Introduction

Gene delivery vectors based on human and non-human adenovirus (Ad) offer broad potential as molecular tools and as therapeutic agents. Ad-based systems are, to date, the most common vector type used in clinical studies worldwide (www.wiley.co.uk/genmed/clinical), representing 26% of all trials. Their ability to mediate highly efficient gene delivery to a broad range of cell types has underpinned their utility. In vivo, both in animal models of human disease and in clinical applications, the route of delivery is fundamental in defining both efficacy and safety of the approach. Local gene transfer can be effective, especially where recipient cells or tissue is permissive for Ad infection: a phenotype dictated by the presence of primary and co-receptors required for virus binding and internalization, respectively. Such local delivery procedures often eliminate (or severely restrict) the dissemination of virus to distant sites because the virus has not been exposed to the bloodstream. Nevertheless, if leakage does occur, Ad accumulation to non-target organs is an important constraint on safety.^1,2 For a number of years it has been the goal of both basic and clinical research to deliver Ad via the minimally invasive intravascular route and so achieve gene transfer to the required target cell type. It has been known for over a decade that the liver is the predominant target organ transduced by Ad based on serotype 5 (Ad5), the most commonly used serotype.³ The unique capacity of Ad for highly efficient transduction of liver cells has been demonstrated in many animal models, including rodent and non-rodent species. Furthermore, evidence from clinical cases of disseminated adenoviral infection of pediatric patients shows manifestation with fatal Ad hepatitis (for example see ref. 4). Collectively, these data provided clear evidence that, following intravascular delivery, Ad particles are efficiently targeted to the liver and are capable of productively interacting with hepatocellular receptors, resulting in cell infection. Based on this, liver-directed gene therapy using Ad has been demonstrated pre-clinically and clinically. However, the need for high doses and the resulting toxicity and immunogenicity elicited by infection limit the application of Ad5, since the therapeutic index remains narrow. This has highlighted the need to develop safer systems of vectors for liver-directed gene therapy, which has included the development of helper-dependent Ad vectors^5,6 or further vector modification to restrict Ad infection to target cells only. In this regard, engineering of Ad capsid proteins offers broad appeal, particularly in the cancer and cardiovascular fields. Most importantly, intravascular administration of oncolytic Ad is highly sought after in the cancer field, particularly in relation to eliminating metastatic disease. Because Ad5 does not optimally fulfil these criteria except for liver-targeted delivery, alternate human and non-human serotypes—as well as Ad5 vectors engineered with targeting systems that mediate a lower propensity for hepatic uptake—are in development. For example, B-group Ad vectors that target high affinity receptors expressed on target cancer cells (such as CD46) are in development.^{7,8,9,10,11,12} Despite many studies that have systematically engineered Ad capsids (particularly Ad5) with targeting systems, including peptides, antibodies, and serotype switching (reviewed in ref. 13), there is a relative paucity of data demonstrating efficient retargeting of Ad vectors to non-hepatic tissues via the intravascular delivery route. Countless studies have shown reduced liver infection by mutated and re-engineered Ad virions, but there has been a clear lack of progress in retargeting. Fundamental to this understanding is the simple but often overlooked difference between Ad particle analysis and subsequent quantification of Ad-mediated cellular transduction. Although the latter is often used as a surrogate marker for the former, cell transduction is the ultimate culmination of particle delivery, cell tethering, internalization, and nuclear trafficking. Measuring Ad particle delivery, particularly at very early time points after injection, provides vital analysis of the acute bio-distribution kinetics of Ad particles that contribute to additional mechanisms, including scavenging by resident macrophages and interactions with blood in vivo.

With respect to vector retargeting strategies, the most convincing study to date used a bi-specific antibody to redirect Ad5 to pulmonary endothelium by instigating virus uptake through the angiotensin converting enzyme. Despite a substantial improvement in lung uptake via angiotensin converting enzyme, the vast majority of virions still remained targeted to liver.^14,15 Transgene expression in the liver hepatocytes was therefore "silenced" by inclusion of an endothelial cell-selective promoter and also by the impressive transduction targeting through angiotensin converting enzyme. Hence it is clear that there are key drivers dictating hepatic and splenic uptake of Ad. Fundamental to the bio-availability of the Ad for receptor binding and uptake systems in defined organs is the interaction with blood components. This encompasses cellular as well as non-cellular components, both of which play a central role in dictating Ad tropism in vivo following intravascular delivery.

