By Maryam Nazir

Neural stem cell gene therapy has only recently come into the spotlight as a potential therapy for a multitude of brain pathologies. Neural stem cells (NSCs) have several characteristics that make them ideal as delivery vehicles for gene therapy in the brain. NSCs are migratory towards areas of brain pathology; this is mediated in part by chemokines and microenvironments produced within the altered brain.(1) NSCs can also be easily isolated and genetically manipulated in vitro, and can be subsequently reintroduced into the central nervous system. The way in which gene therapy is used depends on the specific disease or damage within the brain; however the general idea is to genetically manipulate cells in order to treat a disease.(1) Stem cell gene therapy uses stem cells as tools for rescuing neurons through the delivery of genes and therapeutic molecules, rather than replacing neurons as in cell-replacement therapy. Genes are normally introduced into NSCs by viral vectors, and these genes are used to restore a missing protein, deliver neurotrophic factors, or overexpress growth factors and neurotransmitters. For brain malignancies such as tumours, a variety of chemotherapeutic techniques have been developed. Most of these involve delivery of stem cells that produce an enzyme which can metabolize a prodrug in order to activate the destruction of a tumour.(1) These new gene therapy techniques are of much interest because they could provide non invasive therapies for diseases for which no curative treatments currently exist, such as Alzheimer’s disease.(2) Neural stem cells possess many promising features for gene therapy, such as the ability to migrate to hard to reach areas in the brain; however, there are still many challenges to the implementation of this type of therapy in clinical trials.(2)

3.1 Tropism towards brain pathology

One of the primary reasons the use of neural stem cells (NSCs) in gene therapy holds such promise is because of the great migratory abilities of NSCs.(1) In countless studies NSCs have been shown to be attracted to areas of brain pathology and to migrate to even hard to reach areas in the brain.(3) This is important since for NSCs to exert their therapeutic qualities in specific diseased tissues, they must first track down these cells; this is one of the greatest challenges of gene therapy. Both endogenous and transplanted NSCs seem to migrate to experimental brain lesions of various etiologies, from cancer to neurodegeneration.(1) In several reports, NSCs were transplanted into animal models of brain neoplasia and were subsequently found to have migrated to metastatic tumour cells far from the site of initial transplantation. Such pathotropism could be especially useful to target hard to attain areas, minute brain metastases, and malignant satellites that have dislodged from the main tumour mass or that remain after tumour removal. Neural stem cells can even migrate from one hemisphere to the other.(1)

NSC migration to experimental brain tumour in mouse model. Source : Muller, F., Snyder, E. Y., & Loring, J. F. (2006). Gene therapy: can neural stem cells deliver? Nat Rev Neurosci, 7(2), 75-82.

In the brain, NSC migration and development are controlled in part by the microenvironments in which they reside. The microenvironments that surround NSCs are occupied by microglia, astroglia, and endothelial cells; these are all important in regulating stem cell migration and differentiation in the normal brain.(1) When the environmental equilibrium is disturbed as a result of brain pathology, stem cells may experience unfamiliar or unusual factors with a potential to affect stem cell behaviour. The exact molecular basis of NSC pathotropism is not clearly defined, and different pathologies involve different factors and signals.(1)

3.1.1 Pathotropic factors

Different brain pathologies emit different chemotactic signals; many of these are cytokines, most commonly of the chemokine superfamily. Chemokine production is an especially common feature of brain lesions, such as stroke and brain malignancy.(1) Cytokines potentially involved in NSC pathotropism: CCL2/MCP-1 (chemokine (C-C motif) ligand 2), CXCL12/SDF-1 (chemokine (C-X-C motif) ligand 12), CX3CL1 (chemokine (C-X3-C motif) ligand 1), SCF (stem cell factor), VEGF (vascular endothelial growth factor), and HGF (hepatocyte growth factor).(1)(4)

3.1.2 Receptors

Receptors expressed by neural stem cells (NSCs) encounter chemotactic signals emanated from brain pathologies, which are often present as gradients.(1) The signals, or ligands, pair with the specific receptors on the surface of NSCs and these interactions direct the NSCs towards the source of the signals by following a gradient. Some of these pairs, as well as the disease models in which they were examined, can be seen in Table 1.(4)

Source: Carney, B. J., & Shah, K. (2011). Migration and fate of therapeutic stem cells in different brain disease models. Neuroscience, 197, 37-47.

3.1.3 Physiological processes shaping migration

There are several physiological processes that shape migration of transplanted neural stem cells, two of which are inflammation and reactive astrocytosis.(1)

One of the most favoured hypotheses for neural stem cell migratory behaviour is that NSCs are attracted by the inflammation that results from brain pathology.(5) Brain inflammation is harmful to the brain in general and microglia are usually the first to respond to damage by producing cytokines. It has been shown that microglia and inflammatory cytokines can direct the migration of NSCs in vitro and in vivo, suggesting that immune cells have a role in NSC mobilization.(1)(6)

Reactive astrocytosis:
When microglia release inflammatory cytokines due to brain pathology, reactive astrocytosis may occur. Reactive astrocytosis involves hypertrophy, hyperplasia, and an increase in glial fibrillary acidic protein (GFAP).(1) Reactive astrocytosis, which is affected by the type of pathology, the type of cytokines produced, and the proximity of the astrocytes, results in the secretion of certain factors. These factors, as well as astrocytes themselves, have been shown to be partially responsible for the directed migration of neural stem cells.(1)

3.2 Neural stem cells as delivery vehicles

Neural stem cells (NSCs) have several advantages which make them suitable as delivery vehicles in gene therapy. NSCs can potentially integrate into the brain without disruption of normal function and they can be propagated for long lengths of time in vitro, allowing for genetic manipulation.(1) The great diversity of neural stem cell lines must be considered, however, to determine which are suitable for use in therapy.(1) Ideally, the chosen cell line would have the following characteristics:
  • Stable in tissue culture
  • Capable of expressing therapeutic molecules in a sustained and regulated manner
  • Predictable differentiation patterns in culture and after transplantation
  • Able to survive long term in vivo without the formation of tumours
  • Responsiveness to chemotactic signals emitted by the pathology of interest
  • Easy delivery
Many questions remain as to the nature of the cells to be used; however, many aspects are dependent on the specific pathology of interest.(1)

3.2.1 Primary cells

Primary cell cultures have a limited lifespan and therefore pose significant limitations for use as delivery vehicles in gene therapy. The chance that a small group of live NSCs would suffice to regenerate damaged tissue is low, therefore cell expansion in culture would be necessary to ensure a sufficient number of viable NSCs.(1) The amount of NSC expansion would need to be determined empirically for each individual disease and expansion would be difficult due to the short lifespan of these cells. Currently, problems also exist in relation to consistency and agreement on a common set of protocols for culture conditions and characterization.(1)(7)

3.2.2 Immortalization

Immortalization normally means the introduction of an oncogene; this allows expansion of neural stem cells (NSCs) beyond the time of usual senescence.(1)(8) Immortalized NSCs have unusual characteristics compared to most NSCs, such as exceptional migratory abilities and a high degree of multipotency, which could increase the probability of tumour development by fast growing NSCs. Sometimes immortalized NSCs pose too great a risk for therapeutic use, but in some cases the therapeutic value of enhanced migratory abilities and invasiveness outweighs the safety concerns.(1) Another advantage of immortalization is that it can enable almost indefinite propagation of NSCs with definable characteristics, which can be useful for analysis and quality control.(1)(8) Growth factors that improve long-term culture of NSCs have now made immortalization less essential and eliminated some safety concerns.(1)

3.2.3 Embryonic stem cells and embryonic stem cell-derived cells

There are many advantages to using embryonic stem cells (ESCs) or embryonic stem cell-derived cells over other types of stem cells for gene therapy. ESCs, derived from cells of the inner cell mass of blastocyst embryos, have immortal characteristics and do not senesce even after long lengths of time in culture.(1) In addition, ESC differentiation is not limited to the nervous system, unlike neural stem cells. For different pathologies of the brain, ESCs can be induced to differentiate along certain lineages, and isolated at a stage resembling somatic NSCs.(1) ESCs and ESC-derived cells can be tracked using fluorescent proteins and, advantageously, there is agreement on characterization of ESCs, meaning researchers can begin with a common population of cells.(1)(9) There are challenges to using ESCs and ESC-derived cells in gene therapy, for example, since the ESC-derived cells do not experience the normal developmental signals, they may not develop the appropriate phenotype, resulting in variable outcomes in vivo.(1)

It is currently believed that ESC-derived cells will be some of the most relevant for gene therapy in the brain; however, there is little data comparing the performance of ESC-derived cells to other NSC lines and there is insufficient evidence to favour one type over another.(1)

3.3 Genetic manipulation

In contrast to cell replacement therapy, gene therapy involves using cells as a means of gene delivery for rescuing neurons rather than replacing them. Gene expression vectors are under investigation for use in various diseases caused by deficiency in some crucial factor, as these vectors have the potential to restore the missing factor.(1)

The concept of using neural stem cells in gene therapy originated with lysosomal storage diseases. In these diseases there is a mutated enzyme which causes a metabolic obstruction in upstream pathways. As a result, substrates and side products build up in the cells and the surrounding tissues, a very destructive process.(1) It was hypothesized, and subsequently demonstrated, that if a functional replacement gene was targeted and expressed in the affected region, it would restore the logjam in the metabolic pathway.(1)

This concept has now been expanded to encompass any genetic manipulation of NSCs to treat a lesion or disease. The exact manner in which NSCs are used in therapy is dependent on the specific pathology, but in general the neural stem cells are isolated and proliferated, followed by genetic transduction and differentiation; all of this happens in vitro.(1) Subsequently, the modified NSCs are reintroduced into the brain. The gene that was newly introduced into the NSC encodes a protein and when this protein is expressed the idea is that it will compensate for some kind of deficit that is contributing to the brain pathology.(1) This “gene delivery” can result in deliverance of enzymes, cytokines, neurotrophic factors, growth factors, neurotransmitters, or other proteins.(1)

One of the most common approaches for pathologies such as tumours is the use of an enzyme that metabolizes a non-toxic prodrug into the active form of the drug locally. This prodrug must also be administered and the active form is then able to exert its effects on the tumour.(1) There are still many challenges to the implementation of this type of therapy, most notably in determining which NSCs are appropriate for each purpose, which genes and chemicals can be delivered, and which pathologies are suitable targets. For these reasons, as well as safety concerns, most therapies are still in preclinical or clinical testing phases.(1)

3.3.1 Viral vectors

In terms of the genetic modification, neural stem cells can be genetically transduced in vitro or in vivo. At present, the most common and efficient manner of introducing genes into NSCs is by use of viral vectors in vitro. The main concerns with this approach are that the inserted transgene may be silenced or it may activate a nearby oncogene.(1)

It is very challenging to introduce DNA into cells without causing harm, however, by using viruses which have a natural capacity for introducing genes into the cells of the host, this can be accomplished efficiently.(10) The use of viral vectors for transducing stem cells has expanded in recent years, and this technology is being continuously developed to increase its safety and practicality. Through genetic engineering, genes involved in pathogenesis and replication are deleted and specific transgenes are inserted, enabling viruses to serve as effective tools for gene transfer.(10) These viral vectors will introduce the transgenes into NSCs, but will fail to replicate. Some replication genes are required for packaging and these are supplied as trans-acting elements.(10)

Production of viral vectors. Source: Thomas, C. E., Ehrhardt, A., & Kay, M. A. (2003). Progress and problems with the use of viral vectors for gene therapy. Nat Rev Gen, 4, 346-358.

Viral vectors have several characteristics which make them particularly attractive for use in manipulation of cells of the brain, such as specificity of target cell type and capability of reporter gene co-expression.(10) Tropism towards, and attachment and infection of, specific NSCs is mediated by the particular proteins present on the viral envelope or capsid. This cell specificity can be modified by pseudotyping, in which the capsid or envelope proteins of one virus type are substituted for those of another.(10) Ideally, a gene delivery vector would package the inserted transgene, then target a certain subpopulation of cells in the brain or in vitro and express the transgene in these cells.(10)

Viral vectors derived from adenovirus, adeno-associated virus, herpes simplex virus type 1, and most commonly lentivirus (such as HIV-1), have been found useful for NSC transduction for gene therapy in the brain.(10) Each virus has its own features and disadvantages such as target cell specificity, duration of transgene expression, immune responses, and biosafety.(10)

After years of preclinical progress, several gene therapies using viral vectors are now in Phase I and Phase II clinical trials, showing some encouraging results.(11) Many of the viral vectors in clinical use maintain lifetime expression in the brain, and currently research is being done to try and find a mechanism that could control the expression of the transgene product through the application of a prodrug; this would be valuable if the continued protein expression were to become problematic long-term.(12)

3.4 Suitable disease targets and therapeutic strategies

3.4.1 Alzheimer’s disease

In Alzheimer’s disease the degeneration of neurons is widespread, and encompasses the hippocampus, amygdala, and cortex. Current gene therapy strategies focus on using stem cells to deliver neurotrophic factors that protect neurons from degeneration; it is also thought that neurotrophic factors might play a role in re-activating damaged circuitry caused by the disease.(1) Nerve growth factor (NGF) has been the subject of much interest as it is important for growth and survival of neurons, and has proven to reduce neural degeneration.(1) Clinical trials using fibroblasts and adeno-associated virus vectors to deliver NGF to the brain have shown some success, and it is now thought that the migratory capabilities of neural stem cells might be utilized to target therapeutic enzymes directly to amyloid plaques.(1)(13)

3.4.2 Parkinson`s disease

Parkinson’s disease (PD) involves the specific degeneration of dopaminergic neurons of the substantia nigra, which leads to loss of dopamine production through the depletion of dopamine synthesizing enzymes. For Parkinson’s disease the main techniques that are currently in Phase I and Phase II clinical trials involve using viral vectors to transport enzymes to the striatum to improve symptoms and possibly provide neuroprotection against the disease.(14) The method of improving in vivo dopamine production by expressing transgenes in neural stem cells and distributing them to specific areas of the brain is of current interest, but is not the focus of treatment strategies.(1)

Adeno-associated virus (AAV) is one of the most popular and common vectors to use in experimental treatments of Parkinson’s disease.(14) It is relatively safe and easy to use, and targets the basal ganglia, therefore eliminating the need for systemic medication (and the side effects that come with it). Various studies use AAV vectors to express different proteins, such as tyrosine hydroxylase and L-amino acid decarboxylase (AADC).(14) Tyrosine hydroxylase results in enhanced production of dopamine at the site of injection and has shown good outcomes in clinical trials. The vector carrying the AADC gene is injected into the striatum to increase the local conversion of exogenous L-DOPA, a precursor to dopamine, to dopamine.(15)
Recently, the concept of cerebral delivery of trophic factors, that function to help protect dopaminergic neurons, has been explored.(1) Production of glial-derived neurotrophic factor (GDNF) has been genetically induced via an AAV vector and has been shown to have protective effects on the survival of the midbrain dopaminergic neurons. This has resulted in mixed outcomes in animal models and humans, with some significant positive findings.(14)

Many gene therapies for Parkinson’s disease have shown sustained behavioural improvement in animal models; subsequently, there have also been positive, significant results in clinical trials with humans.The viral mediated gene therapy approaches described above could also make use of the migratory abilities of neural stem cells, however, currently there is insufficient experiment data on this subject. (14)

3.4.3 Lysosomal storage diseases

Lysosomal storage diseases are a group of rare brain metabolism disorders resulting from genetic defects in lysosomal function. These were some of the first diseases in the brain to be considered for treatment with gene therapy and are some of the most promising targets for this type of therapy.(1) In lysosomal storage diseases there is a mutated enzyme which causes a metabolic obstruction in upstream pathways. As a result, substrates and side products build up in the cells and the surrounding tissues, a very destructive process.(1) It was shown that if a functional replacement gene was targeted and expressed in the affected region, it would restore the logjam in the metabolic pathway. Genetically normal neural stem cells have the ability to replace the missing enzymatic activities in these diseases and dispersion of these cells throughout the brain can result in widespread enzymatic change.(1) Treatment using this type of technique has shown good outcomes and now the timing of treatment is in question; this therapy could even potentially be carried out in utero.(1)(16)

3.4.4 Brain malignancy

Neural stem cells have excellent migratory capabilities and they are attracted to brain malignancies including primary and secondary invasive gliomas, melanoma brain metastases, medulloblastoma, and neuroblastoma.(2)(3) For this reason, brain tumours are one of the most promising targets of NSC gene therapy. The genetically modified, tumour-tropic NSCs are able to naturally target specific primary and metastatic cancers in the brain and provide chemotherapy to these regions with significant beneficial effects.(1) Many questions remain for the clinical implementation of this type of technology, such as choice of neural stem cell type and its migratory targets, and specific genetic modifications applied to the stem cells. There are other considerations as well, such as whether the genetic transformations will lead to a successful therapeutic outcome in vivo.(1) Many approaches have been taken for these types of disease targets: prodrug-converting enzymes, immunomodulatory cytokines, cytokines with anti-tumoural activity, and even lytic viruses to destroy tumour cells.(1)

Stem cell migration to tumour in mouse model. Source: Weissleder, R., & Ntziachristos, V. (2003). Shedding light onto live molecular targets. Nat Med, 9 (1), 123-8.

In a recent study, human neural stem cells that were retrovirally transduced with the enzyme cytosine deaminase (CD) gene showed successful migration to experimental gliomas in the brain.(17) After application of a prodrug, 5-fluorocytosine (5-FC), they showed outstanding bystander killer effect, a process in which toxic metabolites or the therapeutic effect are transferred to neighbouring cancer cells.(2)(17) A subsequent investigation showed that using NSCs genetically altered to express both CD and interferon-β, a cytokine known for its antitumour activity, resulted in an additive effect; the destruction of the tumours was significantly increased. Such studies are normally performed in animal models and before clinical trials are carried out many safety issues must be addressed; however, these studies provide evidence that a multimodal gene therapy approach could have great potential for treating tumours of the brain.(17)

3.4.4a Molecular and immunologic strategies
The gene therapy strategies for treatment of brain tumours can be grouped into two categories: molecular and immunologic.(2)

Molecular gene therapy:
This type of gene therapy is focused on the conversion of a non-toxic prodrug into an active drug that has a local therapeutic, anticancer effect. Genetically modified NSCs introduce enzymes that carry out the prodrug conversion and the prodrug itself must be administered separately. The goal is to inhibit the proliferation of cancer cells or to kill off the cancer cells by apoptosis.(2)

Immunologic gene therapy:
This type of gene therapy focuses on expression of immune-stimulating cytokines such as interleukin-4, interleukin-12, and TRAIL (tumour necrosis factor-related apoptosis-inducing ligand) by genetically modified neural stem cells. The main goal is to augment the T-cell mediated immune response against the tumour cells, resulting in their destruction.(2

Molecular and immunologic strategies. Source: Muller, F., Snyder, E. Y., & Loring, J. F. (2006). Gene therapy: can neural stem cells deliver? Nat Rev Neurosci, 7(2), 75-82.

  1. Muller, F., Snyder, E. Y., & Loring, J. F. (2006). Gene therapy: can neural stem cells deliver? Nat Rev Neurosci, 7(2), 75-82.
  2. Kim, S. U. (2011). Neural stem cell-based gene therapy for brain tumors. Stem Cell Rev and Rep, 7, 130-140.
  3. Kim, J. H., Lee, J. E., Kim, S. U., & Cho, K. G. (2010). Stereological analysis on migration of human neural stem cells in the brain of rats bearing glioma. Neurosurgery, 66(2), 333-42
  4. Carney, B. J., & Shah, K. (2011). Migration and fate of therapeutic stem cells in different brain disease models. Neuroscience, 197, 37-47.
  5. Ferrari, D., Zalfa, C., Rota Nodari, L., Gelati, M., Carlessi, L., Delia, D., Vescovi, A. L., & De Filippis, L. (2012). Differential pathotropism of non-immortalized and immortalized human neural stem cell lines in a focal demyelination model. Cell Mol Life Sci, 69(7), 1193-210.
  6. Popovich , P. G., & Longbrake, E. E. (2008). Can the immune system be harnessed to repair the CNS? Nat Rev Neurosci, 9, 481-493.
  7. Schinzel, R. T., Ahfeldt, T., Lau, F. H., Lee, Y. K., Cowley, A., Shen, T., Peters, D., Lum, D. H., Cowan, C.A. (2011). Efficient culturing and genetic manipulation of human pluripotent stem cells. PLoS One, 6(12), e27495.
  8. De Filippis, L., Ferrari, D., Rota Nodari, L., Amati, B., Snyder, E., & Vescovi, A. L. (2008). Immortalization of human neural stem cells with the c-myc mutant T58A. PLoS One, 3(10), 3310.
  9. Dhara, S. K., Gerwe, B. A., Majumder, A., Dodla, M. C., Boyd, N. L., Machacek, D. W., Hasneen, K., & Stice, S. L. (2009). Genetic manipulation of neural progenitors derived from human embryonic stem cells. Tissue Eng Part A, 15(11), 3621-34.
  10. Edry, E., Lamprecht, R., Wagner, S., & Rosenblum, K. (2011). Virally mediated gene manipulation in the adult CNS. Front. Mol. Neurosci, 4, 57.
  11. Nobre, R. J., & Almeida, L. P. (2011). Gene therapy for Parkinson's and Alzheimer's diseases: from the bench to clinical trials. Curr Pharm Des, 17 (31), 3434-45.
  12. Manfredsson, F. P., Bloom, D. C., & Mandel, R. J. (2012). Regulated protein expression for in vivo gene therapy for neurological disorders: Progress, strategies, and issues. Neurobiol. Dis, 1-10.
  13. Wu, S., Sasaki, A., Yoshimoto, R., Kawahara, Y., Manabe, T., Kataoka, K., Asashima, M., & Yuge, L. (2008). Neural stem cells improve learning and memory in rats with Alzheimer's disease. Pathobiology, 75(3), 186-94.
  14. Hickey, P., & Stacy, M. (2011). Available and emerging treatments for Parkinson’s disease: a review. Drug Des Devel Ther, 5, 241 - 254.
  15. Rodnitzky, R. L. (2011). Upcoming treatments in Parkinson’s disease, including gene therapy. Parkinsonism & Related Disorders, 18, S37–S40.
  16. Lau, A. A., Rozaklis, T., Ibanes, S., Luck, A. J., Beard, H., Hassiotis, S., Mazouni, K., Hopwood, J. J., Kremer, E.J., & Hemsley, K.M. (2012). Helper-dependent canine adenovirus vector-mediated transgene expression in a neurodegenerative lysosomal storage disorder. Gene, 491(1), 53-7.
  17. Ito, S., Natsume, A., Shimato, S., Ohno, M., Kato, T., Chansakul, P., Wakabayashi, T., & Kim, S. U. (2010). Human neural stem cells transduced with IFN-β and cytosine deaminase genes intensify bystander effect in experimental glioma. Cancer Gene Therapy, 17, 299-306.