HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) All Versions of this Article: 15/12/2283 01-0321fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by SOUDAIS, C. Articles by KREMER, E. J. Search for Related Content PubMed PubMed Citation Articles by SOUDAIS, C. Articles by KREMER, E. J.

Preferential transduction of neurons by canine adenovirus vectors and their efficient retrograde transport in vivo ¹

Généthon III/CNRS 1923, Evry, France; and
^* Embryologie Moléculaire, Institut Pasteur, Paris, France

²Correspondence: Institute de Génétique Moléculare de Montpellier UMR-CNRS 5533, 1919 Route de Mende, 34293 Monpellier, France. E-mail: ekremer@genethon.fr or kremer{at}igm.cnrs-mop.fr

SPECIFIC AIMS

The initial goal of our study was to characterize the in vivo and ex vivo efficacy of a novel viral vector derived from canine adenovirus serotype 2 (CAV-2). After identifying the preferential transduction of neurons, we propose a molecular basis for the targeting in the tissues tested.

PRINCIPAL FINDINGS

1. CAV-2 preferentially transduce neurons in vitro, in vivo in rodents, and ex vivo in human brain biopsies
We previously described the transduction of mouse distal airway epithelial cells via deep nasal installation using a CAV-2 vector expressing GFP (CAVGFP). While continuing the characterization of CAV-2 vectors, we moved up the respiratory tract and assayed transduction in the rat olfactory neuroepithelium. In the nasal cavity, the ethmoid labyrinth is teeming with columnar epithelial cells that sandwich olfactory neurons. We found no difference between GFP expression of the axon bundles in the olfactory bulb using CAVGFP or a human adenovirus serotype 5 (Ad5) vector expressing eGFP (AdGFP). However, a difference was seen in the ethmoid labyrinth: AdGFP transduced olfactory neurons and epithelial cells whereas CAVGFP appeared to transduce exclusively the former. Immunohistochemical (IH) staining with anti-OMP antibody confirmed that CAVGFP-transduced cells were olfactory neurons.

We then generated and transduced primary neuron cultures from rat spinal cords with CAVGFP, CAVßgal, and AdGFP. Using IH staining, we found that CAVGFP efficiently transduced the majority of neurons in the cultures. We also show that CAVGFP poorly transduced astrocytes and oligodendrocytes. As a control, we demonstrated that a comparable Ad5 vector readily transduced neurons, astrocytes, and oligodendrocytes in spinal cord cultures.

We tested CAV-2 transduction by injection into the rat striatum. As an internal control, AdRFP (expressing red fluorescent protein) was coinjected. Absent from the injection site was the cluster of CAVGFP-transduced cells found with AdRFP, described previously as being 60% glia with Ad5 vectors. There were 20-fold more RFP⁺ than GFP⁺ cells at the injection site. This was the only region (see below) that had a higher percentage of AdRFP vs. CAVGFP-transduced cells.

Preclinical animal testing is indispensable during the development of viral vectors, but may not reflect the result obtained in humans. Adßgal transduction of neurons in vitro in rodents and nonhuman primates and ex vivo in human brain tissue repeatedly demonstrated that Ad5 preferentially transduced glia. To be as clinically relevant as possible, we assayed CAVGFP transduction in human epileptogenic brain biopsy from 12 patients ranging from 2 to 17 years old. After CAVGFP transduction of temporal cortex from a 13-year-old patient, numerous discrete GFP⁺ cells were found. To identify unequivocally the transduced cells, IH staining showed that CAVGFP preferentially transduced neurons (90%). Approximately 60% of the AdGFP-transduced cells were glia. These data demonstrated that in human brain CAV-2 preferentially transduced neurons.

2. Efficient axonal retrograde transport of CAV-2 in the central nervous system (CNS) and when injected into skeletal muscle
After injection into the rat striatum, we asked whether CAVGFP was transported to afferent structures such as the cortex, the substantia nigra, and the thalamus (centromedian nucleus) (Fig. 1) . In the ipsilateral cortex, GFP⁺ cells were found in every section from +3 to -9 mm (relative to the bregma). Sections with an equivalent number of GFP⁺ cells shown in Fig. 1 B were detected in at least 20 consecutive sections (3.0 mm total distance). A conservative estimate of the total number of CAVGFP-transduced cells in the ipsilateral cortex would be greater than 10⁴ cells. RFP⁺ cells were also found in the ipsilateral cortex, but significantly less (between 10- and 50-fold). We did not detect an afferent region transduced by AdRFP remotely similar to that shown in Fig. 1B . Akin to the results found in the cortex, GFP⁺ cells were found in each section (from -4.75 to -6.5 mm) throughout the substantia nigra (>10²). There were few RFP⁺ cells detected in the substantia nigra, and we did not detect neurons that were both GFP⁺ and RFP⁺. Finally, random, isolated GFP⁺ and RFP⁺ cells were found throughout the centromedian nucleus, with fivefold more GFP⁺ cells. Like E1-deleted Ad5 vectors, transgene expression from CAVGFP could be detected for at least 8 wk.

View larger version (13K): [in this window] [in a new window]   Figure 1. Stereotactic injection of CAVGFP and AdRFP in the rat striatum. Rat brains were cut rostrocaudal into 150 µm thick sections, and screened for GFP and RFP expression. Site of the injection in the A) striatum showing AdRFP-transduced astrocytes and neurons (red) and colocalization with CAVGFP-transduced cells (white arrows). Retrograde axonal transport and expression from CAVGFP-transduced neurons in the B) cortex, C) substantia nigra, and D) centromedian nucleus. Schematic inserts show midsagittal and frontal cross sections of the rat brain and the location of the injection site relative to the transduced cells.

We asked whether CAVGFP could transduce motoneurons in vivo after injections in the mouse gastrocnemius. Intramuscular (i.m.) injection of CAV-2 poorly transduced skeletal muscle in adult and newborn mice. As a control, we injected AdGFP in the contralateral muscle and found a significant level of Ad-mediated gene transfer in myofibers. In the ipsilateral anterior horns of the sacral dorsolumbar rachis, we detected a fluorescent signal due to the CAVGFP transduction; in the contralateral anterior horns, essentially no fluorescence was detected. Sections of spinal cord were scanned, and we found several clusters of GFP⁺ cells with the morphology of motoneurons. These results demonstrate that after i.m. injection, CAV-2 preferentially transduced motoneurons and the retrograde transport of CAV-2 appeared to be 10-fold greater than Ad5 vectors.

3. The molecular basis for the neural tropism: CAR expression on neurons
We recently characterized the cell surface molecules used by CAV-2 in vitro and demonstrated that CAV-2 bound human CAR (coxsackie adenovirus receptor) and used it to transduce cells. More relevant, we showed that unlike Ad2/5, CAV-2 transduction was independent of 1) _Mß₂ integrins, 2) the ₂ domain of the MHC class I molecule, and probably 3) _vß_3/5 integrins. We compared the expression and distribution of CAR on differentiating spinal cord cultures, rat nasal cavity, and mouse skeletal muscle (Fig. 2 ).In primary spinal cord cultures, we found that CAR colocalized with MAP-2 (neurons): notably, CAR expression was primarily, if not exclusively, on neurons (i.e., not on astrocytes or oligodendrocytes). The differential and low level of CAR expression on airway epithelial cells in vivo is one of the reasons Ad-mediated gene transfer is inefficient in certain areas of the respiratory tract. We assayed the rat olfactory neuroepithelium for CAR expression and found that olfactory neurons express a significant level of CAR near the olfactory receptors. We did not detect CAR expression on sustentacular epithelial cells. CAR expression on mature skeletal muscle is low or undetectable. Due to the efficient transduction of motoneurons, we asked whether CAR was expressed near the neuromuscular junction. Similar to the spinal cord cultures, a significant level of CAR was found on the neurites of the innervating motoneurons. We also assayed adult and newborn rat brains for CAR distribution. CAR colocalized with NeuN, but a further definitive location was not possible. Our results suggested that CAR distribution in vivo resembled the pattern in primary cultures. These data strongly support our hypothesis that CAR distribution on the soma and neurites permits CAV-2 entry and axonal transported to other regions of the CNS.

View larger version (70K): [in this window] [in a new window]   Figure 2. CAR distribution in vitro and in vivo. To assay CAR expression and distribution, we costained A–C) spinal chord cultures with anti-MAP-2 (insert antisynapsin green, GFP in blue). D–F) The rat nasal epithelium with anti-OMP antibody (insert shows CAR/OMP colocalization on soma). G–I) Mouse diaphragm with antineurofilament (insert -bungarotoxin in green). Top is CAR in red; middle is test antibody in green and bottom row is the merge.

CONCLUSION

If one wanted to treat neurodegenerative diseases by modifying terminally differentiated neurons in the CNS, an attractive vector would preferentially transduce neurons, be transported efficiently to afferent structures, and be capable of long-term expression. Ideally, the vectors would have a large cloning capacity and be produced to high titers. Most important, the vector must be safe—i.e., nontoxic and not inhibited by, or able to exacerbate, a preexisting humoral or cellular immunity. The ability to express a transgene throughout the substantia nigra via injection into the striatum may be essential to potential therapy for certain neurodegenerative diseases. Diseases similar to MPS VII, where ß-glucuronidase expression is needed throughout the CNS and the enzyme can be secreted and sequestered by neighboring cells, may be especially attractive targets for CAV-2 vectors.

Initial trials with viral vectors in vitro and in animal models often appear promising but do not a priori translate into preferential transduction of human cells or tissue. The ex vivo transduction pattern of CAV-2 in human epileptogenic brain biopsies is therefore extremely encouraging and clinically relevant. A significant amount of interest is being devoted to the modification of viral vector tropism. Compared with Ad5, this is what CAV-2 does by targeting only one of the possible adenovirus receptors (i.e., CAR). The selective expression of CAR by neurons (in the CNS, muscle and nasal cavity) presumably enabled CAV-2 to preferentially target these cells. Testing CAV-2 vectors in CAR-deficient mice, if they can be generated, may allow a definitive response.

How the lack of transduction of glia and oligodendrocytes will affect the induced immune response is unclear. At high multiplicity of infections, Ad5 vectors are toxic to neurons and glia in vitro and in vivo and induce a T cell-independent and a T cell-mediated response in naive animals. Neuron-specific transduction may also avoid side effects resulting from the ectopic expression of the therapeutic genes and the MHC II presentation of the capsid proteins via transduced glia. In addition, approaches such as cell-specific restrictive silencers may also be incorporated in CAV-2 vectors in order to express transgenes in a subset of neurons (e.g., dopaminergic or adrenergic). Although Ad5 vectors generate relatively stable expression in the brain, peripheral infections activate macrophages/microglia and T lymphocytes and then target transduced cells. Furthermore, the extent of the immune response generated against the Ad capsid in patients who have been infected repeatedly with many different Ad serotypes, and therefore have a large repertoire of proliferative memory T cells, is unclear. In our opinion, the most clinically relevant test would be to immunize a nonhuman primate with several human Ad serotypes, inject with a gutless or high-capacity CAV-2 vector, followed by a peripheral infection with human Ads. The CAV-2 vectors used in this study may be effective in certain therapeutic applications, but it is unlikely that they will be an appropriate choice for treatment of diseases that demand long-term expression. Gutless, or high-capacity, vectors are the state-of-the-art with Ad vectors: they are deleted in the entire viral coding region and have a 36 kb cloning capacity. We recently generated and are now optimizing the production of gutless CAV-2 vectors (unpublished results). These gutless vectors may eventually be considered improved tools for the treatment of neurological disorders. Nonetheless, the CAV-2 vectors described here could be powerful tools to respond fundamental questions of virology, neurobiology, cell biology, and immunology concerning viral vectors.

