Erica Forsberg (Presenter)
The Scripps Research Institute
Bio: Erica Forsberg is a Post-Doctoral Research Associate at The Scripps Research Institute in the Center for Metabolomics and Mass Spectrometry. She has fostered a passion for working with analytical instrumentation, in particular mass spectrometry, since her undergraduate degree at McMaster University in Hamilton, ON, Canada. This drive allowed her to pursue a direct entry Ph.D. at McMaster where she developed high-throughput drug discovery assays for complex mixtures including natural product extracts. An interest in small molecule discovery and information intense data analysis brought her to the field of metabolomics and her current position in Gary Siuzdak’s lab. She is currently studying immunomodulator function in Parkinson’s Disease and cancer models as well as developing novel nanostructured imaging mass spectrometry platforms.
Authorship: Erica M. Forsberg (1), Katherine E. Olson (2), Charles R. Schutt (2), R. Lee Mosley (2), Tao Huan (1), Mingliang Fang (1), Howard E. Gendelman (2), Gary Siuzdak (1)
(1) The Scripps Research Institute, La Jolla, CA, 92130; (2) University of Nebraska Medical Center, Omaha, NE, 68198
| Short Abstract Parkinson’s Disease (PD) is a neurodegenerative disorder involving progressive loss of nigrostriatial neurons and altered effector T cell functions. Research in animal models demonstrated neurodestructive effector T cells can be pharmacologically transformed to regulatory neuroprotective T cells. To translate animal models to human we used sargramostim for PD treatment to assess immune modulatory functions. A phase I double blind placebo controlled study demonstrated this transformation with improvements in cortical neurophysiological activities. This incited mechanistic questions on whether a metabolomics disease signature exists. Our works demonstrated that global metabolomics isolated tryptophan and vitamin D metabolism as key treatment pathways. Targeted metabolomics was used to quantify dysregulated metabolites, providing insights into potential neuroprotective actions. |
Long Abstract
Introduction
Parkinson’s Disease (PD) is a debilitating neurodegenerative disorder that effects millions worldwide. Symptoms are linked to the loss of substantia nigra pars compacta neurons and reduced striatal dopamine. Release of aggregated and nitrosylated α-synuclein from dead or dying neurons released into the extraneuronal microenvironment activates microglial and affects the numbers of circulating effector T cell (Teff) cells (1). The cause of PD along with therapeutic disease options are less than well established. A newly formed model for disease revolves around immune transformation of Teff to regulatory T cells (Treg). A number of drugs have shown their abilities to perform this function in animal models of inflammatory and degenerative diseases. One is sargramostim (granulocyte macrophage colony-stimulating factor, Sanofi-Genzyme US) which was initially developed for repopulation of myeloid cell populations after chemotherapy and organ transplantation (2). It was recently used in a phase I clinical trial for safety and efficacy in PD patients. Over an 8 week observational period it showed potential neurological benefit (3). Here we probe the mechanism of sargramostim induced neuroprotection through global and targeted metabolomics. Assays were performed on patient serum samples before, during and after the course of a treatment. Global metabolomics is a method of analyzing the entire spectrum of metabolites present in different sample classes in an unbiased manner. Using LC-TOF-MS and the XCMS Online (xcmsonline.scripps.edu) bioinformatics platform, chromatographic features between sample classes are compared for statistically significant differences. Accurate masses of these features are then compared to known masses in the METLIN metabolite database. Using the mummichog pathway analysis tool, matched metabolites are overlaid with known metabolic pathways. Significant pathways are then probed in targeted metabolomics methods using LC-MRM-MS to confirm dysregulation.
Methods
Serum samples were obtained from 22 PD patients treated with sargramostim or placebo during visits before (two visits), during (four and six weeks during), and after treatment cessation (one month after drug stoppage). Sera metabolites were extracted using 50:50 methanol:acetonitrile. Global metabolomics analysis was performed on central carbon metabolites using HILIC-LC-QTOF-MS in negative mode and lipophilic metabolites using RP-UPLC-QTOF-MS in positive mode using a Bruker Impact II QTOF. Pooled samples were run in autonomous (data dependent) MS/MS every ten samples for initial confirmation of analyte identification. Blanks were also run every ten samples to ensure no carry-over was occurring. Resulting chromatograms were uploaded to XCMS Online for peak picking, retention time alignment and statistical analysis. To fully analyze global metabolic features and focus on sargramostim effects on the patients, pairwise jobs were performed using pretreatment versus on-treatment, pretreatment versus treatment cessation for both sargramostim and placebo and placebo versus sargramostim. A meta-analysis between placebo and pretreatment versus on-treatment was also performed. These analyses were performed for both HILIC and RP datasets. Metabolic pathway analysis was performed on statistically significant features (p-value < 0.01, fold change > 1.5) for each pairwise job. The identified pathways unique to pretreatment versus on-treatment for sargramostim treated patients were used for targeted metabolomics studies. A total of 18 tryptophan pathway metabolites were analyzed on all extracted serum samples using a T3 C18 column coupled to an Agilent 6495 triple quadrupole MS. For 21 vitamin D and cortisol metabolites a Poroshell 120 EC-C18 column was used. Quantitative analysis was done using Agilent MassHunter Quantitative Analysis vB.07.00. Statistical analysis of metabolite quantities between sample classes was performed on Prism GraphPad v7.
Results
Global metabolomics revealed significant differences between sample classes with sargramostim had significantly more dysregulated features than with placebo. From the HILIC analysis on the placebo group comparing pretreatment versus on-treatment, 337 dysregulated features were detected, while the same comparison in the sargramostim group identified 600 dysregulated features. This indicates that some placebo affects and/or normal baseline changes are occurring, yet there is an 80% increase in metabolic changes observed in sargramostim compared to placebo. After performing meta-analysis on these datasets (p-value < 0.05, fold change >2), only 63 features were overlapped between the sargramostim and placebo groups, while 601 features were unique to sargramostim treated patients and 483 features unique to the placebo group. This overlap serves as the baseline metabolic changes, while the sargramostim features were used as a control set of features to confirm the results of the pathway analysis tool thereby ensuring changes were specific to the drug and not baseline or placebo. The metabolic pathway analysis of sargramostim group comparing pretreatment versus on-treatment resulted in three pathways with p-value < 0.01: tRNA charging (p-value = 0.0074), tryptophan degradation (p-value = 0.0068) and tyrosine biosynthesis (p-value = 0.0039). Although tryptophan degradation was shown to be undergoing change in the placebo group, the changes were not as significant (p-value = 0.0204). Connections with tryptophan metabolism and neurodegeneration in PD have been previously reported (4) and therefore a targeted method was developed to quantify variations in tryptophan metabolites in the serum samples. Of the 18 metabolites analyzed, three metabolites underwent significant alteration specifically between the pretreatment/treatment stoppage versus on-treatment and placebo versus sargramostim sample classes. In the sargramostim group L-kynurenine showed a 2.3- and 3.0-fold increase compared to pretreatment or placebo, quinolinic acid showed a 2.4-fold increase in both comparisons, while serotonin decreased 2.5- and 2.2-fold compared to pretreatment or placebo. Analysis of the lipophilic compounds from the RP data was performed in a similar manner to the HILIC analysis. In the sargramostim group comparing pretreatment versus on-treatment, there were 1788 dysregulated features. After performing meta-analysis and pathway analysis, there were two major pathways identified as significantly dysregulated and specific to the sargramostim treatment: vitamin D3 metabolism (p-value = 0.0000) and bile acid biosynthesis (p-value = 0.0008). There have also been previous reports of vitamin D involvement in PD (5) and as such a targeted method was developed to analyze a selection of 21 vitamin D and associated cortisol metabolites.
Conclusion
By performing global metabolomics, a link between tryptophan metabolism and treatment of PD patients with sargramostim has been established. Analysis of the pathway by targeted metabolomics showed an increase in L-kynurenine and quinolinic acid along with a decrease in serotonin. Changes in these particular metabolites indicate an association with tryptophan-2,3-dioxygenase (TDO) or indoleamine-2,3-dioxygenase activity (IDO). Since the neurodegeneration that occurs in PD can illicit an immune response associated with Teff dysfunction and further downstream effects on tryptophan metabolism, we posit that the tryptophan pathway is being modulated by upstream events brought on by sargramostim acting on the immune response. As sargramostim is a known immunomodulator, there may be a compensating effect of the drug to overcome the dysregulation of TDO/IDO levels by increasing Treg thereby controlling the Teff response. Further insights into the mechanism of sargramostim mode of action will be elucidated from the targeted analysis of vitamin D and cortisol metabolites. There is already a known connection between increased TDO and cortisol (4) that may provide a link between these pathways and immune response. Although more research must be done to confirm the mechanisms, global and targeted metabolomics have already provided a greater understanding of the improved neurophysiological activity in PD patients treated with sargramostim.
References & Acknowledgements:
(1) Benner EJ, Banerjee R, Reynolds AD, et al. Nitrated alpha-synuclein immunity accelerates degeneration of nigral dopaminergic neurons. PLoS One 2008;3:e1376
(2) Olanow CW, Stern MB, Sethi K. The scientific and clinical basis for the treatment of Parkinson disease. Neurology 2009;72 (Suppl 4):S1-S136
(3) Gendlman HE, Zhang Y, Santamaria P, et al. Sargramostim Safety and Immune Profiling in Parkinson’s Disease. Movement Disorders submitted August 2016.
(4) Anderson G, Maes M. CNS & Neurological Disorders – Drug Targets 2014; 13:137-149.
(5) Newmark HL, Newmark J. Vitamin D and Parkinson’s disease – A hypothesis. Movement Disorders 2007; 22(4):461-468
Acknowledgments
This study was funded by Sanofi US; by community support received through the Carol Swarts Neuroscience Research Laboratory, the Frances and Louis Blumkin Foundation, and the Nebraska Neuroscience Alliance Endowed Funds; by the Vice-Chancellor’s Office of the UNMC for Core Facility Developments, and by DoD grant W81XWH11-1-0700 and NIH grants R01-NS034139 and R01-NS070190.
| Description | Y/N | Source |
| Grants | no | |
| Salary | no | |
| Board Member | no | |
| Stock | no | |
| Expenses | no |
IP Royalty: no
| Planning to mention or discuss specific products or technology of the company(ies) listed above: | no |