First printed in R&D Systems' 1996 Catalog.
Since the discovery of nerve growth factor1, many neurotrophic agents have been characterized. In a very general sense, these neurotrophins can be classified according to their receptors; with one group utilizing receptors that associate with cytoplasmic tyrosine kinases (CNTF and LIF)2,3 and another group utilizing receptors that possess intrinsic kinase activity, either tyrosine kinase activity (IGF, FGF, EGF, and members of the neurotrophin family)4-10, or serine/theronine kinase activity (TGF beta family).11,12 Although almost all of the above factors can be shown to impact a range of neuronal phenotypes and/or developmental stages, some phenotypes are more commonly associated with particular neurotrophins. Glial-cell line derived neurotrophic factor (GDNF) is an approximately 20 kDa glycosylated polypeptide that has distant, but significant, homology to members of the TGF beta superfamily.13 For members of the TGF beta superfamily, the neuronal phenotype of most interest experimentally seems to be that of the dopamine-containing neuron.12,14 This also appears to be the case for GDNF. While recent reports have materially expanded the type of target cell for GDNF, the originally-reported effects of GDNF on dopaminergic neurons still are the subject of the most intense interest. Indeed, GDNF, along with TGF beta 2 and 3, are now considered keys to an understanding of the general development and clinically-relevant iatrogenic survival of the dopaminergic system.14,15
GDNF is an approximately 20 kDa, glycosylated polypeptide that exists in its native form as a homodimer.13 The gene for GDNF has been mapped to human chromosome 5 at p12-p13.116, and gives rise to two alternatively-spliced forms that code for prepropeptides of 211 and 185 amino acid (aa) residues respectively.17,18 The 26 aa residue deletion in the short form occurs in the pro-segment of the precursor. Thus both the long (alpha) and short (beta) forms yield equivalent 134 aa residue mature forms after proteolytic cleavage.13,16 As noted above, GDNF is now considered a member of the TGF beta superfamily.13 Structural characteristics found in GDNF and common to all TGF beta members include seven conserved cysteine residues and the capacity for formation of disulfide-bonded homodimers.13,19 Although the native molecule contains two potential glycosylation sites, non-glycosylated recombinant GDNF has full biological activity.13 Rat to human, rat GDNF shows remarkable aa residue sequence conservation in the long or alpha form, with 100% identity noted in the 19 aa residue signal sequence, 90% identity seen in the 58 aa residue pro-segment, and 93% identity recorded in the 134 aa residue mature peptide sequence.13 Not surprisingly, GDNF shows considerable species cross-reactivity.13,20 Cells known to express GDNF include Sertoli cells (21), type 1 astrocytes22,23, Schwann cells17,24, neurons25,26, pinealocytes27, and (most likely) skeletal muscle cells.17,24
To date, no receptor has been identified for GDNF. However, preliminary binding studies using chick sympathetic neurons show a population of GDNF receptors with a Kd = 1-5 nM.21 Compared to other known TGF beta receptors, a Kd of 1 nM (1000 pM) for GDNF would be considerably lower than the Kds for TGF beta 1's binding to its 80 kDa Type II receptor (Kd = 25 pM) and 110 kDa Type III receptor (Kd = 200 pM).11
The activity most often attributed to GDNF is that of promoting neuron survival. This is manifested in many forms by many different neuron cell types. For example, dopaminergic neurons removed from embryonic midbrains can be rescued from destruction in vitro by GDNF.12 In addition, GDNF can interrupt the apoptotic programs of embryonic avian motor neurons in vivo.20 Furthermore, autonomic (involuntary) motor neurons of both sympathetic and parasympathetic systems also respond to GDNF by maintaining their numbers in vitro at significant levels relative to controls.27 In vivo, following transection of (voluntary) facial motor neuron axons, exogenously applied GDNF has been shown to rescue virtually all damaged neurons from death, while transected, non-treated neurons almost all perish.28 During neuronal development, GDNF appears to cooperate with TGF beta 2 and 3 in providing target-specific survival factors for migrating dopaminergic axons. Remarkably, it has been shown that at embryonic day 15.5 in the rat (day 47 in the human), developing dopaminergic neurons in the midbrain have access to, and apparently utilize both GDNF and TGF beta 2 for trophic support. As these neurons mature, they send out processes which reach their targets by birth. By this time, GDNF and TGF beta 2 production has ceased in the midbrain, but has reappeared in the appropriate target organ(s) that each neuron should connect with. This suggests there are (temporarily?) subpopulations of dopamine neurons that show a time-dependency for select neurotrophins. Those requiring GDNF for survival will only receive it in the corpus striatum, and any GDNF-dependent axons which accidentally migrate to the TGF beta 2 expressing amygdaloid complex will die, due to a lack of molecule-specific trophic support. Thus, "mis-wiring" is avoided, and the proper connections are insured through selective survival.14
The activity of GDNF as a survival factor for dopaminergic neurons suggests the potential for use of GDNF in the treatment of Parkinson's disease. The action of GDNF as a neurotrophic factor for motoneurons suggests the potential usefulness of this factor in treating diseases affecting motor neurons (e.g., amyotrophic lateral sclerosis).