A new method for the rapid and long term growth of human neural precursor cells

https://doi.org/10.1016/S0165-0270(98)00126-5Get rights and content

Abstract

A reliable source of human neural tissue would be of immense practical value to both neuroscientists and clinical neural transplantation trials. In this study, human precursor cells were isolated from the developing human cortex and, in the presence of both epidermal and fibroblast growth factor-2, grew in culture as sphere shaped clusters. Using traditional passaging techniques and culture mediums the rate of growth was extremely slow, and only a 12-fold expansion in total cell number could be achieved. However, when intact spheres were sectioned into quarters, rather than mechanically dissociated, cell–cell contacts were maintained and cellular trauma minimised which permitted the rapid and continual growth of each individual quarter. Using this method we have achieved a 1.5 million-fold increase in precursor cell number over a period of less than 200 days. Upon differentiation by exposure to a substrate, cells migrated out from the spheres and formed a monolayer of astrocytes and neurons. No oligodendrocytes were found to develop from these human neural precursor cells at late passages when whole spheres were differentiated. This simple and novel culture method allows the rapid expansion of large numbers of non-transformed human neural precursor cells which may be of use in drug discovery, ex vivo gene therapy and clinical neural transplantation.

Introduction

There is at present no source of non-transformed human neurons other than primary foetal tissue. This has posed major limitations to both research and industry with regard to studying the basic biology and drug responsiveness of human neurons, and has limited clinical neural transplantation programmes to very small numbers of patients. The only strategy currently available for obtaining large amounts of well characterised human neurons is the use of cell lines. There have been two such human lines described previously, one derived from a tetratocarcinoma and the other oncogenically transformed (Pleasure and Lee, 1993, Sah et al., 1997). Although of great interest, transformed cell lines may not share the exact features of primary human neural tissue, and their oncogenic status makes them less attractive as a source of tissue for clinical transplantation. An alternative approach to producing large amounts of neural tissue is to isolate and expand neural precursor cells from the CNS.

The ideal human precursor cell to expand would be a neural stem cell. A stem cell can be most simply defined as any cell which is capable of self renewal for extended periods of time, the progeny from which are capable of forming the components of a defined tissue. Asymmetric division allows stem cells to generate a progenitor cell, in addition to another stem cell, which have a limited potential for self renewal and often spontaneously stop dividing and differentiate. Stem cells have been most extensively studied in haemopoetic, epidermal and intestinal tissues which require frequent cell replacement throughout life (Hall and Watt, 1989). Recent studies have shown that specific regions of both the developing and adult rodent brain harbour cells which divide in response to mitogens, while retaining the capacity to differentiate into neurons and glia, and as such may represent neural stem cells (Weiss et al., 1996, McKay, 1997, Palmer et al., 1997), although stem cell status is often debated and the term neural precursor may better describe cells within these heterogeneous cultures. Neural precursor cells from the rodent respond to both epidermal growth factor (EGF) and fibroblast growth factor (FGF-2) (for reviews see Gage et al., 1995, McKay, 1997) and can be grown as either monolayer cultures or as free floating spherical aggregates termed `neurospheres' (Reynolds et al., 1992). The short term growth (<60 days) of similar human CNS precursors has recently been reported (Buc-Caron, 1995, Svendsen et al., 1996, Chalmers-Redman et al., 1997, Murray and Dubois-Dalcq, 1997) and in some cases these can survive, migrate, differentiate and restore function following transplantation into rat models of Parkinson's disease (Svendsen et al., 1997a). However, we and others have also shown that human neurospheres are difficult to expand in vitro over long periods of time (Svendsen et al., 1997a, Quinn et al., 1997). Furthermore, we have also shown that rat and mouse neurospheres, grown using identical methods, have very different long term expansion potentials with the rat cells entering senescence within 3–4 weeks of expansion (Svendsen et al., 1997b). Thus, there may be a significant species difference when developing methods for the growth and differentiation of these cells. Clearly, if neural precursor cells are to become a source of tissue for basic neuroscience and clinical programmes it would be a major advantage if they could be expanded for long periods of time.

When attached cells or free floating aggregates reach the end of a growth cycle, they must be mechanically broken up or `passaged', often using digestion enzymes, to avoid contact mediated growth arrest or lack of nutrient diffusion. We postulated that these standard passaging techniques may lead to cellular trauma, strip receptors, deprive cells of contact mediated factors and remove vital tight junctions known to hold tissues together. This may lead to either the terminal differentiation of precursor cells, or a lack of response to mitogens for the rat and human cells. We therefore attempted to adapt the passaging technique such that enzymatic or mechanical disturbance to the cells was minimised and then assess the ability of the human neural precursor cells to continuously renew over time.

Section snippets

Tissue collection

Human fetal tissue (between 7 and 21 weeks post conception) was collected from two different sources: via the Uniform Anatomical Gift Act of the United States or from a local hospital. The methods of collection conform with the arrangements recommended by the Polkinghorne Committee for the collection of such tissues and the guidelines set out by the Department of Health in the United Kingdom.

Cell culture

Tissues collected locally (see Table 1 for details) were dissected in chilled sterile phosphate buffered

Expansion of human precursors

Following seeding into growth medium, aggregates of dividing cells formed into spheres which grew in size over time in response to the mitogens EGF and FGF-2. Between 14 and 21 days of growth the spheres could be gently dissociated to a mixed suspension of single cells and sphere remnants before re-plating into growth medium. Using this technique we have previously reported a 3-4-fold increase in cell number over the first few weeks in vitro, after which the absolute number of cells harvested

Discussion

The present study has demonstrated a new method for the long-term exponential expansion of non immortalised or transformed human neural precursor cells, which maintained the capacity to generate a high percentage of neurons (see Fig. 7 for a schematic of the technique).

There are two main methods in the literature commonly used to generate populations of neural precursor cells. The first uses FGF-2 and a substrate to expand colonies of cells which grow attached to the culture flask while the

Acknowledgements

We thank S.B. Dunnett for his continual support, Biorad Inc. for preparation of the confocal images, and Ziggy Zhang and Irena Sarel of Blowhittiker Inc. for providing human neurospheres. The authors would also like to thank Dr Scott Whittemore for critically appraising an early version of this manuscript. This research was funded by a Wellcome Fellowship to CNS and by the MRC.

References (36)

  • Y Arsenijevic et al.

    Insulin-like growth factor-I is a differentiation factor for post-mitotic CNS stem cell-derived neuronal precursors: distinct actions from those of brain-derived neurotrophic factor

    J Neurosci

    (1998)
  • A Bjorklund

    Better cells for brain repair

    Nature

    (1993)
  • Caldwell MA, Svendsen CN. Potentiation of FGF-2 induced proliferation of mesencephalic neural precursors by heparin....
  • Ciccolini F, Svendsen CN. FGF-2 promotes acquisition of EGF responsiveness in mouse striatal precursor cells:...
  • Gage FH, Ray J, Fisher LJ. Isolation, characterisation, and use of stem cells from the CNS. In: Cowan WM, Shooter EM,...
  • A Gritti et al.

    Multipotential stem cells from the adult mouse brain proliferate and self-renew in response to basic fibroblast growth factor

    J Neurosci

    (1996)
  • P.A Hall et al.

    Stem cells: the generation and maintenance of cellular diversity

    Development

    (1989)
  • K Joh et al.

    Single factors direct the differentiation of stem cells from the fetal and adult central nervous system

    Genes Dev

    (1996)
  • Cited by (504)

    • A novel treatment for Parkinson's disease and ALS: Combined cell and gene therapies

      2020, Neurotherapeutics in the Era of Translational Medicine
    View all citing articles on Scopus
    View full text