Syntheses of (R)- and (S)-2- and 6-fluoronorepinephrine and (R)- and (S)-2- and 6-fluoroepinephrine: Effect of stereochemistry on fluorine-induced adrenergic selectivities

Article abstract of DOI:10.1021/jm990599h

Several routes to the enantiomers of fluoronorepinephrines (1) and fluoroepinephrines (2) were explored. A catalytic enantioselective oxazaborolidine reduction and a chiral (salen)Ti(IV) catalyzed asymmetric synthesis of silyl cyanohydrins proved efficaci

Full text of DOI:10.1021/jm990599h

J . Med. Chem. 2000, 43, 1611-1619
1611
Syn th eses of (R)- a n d (S)-2- a n d 6-F lu or on or ep in ep h r in e a n d (R)- a n d (S)-2- a n d
6-F lu or oep in ep h r in e: Effect of Ster eoch em istr y on F lu or in e-In d u ced
Ad r en er gic Selectivities
Shou-fu Lu,† B. Herbert,† Gu¨nter Haufe,‡ Klaus Wilhelm Laue,‡ William L. Padgett,† O. Oshunleti,†
J ohn W. Daly,† and Kenneth L. Kirk*,†
Laboratory of Bioorganic Chemistry, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of
Health, Bethesda, Maryland 20893, and Organisch-Chemisches Institut, Universita¨t Mu¨nster, D-48149 Mu¨nster, Germany
Received December 8, 1999
Several routes to the enantiomers of fluoronorepinephrines (1) and fluoroepinephrines (2) were
explored. A catalytic enantioselective oxazaborolidine reduction and a chiral (salen)TiIV catalyzed
asymmetric synthesis of silyl cyanohydrins proved efficacious in the key stereo-defining steps
of two respective routes. Binding studies of the catecholamines with R1-, R2-, â1-, and
â2-adrenergic receptors were examined. The assays confirmed that fluorine substitution had
marked effects on the affinity of (R)-norepinephrine and (R)-epinephrine for adrenergic
receptors, depending on the position of substitution. Thus, a fluoro substituent at the 2-position
of (R)-norepinephrine and (R)-epinephrine reduced activity at both R1- and R2-receptors and
enhanced activity at â1- and â2-receptors, while fluorination at the 6-position reduced activity
at the â1- and â2-receptors. The effects of fluorine substitution on the S-isomers were less
predictable.
In tr od u ction
the â-selective agonists isoproterenol (3b,d )5 and tert-
butylphenoxypropanolamine (4b,d ),6 and the R-selective
phenylephrine (5b-d ).7 In all cases, fluorine in the
2-position diminished R-adrenergic activity of com-
pounds possessing R-adrenergic activity and fluorine in
the 6-position reduced â-adrenergic activity in com-
pounds possessing â-adrenergic activity. In contrast,
fluorine had relatively little effect on the weak adren-
ergic activity of dopamine (6b-d ).8,9 The fluorinated
norepinephrines and epinephrines have proven to be
very useful biological tools. For example, these selective
agonists can be produced in vivo by biosynthetic en-
zymes, are metabolized by catecholamine-processing
enzymes, and are efficiently translocated by transport
mechanisms.2 Although there are qualitative differences
in behavior with enzymatic and transport processes,
striking differences are only observed for adrenergic
receptor selectivities.
The use of fluorine substitution to modulate biological
processes has been a strategy used effectively by me-
dicinal chemists for nearly five decades. A major factor
for the effectiveness of fluorine substitution in the
design of analogues of biologically important molecules
is the steric similarity of the C-F bond and the C-H
or C-OH bond that facilitates the acceptance of a
fluorinated analogue by a biological recognition site. On
the other hand, the high electronegativity of fluorine
often imparts altered physicochemical properties to the
analogue that influence the consequences of biological
recognition. These and other factors have been reviewed
in several reports.1
Our own work in this area has included an examina-
tion of the effects of fluorine substitution on a series of
catecholamines and amino acids. We felt at the begin-
ning of this work that the electron-rich catechol system
should be quite susceptible to electronic perturbations
caused by a fluorine atom on the ring. A particularly
rewarding aspect of this research was the discovery of
the effects of fluorine substitution on the adrenergic
receptor selectivities of adrenergic agonists.2 For ex-
ample, racemic 2-fluoronorepinephrine (1b, Chart 1) has
markedly reduced activity at R-adrenergic receptors, but
retains full â-adrenergic activity.3 In contrast, racemic
6-fluoronorepinephrine (1d ) has markedly reduced ac-
tivity at â-adrenergic receptors, but retains full activity
at R-adrenergic receptors. Fluorine substitution at the
5-position had relatively little effect on adrenergic
activity. These initial results were extended to include
fluorinated analogues of epinephrine (2b,d ),4 as well as
We have carried out several studies in an attempt to
determine the mechanism(s) of the fluorine-induced
adrenergic selectivities, but we have not been able to
reach concrete mechanistic conclusions. Since no adren-
ergic receptor selectivity was observed with fluorinated
analogues of dopamine 6a , the stereogenic benzylic
hydroxyl group appeared to be an essential factor
influencing the observed selectivities of norepinephrine
and related agonists. We have considered models de-
picting hydrogen bonding or repulsion between fluorine
and the benzylic hydroxyl group for the origin of the
intramolecular interactions resulting in these effects.2
Altered interactions of the catechol ring with the recep-
tor protein have also been considered.2a
Classical pharmacological studies demonstrated that
the R-isomer of 1 is the more active isomer and that
the S-isomer has activity comparable to dopamine at
adrenergic receptors. If selectivity requires fluorine on
* To whom correspondence should be addressed. Tel: 301-496-2619.
Fax: 301-402-4182. E-mail: kennethk@bdg8.niddk.nih.gov.
† National Institute of Diabetes and Digestive and Kidney Diseases.
‡ Universita¨t Mu¨nster.
10.1021/jm990599h CCC: $19.00 © 2000 American Chemical Society
Published on Web 04/01/2000

1612 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 8
Lu et al.
Ch a r t 1
the catechol ring plus the benzylic hydroxyl group in
the proper configuration, then the enantiopure R-
fluoronorepinephrines may exhibit significantly greater
selectivity patterns than the racemates. In other words,
it seemed possible that the S-enantiomers could make
major contributions to the binding of (()-1b at the
R-adrenergic receptors and (()-1d at the â-adrenergic
receptors. Therefore, clarification of the role of the
stereogenic center could add to our understanding of the
effects of fluorine on the selectivity and to assist us in
solidifying a mechanistic understanding. To that end,
we have prepared the enantiomers of 2- and 6-fluoro-
norepinephrine [(R)- and (S)-1b,d ] and 2- and 6-fluoro-
epinephrine [(R)- and (S)-2b,d ] and have determined
their affinities at R1-, R2-, â1-, â2-adrenergic receptors.
of HCN under usual conditions failed. In contrast, the
ZnI2-catalyzed addition of trimethylsilyl cyanide was
facile and provided ready access to racemic fluoronore-
pinephrines by side-chain elaboration.3
The lability of the benzylic OH group in fluorocate-
cholamino alcohols, and the low reactivity of precursor
benzaldehydes placed certain limitations on synthetic
approaches. However, recent advances have provided
powerful new procedures for asymmetric syntheses, and
we have applied two of these new procedures to the
preparation of the R- and S-enantiomers of amines 1b,d
and 2b,d .
Asym m etr ic Ca r bon yl Red u ction . The enzyme-
like CBS catalysts (e.g., 10) developed by Corey12 have
been applied to the enantioselective synthesis of aryl-
ethanolamines.13 The remarkably high enantioselectivi-
ties obtained with these catalysts provided the basis for
one approach to R-fluoronorepinephrines. Chloroketones
9b,d were prepared from aldehydes 7b,d in 57% and
49% yields, respectively, by the sequence depicted in
Scheme 1. The reaction of ketones 9b,d with BH3‚THF
in the presence of CBS catalyst 10 gave R-chloro
alcohols 11b (80%) and 11d (85%). A single recrystal-
lization provided the chloro alcohols enriched to g 95%
e.e., as determined by chiral HPLC. After conversion
to the azides (12b,d ), catalytic hydrogenation in the
presence of oxalic acid gave the oxalates of (R)-1b (76%)
and (R)-1d (51%). In both cases, the enantiomeric
excesses of the final product were identical to those of
chloro alcohols 11.
Asym m etr ic Cya n oh yd r in F or m a tion . The key
step in our original preparation of racemic fluorinated
amines 1b and 1d was the ZnI2-catalyzed addition of
trimethylsilyl cyanide to fluorinated bis-benzyloxy-
benzladehydes 7b and 7d . A recent report amply
demonstrated that chiral Lewis acid (salen)TiIV com-
plexes (13) catalyze the addition of TMSCN to a variety
of aldehydes in good yield and high enantioselectivity.14
Resu lts a n d Discu ssion
Syn th esis of th e En a n tiom er s of F lu or on or ep i-
n ep h r in es. At the outset of this work, we recognized
certain complications. First, the benzylic position of
fluorinated catecholamines is very susceptible to sol-
volysis. Previous attempts to recrystallize the hydro-
chloride salt of 1d from methanol/ether led to extensive
methyl ether formation at the benzylic position.10 Ding,
Fowler, and co-workers reported that after separation
of the enantiomers of 1d by chiral HPLC, concentration
of an acidic solution of the individual isomers led to
extensive racemization.11 Early attempts to resolve
either 1b or 1d by recrystallization of diasteromeric
salts were also unsuccessful. From this, it became clear
that acid treatment as a final synthetic manipulation
is not compatible with the high reactivity of the benzylic
position. The catecholamines also are rapidly oxidized
in basic medium. Another problem we encountered in
our early work was the low electrophilicity of the
carbonyl group of fluorinated 3,4-dibenzyloxybenzalde-
hydes (7b ,d ) and fluorinated veratraldehydes. For
example, attempts to prepare cyanohydrins by addition

Synthesis of Fluoronorepinephrines and Fluoroepinephrines
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 8 1613
Sch em e 1
Sch em e 2
by hydrogenolysis to give (S)-1b and (S)-1d in 68% and
87% yield, respectively. Both of the final products were
obtained in >95% e.e. As predicted, the configuration
of the product was opposite that of the material pro-
duced through CBS-catalyzed carbonyl reduction, as
evidenced by the chiral HPLC retention times and the
sign of the rotations.
Syn th esis of th e En a n tiom er s of F lu or oep in ep h -
r in es. In the initial examination of enantioselective
trimethylsilyl cyanohydrin formation, we chose the
(R,R)-salen catalyst for the production of the antipodes
of the amino alcohols 1b,d that were synthesized by the
asymmetric carbonyl reduction strategy. Subsequently,
this shorter route was used to prepare both the R- and
the S-isomers of the intermediate fluorinated bis-
benzyloxyphenethanolamines (15b,d ) for elaboration to
both enantiomers of 2- and 6-fluoroepinephrines (2b,d ).
Following the procedure used to prepare racemic 2b,d ,4
amines 15b and 15d were heated with ethyl formate to
provide formamides 16b,d . Reduction of the amides
with lithium aluminum hydride gave the N-methyl
intermediates 17b,d . Hydrogenolysis in the presence of
oxalic acid produced the salts of (R)-2b,d and (S)-2b,d
(Scheme 3). The enantiomeric excess of each isomer was
determined to be >95% ee.
Biologica l Activity. Affinities for R1-, R2-, â1-, and
â2-adrenergic receptors of dopamines, norepinephrins,
and epinephrines were determined by displacement of
specific radioligands with rat brain membranes. For
further details, see Experimental Section. The results
are presented in Table 1.
r1-Ad r en er gic Recep tor . A 2-fluoro substituent
markedly reduced the affinity of (R)-norepinephrine (8-
fold) and (R)-epinephrine (16-fold) for R1-adrenergic
receptors, while having little effect on the affinity of
dopamine, (S)-1a , and (S)-2a (Table 1). These results
suggest that fluorine at the 2-position of the aromatic
ring may prevent the benzylic hydroxyl group from
proper interaction at the receptor site; i.e., for 1b and
2b the R-enantiomer now interacts poorly, and like the
This report prompted us to examine the application of
the method to the less electrophilic fluorinated bis-
benzyloxybenzaldehydes 7b and 7d . The fluorinated
aldehydes 7b and 7d were allowed to react with TMSCN
in the presence of titanium tetraisopropoxide and (R,R)-
13 at -50 °C. The resultant silyl cyanohydrins were
separated from the salen ligand and subjected to reduc-
tion with LiAlH4 to produce the corresponding phen-
ethanolamines S-15b and S-15d in 22-46% yield
(Scheme 2). The seemingly low yield for this transfor-
mation was the result of the instability of the silyl-
cyanohydrins to silica gel chromatography. We discov-
ered that a single recrystallization provided the products
in >95% e.e. The benzyl protecting groups were removed

1614 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 8
Lu et al.
Sch em e 3
Ta ble 1. Effect of Fluorine Substituents on the Affinity of
Dopamines (6), Norepinephrines (1), and Epinephrines (2) at
Adrenergic Receptors
A 6-fluoro substituent had no effect on the affinity of
(R)-1a and (R)-2a at R1-adrenergic receptors and had
only slight effects on the affinity of dopamine, (S)-1a ,
and (S)-2a . Thus, a 6-fluoro substituent had little or no
effect on the proper interaction of (R)-norepinephrine
and (R)-epinephrine, while it either slightly reduced or
slightly enhanced the interaction of dopamine and the
S-enantiomers. Each of the norepinephrines had affini-
ties similar to that of the corresponding epinephrines
for the R1-adrenergic receptors, which indicated that the
methyl group of the epinephrines made little contribu-
tion to the binding.
receptor affinity (Ki, µM, or percent inhibition)a
amine
R1
R2
â1
â2
6a
6b
6d
14 ( 1
16 ( 2b
6 ( 1b
0.058 ( 0.010b
0.12 ( 0.02b
0.040 ( 0.010b
0.020 ( 0.018
0.34 ( 0.02
0.030 ( 0.006
0.55 ( 0.03
0.32 ( 0.02
10 ( 1b
19 ( 5b
41 ( 10b
1.3 ( 0.2
26 ( 6
50 ( 6
∼70c
∼130c
5 ( 2
(R)-1a
(S)-1a 18% (100 µM)
(()-1a
(R)-1b
3.0 ( 0.7
27 ( 1
11 ( 3
6.5 ( 0.9
2.7 ( 0.9
0.13 ( 0.03 0.067 ( 0.017
∼25c
(S)-1b 12% (100 µM)
52 ( 9
0.21 ( 0.03
24 ( 3
27 ( 2
0.14 ( 0.03
28 ( 3
(()-1bd
(R)-1d
(S)-1d
∼50c
0.60 ( 0.02
0.020 ( 0.004
0.36 ( 0.03
The configuration of the benzylic hydroxyl group was
profoundly influential on the affinity of catecholamines
at the R1-adrenergic receptors. For example, (R)-1a had
a multiple-fold higher affinity than (S)-1a , and (R)-2a
had a 12-fold higher affinity than (S)-2a . The R1-
adrenergic receptor also exhibited a high degree of
enantioselection for the 2-fluoronorepinephrines, but not
for the 2-fluoroepinephrines. Only a 2.5-fold selectivity
was observed between the enantiomers of 1d , while the
6-fluoroepinephrines exhibited a 30-fold selectivity; i.e.,
the 6-fluoro substituent reduced the R1-adrenergic
receptor enantioselectivity for norepinephrines, while
it increased the receptor enantioselectivity for epineph-
rines.
3.7 ( 0.7
∼12c
∼120c
∼270c
(()-1d 20% (100 µM) 0.028 ( 0.003
63 ( 8
58 ( 11
0.66 ( 0.02
3.7 ( 0.5
1.1 ( 0.2
0.11 ( 0.03
24 ( 7
(R)-2a
(S)-2a
(()-2a
(R)-2b
(S)-2b
(()-2b
(R)-2d
(S)-2d
(()-2d
1.6 ( 0.2
0.013 ( 0.002
0.045 ( 0.003
0.015 ( 0.002
1.1 ( 0.3
13 ( 2
2.1 ( 0.6
∼20c
2.8 ( 0.3
∼25c
0.067 ( 0.003 0.56 ( 0.03
∼40c
0.053 ( 0.001
0.060 ( 0.001
0.013 ( 0.002
0.078 ( 0.001
0.014 ( 0.002
∼150c
1.8 ( 0.4
∼100c
∼25c
0.26 ( 0.05
38 ( 7
2.5 ( 0.1
∼70c
∼400c
∼90c
5.0 ( 0.7
∼140c
50 ( 3
a
KI values were derived by the Cheng-Prusoff equation from
IC50 values (n ) 3). Values for dopamines from Nie et al.9 c KI
b
d
values estimated from extrapolated inhibition curve. The values
reported previously15 for (()-1b were significantly higher than
these values. The reason is not apparent.
r2-Ad r en er gic Recep tor s. A 2-fluoro substituent
reduced the affinity of (R)-norepinephrine by 25-fold for
the R2-adrenergic receptors, while it reduced the affinity
of (R)-epinephrine by only 5-fold. A 2-fluoro substituent
had little or no effect on the affinity of dopamine, (S)-
1a , and (S)-2a for the R2-adrenergic receptors. As was
S-enantiomer, has a low affinity for R1-adrenergic
receptors. Thus, the 2-fluoro substituent is poorly toler-
ated with respect to interaction of catechol amines with
an R-configuration at the benzylic hydroxyl position
with the R1-adrenergic receptors.

Synthesis of Fluoronorepinephrines and Fluoroepinephrines
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 8 1615
the case for R1-adrenergic receptors, a 2-fluoro substitu-
ent might have prevented the proper interaction of the
benzylic hydroxyl group of (R)-1 at the receptor site and
to a lesser extent for (R)-2. Thus, the presence of the
N-methyl to some extent overcame the negative effect
of the 2-fluoro substituent on proper interaction of the
R-enantiomers.
â2-Ad r en er gic Recep tor s. A 2-fluoro substituent
increased the affinity of (R)-norepinephrine for the â2-
adrenergic receptors by a remarkable 60-fold and it
increased the affinity of (R)-epinephrine by 6-fold. A
2-fluoro substituent had no effect on the affinity of
dopamine or (S)-1a , whereas it reduced the affinity of
(S)-2a for â2-adrenergic receptors. Both (R)-1b and (R)-
2b had the same affinity for â2-adrenergic receptors.
This is in marked contrast to the relative activity of (R)-
norepinephrine and (R)-epinephrine at â2-adrenergic
receptors, where the affinity of (R)-2a is much greater
than that of (R)-1a . Thus, the affinity of (R)-norepi-
nephrine for the â2-receptor was enhanced much more
than it was for (R)-epinephrine by the presence of a
2-fluoro substituent. The N-methyl group of (R)-epi-
nephrine appeared to be an important factor in deter-
mining the effect of a 2-fluoro substituent, with the
2-fluoro substituent apparently reducing the proper
interaction of the N-methyl group.
A 6-fluoro substituent had little or no effect on the
affinity of dopamine, (R)- and (S)-norepinephrine, and
(R)- and (S)-epinephrine for R2-adrenergic receptors.
Thus, a 6-fluoro had little or no effect on the proper
interaction of (R)-1a or (R)-2a .
The configuration of the benzylic hydroxyl group was
quite important for enantioselectivity of the R2-adren-
ergic receptors for norepinephrines with (R)-norepi-
nephrine having a 17-fold higher affinity than (S)-
norepinephrine. This receptor enantioselectivity was
retained in 6-fluoronorepinephrines, where the R-enan-
tiomer had an 18-fold higher affinity than the S-
enantiomer. The configuration of the benzylic hydroxyl
group was less important for epinephrine, where the (R)-
2a had only a 4-fold higher affinity than (S)-2a . Simi-
larly, (R)-2d had only a 6-fold higher affinity than (S)-
2d . The R2-adrenergic receptor exhibited little or no
enantiosectivity for the 2-fluoronorepinephrines and
2-fluoroepinephrines.
A 6-fluoro substituent greatly reduced the affinity of
(R)-epinephrine (60-fold) and (S)-epinephrine (24-fold)
for â2-adrenergic receptors. The 6-fluoro substituent
caused smaller reductions in the affinities of dopamine
(2.6-fold), (R)-1a (6-fold), and (S)-1a (10-fold). Thus, in
norepinephrine and epinephrine the negative effect of
the 6-fluoro substituent at â2-adrenergic receptors ap-
peared relatively unrelated to the configuration of the
benzylic hydroxyl group. A 6-fluoro substituent had a
much greater negative effect on the affinity of epineph-
rines than for norepinephrines, suggesting that the
proper interaction of the N-methyl of the 2a is reduced
by a 6-fluoro substituent.
â1-Ad r en er gic Recep tor . A 2-fluoro substituent
slightly decreased the affinity of dopamine and (S)-
norepinephrine for the â1-adrenergic receptors, while it
markedly decreased the affinity for (S)-epinephrine. In
contrast, a 2-fluoro substituent increased the affinity
of (R)-1a for â1-adrenergic receptors by 10-fold and
increased the affinity of (R)-2a by 2-fold. Thus, the
presence of a 2-fluoro substituent enhanced the proper
interaction of the benzylic hydroxyl group at the â1-
adrenergic receptor, particularly in the case of (R)-
norepinephrine, and to a lesser extent for (R)-epineph-
rine.
The â2-adrenergic receptors exhibited a 5-fold enantio-
selectivity between (R)-norepinephrine and (S)-norepi-
nephrine, and a 6-fold preference for (R)-epinephrine
over the S-enantiomer. The receptor enantioselectivity
for the 2-fluoronorepinephrines was increased to 400-
fold, while for 2-fluoroepinephrines it was increased to
200-fold. The receptor preference for the enantiomers
of 6-fluoronorepinephrines was 10-fold, while that for
the 6-fluoroepinephrines was only 3-fold. Dopamine had
a 10-fold lower affinity than (R)-1a at â2-receptors.
Thus, the presence and configuration of the benzylic
hydroxyl group was important for the binding of nor-
epinephrines and epinephrines at â2-adrenergic recep-
tors, but the presence of an N-methyl group in 2b and
2d decreased the enantioselectivity compared to that
of the corresponding norepinephrines.
A 6-fluoro substituent markedly reduced the affinity
of (R)-1a (17-fold), (R)-2a (100-fold), and (S)-2a (34-fold),
while slightly reducing the affinity of dopamine (4-fold)
and (S)-1a (2-fold). Thus, the 6-fluoro substituent was
poorly tolerated with respect to interaction of catechol-
amines with â1-adrenergic receptors. The negative effect
of the 6-fluoro substituent is greatest for (R)-2a and (S)-
2a . Thus, although a negative effect of the 6-fluoro
substituent on the proper interaction of the benzylic
hydroxyl group appears likely to be the major factor for
norepinephrines, the proper interaction of the N-methyl
of the epinephrines with â1-adrenergic receptors also
seems to be reduced by the presence of a 6-fluoro
substituent.
(R)-2-Flu or on or epin eph r in e an d (R)-2-Flu or oepi-
n ep h r in e: “Selective” Agon ists for â-Ad r en er gic
Recep tor s. A 2-fluoro substituent markedly reduced
the affinity of (R)-norepinephrine and (R)-epinephrine
for R1- and R2-adrenergic receptors and enhanced the
affinity for â1- and â2-adrenergic receptors. The effect
was most dramatic for (R)-1a (Table 1). Thus, (R)-1b
became a more selective â-adrenergic agonist than (R)-
1a , exhibiting a 4-fold and 8-fold higher affinity for â1-
and â2-adrenergic receptors, respectively, than for the
R1-adrenergic receptors. However, because of the very
high affinity of both (R)-1a and (R)-2a for R2-adrenergic
receptors, (R)-2-fluoroepinephrine had an 8-fold higher
affinity for R2-receptors than for â1-receptors and a
2-fold higher affinity for R2-receptors than for â2-
receptors.
The â1-adrenergic receptors exhibited a 20-fold pref-
erence for (R)-norepinephrine over (S)-norepinephrine
and a 12-fold preference for (R)-epinephrine over the
S-enantiomer. The receptor enantioselectivity for 2-fluo-
ronorepinephrines was increased to over 300-fold, while
that of 2-fluoroepinephrines was increased to nearly
300-fold. In contrast, the receptor enantioselectivity
exhibited for 1d and 2d were only 5- and 4-fold,
respectively; i.e., the configuration of the benzylic hy-
droxyl group became less important when a 6-fluoro
substituent was present.

1616 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 8
Lu et al.
(R)-6-F lu or on or ep in ep h r in e a n d (R)-6-F lu or o-
n ep in ep h r in e: Selective Agon ists for r-Ad r en er -
gic Recep tor s. A 6-fluoro substituent markedly re-
duced the affinity of (R)-norepinephrine and (R)-
epinephrine for â1- and â2-adrenergic receptors (Table
1). Thus, these 6-fluoro analogues became more selective
for R-adrenergic receptors than the parent (R)-1 and (R)-
2. (R)-6-Fluoronorepinephrine (1d ) had a 7-fold higher
affinity for R1-receptors than for â1- and â2-adrenergic
receptors, while having a 1200-fold higher affinity for
R2 receptors than for â1- and â2-adrenergic receptors.
(R)-6-Fluoroepinephrine (2d ) had a 40-fold higher af-
finity for R1-receptors than for â1-adrenergic receptors
and a 15-fold higher affinity for R1-receptors than for
â2-adrenergic receptors. At R2-adrenergic receptors (R)-
1d had an 8000- and 3000-fold higher affinity than for
â1- and â2-receptors, respectively.
hydroxyl group even in the less active S-enantiomers
can influence the effects caused by fluoro substituents.
Exp er im en ta l Section
Gen er a l. All reactions were performed under an atmo-
sphere of N2 in flame or oven (90 °C) dried glassware unless
otherwise stated. Anhydrous solvents used for reactions were
purchased from Aldrich and used as received. Chromatography
refers to the flash chromatographic method of Still17 and was
performed on 230-400 mesh E. Merck silica gel purchased
from Bodman. Preparative thin-layer chromatography was
1
performed on 2000 µm silica gel plates from Analtec. H (300
MHz) and 19F NMR (282.3 MHz) spectra were obtained on a
Varian Gemini 300 spectrometer. Coupling constants J are
given in hertz (Hz). Chemical shifts are given in ppm (δ)
relative to the following internal standards: TMS and HOD
for 1H NMR and CFCl3 for 19F NMR unless otherwise noted.
Melting points (mp) were determined on a Thomas-Hoover
capillary melting point apparatus and are uncorrected. Optical
rotations were determined on a Perkin-Elmer Polarimeter 241.
Electron impact (EI) and chemical ionization (CI) mass spectra
were recorded on a Finnigan MAT 312. Chiral HPLC was
performed on a HPLC system from GBC using an analytical
CHIRALPAK AD column with UV monitoring (λ ) 273 nm)
unless otherwise noted.
Su m m a r y
Several routes to the enantiomers of fluoronorepi-
nephrines (1) and fluoroepinephrines (2) were explored.
A catalytic enantioselective oxazaborolidine reduction
was used to control the formation of the stereogenic
center in the preparation of norepinephrines (R)-1b and
(R)-1d . The chiral (salen)TiIV catalyzed asymmetric
synthesis of silyl cyanohydrins was used in the key
stereo-defining steps of the enantiomeric series (S)-1b
and (S)-1d . The versatility of the asymmetric trimeth-
ylsilyl hydrocyanation was demonstrated by the prepa-
ration of (R)-fluoronorepinephrines 1b and 1d . Both
enantiomers of fluoroepinephrines 2b and 2d were
prepared from a common intermediate from the syn-
thesis of the fluoronorepinephrines, namely amino al-
cohols 15b and 15d . Both asymmetric routes produced
the products in excellent (g95%) enantiomeric excess.
Biologica l Assa y. Affinities for R1-adrenergic receptors
were determined through inhibition of binding of [3H]prazosin
to rat cerebral cortical membranes.9 Affinities for R2-adrenergic
receptors were determined through inhibition of binding at
[3H]clonidine to rat cerebral cortical membranes.9 Affinities
for â1-adrenergic and â2-adrenergic receptors were determined
through inhibition of binding of [3H]dihydroalprenolol to rat
cerebral cortical and cerebellar membranes, respectively.9
1-Ch lor o-2-(3,4-d ib en zyloxy-6-flu or op h en yl)et h a n -2-
on e (9d ). A solution of dichloromethyl phenylsulfoxide (587
mg, 2.75 mmol, 1.0 eqiv) in THF (5 mL) was added dropwise
to a -78 °C solution of LDA (1.5 M in THF/cyclohexane, 2.2
mL, 3.3 mmol, 1.2 eq.) in THF (10 mL). After 15 min, aldehyde
7d (1.0 g, 3.0 mmol, 1.1 eq.) was added to the reaction flask,
and the reaction mixture was allowed to stir at -78 °C for 1
h before it was quenched with 2.5 N HCl. The mixture was
extracted with ethyl acetate (3 × 50 mL), and the combined
organic layers were washed with brine, dried over anhydrous
sodium sulfate, and concentrated to afford crude 8d as a light
yellow, amorphous solid which was used for the next reaction
directly.
These isomers were examined for the effects of ster-
eochemistry and the position of fluorine substitution on
the binding affinity at a series of R- and â-adrenergic
receptors. Our expectation that the modest activity of
2-fluoro racemates at R-receptors would be due mainly
or entirely to contributions of the S-enantiomer and that
the modest activity of 6-fluoro racemates at â-receptors3
would be due mainly or entirely to contributions of the
S-enantiomer was not realized. Instead, the (R)-2-fluoro
enantiomers retained low but significant activity at
R-receptors, while (R)-6-fluoro enantiomer retained low
but significant activity at â-receptors. The affinities of
norepinephrines and epinephrines for adrenergic recep-
tors were markedly affected by 2- and 6-fluoro substit-
uents in a manner that was predictable for the R-enan-
tiomers. Thus, a 2-fluoro substituent inhibited proper
interaction of the R-enantiomers at R1-adrenergic recep-
tors, while a 6-fluoro substituent had little effect. A
2-fluoro substituent enhanced the proper interaction of
the R-enantiomers at â-adrenergic receptors, particu-
larly for (R)-norepinephrine, while a 6-fluoro substituent
inhibited the proper interaction. The presence of a
N-methyl substituent, as in (R)-epinephrine, altered the
magnitude of the effects of fluoro substituents. Fluorine
substituents generally had no effect on the binding
affinity of the S-enantiomers. In a few cases, however,
fluorine markedly reduced the binding affinity and in
one case enhanced it. Thus, the presence of a benzylic
To a -78 °C solution of crude 8d in THF (25 mL) under an
atmosphere of argon was added ethylmagnesium bromide (3.0
M in Et2O, 5.0 mL, 15 mmol, 5 eq.). The reaction mixture was
allowed to stir for 1 h at -78 °C, then for 1 h at -45 °C, and
was then quenched by the addition of 2.5 N HCl. The volatile
solvents were removed under reduced pressure, and the
resultant aqueous suspension was extracted with ethyl acetate
(3 × 40 mL). The combined organic layers were washed with
brine, dried over anhydrous sodium sulfate, and concentrated.
The residue was purified by chromatography (petroleum ether/
ethyl acetate, 95/5) to afford 560 mg (49%) of chloroketone 9d
1
as colorless needles. Data for 9d : mp 121-122 °C; H NMR
(300 MHz, CDCl3) δ 4.67 (d, J ) 3.9, 2H, COCH2), 5.17 and
5.21 (two s, 4H, two ArCH2), 6.67 (d, J ) 11.7, 1H, 3- or 6-ArH),
7.31-7.46 (m, 10H, two ArH), 7.54 (d, J ) 6.9, 1H, 3- or
6-ArH); 19F NMR (282 MHz, CDCl3) δ -35.47 (d, J ) 8.2); MS
(CI, CH4) m/z 402 (M+ + NH4+). Anal. (C22H18ClFO3) C, H, F,
Cl.
1-Ch lor o-2-(3,4-d ib en zyloxy-2-flu or op h en yl)et h a n -2-
on e (9b). The same procedure used for the preparation of
ketone 9d provided ketone 9b in 57% yield from aldehyde 7b.
Data for 9b: mp 86-87 °C; 1H NMR (300 MHz, CDCl3) δ 4.66
(d, J ) 3.0, 2H, COCH2), 5.11 and 5.20 (two s, 4H, two ArCH2),
6.86 (d, J ) 7.8, 1H, 5- or 6-ArH), 7.31-7.41 (m, 10H, ArH),
7.69 (t, J ) 8.7, 1H, 5- or 6-ArH); 19F NMR (282 MHz, CDCl3)
δ -48.43 (d, J ) 8.2); MS (CI, CH4) m/z 402 (M+ + NH4+).
Anal. (C22H18ClFO3) C, H, F, Cl.

Synthesis of Fluoronorepinephrines and Fluoroepinephrines
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 8 1617
(1R)-2-Ch lor o-1-(3,4-d iben zyloxy-6-flu or op h en yl)eth a -
n ol (11d ). A solution of (R)-2-Me-CBS (10) (1 M in toluene,
26 µL, 0.026 mmol, 0.1 eq.) and BH3‚THF (1 M in THF, 26
µL, 0.026 mmol, 0.1 eq.) in THF (1 mL) was allowed to stir for
10 min at ambient temperature. BH3‚THF (1 M in THF, 130
µL, 0.13 mmol, 0.5 eq.) and a solution of ketone 9d (66 mg,
0.26 mmol) in THF (1 mL) were added simultaneously, and
the reaction mixture was allowed to stir for an additional 10
min. The reaction mixture was cooled to 0 °C and quenched
by the addition of MeOH (0.15 mL). After 5 min, the reaction
mixture was diluted with 1 N HCl in ethyl ether (5 mL),
thereby producing a white precipitate in 30 min. The solvent
was removed, and the residue was subjected to preparative
thin-layer chromatography to afford 57 mg (85%) of alcohol
11d as colorless needles. A single recrystallization from EtOAc/
(cyclohexane/ether, 1/1). The cyanohydrin was separated from
the remaining salen by chromatography (cyclohexane/ether,
4/1 to 1/1).
The trimethylsilyl cyanohydrin was dissolved in ether (10
mL) and added to a cold (0 °C) suspension of LiAlH4 (90 mg,
2.5 mmol) in ether (10 mL). After the reaction was allowed to
stir at room temperature for 3 h, 90 µL of H2O, 90 µL of NaOH
(15%), 270 µL of H2O, and a small amount of MgSO4 were
added. The suspension was filtered, and the filter cake was
washed three times with hot ethyl acetate. The solvent was
evaporated, and the product was purified by chromatography
(CH2Cl2/MeOH, 9/1-7/3) and recrystallization (ethyl acetate/
hexanes) to provide the (1S) isomer of amino alcohol 15.
Substitution of salen (S,S)-13 in this procedure provides the
(1R) isomers.
(1S)-2-Am in o-1-(3,4-d iben zyloxyp h en yl)eth a n ol [(S)-
15a ]. Using the enantioselective cyanohydrin procedure with
salen (R,R)-13, amino alcohol 15a was obtained as colorless
needles in 25% yield from 3,4-dibenzyloxybenzaldehyde. Data
for (+)-15a : mp 89-92 °C; [R]D20 +16.38° (CH2Cl2, c ) 1.007);
1H NMR (300 MHz, CDCl3) δ 2.08 (br s, 1H, OH), 3.00-2.65
(m, 2H, 2-H2), 4.57 (m, 1H, 1-H), 5.12 and 5.16 (two s, 4H,
two ArCH2), 7.00-6.75 (m, 3H, 2,5,6 ArH), 7.52-7.18 (m, 10H,
two Ph-H); MS (EI, 70 eV) m/z 349 (M+); tR (-)-15a , 29.2 min
(4%); tR (+)-15a , 27.0 min (96%) [Phenominex 3022, hexane/
(120/20/1) dichloroethane/MeOH/TFA, 75/25, 1.0 mL/min].
hexane provided 11d in 95% enantiomeric excess. Data for (R)-
1
11d : mp 83-85 °C; [R]20 -12.2° (CHCl3, c ) 1.02); H NMR
D
(300 MHz, CDCl3) δ 2.61 (d, J ) 3.9, 1H, -OH), 3.54 (dd, J )
8.7, 10.5, 1H, 2-Ha), 3.74 (dd, J ) 3, 10.8, 1H, 2-Hb), 5.09 (m,
1H, 1-H), 5.12 (s, 4H, two ArCH2), 6.66 (d, J ) 11.7, 1H, 3- or
6-ArH), 7.09 (d, J ) 6.9, 1H, 3- or 6-ArH), 7.44-7.31 (m, 10H,
2ArH); 19F NMR (282 MHz, CDCl3) δ -49.26 (d, J ) 16.4);
MS (CI, CH4) m/z 404 (M+ + NH4+); tR (-)-11d , 46.2 min
(97.5%); tR (+)-11d , 50.5 min (2.5%) (hexane/2-propanol, 90/
10, 0.5 mL/min). Anal. (C22H20ClFO3) C, H, F, Cl.
(1R)-2-Ch lor o-1-(3,4-d iben zyloxy-2-flu or op h en yl)eth a -
n ol (11b). The same procedure used for the preparation of
alcohol 11d provided alcohol 11b in 80% yield from ketone 9b.
(1S)-2-Am in o-1-(3,4-d iben zyloxy-6-flu or op h en yl)eth a -
n ol [(S)-15d ]. Using the enantioselective cyanohydrin proce-
dure with salen (R,R)-13, amino alcohol 15d (172 mg, 47%)
Data for 11b: mp 61-62 °C; [R]20 -20.7° (CHCl3, c ) 7.2);
D
1H NMR (300 MHz, CDCl3) δ 2.61 (d, J ) 3, 1H, OH), 3.60
(dd, J ) 8.8, 10.5, 1H, 2-Ha), 3.76 (dd, J ) 2.7, 10.5, 1H, 2-Hb),
5.10 and 5.12 (two apparent s, 5H, ArCH2 and 1-H), 6.78 (d, J
) 9, 1H, 5- or 6-ArH), 7.14 (t, J ) 7.8, 1H, 5- or 6-ArH), 7.30-
7.41 (m, 10H, ArH); 19F NMR (282 MHz, CDCl3) δ -59.09 (d,
J ) 8.5); MS (CI, CH4) m/z 404 (M+ + NH4+); tR (-)-11b, 15.3
min (99.5%); tR (+)-11b, 19.0 min (0.5%) (hexane/2-propanol,
90/10, 0.5 mL/min). Anal. (C22H20ClFO3) C, H, F.
was obtained from aldehyde 1d (336 mg, 1 mmol). Data for
1
(+)-15d : mp 95-99 °C; [R]20 +11.36° (CH2Cl2, c ) 1.04); H
D
NMR (300 MHz, CDCl3) δ 2.20 (br s, 3H, OH, NH2), 2.78 and
2.97 (two br m, 2H, 2-H2), 4.85 (m, 1H, 1-H), 5.17 and 5.20
(two s, 4H, ArCH2), 6.64 (d, J ) 11.4, 3-ArH), 7.08 (d, J ) 7.2,
6-ArH), 7.21-7.44 (m, 10H, 2 ArH); 19F NMR (282 MHz) δ
-126.22 (dd, J ) 11.4, 7.2); MS (EI, 70 eV) m/z 367 (M+); tR
(-)-15d , 33.7 min (4%); tR (+)-15d , 31.1 min (96%) [Phenom-
inex 3022, hexane/(120/20/1) dichloroethane/MeOH/TFA, 75/
25, 1.0 mL/min].
(1R)-2-Azid o-1-(3,4-d ib en zyloxy-2-flu or op h en yl)et h a -
n ol (12d ). A mixture of chloro alcohol 11d (196 mg, 0.51
mmol), 120 mg of NaN3 (120 mg, 1.9 mmol, 3.7 eq.), and KI
(50 mg, 0.3 mmol, 0.6 eq.) in DMF (5 mL) was heated at 110
°C for 8 h under an atmosphere of argon. The reaction mixture
was concentrated in vacuo, and the residue was dissolved in
ethyl acetate (100 mL). The organic phase was washed with
brine, dried over anhydrous sodium sulfate, and concentrated.
The residue was purified by chromatography (petroleum ether/
ethyl acetate, 4/1) to afford 175 mg (88%) of azide 12d as a
(1S)-2-Am in o-1-(3,4-d iben zyloxy-2-flu or op h en yl)eth a -
n ol [(S)-15b]. Using the enantioselective cyanohydrin proce-
dure with salen (R,R)-13, amino alcohol 15b (170 mg, 46%)
was obtained from aldehyde 7b (336 mg, 1 mmol). Data for
(+)-15b: mp 131 °C; [R]20D +30.19° (CH2Cl2, c ) 1.02); 1H NMR
(300 MHz, CDCl3) δ 2.40 (br s, 3H, OH, NH2), 2.78 and 3.00
(two br m, 2H, 2-H2), 4.86 (m, 1H, 1-H), 5.08 (s, 4H, two
ArCH2), 6.75 (dd, 1H, J ) 8.1, 17, 5-ArH), 7.09 (dd, 1H, J )
8.1, 7.9, 6-ArH), 7.43-7.23 (m, 10H, two PhH); 19F NMR (282
MHz) δ -135.65 (d, J ) 7.9); MS (EI, 70 eV) m/z 367 (M+); tR
(-)-15b, 37.8 min (0.5%); tR (+)-15b, 40.8 min (99.5%) [Phe-
nominex 3022, hexane/(120/20/1) dichloroethane/MeOH/TFA,
75/25, 1.0 mL/min].
light yellow oil. Data for 12d : [R]20 -30.7° (CHCl3, c ) 1.1);
D
1H NMR (300 MHz, CDCl3) δ 3.36 (dd, J ) 7.8, 12.6, 1H, 2-Ha),
3.42 (dd, J ) 3.9, 12.6, 1H, 2-Hb), 5.07 (m, 1H, 1-H), 5.11 (s,
4H, ArCH2), 6.65 (d, J ) 10.8, 1H, 3- or 6-ArH), 7.07 (d, J )
7.8, 1H, 3- or 6-ArH), 7.31-7.44 (m, 10H, 2ArH); 19F NMR
(282 MHz, CDCl3) δ -49.45 (dd, J ) 8.2, 12.4); MS (CI, CH4)
m/z 411(M+ + NH4+). Anal. (C22H20FN3O3) C, H, F, N.
(1R)-2-Am in o-1-(3,4-d iben zyloxy-6-flu or op h en yl)eth a -
n ol [(R)-15d ]. Using the enantioselective cyanohydrin proce-
dure with salen (S,S)-13, amino alcohol 15d was obtained in
(1R)-2-Azid o-1-(3,4-d ib en zyloxy-2-flu or op h en yl)et h a -
n ol (12b). The same procedure used for the preparation of
azide 12d provided azide 12b in 57% yield from chloro alcohol
22% yield from aldehyde 7d . Data for (-)-15d : mp 77-80 °C;
1
[R]20 -9.1° (CH2Cl2, c ) 0.5); H NMR (300 MHz, CDCl3) δ
D
11b. Data for 12b: mp 65-66 °C; [R]20 -49.9° (CHCl3, c )
2.13 (broad s, NH2), 2.73 (dd, J ) 7.8, 12.6, 1H, 2-Ha), 2.97
(dd, J ) 3.9, 12.6, 1H, 2-Hb), 4.85 (dd, J ) 3.9, 7.8, 1H, 1-H),
5.10 (s, 4H, 2-PhCH2), 6.63 (d, J ) 12.0, 1H, 3-ArH), 7.09 (d,
J ) 6.9, 1H, 6-ArH), 7.29-7.44 (m, 10H, ArH); 19F NMR (282
MHz, CDCl3) δ -50.00 (dd, J ) 6.2, 10.4); MS (CI, CH4) m/z
368 (M+ + 1); tR (-)-15d , 33.7 min (4%); tR (+)-15d , 31.1 min
(96%) [Phenominex 3022, hexane/(120/20/1) dichloroethane/
D
1
3.5); H NMR (300 MHz, CDCl3) δ 2.31 (d, J ) 3.9, 1H, OH),
3.39-3.50 (m, 2H, 2-Ha and 2-Hb), 5.10-5.13 (m, 5H, two
ArCH2 and 1-H), 6.78 (d, J ) 9, 1H, 5- or 6-Ar-H), 7.13 (dd, J
) 7.8, 8.1, 1H, 5- or 6-ArH), 7.31-7.43 (m, 10H, ArH); 19F NMR
(282 MHz, CDCl3) δ -59.26 (d, J ) 8.5 Hz); MS (CI, CH4) m/z
411 (M+ + NH4+). Anal. (C22H20FN3O3) C, H, F, N.
MeOH/TFA, 75/25, 1.0 mL/min]. HRMS (CI): Calcd for C22H22
FNO3, 367.1584; Found, 367.1575.
-
Gen er a l P r oced u r e for th e P r ep a r a tion of Non r a ce-
m ic Am in o Alcoh ols 15 by En a n tioselective Tr im eth yl-
silyl Cya n oh yd r in P r ep a r a tion /Cya n oh yd r in Red u ction .
Ti(O-i-Pr)4 (25 µL, 0.013 mmol) was added to a solution of salen
(R,R)-13 (85 mg, 0.155 mmol) in CH2Cl2 (3 mL), and the
solution was allowed to stir for 1.5 h at room temperature.
The reaction mixture was cooled to -50 °C, and a solution of
the aldehyde (1 mmol) and TMSCN (400 mg, 3.6 mmol) in
CH2Cl2 (3 mL) was added. After 5 days at -50 °C, the catalyst
was removed by filtration through a column of silica gel
(1R)-2-Am in o-1-(3,4-d iben zyloxy-2-flu or op h en yl)eth a -
n ol [(R)-15b]. Using the enantioselective cyanohydrin proce-
dure with salen (S,S)-13, amino alcohol 15b was obtained as
colorless needles in 33% yield from aldehyde 7b. Data for (-)-
15b: mp 123-125 °C; [R]20D -26° (CH2Cl2, c ) 0.25); 1H NMR
(300 MHz, CDCl3) δ 1.78 (br s, 2Η, ΝΗ2), 2.76 (dd, J ) 12.6,
7.8, 1H, 2-Ha), 3.01 (dd, J ) 12.6, 3.0, 1H, 2-Hb), 4.86 (dd, J )
6.9, 3.9, 1H, 1-H), 5.10 and 5.11 (two s, 4H, two ArCH2), 6.76

1618 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 8
Lu et al.
(dd, J ) 8.7, 1.8, 1H, 5-ArH), 7.11 (dd, J ) 8.7, 7.8, 1H, 6-ArH),
7.30-7.44 (m, 10H, ArH); 19F NMR (282 MHz, CDCl3) δ -59.82
(d, J ) 8.2); MS (CI, CH4) m/z 368 (M+ + 1); tR (-)-15b, 37.8
min (99.5%); tR (+)-15b, 40.8 min (0.5%) [Phenominex 3022,
hexane/(120/20/1) dichloroethane/MeOH/TFA, 75/25, 1.0 mL/
min]. HRMS (CI): Calcd for C22H22FNO3, 367.1584; Found,
367.1586.
δ -59.95 (d, J ) 8.5); tR (-)-17b, 20.6 min (0%); tR (+)-17b,
23.0 min (100%) (hexane/2-propanol, 80/20, 0.5 mL/min); MS
(CI, CH4) m/z 382 (M+ + 1). HRMS (CI) Calcd for C23H24FNO3,
381.1740; Found, 381.1740.
(1R)-2-Am in o-1-(3,4-d ih yd r oxy-6-flu or op h en yl)et h a -
n ol Oxa la te Sa lt: (R)-6-F lu or on or ep in ep h r in e Oxa la te
Sa lt [(R)-1d ]. A mixture of azide (R)-12d (414 mg, 1.05 mmol),
oxalic acid (45 mg, 0.5 mmol, 0.5 eq.), and 10% Pd-C in
methanol (5 mL) was allowed to stir under an atmosphere of
H2 (45 psi) for 24 h. The reaction mixture was filtered under
a blanket of N2, and the solution was concentrated in vacuo.
The solid residue was recrystallized from methanol to afford
185 mg (76%) of oxalate (R)-1d as an off-white solid. Data for
(1R)-1-(3,4-Diben zyloxy-6-flu or op h en yl)-2-(N-m eth yl-
a m in o)eth a n ol [(R)-17d ]. A solution of (R)-15d (662 mg, 1.8
mmol) in ethyl formate (20 mL) was heated to reflux under
an atmosphere of argon. After 8.5 h, the volatiles were removed
in vacuo to afford the formamide 16d (676 mg, 93%) as a
colorless solid. To a cold (0 °C) suspension of LiAlH4 (376 mg)
in THF (5 mL) under argon was added dropwise a solution of
16d (676 mg, 1.7 mmol) in THF (20 mL). The mixture was
heated at reflux for 3 h and then cooled to 0 °C and quenched
by the addition of H2O (0.36 mL), 15% NaOH (0.36 mL), and
H2O (1.1 mL). Anhydrous MgSO4 (16 g) was added, and the
solution was stirred vigorously for 15 min. The solid was
filtered and washed with hot EtOAc (5 × 15 mL). The filtrate
was concentrated, and the residue was diluted with CH2Cl2
and filtered through a short column of silica gel (CH2Cl2/
MeOH, 7/3) to afford 17d (409 mg) as a white solid. Recrys-
tallization from hexanes-EtOAc resulted in a low recovery of
mass with low optical purity (e.e. ) 14%). However, evapora-
tion of the mother liquor gave 185 mg (26% from 15d ) of (R)-
17d with 92% e.e. Data for (-)-17d : mp 95-97 °C; [R]20D -2.8°
(-)-1d : mp 165-166 °C (decomp.); [R]20 -27.5° (MeOH, c )
D
0.3);18 1H NMR (300 MHz, D2O) δ 3.40-3.22 (m, 2H, 2-Ha and
2-Hb), 5.15 (dd, J ) 5.1, 7.8, 1H, 1-H), 6.74 (d, J ) 11.7, 1H, 2-
or 5-ArH), 6.97 (d, J ) 6.9, 1H, 2- or 5-ArH); 19F NMR (282
MHz, D2O, CF3CO2H as external standard) δ -51.79 (dd, J )
5 0.9, 12.4); MS (CI, CH4) m/z 188 (M+ + 1), 205 (M+ + NH4+);
tR (-)-1d , 18.4 min (99.5%); tR (+)-1d , 23.0 min (0.5%)
(CROWNPAK CR (+), 0.02 M HClO4, 0.7 mL/min).
(1R)-2-Am in o-1-(3,4-d ih yd r oxy-2-flu or op h en yl)et h a -
n ol Oxa la te Sa lt: (R)-2-F lu or on or ep in ep h r in e Oxa la te
Sa lt [(R)-1b]. The same procedure used for the preparation
of oxalate (R)-1d provided oxalate (R)-1b in 51% yield from
azide (R)-12b. Data for (-)-1b: mp 70-73 °C; [R]20 -35.9°
D
(MeOH, c ) 1.9);18 1H NMR (300 MHz, D2O) δ 3.33-3.28 (m,
2H, 1-Ha and 1-Hb), 5.17 (m, 1H, 2-H), 6.70-6.91 (m, 2H, 5-
and 6-ArH); 19F NMR (282 MHz, D2O, CF3CO2H as external
standard) -65.43 (m) ppm; MS (CI, CH4) m/z 188 (M+ + 1),
205 (M+ + NH4+); tR (-)-1b, 10.1 min (99.5%); tR (+)-1b, 12.8
min (0.5%) (CROWNPAK CR (+), 0.02 M HClO4, 0.7 mL/min).
1
(CH2Cl2, c ) 0.4); H NMR (300 MHz, CDCl3) δ 2.44 (s, 3H,
CH3), 2.63 (dd, J ) 8.7, 11.7, 1H, 2-Ha), 2.81 (dd, J ) 3.0, 11.7,
1H, 2-Hb), 4.95 (dd, J ) 2.7, 7.8, 1H, 1-H), 5.11 and 5.12 (two
s, 4H, two ArCH2), 6.64 (d, J ) 11.7, 1H, 3-ArH), 7.12 (d, J )
7.8 Hz, 1H, 6-ArH), 7.26-7.45 (m, 10H, Ar-H); 19F NMR (282
MHz, CDCl3) δ -50.30 (dd, J ) 7.3, 11.3); MS (CI, CH4) m/z
382 (M+ + 1); tR (-)-17d , 21.5 min (97%); tR (+)-17d , 30.0 min
(3%) (hexane/2-propanol, 80/20, 0.5 mL/min). HRMS(CI):
Calcd for C23H24FNO3, 381.1740; Found, 381.1732.
(1S)-2-Am in o-1-(3,4-d ih yd r oxy-6-flu or op h en yl)et h a -
n ol Oxa la te Sa lt: (S)-6-F lu or on or ep in ep h r in e Oxa la te
Sa lt [(S)-1d ]. The same procedure used for the preparation
of oxalate (R)-1d provided oxalate (S)-1d in 87% yield from
(1R)-1-(3,4-Diben zyloxy-2-flu or op h en yl)-2-(N-m eth yl-
a m in o)eth a n ol [(R)-17b]. The same procedure used for the
preparation of N-methylamine (R)-17d provided N-methyl-
amine (R)-17b in 65% yield from amine (R)-15b. Data for (-)-
17b: mp 100-101 °C; [R]20D -30.5° (CH2Cl2, c ) 0.2); 1H NMR
(300 MHz, CDCl3) δ 2.45 (s, 3H, CH3), 2.68 (dd, J ) 9.0, 12.0,
1H, 2-Ha), 2.83 (dd, J ) 3.9, 12.9, 1H, 2-Hb), 4.97 (dd, J ) 3.9,
8.7, 1H, 1-H), 5.09 and 5.10 (two s, 4H, two ArCH2), 6.75 (dd,
J ) 8.7, 1.8, 1H, 5-ArH), 7.13 (dd, J ) 7.8, 8.7, 1H, 6-ArH),
7.28-7.44 (m, 10H, Ar-H); 19F NMR (282 MHz, CDCl3) δ
-59.93 (d, J ) 8.5); MS (CI, CH4) m/z 382 (M+ + 1); tR (-)-
17b, 20.6 min (100%); tR (+)-17b, 23.0 min (0%) (hexane/2-
amine (S)-15d . Data for (+)-1d : [R]20 +36.3° (MeOH, c )
D
0.21);18 1H NMR (300 MHz, D2O) δ 3.23-3.35 (m, 2H, 2-Ha
and 2-Hb), 5.15 (dd, J ) 5.1, 7.8, 1H, 1-H), 6.75 (d, J ) 11.7,
1H, 2- or 5-ArH), 6.97 (d, J ) 6.7, 1H, 2- or 5-ArH); 19F NMR
(282 MHz, D2O, CF3CO2H as external standard) δ -51.67 to
-51.73 (m); MS (CI, CH4) m/z 188 (M+ + 1), 205 (M + NH4+);
t
R (-)-1d , 18.4 min (1.5%); tR (+)-1d , 23 min (98.5%) (CROWN-
PAK CR (+), 0.02 M HClO4, 0.7 mL/min).
(1S)-2-Am in o-1-(3,4-d ih yd r oxy-2-flu or op h en yl)et h a -
n ol Oxa la te Sa lt: (S)-2-F lu or on or ep in ep h r in e Oxa la te
Sa lt [(S)-1b]. The same procedure used for the preparation
of oxalate (R)-1d provided oxalate (S)-1b in 68% yield from
amine (S)-15b. Data for (+)-1b: [R]20 +41.8° (MeOH, c )
propanol, 80/20, 0.5 mL/min). HRMS (CI) Calcd for C23H24
FNO3, 381.1740; Found, 381.1749.
-
D
0.76);18 1H NMR (300 MHz, D2O) δ 3.34-3.24 (m, 2H, 2-Ha
and 2-Hb), 5.16 (dd, J ) 4.8, 7.8, 1H, 1-H), 6.76-6.92 (m, 2H,
5- and 6-ArH); 19F NMR (282 MHz, D2O, CF3CO2H as external
standard) -65.38 (m); MS (CI, CH4) m/z 188 (M+ + 1), 205 (M
+ NH4+); tR (-)-1b, 10.1 min (2.5%); tR (+)-1b, 12.8 min (97.5%)
(CROWNPAK CR (+), 0.02 M HClO4, 0.7 mL/min).
(1S)-1-(3,4-Dib en zyloxy-6-flu or op h en yl)-2-(N-m et h yl-
a m in o)eth a n ol [(S)-17d ]. The same procedure used for the
preparation of N-methylamine (R)-17d provided N-methyl-
amine (S)-17d in 38% yield from amine (S)-15d . Data for (+)-
1
17d : mp 97-99 °C; [R]20 +3.5° (CH2Cl2, c ) 0.3); H NMR
D
(300 MHz, CDCl3) δ 2.44 (s, 3H, Me), 2.63 (dd, J ) 8.7, 11.7,
1H, 2-Ha), 2.82 (dd, J ) 3, 11.7, 1H, 2-Hb), 4.95 (dd, J ) 3.9,
9.0, 1H, 1-H), 5.11 and 5.12 (two s, 4H, two ArCH2), 6.64 (d, J
) 11.7, 1H, 3-Ar-H), 7.12 (d, J ) 6.9, 1H, 6-ArH), 7.30-7.45
(m, 10H, Ar-H); 19F NMR (282 MHz, CDCl3) δ -50.31 (dd, J
) 7.3, 11.3); MS (CI, CH4) m/z (M+ + 1); tR (-)-17d , 21.5 min
(1.5%); tR (+)-17d , 28.5 min (98.5%) (hexane/2-propanol, 80/
20, 0.5 mL/min). HRMS (CI): Calcd for C23H24FNO3, 381.1740;
Found, 381.1742.
(1S)-1-(3,4-Dib en zyloxy-2-flu or op h en yl)-2-(N-m et h yl-
a m in o)eth a n ol [(S)-17b]. The same procedure used for the
preparation of N-methylamine (R)-17d provided N-methyl-
amine (S)-17b in 94% yield from amine (S)-15b. Data for (+)-
17b: mp 99-100 °C; [R]20D +29.0° (CH2Cl2, c ) 0.36); 1H NMR
(300 MHz, CDCl3) δ 2.46 (s, 3H, CH3), 2.69 (dd, J ) 8.7, 11.7,
1H, 2-Ha), 2.85 (dd, J ) 3.0, 12.6, 1H, 2-Hb), 4.97 (dd, J ) 3.0,
8.7, 1H, 1-H), 5.09 and 5.10 (two s, 4H, two ArCH2), 6.75
(pseudo d, J ) 8.7, 1H, 5-ArH), 7.13 (dd, J ) 7.8, 8.7, 1H,
6-ArH), 7.30-7.43 (m, 10H, ArH); 19F NMR (282 MHz, CDCl3)
(1R)-1-(3,4-Dih ydr oxy-6-flu or oph en yl)-2-(N-m eth ylam i-
n o)eth a n ol Oxa la te Sa lt: (R)-6-F lu or oep in ep h r in e Ox-
a la te Sa lt [(R)-2d ]. The same procedure used for the prepa-
ration of oxalate (R)-1d provided oxalate (R)-2d in 47% yield
from N-methylamine (R)-17d . Data for (-)-2d : [R]20 -46.7°
D
(MeOH, c ) 0.12);18 1H NMR (300 MHz, D2O) δ 2.79 (s, 3H,
CH3), 3.29-3.39 (m, 2H, 2-Ha and 2-Hb), 5.19 (dd, J ) 4.8, 7.8,
1H, 1-H), 6.75 (d, J ) 10.8, 1H, 5-ArH), 6.96 (d, J ) 7.8, 1H,
2-ArH); 19F NMR (282 MHz, D2O, CF3CO2H as external
standard) δ -51.63 (dd, J ) 7.6, 10.7); tR (-)-2d , 18.1 min
(100%); tR (+)-2d , 27.9 min (0%) (hexane/EtOH/TFA/diethyl-
amine, 93/7/0.13/0.08, 1 mL/min); MS (CI, CH4) m/z 202 (M+
+ 1). HRMS (CI): Calcd for C9H12FNO3, 201.0801; Found,
201.0800.
(1R)-1-(3,4-Dih ydr oxy-2-flu or oph en yl)-2-(N-m eth ylam i-
n o)eth a n ol Oxa la te Sa lt: (R)-2-F lu or oep in ep h r in e Ox-
a la te Sa lt [(R)-2b]. The same procedure used for the prepa-
ration of oxalate (R)-1d provided oxalate (R)-2b in 91% yield
from N-methylamine (R)-17b. Data for (-)-2b: [R]20 -40°
D

Synthesis of Fluoronorepinephrines and Fluoroepinephrines
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 8 1619
(MeOH, c ) 0.14);18 1H NMR (300 MHz, D2O) δ 2.79 (s, 3H,
CH3), 3.29-3.42 (m, 2H, 2-Ha and 2-Hb), 5.21 (dd, J ) 3.9, 7.8,
1H, 1-H), 6.95-6.76 (m, 2H, 5-ArH and 6-ArH); 19F NMR (282
MHz, D2O, CF3CO2H as external standard) δ -65.34 (d, J )
7.6); tR (-)-2b, 20.9 min (100%); tR (+)-2b, 18.6 min (0%)
(hexane/EtOH/TFA/diethylamine, 93/7/0.13/0.08, 1 mL/min);
(4) Adejare, A.; Gusovsky, F.; Creveling, C. R.; Daly, J . W.; Kirk,
K. L.: Synthesis and Adrenergic Activities of Ring-Fluorinated
Epinephrines, J . Med. Chem. 1988, 31, 1972-1977.
(5) Kirk, K. L.; Cantancuzene, D.; Collins, B.; Chen, G. T.; Nimit,
Y.; Creveling, C. R. Syntheses and Adrenergic Agonist Properties
of Ring-Fluorinated Isoproterenols. J . Med. Chem. 1982, 25,
680-684.
(6) Adejare, A.; Nie, J .-y.; Hebel, D.; Brackett, L. E.; Choi, O.;
Gusovsky, F.; Padgett, W. L.; Daly, J . W.; Creveling, C. R.; Kirk,
K. L. Effect of Fluorine Substitution on the Adrenergic Properties
of 3-(tert-Butylamino)-1-(3,4-dihydroxyphenoxy)-2-propanol. J .
Med. Chem. 1991, 34, 1063-1068.
(7) Kirk, K. L.; Olubajo, O.; Buchhold, K.; Lewandowski, G. A.;
Gusovsky, F.; McCulloh, D.; Daly, J . W.; Creveling, C. R.
Synthesis and Adrenergic Activity of Ring-Fluorinated Phenyl-
ephrines. J . Med. Chem. 1986, 29, 1982-1988.
(8) Nimit, Y.; Cantacuzene, D.; Kirk, K. L.; Creveling, C. R.; Daly,
J . W. The Binding of Fluorocatecholamines to Adrenergic and
Dopaminergic Receptors in Rat Brain Membranes. Life Sci.
1980, 27, 1577-1585.
(9) Nie, J .-Y.; Shi, D.; Daly, J . W.; Kirk, K. L. Synthesis of
Fluorodopamines: Effect of Aryl Fluoro Substituents on Affini-
ties for Adrenergic and Dopaminergic Receptors. Med. Chem.
Res. 1996, 6, 318-331.
(10) Cantacuzene, D.; Kirk, K. L. Unpublished results.
(11) Ding, Y. S.; Fowler, J . S.; Gatley, S. J .; Dewey, S. L.; Wolf, A. P.
Synthesis of High Specific Activity (+) and (-)-6-[18F]Fluoro-
norepinephrine via the Nucleophilic Aromatic Substitution
Reaction. J . Med. Chem. 1991, 34, 767-771.
(12) Corey, E. J .; Bakshi, R. K.; Shibata, S. Highly Enantioselective
Borane Reduction of Ketones Catalyzed by Chiral Oxazaboro-
lidines. Mechanism and Synthetic Implications. J . Am. Chem.
Soc. 1987, 109, 5551-5553.
MS (CI, CH4) m/z 202 (M+ + 1). HRMS (CI): Calcd for C9H12
-
FNO3, 201.0801; Found, 201.0797.
(1S)-1-(3,4-Dih yd r oxy-6-flu or op h en yl)-2-(N-m eth yla m i-
n o)eth a n ol Oxa la te Sa lt: (S)-6-F lu or oep in ep h r in e Ox-
a la te Sa lt [(S)-2d ]. The same procedure used for the prepa-
ration of oxalate (R)-1d provided oxalate (S)-2d in 58% yield
from N-methylamine (S)-17d . Data for (+)-2d : [R]20 +35°
D
(MeOH, c ) 0.13);18 1H NMR (300 MHz, D2O) δ 2.79 (s, 3H,
CH3), 3.28-3.39 (m, 2H, 2-Ha and 2-Hb), 5.19 (dd, J ) 4.8, 7.8,
1H, 1-H), 6.74 (d, J ) 10.8, 1H, 5-ArH), 6.96 (d, J ) 6.9, 1H,
2-ArH); 19F NMR (282 MHz, D2O, CF3CO2H as external
standard) δ -51.60 (dd, J ) 5.9, 10.4); tR (-)-2d , 18.1 min (0%);
tR (+)-2d , 27.9 min (100%) (hexane/EtOH/TFA/diethylamine,
93/7/0.13/0.08, 1 mL/min); MS (CI, CH4) m/z 202 (M+ + 1).
HRMS (CI): Calcd for C9H12FNO3, 201.0801; Found, 201.0795.
(1S)-1-(3,4-Dih yd r oxy-2-flu or op h en yl)-2-(N-m eth yla m i-
n o)eth a n ol Oxa la te Sa lt: (1S)-2-F lu or oep in ep h r in e Ox-
a la te Sa lt [(S)-2b]. The same procedure used for the prepa-
ration of oxalate (R)-1d provided oxalate (S)-2b in 73% yield
from N-methylamine (S)-17b. Data for (+)-2b: [R]20 +43°
D
(MeOH, c ) 0.17);18 1H NMR (300 MHz, D2O) δ 2.79 (s, 3H,
CH3), 3.27-3.42 (m, 2H, 2-Ha and 2-Hb), 5.22 (dd, J ) 3.9, 7.8,
1H, 2-H), 6.76-6.95 (m, 2H, 5-ArH and 6-ArH); 19F NMR (282
MHz, D2O, CF3CO2H as external standard) δ -65.32 (d, J )
7.6); tR (-)-2b, 20.9 min (0%); tR (+)-2b, 18.6 min (100%)
(hexane/EtOH/TFA/diethylamine, 93/7/0.13/0.08, 1 mL/min);
(13) Corey, E. J .; Link, J . O. A Catalytic Enantioselective Synthesis
of Denopamine, a Useful Drug for Congestive Heart Failure. J .
Org. Chem. 1991, 56, 442-444.
(14) Belekon’, Y.; Flego, M.; Ikonnikov, N.; Moscalenko, M.; North,
M.; Orizu, C.; Tararov, V.; Tasinazzo, M. Asymmetric Addition
of Trimethylsilyl Cyanide to Aldehydes Catalyzed by Chiral
(Salen)TiIV Complexes. J . Chem. Soc., Perkin Trans. I 1997,
1293-1295.
MS (CI, CH4) m/z 202 (M+ + 1). HRMS (CI): Calcd for C9H12
FNO3, 201.0801; Found, 201.0793.
-
Su p p or tin g In for m a tion Ava ila ble: Combustion analy-
sis results. This material is available free of charge via the
Internet at http://pubs.acs.org.
(15) Chen, B.-H.; Daly, J . W.; Kirk, K. L. Investigation of the
Relationship Between Phenol Ionization and Affinity of Nor-
epinephrine for Adrenergic Receptors using Ring-fluorinated
Analogues. Med. Chem. Res. 1993, 3, 589-599.
(16) Lands, A. M.; Arnold, A.; McAuliff, J . P.; Ludnena, F. P.; Brown,
T. G.; Differentiation of Receptor Systems Activated by Sym-
pathomimetic Amines. Nature 1967, 215, 597-598.
(17) Still, W. C.; Kahn, M.; Mitra, A. Rapid Chromatographic
Technique for Preparative Separations with Moderate Resolu-
tion. J . Org. Chem. 1978, 43, 2923-2925.
(18) In our previous work, racemic 2- and 6-fluoronorepinephrines
and epinephrines were isolated consistently as the hemi-
oxalates. In this study, we used NMR, HRMS, and chiral HPLC
to confirm identity and purity, including enantio-purity, of the
target compounds, initially assuming that the hemi-oxalates
again would be obtained. However, comparison of apparent
specific rotations of R- and S- 1b, 1d , and 2d made it clear that
R-1b, R-1d , and S-2d were present as the acid oxalates.
Refer en ces
(1) Reviews: (a) Kirk, K. L.; Filler, R. Biological Properties. In
Biomedical Frontiers of Fluorine Chemistry; Ojima, I., McCarthy,
J . R., Welch, J . T., Eds.; ACS Symposium Series 639; American
Chemical Society: Washington, D.C., 1996; pp 1-24. (b) Welch,
J . T.; Eswarakrishnan, S. Fluorine in Bioorganic Chemistry;
J ohn Wiley & Sons: New York, 1991. (c) Kirk, K. L. Biochemistry
of Halogenated Organic Compounds, Biochemistry of the Ele-
ments, Vols. 9A, 9B; Frieden, E., Series Ed.; Plenum: New York,
1991.
(2) (a) Kirk, K. L. Fluorine Substituted Neuroactive Amines. In
Selective Fluorination in Organic and Bioorganic Chemistry;
Welch, J . T., Ed.; ACS Symposium Series 456; American
Chemical Society: Washington, D.C., 1991; pp 136-155. (b)
Kirk, K. L. Chemistry and Pharmacology of Ring-Fluorinated
Catecholamines. J . Fluorine Chem. 1995, 72, 261-266.
(3) Kirk, K. L.; Cantacuzene, D.; Nimitkitpaisan, Y.; McCulloh, D.;
Padgett, W. L.; Daly, J . W.; Creveling, C. R. Synthesis and
Biological Properties of 2, 5, and 6-Fluoronorepinephrines. J .
Med. Chem. 1979, 22, 1493-1497.
J M990599H