100 µg of bCII on day 0, boosted with an i.p. injection of 100 µg of bCII on day 21. This resulted in the onset of CIA between D30–32 as described previously. Due to the expected broad viral tropism of the Ad5GFP-R7-knob or Ad5GFP{Delta}knob, targeted infection of synovial cells was achieved by intra-articular (i.a) injection. To determine transduction efficiency in uninflamed knee joints, mice with no clinical arthritis received i.a injections in both knee joints on day 32 with 106 PFU/ml Ad5GFP-R7-knob, Ad5GFP{Delta}knob, or Ad5GFP-FiWT. To determine transduction efficiency of vectors in inflamed knee joints, mice with at least one inflamed paw received i.a injections in both knee joints on day 32 with 1 x 106 PFU/ml Ad5GFP-FiWT, Ad5GFP-R7-knob, or Ad5GFP{Delta}knob. Forty-eight hours later, whole knee joints of mice were removed. Arthritis development was monitored in the knees and paws macroscopically until day 32. Mice were considered to have arthritis when considerable changes in redness and/or swelling were noted in digits or other parts of the paws. Clinical severity was graded on a scale of 0 to 2 for each paw by two independent observers, according to changes in redness and swelling: 0, no changes; 0.25, 1–2 toes affected; 0.5, 3–5 toes affected; 0.5, ankle swelling; 0.5, footpad swelling; 0.5, severe overall swelling and ankylosis. Histological severity was graded on a scale of 0–3, depending on the amount of inflammatory cells in the synovial cavity (exudate) and synovial tissue (infiltrate).

GFP quantification

Human synovial tissue or mouse joints were fixed in 10% formalin, and mouse joints were decalcified. Tissue was embedded in paraffin and GFP, CD3 and CD14 expression was measured by immunohistochemistry, and sections were counterstained with hematoxylin (41, 42). Briefly, for single staining endogenous peroxidase was blocked with 3% hydrogen peroxide. Sections were incubated with a mouse mAb against GFP (IgG1; Chemicon International) for 30 min, followed by biotinylated anti-mouse IgG, streptavidin-peroxidase, and developed using 3-amino 9-ethylcarbazole (Microm Microtech) or diaminobenzidine (DakoCytomation), and counterstained with hematoxylin (DakoCytomation). For double staining, sections were incubated with mouse mAbs against either CD3 or CD14 (IgG2a; Microm Microtech) overnight at 4°C, then biotinylated anti-mouse IgG, followed by avidin-alkaline phosphatase (Microm Microtech), and developed using Fast Blue (Vector Laboratories). In negative control sections, an IgG isotype control Ab was used. Cells positive for GFP or double positive for GFP and CD3 or CD14 were quantified by counting positive cells in 10 consecutive high-power fields (x200). The number of positive cells per high-power field was averaged, and the results were expressed as the number of GFP-positive cells per mm2 using Histolab microvision version 5.9.2.

Statistical analysis

Results are expressed as the mean ± SEM. Differences between groups were calculated with the nonparametric Mann-Whitney U test, and values of p < 0.05 were considered statistically significant.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Gene transduction efficiency of short fiber-modified Ad5 vectors in RA FLS

We investigated the efficiency of gene transfer to RA FLS using short fiber-modified vectors Ad5GFP-R7-knob and Ad5GFP-R7-RGD, compared with Ad5GFP-FiWT, which carried its natural fibers. In comparison to Ad5GFP-FiWT, gene transduction of RA FLS was 40- to 50-fold more efficient with Ad5GFP-R7-knob than with Ad5GFP-FiWT, and 25-fold with Ad5GFP-R7-RGD at a MOI of 200 (Fig. 2, A and B).



View larger version (41K):
[in this window]
[in a new window]
  		FIGURE 2. Gene transduction efficiency of short fiber-modified Ad5 vectors compared with WT Ad5 in RA FLS. A, A representative fluorescent microscopic picture of GFP expression (green), double labeled with DAPI (blue) is shown in Ad5GFP-FiWT, Ad5GFP-R7-RGD, and Ad5GFP-R7-knob-infected RA FLS at a MOI of 200 PFU/cell at 48 h postinfection (p.i.). B, RA FLS were infected with Ad5GFP-FiWT, Ad5GFP-R7-RGD, and Ad5GFP-R7-knob at a MOI of 200 PFU/cell, and GFP-positive cells were counted at 48 h p.i. The results are expressed as the percentage of GFP-positive cells and represent the mean ± SEM of three separate experiments. *, p < 0.05 percent of GFP-positive cells transduced by Ad5GFP-R7-RGD or Ad5GFP-R7-knob compared with Ad5GFP-FiWT. C, Dose-response curves of Ad5GFP-FiWT and Ad5GFP-R7-knob vectors at MOI of 2 to 2000 PFU/cell for 48 h were examined in RA FLS and A549 control cells. The results are expressed as the percentage of GFP-positive cells and represented as the mean ± SEM of three separate experiments. *, p < 0.05 percent of GFP-positive cells transduced by Ad5GFP-R7-knob compared with Ad5GFP-FiWT. D, The time course of GFP transduction by Ad5GFP-FiWT and Ad5GFP-R7-knob at a MOI of 200 was examined at 24, 48, and 72 h. The results are expressed as the percentage of GFP-positive cells and represented as the mean ± SEM of three separate experiments. *, p < 0.05 percent of GFP-positive cells transduced by Ad5GFP-R7-knob compared with Ad5GFP-FiWT.

 
We next examined the dose-response curves of gene transduction of RA FLS by Ad5GFP-R7-knob and Ad5GFP-FiWT at various MOI (2, 20, 200, and 2000), and in comparison with A549 cells (Fig. 2C). A549 cells are fully permissive to Ad5 infection and were used as a positive control for gene transduction (22). At least two different batches of each vector were used. Ad5GFP-R7-knob transduced RA FLS with a 40- to 50-fold higher efficiency than Ad5GFP-FiWT, and in a dose-dependent manner at MOI between 20 and 200. These differences could not be attributed to differences in batch quality because A549 cells were permissive to both vectors (Fig. 2C, right panel). Consistent with previous studies (9, 14), FLS were found to be poorly permissive to Ad5GFP-FiWT, with only 10% of FLS expressing GFP at a MOI of 200, and 58% at a MOI of 2000 at 48 h. In A549 cells, however, 78% of cells transduced by Ad5GFP-FiWT were GFP-positive at a MOI of 200, and 100% at a MOI of 2000.

The time course of Ad5GFP-R7-knob-mediated GFP gene transduction of RA FLS was examined over 24, 48, and 72 h at a MOI of 200. At all time points, Ad5GFP-R7-knob transduced RA FLS with a 40- to 50-fold higher efficiency than Ad5GFP-FiWT (Fig. 2D).

Effect of proinflammatory cytokines on gene transduction efficiency of RA FLS by Ad5GFP-R7-knob

In vivo, FLS are exposed to proinflammatory cytokines such as IL-1 and TNF, which are key factors in the pathogenesis of RA. We next determined whether Ad-mediated gene transduction was influenced, positively or negatively, in the presence of these cytokines. FLS were pretreated with IL-1{beta} (10 ng/ml) or TNF-{alpha} (10 ng/ml) for 48 h, then infected with Ad5GFP-R7-knob or Ad5GFP-FiWT at various MOI for 24 h. Ad5GFP-R7-knob-infected cells showed a 2-fold increase in gene transduction efficiency at a MOI of 20 for IL-1- or TNF-treated cells (Fig. 3). Furthermore, both fluorescence microscopy and flow cytometry showed a proportion of Ad5GFP-R7-knob-infected FLS, displaying a higher mean fluorescence intensity when pretreated with IL-1 or TNF (data not shown). In contrast, pretreatment with IL-1 or TNF had either no effect, or occasionally a negative effect, in Ad5GFP-FiWT-mediated transduction of FLS.



View larger version (15K):
[in this window]
[in a new window]
  		FIGURE 3. Effect of proinflammatory cytokines on Ad5GFP-FiWT and Ad5GFP-R7-knob GFP transduction in RA FLS. RA FLS were preincubated with IL-1 (10 ng/ml) or TNF (10 ng/ml) for 48 h, then infected with Ad5GFP-FiWT or Ad5GFP-R7-knob at varying MOI (2 to 2000 PFU/cell) for 24 h. Results are expressed as the percentage of GFP-positive cells and are the mean ± SEM of three separate experiments. *, p < 0.05 percent of GFP-positive cells transduced by Ad5GFP-R7-knob at a MOI of 20 in the presence of either IL-1 or TNF compared with Ad5GFP-R7-knob at a MOI of 20 alone.

