A lipophilic thioflavin-T derivative for positron emission tomography (PET) imaging of amyloid in brain

Article abstract of DOI:10.1016/S0960-894X(01)00734-X

The synthesis of a new lipophilic thioflavin-T analogue (2-[4′-(methylamino)phenyl]benzothiazole, 6) with high affinity for amyloid is reported. Intravenous injection of [¹¹C]-labeled 6 in control mice resulted in high brain uptake. Amyloid dep

Full text of DOI:10.1016/S0960-894X(01)00734-X

Bioorganic & Medicinal Chemistry Letters 12 (2002) 295–298
A Lipophilic Thioflavin-T Derivative for Positron Emission
Tomography (PET) Imaging of Amyloid in Brain
a,
b
b
b
Chester A. Mathis, * Brian J. Bacskai, Stephen T. Kajdasz, Megan E. McLellan,
a
c
b
a
Matthew P. Frosch, Bradley T. Hyman, Daniel P. Holt, Yanming Wang,
a
d
d
Guo-Feng Huang, Manik L. Debnath and William E. Klunk
aPET Facility, Department of Radiology, University of Pittsburgh, Pittsburgh, PA 15213, USA
bAlzheimer’s Research Unit, Massachusetts General Hospital, Charlestown, MA 02129, USA
cDepartment of Pathology, Brigham and Women’s Hospital, Harvard Medical School,
Boston, MA 02115, USA
d
Department of Psychiatry, University of Pittsburgh, Pittsburgh, PA 15213, USA
Received 24 July 2001; accepted 29 October 2001
0
Abstract—The synthesis of a new lipophilic thioflavin-T analogue (2-[4 -(methylamino)phenyl]benzothiazole, 6) with high affinity
1
1
for amyloid is reported. Intravenous injection of [ C]-labeled 6 in control mice resulted in high brain uptake. Amyloid deposits
were imaged with multiphoton microscopy in the brains of living transgenic mice following the systemic injection of unlabeled 6.
C]6 is a promising amyloid imaging agent for Alzheimer’s disease. # 2002 Elsevier Science Ltd. All rights reserved.
1
1
[
The development of small radiolabeled molecules for
use as biological markers of b-amyloid (Ab) deposits in
Alzheimer’s disease (AD) has been a goal of researchers
dyes (1–3) with high affinity for Ab and good brain
10
uptake in rodents. Here, we report the results of struc-
ture-brain uptake studies that show [N-methyl- C]2-[4 -
11
1
1
0
(methylamino)phenyl]benzothiazole ([ C]6) has char-
1
for several years. The advent of therapies that may
2
affect Ab deposition in AD brain has added new sig-
nificance to this pursuit. A variety of radiolabeled
acteristics more promising than previously reported
10,11
3
compounds.
To demonstrate that systemically
Congo red, Chrysamine-G, and other analogues have
been synthesized and evaluated as potential in vivo
positron emission tomography (PET) and single photon
emission computed tomography (SPECT) imaging
administered, unlabeled, fluorescent 6 (Fig. 1) can
resolve individual Ab plaques and cerebrovascular
1
2
amyloid in living transgenic mice, we employed a
novel high-resolution imaging technique, multiphoton
microscopy. Future studies will include imaging
amyloid load in transgenic mice using newly developed
probes of amyloid deposition in AD brain.4 While
many of these radiotracers displayed high affinity for
Ab, good brain uptake has been hindered by the pre-
ꢀ9
1
3
high-resolution microPET, a technology that will
provide a direct transition to PET imaging studies in
human subjects.
1
0
sence of highly polar functional groups. We reported
the synthesis and initial characterization of neutral
analogues of the amyloid dye thioflavin-T (ThT; Fig. 1),
and a neutral radioiodinated ThT analogue also has
1
1
been described.
Removal of the methyl group from the heterocyclic
nitrogen of ThT eliminated the positive charge from the
benzothiazole ring and provided a series of lipophilic
*Corresponding author. Tel.:+1-412-647-0736; fax:+1-412-647-0700;
e-mail: mathisca@msx.upmc.edu
Figure 1. Structural representations of the amyloid dyes thioflavin-T
and neutral ThT analogues 1–3 and 6.
