Amide-based derivatives of β-alanine hydroxamic acid as histone deacetylase inhibitors: Attenuation of potency through resonance effects

Article abstract of DOI:10.1016/j.bmcl.2012.08.006

A library of amide-linked derivatives of β-alanine hydroxamic acid were prepared (2-7) and the activity as inhibitors of Zn(II)-containing histone deacetylases (HDACs) determined in vitro against HDAC1 and the anti-proliferative activity determined in BE(2)-C neuroblastoma cells. The IC₅₀ values of the best-performing compounds (3-7) against HDAC1 ranged between 38 and 84 μM. The least potent compound (2) inhibited a maximum of only 40% HDAC1 activity at 250 μM. The anti-proliferative activity of 2-7 at 50 μM against BE(2)-C neuroblastoma cells ranged between 57.0% and 88.6%. The structural similarity between the potent HDAC inhibitor trichostatin A (TSA, 1; HDAC1, IC₅₀ 12 nM) and the present compounds (2-7) was high at the Zn(II) coordinating hydroxamic acid head group; and in selected compounds (2, 5), at the 4-(dimethylamino)phenyl tail. The significantly reduced potency of 2-7 relative to 1 underscores the rank importance of the linker region as part of the HDAC inhibitor pharmacophore. Molecular modeling of 1-7 using HDAC8 as the template suggested that the conformationally constrained 4′-methyl group of 1 may contribute to HDAC inhibitor potency through a sandwich-like interaction with a hydrophobic region containing F152 and F208; and that the absence of this group in 2-7 may reduce potency. The close proximity of the 5′-carbonyl oxygen atom in 2-7 to the sulfur atom of Met274 in HDAC8 or the corresponding isobutyl group of Leu274 in HDAC1 may attenuate potency through repulsive steric and dipole-dipole forces. In a unique resonance stabilized form of 2, this interaction could manifest as stronger ion-dipole repulsive forces, resulting in a further decrease in potency. This work suggests that resonance structures of HDAC inhibitors could modulate intermolecular interactions with HDAC targets, and potency.

Full text of DOI:10.1016/j.bmcl.2012.08.006

Bioorganic & Medicinal Chemistry Letters 22 (2012) 6200–6204
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Amide-based derivatives of b-alanine hydroxamic acid as histone
deacetylase inhibitors: Attenuation of potency through resonance effects
Vivian Liao a, Tao Liu b, Rachel Codd a,
⇑
a School of Medical Sciences (Pharmacology) and Bosch Institute, University of Sydney, New South Wales 2006, Australia
b Children’s Cancer Institute Australia, Sydney Children’s Hospital, New South Wales 2031, Australia
a r t i c l e i n f o
a b s t r a c t
Article history:
A library of amide-linked derivatives of b-alanine hydroxamic acid were prepared (2–7) and the activity
as inhibitors of Zn(II)-containing histone deacetylases (HDACs) determined in vitro against HDAC1 and
the anti-proliferative activity determined in BE(2)-C neuroblastoma cells. The IC50 values of the best-per-
Received 13 June 2012
Revised 23 July 2012
Accepted 1 August 2012
Available online 9 August 2012
forming compounds (3–7) against HDAC1 ranged between 38 and 84
inhibited a maximum of only 40% HDAC1 activity at 250 M. The anti-proliferative activity of 2–7 at
50 M against BE(2)-C neuroblastoma cells ranged between 57.0% and 88.6%. The structural similarity
lM. The least potent compound (2)
l
l
Keywords:
between the potent HDAC inhibitor trichostatin A (TSA, 1; HDAC1, IC50 12 nM) and the present com-
pounds (2–7) was high at the Zn(II) coordinating hydroxamic acid head group; and in selected com-
pounds (2, 5), at the 4-(dimethylamino)phenyl tail. The significantly reduced potency of 2–7 relative
to 1 underscores the rank importance of the linker region as part of the HDAC inhibitor pharmacophore.
Molecular modeling of 1–7 using HDAC8 as the template suggested that the conformationally con-
strained 40-methyl group of 1 may contribute to HDAC inhibitor potency through a sandwich-like inter-
action with a hydrophobic region containing F152 and F208; and that the absence of this group in 2–7
may reduce potency. The close proximity of the 50-carbonyl oxygen atom in 2–7 to the sulfur atom of
Met274 in HDAC8 or the corresponding isobutyl group of Leu274 in HDAC1 may attenuate potency
through repulsive steric and dipole–dipole forces. In a unique resonance stabilized form of 2, this inter-
action could manifest as stronger ion-dipole repulsive forces, resulting in a further decrease in potency.