Interactions of ad with blood cells

Ad interaction with blood cells may dramatically affect virus bio-distribution and toxicity in vivo. Recently, Lyons et al.¹⁶ analyzed Ad5 interaction with human and mouse blood cells ex vivo and found that, after a short incubation with blood cells, over 90% of the applied virus dose stably associated with human (but not murine) erythrocytes. They further demonstrated that, in samples of human blood obtained from a patient administered with Ad vector during a clinical trial, more than 98% of Ad genomes were also associated with blood cells.¹⁶ Ultimately, Ad association with red blood cells resulted in reduced virus infectivity for susceptible cell lines. A similar effect was observed when virus infection was conducted in the presence of high levels of neutralizing antibodies in human plasma.¹⁶ In contrast, the presence of low levels of neutralizing antibodies enhanced Ad infection of human monocytes. The authors concluded that Ad interaction with blood cells upon intravascular delivery may significantly decrease virus access to extravascular target cells and tissues, thus limiting the resulting transgene expression in target tissue. In turn, higher doses of Ad may be required for efficacy—with the caveat that toxicity issues become highly relevant. They also speculated that Ad interaction with low levels of neutralizing antibodies in the bloodstream may enhance Ad delivery to circulating monocytes and increase antigen presentation.¹⁶ In agreement with the latter conclusion, an elegant study by Leopold et al.¹⁷ demonstrates that formation of Ad complexes with neutralizing antibodies can, in fact, facilitate virus infection via FcR, which is present on antigen presenting cells. Neutralizing antibodies reduced the virus ability to infect cells via the coxsackievirus and Ad receptor (CAR)-dependent pathway, but when target cells were engineered to express FcR, the virus could efficiently deliver its DNA to the cell nucleus following initial attachment to FcR.¹⁷

The mechanisms mediating Ad interactions with neutrophils, the primary effector cells of the innate immune system, were described by Cotter et al.¹⁸ Following intravascular Ad delivery to mice, neutrophils were rapidly recruited to the liver: 25% interacted with Ad particles. The authors used fluorescent and electron microscopy to demonstrate that, upon incubation with primary human neutrophils, Ad particles bind and internalize. The interaction with neutrophils was associated with marked reduction in Ad infectivity and depended on the presence of complement and antibodies. Ad–neutrophil interactions could be reduced by blocking CD35 (complement receptor 1) or the FcRs CD64, CD32, or CD16. A combined CD35 and FcRs blockade synergistically inhibited Ad–neutrophil interactions to baseline, demonstrating that there are diverse and likely redundant pathways evolved by the host to ensure clearance of virus pathogens from the blood.¹⁸

Whereas neutrophils are present in the circulation at high levels only during times of acute host inflammatory responses, platelets are constantly present in the blood in high numbers (1.5–4.5 10⁵platelets/ml) and are an essential part of the blood coagulation system. Although earlier reports demonstrated that Ad5-based vectors do not induce, inhibit, or potentiate platelet aggregation in vitro,¹⁹ recent studies have provided clear evidence that Ad particles do interact with platelets in vivo. Othman et al.²⁰ reported detailed studies demonstrating that, upon intravascular delivery, Ad5 virus activates platelets and induces platelet–leukocyte aggregate formation in mice. The formation of platelet–leukocyte aggregates dependent on von Willebrand factor and Ad-induced thrombocytopenia was markedly reduced in von Willebrand factor knockout mice. Seventy-eight percent of human platelets stained positive for the Ad5 receptor CAR; this suggests that, in humans, platelets may sequester Ad particles after intravascular delivery. In a recent study by Stone et al.,²¹ direct Ad–platelet interactions in blood caused virus sequestration to the reticulo-endothelial system of liver. After intravenous delivery to mice, Ad5 was found to rapidly bind to circulating platelets, leading to platelet activation, aggregation, and subsequent trapping in liver sinusoids. It was proposed that virus-platelet aggregates are further taken up by Kupffer cells and degraded. Ad sequestration to different organs was reduced by platelet depletion before intravascular Ad administration, demonstrating the importance of this interaction in vivo.²¹ Although human Ad serotype 5 and serotype 11 could both be found in association with platelets following intravascular delivery,^22,23 quantitative studies have suggested that human Ad serotype 11 (Ad11) in mice associate with blood cells, and with platelets in particular, to a much higher degree.^22,24 The high-efficiency Ad11 interaction with platelets in vivo occurs in a fiber-independent manner, suggesting that a novel and still uncharacterized mechanism likely mediates Ad interactions with circulating blood cells.

