Development of a scalable synthesis of P7C3-A20, a potent neuroprotective agent

Article abstract of DOI:10.1016/j.tetlet.2013.06.024

A scalable synthesis of the neuroprotective agent P7C3-A20 is described. The synthesis has provided hundred-gram batches of the final compound for biological evaluation in rodents primates. The synthesis can be performed without chromatographic purificati

Full text of DOI:10.1016/j.tetlet.2013.06.024

Accepted Manuscript
Development of a Scalable Synthesis of P7C3-A20, a Potent Neuroprotective
Agent
Jacinth Naidoo, Christopher J. Bemben, Shawn R. Allwein, Jue Liang, Andrew
A. Pieper, Joseph M. Ready
PII:
S0040-4039(13)00980-5
DOI:
http://dx.doi.org/10.1016/j.tetlet.2013.06.024
TETL 43077
Reference:
To appear in:
Tetrahedron Letters
Received Date:
Revised Date:
Accepted Date:
7 May 2013
3 June 2013
6 June 2013
Please cite this article as: Naidoo, J., Bemben, C.J., Allwein, S.R., Liang, J., Pieper, A.A., Ready, J.M., Development
of a Scalable Synthesis of P7C3-A20, a Potent Neuroprotective Agent, Tetrahedron Letters (2013), doi: http://
dx.doi.org/10.1016/j.tetlet.2013.06.024
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered.
Development of a scalable synthesis of P7C3-
A20, a potent neuroprotective agent
Leave this area blank for abstract info.
Jacinth Naidoo, Christopher J. Bemben, Shawn R. Allwein, Jue Liang, Andrew A Pieper, and Joseph M. Ready

1
Tetrahedron Letters
journal homepage: www.elsevier.com
Development of a Scalable Synthesis of P7C3-A20, a Potent Neuroprotective Agent
Jacinth Naidooa, Christopher J. Bembenb, Shawn R. Allweinb, Jue Lianga, Andrew A Pieper,c and Joseph M.
Readya
aDepartment of Biochemistry, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd. Dallas, Texas, 75390-9038
bSouthwest Research Institute, P.O. Drawer 28510, San Antonio, TX 78228-0510
cDepartments of Psychiatry and Neurology, University of Iowa Carver College of Medicine, 200 Hawkins Ave, Iowa City, IA 52242
ARTICLE INFO
ABSTRACT
Article history:
Received
Received in revised form
Accepted
A scalable synthesis of the neuroprotective agent P7C3-A20 is described. The synthesis has
provided hundred-gram batches of the final compound for biological evaluation in rodents
primates. The synthesis can be performed without chromatographic purification of intermediates
or the final product.
Available online
2009 Elsevier Ltd. All rights reserved.
Keywords:
Neuroprotective
Synthesis
Carbazole
Fluorination
1. Introduction
low solubility of several intermediates may have complicated
large-scale synthesis. Here we describe a new synthesis that
The aminopropyl carbazole P7C3 was discovered in an
unbiased, in vivo assay for neuroprotective and pro-neurogenic
compounds.1,2 A medicinal chemistry campaign led to the
discovery of P7C3-A20, an analog displaying increased activity
and an improved toxicity profile compared to P7C3.3 In mice,
both compounds were found to 1) protect mature spinal cord
motor neurons from cell death in the G93A SOD1 transgenic
mouse model of amyotrophic lateral sclerosis (ALS),4 2) protect
dopaminergic neurons in the substantia nigra from toxicity
associated with 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine
(MPTP) in a model of Parkinson’s disease,5 3) protect neuronal
function and survival in a mouse model of traumatic brain
injury,6 and 4) enhance hippocampal neurogenesis.1 P7C3 has
also proven to be protective in a rat model of age-related
cognitive decline1 and a zebrafish model of retinal degeneration.7
requires no chromatography and has provided hundreds of grams
of P7C3-A20.
2. Results and Discussion
The initial synthesis of P7C3-A20 involved the fluorination of
a secondary alcohol in which an aniline NH was protected with a
4-nitrobenzene sulfonyl (4-Ns) group (3c, Scheme 1).
Fluorination on substrates possessing the aniline NH (3a) proved
problematic, but the Ns group endowed intermediates with poor
solubility and led to reaction mixtures requiring chromatographic
purification. Accordingly, our strategy to optimize the synthesis
of P7C3-A20 centered on the identification of an aniline
protecting group that would facilitate fluorination. To this end, a
variety of N-protected amino alcohols were prepared as shown in
To further study the neuroprotective effects and potential
therapeutic utility of the P7C3 class of chemicals, multi-gram
quantities of P7C3-A20 were required. However, our original
synthesis yielded only hundreds of milligrams of final product
and was not optimized for multi-gram preparations.3 The
requirement for multiple chromatographic purifications and the
———
Corresponding author. Tel.: 214-648-0313; e-mail: joseph.ready@utsouthwestern.edu

2
Tetrahedron
Scheme 1. Condensation of dibromocarbazole with
activated alcohol with the carbonyl of the protecting group.
epibromohydrin proceeded cleanly, and epoxide 1 was isolated in
high yield on large scale by trituration with ethyl acetate.
Finally, we explored conditions to remove the sulfonyl
protecting groups. Previously, we described removal of the 4-Ns
group with mercaptoacetic acid,3 but, as discussed above, the Ns-
containing intermediates were not ideal for large-scale
preparations. The mesyl (6d) and methoxybenezene sulfonyl (6f)
groups were resistant to all acidic, basic or reductive removal
conditions explored. However, treatment of triflate 6e with Red-
Epoxide 1 was opened with m-anisidine (2a) and a variety of
protected variants thereof. Optimal conditions involved a slight
excess of BuLi in dioxane under elevated temperatures. As
described below, the triflate 2e ultimately proved optimal. It was
prepared from m-anisidine and triflic anhydride in 77% yield on
an 18 g scale and 73% yield on a 510 g scale. Addition to
epoxide 1 was performed at 90 oC in the presence of 1.3 equiv of
triflate 2e and 1.8 equiv of BuLi. With less base we observed
substantial quantities of a side product that we characterized as
aziridine 4.8 On moderate scale, we isolated 23 g of 3e (55%)
through precipitation from the crude reaction mixture. An
additional 10 g (23%) was isolated chromatographically. On a
larger scale the reaction also proceeded smoothly, affording the
ring-opened product in 66% yield after trituration with
CH2Cl2/Hexanes (3:2 v:v, 850 g). Epoxide opening with most
other anisidine derivatives and anisidine itself generally
proceeded smoothly. However, 2-Ns anisidine (2b) provided
nearly equal quantities of the desired ring-opened product (3b)
and a cyclized side product 5, which presumably arises from
intramolecular SNAr substitution of the nitro group.9
Al (sodium bis(2-methoxyethoxy)aluminum hydride) at 60
oCcleaved the triflate but generated several side products
(Scheme 2). Acidification of the crude reaction mixture generated
the HCl salt of P7C3-A20, and allowed the subsequent removal
of unreacted starting material, dibromocarbazole, and
a
compound we assigned as an elimination product (9). This salt
was free-based with saturated bicarbonate solution, and the
resulting solid was triturated with CH2Cl2/Hexanes (3:7 v:v) to
remove des-brominated impurities (10). In this way, we obtained
69% yield of 98% pure material on a 10.6 g scale. On an 850 g
scale the free-based P7C3-A20 was initially precipitated from
ethyl acetate/hexanes (1:6 v:v). The resulting product was
recrystallized from hot ethanol, which afforded the product in
25% yield (168 g). On this larger scale, which was attempted
only once, a significant exotherm was noted such that the internal
o
temperature exceeded 90 C. Side products formed under these
With access to a variety of N-protected amino alcohols (3b-g)
we evaluated fluorination of the secondary alcohols with
morphoDAST (Table 1). While unprotected amino alcohol 3a
produced complex mixtures in the presence of mophoDAST,
several sulfonamides were fluorinated in high yields. In
particular, triflate 6e was isolated in excellent yield after simple
aqueous workup on both a 22 g and 905 g scale. In contrast, an
alkyl protected analog, 3g, yielded a complex mixture under
these reactions conditions. Likewise, the corresponding
acetamide (3h)10 and carbamate (3i) were converted to the O-
acetate (7) and cyclic urea (8), likely through substitution of an
conditions likely account for the lower yield on the 850 g scale
reaction. Accordingly, caution is warranted during the addition of
Red-Al on large scale.
The synthetic material made from the route describe herein
was evaluated for its effect on adult hippocampal neurogenesis.
Mice treated intraperitoneally with 20 mg/kg/d for 7 days were
found to have approximately twice the number of new
hippocampal neurons as untreated mice, consistent with
observations from the original synthetic route.1-3,11

3
3. Conclusion
The P7C3 class of compounds displays encouraging
neuroprotective activity. They have proven effective in several
animal models of neurodegenerative disease, and P7C3-A20
represents a valuable tool compound to explore the utility of this
class of chemicals. The scalable synthesis described here should
facilitate those endeavors.
Acknowledgments
This work was supported by the Edward N. and Della C.
Thome Memorial Foundation and the Welch Foundation (I-1612)
(to JMR). Additional support was received from The Hartwell
Foundation and NIH (NIMH R01 MH087986) to AAP.
Supplementary Data
Supplementary data associated with this article, including
experimental procedures can be found in the online version.
1.
Pieper, A. A.; Xie, S.; Capota, E.; Estill, S. J.; Zhong, J.; Long, J.
M.; Becker, G. L.; Huntington, P.; Goldman, S. E.; Shen, C.-H.;
Capota, M.; Britt, J. K.; Kotti, T.; Ure, K.; Brat, D. J.; Williams,
N. S.; MacMillan, K. S.; Naidoo, J.; Melito, L.; Hsieh, J.; De, B.
J.; Ready, J. M.; McKnight, S. L. Cell 2010, 142, 39.
McKnight S.L.; Pieper A.A.; Ready J.M.; De Brabander, J. US
Patent 8,362,277, 2010.
MacMillan, K. S.; Naidoo, J.; Liang, J.; Melito, L.; Williams, N.
S.; Morlock, L.; Huntington, P. J.; Estill, S. J.; Longgood, J.;
Becker, G. L.; McKnight, S. L.; Pieper, A. A.; De Brabander, J.
K.; Ready, J. M. J. Am. Chem. Soc. 2011, 133, 1428.
Tesla, R.; Wolf, H. P.; Xu, P.; Drawbridge, J.; Estill, S. J.;
Huntington, P.; McDaniel, L.; Knobbe, W.; Burket, A.; Tran, S.;
Starwalt, R.; Morlock, L.; Naidoo, J.; Williams, N. S.; Ready, J.
M.; McKnight, S. L.; Pieper, A. A. Proc. Natl. Acad. Sci. 2012,
109, 17016.
2.
3.
4.
5.
De Jesus-Cortes, H.; Xu, P.; Drawbridge, J.; Estill, S. J.;
Huntington, P.; Tran, S.; Britt, J.; Tesla, R.; Morlock, L.; Naidoo,
J.; Melito, L. M.; Wang, G.; Williams, N. S.; Ready, J. M.;
McKnight, S. L.; Pieper, A. A. Proc. Natl. Acad. Sci. 2012, 109,
17010.
6. Blaya1, M. O.; Bramlett, H. M.; Naidoo, J.; Pieper, A. A.;
Dietrich, W. D. submitted for publication.
7.
Asai-Coakwell, M.; March, L.; Dai, X. H.; DuVal, M.; Lopez, I.;
French, C. R.; Famulski, J.; De Baere, E.; Francis, P. J.;
Sundaresan, P.; Sauve, Y.; Koenekoop, R. K.; Berry, F. B.;
Allison, W. T.; Waskiewicz, A. J.; Lehmann, O. J. Hum Mol
Genet 2013, 22, 1432.
8.
9.
The mechanism by which excess BuLi prevents aziridination is
unclear. Formation of aziridine 4 may involve cyclization of an O-
triflate. In this scenario, BuLi may cleave any O-triflate prior to
cyclization, although no data exists to support this proposal.
Cleator, E.; Baxter, C. A.; O’Hagan, M.; O’Riordan, T. J. C.;
Sheen, F. J.; Stewart, G. W. Tetrahedron Lett 2010, 51, 1079.
10. Morcuende, A.; Ors, M.; Valverde, S.; Herradon, B. J. Org. Chem.
1996, 61, 5264.
11. Mice (n=6) treated with 20 mg/k/d P7C3-A20 prepared as
described herein were found to have 31.6±1.1 x106 new
neurons/mm3 in the dentate gyrus. Vehicle treated mice had
14.5±1.1 x 106 neurons/mm3. See reference 1 for experimental
details.