Trifluoromethyl ketone-based inhibitors of apoptosis in cerebellar granule neurons

Article abstract of DOI:10.1248/bpb.24.1335

A variety of aromatic trifluoromethyl ketone derivatives has been studied as inhibitors of apoptosis in cerebellar granule neurons (CGNs). Among them, α-trifluoromethyl diketone (2) and benzyl trifluoromethyl ketone (11) were found to be apoptosis inhibit

Full text of DOI:10.1248/bpb.24.1335

November 2001
Notes
Biol. Pharm. Bull. 24(11) 1335—1337 (2001)
1335
Trifluoromethyl Ketone-Based Inhibitors of Apoptosis in Cerebellar
Granule Neurons
b
Masami KAWASE,*,a Katsuyoshi SUNAGA,a Satoru TANI,a Masayuki NIWA,b and Toshihiko UEMATSU
Faculty of Pharmaceutical Sciences, Josai University,a Sakado, Saitama 350–0295, Japan and Department of
Pharmacology, Gifu University School of Medicine,b Gifu 500–8705, Japan.
Received May 23, 2001; accepted August 17, 2001
A variety of aromatic trifluoromethyl ketone derivatives has been studied as inhibitors of apoptosis in cere-
bellar granule neurons (CGNs). Among them, a-trifluoromethyl diketone (2) and benzyl trifluoromethyl ketone
(11) were found to be apoptosis inhibitors which can prevent a neurodegenerative disease. Compounds 2 and 11
showed neuroprotection effect on low K؉-induced apoptosis in CGNs. Furthermore, these compounds effectively
suppressed DNA fragmentation accompanied with apoptosis. The neuroprotection mode of 2 and 11 was not re-
lated to inhibition of caspase-3.
Key words trifluoromethyl ketone; apoptosis; cerebellar granule neuron; neuroprotection; DNA fragmentation
Apoptosis is a genetically programmed form of cell death in ethanol, and 200-fold concentrated compounds were
which is essential for normal development and health.1) Neu- added directly to the low Kϩ/serum-free medium. Control
rodegenerative diseases involve apoptotic cell death.2) Recent cultures were maintained in serum-free BME medium sup-
studies have suggested that the compounds which inhibit the plemented with 25 mM KCl (HK). After 20—24 h maintained
apoptosis process can prevent a neurodegenerative disease.3,4) in LK or HK, neuronal survival was determined by 3-(4,5-di-
Consequently, apoptosis inhibitors can be considered as po- methyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium
bromide
tential neurorescuing agents. In addition, specific apoptosis (MTT) assay. The assay relies on the ability of the mitochon-
inhibitors might be useful tools to investigate the molecular dria of live cells to reduce MTT to a water-insoluble blue for-
events involved in cell death.
mazan product. In brief, cultures were washed twice and
Trifluoromethyl ketones (TFMKs) are interesting com- BME medium containing 500 mg/ml MTT was added. After
pounds in the design of enzymatic inhibitors.5—7) The strong 90-min incubation at 37 °C, the reaction was stopped by
electron-withdrawing effect of the fluorine atom causes adding a lysing buffer [20% sodium dodecyl sulfate (SDS) in
TFMKs to form stable hydrates or hemiketals in aqueous so- 50% aqueous N,N-dimethylformamide solution, pH 4.7]. The
lution, whose tetrahedral geometry resembles the transition absorbance was measured spectrophotometrically at 570 nm
state of the water addition process to the carbonyl group of a after a further overnight incubation at 37 °C. The percent sur-
peptide substrate. Consequently, TFMKs are potent in- vival was defined as [absorbance (experimental-blank)/ab-
hibitors of a variety of serine esterases, juvenile hormone es- sorbance (HK-blank)]ϫ100, and the blank was the value
terase, mammalian carbonyl esterases, or antennal esterases taken from wells without cells. Results obtained were ana-
in insects.8)
lyzed by a one-way analysis of variance (ANOVA).
Continuing our efforts in the development of new syn-
DNA Fragmentation Analysis Total genomic DNA was
thetic methods9—12) and new biological activity13,14) of extracted and the extent of DNA fragmentation was analyzed
TFMKs, we now want to report on a neuroprotective effect of by agarose gel electrophoresis as described.16) After treat-
TFMKs on apoptosis induced by low extracellular Kϩ in cul- ment with RNase A (50 mg/ml) and Proteinase K (0.1 mg/ml)
tured cerebellar granule neurons (CGNs).
at 37 °C for 30 min and 0.2% SDS at 56 °C for 60 min, solu-
ble DNA was subjected to electrophoresis in a 1.2% agarose
gel and visualized by ethidium bromide staining.
