Brief Report
1 November 2006

Fiber and Penton Base Capsid Modifications Yield Diminished Adenovirus Type 5 Transduction and Proinflammatory Gene Expression with Retention of Antigen-Specific Humoral Immunity


Fiber and penton base capsid proteins of adenovirus type 5 (Ad5) mediate a well-characterized two-step entry pathway in permissive tissue culture cell lines. Fiber binds with high affinity to the cell surface coxsackievirus-and-adenovirus receptor (CAR), and penton base facilitates viral internalization by binding αv integrins through an RGD motif. In vivo, the entry pathway is complicated by interactions of capsid proteins with additional cell surface molecules and blood factors. When administered systemically in mice, adenovirus vectors (Adv) localize primarily to hepatic tissue, resulting in efficient gene transduction and potent activation of the host antiviral immune response. The goal of the present study was to detarget Adv uptake through fiber and penton base capsid protein manipulations and determine how detargeted vectors influence transduction efficiency, inflammatory activation, and activation of the adaptive arm of the immune system. By manipulating fiber and the penton base, we have generated highly detargeted vectors (up to 1,200-fold reduction in transgene expression in vivo) with reduced macrophage stimulatory activity in vitro and in vivo. In spite of the diminished transduction and macrophage activation, the detargeted vectors induce strong neutralizing immunity as well as efficient antitransgene antibody. Three of the modified vectors produce antitransgene humoral immunity at levels that exceed or are equal to that seen with an unmodified Ad5-based vector. The fiber-pseudotyped and penton base constructs with RGD deleted have attributes that could be important enhancements in a number of vaccine applications.
The utility of adenovirus (Ad) vectors (Adv) in systemic gene transfer applications is limited in part by a lack of cell-specific targeting and a potent activation of host antiviral immunity. A goal in Adv biology is the generation of vectors that avoid immune recognition and target a specific cell or tissue type. Controlling the in vivo distribution of Adv requires a thorough understanding of the virus-host cell interactions that mediate both infection and immune activation. The major capsid proteins fiber, penton base, and hexon are principle mediators of these interactions.
In permissive tissue culture cell lines, Ad5 vectors undergo a two-step entry process that is initiated by high-affinity binding of the capsid protein fiber to the cell surface coxsackievirus-and-adenovirus receptor (CAR) (2, 4, 26, 34). Subsequent interaction of the penton base RGD motif with cellular αv integrins facilitates virus internalization (39). This canonical Ad entry pathway permits efficient transgene delivery into cell lines expressing high levels of CAR, including hepatocytes. In cells where CAR expression is minimal, such as macrophages, the level of transgene expression is low when transduced by Ad5-based vectors (14). Although CAR and integrin binding are the dominant elements that contribute to Ad5-mediated transduction in vitro, interactions of Ad capsid with major histocompatibility complex molecules and heparan sulfate glycosaminoglycans (HSG) have also been demonstrated (6, 10).
When administered systemically, Adv are primarily sequestered in hepatic tissue and mediate high levels of gene transduction (41). Studies in rodents and primates suggest that the mechanisms of virus sequestration and gene transduction in liver do not conform to the canonical two-step entry pathway. Ad5 vectors expressing capsid mutants that abolish fiber-CAR and/or penton-integrin binding fail to reduce Ad-mediated gene delivery to liver (1, 21, 31). In contrast, Adv mutated in the fiber KKTK motif, a putative HSG binding domain, show marked reductions in liver transduction in mice and nonhuman primates (33, 38). Chimeric Ad5-based vectors that have been psuedotyped with either CAR-binding fibers lacking a KKTK motif or CAR-nonbinding fibers also exhibit reduced liver transduction and localization (16, 23, 24, 28, 29). The binding of fiber to blood factors such as factor IX has also been suggested to facilitate virus uptake in liver (30). Taken together, these studies highlight a role for fiber in liver sequestration and transduction that is primarily CAR independent.
The penton base RGD motif contributes to adenovirus internalization by binding to membrane integrins and stimulating endocytosis. In cells that possess CAR, the combination of fiber-CAR and penton base-integrin interactions results in a highly effective delivery of viral genome into the cell. In the absence of CAR binding, integrins can serve as both the primary receptor and activator of endocytosis (12). The majority of monocytic cells lack high levels of CAR, and penton base RGD has been shown to facilitate virus entry into a variety of hematopoietic cells (11, 12). This interaction has also been exploited by generating Ad variants that include the penton base RGD motif in additional locations in capsid proteins, notably fiber (7, 36). Insertion of the RGD motif into the fiber knob has been shown to enhance transduction and activation of dendritic cells (40). In association with enhanced vector internalization, the penton base RGD motif contributes to Ad-mediated stimulation of professional antigen-presenting cells, macrophages, and dendritic cells. Bone marrow-derived dendritic cells are stimulated by Ad in an RGD-dependent manner, in part through a phosphatidylinositol 3-kinase (PI3K)-mediated tumor necrosis factor alpha (TNF-α) induction pathway (25). In kidney epithelium-derived (REC) cells, vectors lacking the RGD motif were defective for activation of the chemokine IP-10. RGD-dependent induction of inflammatory genes in REC cells occurred through an Akt/PI3K pathway (18). Liu et al. also established a role for RGD in activation of endothelial cells in vitro and in vivo (19). Taken together, the evidence suggests that penton RGD is an important component of early inflammation stimulated by Ad infection.
The majority of systemically administered Adv localized to the liver is degraded within 24 h of infection, reflecting the efficiency of innate pathways of virus clearance(41). The early inflammatory response to Ad is characterized by the production of proinflammatory cytokines and chemokines and is influenced by the dose of virus used in the initial administration (43). A variety of cell types contribute to the secretion of proinflammatory molecules, including endothelial cells, hepatic macrophages (Kupffer cells), and dendritic cells (19, 43). Additionally, at moderate to high doses, systemic Ad administration results in liver toxicity, alterations in blood pressure, and Kupffer cell necrosis (17, 20, 27). Following induction of the antiviral inflammatory response, the host induces a strong cellular immune response to Adv that features cytotoxic T-lymphocyte (CTL)-mediated killing of infected cells (42), followed by production of antiviral humoral immunity (reviewed in reference 3). In addition to a humoral immune response to the virus and capsid elements, transgenes expressed from Ad5-based vectors are targets of the adaptive immune response, resulting in antitransgene-specific immunity (13). Issues of immune activation are major obstacles impeding the use of Adv in gene therapy applications, but they are also the main feature making Adv attractive vaccine vectors.
The previous studies highlight the complexity of adenovirus-host cell interactions in vitro and in vivo. A prerequisite for successful Ad targeting is the generation of detargeted vectors that avoid liver sequestration and exhibit a diminished immune activation profile. In the present study, we characterize the impact of fiber replacements in combination with an RGD deletion on vector transduction and immune activation in vitro and in vivo. Our Ad5-based vectors follow canonical Ad binding pathways in tissue culture cell lines. In vivo, Ad5 fiber contributes to maximal liver transduction in a CAR-independent manner. The penton base RGD motif plays a significant role in liver transduction, but only with fiber-pseudotyped vectors, which lack putative HSG binding motifs. Both fiber and penton base RGD contribute to induction of proinflammatory cytokines and chemokines in liver. In spite of greatly reduced transgene expression and reduced inflammatory activity in vitro and in vivo, the novel vectors presented in this study are equal to or better than Ad5-based vectors at stimulating transgene-specific humoral immunity.


Adenovirus vector stocks and cell lines.

Six Ad5-based vectors with E1/E3 deleted were used in this study: Ad5, F41T, and F7F41S and their counterparts with RGD deleted, Ad5Δ, F41TΔ, and F7F41SΔ. All vectors contain a cytomegalovirus (CMV)-driven chloramphenicol acetyltransferase (CAT) internal ribosome entry site (IRES)-green fluorescent protein (GFP) (CATiresGFP [CiG]) dual-reporter cassette in the E1 region. To construct the reporter, the pAdCMVCAT plasmid containing left-end Ad sequences was modified by subcloning a CMV promoter driving the CAT reporter gene followed by an IRES-driven enhanced GFP cassette from the pIRES2-EGFP plasmid (Clontech) into the BamHI site downstream of CAT. This construct, pAdCMVCATiresGFP, was cotransfected into 293 cells with an XbaI-digested right-end large fragment from the dlAd5NCAT vector (9) to create Ad5CiG by overlapping homologous recombination. The Ad5CiG vector with RGD deleted was constructed from a parental Ad5βgalΔRGD virus. The left end of Ad5βgalΔRGD was released by ClaI digestion, and the right-end large fragment was recombined with pAdCMVCATiresGFP in 293 cells to generate AdCiGΔ. To create fiber-modified CiG vectors, Ad5CiG and Ad5CiGΔ were digested with PsiI to release right-end fiber-containing fragments. The left-end large fragments were purified and recombined in 293 cells with right-end-containing plasmid pAd70-100.dlE3.F41T or pAd70-100.dlE3.F7F41S (28) to give AdCiG.F41T, AdCiG.F7F41S, AdCiG.F41TΔ, and AdCiG.F7F41SΔ. Viruses were subjected to two rounds of plaque purification and amplified in 293 cells. Viruses were grown large scale in 293 cells, followed by two rounds of CsCl banding purification and dialysis against storage buffer (4% sucrose, 50 mM Tris, pH 8.0, 2 mM MgCl2) to remove salts. Viral particle numbers were quantified by spectrophotometric detection of intact virions at the optical density at 260 nm (OD260) (1012 particles/OD260 unit). HEK-293 monolayer cultures and MS1 endothelial cells (ATCC no. CRL-2279) were kept in Dulbecco's modified Eagle's medium (DMEM) plus 5 to 10% cosmic calf serum (HyClone). FL83B murine hepatocyte (ATCC no. CLR-2390) cultures were maintained in F12K plus 10% fetal bovine serum (FBS). RAW246.7 murine macrophages were generously provided by Carl Nathan and kept in DMEM plus 10% FBS.

In vitro transduction assays.

Cell lines used in this study were seeded into 24-well plates and infected with 1,000 particles/cell in medium without serum for 30 min at 37°C. After infection, virus was aspirated and fresh medium with serum was added back. Infected cells were harvested 24 h postinfection in TEN scrape buffer (40 mM Tris HCl, pH 7.4, 1 mM EDTA, 150 mM NaCl) and resuspended in 0.25 M Tris, pH 7.8. Lysates were subjected to three freeze-thaw cycles followed by a 10-min incubation at 65°C to inactivate cellular deacetylases. CAT activity assays were performed as described previously (9).

BMMO isolation and infection.

Bone marrow macrophages (BMMO) were generated by culturing bone marrow cells in DMEM plus 20% FBS and 30% supernatant derived from L929 confluent cells, replacing two-thirds of the culture volume with fresh macrophage colony-stimulating factor-containing medium every third day. Cells were used after 7 to 9 days, and all stained positive for CD11b but were negative for CD11c and B220. Cultures of BMMO were infected with the indicated virus at 5,000 particles/cell. For transduction experiments, cells were harvested 24 h postinfection for CAT assay. For the maturation phenotype, cells were harvested 36 h postinfection and stained on ice with phycoerythrin-conjugated anti-CD86 mouse antibody (Becton Dickson) for 30 min in phosphate-buffered saline (PBS) containing 1% bovine serum albumin. Cells were washed in PBS and analyzed on an EPICS XL flow cytometer (Beckman Coulter).

In vivo transduction assays.

Six-week-old female B6129/J mice were obtained from Taconic and maintained in compliance with institutional protocols. Mice were injected retro-orbitally with virus or PBS vehicle alone in a total volume of 100 μl. At indicated time points, animals were sacrificed and livers were harvested, weighed, and resuspended in 2 volumes PBS by weight. Tissue lysates were prepared by homogenizing livers and centrifugation for 30 min at 3,000 × g, followed by a 10-min incubation at 65°C. Total liver CAT activity was assayed as described above.

DNA isolation and Southern analysis.

Livers from Ad- or mock-infected animals were harvested, and a portion was used for DNA isolation. Fresh tissue slices were pulverized mechanically with sterile pellet pestles in Eppendorf tubes and incubated in lysis buffer (10 mM Tris, pH 8.0, 400 mM NaCl, 2 mM EDTA, 0.5% sodium dodecyl sulfate, 0.4 mg/ml proteinase K) overnight at 55°C. Samples were treated with 2 μl RNase A (10 mg/ml) for 30 min at 37°C. DNA was purified by two rounds of phenol-CHCl3 extraction and a CHCl3-only extraction, followed by ethanol precipitation. For Southern analysis, 10 μg DNA was digested with EcoRI and electrophoresed on a 1% agarose gel. Samples were transferred to a nylon membrane, and blots were probed with either EcoRI-digested Ad5 vector DNA or Rad50 as a cellular control. DNA levels were quantified by PhosphorImager analysis.

RNA isolation and RNase protection assay.

Bone marrow-derived macrophages and liver sections from from Ad- or mock-infected animals were harvested for RNA isolation with Trizol reagent (Invitrogen) according to the manufacturer's instructions. Five micrograms (BMMO) or 10 μg (liver) of total RNA was hybridized to 32P-labeled probes generated from either the mCK3b (cytokine) or mCK5c (chemokine) multiprobe template sets (Pharmingen). A commercially available RNase protection assay kit (Pharmingen) was used to RNase treat samples and recover protected fragments. Resolution of protected fragments was carried out by electrophoresis on 10% polyacrylamide sequencing gels. Gels were dried and exposed to a PhosphorImager screen, and mRNA species were quantified by PhosphorImager analysis.

Anti-Ad neutralizing antibody blocking assay.

Neutralizing antibody titers were determined by a functional in vitro blocking assay in HeLa cells. Serum samples from mice infected with the various vectors were taken at day 28 and serially diluted in DMEM. Serum dilutions were incubated with 5 × 105 particles of the F41T vector for 1 h at 37°C. The serum-virus mixture was then added to 105 HeLa cells for 30 min at 37°C. Cells were washed with DMEM, and fresh medium was added back. CAT activity in cell extracts was scored 24 h later as described above.

Anti-CAT antibody ELISA.

Serum levels of antitransgene antibody were detected by enzyme-linked immunosorbent assay (ELISA). CAT protein was expressed in the bacterial pET16b vector system (Stratagene) according to the manufacturer's instructions. Maxisorp ELISA plates (Nalge Nunc International) were coated with recombinant CAT in 50 mM sodium carbonate buffer, pH 9.6, for at least 16 h at room temperature. The CAT solution was removed, and 1% casein (Sigma) in PBS was added as a blocking buffer for 1 h at room temperature. Plates were washed three times in blocking buffer plus 0.05% Tween 20. Serum samples were serially diluted in blocking buffer and added to ELISA plates for 1.5 h at room temperature. Plates were washed three times as described above. A secondary horseradish peroxidase-conjugated anti-mouse antibody (Amersham) was diluted in blocking buffer and added to plates for 1.5 h at room temperature. Plates were then washed four times and incubated with tetramethylbenzidine peroxidase substrate (Sigma). Reactions were stopped with 1 N H2SO4, and plates were read at 450 nm in a spectrophotometer.


Capsid-modified Adv construction and characterization.

Six novel Ad5-based vectors were constructed to characterize the role of fiber and the penton base RGD motif in vector transduction and immune activation. All vectors are based on an Ad5 parent with E1/E3 deleted and express a bicistronic reporter gene cassette with a CMV promoter driving the CAT reporter gene followed by an IRES-driven EGFP (CATiresGFP [CiG]). Ad5 serves as our control vector with respect to binding functions. Ad5 binds CAR via fiber and integrins via penton RGD and contains the fiber shaft KKTK motif that has been shown to contribute to liver uptake in vivo. All capsid-modified vectors are devoid of one or more of these three interactions (Fig. 1A). The fiber-modified RGD-containing vectors F41T and F7F41S are based on previously described vector constructs (28). Briefly, F41T expresses both long (41L) and short (41S) fibers from Ad41 in a tandem cassette from the L5 region of the major late transcription unit. The nonbinding fiber 41S is expressed on F41T capsids at levels equal to or slightly higher than those of fiber 41L. F7F41S expresses the Ad7 fiber and fiber 41S, but fiber incorporation into virions is highly biased toward fiber 41S (28, 29). In F7F41S vectors, we routinely detect little to no fiber 7 expression by Western blotting or Coomassie staining (data not shown). The counterparts of Ad5, F41T, and F7F41S with RGD deleted (Fig. 1A) were generated by homologous recombination as described in Materials and Methods. To characterize capsid protein content, 1010 particles of each vector were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and analyzed by Western blotting with antihexon antibody. The blot was then probed with antifiber antibody (4D2; Neomarkers) to verify the fiber content of each of the vectors (data not shown). Limiting dilution PCR of viral DNA was used to confirm that DNA content was consistent with the OD260 and Western blot analysis of capsid proteins (data not shown).

Adv transduction in murine cell lines.

When administered systemically in a mouse model, Adv are primarily sequestered in the liver and come into contact with a number of cell types, including hepatocytes, endothelial cells, and Kupffer macrophages (22). To assess in vitro transduction functions in representative cells, murine hepatocyte (FL83B), endothelial (MS1), and macrophage (RAW264.7) cell lines were infected with each vector at 103 Ad particles/cell for 30 min. Cells were incubated for 24 h and harvested for CAT reporter gene activity. Under identical infection conditions, all Adv except F7F41SΔ transduced hepatocytes more efficiently than endothelial or macrophage cell lines (Fig. 1B). In FL83B, the highest levels of CAT activity were observed with vectors expressing CAR-binding fibers, consistent with the observation that hepatocytes express high levels of CAR (8). The RGD penton base deletion had no significant impact on Ad5 transduction under these infection conditions. The F41T and F41TΔ constructs were diminished in transduction by approximately two- and fivefold, respectively, compared to Ad5. CAR binding fiber 41L present in F41T was the dominant factor in efficient entry, and penton RGD deletion had a modest but significant influence on the efficiency of hepatocyte transduction. The F7F41S lacks known high-affinity binding activity in murine cells, and FL83B transduction was decreased by approximately 50-fold compared to Ad5, consistent with previous findings (29). When the penton base RGD deletion was combined with F7F41S, an additional sixfold decrease was obtained, resulting in transduction levels approaching those found with mock infections. Infection of the endothelial MS1 cell line by each Adv resulted in a transduction profile dominated by CAR binding fiber functions. However, transduction levels were greatly reduced in these cells compared to those in FL83B hepatocytes. This is most likely due to the diminished presence of CAR in endothelial cell lines (5, 37). In RAW macrophages, transduction levels by all vectors were extremely low. CAT expression was marginally above background, and the contribution of fiber content or penton base RGD to transduction was negligible (Fig. 1B). The in vitro transduction assays reveal a clear functional role of fiber and penton base RGD in transduction efficiency and establish F7F41SΔ as a markedly detargeted vector in all cell lines tested.

Transduction and activation of bone marrow-derived macrophages.

We previously established that primary BMMO were more permissive to infection by Adv than the RAW264.7 or P6 macrophage cell lines and demonstrated that Ad5, F41T, and F7F41S were able to equally activate BMMO to induce expression of costimulatory molecules such as CD86 as well as induce expression of proinflammatory cytokines and chemokines (29). To assess the impact of a penton base RGD deletion alone or in combination with fiber replacements in BMMO transduction and activation, we infected primary BMMOs with 5,000 particles/cell of each vector. Thirty-six hours postinfection, BMMO were either harvested for CAT assay or stained with a phycoerythrin-conjugated anti-CD86 antibody and analyzed by flow cytometry for both GFP and CD86 (Fig. 2A). Flow cytometry results revealed that 6% or less of cells infected with fiber-modified vectors expressed detectable levels of GFP, and consistent with our previous observations, fiber replacements had no influence on transduction in BMMO (Fig. 2A and B). Deletion of penton base RGD resulted in the reduction of cells expressing detectable GFP to mock levels (in Fig. 2A and B, for example, compare F41T to F41TΔ). The CAT assay offers a greater sensitivity than GFP as a measure of gene expression, and for all fiber-modified constructs, the levels of CAT expression were qualitatively consistent with the levels of GFP expression (Fig. 2C). In agreement with the GFP analysis, CAT transduction was largely dependent on the penton base RGD motif; however, CAT levels with RGDΔ constructs were above the levels found with mock infection. These results indicate a low level of vector uptake occurs in BMMO independent of the fiber and penton modifications employed in these studies.
One measure of BMMO activation involves the upregulation of costimulatory molecules such as CD40, CD80, and CD86. Flow analysis demonstrates that Adv induce expression of CD86 in primary macrophages in a manner that is largely independent of fiber (Fig. 2A, compare the percentages of CD86-positive cells in the lower right of each panel, for Ad5, F41T, and F7F41S). In each RGDΔ construct, there is a diminished percentage of cells fractionated as CD86 positive. Surprisingly, of the three RGDΔ constructs, the least responsive was F41TΔ, followed by Ad5Δ and F7F41SΔ. In addition, for each infection, the GFP-positive cells were equally distributed between CD86+ versus CD86 cells. Detectable GFP expression does not correlate with induction of CD86.
The penton base RGD motif deletion has an impact on transduction efficiency as well as induction of BMMO maturation/activation markers such as CD86 at 24 to 48 h postinfection (Fig. 2). To establish the effect of penton base RGD deletion on the expression of a broad set of proinflammatory cytokines and chemokines shortly after virus exposure, we assessed mRNA levels of cytokines and chemokines in BMMO 5 h postinfection. Total RNA was harvested following infection and assayed for transcript levels by RNase protection assay using multitemplate probe sets (as described in Materials and Methods). Chemokine mRNAs MIP-1β, IP-10, TCA3, MIP-1α, and RANTES were all induced following infection with fiber-modified Adv, and the cytokine mRNAs corresponding to TNF-α and beta interferon (IFN-β) were induced over mock infection, but levels were modest (Fig. 3). Treatment of BMMO with bacterial endotoxin lipopolysaccharide was included as a positive control. Analysis of RNA following infection with each of the vectors with RGD deleted demonstrated a much weaker level of chemokine and cytokine induction compared to those of RGD-containing counterpart vectors. Results from the BMMO studies (Fig. 2 and 3) confirm our previous findings that fiber replacements do not impact on macrophage activation (29) and highlight a specific role for penton base RGD in Ad-mediated transduction and activation of this cell type.

Influence of fiber and penton base-modified Adv on transduction and localization following a systemic administration.

F7F41SΔ represents a viral vector that has been depleted of the primary viral cell-targeting mechanisms that mediate the two-step entry mechanism used by adenovirus for efficient infection in tissue culture settings. To determine the influence of Adv detargeting following a systemic administration, B6129/J mice were injected through retro-orbital administration of 1010 particles of each Adv. Groups of five animals were infected with Adv, and animals were sacrificed at 5 h, 24 h, and 28 days postinfection. An emphasis was placed on analyzing hepatic transduction since previous studies demonstrated that over 90% of Ad5 vector localizes to liver (41). The CAT expression profile at 5 h (Fig. 4A) was similar to that found in the FL83B hepatocyte cell lines (Fig. 1B), where fiber-directed CAR binding (Ad5 and Ad41T) had the largest influence on CAT expression and penton base-mediated integrin binding (Ad5Δ, Ad41TΔ, and AdF7F41SΔ) makes a significant contribution to levels of CAT gene expression in the pseudotyped constructs. At 24 h, the magnitude of transduction for all vectors increased and striking differences with respect to the role of penton base RGD in transgene delivery were revealed. Ad5 transduced liver cells to the highest level, and the RGD deletion in Ad5Δ had little effect on delivery of the reporter gene. These data indicate that liver transduction by Ad5 relies primarily on the non-CAR binding functions of fiber 5 and does not depend on a penton-integrin interaction. The F41T and F41TΔ vectors were impaired in transduction by approximately 8- and 500-fold, respectively, compared to Ad5. CAT expression levels following administration of F7F41S and F7F41SΔ were diminished by approximately 15- and 1,200-fold, respectively, compared to Ad5. For each of the fiber-modified vectors, deletion of the RGD motif conferred an additional 60-to 80-fold reduction in liver CAT expression. Taken together, the data suggest that in the absence of binding functions specific to fiber 5, Adv such as F41T and F7F41S depend primarily on the penton base to facilitate vector transduction. This occurs in spite of a functional CAR binding fiber present in F41T. Infection with the F41TΔ and F7F41SΔ vectors resulted in liver CAT activity that was over background levels. Residual transduction by F41TΔ and F7F41SΔ is indicative of transgene delivery that is both CAR and integrin independent.
CAT transduction provides an indication of successful virus entry and delivery of vector genome to the nucleus for transcription by the host transcriptional apparatus. In the case of Adv, only a small fraction of administered virus contributes to gene expression, and the remainder is eliminated through degradative processes associated with Kupffer cell uptake of virus (17, 41). To determine the impact of penton and fiber manipulations on hepatic localization, total DNA was isolated from livers of mock- or Adv-infected mice 5 h postinjection. DNA was digested and characterized by Southern blot analysis, probing membranes with either 32P-labeled total Ad5 viral DNA or a Rad50 probe as a cellular control. Of the six vectors, Ad5 DNA was present at the highest levels and Ad5Δ DNA was decreased by about 25% in comparison (Fig. 4B). All other vectors had approximately 60% less viral DNA associated with the liver compared to Ad5. These data are consistent with a model in which both fiber 5 and penton base RGD cooperate in mediating maximal Ad5 uptake in liver. The data also indicate that although the modified vectors are diminished in transduction by several orders of magnitude (Fig. 4A), the decrease in viral localization to liver is less dramatic (Fig. 4B). Studies characterizing redistribution to other tissues (primarily spleen, blood, or lung) do not provide quantitative values that account for lost viral vector (15, 28; data not shown).

Early inflammatory response to capsid-modified vectors.

The two viral vectors F41TΔ and F7F41SΔ illustrate two desirable features: they are compromised in virus entry in vitro as well as in vivo, and when used to infect primary BMMO, they are less potent in activating an inflammatory cascade. We next asked if the combination of detargeting and diminished macrophage activation results in a reduced level of early immune activation in vivo. Total RNA was harvested from liver slices 5 h postinjection. Ten-microgram samples of RNA were used in RNase protection assays as previously described. At 1010 particles/mouse, minimal cytokine (TNF-α) induction with Ad5 was found (data not shown) and all other vectors were similar to the mock vector with respect to cytokine upregulation. Induction of chemokines was highest with the penton-containing CAR binding vectors Ad5 and F41T (Fig. 4C). Deletion of the RGD motif significantly reduced chemokine mRNA induction by the Ad5Δ and F41TΔ vectors to nearly mock levels. F7F41S and F7F41SΔ were essentially at background with respect to chemokine mRNA upregulation. Consistent with in vitro assays and in vivo transduction assays at 5 h, the induction of inflammatory mRNAs in vivo is dominated by a penton base RGD capsid function that is enhanced by fiber-dependent localization or transduction functions.
Ad5-based vectors can induce hepatotoxicity when administered systemically at doses of 2 × 1011 particles/animal (35). To determine if our vectors cause liver injury at a dose of 1010 particles/mouse, we assayed for serum levels of alanine aminotransferase 24 h postinfection using a commercially available kit (Teco Diagnostics). Under these conditions, no significant elevation in serum alanine aminotransferase levels was observed with any of our vectors compared to mock-infected animals (data not shown).

Adaptive immune response to capsid-modified vectors.

Long-term transgene persistence following systemic administration of Adv in an immune-competent host is precluded by a well-characterized CTL response (42). To determine if vectors with RGD deleted were impaired in activation of cell-mediated immunity, we assayed liver extracts for CAT transgene persistence 28 days postinfection. With all vectors tested, liver CAT levels were not significantly elevated above mock infection (Fig. 4A). These data indicate that at 1010 particles/mouse, our vectors are capable of inducing a robust CTL-mediated clearance of infected cells, regardless of capsid modifications.
One of the major impediments and virtues of Adv is the induction of a strong humoral immune response against the virus and virally expressed antigen. To determine the influence of fiber and penton base manipulations on the humoral immune response, serum samples from 28-day-infected mice were used in an in vitro neutralization assay against F41T as described in Materials and Methods. When administered at a dose of 1010 particles/mouse, each virus induced detectable serum levels of anti-Ad neutralizing antibodies (Fig. 5A). A second and more biologically relevant method to characterize anti-Ad neutralizing antibodies involves an in vivo readministration assay. The efficiency of protective immunity is determined by the inhibition of gene transduction following a readministration of Adv. Animals were injected with 1010 particles of each virus or mock infected with buffer. On day 25 postinjection, when CAT expression should be near baseline (29), animals were boosted with 1010 particles of Ad5. Livers were harvested and CAT transgene levels were assayed either 24 h or 3 days postinfection. The 24-h time point was included because reactivation of memory CTL may result in clearance of infected cells before the 3-day end point. Regardless of the vector used in the original administration, all animals developed neutralizing humoral immunity capable of blocking transduction by the Ad5 booster (Fig. 5B).
Since the capsid-modified Adv used in this study were compromised at the level of gene transduction, we next asked if the decrease in transduction efficiency influenced the induction of antigen-specific (anti-CAT) humoral immunity. An anti-CAT ELISA was used to assess serum levels of anti-CAT antibody in each animal (Fig. 5C). The data indicate that of all vectors assayed, F41T primary infection resulted in the highest level of antigen-specific antibody. Surprisingly, Ad5 and Ad5Δ, the vectors that yield the highest level of hepatic CAT expression, were less effective than several of the detargeted constructs with respect to induction of antigen-specific antibody.
To determine if the detargeted vectors have different dosing thresholds for induction of antiviral/antigen humoral immunity, mice were infected with 109, 108, and 107 particles of F41TΔ and F7F41SΔ, with F41T as the control and booster vector. On day 25 postinfection, each animal received 1010 particles of F41T. Livers were harvested on day 28 and assayed for CAT reporter expression. At a dose of 109 particles/mouse, all vectors induced sufficient neutralizing antibody to block transduction by the F41T booster (Fig. 6A). At 108 particles/mouse, we observed low but detectable levels of CAT in livers from boosted mice and CAT expression was similar for each Adv tested. At the lowest dose tested, 107 particles/mouse, CAT expression was detected following each readministration. Mice receiving F41T had significantly higher levels of liver CAT and therefore lower levels of anti-Ad neutralizing antibody, than animals infected with F7F41SΔ and F41TΔ.
We next determined the level of anti-CAT antibody generated by the detargeted vectors and F41T following the readministration of F41T (Fig. 6B). The data indicate that a dose of 109 was able to induce anti-CAT antibody with F41T and to a lesser degree with F7F41SΔ and F41TΔ. At 108 or 107 particles/mouse, we were below the detection sensitivity of our anti-CAT ELISA for all constructs. The data presented in Fig. 6 demonstrate two distinct patterns of humoral immune response to modified vectors. The penton RGD-detargeted constructs effect a strong antiviral blocking immunity and a comparatively weak antigen-specific response. The F41T construct induced a weaker antiviral blocking immunity and the highest level of antigen-specific antibody.


The goal of this study was to characterize Ad vectors that have been compromised in their ability to transduce cells through the well-characterized two-step entry process mediated by fiber and penton base. Detargeting adenovirus is an important first step in developing a retargeting strategy to allow cell-specific targeting of viral vectors intended for anticancer or gene therapy applications. Additionally, since Adv are currently being considered as important vaccine vectors, the assays used to characterize the biology of fiber- and penton base-modified vectors contribute to development of enhanced vectors for vaccine applications.
Previous studies have demonstrated that fiber-pseudotyped vectors, F41T and F7F41S, alter Adv infectivity in a predictable manner when used to transduce cells in vitro (28, 29). F41T transduction occurs through CAR but not through the secondary functions associated with the KKTK motif of Ad5 fiber shaft. F7F41S lacks CAR or known murine fiber-dependent binding functions, and the low level of transduction in the FL83B hepatocyte cell line was attributed to secondary binding of penton base to membrane integrins. The F41TΔ and F7F41SΔ constructs with a deletion of the integrin binding RGD motif in a penton base combined with the fiber modifications were predicted to further compromise transduction compared to that of Ad5. In vitro the double-capsid mutant F7F41SΔ transduces murine hepatocytes with 500-fold less efficiency than Ad5. By our assays, F7F41SΔ is a successfully detargeted vector. In contrast, F41TΔ is a vector that is specifically targeted to the CAR receptor. Transduction efficiency in the endothelial cell line is consistent with the pattern found with hepatocytes, although overall transduction yield is diminished, presumably due to low levels of CAR receptor in endothelial cells. All vectors were similarly compromised when used to infect the RAW macrophage cell line with levels of CAT only slightly above background.
In contrast to transduction into the RAW cell line, when Adv was used to infect more permissive primary murine macrophages, the penton base RGD deletion vectors had reduced levels of CAT and GFP expression to near background levels (Fig. 2). Coincident with the diminished level of transduction, the ΔRGD constructs were less effective at inducing expression of inflammatory chemokines in BMMO. An unexpected observation coming from this study was that macrophage transduction and activation are not functionally coupled on a cell-per-cell basis. Vectors expressing RGD transduce only 5 to 7% of BMMO but activate approximately 50% of cells. Less than half of GFP-expressing cells upregulate CD86. The most straightforward explanation is that cells that are CD86 positive but not GFP positive have taken up sufficient virus to activate the cell, but not enough virus to allow detection of a low-level GFP expression. Several alternative explanations for these observations can be entertained as well. In the first, a small number of cells are productively transduced by Ad in an RGD-dependent manner and are subsequently activated. The pool of activated cells stimulates neighboring cells in a paracrine fashion, resulting in CD86 upregulation by a larger percentage of the total population. In a second model, macrophage uptake may be occurring in the majority of cells, but only a small fraction of vector is able to transmigrate to the nucleus to mediate gene expression and the remaining vector is shunted into a degradation pathway that may also serve to stimulate activation of the BMMO. Finally, activation may require only extracellular stimulation via the RGD motif, but not internalization of the vector. We are currently carrying out studies to test these possibilities.
Our data also indicate that activation of BMMO is not strictly RGD dependent (Fig. 2). We detected upregulation of CD86 and induction of inflammatory genes above background by the vectors with RGD deleted, suggesting at least one alternative pathway of activation. With the Ad5Δ and F41TΔ vectors, our results indicate that alternate activation pathways may be unrelated to penton RGD or fiber. The F7F41SΔ vector was only marginally less effective than F7F41S at stimulating BMMO. In the absence of favorable RGD-integrin interactions, nonspecific uptake of Adv into macrophages, by macropinocytosis, for example, may be sufficient to stimulate recognition of viral pathogen. Such a pathway would be predicted to be responsive to increasing concentrations of infecting virus.
Studies using fiber 5 mutant vectors, penton base RGD deletions, or a combination of these mutations have yielded a variety of results with respect to the relative contributions of fiber and penton binding to transduction in vivo (1, 15, 16, 21, 28, 31-33). Consistent with our predictions, the deletion of the RGD motif in F7F41SΔ had an obvious and dramatic influence on vector transduction in vivo (Fig. 4A). A 1,200-fold transduction decrease compared to Ad5 was observed; however, we were still able to detect low levels of CAT gene expression. These results are consistent with the in vitro BMMO studies where we were also able to identify very low levels of background transduction by the detargeted constructs. It is apparent that CAR-mediated transduction of hepatic tissue in vivo is overshadowed by penton RGD-mediated transduction (Fig. 4A, compare F41T to F41TΔ and F7F41S). Additionally, both penton RGD binding and CAR binding are rendered irrelevant by fiber 5 shaft/knob binding functions which are not present in fiber 41L. Fiber 41L is structurally very similar to F5 with respect to shaft length and CAR binding efficiency but lacks the KKTK motif presumably associated with HSG binding. It is also clear that in all constructs, the RGDΔ mutation results in a diminished level of chemokine activation following systemic administration of 1010 viral particles. Based on these observations, the tandem fiber constructs F41T and F7F41S when combined with the penton base RGDΔ mutation offer greatly reduced in vivo transduction and a diminished induction of the early antiviral inflammatory response.
The next issue addressed in the present study was how fiber- and penton base-modified vectors that possess distinct mechanisms for interacting with host cells and present a milder inflammatory response are managed by the adaptive arm of the immune system. Elimination of infected cells by cytotoxic T cells results in the loss of vector-mediated transgene expression over time (42). For each Adv, transgene expression was completely eliminated by 28 days postinfection, indicating an intact cellular response to each construct. Similarly, when serum from mice infected for 28 days was characterized for anti-Ad5 neutralizing antibody, or when each vector was characterized in a readministration assay, there was no indication that the fiber or penton base modifications influenced antiadenovirus humoral immunity. The threshold of viral recognition required for adaptive immune activation was reached by each vector at a dose of 1010 particles/animal. To confirm that the compromise in inflammatory activation by the RGDΔ constructs was irrelevant to the ability of the host to mount a strong antiviral response, a readministration threshold titration curve for F41T, F41Δ, and F7F41Δ was established. The data demonstrate that at a dose of 108 particles/animal, these constructs were equally effective at inducing blocking immunity. At the lowest dose, 107 particles/animal, the RGDΔ constructs induced a more robust anti-Ad immunity than F41T.
In contrast to the relatively uniform anti-Ad neutralizing response, the magnitude of antibody developed against the CAT transgene was influenced by capsid modifications. The most effective vector for vaccine considerations was F41T. Although transduction efficiency was less than Ad5 (Fig. 4A), it generated the highest anti-CAT titer (Fig. 5C). The most surprising observation was that F41TΔ and F7F41SΔ were as good as Ad5 at inducing CAT-specific antibodies (Fig. 5C). These constructs are compromised in gene transduction by 500- and 1,200-fold, respectively, and induce a greatly diminished inflammatory response compared to Ad5, yet still induce a potent antitransgene immunity. Based on the threshold titrations carried out, vector doses below 109 particles/animal are unproductive in generating anti-CAT antibody.
The efficiency of anti-CAT humoral immunity generated by the detargeted vectors was unexpected. Based on the reduced level of transduction into BMMO and the diminished inflammatory response induced by these vectors, we anticipated that they would evoke a compromised level of antitransgene humoral immunity compared to an unmodified Ad5 capsid vector. These data clearly indicate that the two-step fiber/penton base mechanism of Adv internalization can be eliminated without a discernible loss in humoral immunity. In this study, we have not elaborated on how these modification impact transgene-specific antigen activation of the CD8 T-cell pathway, but since transgene-expressing cells were efficiently eliminated based on CAT expression after 28 days, we anticipate the cellular response to the detargeted vectors has been left intact.
FIG. 1.
FIG. 1. Capsid-modified adenovirus vector characterization and transduction in vitro. (A) Vector genome configuration of the Ad5-based constructs used in this study. All vectors were generated by homologous recombination in 293 cells and express a CMV-driven CATiresGFP (CiG) cassette in the E1 region. In the fiber-modified vectors, fiber 5 was genetically replaced with alternate fibers, as indicated by the shaded boxes. Counterparts (with RGD deleted [Δ]) to each vector were also generated. Three potential interactions for Ad5 binding are depicted with plus signs: CAR, integrin, and HSG (via the KKTK motif) binding. Capsid-modifed vectors are devoid of one or more of these interactions (depicted with minus signs). Relative fiber content as determined by Western blotting and Coomassie staining is given in the right column. (B) Diminished in vitro transduction by capsid-modified vectors. Approximately 2 × 105 murine FL83B hepatocytes, MS1 endothelial cells, or RAW macrophages were infected with each vector at 1,000 particles/cell for 30 min. Twenty-four hours postinfection, cell lysates were harvested for CAT activity. Data are representative of the average CAT activity ± standard error of the mean from at least two experiments performed in triplicate.
FIG. 2.
FIG. 2. Activation and transduction of BMMO by modified vectors. BMMO were infected with 5,000 particles/cell or mock infected with viral storage buffer. At 36 h postinfection, cells were harvested and stained with anti-CD86 antibody. Flow cytometry was used to assess cell surface expression of CD86 (A) and transduction levels by GFP (A and B). The numbers in each quadrant represent the percentage of total cells present in the quadrant. In parallel, BMMO cell lysates were harvested for CAT activity (C). Data are representative of average CAT or GFP expression ± standard error of the mean from three experiments performed in triplicate.
FIG. 3.
FIG. 3. Ad-mediated induction of proinflammatory cytokines and chemokines in BMMO. A total of 4 × 106 BMMO were infected with 5,000 particles/cell and harvested 5 h postinfection for RNA isolation. Total RNA was analyzed for induction of chemokine (A) or cytokine (B) mRNAs by RNase protection assay as described in Materials and Methods. Each lane represents RNA from an individual preparation. Transcript levels were quantitated by PhosphorImager analysis and normalized to ribosomal protein L32 mRNA levels (C). Data are means of triplicates ± standard error of the mean. MIP-1β/1α, macrophage inflammatory protein; IP-10, interferon-gamma-inducible protein 10; TCA-3, T-cell activation gene 3; IL-6, interleukin-6; TGFβ1, transforming growth factor β1; MIF, macrophage-inducing factor; GAPDH, glyceraldehyde phosphate-3 dehydrogenase.
FIG. 4.
FIG. 4. Diminished in vivo transduction, vector uptake, and innate immune activation by capsid-modified vectors. Groups of five female B6129/J mice were injected with 1010 virus particles per mouse and sacrificed at the indicated time points. (A) Liver samples were assayed for total liver CAT activity as described in Materials and Methods. Data are means of five animals ± standard error of the mean. (B) (Lower) DNA was extracted from liver tissue and digested with EcoRI. Ten micrograms of digested DNA was separated on a 1% agarose gel and transferred to a nylon membrane, and blots were probed with either EcoRI-digested Ad5 vector DNA or Rad50. The band shown corresponds to the largest EcoRI fragment. (Upper) Ad DNA levels were quantitated by PhosphorImager analysis and normalized to cellular Rad50 levels. Each lane/bar represents DNA from an individual animal. (C) Total RNA was isolated and analyzed for chemokine mRNA levels by RNase protection assay. Each lane represents RNA from an individual animal (n = 4 for F7F41S and n = 5 for all others).
FIG. 5.
FIG. 5. Induction of anti-Ad neutralizing and antitransgene antibodies by capsid-modified vectors. (A) Anti-Ad neutralizing antibody titer assay. Serum levels of anti-Ad neutralizing antibody from mice infected with the indicated virus or mock infected were determined by a functional in vitro transduction blocking assay as described in Materials and Methods. Data are means of three (Ad5, Ad5Δ, and F7F41S) or five (F41T, F41TΔ, F7F41SΔ, and mock) mice per group. (B) Efficacy of neutralizing antibodies in blocking Ad5 vector readministration. Three groups of three mice were injected on day 0 with 1010 particles of the indicated virus (on the x axis) or vehicle buffer (mock). Twenty-five days postinjection, two of the three groups received a booster of 1010 particles of Ad5CiG. One group remained as a nonboost control (white bars) and was harvested on day 28. The two boosted groups were harvested either 24 h (gray bars) or 3days (white bars) postinfection, and livers were scored for CAT activity. Data are means ± standard error of the mean. (C) Anti-CAT ELISA. Serum samples from 28-day nonboosted mice were assayed for anti-CAT antibody levels by ELISA. Data are means of three (Ad5, Ad5Δ, and F7F41S) or five (F41T, F41TΔ, F7F41SΔ, and mock) mice per group.
FIG. 6.
FIG. 6. Threshold titration of Ad-mediated adaptive immune activation in vivo. (A) Neutralizing antibody dose-response titration. Groups of three mice were injected on day 0 with various doses (109, 108, or 107 particles [p]) of F41T, F41TΔ, or F7F41SΔ. Twenty-five days postinjection, mice received 1010 particles of F41T booster and were harvested 3 days later for liver CAT activity. One group of three mice received only vehicle buffer throughout the protocol as a negative control (mock). Another group received vehicle buffer in the primary administration but was boosted on day 25 with F41T and harvested on day 28 as a positive control for maximal 3-day liver CAT activity (boost). Data are means of three mice ± standard error of the mean. (B) Anti-CAT ELISA. Serum samples were harvested from mice on day 28 and assayed for anti-CAT antibody levels by ELISA. Data are means of three samples per group.


This work was supported by NIH grant AI-63142 to E.F.-P.
GenVec has licensing agreements with Cornell Research Foundation for technology developed in collaboration with E. Falck-Pedersen.


Alemany, R., and D. T. Curiel. 2001. CAR-binding ablation does not change biodistribution and toxicity of adenoviral vectors. Gene Ther.8:1347-1353.
Bergelson, J. M., J. A. Cunningham, G. Droguett, E. A. Kurt-Jones, A. Krithivas, J. S. Hong, M. S. Horwitz, R. L. Crowell, and R. W. Finberg. 1997. Isolation of a common receptor for Coxsackie B viruses and adenoviruses 2 and 5. Science275:1320-1323.
Bessis, N., F. J. GarciaCozar, and M. C. Boissier. 2004. Immune responses to gene therapy vectors: influence on vector function and effector mechanisms. Gene Ther.11(Suppl. 1):S10-S17.
Bewley, M. C., K. Springer, Y. B. Zhang, P. Freimuth, and J. M. Flanagan. 1999. Structural analysis of the mechanism of adenovirus binding to its human cellular receptor, CAR. Science286:1579-1583.
Carson, S. D., J. T. Hobbs, S. M. Tracy, and N. M. Chapman. 1999. Expression of the coxsackievirus and adenovirus receptor in cultured human umbilical vein endothelial cells: regulation in response to cell density. J. Virol.73:7077-7079.
Dechecchi, M. C., P. Melotti, A. Bonizzato, M. Santacatterina, M. Chilosi, and G. Cabrini. 2001. Heparan sulfate glycosaminoglycans are receptors sufficient to mediate the initial binding of adenovirus types 2 and 5. J. Virol.75:8772-8780.
Dmitriev, I., V. Krasnykh, C. R. Miller, M. Wang, E. Kashentseva, G. Mikheeva, N. Belousova, and D. T. Curiel. 1998. An adenovirus vector with genetically modified fibers demonstrates expanded tropism via utilization of a coxsackievirus and adenovirus receptor-independent cell entry mechanism. J. Virol.72:9706-9713.
Fechner, H., A. Haack, H. Wang, X. Wang, K. Eizema, M. Pauschinger, R. Schoemaker, R. Veghel, A. Houtsmuller, H. P. Schultheiss, J. Lamers, and W. Poller. 1999. Expression of coxsackie adenovirus receptor and alphav-integrin does not correlate with adenovector targeting in vivo indicating anatomical vector barriers. Gene Ther.6:1520-1535.
Gall, J., A. Kass-Eisler, L. Leinwand, and E. Falck-Pedersen. 1996. Adenovirus type 5 and 7 capsid chimera: fiber replacement alters receptor tropism without affecting primary immune neutralization epitopes. J. Virol.70:2116-2123.
Hong, S. S., L. Karayan, J. Tournier, D. T. Curiel, and P. A. Boulanger. 1997. Adenovirus type 5 fiber knob binds to MHC class I alpha2 domain at the surface of human epithelial and B lymphoblastoid cells. EMBO J.16:2294-2306.
Huang, S., R. I. Endo, and G. R. Nemerow. 1995. Upregulation of integrins αvβ3 and αvβ5 on human monocytes and T lymphocytes facilitates adenovirus-mediated gene delivery. J. Virol.69:2257-2263.
Huang, S., T. Kamata, Y. Takada, Z. M. Ruggeri, and G. R. Nemerow. 1996. Adenovirus interaction with distinct integrins mediates separate events in cell entry and gene delivery to hematopoietic cells. J. Virol.70:4502-4508.
Juillard, V., P. Villefroy, D. Godfrin, A. Pavirani, A. Venet, and J. G. Guillet. 1995. Long-term humoral and cellular immunity induced by a single immunization with replication-defective adenovirus recombinant vector. Eur. J. Immunol.25:3467-3473.
Kaner, R. J., S. Worgall, P. L. Leopold, E. Stolze, E. Milano, C. Hidaka, R. Ramalingam, N. R. Hackett, R. Singh, J. Bergelson, R. Finberg, E. Falck-Pedersen, and R. G. Crystal. 1999. Modification of the genetic program of human alveolar macrophages by adenovirus vectors in vitro is feasible but inefficient, limited in part by the low level of expression of the coxsackie/adenovirus receptor. Am. J. Respir. Cell Mol. Biol.20:361-370.
Koizumi, N., K. Kawabata, F. Sakurai, Y. Watanabe, T. Hayakawa, and H. Mizuguchi. 2006. Modified adenoviral vectors ablated for coxsackievirus-adenovirus receptor, alphav integrin, and heparan sulfate binding reduce in vivo tissue transduction and toxicity. Hum. Gene Ther.17:264-279.
Koizumi, N., H. Mizuguchi, F. Sakurai, T. Yamaguchi, Y. Watanabe, and T. Hayakawa. 2003. Reduction of natural adenovirus tropism to mouse liver by fiber-shaft exchange in combination with both CAR- and αv integrin-binding ablation. J. Virol.77:13062-13072.
Lieber, A., C.-Y. He, L. Meuse, D. Schowalter, I. Kirillova, B. Winther, and M. A. Kay. 1997. The role of Kupffer cell activation and viral gene expression in early liver toxicity after infusion of recombinant adenovirus vectors. J. Virol.71:8798-8807.
Liu, Q., L. R. White, S. A. Clark, D. J. Heffner, B. W. Winston, L. A. Tibbles, and D. A. Muruve. 2005. Akt/protein kinase B activation by adenovirus vectors contributes to NFκB-dependent CXCL10 expression. J. Virol.79:14507-14515.
Liu, Q., A. K. Zaiss, P. Colarusso, K. Patel, G. Haljan, T. J. Wickham, and D. A. Muruve. 2003. The role of capsid-endothelial interactions in the innate immune response to adenovirus vectors. Hum. Gene Ther.14:627-643.
Manickan, E., J. S. Smith, J. Tian, T. L. Eggerman, J. N. Lozier, J. Muller, and A. P. Byrnes. 2006. Rapid Kupffer cell death after intravenous injection of adenovirus vectors. Mol. Ther.13:108-117.
Martin, K., A. Brie, P. Saulnier, M. Perricaudet, P. Yeh, and E. Vigne. 2003. Simultaneous CAR- and alpha V integrin-binding ablation fails to reduce Ad5 liver tropism. Mol. Ther.8:485-494.
Muruve, D. A. 2004. The innate immune response to adenovirus vectors. Hum. Gene Ther.15:1157-1166.
Nakamura, T., K. Sato, and H. Hamada. 2003. Reduction of natural adenovirus tropism to the liver by both ablation of fiber-coxsackievirus and adenovirus receptor interaction and use of replaceable short fiber. J. Virol.77:2512-2521.
Ni, S., K. Bernt, A. Gaggar, Z. Y. Li, H. P. Kiem, and A. Lieber. 2005. Evaluation of biodistribution and safety of adenovirus vectors containing group B fibers after intravenous injection into baboons. Hum. Gene Ther.16:664-677.
Philpott, N. J., M. Nociari, K. B. Elkon, and E. Falck-Pedersen. 2004. Adenovirus-induced maturation of dendritic cells through a PI3 kinase-mediated TNF-alpha induction pathway. Proc. Natl. Acad. Sci. USA101:6200-6205.
Roelvink, P. W., A. Lizonova, J. G. M. Lee, Y. Li, J. M. Bergelson, R. W. Finberg, D. E. Brough, I. Kovesdi, and T. J. Wickham. 1998. The coxsackievirus-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.
Schiedner, G., W. Bloch, S. Hertel, M. Johnston, A. Molojavyi, V. Dries, G. Varga, N. Van Rooijen, and S. Kochanek. 2003. A hemodynamic response to intravenous adenovirus vector particles is caused by systemic Kupffer cell-mediated activation of endothelial cells. Hum. Gene Ther.14:1631-1641.
Schoggins, J. W., J. G. D. Gall, and E. Falck-Pedersen. 2003. Subgroup B and F fiber chimeras eliminate normal adenovirus type 5 vector transduction in vitro and in vivo. J. Virol.77:1039-1048.
Schoggins, J. W., M. Nociari, N. Philpott, and E. Falck-Pedersen. 2005. Influence of fiber detargeting on adenovirus-mediated innate and adaptive immune activation. J. Virol.79:11627-11637.
Shayakhmetov, D. M., A. Gaggar, S. Ni, Z.-Y. Li, and A. Lieber. 2005. Adenovirus binding to blood factors results in liver cell infection and hepatotoxicity. J. Virol.79:7478-7491.
Smith, T., N. Idamakanti, H. Kylefjord, M. Rollence, L. King, M. Kaloss, M. Kaleko, and S. C. Stevenson. 2002. In vivo hepatic adenoviral gene delivery occurs independently of the coxsackievirus-adenovirus receptor. Mol. Ther.5:770-779.
Smith, T. A., N. Idamakanti, J. Marshall-Neff, M. L. Rollence, P. Wright, M. Kaloss, L. King, C. Mech, L. Dinges, W. O. Iverson, A. D. Sherer, J. E. Markovits, R. M. Lyons, M. Kaleko, and S. C. Stevenson. 2003. Receptor interactions involved in adenoviral-mediated gene delivery after systemic administration in non-human primates. Hum. Gene Ther.14:1595-1604.
Smith, T. A., N. Idamakanti, M. L. Rollence, J. Marshall-Neff, J. Kim, K. Mulgrew, G. R. Nemerow, M. Kaleko, and S. C. Stevenson. 2003. Adenovirus serotype 5 fiber shaft influences in vivo gene transfer in mice. Hum. Gene Ther.14:777-787.
Tomko, R. P., R. Xu, and L. Philipson. 1997. HCAR and MCAR: the human and mouse cellular receptors for subgroup C adenoviruses and group B coxsackieviruses. Proc. Natl. Acad. Sci. USA94:3352-3356.
Varnavski, A. N., R. Calcedo, M. Bove, G. Gao, and J. M. Wilson. 2005. Evaluation of toxicity from high-dose systemic administration of recombinant adenovirus vector in vector-naive and pre-immunized mice. Gene Ther.12:427-436.
Vigne, E., I. Mahfouz, J.-F. Dedieu, A. Brie, M. Perricaudet, and 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.
Vincent, T., R. F. Pettersson, R. G. Crystal, and P. L. Leopold. 2004. Cytokine-mediated downregulation of coxsackievirus-adenovirus receptor in endothelial cells. J. Virol.78:8047-8058.
Von Seggern, D. J., S. Huang, S. K. Fleck, S. C. Stevenson, and G. R. Nemerow. 2000. Adenovirus vector pseudotyping in fiber-expressing cell lines: improved transduction of Epstein-Barr virus-transformed B cells. J. Virol.74:354-362.
Wickham, T. J., P. Mathias, D. A. Cheresh, and G. R. Nemerow. 1993. Integrins alpha v beta 3 and alpha v beta 5 promote adenovirus internalization but not virus attachment. Cell73:309-319.
Worgall, S., A. Busch, M. Rivara, D. Bonnyay, P. L. Leopold, R. Merritt, N. R. Hackett, P. W. Rovelink, J. T. Bruder, T. J. Wickham, I. Kovesdi, and R. G. Crystal. 2004. Modification to the capsid of the adenovirus vector that enhances dendritic cell infection and transgene-specific cellular immune responses. J. Virol.78:2572-2580.
Worgall, S., G. Wolff, E. Falck-Pedersen, and R. G. Crystal. 1997. Innate immune mechanisms dominate elimination of adenoviral vectors following in vivo administration. Hum. Gene Ther.8:37-44.
Yang, Y., and J. M. Wilson. 1995. Clearance of adenovirus-infected hepatocytes by MHC class I-restricted CD4+ CTLs in vivo. J. Immunol.155:2564-2570.
Zhang, Y., N. Chirmule, G. P. Gao, R. Qian, M. Croyle, B. Joshi, J. Tazelaar, and J. M. Wilson. 2001. Acute cytokine response to systemic adenoviral vectors in mice is mediated by dendritic cells and macrophages. Mol. Ther.3:697-707.

Information & Contributors


Published In

cover image Journal of Virology
Journal of Virology
Volume 80Number 211 November 2006
Pages: 10634 - 10644
PubMed: 16943295


Received: 27 June 2006
Accepted: 17 August 2006
Published online: 1 November 2006


Request permissions for this article.



John W. Schoggins
Weill Medical College of Cornell University, Hearst Research Foundation, Department of Microbiology and Immunology, Molecular Biology Graduate Program, New York, New York 10021
Erik Falck-Pedersen [email protected]
Weill Medical College of Cornell University, Hearst Research Foundation, Department of Microbiology and Immunology, Molecular Biology Graduate Program, New York, New York 10021

Metrics & Citations



  • For recently published articles, the TOTAL download count will appear as zero until a new month starts.
  • There is a 3- to 4-day delay in article usage, so article usage will not appear immediately after publication.
  • Citation counts come from the Crossref Cited by service.


If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. For an editable text file, please select Medlars format which will download as a .txt file. Simply select your manager software from the list below and click Download.

View Options

Figures and Media






Share the article link

Share with email

Email a colleague

Share on social media

American Society for Microbiology ("ASM") is committed to maintaining your confidence and trust with respect to the information we collect from you on websites owned and operated by ASM ("ASM Web Sites") and other sources. This Privacy Policy sets forth the information we collect about you, how we use this information and the choices you have about how we use such information.
FIND OUT MORE about the privacy policy