Neurotrophic factors GDNF and NRTN : from basic properties to clinical trials

Parkinson s disease (PD) is a neurodegenerative disorder affecting seven to ten million people worldwide. The average age of diagnosis is 60, but some forms can affect even young adults. In the US alone the direct and indirect expenses for PD exceed $25 billion each year. PD is best characterized by the death of dopaminergic neurons of the substantia nigra pars compacta, which causes symptoms ranging from rigidity to postural instability. As the disease progresses, other areas of the brain become affected, generating psychiatric and cognitive dysfunctions. Current therapies effectively reduce motor symptoms of PD, but do not stop its progression. Neurotrophic factors regulate neuronal growth, differentiation, and survival, and several of them have been shown to protect and regenerate dopaminergic neurons in animal models of PD. The glial cell line-derived neurotrophic factor (GDNF) and neurturin (NRTN) have reached clinical trials, but they did not replicate the promising results of the preclinical studies. Several reasons, including stability of recombinant proteins and their diffusion in the brain tissue, could explain the results of the clinical trials. Stability might have been a problem especially with GDNF, which has been delivered mostly as bacterially-produced recombinant protein. In this work we analysed GDNF produced in mammalian cells and compared it to bacterially-produced GDNF. E. coli produced-GDNF is less stable than mammalian GDNF. This difference is probably due to the purification/renaturation method used with the E. coli-produced factor. Processing and stability of GDNF are affected also by cell line and medium used for its production. In mammalian cells glycosylation of GDNF is fundamental for its processing into the mature molecule. The diffusion problem affects both GDNF and NRTN, which do not diffuse far enough from the infusion site because of their heparin-binding properties. Heparin and the closely related heparan sulphates are abundant in the extracellular matrix and on the cell surface, and hinder the diffusion of GDNF and NRTN. The diffusion issue might not be a significant problem in the animal experiments, but might limit the results achieved with humans, who have significantly bigger brain size compared to rats and monkeys. In this work we have developed NRTN mutant variants with lower affinity for heparin and characterized their activity in vitro and in a unilateral 6-OHDA rat model of PD. All NRTN variants were biologically active. Especially the variant N4 showed better diffusion and rescued a higher number of dopaminergic fibres than E. coli-produced GDNF. Toxin-treated rats administered with N4 also showed functional recovery in behavioural assays. However, as a caveat the mutations introduced could have drawbacks influencing NRTN recycling/degradation and signalling. In this respect lack of heparin-binding could affect NRTN accumulation on the cell surface and inside the cells, therefore causing a slower initiation of the signal. Taken together our results help understanding basic features of GDNF and NRTN, such as the roles of glycosylation and of heparin binding. They also point out several important features that have to be taken into account when producing and/or modifying growth factors for clinical use, and underlines that mammalian molecules with reduced heparin binding could be beneficial for treating PD patients.