0
960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00734-X

296
C. A. Mathis et al. / Bioorg. Med. Chem. Lett. 12 (2002) 295–298
Scheme 1. Synthesis of compounds 6 and [11C]6.
The syntheses of unlabeled14 and carbon-11-labeled15
are shown in Scheme 1 and were achieved using litera-
6
12% at 30 min. All of the radiolabeled plasma meta-
1
20
1
bolites of [ C]6 were polar species and were not expec-
ted to enter brain. To confirm this, mice were injected
with [ C]6, and their brains were removed at 2 and 30
1
0,16
ture-based methods.
Compound 6 was synthesized
1
1
by direct coupling of 4-(methylamino)benzoic acid (5)
with 2-aminothiophenol (4) in the presence of poly-
phosphoric acid at elevated temperature. The primary
2
1
min after injection and homogenized. Radiolabeled
species were extracted and assayed, and >95% of the
radioactivity in the homogenates was found to be
1
1
amine precursor for [ C]-labeling (8) was synthesized
via reduction of the intermediate 2-(4 -nitrophenyl)-
0
benzothiazole, which was prepared from the coupling of
11
unmetabolized [ C]6, indicating very low brain uptake
1
1
of radiolabeled metabolites of [ C]6.
4
-nitrobenzoyl chloride (7) with 4.
In other experiments, three 17-month-old PS1/APP
1
2
The results of in vitro competition binding assays using
3
transgenic mice were injected intraperitoneally (ip)
with a single dose of 10 mg/kg of compound 6 in a
solution of DMSO, propylene glycol, and pH 7.5 PBS
(10/45/45 v/v/v). Twenty-four hours later, multiphoton
17
[
N-methyl- H]6 and synthetic aggregated Ab(1–40)
10
fibrils are shown in Table 1 and demonstrate that the
neutral analogues 2 and 6 possessed higher affinity for
Ab(1–40) than the charged parent compound, ThT.
Octanol–water partition coefficients (Poct) of 2 and 6
2
2,23
fluorescence microscopy
high-resolution images in the brains of these living mice
using a cranial window technique.
images of compound 6 in a living PS1/APP mouse are
shown in Figure 2, and plaques and cerebrovascular
amyloid are clearly distinguishable. After imaging with
was employed to obtain
1
0
24ꢀ26
were determined at pH 7.4, and the logPoct values are
1
Typical in vivo
1
given in Table 1. The logPoct value of [ C]6 is near the
1
8,19
range of values predicted to result in optimal brain
uptake at early times following intravenous (iv) bolus
1
1
injection, while the logPoct value of [ C]2 is somewhat
beyond the optimum range of 1.5–2.5. Whole-brain
1
0
11
uptake values at 2 min post iv injection of [ C]2 and
1
1
[ C]6 in wild-type (control) female Swiss–Webster mice
are shown in Table 1, and the nearly 2-fold higher brain
1
1
11
uptake of [ C]6 over [ C]2 supported the logPoct-
based prediction. The egress of radioactivity from nor-
mal mouse brain tissue was also determined by
measurement of whole brain radioactivity at 30 min
following bolus iv injection (Table 1). Injection of the
1
1
less lipophilic compound, [ C]6, resulted in more rapid
clearance of radioactivity from the brain and provided a
Figure 2. Multiphoton microscopic images of amyloid deposits in liv-
ing PS1/APP mouse brain 24 h after injection of compound 6 (10 mg/
kg). These cortical images were obtained slightly lateral to the sagittal
suture and just posterior to the coronal suture through a cranial win-
dow preparation. Fluorescence was excited at 750 nm and collected at
380–480 nm. The views are projections ꢂ300 mm deep. The image on
the left reveals several bright plaques (arrows) and the faint outlines of
cerebrovascular amyloid (upper left of panel). The image on the right
shows a distinct amyloid-laden vessel. Note the absence of fluores-
cence emanating from within the vessels 24 h after the injection of
compound 6. The scale bar is 100 mm.
2
–30 min brain concentration ratio nearly 3-fold higher
1
1
than that of [ C]2, indicative of good clearance of free
and nonspecifically bound radioactivity from control
mouse brain tissue.
