Note: Descriptions are shown in the official language in which they were submitted.
L252~8~
- l - 20388-1585D
This is a divisional application of Application Serial
No. 503,659 filed March 10, 1986.
The subject matter of this application is closely rela-
ted to that of Canadian Patent Application Serial No. 452, 603.
An aspect of this divis.ional application is directed to
a compound of the formula 2 described hereinafte~, which is use-
ful as a ligand for preparing a complex claimed in the parent
application.
Another aspect of this application is directed to a
process for producing the compound of the formula 2. This process
comprises:
reacting a dione monoxime of the formula:
R \ ~ O
R N
OH
with a diamine of the formula:
R2 R2
R3 ~ R4
H2N NH2
in the presence of an acid dehydration catalyst, thereby producing
a diimine of the formula: R ~ / R4
R3 ~ ¦_`~R4
Rl ~ N N ~ R
R N ~N ~ R
OH OH
and then reducing the imino double bonds in the diimine.
` ~5248~
- la - 20388-1585D
It should be understood that the expression "this
invention includes the subject matters of the parent application
as well as this divisional application.
Technetium-99m (Tc-99m~ is the favoured radio-nuclide
for organ imaging and other forms of in vivo diagnosis. Complexes
of Tc-99m have been used for investigating most parts of the
body.
This invention relates to complexes of Technetium-99m
useful as diagnostic pharmaceuticals, and in particular to com-
plexes which are capable of crossing the blood-brain barrier and
being retained in the brain for a time to permit diagnosis.
European Patent Specification 123504 provides a lipophi-
lic macrocyclic complex of Technetium-99m useful as a diagnostic
radiopharmaceutical which can be formed by complexing in aqueous
solution Tc-99m pertechnetate under reducing conditions with an
alkylene amine oxime containing 2 or 3 carbon atoms in the
alkylene group, which group is unsubstituted or substituted, the
complex having a core with a zero net charge, containing an 0-H-0
ring closure bond, and being sufficiently stable for parenteral
administration and imaging by scintilation scanning, any alkylene
substituents present being of the kind useful for adapting radio-
nuclide ligands for body imaging applications. Preferred comple-
xes are believed tc have the formula:
5;~8~
~2 R3
H2~ C1~2
R4R5~ / `C R4R5
J~N/ 4N~C~,
0~ 0
where each R1, R4, and R5 is hydrogen or Cl to C12
alkyl, and each of R2 and R3 is hydrogen, hydroxyl, C1
to C12 alkoxyl, C1 to C22 hydrocarbon which may be
alkyl, alkenyl, alkaryl, aralkyl or aryl, or tertiary
amine with 1 to 20 carbon atoms, or R2 and R3 form,
together with the carbon atom to which they are
attached, a cycloaliphatic group which may be amine
substituted.
The present invention relates to a family o~
complexes, and the associated ligands, falling within
the scope of the invention of the aforesaid European
patent specification but not specifically described
therein, which complexes show interesting properties
particularly as regards brain retention, An important
member of the family is a lipophilic macrocyclic
complex, useful as a diagnostic radiopharmaceutical,
of Technetium-99m with a propylene amine oxime ligand
having the formula l:-
'~2~
. . ~
H3~ ~H3
~6~
5 H~C~NH HN~C~t3
H3C I Nl CH3 1-
H~ ~H
The compound of formula 1, together with some
other propylene amine oximes according to the aforesaid
European Patent Specification, have two asymmetric
carbon atoms and thus exist in the form of three
stereoisomers. Another important aspect of this
invention results from the unexpected discovery that
there are significant differences between the in vivo
properties of Tc-99m complexes of these stereoisomers.
The present invention provides a lipophilic macro-
cyclic complex, useful as a diagnostic radiopharmaceutical,
of Technetium-99m with a propylene amine oxime ligand
having the formula 2:-
~R4
R~N N~l~ R1
R~ I I ~ R
~H 0~
where R is H, alkyl, aryl, cycloalkyl, CF3, CONX2
6 ~ 3
R is alkyl, aryl, cycloalkyl or CF3,
or R and R together form part of a cycloalkyl
.
~Z5~8~L
ring,
each of R2, R3 and R4 is H, alkyl, aryl, cycloalkyl
or CF3, and
s H, CH3, C2H5 or C6H5~
wherein the ligand of formula 2 is in the form of
a single stereoisomer or of a mixture of 2 or more such
stereoisomers,
provided that, where R is methyl and R1 is methyl
and R2, R3 and R4 are hydrogen, a mixture
of 3 such stereoisomers contains an artificially high
concentration of one of them.
The ligands of general formula 2. include a pair
of groups R, a pair of groups Rl, a pair of groups R2,
a pair of groups R3 and a pair of groups R4, five pairs
in all. The two groups constituting each pair may be
the same or different. When they are the same (and
when R3 is the same as R4), as is the case with the
compound of formula 1., the ligands exist in the form
of three stereoisomers, by virtue of the two asymmetric
carbon atoms to which the groups R1 are attached.
When the two groups constituting one or more of the
five pairs are different from one another (or when R3
is different from R ), the stereochemistry becomes more
complex. ~or simplicity hereafter, the two groups
constituting each of the five pairs will be treated as
being the same. But it should be understood that the
invention is not so limited.
There follows, by way of example, a list of
compounds which are propylene amine oxime ligands of
3o formula 2 in which the two groups oonstituting each
pair are the same (except in the case of compound 11
where the R2's are different; and compound 12 wherein
both the R3's and the R4's are different):-
-- 5
~ . ._ .. . .
