ACS Combinatorial Science
TECHNOLOGY NOTE
Scheme 1. Previous and Improved Approaches for Disulfide
disulfide 6 and a second symmetrical disulfide 7 containing the
hydrophilic ‘capping’ group (amino or dimethylamino). By using
an excess of the “capping” disulfide 7, the equilibrium would be
driven to favor formation of the desired disulfidefragment8bearing
the soluble cap. This material would then be separated chromato-
graphically (HPLC) from the much more hydrophilic diamine
homodimer 7 and any remaining fragment homodimer (6).
The two steps of the proposed synthetic route were explored
separately before attempting the desired one-pot reaction pro-
tocol. Hence, we started by coupling 2,6-dimethoxy nicotinic
Fragment Synthesisa
acid (9) to cystamine dihydrochloride salt (2 2HCl), using
3
1-ethyl-3-(3-dimethylaminopropyl) carbodiimide) hydrochlor-
ide (EDC) and 1-hydroxybenzotriazole (HOBt) in methylene
chloride (Scheme 2). The desired dimeric fragment 10 was
purified by flash chromatography and isolated in reasonable yield
(57%). Our initial attempts at effecting disulfide exchange
involved reacting purified disulfide 10 with an excess of the
dimeric capping reagent 2 in the presence of a free thiol initiator
(cysteamine HCl, 11, 0.5 equiv) and an excess of either diiso-
3
propylethylamine or sodium hydroxide, in DMF or DMSO. We
were happy to find that each of these procedures indeed
produced a reasonable mixture of hetero and homo disulfides
after stirring overnight. Three disulfide species (2, 10, and 12)
could be identified by LC/MS analysis and these were readily
separated via preparative HPLC. Ultimately we found that
increasing the amount of capping disulfide (to 5 equiv12) gave
improved conversion to the desired disulfide (12), that use of
0.1 equivalents of thiol initiator was sufficient, and that
triethylamine was as effective as diisopropylethylamine or
aqueous NaOH.
a The new one-pot method eliminates all work ups and requires only a
single reverse-phase chromatographic purification step.
Scheme 2. Synthesis of a Disulfide Fragmenta
To operationally simplify the procedure, we next explored the
feasibility of performing the exchange chemistry directly on the
crude reaction mixture obtained in the initial coupling. In fact,
this two step, one-pot procedure worked well after some
optimization of the reaction conditions. We switched to di-
methylformamide (DMF) for the coupling reaction, which in
turn required that a small amount of water be included to dissolve
the cystamine dihydrochloride salt. And although most of the
coupling reactions proceed to completion with only a small
excess of EDC, we found that using 2 equiv gave more consistent
results over a broader array of substrates and this, significantly,
allowed us to eliminate monitoring of individual reactions prior
to initiating the exchange reaction. The inclusion of water in both
the coupling and exchange reactions also insures that any excess
EDC is quenched prior to chromatographic purification. We also
switched from a cysteamine (ethylamino) cap (as in 5) to a
dimethylamino cap (as in 8). Not only does this improve
chromatographic separations, it eliminates a nucleophilic pri-
mary amine from the reaction sequence. A significant improve-
ment to the exchange step was realized by eliminating the
exchange initiator cysteamine 11, which is prone to air oxidation
over time, and therefore requires either careful handling or use of
fresh reagent for the most consistent results. In the improved
procedure, a reactive thiol is produced in situ by including the
water-soluble phosphine tris-(2-ethylcarboxy)phosphine hydro-
chloride (TCEP, 0.1 equivalent) in the exchange reagent solution
containing bis([2-(N,N-dimethylamino)ethyl]disulfide 7.
This modification eliminates the formation of multiple disul-
fide products as is observed when using cysteamine as the
initiator.
a Conditions: (a) 2 2HCl, EDC, HOBt, CH2Cl2; (b) 3 equiv. 2 2HCl,
3
3
0.1 equiv. 11 HCl, Et3N, DMF.
3
grow our existing disulfide fragment library, drove us to develop
just such a synthetic manifold. Herein, we report a very simple,
one-pot synthetic methodology(Scheme 1) thatcan beperformed
on the benchtop, in a parallel fashion without the use of air-free
techniques, significantly reducing the use of toxic or hazardous
materials, suchasTFA, and generating a verymodestwaste stream.
In the original synthesis of disulfide fragments, scrambling of
disulfide intermediates and side products was a significant problem
that had to be carefully monitored. We reasoned that this liability
might be used to advantage by altering the synthetic approach.
Hence, rather than coupling a partially protected (and non-
symmetric) disulfide (e.g., 3) to the acid building block, we
envisioned coupling the acid to both ends of a symmetric disulfide
(e.g., 2). The desired fragments would then be obtained by
promoting free exchange between the symmetrical fragment
The bottleneck in most parallel synthesis procedures is the
workup and/or purification of the final products. We therefore
206
dx.doi.org/10.1021/co200038g |ACS Comb. Sci. 2011, 13, 205–208