Proteome characterization of ocular microvascular beds is pivotal for in-depth understanding of many ocular pathologies in humans. This study demonstrates an effective, rapid, and robust method for protein extraction and sample preparation from small blood vessels employing the porcine short posterior ciliary arteries as model vessels for mass-spectrometry-based proteomics analyses.
The use of isolated ocular blood vessels in vitro to decipher the pathophysiological state of the eye using advanced technological approaches has greatly expanded our understanding of certain diseases. Mass spectrometry (MS)-based proteomics has emerged as a powerful tool to unravel alterations in the molecular mechanisms and protein signaling pathways in the vascular beds in health and disease. However, sample preparation steps prior to MS analyses are crucial to obtain reproducible results and in-depth elucidation of the complex proteome. This is particularly important for preparation of ocular microvessels, where the amount of sample available for analyses is often limited and thus, poses a challenge for optimum protein extraction. This article endeavors to provide an efficient, rapid and robust protocol for sample preparation from an exemplary retrobulbar ocular vascular bed employing the porcine short posterior ciliary arteries. The present method focuses on protein extraction procedures from both the supernatant and pellet of the sample following homogenization, sample cleaning with centrifugal filter devices prior to one-dimensional gel electrophoresis and peptide purification steps for label-free quantification in a liquid chromatography-electrospray ionization-linear ion trap-Orbitrap MS system. Although this method has been developed specifically for proteomics analyses of ocular microvessels, we have also provided convincing evidence that it can also be readily employed for other tissue-based samples.
The advancement in the field of proteomics, which permits integrated and unsurpassed data collection power, has greatly revolutionized our understanding of the molecular mechanisms underlying certain disease conditions as well as in reflecting the physiological state of a specific cell population or tissue1,2,3,4. Proteomics has also proved to be an important platform in ophthalmic research owing to the sensitivity and unbiased analysis of different ocular samples that facilitated identification of potential disease markers for eventual diagnosis and prognosis, as evidenced elegantly by many studies in recent years, including some of ours1,5,6,7,8,9,10. However, it is often difficult to obtain human samples for proteomic analyses due to ethical reasons, especially considering the need for control material from healthy individuals for reliable comparative analyses. On the other hand, it is also challenging to obtain sufficient amount of samples for optimal and reliable mass spectrometric analyses. This is particularly crucial for mass-limited biological materials such as the micro-blood vessels of the eye. One such major retrobulbar blood vessel that plays pivotal roles in the regulation of ocular blood flow is the short posterior ciliary artery (sPCA). Any perturbation or anomalies in this vascular bed may result in severe clinical repercussions, which can lead to the pathogenesis of several sight-threatening diseases such as glaucoma and nonarteritic anterior ischemic optic neuropathy (NAION)11,12. However, there is a lack of studies elucidating the proteome changes in this arterial bed due to the above-mentioned drawbacks. Therefore, in recent years, the house swine (Sus scrofa domestica Linnaeus, 1758) has emerged as a good animal model in ophthalmic research owing to the high morphologic and phylogenetic similarities between humans and pigs13,14,15. Porcine ocular samples are easily available and most importantly, are more accurate representation of human tissues.
Considering the important role of these blood vessels in the eye, as well as the dearth of methodology catered for efficient protein extraction and analyses from these microvessels, we have previously characterized the proteome of the porcine sPCA using an in-house protocol that resulted in the identification of a high number of proteins16. Based on this study, we have further optimized and described in-depth our methodology in this article, which allows proteome analysis from minute amounts of samples using the porcine sPCA as model tissue. Albeit the main aim of this study was to establish a MS-compatible methodology for mass-limited ocular blood vessels, we have provided substantial experimental evidence that the described workflow can also be broadly applied to various tissue-based samples.
It is envisioned that this workflow will be instrumental for preparation of high-quality MS-compatible samples from small quantities of materials for comprehensive proteome analyses.
All experimental procedures using animal samples were performed in strict adherence to the Association for Research in Vision and Ophthalmology (ARVO) Statement for the Use of Animals in Ophthalmic and Vision Research and by institutional guidelines. This study was conducted and approved at the Department of Ophthalmology, University Medical Centre Mainz.
NOTE: Porcine eyes together with optic nerve and extraocular tissues were obtained fresh from the local abattoir immediately post-mortem. Enucleated eyes were transported to the laboratory in ice-cold phosphate buffered saline (PBS) and used immediately. The schematic overview of the workflow employed is as depicted in Figure 1.
1. Solutions
2. Isolation of Short Posterior Ciliary Arteries
NOTE: The porcine eye is basically divided into the anterior (Figure 2A) and posterior sections (Figure 2B).
3. Sample Preparation
4. Pellet Digestion
5. Sample Cleaning and Buffer Exchange
6. Protein Measurement
7. One-dimensional Gel Electrophoresis (1DE)
8. In-gel Tryptic Digestion
NOTE: This protocol is according to the method by Shevchenko et al.17, with slight modifications. This procedure should be carried out in a laminar flow hood and use dedicated set of pipettes, tips, tubes, and glassware specifically for this purpose. Wear gloves and appropriate lab apparel at all times to prevent keratin and other contamination. Prepare all solutions and reagents used in this procedure shortly before use.
9. Peptide Purification
NOTE: This peptide sample desalting and purification procedure is carried out with the use of C18 pipette tips (see C15 in the Table of Materials). Use a new tip for each sample.
10. Liquid chromatography-electrospray Ionization-MS/MS Analyses
NOTE: Label-free quantitative proteomics analysis is performed on a liquid chromatography-electrospray ionization-linear ion trap-Orbitrap (LC-ESI-LTQ-Orbitrap) MS system. The LC is composed of Rheos Allegro quaternary pump equipped with an online degasser (coupled to an HTS PAL autosampler, and the system comprises a 30 mm x 0.5 mm C18 pre-column connected to a 150 mm x 0.5 mm C18 column. Use reverse phase aqueous solvent A consisting of LC-MS grade water with 0.1% (v/v) formic acid and organic solvent B consisting of LC-MS grade acetonitrile with 0.1% (v/v) formic acid. Use the gradient with a running time of 60 min per gel band, as described in detail in our previous studies6,16.
Limited sample availability is one of the major drawbacks in ophthalmic research. Correspondingly, extraction methods for optimum protein yield from small amounts of samples such as ocular blood vessels are often debatable. To date, there is a paucity of methods catered particularly for protein extraction from retrobulbar blood vessels. Therefore, as a first step in method optimization and as a proof-of-principle to compare the efficacy and robustness of several commonly employed protein extraction detergents to a relatively new reagent, T-PER, we carried out a pilot study using cardiac tissues from mice (due to easy and sufficient sample availability for optimization steps). We compared the protein yield by comparing the total protein concentrations and total proteins identified using the following reagents: T-PER, 0.02% n-dodecyl-β-D-maltoside (DDM), 1% 3-[(3-cholamidopropyl)-dimethylammonio]-1-propane sulfonate (CHAPS), 1% amidosulfobetaine-14 (ASB-14) and a mixture of ACN and TFA (20% ACN/ 1% TFA). The total amount of proteins in samples extracted with each of these detergents is as depicted in Figure 6A, with the highest yield from tissues extracted with T-PER (56.4 µg/mg tissue), followed by DDM (22.88 µg/mg tissue), CHAPS (16.01 µg/mg tissue), ASB-14 (11.56 µg/mg tissue) and the lowest yield from ACN/TFA (4.38 µg/mg tissue). Consistently, the total proteins identified were also the highest in the T-PER-extracted sample (1649 proteins) > DDM (1310 proteins) > CHAPS (1319 proteins) > ASB-14 (1121 proteins) > ACN/TFA (924 proteins), as shown Figure 6B. Based on these results, we proceeded with the protein extraction from sPCA samples with T-PER.