1
1
The peripheral metabolism of [ C]6 in control mice was
1
1
rapid, with unmetabolized [ C]6 representing 92% of
total plasma radioactivity at 2 min after injection and
Table 1. Comparison of the in vitro and in vivo properties of thioflavin-T and neutral thioflavin-T analogues 2 and 6
Compd
ThT
2
6
a
K
logPoct (pH 7.4)
2
3
i
(nM) Ab(1–40) fibrils
580 (ꢁ38)
10 (ꢁ0.5)
11 (ꢁ1)
2.7
12.9 (ꢁ4.2)
1.7 (ꢁ0.3)
7.6
b
b
0.57
3.4
min mouse brain uptake (%ID/g)c
0 min mouse brain uptake (%ID/g)c
—
—
—
7.6 (ꢁ0.3)
2.8 (ꢁ0.1)
2.7
b
b
Ratio of 2-to-30 min mouse brain uptakes
a
Values are means (ꢁSD) of three experiments.
b
c
10
Previously published values provided for reference.
Values are means (ꢁSD) of 3–5 measurements.

C. A. Mathis et al. / Bioorg. Med. Chem. Lett. 12 (2002) 295–298
297
2
7
6, one of the mice also received an iv injection of Texas
Red-dextran (Mr 70,000, Molecular Probes), a red
animals deposit large amounts of Ab in their brain. They are
derived by a cross between transgenic mice containing the
2
8
29,30
Tg2576 APPsw and the PS1(M146L)
13. Cherry, S. R. J. Clin. Pharmacol. 2001, 41, 482.
4. The procedures used to prepare 6 and precursor 8 for
subsequent radiolabeling were as follows: 2-Aminothiophenol
2.5 g, 20 mmol) (4) and 4-(methylamino)benzoic acid (5) (3.0
transgenes.
fluorescent dye that does not cross an intact blood–
2,23
2
brain barrier (BBB).
Subsequent imaging showed no
1
leakage of Texas Red-dextran across the cerebral vas-
culature, indicating that the BBB remained intact.
(
g, 20 mmol) were heated in polyphosphoric acid (10 g) at
ꢃ
The multiphoton microscopy studies demonstrated the
in vivo specificity of compound 6 for Ab in the brains of
living PS1/APP transgenic mice and provided the first
reported in vivo images of amyloid deposits in the brain
of transgenic mice following systemic injection of an
amyloid imaging probe. This technique likely will pro-
vide a valuable experimental complement to future
microPET imaging studies of [ C]6 in PS1/APP trans-
genic mice and assist in the interpretation of the lower
resolution (2 mm FWHM) PET studies.
1
80 C for 4 h. After cooling, the reaction mixture was poured
2 3
into 10% Na CO solution (50 mL). The precipitated product
was collected by filtration under vacuum and was purified by
flash chromatography (hexane/ethyl acetate, 9:1). Subsequent
de-colorization using activated carbon gave 1.9 g (40%) of 6.
1
H NMR (300 MHz, acetone-d
6
) d 7.92 (d, J=7.7 Hz, 1H),
.88 (d, J=7.7 Hz, 1H), 7.86 (d, J=7.9 Hz, 2H), 7.44 (dt,
=7.7 Hz, J =2.1 Hz, 1H), 7.33 (dt, J =7.7 Hz, J =1.9 Hz,
H), 6.70 (d, J=7.9 Hz, 2H), 2.85 (s, 3H). HRMS m/z calcd
7
J
1
1
1
1
2
1
2
+
12 2 1
for C14H N S (M ) 240.0721, found 240.0721. For 8: 4-
Nitrobenzoyl chloride (7) (1.5 g, 8.0 mmol) and 2-aminothio-
phenol (4) (1.0 g, 8.0 mmol) in anhydrous benzene (20 mL)
were stirred at room temperature for 16 h. Following extrac-
tion with ethyl acetate from water, the solvent was evaporated,
and the residue was purified by flash chromatography (hexane:
In summary, we report the synthesis of a new lipophilic
thioflavin-T analogue, compound 6, with high affinity
1
1
for Ab. Injection of [ C]6 in control mice resulted in
high brain uptake at early time points and relatively
rapid egress of radioactivity from normal brain tissue.
Injection of unlabeled 6 in PS1/APP mice resulted in the
visualization of both cerebral plaques and cere-
0
ethyl acetate, 85:15) to give 1.5 g (73%) of 2-(4 -nitro-
0
phenyl)benzothiazole as a pale-yellow solid. 2-(4 -Nitrophenyl)-
benzothiazole (0.10 g, 0.40 mmol) and tin(II) chloride dihydrate
(
N for 4 h. Ethanol was removed, and the residue was dissolved
0.20 g, 0.90 mmol) in ethanol (20 mL) were then refluxed under
1
1
2
brovascular amyloid deposits. Thus, [ C]6 is a promis-
ing lead radioligand for further development as a PET
amyloid imaging agent.
into ethyl acetate (20 mL). The solution was washed with 1 N
NaOH solution (3ꢄ20 mL) and water (3ꢄ20 mL) followed by
1
evaporation of the solvent to give 0.10 g (97%) of 8. H NMR
(
300 MHz, acetone-d ) d 7.7–8.2 (m, 4H), 7.4–7.6 (m, 1H), 7.2–
6
7.4 (m, 1H), 6.81 (d, J=8.4 Hz, 2H).
15. Typical radiolabeling procedures were as follows: Com-
Acknowledgements
pound 8 (1 mg) was dissolved in DMSO (0.4 mL), and 5 N
1
1
3
NaOH (10 mL) was added. High specific activity [ C]CH I
Financial support was provided in part by grants from
the Alzheimer’s Association (Pioneer Award to B.T.H.
and NIRG-00-2335) and the National Institute on
Aging (AG08487, AG15379, and AG18402).
3
1
was produced, and the entire amount was bubbled through
ꢃ
the solution. The reaction mixture was heated at 95 C for 5
min, and the crude reaction mixture was purified by semi-pre-
parative HPLC (Prodigy ODS-Prep) eluted with acetonitrile/
aqueous buffer (55/45 v/v, 50 mM triethylammonium phos-
1
1
phate buffer, pH 7.2). The fraction containing [ C]6 was col-
lected, and the product was isolated from the HPLC eluent
using a C8 SepPak Plus cartridge (Waters Corp.). The car-
tridge was washed with 10 mL water and eluted with 1 mL
References and Notes
1
2
. Klunk, W. E. Neurobiol. Aging 1998, 19, 145.
. Cutler, N. R.; Sramek, J. J. Prog. Neuro-Psychopharm.
1
1
ethanol to yield [ C]6 with a radiochemical purity >95% and
a specific activity of 55–110 GBq/mmol at the end of synthesis
(EOS). Only trace amounts of the dimethylated product were
produced, as the amine precursor 8 was in ꢂ100-fold stoi-
Biolog. Psych. 2001, 25, 27.
3
. Selkoe, D. J. Nature Biotechnol. 2000, 18, 823.
. Klunk, W. E.; Debnath, M. L.; Pettegrew, J. W. Neurobiol.
4
1
1/12
Aging 1994, 15, 691.
. Mathis, C. A.; Mahmood, K.; Debnath, M. L.; Klunk,
W. E. J. Labelled Compd. Radiopharm. 1997, 40, 94.
. Zhen, W.; Han, H.; Anguiano, M.; Lemere, C. A.; Cho,
C. G.; Lansbury, P. T. J. Med. Chem. 1999, 42, 2805.
. Dezutter, N. A.; Dom, R. J.; de Groot, T. J.; Bormans,
G. M.; Verbruggen, A. M. Eur. J. Nucl. Med. 1999, 26, 1392.
. Skovronsky, D. M.; Zhang, B.; Kung, M.-P.; Kung, H. F.;
chiometric excess over [
C]CH I, and this byproduct was
3
5
readily separated by the semi-preparative HPLC conditions.
1
1
Analytical HPLC of [ C]6 was performed using reverse-phase
chromatography [Phenomenex Prodigy ODS(3) 5 mm 250 ꢄ
4.6 mm column eluted with acetonitrile/aqueous buffer (55/45
v/v, 50 mM triethylammonium phosphate buffer, pH 7.2)].
6
7
0
11
The k value of [ C]6 under these conditions was 5.7, and
1
1
8
[ C]6 co-eluted with a sample of authentic 6. In addition,
normal-phase chromatography was utilized to confirm the co-
Trojanowski, J. Q.; Lee, V. M.-Y. Proc. Natl. Acad. Sci.
