Improvement of solid material for affinity resins by application of long PEG spacers to capture the whole target complex of FK506

Article abstract of DOI:10.1016/j.bmcl.2015.05.014

Solid materials for affinity resins bearing long PEG spacers between a functional group used for immobilization of a bio-active compound and the solid surface were synthesized to capture not only small target proteins but also large and/or complex target proteins. Solid materials with PEG1000 or PEG2000 as spacers, which bear a benzenesulfonamide derivative, exhibited excellent selectivity between the specific binding protein carbonic anhydrase type II (CAII) and non-specific ones. These materials also exhibited efficacy in capturing a particular target at a maximal amount. Affinity resins using solid materials with PEG1000 or PEG2000 spacers, bear a FK506 derivative, successfully captured the whole target complex of specific binding proteins at the silver staining level, while all previously known affinity resins with solid materials failed to achieve this objective. These novel affinity resins captured other specific binding proteins such as dynamin and neurocalcin δ as well.

Full text of DOI:10.1016/j.bmcl.2015.05.014

Bioorganic & Medicinal Chemistry Letters 25 (2015) 2788–2792
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Improvement of solid material for affinity resins by application
of long PEG spacers to capture the whole target complex of FK506
Miyuki Mabuchi a,b, Tadashi Shimizu a, Masahiro Ueda a, Kuniko Mitamura c, Shigeo Ikegawa c,
Akito Tanaka a,b,
⇑
a Department of Pharmacy, Hyogo University of Health Sciences, 1-3-6 Minatojima, Chuo-ku, Kobe 650-8530, Japan
b KOBE Chemical Genetics., Inc. 2-2-2 Minatojima-nakamachi, Chuo-ku, Kobe 650-8530, Japan
c Faculty of Pharmaceutical Sciences, Kinki University, 3-4-1 Kowakae, Higashi-Osaka 577-8502, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Solid materials for affinity resins bearing long PEG spacers between a functional group used for immobi-
lization of a bio-active compound and the solid surface were synthesized to capture not only small target
proteins but also large and/or complex target proteins. Solid materials with PEG1000 or PEG2000 as spac-
ers, which bear a benzenesulfonamide derivative, exhibited excellent selectivity between the specific
binding protein carbonic anhydrase type II (CAII) and non-specific ones. These materials also exhibited
efficacy in capturing a particular target at a maximal amount. Affinity resins using solid materials with
PEG1000 or PEG2000 spacers, bear a FK506 derivative, successfully captured the whole target complex
of specific binding proteins at the silver staining level, while all previously known affinity resins with
solid materials failed to achieve this objective. These novel affinity resins captured other specific binding
proteins such as dynamin and neurocalcin d as well.
Received 11 January 2015
Revised 12 April 2015
Accepted 7 May 2015
Available online 15 May 2015
Keywords:
Affinity resins
PEG
Target identification
Carbonic anhydrase II
FKBP12
Ó 2015 Elsevier Ltd. All rights reserved.
Benzenesulfonamide
FK506
Calcineurin
Target complex
Identification of target proteins for bio-active compounds, such
as drugs, natural products and toxins, is an essential component of
modern pharmaceutical sciences and chemical biology. Phenotype
assays of cells or organs without target protein identification have
recently been used in screening studies for drug discovery.1 During
these screening studies, the target compounds are unknown, and
further studies are necessary to identify them. While affinity chro-
matography matrices bearing such bioactive compounds are one of
the major tools for achieving this purpose,2,3 it is often difficult to
identify specific binding proteins on affinity resins owing to a
preponderance of non-specific binding proteins.4 Therefore, the
successful isolation of target proteins by affinity matrices depends
on the synthesis of polymeric resins that can capture the cellular
target with maximum selectivity and efficiency. Affigel™,5a an
agarose derivative, is one of the most popular matrices for this
purpose, and its hydrophilic character helps reduce non-specific
protein absorption.4 However, Affigel™ becomes easily denatured
under organic synthesis conditions; thus, its use in chemical stud-
ies is often restricted. Synthetic resins consist of methacrylates or
stylenes as the core moiety and are widely known. They are usually
synthesized by two steps as follows: first, the hydrophobic core
structure is constructed by polymerization of hydrophobic mono-
mers, such as glycidyl methacrylate5b or stylene,5c and then, the
core structure is covered by hydrophilic spacers to reduce non-
specific protein absorption. However, the coverage by hydrophilic
spacers is usually not sufficient due to difficulty in the additive
reaction of the hydrophilic spacers in the second step caused by
steric hindrance on the surface. The lack of hydrophilicity of
synthetic resins often causes high levels of non-specific binding.4
In previous a Letter,6 we reported the development of a novel
poly(methacrylate) solid material for affinity resins with a mono-
lithic structure, AquaFirmus™ (1), that consists of only hydrophilic
monomers and is hydrophilic enough to reduce non-specific pro-
tein absorptions (Fig. 1). However, it is still difficult for affinity
resins using previously generated solid materials that bear FK506
to identify the whole target complexes of FK506, which include
immunosuppressive complexes and complexes with FKBP12,
calcineurin A/B and calmodulin.7 We hypothesize that the inability
of affinity resins to capture large-sized target complexes is due to
steric hindrance between the captured proteins and the surfaces of
the solid material. In this study, we demonstrate the applications
and effectiveness of long spacers, such as PEG1000 or PEG2000,
for reducing steric hindrance (Fig. 1).
⇑
Corresponding author. Tel.: +81 78 304 3067; fax: +81 78 304 2767.
E-mail address: tanaka-a@huhs.ac.jp (A. Tanaka).
http://dx.doi.org/10.1016/j.bmcl.2015.05.014
0960-894X/Ó 2015 Elsevier Ltd. All rights reserved.

M. Mabuchi et al. / Bioorg. Med. Chem. Lett. 25 (2015) 2788–2792
2789
Spacer
Component
A
1) Polymerization
2) De-protection
O
NHBoc
O
O
AquaFirmusTM (1)
Large target protein(s)
5
NH2
2
O
O
Small target protein
O
B
C
O
3
O
O
Solid material
Bioactive compound
O
5
O
O
4
Long PEG spacer
O
O
O
O
O
O
O
NH2
n
Applications of long PEG spacers
Figure 1. Improvement of affinity resins for capture of large proteins or complexes, 1 was prepared by polymerization of three components (A:B:C = 1:1:9, molar ratio) and
subsequent de-protection of the Boc group by 10% H2O/TFA. 1 contained hexaethylene glycol as a spacer between an amine group for immobilization of bioactive compounds
and for backbone structure. We hypothesized that the capture of large proteins and protein complexes was disturbed by the steric hindrances between captured proteins and
the surfaces of the solid materials (red arrow). Long PEG spacers were adopted as the spacers in this study to capture both small, large proteins and protein complexes.
mixed with lysate from rat brains. Benzenesulfonamide and its
derivatives are known as specific binding inhibitors of carbonic
Design and synthesis
anhydrase type II (CAII) with a Kd of 0.32–1.25 l
M.8 Because the
A previously reported solid material (1) was synthesized by co-
polymerization of three hydrophilic methacrylates: (a) a derivative
with hexaethylene glycol as a spacer bearing a protected amino
group for immobilization of the bioactive compound at the termi-
nal position (component A); (b) a co-polymerization monomer for
adjusting the conditions on the surface (component B); and (c) a
cross-linker monomer (component C) (A:B:C = 1:1:9 in molar
ratio). The addition of a large amount of the cross-linker is vital
for the polymerization reactions of a wide range of methacrylate
monomers bearing flexible PEG spacers.6b We designed and
synthesized methacrylate monomers bearing long PEG spacers,
such as PEG400, PEG1000, PEG2000, PEG4000, PEG8000 and
PEG10000 (5–10, Fig. 2), as component A to reduce the steric
interactions between the captured proteins and the surfaces of
the solid material. In order to examine effects of components B
and C on captures of specific binding protein and non-specific
absorbed ones, additive monomers 11–12 for component B and
cross-linker monomers 13–15 for component C were synthesized
in similar manners (Fig. 2).
In this study, components A, B and C were co-polymerized
under the same conditions as those for 1 and were subsequently
de-protected in 10% H2O/TFA (16–28, Table 1). The resulting solid
materials were stable under acidic conditions. A scanning electro-
micrograph (SEM) analysis showed that the solid materials con-
sisted of relatively shorter PEG spacers (16–17 and 22–26),
which had monolithic structures, similar to 1, while those with
long PEG spacers (18–21 and 27–28) did not (Fig. S1).
molecular weight of CAII (29 kDa) differs significantly from those
of representative non-specific binding proteins, such as actin
(42 kDa) and tubulin (50 kDa), it is easy to estimate the amounts
of specific nonspecific binding proteins. Binding proteins on
29–42 were completely eluted with SDS sample buffer (Fig. 3) after
carefully washing the affinity resins with lysate buffer (0.25 M
sucrose, 0.3 mM N,N-diethylthiocarbamate, 25 mM Tris (pH = 7.6)
with/without 0.1% Tween20) (Fig. S2). In this study, we prepared
two lysate buffers because the amounts of non-specific absorbed
proteins were generally reduced by the addition of detergents,
such as Tween20.