Next, optimization of sample preparation protocols prior to pre-fractionation in 1DE is a very crucial step to obtain well-separated protein bands and highly reproducible results between samples and replicates. The importance of these steps is reflected and highlighted in the results of 1DE. Figure 7 shows the comparison of the 1DE protein profiles of sPCA before and after subjected to the optimized sample preparation and cleaning steps. Overall, a high degree of smearing and poor separation of the protein bands was observed at lane 3. This profile demonstrates that the samples may contain extraction reagent and also contaminants such as lipids and cellular debris. However, the sPCA samples that were separated into supernatant and pellet and, subjected to the optimized protocol resulted in exemplary 1DE profiles, as represented in lane 1 and 2 (Figure 7).
In gist, based on these promising results, the optimized method for rapid, robust and efficient soluble protein extraction from ocular microvessels is employing tissue protein extraction reagent (T-PER) and using Extraction Buffer 2A (TM-PEK) to extract membrane-based proteins found in the sample pellet. Subsequently, sample homogenates (the T-PER fraction) are subjected to buffer exchange and sample cleaning with the 3 kDa centrifugal filter units prior to 1DE. The optimum sample concentration is 50 µg per well. On the other hand, it has to be highlighted here that this protocol is not only applicable for small tissues such as blood vessels, but is also feasible for other tissue-based samples. This is evidenced by the 1DE profiles of murine brain and heart tissues, which demonstrated that there are large numbers of proteins that can be extracted from both supernatant (Figure 8A,B) and pellet fractions (Figure 8C,D).
Figure 1: Workflow overview. A schematic representation of the protocols employed for label-free quantitative proteome analysis of the porcine sPCA. In general, this procedure is divided into two major sections comprising microvessel sample preparation steps and MS-based proteomics approach. Please click here to view a larger version of this figure.
Figure 2: Representative photographs of the porcine eyes. (A) Lateral view of the eye globe shows the cornea, which is located in the anterior part of the eye, the sclera, surrounding muscle and optic nerve. (B) Posterior view of the eye shows the optic nerve. (C) Branches of sPCA seen at the back of the eye globe composed of the paraoptic and distal branches. (D) Isolated sPCA with surrounding fat and connective tissues. Please click here to view a larger version of this figure.
Figure 3: Sample homogenization. Representative photographs showing the sample (A) before and (B) after homogenization. (C) The homogenized samples were further separated into supernatant containing soluble proteins and pellet containing insoluble, transmembrane-based proteins. Please click here to view a larger version of this figure.
Figure 4: Pellet protein extraction procedure. In order to prevent protein denaturation, pellet samples are extracted on ice. Please click here to view a larger version of this figure.
Figure 5: Sample cleaning. Buffer exchange is carried out with 3 kDa cutoff centrifugal filters to clean the samples prior to MS analysis. Representative photographs showing the homogenate (A) before and (B) after filtration. Please click here to view a larger version of this figure.
Figure 6: Comparison of protein extraction efficacy and robustness between T-PER and four common protein extraction detergents. Bar charts depicting the (A) protein amounts and (B) total number of proteins identified using T-PER, 0.02% DDM, 1% CHAPS, 1% ASB-14 and 20% ACN/ 1% TFA. Please click here to view a larger version of this figure.
Figure 7: Representative 1DE gel of the porcine sPCA protein profiles before and after optimization. Lane 3 depicts the 1DE profile of sample that was not subjected to any prior cleaning steps. Lane 1 and 2 show the exemplary protein profiles of sPCA supernatant and pellet, respectively, after subjecting the samples to the optimized sample preparation and cleaning steps. Gel was stained with colloidal blue staining kit. M = Marker. Please click here to view a larger version of this figure.
Figure 8: Representative colloidal blue-stained 1DE gel of exemplary tissue-based samples. Protein profiles of supernatant (A,B) and pellet (C,D) of murine brain and cardiac tissue samples, respectively, at 50 µg per well. Supernatant and pellet proteins were extracted employing the T-PER and transmembrane protein extraction kit, respectively. M = Marker. R1-R3 represent three replicates. Please click here to view a larger version of this figure.
Component | Volume (µL) |
Sample (supernatant or pellet) | x |
LDS Sample Buffer (4x) | 2.5 |
Reducing agent (10x) | 1 |
Deionized water | Up to 6.5 (depending on sample volume) |
Total volume per sample | 10 |
Note: x is calculated based on the protein concentration (50 µg total protein per sample). |
Table 1: 1 Dimensional Gel Electrophoresis (1DE). The details of the components required for sample preparation to perform 1DE.
Solution | For 1 gel (mL) | For 2 gels (mL) | For 4 gels (mL) |
Deionized water | 40 | 80 | 160 |
Methanol | 50 | 100 | 200 |
Acetic acid | 10 | 20 | 40 |
Table 2: Gel fixation. The details of the components required to prepare fixing solution for the 1DE.
Solution | For 1 gel (mL) | For 2 gels (mL) | For 4 gels (mL) |
*Deionized water | 55 | 110 | 220 |
*Methanol | 20 | 40 | 80 |
*Stainer A | 20 | 40 | 80 |
Stainer B | 5 | 10 | 20 |
Table 3: Gel staining. The details of the components for preparation of the Colloidal Blue staining solution.
Solution (for) | Composition | ACN** | H2O** | TFA | Total volume* |
Wetting | 100% ACN | 2 mL | 2 mL | ||
#Washing and equilibration | 0.1% TFA | 10 mL | 10 µL | ~ 10 mL | |
Peptide elution | 0.1% TFA in 60:40= ACN: H2O | 6 mL | 4 mL | 10 µL | ~ 10 mL |
Note: * The total volume should be adjusted according to the total number of samples to be subjected to Zip Tip cleaning. | |||||
** Use HPLC-grade or LC-MS-grade. | |||||
# Prepare two separate Eppendorf tubes for washing and equilibration, respectively. |
Table 4: Peptide purification. The details of the components and their respective compositions for peptide purification procedure using the C18 pipette tips.
Comprehensive proteome profiling of a diverse range of ocular samples is an important and indispensable first step to elucidate the molecular mechanisms and signaling pathways implicated in health and disease. In order to obtain high quality data and to ensure the reproducibility of results obtained from these analyses, the preceding sample preparation steps are crucial, as highlighted in a review by Mandal et al. that discussed in-depth the sample processing procedures for different parts of the eye employing two-dimensional gel electrophoresis and mass spectrometry strategy1. In line of these investigations, our current study provides an optimized step-by-step protocol for rapid, robust and highly efficient MS-compatible sample preparation using the porcine sPCA as model ocular microvessels. This investigation was initiated following the paucity of specific methodology to extract sufficient amounts of proteins from quantity-limited arterial samples to generate high quality MS data. Our method is also an endeavor to contribute to the existing body of knowledge on the use of micro-scale techniques that enable excellent proteome mapping18.
There are several critical aspects in this experimental protocol that need to be taken into consideration for optimal performance for a quantitative proteome analysis. First, it is important that the samples, regardless of the amounts, are subjected to complete homogenization to ensure optimal protein extraction. In our methodology, the use of a mixture of different sized beads and bullet blender homogenizer was instrumental for complete tissue lysis. The type and size of beads used depend on the sample type and amount. Beads with higher densities, such as the currently utilized ZrO2 and stainless steel, are suitable for medium- tough tissues and worked especially well for blood vessels.
Second, it is imperative to separate the supernatant from pellet and, to subject the latter to digestion and extraction using the specified kit. This step is pivotal to extract high molecular weight proteins such as transmembrane proteins, which are otherwise difficult to homogenize using mild detergents19,20,21. Precipitated pellet is best dissolved using sonication to avoid sample loss incurred by splash-up introduced during vigorous agitation or shaking methods.
Third, it is noteworthy that all sample preparation procedures are carried out at low temperature (4 °C), unless otherwise indicated in the methodology. This is to ensure minimal protein denaturation during the extraction procedures. Fourth, repeated freeze-thawing of samples should be avoided to prevent protein degradation and deterioration of sample quality.
