Design of a novel screen for functional detection antibodies for lysate microarrays. A, General strategy for identifying detection antibodies, combining a primary high-throughput screen of candidate detection reagents with a secondary gold-standard assay. Validation of antibodies depends on concordance between the measurements from both techniques. B, Schematic of microarray screen. Lysates from many different biological contexts are arrayed onto nitrocellulose pads, assembled into microtiter plates, and probed with each of a large collection of primary antibodies. Microarray “hits” are assigned based on the statistical significance of the maximum observed signal spread, ΔI, across each lysate set. C, Breakdown of antibodies included in the screen by target epitope class and host species.

Identification of microarray hits based on observed assay variability. A, Sample of ten representative lysate microarrays generated in our screen. Each array was probed with a different primary antibody (red arrays), pooled with a β-actin-detecting antibody of a different species (green arrays) to normalize for protein concentration. Top row: five antibodies detecting post-translational modifications. Bottom row: five antibodies detecting total levels of target proteins (pan-specific antibodies). B, Distribution of biological noise in the screen. The vertical green line marks the threshold for statistical significance of signal spread at a significance level of α = 0.01, corrected for multiple testing. C, Plot of rank-ordered, normalized signal spread, ΔI, for all antibody-context pairs. Positive values indicate primarily up-regulation of antigen levels and negative values indicate primarily down-regulation. Red: significant down-regulation of signal. Blue: statistically insignificant change in signal. Green: significant up-regulation of signal. Insets show two examples of signal up- and down-regulation across time course treatments.

Secondary validation by quantitative immunoblotting reveals key determinants of antibody performance. A, Representative Western blot, showing moderate cross-reactivity. Lysates of HeLa cells treated with anisomycin, simultaneously probed with an antiphospho-ATF-2 (T71) antibody (red) and an anti-β-actin antibody (green). β-actin intensity was used to normalize signals for differences in gel loading. B, Lysate microarray images using the same lysates and antibodies as in (A). C, Scatter plot of signal intensities from (A) and (B) reveals degree of correlation between microarray and Western blotting data. D, Determinants of antibody reactivity, derived from an exhaustive Western blotting validation of microarray hits across four time course treatments (A431 + EGF, HT-29 + insulin, HeLa + anisomycin, and HeLa + TNFα). Blue: mouse antibodies, pan-specific antibodies, and pTyr-detecting antibodies generally exhibited lower validation rates. Green: rabbit, PTM-specific, and pSer/pThr-specific antibodies generally showed higher rates of validation. E, Determinants of antibody reactivity, derived from Western blotting validation of a subset of microarray hits across all time course treatments. Results were similar to (D). F, Validation rate of pan-specific antibodies across untreated cell lines. G, Prior validation of an antibody in one or more biological contexts increases the probability that it will perform well in additional biological contexts.

Microarray reactivity is antibody-dependent, but not batch- or antigen-dependent. A, Comparison of microarray data using different batches of the same antibodies. Top: Venn diagram showing substantial overlap between microarray hits. Bottom: Scatter plot of signal spreads (ΔI) across all time course microarray data, showing high degree of correlation. Data are represented on a double-logarithmic scale for ease of visualization. B, Comparison of microarray data from different antibodies directed at the same antigens, showing minimal overlap of hits and low degree of correlation. Top and bottom as in (A).

Self-organizing map analysis reveals three primary classes of time courses. Shown in the center is a visual representation of the average distances between neighboring map units in a self-organizing map of the time course profiles (U-matrix). Dark blue regions represent clusters of similar dynamic profiles, and regions of early, sustained, and late signaling dynamics (clusters 1, 2, and 3) are highlighted. Early and sustained signals are connected by a transition region of moderately clustered time courses (blue). Shown in panels on the sides are the identities, temporal profiles (dashed lines) and average temporal profiles (solid line) of time courses within each cluster.