Here, we present a protocol for a free-floating indirect immunofluorescence assay on skin biopsy sections that allows for the identification of disease specific conformation variants of alpha synuclein involved in Parkinson disease and multiple proteins of the peripheral nervous system.
To date, for most neurodegenerative diseases only a post-mortem histopathological definitive diagnosis is available. For Parkinson's disease (PD), the diagnosis still relies only on clinical signs of motor involvement that appear later on in the disease course, when most of the dopaminergic neurons are already lost. Hence, there is a strong need for a biomarker that can identify patients at the beginning of disease or at the risk of developing it. Over the last few years, skin biopsy has proved to be an excellent research and diagnostic tool for peripheral nerve diseases such as small fiber neuropathy. Interestingly, a small fiber neuropathy and alpha synuclein (αSyn) neural deposits have been shown by skin biopsy in PD patients. Indeed, skin biopsy has the great advantage of being an easily accessible, minimally invasive and painless procedure that allows the analysis of peripheral nervous tissue prone to the pathology. Moreover, the possibility of repeating the skin biopsy in the course of the follow-up of the same patient allows studying the longitudinal correlation with the disease progression. We set up a standardized reliable protocol to investigate the presence of αSyn aggregates in skin nerve fibers of the PD patient. This protocol involves few short fixation steps, a cryotome sectioning and then a free-floating immunofluorescence double-staining with two specific antibodies: anti Protein Gene Product 9.5 (PGP9.5) to mark the cutaneous nerve fibers and anti 5G4 for detecting αSyn aggregates. It is a versatile, sensitive and easy to perform protocol that can also be applied for targeting other proteins of interest in skin nerves. The ability to mark αSyn aggregates is another step forward to the use of skin biopsy as a tool for establishing a pre-mortem histopathological diagnosis of PD.
Skin biopsy has acquired a great importance as the diagnostic and research tool in the field of neurological disorders1. Indeed, epidermis and dermis contain abundant somatic sensory nerve fibers (myelinated and unmyelinated), nociceptive free nerve endings, sensory receptors and autonomic innervation of sweat glands, vessels, sebaceous glands and muscle arrector pilorum2.
In the mid-20th century, the setup for immunohistochemistry of PGP9.5 antibody allowed the evidence of an extensive innervation of human epidermis mammalian skin3. PGP9.5 is a carboxyl-terminal hydrolase equally distributed along axons of both the central and peripheral nervous system (PNS). The availability of this antibody allowed not only to clarify the morphology and anatomy of PNS in the skin but also implemented the study of diseases associated with it3,4. Skin biopsy contributed to defining a new clinical entity: the small fiber neuropathy. Several international groups demonstrated the association between the loss of intraepidermal nerve fibers and symptoms/signs of small fiber neuropathy5 by skin biopsy analysis and provided standardized protocols for nerve morphometry as well as normative reference values to be used in the clinical practice6,7,8.
Recently a large amount of evidence has shown that neurodegenerative diseases, characterized by misfolded protein accumulations in the central nervous system, are multi-system pathologies9. Indeed, PD is characterized by αSyn accumulation in the dopaminergic neuron of substantia nigra, but it has been demonstrated that αSyn and its pathological form, phosphorylated αSyn (P-αSyn), could be detected also in the peripheral tissues. Gastro-intestinal mucosa10, salivary glands11, skin autonomic fibers surrounding sweat glands and pilomotor muscles12,13,14, show immunoreactivity to pathogenic forms of αSyn, in accordance with Braak hypothesis that intriguingly postulate that αSyn pathology may begin in PNS well in advance, before its accumulation in the brain15. Further, the presence of p-αSyn has been demonstrated in skin nerves of patients with REM Behavior Disorders that are considered prodromal PD16,17 thus skin pathologic αSyn can be considered a promising early peripheral histopathological marker of synucleinopathy.
The association of small fiber neuropathy in PD has been demonstrated previously and it has been found that intraepidermal nerve fibers density reflects disease progression18,19. Hence, the skin biopsy is a useful tool for studying neurodegeneration in PD and for establishing a pre-mortem histopathological diagnosis of the disease. Indeed, skin biopsy has a great advantage of being an easily accessible and minimally invasive procedure, allowing the analysis of nervous tissue prone to the pathology. Finally, the possibility of repeating the skin biopsy in the course of follow-up of the same patients allows studying the longitudinal correlation with disease progression.