AquaFirmus™ (1), whose spacer is hexaethylene glycol, which
bears the benzenesulfonamide derivative (29), successfully cap-
tured the specific binding protein CAII with a small amount of
non-specific binding proteins found in the lysate buffers with
and without 0.1% Tween20. We next examined affinity resins
bearing a variety of PEG spacers, from PEG400 to PEG10000, in
component A (30–35). Interestingly, for the affinity resins with
PEG1000 (31), PEG2000 (32), or PEG4000 (33), the amount of
CAII was greatly increased compared to that on the affinity resins
with 29, while that on the affinity resin bearing PEG400 as a
spacer (30) was almost same as that with 29. These results indi-
cate that, at a minimum, PEG1000 was required as the spacer
moiety for the maximum capture of the target protein, even for
capture of the relatively smaller protein CAII. On the other hand,
the amount of CAII on affinity resins with PEG8000 (34) and
PEG10000 (35) was lower than on those with 29. We hypothesize
that this decrease was due to the relatively low density of the
amino group, which is necessary for the immobilization of the
compound (Table 1).
Selective capture of specific binding protein
To validate novel solid materials (16–28), we immobilized
4-carboxyl-benzensulfonamide on those solid materials (29–42,
Table 1), and binding proteins on them were analyzed after being
Next, we analyzed the binding of proteins on solid materials
that did not contain component B after immobilization of the

2790
M. Mabuchi et al. / Bioorg. Med. Chem. Lett. 25 (2015) 2788–2792
O
NH
O
PEG length
O
OH
O
N
O
O
OH
Component A
HO
OH
H2N
n
n=5
PEG400
PEG1000
PEG2000
PEG4000
PEG8000
PEG10000
n
2
5
PPh3, DEAD
NH2NH2
O
n
6
7
8
O
O
9
Cl
O
O
OH
BocHN
10
O
BocHN
n
Me2N(CH2)3NMe2
n
Boc2O, pH=9~9.5
R
O
O
Cl
-(CH2CH2O)2CH3
-(CH2CH2O)3CH3
-(CH2)5CH3
3
11
12
Component B
Component C
O
HO
HO
R
R
Me2N(CH2)3NMe2
PEG length
O
O
O
Cl
O
O
n=5
4
OH
n
O
n
O
PEG200
PEG600
PEG1000
13
14
15
Me2N(CH2)3NMe2
Immobilization of bioactive compounds
HOOC
SO2NH2
O
NHCO
SO2NH2
NH2
HOOC
O
HBTU, NEt3
H3CO
CH3
29 - 42
1, 14-28
H3C
N
O
1)FK506-COOH:
O
OH
1)
FK506-COOH
HBTU, NEt3
O
CH3
NHCO
FK506
O
NH2
O
OH
O
CH3
H3C
43 - 46
1, 16-17
OCH3
OCH3
Figure 2. Synthesis of monomers and affinity resins in this study. Synthetic conditions are described in detail in the Supplemental materials.
benzenesulfonamide derivative (36, 37). The amount of CAII on
solid materials with 36 was lower than for those with 29 and
shows a great amount binding of non-specific proteins, and the
amount of CAII on solid materials with 37 was almost the same
as on those with 29. These results suggest that component B is nec-
essary for the specific capture of the target protein (36) and that
increasing the amine density on the solid surface is not important
for capturing target proteins because component A was used twice
for the synthesis of 37 and no increase in captured CAII was
observed. Solid material with one additional PEG unit (24) and a
hydrophobic hexyl (25) instead of hexaethylene glycol in
component B in 1 were prepared, and the benzensulfonamide
derivative was immobilized on that solid material (38, 39). The
amounts of CAII on solid materials with 38 and 39 were almost
the same as for those with 29. On the other hand, there was a great
amount of non-specific binding proteins on solid materials with
39. We thought this increases of non-specific absorption was due
to an increase of hydrophobicity on the surface of the solid
material4 and/or hyper-density of the amino group on them
Capturing the whole target complex of FK506
Based on the results described above, we selected 17 and 18 for
further study of solid materials with an immobilized FK506 deriva-
tive (45, 46). Some affinity resins with 1, 16, AffiGel™, and
Toyopearl™ bearing the FK506 derivative were also prepared.
FK506 is an immunosuppressive drug that targets FKBP12 with a
Kd of 0.4 nM.9 The structure and function of the complex of
FK506 with its targets, including FKBP12, have been well charac-
terized at the molecular level9, and a linker moiety was introduced
at the hydroxyl group at the 32 position of FK506 to connect it to
the resin because crystal structure studies have indicated that this
position is not involved in the binding of FK506 with FKBP12 or
calcineurin A/B (Fig. 2). If affinity resins bearing FK506 can capture
the whole target complex, they are valuable as the standard solid
material for affinity resins because it is difficult for previously used
solid materials to capture the whole complex at the silver staining
level. After being mixed with lysate from rat brains, the binding
proteins on affinity resins were analyzed. Affinity resins 45 and
46 successfully captured the whole proteins of the FK506 target
complex at the silver staining level with a small amount of non-
specific binding proteins, and this was difficult to achieve using
other affinity resins (Fig. 3). The amounts of specific binding pro-
teins (band 1–5) were much smaller than that of FKBP12 in
(50.3 lmol/mL, Table 1).
Finally, component C was modified (40–41). Interestingly, the
use of a longer PEG unit for the spacer adversely affected the
capture of specific binding proteins (Fig. 3). We currently have
no explanation for this result.

M. Mabuchi et al. / Bioorg. Med. Chem. Lett. 25 (2015) 2788–2792
2791
Table 1
Solid materials studied in this study
Component for polymerizationa
Synthesized solid material
NH2
Affinity resins
Immobilized
A
B
C
bioactive compound
PEG
PEG
O R
NO
NH2 densityb
18.8
BSA
FK506
O
NHBoc
O
O
O
O
O
O
2
3
4
1 (AquaFimus™)
29
43
5
6
7
8
9
10
3
3
3
3
3
3
4
4
4
4
4
4
16
17
18
19
20
21
9.8
12.3
6.3
7.2
0.56
0.31
30
31
32
33
34
35
44
45
46
2
2
2
2
None
None
11
4
4
4
4
22c
23d
24
21.7
28.0
28.6
50.3
36
37
38
39
12
25
2
2
2
3
3
3
13
14
15
26
27
28
1.6
11.4
5.18
40
41
42
a
b
c
Molar ratio were A:B:C = 1:1:9 (except for those of 22 and 23).
Density of NH2 on solid material were estimated by the ninhydrin test (mmol/mL, UV absorption at 570 nm of 1 was the reference value.
Molar ratio for synthesis of 22 was 2:4 = 1:9 (no component B).
d
Molar ratio for synthesis of 23 was 2:4 = 2:9 (no component B).
a. Benzenesulfonamide
Without Tween20
With 0.1% Tween20
29 30 31 32 33 34 35 36 37 38 39 40 41 42
29 30 31 32 33 34 35 36 37 38 39 40 41 42
150
100
75
tubulin
actin
50
37
CA II
25
20
15
10
CBB stained
b. FK506
43
44
45
46
Competition
150
᧩
+
᧩
+
᧩
+
᧩
+
᧩
+
᧩ +
100
75
1
2
1
2
1:Dynamin
2: Calcineurin A
3: Neurocalcin
4: Calmodulin
5: Calcineurin B
6: FKBP12
50
37
25
20
3
3
15
6
4
4
6
5
6
5
6
10
Silver stained
6
6
Western blotting
(anti-CnA)
50
Figure 3. Binding proteins on solid materials bearing (a) benzensulfonamide and (b) FK506. Each resin bearing p-carboxyl-benzensulfonamide (a) or FK506 (b) was mixed
with lysate from rat brains, and binding proteins were eluted with SDS sample buffer (Nacalai #30566-22) after carefully being washed with lysate buffer (0.25 M sucrose,
0.3 mM N,N-diethylthiocarbamate, 1 mM CaCl2, 1 mM MgCl2, 25 mM Tris (pH = 7.6)) with or without 0.1% Tween20. Specific binding proteins, CA II (carbonic anhydrase type
II) for 29–42 and FKBP12 (band 6) and other proteins (bands 1–5) for 43–46, and nonspecific absorbed proteins were analyzed.

2792
M. Mabuchi et al. / Bioorg. Med. Chem. Lett. 25 (2015) 2788–2792
Figure 3. We thought the reason for this stoichiometrical difference
due to their population in cells and the strong affinity between
FK506 and FKBBP12 compared to the others.
Acknowledgment
This work was supported by Grant-in-Aid for Scientific
Research (C) (NO. 19599025).
Interestingly, dynamin and neurocalcin d were also identified as
specific binding proteins on 45 and 46. Dynamin is a large protein
(97 KDa) and known as an associated protein of calcineurin in
synaptic cells and this interaction is vital for exocytosis of synaptic
vesicles.10 Neurocalcin d is a neuronal calcium-binding protein like
the calmodulin in mammalian brains.11 Roles of these specific
bindings in bioactivities of FK506 were now unclear while neuro-
protective effect in vivo of FK506 has been reported.12 The speci-
ficities of each binding protein were confirmed by the
competition method, and identifications of FKBP12, calmodulin
calcineurin A, Calcineurin B, dynamin, and neurocalcin d were car-
ried out by a MS/MS ion search method based on ESI ion trap mass
spectrometry after in-gel digestion (Figs. S3–S4).
In conclusion, some solid materials for affinity resins were syn-
thesized for the target identification of bio-active compounds and
were intended to be effective at identifying both small target pro-
teins and large target complexes (Fig. 1). Solid materials with
PEG1000 (17) or PEG2000 (18) spacers, which both bear a benzen-
sulfonamide derivative (31, 32), captured the target protein, CAII,
at a maximal amount with low amounts of non-specific binding
proteins (Fig. 3a). These results indicated that, at a minimum,
PEG1000 was required to reduce the steric hindrance between
the captured proteins and the solid surfaces, even to capture the
Supplementary data
Supplementary data (experimental procedures and Figs. S1–S4)
associated with this article can be found, in the online version, at
http://dx.doi.org/10.1016/j.bmcl.2015.05.014.
References and notes
1. Gershell, L. J.; Atkins, J. H. Nat. Rev. Drug Discov. 2003, 2, 321.
2. Harding, M. W.; Galat, A.; Uehling, D. E.; Schreiber, S. L. Nature 1989, 341, 758.
3. Taunton, J.; Hassig, C. A.; Schreiber, S. L. Science 1996, 272, 408.
4. Tamura, T.; Terada, T.; Tanaka, A. Bioconjugate Chem. 2003, 14, 1222.
5. (a) http://www.bio-rad.com/; (b) TOYOPEARL; http://www.tosho.cc/en/; (c)
Shimizu, N.; Sugimoto, K.; Tang, J.; Nishi, T.; Sato, I.; Hiramoto, M.; Aizawa, S.;
Hatakeyama, M.; Ohba, R.; Hatori, H.; Yoshikawa, T.; Suzuki, F.; Oomori, A.;
Tanaka, H.; Kawaguchi, H.; Watanabe, H.; Handa, H. Nat. Biotechnol. 2000, 18,
877.
6. AquaFirmus™; http://www2.huhs.ac.jp/~h070016a/AQUAFIRMUS.pdf; (b)
Mori, T.; Takahashi, T.; Shiyama, T.; Tanaka, A.; Hira, N.; Tanaka, N.; Hosoya,
K. Bioorg. Med. Chem. 2006, 14, 5549.
7. Liu, J.; Farmer, J. D.; Lane, W. S.; Friedman, J.; Weissman, I.; Schreiber, S. L. Cell
1991, 66, 807.
8. (a) Lahiri, J.; Isaacs, L.; Tien, J.; Whitesides, G. M. Anal Chem. 1999, 71, 777; (b)
Svedhem, S.; Enander, K.; Karlsson, M.; Sjöbom, H.; Liedberg, B.; Löfås, S.;
Mårtensson, L. G.; Sjöstrand, S. E.; Svensson, S.; Carlsson, U.; Lundström, I. Anal.
Biochem. 2001, 296, 188.
9. Siekierka, J. J.; Hung, S. H.; Poe, M.; Lin, C. S.; Sigal, N. H. Nature 1989, 341, 755.
10. Lai, M. M.; Hong, J. J.; Ruggiero, A. M.; Burnett, P. E.; Slepnev, V. I.; Camilli, P. D.;
Snyder, S. H. J. Biol. Chem. 1999, 274, 25963.
11. Wang, W.; Zhou, Z.; Zhao, W.; Huang, Y.; Tang, R.; Ying, K.; Xie, Y.; Mao, Y.
Biochim. Biophys. Acta 2001, 1518, 162.
12. Sharkey, J.; Butcher, S. P. Nature 1994, 371, 336.
small target protein CAII (29 KDa). Affinity resins bearing
a
FK506 derivative (45, 46) successfully captured the whole target
complex, while all of the commercially available solid materials
were unable to do so (Fig. 3). On 45 and 46, two novel specific
binding proteins of FK506, dynamin and neurocalcin, were also
identified while their role were unclear. Based on these results,
we believe that 17 and 18 were effective solid materials for target
identification studies examining
compounds.
a wide range of bioactive