Biomedical Engineering Reference
In-Depth Information
cell type, in response to a specific cytokine. For example, one signaling node was the
Stat1 response in B cells to stimulation with IFN
a
. We found that single signaling
nodes in isolation were imperfect predictors of flare. However, when we used an
algorithm that looked at each pair of nodes in the data and determined whether they
accurately distinguished patients who would experience a flare within 90 days,
we found several pairs of nodes that appeared highly accurate at predicting flare.
Importantly, the classifiers most predictive of flare paired a decrease in a pro-tolerance
response in one leukocyte subset with a pro-inflammatory or pro-proliferative
response in another leukocyte population. The relative activation of the nodes making
up each two-dimensional classifier can be quantified as a signaling ratio. For many of
the classifiers, patients who stayed in remission had a positive signaling ratio such
that the protolerance signaling response balanced the proinflammatory signaling
response. The patients who went on to experience flare within 90 days had a signaling
ratio significantly below a value of 1. Inverted signaling ratios that predicted flare
were termed pathogenic signaling ratios (Figure 15.3b). The observed balance of
cell type-specific signaling highlights the importance of designing drugs, such as
kinase inhibitors, that are not only pathway specific but also act on the appropriate
cell type(s).
15.4 HIGH-THROUGHPUT SCREENING
Flow cytometry is not often associated with high-throughput screening of large
(
100,000 molecule) compound libraries. Instead, along with microscopy platforms,
flow cytometry is typically used for “high-content” screening applications. In high-
content screening, many parameters are measured simultaneously to obtain a richer
understanding of a drug candidate's effects on the cell as a whole, not just on the
particular target molecule. However, recent advances in both assays and instrumenta-
tion are enabling the flow cytometer to simultaneously be used as a high-throughput
and a high-content platform. To improve throughput, we recently developed a
technique called fluorescent cell barcoding or FCB, which enables one to combine
many different samples together into one tube prior to adding antibodies or other
measurement reagents [10]. This dramatically reduces reagent consumption and
improves data robustness. On the hardware side, an autosampler has been developed
that enables data acquisition from 384-well plates in approximately 15min. The
combination of these two advances will push flow cytometry-based high-content,
multiparameter analyses to staggering throughput rates.
>
15.4.1 Fluorescent Cell Barcoding
One of the major limitations of many high-content analysis platforms is the cost
associated with the reagents that are used to make the measurements. While the
typical high-throughput analysis uses fluorescent dyes or genetically encoded reporter
constructs that are extremely inexpensive on a per well basis, high-content assays
typically use antibodies for staining of cellular antigens. Antibody costs can be
Search WWH ::
Custom Search