G. Mengheres et al.
Bioorganic & Medicinal Chemistry Letters 34 (2021) 127761
activated microglia. Microglia, the resident macrophage cells of the
CNS, are activated by various noxious agents and release pro-
inflammatory cytokines and other mediators, causing inflammation. If
at the beginning of Aβ plaque formation microglia contribute to Aβ
plaque phagocytosis and clearance, continuous activation causes
microglia to surround the Aβ plaques and contribute to Aβ plaque for-
mation and subsequent neuronal damage.18 It was noted that upon >
95% microglia depletion for > 6 months in the 5xFAD mouse model of
Alzheimer Disease (AD), Aβ plaque formation was prevented. Thus,
removing the activated and dysfunctional microglia may prevent Aβ
plaque formation. Therefore, anti-inflammatory drugs that prevent
microglia activation and subsequent production of pro-inflammatory
cytokines and mediators may have a beneficial effect in reducing the
inflammation and neurodegeneration, and could be a potential treat-
ment for neurodegenerative diseases as the healthy microglia cells
would be able to engage in phagocytosis and remove the harmful agents.
As discussed, isoflavones and benzo-δ-sultams have been shown by
several studies to exhibit anti-inflammatory activity and neuro-
protective effects by inhibiting neuroinflammation in BV2 microglia, but
there was no clear relationship between the chemical structure of the
compounds and neuroprotective activity.19–21 Also, due to the
complexity of the neurodegenerative diseases, the mechanism by which
isoflavones and benzo-δ-sultams gain their neuroprotective effects is not
yet fully understood. In this work, we describe the synthesis of a series of
isoflavone/benzo-δ-sultam hybrids with potential anti-inflammatory
and neuroprotective activity. The products were tested for their ability
Fig. 2. The chemical structures of some bioactive benzosultam derivatives.
A second approach to isoflavone/benzo-δ-sultam hybrids was also
used. For this, the alkyne 13 was cross-coupled with various 2-haloben-
zenesulfonamide derivatives 14–18. Cyclisation of intermediates 19–21
in the presence of Cu(I) or Ag(I) allowed us to obtain the isoflavone/
benzo-δ-sultam hybrids 22–24 (See Table 1). The synthesis of in-
termediates 19–21, and hybrids 22–24, as can be seen in Table 1, was
influenced by several factors such as solvent, catalyst, halide, the
degassing process and the method of addition of alkyne 13. An opti-
mization reaction between 13 and 14/16 to obtain 22 was initially
attempted (Table 1, methods a-e). As catalyst, Pd(PPh3)2Cl2 worked
best, and as solvent, DMF. Also, the slow addition of 13 (dissolved in
DMF) facilitated the cross-coupling. Method (c) gave the best yield of
the desired product 22 (44% yield). When method (c) was applied to the
synthesis of compound 23, only starting material 15 was recovered.
Changing the solvent to THF, and the halide to the iodide 17 furnished
23 in 27% yield (method f). Changing the catalyst to 10% Pd/C, adding
PPh3 as ligand, and using 2-iodobenzenesulfonamide derivatives 16–18
(method g),28 led to the synthesis of both intermediates 19–21, and
hybrids 22–24 in different ratios. The separation of the intermediates
from the hybrids was possible for compounds 19 and 22, but not for 20/
23, and 21/24, these being obtained as mixtures. Treating the inter-
mediate 19 and the mixtures 20/23, 21/24 with AgNO3 and Et3N in
to inhibit NO and TNF-
cells.
α production in LPS-stimulated BV2 microglial
The synthesis of the hybrids was achieved through a cascade process
involving cross-coupling and regioselective 6-endo-dig cyclisa-
tion.12,22,23 First, the benzo-δ-sultam was built on the (B) ring of the
isoflavone (Scheme 1). Reacting amine 1 with 2-bromobenzenesulfonyl
chloride 2 in pyridine gave the previously unreported benzenesulfona-
mide 3. One-pot Sonogashira coupling of 3 with phenylacetylene 4 and
subsequent 6-endo-dig cyclization through hydroamination,22 led to the
synthesis of isoflavone/benzo-δ-sultam hybrid 6 alongside its precursor
5, the Sonogashira coupling product (method (b): 6, 10%; 5, 20%), both
of which are new molecules. The slow addition of the alkyne 4 (to avoid
potential homocoupling of the alkyne), and a better degassing of the
solvent and reaction mixture (to avoid potential oxidation of the cata-
lysts) led to compound 6 being obtained in greater yield (54%, method
(c)). It can be noted that the coupling step is the step that determines the
course of the reaction, so that improving the coupling reaction may lead
to a better yield. When THF was used as the solvent instead of DMF,
compound 6 was obtained only in trace amount.
EtOH,23 afforded the cyclised compounds 22, 23, 24 in excellent yields
29
–
(for the C N bond forming reaction).
The known alkyne 13 was synthesised by a modified literature
method which involved coupling of (trimethylsilyl)acetylene 25 or
(triethylsilyl)acetylene 26 with 3-iodo-7-methoxy-4H-chromen-4-one
10 to give the protected derivatives 27 and 28 (Scheme 3). Subsequent
deprotection of 27 and 28 by using TBAF and D-camphor-10-sulfonic
acid (CSA) in THF provided a high yield of the desired 3-ethynyl-7-
methoxy-4H-chromen-4-one 13.30 When TBAF was used without CSA,
and a mixture of methanol and THF (1:1) was used as solvent, the
desired product was not obtained. It was reported in the literature that
the deprotected derivative 13 undergoes hydrolysis and acetal forma-
tion in the presence of alcohols.31 Also when only TBAF in THF was
used, literature reported that the deprotected alkyne was obtained in
27% yield.32 D-camphor-10-sulfonic acid (CSA) was synthesised as pre-
viously reported.33
The known amine 1 was synthesised starting with the selective
protection of 2,4-dihydroxyacetophenone 7 at the 4-hydroxy group
using iodomethane to give 8,24 as shown in Scheme 1. The protected
compound 8 was treated with N,N-dimethylformamide dimethyl acetal
(DMF-DMA) to furnish the enamino ketone 9, which was subsequently
cyclised using pyridine and iodine to give the desired 3-iodo-4H-chro-
men-4-one 10, a known compound.25,26 Suzuki-Miyaura coupling of 3-
iodochromone 10 with 4-nitrophenylboronic acid 11 in the presence
of [Pd(dppf)Cl2] as catalyst and Na2CO3 as base, led to formation of the
nitro-isoflavone 12. The nitro-isoflavone 12 was reduced to amine 1 in
the presence of iron powder and NH4Cl in ethanol (Scheme 2).27
The required known sulfonamides 14–17 were readily obtained from
2-halobenzenesulfonyl chloride 2 or 29 and the corresponding amines
30 and 31 (Scheme 4). With the hope of obtaining a benzo-δ-sultam with
an NH, the previously unreported sulfonamide 18, with PMB as a pro-
tecting group, was also synthesised.
A mechanism for the Cu- or Ag-mediated cyclization of the alkynes 5,
and 19–21 to give the benzo-δ-sultam 6, and 22–24, can be proposed
based upon work by Barange et al., and starts with the coordination of
Fig. 1. The chemical structures of some bioactive isoflavones.
2