Correlation of Plasma EGF with Striatal Dopamine Transporter Availability in Healthy Subjects

We aimed to evaluate the association between plasma epidermal growth factor (EGF) and the availability of dopamine transporter (DAT) measured from 123I-FP-CIT single-photon emission computed tomography in healthy controls in this study. Volume of interest template was applied to measure specific binding ratios (SBRs) of caudate nucleus, putamen, and striatum representing DAT availability as follows; SBR = (target– cerebellum)/cerebellum. Plasma EGF was negatively correlated with the availabilities of both caudate nucleus (r = −0.261, p = 0.019), and putamen (r = −0.341, p = 0.002). After dividing subjects according to Apo E genotyping, DAT availability of caudate nucleus of Apo e4 non-carriers (n = 60) showed the positive correlation with cerebrospinal fluid (CSF) α-synuclein (r = 0.264, p = 0.042). Plasma EGF was negatively correlated with DAT availabilities of Apo e4 non-carriers. Further studies are needed to clarify underlying mechanisms of this phenomenon.

might interfere with DAT SPECT scans, anticoagulants that might preclude safe completion of the lumbar puncture, or investigational drugs, and a condition that precludes the safe performance of routine lumbar puncture were excluded. Subjects without CSF biomarkers, plasma EGF, and Apo E genotyping were excluded. Medical history, cerebrospinal fluid (CSF), plasma EGF, and 123 I-FP-CIT SPECT scans were downloaded. The PPMI study was approved by the local Institutional Review Boards of all participating sites (Institute for Neurodegenerative Disorders, University of Pennsylvania, University of California, Los Angeles, Coriell Institute for Medical Research, Clinical Trials Coordination Center, Laboratory of Neurogenetics; National Institute on Aging NIH, Institute for Neurodegenerative Disorders, Clinical Trials Statistical and Data Management Center, University of Iowa) and written informed consent was obtained from each subject at the time of enrollment for imaging data and clinical questionnaires. All methods were performed in accordance with the relevant guidelines and regulations.
Cerebrospinal fluid biomarkers and Plasma EGF. CSF biomarkers of Aβ 1-42 , tau protein phosphorylated at the threonine 181 position (p-Tau 181 ), and total tau were analyzed with multiplex xMAP Luminex platform (Luminex Corp, Austin, TX, USA), and Innogenetics immunoassay kits (Innogenetics/Fujirebio, Ghent, Belgium). α-synuclein were analyzed using an enzyme-linked immunosorbent assay (Covance Research Products Inc., Denver, PA, USA). Plasma levels of EGF were measured by enzyme-linked immunosorbent assay (ELISA, R&D Systems, Minneapolis, MN, USA) according to manufacturer instructions. Samples were run in duplicate and data used for this study met quality control measures for technical performance.
Apo E Genotyping. Apo E genotyping was performed on DNA samples. Two non-synonymous single nucleotide polymorphisms, rs429358, and rs7412 were genotyped in each sample to distinguish between Apo e2, e3, and e4 alleles using TaqMan assays (Applied Biosystems, Foster City, CA, USA).

123
I-FP-CIT SPECT. Protocol. 123 I-FP-CIT SPECT was performed during the screening visit for all subjects.
SPECT scans were acquired 4 ± 0.5 hrs after injection of 111-185 MBq of 123 I-FP-CIT. Subjects were pretreated with iodine solution or perchlorate prior to injection to block thyroid uptake. Raw data were acquired into a 128 × 128 matrix stepping each 3 or 4 degrees for the total projections. Raw projection data were reconstructed using iterative ordered subset expectation maximization with HERMES (Hermes Medical Solutions, Stockholm, Sweden). The reconstructed images were transferred to pmod (PMOD Technologies LLC, Zürich, Switzerland) for subsequent processing including attenuation correction.
Image analysis. Downloaded scans were loaded using pmod v3.6 (PMOD Technologies LLC, Zürich, Switzerland) with 123 I-FP-CIT template 11 . Specific binding of 123 I-FP-CIT regarding DAT was calculated using a region of interest analysis. A standard set of volume of interest (VOI) defining caudate nucleus, putamen, and striatum (caudate nucleus + putamen) based on the Automated Anatomical Labeling (AAL) atlas 12 . The cerebellum was chosen as a reference region. VOI template was applied to measure specific binding ratios (SBRs) of caudate nucleus, putamen, and striatum representing DAT availability as follows; SBR = (target-cerebellum)/cerebellum. Statistical Analysis. Normality was examined using D' Agostino-Pearson omnibus test. Spearman correlation was used to measure the relationship of SBRs with CSF biomarkers, and plasma EGF. Mann-Whitney test was applied to compare SBRs, CSF biomarkers, and plasma EGF between Apo e4 non-carriers and carriers. Statistical analyses were performed using GraphPad Prism 7 for Mac OS X (GraphPad Software Inc, San Diego, CA, USA).

Results
Subjects' characteristics. 81 healthy subjects (49 male, 32 female) were included in this study. Mean age was 62.3 years. Mean BMI was 26.5 kg/m 2 . Twenty-one subjects were Apo e4 carriers (25.9%). DAT availabilities in caudate nucleus (r = −0.313, p = 0.004) showed a reduction with aging as expected. When subjects were divided according to Apo e4 genotyping, Aβ 1-42 was higher in Apo e4 non-carriers than Apo e4 carriers. However, age, sex, BMI, α-synuclein, p-Tau 181 , total tau, and plasma EGF showed no significant differences between Apo e4 non-carriers and Apo e4 carriers. Subjects' characteristics are summarized in Table 1.
Data availability. Data used in the preparation of this article were obtained from PPMI database (www. ppmi-info.org/data).

Discussion
In this study, plasma EGF was negatively correlated with striatal DAT availability. When subjects were categorized according to Apo E genotyping, negative correlation was observed in Apo e4 non-carriers. α-synuclein was positively correlated with DAT availability in Apo e4 non-carriers.  Table 2. Correlations of the availability of DAT with age, BMI, cerebrospinal fluid biomarkers, and plasma EGF. * DAT, dopamine transporter; BMI, body mass index; EGF, epidermal growth factor. * P < 0.05 * , P < 0.1 # .

Figure 2. Correlation between DAT availability and CSF α-synuclein in subjects with Apo e4 non-carriers.
EGF is known to have protective effect in dopaminergic neuron 13 and stimulate the uptake of dopamine 14 . Chen-Plotkin et al. reported that when the PD patients were divided by quartile according to plasma EGF values, the lowest quartile of the PD patients showed the highest conversion rate to Parkinson disease dementia and they demonstrated that plasma EGF was an independent variable predicting cognitive decline in PD patients 4 . Lim et al. also reported that low baseline plasma EGF predicted cognitive decline in PD patients and conversion from amnestic mild cognitive impairment (MCI) to AD 5 . Jiang et al. demonstrated that plasma EGF was decreased in the early stage of PD and there was no significant difference of plasma EGF between advanced PD and normal control 15 . In study regarding EGF and dopamine uptake in cultured rat astrocytes, they suggested the existence of Na + -dependent and Na + -independent dopamine uptake in cultured rat astrocytes and they concluded that EGF might stimulate the expression and translocation of the extraneuronal DAT 16 . In study assessing EGF-ErbB1 action on developing midbrain dopaminergic neuron, authors showed that EGF elevated DAT level in mesencephalic cultures and they reported that EGF receptor inhibitor reduced DAT level in the striatum, nucleus accumbens, and globus pallidum in neonatal rats 17 . EGF levels in both striatum and serum of patients with chronic schizophrenia were reduced comparing with those of normal controls in postmortem study 18 . Laakso et al. showed that striatal DAT availability measured by 18 F-CFT was reduced in patients with chronic schizophrenia as compared with normal controls 19 . Thus, we expected that the DAT availability would be decreased if the level of plasma EGF was decreased. Unlike our expectations, negative correlation between plasma EGF and DAT availability was observed in healthy subjects. There was no relevant study to explain this negative correlation.
α-synuclein, 14 kDa protein, consists of three domains with N-terminal lipid-binding α-helix, amyloid-binding central domain, and C-terminal acidic tail 20,21 , which has a function in suppression of apoptosis, regulation of glucose levels, modulation of calmodulin activity, chaperone activity, and regulation of dopamine biosynthesis 20 . Also, it has been known to be associated with neurodegenerative disease such as PD, dementia with Lewy bodies, MSA, and AD 21 . Especially in PD, α-synuclein interacts with tubulin, parkin, dopamine receptor, synphilin-1, phospholipase, and small ubiquitin related modifiers 20 . Previous studies showed that α-synuclein was not different between Apo e4 carriers and Apo e4 non-carriers in normal subjects, MCI, and AD, consistent with this study 22,23 . However, molecular linkage between Apo E and α-synuclein was demonstrated in one study using A30P and A53T transgenic mice. α-synuclein induces neuronal degeneration leading to Apo E deposition in spinal cords, astrocytes, and activated microglia of transgenic mice 24 . Astrocyte-secreted Apo E reduced α-synuclein uptake, and the effect was seen in Apo e4, followed by e3, and e2 25 . Wersinger et al. reported that α-synuclein reduced DAT activity by recruitment of DAT from plasma membrane to cytoplasm in their in vitro study 26 . The decreased activity of DAT was caused by reduced dopamine uptake velocity, not by decreased DAT expression 27 . Kovacs et al. showed that the density of DAT, identified by immunochemistry, inversely correlated with the density of α-synuclein in the substantia nigra of patients with Lewy body disease and PD 28 . However, Bellucci et al. demonstrated that transgenic mice producing human α-synuclein had increased levels of striatal DAT 29 .
Apo E has three isoforms which are known to Apo e2, Apo e3, and Apo e4 30 . Among them, Apo e4 is known to be the greatest risk factor for AD, followed by Apo e3, contrary to protective effect of Apo e2 31 . Apo E has a major role in metabolism of Aβ, which is abundant in brain of Apo e4 carriers than Apo e4 non-carriers 31 . In this study, CSF Aβ 1-42 was higher in Apo e4 non-carriers than Apo e4 carriers. Consistent with this study, Prince et al. showed decreased amount of CSF Aβ in Apo e4 carriers in both AD and normal control groups 32 . α-synuclein attenuated the effect of EGF by showing decreased luciferase activity in a study by Iwata et al. 33 . Therefore, Apo e4 non-carriers may have higher CSF Aβ 1-42 , CSF α-synuclein, and CSF α-synuclein leading to both lower plasma EGF and higher DAT availability.
This is the first study that investigated the association between Apo E genotyping, Aβ 1-42 , α-synuclein, plasma EGF, and DAT availability in healthy controls. However there are several limitations in this study. First, subjects included in this study was collected from PPMI database. Second, the number of subjects was small in Apo e4 carriers. Third, as PPMI database was collected from multiple institutes, the difference in image acquisition may affect the results.
In conclusion, plasma EGF was negatively correlated with DAT availabilities of Apo e4 non-carriers. DAT availability was positively correlated with α-synuclein in Apo e4 non-carriers. Further studies are needed to clarify underlying mechanisms of this phenomenon.  Table 3. Correlations of the availability of DAT with age, BMI, cerebrospinal fluid biomarkers, and plasma EGF according to Apo E genotyping. *DAT, dopamine transporter; BMI, body mass index; EGF, epidermal growth factor. *p < 0.05*, p < 0.1#.