Overview

AlphaLISA^® is a bead based assay technology that can be used for immunoassays in a microplate format. A key feature of this assay is that it is a homogeneous, or a no wash assay. This feature combined with its sensitivity and high throughput capabilities make Alpha an assay format choice for many research applications. As many researchers wish to apply this technology to samples containing serum, the below content will serve as a guide to optimizing this type of assay.

Some topics that will be considered include:

• Clinical and pre-clinical studies using serum or blood samples
• Hemolyzed samples and potential for sample interference
• Finding a suitable matrix for your standard curve

Both serum and plasma originate from whole blood. Plasma is richer than serum as it contains everything but cells, including fibrinogen and coagulation factors. Researchers may wish to harvest plasma and use agents such as heparin or chelators such as EDTA to prevent blood clotting. Both EDTA and heparin were shown not to affect the Alpha Technology. Serum is more stable and more uniform. Other biological fluids such as cerebrospinal fluid (CSF), bronchoaveolar lavage fluid (BALF), amniotic fluid (AF), saliva or urine are often used to measure biomarkers. Each of these fluids has a unique protein composition and will require specific validation. The goal of the document is to guide you through this specific validation approach.

Working with a complex sample such as serum or CSF may sometimes result in decreased assay sensitivity. However, in most cases, working with these biological matrices is successful if the appropriate procedures are followed.

Top

Matrix Importance

For quantitation of an analyte in your samples, it is necessary to create a standard curve of your analyte in a matrix that closely matches your samples. For example, if you are working with serum samples, you will need to create a standard curve of your analyte in analyte-depleted serum or something similar, such as fetal bovine serum (FBS). A spike-and-recovery experiment should be performed to assess whether a proposed matrix is suitable for your sample type. Table1 provides some examples of possible matrices.

There is a correlation between recovery and matrix. When the matrix for your standard matches the sample matrix, the percentage of recovery is better. Figures 1 & 2 illustrate matrix importance and percent recovery. It is important that the matrix should be as close to the sample as possible.

 

Sample in immunoassay buffer	Spiking conc. (pg/mL)	Detected (pg/mL)	% recovery
1	50,000	39,996	80
2	10,000	10,017	100
3	100	62	62
Sample in RPMI + 10% FBS	Spiking conc. (pg/mL)	Detected (pg/mL)	% recovery
4	50,000	20,016	40
5	10,000	4,847	48
6	100	not detectable

Figure 1: A hEGFR detection assay with a standard curve prepared with Immunoassay Buffer (IAB). Data show comparative values of percent recovery with two different sample matrices. As the standard curve was prepared in IAB, the percent recovery is better for the IAB samples, as compared to the sample prepared with RPMI + 10% FBS.

In Figure 2, the standard curve was instead run in RPMI + 10% FBS. Data show percent recoveries are better with spike-ins prepared in RPMI + 10% FBS.

 

Sample in immunoassay buffer	Spiking conc. (pg/mL)	Detected (pg/mL)	% recovery
1	50,000	74,943	150
2	10,000	15,331	153
3	100	161	161
Sample in RPMI + 10% FBS	Spiking conc. (pg/mL)	Detected (pg/mL)	% recovery
4	50,000	48,440	97
5	10,000	9,935	99
6	100	57	57

Figure 2. The standard curve prepared with RPMI + 10% FBS. Data show only the spike-in concentrations made with RPMI + 10% FBS showed acceptable recoveries; the ones prepared in IAB did not.

In a third experiment, a standard curve for human IL6 was performed in both IAB and FBS. The results are shown in Figure 3.

 
Figure 3. The human IL-6 standard curve was prepared and plotted in both immunoassay buffer and FBS. The curve in FBS shifted to the right, which indicates the reason for different percent recoveries in the different matrices.

There are several standard curve matrix options that can be used when working with serum samples in an Alpha format. These include:

• Using FBS with or without sample dilution
• Preparing analyte-depleted serum
• Buying or custom requesting analyte depleted serum (companies such as Scipac^® sell analyte-depleted serum)
• Using charcoal-stripped serum as a matrix for some analytes
• Pooled normal serum may be used (Petersen et al. 2010) only if it does not contain a detectable level of analyte being tested

Please refer to our application note regarding Analyte Depletion of Serum for AlphaLISA.

Top

Potential For Sample Interference

When working with serum or plasma, hemolysis is a factor to consider. Hemoglobin absorbs light in the same range as Alpha beads emit light, and can cause interference.

Figure 4 shows different wells with varying amounts of hemoglobin spiked in immunoassay buffer. Based on normal red blood cell (RBC) counts and hemoglobin content in a RBC, the spiked hemoglobin amounts were converted to the approximate hemolysis percentage of RBC. This is a visual representation comparing the amount of hemoglobin and the percent hemolysis in regards to well color.

Figure 4. Comparison between the amount of hemoglobin added as it correlates to the percentage of hemolysis with an AlphaLISA assay.

In order to better determine the effect of hemoglobin on an AlphaLISA assay, a sample was spiked with increasing concentrations of hemoglobin and was measured using the Human Amyloid β 1-40 Immunoassay Kit. The assay was performed as stated in the manual using 5 μl of sample, 20 μl AlphaLISA Acceptor beads/biotinylated antibody and 25 μl of Donor beads.

• The buffer was spiked with 16,000 pg/mL of Aβ 1-40 and increasing concentrations of hemoglobin
• The plates were read with an EnVision^® 2104 Multilabel plate reader HTS with Alpha option
• The interpolated values (as shown in Table 2) were determined from the Immunoassay Buffer (IAB) standard curve

For key AlphaLISA references where serum samples were used, please refer to the References and Citations section following.

Other components of a sample matrix can also potentially interfere with the assay. Interference may result from cross-reactivity of antibodies to a component in the serum sample, depletion of the spiked analyte, or variable levels of IgG, albumin or other proteins.

Serum often contains high concentrations of many components including:

• Hemoglobin
• Bilirubin
• Transferrin
• IgG and other immunoglobulins
• Triglycerides and other lipids
• Biotin and biotinylated proteins
• Binding proteins/Receptors
• Related immunogens detected by either one antibody or two antibodies

Reducing sample to sample variability is a critical step in optimizing an AlphaLISA assay using serum samples. There are several factors to examine when attempting to improve percent recovery for an assay. Some suggestions to improve percent recovery include but are not limited to:

• Diluting the serum samples 2X with FBS
• Removing IgG
• Reducing heterophilic interference by using commercial blockers

Figure 5. The good recoveries obtained for TNFα with a 1/2 diluted pooled serum as a matrix.

The above experiment was repeated using insulin instead of TNFα, as shown in Figure 6.

Figure 6. Good insulin recovery obtained using 1/2 diluted serum as a matrix.

As we have previously mentioned, serum is made up of many biological components that could potentially interfere with an assay. When using serum samples with the AlphaLISA kits, dilutions may need to be made based on kit parameters. In most cases, potential interference issues can be resolved by diluting the sample or changing the assay buffer.

Top

Identifying Diluents For Any Matrix

The following section describes how to identify a correct diluent for complex sample matrices based on examples from various animal fluids. There were three cytokines selected: IL6, TNFα, and MCP-1 (CCL2). The four sample matrices selected were mouse serum, mouse bronchial lavage fluid (BALF), rat cerebrospinal fluid (CSF) and rat amniotic fluid (AF). These results may also be found in the poster Detecting cytokines in animal fluids.

The assay was performed as indicated in the manual. The protocol is listed below along with the calibration curve for mouse MCP-1.

 
Protocol

To begin assay development using a new matrix, one must first identify an appropriate matrix (or "diluent") for the standards. Linearity experiments and spike-and-recovery experiments were performed to assess whether each proposed matrix is suitable for these sample types. A protocol example may be found in the Application Notes and Poster section below. In Figure 7, the linearity of dilutions in five diluents for spiked mouse BALF (3 ng/mL mTNFα) were examined.

 
Figure 7. Linearity of five different diluents was determined.

A good diluent will exhibit good recovery (values between 70% and 130%) and linearity (no change upon dilution). In this case, the two best choices are PBS + 0.1% BSA and PBS + 0.01% BSA as demonstrated in Table 3.

To decide which of these two diluents would work best, a spike-and-recovery experiment was performed. Shown in Table 4 are the spike-and-recovery results using PBS + 0.1% BSA as diluents.

Table 4: Excellent recovery was obtained for all four spike concentrations tested. The recovery percentage is equal to the recovery compared to the spiked diluent control.

For certain sample types, dilutions may be required to achieve good linearity results. In another set of experiments, five diluents were examined for linearity when spiked with 3 ng/mL of mMCP-1 rat AF. In the below chart, the rat amniotic fluid required further sample dilution in order to achieve good linearity. This sample differs from the mouse BALF as shown in Figure 7 which did not require a 2x dilution for good results. It may be necessary to use dilutions to make the samples comparable to the sample being tested.

 
Figure 8. Linearity of five different diluents was determined.

Good recovery (values between 70% and 130%) and linearity (no change upon dilution) are expected once the correct dilution has been determined. The Table 5 summarizes the recovery results obtained.

Not a single diluent gave acceptable results from the neat sample, but upon a 1:2 dilution, the two diluents with the best results are HiBlock buffer and PBS with 0.1% BSA, as highlighted by stable recoveries upon dilution and values close to 100%. The diluent showing the best spike-and-recovery was further examined. The concentration for 300-3000 pg/mL spikes is equal to the measured concentration minus the no spike value. The percent recovery is calculated by dividing the value in the sample by the value obtained in the spiked diluent control.

In summary, three cytokines were examined using various sample fluids and diluents. As seen in the below table, lower detection limits were determined along with dilution, spike range and physiological level measured.

Previously published data comparing different serum matrices is included in the table below. Specifically, the physiological level of serum as shown in the above AlphaLISA data in Figure 16 has been compared with the physiological range as reported in the literature. For most sample types, the AlphaLISA results match well with the physiological ranges reported in the literature.

Top

Summary

While assay interference is not a general issue with AlphaLISA technology, there is potential that components in complex biological matrices may impact the measured values. The above information is meant to serve as an assay development guide for using Alpha technology with such complex matrices. The assay may require some optimization, including verification that the control matrix is as similar to the sample as possible, selecting appropriate diluent buffer, as well as optimizing the spiking concentration for accurate recovery determinations.

The key points to note when working with serum and other biological matrices in an AlphaLISA assay are as follows.

• Components of a sample matrix may potentially interfere with an assay.
• The standard curve of your analyte should be created in a matrix that closely matches the samples.
• Linearity and spike-and-recovery experiments should be performed to determine if a suggested matrix is suitable for the samples.
• Sample dilutions may be required in some cases to achieve good results.

Top

Application notes, posters, and protocols

• Protocols for linearity & spike-and-recovery experiments
• Detecting cytokines in animal fluids
• Analyte depletion of serum for quantitative AlphaLISA assays
• Detection of inflammation biomarkers in donor serum from asthmatic patients
• Detection of TNFR1 in cultured cells, PBMCs, disease-state human serum, and disease-state human urine

Top

References

Poulsen F and Jensen KB. A Luminescent Oxygen Channeling Immunoassay For The Determination Of Insulin In Human Plasma. J. Biomol. Screen. 12(2), 2007. Link

Petersen, SB et al. Comparison Of Luminescent Oxygen Channeling Immunoassay And ELISA For Detecting Insulin Aspart In Human Serum. J. Pharmaceutical and Biomedical Analysis 51 (2010). Link

Top

Custom conjugation and custom assay development at 爱游戏平台注册登录

爱游戏平台注册登录 offers custom bead conjugation services as well as custom assay development. If you are interested in having your biomolecule custom-conjugated to a bead, or in custom assay development, please contact our custom teams:

ON>POINT^® Custom Labeling and Conjugation Services
ON>POINT^® Custom Assay Development Services