MATERIALS AND METHODS
Preparation of Neutrophils Human neutrophils in
Cell Culture Culture enriched in granule neurons were blood were isolated by Dextran-Histopaque method as previ-
obtained from 8-d-old Wistar rats as described previously.15) ously described17) with minor modifications.18) Purification of
Cells were plated in basal medium Eagle (BME) supple- neutrophils was performed to minimize exposure of the cells
mented with 10% fetal bovine serum, 2 mM L-glutamine, to bacterial endotoxin. Purity of neutrophils was greater than
25 mM KCl, and 50 mg/ml gentamicin on 48-well plate or 60- 95%. Cell number was counted by a Coulter counter model
mmf dish coated with poly-L-lysine. Cells were plated at a ZM (Coulter, U.S.A.), and diluted in RPMI 1640 medium to
density of 2.5ϫ105/cm2. Cytosine-b-D-arabinofuranoside the final required concentrations and kept on ice until exam-
(AraC, 10 mM) was added to the culture medium 18—22 h ined.
after plating to prevent proliferation of nonneuronal cells.
Measurement of Caspase-3 Activity Neutrophils were
Treatment of Cultures and Assessment of Neuronal harvested after being exposed to TNF-a (100 ng/ml)/cyclo-
Survival After 12—13 d in vitro in 25 mM KCl medium, heximide (1 mg/ml) for 3 h, and resuspended in hypotonic
the culture medium was replaced with serum-free BME lysis buffer (25 mM HEPES, pH 7.5, containing 5 mM MgCl2,
medium containing 5 mM KCl and supplemented with L-glut- 5 mM EDTA, 5 mM EGTA, 5 mM dithiothreitol, 2 mM phenyl-
amine, gentamicin and AraC at the concentrations indicated methylsulfonyl fluoride, 10 mg/ml pepstatin A and 10 mg/ml
above (LK). The test compounds were dissolved and diluted leupeptin). Then cells were lysed by subjecting them to four
To whom correspondence should be addressed. e-mail: kawasema@josai.ac.jp
© 2001 Pharmaceutical Society of Japan

1336
Vol. 24, No. 11
carboxylic acid,26) and caspase-3 inhibitors.27)
cycles of freezing and thawing. After centrifugation
(15000ϫg, for 20 min at 4 °C) of the cell lysates, supernatant
was used as caspase-3 activated fraction. The fraction was in-
cubated with test compound at 30 °C for 10 min, then cas-
pase-3 substrate, as described previously.19,20) Caspase-3 ac-
tivities were expressed as the amount of liberated AMC (7-
amino-4-methylcoumarin) cleaved from Ac-L-aspartyl-L-glu-
tamyl-L-valyl-L-aspartyl (DEVD)-AMC, measured by using
spectrofluorometer (Fluoroskan, Dainippon Pharmaceutical
Co., Ltd., Japan).
Materials Trifluoromethyl ketones (4, 12) were syn-
thesized by reactions of mandelic acid or 3-phenyllacetic
acid with trifluoroacetic anhydride (TFAA).10) Compounds
(13, 14) were prepared by treatment of the corresponding
N-alkyl-N-methoxycarbonylphenylalanines with TFAA.11)
Compound 15 (mp 81—84 °C) was obtained in 91% yield by
the catalytic reduction of N-(2,6-dichlorobenzoyl)-2-trifluo-
roacetylpyrrolidine.9) Other compounds (1—3, 5—11) are
commercially available.
As a result of screening to obtain possible lead structures
bearing a trifluoroacetyl group, a-trifluoromethyl diketone
(2) and trifluoromethyl ketone (11) were identified as the
most potent inhibitor of apoptosis in CGNs. In an effort to
further define this novel inhibitory activity, a series of struc-
turally similar compounds was screened in order to deter-
mine the importance of the a-trifluoromethyl diketone
group. The trifluoromethyl ketones tested were classified as
follows: ketones (1, 11), a-diketone (2), b-diketone (3), a-
hydroxyketones (4, 12), a-amido ketones (13—15), and al-
cohols (10) (Table 1). Among them, a-diketone (2) and
PhCH2COCF3 (11) rescued most CGNs from death caused
by low Kϩ and its protection potency was similar to those of
Act-D and CHX.24)
With the COCOR series, the non-fluorinated ketones were
as follows: PhCOCOCH3 (5), PhCOCHO (6), PhCOCO2H
(7), PhCOCO2CH3 (8), PhCOCH2COCH3 (9). The protection
potency decreased in the order: 8Ͼ7ϾϾ5, 6, 9. It is noted that
non-fluorinated a-diketone (5) did not exhibit protection ac-
tivity. These data suggest the importance of the a-trifluo-
romethyl diketone moiety for the apoptosis-inhibition. The
EC50 values of 2, 4, 7, 8, and 11 were 1.20, 2.26, 2.03, 0.44,
and 1.73 mM, respectively.
RESULTS AND DISCUSSION
We screened our in-house library of TFMKs, most of them
previously synthesized in our laboratory.9—12) To our knowl-
edge, this is the first report on the screening of TFMKs for
their neuroprotective activity (Table 1 and Fig. 1). Potassium
(Kϩ) deprivation-induced apoptosis of CGNs represents one
of the best in vitro models of neuronal apoptosis.21,22) The
cultures of CGNs are formed by a homogenous population of
granule neurons which can survive up to 15 d when they are
maintained in fetal bovine serum-containing BME supple-
mented with 25 mM Kϩ.23) Apoptosis of CGNs can be in-
duced by lowering the extracellular Kϩ concentration from
25 to 5 mM, as evidenced by morphological and biochemical
methods.21,22,24) A switch to low Kϩ concentration decreases
viability of CGNs by Ͼ50% when measured after 24 h. Neu-
ronal survival was determined by the MTT assay. The apop-
tosis after Kϩ deprivation was reported previously to be
blocked by treatment with actinomycin D (Act-D),24) cyclo-
heximide (CHX),24) forskolin,25) C3-fullero-tris-methanodi-
A electrophoresis pattern of genomic DNA extracted from
LK-exposed CGNs for 24 h were drastically fragmented and
effectively suppressed by treatment of compounds 2 (30 mM)
and 11 (30 mM) (Fig. 2). The results suggest that these com-
pounds suppress the apoptosis induced by lowering extracel-
Fig. 1. Structures of TFMKs (1—15)
Table 1. Effects of Trifluoromethyl Ketones on Kϩ Deprivation-Induced Apoptosis of CGNs
Viability (%) normalized to high Kϩ (25 mM)
Compound
0
1 mM
3 mM
10 mM
30 mM
100 mM
1
2
3
4
5
6
7
8
9
53.8Ϯ4.8
50.0Ϯ2.1
43.4Ϯ2.8
48.9Ϯ4.1
51.8Ϯ3.4
50.8Ϯ3.9
50.2Ϯ3.6
50.7Ϯ3.0
53.2Ϯ4.5
50.0Ϯ4.6
50.7Ϯ2.7
47.0Ϯ6.5
45.5Ϯ8.7
45.5Ϯ8.7
47.0Ϯ6.5
43.4Ϯ2.8
56.1Ϯ4.2
54.1Ϯ5.5
42.1Ϯ10.6
52.8Ϯ4.8
52.3Ϯ4.7
53.4Ϯ6.4
54.5Ϯ6.0
60.4Ϯ2.8
54.0Ϯ4.0
57.1Ϯ6.5
58.3Ϯ3.2
51.9Ϯ0.2
51.5Ϯ8.1
45.4Ϯ0.0
46.9Ϯ5.1
45.8Ϯ2.9
55.5Ϯ5.1
65.7Ϯ7.5
48.2Ϯ6.2
58.7Ϯ4.3
51.4Ϯ4.8
52.7Ϯ6.0
52.9Ϯ7.7
60.3Ϯ4.3
48.5Ϯ5.3
52.6Ϯ6.0
70.5Ϯ5.6
54.4Ϯ5.9
47.4Ϯ9.2
48.7Ϯ10.5
47.1Ϯ4.2
45.7Ϯ4.1
55.5Ϯ4.3
71.4Ϯ6.2
51.5Ϯ4.7
62.6Ϯ3.8
53.0Ϯ5.1
49.6Ϯ10.0
60.9Ϯ5.0
65.4Ϯ5.9
40.3Ϯ3.4
57.3Ϯ5.5
74.8Ϯ5.7
53.8Ϯ9.1
58.3Ϯ11.0
46.0Ϯ9.7
45.4Ϯ3.1
9.4Ϯ22.4
56.7Ϯ4.3
72.0Ϯ5.5
52.8Ϯ11.4
58.4Ϯ1.9
49.8Ϯ5.1
36.9Ϯ7.1
61.2Ϯ5.7
61.9Ϯ4.7
24.9Ϯ1.0*
48.9Ϯ4.4
78.1Ϯ3.8
57.7Ϯ2.2
62.3Ϯ1.3
42.2Ϯ6.9
39.8Ϯ17.6
3.9Ϯ17.9
58.9Ϯ4.7
66.0Ϯ4.0
38.9Ϯ7.6
64.1Ϯ6.5
50.1Ϯ5.2
19.6Ϯ2.0*
57.6Ϯ3.9
61.7Ϯ4.9
23.0Ϯ0.8*
51.8Ϯ5.0
77.3Ϯ3.1
43.8Ϯ5.9
33.1Ϯ15.2
41.1Ϯ13.0
34.8Ϯ26.7
9.3Ϯ37.5
10
11
12
13
14
15
TPCK
* pϽ0.05 compared with the vehicle control, using ANOVA analysis.

