Blood-based biomarkers for neurodegenerative diseases are essential for implementing large-scale clinical studies. A reliable and validated blood test should require a small sample volume as well as be a less invasive sampling method, affordable, and reproducible. This paper demonstrates that high-throughput capillary electrophoresis immunoassay satisfies criteria for potential biomarker development.
Capillary electrophoresis immunoassay (CEI), also known as capillary western technology, is becoming a method of choice for screening disease relevant proteins and drugs in clinical trials. Reproducibility, sensitivity, small sample volume requirement, multiplexing antibodies for multiple protein labeling in the same sample, automated high-throughput ability to analyze up to 24 individual samples, and short time requirement make CEI advantageous over the classical western blot immunoassay. There are some limitations of this method, such as the inability to utilize a gradient gel (4%–20%) matrix, high background with unrefined biological samples, and commercial unavailability of individual reagents. This paper describes an efficient method for running CEI in a multiple assay setting, optimizing protein concentration and primary antibody titration in one assay plate, and providing user-friendly templates for sample preparation. Also described are methods for measuring pan TDP-43 and phosphorylated TDP-43 derivative in platelet lysate cytosol as part of the initiative in blood-based biomarker development for neurodegenerative diseases.
The overall goal of CEI as described here is to provide an updated stepwise protocol for analyzing target proteins in human platelets. Assignment of a blood-based signature molecule is one of the most important tasks in the field of biomarker development in human neurodegenerative diseases, such as Alzheimer's disease (AD), amyotrophic lateral sclerosis (ALS), frontotemporal lobar degeneration (FTLD), Parkinson's disease (PD), inclusion body myositis (IBM), and other protein-aggregation relevant pathologic conditions. The detection of minute amounts of such signature proteins in large volumes of blood with many interfering agents is a challenge. Therefore, the specificity, sensitivity, ability to handle large number of samples, and reproducibility of the selected method are crucial.
Human platelets can serve as a milieu to identify and assign potential biomarker proteins for neurodegenerative disease. Platelets provide the opportunity to serve as a surrogate primary cell model, which reflect some features of neuronal cells1,2,3. There are certain features that make platelets one of the preferred means to analyze biomarker candidate proteins and their chemical derivatives. First, platelets can be easily acquired using a less invasive approach by collecting blood from donors (i.e., venipuncture) or in large volumes from community blood banks. Second, platelets can be easily isolated from the whole blood with minimal preparatory work in minimally equipped laboratories4,5. Third, platelets do not have nuclei; therefore, they are a good model cell to study alterations in metabolism without transcriptional regulation. Fourth, the biomolecule content of platelets is encapsulated; therefore, the platelet microenvironment protects its contents from serum-interfering substances (i.e., proteases). Fifth, platelet-enriched plasma can be stored at room temperature for 7–8 days without losing metabolic activity. Therefore, platelets provide a working model in which external factors are minimized and controlled.
Traditional immunoassay techniques such as immunoblotting (e.g., western blotting) and enzyme-linked immunosorbent assay (ELISA) are more widely used in specific protein analysis. However, these two methods have several disadvantages, including multiple assay steps, requirement of hazardous chemicals and reagents, large sample size, issues with assay reproducibility, and inter-run data variabilities. These prompted the development of a method that is simpler with fewer steps and achievable in a relatively short period. Although the classical western blot technique will remain a popular laboratory method, its multi-step procedure, supplies, toxic waste (i.e., acrylamide, methanol, etc.) and assay time are becoming less desirable when performing high-throughput quantitative protein analysis.
An automated CEI approach is gradually becoming a method of choice for laboratories that conduct high-throughput protein assays6. CEI eliminates the need for gels, gel electrophoresis apparatuses, membranes, electrophoresis and electro-transfer devices, and more physical handling involvements. If designed well, a CEI assay should be completed within approximately 3.5 h, including quantitative data analysis, publication quality electropherogram, and graphs with statistical analysis. Another superiority of the CEI system is its requirement of 10x–20x less protein concentration, making it ideal for use in human samples used in clinical trials7,8.
The most critical part of CEI is optimizing the assay conditions for each antibody purchased from different vendors, type of antibody (monoclonal vs. polyclonal), optimum protein concentrations, sample preparation, sample denaturation temperature, and electrophoresis voltage applied on the capillaries. We have developed a single-assay format optimization method for the CEI that should be implemented before any new assays, which will save time and resources. This optimization step is followed by an automated quantitative assessment of both total and phosphorylated derivative of transactivation response DNA/RNA binding protein (TARDP). Due to its size (43 kDa), the acronym TDP-43 will be used throughout this paper. Here, TDP-43 protein in human platelet lysate obtained from ALS patients are assessed to help develop predictive phosphorylation value (PPV) as a potential prognostic biomarker.
TDP-43 is a new potential disease biomarker candidate for ALS. TDP-43 is an omnipresent protein in all nucleated cells; therefore, the functions of TDP-43 during various normal cellular events and in neurodegenerative disease have been investigated9,10,11,12,13,14. Although TDP-43 is a nuclear protein15, it has the ability to shuttle in and out between the nucleus and cytoplasm due to the presence of nuclear localization and nuclear export sequences16,17,18,19. Cytoplasmic TDP-43 is involved in various cellular events, such as mRNA stability and transport, the stress response, mitochondrial function, autophagosome20. However, not much is known about the role of phosphorylated derivatives of TDP-43 other than their involvement in the pathogenesis of neurodegenerative disease21.
This protocol illustrates how to optimize the assay conditions to analyze the contents of TDP-43 and its phosphorylated derivative in platelets using the CEI approach. Since phosphorylated TDP-43 is not commercially available, it is proposed to use a predictive phosphorylation value (PPV) to assess TDP-43 profiles in ALS patients. This CEI system utilizes a small volume of sample mixture (2.5–3.0 µL per capillary). Total assay volume set-up is 8.0 µL per capillary based on the manufacturer's protocol; hence, researchers can utilize one sample mixture preparation for two separate runs. The manufacturer designed the assay protocol so that any pipetting errors are minimized, if not entirely eliminated. The 24 individual human platelet lysate sample mixtures are divided into half-volumes (i.e., 2.5–3.0 µL per sample) and consecutively analyzed those by two different antibodies within ~7 h. The CEI system described here provides a desirable high-throughput assay modality. Users need to test antibodies from different vendors and sample preparation modalities for the target protein before performing large-scale screening.
All protocols concerning the processing of human platelets follow the guidelines of both the University of Kansas Medical Center and Kansas City University of Medicine and Biosciences IRB committees.
1. Preparation of buffers and reagents
NOTE: Prepare all samples as per the manufacturer's guidelines. Wear personal protection equipment (lab coats, gloves, and goggles) during this procedure.
2. Platelet isolation
3. Preparation for CEI
NOTE: 100 µL of human platelet lysate was combined from ALS patients (n = 8–10), and healthy subjects (n = 8–10) were separately pooled and used for the assay optimization.
Figure 1: Assay layout. Both primary antibody and target protein sample optimization can be performed in one assay. Capillaries 2–7, 8–13, 14–19, and 20–24 represent various protein concentration and primary antibody range. Capillary 25 represents positive control. Anti-ERK antibody was used; however, any appropriate positive control can be included. Please click here to view a larger version of this figure.
4. Performing the CEI on plate 1
5. Performing the CEI on plate 2
NOTE: This plate is set for analyzing phosphorylated TDP-43 levels.
6. Data analysis
Optimization of platelet cytosolic protein concentration and primary antibody titration
It is important to establish a linear dynamic range of platelet cytosolic proteins in the assay, since changes in signal are directly proportional to changes in protein in the platelet cytosol. Use of whole platelet lysate mixture in the assay may reduce the signal intensity of the target proteins (TDP-43 and P(S409-412) TDP-43) and contribute to a high background signal. Therefore, in this assay, the clear supernatant was used (cytosolic fraction) after rupturing platelets (Figure 2).
Figure 2: Signal clarity depends on the sample quality. (A) Whole platelet lysate homogenate interferes with anti-TDP-43 antibody binding; therefore, a noisy electropherogram was observed. (B) Platelet cytosolic fraction was obtained from subjecting the whole lysate to centrifugation (16,000 x g for 30 min). Most of the membranous proteins were removed; hence, anti-TDP-43 antibody binding to TDP-43 protein was enhanced (blue line trace). Please click here to view a larger version of this figure.
A linear dynamic range for platelet cytosol protein concentration was established at 0.2–0.8 mg/mL. An assay template was adopted so that both protein concentration and primary antibody titration were able to be performed in one assay (Figure 3).
Figure 3: Linear dynamic range for platelet cytosol protein concentration. (A) Both protein concentrations and antibody titrations were optimized in one plate during the same run. (B) The linear working range (0.2–0.8 mg/mL) for protein was established. A 0.5 mg/mL protein load was labeled by a-ERK antibody as the positive control (capillary 25). Please click here to view a larger version of this figure.
It should be noted that glycerol content in the sample preparation tube should be less than 20% (final), otherwise the high glycerol concentration will adversely affect the primary antibody binding.
Determining optimum exposure time
In the older version of the software, the optimum exposure time had to be determined by plotting peak area against protein concentration (mg/mL). The new version provides a new tool named the high dynamic range (HDR) detection profile (Figure 4). Using the image panel provided the option to view all exposure times (i.e., 5 s, 15 s, 30 s, 60 s, 120 s, 240 s, 480 s) together. Computer software analyzed all exposure times and automatically identified the best exposure time (HDR). HDR detection profile delivered a significantly wider dynamic range due to the greater sensitivity of CEI, which means better detection and quantitation over a larger sample concentration range. However, users still have the option to choose any exposure time that satisfies the experimental goal. Using this feature, optimum exposure time was found for TDP-43 protein. The peak represents the optimum exposure time (Figure 4A). A single exposure time (4 s) was defined for this antibody after reviewing all nine exposure times ranging between 1–512 s (Figure 4B).
Figure 4: High dynamic range (HDR) detection profile for a target protein. (A) TDP-43 protein peak represents the optimum exposure time for the target protein. a-TDP-43 Ab titration was 1:300, and platelet cytosol protein concentration was 0.5 mg/mL. The software-defined blue line indicates the optimum exposure time. (B) This figure represents a user-defined single exposure time (4 s) after reviewing all nine exposure times ranging between 1–512 s. Please click here to view a larger version of this figure.
TDP-43 levels in human platelet cytosol of ALS patients
A blood base biomarker development was performed. Using optimized assay conditions, platelet lysate cytosolic fractions obtained from ALS patients were analyzed using two sets of antibodies (i.e., anti-TDP-43 [Pan] antibody, an antibody that recognizes phosphorylated derivatives of TDP-43 protein; here, a-P [S409/410-2] TDP-43 was used). In this demonstration, disease-specific TDP-43 and its phosphorylated derivative are presented (Figure 5).
Figure 5: A representation of predictive phosphorylation value (PPV) of TDP-43. Absolute amount of phosphorylated TDP-43 and pan TDP-43 alone did not show much difference between the ALS and control groups. However, PPV indicated a slight increase in the ALS cohort, although there was no statistical difference between the two groups due to insufficient numbers of subjects (ALS = 25, control = 27). A low Cohen's d value between the means of ALS and control group showed a low effect size between the two groups due to small sample size (control = 25, ALS = 27). Please click here to view a larger version of this figure.
Total TDP-43 was quantified using the calibration curve (Figure 6)
Figure 6: Standard calibration curve. Commercially purchased recombinant TDP-43 protein concentrations were used to construct a standard curve. Each data represents the average of triplicates. The protein band intensities were concentration dependent (Inset). Please click here to view a larger version of this figure.