 
Role of fiber-knob domain in Ad5GFP-R7-knob attachment and entry of FLS

Because Ad5GFP-R7-knob most efficiently transduced RA FLS compared with RGD-liganded or conventional WT vectors, we next sought to identify the role of different capsid proteins in attachment of the virus to the cell surface and viral entry. The knob domain of Ad5 is known to bind other attachment receptors apart from CAR, which includes heparin sulfate glycosaminoglycans and the MHC class I {alpha}2 domain (23, 43, 44). We investigated the role of the knob domain in Ad5GFP-R7-knob entry of FLS by cell binding competition assays. RA FLS or control A549 cells were infected with Ad5GFP-R7-knob at a MOI of 200 in the presence or absence of increasing quantities of recombinant fiber-knob protein. Because the numbers of cell surface receptors for Ad5 is not known, the amounts of knob competitor represented the equivalent of 102-, 103-, and 104-fold excess over the total number of knob protein present in the infectious virus inoculum. As shown in Fig. 4A (left panel), Ad5GFP-R7-knob-mediated gene transduction of RA FLS was not affected by competition with fiber-knob protein. In contrast, in control CAR-positive and integrin-rich A549 cells, Ad5GFP-R7-knob-mediated gene transduction was significantly reduced by competition with fiber-knob protein in a dose-dependent way, 2- to 5-fold with a 103- to 104-fold excess (Fig. 4A, right panel).



View larger version (41K):
[in this window]
[in a new window]
  		FIGURE 4. Role of knob, Pb, or hexon proteins in Ad5GFP-R7-knob attachment and entry to FLS. A, Competition with knob protein. RA FLS and control A549 cells were incubated in the presence or absence or recombinant fiber-knob protein at various concentrations corresponding to 102-, 103-, and 104-fold excess over the amount of capsid protein knob present in the infectious virus inoculum. Cells were then infected with Ad5GFP-R7-knob at a MOI of 200 for 24 h. Results are expressed as the percentage of GFP-positive cells in the absence of competitor and are the mean ± SEM of three separate experiments. *, p < 0.05 percent of GFP-positive cells transduced by Ad5GFP-R7-knob in the presence of fiber-knob protein compared with the absence of fiber-knob protein. B, Effect of fiber deknobbing on gene transduction. RA FLS cells were infected with Ad5GFP-R7-knob or Ad5GFP{Delta}knob at a MOI of 2 and 20 for 24 h. Results are expressed as the percentage of GFP-positive cells and are the mean ± SEM of three separate experiments. *, p < 0.05 percent of GFP-positive cells transduced by Ad5GFP{Delta}knob compared with Ad5GFP-R7-knob. C, Competition with Pb and hexon proteins. RA FLS were incubated in the presence or absence of increasing amounts of Pb or hexon protein as indicated, then infected with Ad5GFP-R7-knob at a MOI of 200 for 24 h. Results are expressed as the percentage of GFP-positive cells in the absence of competitor. Results are the mean ± SEM of three separate experiments. *, p < 0.05 percent of GFP-positive cells transduced by Ad5GFP-R7-knob in the presence of Pb compared with the absence. D, Competition with anti-Pb Abs. Ad5GFP-R7-knob at a MOI of 200 was preincubated with preimmune rabbit serum or rabbit polyclonal anti-Pb, then added to RA FLS, and infection was allowed to proceed at 37°C for 24 h. Results are expressed as the percentage GFP-positive cells in the absence of anti-Pb Ab or preimmune rabbit serum. Results are expressed as the mean ± SEM of three separate experiments. *, p < 0.05 percent of GFP-positive cells transduced by Ad5GFP-R7-knob in the presence of anti-Pb Ab compared with preimmune serum. E, Competition with WT and mutant Pb. RA FLS were incubated in the presence or absence of increasing amounts of WT Pb, R340E, or D288K mutant as indicated, then infected with Ad5GFP-R7-knob at a MOI of 200. Results are expressed as the percentage of GFP-positive cells in the absence of competitor and represented the mean ± SEM of three separate experiments. *, p < 0.05 percent of GFP-positive cells transduced by Ad5GFP-R7-knob in the presence of WT or mutant Pb compared with the absence.

 
To confirm the lack of involvement of the fiber-knob domain in Ad5GFP-R7-knob transduction of RA FLS, we compared gene transduction between Ad5GFP-R7-knob and its knobless version Ad5GFP{Delta}knob at a MOI of 2 and 20 (33), and anticipated that gene transduction would not differ between both vectors. Unexpectedly, we observed an increase in transduction of RA FLS by Ad5GFP{Delta}knob, compared with Ad5GFP-R7-knob, with a 15-fold higher efficiency at a MOI of 2, and a 5-fold higher efficiency at a MOI of 20 (Fig. 4B). This suggested that not only was the fiber-knob domain not required by Ad5GFP-R7-knob in FLS attachment and infection, but that it may impede viral infection.

Role of viral capsid proteins hexon and Pb in Ad5GFP-R7-knob attachment and entry to FLS

We next examined whether other capsomers such as hexon or Pb were involved in Ad5GFP-R7-knob transduction of RA FLS by cell binding competition assays using increasing amounts of viral proteins as competitors. No competition effect was observed with hexon protein, even at amounts equivalent to a 100-fold excess over the theoretical numbers of hexon capsomers present in the virus inoculum (Fig. 4C). However, Ad5GFP-R7-knob gene transduction was significantly diminished in the presence of recombinant WT Pb protein, in a dose-dependent way. WT Pb reduced Ad5GFP-R7-knob gene transduction by 41% at a 30-fold excess, and by 71% at a 300-fold excess over the Pb capsomers (Fig. 4C). This suggested that Pb protein, but not hexon, was involved in cell attachment and viral entry.

To confirm involvement of the Pb capsomers in Ad5GFP-R7-knob infection of FLS, we examined the effect of anti-Pb Ab on gene transfer mediated by Ad5GFP-R7-knob. RA FLS were infected with Ad5GFP-R7-knob at a MOI of 200 in the presence of anti-Pb antiserum or corresponding preimmune serum. Ad5GFP-R7-knob gene transduction was significantly reduced in a dose-dependent manner in the presence of anti-Pb Ab but not by preimmune serum. A 66% reduction was observed with 3 µg/ml, and 86% with 30 µg/ml anti-Pb Ab (Fig. 4D).

Ad5 endocytosis is known to be mediated by the binding of cellular integrins to specific recognition motifs on viral proteins, primarily RGD in Pb capsomers (5). The LDV sequence binds a different spectrum of integrins (36, 45). To identify which integrin binding sequences in the Pb capsomer were involved in Ad5GFP-R7-knob entry of RA FLS, we used Pb mutants in the RGD340 (R340E) or LDV288 (D288K) motif in cell binding competition experiments with Ad5GFP-R7-knob. When compared with WT Pb, Ad5GFP-R7-knob gene transduction was also reduced by R340E or D288K mutant, although only at high doses (i.e., at 300-fold excess) (Fig. 4E). This suggested that Ad5GFP-R7-knob used both RGD and LDV motifs, as well as additional peptide motifs of lesser-defined sequences (38), and hence a broader spectrum of integrins in attachment and entry of RA FLS.

Surface expression and role of integrin receptors in Ad5GFP-R7-knob infection of RA FLS

Because both RGD and non-RGD sequences were observed to be involved in Ad5GFP-R7-knob entry into RA FLS, we examined the expression of integrins on RA FLS, which may be potential receptors for Ad5 by flow cytometry. Integrins {alpha}v{beta}3 and {alpha}v{beta}5 bind to RGD sequences, whereas {alpha}3{beta}1 mediates binding to both RGD and non-RGD sequences, and {alpha}4{beta}1 binds to LDV motifs (5, 38). Consistent with the literature (10, 11, 46), {alpha}v{beta}3, {alpha}v{beta}5, and {alpha}v integrins were expressed in RA FLS as were {alpha}3 and {beta}1 (Fig. 5A).



View larger version (38K):
[in this window]
[in a new window]
  		FIGURE 5. Role of integrin receptors in Ad5GFP-R7-knob infectivity of RA FLS. A, Cell surface integrin expression. RA FLS were analyzed for expression of {alpha}v{beta}3, {alpha}v{beta}5, {alpha}v, {alpha}3, and {beta}1 by flow cytometry. Cells stained with FITC-conjugated goat anti-mouse Ab alone was used as a control for background fluorescence. B, Effect of inhibition of integrin function on gene transduction. RA FLS were incubated with or without function-blocking mAbs against integrins {alpha}v{beta}3, {alpha}v{beta}5, {alpha}v, {alpha}3, and {beta}1 at a concentration of 40 µg/ml, then infected with Ad5GFP-R7-knob, and GFP expression was measured by flow cytometry. The results are expressed as the percentage of GFP-positive cells in the absence of anti-integrin Abs and expressed as mean ± SEM of three separate experiments. *, p < 0.05 percent of GFP-positive cells transduced by Ad5GFP-R7-knob in the presence of anti-integrin mAbs compared with the absence.

 
We next examined the requirement for these integrin species in Ad5GFP-R7-knob infection of RA FLS by cell binding competition assays with mAbs against integrins {alpha}v{beta}3, {alpha}v{beta}5, {alpha}v, {alpha}3, or {beta}1. RA FLS were infected with Ad5GFP-R7-knob at a MOI of 200 in the presence or absence of anti-integrin mAbs, and GFP-positive cells were analyzed after 24 h. Integrins {alpha}v{beta}3, {alpha}v{beta}5, {alpha}v, and {beta}1, but not {alpha}3, participated in Ad5GFP-R7-knob-mediated gene transduction of FLS (Fig. 5B).

Effect of SF on Ad5GFP-R7-knob-mediated gene transduction

To investigate the clinical relevance of Ad5GFP-R7-knob as a vector in human RA, the effect of SF on gene transduction was examined. SF from 70% of RA patients are known to contain anti-Ad5-neutralizing Abs, which limit the efficacy of unmodified conventional Ad5 vectors in treatment of human disease (14, 47). To examine whether Ad5GFP-R7-knob remained efficacious in the presence of SF, even at a lower MOI compared with Ad5GFP-FiWT, Ad5GFP-R7-knob at a MOI of 200 was compared with Ad5GFP-FiWT at a MOI of 2000. Vectors were preincubated with different dilutions of SF from six different RA patients for 1 h, then the mixture was added to RA FLS, and GFP expression was measured after 24 h. Ad5GFP-FiWT was completely neutralized by SF at 1/10, 1/100, and 1/1,000 dilutions, even though it was used at a higher MOI of 2000 (Fig. 6A, left panel). In comparison, in the presence of SF at 1/10, 1/100, and 1/1000 dilutions, Ad5GFP-R7-knob at a MOI of 200 transduced 26–50 (mean, 36%), 33–60 (mean, 42%), and 30–53% (mean, 37%) of FLS, respectively (Fig. 6A, right panel). This corresponded to 58, 39, and 35% inhibition of Ad5GFP-R7-knob-mediated gene transfer in the presence of SF at dilutions of 1/10, 1/100, and 1/1000, respectively. Thus, in the absence of SF, infection by Ad5GFP-R7-knob was 40- to 50-fold higher in comparison to Ad5GFP-FiWT (refer to Fig. 2), and remained 35- to 40-fold higher in the presence of SF despite partial viral neutralization.



View larger version (38K):
[in this window]
[in a new window]
  		FIGURE 6. Effect of SF on Ad5 infectivity of RA FLS. A, Ad5GFP-FiWT at a MOI of 2000 and Ad5GFP-R7-knob at a MOI of 200 were incubated with SF from six different RA patients at 1/10, 1/100, and 1/1000 dilution for 1 h, then added to RA FLS for a further 24 h. Results are expressed as the percentage of GFP-positive cells and represent the mean of three separate experiments. B, Ad5GFP-FiWT at a MOI of 2000 was incubated with IgG-depleted SF (SF+G) or undepleted SF (SF) at different dilutions then added to RA FLS as described above. The result is representative of three separate experiments. C, Ad5GFP{Delta}knob at a MOI of 20 was incubated with SF from two representative samples of SF (patients 4 and 6) as described above, and results were expressed as the percentage of GFP-positive cells. D, Ad5GFP{Delta}knob at a MOI of 20–200 was incubated without SF (–SF) or with SF (+SF) from two representative samples of SF (patients 4 and 6) at a dilution of 1/10, then added to RA FLS. Results are expressed as the percentage of GFP-positive cells and are the mean of two experiments performed in duplicate.

 
As a control, we examined the neutralizing effect of IgG-depleted SF vs undepleted SF on Ad5GFP-FiWT transduction of RA FLS (Fig. 6B). The IgG-depleted SF inhibited Ad5GFP-FiWT-mediated transgene expression significantly less compared with undepleted SF. We were able to confirm that the inhibitory effect of SF is almost entirely due to the presence of anti-Ad5 IgG Abs, as has been published previously (47). We similarly noted a slight inhibition of transgene expression, an effect that could not be attributed to cells or complement, because these components were removed or inactivated before incubation with vectors, but might be due to incomplete depletion of IgG or possibly other inhibitory factors in the SF.

Anti-Ad5-neutralizing Abs from sera of patients with lung and liver cancer, and ascitic fluid from ovarian cancer patients have been demonstrated to be mainly directed against fiber and Pb capsomers (39, 48, 49). We investigated the hypothesis that anti-Ad5 Ab from SF of RA patients were primarily directed against both fiber and Pb, and not the knob protein, by examining the effect of SF on transduction of RA FLS by the knob-deleted vector Ad5GFP{Delta}knob. Ad5GFP{Delta}knob at an efficacious MOI of 20 was incubated with SF from two representative RA patients (patients 4 and 6) as described above, and GFP expression was measured at 24 h (Fig. 6C). In the presence of SF at 1/10, 1/100, and 1/1000 dilutions, Ad5GFP{Delta}knob transduced a mean of 25, 37, and 40% of FLS. This corresponded to 76, 60, and 55% inhibition of Ad5GFP{Delta}knob-mediated gene transfer, respectively. This suggested that that the contribution of the fiber-knob domain in virus neutralization by SF was negligible, and that anti-Ad5 Abs in SF were mainly directed against capsid proteins and domains other than the fiber-knob, such as the Pb protein or the fiber shaft domain.

We next investigated whether viral neutralization could be overcome by using increasing amounts of the most efficacious fiber-modified Ad5, Ad5GFP{Delta}knob. Ad5GFP{Delta}knob at a MOI of 20–200 was preincubated with SF from the same two representative RA patients (patients 4 and 6) at the lowest dilution (1/10), and RA FLS were infected as described above (Fig. 6D). In the presence of SF at 1/10, Ad5GFP{Delta}knob transduction of RA FLS was inhibited by 54% at a MOI of 20, and 21% at a MOI of 50. At a higher MOI of 100 and 200, however, Ad5GFP{Delta}knob-mediated gene transfer was not inhibited in the presence of SF.

Gene transduction efficiency of short fiber-modified Ad5 compared with Ad5GFP-FiWT in RA synovium explants

To extend these studies to the actual joint situation, we examined gene transduction efficiency of Ad5GFP-R7-knob or Ad5GFP{Delta}knob compared with Ad5GFP-FiWT in an ex vivo synovium explant model. Freshly harvested 3-mm3 synovium pieces were placed in 6-well tissue culture plates in quadruplicate for 4 days, then infected with Ad5GFP-R7-knob, Ad5GFP{Delta}knob, or Ad5GFP-FiWT at a MOI of 200 for 24, 48, and 72 h. Explant tissue was infected at 4 days for investigation of infection efficiency of the vectors in both synovial explants and in explant outgrowth because it takes 3–4 days for cells to migrate out of the tissue. There was no difference in the infection efficiency or rapidity of infection in synovial tissue, after immediate infection compared with 4 days later (n = 4 from four different RA donors; data not shown). GFP expression was determined by fluorescence microscopy. Control tissue showed minimal autofluorescence. Cellular outgrowth from synovium explants are FLS and have the characteristic spindle-like morphology (Fig. 7B) and flow cytometry findings consistent with standard measures of isolation and identification of FLS following synovial tissue digestion established in our laboratories and elsewhere, as described in Materials and Methods. FLS from synovial outgrowth were transduced by Ad5GFP-R7-knob but not by Ad5GFP-FiWT (Fig. 7, A and B). FLS from synovial explants were grown to confluence and infected by short fiber-modified vectors compared with Ad5GFP-FiWT. Gene transfer by Ad5GFP-R7-knob was also 40- to 50-fold higher than Ad5GFP-FiWT in RA FLS from synovial explant outgrowth and was not different compared with gene transfer in FLS derived by enzyme digestion (data not shown). Explant tissue was more rapidly transduced by Ad5GFP-R7-knob or Ad5GFP{Delta}knob compared with Ad5GFP-FiWT (Fig. 7C). At 24 h, GFP-positive cells were observed in synovial lining and sublining tissue infected by Ad5GFP-R7-knob or Ad5GFP{Delta}knob but not by Ad5GFP-FiWT. At 48 h, GFP-positive cells increased in Ad5GFP-R7-knob or Ad5GFP{Delta}knob infected synovial lining and sublining tissue, and was observed in Ad5GFP-FiWT sublining tissue. At 72 h, GFP-positive cells were observed in Ad5GFP-R7-knob, Ad5GFP{Delta}knob, and Ad5GFP-FiWT synovial lining cells. The delay observed in gene transfer by Ad5GFP-FiWT, compared with fiber-modified Ad5GFP-R7-knob, was consistent with our in vitro experiments in RA FLS. To confirm immunofluorescence data, we quantified GFP expression in synovial explants after 72 h by immunohistochemistry (Fig. 7E). Gene transfer by Ad5GFP-R7-knob or Ad5GFP{Delta}knob was ~3-fold higher than Ad5GFP-FiWT in synovial tissue explants. Unexpectedly, there was no advantage of the knobless version of Ad5GFP-R7-knob in synovial explants compared with FLS.



View larger version (41K):
[in this window]
[in a new window]
  		FIGURE 7. Gene transduction efficiency of short fiber-modified vectors compared with Ad5GFP-FiWT in RA synovium explants. Freshly harvested 3-mm synovial tissue pieces were placed in culture for 4 days then infected with Ad5GFP-FiWT, Ad5GFP-R7-knob, or Ad5GFP{Delta}knob at a MOI of 200 for 24, 48, and 72 h. GFP-positive cells in cellular outgrowth from synovium explant tissue (A, x20 and B, x100) and in synovium explant tissue (C and D) were analyzed by fluorescence microscopy (left panels) and phase contrast microscopy (right panels) at low (C, x20) and high magnification (D, x100). Results are representative of at least three separate experiments in three different RA donors. E, Seventy-two hours p.i., synovial explants were fixed, paraffin embedded, and sections were examined for GFP expression by immunostaining using 3-amino 9-ethylcarbazole (red) and counterstained with hematoxylin; magnification, x400. GFP expression was quantified in 10 high-power fields, averaged, and expressed as the number of GFP-positive cells/mm2. *, p < 0.05.

 
Rheumatoid synovium is characterized by chronic inflammatory cell infiltrates consisting mainly of macrophages, T lymphocytes, and resident FLS. Similar to FLS, human macrophages and lymphocytes are known to be mainly CAR negative (6). Reports in murine macrophage and lymphocyte cell lines show that these cell types are CAR negative, but {alpha}v integrin positive (3). To identify infected cell types in synovial tissue explants, double staining was performed for GFP+CD14+ or GFP+CD3+ cells by immunohistochemistry. In Ad5GFP-R7-knob-infected synovial tissue, GFP+CD14+ and GFP+CD3+ cells and GFP+ cells in synovial lining were identified (Fig. 8A). Gene transfer was significantly higher in synovial macrophages and lymphocytes by Ad5GFP-R7-knob compared with Ad5GFP-FiWT (Fig. 8B; p < 0.05). Ad5GFP-R7-knob or Ad5GFP{Delta}knob transduced macrophages more efficiently than lymphocytes. RA synovial tissue was also enzyme digested, and whole cells were infected with Ad5GFP-R7-knob compared with Ad5GFP-FiWT at a MOI of 200 and incubated with PE-conjugated CD14 or CD3. More than 90% of FLS were transfected by Ad5GFP-R7-knob compared with 80% of GFP+CD14+ cells and 23% of GFP+CD3+ cells, and 1–10% of subtypes by Ad5GFP-FiWT (data not shown).



View larger version (60K):
[in this window]
[in a new window]
  		FIGURE 8. GFP expression in Ad5-transduced synovial macrophages and lymphocytes. Synovial explants infected with Ad5GFP-FiWT, Ad5GFP-R7-knob, or Ad5GFP{Delta}knob at a MOI of 200 for 72 h were fixed, paraffin embedded, and sections were examined by immunostaining. A, Sections were double stained for GFP expression (red) and CD14 (blue) or CD3 (blue); magnification, x200. B, GFP+CD14+ cells or GFP+CD3+ cells were quantified in 10 high-power fields, averaged, and expressed as the number of GFP-positive cells/mm2. *, p < 0.05.

 
Gene transduction efficiency of short fiber-modified Ad5 compared with Ad5GFP-FiWT in murine CIA

Due to the expected broad viral tropism of the Ad5GFP-R7-knob or Ad5GFP{Delta}knob vectors, targeted infection of synovial cells was achieved by i.a injection. To demonstrate efficacy of our short fiber-modified vectors even at a lower viral titer we used 1 x 106 PFU/ml, which is reduced compared with protocols used by other authors (1 x 107 PFU/ml), for conventional Ad5 vectors (13).

To determine whether transduction efficiency of short fiber-modified vectors was enhanced in the presence of inflammation in vivo, we examined transduction efficiency in inflamed knee joints in murine CIA. Macroscopic scores in the knees were no different between vector groups (1.5). Histological scores in frontal sections of the knees were likewise no different between vector groups (1.25–1.5). GFP expression was measured by immunohistochemistry in murine-inflamed knee joints (Fig. 9, A–C; n = 8 mice, two mice per vector group, both knee joints injected per mouse). We observed increased GFP expression in short fiber-modified vector-infected synovial tissue. This was observed both throughout the synovium and at the pannus-cartilage junction. Gene transfer was significantly more efficient by Ad5GFP-R7-knob or Ad5GFP{Delta}knob compared with Ad5GFP-FiWT in inflamed murine knee joints (Fig. 9D; p < 0.05).



View larger version (82K):
[in this window]
[in a new window]
  		FIGURE 9. Gene transduction efficiency‘ of short fiber-modified vectors compared with Ad5GFP-FiWT in murine CIA. CIA was induced, and on day 32 control saline or 106 PFU/ml Ad5GFP-FiWT, Ad5GFP-R7-knob, or Ad5GFP{Delta}knob were injected into both knee joints (n = 8 mice, 2 mice per vector group). Forty-eight hours after injection of the viruses, knee joints were fixed, paraffin embedded, and decalcified. Frontal sections of the knee joints were examined for GFP expression (arrowheads) and counterstained with hematoxylin in inflamed knee joints. Control uninfected knee joint (top left panel), conventional Ad5GFP-FiWT-transduced knee joint (top right panel), Ad5GFP-R7-knob-transduced knee joint (bottom left panel), Ad5GFP{Delta}knob-transduced knee joint (bottom right panel). b = bone, c = cartilage, s = synovitis at a magnification of x40 (A) and x200 (B). C, A representative image of GFP-positive staining in Ad5GFP{Delta}knob-infected murine synovium; magnification, x400. D, GFP expression was analyzed in murine inflamed knee joints (left panel) and uninflamed knee joints (right panel). GFP expression was quantified in 10 high-power fields, averaged, and expressed as the number of GFP-positive cells/mm2. *, p < 0.05.

 
We next examined in vivo transduction efficiency of the vectors in uninflamed knee joints in murine CIA. GFP expression was measured in mice with no evidence of arthritis at the time of i.a vector injection (n = 8 mice, two mice per vector group, one knee injected per mouse, macroscopic score 0). GFP expression was significantly higher in Ad5GFP-R7-knob or Ad5GFP{Delta}knob-infected knee joints compared with Ad5GFP-FiWT-infected uninflamed knee joints (Fig. 9E; p < 0.05). Gene transfer by Ad5GFP-R7-knob, Ad5GFP{Delta}knob was significantly higher in the presence of inflammation compared with uninflamed knee joints (2.8- and 2.5-fold higher, respectively). This is consistent with in vitro data in which proinflammatory cytokines enhanced transduction efficiency of short fiber-modified vectors. We also observed a trend to increased transduction efficiency of Ad5GFP-FiWT-infected knee joints (1.5-fold increase); however, this was not statistically significant. Similar to data in human synovium, in murine synovium there was no advantage of the knobless vector compared with Ad5GFP-R7-knob. Increased efficacy of the short fiber-modified vectors compared with Ad5GFP-FiWT is consistent with human in vitro and ex vivo data.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
Ad5 vectors have been extensively used in animal models of arthritis due to their ability to infect both proliferating and quiescent cells, ease of propagation, nonintegrating nature, and relative safety. However, both murine and human RA FLS have been reported to be poorly permissive to Ad5 vectors due to the absence of the high-affinity receptor CAR (9, 14). Although, low levels of transgene expression obtained with Ad5 can be compensated by the use of high MOI, this may induce toxicity and arthritis flares in murine models (2, 50). Furthermore, the high prevalence of anti-Ad5-neutralizing Abs in SF may limit the use of such vectors in human disease. Hence, strategies to improve Ad5 viral tropism would be desirable. In this study, we observed that genetically modified Ad5 vectors with short-shafted fibers (7 shaft repeats instead of 22 in the WT fiber) were highly efficient in transduction of RA FLS and human and murine synovium.

In RA FLS, this was enhanced by deletion of the knob domain and was more efficient than insertion of RGD motifs in the shaft. We demonstrated for the first time that in the context of the short fiber modification, Ad5 used a spectrum of both RGD and non-RGD-binding integrins to enhance viral tropism in RA FLS. In RA FLS, synovial macrophages, and lymphocytes, we observed that Ad5 vectors with short fibers were significantly more efficient in gene transduction than the conventional vector with WT fibers. This was more striking for RA FLS in which Ad5GFP-R7-knob was 40- to 50-fold, and Ad5GFP-R7-RGD 25-fold more efficient than Ad5GFP-FiWT. Furthermore, Ad5GFP-R7-knob more rapidly transduced cells within 24 h, compared with Ad5GFP-FiWT in both RA FLS and synovium. Finally, these vectors show promise in effective gene delivery even in the presence of neutralizing anti-Ad5 Abs in SF.

Shortened fiber length may be associated with enhanced viral tropism in CAR-deficient cells (19, 51, 52). Magnusson et al. (19) reported an increased viral tropism of Ad5GFP-R7-RGD in CAR-negative RD cells compared with Ad5GFP-FiWT. Subgroup B serotype Ad35 more efficiently transduced RA FLS compared with Ad5GFP-FiWT in RA FLS (14). Similarly, chimeric Ad5 vectors carrying subgroup B short fibers such as serotype 16 or 35 showed enhanced gene delivery to RA FLS compared with an Ad5 vector with WT fibers (9). Ad vectors with short fibers of <8 shaft repeats such as Ad9 also showed increased infection efficiency in CAR-deficient cells (51, 52). The authors (53) suggested that vectors with less than a critical number of shaft repeats may permit close proximity and accessibility of the viral Pb to integrin receptors on the cell surface. Integrin molecules protrude above the cell surface and are approximately the same height (53) as a fiber with a shaft comprised of seven repeats of ~13.5 Å each. Integrin receptors could thus directly serve as both attachment and endocytic receptors by direct binding of the Pb. In CAR-deficient RA FLS, we observed that shortened fiber length of seven shaft repeats was a key determinant of Ad5 tropism and required integrin binding motifs in the Pb.

In assays to identify receptor usage, we observed that Ad5GFP-R7-knob attached to RA FLS in a knob-independent fashion. This was consistent with previous studies of Ad5GFP-R7-knob showing CAR-independent attachment and infection in other CAR-deficient cell types (19, 22). Furthermore, a deknobbed version of Ad5GFP-R7-knob, Ad5GFP{Delta}knob lead to an unexpected increase in gene transduction in RA FLS compared with Ad5GFP-R7-knob. It has been suggested that steric hindrance provoked by the knob domain or repulsive acidic charges at the cellular surface may interfere with the infectivity of Ad5 vectors with short fiber shafts (27). This could be the case for Ad5GFP-R7-knob, and may explain the higher level of gene transduction observed with its knobless version, Ad5GFP{Delta}knob.

Receptor specificity of the Pb is crucial in viral tropism (38, 54). The incorporation of the RGD motif into the HI loop or at the C-terminal end of the Ad5 fiber or hexon protein has previously been shown to enhance viral tropism in CAR-deficient cells, by retargeting the virus to cellular integrins (19, 20, 21, 55). We observed that enhanced transduction of RA FLS by Ad5GFP-R7-knob was knob-independent and did not exclusively occur via the RGD-dependent integrin pathway. The LDV288 motif and probably other less-defined sequences in the Pb may mediate binding to both {alpha}v-integrins and other species. This was suggested by cell-binding competition assays, in which the RGD340 mutant of Pb had a significant competing effect, lower than WT Pb and trended to being less pronounced than LDV288, although this was not statistically significant. This was also consistent with the lesser gene-transducing activity of Ad5GFP-R7-RGD, the RGD-liganded short-shafted vector compared with Ad5GFP-R7-knob. In addition, {beta}1 integrin, which is the most highly expressed integrin on FLS, may bind both RGD and non-RGD motifs.

In the presence of IL-1 and TNF, Ad5GFP-R7-knob transduction of GFP was enhanced but was unchanged or at times reduced with Ad5GFP-FiWT. In addition, in cytokine-stimulated Ad5GFP-R7-knob-transduced cells, the mean fluorescence intensity of GFP was increased compared with unstimulated cells. This may be due to increased efficiency of postentry steps (9, 22). Similarly, in murine CIA we observed increased transduction efficiency of the short fiber-modified vectors in inflamed compared with uninflamed knee joints. This suggests that Ad5GFP-R7-knob and Ad5GFP{Delta}knob-mediated gene delivery may be enhanced in inflammatory arthritis. Vectors up-regulated in the presence of inflammation may be advantageous due to the relapsing and remitting nature of RA. This concept has been extended to disease-regulated therapeutic transgene expression (56, 57, 58). Cytokines have been reported to down-regulate {alpha}v integrins in RA FLS and synovium (12, 46). However, integrin expression levels do not correlate with Ad5-mediated transgene expression (13, 59). Increased efficacy of Ad5GFP-R7-knob, despite down-regulation of cellular integrins, could be attributed to an increased spectrum of integrin receptor choice and affinity for ligands other than RGD in our short-shafted vectors, rather than the absolute level of integrin expression.

Ad5GFP-R7-knob or Ad5GFP{Delta}knob more efficiently transduced both human and murine synovium compared with Ad5GFP-FiWT. In contrast to RA FLS, there was no advantage of the knobless vector compared with Ad5GFP-R7-knob. There was a trend to increased efficacy of Ad5GFP-R7-knob compared with Ad5GFP{Delta}knob, although this was not statistically significant. The difference between transduction efficiency of these two vectors in RA FLS and synovial tissue may be explained by different integrin expression in FLS compared with cellular infiltrates in human or mouse synovium. RA FLS and tissue are known to express multiple integrins (10). Among these, RGD-binding integrins such as {alpha}v{beta}3, {alpha}v{beta}5, {alpha}5{beta}1, non-RGD-binding integrins such as {alpha}3{beta}1, and LDV-binding integrins {alpha}4{beta}1 have been described (10, 12, 46). In murine synovial cells and tissue, CAR is also notably absent, and {alpha}v{beta}3 and {alpha}v{beta}5 integrins are expressed (13). Similarly, in murine lymphocyte and RAW macrophage cell lines, CAR is absent and, {alpha}v integrins are expressed. In CAR-negative cells in synovium such as FLS, macrophages, and lymphocytes we observed increased transduction efficiency of the short fiber-modified vectors compared with the conventional vector. Thus, in the absence of CAR, several integrins may act together as coreceptors to increase viral tropism of Ad5GFP-R7-knob in RA. Thus, availability of Ad5 receptors may determine differential efficacy and selectivity of the short fiber-modified vectors. The differences in selectivity of Ad5GFP-R7-knob or Ad5GFP{Delta}knob could be used advantageously depending on the cellular target in gene therapy. It could be envisaged that Ad5GFP{Delta}knob may be used predominantly in targeting RA FLS for delivery of proapoptotic therapies in which a "genetic synovectomy" effect is desired. In contrast, more broad-spectrum anti-inflammatory effects may be desirable by use of either Ad5GFP-R7-knob or Ad5GFP{Delta}knob in delivery of anti-inflammatory therapies.

Immunogenicity due to the high prevalence of anti-Ad5 Abs may decrease efficiency and feasibility of Ad5 vectors in the clinical setting. This major disadvantage has prompted use of other Ad serotypes such as Ad35, chimeric Ad5 vectors, or alternative viral vectors such as retroviral or adenoassociated vectors in gene therapy of arthritis (3, 14, 47). As much as 70% of RA patients contain neutralizing anti-Ad5 Ab in the SF (47). We observed that Ad5GFP-R7-knob maintained up to a 40-fold increase in gene transduction compared with Ad5GFP-FiWT in the presence of SF. In contrast, Ad5GFP-FiWT was completely neutralized. Furthermore, Ad5GFP{Delta}knob was also partially neutralized in the presence of SF at low MOI of 20–50, suggesting that a significant proportion of anti-Ad5-neutralizing Abs were directed against the Pb and possibly fiber shaft, and not the knob domain. However, at higher MOI of 100–200, Ad5GFP{Delta}knob completely over-rode the neutralizing effects of SF. This suggests that epitopes in the native Pb, which are crucial for Ad5 infection, may be different from those targeted by neutralizing Ad5 Abs, and may have a higher affinity for integrin receptors. This may be consistent with a previous study demonstrating that RGD-specific Ad5 Abs were not promiscuous in human subjects and had weak neutralizing activity, whereas other Pb epitopes were found to be highly neutralizing (48). Other studies have shown that RGD-modified viruses may partially escape viral neutralization (60, 61). This suggests that an additional advantage of broadened viral tropism and integrin targeting by Ad5GFP-R7-knob and Ad5GFP{Delta}knob may be either partial or complete avoidance of viral neutralization by anti-Ad5 Abs in SF, respectively. In vivo, more prolonged transgene expression may be additionally achieved by the possibility of i.a lavage to remove neutralizing Ad5 abs in SF before local i.a vector injection.

We attempted to address the question of whether pre-existing immunity against Ad5 may affect transgene duration in human in vitro studies, because we believe this is more relevant to preclinical human applications. We have shown that there is a clear advantage in use of the short fiber-modified vectors over the conventional WT in avoidance of viral neutralization. Our observation that short fiber-modified vectors circumvent viral neutralization suggests they should have a significant advantage in duration of transgene expression in the human context. In the absence of viral neutralization, Ad5 transgene expression may persist for as long as 6 wk (1, 2). It would be interesting to determine whether the short fiber-modified vectors induced a reduced immune response (Ab formation) against the virus and more prolonged transgene expression in vivo. We are currently generating therapeutic vectors, and in our next study we will examine the question of prolonged therapeutic benefit of the vectors and immune response, which we believe would be more relevant to clinical applications.

In conclusion, Ad5GFP-R7-knob and its deknobbed version Ad5GFP{Delta}knob are two potential powerful novel vectors for local i.a gene delivery in human and murine arthritis. Key determinants of Ad5 tropism in RA FLS include shortened viral fiber length and receptor specificity of the Pb to a spectrum of non-RGD and RGD-binding integrins. Furthermore, knob ablation further enhanced efficacy in RA FLS, which in turn would reduce the required vector dose and potential for induction of arthritis flares. In addition, in the event of even minimal leakage of the vector into the systemic circulation postarticular injection, knob ablation would detarget the vector from CAR-positive liver cells and hence reduce potential liver toxicity. In contrast to the previous focus on altering the knob domain in retargeting strategies, tropism of the native Pb may optimize use of integrins as both attachment and endocytic receptors to enhance gene transduction in RA FLS. An additional advantage of highly efficacious gene transfer using Ad5GFP{Delta}knob was the ability to completely circumvent viral neutralization by SF. Use of such fiber-modified Ad5 may revive the potential clinical usefulness of Ad5 vectors in gene therapy of human RA patients. Use of these vectors in delivery of a therapeutic transgene would be highly desirable, and work has commenced to examine the functional sequelae of the improved constructs compared with the conventional WT in our in vitro, ex vivo, and in vivo models targeting IL-17.


   Acknowledgments
 
We thank Marine Poncherot and Nicolas Gadot for technical advice and assistance. Eric Morand, Bruno Mougin, and Manuel Rosa-Calatrava are gratefully acknowledged for their support, valuable advice, and fruitful discussions. We are indebted to Charles Dumontet and Adriana Plesa for their invaluable aid in flow cytometry.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
The authors have no financial conflict of interest.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 This work was supported by grants from the Association Française contre les Myopathies (AFM), Institut de Myologie, Paris (Subvention AFM-Groupe E/2003/11845-9481), the Association Vaincre la Mucoviscidose (Paris, France), and the Région Rhône Alpes. M.-L.T. was successively supported by grants from the National Health and Medical Research Council of Australia, Arthritis Australia Foundation Fellowship, and the Région Rhône Alpes. Back

2 Address correspondence and reprint requests to Dr. Pierre Miossec, Clinical Immunology Unit, Departments of Immunology and Rheumatology, Hospital Edouard Herriot, 5 place d’Arsonval, 69437 Lyon Cedex 03, France; E-mail address: pierre.miossec{at}univ-lyon1.fr or Dr. Pierre Boulanger, Laboratoire de Virologie and Pathogenèse Virale, Centre National de la Recherche Scientifique-University of California Biomechanics Laboratory UMR-5537, Faculté de Médecine RTH Laennec, 7 rue Guillaume Paradin, 69372 Lyon Cedex 08, France; E-mail address: Pierre.Boulanger{at}sante.univ-lyon1.fr Back

3 Abbreviations used in this paper: RA, rheumatoid arthritis; Ad, adenovirus; Ad5, Ad serotype 5; CAR, coxsackie-Ad receptor; RGD, arginine-glycine-aspartic acid; Pb, penton base; FLS, fibroblast-like synoviocyte; CIA, collagen-induced arthritis; SF, synovial fluid; WT, wild type; MOI, multiplicity of infection; DAPI, 4',6'-diamidino-2-phenylindole; bCII, bovine type II collagen; i.a, intra-articular; p.i., postinfection. Back

