SLAS Discovery - Special Issue on Advances in MALDI Mass Spectrometry for Drug Discovery - Volume 22 Issue 10 - December 2017 - 1216

1216
1×106
8×105

B4, 302 Da
HTRF IC50
1800 nM

6×105
4×105

-BACE

+ BACE

-BACE

+ BACE

-BACE

2×105
0

B

B5, 404 Da
HTRF IC50
14 nM

B1, 374 Da
HTRF IC50
6 nM

+ BACE

A
CompoundAUC

larger AUC compared with the weaker binding compounds.
The format detected potent inhibitors in the hundreds of nM
range with 25-fold signal over the no-protein control
samples.
Similar to other affinity capture methodologies, saturating the capture surface with target protein on SAMDI (typically in the range of 500 nM to single-digit micromolar)
will maximize the detection sensitivity and uniformity. The
need for high protein concentration is balanced by the small
assay volume and even smaller volume for the actual detection on SAMDI (as low as 300 nL). To understand the sensitivity limit of the SAMDI small-molecule binding assay
in detecting weak binders from a mixture, 650 nM biotinylated DLK and a chemically diverse set of compounds were
used to mimic conditions of a screen against a randomized
and unbiased library.23 We pooled three ATP-competitive
DLK inhibitors, D01, D02, and D03, that ionize efficiently
using straight MALDI and had weak biochemical IC50 values of 2 µM, 5 µM, and 10 µM, respectively, in a total of 20
inactive compounds with IC50 values all >25 µM. The concentration of each compound in the pool was 5 µM.
Compound D01 (DLK-HTRF IC50 = 2 µM) revealed a
15-fold signal increase in samples with protein compared
with samples without protein and was easily identified as a
positive hit from the pool of 20 compounds (Fig. 3B).
However, the two weaker DLK compounds with HTRF
IC50 = 5 µM and 10 µM did not show significant signal in
the samples with DLK compared with samples without
DLK. This result was consistent with the hypothesis that the
DLK protein was fully occupied by the most potent inhibitor in the pool, given that the weaker compounds present
were known to bind to the same site of the protein and were
not detected as binders. Although these data demonstrated
that hits with single-digit micromolar binding potency were
readily identified in this binding assay format from a compound pool, thoughtful planning of the assay conditions that
consider the expected hit rate and anticipated potencies of
hits would minimize false-negatives in a screen due to competitive binding of multiple compounds in a pool.
Compound binders with single-digit micromolar affinity
are common starting points for new chemical matter identified from HTS. Experimental parameters to test prior to a
screen that may maximize detection of weaker binding
compounds, especially in a compound mixture, include (1)
incubating the plates at 4 °C for a longer time period rather
than at room temperature to preserve weak complexes, (2)
increasing the screening compound concentration if the
anticipated hit rate and potency are low, and (3) decreasing
the compound pool size to reduce the competition for available binding sites for low-affinity molecules to the protein
targets.
In summary, affinity capture of small-molecule binders using SAMDI potentially applies to a wide variety of
protein target classes, especially those with limited

SLAS Discovery 22(10)

D01
DLK HTRF IC50 = 2.37uM
SAMDI fold signal = 14.3
Compound MW = 501 Da

D02
DLK HTRF IC50 = 5.14uM
SAMDI fold signal = 1.14
Compound MW = 468 Da
D03
DLK HTRF IC50 = 10.8uM
SAMDI fold signal = 0.2
Compound MW = 290 Da

Figure 3. Self-Assembled Monolayers and matrix-assisted laser
Desorption Ionization (SAMDI) detection of compound binders
from pooled compounds. (A) SAMDI captures 1 µM BACE
incubated with a pool of three compounds, each at 10 µM. The
most potent compound shows a 25-fold signal in the SAMDI
binding format. (B) SAMDI captures 650 nM Dual Leucine
Zipper Kinase incubated with a pool of 20 compounds, each at 5
µM. A fivefold window is defined for distinguishing hits between
the samples with and without protein. The compound identified
as a binding hit, D01, has a 15-fold signal in the SAMDI binding
format and HTRF IC50 = 2 µM.

reagent options or if the screen aims to capture hits with
diverse binding epitopes on the target protein. In this
study, we demonstrated proof of concept that the SAMDI
platform can identify binders with affinities in the micromolar range and below, the output was robust and reproducible, and data correlated with biochemical activity for
two different protein targets. Given the ability for MS to
accurately identify individual hits from a mixture based
on the mass identity, screening compound libraries in a
pooled mixture using SAMDI can further increase the
throughput. Because not all compounds ionize well and
Table of Contents for the Digital Edition of SLAS Discovery - Special Issue on Advances in MALDI Mass Spectrometry for Drug Discovery - Volume 22 Issue 10 - December 2017
