100
J. W. Black et al. / Bioorg. Med. Chem. Lett. 24 (2014) 99–102
chains.17 We also noted that the nature of the counterion was less
influential.
Presently, we have chosen to focus on a commonplace series of
bis-amine structures as our synthetic core, starting with N,N,N0,N0-
tetramethylethylenediamine (TMEDA), which is available at a cost
of approximately $20 per mol.19 Analogous structures with in-
creased linker distance between the amines, as well as those with
an increased number of amines such as spermidine and spermine
(Scheme 1), are also available at reasonable cost. The Clardy group
has been investigating protonated polyamine compounds for
biological activity, reporting the biofilm disrupting capability of
norspermidine derivatives.20 In this work we will detail the prepa-
ration of the compounds designated (m,2,n), where 2 is the linker
between two amines, and m and n are the number of carbons of
each n-alkyl hydrophobic side chain.
Scheme 2. Preparation of monoalkylated TMEDA derivatives (m,2,0).
While the preparation of symmetric (gemini)21 bis-alkylated
derivatives of TMEDA is well-precedented,22–24 and select antimi-
crobial activities have been reported,25,26 we found that the prep-
aration of asymmetric bis-alkylated TMEDA derivatives has
scarcely been published.27 Optimization of antimicrobial activity
for such asymmetric derivatives has remained uninvestigated.
Since we envisioned expedient syntheses of such compounds and
anticipated powerful antimicrobial activities thereof,16 we set out
to prepare a series of such asymmetric bis-alkylated TMEDA
derivatives.
Scheme 3. Yields of synthetic preparation of asymmetric bis-alkylated TMEDA
derivatives. Asterisk indicates a water-insoluble compound. See Supporting infor-
mation for full experimental detail of all compounds.
4,40-bipyridinium series, wherein we saw that we could optimize
efficiency by judicious choice of chain length.
Monoalkylation of TMEDA can be accomplished in a straightfor-
ward and atom-economical manner, with exposure of an excess
(2 molar equivalents) of the bisamine to a variety of alkyl bromides
in minimal-solvent conditions (Scheme 2; see Supporting informa-
tion for full synthetic detail of all compounds). Simple removal of
excess TMEDA in vacuo led to pure (>98%) monoalkylated crystal-
line products in nearly quantitative yields, without workup or
chromatography.
Subsequent exposure to a different alkyl bromide, again in high
concentration (ꢀ2 M in acetonitrile), followed by filtration, led to
good yields (43–92%) of the desired asymmetric biscationic amphi-
philes, as shown in Scheme 3. Recrystallization was performed as
necessary to ensure compound purity >98%, as determined by
NMR. It was found to be operationally advantageous to start with
the longer-chained monocationic compounds for installation of
the second chain, i.e., preparing (20,2,10) from (20,2,0) and not
from (10,2,0). This perhaps reflects the hygroscopic nature of the
smaller-chained compounds. It was noted that the largest com-
pound prepared, (20,2,18), suffered from poor water solubility; it
was thus not evaluated for bioactivity.
Finally, for comparative purposes, symmetrical TMEDA amphi-
philes were prepared according to literature precedent28
(Scheme 5). Thus, exposure of TMEDA to excess alkyl bromide
(3 equiv) in acetonitrile, followed by filtration, led to (n,2,n)
compounds, which were recrystallized as necessary.
With a series of 36 amphiphiles in hand, MIC values against the
Gram-positive Staphylococcus aureus and Enterococcus faecalis and
Gram-negative Escherichia coli and Pseudomonas aeruginosa were
determined by standard methods (Table 1). Comparison was made
to a commercially-available benzalkonium chloride solution. The
broth microdilution for determining the MIC of the compounds
was performed as previously reported;17 details are reported in
the Supporting information.
Examination of the antibacterial activity of the prepared amphi-
philes revealed strong antimicrobial activity in many cases, reach-
ing MIC levels as low as 1 lM, with 14 compounds showing
activity superior to that of benzalkonium chloride. Some clear
trends were also uncovered. First, monocationic compounds were
generally less effective at inhibiting the Gram negative bacteria
tested (E. coli and P. aeruginosa) as compared to the bis-alkylated
counterparts. For example, (18,2,0) highlighted this trend, display-
Additionally, two compounds with odd numbers of carbons in
one chain [(13,2,10) and (11,2,10)] were prepared from (10,2,0);
yields were comparable to the other preparations (Scheme 4).
These preparations were prompted by observations from our
ing MIC values of 2–4
63 M versus both Gram negative species.
In accordance with literature precedent,17 compounds with an
lM versus the Gram positive bacteria and
l
aggregate of 20–24 side chain carbons displayed optimal activity.
Six of these presented MIC values strictly in the single digit micro-
molar range. Accordingly, (16,2,8), (14,2,10), and (12,2,12) were
highly potent ‘24-carbon’ compounds; (14,2,8) and (12,2,10) were
bioactive ‘22-carbon’ compounds; (12,2,8) was the most active of
the compounds with 20 carbons in the side chains. We were sur-
prised, however, to see a relative uniformity of bioactivity, as many
of these strongly inhibitory compounds showed nearly identical
MIC values. Furthermore, while we generally saw preferential
activity of many compounds against the Gram positive bacteria
tested (S. aureus and E. faecalis), there was little differentiation in
activity for the strongest compounds between Gram positive and
Gram negative bacteria.
Two compounds with an odd number of side chain carbons
were prepared: (13,2,10) and (11,2,10), which allowed for exami-
nation of compounds with 23 and 21 carbons in the side chains.
Gratifyingly, this led to our most potent compound: (13,2,10),
Scheme 1. Commercially available polyamines, and some ammonium derivatives
thereof.