5 Cu~ d _ ~ R1 ~ R3,R~
1 CH3 CH3CH3 H
2 CH3 CH3 H H
3 CH3 C2H5 H H
4 C2H5 CH3 H H
3 7 CH3 H H
6 C2H5 CH3 CH3; H
7 CH3 CH3 C2H5 H
8 3 7 CH3 CH3 H
9 H CH3 H H
CH3 Ph H H
11 CH3 CH3 H,CH3 H
12 CH3 CH3 H CH3~H
13 i-C3H7 CH3 H H
Compound 2 is mentioned in European Patent Specification
123504. The remaining compounds and their Tc-99m
complexes are believed new. Additionally, single
stereoisomers, and mixtures of two or more such
stereoisomers including an artificially high
concentration of one of them, of all the compounds
(including compound 2), and their Tc-99m complexes are
believed new. All the compounds are capable of
forming complexes with radioactive and non-radioactive
isotopes of metals other than teohnetium, for example,
palladium and platinum, which complexes may have useful
properties for therapy and diagnosis. The following
diagram shows how compound 1, by way of example, exists
in the form of an optically inactive meso-
diastereoisomer and of optically active d- and
l-enantiomers.
~Z~2~
V ~ .,
I ~ ~
~ rZ ~--o
S~Z Z - o
I ~ 5_
~U ~ O
~ r;z 2--o
Z~ o
r2 a ~
D \ ~1/\_S z :r
o
~, ;~f 2--
r
~ ~Z ~
.~
~2~z~a8
-- 7 --
The isomerism arises because the 3- and 9- carbon
atoms are asymmetric. The meso-diastereoisomer has
distinctly different properties from the dl-diastereo-
isomer (a racemic mixture of the d- and l-enantiomers).
For example, they have different melting points, and
their retention times on HPLC systems can also differ.
The Tc-99m complexes from the two diastereoisomers have
markedly different in vivo properties (% i.d. to be
taken up in brain of rats). From these observations,
it is to be expected that similar differences will
occur in other compounds of formula 2 shown above;
i.e. the two diastereoisomers derived from any one
compound will have different physical properties, and
the Tc-99m complexes derived from the two diastereo-
isomers (from a single compound) will display differencesin in vivo distribution.
It has further been determined that the d- and l-
enantiomers of compound 1 have properties (physical and
biological) which differ from each other, and from a
racemic mixture of the two enantiomers, and from the
meso-diastereoisomer. Again, it i9 to be expected
that similar differences will occur in other compounds
of formula 2 shown above.
The different in vivo properties of Tc-99m
complexes of the two diastereoisomers is surprising.
These complexes are thought to cross the blood-brain
barrier by a passive diffusion mechanism. The
ability of molecules to diffuse through membranes is
related in a positive way to lipophilicity and in a
3o negative way to molecular weight. On this basis only
relatively small differences in the amount of
radioactivity in the brain would be anticipated since
the molecular weights of the two isomers are identical
and the lipophilicities of the Tc-99m complexes are
thought to be very similar. Consequently the
magnitude of the difference in brain uptake
~25~
of the two isomers is surprising - a factor of
2.3 is involved in the case of compound 1. The better
of the two isomers is the dl-isomer.
Similarly, there is a surprising difference in
brain uptake of the d- and l-enantiomers of compound 1
- a factor of 1.6 has been found in studies with rats.
~ ach of the three stereoisomers so far
mentioned can itself exist in three different isomeric
forms by virtue of the restricted rotation about their
C=N bonds of the two oxime groups. The isomers may
have different physical properties (m.p., b.p.) and can
be separated by chromatographic techniques (TLC and
HPLC). Their interconversion is generally facile and
is catalysed by mineral or Lewis acids, bases or metal
ions.
Compound 1 has two oxime groups so 3 isomers are
possible:-
25 ~ ~ ~N~ ~A~
N N N ~1~ N ~J
I
o~ ~ OH
, EE EZ Z~
j 35
~25:~8~L
Note that EZ is identical to ZE. The sameconsiderations apply to both 'd' and 'l' isomers, i.e.
each has three (EE, ZZ and EZ) oxime isomers. The
thermodynamically preferred isomer is expected to be EE
since that has the maximum separation of the bulkier
groups. HPLC analysis of compound 1 (Example 11A)
displays two major peaks, identified as the EE oxime
isomers of the meso- and d,l-diastereoisomers, plus
four minor peaks, provisionally assigned as the EZ and
ZZ oxime isomers of each diastereoisomer. It is
probable that the EE, EZ and ZZ isomers may form Tc-99m
complexes having different biodistribution properties.
The propylene amine oxime ligands may be prepared
by standard chemical routes, as generally described in.
the aforesaid European Patent Specification 123504.
A preferred preparative route is described below in
Example 1. (When the two groups R, and/or the two
groups R1, are different, the ligands can be prepared
- by an analogous route involving: reaction of a mono-
protected propanediamine with one molar equivalent of a
suitable dione monoxime; reduction of the resulting
imine; deprotection; reaction of the resulting primary
amine with a different dione monoxime followed by
reduction of the resulting second imino group.)
The compounds are generally obtained in the form
of a mixture of all three isomers. The meso-isomer
can be separated from the dl-mixture by standard
techniques such as repeated fractional crystallisation
from a solvent in which their solubilities are
3o different; we have used acetonitrile and ethyl acetate
with success. The analytical or preparative
separation of the isomers can also be effected by HPLC.
The d- and l-enantiomers can be separated by standard
techniques involving the use of optical isomers of an
organic acid such as tartaric acid. Separation
techniques are detailed in Examples 14 to 20 below.
5~
,
- 10 -
The complexation reaction between the propylene
amine oxime ligand and pertechnetate (Tc04 from
generator eluate) may be carried out in âqueous or
aqueous/organic solution under reducing conditions.