Received for publication January 11, 2005. Accepted for publication September 6, 2005.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Robbins, P., C. Evans, Y. Chernajovsky. 2003. Gene therapy for arthritis. Gene Ther. 10: 902-911. [Medline]
  2. Nita, I., S. Ghivizzani, J. Galea-Lauri, G. Bandara, H. Georgescu, P. Robbins, C. Evans. 1996. Direct gene delivery to synovium: an evaluation of potential vectors in vitro and in vivo. Arthritis Rheum. 39: 820-828. [Medline]
  3. van de Loo, F., R. Smeets, W. van den Berg. 2004. Gene therapy in animal models of rheumatoid arthritis: are we ready for the patients?. Arthritis Res. Ther. 6: 183-196. [Medline]
  4. Bergelson, J., J. Cunningham, G. Droguett, E. Kurt-Jones, A. Krithivas, J. Hong, M. Horwitz, R. Crowell, R. Finberg. 1997. Isolation of a common receptor for Coxsackie B viruses and adenovirus 2 and 5. Science 275: 1320-1323. [Abstract/Free Full Text]
  5. Wickham, T., P. Mathias, D. Cheresh, G. Nemerow. 1993. Integrins {alpha}v{beta}3 and {alpha}v{beta}5 promote adenovirus internalization but not virus attachment. Cell 73: 309-319. [Medline]
  6. Huang, S., R. Endo, G. Nemerow. 1995. Upregulation of integrins {alpha}v{beta}3 and {alpha}v{beta}5 on human monocytes and T lymphocytes facilitates adenovirus-mediated gene delivery. J. Virol. 69: 2257-2263. [Abstract]
  7. Li, E., S. Brown, D. Stupack, X. Puente, D. Cheresh, G. Nemerow. 2001. Integrin {alpha}v{beta}1 is an adenovirus coreceptor. J. Virol. 75: 5405-5409. [Abstract/Free Full Text]
  8. Wu, E., L. Pache, D. Von Seggern, T. Mullen, Y. Mikyas, P. Stewart, G. Nemerow. 2003. Flexibility of the adenovirus fiber is required for efficient receptor interaction. J. Virol. 77: 7225-7235. [Abstract/Free Full Text]
  9. Goossens, P., M. Havenga, E. Pieterman, A. Lemckert, F. Breedveld, A. Bout, T. Huizinga. 2001. Infection efficiency of type 5 adenoviral vectors in synovial tissue can be enhanced with a type 16 fiber. Arthritis Rheum. 44: 570-577. [Medline]
  10. Nakayamada, S., K. Saito, K. Fujii, M. Yasuda, M. Tamura, Y. Tanaka. 2003. {beta}1 Integrin-mediated signaling induces intercellular adhesion molecule 1 and fas on rheumatoid synovial cells and fas-mediated apoptosis. Arthritis Rheum. 48: 1239-1248. [Medline]
  11. Pirila, L., J. Heino. 1996. Altered integrin expression in rheumatoid synovial lining type B cells: in vitro cytokine regulation of {alpha}1{beta}1, {alpha}6{beta}1, and {alpha}v{beta}5 integrins. J. Rheumatol. 23: 1691-1698. [Medline]
  12. Rinaldi, N., T. Barth, D. Weis, M. Schwarz-Eywill, A. Pezzutto, M. Lukoschek, D. Brocai, B. Brado. 1998. Loss of laminin and of the laminin receptor integrin subunit {alpha}6 in situ correlates with cytokine induced down regulation of {alpha}6 on fibroblast-like synoviocytes from rheumatoid arthritis. Ann. Rheum. Dis. 57: 559-565. [Abstract/Free Full Text]
  13. Bakker, A., F. Van de Loo, L. Joosten, M. Bennink, O. Arntz, I. Dmitriev, E. Kashentsera, D. Curiel, W. van den Berg. 2001. A tropism-modified adenoviral vector increased the effectiveness of gene therapy for arthritis. Gene Ther. 8: 1785-1793. [Medline]
  14. Vogel, R., D. Zuijdgeest, R. van Rijnsoever, E. Hartkoorn, I. Damen, M.-P. de Bethune, S. Kostense, G. Penders, N. Helmus, W. Koudstaal, et al 2003. Replication-deficient human adenovirus type 35 vectors for gene transfer and vaccination: efficient human cell infection and bypass of preexisting adenovirus immunity. J. Virol. 77: 8263-8271. [Abstract/Free Full Text]
  15. Kasono, K., J. Blackwell, J. Douglas, I. Dmitriev, T. Strong, P. Reynolds, D. Kropf, W. Carroll, G. Peters, R. Bucy, et al 1999. Selective gene delivery to head and neck cancer cells via an integrin targeted adenoviral vector. Clin. Cancer Res. 5: 2571-2579. [Abstract/Free Full Text]
  16. Havenga, M., A. Lemckert, J. Grimbergen, R. Vogels, L. Huisman, D. Valerio, A. Bout, P. Quax. 2001. Improved adenovirus vectors for infection of cardiovascular tissues. J. Virol. 75: 3335-3342. [Abstract/Free Full Text]
  17. Attur, M. G., M. Dave, M. Leung, C. Cipoletta, M. Meseck, S. Woo, A. Amin. 2002. Functional genomic analysis of type II IL-1{beta} decoy receptor: potential for gene therapy in human arthritis and inflammation. J. Immunol. 168: 2001-2010. [Abstract/Free Full Text]
  18. Schmidt, M., B. Piekos, M. Cabatingan, R. Woodland. 2000. Expression of a human Coxsackie-adenovirus receptor transgene permits adenovirus infection of primary lymphocytes. J. Immunol. 165: 4112-4119. [Abstract/Free Full Text]
  19. Magnusson, M., S. Hong, P. Boulanger, L. Lindholm. 2001. Genetic retargeting of adenovirus: novel strategy employing "deknobbing" of the fiber. J. Virol. 75: 7280-7289. [Abstract/Free Full Text]
  20. Michael, S., J. Hong, D. Curiel, J. Engler. 1995. Addition of a short peptide ligand to the adenovirus fiber protein. Gene Ther. 2: 660-668. [Medline]
  21. Dmitriev, I., V. Krasnykh, C. Miller, M. Wang, E. Kashentseva, G. Mikheeva, N. Belousova, D. Curiel. 1998. An adenovirus vector with genetically modified fibers demonstrates expanded tropism via utilization of a cocksackie and adenovirus receptor-independent cell entry mechanism. J. Virol. 72: 9706-9713. [Abstract/Free Full Text]
  22. Gaden, F., L. Franqueville, M. Magnusson, S. Hong, M. Merten, L. Lindholm, P. Boulanger. 2004. Gene transduction and cell entry pathway of fiber-modified adenovirus type 5 vectors carrying novel endocytic peptide ligands selected on human tracheal glandular cells. J. Virol. 78: 7227-7247. [Abstract/Free Full Text]
  23. Gaden, F., L. Franqueville, S. Hong, V. Legrand, C. Figarella, P. Boulanger. 2002. Mechanism of restriction of normal and cystic fibrosis transmembrane conductance regulator-deficient human tracheal gland cells to adenovirus infection and Ad-mediated gene transfer. Am. J. Respir. Cell Mol. Biol. 27: 628-640. [Abstract/Free Full Text]
  24. Kanerva, A., A. Hemminki. 2004. Modified adenoviruses for cancer gene therapy. Int. J. Cancer 110: 475-480. [Medline]
  25. Perlman, H., H. Liu, C. Georganas, J. Woods, M. Amin, A. Koch, T. Wickham, I. Kovesdi, T. Mano, K. Walsh, et al 2002. Modifications in adenoviral coat fiber proteins and transcriptional regulatory sequences enhance transgene expression. J. Rheumatol. 29: 1593-1600. [Abstract/Free Full Text]
  26. Smeets, R., F. van de Loo, L. Joosten, O. Arntz, M. Bennink, W. Loesberg, I. Dmitriev, D. Curiel, M. Martin, W. van den Berg. 2003. Effectiveness of the soluble form of the interleukin-1 receptor accessory protein as an inhibitor of interleukin-1 in collagen-induced arthritis. Arthritis Rheum. 48: 2949-2958. [Medline]
  27. Shayakhmetov, D., A. Lieber. 2000. Dependence of adenovirus infectivity on length of the fiber shaft domain. J. Virol. 74: 10274-10286. [Abstract/Free Full Text]
  28. Arnett, F., S. Edworthy, D. Bloch, D. McShane, J. Frie, N. Cooper, L. A. Healey, S. P. Kaplan, M. H. Liang, H. S. Luthra, et al 1988. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum. 31: 315-324. [Medline]
  29. Toh, M., Y. Yang, M. Leech, L. Santos, E. Morand. 2004. Expression of mitogen-activated protein kinase phosphatase 1, a negative regulator of the mitogen-activated protein kinases, in rheumatoid arthritis: upregulation by interleukin-1{beta} and glucocorticoids. Arthritis Rheum. 50: 3118-3128. [Medline]
  30. Chabaud, M., F. Fossiez, J. Taupin, P. Miossec. 1998. Enhancing effect of IL-17 on IL-1 induced IL-6 and leukemia inhibitory factor production by rheumatoid arthritis synoviocytes and its regulation by Th2 cytokines. J. Immunol. 161: 409-414. [Abstract/Free Full Text]
  31. Chabaud, M., P. Miossec. 2001. The combination of tumor necrosis factor {alpha} blockade with interleukin-1 and interleukin-17 blockade is more effective for controlling synovial inflammation and bone resorption in an ex vivo model. Arthritis Rheum. 44: 1293-1303. [Medline]
  32. Miossec, P., J. Briolay, J. Dechanet, J. Wijdenes, H. Martinez-Valdez, J. Banchereau. 1992. Inhibition of the production of proinflammaotry cytokines and immunoglobulins by interleukin-4 in an ex vivo model of rheumatoid arthritis. Arthritis Rheum. 35: 874-883. [Medline]