View larger version (16K): [in this window] [in a new window]   Figure 3. CARs: CAV-2 TARGETING OF ENURONS IN VIVO AND EX VIVO

FOOTNOTES

¹ To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.01-0321fje; to cite this article, use FASEB J. (August 17, 2001) 10.1096/fj.01-0321fje

This article has been cited by other articles:

				G. Schoehn, M. El Bakkouri, C. M. S. Fabry, O. Billet, L. F. Estrozi, L. Le, D. T. Curiel, A. V. Kajava, R. W. H. Ruigrok, and E. J. Kremer Three-Dimensional Structure of Canine Adenovirus Serotype 2 Capsid J. Virol., April 1, 2008; 82(7): 3192 - 3203. [Abstract] [Full Text] [PDF]

				M. Perreau, F. Mennechet, N. Serratrice, J. N. Glasgow, D. T. Curiel, H. Wodrich, and E. J. Kremer Contrasting Effects of Human, Canine, and Hybrid Adenovirus Vectors on the Phenotypical and Functional Maturation of Human Dendritic Cells: Implications for Clinical Efficacy J. Virol., April 1, 2007; 81(7): 3272 - 3284. [Abstract] [Full Text] [PDF]

				E. Seiradake, H. Lortat-Jacob, O. Billet, E. J. Kremer, and S. Cusack Structural and Mutational Analysis of Human Ad37 and Canine Adenovirus 2 Fiber Heads in Complex with the D1 Domain of Coxsackie and Adenovirus Receptor J. Biol. Chem., November 3, 2006; 281(44): 33704 - 33716. [Abstract] [Full Text] [PDF]

				T. S. Hnasko, F. A. Perez, A. D. Scouras, E. A. Stoll, S. D. Gale, S. Luquet, P. E. M. Phillips, E. J. Kremer, and R. D. Palmiter Cre recombinase-mediated restoration of nigrostriatal dopamine in dopamine-deficient mice reverses hypophagia and bradykinesia PNAS, June 6, 2006; 103(23): 8858 - 8863. [Abstract] [Full Text] [PDF]

				A. Keriel, C. Rene, C. Galer, J. Zabner, and E. J. Kremer Canine Adenovirus Vectors for Lung-Directed Gene Transfer: Efficacy, Immune Response, and Duration of Transgene Expression Using Helper-Dependent Vectors J. Virol., February 1, 2006; 80(3): 1487 - 1496. [Abstract] [Full Text] [PDF]

				M. Perreau and E. J. Kremer Frequency, Proliferation, and Activation of Human Memory T Cells Induced by a Nonhuman Adenovirus J. Virol., December 1, 2005; 79(23): 14595 - 14605. [Abstract] [Full Text] [PDF]

				Y. Li, Y. Yang, and S. Wang Neuronal gene transfer by baculovirus-derived vectors accommodating a neurone-specific promoter Exp Physiol, January 1, 2005; 90(1): 39 - 44. [Abstract] [Full Text] [PDF]

				S. A. Kelkar, K. K. Pfister, R. G. Crystal, and P. L. Leopold Cytoplasmic Dynein Mediates Adenovirus Binding to Microtubules J. Virol., September 15, 2004; 78(18): 10122 - 10132. [Abstract] [Full Text] [PDF]

				K. Al-khatib, B. R. G. Williams, R. H. Silverman, W. Halford, and D. J. J. Carr Distinctive Roles for 2',5'-Oligoadenylate Synthetases and Double-Stranded RNA-Dependent Protein Kinase R in the In Vivo Antiviral Effect of an Adenoviral Vector Expressing Murine IFN-{beta} J. Immunol., May 1, 2004; 172(9): 5638 - 5647. [Abstract] [Full Text] [PDF]

				J. Morante-Oria, A. Carleton, B. Ortino, E. J. Kremer, A. Fairen, and P.-M. Lledo Subpallial origin of a population of projecting pioneer neurons during corticogenesis PNAS, October 14, 2003; 100(21): 12468 - 12473. [Abstract] [Full Text] [PDF]

This Article Full Text (PDF) All Versions of this Article: 15/12/2283 01-0321fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by SOUDAIS, C. Articles by KREMER, E. J. Search for Related Content PubMed PubMed Citation Articles by SOUDAIS, C. Articles by KREMER, E. J.