Gene Therapy for Metabolic Disease

Anthony T. Cheung

Timothy J. Kieffer

The revolution of molecular genetics is ushering in new paradigms in the treatment of disease that could previously only be imagined. Gene therapy is eagerly anticipated to play a dominant role in future treatments for numerous diseases. Gene therapy involves modifying the expression of an individual’s genes or correction of abnormal genes to treat disease. It is currently being applied to several diseases, including cystic fibrosis, cardiovascular disease, infectious diseases, and cancer. Metabolic diseases that are treatable by the secretion of therapeutic proteins into the blood for distribution throughout the body are particularly attractive candidates for gene therapy since there are many possible target tissues for genetic engineering. Automatic physiologic release of therapeutic proteins by cells within the body would be a welcome replacement for conventional protein delivery methods such as needle injections. Although currently an experimental form of medical treatment, there continue to be significant advances in this exciting technology such that treating metabolic diseases by gene therapy appears increasingly possible. However, several significant hurdles remain, not the least of which is the development of safe and effective vectors for gene delivery. Metabolic diseases such as type 1 diabetes mellitus (DM) present additional challenges, because although the disease is ultimately curable by the production of insulin, both the quantity and timing of insulin release into the blood must be exquisitely coordinated with meals in order to reestablish normal glucose homeostasis. This chapter reviews the gene delivery systems under development for gene therapy and focuses on strategies being investigated to treat DM by insulin replacement.

Gene Delivery Technology

The clinical success of gene therapy fundamentally depends on the availability of effective and safe gene delivery systems. There are two basic approaches for delivery of genes: in vivo and ex vivo methods. With the in vivo strategy, vectors carrying the therapeutic gene are administered directly into the circulation or target tissues. Alternatively, in ex vivo methods, the target cells or tissues are first isolated from the patient, cultured, and transfected with the therapeutic gene. The transduced cells are then selected and returned to the patients. Gene vector development has been a primary focus of gene therapy research during the past decade. This effort has generated a broad arsenal of gene transfer systems that continue to be optimized for various clinical applications. Currently, there is no single universal vector system for gene therapy; each vector system has its unique strengths and weaknesses. The selection of a suitable vector depends on criteria such as therapeutic strategy (e.g., ex vivo vs. in vivo), target tissue, transgene size, and the desired duration and levels of gene expression. Current gene delivery systems can be broadly segmented into viral- and nonviral-based vectors. Salient characteristics of the different vector systems are summarized in Table 49.1.

Viral Vectors

Viruses are naturally evolved carriers of genetic material into cells. This property has been exploited by converting them into vehicles for gene therapy. To generate vectors, wild-type viruses are genetically modified to be replication incompetent and nonpathogenic while maintaining their infectivity. To achieve this, some or all of the critical genes required for viral replication and particle assembly are deleted and subsequently replaced with a therapeutic gene. The recombinant viral vectors are assembled in so-called packaging cells in which the deleted viral genes are provided in trans. Viral vectors are then harvested from the cells and purified. There are four main types of viruses currently in use for gene therapy: retroviruses, adeno-associated viruses (AAVs), herpes simplex virus (HSV), and adenoviruses.

Retroviral Vectors

Retrovirus-based vectors are presently the most widely used vectors in clinical trials, with use in over 1,500 human patients (1). Attractive features of retrovirus include the relatively large capacity for foreign DNA (∼7 kb) and the ability to provide stable integration of genetic material into the host genome, thereby allowing for long-term transgene expression (reviewed in reference 2). There are three types of retroviruses, namely, oncoretrovirus, lentivirus, and foamy virus. The most commonly used retroviral vector is based on the murine leukemia virus (MLV), which is an oncoretrovirus. MLV-based vectors

P.748

P.749

require cell division for successful transduction and thus are restricted to target cells that are actively proliferating (2). The recent development of lentiviral vectors, which are capable of transducing both dividing and nondividing cells, has shown promise in overcoming this restriction. The two most widely studied lentiviral vectors are based on feline immunodeficiency virus (FIV) (3) and human immunodeficiency virus (HIV) (4). Another attractive feature of retroviral vectors is that their envelop glycoproteins (Env), which govern their tissue tropism through interactions with specific membrane receptors on target cells, can be replaced by other viral Env in a process known as pseudotyping. With this approach, the host range of retroviral vectors can be expanded or restricted by supplying the env sequences from other viruses during viral assembly in packaging cells (5).

Table 49.1. Salient features of gene delivery vectors commonly used for gene therapies

Vector	Characteristics	Insert size	Stability of expression	Potential disadvantages
Oncoretrovirus (MLV-based)	Titers: 10⁷–10⁹ cfu/mL Broad cell tropism Only infect dividing cells Low immunogenicity and toxicity Used in over 1,500 patients	< 10 kb	Stable due to integration into host genome	Risk for mutagenesis due to random genome insertion Possible formation of RCR Prone to inactivation by complements
Lentivirus (HIV-based)	Titers: 10⁷–10⁹ cfu/mL Broad cell tropism Infect both dividing and quiescent cells Low immunogenicity and toxicity Yet to be tested in humans	< 10 kb	Stable due to integration into host genome	Risk for mutagenesis due to random genome insertion Possible formation of RCR Serum conversion to HIV
Lentivirus (FIV-based)	Titers: 10⁷–10⁹ cfu/mL Broad cell tropism Infect both dividing and quiescent cells Low immunogenicity and toxicity Nonpathogenic in humans Yet to be tested in humans	< 10 kb	Stable due to integration into host genome	Risk for mutagenesis due to random genome insertion Possible formation of RCR
AAV	Titers: 10¹²–10¹³ particle/mL Broad cell tropism Infect dividing and quiescent cells Simple viral genome Widely tested in humans	< 5 kb	Stable; unclear if DNA integrates in vivo	Limited packaging capacity for transgene Risk for mutagenesis due to the random genome insertion Immunogenic
HSV	Titers: 10⁸–10¹⁰ cfu/mL Infects dividing and nondividing cells Strong tropism for neuronal cells Does not integrate into host genome Limited testingi n humans	> 25 kb	Stable; DNA maintained episomally	Immunogenic Complex viral genome to manipulate Restricted tropism
Adenovirus	Titers: 10¹¹–10¹² pfu/mL Broad cell tropism Infects dividing and quiescent cells Rapid and high levels of gene expression Widely tested in humans	∼8 kb; early generations E1-deleted vectors ∼30 kb; gutless vectors	Transient (3–4 wk); DNA maintained episomally	Immunogenic Transient gene expression Complicated viral genome
Liposomes	Transfect dividing and quiescent cells Noninfectious Low toxicity and nonpathogenic Suitable for delivery of oligonucleotides Easily produced in large scale	No limit	Transient	Low transfection efficiency in vivo Transient gene expression Immunogenic risk due to reactions to unmethylated CpG motifs Risk of insertional mutagenesis
MLV, murine leukemia virus; cfu, colony-forming unit; RCR, replication-competent retrovirus; HIV, human immunodeficiency virus; pfu, plaque-forming unit.

There are two major safety concerns in using retroviral vectors for gene therapy in humans: (a) potential formation of replication competent retroviruses (RCRs), and (b) potential insertional mutagenesis. Recombination events that occur during vector packaging can lead to generation of RCRs. Nevertheless, this risk has been reduced significantly through refinements of the vector packaging system. For example, the retroviral genome has been broken down into multiple plasmids with homologous sequences removed. Additionally, vector stocks are produced by transient transfection of packaging cells. Thus, formation of RCR would require the unlikely simultaneous rearrangement among multiple plasmids in a specific orientation in a short period of time. However, it remains possible that RCRs could be formed by recombination between vectors and human endogenous retroviruses in patients. Thus far, no such incidence has been reported, and deleting key viral genes from the retroviral vector sequence can further minimize this risk. A second safety concern is related to the random insertion of retroviral vectors into host genomes. This property has the potential to alter the expression of oncogenes and tumor suppressor genes, which could lead to neoplasia. A study exploring this risk found that the transformation rate in mouse fibroblasts transduced with retroviral vectors was not different from the frequency seen in untreated cells (6). However, the development of a leukemia-like disease in two children with the deadly X-linked severe combined immune deficiency (X-SCID), treated with early generation MLV-based vectors, has heightened concerns over the safety of integrative viruses (7). The rare devastating disease in these subjects results from disrupted expression of the γ-chain cytokine receptor subunit (γc) due to a genetic mutation. The resulting impairment in immune system function causes severe predisposition to infection and is fatal without successful reconstitution of the child’s immune system. Pioneering gene therapy trials have demonstrated that hematopoietic progenitor cells from these patients can be corrected by retrovirally mediated ex vivo transfer of the γc gene (8). Although retroviral vector–related activation of a protooncogene has been detected in subjects from one trial (9), it is currently premature to conclude that the carcinogenic events in this trial are solely attributed to the vector-induced mutagenesis (7,10). There are factors specific to this trial that suggest insertional mutagenesis alone is probably not enough to induce neoplasia, because carcinogenesis is a multifactorial process. For example, other trials in SCID using the same ex vivo therapeutic strategy and viral vector, but different transgene, have not observed any such adverse event (11,12). The transgene used in the X-SCID trial, which encodes a T-cell growth-promoting cytokine receptor, together with the phenomenon of selective advantage inherent in the ex vivo therapeutic strategy, may have promoted the transformation (13). Additionally, one of the patients who developed leukemia has two relatives with childhood cancers, indicating potential inherited susceptibility. More in-depth studies and monitoring will need to be performed to better manage the risks of using retroviral vectors in human gene therapy.

Adeno-Associated Viral Vectors

Adeno-associated viruses are small single-stranded DNA viruses from the human parvovirus family that require a helper virus, such as adenovirus or herpes simplex virus for replication (reviewed in reference 14). There are currently at least eight known serotypes of AAV, each with somewhat different tropisms. Although the majority of the population is seropositive for AAV, no pathology or toxicity has been associated with it in humans, making it an attractive vector candidate for gene therapy. The AAV genome is relatively simple, consisting of two genes (rep and cap) flanked by inverted terminal repeats (ITRs). AAV-based vectors are produced by replacing the two AAV genes with a therapeutic gene cassette. Vector packaging is then carried out in cells cotransfected with plasmids encoding the rep and cap genes, the essential adenoviral helper genes, and the therapeutic gene cassette. Because AAV vectors lack viral coding sequence, they have an excellent safety profile. In fact, no vector toxicity has been linked to an AAV vector to date. Wild-type AAVs infect both dividing and nondividing cells and incorporate into a defined region on host chromosome 19 (15). However, recombinant AAV vectors, in the absence of Rep proteins, do not integrate site specifically (15). Despite observations of stable transgene expression by AAV vectors for months in vivo (16), it is still unclear if these vectors integrate randomly into the host genomes or are maintained episomally (17). A current major limitation of AAV vectors is their relatively small packaging capacity (<5 kb). In addition, capsids of AAV vectors have been shown to elicit neutralizing antibodies that may limit their application in therapies in which readministration is required (18). Because AAV vectors have the potential to randomly integrate into host genome, the risk of insertional mutagenesis also exists. Similar to other viral vector systems, the potential for generating wild-type AAVs by recombination during vector production remains a risk. Although AAVs may not be suitable for delivery of genes to all cell types, therapeutically significant levels of transgene expression in vivo have been achieved in muscle (19), liver (20), brain (21), lung (22), and intestine (23). Notably, human clinical trials with AAV vectors for treating chronic diseases such as cystic fibrosis, hemophilia, and muscular dystrophy are under way.

Herpes Simplex Viral Vectors

Herpes simplex virus is a human neurotropic virus with the unique ability to persist latently within infected neuronal cells,

P.750

without chromosomal integration and without being rejected by the immune system (24). Consequently, there is a growing interest in converting HSV into vectors for persistent gene transfer to the nervous system. Despite its tropism for neuronal tissue, HSV can infect other tissues, including skin, muscle, and liver (25). The HSV genome is a large (∼152 kb) complex linear double-stranded DNA molecule consisting of over 80 gene products, of which over half are not required for viral replication in cell culture (2). After deletion of these nonessential genes, HSV vector can accommodate up to 50 kb of foreign genetic material. Such an impressive genetic payload may be advantageous in gene therapy for multigenic diseases such as diabetes, where simultaneous expression of several genes may be required for curative effects. HSV vectors have shown promise in gene therapy for neurologic disorders and pain control (26) and have been successfully deployed in delivering genes to treat diabetic neuropathy in rats (27). Currently, information on the safety and effectiveness of HSV vectors in humans is not available.

Adenoviral Vectors

Adenoviruses consist of a linear DNA genome of approximately 36 kb that is encapsidated with an icosahedral-shaped protein shell. There are more than 50 different serotypes of mammalian adenovirus (reviewed in reference 28). Current adenoviral vectors are mostly based on serotypes 2 and 5 because they, unlike other serotypes, are not associated with any severe pathologies and do not cause cancers in humans. Early first-generation adenoviral vectors are rendered replication defective by replacing the E1 region in the viral genome with the foreign gene of interest. The E1 region encodes several early gene products essential for initiation of the viral life cycle. Viral vectors are packaged by transfecting the E1-deleted viral genome containing the transgene into cell lines stably expressing the E1 gene products (29). The first clinical application of adenoviral vectors was performed in 1995 in patients with cystic fibrosis (30). Currently, this vector is being used in more than 70 human clinical trials worldwide for treating diseases ranging from cancer to cardiovascular disorders. The ability of adenoviral vectors to provide fast and robust transgene expression in a broad spectrum of cells, regardless of their proliferative state, makes them an attractive vector for gene therapy. Other advantages of this viral vector system include the capability to produce high-titer viral stocks and the large packaging capacity for therapeutic genes. The strong hepatic tropism of serotype 5 adenoviral vectors also makes this vector attractive to gene therapy for metabolic diseases associated with the liver.

There are several drawbacks that restrict the applications of adenoviral vectors in human gene therapy. First, the viral particles are highly immunogenic in immunocompetent hosts. In addition to inducing potentially dangerous inflammatory responses, vector immunogenicity can result in destruction of the transduced cells, leading to loss of transgene expression, and also can prevent readministration of the same vector (31). This immunogenicity derives largely from the expression of adenoviral proteins in transduced cells (32,33,34). Notably, the first reported fatality in gene therapy was caused by excessive immune reaction to the adenoviral vectors used in a patient with ornithine transcarbamylase deficiency (35). To decrease immune response, “gutless” adenoviral vectors have been generated whereby genes encoding adenoviral proteins in addition to E1 have been removed (36). These gutless vectors show significant reduction in toxicity and immunogenicity in immunocompetent animals, but they have not resulted in stable transgene expression in vivo (37,38,39). Some of the adenoviral proteins may be required for extrachromosomal maintenance of the vector DNA in cells. Although gutless vectors show substantial improvement over earlier generation of adenoviral vectors, their application is somewhat limited by their cumbersome production and purification requirements. The use of helper virus in gutless vector generation also increases the risk for recombination and formation of replicative viral particles (37). Secondly, adenoviral vectors are nonintegrative; therefore, they only permit transient transgene expression. The inability of adenoviral vector to maintain stable gene expression, as a result of immunogenicity and lack of genome integration, impedes the application of this viral vector in many chronic metabolic diseases.

Nonviral Vectors

Nonviral systems are not infectious, exhibit low toxicity, and circumvent the risk for replication-competent virus formation associated with the use of viral vectors (reviewed in reference 40). Nonviral vectors are particularly suited to deliver small nucleotide sequences, which render them excellent vectors for therapeutic strategies involving small interfering RNA (siRNA) and antisense (41,42). Gene delivery systems based on nonviral vectors can be categorized into two general techniques: (a) mechanical administration of naked DNA, and (b) chemical carrier-mediated delivery. The simplest method of gene delivery is direct injection of naked DNA plasmids into the target tissue or the circulation. However, due to the presence of nucleases and the scavenging mechanism of the immune system, transgene expression achieved by this simple strategy in vivo is generally not as effective as that obtained by viral-mediated gene delivery systems. Various mechanical methods have been used to improve the delivery efficiency for naked DNA into target tissues. Controlled electric fields have been used in a process known as electroporation to enhance DNA uptake into cells (43). In addition, a bioballistic approach (e.g., gene gun) has been applied to bombard gold particles coated with the DNA of interest directly into target cells (44). More recently, ultrasonography has been used to increase the permeability of cell membranes to plasmid DNA, resulting in enhanced gene expression (45). Because ultrasonography is safe, noninvasive, and flexible, it offers great potential in clinical gene delivery applications. Chemical carriers are also commonly used in combination with plasmid DNA to achieve functional gene delivery to cells in vivo. Among these carriers, liposome-based reagents are the most widely used. Increasing effort is now being placed on optimizing the formulation of existing liposomes for specific in vivo applications. For instance, redox-sensitive thiol groups have been incorporated into lipid molecules such that the DNA-containing liposomes disintegrate under the reductive environment in

P.751

the cytosol, thus increasing the delivery of DNA inside the cell (46). Liposomes have also been used in combination with viral vectors to improve gene delivery efficiency (47). Other newer DNA chemical carriers under investigation include low toxicity and biodegradable polymers (48). Although many of these new derivatives show improved transfection efficiency in vitro, their pharmacokinetics, stability, and distribution in vivo are yet to be carefully characterized.

The levels of gene expression achieved by nonviral methods of gene delivery in vivo often fall short for many therapeutic applications. Low transfection efficiency and lack of sustained gene expression continue to impede the use of nonviral vectors in clinical gene therapy. Although it is generally accepted that nonviral systems produce fewer immune responses and have lower risk for insertional mutagenesis than viral vectors, these problems still exist. Plasmid DNA can elicit strong immune responses (49), which can be enhanced by liposome-induced induction of proinflammatory cytokine production (50). Indeed, an immune response to lipid-DNA conjugates is believed to be one of the main causes of failure in some gene therapy trials for cystic fibrosis that used liposome-mediated delivery of DNA (51,52). Some DNA plasmids can integrate spontaneously into the host genome, thus posing the risk for insertional mutagenesis (53). However, this risk can be reduced by using nonintegrative or episomal plasmids (e.g., Epstein-Barr virus–based plasmids).

Future Directions in Vector Development

To make vectors safer and more efficient for clinical applications, a number of promising strategies and concepts have recently been introduced. Future directions in vector development are generally focused on minimizing the risk for insertional mutagenesis, improving gene expression, and generating cell-specific targeting vectors. Site-directed integration of a transgene into the host genome could provide long-term gene expression while minimizing the risk for mutagenesis. To this end, integrase from bacteriophage φC31 has been successfully modified by directed evolution to achieve integration of therapeutic genes in defined location within mammalian genomes (54). This genomic integration system has been applied to deliver the human clotting factor IX (hFIX) gene to mouse liver using nonviral high-pressure tail vein injection methods (55). High efficiency of integration was achieved in this study, resulting in therapeutic plasma hFIX levels that remained stable for over 8 months. Notably, this site-specific DNA integrating system can be incorporated into other nonviral systems and nonintegrative viral vectors (e.g., adenoviral and HSV vectors). Another approach to achieve site-specific genomic integration is by making hybrid viral vectors. For example, nonintegrative adenovirus or HSV vectors could be used to deliver a transgene packaged in AAV plasmid to cells while simultaneously providing transient expression of AAV Rep proteins (56). In this way, the transgene-carrying AAV vector can be integrated into a specific region in human chromosome 19. However, high levels of intracellular Rep proteins can be cytotoxic (57). Therefore, both proper timing and expression levels of Rep proteins need to be identified before such hybrid vectors can be applied in human gene therapy.

Inefficient cellular uptake of vectors is a major hurdle in the development of gene delivery systems. To achieve better cell entry, peptide ligands that bind to cell surface receptors have been incorporated into nonviral (58) and viral (59) systems to improve vector uptake into tissues. Recently, small polybasic peptides derived from transducing domains (TDs) of certain proteins, such as the HIV Tat and Drosophila antennapedia homeobox proteins, have been shown to enter a variety of cells by a receptor- and energy-independent process (60). These cell-permeable peptides have been deployed to carry DNA (60) and viral vectors (61) into the cytoplasm of cells both in vitro and in vivo. Furthermore, Mi et al. have succeeded in using phage display technology to evolve new TD peptides that were able to transduce specific cell types with higher efficacy than the wild-type TD from Tat protein (62). Notably, a specific 12-mer TD peptide that was able to transduce human islets efficiently in culture was identified with the same strategy (62). In addition to improving gene transduction efficiency, this peptide-mediated gene delivery approach also has the potential to provide cell-specific gene transfer.

Gene therapy would be greatly facilitated by the availability of targeted vectors that could specifically restrict gene delivery or expression to certain cells or tissues in vivo. These vectors are expected to avoid undesirable side effects due to ectopic transgene expression and limit germ-line transmission of the transgene. Targeted in vivo gene delivery can be achieved by two different approaches. A tissue-specific promoter attached upstream to the transgene can be used to restrict expression of therapeutic genes to specific cell types. Alternatively, gene vectors can be engineered such that they selectively bind to certain target cells. Various genetic or immunologic methods have been devised to alter the binding specificity of viral vectors (reviewed in references 63 and 64). With genetic strategies, cellular receptor binding domains on the surface of viral particles are either genetically modified or replaced. Immunologic retargeting strategies are based on the use of bifunctional adapter conjugates, typically between an antibody that interacts with a viral coat protein and a targeting ligand or an antibody that binds to the target cells. Although these gene-targeting strategies have been shown feasible in cellular and small animal models, their specificities still require improvement and the practicality to produce these vectors for clinical use in humans has yet to be addressed.

Gene Therapy for Metabolic Diseases

The first human gene therapy trial was conducted in 1989, a study in which tumor-infiltrating lymphocytes were modified to express a marker gene (neomycin) by retroviral gene transduction before infusion into patients with metastatic melanoma (65). Many other human gene transfer studies followed for a variety of diseases. Currently, there are more than 600 gene therapy clinical protocols under investigation, involving more than 3,000 patients worldwide (1). Although gene therapy has

P.752

been tested as a treatment regimen for a broad spectrum of human diseases, the majority (∼63%) of ongoing clinical trials target cancer. Thus far, the number of clinical protocols in gene therapy for metabolic diseases is limited. Furthermore, the targeted metabolic diseases are monogenic disorders that can be corrected by replacing the nonfunctional gene. One of the first metabolic diseases tested for gene therapy in humans was familial hypercholesterolemia (FH). FH is an autosomal-codominant disorder resulting in low-density lipoprotein (LDL) receptor deficiency that affects 1 in every 500 individuals in North America and Europe. The compromised LDL receptor function leads to poor clearance of LDL from the liver of affected individuals. As a result, FH causes severe hypercholesterolemia and premature cardiovascular disease leading to death by early adulthood in the absence of treatment. Owing to the lack of an effective treatment and its well-defined etiology and lethality, FH is a good candidate for gene therapy. The defect in FH is amenable to both in vivo transfer of LDL receptor gene to the liver and implantation of autologous hepatocytes transfected with the receptor gene. The latter ex vivo approach was used in conjunction with retroviral vectors in the first human clinical trials for FH; and a 6% to 23% reduction of serum LDL levels was achieved in three of the five treated subjects (66). Liver-directed in vivo delivery of the LDL receptor gene is yet to be attempted in humans, but therapeutic efficacy of this approach has been shown in animal models of FH using either nonviral or adenoviral vector systems (67). However, the therapeutic effects achieved by these in vivo strategies were short-lived due to either poor transfection efficiency or host immune response to the viral vectors. In light of recent reports on the efficiency of integrative lentiviral and AAV vectors in providing stable gene expression in the liver through in vivo delivery, the safety and feasibility of using such vectors in FH gene therapy should be further evaluated.

Other metabolic diseases currently undergoing clinical trials in humans are Gaucher disease and Hunter syndrome (mucopolysaccharidosis II). These studies are in early phases of clinical trials; therefore, their efficacies in humans are still unclear. Gaucher disease is an inherited disorder in which harmful quantities of a fatty substance called glucocerebroside accumulate in the spleen, liver, lungs, bone marrow, and, in some cases, the brain, leading to an abnormal increase in organ volume and organ dysfunction. Present clinical protocol uses a retroviral vector to transfer the human glucocerebrosidase (GC) gene into peripheral blood stem cells (PBSCs) obtained from patients with Gaucher disease (68). Transduced PBSCs with normal GC activity are selected and transplanted back to the patients. A similar ex vivo gene therapy protocol is currently being conducted in patients with mild Hunter syndrome. This metabolic disorder is characterized by the deficiency of the lysosomal enzyme, iduronate-2-sulfatase (IS), resulting in defective degradation of glycosaminoglycans. Clinical manifestations of this disease include hepatosplenomegaly, coarse facial features, and elevated urine excretion of glycosaminoglycans. In this trial, T-lymphocytes are harvested from test patients, transduced with retroviral vectors carrying the IS gene, and then reinfused back into the patients (68).

An increasing number of preclinical studies show that gene therapy may be suitable for other monogenic metabolic diseases. Fabry disease is caused by deficient activity of the lysosomal enzyme α-galactosidase A (α-gal A). Onset typically occurs during childhood or adolescence, with episodes of severe pain in the extremities, skin lesions, and cornea problems. Affected individuals usually die in midlife from kidney, heart, or cerebrovascular complications. Takahashi et al. recently demonstrated that muscle-directed in vivo transfer of the α-gal A gene using AAV vectors showed long-term therapeutic effects in a mouse disease model (69). In addition, Ponder et al. succeeded in using gene therapy to provide effective long-term treatment in a canine model of mucopolysaccharidosis VII (MPS VII) (70). Retroviral vectors expressing the canine β-glucuronidase were injected intravenously into neonatal dogs with MPS VII. Treated animals showed significant health improvements that lasted for 17 months (70).

Gene Therapy for Diabetes

Diabetes mellitus is a complex multigenic disease characterized by the body’s inability to produce sufficient amounts of insulin to maintain normal glycemia (71,72). In type 1 DM, there is an absence of endogenous insulin production as a result of autoimmune destruction of the pancreatic β-cells. Type 2 DM is characterized by at least two major metabolic perturbations: insulin deficiency and insulin resistance. Although there are multiple treatment options available today for treating DM, death rates from diabetes have paradoxically increased and the life expectancy of people with diabetes remains 10 to 15 years shorter with overall mortality more than twice that of the general population (73). This disappointing outcome is partly due to the lack of significant improvement in therapies for diabetes at a time when the prevalence of diabetes continues to increase sharply. Improved understanding in the molecular basis of DM and the availability of genetic engineering technologies could significantly impact future strategies to treat the disease. Indeed, there are several gene therapy approaches for DM currently under investigation, including preservation of pancreatic β-cells (74,75), improvement of insulin sensitivity (76,77,78), treatment of diabetic complications (27), and insulin replacement (79). By far the most extensively pursued approach has been restoration of normal glucose homeostasis through the production of insulin from surrogate cells. Although insulin replacement by this method can readily be achieved by gene therapy techniques, the hurdle remains delivering the appropriate amount of insulin at the correct time.

The explicit goal of treating diabetes is to achieve blood glucose as close to normal as possible. The validity of this approach was confirmed by the Diabetes Control and Complications Trial (DCCT) and United Kingdom Prospective Diabetes Study (UKPDS), which convincingly demonstrated that good glycemic control significantly reduces complications in patients with diabetes (80,81). More recently, there is growing clinical evidence indicating that stringent management of postprandial glycemia in patients with diabetes provides a better outcome than controlling fasting hyperglycemia alone (82,83). These findings underscore the importance of early postprandial insulin secretion in minimizing glycemic excursions and diabetic complications. Therefore, future therapies for diabetes should

P.753

provide effective control of both fasting and postprandial hyperglycemia. In current clinical practice, however, achieving normal glycemia is a formidable challenge and controlling postprandial hyperglycemia is even more difficult. In patients with type 1 DM, proper blood glucose control is only attainable by insulin replacement therapy. In patients with type 2 DM, additional treatment options are available (84). However, these therapies are suboptimal in fighting against the long-term devastations of diabetes and ineffective in providing proper glycemia management in advanced-stage diabetic patients. As a result, approximately one third of patients with type 2 DM require regular insulin injections in an attempt to best manage their condition.

Although the use of injectable insulin has revolutionized the care of diabetes, optimal insulin therapy is seldom feasible outside of well-monitored study settings. It demands a certain degree of compromise of patients’ quality of life and requires a great deal of self-discipline and effort to perform frequent blood sugar tests and coordinate insulin injections with meals. Furthermore, it carries real risks for hypoglycemia. For example, 25% of patients practicing intensive insulin therapy suffer at least one episode a year of severe, temporarily disabling hypoglycemia, often with seizures or coma, resulting in persistent psychological morbidity in many patients with type 1 diabetes (85). Weight gain is another negative side effect of intensive insulin therapy. In the UKPDS, insulin-treated obese patients gained on average 4.0 kg more than patients assigned to diet therapy alone (86). Since obesity causes insulin resistance and represents an independent risk factor for cardiovascular disease (87), weight gain is most certainly an undesirable side effect of insulin therapy. Weight gain induced by insulin therapy also imposes psychosocial distress on some diabetic patients. Notably, 31% of women with diabetes intentionally omit insulin injections occasionally, while 8.8% do so regularly to avoid weight gain (88). In addition to these risks, treating diabetes with insulin injections is still suboptimal because diabetic patients still suffer from debilitating complications, despite their intensive efforts to manage glycemia (80,81). These findings emphasize the need to design and develop better methods for insulin replacement in patients with diabetes.

An attractive and potentially more physiologic alternative to insulin injection is transplantation of human insulin-producing tissues. Whole pancreas transplantation is an accepted therapy for patients with type 1 DM who simultaneously undergo kidney transplantation. However, widespread use of this procedure is still limited by the cost and risk of the surgery, shortage of donor tissue, and side effects associated with long-term use of immunosuppressive drugs. Pancreatic islet transplantation holds greater promise as a cure for diabetes as evidenced by the recent breakthrough of the Edmonton Protocol (89). However, for this procedure to have a major clinical impact, the barriers of limited islet source and graft rejection must first be overcome (90). Numerous laboratories are trying to derive islet tissue from potential precursor cells within the pancreas as well as from adult and embryonic stem cells (Fig. 49.1). These approaches are considered by many to be the most promising to solve the supply problem (91,92,93). Efforts are thus under way to better understand islet development in order to facilitate orchestrating transcriptional events to culminate in the generation of stable fully functional β-cells, an extraordinarily difficult task. In addition to the restriction of tissue shortage, transplanting islets to cure diabetes is still limited by graft rejection and the threat of recurring autoimmune reaction. Several gene therapy approaches are under development to prevent immune rejection by induction of tolerance and to provide β-cells resistance against immune attack (94). Preventing immune destruction of β-cells by genetic manipulation will require considerably more research in light of the complexity of graft rejection and autoimmunity in type 1 DM, together with our incomplete knowledge on these processes.

Ex Vivo Gene Transfer for Insulin Replacement

One strategy to avoid the recipient’s immune system is to engineer non–β-cells from the patient to produce insulin. In theory, non–β-cells from a patient could be removed, genetically modified in order to produce insulin, selected, and then expanded by in vitro culture. Cells with the appropriate functional characteristics could then be transplanted back to the patient as a source of insulin replacement (Fig. 49.1). There are several potential advantages of this ex vivo somatic cell gene therapy approach (95). Because the target tissue is autologous and not a β-cell, this therapy is expected to provide a steady supply of insulin while evading graft rejection and recurring autoimmunity. By using the patients’ own cells, there are no supply limitations; conceivably cells could even be cryopreserved for subsequent transplants. Importantly, the automatic physiologic release of insulin from surrogates might eliminate diabetic complications and considerably improve the quality of life of subjects with diabetes. For these reasons, engineering surrogate cells to replace production of insulin in diabetes patients was proposed shortly after the cloning of the insulin gene (96,97).

Selden and colleagues first examined the feasibility of this approach by transplanting diabetic mice with cultured mouse fibroblasts transfected with a fusion gene designed to express human preproinsulin (98). Normoglycemia was restored in some mice within 2 weeks of transplantation, but this was a result of the production of proinsulin as opposed to more potent mature insulin. Insulin is made as a larger prohormone molecule that is processed to mature and fully bioactive insulin by prohormone convertases (PCs) found in neuroendocrine cells (99,100,101) (Fig. 49.2A). To overcome the absence of PCs in other target surrogate cells, the two PC enzyme recognition sites of the proinsulin molecule have been mutated into a sequence cleavable by the more widely expressed enzyme furin (Fig. 49.2B). Numerous studies demonstrated that when the mutated proinsulin gene having the furin cleavable sites is introduced into cultured non-neuroendocrine cells such as hepatocytes, myocytes, epithelial cells, and fibroblasts, mature bioactive insulin is produced (102,103,104,105,106).

As an alternative approach to circumvent the limitation of missing appropriate PCs in nonendocrine tissue, Lee et al. recently developed a single-chain insulin analogue that does not require proteolytic processing to become bioactive (Fig. 49.2C). The “designer” insulin retains 20% to 40% of the activity of native insulin, and when the gene encoding this insulin analog

P.754

was expressed in liver, it effectively ameliorated diabetes in rodents (107). However, the potential therapeutic benefits of C-peptide (108) will not be available from the use of single-chain insulin analogue. Furthermore, the properties of this new molecule (e.g., immunogenicity, specificity, etc.) require further characterization before it can be implemented for human use.

Figure 49.1. Hypothetical tissue engineering approaches to treat diabetes. Patient stem cells can be harvested, purified, and expanded in culture (left). These cells can be differentiated through the appropriate culture conditions to generate insulin-producing cells that may be selected for transplantation either functionally or by introduction of a selection marker (e.g., insulin promoter driving antibiotic resistance enabling cells to survive the selection regimen or a marker enabling cell sorting). Cells may also be encouraged to form into islet-like structures by further culture before characterization for insulin secretion and reimplantation into the patient. Gene therapy techniques can be applied to improve the yield of precursor cells, for purification, to promote differentiation of precursors into β-cells, and for selection/concentration of newly differentiated β-cells. Alternatively, non–β-cells (stem cells or non–stem cells) can be isolated from a patient tissue sample and engineered in culture to produce insulin from a suitable promoter (right). Insulin-producing clones can then be selected as above, expanded, and functionally characterized with respect to insulin secretion before reimplantation back to the patient. If the target cells are not capable of significant expansion, then a “reversible immortalization” step may be required.

Although production of mature insulin in surrogate non–β-cells is now possible, the main obstacle limiting effective deployment of cell therapy for insulin replacement is the difficulty in recreating glucose-regulated insulin secretion. The importance of tight regulation is underscored by the finding that diabetic mice transplanted with fibroblasts constitutively secreting proinsulin paradoxically died of hypoglycemia (98). Pancreatic β-cells are remarkably specialized and uniquely able to achieve insulin secretion tuned to the body’s minute-to-minute requirements. Thus, for this approach to be successfully applied, several unique cellular functions of β-cells must be reproduced in surrogate cells, particularly regulated synthesis of proinsulin to maintain stores consistent with demand, storage of mature insulin, and acute secretion in response to glucose. These daunting requirements for ideal surrogate cells have recently been scrutinized (109). Considerable progress has been

P.755

made toward these criteria, particularly in studies aimed more at in vivo gene therapy approaches, as reviewed below.

Figure 49.2. Proinsulin processing pathways. A: In neuroendocrine cells such as β-cells, prohormone convertases (PC1/3 and PC2) cleave the proinsulin molecule at two dibasic sites. The dibasic amino acid residues on C-termini of the products are then removed by carboxypeptidase (CP) to yield mature insulin and C-peptide, which are cosecreted by the cell upon appropriate stimulation. B: In non-neuroendocrine cells such as hepatocytes, where PCs are not expressed, the two dibasic sites flanking the C-peptide in proinsulin can be mutated to furin recognition sites consisting of four dibasic residues (open squares). Following furin cleavage, CP removes four C-terminal basic amino acid residues, liberating native insulin and C-peptide. C: To circumvent the need for PC-mediated conversion of proinsulin to bioactive insulin, a single-chain insulin analogue can be created.

In Vivo Insulin Gene Transfer for Insulin Replacement

An alternate strategy to provide endogenous insulin therapy for diabetes is to induce non–β-cells or surrogate cells to produce insulin by in vivo gene transfer. This approach has all the advantages outlined above for ex vivo therapy, without requiring the cumbersome processes for cell isolation, expansion, and transplantation. However, as for ex vivo approaches, surrogate cells must closely mimic many features of β-cells for appropriate maintenance of glucose homeostasis. This includes regulated insulin gene expression, intracellular storage, and glucose-regulated insulin secretion. In the remainder of this chapter, we will discuss different strategies used to reconstruct these features in surrogate cells in vivo and remaining challenges that need to be overcome before insulin gene transfer can be used clinically to cure human diabetes.

Regulated Insulin Gene Expression in Non–β-Cells

Initial attempts at in vivo insulin gene transfer targeted liver with the native preproinsulin gene (110,111). The liver appears to be an appealing candidate as a surrogate organ for insulin production because hepatocytes have the intrinsic ability to respond to changes in glycemia. Like the proinsulin gene in β-cells, expression of many liver-specific genes is coupled to changes in blood glucose levels (112,113). Several investigators have used the upstream regulatory sequences or promoters associated with these genes to achieve liver-specific glucose-responsive insulin expression. Hepatic proinsulin expression driven off the liver-type pyruvate kinase (LPK) promoter in transgenic mice was upregulated by a high-carbohydrate diet (114). However, these mice were not protected from streptozotocin (STZ)-induced diabetes, indicating that the LPK promoter used was too weak to provide sufficient insulin expression. To strengthen the LPK promoter, Lee et al. incorporated an SV40 early enhancer and attained hepatic insulin production in mice sufficient to cause remission of STZ-induced diabetes (107). Notably this was effective in controlling both fasting and postprandial glycemia in rodents, despite almost a 2-hour delay in insulin production (107). Equally puzzling in this study is the lack of hypoglycemia in the face of excess insulin production almost 6 hours following the glucose challenge. Auricchio et al. placed the expression of insulin under the control of a dimerizer-inducible transcription system in order that hepatic insulin expression could be regulated pharmacologically. Insulin secretion from engineered diabetic mouse liver was induced in a dose-dependent manner following administration of the dimerizer drug rapamycin and reversed following drug withdrawal (115). As anticipated with regulation of expression achieved at the transcriptional level, the induction of insulin release was delayed several hours and persisted beyond return to normoglycemia (115). These insulin release kinetics are unlikely to be as acceptable in humans (116).

In an attempt to improve upon the regulatory characteristics, other custom promoters have been developed that are stimulated by glucose, yet suppressed by insulin, thereby providing an added degree of safety to prevent hypoglycemia. One configuration consisted of a furin-cleavable proinsulin transgene containing glucose-responsive elements from the LPK

P.756

promoter and an inhibitory insulin response sequence from the insulin-like growth factor binding protein-1 promoter (117,118). Although the secretory dynamics were not reported, treatment of STZ-induced diabetic rats with this insulin transgene enabled animals to reduce blood glucose following a glucose load, and occurrences of hypoglycemia were infrequent during a fast. Similar findings were obtained with an aldolase B enhancer–incorporated glucose-6-phosphatase (G6P) promoter driving expression of proinsulin. Adenoviral delivery of the transgene to STZ-induced diabetic rats accelerated glucose clearance following a glucose challenge, albeit not to the levels of nondiabetic control rats, and no fasting hypoglycemia was observed (119). Collectively, these studies demonstrate the potential for autoregulated hepatic insulin expression as a treatment modality for type 1 DM.

An alternative approach to render additional β-cell characteristics to hepatocytes was recently developed by Ferber and colleagues. In vivo adenovirus-mediated transfer of the pancreatic and duodenal homeobox-1 (PDX-1) gene to the livers of mice with diabetes led to induction of the otherwise silent genes for mouse insulin and PC1/3, and ameliorated hyperglycemia (120). PDX-1 is a transcription factor that plays a key regulatory role in pancreatic development and islet cell differentiation (121,122). Although the exact identity of the liver cell type that was transformed is still undefined, it is likely that there are putative stem cells in the liver that possess the plasticity to differentiate into a β-cell phenotype. Although these findings offer hope that certain adult stem cells may be converted into a β-cell phenotype in situ by a simple gene transfer protocol, the exact mechanism of this complex process and the safety of this method in humans are yet to be addressed. More recently, it has been demonstrated that intestinal cells may also be induced toward a β-cell phenotype by expressing PDX-1 and ISL-1 (123,124). However, based on the results presented thus far, it is still premature to decipher if these methods would produce sufficient and appropriately glucose-responsive surrogate cells for optimal insulin therapy.

Although regulated insulin production from liver can be achieved with promoters to drive proinsulin gene transcription, the relatively slow time course of transcriptional regulation may limit clinical application (Fig. 49.3). A delay in turning off proinsulin expression resulting in prolonged insulin production could cause dangerous hypoglycemic episodes. Hyperinsulinemia associated with fasting, exercise, or prolonged illness must be reduced to near zero before this new treatment modality could find acceptance in clinical applications. Conversely, the slow onset (hours) of de novo insulin biosynthesis is unlikely to be effective in controlling postprandial hyperglycemia, especially in humans. The inability of transcriptional control to provide prompt postmeal insulin production will thus certainly compromise its utility in gene therapy for diabetes. However, hepatic insulin production could provide sustained basal insulin secretion needed to improve metabolic regulation in patients with diabetes, such as normalization of postabsorptive glycemia, inhibition of excessive lipolysis, protein wasting, and uncontrolled hepatic glycogenolysis and gluconeogenesis. These metabolic improvements would definitely be advantageous to diabetes management. In fact, several studies using in vivo gene transfer approaches in diabetic rodents have already demonstrated the many therapeutic benefits of basal insulin production (105,119,125,126,127,128,129,130,131).

Intracellular Insulin Storage

Nature’s solution to rapid insulin production is to synthesize insulin in advance and store it in specialized secretory granules in the cytoplasm, ready to be released into the circulation immediately when needed. Therefore, one important characteristic of an ideal surrogate cell is the ability to store and release premade insulin in a regulated fashion. Although this feature is naturally found only in secretory (endocrine and exocrine) cells, Rivera et al. devised a way of recreating it in other cell types (132). To achieve this, they expressed proinsulin as a fusion protein with a reversible aggregation domain of FKBP12. Aggregates accumulate in the endoplasmic reticulum (ER) and upon exposure to a small molecule drug dissolve, permitting export into the constitutive secretory pathway where bioactive insulin is liberated by furin (Fig. 49.3). When engineered cells expressing FKBP12-proinsulin fusion proteins were implanted intramuscularly into diabetic mice, rapid and transient insulin secretion was achieved following intravenous administration of a synthetic drug, resulting in transient glycemic correction. This study demonstrated for the first time that the unique hormone storage and rapid secretion capability of endocrine cells can be duplicated in virtually any type of cell. However, a critical limitation of this technique for insulin replacement is that the secretion is induced by a synthetic compound rather than glucose. Therefore, in practice it will be challenging to find the correct dosage of drug to accommodate the body’s frequently changing requirements for insulin. In addition, the effects and safety of long-term protein aggregation in the ER compartment or surrogate cells will have to be carefully examined before this new technology can be implemented in human gene therapy.

To avoid the complexity of having to engineer in surrogate cells the machinery required for insulin storage, many investigators have turned to using secretory cells for insulin gene therapy. Moore and colleagues were the first to engineer insulin production in endocrine cells (133). AtT20 cells, an anterior pituitary tumor cell line, were selected as surrogate cells because they possess the secretory pathway needed for regulated hormone release in response to stimuli and express the necessary processing enzymes. When transfected with a human proinsulin complementary DNA (cDNA) controlled by a constitutive viral promoter, AtT20 cells were able to process proinsulin to mature insulin. Lipes et al. targeted proinsulin gene expression to the intermediate lobe pituitary cells in nonobese diabetic (NOD) mice using a pro-opiomelanocortin (POMC) promoter (134). The resulting transgenic NOD mice expressed proinsulin messenger RNA and insulin protein in their pituitary glands, and the ectopically produced insulin was secreted into the circulation. Notably, in contrast to pancreatic β-cells, the insulin-producing pituitary cells were not destroyed by the autoimmunity in NOD mice. This important finding offers hope that using non–β-cells as surrogate cells for insulin production represents a viable means to provide long-term insulin replacement in subjects with type 1 diabetes because they may be spared by the autoimmune response.

Figure 49.3. Insulin secretion pathways and kinetics. Top panels are schematic representations of insulin synthesis and secretion mechanisms in β-cells or engineered surrogate cells, with the corresponding hypothetical secretion kinetics shown below. A: In β-cells, elevation in the ambient glucose concentration leads to an increase in the intracellular glycolytic flux, which ultimately activates the exocytosis of secretory vesicles and cosecretion of insulin and C-peptide. This process also activates expression of preproinsulin by activating specific elements within the insulin promoter region (Ins Pro) in order to replenish insulin stores. Newly synthesized proinsulin in the rough endoplasmic reticulum (RER) is translocated through the Golgi apparatus to the trans-Golgi network, where proinsulin molecules are sorted into nascent β-granules. In this compartment, proinsulin molecules undergo proteolytic processing, and the resulting insulin and C-peptide are stored in secretory vesicles in preparation for rapid (within minutes) release of insulin in response to a stimulus, such as glucose (bottom). B: In nonendocrine cells, insulin production can be achieved by driving preproinsulin expression off a constitutively active promoter (Con Pro; e.g., cytomegalovirus). With this system, insulin secretion occurs via a constitutive pathway and is thus not regulated by glucose (bottom). C: In nonendocrine cells such as hepatocytes, preproinsulin expression can also be driven off a glucose-responsive promoter (GR Pro; e.g., liver-type pyruvate kinase). In this manner, expression of insulin is coupled to glucose metabolism. Constitutive secretion of insulin is therefore regulated indirectly by glucose metabolism, but over a timescale of hours, with delays both in initiation and cessation following transient exposure to glucose (bottom). D: Drug-induced insulin secretion in non β-cells can be achieved through controlled aggregation in the endoplasmic reticulum (ER). A fusion protein consisting of proinsulin linked to a conditional aggregation domain of FKBP12 is expressed under the control of a constitutive promoter. This chimeric protein forms aggregates that are retained in the ER. Addition of a small molecule drug (Drug ‘X’) triggers rapid disaggregation of chimeric proinsulin molecules, releasing them into the secretory network. Furin in the vesicular compartment removes the FKBP12 domain (star) and converts the modified proinsulin molecule to mature insulin and C-peptide, which are then secreted. The secretion of insulin from the modified cells is thus sensitive to the drug, but not glucose (bottom).

P.757

One limitation of pituitary cells as surrogate cells is that unlike β-cells, pituitary cells are not glucose responsive. This can be somewhat resolved by overexpression of both glucose transporter 2 (GLUT-2) and glucokinase (GK), suggesting that glucose-induced secretion can be installed in non–glucose-responsive endocrine cells by the simple introduction of two key glucose sensors (135). However, a recent report by Faradji et al. questioned this notion (136). Using primary pituitary cells, they showed that some degree of glucose-sensing capability can be introduced by coexpression of GK and GLUT-2, but these manipulations also result in glucose-induced toxicity and severe apoptosis in the cells. This finding highlights the complexity of engineering glucose sensing systems in surrogate cells and raises awareness that expression levels of glucose sensors need to be finely tuned, perhaps with additional downstream elements, to prevent cell toxicity. Coupling glucose sensing to the secretory pathway in non–glucose-responsive cells may also prove extremely challenging.

Pancreatic exocrine tissue has also been explored as a surrogate organ for insulin replacement (137). Goldfine et al. capitalized on the realization that exocrine glands are “duocrine” organs capable of secreting proteins into the circulation as well as the external environment via the gland’s duct system. Remarkably, when an expression plasmid containing a modified human proinsulin cDNA with furin processing sites linked to a cytomegalovirus promoter was delivered into diabetic rats via retrograde perfusion of the pancreatic duct, the circulating insulin levels achieved were effective in maintaining normoglycemia (137).

P.758

Furthermore, when a human growth hormone (hGH) gene was delivered by the same method, secretion of hGH was shown to be rapidly inducible by a cholinergic agonist, indicating that exocrine cells contain a secretagogue-responsive pool. Unfortunately, insulin or hGH secretion in response to a meal or an oral glucose load was not reported, so it is unclear if in vivo insulin production by this method is superior to nonregulated basal insulin production. Although pancreatic exocrine secretion is controlled by some factors that also govern insulin secretion in β-cells, future studies will need to determine if endocrine secretion from exocrine cells, for example by glucose, will be appropriate enough to normalize glycemia in diabetes.

More recently, gastric G-cells were examined as potential surrogate cells capable of producing insulin. These endocrine cells are located predominantly in the gastric antrum and secrete the gastric acid stimulatory hormone gastrin into the blood (138). Zhukova et al. targeted the expression of human insulin to these cells by injecting fertilized mouse oocytes with a construct containing the human insulin gene spliced into a hybrid rat/human gastrin chimeric minigene (139). The resulting transgenic mice expressed processed insulin specifically in antral G-cells. Although not reported, human insulin would presumably be released from these cells into the portal circulation. However, unlike β-cells, gastric G-cells are not glucose responsive, but rather secrete gastrin in response to stimuli such as contact with digestion products of proteins and neutralization of gastric contents (138). Therefore, the ultimate kinetics of insulin release from engineered G-cells may not adequately mimic that of insulin release from normal β-cells.

Glucose-Regulated Insulin Secretion in Non–β-Cells

Another gastrointestinal endocrine cell type explored as potential surrogate β-cells are K-cells. Unlike G-cells, K-cells located predominantly in the duodenum and jejunum are glucose responsive (140), secreting the hormone glucose-dependent insulinotropic polypeptide (GIP), which functions to potentiate insulin release following a meal (141). K-cells are remarkably similar to β-cells in that both are endocrine cells that synthesize and secrete hormones into the portal circulation in a carbohydrate-dependent manner. More importantly, the secretion kinetics of GIP and insulin following glucose ingestion are surprisingly comparable, fitting with their physiologic roles in glucose homeostasis (142,143,144). The similarity in circulating profiles extends throughout the day in subjects consuming mixed meals (143). Taking advantage of the similarity between the two cell types, Cheung et al. recently tested the feasibility of using K-cells as surrogate cells for insulin gene therapy (145). To do so, transgenic mice expressing human insulin under the control of the rat GIP promoter were generated. Human insulin was detected in the gastrointestinal K-cells and circulation of transgenic mice, and the levels of human insulin produced were sufficient to prevent mice from developing STZ-induced diabetes. Interestingly, transgenic mice lacking the majority of endogenous β-cells disposed of oral glucose at the same rate as normal mice, suggesting that insulin secretion from K-cells is appropriately glucose responsive. Indeed, in response to a mixed meal challenge, a significant increase in plasma human insulin levels in transgenic mice was observed within 30 minutes (146). This is the first study to achieve rapid meal-dependent insulin secretion from non–β-cells. These findings support K-cells as good candidate surrogate cells for insulin gene therapy, because they appear to have the capacity to synthesize and release mature insulin in response to glucose in a pattern comparable with insulin secretion from the pancreas.

Although results from transgenic mice indicate that K-cells fulfill the many criteria of ideal surrogate cells for insulin gene therapy, an in vivo gene delivery system to provide sufficient and sustained amounts of insulin to treat diabetes with this strategy is yet to be developed. Because the life span of gut epithelial cells is relatively short, gene delivery systems must target stem cells of the intestinal crypts of Lieberkühn for long-term insulin production. Several reports on effective and long-term gene transfer to the intestine support the feasibility of this approach (23,147,148). Another caveat of using K-cells as surrogate cells for insulin production is the responsiveness of GIP secretion to fat ingestion (141,143). If insulin is ectopically produced in K-cells, it is anticipated that ingestion of fat alone would lead to insulin secretion in the absence of hyperglycemia, which, if beyond the counterregulatory responses, could produce hypoglycemia. However, this may not be a critical setback because humans rarely consume large meals consisting entirely of fat. Therefore, given that K-cells require relatively little genetic manipulation with natural elements to achieve meal-regulated insulin production and secretion, gene transfer to these cells in vivo should be developed. It is considered that insulin production from K-cells holds promise as a means to physiologically replace insulin production in patients with diabetes, thus ending the requirement for insulin injections and the debilitating complications that result from poor blood glucose control (79,149,150).

Prospects of Gene Therapy for Diabetes

Automatic physiologic insulin secretion from engineered surrogate tissues is an attractive approach to restore physiologic insulin delivery that could ultimately reduce or even eliminate the risk for developing the debilitating complications associated with poor glycemic control. During the past few years, several potential methods have been identified to generate insulin-producing tissues for transplantation. However, generating sufficient quantities of suitably glucose-responsive tissue remains a challenge. Efforts to protect this tissue from recipient immune destruction must continue. There have also been considerable advancements in the development of modified therapeutic insulin genes for delivery into diabetic subjects. The ultimate success of gene therapy approaches depends on the availability of a safe and efficient gene delivery system. In this regard, numerous viral and nonviral vector systems are already available for gene delivery, and improvements continue to be developed (2,151). The optimal vector system for insulin gene therapy is dependent on the type of surrogate cells targeted and the size of the genetic material to be transferred. Future efforts should also be directed toward developing gene treatments that can be modulated

P.759

as needed or even terminated to deal with unforeseen outcomes. Because diabetes is already a treatable disease, risks associated with gene therapy for diabetes (e.g., hypoglycemic episodes, vector toxicity, etc.) must be substantially minimized before it could become accepted in clinical practice. However, the greatest obstacle to a gene therapy approach to treat diabetes remains the achievement of long-term, sufficient insulin production that is suitably coupled to changes in blood glucose concentrations. Although it remains uncertain if such a therapy will be developed for diabetes, advances in this area could lead to more generalized treatments of other metabolic diseases in which such precise regulation of production of a therapeutic protein is less critical. Nevertheless, the ever-growing diabetic population will continue to fuel the search for better treatment options, and tissue engineering approaches should continue to be explored.

References

1. Data from the Journal of Gene Medicine, http://www.wiley.co.uk/ genetherapy/clinical/, updated in 2002.

2. Kay MA, Glorioso JC, Naldini L. Viral vectors for gene therapy: the art of turning infectious agents into vehicles of therapeutics. Nat Med 2001;7:33–40.

3. Johnston JC, Gasmi M, Lim LE, et al. Minimum requirements for efficient transduction of dividing and nondividing cells by feline immunodeficiency virus vectors. J Virol 1999;73:4991–5000.

4. Naldini L, Blomer U, Gallay P, et al. In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector. Science 1996;272:263–267.

5. Palu G, Parolin C, Takeuchi Y, et al. Progress with retroviral gene vectors. Rev Med Virol 2000;10:185–202.

6. Powell SK, Kaloss M, Burimski I, et al. In vitro analysis of transformation potential associated with retroviral vector insertions. Hum Gene Ther 1999;10:2123–2132.

7. Kaiser J. Gene therapy: seeking the cause of induced leukemias in X-SCID trial. Science 2003;299:495.

8. Hacein-Bey-Abina S, Le Deist F, Carlier F, et al. Sustained correction of X-linked severe combined immunodeficiency by ex vivo gene therapy. N Engl J Med 2002;346:1185–1193.

9. Hacein-Bey-Abina S, von Kalle C, Schmidt M, et al. A serious adverse event after successful gene therapy for X-linked severe combined immunodeficiency. N Engl J Med 2003;348:255–256.

10. Verma IM. A voluntary moratorium? Mol Ther 2003;7:141.

11. Aiuti A, Slavin S, Aker M, et al. Correction of ADA-SCID by stem cell gene therapy combined with nonmyeloablative conditioning. Science 2002;296:2410–2413.

12. Blaese RM, Culver KW, Miller AD, et al. T lymphocyte–directed gene therapy for ADA-SCID: initial trial results after 4 years. Science 1995;270:475–480.

13. Check E. A tragic setback. Nature 2002;420:116–118.

14. Monahan PE, Samulski RJ. AAV vectors: is clinical success on the horizon? Gene Ther 2000;7:24–30.

15. Young SM Jr, McCarty DM, Degtyareva N, et al. Roles of adeno- associated virus Rep protein and human chromosome 19 in site- specific recombination. J Virol 2000;74:3953–3966.

16. Duan D, Sharma P, Yang J, et al. Circular intermediates of recombinant adeno-associated virus have defined structural characteristics responsible for long-term episomal persistence in muscle tissue. J Virol 1998;72:8568–8577.

17. Schnepp BC, Clark KR, Klemanski DL, et al. Genetic fate of recombinant adeno-associated virus vector genomes in muscle. J Virol 2003;77:3495–3504.

18. Monahan PE, Samulski RJ. Adeno-associated virus vectors for gene therapy: more pros than cons? Mol Med Today 2000;6:433–440.

19. High K. Gene-based approaches to the treatment of hemophilia. Ann NY Acad Sci 2002;961:63–64.

20. Snyder RO, Miao CH, Patijn GA, et al. Persistent and therapeutic concentrations of human factor IX in mice after hepatic gene transfer of recombinant AAV vectors. Nat Genet 1997;16:270–276.

21. Kaplitt MG, Leone P, Samulski RJ, et al. Long-term gene expression and phenotypic correction using adeno-associated virus vectors in the mammalian brain. Nat Genet 1994;8:148–154.

22. Engelhardt JF. The lung as a metabolic factory for gene therapy. J Clin Invest 2002;110:429–432.

23. During MJ, Xu R, Young D, et al. Peroral gene therapy of lactose intolerance using an adeno-associated virus vector. Nat Med 1998; 4:1131–1135.

24. Wolfe D, Goins WF, Yamada M, et al. Engineering herpes simplex virus vectors for CNS applications. Exp Neurol 1999;159:34–46.

25. Burton EA, Wechuck JB, Wendell SK, et al. Multiple applications for replication-defective herpes simplex virus vectors. Stem Cells 2001;19:358–377.

26. Glorioso JC, Fink DJ. Use of HSV vectors to modify the nervous system. Curr Opin Drug Discov Dev 2002;5:289–295.

27. Goss JR, Goins WF, Lacomis D, et al. Herpes simplex–mediated gene transfer of nerve growth factor protects against peripheral neuropathy in streptozotocin-induced diabetes in the mouse. Diabetes 2002;51:2227–2232.

28. Horwitz MS. Adenoviruses. In: Knipe DM, Howley PM, eds. Fields virology. Vol. 2. Philadelphia: Lippincott Williams & Wilkins, 2001: 2301–2326.

29. Hitt MM, Graham FL. Adenovirus vectors for human gene therapy. Adv Virus Res 2000;55:479–505.

30. Crystal RG, Jaffe A, Brody S, et al. A phase 1 study, in cystic fibrosis patients, of the safety, toxicity, and biological efficacy of a single administration of a replication deficient, recombinant adenovirus carrying the cDNA of the normal cystic fibrosis transmembrane conductance regulator gene in the lung. Hum Gene Ther 1995;6:643–666.

31. Kovesdi I, Brough DE, Bruder JT, et al. Adenoviral vectors for gene transfer. Curr Opin Biotechnol 1997;8:583–589.

32. Yang Y, Li Q, Ertl HC, et al. Cellular and humoral immune responses to viral antigens create barriers to lung-directed gene therapy with recombinant adenoviruses. J Virol 1995;69:2004–2015.

33. Yang Y, Jooss KU, Su Q, et al. Immune responses to viral antigens versus transgene product in the elimination of recombinant adenovirus–infected hepatocytes in vivo. Gene Ther 1996;3:137–144.

34. Dai Y, Schwarz EM, Gu D, et al. Cellular and humoral immune responses to adenoviral vectors containing factor IX gene: tolerization of factor IX and vector antigens allows for long-term expression. Proc Natl Acad Sci USA 1995;92:1401–1405.

35. Weiss R, Nelson D. Teen dies undergoing gene therapy. Washington Post, 1999:A1.

36. Kochanek S, Schiedner G, Volpers C. High-capacity “gutless” adenoviral vectors. Curr Opin Mol Ther 2001;3:454–463.

37. Kochanek S. High-capacity adenoviral vectors for gene transfer and somatic gene therapy. Hum Gene Ther 1999;10:2451–2459.

38. Reddy PS, Sakhuja K, Ganesh S, et al. Sustained human factor VIII expression in hemophilia A mice following systemic delivery of a gutless adenoviral vector. Mol Ther 2002;5:63–73.

39. Balague C, Zhou J, Dai Y, et al. Sustained high-level expression of full-length human factor VIII and restoration of clotting activity in hemophilic mice using a minimal adenovirus vector. Blood 2000;95: 820–828.

40. Nishikawa M, Huang L. Nonviral vectors in the new millennium: delivery barriers in gene transfer. Hum Gene Ther 2001;12:861–870.

41. Akhtar S, Hughes MD, Khan A, et al. The delivery of antisense therapeutics. Adv Drug Deliv Rev 2000;44:3–21.

42. Paul CP, Good PD, Winer I, et al. Effective expression of small interfering RNA in human cells. Nat Biotechnol 2002;20:505–508.

43. Somiari S, Glasspool-Malone J, Drabick JJ, et al. Theory and in vivo application of electroporative gene delivery. Mol Ther 2000;2:178–187.

44. Lin MT, Pulkkinen L, Uitto J, et al. The gene gun: current applications in cutaneous gene therapy. Int J Dermatol 2000;39:161–170.

P.760

45. Lu QL, Liang HD, Partridge T, et al. Microbubble ultrasound improves the efficiency of gene transduction in skeletal muscle in vivo with reduced tissue damage. Gene Ther 2003;10:396–405.

46. Dauty E, Remy JS, Blessing T, et al. Dimerizable cationic detergents with a low cmc condense plasmid DNA into nanometric particles and transfect cells in culture. J Am Chem Soc 2001;123:9227–9234.

47. Wu AG, Liu X, Mazumder A, et al. Improvement of gene transduction efficiency in T lymphocytes using retroviral vectors. Hum Gene Ther 1999;10:977–982.

48. Lim YB, Han SO, Kong HU, et al. Biodegradable polyester, poly[alpha-(4-aminobutyl)-L-glycolic acid], as a non-toxic gene carrier. Pharm Res 2000;17:811–816.

49. Krieg AM, Yi AK, Matson S, et al. CpG motifs in bacterial DNA trigger direct B-cell activation. Nature 1995;374:546–549.

50. Tousignant JD, Gates AL, Ingram LA, et al. Comprehensive analysis of the acute toxicities induced by systemic administration of cationic lipid:plasmid DNA complexes in mice. Hum Gene Ther 2000; 11:2493–2513.

51. Li S, Wu SP, Whitmore M, et al. Effect of immune response on gene transfer to the lung via systemic administration of cationic lipidic vectors. Am J Physiol 1999;276:796–804.

52. Ruiz FE, Clancy JP, Perricone MA, et al. A clinical inflammatory syndrome attributable to aerosolized lipid-DNA administration in cystic fibrosis. Hum Gene Ther 2001;12:751–761.

53. Nakai H, Montini E, Fuess S, et al. Helper-independent and AAV-ITR–independent chromosomal integration of double-stranded linear DNA vectors in mice. Mol Ther 2003;7:101–111.

54. Thyagarajan B, Olivares EC, Hollis RP, et al. Site-specific genomic integration in mammalian cells mediated by phage phiC31 integrase. Mol Cell Biol 2001;21:3926–3934.

55. Olivares EC, Hollis RP, Chalberg TW, et al. Site-specific genomic integration produces therapeutic factor IX levels in mice. Nat Biotechnol 2002;20:1124–1128.

56. Heister T, Heid I, Ackermann M, et al. Herpes simplex virus type 1/adeno-associated virus hybrid vectors mediate site-specific integration at the adeno-associated virus preintegration site, AAVS1, on human chromosome 19. J Virol 2002;76:7163–7173.

57. Schmidt M, Afione S, Kotin RM. Adeno-associated virus type 2 Rep78 induces apoptosis through caspase activation independently of p53. J Virol 2000;74:9441–9450.

58. Metselaar JM, Mastrobattista E, Storm G. Liposomes for intravenous drug targeting: design and applications. Mini Rev Med Chem 2002;2:319–329.

59. Wu H, Seki T, Dmitriev I, et al. Double modification of adenovirus fiber with RGD and polylysine motifs improves coxsackievirus-adenovirus receptor–independent gene transfer efficiency. Hum Gene Ther 2002;13:1647–1653.

60. Snyder EL, Dowdy SF. Protein/peptide transduction domains: potential to deliver large DNA molecules into cells. Curr Opin Mol Ther 2001;3:147–152.

61. Gratton JP, Yu J, Griffith JW, et al. Cell-permeable peptides improve cellular uptake and therapeutic gene delivery of replication-deficient viruses in cells and in vivo. Nat Med 2003;9:357–363.

62. Mi Z, Mai J, Lu X, et al. Characterization of a class of cationic peptides able to facilitate efficient protein transduction in vitro and in vivo. Mol Ther 2000;2:339–347.

63. Curiel DT. Considerations and challenges for the achievement of targeted gene delivery. Gene Ther 1999;6:1497–1498.

64. Russell SJ, Cosset FL. Modifying the host range properties of retroviral vectors. J Gene Med 1999;1:300–311.

65. Rosenberg SA, Aebersold P, Cornetta K, et al. Gene transfer into humans—immunotherapy of patients with advanced melanoma, using tumor-infiltrating lymphocytes modified by retroviral gene transduction. N Engl J Med 1990;323:570–578.

66. Grossman M, Rader DJ, Muller DW, et al. A pilot study of ex vivo gene therapy for homozygous familial hypercholesterolaemia. Nat Med 1995;1:1148–1154.

67. Wilson JM, Grossman M, Wu CH, et al. Hepatocyte-directed gene transfer in vivo leads to transient improvement of hypercholesterolemia in low density lipoprotein receptor–deficient rabbits. J Biol Chem 1992;267:963–967.

68. National Institutes of Health. http://www.clinicaltrials.gov, updated 2003.

69. Takahashi H, Hirai Y, Migita M, et al. Long-term systemic therapy of Fabry disease in a knockout mouse by adeno-associated virus–mediated muscle-directed gene transfer. Proc Natl Acad Sci USA 2002;99:13777–13782.

70. Ponder KP, Melniczek JR, Xu L, et al. Therapeutic neonatal hepatic gene therapy in mucopolysaccharidosis VII dogs. Proc Natl Acad Sci USA 2002;99:13102–13107.

71. Taylor SI. Deconstructing type 2 diabetes. Cell 1999;97:9–12.

72. Kahn BB. Type 2 diabetes: when insulin secretion fails to compensate for insulin resistance. Cell 1998;92:593–596.

73. Roper NA, Bilous RW, Kelly WF, et al. Excess mortality in a population with diabetes and the impact of material deprivation: longitudinal, population based study. BMJ 2001;322:1389–1393.

74. Efrat S. Preventing type 1 diabetes mellitus: the promise of gene therapy. Am J Pharmacogenomics 2002;2:129–134.

75. Yamaoka T. Regeneration therapy of pancreatic beta cells: towards a cure for diabetes? Biochem Biophys Res Commun 2002;296:1039–1043.

76. Freeman DJ, Leclerc I, Rutter GA. Present and potential future use of gene therapy for the treatment of non-insulin dependent diabetes mellitus [Review]. Int J Mol Med 1999;4:585–592.

77. Dhillon H, Kalra SP, Kalra PS. Dose-dependent effects of central leptin gene therapy on genes that regulate body weight and appetite in the hypothalamus. Mol Ther 2001;4:139–145.

78. Desai UJ, Slosberg ED, Boettcher BR, et al. Phenotypic correction of diabetic mice by adenovirus-mediated glucokinase expression. Diabetes 2001;50:2287–2295.

79. Giannoukakis N, Robbins PD. Gene and cell therapies for diabetes mellitus: strategies and clinical potential. Biodrugs 2002;16:149–173.

80. U.K. Prospective Diabetes Study (UKPDS) Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993;329:977–986.

81. U.K. Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 1998;352:837–853.

82. Garber AJ. The importance of early insulin secretion and its impact on glycaemic regulation. Int J Obes Relat Metab Disord 2000;24 (suppl 3):32–37.

83. Bastyr EJ 3rd, Stuart CA, Brodows RG, et al. Therapy focused on lowering postprandial glucose, not fasting glucose, may be superior for lowering HbA1c. IOEZ Study Group. Diabetes Care 2000;23: 1236–1241.

84. DeFronzo RA. Pharmacologic therapy for type 2 diabetes mellitus. Ann Intern Med 1999;131:281–303.

85. Cryer PE, Fisher JN, Shamoon H. Hypoglycemia. Diabetes Care 1994;17:734–755.

86. U.K. Prospective Diabetes Study (UKPDS) Group. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). Lancet 1998;352: 854–865.

87. Reaven GM. Insulin resistance: a chicken that has come to roost. Ann NY Acad Sci 1999;892:45–57.

88. Polonsky WH, Anderson BJ, Lohrer PA, et al. Insulin omission in women with IDDM. Diabetes Care 1994;17:1178–1185.

89. Shapiro AM, Lakey JR, Ryan EA, et al. Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen. N Engl J Med 2000;343:230–238.

90. Weir GC, Bonner-Weir S. Scientific and political impediments to successful islet transplantation. Diabetes 1997;46:1247–1256.

91. Bonner-Weir S, Sharma A. Pancreatic stem cells. J Pathol 2002;197: 519–526.

92. Serup P, Madsen OD, Madrup-Poulsen T. Islet and stem cell transplantation for treating diabetes. BMJ 2001;322:29–32.

93. Lechner A, Habener JF. Stem/progenitor cells derived from adult tissues: potential for the treatment of diabetes mellitus. Am J Physiol 2003;284:E259–E266.

P.761

94. Efrat S. Prospects for gene therapy of insulin-dependent diabetes mellitus. Diabetologia 1998;41:1401–1409.

95. Bailey CJ, Davies EL, Docherty K. Prospects for insulin delivery by ex-vivo somatic cell gene therapy. J Mol Med 1999;77:244–249.

96. Laub O, Rutter WJ. Expression of the human insulin gene and cDNA in a heterologous mammalian system. J Biol Chem 1983;258: 6043–6050.

97. Permutt MA, Chirgwin J, Rotwein P, et al. Insulin gene structure and function: a review of studies using recombinant DNA methodology. Diabetes Care 1984;7:386–394.

98. Selden RF, Skoskiewicz MJ, Russell PS, et al. Regulation of insulin-gene expression: implications for gene therapy. N Engl J Med 1987; 317:1067–1076.