  33. Molinier-Frenkel, V., A. Prevost-Blondel, S. Hong, R. Lengagne, S. Boudaly, M. Magnusson, P. Boulanger, J.-G. Guillet. 2003. The maturation of murine dendritic cells induced by human adenovirus is mediated by the fiber knob domain. J. Biol. Chem. 278: 37175-37182. [Abstract/Free Full Text]
  34. Defer, C., M. Belin, M. Caillet-Boudin, P. Boulanger. 1990. Human adenovirus-host cell interactions: comparative study with members of subgroups B and C. J. Virol. 64: 3661-3673. [Abstract/Free Full Text]
  35. Karayan, L., B. Gay, J. Gerfaux, P. Boulanger. 1994. Oligomerization of recombinant penton base of adenovirus type 2 and its assembly with fiber in baculovirus-infected cells. Virology 202: 782-795. [Medline]
  36. Karayan, L., S. Hong, B. Gay, J. Tournier, D. D’Angeac, P. Boulanger. 1997. Structural and functional determinants in adenovirus type 2 penton base recombinant protein. J. Virol. 71: 8678-8689. [Abstract]
  37. Molinier-Frenkel, V., R. Lengagne, F. Gaden, S. S. Hong, J. Choppin, H. Gahery-Segard, P. Boulanger, J. G. Guillet. 2002. Adenovirus hexon protein is a potent adjuvant for activation of a cellular immune response. J. Virol. 76: 127-135. [Abstract/Free Full Text]
  38. Salone, B., Y. Martina, S. Piersanti, E. Cundari, G. Cherubini, L. Franqueville, C. Failla, P. Boulanger, I. Saggio. 2003. Integrin {alpha}3{beta}1 is an alternative cellular receptor for adenovirus serotype 5. J. Virol. 77: 13448-13454. [Abstract/Free Full Text]
  39. Stallwood, Y., K. Fisher, P. Gallimore, V. Mautner. 2000. Neutralization of adenovirus infectivity by ascitic fluid from ovarian cancer patients. Gene Ther. 7: 637-643. [Medline]
  40. Smeets, R., L. Joosten, O. Arntz, M. Bennink, N. Takahashi, H. Carlsen, M. Martin, W. van den Berg, F. van de Loo. 2005. Soluble interleukin-1 receptor accessory protein ameliorates collagen-induced arthritis by a different mode of action from that of interleukin-1 receptor antagonist. Arthritis Rheum. 52: 2202-2211. [Medline]
  41. Dufour, J., R. Hemendinger, C. Halberstadt, P. Gores, D. Emerich, G. Korbutt, R. Rajotte. 2004. Genetically engineered sertoli cells are able to survive allogeneic transplantation. Gene Ther. 11: 694-700. [Medline]
  42. Page, G., P. Miossec. 2004. Paired synovium and lymph nodes from rheumatoid arthritis patients differ in dendritic cell and chemokine expression. J. Pathol. 201: 28-38.
  43. Dechecchi, M., P. Melotti, A. Bonizzato, M. Santacatterina, M. Chilosi, G. Cabrini. 2001. Heparin sulfate glycosaminoglycans are receptors sufficient to mediate the initial binding of adenovirus types 2 and 5. J. Virol. 75: 8772-8780. [Abstract/Free Full Text]
  44. Hong, S. S., L. Karayan, J. Tournier, D. T. Curiel, P. Boulanger. 1997. Adenovirus type 5 fiber knob binds to MHC class I {alpha}2 domain at the surface of human epithelial and B lymphoblastoid cells. EMBO J. 16: 2294-2306. [Medline]
  45. Humphries, M.. 1990. The molecular basis and specificity of integrin-ligand interactions. J. Cell Sci. 97: 782-796.
  46. Rinaldi, N., D. Weis, B. Brado, M. Schwarz-Eywill, M. Lukoschek, A. Pezzutto, U. Keilholz, T. Barth. 1997. Differential expression and functional behaviour of the {alpha}v and {beta}3 integrin subunits in cytokine stimulated fibroblast-like cells derived from synovial tissue of rheumatoid arthritis and osteoarthritis in vitro. Ann. Rheum. Dis. 56: 729-736. [Abstract/Free Full Text]
  47. Goossens, P., R. Vogels, E. Pieterman, M. Havenga, A. Bout, F. Breedveld, D. Valerio, T. Huizinga. 2001. The influence of synovial fluid on adenovirus-mediated gene transfer to the synovial tissue. Arthritis Rheum. 44: 48-52. [Medline]
  48. Hong, S., N. Habib, L. Franqueville, S. Jensen, P. Boulanger. 2003. Identification of adenovirus (Ad) penton base neutralizing epitopes by use of sera from patients who have received conditionally replicative Ad (Addl1520) for treatment of liver tumors. J. Virol. 77: 10366-10375. [Abstract/Free Full Text]
  49. Gahery-Segard, H., F. Farace, D. Godfrin, J. Gaston, T. Lengagne, T. Tursz, P. Boulanger, J.-G. Guillet. 1998. Immune response to recombinant capsid proteins of adenovirus in humans: anti-fiber and anti-penton base antibodies have a synergistic effect on neutralizing activity. J. Virol. 72: 2388-2397. [Abstract/Free Full Text]
  50. Ghivizzani, S., R. Kang, C. Hatton, C. Tio, E. Lechman, A. Emmert, P. D. Robbins, C. H. Evans. 1996. In vivo delivery of genes encoding soluble receptors for IL-1 and TNF{alpha} results in a synergistic therapeutic effect in antigen-induced arthritis in the rabbit knee. Arthritis Rheum. 39: S308 (Abstr).
  51. Hidaki, C., E. Milano, P. Leopold, J. Bergelson, N. Hackett, R. Finberg, T. Wickham, I. Kovesdi, P. Roelvink, R. Crystal. 1999. CAR-dependent and CAR-independent pathways of adenovirus vector-mediated gene transfer and expression in human fibroblasts. J. Clin. Invest. 103: 579-587. [Medline]
  52. Roelvink, P., I. Kovesdi, T. Wickham. 1998. Comparative analysis of adenovirus fiber-cell interaction: adenovirus type 2 (Ad2) and Ad9 utitlize the same cellular fiber receptor but use different binding strategies for attchment. J. Virol. 70: 7614-7621.
  53. Roelvink, P., A. Lizonova, J. Lee, Y. Li, J. Bergelson, R. Finberg, D. Brough, I. Kovesdi, T. Wickham. 1998. The cocksackievirus-adenovirus receptor protein can function as a cellular attachment protein for adenovirus serotypes from subgroups A, C, D, E, and F. J. Virol. 72: 7909-7915. [Abstract/Free Full Text]
  54. Wickham, T., M. Carrion, I. Kovesdi. 1995. Targeting of adenovirus penton base to new receptors through replacement of its RGD motif with other receptor-specific peptide motifs. Gene Ther. 2: 750-756. [Medline]
  55. Vigne, E., I. Mahfoux, F. Dedieu, A. Brie, M. Perricaudet, P. Yeh. 1999. RGD inclusion in the hexon monomer provides adenovirus type 5-based vectors with a fiber knob-independent pathway for infection. J. Virol. 73: 5156-5161. [Abstract/Free Full Text]
  56. Miagkov, A., A. Varley, R. Munford, S. Makarov. 2002. Endogenous regulation of therapeutic transgene restores homeostasis in arthritic joints. J. Clin. Invest. 109: 1223-1229. [Medline]
  57. van de Loo, F., A. de Hooge, R. Smeets, A. Bakker, M. Bennink, O. Arntz, L. Joosten, H. van Beuningen, P. van der Kraan, A. Varley, et al 2004. An inflammation-inducible adenoviral expression system for local treatment of the arthritic joint. Gene Ther. 11: 581-590. [Medline]
  58. van de Loo, F.. 2004. Inflammation-responsive promoters for fine-tuned gene therapy in rheumatoid arthritis. Curr. Opin. Mol. Ther. 6: 537-545. [Medline]
  59. Reynolds, P., I. Dmitriev, D. Curiel. 1999. Insertion of an RGD motif into the H1 loop of adenovirus fiber protein alters the distribution of transgene expression of the systemically administered vector. Gene Ther. 6: 1336-1339. [Medline]
  60. Kanerva, A., M. Wang, G. Bauerschmitz, J. Lam, R. Desmond, S. Bhoola, M. Barnes, R. Alvarez, G. Siegal, D. Curiel, et al 2002. Gene transfer to ovarian cancer versus normal tissues with fiber-modified adenoviruses. Mol. Ther. 5: 695-704. [Medline]
  61. Blackwell, J., H. Li, J. Gomez-Navarro, I. Dmitriev, V. Krasnykh, C. Richter, D. Shaw, R. Alvarez, D. Curiel, T. Strong. 2000. Using a tropism-modified adenoviral vector to circumvent inhibitory factors in ascites fluid. Hum. Gene Ther. 11: 1657-1669. [Medline]



This article has been cited by other articles:


Home page
 J. Virol.Home page
O. Granio, M. Porcherot, S. Corjon, K. Kitidee, P. Henning, A. Eljaafari, A. Cimarelli, L. Lindholm, P. Miossec, P. Boulanger, et al.
Improved Adenovirus Type 5 Vector-Mediated Transduction of Resistant Cells by Piggybacking on Coxsackie B-Adenovirus Receptor-Pseudotyped Baculovirus
J. Virol., June 15, 2009; 83(12): 6048 - 6066.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Toh, M.-L.
Right arrow Articles by Miossec, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Toh, M.-L.
Right arrow Articles by Miossec, P.
Right arrowPubmed/NCBI databases
*Compound via MeSH
*Substance via MeSH
Medline Plus Health Information
*Genes and Gene Therapy
*Joint Disorders
*Rheumatoid Arthritis

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS