SOUTH SAN FRANCISCO, Calif. — Scientists at biotech grandaddy Genentech have used an electrochemiluminescent assay to detect surface proteins in live-suspension and adherent cells. They found the quantitative and high-throughput process to be easier and less time-consuming than flow cytomtery and enzyme-linked immunosorbent assays, a Genentech official said at a conference here last week.
The ECL assay can also be used to identify and rank neutralizing antibodies for drug development, or to compare the activity of different lots of antibodies for quality control purposes. “We think that it can also be used to identify small molecule inhibitors,” Gloria Meng, a senior scientist in automation and assay technology at Genentech and a co-author of the study, told CBA News in a follow-up e-mail this week.
The team developed the ECL assay using a CCD-based detection system and microwell plates with built in carbon surface electrodes from Meso Scale Discovery in Gaithersburg, Md. They wanted to evaluate how I-CAM expression can be inhibited in human umbilical vein endothelial cells stimulated by different anti-lymphotoxin antibodies, said Meng.
She presented the study findings at last week’s Third Annual Assay Development and Screening Technologies Conference.
The researchers found that the HUVECs can be grown, activated, and assayed in the same carbon surface plate, and Meng said the results may have “significant implications” for detecting surface protein expression on adherent cells.
Meng said that for suspension cells, her team found that the ECL assay method had a higher throughput than cell-based ELISA assays because the carbon surface plates bind cells tightly enough to allow plates to be washed. Thus, the centrifugation steps used in these assays were avoided.
The ECL assay method “has a much lower background than a corresponding cell-based ELISA,” and is much simpler than a corresponding flow cytometry assay, which requires cell detachment steps, said Meng.
In addition, because the ECL methodology relies on live cells, it sidesteps the need to fixate cells, which could alter surface proteins.
The research was published online in Current Opinion in Pharmacology last September.
State of Suspension
Meng and her colleagues performed four experiments to evaluate whether ECL can detect surface proteins in live suspension cells or detached adherent cells in suspension. To evaluate the use of ECL to gauge relative levels of surface proteins, the researchers detected CD20 expression on transfected Chinese hamster ovary cells and BJAB cells.
They plated the cells in a 96-well ECL assay plate and added serial dilutions of Rituxan followed by ruthenium-conjugated anti-human Fc. The scientists found that ECL counts increased in tandem with increased CD20 molecules per cell, suggesting that the ECL method can be used to estimate the relative amount of CD20 expression on cells.
In a second experiment, the investigators treated human CD20 transgenic mice with anti-CD20 or antiragweed IgG isotype control. The researchers drew blood three days later and removed the constituent erythrocytes.
“This assay has a much lower background than a corresponding cell-based ELISA, and is much simpler than a corresponding flow cytometry assay.”
The cells were then plated in 96-well ECL assay plates at 25,000 cells per well. Serially diluted biotinylated anti-B220 antibody was added followed by ruthenium-conjugated streptavidin to detect B-cells. The researchers observed “much higher” ECL counts in the ragweed control group, compared to those in the anti-CD20-treated group.
The researchers used flow cytometry to confirm that few peripheral blood B-cells remained in the anti-CD20-treated mice. These results indicate that ECL can be used to detect peripheral blood B-cell reduction following anti-CD20 treatment.
To determine if ECL can be used to detect CXCR4 expression on cells, the researchers added serially diluted biotinylated anti-CXCR4 antibody to transfected human embryonic kidney 239 cells expressing CXCR4. The cells were plated in a 96-well ECL assay plate at 25,000 cells per well.
Bound antibody was detected using ruthenium-conjugated streptavidin and a dose-response curve was obtained (ECL counts 144; 513; 1,941; and 3,178 at 0; 8; 222; and 2,000 ng/mL, respectively). These results indicate that ECL can be used to detect CXCR4 expression in cells to support the development of CXCR4-targeted drugs.
The authors detected expression of IL-2Rα and ICAM-1 on activated CD4 cells in their final experiment. The T cells were cultured for 40 hours in the presence either of phytohemagglutinin-L, anti-CD3, or anti-CD28. The cells were then harvested and plated in a 96-well ECL assay plate.
IL-2Rα and ICAM-1 expression was detected using biotinylated goat anti-IL-2Rα or mouse monoclonal anti-ICAM-1, followed by ruthenium-conjugated streptavidin. The researchers found that CD4 cells activated with 5 µg/mL PHA-L had sixfold higher signals versus non-activated cells for both IL-2Rα and ICAM-1 expression. CD4 cells treated with anti-CD3 and anti-CD28 had sevenfold and 12-fold higher signals for IL-2Rα and ICAM-1 expression, respectively.
These findings suggest that ECL can be used to detect changes in surface protein expression on activated CD4 cells. Flow cytometry on duplicate samples confirmed increased IL-2Rα and ICAM-1 expression on activated CD4 cells.
Meng then discussed the assay she and her colleagues developed to determine if adherent cells can be activated when immobilized on an ECL assay plate. HUVECs in a 96-well assay plate were stimulated with serially diluted TNFα, lymphotoxin α3, or lymphotoxin α1β2. After incubating overnight, ICAM-1 expression was detected using biotinylated anti-ICAM-1 antibody, followed by ruthenium-conjugated streptavidin.
Dose-response curves were seen in TNFα- and LTα3-stmulated cultures, and to a lesser extent in LTα1β2-stimulated cultures. Hepatocyte growth factor was used as a negative control.
Serially diluted TNFα, LTα3, or LTα1β2 coated directly on the ECL plate in the absence of HUVECs only gave background signals.
This confirms that the dose response curves observed were due to increased ICAM-1 expression and not cross-reactivity of the anti-ICAM antibody to serially diluted cytokines. The authors said that the assay has been reproducible for the past year.
A corresponding ELISA was performed in tissue culture plates using biotinylated anti-ICAM-1 followed by horseradish peroxidase-conjugated streptavidin to detect ICAM-1 expression on TNFα-stimulated HUVECs. This assay gave poor signal-to-noise ratio and high background.
The investigators used the induction of ICAM-1 expression assay that they developed to assess the neutralizing activity of anti-LTα3 antibodies. A panel of serially diluted anti-LTα3 antibodes was incubated with a fixed concentration (20 ng/mL) of LTα3 at room temperature for one hour before addition to HUVECs in a 96-well ECL assay plate.
The researchers found that most antibodies produced a dose-dependent inhibition of ICAM-1 expression. For comparison, some anti-LTα3 antibodies were also analyzed by flow cytometry. The ranking of the neutralizing activity of the LTα3 antibodies corroborated that determined using the ECL assay.
Although ECL is easier to perform than a corresponding cell-based ELISA for B cells and CD4 cells, other suspension cells still need to be evaluated, the authors said. In addition, more comparison studies are required to be certain that cells immobilized on the ECL plate carbon surface are activated in a manner comparable to that of cells in tissue culture plates.