Stannous salts are convenient reducing agents, but other
reducing agents are well known for this type of
reaction and can be used. Since the complexes of this
invention contain Tc-99m bound rather strongly, they
can alternatively be prepared by a process of ligand
exchange. The preparation of Tc-99m complexes by
reducing pertechnetate in the presence of a complexing
ligand is well-known; the conditions for such general
reactions are also well-known and can be used in the
particular instance of this invention. These complexes
are presently believed to have the structure 3:-
R3~R4
R3~1 R4
. R~ 01~N,~ 3
R N Pll R
O---H--O
The Tc-99m complex of compound 1 (an unseparated dl/meso
mixture) displays:
1) Cood uptake in the brain of rats and humans.
2) Slow removal of radioactivity from brain in
humans t âllowing tomographic imaging to be performed
with conventional rotating gamma camera equipment.
Its advantages over the Tc-99m complex of compound 2
are:-
~L25;~
1) Far better in vitro stability (see Example 21
below).
2) Better blood and tissue background clearance
of activity in humans.
3) The complex of compound 2 may not be suitable for
radiopharmaceutical applications because of rapid in
vitro degradation, but the complex of compound 1 is
ideally suited as it displays only slow in vitro
degradation. (See Example 26 below).
The complex of compound 1 (dl-isomer) is
surprisingly superior to the complex of the meso-
isomer, so far as brain retention and brain imaging are
concerned. ~y contrast, the complexes o~ compound 2
(dl-isomer) and compound 2 (meso-isomer) do not show
~5 any marked difference in brain uptake (see Example 2
below).
The complex of compound 1 (both the unseparated
dl/meso mixture and the dl-isomer) also show other
interesting and unexpected properties:-
- They are good tumour blood flow agents
tunpublished article of V.R.McCready et al submitted to
Journal of Nuclear Medicine).
- They are good agents for labelling blood cells,
particularly leucocytes. (See Example 27 below3.
- The biodistribution properties of compounds 3
to 13 (see Example 22 below) also indicate various
uses for these compounds as diagnostic pharmaceuticals,
both in the form of their unseparated dl/meso isomer
mixtures and in the form of their separated isomers.
- They are useful for myocardial perfusion
imaging.
The following Examples illustrate the invention.
:
i2~8~l.
- 12 -
Example 1
Preparation of the Compound 1
N2N IJH2 ~N N~
OH OH OH OH OH
1A. 4,8-Diaza-3,6,6,9-tetramethyl-3,~-undecadiene-
2,10-dione bis oxime
2,3-Butanedione monoxime (11.66g, 115.4mmol) was
dissolved in benzene (50cm3) containing acetic acid
- (75~1), and the solution was brought to reflux in an
apparatus fitted with a Dean-Stark trap and in a
nitrogen atmosphere. To this was added a solution of
2,2-dimethyl-1,3-propanediamine (5.00g, 5.B8cm3, 49mmol)
in benzene (100cm3) over a period of 5 hours. The
resulting yellow-brown solution was refluxed for a
further 16 hours under nitrogen, then allowed to cool
to room temperature. The resulting solid was
filtered off under suction and washed with a little
cold (-40C) acetonitrile giving the product as a fine
white powder. Aftér drying under high vacuum for 2
hours 7.9g t60% yieid) of product was obtained, m.p.
131C 134C. The product thus obtained contains a
trace (ca. 2% - 5%) of starting ketooxime (as seen by
H NMR), but this can be removed completely by a single
recrystallisation from benzene, giving a product
melting at 132C - 135C. NMR(1H,60MHz,CDCl3):
- ~3.3(4H,brs,CH2N), 2.1(6H,s,CH3-C=N),
2.0 (6H,s,CH3-C=N), 1.1(6H,s,(CH3)2C)ppm.
~LZ5;~8~
- 13 -
1. 4,8-diaza-3,6,6,9-tetramethylundec _e-2,10-dione
bis oxime
The diimine (75g, 287mmol) was slurried in 95%
aqueous ethanol (690cm3) at 0C. Sodium borohydride
(10.9g, 287mmol) was added in portions over ~ hour and
the mixture stirred at 0C for 2 hours. Water
(230cm3) was added and the mixture stirred well for a
further 2 hours. The ethanol was removed in vacuo
and more water (140cm3) added. The pH was adjusted to
around 11 then the resulting white solid was filtered,
washed with a little water and dried in vacuo giving
the crude product. Double recrystallisation from hot
acetonitrile gave pure product 62.5g, (80%) m.p. 144C.
_ 145C.
NMR (1H,200MHz,DMS0):
~10.24(2H,s,OH)3.13(2H,q,CHMe), 2.12(14,m,CH2N),
1.65(6H,s,MeC=N), 1.07(6H,d,CHMe),0.78(6H,s,CMe2)ppm.
Examples 2 to 13
Preparation of 2,3-pentanedione-3-oxime.
Methyl nitrite was bubbled, at a rate sufficient
to maintain vigorous reflux, into a well-stirred
mixture of 2-pentanone (102g) ether (400cm3) and
concentrated hydrochloric acid (15cm3). The methyl
nitrite gas was generated by the dropwise addition of a
mixture of concentrated sulphuric acid (100cm3) and
water (95cm3) onto a stirred slurry of` sodium nitrite
(112g), methanol (66g) and water (75cm3). ~fter the
addition was complete the mixture was neutralised with
3o saturated aqueous sodium bicarbonate (32g in 300cm3).
The ether layer was separated and the aqueous layer
extracted with more ether. The combined organic layers
were dried and concentrated _ vacuo giving a yellow
oil which crystallised on standing. Recrystallisation
- 35 from hot hexane gave pure product (67g), m.p. 54-5C.
The following were prepared in a similar manner
~SZ~8~
- 14 -
(yield and melting point given):-
2,3-Pentanedione-2-oxime (46%, mp 59-61C)
2,3-Hexanedione-3-oxime (64%, mp 42-3C)
1,2-Propanedione-1-oxime (18%, mp 61-5C)
2-methyl-3,4-pentanedione-3-oxime (45% m.p. 72-
75C).
Preparation of Compounds 2A to 13A
The following diimines were prepared in a similar
manner to Example 1 (yields and melting points given):-
- 4,8-Diaza-3,9-dimethyl-3,8-undecadiene-2,10-dione
bis oxime (2A), (51%, m.p. 91-2C).
- 4,8-Diaza-3,9-diethyl-3,8-undecadiene-2,10-dione
bis oxime (3A), (68%, m.p. 76-8C).
- 5,9-Diaza-4,10-dimethyl-4,9-tridecadiene-3,11-dione
bis oxime (4A), (31%, m.p. 143.5-144.5C).
- 6,10-Diaza-5,11-dimethyl-5,10-pentadecadiene-4,12-
dione bis oxime (5A), (30%, m.p. 130-1C.
- 5,9-Diaza~4,7,7,10-tetramethyl-4,9-tridecadiene-
3,11-dione bis oxime (6A) (49%, m.p. 79-82C).
- 4,8-Diaza-6,6-diethyl-3,9-dimethyl-3,8-undecadiene
-2,10-dione bis oxime (7A) (67%, m.p. 168-9C).
- 6,10-Diaza-5,8,8,11-tetramethyl-5,10-
pentadecadiene-4,12-dione bis oxime (8A) (32%,
m.p. 91-3C).
- 3,7-Diaza-2,8-dimethyl-2,7-nonadiene-
1,9-dione bis oxime (9A) (87%, m.p. 124C, dec)
- 4,8-Diaza-3,9-diphenyl-3,8-undecadiene-2,10-dione
bis oxime (lOA) (59%, m.p. 169-172C).
- 4,8-Diaza-3j6,9-trimethyl-3,8-undecadiene-2,10-
dione bis oxime (11A), (70%, oil).
- 4,8-Diaza-3,5,7,9-tetramethyl-3,8-undecadiene-
2,10-dione bis oxime (12A), (98%, oil).
- 5,9-Diaza-2,4,10,12-tetramethyl-4,9-tridecadiene-
3,11-dione bis oxime (13A), (27%, m.p. 120C).
~5;~8~
Preparation of Compounds 2 to 13
The following ligands were prepared in a similar
manner to Example 1 (yields, melting points and NMR
data given):
- 4,8-Diaza-3,9~dimethylundecane-2~10-dione bis
oxime (2), (18%, m.p. 119-122C).
NMR (1H,200MHz, d6-DMSO): 610.2(2H,s,OH),
3.2(2H,q,CH),2.3(4H,t,CH2N) 1.65(6H,s,N=CMe),
1.44(2H,m,CH2),1.04(6H,d,CH)ppm.
- 4,8-Diaza-3,9-diethylundecane-2,10-dione bis oxime
~3), (76%, oil)
NMR(1H,200MHz,d6-DMSO:610.28(2H,s,OH),2.90
(2H,m,CH),2.3(4H,brm,CH2N),2.19(4H,q,CH2Me),1.61
(6H,s,N=CMe),1.4(2H,m,CH2),0.98 and 0.76
(6H,t,CH3)ppm
- 5,9-Diaza-4,10-dimethyltridecane-3,11-dione bis
oxime (4), (59%, oil).
NMR(1H,200MHz,CDCl3): ~3.33(2H,q,CH),2.61
(4H,q,N=CCH2),2.18(4H,m,CH2N), 1.68(2H,m,CH2),1.24
and 1.13(12H,m,CH2)ppm.
6,10-Diaza-5,11-dimethylpentadecane-~1,12-dione bis
oxime (5), (57%, oil).
NMR(lH, 200MHz,CDC13):63.31(2H,q,CH),
2.62(4H,m,N=CCH2), 2.40(4H,m,CH2N),1.57(6H,m,CH2CH2).
1.23(6H,m,CHMe), 0.97(6H,t,CH3)ppm.
- 5,9-Diaza-4,7,7,10-tetramethyltridecane-3,11 -dione
bis oxime (6), (33%, m.p. 120-1C).
- NMR('H,200MHz, d6-DMSO): ~10.2(2H,s,0H),
3.15(2H,q,CH),2.2(8H,brm,CH2N~N=CCH2),1.13(6H,d,CH3),
1.05(6H,t,CH3), 0.78(6H,s,CMe2)ppm.
- 4,8-Diaza-6,6-diethyl-3,9-dimethylundecane-2,
10-dione bis oxime (7), (10%, m.p. 142-4C).
NMR(1H,200MHz,CDCl3):~ 3 36(2H,m,CH), 2.39
(4H,s,CH2N),1.85~6H,s,C=NMe),1.25(10H,m,CH2Me+CHCH3),
0.76(6H~m,cH3)ppm~
- 6,10-Diaza-5,8,8,11-tetramethylpentadecane-4,12-
~25~8~
., .
- 16 -
dione bis oxime (8), (5%, m.p. 134-5C).
NMR(1H,200MHz,d6-DMSO): ~3.19(2H,q,CH),2.30
(4H,s,CH2N),2.25(4H,m,N=CCH2), 1.49(4H,brm,CH2CH3),
1.13(6H,d,CHMe),0.90(12H,m,CH3)ppm.
- 3,7-Diaza 2,8-dimethylnonane-1,g-dione bis oxime
(9), (13%, m.p. 111-4C).
NMR(1H,200MHz,d6-DMSO: 61o~46(2H~brs~oH~ 7.05 and
6.45 (2H,d,N=CH), 3.15(2H,m,CH), 2.44 (4H,brm,CH2N),
1.47t2H,m,CH2CH2~, 1.07(6H,d,Me)ppm.
- 4,8-Diaza-3,9-diphenylundecane-2,10-dione bis oxime
(10), (22%, m.p. 101-5C).
NMR( 1H,200MHz,d6-DMSO): ~10.50(2H,brs,OH),
7.3(10H,m,Ph), 4.29~2H,s,CHPh), 2.48(4H,m,CH2N),
1.63(2H,brm,CH2CH2), 1.54(6H,s,CH3)ppm.
- 4,8-Diaza-3,6,9-trimethylundecane-2,10-dione bis
oxime (11), (40% oil).
NMR ( H, 200MHz, d6-DMSO): 63.3 (2H, q, CH) 2.4
(4H, m, CH2), 1.8 (6H, s, CH3), 1.16 (6H, d, CH3),
0.7-1.4 (lH, m, CH), 0.85 (3H, d, CH3)ppm.
- 4,8-Diaza-3,5,7,9-tetramethylundecane-2,10 bis
oxime (12), (5%, m.p. 138-145C).
NMR ( H, 200MHz, d6-DMSO) 62.5 (2H, m, CH), 1.65
(6H, s, CH3), 1.23 (2H, t, CH3), 1.01 (6H, d,
CH3), 0.85 (6H, m, CH3)ppm.
- 5,9-Diaza-2,4,10,12-tetramethyltridecane-3,11-
dione bis oxime (13), (5.4%, m.p. 127-128C).
NMR (lH, 200MHz, CDC13)~ 3.4-3.6 (2H, 9, CH), 2.5-
2.8 (4H, m, CH2), 2.3-2.5 (2H, m, CH), 1.6-1.8
(2H, m, CH2), 1.6-1.8 ~2H~ m, CH2), 1.4 (6H, d,
CH3), 1.15 (12H, d, CH3)ppm.
Example 14.
SeDaration of meso- and d,l-stereoisomers of Compound 1
_ . _ . . _ _ _ _ _ . . . _ . _ . _ . _ .
. HPLC.
The analytical separation of the meso- and d,l-
diastereoisomers was accomplished by normal-phase HPLC
using a 250x4.6mm stainless steel column packed with
~SZ~8
-- 17 --
. .
5~m silica gel microspheres connected to a commercial
dual pump chromatographic system. Detection was via a
variable UV detector set at 210nm, and output from the
detector was directed to a chart recorder and
microcomputer programmed for peak integration.
The solvent system used throughout consisted of a
mixture of 85% methanol and 15% 0 04M aqueous
ammonia (v/v ). In order to exclude the possibility Or
column degradation using this solvent system, a
1~ precolumn packed with silica gel 15~m -25~m
particles was placed in the solvent line before the
sample injector. The flow rate used throughout was
1ml min 1.
Samples consisting of a mixture of diastereoisomers
of Compound 1 were dissolved in methanol at a concentration
of 10mg ml 1, and 10~1 aliquots were analysed.
! Baseline resolution was obtained with no evidence of
tailing, and retention times of 8.90 min and 9.87 min
were recorded for the meso-E,E- and d,l-E,E-isomers
respectively.
B. Fractional Crystallisation.
meso-4,8-Diaza-3,6,6,9-tetramethylundecane-2,10-
dione bis oxime
A sample of crude product, obtained direct from
the aqueous work up (38g, ratio 60:40, meso:dl) was
recrystallised ~our times successively from hot
acetonitrile giving pure meso isomer as fine white
needles (10.5g), m.p. 149.5-150C.
dl-4t8-Diaza-3L6,6,9-tetramethylundecane~2~10-
dione bis oxime
A sample of crude product (9.8g, ratio 50:50) wasdoubly recrystallised from hot acetonitrile giving dl-
enriched material (4.6g, ratio 47:53 meso:dl). The
filtrate from the second crystallisation was set aside
at roo~ temperature. A small crop of crystals (220 mg,
ratio 20:80, meso:dl) was removed and then the filtrate
~25~8~
. .
- 18 _
concentrated in vacuo giving dl-enriche~ material
(1.41g, ratio 22:78, meso:dl). Slow recrystallisation
from ethyl acetate gave pure dl isomer as large clear
crystals (82mg), m.p. 129-130C.
Example 15
X ray Crystallography.
Crystals suitable for X-ray crystallography were
obtained by crystallisation of the separated diastereo-
isomers from methanol. Details of the structuredetermination are given below. The determinations
demonstrate that the diastereoisomer with an HPLC
retention time of 8.90 minutes (Example 14.A) is the
meso-diastereoisomer, while the diastereoisomer with a
retention time of 9.87 minutes is the dl-diastereoisomer.
In both cases, the configuration of the oxime
functionalities is EE.
a) Meso isomer
C13H28N402 yLH20~ M=281, orthorhombic, space group
P2~212, a = 16.946, b = 15.565, c = 6.318.~,
V = 1666.46 A3, Z = 4~ Dc = 1.119 gcm 3, F(000) = 618,
~(Mo-Kx) = 0.47cm 1. 1716 intensities were recorded
(3<0<25) on a Philips PW1100 diffractometer. R 0.064
for 1036 reflections with F>6~F.
b) dl isomer
C13H28N402, M = 272, monoclinic, space group C2/C,
a = 6.763, b = 10.92, c = 23.863a, V = 1600.79A3,
~ = 4~ Dc = 1.128 gcm 3, F(000) = 600, (Mo-K ) =
0.49cm 1 1448 intensities were recorded (3<0<25) on
a Philips PW1100 diffractometer. R 0~071 for 861
reflections with F>6~F.
~L25~8~
- 19 -
Exarnple 16.
The 1H and 13~ NMR Spectra of the meso- and
dtl-diastereoisomers of Compound 1.
1. 1H NMR Spectra
1H NMR spectra were run in d6-DMSO at 500MHz using a
Bruker AM-500 FT NMR spectrometer. The following
assignments were made. (For the carbon atom numbering,
see the diagram above entitled "Diastereoisome~rs of
Compound 1"):-
5h1rt Multiplicity Assignment
. _
0.7748 s (CH3)2-C6
(meso~)
0.7779 s
0.7835 s (d,l-)
1.06905 d,J-6.73 CH3-C3+CH3-C9
(d,l-)
1.0660 d,J=6.74 CH3-c3+cH3-c9
(meso-)
1,6437 s CH3-C2+CH3-C10
2.1112 d of AB q's H's at C5~C7
25 3.1235 ABq,br H's at C3~Cg
3.30 s,br NH's
10.2495 s Oxime OH's
10.2511 s
In the d,l-diastereoisomer, the methyl
groups attached to C6 are in equivalent environments,
and so should give a single signal. This has been
assigned to the singlet at 0.7835 ppm. In the meso-
diastereoisomer, the methyl groups attached to C6 are
~2S2~B~
~ 20 -
in different environments, and so two singlets should
be seen. These have been assigned to the singlets at
0.774B and 0.7779 ppm.
2.13C NMR Spectra
_
'~C NMR spectra were run in d6-DMS0 or dll-MeOH
using a Bruker AM-250 FT NMR spectrometer or a Jeol
FX-200 FT NMR spectrometer. The following assignments
were made:
Chemical shift/ppm Assignment
1 0
8.9485 CH3-c2+cH3-clo
19.3809 CH3-C3+CH3_C9 (meso-)
19.4037 3 3 3 9 (d,l )
1525.0265 ) CH3-C6 (meso-)
25.0421
25.0794 CH3-C6 (d,l-)
35.3182 C6 (d,l-)
35.3731 C6 (meso-)
2057.8070 C5+C7 (meso-)
58.3569 C5-~C7 (d~1-)
58.9974 C3+Cg (meso-)
59 .0111 C3+Cg (d,l-)
161.3406 C2+C10 (d,l-)
25161.4497 C2~C10 (meso-)
The assigments were supported by off-resonance and
selective proton decoupling experiments.
As with the proton spectrum, the signals for the
gem-dimethyl groups at C6 are distinct for the two
isomers. The clearest distinction between the isomers
can be seen in the signals for the carbons C5 and C7,
the signal for the d,l-diastereoisomer being seen
approximately 0.55ppm downfield of the corresponding
signal for the meso-diastereoisomer. The relative
integrations for these signals corresponded closely to
~l2~i;2~8~
.
- 21 -
the values obtained for the isomer ratios by HPLC for a
wide variety of samples.
Example 17
SEPARATION OF d- and l- ENANTIOMERS OF COMPOUND 1
The diastereoisomer of Compound 1 was treated with
an equivalent of L-(+)-tartaric acid in hot ethanolic
solution. The solution was allowed to cool, and the
white solid was filtered off and recrystallised three
times to give a (+)-tartrate salt of one enantiomer
[~]D25 _ 28 .o8 (c=2.5,H20). M.p. 173-175C.
The filtrate from the above preparation was
concentrated, and the salt was decomposed to give
Compound 1 (of unknown enantiomer proportions) by
dissolving in water, basifying to pH9, and filtering
off the white solid. This was recrystallised from
ethyl acetate to give white crystals. This sample was
treated with an equivalent of D-~-)-tartaric acid in
hot ethanolic solution, and the resulting white solid
was recrystallised three times to give the (-)-tartrate
of the other enantiomer, [~]D25 = 27.67
(c=2.5,H20). M.pt. 167.5-168 C.
The samples of the tartrate salts thus obtained
were converted into the free bases by the method given
above to give samples of d- and l- Compound 1. The
(+)-tartrate salt gave l-Compound 1, [~D25 = -2.52
(c=4, MeOH), and the (-)-tartrate salt gave d-Compound
t, [~]D25 = +2.51 (c=4, MeOH).
Example 18
SEPARATION OF COMPOUND 2 INTO ITS meso- AND d,l FORMS
_ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _
The meso- and d,l- forms of Compound 2 were
separated by HPLC using a modification of the H~LC
conditions employed in Example 14. The only
differences were t~at the solvent composition was 98%
MeOH and 2% ( 0.04 M ) aqueous ammonia, instead Or 85%
~L~52~8~
and 15% respectively, and that the flow rate was 2ml
min 1 with a preparative column. Under these
conditions the faster-running isomer had a retention
time of approximately 24 minutes, and that of the
slower-running isomer was approximately 26 minutes.
The isomers were separated by preparative HPLC, giving
samples of approximately 90% purity as estimated by
HPLC. From the H NMR spectra the diastereoisomer
eluted first was designated meso-, and the fraction is
the d,l-form.
Example 19
SEPARATION OF THE d,l and the meso ISOMERS OF
COMPOUND 4
.
Partial separation of the d,l- and meso-isomers of
Compound 4 was achieved using HPLC with the same
conditions as were used for Compound 2. The results
indicated that although the isomer separation was poor,
samples of -70%:30% proportions were obtained.
As oxime isomerism occurred very rapidly t it was not
possible to isolate samples of greater purity.
Example 20
SEPARATION AND LABELLING OF OXIME ISOMERS OE COMPOUND 1
-
Due to rapid equilibration effects, the (E,Z)-
oxime isomer of Compound 1 had proved impossible to
characterize as an isolated component. However, the
labelling of the (E,Z) oxime isomer was studied, by
isolation of the HPLC peak corresponding to the (E,Z)
oxime isomer with immediate labelling of the resulting
solution. Experiments using the (E,E) isomer showed
that labelling under these conditions proceeded
smoothly. Using the (E,Z)-isomer it was not possible
to obtain consistent results, due in part to
irreproducibility of the HPLC method and in part to the
equilibration process; however, a number of complexes
were obtained, including a lipophilic species (usually
~5Z~8~
,
- 23 -
~20%, and probably derived from the (E,E)-isomer from
re-equilibration of the (E,Z)-isomer) and several more
hydrophilic species.
Example ? 1
Formation of the Tc-99m Complex of Compound 1
A sterile freeze dried formulation of 1.0mg of the
compound 1 (a dl/meso mixture) and 15 mg of stannous
tartrate in a sealed 10 ml glass vial containing a
nitrogen atmosphere, was reconstituted with 3-8 ml of
eluate of Tc-99m pertechnetate, obtained from a Mo-
99/Tc-99m generator system. Analysis of the resultant
mixture indicated that reduction of pertechnetate (to a
lower oxidation form of technetium), and complexation
of the reduced technetium by the ligand is complete
after standing at ambient temperatures for 1 minute.
The Tc-99m complexes of Compounds 2 to 13 and
individual isomers and enantiomers thereof were
formed similarly~
An alternative and currently preferred formulation
consists of 0.5mg of compound 1 (dl-isomer), Il.5mg of
sodium chloride, and 7.6mg of stannous chloride
dihydrate.
Analysis of the Tc-99m complexes
Thin Layer Chromatography
.
Glass fibre strips impregnated with silica gel
form the stationary phase of a fast and accurate
analytical system. Two strips, each measuring
20 cm x 2 cm were used in each analysis. Approximately
five microlitres of the solution containing the complex
was applied 1 cm from the base of each strip, and one
strip developed with saline, the other with methylethyl~
ketone (MEK).
Determination of the distribution of radioactivity
- ~2S2~
- 24 -
along each strip was conducted by means of a 100 channel
analysis system interfaced to a Nova computer, programmed
for peak integration. The Table below indicates the
RE values of the major components of the Tc-99m solutions.
The observed radiochemical purity of the Tc-99m complexes
was generally greater than 80%.
Silica gel on glass fibre chromatography
Observed RF values
Eluent: MEK Saline
Tc-99m colloid O O
Tc-99m pertechnetate 1.01.0
15 Tc-99m complexes 0.9-1.00.0-0.2
Stability of the complexes
Radiochemical purity determinations by thin layer
chromatography were carried out at several time points
following formation of the technetium complexes of
compounds 1 and 2 (a dl/meso mixture in each case) to
determine the stability of the complexes. Typical
results are shown in the table below.
% of the desired Tc-99m complex
Time post formation
. .
!
Tc-99m complex of: 2 min 60 mln 120 min
3o Compound 1 92 87 79
Compound 2 82 60 47
_ __
The in vitro stability of the complex of compound
2 is seen to be greatly inferior to that of the
complex of compound 1. These differences are important.
The percent of the Tc-99m not in the form of the
2~a8~
- 25 -
; complex is at all times more than twice as great with
Compound 2 as with Compound 1. This radiochemical
impurity would increase the background radiation in
vivo to a point at which useful information would be
more difficult or even impossible to obtain.
Example 22
Animal biodistribution data
0.1 ml of the Tc-99m complex solution was
administered by intravenous injection (lateral tail
vein) to each of five or six rats (140-220g). The
injected dose was equivalent to approximately lmCi of
Tc-99m. Three rats were sacrificed at 2 minutes post
in~ection, and 3 at 1 hour or 2 at 2 hours post
injection. At dissection, the organs and blood
samples shown in the following table were taken, and
assayed for radioactivity. The uptake in each organ
or tissue was calculated as a percentage of the total
activity recovered,
The ratio of diastereoisomers in the Tc-99m
complexes used in this experiment is not known.
5~
-- 26 -
Biodistribl~tion data o~ ~c-99m conplexes o~ Iigands (no ssparation of
stereoisomers )
_ _ _ _
Tc-99m % id/or~al In rats
on plex
Df 2 min pi 1 Dr 2 hr pi
l OL ig and Nr ain de~rt LI . er Gl ood NraIc Neart Li ver B I ~d
1 1.50.4 ~.4 16 .7 1.20.2 5.4 19.6 (lhr )
2 1.60.5 10 .914 .6 1.40 .S~ .912 .~ (lhr )
3 1.10.2 27.~ 6.8 0.70.117.0 2.1 (2hr )
4 1.6 _ 15 .613 .4 1.5 _ 12 .7 5.7 (lhr )
0.8 _ la.0 10.5 O.6 _ 12.1 2.8 ~1hr )
6 1.20.4 19.6 5.a 0.~0.213.5 2.6 (2hr )
7 1.0 _ 20.6 7.4 0.5 _ 36.5 3.3 (lhr )
0 .4 o .5 32 .9 1~ .0 0 .6 0 .3 16 .9 2 .7 (2hr )
20 11 1.30.8 12.~ 11.9 1.00.410.2 4 .7 (lhr )
12 1.00.5 23.9 8.6 0.90.319.7 6 .0 (lhr )
13 0 .5 0 .6 15.9 15 .Z 0 .Z 0 .2 12 .5 7 .2 (1hr )
2rj The results at 2 minutes and 1 hour p.i. are the mean
of 3 animal~.
The results at 2 hours p.i. are the mean of 2 animals.
The experiment of Example 22 was repeated using the
~eparated diastereoisomers of compound 1. The results are
given in the Table below and should be compared to thoAe
ror compound 1 (the mixture) in the Table in Example 15.
2 ~ 8
-27-
Biodistribution data on the Tc-99m complexes rrom
r
; d,l- and meso-diastereoisomers Or compound 1.
. ... .
% id/organ in rats
Tc-99m 2 min p.i. 1 hour p.i.
complexes
of ~rain Liver Blood Brain Liver Blood
I
d.l 1.99 9.42 11.42 1.90 9.25 10.17
meso- 0.78 26.86 5.33 0.53 24.24 3.17
The reasons for the differences in brain uptake
and retention are not understood, but do not appear to
result from any difference in lipophilicity between the
Tc-99m complexes of the two isomers. We have compared
the lipophilicities o~ the two complexes, by a special
HPLC method developed by ourselves for this purpose,
and have concluded that they are indistin~uishable.
Example 24
RAT BIODISTRIBUTION DATA ON THE Tc-99m COMPLEXES FROM
ISOLATED STEREOISOMERS
a) dI and meso diastereoisomers
%id/o~ninrats
BompDund Ster~o 2 ~ins pi 1 hr pl
~o. Iso~er Brain Heart Llver 310Od Brain ~p~er ~IQQd
1 aeso 1.1 0.6 19.9 5.0 0.7 0.223.7 3.1
dl 2.1 1.2 11 3 1100 1.6 0,711.9 8.6
2 eSD 1.7 0.6 17.9 6.9 1.4 0.220.7 ?~8
dl ~.a 1.1 11.6 10.5 1.5 0.7 9.~ 7.~
2 ~
- 28 -
b) d and l enantiomers of Compound 1
_ . . . .
% id/organ in rats
.
2 min pi 1 hr Pi
5 Enantiomer Brain Liver Blood Brain Liver Blood
d 1.6 13.4 11.4 1.4 9.7 10.2
l 2.6 9.6 11.0 2.2 7.7. 11.1
Data in Tables (a) and (b) are the mean from 3 animals
at each time point. They were obtained at a different
time and using different formulations from the data in
Example 23.
Example 25
CLINICAL STUDIES
All clinical studies, involving comparisons
between the Tc-99m complexes of dl and meso, and d,l
and dl stereoisomers of Compound 1 were conducted in
normal volunteer subjects at Aberdeen Royal Infirmary.
a) Comparison of dl and meso diastereoisomers of
Compound 1
This study was published:
99mTc HM-PA0 Stereoisomers as Potential Agents for
Imaging Regional Cerebral Blood Flow - Human Volunteer
~5 Studies. Sharp PF, Smith FW, Cemmell HC et al.
J.Nucl.Med., 1986, 27, pages 171-177.
The following table demonstrates the mean
percentage of total activity injected per organ at 20
minutes p.i. Data is then taken from area of
interest studies employing a whole body scanning
device.
- ~2SZ~
- 29 _
Tc-99m complex of % injected activity
Stereoisomer of Bladder
Compound 1 Brain Liver Kidneys +Urine
meso 1.85 16.6 1.3 0.6
dl 4.22 10.22 3.50 2.28
Mixture 1.95 13.0 1.95 1.1l
b) Comparison of d,l and dl stereoisomers of Compound
. _
The following table gives the mean of three
studies for each stereoisomer, of percentage of
injected dose in normal volunteers at 30 minutes p.i.
BrainBiver Kidney
d 3.7612.66 1.46
l 4.308.o6 4.0
dl 4.1610.73 3.03
Clinical studies in normal volunteer sub~ects
-
The following data allows comparison of the
relative performance of the complexes of compounds 1
and 2 (an unseparated dl/meso mixture in each case) in
30 human volunteer subjects.
~2SZ~
- 30 - 20388-1585D
i) Blood clearance
Time post% id in blood
injection
(minutes)Compound lCompound 2
14 9.54 11.06
7.73 6.9
6.38 5.93
250 5.26 3.9
ii) Whole body distribution
Whole body distribution at
2 hrs p.i.
Compound 1 Compound 2
Brain 3.70 3.48
Liver 23.35 13.21
Bladder 3.9 21.02
Tomographic Imaging Studies in man
rromographic images o~ the brain were determined using
the same complexes in normal volunteers. The device used was a
single head rotating gamma camera and minicomputer system. 64
25-second images were accumulated by the gamma camera during
~Z5;~
- 31 - 20388-1585D
360 circular ro-tation of the head and shoulders of the
volunteer.
Good quality tomographic images of the brain were
obtained by the reconstruction of the images obtained using -the
complex of Compound 1 by the minicomputer. The images ob-tained
using the complex of Compound 2 were less good, due to a higher
level of background radiation, resulting from a higher uptake in
soft tissue regions.
Example 27
_n vitro labelling of leucocytes
A solution of the technetium-99m complex of compound 1
was made as follows. A vial contained 0.5mg compound 1; 7.5mg
stannous chloride dihydrate; and 4.5mg sodium chloride;
freeze-dried and sealed under nitrogen. To this was added 5.0ml
of sodium pertechnetate eluate from a 135mCi technetium generator.
Mixed leucocytes were obtained from 34ml of acid citrate
an-ticoagulated blood by dextran sedimentation. These were washed
twice and resuspended in 2.Oml phosphate buffered saline
containing approximately 0.25mg/ml prostaglandin El. The
suspension was incuba-ted with 0.2ml of the solution of the
technetium-99m complex of compound 1. Incubation was at ambient
temperature, and samples were removed at intervals for analysis.
After two minutes, 60% of the radioactivity was associated with
blood cells; after 5 minutes, 83% and after 10 minutes, 89%.
Tc-99m labelled leucocytes prepared by this method were
in~ected into rats bearing abscesses produced by implantation of
sponge impregnated with faecal extract. Abscess uptake was
identical to that for In-lll labelled leucocytes.
