Organic Letters
Letter
Figure 3. Evaluation of T cell anti-CD3/CD28-mediated proliferation
in the presence of high doses of synthetic BraL 2. (A) Expanded dose
titration of BraL 2 on T cell proliferation as determined by quantification
of mean fluorescence intensity (MFI). (B) Inset depicting fluorescence
levels of unstimulated T cells (black histogram), bead-stimulated alone
T cells (red histogram), bead-stimulated T cells in the presence of 100
mM BraL 2 (blue histogram), or bead-stimulated T cells in the presence
of 100 nM BraA 1 (mustard histogram).
(MLR) and demonstrated an IC50 of approximately 63 nM.3
Here, in order to evaluate the efficacy of BraA 1 and BraL 2 in
human cells, an in vitro CD4+ T cell activation system was
developed based upon anti-CD3/CD28 bead stimulation and
CFSE fluorescent signal dilution as a means of monitoring
cellular proliferation. In this system, bead stimulation of normal
donor human CD4+ T cells activates the vast majority to
replicate (Figure 2a, red histogram) compared to unstimulated
cells (black histogram). Incubation of activated cells with BraL 2
(up to 400 nM) did not prevent proliferation (Figure 2b), while
cells incubated in the presence of natural BraA 1 were
significantly inhibited at concentrations as low as 25 nM (Figure
2c). Similar to previous reports, cyclosporine A treatment
demonstrated modest activity at concentrations approximating
100 nM in this assay system (Figure 2d).20 Dose titration analysis
of mean fluorescence intensity (MFI) as a function of compound
concentration estimates the IC50 of BraA 1 as approximately 65
nM, which is consistent with the previously reported results in
mouse cells (Figure 2e). BraL 2 was further tested at
concentrations between 100 nM to 100 μM in the assay system
described above in order to estimate an IC50 (Figure 3A). While
BraL 2 did not inhibit T cell proliferation at these concentrations,
the addition of 100 nM natural BraA 1 was found to be inhibitory
(p < 0.0064), reproducing our earlier observations (Figure 3B).
The inhibition of a CD3/CD28-dependent activation signal by
natural BraA 1 is quite significant since this is one of the most
potent in vitro activation signals available to immunologists.
Using only a CD3 activation, which induces some proliferation
and can be inhibited by CsA, activation by both a primary (via
CD3) and a secondary signal (via CD28) results in massive
proliferation and autocrine secretion of the T cell growth factor
IL-2 as well as its receptor.
Figure 2. Characterization of BraL 2 activity in inhibiting CD4+ T cell
proliferation. (A) Comparison of T cell cellular fluorescence (FL1-H) in
the absence (black histogram) or presence (red histogram) of anti-
CD3/CD28 bead stimulation. Flow cytometric evaluation of T cell
proliferation in the presence of increasing concentrations of (B) BraL 2,
(C) BraA 1, and (D) cyclosporine A (CsA). (E) Mean fluorescence
intensity (MFI) comparison of anti-CD3/CD28 bead stimulated CD4+
T cells in the presence of increasing concentrations of BraL 2 (open
circles), BraA 1 (closed circles), and CsA (open squares).
Although the anomers could not be separated at this stage by
silica gel chromatography, final HPLC purification of 2 was
effective in isolating the desired anomer. Deprotection of all of
the five acetate protecting groups of 15 was achieved in 53% yield
with a stoichiometric amount of LiOH·H2O in THF:H2O, which
left the benzhydryl ester and the significantly more stable
benzoate moieties intact, 16. It is important to note that these
conditions did not cause any racemization or elimination of the
protected β-methoxy amino acid derivative. We had already
reported on selective hydrolysis of the acetates without affecting
the hindered benzoate functionality of the disaccharide unit.18
Final deprotection of both the Cbz and benzhydyl esters using
catalytic hydrogenation with palladium on charcoal in methanol
proceeded without complication, and the BraA analogue, the
brasililogue BraL, 2, was isolated by preparative HPLC
chromatography in 79% yield as a single anomer as the TFA
salt. Thus, this compound was synthesized in only 10 steps from
the optically active diol 6 in fair overall yield (∼10%).
In conclusion, we have synthesized a BraA mimic in which the
natural tricyclic core structure was replaced with a chiral tetralin
unit. Although the BraA analogue, BraL, 2, like the natural BraA,
possesses the same amino acid side chain and the complex
disaccharide, it did not display the immunosuppressive activity of
BraA 1, shown here for the first time in human T cells. This result
demonstrates that merely approximating the distance between
the sugar and the amino acid portions of BraA is not sufficient for
bioactivity. Additionally, this work presents a model system that
The immunosuppressive activity of BraA was previously
studied using a one-way mouse mixed lymphocyte reaction
C
Org. Lett. XXXX, XXX, XXX−XXX