 |
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
|
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
Gene Therapy Laboratory, CHU Hotel-Dieu, Nantes, France.
Correspondence to Dr. Philippe Moullier, Laboratoire de Thérapie Génique, CHU Hotel-Dieu, 44000 Nantes, France. Phone: 33-240087490; Fax: 33-240087491; E-mail: moullier{at}sante.univ-nantes.fr
 |
Abstract |
|---|
 Top Abstract IntroductionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionReferences |
|---|
(TGF-), atrial natriuretic
peptide, and lipoxins] in the establishment and maintenance of pathologic
glomerular states. In vivo or ex vivo delivery of genes to the kidney has been well described (9). Gene delivery vehicles include (1) viral vectors derived from infectious but nonreplicating viruses, (2) nonviral vectors derived from liposome-based chemical structures, and (3) genetically modified cells that are administered to the animals, possibly with prior selection. The latter vehicles are not discussed here, but recent studies describe this approach (10, 11).
The purpose of using viral vectors is to retain their infectivity potential, but with significant attenuation of their replication potential. These results are obtained by deleting essential viral genes from the wild-type genome and inserting the therapeutic gene(s). The transgene can be expressed under an independent constitutive, tissue-specific, or regulatable promoter, which adds considerable flexibility. In many instances, more than one transgene can be included (expressed independently or linked by an internal ribosomal entry site sequence). Antisense mRNA can also be generated after cloning of the transgene, in reverse orientation, downstream of the promoter.
Retroviral vectors require the target cell to divide after transduction. The remaining viral motifs and the presence of the reverse transcriptase and integrase peptides within the recombinant vector particle allow stable integration into host chromosomes, with subsequent expression of the transgene. Insulators can even be cloned on both sides of the transgene to reduce chromosomal interference at the integration locus (12), resulting in long-lasting transgene expression in vivo. Because division of the target cells is required for efficient retroviral transduction and because the mature kidney has a low mitotic index, this vector has not been popular to achieve direct gene transfer in the kidney. Even during folic acid-induced tubular regeneration, retrovirus-mediated transduction was poor and strictly localized to the site of administration (1).
Adenoviral vectors have opposite characteristics, because they can infect quiescent cells and the recombinant genome remains extrachromosomal. However, because of the complexity of the wild-type viral genome, deletion of the essential viral genes is only partial in the first generations of recombinant adenoviral vectors. This results in the expression of viral epitopes in the genetically modified organs, with subsequent cytotoxic T lymphocyte activation and rapid elimination of the transduced cells. Importantly, in immunodeficient mice, transgene expression persists longer in the liver after a single intraportal administration (13), suggesting that, although the foreign DNA remains external to the host genome, it displays remarkable stability, at least in murine liver. In addition to these molecular features, adenoviral binding and internalization require specific receptors and co-receptors on the target cell surface (14, 15). Because their distribution and density may not be optimal, adenoviral vectors often require high multiplicities of infection, i.e., high concentrations of virions per target cell, to obtain significant transduction. This is an important point to consider, because the adenoviral particle itself produces toxic effects by disturbing cellular integrity and triggering coagulation/complement disorders when administered in vivo at high concentrations (16). However, several initial studies have documented limited transduction of rat kidney after both intrarenal and retrograde delivery (2, 17). The combination of cold incubation and administration of a vasoactive drug with the adenoviral vector allows retargeting of the transduction from cortical to medullary compartments (17). A significant improvement in transduction could be obtained with isolated perfusion of pig kidney for at least 2 h with a preservative solution containing the recombinant adenovirus. This resulted in transduction of a reporter gene in approximately 85% of the glomeruli but not in tubular epithelial cells (18).
Although adenoviral vectors in their current design exhibit too many limitations, such as the requirement for high-titer stocks for transient and limited transduction of renal tissue, the refinement of adenoviral vectors may resolve these difficulties. Recent improvements include the modification of adenoviral tropism (19, 20) and the generation of recombinant adenovirus with complete deletion of all viral open reading frames (21). Another approach is to generate in vitro an adenovirus-polylysine-DNA complex that relies on adenovirus inactivation and the preserved ability of the viral particle to disrupt endocytic vesicles in the target cells. This complex resulted in transduction of proximal tubular epithelial cells when perfused into isolated human kidneys (22).
Adeno-associated virus (AAV) is a parvovirus that has recently emerged as a
particularly attractive vector because it lacks in vivo toxicity and
has the ability to sustain long-term transduction of skeletal myofibers,
neurons, and retina (for review, see reference
(23). Several groups have
reported preclinical data for nonhuman primates
(24,
25) and, more recently, for
patients with hemophilia (26),
suggesting that AAV is indicated when a therapeutic protein is required
systemically. Despite the increasing use of AAV vectors in various organs and
disease models, only one report described in vivo AAV delivery by
renal intraparenchymal injection in mice, which was associated with limited
transduction (for at least 3 mo) in tubule epithelial cells surrounding the
injection site (27). However,
in a murine model of a lysosomal storage disease, an important result was
obtained after neonatal gene transfer using an AAV vector, showing that
transgene cDNA and -glucuronidase activity persisted for at least 16 wk
in multiple organs, including kidney
(28). This result suggests
that, at least at the mouse neonatal stage, one or several renal compartments
are susceptible to long-term AAV transduction. Whether recombinant AAV
actually represents a relevant alternative to adenovirus for gene transfer in
the kidney remains to be explored in a more comprehensive manner.
The same remark can be made regarding the recently developed lentiviral vectors, with which safety is significantly improved because (1) they can be derived from several different species, (2) the majority of the original viral open reading frames are removed, and (3) the original envelope protein is deleted and recombinant particles can be pseudotyped with the vesicular stomatitis virus envelope G protein. Although lentiviral vectors have recently exhibited success in vivo in murine hepatocytes, central nervous system, and lung, there is no evidence that the kidney would also be a potential target. In conclusion, efficient and clinically relevant vectors for in situ gene transfer in the kidney still need to be developed and tested using animal models, to improve transduction and limit immunologic side effects.
A possible alternative for in situ gene therapy of renal diseases
involves delivery of the therapeutic protein from an extrarenal site to the
renal compartments. Mice with spontaneous or induced hypertension were treated
using adenoviral vectors containing kallistatin cDNA
(29) or kallikrein cDNA
(30,
31). The vectors were
administered via a peripheral vein, with subsequent partial and transient
phenotypic correction. Atrial natriuretic peptide cDNA was also expressed from
an adenoviral vector and attenuated gentamicin-induced nephrotoxicity in rats
(5). Renal pathologic features
observed in lysosomal storage diseases were improved after intravenous
delivery of recombinant adenovirus containing the -glucuronidase
(32) and 
-galactosidase
(33) genes, in the
mucopolysaccharidosis type VII and Fabry mouse models, respectively. In all of
these reports, the biologic effects were likely obtained through transduction
of the easily accessible liver, which acted as a source for the therapeutic
protein.
Intraperitoneal administration of a recombinant adenovirus encoding the human erythropoietin (Epo) gene transiently improved renal anemia in DBA/2FG-pcy mice with chronic renal failure (34). Importantly, long-term regulated expression of the murine Epo gene was achieved after a single intramuscular injection of recombinant AAV containing Epo cDNA under the control of a tetracycline-inducible system (35).
A special comment should be made regarding the hemagglutinating virus of
Japan (HVJ) liposome vector. It is considered a chimeric vector, consisting of
a lipid mixture associated with the inactivated Sendai virus. The viral
epitopes provide effective ways to bind to the ubiquitously distributed HVJ
receptors and to fuse with the lipid bilayer at the cell surface and in
intracytoplasmic vesicles
(36). In addition,
high-mobility group 1 protein (a nonhistone chromosomal protein) is complexed
to the foreign DNA before in vitro HVJ liposome assembly, to
facilitate nuclear translocation
(37). HVJ liposome vectors
have been relatively well evaluated for therapeutic approaches to experimental
renal diseases. Glomerulosclerosis is suspected to be triggered by
dysregulation of various growth factors and cytokines in the kidney, including
TGF-, platelet-derived growth factor, fibroblast growth factor, and
endothelin-1 (38). In
particular, TGF- plays a critical role in the progression of
glomerulosclerosis, and inhibition of TGF- by the proteoglycan decorin
suppressed extracellular matrix accumulation in the anti-Thy-1 model of
experimental glomerulonephritis
(6). Decorin cDNA, in HVJ
liposomes, was injected into the skeletal muscle, which acted as an external
source of decorin after transduction. That study provided immunofluorescence
evidence of the circulating decorin in the glomeruli, in association with
reduced extracellular matrix accumulation and proteinuria. Those results were
in agreement with previous reports that demonstrated protection against
scarring in experimental kidney disease after inhibition of TGF- by
antisense oligonucleotides
(39) or injection of
recombinant decorin and anti-TGF- antibodies
(40). This approach had the
advantage of being highly relevant for the clinical development of a novel
treatment for fibrotic diseases caused by TGF-1. More recently, a
similar approach was used to create an efficient inhibitor with the capacity
to block the binding of TGF- to type II receptors, thus preventing
subsequent signal transduction
(7). A cDNA construct encoding
chimeric TGFRII/Fc (consisting of the extracellular portion of the
TGF- type II receptor fused to an Ig heavy-chain Fc fragment) was placed
in HVJ liposomes and injected into the gluteal muscles of nephritic rats.
Genetically modified muscle was able to sustain secretion of the
TGFRII/Fc protein, with concomitant suppression of glomerular TGF-
production and reduction of extracellular matrix accumulation.
HVJ liposomes were also used to directly transduce rat glomeruli for approximately 7 d (41). The renin and angiotensinogen genes were expressed in glomeruli, to investigate the local effects of angiotensin II upregulation. Thirty percent of the glomeruli were transduced after 3 d, with simultaneous extracellular matrix expansion of the mesangial area. That study is in agreement with a more recent one in which HVJ-mediated transduction of rat kidneys with 15-lipoxygenase cDNA suppressed inflammation and preserved function in passive and accelerated anti-glomerular basement membrane-induced nephritis models (8). On the basis of these reports and others, HVJ liposomes seem to be a fairly powerful tool for transient expression of foreign genes, for therapeutic intervention or investigation of renal pathophysiologic processes.
For a better understanding of the difficulties presented by the use of viral vectors in the kidney, a few comments regarding the recent development of nonviral vectors are necessary. An important report recently illustrated the requirement for glomerular filtration for transduction of proximal tubular cells in rat kidneys after injection of nonviral DNA complexes into the renal artery (42). FITC-pLys-containing lipopolysaccharide complexes were used to detect DNA particles in vivo. This elegant method established that nonviral vectors are able to transduce proximal tubular cells only when filtered through the glomerular basement membrane. Furthermore, DNA complexes that do not undergo glomerular filtration remain unable to transduce the nephron via the peritubular capillaries. It is tempting to extend these findings by suggesting a model in which DNA complexes that have reached the urinary space are actively reabsorbed in the proximal tubules, which would favor random transduction events. In other words, one simple strategy to genetically modify the nephron would be to establish the physicochemical properties of DNA complexes for optimal glomerular filtration and subsequent tubular reabsorption. A paradox could be that the proximal nephron is an appropriate target for nonviral vector-mediated gene transfer [or transduction of small (<70-nm) viral particles, such as AAV]. Indeed, glomerular filtration represents the first concentration step leading to high vector loads at the proximal tubule level, where reabsorptive activity is naturally important. Therefore, vector uptake may be optimal at this level. This hypothesis would be in agreement with several studies that unambiguously demonstrated that renal artery administration or intra-renal pelvic infusion of cationic lipid-based vectors encoding a reporter gene resulted in transient transduction of the tubules but not the glomerular, vascular, or interstitial compartments of the kidney (43, 44). An important study demonstrated that retrograde injection of cationic liposomes complexed with carbonic anhydrase II (CAII) cDNA into the renal pelvis of CAII-deficient mice resulted in expression of CAII in the kidney for up to 1 mo. Furthermore, after gene transfer, CAII-deficient mice regained the ability to acidify urine after oral administration of ammonium chloride (45). A careful immunohistochemical evaluation using an anti-CAII antibody demonstrated that the CAII gene was expressed only in tubular cells of the outer medulla and corticomedullary junction.
Gene transfer into the kidney has presented novel avenues for nephrologists interested in the pathophysiologic processes of renal diseases. It has also triggered dreams of new therapeutic approaches. Obviously, there are still major obstacles to therapeutic applications, and development of a large database on vectors and transfer methods in animal models remains a priority. The most recently developed viral vectors, including recombinant AAV and lentiviruses, have not yet been evaluated in the kidney. On the basis of their characteristics in other tissues, these vectors may allow transduction of quiescent renal cells and may provide long-term expression of therapeutic genes.
|
Acknowledgments |
|---|
|
References |
|---|
|
TopAbstractIntroductionReferences |
|---|
1
expression by antisense oligonucleotides suppressed extracellular matrix
accumulation in experimental glomerulonephritis. Kidney
Int 50: 148-155,1996[Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
HOME
CURRENT ISSUE
ARCHIVES
JASN Express
ONLINE SUBMISSION
AUTHOR INFO
EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP |
