Abstract
Background:
Alpha-synuclein containing inclusions in neurons, the characteristic pathological lesions of Parkinson’s disease (PD), are not limited to the central nervous system, but also affect the enteric nervous system (ENS). This suggests that the ENS offer some potential as a surrogate of central nervous system pathology and that it may represent an original source of biomarkers for PD. However, the usefulness of α-synuclein detection in gastrointestinal biopsies as a biomarker for PD is still unclear, as the different immunohistochemical methods employed to date have led to conflicting results.
Objective:
Our aim is to propose an optimized immunoblotting method for the detection of endogenous α-synuclein in the healthy ENS that may be used to supplement the immunohistochemical analysis.
Methods:
Primary culture of rat ENS and homogenates of human small intestine were analyzed by Western Blot using seven different α-synuclein and phospho-α-synuclein antibodies along with two methods that increase α-synuclein retention on blot membranes, namely incubation of the membranes with paraformaldehyde (PFA) or treatment of samples with the crosslinker dithiobis[succinimidylpropionate] (DSP).
Results:
A moderate improvement in the detection of endogenous enteric α-synuclein was observed following membrane fixation with PFA for only two of the seven antibodies we tested. Immunodetection of total and phosphorylated α-synuclein in the ENS was markedly improved when samples were treated with DSP, regardless of the antibody used.
Conclusions:
Our results demonstrate that the detection of α-synuclein in the gut by Western Blot can be optimized by using methods for enhanced membrane retention of the protein along with the appropriate antibody. Such an optimized protocol opens the way to the development of novel biomarkers for PD that will enable a quantification of α-synuclein in gastrointestinal biopsies.
INTRODUCTION
α-Synuclein is a small 140-amino acid protein that has special relevance for the neuropathological assessment of Parkinson’s disease (PD). In the late 1990s, Lewy bodies and neurites, the characteristic pathological lesions of PD, were discovered to have a strong α-synuclein and phosphorylated α-synuclein immunoreactivity [1]. Since then, α-synuclein and phosphorylated α-synuclein immunostaining has become the method of choice for the detection of PD pathology [2, 3] leading to a reassessment of the distribution of Lewy pathology in the central nervous system and in peripheral autonomic systems by several different groups [4, 5]. Using this approach, it has been demonstrated that the vast majority of PD patients exhibit α-synuclein or phospho-α-synuclein immunoreactive lesions within their enteric nervous system (ENS) [4], an autonomic neuronal network that is readily and repeatedly accessible to routine gastrointestinal biopsies [6]. These specific features suggest that the ENS offers some potential as a surrogate of brain pathology and that it might represent an original source of biomarkers for PD[7, 8].
We have shown that the analysis of a single microdissected gastrointestinal biopsy by immunohistochemistry enables the detection of Lewy pathology in nearly ¾ of PD patients [6]. Although the microdissection technique we have developed provides outstanding information on the morphology of enteric neurons, it nonetheless has several limitations as it needs to be performed immediately after the endoscopic procedure and requires a skilled technician [6]. A subsequent study that used deparaffinized biopsy samples without microdissection demonstrated the presence of α-synuclein aggregates in the sigmoid colon of PD patients with a high sensitivity and specificity [9] but other groups did not confirm these results [10]. Furthermore, in all histological techniques employed to date, quantification of α-synuclein deposition and subsequent determination of a positive signal remains somewhat ambiguous [9, 11–14]. In this context, it is worth noting that none of the available studies on the expression and distribution of α-synuclein in the gut have been performed using standard Western blotting [12, 16], despite the generally higher sensitivity of the latter technique compared to immunohistochemistry. Two recent papers elegantly demonstrated that the detection of α-synuclein in neuronal cell lines and brain samples by Western blotting was more delicate and difficult than previously thought as the protein is easily washed off the membrane when a routine blotting procedure is used [17, 18]. They have therefore developed two simple methods to increase Western blot signals of α-synuclein by improving its retention on blot membranes using either incubation of the membranes with paraformaldehyde (PFA) [17] or treatment of the samples with the crosslinker dithiobis[succinimidylpropionate] (DSP) [18]. In order to optimize the detection of endogenous α-synuclein by immunoblotting in the ENS, we evaluated these two approaches in gut samples using several routinely used α-synuclein and phosphorylated-α-synucleinantibodies.
MATERIAL AND METHODS
Recombinant α-synuclein, primary culture of ENS and human tissues
Recombinant α-synuclein was from Sigma (Sigma, Saint Quentin Fallavier, France). Primary cultures of ENS was generated and cultured as previously described [19]. Human brain samples were obtained from a coronal frozen section of a human brain devoid of neurodegeneration. A specimen of human small intestine (ileum) was obtained from one neurologically unimpaired subject who underwent intestinal resection for volvulus. Tissues samples used for the current study were from macroscopically normal segments of uninvolved resection margins. The study was approved by the Fédération des biothèques of the University Hospital of Nantes, according to the guidelines of the French Ethics Committee for Research on Human, and registered under the no. DC-2008-402.
Sample preparation from cultured cells and tissues
At 13 days, primary culture of rat ENS was harvested in NETF buffer (100 mM NaCl, 5 mM EDTA, 50 mM Tris-Cl, pH 7.4, and 50 mM NaF) containing 1% (v/v) Igepal CA-630 (Sigma-Aldrich, Saint-Quentin Fallavier, France), phosphatase inhibitor cocktail II (Roche, Neuilly sur Seine, France) and protease inhibitors cocktail (Roche). Tissues were lysed NETF/Igepal buffer, homogenized using the “Precellys 24” tissue homogenizer (Bertin technologies, Saint Quentin-en-Yvelines, France), sonicated and centrifugated at 16000 g for 20 minutes at 4°C.
Electrophoresis and immunoblotting
Total protein concentrations were determined by Bradford assay (Pierce, Thermo Scientific, Illkirch, France) using Varioskan™ (Thermo Scientific, Waltham, Massachusetts, United States). The crosslinker dithiobis[succinimidylpropionate] (DSP) (Sigma-Aldrich) was dissolved in DMSO at 25X final concentration, added to the sample to obtain the desired final concentration and incubated for 30 min at 37°C with rotation [18]. Controls were treated with an equivalent volume of DMSO. Samples were further prepared for electrophoresis by diluting with NuPAGE sample buffer with or without 50 mM dithiothreitol (DTT), then heated at 70°C for 10 min. Lysates were separated using the NuPAGE 4–12% Bis-Tris gels MiniGels™ (Life technologies, Cergy-Pontoise, France) together with the 2-(N-morpholino)ethanesulfonic acid/sodium dodecyl sulfate (MES-SDS) running buffer before electrophoretic transfer to PVDF membranes with the iBlot™ Dry Blotting System also from Invitrogen. Post-transfer membranes were fixed either with 10% (v/v) acetic acid for 10 min or 0.4% PFA (v/v) for 30 min. Membranes were then blocked for one hour at room temperature in Tris-buffered saline (TBS) (100 mM NaCl, 10 mM Tris, pH 7.5) with 0.1% (v/v) Tween-20 and 5% (w/v) non fat dry milk and incubated overnight at 4°C with the primary antibodies. Details on the primary antibodies against α-synuclein are summarized in Table 1; mouse monoclonal anti-β-actin antibody was from Sigma and used at 1:5000. Bound antibodies were detected with horseradish peroxidase-conjugated anti-rabbit or anti-mouse antibodies (Life technologies, diluted 1:5000) and visualized by enhanced chemiluminescent detection (Clarity ECL, Biorad, Marnes-la-Coquette, France). When necessary, membranes were incubated in stripping buffer containing 500 mM Tris-HCl with 0.1M β-mercapto-ethanol and 2% (v/v) sodium dodecyl sulfate for 20 min at 60°C, followed by extensive washing in TBS before reblocking for 30 min in TBS with 5% non-fat dry milk and reprobing. The relevant immunoreactive bands were captured using ChemiDoc MP and analyzed with Image Lab (Biorad). The value of α-synuclein or phosphorylated α-synuclein was normalized to the amount of β-actin, for comparison between samples and expressed as a fold change relative to controlcondition.
Statistics
All data are given as the mean ± standard error of the mean (SEM). For experiments with DSP, data were analyzed using non-parametric Kruskal Wallis test with Dunn’s post test. Differences were considered statistically significant if p < 0.05.
PFA FACILITATES THE DETECTION OF α-SYNUCLEIN IN THE ENS WHEN (C20)-R AND SYN-1 ANTIBODIES ARE USED
Previously, Lee and Kamitani [17] reported that endogenous α-synuclein from various cell lines can be efficiently detected by immunoblotting when blotted membranes are fixed with 0.4% PFA. We therefore compared Western blotting with a conventional method of membrane fixation (10% acetic acid) [20, 21] with PFA fixation to detect endogenous α-synuclein in rat primary culture of ENS, samples of human small intestine and cerebral cortex using four different antibodies to total α-synuclein as well as three different phospho-α-synuclein antibodies (Table 1).
All four α-synuclein antibodies detected endogenous monomeric 15-kDa α-synuclein from human brain when blotting membranes were fixed with acetic acid (Fig. 1A–D), a signal that was further enhanced when membranes were PFA-fixed (Fig. 1A–D). Endogenous α-synuclein in primary culture of rat ENS and human small intestine was barely or not detected after fixation of membranes with acetic acid, regardless of the antibody used (Fig. 1A–D). Membrane fixation with PFA enabled to detect endogenous monomeric α-synuclein in these two intestinal tissues when (C20)-R and Syn-1, but not LB509 and Syn211 antibodies were used (Fig. 1A–D). Densitometry of these immunoblots showed that the α-synuclein signal intensity obtained with (C20)-R and Syn-1 in human small intestine was respectively ≈1.5 and ≈2.75 higher when membranes were PFA- rather than acetic acid-fixed. Brain and gut monomeric α-synuclein was not detected with any of the three phospho-α-synuclein antibodies tested, even when membranes were PFA-fixed (Fig. 1E, F and data not shown for EPI1536Y).
DSP ALLOWS OPTIMAL DETECTION OF TOTAL AND PHOSPHORYLATED α-SYNUCLEIN IN THE ENS
Newman et al. recently showed that treatment of cell lysates and human brain homogenates with DSP followed by cleavage with a reducing agent enhanced α-synuclein immunoreactivity [18]. In a first set of experiments, we applied this method to the detection of endogenous α-synuclein in the ENS using antibodies to total α-synuclein (Table 1). Small intestine homogenates were treated with vehicle alone (DMSO) or with concentration gradients of 0.5, 1.0 and 2.5 mM of DSP in the absence or presence of the reducing agent agent dithiothreitol (DTT). Treatment of small intestine homogenates with DSP and DTT resulted in an increased 15-kDa monomeric α-synuclein signal that was maximal at 2.5 mM when (C20)-R (Fig. 2A, B), LB509 (Fig. 2E,F) and Syn211 (Fig. 2 G, H) were used and at 1 mM when Syn-1 immunoblotting was performed (Fig. 2 C, D). A similar marked enhancement of monomeric α-synuclein signal by DSP/DTT, peaking at 2.5 mM, was also observed with antibodies specific for α-synuclein phosphorylated at serine 129 (Fig. 3). The specificity of the cross-linking was supported by the lack of effect of DSP in the absence of DTT (Figs. 2A, C, E, G, 3A, C, D). The improved detection of endogenous α-synuclein obtained with DSP/DTT was not significantly enhanced further by additional treatments of the membranes with PFA (data not shown)
In the absence of DSP, all tested antibodies detected a predominant 60-kDa immunoreactive band (Figs. 2A, C, E, G, 3A, C, D). Treatment with DSP and DTT induced a decrease in the immunoreactivity of this 60-kDa band together with a corresponding increase in the levels of 15-kDa monomeric α-synuclein (Figs. 2A, C, E, G, 3A, C, D). Such a pattern observed after in vitro cross-linking is consistent with the existence of tetrameric α-synuclein in the ENS, as already reported in central nervous system neurons [18].
DISCUSSION
Here we show that the immunodetection of endogenous α-synuclein in the ENS is improved when methods for enhanced membrane retention of the protein, such as incubation of the membranes with PFA or treatment of the samples with DSP, are applied [17, 18]. As already reported [18], these two retention methodswere not additional, as the combined use of PFA and DSP did not improve immunodetection beyond the effect of DSP alone. Nevertheless, our results strongly suggest that incubating blot membrane in PFA is less effective than treating the samples with the cross linker DSP for the detection of endogenous α-synuclein in enteric neurons. While Syn-1 and (C20)-R signals in gut samples were increased with PFA, LB509 and Syn211, two other widely used α-synuclein antibodies, failed to detect α-synuclein in the ENS when associated with this retention membrane method. By contrast, the 15-kDa monomeric α-synuclein signal obtained in gut samples using all four total α-synuclein antibodies was markedly enhanced when samples were treated with DSP and β-mercaptoethanol. When it comes to phosphorylated α-synuclein, the difference in effectiveness between the two methods is even more striking, as DSP was the only method capable of detecting monomeric α-synuclein phosphorylated at serine 129 not only in gut but also in brain. These results are in accordance with a recent survey, which reported that PFA fixation in Western blotting using EP1536Y antibody facilitated the visualization of signals for endogenous phosphorylated α-synuclein monomers in neuronal cell lines, but not in brain [22].
As a whole, our results demonstrate that the detection of total and phosphorylated α-synuclein can be markedly increased in the gut, which express low level of the protein. As an optimized protocol, we suggest that gut homogenates should be treated with 2.5 mM DSP and that the immunoblot should be performed with (C20)-R antibodies for the detection of total α-synuclein and either p-α-syn, EP1536Y or pSyn#64 for the detection of the phosphorylated form of the protein. Such an optimized protocol opens the way to the development of new and original biomarkers for PD that will enable a quantification of pathological α-synuclein, either aggregated and/or phosphorylated in gastrointestinal biopsies.
CONFLICT OF INTEREST
The authors declare no actual or potential conflict of interest.
Footnotes
ACKNOWLEDGMENTS
This work was supported by a grant from centre d’entraide et de coordination des associations de parkinsoniens (CECAP) to PD. GB is supported by a Master’s grant of DHU2020. AGC is a recipient of a poste d’accueil Inserm.