Bioinformatics and Immunohistochemical Analyses Support Preserved Expression of Glial Cell Line-Derived Neurotrophic Factor Receptor RET in Parkinson's.

Intracerebral administration of glial cell line–derived neurotrophic factor (GDNF) as a disease-modifying therapy has been trialed in Parkinson's disease (PD)1, 2; however, the potent effects seen in animal models have not been reproduced clinically. One reason proposed for this lack of efficacy in humans is downregulation of the GDNF receptor RET by α-synuclein (αSyn) in the PD substantia nigra (SN),3 which is supported by some,4, 5 but not all,6 preclinical studies. The recent description of off-state dyskinesia correlating with GDNF administration in a clinical trial participant7 may be evidence of functional pharmacological effects of GDNF, even in advanced PD. In this letter, we consider these findings in light of gene expression analysis using open-source transcriptome data and immunohistochemistry of RET expression in the SN of patients with PD. We used the R package limma to identify differentially expressed genes (DEGs) with log2FC (fold change) of ±0.4 and false discovery rate (FDR) adjusted P ≤ 0.05, between PD and control SN, using a combined dataset comprising data from five microarray studies: GSE8397, GSE7621, GSE49036, GSE2029, and GSE20186 (see Supporting Information Methods in Data S1). Of the 142 samples, 79 were PD cases and 63 were controls. The most significant DEG was the dopamine transporter SLC18A2 (VMAT2), with a logFC value of −2.34 in PD SN samples compared with controls (FDR = 9.57E−10). Notably, throughout the DEG list there was significant downregulation of dopaminergic markers in the PD group, including downregulation of RET (logFC = −1.62, FDR = 1.08E−07) (Fig. 1A,B). To investigate whether this downregulation of RET expression in PD was due to, or independent of, the loss of dopamine neurons, we examined the coexpression of RET and seven dopaminergic markers: SLC18A2, TH, DDC, SLC6A3, KCNJ6, ALDH1A1, and NR4A2. For the majority of comparisons, there was significant moderate to strong correlation between RET and a given marker in controls, and this correlation was maintained in PD (Fig. 1C,D). This suggests that the downregulation of RET observed in the PD SN is likely due to dopaminergic neuronal loss, rather than downregulation of RET transcript expression. To extend these findings, we performed differential expression analysis using data generated from laser-captured microdissection (LCM) of individual SN neurons in postmortem brains of 10 PD and 8 control cases (GSE20141). In agreement with the coexpression data, RET was not significantly downregulated in the LCM dataset (Fig. 1E). Our immunohistochemical analysis found that RET expression was reduced in SN neurons in PD compared with controls, but that RET continued to be expressed even in late-stage PD (Fig. 1F,G). These data are in agreement with a recent study3 and suggest that the receptor needed for GDNF to be effective is intact in the PD brain, even at late-stage disease. The report by Lloyd et al7 suggested that intracerebral GDNF is capable of inducing sprouting, even in cells carrying αSyn. To test this, we transduced primary cultures of embryonic day 14 rat ventral mesencephalon with AAV-GFP or AAV-αSyn and found that, although αSyn significantly reduced neurite growth after 10 days, GDNF treatment led to a significant increase in neurite growth (Fig. 1H). Our study shows sustained, albeit reduced, expression of RET in the PD SN and is consistent with the report Lloyd et al,7 whose results suggest that GDNF may induce sprouting of dopaminergic terminals throughout the putamen. Our finding that RET expression is retained in the PD SN, albeit at a lower level than in controls, provides further support for the future of GDNF therapy. Nevertheless, early administration of GDNF, when extensive loss of RET-expressing dopaminergic neurons has not yet occurred, may lead to better outcomes in patients with PD. We are grateful to the Parkinson's UK Brain Bank at Imperial College London for the brain tissue samples and associated clinical and neuropathological data. Open access funding provided by IReL. This study was designed by A.M.S., G.W.O., and L.M.C. C.G.D. completed the bioinformatics analysis. F.W. and M.M. performed the immunohistochemical work. S.R.G. performed the in vitro studies. A.M.S. obtained ethical approval. G.W.O., L.M.C., and A.M.S. supervised the work. G.W.O. and A.M.S. wrote the manuscript. The final manuscript was approved by all authors. A.M.S. and G.W.O. were supported by Cure Parkinson's Project Grant (Grant CP:GO01). G.W.O. was supported by Science Foundation Ireland (Grant 19/FFP/6666). A.M.S., L.M.C., and G.W.O. were supported by a Daniel and Margaret Cronin Advancing Access Scholarship. G.W.O. and A.M.S. were supported by the Marie Skłodowska-Curie Fellowship programme (Grant MSCA-IF-2019 890290). The data that support the findings of this study are available from the corresponding authors upon reasonable request. Data S1 Supporting Information Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.