99. Dodson G, Steiner D. The role of assembly in insulin’s biosynthesis. Curr Opin Struct Biol 1998;8:189–194.

100. Hutton JC, Rhodes CJ. The enzymology of proinsulin conversion. Cell Biophys 1991;19:57–62.

101. Steiner DF. The proprotein convertases. Curr Opin Chem Biol 1998; 2:31–39.

102. Yanagita M, Hoshino H, Nakayama K, et al. Processing of mutated proinsulin with tetrabasic cleavage sites to mature insulin reflects the expression of furin in nonendocrine cell lines. Endocrinology 1993; 133:639–644.

103. Arcelloni C, Falqui L, Martinenghi S, et al. Processing and release of human proinsulin-cleavage products into culture media by different engineered non-endocrine cells: a specific assessment by capillary electrophoresis. J Endocrinol 2000;166:437–445.

104. Yamasaki K, Sasaki T, Nemoto M, et al. Differentiation-induced insulin secretion from nonendocrine cells with engineered human proinsulin cDNA. Biochem Biophys Res Commun 1999;265:361–365.

105. Falqui L, Martinenghi S, Severini GM, et al. Reversal of diabetes in mice by implantation of human fibroblasts genetically engineered to release mature human insulin. Hum Gene Ther 1999;10:1753–1762.

106. Simonson GD, Groskreutz DJ, Gorman CM, et al. Synthesis and processing of genetically modified human proinsulin by rat myoblast primary cultures. Hum Gene Ther 1996;7:71–78.

107. Lee HC, Kim SJ, Kim KS, et al. Remission in models of type 1 diabetes by gene therapy using a single-chain insulin analogue. Nature 2000;408:483–488.

108. Wahren J, Ekberg K, Johansson J, et al. Role of C-peptide in human physiology. Am J Physiol Endocrinol Metab 2000;278:759–768.

109. Halban PA, Kahn SE, Lernmark A, et al. Gene and cell-replacement therapy in the treatment of type 1 diabetes. How high must the standards be set? Diabetes 2001;50:2181–2191.

110. Nicolau C, Le Pape A, Soriano P, et al. In vivo expression of rat insulin after intravenous administration of the liposome-entrapped gene for rat insulin I. Proc Natl Acad Sci USA 1983;80:1068–1072.

111. Soriano P, Dijkstra J, Legrand A, et al. Targeted and nontargeted liposomes for in vivo transfer to rat liver cells of plasmid containing the preproinsulin I gene. Proc Natl Acad Sci USA 1983;80:7128–7131.

112. Rencurel F, Girard J. Regulation of liver gene expression by glucose. Proc Nutr Soc 1998;57:265–275.

113. Mitanchez D, Doiron B, Chen R, et al. Glucose-stimulated genes and prospects of gene therapy for type 1 diabetes. Endocr Rev 1997; 18:520–540.

114. Mitanchez D, Chen R, Massias JF, et al. Regulated expression of mature human insulin in the liver of transgenic mice. FEBS Lett 1998; 421:285–289.

115. Auricchio A, Gao GP, Yu QC, et al. Constitutive and regulated expression of processed insulin following in vivo hepatic gene transfer. Gene Ther 2002;9:963–971.

116. Olefsky JM. Diabetes. Gene therapy for rats and mice. Nature 2000; 408:420–421.

117. Thule PM, Liu JM. Regulated hepatic insulin gene therapy of STZ-diabetic rats. Gene Ther 2000;7:1744–1752.

118. Thule PM, Liu J, Phillips LS. Glucose regulated production of human insulin in rat hepatocytes. Gene Ther 2000;7:205–214.

119. Chen R, Meseck ML, Woo SL. Auto-regulated hepatic insulin gene expression in type 1 diabetic rats. Mol Ther 2001;3:584–590.

120. Ferber S, Halkin A, Cohen H, et al. Pancreatic and duodenal homeobox gene 1 induces expression of insulin genes in liver and ameliorates streptozotocin-induced hyperglycemia. Nat Med 2000;6: 568–572.

121. Stoffers DA, Zinkin NT, Stanojevic V, et al. Pancreatic agenesis attributable to a single nucleotide deletion in the human IPF1 gene coding sequence. Nat Genet 1997;15:106–110.

122. Sander M, German MS. The beta cell transcription factors and development of the pancreas. J Mol Med 1997;75:327–340.

123. Kojima H, Nakamura T, Fujita Y, et al. Combined expression of pancreatic duodenal homeobox 1 and islet factor 1 induces immature enterocytes to produce insulin. Diabetes 2002;51:1398–1408.

124. Yoshida S, Kajimoto Y, Yasuda T, et al. PDX-1 induces differentiation of intestinal epithelioid IEC-6 into insulin-producing cells. Diabetes 2002;51:2505–2513.

125. Kolodka TM, Finegold M, Moss L, et al. Gene therapy for diabetes mellitus in rats by hepatic expression of insulin. Proc Natl Acad Sci USA 1995;92:3293–3297.

126. Muzzin P, Eisensmith RC, Copeland KC, et al. Hepatic insulin gene expression as treatment for type 1 diabetes mellitus in rats. Mol Endocrinol 1997;11:833–837.

127. Dong H, Morral N, McEvoy R, et al. Hepatic insulin expression improves glycemic control in type 1 diabetic rats. Diabetes Res Clin Pract 2001;52:153–163.

128. Short DK, Okada S, Yamauchi K, et al. Adenovirus-mediated transfer of a modified human proinsulin gene reverses hyperglycemia in diabetic mice. Am J Physiol 1998;275:748–756.

129. Zhang W, Lu D, Kawazu S, et al. Adenoviral insulin gene therapy prolongs survival of IDDM model BB rats by improving hyperlipidemia. Horm Metab Res 2002;34:577–582.

130. Riu E, Mas A, Ferre T, et al. Counteraction of type 1 diabetic alterations by engineering skeletal muscle to produce insulin: insights from transgenic mice. Diabetes 2002;51:704–711.

131. Martinenghi S, Cusella De Angelis G, Biressi S, et al. Human insulin production and amelioration of diabetes in mice by electrotransfer-enhanced plasmid DNA gene transfer to the skeletal muscle. Gene Ther 2002;9:1429–1437.

132. Rivera VM, Wang X, Wardwell S, et al. Regulation of protein secretion through controlled aggregation in the endoplasmic reticulum. Science 2000;287:826–830.

133. Moore HP, Walker MD, Lee F, et al. Expressing a human proinsulin cDNA in a mouse ACTH-secreting cell: intracellular storage, proteolytic processing, and secretion on stimulation. Cell 1983;35:531–538.

134. Lipes MA, Cooper EM, Skelly R, et al. Insulin-secreting non-islet cells are resistant to autoimmune destruction. Proc Natl Acad Sci USA 1996;93:8595–8600.

135. Hughes SD, Johnson JH, Quaade C, et al. Engineering of glucose-stimulated insulin secretion and biosynthesis in non-islet cells. Proc Natl Acad Sci USA 1992;89:688–692.

136. Faradji RN, Havari E, Chen Q, et al. Glucose-induced toxicity in insulin-producing pituitary cells that coexpress GLUT2 and glucokinase: implications for metabolic engineering. J Biol Chem 2001;6:6.

137. Goldfine ID, German MS, Tseng HC, et al. The endocrine secretion of human insulin and growth hormone by exocrine glands of the gastrointestinal tract. Nat Biotechnol 1997;15:1378–1382.

138. Walsh JH, Grossman MI. Gastrin (first of two parts). N Engl J Med 1975;292:1324–1334.

139. Zhukova E, Afshar A, Ko J, et al. Expression of the human insulin gene in the gastric G cells of transgenic mice. Transgenic Res 2001; 10:329–341.

140. Kieffer TJ, Buchan AM, Barker H, et al. Release of gastric inhibitory polypeptide from cultured canine endocrine cells. Am J Physiol 1994;267:489–496.

141. Wolfe MM, Boylan MO, Kieffer TJ, et al. Glucose-dependent insulinotropic polypeptide (GIP): incretin vs enterogastrone. In: Greeley GHJ, ed. Gastrointestinal endocrinology. Vol. 8. Totowa, NJ: Humana Press, 1999:439.

142. Cataland S, Crockett SE, Brown JC, et al. Gastric inhibitory polypeptide (GIP) stimulation by oral glucose in man. J Clin Endocrinol Metab 1974;39:223–228.

P.762

143. Elliott RM, Morgan LM, Tredger JA, et al. Glucagon-like peptide-1 (7–36)amide and glucose-dependent insulinotropic polypeptide secretion in response to nutrient ingestion in man: acute post-prandial and 24-h secretion patterns. J Endocrinol 1993;138:159–166.

144. Hampton SM, Morgan LM, Tredger JA, et al. Insulin and C-peptide levels after oral and intravenous glucose. Contribution of enteroinsular axis to insulin secretion. Diabetes 1986;35:612–616.

145. Cheung AT, Dayanandan B, Lewis JT, et al. Glucose-dependent insulin release from genetically engineered K cells. Science 2000;290: 1959–1962.

146. Cheung AT, Lewis JT, Dayanandan B, et al. Meal-regulated insulin production from intestinal K-cells. Diabetes 2001;50(suppl 2):A6.

147. Dumon KR, Ishii H, Fong LY, et al. FHIT gene therapy prevents tumor development in FHIT-deficient mice. Proc Natl Acad Sci USA 2001;98:3346–3351.

148. Jacomino M, Lau C, James SZ, et al. Gene transfer into fetal rat intestine. Hum Gene Ther 1996;7:1757–1762.

149. Corbett JA. K cells: a novel target for insulin gene therapy for the prevention of diabetes. Trends Endocrinol Metab 2001;12:140–142.

150. Dong H, Anthony K, Morral N. Challenges for gene therapy of type 1 diabetes. Curr Gene Ther 2002;2:403–414.

151. Anderson WF. Human gene therapy. Nature 1998;392:25–30.

	Diabetes Mellitus: A Fundamental and Clinical Text 3rd Edition
	© 2004 Lippincott Williams & Wilkins

Diabetes Mellitus: A Fundamental and Clinical Text3rd Edition

Diabetes Mellitus: A Fundamental and Clinical Text
3rd Edition