This work suggests that resonance structures of HDAC inhibitors could modulate intermolecular interac-
tions with HDAC targets, and potency.
Histone deacetylase inhibitors
Trichostatin A
Ó 2012 Elsevier Ltd. All rights reserved.
Some of the most potent inhibitors of Class I/II and IV of the his-
tone deacetylases (HDACs), which are targets at the forefront of
cancer research, contain a hydroxamic acid group that inactivates
the Zn(II) ion at the HDAC active site. In concert with histone
acetyltransferases, HDACs modulate the acetylation status of the
vorinostat; HDAC1, IC50 48 nM) for the treatment of cutaneous
T-cell lymphoma, a notable example.3,6 Even more potent than
SAHA, is the natural hydroxamic acid trichostatin A (2E,4E,6R)-7-
[4-(dimethylamino)phenyl]-N-hydroxy-4,6-dimethyl-7-oxohepta-
2,4-dienamide (1) produced by Streptomyces platensis, Streptomyces
hygroscopicus Y-50 or Streptomyces sioyaensis (HDAC1, IC50
12 nM).7,8 This compound is costly to access from culture and is
difficult to synthesize,9 which calls for the design of analogues of
1 as potential HDAC inhibitors.
The pharmacophore of hydroxamic acid-based HDAC inhibitors
has been defined as comprising three elements: a Zn(II) coordinat-
ing hydroxamic acid group, a terminal aromatic group, and a linker
region.10–12 In order to better understand the contribution of the
linker region to the pharmacophore, we prepared a library of
amide-linked derivatives of b-alanine hydroxamic acid, some
candidates which shared with 1, the first two elements of the
pharmacophore.
e-amino group of surface-exposed lysine residues in nucelosomal
histones, which affects chromatin topology and transcriptional
activity. Hypo-acetylated histones that result from upregulated
HDAC activity or downregulated histone acetyltransferase activity
increase the relative concentration of condensed chromatin which
represses the transcription of tumor suppressor genes and pro-
motes the onset and progression of tumors.1–3 The metal coordi-
nating ability of the hydroxamic acid functional group has broad
utility in clinical settings,4,5 with hydroxamic acid-based HDAC
inhibitors such as suberoylanilide hydroxamic acid (SAHA or
Abbreviations: HDAC, histone deacetylase; TSA, trichostatin A; SAHA, suberoy-
lanilide hydroxamic acid.
Amide-linked precursors were prepared from the HOBt-based,
EDC-activated conjugation of para-substituted (X = dimethylamino,
methyl, nitro, H) derivatives of cinnamic acid, 3-phenylpropanoic
⇑
Corresponding author. Tel.: +61 (0)2 9351 6738; fax: +61 (0)2 9351 4717.
E-mail address: rachel.codd@sydney.edu.au (R. Codd).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.08.006

V. Liao et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6200–6204
6201
acid, 2-phenylacetic acid or 2-phenoxyacetic acid; and b-alanine
ethyl ester hydrochloride. Following hydrolysis of the ethyl ester,
the hydroxamic acid group was installed from the reaction between
the mixed anhydride formed in situ from chloroethylformate and
N-methylmorpholine; and hydroxylamine (Scheme 1). After work-
up, when sprayed with an ethanol solution of Fe(III), the final spot
of the product on a TLC plate turned pale pink, as evidence of the
metal-coordinating ability of the hydroxamic acid group. The prod-
ucts (Table 1) were characterised by 1H and 13C NMR spectroscopy,
ESI-MS and microanalysis (Supplementary data). Relative to the free
carboxylic acid, the signals in the 1H NMR spectra assigned to the
coordinate bonds between the hydroxamic acid group and the
Zn(II) ion were re-installed. This method assumes that the biden-
tate hydroxamate coordination between 2 and 7 and Zn(II) will
not differ significantly from the 1-Zn(II) coordination observed in
the X-ray crystal structure. The method allows for the structure
of 2–7 distal to the active site Zn(II) to be optimized within the
constraints of the HDAC8 active site cavity. The veracity of the
method was supported by the close agreement between the con-
formation of native 1 and 1 which was minimized within the
HDAC8 active site as described for 2–7 (Fig. 2, far left). The overlaid
structures of minimized 1 and each of 2–7 are shown in Figure 2.
Despite the variation in the number of main-chain atoms in the
linker separating the hydroxamic acid head group from the
phenyl-substituted tail (six atoms: 2–4, 7; five atoms: 5, 6), the
phenyl group in most cases minimized to a position that coalesced
closely with the phenyl group of 1. The poorest overlay of phenyl
groups occurred between 1 and 4 (Fig. 2, middle). In this case,
the ethylene group proximal to the tolyl group was sufficiently
flexible to mine a local hydrophobic pocket in the protein, which
re-directed the orientation of the tolyl group. This local conforma-
tional flexibility was not available to other candidates, due to
unsaturation (2, 7), an insufficient number of methylene groups
(5, 6) or the presence of a methoxy group subject to electron
delocalization effects (3).
The linker region of 1 is more complex than 2–7, with two
methyl substituents, one chiral carbon (6R) and two unsaturated
carbon–carbon bonds (2E, 4E). In the X-ray crystal structure of 1
bound to HDAC8, the 40-methyl group is positioned in a sand-
wich-like fashion between the aromatic rings of F152 and F208
with the distance from each plane of the aromatic ring to the
methyl group carbon atom equal at 3.84 Å (Fig. 3). This hydropho-
bic interaction was not available to 2–7 (Fig. 3, shown for 2) and
may be a factor that contributed to their low potency as HDAC1
inhibitors. The structures of 2–7 as minimized in the HDAC8 ac-
tive site cavity revealed a close distance between the 50-carbonyl
oxygen atom and the sulfur atom of Met274. In 2 and 7, this dis-
tance was 3.5 Å, which is close to the sum of the sulfur and oxy-
gen van der Waals radii (3.3 Å)22 and raises the possibility of the
carbonyl oxygen atom-mediated oxidation of the Met274 sulfur
atom in HDAC8, as shown to occur in amyloid b-peptide.23,24
Inhibitor potency was determined against HDAC1, which instead
of the four amino acid sequence P273-M274-C275-S276 in
HDAC8, contains R273-L274-G275-C276.25 In silico mutation of
PMCS(HDAC8) ? RLGC(HDAC1) followed by minimization of 2–7
and RLGC only, with the remaining protein constrained, showed
that the 50-carboxyl oxygen atom of 2 was in close proximity to
the isobutyl side chain of Leu274, with one oxygen–hydrogen
(methyl) distance of 2.8 Å. It is proposed that the proximity of
the 50-carbonyl oxygen atom in 2-7 to Leu274 in HDAC1 is a con-
tributing factor to the low potency of this group of compounds.
Reduced binding to the active site would arise from steric repul-
sion and from repulsive dipole–dipole interactions, since the
a-methylene protons shifted upfield by about 0.2 ppm in the
corresponding hydroxamic acid.
The IC50 values of 2–7 against HDAC1 were determined using a
fluorometric assay and the anti-proliferative effects of the com-
pounds were determined against BE(2)-C neuroblastoma cells
using the Alamar blue assay (Fig. 1, Table 1).13–15 Compared to
the IC50 value of 1 against HDAC1 (IC50 12 nM), 2–7 had signifi-
cantly decreased potency (38–84
a maximum of 40% HDAC1 activity at 250
l
M). Compound 2 inhibited only
M, and was the least
l
potent compound. The IC50 value of SAHA against HDAC1 deter-
mined in the current work (110.8 1.1 nM) calibrated the results
to literature IC50 values for SAHA;16 and supported the veracity
of the data for 2–7. Given the similarity between 2–7 and 1, with
respect to the Zn(II) coordinating hydroxamic acid group; and be-
tween 2, 5 and 1, with respect to the 4-(dimethylamino)phenyl
group, the low potency against HDAC1 was unexpected. Most sur-
prising was the extraordinarily low potency of 2 against HDAC1.
The whole cell-based assay system gave an activity profile that dif-
fered from the profile established using isolated HDAC1. At 50 lM
the anti-proliferative activity of 2–7 against BE(2)-C neuroblas-
toma cells ranged between 57.0 and 88.6%, with 2 the most potent
compound. Other work has shown differences between the activi-
ties of compounds screened using isolated HDACs or whole-cell
assay systems, with belinostat and SAHA a notable example
(belinostat:SAHA; biochemical activity (ꢀ1:1), anti-proliferative
activity against a selection of cancer cell lines (ꢀ1.7–7:1)).17 This
is likely due to the higher complexity of the whole cell-based assay
system with regard to the involvement of HDACs in regulating the
acetylation of non-histone proteins (a-tubulin, tumor suppressor
p53) and the optimal activity of most HDACs requiring the forma-
tion of multiprotein complexes. The IC70 values of 1 or SAHA were
determined in BE(2)-C neuroblastoma cells as 0.05
lM or 2 lM,
respectively. Compound 7 has been reported to inhibit 19% of
HDAC activity in HeLa extract at 1
oped as a lead.18
lM, but was not further devel-
Molecular modelling (HyperChem 7.5, MM+ force field for en-
ergy minimization) was used to inform the IC50 results of 2–7
against HDAC1. Three X-ray crystal structures of human HDACs
with 119,20 and one structure of a prokaryotic HDAC homologue
with 1 have been solved.21 The X-ray crystal structure of 1 bound
to human HDAC819 was used as the template. The coordinate
bonds between 1 and Zn(II) were deleted and the isolated fragment
was used to build models of each of 2–7 with zero charge. The pro-
tein and the hydroxamic acid group (C(O)NH(OH)) of the model of
2–7 were frozen and the remaining structure of the free ligand was
minimized. At the completion of the minimization process, the
molecular dipole in leucine is orientated from the
a-carbon atom
(positive) along the isobutyl tail (negative). Due to the presence of
the electron donating dimethylamino substituent (rp –0.83) in 2
and the conjugated linker region, this compound could comprise
different contributing resonance structures (Fig. 4a). In the
Scheme 1. Synthesis of amide-linked derivatives of b-alanine hydroxamic acid: reagents and conditions (shown for 2). (i) HOBt, EDCꢁHCl, DIPEA, DMF, RT, 16 h; (ii) NaOH,
MeOH, RT, 30 min; (iii) EtOCOCl, NMM, NH2OH, MeOH, RT, 15 min.

6202
V. Liao et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6200–6204
Table 1
Synthesis and activity measurements of 2–7
Structure
Activity
IC50 HDAC1a
(lM)
Neuroblastoma cellsb (%)
c
1
0.012
2
>250d
57.0 (1.1)
3
4
84.0 (4.6)
80.0 (1.5)
80.4 (1.8)
72.4 (1.7)
5
38.0 (1.9)
62.8 (2.4)
6
7
39.4 (0.3)
67.0 (6.7)
87.9 (3.8)
88.6 (1.6)
a
Results reported as an average of n = 3 with standard error of the mean in brackets.
b
Results from a concentration of 50
lM reported as an average of n = 9 with standard error of the mean in brackets.
c
The IC70 values of TSA or SAHA were determined in BE(2)-C neuroblastoma cells as 0.05 M or 2 M, respectively.
Maximal inhibition of HDAC1 of 40% at 250 lM.
l
l
d
the dimethylamino group is absent) to an ion-dipole force (2).
This provides a rationale that accommodates for both the low po-
tency of 2–7 as a group relative to 1; and the extraordinarily low
potency of 2, relative to 3–7. The distance between the 70-car-
bonyl oxygen atom of 1 which could exist in two resonance forms
and atoms of the corresponding side chains (Met274, Leu274) is
sufficient to reduce these intermolecular repulsions (Fig. 4c). In
summary, the absence of the 40-methyl group in 2–7 prevents
favorable van der Waals binding effects with F152 and F208, as
posited for 1. As a group, there may be additional reductions in
HDAC inhibitor potency in 2–7 due to repulsive steric and di-
pole–dipole effects resulting from the close proximity of the 50-
carbonyl oxygen atom and the isobutyl group of Leu274 (HDAC1)
or the sulfur atom in Met274 (HDAC8). The existence of unique
resonance stabilized forms of 2 could manifest as further de-
creases in potency, due to the increased repulsive ion-dipole force
between the anionic 50-carbonyl oxygen atom and these residues.
The work has demonstrated that the positioning of the amide
bond in the linker region of HDAC inhibitors has a significant ef-
fect on potency and that resonance stabilized structures of com-
pounds should be factored into the drug design process. In this
case, the resonance stabilization appears to have attenuated po-
tency. It remains conceivable that this effect could be used to pos-
itively modulate potency in HDAC inhibitors.
Figure 1. Anti-proliferative activity of 2–7 against BE(2)-C neuroblastoma cells.
resonance structure of 2 which features a negative charge on the
50-carboxyl oxygen atom, the intermolecular repulsive force with
Leu274 (HDAC1) or Met274 (HDAC8) would increase in strength
from a dipole–dipole force (3–7) (Fig. 4b, shown for 7, in which

V. Liao et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6200–6204
6203
Figure 2. The overlay of 1 as from the X-ray crystal structure of 1-bound HDAC8,19 and 1 as minimized in the HDAC8 active site. The models of 2–7 as minimized within the
HDAC8 active site are overlaid with the model of 1. Excluding the atoms comprising the hydroxamic acid group (–C(O)NHOH), the compounds were minimized in the
presence of the static HDAC8 enzyme. Hydrogen atoms have been omitted for clarity.
Figure 3. Overlaid structures of minimized models of 1 or 2 bound to the active site of HDAC8 (a) or HDAC1 (as generated by in silico mutation of HDAC8) (b). Apart from
PMCS (HDAC8) and RLGC (HDAC1), the residues located within 5 Å of the Zn(II) ion are conserved between HDAC8 and HDAC1. The close distance between the 50-carbonyl
oxygen atom in 2–7 and the sulfur atom in Met274 (HDAC8) or the isobutyl group of Leu274 (HDAC1) may attenuate HDAC inhibitor potency.
Figure 4. The ion-dipole repulsion between the negatively charged 50-carbonyl oxygen atom of the zwitterionic resonance stabilized form of 2 and the sulfur atom of M274
(HDAC8) or the isobutyl group of Leu274 (HDAC1) (RHS, (a)) is posited to contribute to its limited potency as an HDAC1 inhibitor. Relative to 2, the weaker dipole–dipole
repulsion in 7 manifests as increased HDAC1 inhibitor potency (b). The distance between the 7’-carbonyl oxygen atom of 1, which could exist in two resonance forms, and
atoms of the corresponding side chains, is sufficient to prevent intermolecular repulsion (c).
2. Liu, T.; Kuljaca, S.; Tee, A.; Marshall, G. M. Cancer Treat. Rev. 2006, 32,
Acknowledgments
157.
3. Bertrand, P. Eur. J. Med. Chem. 2010, 45, 2095.
4. Marmion, C. J.; Griffith, D.; Nolan, K. B. Eur. J. Inorg. Chem. 2004, 3003.
5. Codd, R. Coord. Chem. Rev. 2008, 252, 1387.
6. Richon, V. M.; Emiliani, S.; Verdin, E.; Webb, Y.; Breslow, R.; Rifkind, R. A.;
Marks, P. A. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 3003.
Funding to R. C. from the Cancer Institute NSW (08/RFG/1-12)
and to T. L. from the National Health and Medical Research Council
Australia (1006002) is kindly acknowledged.
7. Codd, R.; Braich, N.; Liu, J.; Soe, C. Z.; Pakchung, A. A. H. Int. J. Biochem. Cell Biol.
2009, 41, 736.
8. Ejje, N.; Lacey, E.; Codd, R. RSC Adv. 2012, 2, 333.
Supplementary data
9. Zhang, S.; Duan, W.; Wang, W. Adv. Synth. Catal. 2006, 348, 1228.
10. Miller, T. A.; Witter, D. J.; Belvedere, S. J. Med. Chem. 2003, 46, 5097.
11. Jung, M.; Brosch, G.; Kölle, D.; Scherf, H.; Gerhäuser, C.; Loidl, P. J. Med. Chem.
1999, 42, 4669.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2012.08.006.
12. Whitehead, L.; Dobler, M. R.; Radetich, B.; Zhu, Y.; Atadja, P. W.; Claiborne, T.;
Grob, J. E.; McRiner, A.; Pancost, M. R.; Patnaik, A.; Shao, W.; Shultz, M.;
Tichkule, R.; Tommasi, R. A.; Vash, B.; Wang, P.; Stams, T. Bioorg. Med. Chem.
Lett. 2011, 19, 4626.
References and notes
1. Bolden, J. E.; Peart, M. J.; Johnstone, R. W. Nat. Rev. Drug Discov. 2006, 5, 769.

6204
V. Liao et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6200–6204
13. HDAC1 assay. The HDAC1 Inhibitor Screening assay kit was purchased from
Cayman Chemical Company. A FLUOstar Omega microplate reader was used at
an excitation wavelength of 345 nm and an emission wavelength of 450 nm.
Samples of 2–7 were prepared in methanol at five concentrations (10, 100, 250,
17. Khan, N.; Jeffers, M.; Kumar, S.; Hackett, C.; Boldog, F.; Khramtsov, N.; Qian, X.;
Mills, E.; Berghs, S. C.; Carey, N.; Finn, P. W.; Collins, L. S.; Tumber, A.; Ritchie, J.
W.; Jensen, P. B.; Lichenstein, H. S.; Sehested, M. Biochem. J. 2008, 409, 581.
18. Watkins, C. J.; Romero-Martin, M.-R.; Moore, K. G.; Ritchie, J.; Finn, P. W.;
Kalvinish, I.; Loza, E.; Starchenkov, I.; Dikovska, K.; Bokaldere, R. M.; Galite, V.;
Vorona, M.; Andrianov, V.; Lolya, D.; Semenikhina, V.; Amolins, A.; Harris, C. J.;
Duffy, J. E. S. 2002, PCT Int. Appl. WO 2002026696.
19. Dowling, D. P.; Gantt, S. L.; Gattis, S. G.; Fierke, C. A.; Christianson, D. W.
Biochemistry 2008, 47, 13554.
20. Schuetz, A.; Min, J.; Allali-Hassani, A.; Schapira, M.; Shuen, M.; Loppnau, P.;
Mazitschek, R.; Kwiatkowski, N. P.; Lewis, T. A.; Maglathin, R. L.; McLean, T. H.;
Bochkarev, A.; Plotnikov, A. N.; Vedadi, M.; Arrowsmith, C. H. J. Biol. Chem.
2008, 283, 11355.
500, 1000
75, 150, 200
in a 96-well microplate according to the instruction manual. The HDAC1
activity was completely inhibited with 1 at 1 M. The IC50 values of 2–7 were
l
M). Additional samples were prepared in methanol for 5 (25, 50,
l
M) and 7 (1, 5, 25, 50 M). The assay was conducted in triplicate
l
l
calculated using Prism software by fitting the data points with the equation:
y = A ꢂ ((A ꢃ x)/B + x), where A = 100% activity and B = IC50. The inhibitory
activity of 2–7 against other HDAC isoforms was not screened in this work.
14. Antiproliferative-activity. BE(2)-C neuroblastoma cells were treated with 2–7
at 50 lM or with different dosages of TSA or SAHA. Seventy-two hours after
treatment, relative numbers of cells were examined by the Alamar blue assay,
measured as optical density (OD) units of absorbance, and expressed as a
percentage of absorbance of the experimental samples over that of control
samples (i e., percentage change in cell number).
21. Finnin, M. S.; Donigian, J. R.; Cohen, A.; Richon, V. M.; Rifkind, R. A.; Marks, P.
A.; Breslow, R.; Pavletich, N. P. Nature 1999, 401, 188.
22. Bondi, A. J. Phys. Chem. 1964, 68, 441.
23. Coles, M.; Bicknell, W.; Watson, A. A.; Fairlie, D. P.; Craik, D. J. Biochemistry
1998, 37, 11064.
24. Tan, Z.; Shang, S.; Danishefsky, S. J. Proc. Natl. Acad. Sci. U.S.A. 2011, 108, 4297.
25. Somoza, J. R.; Skene, R. J.; Katz, B. A.; Mol, C.; Ho, J. D.; Jennings, A. J.; Luong, C.;
Arvai, A.; Buggy, J. J.; Chi, E.; Tang, J.; Sang, B.-C.; Verner, E.; Wynands, R.;
Leahy, E. M.; Dougan, D. R.; Snell, G.; Navre, M.; Knuth, M. W.; Swanson, R. V.;
McRee, D. E.; Tari, L. W. Structure 2004, 12, 1325.
15. Marshall, G. M.; Liu, P. Y.; Gherardi, S.; Scarlett, C. J.; Bedalov, A.; Xu, N.; Iraci,
N.; Valli, E.; Ling, D.; Thomas, W.; van Bekkum, M.; Sekyere, E.; Jankowski, K.;
Trahair, T.; MacKenzie, K. L.; Haber, M.; Norris, M. D.; Biankin, A. V.; Perini, G.;
Liu, T. PLoS Genet. 2011, 7, e1002135.
16. Suzuki, T.; Kouketsu, A.; Matsuura, A.; Kohara, A.; Ninomiya, S.-I.; Kohda, K.;
Miyata, N. Bioorg. Med. Chem. Lett. 2004, 14, 3313.