INTRODUCTION

Severe adverse events associated with high-dose systemic administration of adenoviral (Ad) vectors for monogenetic disorder treatment has placed greater emphasis on local delivery (e.g., intradermal or intramuscular) or ex vivo transduction of target cells for vaccine administration and immunotherapy (see refs. 1–4 for reviews). Ad vector immunomodulation requires efficient transduction of hematopoietic cells; yet, most blood or bone marrow-derived cells are refractory to Ad5-based vectors, owing to the absence of their primary attachment protein, coxsackie and adenovirus receptor (CAR).⁵ In particular, Ad5 transduction of freshly isolated human peripheral blood mononuclear cells (PBMCs) is quite poor.^{6, 7, 8} Dendritic cells (DCs) are potent antigen-presenting cells (APCs) and have been the main antigen delivery target via Ad-based immunotherapy. However, the lack of DC CAR expression has required Ad5 vector modifications for efficient transduction, such as pseudotyping with CD46-utilizing fibers⁹ or insertion of an Arg-Gly-Asp (RGD) sequence in the fiber knob (FK).¹⁰ While allowing DC transduction, they also have the unwanted side effect of broadening the range of cell types transduced.

Currently, there are 51 known Ads classified into six species, A through F, based on immunological and genetic criteria, with species B divided into two subspecies, B1 and B2. Generally, certain Ad species exhibit specific tissue tropism, such as in the eye (D), respiratory tract (B1, C, and E), gastrointestinal tract (A, D, and F), and kidneys and urinary tract (B2). Tropism is mainly due to the different Ad fiber proteins that mediate virus binding to specific receptors broadly distributed on specific cell types. All Ad fiber proteins form trimeric structures that share a common arrangement of an N-terminal tail that anchors the fiber to the capsid, a shaft of varying length and flexibility, and a C-terminal globular knob region, the primary site of receptor binding. Subsequent interactions between _v integrins and viral penton base RGD motifs trigger cellular signals for particle internalization via clathrin-coated pits and endosomes.¹¹

Several Ad attachment receptors have been identified, including CAR, CD46, CD80, CD86, 2 integrins, sialic acid, and heparan sulfate glycosaminoglycans (see ref. 12 for a review). CAR is the most extensively studied, due in part to its usage by Ads of different species, including A (Ad12), C (Ad2 and Ad5), E (Ad4), and F (Ad41 long fiber). CD46 has been shown to be a receptor for most species B Ads.^{13, 14, 15} We have shown that species D Ad37 also uses CD46 as an attachment receptor.¹⁶ Ad37 has also been reported to use sialic acid for binding.^{17, 18, 19}

To investigate Ad binding and transduction of different hematopoietic cell types, we generated soluble Ad FK domains for use in competition studies with green fluorescent protein (GFP) transgene-containing Ads with either the endogenous Ad5 fiber (Ad5.F5), or genetically engineered to express fibers from serotype 16 (Ad5.F16), or 37 (Ad5.F37). Our studies revealed an unanticipated mode of virus interaction with human myeloid cells that could foster the development of improved methods of Ad-mediated gene delivery to specific human blood cells.

RESULTS

37FK enhances Ad transduction

CD46-utilizing Ads such as Ad37 have an increased capacity to transduce human PBMCs;^{6, 20, 21} however, the precise receptors (i.e., sialic acid or CD46) used by Ad37 on these cells have not been defined. Studies were therefore undertaken to examine the mode of Ad transduction of a CAR-negative, human monocytic cell line, U-937, in the presence of different soluble Ad FKs (Figure 1a). Quite remarkably, the Ad37 FK (37FK) caused a significant enhancement of Ad5.F5, Ad5.F16, and Ad5.F37 transduction. In contrast, the FK from Ad16 did not exhibit this enhancing effect, but as anticipated inhibited Ad5.F16 transduction. The increased transduction by 37FK was dose dependent and independent of the multiplicity of infection (MOI) (Figure 1b). Interestingly, the extent of virus transduction was significantly greater following incubation on ice than when incubated at 37°C (Figure 1c), suggesting that 37FK promotes Ad binding to cells. We performed further studies to characterize the specific cell types that support 37FK-enhanced gene delivery and the mechanisms involved.

Figure 1.

Enhanced Ad transduction of human myeloid cells. (a) 37FK enhances cell transduction by Ad5 vectors displaying distinct fiber proteins. U-937 cells were incubated with buffer (white bars) or 50 g/ml of 16FK (gray bars) or 37FK (black bars) and 1,000 Ad5.F5, 100 Ad5.F16, or 200 Ad5.F37 p/cell for 1 h on ice. Cells were washed and then incubated for 16–20 h at 37°C before assaying for GFP expression by flow cytometry as described under Materials and Methods. (b) Enhancement of 37FK-mediated U-937 transduction is dose dependent. U-937 cells were incubated with buffer or with increasing amounts of 37FK and 100, 300, or 1,000 Ad5.F5 p/cell, and transduction was assayed as described above. (c) Effect of incubation temperature on transduction efficiency. U-937 cells were treated with buffer or increasing amounts of 37FK and 1,000 p/cell of Ad5.F5 for 1 h at 37°C (white bars) or on ice (black bars). Cells were then washed and assayed for transduction as above. All data shown are the mean percentages of transduced cellsSD (n=3).

Full figure and legend (68K)

37FK-enhanced gene delivery is restricted to myeloid cells

To determine if the increased transduction also extends to primary cells, we obtained monocytes from normal human donors by magnetic-activated cell sorter (MACS) immunodepletion and examined their susceptibility to Ad transduction with 37FK (Figure 2a). Whereas CAR-deficient monocytes were non-permissive to Ad5.F5 transduction, 33% of the monocytes were transduced in the presence of 37FK. Similarly, the percentage of cells transduced by Ad5.F16 increased from 29 to 60% in the presence of 37FK. Moreover, the geometric mean fluorescence intensity of the Ad5.F16-transduced cells increased over 5-fold (from 190 to 1,046) in 37FK-treated cells, indicating a significantly higher level of transgene expression in the transduced cells (data not shown). As anticipated from previous reports,^{14, 15} the 16FK at equal amounts to the 37FK was able to strongly inhibit Ad5.F16 transduction of monocytes but had no effect on Ad5.F5 transduction. In striking contrast to the monocytes, lymphocytes were refractory to Ad transduction by either Ad5.F5 or Ad5.F16 at the same MOI used for the monocytes, and the addition of 37FK did not alter transduction of these cells (Figure 2b). At higher doses of Ad5.F16 that allowed up to 25% transduction of the untreated lymphocytes, 37FK did not enhance either the percentage of cells transduced or the geometric mean fluorescence intensity (data not shown).

Figure 2.

37FK enhances Ad uptake in myeloid, but not lymphoid cells. (a, b) Human monocytes are preferentially susceptible to 37FK-enhanced Ad transduction. MACS isolation was used to negatively and positively select monocytes and lymphocytes, respectively, from human PBMCs. Cells were incubated with buffer (white bars), or 50 g/ml of 16FK (gray bars) or 37FK (black bars) and 1,000 p/cell of Ad5.F5 or 100 p/cell of Ad5.F16 for 1 h on ice, and then assayed for transduction as described in Figure 1. Contaminating cell types in each subpopulation were excluded from the analyses by appropriate gating based on forward and side scatter parameters. (c, d) 37FK does not enhance Ad transduction of human B- and T-cell lines. The HSB-2 human T lymphoblastoid and JR2 human B lymphoblastoid cell lines were incubated with buffer or FKs and Ads and assayed for transduction as described above. All data shown are mean percentages of transduced cellsSD of triplicate samples of one experiment and are representative of at least three experiments similarly performed.

Full figure and legend (62K)

We also tested the HSB-2 T-cell leukemia and JR2 B-cell lymphoma human cell lines for their susceptibility to 37FK-enhanced virus transduction (Figure 2c and d). Neither 16FK nor 37FK enhanced the percentage of cells transduced by either virus, nor did they enhance the geometric mean fluorescence intensity (not shown). Taken together, these results indicate that the 37FK enhancement of Ad transduction is limited to blood cells of myeloid cell lineage.

Susceptibility of monocyte-derived DCs to Ad transduction

Ad vectors have proved useful for the delivery of foreign antigens to DCs for cell-based immunotherapy, particularly for the treatment of cancer (reviewed in ref. 22); however, the absence of CAR on the surface of these cells²³ has hindered this approach. We therefore sought to determine whether 37FK could also improve Ad5-mediated gene transfer to both immature and mature monocyte-derived DCs (MDDCs) (Figure 3). Human monocytes were initially treated with granulocyte–macrophage colony-stimulating factor and interleukin-4 (IL-4) to generate immature MDDCs and then differentiated to maturity by the addition of tumor necrosis factor-. Flow cytometric analysis confirmed monocyte differentiation to immature MDDCs as indicated by decreases in both CD14 and CD86 expression, whereas MDDC maturation was indicated by substantial increases in CD80, CD83, and CD86 expression (Figure 3a). At an Ad5.F5 MOI of 1,000, the number of transduced immature MDDCs reached 78% in the presence of 37FK compared to no transduction with the buffer control or 16FK. Similarly, Ad5.F16 transduction increased from 2 to 51% at an MOI of 100. Transduction of mature MDDCs also increased, albeit to a lower level (44%), with Ad5.F5 and 37FK. Ad5.F16, which had a basal transduction level of 18%, increased to 43% with 37FK. Thus, 37FK not only has the capacity to promote gene delivery to freshly isolated monocytic cells but also to both immature and mature MDDCs.

Figure 3.

Human MDDCs are efficiently transduced by Ads in the presence of 37FK. Human monocytes were isolated by MACS negative selection and cultured in the presence of IL-4 and granulocyte–macrophage colony-stimulating factor for 5 days to generate immature MDDCs and an additional 3 days with tumor necrosis factor- to induce maturation. (a) Surface expression of monocyte and DC markers on monocytes (white bars), immature MDDCs (gray bars), and mature MDDCs (black bars). Cells were incubated with phycoerythrin-conjugated control immunoglobulin G antibody or antibodies against the indicated proteins. Cells were assayed by flow cytometry and the geometric mean fluorescence intensity of the viable cell populationSD of triplicate samples is shown. (b) Immature and (c) mature MDDCs were incubated with buffer (white bars), or 50 g/ml of 16FK (gray bars) or 37FK (black bars) and 1,000 p/cell of Ad5.F5 or 100 p/cell of Ad5.F16 for 1 h on ice, and then assayed for transduction as described in Figure 1. Data shown are mean percentages of transduced cellsSD of triplicate samples of one experiment and are representative of at least three experiments similarly performed.

Full figure and legend (72K)

Enhanced gene delivery by 37FK involves interactions with sialic acid

In addition to CD46,¹⁶ sialic acid has also been reported to mediate initial interactions of Ad37 with the cell surface.^{17, 18, 24} Binding of wild-type Ad37 and of Ad37F, a pseudotyped Ad5 vector, was reported to be inhibited by sialic acid cleavage using neuraminidase. Furthermore, Maackia amurensis agglutinin (MAA), a lectin that binds specifically to sialic acid residues in an (2–3)-linked glycosidic bond (2,3-SA), was reported to inhibit Ad37 binding to cells more effectively than Sambucus nigra agglutinin (SNA), which binds to (2–6)-linked sialic acid (2,6-SA), indicating a preference of Ad37 particles for sialic acid residues in one particular glycosidic linkage. To determine if 37FK-mediated Ad transduction of monocytic cells also involves sialic acid, we treated monocytic cells with neuraminidase (Figure 4). Removal of sialic acid dose-dependently decreased 37FK-mediated Ad transduction. In contrast, the enzyme did not inhibit transduction by CD46-utilizing Ad5.F16. In fact we observed slightly enhanced transduction, similar to that seen with Ad5 infection of neuraminidase-treated A549 cells.¹⁸ This finding suggests that the 37FK enhances Ad-mediated gene delivery via direct interactions with cell-associated sialic acid residues. To further confirm this, we performed Ad transductions in the presence of soluble lectins with sialic acid specificity (Figure 5). Somewhat surprisingly, we observed that the 2,6-SA binding SNA strongly and dose-dependently inhibited 37FK-mediated Ad gene delivery, whereas neither the 2,3-SA binding MAA nor control lectin from Pisum sativum that binds terminal -D-mannosyl residues inhibited transduction. Wheat germ agglutinin, which binds to sialic acid residues irrespective of glycosidic linkage, also reversed the enhanced transduction by 37FK, albeit at higher concentrations.

Figure 4.

Neuraminidase treatment of cells to remove sialic acid inhibits 37FK-enhanced Ad transduction. U-937 cells were treated with the indicated concentrations of neuraminidase. After 30 min at 37°C, cells were put on ice and incubated for 1 h with 50 g/ml of 37FK and 1,000 p/cell of Ad5.F5 (), or with buffer and 100 p/cell of Ad5.F16 (). Cells were then washed, incubated overnight at 37°C, and assayed for transduction as described in Figure 1. Data shown are mean percentages of transduced cellsSD of triplicate samples of one experiment and are representative of at least three experiments similarly performed.

Full figure and legend (10K)

Figure 5.

Lectins that bind (2–6)-linked sialic acids inhibit 37FK-mediated transduction. U-937 cells were treated with buffer or increasing doses of lectin agglutinins from Pisum sativum (), MAA (), SNA (), or Wheat germ agglutinin (). After 30 min on ice, 50 g/ml 37FK was added with 1,000 p/cell of Ad5.F5 and further incubated for 1 h on ice. Cells were washed and incubated overnight at 37°C before assaying for cell transduction as described in Figure 1. Data shown are mean percentages of transduced cellsSD of triplicate samples of one experiment and are representative of at least three experiments similarly performed.

Full figure and legend (11K)

37FK binds to Ad5.F5 particles via electrostatic interactions

One question that arose from the above studies was how 37FK association with sialic acid promotes Ad gene delivery. As CAR-deficient myeloid cells do not support CAR-dependent Ad5.F5 attachment, one possible explanation is that the 37FK acts as a bridge for the indirect association of virions with cell-associated sialic acid residues. To address this, we incubated FKs with virus particles, banded the samples via ultracentrifugation in Histodenz step gradients, and then analyzed the isolated virus for the presence of bound FK by immunoblotting (Figure 6a). Although a small amount of 37FK was detected at the gradient interface in the absence of virus, a significantly greater amount was detected upon co-incubation with Ad5.F5 (compare lanes 6 and 9), indicating that the 37FK is capable of binding directly to virus. In contrast, neither FK from Ad5 (5FK) nor 16FK was able to bind to the virus particles (lanes 7 and 8), demonstrating the specificity of the 37FK association. To investigate the nature of the interaction between 37FK and virions, increasing amounts of NaCl were added to the reaction mixtures (Figure 6b). A substantial inhibition of binding occurred when the salt concentration was increased to 250 mM and above (lanes 3–6), consistent with an electrostatic interaction occurring between the highly positively charged 37FK and the negatively charged hexon proteins of the viral capsids.²⁵

Figure 6.

37FK interacts directly with Ad virions in a charge dependent manner. (a) 37FK binds to Ad5.F5 particles. Forty micrograms of 5FK, 16FK, or 37FK were mixed with buffer alone (lanes 4–6) or with 40 g of Ad5.F5 (lanes 7–9) in complete RPMI medium for 1 h on ice and then ultracentrifuged through 40% Histodenz onto an 80% Histodenz cushion. Samples were taken from the interface, mixed with SDS sample buffer, and separated on a 4–20% SDS–polyacrylamide gel electrophoresis gel. After transfer to nitrocellulose, the blot was probed with an anti-His-tag antibody (top) and the proteins detected by enhanced chemiluminescence. The blot was then stripped, and re-probed with the 4D2 anti-Ad fiber mAb (bottom). As an internal reference control, 125 ng of each FK was loaded on the gel (lanes 1–3). Molecular weight marker positions are indicated on the right. (b) The 37FK–virus interaction is charge dependent. Forty micrograms of 37FK were mixed with buffer (lane 7) or 40 g of Ad5.F5 in the presence of increasing amounts of NaCl (lanes 2–6) and treated as above in a. Fifty nanograms of 37FK were loaded on the gel as a control (lane 1). Molecular weight marker positions are indicated on the right.

Full figure and legend (122K)

37FK promotes virus–cell association

To determine if binding of 37FK to virus particles leads to increased cell association, ¹²⁵I-labled Ad5.F5 particles were incubated with monocytic cells in the presence of increasing amounts of 16FK or 37FK (Figure 7). 37FK enhanced Ad5.F5 association with monocytic cells in a dose-dependent manner. At the highest dose of 37FK, we observed a 30-fold increase in virus association. This was substantially reduced by a 100-fold excess of unlabeled virus, demonstrating the specificity of the virus–cell interaction. No enhanced cell association was seen upon addition of 16FK at any dose. Together, these results indicate that 37FK enhances both binding and entry of Ads into myeloid cells and that the increased transduction is not merely due to enhanced GFP expression at the transcriptional or translational level. Thus, we conclude from these investigations that the 37FK is capable of mediating the indirect association of Ad5 particles with sialic acid residues on monocytic cells, consequently greatly enhancing viral gene delivery.

Figure 7.

37FK enhances Ad5.F5 association with cells. For each sample, 2 10⁵ U-937 cells were washed with ice-cold complete RPMI medium and mixed with buffer or increasing amounts of 16FK or 37FK. ¹²⁵I-Ad5.F5 (500 p/cell) was added in the presence (+) or absence of a 100-fold excess of unlabeled Ad5.F5. After incubation for 1 h at 37°C, cells were centrifuged through a cushion of silicone and mineral oil to separate free virus from the cell-associated virus. Both cell pellets and supernatants were counted on a gamma counter. Data shown are the mean percentages of cell-associated CPM calculated by dividing the CPM of the cell pellet by the sum of the CPM of the pellet and supernatant and multiplying by 100. Error bars depict SDs of triplicate samples. One experiment representative of four is shown.

Full figure and legend (11K)

DISCUSSION

Both the innate and adaptive immune responses to Ad vectors have frustrated their clinical use as gene replacement vectors. In contrast, these host immune responses make Ads ideal vehicles for antigen delivery in vaccine development; however, the lack of CAR on APCs remains an impediment to efficient gene delivery to these cells. In previous studies, efficient transduction of APCs required fiber protein modifications of Ad5-based vectors. Although these modifications do allow for transduction of human DCs, they also have the disadvantage of broadening the range of cells transduced by these vectors.

Here, we demonstrate that both human monocytes and MDDCs can be effectively and specifically transduced with Ad vectors when used in combination with 37FK. This enhancement occurs irrespective of attachment receptor usage, as transduction by both CD46- and sialic acid-utilizing viruses is improved. More importantly, 37FK facilitates efficient transduction by a CAR-utilizing Ad5 vector specifically in cells that do not express CAR and are thus ordinarily poorly transduced by this virus. Interestingly, the enhanced gene transfer is limited to blood cells of myeloid origin, as CAR-negative B and T cells and cell lines do not exhibit this effect.

It is well known that APC are exquisitely sensitive to activation by bacterial lipopolysaccharide. Therefore, we considered the possibility that the enhanced transduction induced by 37FK was due to an alternate signaling response arising from residual endotoxin present in the bacterially produced recombinant 37FK. However, this was largely excluded by several criteria including the fact that the identically produced 16FK did not have this enhancing effect (Figures 1a, 2a, 3b, and c). Furthermore, the lipopolysaccharide-inhibiting peptide polymyxin B had no effect on the 37FK-mediated enhancement of Ad transduction, and addition of up to 10 g/ml of bacterial lipopolysaccharide to the cells did not significantly enhance Ad5.F5 or Ad5.F16 transduction (data not shown). Thus, we considered other modes of 37FK enhancement of gene transfer.

Significant insights into the mechanism were gained by treating cells with neuraminidase (Figure 4) or by addition of sialic acid-binding lectins (Figure 5) that reduced the 37FK-enhanced Ad gene transfer. Furthermore, we demonstrated that 37FK can bind directly to Ad5 particles to mediate enhanced association of viral particles with myeloid cells (Figures 6 and 7). This suggests that the 37FK may be acting as a bridge between virus and cell, essentially pseudotyping the Ad5-based virus with an Ad37 fiber. The interaction between FK and virus particle likely occurs through an electrostatic interaction between the highly positively charged 37FK (pI 9.39) and the negatively charged Ad5 hexon protein (pI 5.15).²⁵ This is supported by the inhibition of this interaction with increasing amounts of NaCl (Figure 6b). Formation of the ternary complex of virus, 37FK, and cell is further supported by the finding that co-incubation of the mixture on ice allows for even greater transduction than if the interactions are allowed to proceed at 37°C (Figure 1c). The most likely explanation for this observation is that 37FK bound to the cell surface is quickly internalized, thereby sequestering both itself and available cell surface receptors from interactions with virus. In contrast, when the interactions are first allowed to proceed on ice and internalization cannot occur, sufficient time ensues to form the ternary complex before internalization is allowed to proceed at 37°C. In further support of this hypothesis, we have performed experiments pre-incubating 37FK with virus alone, and then adding the complexes to warm cells. The pre-incubation, whether on ice or at 37°C, greatly enhances virus transduction compared to adding all components at the same time (data not shown), indicating that complex formation between virus and 37FK is necessary.

The enhanced transduction induced by 37FK is sensitive to inhibition by the 2,6-SA binding SNA lectin, rather than the 2,3-SA binding MAA (Figure 5). This is in contrast to previous findings that both Ad37 virus attachment^{18, 24} and transduction¹⁷ are preferentially mediated through sialic acid with (2–3)-glycosidic linkages. However the crystal structure of the 37FK in complex with sialyl-lactose with both linkages suggests that sialic acid binding can occur irrespective of glycosidic linkage owing to the position of the adjoining galactose residues pointing away from the surface of the FK.¹⁹ The switch to using (2–6)- rather than (2–3)-glycosidic linkages by the 37FK may contribute to why this protein enhances, rather than inhibits, Ad37 transduction (Figure 1a).

We have previously demonstrated that CD46 is utilized as a receptor for Ad37.¹⁶ Protease treatment of cells has also implicated a protein component in sialic acid-mediated Ad37 binding,^{17, 18} but thus far no other cell surface glycoproteins have been identified as an Ad37 receptor. Human lymphocytes express CD46, as demonstrated by their ability to be infected by Ad5.F16 at high MOI.⁶ However, they do not respond to 37FK-mediated enhanced transduction (Figure 2b). It is possible that CD46 expression per se is not sufficient to mediate this response, or that another sialic acid-containing surface protein selectively expressed on monocytic cells mediates enhancement. Alternatively, it is possible that lymphoid cells do not have the molecular machinery to respond with enhanced uptake when engaging CD46 as a receptor.

One of the interesting implications of our findings is that 37FK-enhanced Ad transduction may reveal a mechanism by which wild-type Ad37 increases its spread in a normal infection. During replication, both Ad2 and Ad5 generate fiber proteins in excess of what is incorporated into virions.^{26, 27, 28} Walters et al.²⁸ proposed that this excess Ad5 fiber is used by the virus to disrupt CAR dimers that mediate epithelial cell adhesions at tight junctions at basolateral surfaces to allow virus escape and spread to the environment. The affinity of 37FK for sialyl-lactose is quite low, with an estimated K_d of 5 mM as determined by Burmeister et al.¹⁹ It is possible that excess free fiber proteins present during an Ad37 infection can interact with de novo viral particles to increase the avidity for cell surface sialic acid residues beyond what is possible with the 12 virus-associated fibers. The pI of the Ad37 hexon protein is 5.15 and, like the Ad5 capsid, has a net negative charge at physiological pH. This electrostatic interaction may explain why 37FK can also bind to wild-type Ad37 virions in the co-incubation and Histodenz sedimentation assay (data not shown). However, it remains to be established whether this mode of enhanced entry exists for Ad37 particles in vivo.

Our findings may also have practical implications for improving gene transfer to specific blood cell types.^{29, 30, 31} In particular, targeting monocytes and DCs is advantageous in that this may be accomplished in a mixed population of cells during ex vivo transduction of PBMCs. Using 37FK, we demonstrate that a much larger proportion of freshly isolated monocytes and MDDCs can be transduced with Ad5-based vectors, compared to virus alone, without further manipulation of existing viral vectors and minimal ex vivo processing of human leukocytes.

MATERIALS AND METHODS

Reagents and cells. Except where noted, tissue culture reagents were obtained from Invitrogen (Carlsbad, CA), and all others were purchased from Sigma Chemical Company (St Louis, MO). The U-937 human histiocytic lymphoma, and HSB-2 human T lymphoblastoid cell lines were from the American Type Culture Collection (Manassas, VA), and maintained in Rosewell Park Memorial Institute medium (RPMI) 1640 supplemented with 10% fetal calf serum (Omega Scientific, Tarzana, CA), 10 mM N-2-hydroxylpiperazine-N'-2-ethanesulfonic acid (pH 7.55), 4 mM L-glutamine, 100 U/ml penicillin, and 100 g/ml streptomycin. JR2, a B lymphoblastoid cell line developed by immortalizing freshly isolated human PBMCs with Epstein–Barr virus,³² were cultured in the same medium. Human 293 cells were also from American Type Culture Collection and propagated in Dulbecco's modified Eagle's medium supplemented as above, with the addition of 1 mM sodium pyruvate and 0.1 mM non-essential amino acids. Human PBMCs were isolated from heparin-treated blood as described previously.⁶ Monocytes were negatively selected from the lymphocytes by two rounds of MACS magnetic cell sorting using the human Monocyte Isolation Kit II (Miltenyi Biotec, Auburn, CA) according to the manufacturer's instructions. MDDCs were generated by culturing the purified monocytes at 1 10⁶ cells/ml in RPMI medium containing 100 ng/ml recombinant human IL-4 and 50 ng/ml recombinant human granulocyte–macrophage colony-stimulating factor (R&D Systems, Minneapolis, MN). Cells were changed to medium with fresh cytokines on day 2 or 3. On day 5, the cells were changed to medium containing 100 ng/ml recombinant human tumor necrosis factor- (BD Biosciences, San Jose, CA), 100 ng/ml IL-4, and 50 ng/ml granulocyte–macrophage colony-stimulating factor, and cultured for an additional 2–3 days. To detect cell surface antigens, 5 10⁵ cells were washed with an ice-cold fluorescence-activated cell sorter (FACS) buffer of 20 mM Tris-buffered saline, 0.2% sodium azide, and 0.2% bovine serum albumin, and mixed with 5 l of phycoerythrin-conjugated monoclonal antibodies (mAbs) against human CD3, CD14, CD19, CD80, CD83, or CD86, or with isotype-matched control antibody (eBioscience, San Diego, CA) for 30 min on ice. After washing, 10,000 cells were acquired on a FACScan, FACSort, or FACSCalibur flow cytometer and analyzed using CellQuest software (BD Biosciences). Only viable cells, based on forward and side scatter parameters, were used in the analyses.

Adenovirus preparation. The replication-defective Ad vectors used in these studies were Ad5-based, E1/E3-deleted vectors containing a cytomegalovirus promoter-driven GFP reporter gene cassette, and expressing either the endogenous Ad5 fiber (Ad5.F5), or genetically engineered to express fibers from serotype 16 (Ad5.F16), or 37 (Ad5.F37) in the virus backbone.¹⁶ Ads were propagated in 293 cells as described previously.¹⁶ The viral protein concentration was determined by the Bio-Rad Protein Assay (Bio-Rad, Richmond, CA) with a bovine serum albumin standard, and used to calculate the viral particle concentration (1 g=4 10⁹ virions).

Recombinant FK preparation. The Ad5 FK (5FK) protein containing an N-terminal hexahistidine (6 His)-Tag was constructed as described previously.³³ Recombinant Ad16 FK (16FK) containing N-terminal 6 His- and T7-Tags were generated as follows. Ad16 fiber DNA was polymerase chain reaction amplified from plasmid pDV156³⁴ using the primers 5'-CGCGGATCCGACTCTTCCAATGCTATCAC-3' (BamHI site underlined) and 5'-TGGTGCGGCCGCTCAGTCATCTTCTCTG-3' (NotI site underlined) and the Expand High Fidelity polymerase chain reaction System (Roche Applied Science, Indianapolis, IN). The amplified DNA fragment, encoding residues 151–353 of the Ad16 fiber protein, was cloned into pCR2.1-TOPO using the TA-Cloning Kit (Invitrogen). The resulting plasmid was digested with BamHI and NotI and the 16FK fragment was subcloned into the bacterial expression vector pET-28a(+) (Novagen, Madison, WI) in frame to add a 6 His-Tag, a 6 amino-acid thrombin cleavage site, and an 11 amino-acid T7-Tag to the N-terminus. Similar to above, a 37FK-containing DNA fragment was subcloned from a previously constructed 37FK plasmid³⁵ into the pET-28a(+) vector. Both FK plasmids were sequenced to verify fidelity, and used to transform BL21(DE3)pLysS cells (Invitrogen). Protein expression was induced with 1 mM isopropyl -D-thiogalactopyranoside for 4 h at 37°C, the bacteria were lysed with BugBuster Protein Extraction Reagent (Novagen), and FKs were purified using the TALON Metal Affinity Resin (BD Biosciences) as recommended by the manufacturer. Protein purity was determined by sodium dodecyl sulfate (SDS)–polyacrylamide gel electrophoresis and SimplyBlue staining (Invitrogen), or by immunoblotting using the T7-Tag antibody horseradish peroxidase conjugate or His-Tag mAb (Novagen). FKs were dialyzed against N-2-hydroxylpiperazine-N'-2-ethanesulfonic acid-buffered saline (pH 7.4), concentrated to 1–2 mg/ml using Centricon filter devices with YM-10 membranes (Millipore, Billerica, MA), and stored at -80°C. Protein concentrations were determined by spectrophotometric analysis at 280 nm using the following conversions: 0.72 mg/ml/A₂₈₀ for 5FK, 1.01 mg/ml/A₂₈₀ for 16FK, and 0.66 mg/ml/A₂₈₀ for 37FK. Conversion factors were based on the amino-acid compositions of the tagged proteins, as well as charge characteristics and pI calculations were made using the protein analysis software of Vector NTI version 9.0 (Invitrogen).

Virus transduction assays. A total of 2 10⁵ cells were plated in 96-well tissue culture plates. Buffer, FKs, and viruses were diluted in RPMI medium and added to the concentrations indicated in the Figure legends to a final volume of 200 l/well. After incubation for 1 h on ice, cells were washed three times, and incubated at 37°C for 16–20 h. The cells were washed and resuspended in FACS buffer, then analyzed for GFP expression by flow cytometry as described above. For consistency, gates to determine positive populations were set to include 1% of the negative control. Specific analyses of MACS sorted monocytes and lymphocytes were performed by appropriate gating based on forward and side scatter parameters. For lectin treatments, agglutinins from MAA, Pisum sativum, SNA, or Triticum vulgaris (Wheat germ agglutinin) were dissolved in N-2-hydroxylpiperazine-N'-2-ethanesulfonic acid-buffered saline, diluted in ice cold medium, added to the cells immediately before the addition of FKs and virus, then incubated for 1 h on ice. Cells were washed three times with cold medium, and then incubated overnight at 37°C before flow cytometric analysis. To cleave sialic acid residues, cells were treated for 30 min at 37°C with the indicated concentrations of neuraminidase from Vibrio cholerae (Roche Applied Science) diluted in medium. Virus was then added in the presence or absence of FKs and incubated for an additional 1 h on ice. The cells were washed three times with medium, and then incubated overnight at 37°C before analyzing for virus GFP transduction by flow cytometry.

Binding assays. To label Ad particles, 50 g of Ad5.F5 was mixed with 1 mCi of Na¹²⁵I (MP Biomedicals, Irvine, CA) diluted in 100 l of PBS (pH 7.4), containing one Iodo-Beads Iodinating Reagent bead (Pierce Biotechnology, Rockford, IL) and incubated for 8 min at room temperature. Labeled virus was separated from free iodine using a 2 ml Zeba desalt spin column according to the manufacturer's instructions (Pierce Biotechnology). For the binding and internalization assays, triplicates of 2 10⁵ U-937 cells were washed with ice-cold medium, and mixed with the indicated amounts of FKs, and ¹²⁵I-Ad5.F5 (500 particles/cell) in the presence or absence of 100-fold excess unlabeled Ad5.F5 in a final volume of 200 l. After incubation for 1 h at 37°C, cells were centrifuged through a 200 l cushion of 86% silicone oil and 14% mineral oil for 5 min at 13 000 g. The tip of the tube containing the cell pellet was cutoff, and both the tip and the remaining tube with the supernatant were counted separately in a gamma counter. The percentage of input CPM that was cell associated was calculated by dividing the CPM of the cell pellet by the sum of the CPM of the pellet and supernatant and multiplying by 100.

To measure Ad interactions with FKs, 40 g of 5FK, 16FK, or 37FK were mixed with 40 g of Ad5.F5, with or without increasing amounts of NaCl in complete RPMI medium in a final volume of 200 l. After incubation for 1 h on ice, the mixture was layered over a step gradient of 290 l of 40% Histodenz and 190 l of 80% Histodenz, both in N-2-hydroxylpiperazine-N'-2-ethanesulfonic acid-buffered saline (pH 8.0). Gradients were centrifuged at 47,000 r.p.m. in a SW55Ti rotor for 2 h at 4°C. Samples banded between the Histodenz layers were extracted with a needle and syringe, mixed with reducing SDS sample buffer, separated by SDS–polyacrylamide gel electrophoresis, and transferred to nitrocellulose membranes for immunoblotting. FKs were detected using the T7-Tag antibody horseradish peroxidase conjugate or His-Tag mAb and the enhanced chemiluminescent SuperSignal Substrate System (Pierce). Following detection, the blots were stripped and re-probed for endogenous Ad5.F5 fiber protein using the anti-fiber mAb 4D2 (NeoMarkers, Fremont, CA), peroxidase-conjugated anti-mouse immunoglobulin G antibody, and enhanced chemiluminescence.

References

1. Shiver, JW and Emini, EA (2004). Recent advances in the development of HIV-1 vaccines using replication-incompetent adenovirus vectors. Annu Rev Med 55: 355–372. | Article | PubMed | ISI | ChemPort |
2. Santosuosso, M, McCormick, S and Xing, Z (2005). Adenoviral vectors for mucosal vaccination against infectious diseases. Viral Immunol 18: 283–291. | Article | PubMed | ChemPort |
3. Kaplan, JM (2005). Adenovirus-based cancer gene therapy. Curr Gene Ther 5: 595–605. | Article | PubMed | ChemPort |
4. Relph, K, Harrington, K and Pandha, H (2005). Adenoviral strategies for the gene therapy of cancer. Semin Oncol 32: 573–582. | Article | PubMed | ChemPort |
5. Philipson, L and Pettersson, RF (2004). The coxsackie-adenovirus receptor—a new receptor in the immunoglobulin family involved in cell adhesion. In: Doerfler W and Böhm P (eds). Adenoviruses: Model and Vectors in Virus–Host Interactions. Springer-Verlag Berlin: Berlin, Germany, pp. 87–111.
6. Iacobelli-Martinez, M, Nepomuceno, RR, Connolly, J and Nemerow, GR (2005). CD46-utilizing adenoviruses inhibit C/EBPbeta-dependent expression of proinflammatory cytokines. J Virol 79: 11259–11268. | Article | PubMed | ChemPort |
7. Horvath, J and Weber, JM (1988). Nonpermissivity of human peripheral blood lymphocytes to adenovirus type 2 infection. J Virol 62: 341–345. | PubMed | ChemPort |
8. Huang, S, Endo, RI and Nemerow, GR (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. | PubMed | ISI | ChemPort |
9. Rea, D et al. (2001). Highly efficient transduction of human monocyte-derived dendritic cells with subgroup B fiber-modified adenovirus vectors enhances transgene-encoded antigen presentation to cytotoxic T cells. J Immunol 166: 5236–5244. | PubMed | ISI | ChemPort |
10. Okada, N et al. (2001). Efficient antigen gene transduction using Arg-Gly-Asp fiber-mutant adenovirus vectors can potentiate antitumor vaccine efficacy and maturation of murine dendritic cells. Cancer Res 61: 7913–7919. | PubMed | ISI | ChemPort |
11. Mathias, P, Wickham, T, Moore, M and Nemerow, G (1994). Multiple adenovirus serotypes use alpha v integrins for infection. J Virol 68: 6811–6814. | PubMed | ISI | ChemPort |
12. Zhang, Y and Bergelson, JM (2005). Adenovirus receptors. J Virol 79: 12125–12131. | Article | PubMed | ChemPort |
13. Marttila, M et al. (2005). CD46 is a cellular receptor for all species B adenoviruses except types 3 and 7. J Virol 79: 14429–14436. | Article | PubMed | ChemPort |
14. Gaggar, A, Shayakhmetov, DM and Lieber, A (2003). CD46 is a cellular receptor for group B adenoviruses. Nat Med 9: 1408–1412. | Article | PubMed | ISI | ChemPort |
15. Segerman, A, Atkinson, JP, Marttila, M, Dennerquist, V, Wadell, G and Arnberg, N (2003). Adenovirus type 11 uses CD46 as a cellular receptor. J Virol 77: 9183–9191. | Article | PubMed | ISI | ChemPort |
16. Wu, E et al. (2004). Membrane cofactor protein is a receptor for adenoviruses associated with epidemic keratoconjunctivitis. J Virol 78: 3897–3905. | Article | PubMed | ISI | ChemPort |
17. Cashman, SM, Morris, DJ and Kumar-Singh, R (2004). Adenovirus type 5 pseudotyped with adenovirus type 37 fiber uses sialic acid as a cellular receptor. Virology 324: 129–139. | Article | PubMed | ISI | ChemPort |
18. Arnberg, N, Edlund, K, Kidd, AH and Wadell, G (2000). Adenovirus type 37 uses sialic acid as a cellular receptor. J Virol 74: 42–48. | Article | PubMed | ISI | ChemPort |
19. Burmeister, WP, Guilligay, D, Cusack, S, Wadell, G and Arnberg, N (2004). Crystal structure of species D adenovirus fiber knobs and their sialic acid binding sites. J Virol 78: 7727–7736. | Article | PubMed | ISI | ChemPort |
20. Jung, D, Neron, S, Drouin, M and Jacques, A (2005). Efficient gene transfer into normal human B lymphocytes with the chimeric adenoviral vector Ad5/F35. J Immunol Methods 304: 78–87. | PubMed | ChemPort |
21. Schroers, R et al. (2004). Gene transfer into human T lymphocytes and natural killer cells by Ad5/F35 chimeric adenoviral vectors. Exp Hematol 32: 536–546. | Article | PubMed | ChemPort |
22. Basak, SK, Kiertscher, SM, Harui, A and Roth, MD (2004). Modifying adenoviral vectors for use as gene-based cancer vaccines. Viral Immunol 17: 182–196. | Article | PubMed | ChemPort |
23. Varnavski, AN, Schlienger, K, Bergelson, JM, Gao, G-P and Wilson, JM (2003). Efficient transduction of human monocyte-derived dendritic cells by chimpanzee-derived adenoviral vector. Hum Gene Ther 14: 533–544. | Article | PubMed | ChemPort |
24. Arnberg, N, Pring-Akerblom, P and Wadell, G (2002). Adenovirus type 37 uses sialic acid as a cellular receptor on Chang C cells. J Virol 76: 8834–8841. | Article | PubMed | ChemPort |
25. Li, QG, Lindman, K and Wadell, G (1997). Hydropathic characteristics of adenovirus hexons. Arch Virol 142: 1307–1322. | Article | PubMed | ChemPort |
26. Boulanger, PA and Puvion, F (1973). Large-scale preparation of soluble adenovirus hexon, penton and fiber antigens in highly purified form. Eur J Biochem 39: 37–42. | Article | PubMed | ChemPort |
27. Huebner, RJ, Rowe, WP, Ward, TG, Parrott, RH and Bell, JA (1954). Adenoidal-pharyngeal-conjunctival agents: a newly recognized group of common viruses of the respiratory system. N Engl J Med 251: 1077–1086. | PubMed | ChemPort |
28. Walters, RW, Freimuth, P, Moninger, TO, Ganske, I, Zabner, J and Welsh, MJ (2002). Adenovirus fiber disrupts CAR-mediated intercellular adhesion allowing virus escape. Cell 110: 789–799. | Article | PubMed | ISI | ChemPort |
29. Thrasher, AJ, Casimir, CM, Kinnon, C, Morgan, G, Segal, AW and Levinsky, RJ (1995). Gene transfer to primary chronic granulomatous disease monocytes. Lancet 346: 92–93. | Article | PubMed | ChemPort |
30. Schneider, SD, Rusconi, S, Seger, RA and Hossle, JP (1997). Adenovirus-mediated gene transfer into monocyte-derived macrophages of patients with X-linked chronic granulomatous disease: ex vivo correction of deficient respiratory burst. Gene Ther 4: 524–532. | Article | PubMed | ChemPort |
31. Huang, S, Kamata, T, Takada, Y, Ruggeri, ZM and Nemerow, GR (1996). Adenovirus interaction with distinct integrins mediates separate events in cell entry and gene delivery to hematopoietic cells. J Virol 70: 4502–4508. | PubMed | ISI | ChemPort |
32. Wilkins, JA, Jordan, Y, Rutherford, WJ and Warrington, RJ (1983). T cell factor-mediated suppression of immunoglobulin secretion by a human B lymphoblastoid cell line. J Immunol 131: 298–302. | PubMed | ChemPort |
33. Henry, LJ, Xia, D, Wilke, ME, Deisenhofer, J and Gerard, RD (1994). Characterization of the knob domain of the adenovirus type 5 fiber protein expressed in Escherichia coli. J Virol 68: 5239–5246. | PubMed | ISI | ChemPort |
34. Denby, L et al. (2004). Adenoviral serotype 5 vectors pseudotyped with fibers from subgroup D show modified tropism in vitro and in vivo. Hum Gene Ther 15: 1054–1064. | PubMed | ISI | ChemPort |
35. Wu, E, Fernandez, J, Fleck, SK, Von Seggern, DJ, Huang, S and Nemerow, GR (2001). A 50-kDa membrane protein mediates sialic acid-independent binding and infection of conjunctival cells by adenovirus type 37. Virology 279: 78–89. | Article | PubMed | ISI | ChemPort |

Acknowledgements

We thank Chunli Cui for excellent technical assistance. This work was supported by NIH Grant 2R01EY011431-09A1 and R24-EY14174 and is manuscript number 18178-IMM of The Scripps Research Institute.