Finally, removal of contaminants and detergents is necessary following protein extraction to prevent downstream interference during in-gel fractionation, enzymatic digestion, and MS analysis18,22. These contaminants often interfere with the resolution of the electrophoretic separation and correspondingly, influence the visualization of the result, as shown in the 1DE profile (Figure 7). To circumvent this issue, the use of centrifugal cutoff filter devices is favored for their ease of use and minimal protein loss.
Although the current experimental procedures provide an in-depth outlook into the important sample preparation steps for optimal label-free quantitative MS analyses, there are two limitations. First, sPCA samples were pooled from two porcine eyes to provide sufficient amounts of tissues for subsequent analysis. Since the eyes obtained from the local abattoir are randomized and therefore, it is not known if the blood vessels are being isolated from the eyes of the same animal, sample pooling mitigates inter-individual variations5,6. However, the current methodology can also be adapted for individual sample preparation depending on the amount of samples available. Second, the presented methodology has been specifically developed for 1DE gel-based fractionation. Although the compatibility of the current method for integration with top-down and other fractionation methods warrant investigation, we opted for 1DE owing to several factors ranging from good reproducibility, ease of quality control to better depth of analyses, especially for complex samples such as the currently exemplified ocular blood vessels1,5,23.
In conclusion, despite the limitations highlighted above, the described workflow represents a simple yet robust approach to stringent sample preparation steps catered specifically for analysis of small amount of blood vessels. It is also important to highlight here that this method can be readily integrated for mass spectrometry-based proteomic analysis of other cell- and tissue-based samples.
The authors have nothing to disclose.
Dr. Manicam is supported by the Internal University Research Funding (Stufe 1) from the University Medical Centre of the Johannes Gutenberg University Mainz and a grant from the Deutsche Forschungsgemeinschaft (MA 8006/1-1).
A. Chemicals | |||
1, 4-Dithiothreitol (DTT) | Sigma-Aldrich | 1.11474 | |
Ammonium bicarbonate (ABC, CH₅NO₃) | Sigma-Aldrich | 5.33005 | |
Calcium chloride dihydrate (CaCl2) | Carl Roth | 5239.1 | 2.5 mM |
Dulbecco's phosphate-buffered saline (PBS) | Thermo Fisher Scientific | 14190169 | |
Formic acid (CH2O2) | AppliChem | A0748 | |
HPLC-grade acetonitrile (ACN, C2H3N) | AppliChem | A1605 | |
HPLC-grade methanol (CH3OH) | Fisher Scientific | M/4056/17 | |
HPLC-grade water | AppliChem | A1589 | |
Iodoacetamide (IAA) | Sigma-Aldrich | I6125 | |
Kalium chloride (KCl) | Carl Roth | 6781.1 | 4.7 mM |
Kalium dihydrogen phosphate (KH2PO4) | Carl Roth | 3904.2 | 1.2 mM |
LC-MS-grade acetic acid | Carl Roth | AE69.1 | |
Magnesium sulphate (MgSO4) | Carl Roth | 261.2 | 1.2 mM |
NuPAGE Antioxidant | Thermo Fisher Scientific (Invitrogen) | NP0005 | |
NuPAGE LDS Sample buffer | Thermo Fisher Scientific (Invitrogen) | NP0007 | 4x |
NuPAGE MES SDS Running Buffer | Thermo Fisher Scientific (Invitrogen) | NP0002 | 20x |
NuPAGE Sample reducing agent | Thermo Fisher Scientific (Invitrogen) | NP0004 | 10x |
SeeBlue Plus2 pre-stained protein standard | Thermo Fisher Scientific (Invitrogen) | LC5925 | |
Sequencing grade modified trypsin | Promega | V5111 | |
Sodium chloride (NaCl) | Carl Roth | 9265.2 | 118.3 mM |
Sodium hydrogen carbonate (NaHCO3) | Carl Roth | 965.3 | 25 mM |
Trifluoroacetic acid (TFA, C2HF3O2) | Merck Millipore | 108178 | |
α-(D)-(+)- Glucose monohydrate | Carl Roth | 6780.1 | 11 mM |
B. Reagents and Kits | |||
0.5mm zirconium oxide beads | Next Advance | ZROB05 | |
1.0mm zirconium oxide beads | Next Advance | ZROB10 | |
Colloidal Blue Staining Kit | Thermo Fisher Scientific (Invitrogen) | LC6025 | To stain 25 mini gels per kit |
NuPAGE 4-12 % Bis-Tri gels | Thermo Fisher Scientific (Invitrogen) | NP0321BOX | 1.0 mm, 10-well |
Pierce Bicinchoninic Acid (BCA) Protein Assay Kit | Thermo Fisher Scientific | 23227 | |
ProteoExtract Transmembrane Protein Extraction Kit, TM-PEK | Merck Millipore | 71772-3 | 20 reactions per kit |
Tissue Protein Extraction Reagent (T-PER) | Thermo Scientific | 78510 | |
C. Tools | |||
96-well V-bottom plates | Greiner Bio-One | 651180 | |
Corning 96-well flat-bottom plates | Sigma-Aldrich | CLS3595-50EA | |
Disposable microtome blades | pfm Medical | 207500014 | |
Disposable scalpels #21 | pfm Medical | 200130021 | |
Dissection pins | Carl Roth | PK47.1 | |
Extra Fine Bonn Scissors | Fine Science Tools | 14084-08 | |
Falcon conical centrifuge tubes (50 mL) | Fisher Scientific | 14-432-22 | |
Mayo scissors, Tough cut | Fine Science Tools | 14130-17 | |
Precision tweezers | Fine Science Tools | 11251-10 | Type 5 |
Precision tweezers, straight with extra fine tips | Carl Roth | LH53.1 | Type 5 |
Self-adhesive sealing films for microplates | Ratiolab (vWR) | RATI6018412 | |
Standard pattern forceps | Fine Science Tools | 11000-12 | |
Student Vannas spring scissors | Fine Science Tools | 91501-09 | |
Vannas capsulotomy scissors | Geuder | 19760 | Straight, 77 mm |
ZipTipC18 pipette tips | Merck Millipore | ZTC18S096 | |
D. Equipment and devices | |||
150 × 0.5 mm BioBasic C18 column | Thermo Scientific, Rockford, USA | 72105-150565 | |
30 × 0.5 mm BioBasic C18 pre-column | Thermo Scientific, Rockford, USA | 72105-030515 | |
Amicon Ultra-0.5 3K Centrifugal Filter Devices | Merck Millipore | UFC500396 | Pack of 96. |
Analytical balance | Sartorius | H51 | |
Autosampler | CTC Analytics AG, Zwingen, Switzerland | HTS Pal | |
BBY24M Bullet Blender Storm | Next Advance | NA-BB-25 | |
Eppendorf concentrator, model 5301 | Sigma-Aldrich | Z368172 | |
Eppendorf microcentrifuge, model 5424 | Fisher Scientific | 05-403-93 | Non-refrigerated |
Heraeus Primo R Centrifuge | Thermo Scientific | 75005440 | Refrigerated |
Labsonic M Ultrasonic homogenizer | Sartorius | BBI-8535027 | |
LC-MS pump, model Rheos Allegro | Thermo Scientific, Rockford, USA | 22080 | |
LTQ Orbitrap XL mass spectrometer | Thermo Scientific, Bremen, Germany | ||
Multiskan Ascent plate reader | Thermo Labsystems | v2.6 | |
Rotator with vortex | neoLab | 7-0045 | |
Titanium probe (Ø 0.5mm, 80mm long) | Sartorius | BBI-8535612 | |
Ultrasonic bath, type RK 31 | Bandelin | 329 | |
Xcell Surelock Mini Cell | Life Technologies | El0001 |