The role of plasma protein interactions with ad in defining bio-distribution and transduction in vivo

The role of CAR, integrins, and heparan sulphate proteoglycans in hepatic uptake of viruses in vivo has been a controversial area of research. Fully defining the mechanisms of hepatic Ad uptake is of fundamental importance if Ad vectors are to be effectively optimized for human gene therapy. It is notable that Ad5 mutants that are devoid of CAR binding capacity do not show reduced hepatocyte transduction following intravenous delivery; nor do integrin binding mutants or, indeed, combined double mutants (ref. 13 for review). Three studies in mice, rats, and non-human primates have described a clear hepatocyte "detargeting" when the putative heparan sulfate proteoglycansbinding motif in the fiber shaft (KKTK) was mutated to GAGA.^25,26,27 Previous in vitro studies have highlighted the potential role for heparan sulfate proteoglycans binding in Ad transduction in vitro.^28,29 Hence it was thought that the KKTK might block transduction mediated through heparan sulfate proteoglycans. However, two recent studies now suggest that Ad5 vectors mutated at this site are in fact dysfunctional for cell transduction even in the presence of knob-engineered retargeting systems.^30,31 In vivo, recent evidence has pinpointed heparan sulfate proteoglycans and low-density lipoprotein-related protein in Ad-mediated liver transduction, an effect resulting from binding of Ad vectors to plasma proteins following intravascular injection. Shayakhmetov et al.³² demonstrated that coagulation factor IX (FIX) and complement factor 4 binding protein had the potential to influence hepatocyte transduction mediated by Ad5 and Ad5/35 (Ad5-based vectors with Ad35-derived fiber knob domain). This was demonstrated in vitro and also in an in situliver perfusion setting that allowed assessment of hepatocyte delivery of Ad in the presence and absence of blood³² (Figure 1). This clearly showed that physiological levels of FIX were able to influence hepatocyte targeting in vivo. Heparinase and lactoferrin blocked transduction, implicating new receptor pathways in vivo. In fact, this study also showed that many proteins were able to bind to the Ad fiber, suggesting complex interplay between Ad and host systems that may influence in vivo virus bio-distribution and cell transduction—rather than the previously accepted two-step receptor binding and internalization mechanism of infection (Figure 1).

Figure 1.

Pathways mediating adenovirus (Ad) infection of cells in vitro and liver cells in vivo following intravascular delivery. Following intravascular injection, plasma proteins bind Ad5; this leads to liver transduction through heparan sulfate proteoglycans (HSPG) and/or lipoprotein-related protein (LRP) binding, effects that can be blocked by pathway inhibitors including warfarin, heparin, heparinase, and lactoferrin.^31,35

Full figure and legend (30K)

Coagulation factors with a defined structure influence in vivo ad tropism

The serine proteases of the blood coagulation network (factors (F) VII, IX, X, and XI, protein C, and prothrombin) are synthesized in the liver and circulate in plasma as inactive zymogens that require limited proteolytic cleavage for activation.³³ FVII, FIX, FX, and protein C share a common protein structure: they are all zymogens of vitamin K–dependent serine proteases with the domain structure Gla-EGF1-EGF2-SP (-carboxylate glutamic acid, epidermal growth factor-1 and -2-like and serine protease) (Figure 2). Although they have distinct functional properties within the coagulation network, analysis of the gene organizations, protein structures, and sequence identities suggests that they have resulted from gene duplication events that occurred before the divergence of teleosts from tetrapods more than 430 million years ago.^34,35 Ad5 was shown to bind to factors sharing the Gla-EGF1-EGF2-SP structure (FVII, FIX, FX, and PC) but not to FXI (which has the domain structure of four "apple" (AP) or PAN domains and a serine protease domain (AP1-AP2-AP3-AP4-SP)).³⁶ Furthermore, these factors had the capacity to enhance Ad transduction of hepatoma cells in vitro. It is interesting that this effect did not require zymogen activation. The study also developed an important in vivo model to assess the role of these Gla domain–containing coagulation factors in Ad liver transduction. The Gla domain consists of a number (9–12) of glutamic acid residues that are posttranslationally modified by the addition of a carboxyl group to the -carbon by a vitamin K–dependent carboxylase. Warfarin, a widely used anticoagulant drug, blocks this posttranslational modification, preventing appropriate folding of the domain and reducing functional levels of these factors.³⁷ Warfarin injected into mice before administration of Ad effectively blocked liver transduction mediated by the CAR-binding deleted Ad vector.³⁶ Furthermore, FX injection at a level that fully reconstituted physiological levels of circulating FX completely restored liver transduction in the warfarin-treated cohort, thereby demonstrating a critical role for FX in vivo.³⁶ This recent work from our laboratories^30,36 thus provides an important new perspective on mechanisms governing in vivo Ad tropism (Figure 1).^32,36 However, many questions related to this newly described infectivity pathway still remain to be answered. What is the influence of warfarin on non-transducing events such as Kupffer cell uptake of Ad? What are the kinetics of Ad5–coagulation factor binding interactions, and precisely where does coagulation factor binding occur? The holy grail for much of intravascular work is to engineer Ad vectors with efficient targeting systems for alternate, non-hepatic tissues. Does modulation of this pathway allow Ad targeting systems to perform efficiently? The finding that warfarin ablates liver transduction firmly implicates this pathway in in vivo Ad tropism and warrants further and substantive investigations. Is this pathway relevant to other Ad human serotypes and, indeed, to non-human serotypes? We already know that serotype 35 fiber knob domain is sensitive to FIX binding,³² so what are the implications of this for targeting CD46 on non-hepatic cells such as metastatic cancer cells? Recent studies have shown that human subgroup D fibers pseudotyped onto Ad5 vectors are very sensitive to FX-mediated transduction of hepatocytes in vitro.³⁸ Analysis of full serotypes rather than fiber pseudotypes will be important.

Figure 2.

Structural homology between coagulation factors and Ad5 binding. (a) Molecular model of the serine protease (blue), epidermal growth factor (EGF)-like (red), and Gla (cyan) domains of human FVIIa. The location of the calcium ions is indicated by green spheres. The active site inhibitor is represented by ball and stick. (b) Alignments of human Gla-EGF1-EGF2-serine protease amino acid sequences were generated using ClustalW and edited in BioEdit (http://www.mbio.ncsu.edu/BioEdit/bioedit.html/). The Gla (cyan), EGF-like (red), and serine protease (blue) domains are indicated by the colored bars. (c) Surface plasmon resonance analysis of FX–adenovirus interactions. AdKO1 10¹¹ virus particles/ml was perfused over FX, and FXI was immobilized onto a CM5-sensor chip in 50 mmol/l Tris pH 7.4; 150 mmol/l NaCl; 5 mmol/l CaCl₂; 0.005% Tween 20 at a flow rate of 20 l/minute at 25°C. Depicted are sensorgrams demonstrating FX specific binding (purple trace) following injection of AdKO1. No binding to the FXI channel is observed (green trace). Injection of 3 mmol/l ethylenediaminetetraacetic acid (EDTA) dissociates the virus from the FX, indicating that this is a calcium-dependent interaction. RU, response units.

Full figure and legend (77K)

In addition to interactions with the coagulation system, human complement component C3 has also been shown to bind to Ad5 virus particles in the presence of either human plasma or a mixture of complement Factors B and D.³⁹ The role of the complement system in modulating in vivo Ad infectivity and toxicity has been further demonstrated by a 99-fold reduction in liver cell transduction by Ad vectors in complement C3 knockout mice compared to control mice.^40,41 Unexpectedly, this reduction was most pronounced only when rather low doses of the virus (2.3 10⁹ virus particles/mouse) were used for intravascular delivery and was less significant at higher administered doses of the virus. Furthermore, the acute inflammatory responses induced by intravenous delivery of Ad were attenuated in mice lacking functional complement component C3.

The role of non-fiber-mediated pathways in determining ad clearance from blood

After intravascular application of human Ad5-based vectors to mice, the well-documented and relatively efficient liver cell infection correlates with a high viral vector DNA load per milligram of liver tissue when analyzed between 24 and 72 hours after virus administration. Somewhat surprisingly, data from analyses of bio-distribution and liver cell transduction by Ad35- and Ad11-based vectors demonstrated that, although liver cell transduction (as assessed by reporter gene expression) was several orders of magnitude lower than for Ad5, the total amounts of virus DNA recovered from the liver shortly after intravenous delivery was comparable to that of Ad5.^22,42 Indeed, Reddy et al.⁴³ reported that both Ad5- and Ad35-based vectors exerted similar kinetics of clearance from blood, and 99% of the virus dose was eliminated from the blood within 3 minutes of intravenous administration.⁴³ Furthermore, in two independent studies, Sakurai et al. reported that the amount of Ad DNA deposited in liver within 3 hours of intravenous administration was similar for both Ad5 and Ad35 vectors, whereas only Ad5 showed efficient hepatocyte transduction (as evaluated by reporter gene expression).^42,44 Similar data was reported by Stone et al., who analyzed in vivo bio-distribution and toxicity of an Ad11-based vector.²² Using Southern blot analysis, the authors demonstrated that the level of vector DNA for both Ad5 and Ad11 was indistinguishable in liver 30 minutes after infusion; however, efficient reporter gene expression in the liver was observed only in Ad5-administered animals. Taken together, these data suggest that, after intravascular delivery, liver tissue (at least in mice) has the capacity to efficiently clear Ad from the blood and that this clearance apparently does not depend on liver cell transduction. To better understand the molecular mechanisms involved in Ad trapping in the liver, we analyzed the correlation between deposition of Ad DNA in the liver 30 minutes after intravenous administration and liver cell transduction for Ad5-based vectors possessing either long Ad5-derived (Ad5/35L) or short Ad9- or Ad35-derived (Ad5/35S) fibers.⁹ Our studies showed that Ad5-based vectors with long fibers efficiently accumulated in the liver 30 minutes after intravenous application, and this correlated well with high-efficiency liver cell transduction. Conversely, for vectors with short-shafted fibers, liver cell transduction was extremely inefficient despite high levels of vector DNA accumulation 30 minutes after intravenous vector delivery. Our subsequent analyses demonstrated that (i) Ad5/35S accumulation in the liver did not require fiber-mediated vector interaction with cellular attachment receptors, and (ii) after deposition in liver, Ad5/35S failed to internalize into cells and could be quantitatively removed after collagenase or trypsin treatment. It is worth noting that, even when large amounts of Ad5/35S vector are cleared by the liver, administration of Ad5/35S (unlike Ad5) is not associated with induction of inflammatory host responses.⁴⁵ This suggests that this mechanism of virus clearance from the blood is likely a decoy pathway that ensures the removal of pathogens from circulation to prevent their dissemination into other tissues. Because the liver accumulates large and comparable amounts of unmodified Ad5 as well as short-shafted Ad5/35S, Ad35, or Ad11 vectors after intravascular application, there is likely a common mechanism mediating Ad sequestration to liver that does not require Ad fiber interaction with cellular receptors and results in clearance from the blood and inactivation of the bulk of the administered vector dose (Figure 3). We are currently only beginning to understand the importance of fiber-independent mechanisms of Ad clearance from the blood. It is clear, however, that preventing decoy removal of the bulk of Ad following intravascular application would represent a major step toward generation of efficient vectors with improved pharmacokinetic and toxicity profiles for applications in vivo. This field is an active area of research, and future studies should show the relevance of all Ad–host interactions found in mice to other animal models and, ultimately, to humans.