The quantification of phosphorylated TDP-43 protein was not possible due to commercially unavailability of this protein. Instead, a predicted phosphorylation value (PPV) was established that defines the percent of the phosphorylated species of TDP-43. PPV was determined from two sequential CEI assays for the same sample using the following equation.
Intra- and inter-run assay variability were tested in pooled human ALS platelet cytosolic fractions (Figure 7)
Figure 7: Assay variations. (A) Two capillaries loaded with the same sample were analyzed by CEI, and intra-run assay variation was calculated as CV% = 14.9. (B) The same sample was analyzed on three different assay days and in three different assay runs. Inter-run assay variation was calculated as CV% = 18.7. Please click here to view a larger version of this figure.
Although intra-run (capillary-to-capillary variation) coefficient variation values fell in the acceptable range (CV% = 14.9), the inter-run assay value was relatively high (CV% = 18.7). It is interpreted that this variation may be due to using capillary cartridges and sample plates from different lots. It is recommended that reproducibility studies should be performed in CEI components that have the same lot number.
Table 1: CEI assay plate loading template. Please click here to download this table.
Table 2: Interactive sample mixture preparation template. After entering the stock protein concentration from unknown samples, interactive cells will automatically calculate how much volume needs to be used for preparing the sample mix. Please click here to download this table.
The capillary electrophoretic-based immunoassay is now the method of choice for high-throughput sample analysis and drug screenings25. Small sample volumes, well-optimized assay components, user-friendly assay platform and instrumentation, reagent expenditure, and low CV percentage are primary advantages26,27. Although there are several methods for separating proteins in different assay modalities, the antibody-based CEI described here can be adapted by small laboratories that are engaged in blood-based biomarker development. The CEI assay technology used here provides reliable, reproducible, and sensitive measurements for TDP-4328 and its phosphorylated derivative5.
The CEI system also provides a multiplexing choice of analyzing TDP-43 and its phosphorylated derivatives simultaneously and providing the direct quantification of the target protein, if purified or recombinant target protein is available. Full-length recombinant TDP-43 protein is commercially available; however, recombinant phosphorylated TDP-43 derivative is not. Since phosphorylated TDP-43 is not commercially available, a predictive phosphorylation value (PPV) was implemented to assess TDP-43 profile in ALS patients. Pan TDP-43 and phosphorylated TDP-43 amounts were permanently labelled with a fluorophore; therefore, the TDP-43 profile remains the same with or without a quantitative unit (i.e., ng/mL, pg/mL, etc.). Although determining the absolute amount of TDP-43 and its phosphorylated derivatives (i.e., P [S409-410-12] TDP-43) provides a more quantitative measurement, calculating PPV eliminates the necessity of the recombinant phosphorylated TDP-43 for standardization, since it is not commercially available.
CEI provides several checkpoints in the assay platform to accurately identify the problem in the case that an assay fails. This eliminates obstacles and provides better experimental design. The assay procedure is fully automated except for filling the sample plate. This is a significant feature compared to standard western blotting analysis. This feature provides consistency from run-to-run. Although every laboratory has unique standard operating procedures, it is important to adhere to practices that minimize human error. For example, it is critical to prepare the luminol-S/peroxide mixture just before plate loading, since adding peroxide into luminol starts the enzymatic reaction and consumes the luminol substrate. Loading samples and primary/secondary antibodies into plate wells without air bubbles are also critically important steps.
Additionally, since the plate wells are small in volume and there is no space between wells, users should use caution while pipetting, which is the most important step since everything else is automated. The loading order of the samples, antibodies, and other reagents is important for consistency of the assay (Figure 1). The process of plate preparation takes about 40–45 min. Therefore, it is recommended to first load the plate with the required assay components and prepare the luminol-S/peroxide mixture just before pipetting. This way, there is a consistent sequence of reagent adding, and consistent luminescence signal strength will be attained. It is not recommended to use an expired luminol-S/peroxide reagent, as it primarily affects the strength of the peroxide. Recent progress in introducing the split-buffer system and including the chemical and detergent compatibility range has enhanced the assay quality and produced more reproducible and predictable results. Now, a new combo-analyzer from the same manufacturer possesses a feature for analyzing the samples labeled with chemiluminescence and fluorescence conjugated antibodies in the same run. This new feature eliminates the need to consecutively run two individual plates and eliminates run-to-run variation.
The assay plates should be stored in ambient temperature. If it is chosen to keep the assay plates in a 4 ˚C refrigerator, the plates must be taken out the night before the assay and brought to ambient temperature. Incorrectly loaded sample wells need to be extensively (4–5 times) washed with buffer provided in kit before adding the correct sample. Each primary antibody and biological samples are unique; therefore, the antibody/protein optimization should be performed before analyzing the samples for target proteins in biological fluids.
Here, the primary antibody incubation time was set for 30 min by default. If the signal is weak, users should consider increasing the primary antibody incubation time until reaching the desired signal strength without fluorescence signal burnout. For human platelets, pooled samples from patients were prepared and used for an optimization assay. The sample pooling better represents the variation among target biomolecules. It is recommended to use clear supernatants rather than total lysate or total homogenate for the CEI.
The high concentration of protein in whole platelet lysate mixture may decrease the signal-to-noise ratio (Figure 2). Repeated freeze-thaw cycles of samples should be avoided, as this adversely affects primary antibody binding. The ingredients of the lysate buffer are important, as some reagents are not compatible with CEI29. It is advised to cross-check the list of compatible reagents provided on the manufacturer's website before sample preparation. This is a limitation of the system that does not tolerate high stringency conditions for sample preparation. It is recommended to optimize the assay run parameters (i.e., primary antibody dilution, protein concentration, primary antibody incubation time, etc.) using pooled samples to subsequently analyze the individual samples.
The authors have nothing to disclose.
This research was sponsored by an intramural grant awarded for A.A. This work was supported by a CTSA grant from NCATS awarded to the University of Kansas Medical Center for Frontiers: The Heartland Institute for Clinical and Translational Research (#UL1TR606381). The contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH or NCATS. We are thankful for The University of Kansas Medical Center ALS clinic personal for obtaining IRB approval for blood sample collection from healthy volunteer and ALS patients. The authors thank Emre Agbas for proofreading of the manuscript.
12-230 kDa Separation kit | ProteinSimple | SM-W004 | Contains pre-filled assay plate and 25-channel capillary cartridge |
3000G Thermocycler | Techne | FTC3G/02 | We used this thermocylcer for heating the sample mix |
Anti-Mouse detection kit | ProteinSimple | 042-205 | Includes HRP-conjugated secondary antibody, buffer, luminol reagent, molecular weight marker |
Anti-P(S409-410) TDP-43 antibody | ProteinTech | 22309-1-AP | Primary antibody that recognizes phosphorylated TDP-43 |
Anti-P(S409-412) TDP-43 antibody | CosmoBio-USA | TIP-PTD-P02 | Primary antibody that recognizes phosphorylated TDP-43 |
Anti-Rabbit detection kit | ProteinSimple | DM-001 | Includes HRP-conjugated secondary antibody, buffer, luminol reagent, molecular weight marker |
Anti-TDP-43 (pan) antibody | ProteinTech | 10782-2-AP | Primary antibody that recognizes whole TDP-43 protein |
Compass for SimpleWestern (SW) | ProteinSimple | Ver.4.0.0. | Compass for SW is the control and data analysis application for SimpleWestern instruments |
Sonic Dismembrator; Model100 | Fisher Scientific | Sonicator. Used to rupture the cell membrane. This model is discontinued (Model XL2000-350) | |
Table top centrifuge | Eppendorf | 22625004 | Model# 5810 with swinging plate bucket |
Wes analyzer | ProteinSimple | 55892-WS-2203 | Performs the capillary gel electrophoresis |