Design and synthesis of mimics of the T7-loop of FtsZ

Article abstract of DOI:10.1021/ol4010068

Mimics of the T7-loop of the bacterial cell division protein FtsZ have been designed and synthesized. The design is based on the X-ray cocrystal structure of P. aeruginosa FtsZ:SulA. Fast Rigid Exhaustive Docking (FRED) was employed to select compounds that can mimic the key interacting residues. Bicyclic oxazolidinones 1a-d were selected for further study and synthesized from a key bicyclic aziridine intermediate, which is synthesized from a readily available unsaturated aldehyde and amides derived from α-amino acids.

Full text of DOI:10.1021/ol4010068

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Design and Synthesis of Mimics of the
T7-loop of FtsZ
Nohemy A. Sorto, Phillip P. Painter, James C. Fettinger, Dean J. Tantillo, and
Jared T. Shaw*
Department of Chemistry, One Shields Ave, University of California, Davis,
California 95616, United States
jtshaw@ucdavis.edu
Received April 10, 2013
ABSTRACT
Mimics of the T7-loop of the bacterial cell division protein FtsZ have been designed and synthesized. The design is based on the X-ray cocrystal
structure of P. aeruginosa FtsZ:SulA. Fast Rigid Exhaustive Docking (FRED) was employed to select compounds that can mimic the key
interacting residues. Bicyclic oxazolidinones 1aÀd were selected for further study and synthesized from a key bicyclic aziridine intermediate,
which is synthesized from a readily available unsaturated aldehyde and amides derived from R-amino acids.
Inhibition of cell division by the SOS response protects
bacterial organisms against antibiotics that target cell wall
synthesis, including β-lactams.1 Although these antibiotics
are among the most commonly used for the treatment of
infections, bacterial resistance has rendered them ineffec-
tive in many cases.2 For this reason, it is imperative to find
new pathways to battle bacterial infections. While devel-
oping new antibiotics is vital, restoring the lethality of
current antibiotics is a strategy that could also be effective.
Researchers at Merck recently reported that certain
β-lactams regain their lethality against MRSA when used
in combination with compound PC190723.3 This com-
pound is known to disrupt the normal function of the
bacterial cell division protein FtsZ. This fascinating result
can be adapted to other pathways related to the cell
division machinery such as the bacterial SOS response,
which triggers DNA repair in bacteria.4 This event acti-
vates the overexpression of SulA, an endogenous inhibitor
of FtsZ. SulA inhibits the polymerization of FtsZ mono-
mers, an essential process leading to bacterial cell division,
by binding to the T7-loop of FtsZ.5 The crucial role that
the T7-loop plays during the SOS response has led us to
hypothesize that mimics of this loop can act as SulA
modulators. The SOS response represents an attractive
target that can be utilized to develop compounds that can
operate in synergy with antibiotics to restore their
lethality.1 Mimics of the T7-loop of FtsZ represent a great
starting point in the search for such compounds. Aided by
computational methods and inspired by the X-ray cocrys-
tal structure of FtsZ and SulA,6 we report the design and
the synthesis of mimics of the T7-loop of FtsZ.
(1) Miller, C.; Thomsen, L. E.; Gaggero, C.; Mosseri, R.; Ingmer, H.;
Cohen, S. N. Science (Washington, DC, U.S.) 2004, 305, 1629–1632.
(2) Perez-Llarena, F. J.; Bou, G. Curr. Med. Chem. 2009, 16, 3740–3765.
(3) (a) Elsen, N. L.; Lu, J.; Parthasarathy, G.; Reid, J. C.; Sharma, S.;
Soisson, S. M.; Lumb, K. J. J. Am. Chem. Soc. 2012, 134, 12342–12345.
(b) Tan, C. M.; Therien, A. G.; Lu, J.; Lee, S. H.; Caron, A.; Gill, C. J.;
Lebeau-Jacob, C.; Benton-Perdomo, L.; Monteiro, J. M.; Pereira, P. M.;
Elsen, N. L.; Wu, J.; Deschamps, K.; Petcu, M.; Wong, S.; Daigneault,
E.; Kramer, S.; Liang, L.; Maxwell, E.; Claveau, D.; Vaillancourt, J.;
Skorey, K.; Tam, J.; Wang, H.; Meredith, T. C.; Sillaots, S.; Wang-
Jarantow, L.; Ramtohul, Y.; Langlois, E.; Landry, F.; Reid, J. C.;
Parthasarathy, G.; Sharma, S.; Baryshnikova, A.; Lumb, K. J.; Pinho,
M. G.; Soisson, S. M.; Roemer, T. Sci. Transl. Med. 2012, 4, 126ra35. (c)
Haydon, D. J.; Stokes, N. R.; Ure, R.; Galbraith, G.; Bennett, J. M.;
Brown, D. R.; Baker, P. J.; Barynin, V. V.; Rice, D. W.; Sedelnikova,
S. E.; Heal, J. R.; Sheridan, J. M.; Aiwale, S. T.; Chauhan, P. K.;
Srivastava, A.; Taneja, A.; Collins, I.; Errington, J.; Czaplewski, L. G.
Science 2008, 321, 1673–1675.
(4) d’Ari, R. Biochimie 1985, 67, 343–7.
(5) Cordell, S. C.; Robinson, E. J. H.; Loewe, J. Proc. Natl. Acad. Sci.
U.S.A. 2003, 100, 7889–7894.
(6) PDB ID: 1OFU Cordell, S. C.; Robinson, E. J. H.; Loewe, J. Proc.
Natl. Acad. Sci. U.S.A. 2003, 100, 7889–7894.
r
10.1021/ol4010068
XXXX American Chemical Society

Figure 2. (Top) Core structures 1 (R1 = CH3, R2 = Bn) and 2
(R1 = CH3, R2 = i-Pr). (Bottom) Overlay of the truncated
T7-loop (green) and mimics (gray).
Figure 1. PDB ID 1OFU. Truncated T7-loop (green). Left to
right; Ile, Met, and Gly showing the interactions and distances
Despite the fact that scaffolds 1 and 2 differ at C3 and C4,
docking experiments showed that both of these compounds
adopted very similar orientations. Retrosynthetic analysis
of 1 revealed that it was an ostensibly easier target, given
that it has fewer stereogenic centers and more tractable
intermediates.9 We envisioned that the oxazolidinone ring
would be installed last from piperazine 3or possibly by an oxi-
dative cyclization from acyclic alkene precursor 4 (Scheme 1).
Alkene 4 could come from unsaturated aldehyde 5 and
diamines 6, derived from R-amino acids. We reasoned that
racemic amino acid precursors would double the number of
compounds that were assessed, once we established an assay
for the disruption of the FtsZ-SulA complex.
˚
(in A) between SulA and FtsZ.
As revealed by the cocrystal structure,6 the interaction
between the T7-loop of FtsZ and SulA is limited to only
a few residues, specifically Pro-204, Gly-205, Met-206,
Ile-207, andAsn-208.5 Upon closerinspection ofthecrystal
structure, we observed that Gly-205, Met-206, and Ile-207
were participating in key hydrogen bonding and hydro-
phobic interactions with the SulA surface (Figure 1). These
key interactions prompted us to utilize these amino acid
residues as a template for the design of T7-loop mimics.
The search for T7-loop mimics began by screening
potential ligands using fast rigid exhaustive docking
(FRED).7,8 Initially, we searched for molecules that fit
into the SulA pocket and placed the functional groups in
the same location as the T7-loop tripeptide shown above.
Although this process led to the identification of some
molecules that were synthetically unfeasible, it provided us
with a starting point for the design of a core structure that
was more tractable. After an extensive search of an in-
house library, we were pleased to find that core structures 1
and 2 mimicked the positioning of the three amino acid
side chains (Figure 2). These scaffolds showed tolerance of
functionalization at R1 and R2. Bulky substituents, such as
tert-butyl at R1, adopted an undesired orientation (not
shown). Less sterically demanding groups on the other
hand, such as methyl and ethyl, adopted the desired
orientation. Substitution at R2 was also explored, and we
found that a simple methyl group again resulted in an
unwanted binding orientation, while longer hydrophobic
chains such as isopropyl and benzyl were well-tolerated.
Scheme 1. Retrosynthesis of Compound 1, (R1 = CH2CH3,
CH3, R2 = Bn, i-Pr)
Synthesis of fragments 5 and 6 was accomplished in
good yield. The R,β unsaturated aldehyde 5 was synthe-
sized by the monoprotection of 1,4-butanediol (7) with
benzyl bromide in the presence of TBAI.10 TEMPO
(7) McGaughey, G. B.; Sheridan, R. P.; Bayly, C. I.; Culberson, J. C.;
Kreatsoulas, C.; Lindsley, S.; Maiorov, V.; Truchon, J.-F.; Cornell,
W. D. J. Chem. Inf. Model. 2007, 47, 1504–1519.
(8) Inquiries about the details of virtual screening should be directed
to Prof. Dean J. Tantillo (djtantillo@ucdavis.edu). See SI for computa-
tional details, including FRED results for diastereomers of 1aÀd.
(9) Preliminary synthetic studies (not shown) toward 2 revealed this
target to be less tractable than 1.
(10) Davis, J. M.; Truong, A.; Hamilton, A. D. Org. Lett. 2005,
7, 5405–5408.
B
Org. Lett., Vol. XX, No. XX, XXXX

oxidation11 followed by a Mannich reaction to install the
exomethylene moiety12 provided the desired aldehyde in
high yield (Scheme 2A). The synthesis of 6 was achieved by
the reaction between N-Boc protected amino acid 10 and
ethyl chloroformate to form a mixed anhydride intermedi-
ate, which upon treatment with benzyl amine yielded 11.13
Although various reducing agents such as borane (THF
and DMS) and LiAlH4 were used for the reduction of 11,
we obtained the best results with Red-Al. This method of
reduction yielded 6 in modest to high yield (Scheme 2B).14
Aldehyde 5 and amine 6 underwent reductive amination in
the presence of NaBH(OAc)3 to produce 12 in moderate
yield.15
Scheme 3. (A) Synthesis of 16aÀb and X-ray Crystal Structure
of 16a, R2 = Bn; (B) Conformational Model to Explain
Diastereoselective Formation of 13 from 12
Scheme 2. (A) Synthesis of R,β-Unsaturated Aldehyde 5;
(B) Synthesis of Amine 6
that, upon N-Boc removal, treatment of the intermediate
amine with I2 resulted in rapid and diastereoselective
formation of aziridines 13a and 13b from 12a and 12b,
respectively. The diastereoselectivity of this process is
estimated to be >85:15.17 The selective formation of
the cis diastereomer can be explained by a six-membered
ring transition state for an exocyclization in which the
two substituents are held in pseudoequatorial positions
(Scheme 3b). Although related reactions of amine-
appended alkenes have been reported18aÀd this is one of
the few examples in which high diastereoselectivity for the
formation of a fused aziridine is observed. Treatment of 13
with benzoic acid yielded 14, which, upon saponification
and addition of triphosgene, yielded compound 15 in good
yield.19,20 Removal of both benzyl-protecting groups with
Pd(OH)/H2 and catalytic HCl yielded 16 in 60À90% yield
(Scheme 3).
The synthesis of compounds 1aÀd was achieved in nine
steps from intermediate 12. We first attempted a cascade
aminocarboxylation of olefin 12 to oxazolidinone 15 by
employing conditions that were recently reported for the
direct aminocarboxylation of monosubstituted olefins.16
Unfortunately, this method failed for our disubstituted
substrate 12, resulting in no reaction or decomposition of
the starting material. Although a one-pot oxazolidinone
formation was not feasible, we were pleased to observe
The target bicyclic lactams were completed in a short
sequence from piperazines 16a and 16b. Amine acylation21
(17) Only one peak with the correct mass is observed by LCMS under
several conditions, suggesting high diastereoselectivity. Although the
complexity of the splitting patterns in the 1H NMR spectrum of the
unpurified reaction mixture prevents unequivocal confirmation of the
LCMS results, the selectivity appears to be at least 85:15.
(11) Packard, G. K.; Hu, Y.; Vescovi, A.; Rychnovsky, S. D. Angew.
Chem., Int. Ed. 2004, 43, 2822–2826.
(12) Pichlmair, S.; de, L. R. M.; Vilotijevic, I.; Paquette, L. A.
Tetrahedron 2006, 62, 5791–5802.
(13) Vaughan, J. R., Jr.; Osato, R. L. J. Am. Chem. Soc. 1951, 73,
5553–5.
(14) Voight, E. A.; Bodenstein, M. S.; Ikemoto, N.; Kress, M. H.
Tetrahedron Lett. 2006, 47, 1717–1720.
(15) Abdel-Magid, A. F.; Carson, K. G.; Harris, B. D.; Maryanoff,
C. A.; Shah, R. D. J. Org. Chem. 1996, 61, 3849–3862.
(16) Borsini, E.; Broggini, G.; Fasana, A.; Galli, S.; Khansaa, M.;
Piarulli, U.; Rigamonti, M. Adv. Synth. Catal. 2011, 353, 985–994.
(18) (a) Tranchant, M.-J.; Dalla, V.; Jabin, I.; Decroix, B. Tetrahe-
dron 2002, 58, 8425–8432. (b) Black, D. S. C.; Doyle, J. E. Aust. J. Chem.
1978, 31, 2247–57. (c) Yoo, S.-E.; Kim, J.-H.; Yi, K. Y. Bull. Korean
Chem. Soc. 1999, 20, 139–140. (d) Zhou, Y.; Murphy, P. V. Org. Lett.
2008, 10, 3777–3780. (d) McCormick, J. L.; Osterman, R.; Chan, T.-M.;
Das, P. R.; Pramanik, B. N.; Ganguly, A. K.; Girijavallabhan, V. M.;
McPhail, A. T.; Saksena, A. K. Tetrahedron Lett. 2003, 44, 7997–8000.
(19) Dangerfield, E. M.; Gulab, S. A.; Plunkett, C. H.; Timmer,
M. S. M.; Stocker, B. L. Carbohydr. Res. 2010, 345, 1360–1365.
(20) Goujon, J.-Y.; Gueyrard, D.; Compain, P.; Martin, O. R.;
Ikeda, K.; Kato, A.; Asano, N. Bioorg. Med. Chem. 2005, 13, 2313–2324.
(21) Deyine, A.; Poirier, J.-M.; Duhamel, P. Synlett 2008, 260–262.
Org. Lett., Vol. XX, No. XX, XXXX
C

Thelateintroductionofthe two sidechains inthissynthesis
will allow for straightforward variation of these two
groups in subsequent SAR studies.
Scheme 4. Completion of Bicyclic Lactams 1aÀd
In summary, we have designed and synthesized
heterocyclic small molecules that will serve as mimics
for the T7-loop of FtsZ. Successful mimicry of this
crucial binding region should provide the basis for
disruption of the interaction of this essential cell
division protein with its natural peptide inhibitor
SulA. Current studies are focused on the development
of biochemical and phenotypic assays to assess the
propensity of 1aÀd to disrupt the interaction of these
two bacterial proteins.
Acknowledgment. N.A.S. thanks The Alfred P. Sloan
Foundation for support. This work was supported by the
National Institutes of Health (R01-AI08093). The authors
acknowledge OpenEye for providing software used for
automated docking. Prof G. Broggini in the Department
ꢀ
ofScienceandHighTechnologyatUniversita dell’Insubria
(Italy) is acknowledged for helpful discussions.
Supporting Information Available. Characterization of
all new compounds, 1H and 13C NMR spectra, and dock-
ing results. This material is available free of charge via the
Internet at http://pubs.acs.org.
followed by mesylation of the alcohol moiety and displace-
ment withsodium methanethiolateyieldedthe desiredfinal
products 1aÀd in 30À40% over three steps (Scheme 4).22
(22) Ekegren, J. K.; Roth, P.; Kallstrom, K.; Tarnai, T.; Andersson,
P. G. Org. Biomol. Chem. 2003, 1, 358–366.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX