Chemistry Reference
In-Depth Information
The mass spectrometry for this analysis was performed on an APEX III 4.7 T FTICR
mass spectrometer (Bruker Daltonics, Billerica, MA, USA) fitted with anApollo ESI source
operated in the negative ion mode. Broadband excitation was used to analyze a mass
range from
m
/
z
100 to 4500, with instrument parameters tuned and optimized for detecting
m
/
z
3000. DCL samples were directly infused into the ESI source at 2 L min
−
1
with
an ESI source pressure of 6.2
∼
×
10
−
7
mbar, a high-vacuum analyzer region pressure of
1.3
10
−
10
mbar and a hexapole ion accumulation time of 1 s. ESI-FTMS analysis of a
solution containing only CA II (
×
29 kDa) yielded the ESI negative ion mass spectrum in
Figure 7.13a. Peaks corresponding to the 8
−
to 10
−
charge states of CA II were observed,
with the 9
−
charge state predominating. This charge state envelope (low charge states and
fewcharge states) is typical for CAII when in a compact, tightly folded native structure.
[
50, 57
]
The mass spectrum of the CA II-DCL solution (prepared from
1
,
2
and
A-E
) is presented in
Figure 7.13b. The same charge state envelope as for free CAII (Figure 7.13a) was observed,
but each charge state now consisted of a grouping of peaks: a peak that corresponded to
native CA II and at higher
m
/
z
value a group of peaks that corresponded to the five different
CA II-hydrazone noncovalent complexes CA II-
1A
...
1E
in addition to a small amount
of CA II-
1
.
Owing to overlapping isotopic envelopes, a consequence of the molecular mass of CA
and the broadband detection mode, the five complexes CA II-
1A
...
1E
were not com-
pletely resolved (Figure 7.13b, inset) and the MS-MS technique was employed to confirm
the identity of the bound ligands from the DCL. MS-MS experiments were performed
by sustained off-resonance irradiation collision-activated dissociation (SORI-CAD) using
argon as the collision gas at an analyzer pressure of
∼
∼
10
−
8
mbar (argon inlet pressure
2.9
10
−
2
mbar). The parent ions bearing the 9
−
charge state were selected by use of
correlated sweep isolation. This was followed by SORI-CAD, resulting in dissociation of
the noncovalent complexes. The collision energy for the experiment was tuned to cause
dissociation of the noncovalent protein-ligand complexes. The result yielded free CA II
(both 8
−
and 9
−
charge states) and, important for the application to DCC screening, singly
charged negative ions for the hydrazone ligands
1A
-
E
, now well resolved by molecular
mass (Figure 7.13c). The masses of these ions were consistent with the [M - H]
−
ions
expected for the DCL sulfonamide hydrazone products
1A
-
E
(Table 7.2). No ions in the
tandem mass spectrum could be attributed to hydrazones
2A
-
E
(lacking the sulfonam-
ide moiety). The hydrazones identified from the DCL with affinity for CA II could only
have been synthesized
in situ
and the result demonstrates that the MS screening approach
was able to identify relevant combinations of fragments whilst in the presence of the tar-
get biomolecule. The sample quantity consumed for these MS-MS experiments was less
than 100 L, the initial ESI-FTMS experiment takes only minutes to perform, while the
MS-MS experiment can be completed within 30 min. Confirmation of the results of this
DCL experiment was then obtained by conducting a conventional solution-phase compet-
itive binding assay for CA II to measure the equilibrium dissociation constants (
K
i
) for
the compounds described in this study. The DCL products
1A
-
E
each exhibited increased
affinity for the enzyme (
K
i
range 10.6-82.3 nM) compared with the scaffold building block
1
(
K
i
=
×
150 nM).
Follow-up ESI-FTMS experiments in our laboratory have been effective with a 10-
fold reduction in CA II concentration (from 30 to 3 M,
4 g protein/DCL based on a
50 L reaction volume) while retaining the ability to detect protein-ligand noncovalent
∼
