H. Shao et al.
Bioorganic & Medicinal Chemistry Letters 41 (2021) 128025
Table 1
10–13) improved microsome stability (~13 to 33% remaining). As in
the previous series, replacement of fluorine with chlorine further
improved metabolic stability. Specifically, 52% and 39% of compounds
Chemical structures and liver microsome stabilities of YM-08
1
5 (JG-23) and 16 were remaining, respectively. Compounds 14 and 17
had poor solubility and were thus excluded from further study (n/t;
Table 1). Based on these results, we chose the top 4 compounds and
measured their T1/2 in mouse liver microsome assays (Fig. 2). The
calculated T1/2 values from these experiments correlated with the single
time point measurements: the half-lives of compounds 5, 6 and 16 are
around 20 min, and JG-23 (15) has the longest T1/2 value (36 min).
Finally, we tested the ability of JG-23 to decrease total tau (t-tau)
levels in the HeLaC3 cell model. Cells were treated with JG-23 for 24 h
at the indicated concentrations, and the t-tau levels were analyzed by
Western blot. We found that JG-23 significantly decreased t-tau levels at
concentrations above 10 µM (Fig. 3a). Also, JG-23 did not activate
cellular stress, as determined by the constant levels of Hsp72 and Hsp90
in the treated cells. Next, we asked whether JG-23 might also reduce t-
tau levels in SH-SY5Y cells. This experiment was important because the
HeLaC3 model over-expresses 0N4R tau, while it is expressed at physi-
ological levels in the SH-SY5Y cells. Satisfyingly, we found that JG-23
had similar effects on t-tau, Hsp72 and Hsp90 in this model (Fig. 3b),
reducing t-tau by ~80%.
analogs.
Compound
R
1
R
2
Percentage atT30min
YM-08
H
2-pyridinyl
2-pyridinyl
2-pyridinyl
2-pyridinyl
2-pyridinyl
2-pyridinyl
2-pyridinyl
2-pyridinyl
2-pyridinyl
4-pyridinyl
4-pyridinyl
4-pyridinyl
4-pyridinyl
4-pyridinyl
4-pyridinyl
4-pyridinyl
4-pyridinyl
4-pyridinyl
–
12
19
5
1
2
3
4
5
6
7
8
9
3-F
4-F
5-F
6-F
3-Cl
4-Cl
5-Cl
6-Cl
H
12
38
37
13
17
6
1
1
1
1
1
1
1
1
0
3-F
4-F
5-F
6-F
3-Cl
4-Cl
5-Cl
6-Cl
33
23
13
30
n/t
52
39
n/t
1
2
3
4
5 (JG-23)
6
7
Conclusions
Allosteric inhibitors of Hsp70, which favor a pro-degradation
conformation of the chaperone, are potentially promising chemical
probes for understanding tau homeostasis. However, the poor metabolic
stability of earlier molecules, such as YM-08, had limited such studies.
Here, guided by the knowledge that P450-mediated metabolism occurs in
the benzothiazole ring, we focused on introducing halogens at the 3-, 4-,
the final products 1–17 in overall yields between 20 and 30% and purity
greater than 97%. The compounds were purified by flash chromatog-
1
raphy and characterized by LC-MS and H-NMR.
To probe which benzothiazole position might be most important for
microsome stability, we first measured the amount of unmodified
compound remaining after incubation with mouse liver microsomes for
5
- and 6-positions. Two series of compounds were designed and syn-
◦
28
3
0 min at 37 C, using dextromethorphan as a positive control . First,
thesized, revealing that a 4-chloro modified analog, JG-23, is 12-fold
more stable than YM-08. One of the interesting aspects of these find-
ings was that shifting the heteroatom in the distal pyridine generally
improves stability, despite the fact that metabolism does not occur in
this ring. It seems possible that the pyridine impacts either P450 binding
or oxidation activity, perhaps mediated by electronic effects through the
conjugated pi system. Importantly, these modifications did not seem to
affect target binding, as compound JG-23 retained the ability to pro-
mote t-tau degradation in two cellular models. Future work will focus on
improving potency and brain exposure, with the goal of creating a next-
generation Hsp70 probe for studying tau turnover in the brain.
we confirmed that YM-08 is rapidly metabolized in this experiment,
such that its levels were below the limit of detection at 30 min (Table 1).
Then, we found that introduction of fluorine at any of the 3-, 4-, 5- or 6-
positions on YM-08 (compounds 1–4) modestly increased this stability
(
~5 to 20% remaining after 30 min). The analogs containing chlorine at
the corresponding positions (compounds 5–8) were generally more
stable. For example, the 5 and 6 were approximately 2- to 3-fold more
stable than 1 and 2 (37 to 38%). In addition, across this series, the
compounds with substituents at the 3- and 4-positions were generally
better than those with halogens at the 5- and 6-positions.
Next, we examined the effects of switching the pyridine nitrogen
from the ortho (YM-08) to para position (compound 9–17). In early ex-
periments, we found that compound 9 was modestly more stable than
YM-08 (Table 1), so it seemed worth exploring this series in more detail.
Introduction of a fluorine at any of the four positions (compounds
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
Fig. 2. Microsomal stability of selected compounds. Dextrometorphan is used as a positive control.
3