U.S.A. 2000, 97, 7609.
9. Agdeppa, E. D.; Kepe, V.; Liu, J.; Shoughi-Jadid, K.;
Satyamurthy, N.; Small, G. W.; Petric, A.; Vinters, H. V.;
1
1
elution of [ C]6 and a sample of authentic 6 [Whatman Par-
tisil 10 250ꢄ4.6 mm 10 mm eluted with ethyl acetate/n-hexane/
0
11
DEA (11.5/88/0.5 v/v/v)]. The k value of [ C]6 and 6 under
these conditions was 7.8. In-line HPLC detectors included UV
detectors (Waters Corp. Model 996) and radio-HPLC detec-
tors (Raytest Corp. Model Gabi).
Huang, S. C.; Barrio, J. R. J. Labelled Compd. Radiopharm.
2
001, 44, S242.
0. Klunk, W. E.; Wang, Y.; Huang, G.-F.; Debnath, M. L.;
1
Holt, D. P.; Mathis, C. A. Life Sci. 2001, 69, 1471.
1. Zhuang, Z. P.; Kung, M. P.; Hou, C.; Skovronsky, D. M.;
Gur, T. L.; Plossl, K.; Trojanowski, J. Q.; Lee, V. M.; Kung,
16. Shi, D. F.; Bradshaw, T. D.; Wrigley, S.; McCall, C. J.;
Lelieveld, P.; Fichtner, I.; Stevens, M. F. J. Med. Chem. 1996,
39, 3375.
1
3
H. F. J. Med. Chem. 2001, 44, 1905.
2. Doubly transgenic PS1/APP mice were utilized and these
17. The synthesis of [N-methyl- H]6 was accomplished in a
1
1
1
manner similar to that of [N-methyl- C]6, except that 5.4

298
C. A. Mathis et al. / Bioorg. Med. Chem. Lett. 12 (2002) 295–298
3
11
GBq of [ H]CH
3
I were utilized rather than [ C]CH
3
I. The
23. Christie, R. H.; Bacskai, B. J.; Zipfel, W. R.; Williams,
R. M.; Kajdasz, S. T.; Webb, W. W.; Hyman, B. T. J. Neu-
rosci. 2001, 21, 858.
24. Yuan, X. Q.; Smith, T. L.; Prough, D. S.; De Witt, D. S.;
Dusseau, J. W.; Lynch, C. D.; Fulton, J. M.; Hutchins, P. M.
Am. J. Physiol. 1990, 258, H1395.
3
final product ([N-methyl- H]6) was obtained in ꢂ10% radio-
chemical yield with a radiochemical purity greater than 97%
and a specific activity of 2.3 GBq/mmol.
1
8. Dishino, D. D.; Welch, M. J.; Kilbourn, M. R.; Raichle,
M. E. J. Nucl. Med. 1983, 24, 1030.
1
9. Gupta, S. P. Chem. Rev. 1989, 89, 1765.
0. The procedure used for plasma analysis was as follows:
25. Yuan, F.; Salehi, H. A.; Boucher, Y.; Vasthare, U. S.;
Tuma, R. F.; Jain, R. K. Cancer Res. 1994, 54, 4564.
26. The cranial window procedure utilized was similar to
2
Whole blood (0.7 mL) containing about 0.05 mL of heparin
was centrifuged at 13,000g for 2 min and 0.3 mL of the
supernatant plasma was removed. An equal volume of aceto-
nitrile was added, the mixture was vortexed for 2 min and
centrifuged at 13,000g for 2 min. Approximately 0.4 mL of the
supernatant solution was injected onto an HPLC system for
analysis (methods were identical to the reverse-phase analy-
2
4,25
other studies.
The PS1/APP mice were anesthetized with
Avertin (250 mg/kg ip), and maintained with ip injections as
required. The skin over the dorsal surface of the skull was
shaved and cleaned with 70% ethanol. The skin was removed
leaving a circular area approximately 15 mm in diameter. The
animal was fixed in a stereotaxic-like device and placed on a
heating pad. The skull was opened using a high-speed drill
(Fine Science Tools) with a fine-toothed burr (Stoelting) 0.6
mm in diameter. The underlying brain surface was kept moist
at all times with small pieces of gelfoam soaked in saline. The
dura was carefully removed from each hemisphere, while
leaving a strip at the midline. An 8.0 mm coverslip (Warner
Instrument Corporation) was carefully placed over the site.