November 2001
1337
mechanisms of apoptosis by cell type and physiological state.
A correlation between the structural and biochemical differ-
ences of 2 and 4 may serve as a model system for under-
standing apoptosis.
In conclusion, the present study demonstrated that a-tri-
fluoromethyl diketone (2) and 1,1,1-trifluoro-3-phenyl-2-
propanone (11) are neuroprotective in an in vitro model of
Kϩ deprivation-induced apoptosis of CGNs and may be at-
tractive lead compounds for further development as a neu-
rorescuing agent. The extensive structure-relationship in this
type of compound, including the inhibition mechanism of
apoptosis, will be reported in due course.
Fig. 2. Effects of Compounds 2 and 11 on Low Kϩ (LK)-Induced DNA
Fragmentation in Cultured Cerebellar Granule Cells
REFERENCES
1) Thompson C. B., Science, 267, 1456 (1995).
Lane 1, unexposed to LK medium; Lane 2, exposed to LK for 24 h; Lane 3 and 4, ex-
posed to LK and treated with compounds 2 (30 mM) and 11 (30 mM), respectively.
2) Pettmann B., Henderson C. E., Neuron, 20, 633—647 (1998).
3) Jacobson M. D., Current Biology, 8, R418—R421 (1998).
4) Davidson F. F., Steller H., Nature (London), 391, 587—591 (1998).
5) Walter M. W., Adlington R. M., Baldwin J. E., Schofield C. J., J. Org.
Chem., 63, 5179—5192 (1998).
6) Abouabdellah A., Begue J.-P., Bonnet-Delpon D., Kornilov A., Ro-
drigues I., Richard C., J. Org. Chem., 63, 6529—6534 (1998).
7) Boger D. L., Sato H., Lerner A. E., Austin B. J., Patterson J. E., Patri-
celli M. P., Cravatt B. F., Bioorg. Med. Chem. Lett., 9, 265—270
(1999).
lular Kϩ concentration in CGNs.
In agreement with previous reports,24) both Act-D
(0.5 mg/ml) and CHX (5 mg/ml) rescued most CGNs from
death caused by low Kϩ. As Act-D and CHX are thought to
act at the upstream of several apoptotic cascades, they may
exhibit powerful neuroprotective effects. Since 2 and 11
showed potent neuroprotection, it is interesting to know
where TFMKs effect this potency.
8) Begue J. P., Bonnet-Delpon D., Tetrahedron, 47, 3207—3258 (1991).
9) Kawase M., Miyamae H., Kurihara T., Chem. Pharm. Bull., 46, 749—
756 (1998).
Caspases, a family of cystein proteases, play a critical role
in execution of apoptosis and are responsible for many of the
biological and morphological changes associated with apop-
tosis.28,29) Recent report suggested that caspase-3 activity is
up-regulated and specific caspase-3 inhibitors moderately
suppressed cell death during low Kϩ-induced apoptosis in
CGNs.27) Thus, the caspase inhibitors, carboxybenzoxy-L-as-
10) Kawase M., Saito S., Kurihara T., Chem. Pharm. Bull., 48, 1338—
1343 (2000).
11) Kawase M., Hirabayashi M., Kumakura H., Saito S., Yamamoto K.,
Chem. Pharm. Bull., 48, 114—119 (2000).
12) Kawase M., Yuki Gosei Kagaku Kyokai Shi, 59, 755—766 (2001).
13) Kawase M., Harada H., Saito S., Cui J., Tani S., Bioorg. Med. Chem.
Lett., 9, 193—194 (1999).
14) Kawase M., Sakagami H., Kusama K., Motohashi N., Saito S., Bioorg.
Med. Chem. Lett., 9, 3113—3118 (1999).
15) Levi G., Aloisi F., Ciotti M. T., Thangnipon W., Kingsbury A., Balazs
R., “A Dissection and Tissue Culture Manual of the Nervous System,”
ed. by Shahar A., de Vellis J., Vernadakis A., Haber B., Alan R. Liss,
New York, 1989, pp. 211—214.
16) Hockenbery D., Nunez G., Milliman C., Schreiber R. D., Korsmeyer S.
J., Nature (London), 348, 334—336 (1990).
17) Boyum A., Scand. J. Clin. Lab. Invest., 97, 77—89 (1968).
18) Niwa M., Al-Essa L. Y., Kohno K., Kanamori Y., Matsuno M., Abe
A., Uematsu T., J. Immunol., 157, 4147—4153 (1996).
19) Nicholson D. W., Ali A., Thornberry N. A., Vaillancourt J. P., Ding C.
K., Gallant M., Gareau Y., Griffin P. R., Labelle M., Lazebnik Y. A.,
Munday N. A., Raju S. M., Smulson M. E., Yamin T.-T., Yu V. L.,
Miller D. K., Nature (London), 376, 37—43 (1995).
partyl-L-glutamyl-L-valyl-L-aspart-1-ylfluoromethane
(Z-
DEVD-FMK, caspase-3 inhibitor) and carboxybenzoxy-L-
valyl-L-alanyl-L-b-methyl-aspart-1-ylfluoromethane (Z-VAD-
FMK, non-selective caspase inhibitor) used at 200 mM, di-
minished death by 50—60%, whereas 200 mM acetyl-Tyr-Val-
Ala-Asp-aldehyde (Ac-YVAD-CHO, caspase-1 inhibitor)
was not protective in our assay systems, as previously
shown.27) These inhibitors are the peptides including the
amino acid sequence with an enzymatic cleavage site. There-
fore, it is useful if the simple compounds (2 and 11) have in-
hibitory activity on caspases. On the other hand, the chy-
motrypsin inhibitor N-tosyl-L-phenylalanyl chloromethylke-
tone (TPCK) had no effect on Kϩ-deprivation-induced apop-
20) Niwa M., Hara A., Kanamori Y., Matsuno H., Kozawa O., Yoshimi N.,
Mori H., Uematsu T., Eur. J. Pharmacol., 371, 59—67 (1999).
tosis at concentrations tolerated by CGNs (Table 1).24) There- 21) D’Mello S. R., Galli C., Ciotti T., Calissano P., Proc. Natl. Acad. Sci.
fore, the apoptosis-inhibitory activity of TFMKs was sus-
pected to be related to the inhibition of caspase-3. We exam-
U.S.A., 90, 10989—10993 (1993).
22) Miller J. M., Johnson E. M., Jr., J. Neurosci., 16, 7487—7495 (1996).
23) Levi G., Aloisi F., Ciotli M. T., Gallo V., Brain Res., 290, 77—86
ined the inhibitory activity of TFMKs against caspase-3 acti-
(1984).
vated fraction from neutrophils.20) However, the neuroprotec-
24) Schulz T. B., Weller M., Klockgether T., J. Neurosci., 16, 4696—4706
(1996).
tive activity of TFMKs was unrelated to the caspase-3 activ-
25) Galli C., Meucci O., Scorziello A., Werge T. M., Calissano P., Schettini
G., J. Neurosci., 15, 1172—1179 (1995).
26) Bisaglia M., Natalini B., Pellicciari R., Straface E., Malorni W., Monti
ity.
Previous studies from our laboratories have shown that a-
trifluoromethyl acyloins (4, 12) can induce apoptosis of
human cancer cells in vitro, whose induction was mediated
by activation of the caspase pathway.14) Two structurally re-
lated PhCOCOCF3 (2) and PhCOCH(OH)CF3 (4) differ in
their apoptosis-modulating activity. These distinct inhibitory
modes of action by 2 and 4 are interesting. It could be sug-
gested the existence of several pathways and/or regulation
D., Franceschi C., Schettini G., J. Neurochem., 74, 1197—1204
(2000).
27) Moran J., Itoh T., Reddy U. R., Chen M., Alnemri E. S., Pleasure D., J.
Neurochem., 73, 568—577 (1999).
28) D’Mello S. R., Kuan C-Y., Flavell R. A., Rakic P., J. Neurosci. Res.,
59, 24—31 (2000).
29) Budihardjo I., Oliver H., Lutter M., Luo X., Wang X., Annun. Rev.
Cell. Dev. Biol., 15, 269—290 (1999).