Final Report Summary - RESCUE (From stem cell technology to functional restoration after spinal cord injury)
The objective of the project was to explore the ability of human stem cells to repair spinal cord traumatic injury. This required the selection of appropriate neural and non neural stem cells, their cultivation and eventual transformation, the grafting in appropriate models of rat spinal cord injury, the functional and anatomical evaluation of the influence of grafted cells.
Worth recalling is that there does not exist today any specific therapy of spinal cord injury, that several trials using neuroprotective molecules have been so far unsatisfactory, and that uncontrolled human therapies using poorly defined 'stem cells' have been carried in China and in Europe without appreciable results. The only controlled study was carried by the group of Mackay-Sim in Australia, a phase I/IIa study using Olfactory bulb ensheathing cells (OBEC), which concluded for the absence of unwanted secondary effects, but did not report any functional improvement (Mackay-Sim, 2008).
The characterisation of resident stem cells in the human and rodent spinal cord was successfully achieved, and the analysis of their fate in control conditions and after injury (for rodent cells) is progressing. This is a long term objective which was not initially planned in the project, but appeared as a logical follow-up in the course of the study.
The expansion on a large scale of adult neural stem / progenitor cells derived from human preoperative biopsies was achieved for one specimen, and transferred to other groups for grafting. It appeared that tumours formed from these grafts, and this part of the project was discontinued. One of the objectives of the project was indeed to explore the safety of this procedure, and this objective was reached.
The differentiation of human neural precursor derived astrocytes was successfully achieved, and the ability of type 1 astrocytes to support axonal growth was evidenced, due both to cell derived extracellular matrix and diffusible released growth factors. These astrocytes were embedded in collagen scaffolds and were found to improve the growth of axons in a paradigm of hemisected dorsal root ganglion explants.
It was found that adipose stromal stem cells provided a better source of cells than the classical bone marrow preparation. Cells formed spheres expressing the neural marker NCAM, and upon dissociation expressed beta3 tubulin. The growth of these cells on biocompatible scaffolds was extensively analysed, and one of these gels was found to be adequate for this purpose. Since human nestine-positive mesenchymal stem cells were found unable to trans-differentiate into neural cell types, the initial project was modified and they were used as substrates to promote axonal elongation. Standard conditions for their culture in absence of foetal calf serum were defined. The production by these cells of diffusible trophic factors was extensively studied, and an important point appeared that there exists a large variability in donor phenotype, thus urging for extensive phenotyping prior to autograft cellular therapy.
The orientation of the cellular differentiation and subsequent cell sorting was carried using the bHLH transcription factors Mash 1 and Neurogenins 1, 2 and 3. Lentiviral vectors were used to drive the expression of these proneural genes under the control of the PGK promoter in neural stem cells extracted from the telencephalic vesicles of human aborted foetuses. The proliferative ability and phenotype of these cells after transfection was extensively analysed, and it was decided that cells transfected with Ngn2 were suitable for use for transplantation studies. Besides, a new kind of non integrative lentiviral vector was developed, which was intended to provide a stable gene expression in quiescent cells, but a transient gene expression in dividing cells. This vector may be particularly suited for clinical applications where maximal safety is required.
The generation of human, conditionally-immortalised NSC lines was successfully obtained. Four lines were generated from eight weeks old human spinal cords, and they were characterized for their developmental potential, the plasticity of their positional specification, and last but not least their allelic expression. The four lines exhibited upon differentiation neuronal, astrocytic and oligodendrocytic markers, and formed neurospheres. One of these lines was found to be tetraploid and was discarded. Regarding positional expression, these cells were found to express only ventral markers. They could be further 'dorsalised' and 'rostralised' by appropriate treatment. Regarding allelic expression, a study not initially planned, it was found that there exists a substantial stochastic monoallelic gene expression in neural stem cells, as found previously in lymphoblastic cells. It is of great interest to investigate the extent in which this monoallelism may contribute to neural stem cell functional diversity.
In conclusion, the main trends that emerge from a considerable number of experiments carried in parallel and / or in cooperation by the nine groups are the following:
- Human stem cells show such a variety of phenotypes and behaviours that only a few carefully identified sets could be investigated in this project. Of further interest in this respect is the variety of mono-allelic expression.
- Stem cells intrinsic to the human spinal cord could be a potential source of therapeutic cells / factors. Further characterisation of phenotypes is clearly needed.
- Grafted human stem cells, whatever their origin and further alterations, were never found to integrate into spinal cord circuitry, at variance with homografts of foetal neurons.
- Grafted human stem cells demonstrated their ability to improve functional recovery, most probably through a trophic influence on the regeneration / sprouting of intrinsic fibre systems. This influence is transient, as illustrated by the almost total absence of survival after a few weeks in most of the instances where they were functionally effective.
- Cell therapies involving in situ transplantation of foreign cells clearly induce aversive effects per se. Great care should be taken for future human clinical trials to weigh carefully the risk / benefit ratio when comparing cells therapies with drug and / or gene therapies.
Worth recalling is that there does not exist today any specific therapy of spinal cord injury, that several trials using neuroprotective molecules have been so far unsatisfactory, and that uncontrolled human therapies using poorly defined 'stem cells' have been carried in China and in Europe without appreciable results. The only controlled study was carried by the group of Mackay-Sim in Australia, a phase I/IIa study using Olfactory bulb ensheathing cells (OBEC), which concluded for the absence of unwanted secondary effects, but did not report any functional improvement (Mackay-Sim, 2008).
The characterisation of resident stem cells in the human and rodent spinal cord was successfully achieved, and the analysis of their fate in control conditions and after injury (for rodent cells) is progressing. This is a long term objective which was not initially planned in the project, but appeared as a logical follow-up in the course of the study.
The expansion on a large scale of adult neural stem / progenitor cells derived from human preoperative biopsies was achieved for one specimen, and transferred to other groups for grafting. It appeared that tumours formed from these grafts, and this part of the project was discontinued. One of the objectives of the project was indeed to explore the safety of this procedure, and this objective was reached.
The differentiation of human neural precursor derived astrocytes was successfully achieved, and the ability of type 1 astrocytes to support axonal growth was evidenced, due both to cell derived extracellular matrix and diffusible released growth factors. These astrocytes were embedded in collagen scaffolds and were found to improve the growth of axons in a paradigm of hemisected dorsal root ganglion explants.
It was found that adipose stromal stem cells provided a better source of cells than the classical bone marrow preparation. Cells formed spheres expressing the neural marker NCAM, and upon dissociation expressed beta3 tubulin. The growth of these cells on biocompatible scaffolds was extensively analysed, and one of these gels was found to be adequate for this purpose. Since human nestine-positive mesenchymal stem cells were found unable to trans-differentiate into neural cell types, the initial project was modified and they were used as substrates to promote axonal elongation. Standard conditions for their culture in absence of foetal calf serum were defined. The production by these cells of diffusible trophic factors was extensively studied, and an important point appeared that there exists a large variability in donor phenotype, thus urging for extensive phenotyping prior to autograft cellular therapy.
The orientation of the cellular differentiation and subsequent cell sorting was carried using the bHLH transcription factors Mash 1 and Neurogenins 1, 2 and 3. Lentiviral vectors were used to drive the expression of these proneural genes under the control of the PGK promoter in neural stem cells extracted from the telencephalic vesicles of human aborted foetuses. The proliferative ability and phenotype of these cells after transfection was extensively analysed, and it was decided that cells transfected with Ngn2 were suitable for use for transplantation studies. Besides, a new kind of non integrative lentiviral vector was developed, which was intended to provide a stable gene expression in quiescent cells, but a transient gene expression in dividing cells. This vector may be particularly suited for clinical applications where maximal safety is required.
The generation of human, conditionally-immortalised NSC lines was successfully obtained. Four lines were generated from eight weeks old human spinal cords, and they were characterized for their developmental potential, the plasticity of their positional specification, and last but not least their allelic expression. The four lines exhibited upon differentiation neuronal, astrocytic and oligodendrocytic markers, and formed neurospheres. One of these lines was found to be tetraploid and was discarded. Regarding positional expression, these cells were found to express only ventral markers. They could be further 'dorsalised' and 'rostralised' by appropriate treatment. Regarding allelic expression, a study not initially planned, it was found that there exists a substantial stochastic monoallelic gene expression in neural stem cells, as found previously in lymphoblastic cells. It is of great interest to investigate the extent in which this monoallelism may contribute to neural stem cell functional diversity.
In conclusion, the main trends that emerge from a considerable number of experiments carried in parallel and / or in cooperation by the nine groups are the following:
- Human stem cells show such a variety of phenotypes and behaviours that only a few carefully identified sets could be investigated in this project. Of further interest in this respect is the variety of mono-allelic expression.
- Stem cells intrinsic to the human spinal cord could be a potential source of therapeutic cells / factors. Further characterisation of phenotypes is clearly needed.
- Grafted human stem cells, whatever their origin and further alterations, were never found to integrate into spinal cord circuitry, at variance with homografts of foetal neurons.
- Grafted human stem cells demonstrated their ability to improve functional recovery, most probably through a trophic influence on the regeneration / sprouting of intrinsic fibre systems. This influence is transient, as illustrated by the almost total absence of survival after a few weeks in most of the instances where they were functionally effective.
- Cell therapies involving in situ transplantation of foreign cells clearly induce aversive effects per se. Great care should be taken for future human clinical trials to weigh carefully the risk / benefit ratio when comparing cells therapies with drug and / or gene therapies.