Saline was slowly injected under the coverslip until the menis-
cus reached the edge of the glass, leaving a thin layer of saline
between the brain and the coverslip. A histocompatible cya-
noacrylic glue was applied over the juncture between the skull
and the edge of the coverslip. The animal remained on the
heading pad until it was fully recovered from the anesthesia.
27. Holcomb, L.; Gordon, M. N.; McGowan, E.; Yu, X.;
Benkovic, S.; Jantzen, P.; Wright, K.; Saad, I.; Mueller, R.;
Morgan, D.; Sanders, S.; Zehr, C.; O’Campo, K.; Hardy, J.;
Prada, C. M.; Eckman, C.; Younkin, S.; Hsiao, K.; Duff, K.
Nat. Med. 1998, 4, 97.
28. Hsiao, K.; Chapman, P.; Nilsen, S.; Eckman, C.; Harigaya,
Y.; Younkin, S.; Yang, F.; Cole, G. Science 1996, 274, 99.
29. Duff, K.; Eckman, C.; Zehr, C.; Yu, X.; Prada, C. M.;
Perez-tur, J.; Hutton, M.; Buee, L.; Harigaya, Y.; Yager, D.;
Morgan, D.; Gordon, M. N.; Holcomb, L.; Refolo, L.; Zenk,
B.; Hardy, J.; Younkin, S. Nature 1996, 96, 710.
30. Berezovska, O.; Frosch, M.; McLean, P.; Knowles, R.;
Koo, E.; Kang, D.; Shen, J.; Lu, F. M.; Lux, S. E.; Tonegawa,
S.; Hyman, B. T. Brain Res. Mol. Brain Res. 1999, 69, 273.
31. Dannals, R. F.; Ravert, H. T.; Wilson, A. A.; Wagner, H. J.
Int. J. Radiat. Appl. Isot. 1986, 37, 433.
1
5
0
tical HPLC system described above ). The k values of the
0
11
radiolabeled metabolites were 0.2–0.7 (k of [ C]6=5.7).
1. The procedures used to assess brain radioactivity were as
follows: Female Swiss–Webster mice (20–26 g) were injected
2
via lateral tail veins with 0.74–3.7 MBq (20–100 mCi) of high
1
1
specific activity (>50 GBq/mmol) [ C]6 contained in 0.1 mL
of isotonic saline solution (containing 5% ethanol). The mice
were anesthetized at 2 min and 30 min post injection and killed
by cardiac excision. The mouse brains were rapidly removed
and divided in half down the midline of the brain. A modified
Folch extraction32 was used to prepare the brain for analysis.
One half of the mouse brain (ꢂ0.2 g) was added to a 1 mL
solution of CHCl /MeOH (2/1 v/v) and homogenized (Poly-
3
tron, Kinematica AG) for ꢂ30 s. The homogenized solution
was filtered through a plug of glass wool. The glass wool is
then washed with an additional 9 mL of CHCl /MeOH (2/1 v/
3
v). An aqueous salt solution (2.5 mL of 0.04% CaCl , 0.034%
2
MgCl
was vortexed for 2 min followed by centrifugation for 2 min at
3,000g. Samples from the aqueous and organic phases (500
2
, 0.74% KCl) was added to the filtrate, and the mixture
1
mL each) and the entire glass wool plug were counted using a
gamma well-counter (Packard Instruments) to determine the
extraction efficiency (<5% of the total radioactivity remained
on the glass wool plug). An authentic standard of 6 was added
to an aliquot of the organic fraction, which was analyzed by
reverse-phase HPLC using methods described above. It was
assumed that all of the radioactivity in the aqueous fraction
<3%) was radiolabeled metabolites.
2. Bacskai, B. J.; Kajdasz, S. T.; Christie, R. H.; Carter, C.;
Games, D.; Seubert, P.; Schenk, D.; Hyman, B. T. Nat. Med.
001, 7, 369.
1
5
(
2
32. Folch, J.; Lees, M.; Sloan-Stanley, G. H. J. Biol. Chem.
1957, 226, 497.
2