Figure 3.

Fiber-dependent and fiber-independent mechanisms of adenovirus (Ad) clearance from the blood following intravascular vector delivery. CAR, Coxsackievirus and adenovirus receptor.

Full figure and legend (19K)

Limitations of animal models

The emerging complexity of Ad interactions with cells and proteins of the blood, when virus is delivered intravenously, may have dramatic ramifications for developing and testing novel Ad-based vectors for human gene therapy applications. Indeed, when novel gene therapy vectors are developed to treat human disease, studies in pre-clinical animal models must initially be conducted to assess the vector performance in vivo. Accumulating research data suggest, however, that currently existing animal models for testing the performance of human Ad-based vectors may likely provide only limited information—with uncertain predictive value regarding the vector performance in humans. For example, the discovery that CD46 is a major receptor for human group B Ads^8,12 precludes the effective use of non-human primates for pre-clinical analyses of group B virus vectors. This is principally due to the fact that, unlike humans, old-world monkeys express CD46 on erythrocytes. Similarly, rodent CD46 expression is restricted to testes, unlike the widespread expression pattern in humans,⁴⁶ and this necessitates the use of transgenic mice expressing human CD46 in a tissue-specific manner in order to provide a model more akin to human CD46 expression. This complicates murine studies because the targeting of CD46-overexpressing human tumors should be investigated, yet this requires the use of immunologically deficient animals for xenograft studies. The higher levels of Ad11 trapping in blood of mice, compared with Ad5, suggest that new and unknown mechanisms of virus clearance may occur even in well-characterized rodent models when new vectors are tested for in vivo applications. Moreover, in currently used animal models, the lack of pre-existing virus neutralizing antibodies—which are prevalent in humans—suggests that clearance mechanisms relying on FcR are not invoked, thus further complicating the translation of vector performance in animals to humans in the setting of either single or repeat injections.

Ad5 interaction with erythrocytes is species sensitive,⁴⁷ human and rat agglutination appears to be mediated by CAR (since CAR-binding mutants do not agglutinate erythrocytes²⁵), and CD46 is ubiquitously expressed on all human nucleated cells.⁴⁸ These are just a few of the species-related issues that affect the relevance and translation from data in certain animal models. Where possible, future Ad-based vectors developed for targeted gene delivery following intravascular application must take account of these complex interactions. This is an absolute requirement if the goal of selectivity in vector transduction is to be achieved effectively. Further definition of the molecular mechanisms and precise modes of interaction between Ad particles and blood cells or proteins represents a significant and challenging area of research. Such research should help us better understand anti-viral host defense mechanisms and also aid in the development of novel and efficient viral vectors for treating human diseases.

Summary and future directions

The balance of the evidence outlined here suggests that blood cells and circulating plasma proteins, such as Gla-EGF1-EGF2-SP-containing coagulation factors, play a central role in defining the in vivo pharmacokinetics of Ad vector transduction when administered via the intravascular route. Defining specific interactions and unravelling any species differences in blood cell–virus interactions is critical to understanding the translational aspects of Ad biology in vivo. Published studies using a number of model systems demonstrate the importance of Ad-coagulation factor interactions. The potential complexity of the interactions of Ad vectors with plasma proteins and blood cells immediately after intravascular delivery requires careful and thorough investigation in appropriate model systems, since this dictates bio-availability and cell/organ targeting by the virus in vivo. In turn, this defines the potential of the vector for successful gene therapy applications.

References

REFERENCES

1. Wang, YH, Hu, JK, Krol, A, Li, YP, Li, CY and Yuan, F (2003). Systemic dissemination of viral vectors during intratumoral injection. Mol Cancer Ther 2: 1233–1242. | PubMed | ISI | ChemPort |
2. Hiltunen, MO, Turunen, MP, Turunen, AM, Rissanen, TT, Laitinen, M, Kosma, VM et al. (2000). Biodistribution of adenoviral vector to nontarget tissues after local in vivo gene transfer to arterial wall using intravascular and periadventitial gene delivery methods. FASEB J 14: 2230–2236. | Article | PubMed | ISI | ChemPort |
3. Huard, J, Lochmuller, H, Ascadi, G, Jani, A, Massie, B and Karpati, G (1995). The route of administration is a major determinant of the transduction efficiency of rat tissues by adenoviral recombinants. Gene Ther 2: 107–115. | PubMed | ISI | ChemPort |
4. Steiner, I, Aebi, C, Ridolfi Luthy, A, Wagner, B and Leibundgut, K (2007). Fatal adenovirus hepatitis during maintenance therapy for childhood acute lymphoblastic leukemia. Pediatr Blood Cancer epub (ahead of print).
5. Oka, K, Pastore, L, Kim, IH, Merched, A, Nomura, S, Lee, HJ et al. (2001). Long-term stable correction of low-density lipoprotein receptor-deficient mice with a helper-dependent adenoviral vector expressing the very low-density lipoprotein receptor. Circulation 103: 1274–1281. | PubMed | ISI | ChemPort |
6. Parks, RJ, Chen, L, Anton, M, Sankar, U, Rudnicki, MA and Graham, FL (1996). A helper-dependent adenovirus vector system: removal of helper virus by cre-mediated excision of the viral packaging signal. Proc Natl Acad Sci USA 93: 13565–13570. | Article | PubMed | ChemPort |
7. Shayakhmetov, D, Li, ZY, Ni, S and Lieber, A (2002). Targeting of adenovirus vectors to tumor cells does not enable efficient transduction of breast cancer metastases. Cancer Res 62: 1063–1068. | PubMed | ISI | ChemPort |
8. 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 |
9. Shayakhmetov, DM, Li, ZY, Ni, S and Lieber, A (2004). Analysis of adenovirus sequestration in the liver, transduction of hepatic cells, and innate toxicity after injection of fiber-modified vectors. J Virol 78: 5368–5381. | Article | PubMed | ISI | ChemPort |
10. Stone, D, Ni, S, Li, ZY, Gaggar, A, DiPaolo, N, Feng, Q et al. (2005). Development and assessment of human adenovirus type 11 as a gene transfer vector. J Virol 79: 5090–5104. | Article | PubMed | ISI | ChemPort |
11. Gaggar, A, Shayakhmetov, DM, Liszewski, MK, Atkinson, JP and Lieber, A (2005). Localization of regions in CD46 that interact with adenovirus. J Virol 79: 7503–7513. | Article | PubMed | ISI | ChemPort |
12. 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 |
13. Nicklin, SA, Wu, E, Nemerow, GR and Baker, AH (2005). The influence of adenovirus fiber structure and function on vector development for gene therapy. Mol Ther 12: 384–393. | Article | PubMed | ISI | ChemPort |
14. Reynolds, PN, Nicklin, SA, Kaliberova, L, Boatman, BG, Grizzle, WE, Balyasnikova, IV et al. (2001). Combined transductional and transcriptional targeting improves the specificity of transgene expression in vivo. Nat Biotechnol 19: 838–842. | Article | PubMed | ISI | ChemPort |
15. Reynolds, PN, Zinn, KR, Gavrilyuk, VD, Balyasnikova, IV, Rogers, BE, Buchsbaum, DJ et al. (2000). A targetable, injectable adenoviral vector for selective gene delivery to pulmonary endothelium in vivo. Mol Ther 2: 562–578. | Article | PubMed | ISI | ChemPort |
16. Lyons, M, Onion, D, Green, NK, Aslan, K, Rajaratnam, R, Bazan-Peregrino, M et al. (2006). Adenovirus type 5 interactions with human blood cells may compromise systemic delivery. Mol Ther 14: 118–128. | Article | PubMed | ISI | ChemPort |
17. Leopold, PL, Wendlan, RL, Vincent, T and Crystal, RG (2006). Neutralized adenovirus-immune complexes can mediate effective gene transfer via an Fc receptor-dependent infection pathway. J Virol 80: 10237–10247. | Article | PubMed | ISI | ChemPort |
18. Cotter, MJ, Zaiss, AK and Muruve, DA (2005). Neutrophils interact with adenovirus vectors via Fc receptors and complement receptor 1. J Virol 79: 14622–14631. | Article | PubMed | ISI | ChemPort |
19. Eggerman, TL, Mondoro, TH, Traci, Lozier JN and Vostal, JG (2002). Adenoviral vectors do not induce, inhibit, or potentiate human platelet aggregation. Hum Gene Ther 13: 125–128. | Article | PubMed | ISI | ChemPort |
20. Othman, M, Labelle, A, Mazzetti, I, Elbatarny, HS and Lillicrap, D (2006). Adenovirus-induced thrombocytopenia: the role of von Willebrand factor and P-selectin in mediating accelerated platelet clearance. Blood 109: 2832–2839. | PubMed | ISI |
21. Stone, D, Liu, Y, Shayakhmetov, D, Li, ZY and Lieber, A (2007). Adenovirus-platelet interaction in blood causes virus sequestration to the reticulo-endothelial system of liver. J Virol 81: 4866–4871. | Article | PubMed | ChemPort |
22. Stone, DN, Ni, S, Li, ZY, Gaggar, A, DiPaolo, N, Feng, Q et al. (2005). Development and assessment of human adenovirus type 11 as a gene transfer vector. J Virol 79: 5090–5104. | Article | PubMed | ISI | ChemPort |
23. Stone, D, Shaoheng, N, Li, ZY, Shayakhmetov, D and Lieber, A (2005). Trapping of adenovirus vectors in blood. Mol Ther 11: 159. | Article |
24. Stone, D, Shaoheng, N, Li, ZY, Shayakhmetov, D and Lieber, A (2006). Intravascular interaction between adenovirus and platelets promotes adenovirus sequestration, degradation and toxicity. Mol Ther 13: S30–S31.
25. Nicol, CG, Graham, D, Miller, WH, White, SJ, Smith, TA, Nicklin, SA et al. (2004). Effect of adenovirus serotype 5 fiber and penton modifications on in vivo tropism in rats. Mol Ther 10: 343–354. | Article | ChemPort |
26. Smith, TA, Idamakanti, N, Rollence, ML, Marshall-Neff, J, Kim, J, Mulgrew, K et al. (2003). Adenovirus serotype 5 fiber shaft influences in vivo gene transfer in mice. Hum Gene Ther 14: 777–787. | Article | PubMed | ISI | ChemPort |
27. Smith, TA, Idamakanti, N, Marshall-Neff, J, Rollence, ML, Wright, P, Kaloss, M et al. (2003). Receptor interactions involved in adenoviral-mediated gene delivery after systemic administration in non-human primates. Hum Gene Ther 14: 1595–1604. | Article | PubMed | ISI | ChemPort |
28. Dechecchi, MC, Melotti, P, Bonizzato, A, Santacatterina, M, Chilosi, M and Cabrini, G (2001). Heparan sulfate glycosaminoglycans are receptors sufficient to mediate the initial binding of adenovirus types 2 and 5. J Virol 75: 8772–8780. | Article | PubMed | ISI | ChemPort |
29. Dechecchi, MC, Tamanini, A, Bonizzato, A and Cabrini, G (2000). Heparan sulphate glycosaminoglycans are involved in adenovirus type 5 and 2-host cell interactions. Virology 268: 382–390. | Article | PubMed | ISI | ChemPort |
30. Bayo-Puxan, N, Cascallo, M, Gros, A, Huch, M, Fillat, C and Alemany, R (2006). Role of the putative heparan sulfate glycosaminoglycan-binding site of the adenovirus type 5 fiber shaft on liver detargeting and knob-mediated retargeting. J Gen Virol 87: 2487–2495. | Article | PubMed | ISI | ChemPort |
31. Kritz, AB, Nicol, CG, Dishart, KL, Nelson, R, Holbeck, S, Von Seggern, DJ et al. (2007). Adenovirus 5 fibers mutated at the KKTK putative HSPG binding site show restricted retargeting capacity when engineered with targeting peptides in the HI loop. Mol Ther 15: 741–749. | PubMed | ISI |
32. Shayakhmetov, DM, Gaggar, A, Ni, S, Li, ZY and Lieber, A (2005). Adenovirus binding to blood factors results in liver cell infection and hepatotoxicity. J Virol 79: 7478–7491. | Article | PubMed | ISI | ChemPort |
33. Gomez, K and McVey, JH (2006). Tissue factor initiated blood coagulation. Front Biosci 11: 1349–1359. | Article | PubMed | ISI | ChemPort |
34. Davidson, CJ, Tuddenham, EG and McVey, JH (2003). 450 million years of hemostasis. J Thromb Haemost 1: 1487–1494. | Article | PubMed | ISI | ChemPort |
35. Davidson, CJ, Hirt, RP, Lal, K, Snell, P, Elgar, G, Tuddenham, EG et al. (2003). Molecular evolution of the vertebrate blood coagulation network. Thromb Haemost 89: 420–428. | PubMed | ISI | ChemPort |
36. Parker, AL, Waddington, SN, Nicol, CG, Shayakhmetov, DM, Buckley, SM, Denby, L et al. (2006). Multiple vitamin K-dependent coagulation zymogens promote adenovirus-mediated gene delivery to hepatocytes. Blood 108: 2554–2561. | Article | PubMed | ISI | ChemPort |
37. Zivelin, A, Rao, LV and Rapaport, SI (1993). Mechanism of the anticoagulant effect of warfarin as evaluated in rabbits by selective depression of individual procoagulant vitamin K-dependent clotting factors. J Clin Invest 92: 2131–2140. | PubMed | ISI | ChemPort |
38. Parker, AL, McVey, JH, Doctor, JH, Lopez-Franco, O, Waddington, SN, Havenga, MJ et al. (2007). Influence of coagulation factor zymogens on the Infectivity of adenoviruses pseudotyped with fibers from subgroup D. J Virol 81: 3627–3631. | Article | PubMed | ISI | ChemPort |
39. Jiang, H, Wang, Z, Serra, D, Frank, MM and Amalfitano, A (2004). Recombinant adenovirus vectors activate the alternative complement pathway, leading to the binding of human complement protein C3 independent of anti-ad antibodies. Mol Ther 10: 1140–1142. | Article | PubMed | ISI | ChemPort |
40. Zinn, KR, Szalai, A, Stargel, A, Krasnykh, V and Chaudhuri, TR (2004). Bioluminescence imaging reveals a significant role for complement in liver transduction following intravenous delivery of adenovirus. Gene Ther 11: 1482–1486. | Article | PubMed | ISI | ChemPort |
41. Kiang, A, Hartman, ZC, Everett, RS, Serra, D, Jiang, H, Frank, MM et al. (2006). Multiple innate inflammatory responses induced after systemic adenovirus vector delivery depend on a functional complement system. Mol Ther 14: 588–598. | Article | PubMed | ISI | ChemPort |
42. Sakurai, F, Mizuguchi, H, Yamaguchi, T and Hayakawa, T (2003). Characterization of in vitro and in vivo gene transfer properties of adenovirus serotype 35 vector. Mol Ther 8: 813–821. | Article | PubMed | ISI | ChemPort |
43. Reddy, PS, Ganesh, S, Limbach, MP, Brann, T, Pinkstaff, A, Kalossa, M et al. (2003). Development of adenovirus serotype 35 as a gene transfer vector. Virology 311: 384–393. | Article | PubMed | ISI | ChemPort |
44. Sakurai, F, Kawabata, K, Koizumi, N, Inoue, N, Okabe, M, Yamaguchi, T et al. (2006). Adenovirus serotype 35 vector-mediated transduction into human CD46-transgenic mice. Gene Ther 13: 1118–1126. | Article | PubMed | ISI | ChemPort |
45. Shayakhmetov, DM, Li, ZY, Ni, S and Lieber, A (2005). Interference with the IL-1-signaling pathway improves the toxicity profile of systemically applied adenovirus vectors. J Immunol 174: 7310–7319. | PubMed | ISI | ChemPort |
46. Inoue, N, Ikawa, M, Nakanishi, T, Matsumoto, M, Nomura, M, Seya, T et al. (2003). Disruption of mouse CD46 causes an accelerated spontaneous acrosome reaction in sperm. Mol Cell Biol 23: 2614–2622. | Article | PubMed | ISI | ChemPort |
47. Cichon, G, Boeckh-Herwig, S, Kuemin, D, Hoffmann, C, Schmidt, HH, Wehnes, E et al. (2003). Titer determination of Ad5 in blood: a cautionary note. Gene Ther 10: 1012–1017. | Article | PubMed | ISI | ChemPort |
48. Cho, SW, Oqlesby, TJ, Hsi, BL, Adams, BM and Atkinson, JP (1991). Characterization of three monoclonal antibodies to membrane co-factor protein (MCP) of the complement system and quantification of MCP by radioassay. Clin Exp Immunol 83: 257–261. | PubMed | ISI | ChemPort |

Acknowledgments

This work was supported by the grants from the European Commission (to A.H.B.), Biotechnology and Biophysical Research Council (to A.H.B.) and the US National Institutes of Health AI062853, AI064882, and AI065429 (to D.M.S). Haemostasis and Thrombosis are supported by the Medical Research Council. S.N.W. is a recipient of the Philip Gray Memorial Fellowship, Katharine Dormandy Trust.