In our laboratory, exploiting a double immuno-staining with PGP9.5 and the conformation-specific monoclonal 5G4 antibody, that recognizes disease specific forms of αSyn including small aggregates20,21, we were able to show the presence of αSyn aggregates in skin nerves with a promising high diagnostic efficiency19. Immunofluorescence analysis of the skin biopsy in conformational diseases stands out as a promising source of biomarkers by combining both the detection of protein aggregates and the measure of the neurodegeneration in vivo. Hereafter, we illustrate an easy and versatile protocol on handling the skin biopsy and performing the free-floating immunofluorescence staining for detecting αSyn aggregates. Moreover, this protocol can be adapted for targeting any other protein of interest expressed in skin PNS.
The following study protocol has been used to evaluate the diagnostic utility of aggregated αSyn analysis in the PNS of PD by skin biopsy19. Inclusion criteria for PD were: a definite clinical diagnosis according to the UK Brain Bank diagnostic criteria, disease duration at least 3 years, no family history, and no major cognitive impairment or major dysautonomic symptoms in the history. Exclusion criteria were known causes of neuropathy (glycated hemoglobin, creatinine, vitamin B12, TSH, serum immunofixation, HIV, HCV, syphilis, and borreliosis). Each subject underwent to 3 mm-diameter skin biopsies at three anatomical sites (neck at C8 dermatomal level, thigh 10 cm above the knee, leg 10 cm above lateral malleolus) on the side, which was clinically more affected. In general, the following protocol is about handling the skin biopsy and performing the free-floating immunofluorescence staining and analysis. Hence it can be adapted and used for the detection of other proteins of interest in skin tissue.
The protocol has been approved by the Cantonal Ethics Committee and all enrolled subjects gave written informed consent to the study.
1. Skin Biopsy Collection
2. Tissue Fixation and Storage
3. Tissue Cut with a Cryotome
4. Immunofluorescence Staining
5. Immunofluorescence Imaging
Antifreeze solution (store at 4 °C for up to 6 months) |
30% Glycerol 30% Ethylene glycol 30% dH2O 10% 2x Phosphate buffer |
Blocking solution (prepare at the moment) |
4% Normal Goat Serum 1% Triton X-100 in Washing solution |
Cryo-protectant (store at 4 °C for up to 6 months) |
20% Glycerol 80% Sorensen's solution |
Disodium hydrogen phosphate solution (store at RT up to 6 months) |
0.45M Disodium hydrogen phosphate (Na2HPO4) in dH2O Filter in a sterile bottle |
Lysine solution (store at 4 °C for up to 3 weeks) |
50% of 0.3M L-Lysine monohydrochloride solution 50% of 0.1M Sorensen's solution pH7.6 pH 7.4, filter in a sterile bottle |
Paraformaldehyde (PFA) 8% (prepare under fume hood and store at 4 °C for up to 1 month) |
2.6M PFA in dH2O (to 55°C_do not exceed 60 °C to avoid formic acid formation) filter in a sterile bottle |
Phosphate buffer 2x (store at 4 °C for up to 6 months) |
6% Monosodium phosphate (NaH2PO4) solution 45% Disodium phosphate (Na2HPO4) solution in dH2O |
PLP fixative solution (prepare at the moment, under fume hood ) |
25% Paraformaldehyde 8% 0.01M Sodium (meta)periodate 75% Lysine solution< |
Sodium Dihydrogen Phosphate Monohydrate (store at RT up to 6 months) |
0.52M Sodium Dihydrogen Phosphate Monohydrate (NaH2PO4*H2O) in dH2O Filter in a sterile bottle. |
Sorensen's solution (store at 4 °C for up to 1 month) |
2.5% Monosodium phosphate solution 18.7% Disodium phosphate solution pH7.6, in dH2O |
Washing solution (prepare at the moment) |
0.25M Trizma base 0.26M NaCl pH7.6, in dH2O |
Working solution (prepare at the moment) |
50% Blocking solution 50% Washing solution |
Table 1: Required solutions. List of required solutions for the protocol and brief description of how to prepare it.
Following the described procedure (Figure 1), we detected αSyn aggregates, labeled with 5G4 antibody, in dermal nerve fascicles innervating autonomic structures of PD patients. Morphology of alpha-synuclein deposits appears as a dotted signal along the axons of dermal nerves (Figure 2). Indeed, exploiting this protocol in 19 PD patients and 17 controls in skin biopsies at three anatomical site (cervical, thigh and distal leg) we found that 5G4 had 81% of sensitivity and 86% of specificity respect to healthy controls and P-αSyn had 56% of sensitivity and 100% of specificity19.
In particular, we found 5G4 and P-αSyn positive deposits mainly in nerves around sweat glands but also in muscle erector pili, small vessels, and subepidermal and dermal plexus, never in intraepidermal nerve fibers.
Generally, within the sweat gland’s lumen, it is possible to observe a nonspecific signal, which could be misinterpreted as 5G4/ P-αSyn positivity. This type of signal is dotted, spherical and it is due most probably to intraluminal auto-fluorescent material, as we demonstrated in technical controls without primary and secondary antibodies (Figure 3). Co-localization with PGP9.5 that marks the nerve fibers, which morphologically are filamentous and elongated, helps to identify the correct signal. Therefore, the specificity of 5G4 signal is highly increased by a double immunostaining with an axonal marker (Figure 4).
An accurate fixation of the biopsy is a mandatory factor for the good quality of immunofluorescence staining, and for the reliable interpretation of the fluorescent signals. If the fixative is not correctly prepared or if an over-fixation has occurred, the result will be a high auto-fluorescence that will mask the signal of nerve fibers crossing the dermal-epidermal junction or innervate the main dermal structures (Figure 5B, D, F). In this case, the inability to visualize correctly the PGP9.5 positive fibers will make difficult to analyze correctly also 5G4.
Finally, this protocol can be used for the detection of any protein of interest, including among others αSyn, P-αSyn or tyrosine hydroxylase (TH) and vasoactive intestinal peptide (VIP) that mark respectively adrenergic and cholinergic subtype of autonomic innervation (Figure 6).
Figure 1: Schematic representation of the protocol. Graphical representation of critical steps in the fixation, cut and staining of skin sections for the visualization of αSyn aggregates in cutaneous peripheral nerve fibers. Please click here to view a larger version of this figure.
Figure 2: Presence of 5G4 aggregates in dermal nervous fibers. Confocal merged Z-stack images of immunofluorescence with PGP9.5 (green) and 5G4 (red) of dermal nerves around sweat glands in PD patients (A–C) and healthy subject (D-F). This figure is modified from 19. 3D visualization of the same sweat gland showed above from PD (G) and healthy (H) subjects. Indicated by white arrows, yellow colocalization of the two markers along axons. Scale bar = 50 µm. Please click here to view a larger version of this figure.
Figure 3: Auto-fluorescent signals: technical controls. Confocal merged Z-stack images of skin section stained with DAPI and without the primary antibody (A-B), without the secondary antibody (C-D), without primary and secondary antibodies (E-F). The images show the presence of non-specific dotted signals in sweat glands structures in all conditions. Scale bar = 50 µm. Please click here to view a larger version of this figure.
Figure 4: Consider nerve fibers morphology to recognize non-specific signal. Confocal merge Z-stack images of immunofluorescence with PGP9.5 (green) and 5G4 (red) of dermal nerves around sweat glands. White arrows indicate positive structures; asterisks indicate unspecific staining in non-neuronal structures. Scale bar = 50 µm. This figure is modified from reference19. Please click here to view a larger version of this figure.
Figure 5: Incorrect fixation compromises the quality of immunofluorescence staining. Confocal merge Z-stack images of immunofluorescence with PGP9.5 (green) and DAPI (blue) of intra epidermal nerves fibers (A-B) and dermal nerves around sebaceous gland (C-D) and sweat glands (E-F). On the left result of correct fixation, on the right result of over fixation. Scale bar = 50 µm. Please click here to view a larger version of this figure.
Figure 6: Examples of other proteins detectable in skin nerves. Microscope images of dermal structures stained using a free-floating immunofluorescence assay. In red αSyn (A), p-αSyn (B), and TH (C), in green VIP (D). Scale bar = 50 µm. Please click here to view a larger version of this figure.
We describe a free-floating immunofluorescence assay for skin biopsies for the diagnosis of PD: it exploits double immunostaining with anti-PGP9.5 antibody, a panaxonal marker, and anti-5G4, a conformation specific antibody that recognizes the aggregated form of αSyn.
The great advantages of skin biopsy for diagnostic purpose in PD and possibly in other protein conformational disorders are: 1) the direct access to nervous tissue prone to disease by a mildly invasive technique and thus with an expected better determination sensitivity for αSyn aggregates than biological fluids like blood or CSF; 2) the opportunity to detect and quantify epidermal and autonomic nerve fiber density as a measure of neurodegeneration; 3) the possibility of repeating skin biopsy in the course of follow-up of the same patients, in order to study longitudinally the correlation with disease progression.
The protocol proposed here is rapid, versatile and has few critical steps. First of all, tissue fixation has to be pursued correctly, otherwise a high auto-fluorescence will mask the specific signal and will prevent the correct analysis of the data. The paraformaldehyde in the fixative solution should be made fresh and pH of 7.4 should be accurately checked. It is extremely important to avoid the formation of formic acid that could damage biopsies. Before storage and cut, if the compactness of the biopsy suggests an incorrect fixation, the biopsy should be rinsed again with Sorensen's solution, and re-incubated with PLP solution O/N at 4 °C. Moreover, at the beginning or at the end of the immunofluorescence staining protocol, a few additional steps can be introduced to counteract auto-fluorescence. At the beginning of the protocol (between steps 4.1 and 4.2) a treatment with sodium borohydride solution (1 mg/mL in TBS; 3 times for 10 min), a treatment with hydrogen peroxide (3%; 15 min) or a treatment with glycine (0.1 M in TBS; 1h) can be performed. In alternative at the end of the staining (between steps 4.8 and 4.9), a treatment with trypan blue (250 μg/mL in TBS; 20 min) or with Sudan Black (0.5% in 70% EtOH; 30 min) can be performed. However, in all cases, in addition to the reduction of auto-fluorescence, a reduction in the specific signal will also occur. Alternatively, biopsies incorrectly fixated can be used to perform classical bright field immunohistochemistry. In this case, however, the double immunostaining for colocalization of 5G4 and PGP9.5 will be more challenging and difficult to analyze.
Another critical step is the insertion of a skin biopsy into the cryomold (step 3.3). The orientation of the biopsy towards the cutting blade is crucial to obtain slices in which epidermis and dermis are both present. The longitudinal axis (epidermis-dermis) has to be parallel to the bottom of the cryomold, so that the side of the biopsy, not the top or the bottom, is facing the operator. The choice of slices thickness, 50 µm, is in accordance with European guidelines for the use of skin biopsy as a diagnostic tool7. Moreover, for a αSyn aggregates detection, 50 µm thickness allows with greater probability to have within the section more dermal structures, where pathological deposit are mainly found in PD. Due to the high thickness, to be sure that the whole section is dyed, it is not recommended to place the sections on the slide for staining, but instead, a free floating staining is mandatory. For this procedure, the major difficulty is to transfer the sections from one well to another using a small brush without damaging the sections. It is recommended to insert the tip of the brush below the biopsy that floats in the liquid, allowing the biopsy to settle on the tip of the brush, gently carry the biopsy from one well to another, immerse the biopsy in the new solution and remove the brush making sure that no biopsy residue is left on the brush. It is important that the operator acquire manual skills with practice.
In conclusion, this is a short and easy protocol for the optimal handling and fluorescent immunostaining of skin biopsy. The advantage of this protocol in respect to previous studies is the use of 5G4 antibody, allowing the detection of αSyn small aggregates for the first time in dermal autonomic nerve fibers of PD patients. It can be potentially used for diagnostic purposes in different types of neurodegenerative disorders involving PNS, especially at early phases, when potential cure can be most effective. Limitations of the method is that sensitivity of P-αSyn and specificity of 5G4 are still suboptimal but a combination of different markers in future studies could certainly improve the diagnostic yield of skin biopsy in PD.
The authors have nothing to disclose.
We thank Parkinson Schweiz and ABREOC (the Scientific Research Advisory Board of the Ente Ospedaliero Cantonale) for their financial support of this study.
5G4 (anti human αSyneclein 5G4) | Analytik Jena Roboscreen | 847-0102004001 | Mouse monoclonal |
AlexaFluor 488 Goat anti Rabbit IgG | Invitrogen | 1971418 | 2mg/ml |
AlexaFluor 594 Goat anti Mouse IgG | Invitrogen | 1922849 | 2mg/ml |
Disodium hydrogen phosphate solution | Merk Millipore | 106586 | |
Ethylene Glycol | Sigma-Aldrich | 324558 | |
Glycerol | Sigma-Aldrich | G7757 | |
L-Lysine monohydrochloride | Sigma-Aldrich | L5626 | |
Paraformaldehyde | Aldrich Chemistry | 441244 | |
PGP9.5 | Abcam | ab15503 | Rabbit polyclonal |
Sodium Chloride | Sigma | S3014 | |
Sodium Dihydrogen Phosphate Monohydrate | Merck Millipore | 106346 | |
Sodium (meta)periodate | Sigma-Aldrich | S1878 | |
Trizma Base | Sigma | T1503 | |
Tryton X-100 | Sigma-Aldrich | X100 | |
Vectashield | Vector Laboratories | H-1000 | Mounting medium |