Here, we present a brain membrane fractionation protocol that represents a robust procedure to isolate proteins belonging to different synaptic compartments.
Assessing the synaptic protein composition and function constitutes an important challenge in neuroscience. However, it is not easy to evaluate neurotransmission that occurs within synapses because it is highly regulated by dynamic protein-protein interactions and phosphorylation events. Accordingly, when any method is used to study synaptic transmission, a major goal is to preserve these transient physiological modifications. Here, we present a brain membrane fractionation protocol that represents a robust procedure to isolate proteins belonging to different synaptic compartments. In other words, the protocol describes a biochemical methodology to carry out protein enrichment from presynaptic, postsynaptic, and extrasynaptic compartments. First, synaptosomes, or synaptic terminals, are obtained from neurons that contain all synaptic compartments by means of a discontinuous sucrose gradient. Of note, the quality of this initial synaptic membrane preparation is critical. Subsequently, the isolation of the different subsynaptic compartments is achieved with light solubilization using mild detergents at differential pH conditions. This allows for separation by gradient and isopycnic centrifugations. Finally, protein enrichment at the different subsynaptic compartments (i.e., pre-, post- and extrasynaptic membrane fractions) is validated by means of immunoblot analysis using well-characterized synaptic protein markers (i.e., SNAP-25, PSD-95, and synaptophysin, respectively), thus enabling a direct assessment of the synaptic distribution of any particular neuronal protein.
Synaptic transmission relies on the physical integrity of the synapse, a concept that was envisaged as early as 1897 by Foster and Sherrington1. Thus, understanding the distribution of key neurotransmission components (e.g., ion channels, receptors, etc.) is essential to elucidate synaptic function, both in normal and pathological conditions. Electron microscopy (EM) has contributed enormously to the current ultrastructural notion of prototypical central nervous system (CNS) synapses. In such a way, EM has finely established the differences between pre- and postsynaptic densities, which are separated by a cleft of a rather uniform distance (~25 nm)2. Interestingly, the postsynaptic apparatus exhibits a relatively continuous, electron-dense thickening below its plasma membrane, the so-called postsynaptic density, or PSD2. Conversely, at the presynaptic apparatus, a noticeable discontinuous cytomatrix network is arranged just beneath the plasma membrane, which is essential to the alignment and docking of synaptic vesicles to the plasma membrane active zone3. Hence, EM constitutes the golden experimental approach to survey the distribution of proteins within structurally preserved CNS synapses. However, the information provided by electron micrographs is static. Indeed, accumulating evidences show that in vivo synapses are extremely dynamic, thus experiencing dramatic structural changes upon sustained synaptic transmission. In addition, the morphology and composition of synapses can change throughout different CNS regions and upon development, maturation, aging, and the development of neuropathological conditions. Overall, a protocol focused on isolating proteins belonging to different synaptic compartments in physiological conditions represents a valuable tool for a more comprehensive study of synaptic functioning.
Here, we describe this kind of complementary experimental approach, which allows for the preparative biochemical enrichment of the different synaptic membrane compartments-namely, extra-, pre- and postsynaptic membrane domains. This membrane fractionation method, first described by Philips et al. (2001)4, is based on a pH shift that weakens the adhesive interactions occurring within the pre- and postsynaptic apparatus. First, by using mild detergents at pH 6.0, it is possible to discern the adherens junction that holds the pre- and postsynaptic apparatus and that is maintained from the extrasynaptic membrane domain, which is solubilized and thus can be extracted from the synaptic contacts. Subsequently, raising the pH from 6.0 to 8.0 in the presence of mild detergents weakens the strength of the adherens junction that keeps the presynaptic active zone tightly bound to the postsynaptic density. Hence, the presynaptic compartment is solubilized and can be separated from the postsynaptic density, which is mostly preserved because the concentration of detergent used does not promote its solubilization4. Interestingly, the fractionation efficiency, eventually higher than 90%, can be confirmed by different subsynaptic markers: i) synaptosomal-associated protein 25 (SNAP-25), from the presynaptic active zone; ii) synaptophysin, from the extrasynaptic fraction (i.e., outside the active zone and including microsomes); and iii) postsynaptic density protein 95 (PSD-95), from the postsynaptic density. Notably, this brain membrane fractionation method has been used successfully. Accordingly, it has been possible to precisely determine the subsynaptic localization of different receptors, such as alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors5, adenosine A1 receptor (A1R)6, adenosine A2A receptor (A2AR)7, adenosine triphosphate (ATP) P2 receptors8, nicotinic acetylcholine receptor subunits9, and Parkinson's disease-associated receptor GPR3710. However, a number of limitations may impede the proper assessment of the synaptic distribution of a particular neuronal protein. Thus, in this procedure, we not only fully describe the entire protocol, but we also highlight some critical points to be considered, such as the rather large amount of tissue needed, the low protein yield, and the mandatory requirement to validate the efficiency of each separation before performing the definite experiment.
All animal experimental procedures were approved by the University of Barcelona Committee on Animal Use and Care (CEEA), in compliance with the guidelines described in the Guide for the Care and Use of Laboratory Animals11 and following the European Community, law 86/609/CCE, FELASA, and ARRIVE guidelines. Thus, mice are housed in standard cages, with ad libitum access to food and water, and are maintained under controlled standard conditions (12 h dark/light cycle starting at 7:30 AM, 22 °C temperature, and 66% humidity).
1. Obtaining Mouse Brain Synaptosomes Using a Discontinuous Sucrose Gradient
Note: This method was reported previously10.
2. Pre-, Post- and Extrasynaptic-isolation
3. Analyze the samples by immunoblot to validate the membrane fractionation
The described methodology has been largely used for the subsynaptic analysis of neuronal proteins in general and for the isolation and biochemical characterization of synaptic receptors5,6,7,8,9 in particular. Interestingly, the representative result displayed here show the usefulness of this experimental procedure for the analysis of the subsynaptic hippocampal distribution of an orphan G protein-coupled receptor- namely, the Parkinson's disease-associated receptor GPR3710. GPR37 was originally identified through searches for homologues for endothelin and bombesin receptors13, although it was found not to bind to endothelins or related peptides. In its place, GPR37 was proposed to be activated by the head activator peptide14,15,16 and, more recently, by the neuropeptides prosaposin and prosaptide17, although these associations are yet to be universally accepted. GPR37 has received the most attention for its linkage to Parkinson's disease18. Thus, there is significant interest in knowing the neurobiology of this intriguing receptor, both under normal and pathological conditions. Therefore, uncovering the GPR37 subsynaptic localization might help to clarify its function within the brain. To this end, the hippocampus was first isolated from C57BL/6J (WT) and GPR37-KO mice at eight weeks of age. Then, the extra-, pre- and post-subsynaptic fractions were purified using the membrane fractionation protocol (see Figure 1 for a schematic overview of the procedure). Subsequently, the purities of these subsynaptic compartments were verified by the segregation of the respective synaptic markers: i) the extrasynaptic vesicular marker (synaptophysin); ii) the presynaptic active zone marker (SNAP-25); and iii) the postsynaptic density marker (PSD95). Accordingly, the enrichments in synaptophysin, SNAP-25, and PSD95 within the extra-, pre- and post-subsynaptic fractions, respectively, were analyzed by immunoblot using specific antibodies against these proteins (Figure 2). Hence, a fractionation efficiency of at least 90% was found for each synaptic marker tested (Figure 3), similar to those described previously6. Interestingly, GPR37 immunoreactivity was more abundant (n = 3; P <0.001) in the extrasynaptic fraction when compared to the presynaptic (20 ± 4%) and the postsynaptic (36 ± 2%) fractions (Figure 3). In addition, our data indicated that, while present at the presynaptic active zone, GPR37 was mainly localized in the postsynaptic density (n = 3; P <0.05) (Figure 3). Overall, the brain membrane fractionation protocol allowed for the assessment of the subsynaptic distribution of GPR37 in the mouse hippocampus, thus providing valuable information for future manipulations of this orphan receptor.
Figure 1: Schematic flow-chart representation of the membrane fractionation protocol. All experimental procedures are described in the left-hand column, while the sample collection is depicted in the right-hand column. Please click here to view a larger version of this figure.
Figure 2: Subsynaptic distribution of GPR37 in the mouse hippocampus. Representative immunoblot showing Synaptophysin, SNAP-25 and PSD-95 as extra-, pre- and postsynaptic specific synaptic markers, as well as GPR37 immunoreactivity in hippocampal synaptic fractions of WT and GPR37-KO mice. Hippocampal synaptosomes (Syn) were fractionated into extrasynaptic (Extra) and presynaptic (Pre) active zones and postsynaptic density (Post) fractions. These were analyzed by immunoblot (20 µg of protein/lane) using the rabbit anti-synaptophysin (1:3,000), mouse anti-SNAP-25 (1:3,000), rabbit anti-PSD95 (1:3,000), and rabbit anti-GPR37 (1 µg/mL) antibodies. The primary bound antibody was detected using either an HRP-conjugated goat anti-rabbit (1/30,000) or a rabbit anti-mouse (1:30,000) antibody. These data are extracted from reference10, with permission. Please click here to view a larger version of this figure.
Figure 3: Relative quantification of GPR37 enrichment in hippocampal extra-, pre-, and postsynaptic fractions. The intensities of the immunoreactive bands on the immunoblotted membranes corresponding to extrasynaptic (Extra; yellow column), presynaptic (Pre, green column), and postsynaptic (Post, red column) fractions, shown in Figure 2, were measured by densitometric scanning. The densities were quantified from non-saturated bands. Values were normalized (in % of the relative densitometric scanning, RDS) using the amount of Synaptophysin, SNAP-25, PSD95, and GPR37 in the most enriched fraction and were presented as the means ± SEM of three independent experiments10. The asterisks denote significantly different data: *p <0.05, ***p <0.001 (1-way ANOVA with a Bonferroni's post hoc test). Please click here to view a larger version of this figure.
The protocol presented here constitutes a powerful biochemical tool for the study of the subsynaptic distribution of specific proteins within any brain region. However, there are some drawbacks inherent to the technique that deserve to be highlighted here. For instance, one of the main limitations is the relatively large amount of tissue needed to purify a reasonable amount of protein in order to perform the immunoblot analysis of all subsynaptic fractions. This issue might be related to the fact that synapses (i.e., synaptosomes) represent only 1-2% of the total hippocampal volume12. Indeed, between 1 and 1.5 g of fresh tissue (i.e., the hippocampus) is needed to perform a successful fractionation; otherwise, the yield is too low to assess the identity, as well as the localization, of the protein being studied.
Conversely, if an excess of brain tissue is used, then the separation procedure will not be optimal. It is also critical to carefully adjust the pH of the solutions to ensure an optimal subsynaptic fractionation. Consequently, every time a brain membrane fractionation is performed, it is mandatory to validate the efficiency of each fraction before any further study. Importantly, the following protocol it is unlikely to be suitable for the subsynaptic fractionation of inhibitory synapses because of its differential structure and distribution. However, all these drawbacks do not overshadow the enormous usefulness of this experimental procedure, which, without any doubt, will become a universal methodology for the study of the synapse.
The authors have nothing to disclose.
This work was supported by Ministerio de Economìa y Competitividad/Instituto de Salud Carlos III (SAF2014-55700-P, PCIN-2013-019-C03-03 and PIE14/00034), Instituciò Catalana de Recerca i Estudis Avançats (ICREA Academia-2010), and Agentschap voor Innovatie door Wetenschap en Technologie (SBO-140028) to F. C. Also, X. M, V. F.-D., and F. C. belong to the “Neuropharmacology and Pain” accredited research group (Generalitat de Catalunya,2014 SGR 1251). The work was also supported by the “Programa Pesquisador Visitante Especial-Ciência sem Fronteiras” from CAPES (Brazil) to F.C.
Sucrose | Pancreac Química SL, Barcelona, Spain | 1,316,211,211 | |
CaCl2 | Pancreac Química SL, Barcelona, Spain | 2,112,211,210 | |
MgCl2·6H2O | Pancreac Química SL, Barcelona, Spain | 1,313,961,210 | |
Protease inhibitor cocktail Set III | Millipore, Darmstadt, Germany | 535140 | |
Trizma Base | Sigma, St. Louis, MO, USA | T1503 | |
Tris-HCl | Pancreac Química SL, Barcelona, Spain | 1,236,541,209 | |
Triton X-100 | Sigma, St. Louis, MO, USA | X100 | |
SDS | Sigma, St. Louis, MO, USA | L3771 | |
Glycerol | Sigma, St. Louis, MO, USA | G5516 | |
Bromophenol Blue | Pancreac Química SL, Barcelona, Spain | 1,311,651,604 | |
Dithiothreitol | Sigma, St. Louis, MO, USA | D0632 | |
Tween 20 | Sigma, St. Louis, MO, USA | P2287 | |
Non fat dry milk | |||
NaCl | Pancreac Química SL, Barcelona, Spain | 1,216,591,211 | |
KCl | Pancreac Química SL, Barcelona, Spain | 1,314,941,210 | |
KH2PO4 | Merck | 4873 | |
Na2HPO4 | Pancreac Química SL, Barcelona, Spain | 1,316,781,211 | |
Basic 20 pH | Crison, Alella, Spain | ||
Polytron VDI 12 Adaptable Homogenizer | VWR, Radnor, PA, USA. | ||
Ultra-Clear Tubes (14x89mm) | Beckman Coulter, Hospitalet de Llobregat, Barcelona | 344059 | Tubes should be filled almost completely when used to prevent collapsing due to ultracentrifugation. |
Amicon Ultra-15 Centrifugal filters Ultracel -10K | Merck Millipore, Darmstadt, Germany | UFC901008 | |
Centrifuge 5430R | Eppendorf, Hamburgo, Germany | ||
Optima L-90K Ultracentrifuge | Beckman Coulter, Hospitalet de Llobregat, Barcelona | ||
Sonifier 250 | Branson, Danbury, Connecticut | ||
Amersham Imager 600 | GE Healthcare Europe GmbH, Barcelona, Spain | ||
Disposable Glass Pasteur Pippetes 230 mm | VWR, Radnor, PA, USA | 612-1702 | |
Compact Balance EK-610 | A&D, Tokyo, Japan | ||
Pierce™ BCA Protein Assay Kit | Pierce Biotechnology, Rockford, IL, USA | ||
SuperSignal west pico chemiluminescent substrate | Thermo Fisher Scientific Inc., Waltham, MA, USA | ||
GR 200 Precision Balance | A&D, Tokyo, Japan | ||
Anti-GPR37 | Homemade antibody anti-GPR37 produced and validated in Francisco Ciruela Laboratory. | Primary antibodies used at a final concentration of 0.250ug/ml | |
Anti-SNAP-25, anti-PSD-95, anti-synaptophysin | Abcam, Cambridge, United Kingdom | Primary antibodies diluted 1:10000 | |
HRP-conjugated goat anti-mouse IgG | Thermo Fisher Scientific, Waltham, MA, USA. | Secondary antibody diluted 1:10000 | |
HRP-conjugated goat anti-rabbit IgG | Thermo Fisher Scientific, Waltham, MA, USA. | Secondary antibody diluted 1:30000 |