Improving the developability profile of pyrrolidine progesterone receptor partial agonists

Article abstract of DOI:10.1016/j.bmcl.2009.10.092

The previously reported pyrrolidine class of progesterone receptor partial agonists demonstrated excellent potency but suffered from serious liabilities including hERG blockade and high volume of distribution in the rat. The basic pyrrolidine amine was in

Full text of DOI:10.1016/j.bmcl.2009.10.092

Bioorganic & Medicinal Chemistry Letters 20 (2010) 371–374
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Improving the developability profile of pyrrolidine progesterone receptor
partial agonists
a,
a
a
a
Lara S. Kallander a, , David G. Washburn , Tram H. Hoang , James S. Frazee , Patrick Stoy ,
Latisha Johnson a, Qing Lu a, Marlys Hammond a, Linda S. Barton a, Jaclyn R. Patterson a,
Leonard M. Azzarano c, Rakesh Nagilla c, Kevin P. Madauss d, Shawn P. Williams d, Eugene L. Stewart d,
Chaya Duraiswami d, , Eugene T. Grygielko b, Xiaoping Xu b, Nicholas J. Laping b, Jeffrey D. Bray b,
Scott K. Thompson a
*
*
a Department of Chemistry, Metabolic Pathways Centre for Excellence in Drug Discovery, GlaxoSmithKline Pharmaceuticals, 709 Swedeland Road, King of Prussia, PA 19406, USA
b Department of Biology, Metabolic Pathways Centre for Excellence in Drug Discovery, GlaxoSmithKline Pharmaceuticals, 709 Swedeland Road, King of Prussia, PA 19406, USA
c Department of Drug Metabolism and Pharmacokinetics, Metabolic Pathways Centre for Excellence in Drug Discovery, GlaxoSmithKline Pharmaceuticals, 709 Swedeland Road,
King of Prussia, PA 19406, USA
d Deparmtent of Computational and Structural Chemistry, Molecular Discovery Research, GlaxoSmithKline Pharmaceuticals, P.O. Box 13398, Research Triangle Park, NC 27709, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The previously reported pyrrolidine class of progesterone receptor partial agonists demonstrated excel-
lent potency but suffered from serious liabilities including hERG blockade and high volume of distribu-
tion in the rat. The basic pyrrolidine amine was intentionally converted to a sulfonamide, carbamate,
or amide to address these liabilities. The evaluation of the degree of partial agonism for these non-basic
pyrrolidine derivatives and demonstration of their efficacy in an in vivo model of endometriosis is dis-
closed herein.
Received 9 September 2009
Revised 20 October 2009
Accepted 21 October 2009
Available online 25 October 2009
Keywords:
Pyrrolidine
Ó 2009 Elsevier Ltd. All rights reserved.
Progesterone
Progesterone receptor
Partial agonist
hERG
Rat DMPK
Rat OVX model
Endometriosis
O
Me
O
Me
OAc
The progesterone receptor (PR) is a ligand-activated transcrip-
tion factor and a member of the nuclear receptor super-family.1
PR and its endogenous ligand, progesterone (1, Chart 1), have
essential roles in ovulation and implantation.2 For this reason, syn-
thetic PR agonists are used for oral contraception, hormone ther-
apy, leiomyomas, and endometriosis.3
For the treatment of endometriosis, medroxyprogesterone ace-
tate (MPA, 2), 1, and other agonists can act by opposing estrogen-
stimulated endometrial tissue proliferation.4 Side-effects associ-
ated with this therapy, including weight gain, break-through
bleeding, and mood disturbances, have been attributed to PR full
agonist activity and/or lack of nuclear receptor selectivity.5 Despite
recent advances in the identification of non-steroidal PR agonists,6
Me
H
Me
H
H
H
H
H
O
O
Me
2, MPA
1, Progesterone
R2
CF3
N
N
N
CN
N
CN
X R1
Cl
3
Cl
X = SO2 (8-20)
PR binding IC50= 16 nM
PR T47D EC50 = 9 nM (65%)
or C(O)O (21-29)
or CO (30-41)
* Corresponding authors.
E-mail address: lara.s.kallander@gsk.com (L.S. Kallander).
Alternative address: PO Box 5089, Collegeville, PA 19426, USA.
Chart 1. Steroidal and non-steroidal PR ligands.
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.10.092

372
L. S. Kallander et al. / Bioorg. Med. Chem. Lett. 20 (2010) 371–374
H
N
these compounds are likely to induce many of the same side-ef-
fects as full PR agonists. However, a selective partial agonist of
PR that would be able to fully suppress the action of estrogen on
endometrial tissue may reduce the side-effects associated with full
agonists.
H
N
H2N
a
b
N
N
N
CN
CN
R1
Boc
X
c
Boc
Cl
Cl
6
4
5
As part of this effort, we recently reported that ligands containing
a (3S)-3-amino-1-pyrrolidine template (e.g., 3) had excellent PR
binding potency and partial agonist activity as measured in a human
T47D-based alkaline phosphatase PR agonist assay, and >100-fold
selectivity over several nuclear receptors, including the glucocorti-
coid receptor (GR), mineralocorticoid receptor (MR), and estrogen
receptor (ER).7 However, in the course of profiling these compounds,
it was determined that compound 3 suffered from potent hERG
c
R2
R2
N
b
N
N
CN
N
CN
Boc
Cl
X
R1
channel blockade (0.78 lM), potent androgen receptor (AR) binding
Cl
7
affinity (250 nM), and poor rat PK parameters including high clear-
ance, a large volume of distribution and low oral exposure. In addi-
tion, compound 3 demonstrated modest inhibition of the human
X = SO2 (8-20)
or C(O)O (21-29)
or CO (30-41)
cytochrome P450 2D6 (1.6 lM) and 2C9 (1.1 lM) isoenzymes. Be-
Scheme 1. Reagents and conditions: (a) 2-chloro-4-fluorobenzonitrile, NaHCO3,
H2O/DMSO (78%); (b) (i) TFA, CH2Cl2; (ii) Et3N/DIEA/pyridine, CH2Cl2, R1XCl [or for
38–41 (2S)-2-{[(1,1-dimethylethyl)(dimethyl)silyl]oxy}propanoic acid, NMM, i-
butyl chloroformate, then TBAF]; (c) NaH, DMF, R2BnBr.
cause compounds with basic amines are known to be associated
with both hERG inhibition and high volumes of distribution, we ex-
plored analogs where the basicity of the pyrrolidine nitrogen was
attenuated.8 We have previously disclosed that conversion of the
pyrrolidine ring to a pyrrolidinone, thereby reducing the basicity
of the ring nitrogen, improved all of the aforementioned liabilities
while maintaining excellent potency and partial agonism.9 We
now report that pyrrolidine sulfonamides, carbamates, and amides
share the same properties while having the added benefit of ease
of synthesis as well as the ability to deliver compounds with a broad
range of agonist efficacy (37–77%).
The short synthetic route to prepare compounds 8–41 is shown
in Scheme 1.13 It is straightforward to diversify at two different
positions, R1 or R2, by bringing in either group in the final step;
steps b and c can be performed in either order. In either case, the
first step involved the primary amine of (S)-3-amino-1-tert-but-
oxycarbonylpyrrolidine displacing the fluoride of 2-chloro-4-fluo-
robenzonitrile to give the secondary amine 5. For diversification
of the aniline nitrogen, step b was conducted next, beginning with
removal of the BOC protecting group from the pyrrolidine 5 under
acidic conditions to yield the amine, which was then reacted with a
sulfonyl chloride, acid chloride or alkyl chloroformate to yield
intermediate 6. Lastly, benzylation of the aniline nitrogen under
basic conditions afforded the target products. Alternatively, diver-
sification of the pyrrolidine nitrogen in the last step began with
incorporation of the substituted benzyl group onto the aniline
nitrogen of 5 to form intermediate 7. BOC removal under acidic
conditions afforded the free pyrrolidine which was then trans-
formed to the target analogs via reaction with a sulfonyl chloride,
acid chloride or alkyl chloroformate.
The sulfonylpyrrolidines were tested initially in a binding assay
for the progesterone receptor; those results are listed in Table 1.
Since the limit of detection in that assay is around 20 nM, most
compounds were also tested in a cellular assay format (T47D)
which reports potency and efficacy relative to progesterone
(100%). It is interesting to note that the potency at the PR receptor
generally decreases by extending the R1 alkyl chain from methyl to
ethyl, propyl, and butyl. In the cellular assay, low nanomolar po-
tency is achieved by both methanesulfonamide 8 and isobutane-
sulfonamide 12. Two of the most important benefits achieved
with the alkanesulfonamides were the improvement in AR selec-
tivity and reduced P450 inhibition, specifically the 2D6 isoform.
Consistent with the SAR observed with the N-alkyl substituted pyr-
rolidines, ortho substitution of the aniline benzyl group afforded
optimal PR binding potency.7 Though the chloro (18) and bromo
(19) substituted compounds are exceptionally potent, they are less
preferred due to their high levels of intrinsic agonism as measured
by the T47D assay (87/88%). Both compounds 17 and 20 have a
very promising overall in vitro profile.
Data for pyrrolidine carbamates is given in Table 2. Similar to
the pyrrolidine sulfonamides, substitution of the benzyl 2-position
led to compounds with improved PR binding affinity and AR selec-
tivity (22–25). Disubstitution of the R2 benzyl ring (26–27) reduces
the cell potency and efficacy and increasing the size of the R1 sub-
Table 1
In vitro data for 3 and sulfonamide compounds (X = SO2) 8–20a
Compd
R1
R2
PR IC50 (nM)
AR/PR
T47D EC50, nM (% P4)
2D6 IC50
(l
M)
2C9 IC50 (lM)
3
8
9
Chart 1
Me
Et
i-Pr
n-Pr
i-Bu
n-Bu
Me
Me
Me
Me
Me
16b
13b
40
40
80
80
63
63
16
480
62
9 (65%)
11 (82)
3160 (26)
>10,000
2630 (41)
10 (65)
NA
843 (53)
NA
1130 (53)
36 (66)
1 (88)
1.6
15
13
25
9.0
13
NA
14
NA
>33
18
31
31
22
1.1
2-CF3
2-CF3
2-CF3
2-CF3
2-CF3
2-CF3
H
3-CF3
2-CN
2-F
0.33
0.35
0.28
0.11
0.16
NA
2.4
NA
1.5
1.7
10
11
12
13
14
15
16
17
18
19
20
62
>125
>125
>160
160
16
>100
315
>1000
>600
250
100
100
20b
6b
2-Cl
2-Br
2-Me
0.87
0.69
2.4
Me
Me
5b
3 (87)
25 (77)
16b
a
Values are the mean of P2 determinations.
Tight-binding limit of the assay.
b

L. S. Kallander et al. / Bioorg. Med. Chem. Lett. 20 (2010) 371–374
373
Table 2
In vitro data for carbamate compounds (X = C(O)O) 21–29a
Compd
R1
R2
PR IC50 (nM)
AR/PR
T47D EC50, nM (% P4)
2D6 IC50
(l
M)
2C9 IC50 (lM)
21
22
23
24
25
26
27
28
29
Me
Me
Me
Me
Me
Me
Me
Et
H
2-CF3
2-F
2-Me
63
50
333
260
260
160
125
250
167
32
NA
8 (66)
165 (51)
4 (64)
0.1 (78)
137 (51)
41 (62)
364 (43)
233 (44)
7.5
8.0
7.3
22.0
11.0
6.7
28.0
22.0
15.0
1.3
0.16
1.4
0.68
1.1
1.2
1.1
1.5
1.1
12b
12b
12b
8b
2-Cl
2-F, 3-F
2-Me, 5-F
2-Me, 5-F
2-Me, 5-F
20
20
30
50
i-Pr
a
Values are the mean of P2 determinations.
Tight-binding limit of the assay.
b
Table 3
In vitro data for amide compounds (X = CO) 30–41a
Compd
R1
R2
PR IC50 (nM)
AR/PR
T47D EC50, nM (% P4)
2D6 IC50
(l
M)
2C9 IC50 (lM)
30
31
32
33
34
35
36
37
38
39
40
41
Me
Et
i-Pr
t-Bu
Et
Et
Et
Et
(S)-CH (OH)CH3
(S)-CH (OH)CH3
(S)-CH (OH)CH3
(S)-CH (OH)CH3
2-CF3
2-CF3
2-CF3
2-CF3
2-Br
2-Me
2-Me, 5-F
2-Cl, 5-Cl
2-CF3
30
25
25
50
333
320
318
120
310
490
320
60
160
>320
>120
160
35 (78)
24 (69)
290 (35)
400 (6)
0.4 (95)
2.1 (66)
75 (69)
163 (81)
208 (47)
3 (60)
17
10
6.0
16
7.5
20
16
13
19
1.8
1.2
0.72
0.44
3.0
4.0
2.9
2.9
2.4
14
8b
16b
25
25
63
32
79
50
2-Me
2-Me, 5-F
2-Cl, 5-Cl
>33
>33
20
183 (35)
341 (37)
11
7.9
a
Values are the mean of P2 determinations.
Tight-binding limit of the assay.
b
stituent also led to reduced cellular potency and AR selectivity
(28–29). In general, carbamate analogs showed a significant reduc-
tion in inhibition of the 2D6 isozyme but only modest improve-
ment in 2C9 inhibition, similar to the sulfonamides. Compound
26 best exemplified the targeted potency, selectivity, and intrinsic
agonism.
In the pyrrolidine amide class, a number of ligands with excel-
lent PR binding affinity and AR selectivity were identified (Table 3).
In general, 2D6 inhibition was less of an issue in this class. Again
consistent with properties of other pyrrolidine derived analogs,
increasing the size of R1 led to reduced agonist potency and effi-
cacy, and increased 2C9 inhibition (30–33). Significant decrease
in 2C9 inhibition was achieved by replacement of the 2-CF3 benzyl
substituent which produced potent PR partial agonists (34–37).
Since P450 inhibition can often be attributed to lipophilicity, an ef-
fort to introduce polarity into these compounds was undertaken.
Since the driving force for making these reduced-basicity ana-
logs was to improve hERG and PK properties, it was gratifying to
see improvement in both areas. Sulfonamide 20, carbamate 26,
and amide 41 all showed reduced blockade of the hERG channel
relative to 3 in a whole cell patch clamp assay (Table 4).12 Addi-
tionally, rat exposure data indicates that the volume of distribution
was reduced by more than threefold for all analogs. All three com-
pounds also showed significant improvement in oral bioavailabil-
ity, while 20 and 26 also showed reduced clearance.
Trp755
Met909
On AF2
Gln725
To this end, incorporation of a hydroxyl group onto the
a-carbon
of the amide moiety (38–41) led to PR partial agonists with further
reduced P450 inhibition at 2C9. Compound 41 was selected for
additional profiling because the intrinsic agonism (37%) was clos-
est to the low end of the target range and was significantly differ-
ent from the other compounds chosen for profiling.
Arg766
An X-ray co-crystal structure of sulfonamide 20 bound to the PR
ligand binding domain was solved to a resolution of 2.0 Å.10 An
overlay of progesterone (1) and compound 20 bound to PR
(Fig. 1) shows that the benzonitrile group of 20 occupies the same
space and interacts with the same residues (Gln725 and Arg766) as
the A-ring carbonyl oxygen of 1. In addition, the pyrrolidine ring of
20 lies perpendicular to the core steroidal backbone of 1 and ex-
tends out towards the AF2 region of the protein. In order to accom-
modate this group, the Met909 and Trp755 are slightly displaced
relative to their position when 1 is bound. The movement of
Met909 is notable as this residue is likely associated with the de-
gree of agonism.11
Figure 1. Superposition of 20 (yellow) & 1 (green) crystal structures. Enzyme
backbone and key backbone residues are shown in the color of the ligand to which
they correspond.

374
L. S. Kallander et al. / Bioorg. Med. Chem. Lett. 20 (2010) 371–374
Table 4
progressed to an in vivo model to measure their ability to reduce
estrogen-stimulated uterine growth and all were efficacious.
Rat pharmacokinetic and hERG channel dataa
Parameter
Compound
Acknowledgments
3b
20b
26c
41b
Dose (iv, po, mg/kg)
CLp, iv (mL/min/kg)
Vdss (L/kg)
Cmax, iv (ng/mL)
t1/2, iv (h)
1.1/2.1
120
39
39
3.6
2.1/4.2
98
12
221
1.6
1.3/0.85
1.2/1.8
150
10
134
1.1
The authors would like to acknowledge David Gray, Margaret
Clackers and Rosemary Sasse for performing the PR and AR binding
assays and Walter Trizna for performing T47D cellular assays.
Additionally, Theresa Roethke, Melanie Nord and Katrina Rivera
ran the rat in vivo studies and P450 in vitro assays. Finally, we
would like to thank Dennis A. Holt for scientific suggestions and
discussions.
59.7
11.7
290
5.0
Oral %F
23
0.78
ꢀ100
ꢀ100
ꢀ100
hERG IC50
(lM)
2.8
4.0
2.2
a
b
c
Values are means of two experiments.
Discrete study.
Cassette study (conducted as a mixture).
References and notes
1. Mangelsdorf, D. J.; Thummel, C.; Beato, M.; Herrlich, P.; Schütz, G.; Umesono,
K.; Blumberg, B.; Kastner, P.; Mark, M.; Chambon, P.; Evans, R. M. Cell 1995, 83,
835.
2. Edgren, R. A.; Sturtevant, F. M. Am. J. Obstet. Gynecol. 1976, 125, 1029.
3. Spitz, I. M. Exp. Rev. Obstet. Gynecol. 2007, 2, 227.
4. Lundeen, S. G.; Zhang, Z.; Zhu, Y.; Carver, J. M.; Winneker, R. C. J. Steroid
Biochem. Mol. Biol. 2001, 78, 137.
5. (a) Mihalyi, A.; Simsa, P.; Mutinda, K. C.; Meuleman, C.; Mwenda, J. M.;
D’Hooghe, T. M. Exp. Opin. Emerg. Drugs 2006, 11, 503; (b) Olive, D. L.; Pritts, E.
A. N. Eng. J. Med. 2001, 345, 266.
6. (a) Kern, J. C.; Terefenko, E. A.; Fensome, A.; Unwalla, R.; Wrobel, J.; Cohen, J.;
Zhu, Y.; Berrodin, T. J.; Yudt, M. R.; Winneker, R. C.; Zhang, Z.; Zhang, P. Bioorg.
Med. Chem. Lett. 2008, 18, 5015; (b) Zhi, L.; Tegley, C. M.; Kallel, E. A.; Marschke,
K. B.; Mais, D. E.; Gottardis, M. M.; Jones, T. K. J. Med. Chem. 1998, 41, 291; (c)
Dols, P. P. M. A.; Folmer, B. J. B.; Hamersma, H.; Kuil, C. W.; Lucas, H.; Ollero, L.;
Rewinkel, J. B. M.; Hermkens, P. H. H. Bioorg. Med. Chem. Lett. 2008, 18, 1461.
7. Thompson, S. K.; Washburn, D. G.; Frazee, J. S.; Madauss, K. P.; Hoang, T. H.;
Lapinski, L.; Grygielko, E. T.; Glace, L. E.; Trizna, W.; Williams, S. P.; Duraiswami,
C.; Bray, J. D.; Laping, N. J. Bioorg. Med. Chem. Lett. 2009, 19, 4777.
8. Gleeson, M. P. J. Med. Chem. 2008, 51, 817.
110
100
90
80
70
60
50
40
30
20
10
0
9. Washburn, D. G.; Hoang, T. H.; Frazee, J. S.; Johnson, L.; Hammond, M.; Manns,
S.; Madauss, K. P.; Williams, S. P.; Duraiswami, C.; Tran, T. B.; Stewart, E. L.;
Grygielko, E. T.; Glace, L. E.; Trizna, W.; Nagilla, R.; Bray, J. D.; Thompson, S. K.
Bioorg. Med. Chem. Lett. 2009, 19, 4664.
10. (a) Williams, S. P.; Sigler, P. B. Nature (London) 1998, 393, 392; (b) Madauss, K.
P.; Deng, S.-J.; Austin, R. J.; Lambert, M. H.; McLay, I.; Pritchard, J.; Short, S. A.;
Stewart, E. L.; Uings, I. J.; Williams, S. P. J. Med. Chem. 2004, 47, 3381; (c)
Compound 20’s PDB code is 3KBA.
11. Madauss, K. P.; Grygielko, E. T.; Deng, S.-J.; Sulpizio, A. C.; Stanley, T. B.; Wu, C.;
Short, S. A.; Thompson, S. K.; Stewart, E. L.; Laping, N. J.; Williams, S. P.; Bray, J.
D. Mol. Endocrinol. 2007, 21, 1066.
None
20
26
41
Compound (10 mg/kg)
Chart 2. Rat Uterus Wet Weight. Line corresponds to reduction for medroxypro-
gesterone acetate (2, 10 mg/kg).
12. Zeng, H.; Lozinskaya, I. M.; Lin, Z.; Willette, R. N.; Brooks, D. P.; Xu, X. J.
Pharmacol. Exp. Ther. 2006, 319, 957.
Previous work has demonstrated that PR partial agonists op-
pose the effects of estrogen-stimulated uterine growth in an ovari-
ectomized rat uterotrophic model.7,9 Three new compounds,
sulfonamide 20, carbamate 26 and amide 41 were tested at a
10 mg/kg dose and the reduction in uterine wet weight relative
to estrogen alone was determined (Chart 2). All three compounds
were efficacious; in fact, compound 41 demonstrated a decrease
in uterine wet weight comparable to the full agonist MPA (2). This
result suggests that compounds with a range of PR agonism levels
could be developed and used to treat endometriosis. Whether such
compounds will achieve an improved side-effect profile compared
to full progestins will likely require clinical evaluation because
there are no validated preclinical models for common PR-related
side-effects such as break-through bleeding.
In summary, pyrrolidine derivatives designed to be less basic by
incorporation of sulfonamide, carbamate or amide substituents
demonstrate reduced hERG channel blockade and improved rat
PK properties. These compounds were found to be potent PR par-
tial agonists with good AR selectivity and reduced activity at two
P450 isozymes. Three high quality compounds 20, 26, and 41 were
13. Example preparation (compound 23); step b (i): TFA (9.2 mL) was added to a
solution of 1-dimethylethyl (3S)-3-[(3-chloro-4-cyanophenyl)amino]-1-
pyrrolidinecarboxylate (6 g, 18.7 mmol) in CH2Cl2 (40 mL) and the mixture
was stirred for 5 h. Toluene (20 mL) was added and the reaction mixture was
concentrated. Saturated aqueous NaHCO3 (50 mL) was added to the residue
and the mixture was extracted with EtOAc (5 Â 100 mL). The organic extracts
were dried over MgSO4 and concentrated to yield the product as a red–brown
solid (6.8 g). MS(ES) m/e 220.0 [M+H]+. (ii): Methyl chloroformate (0.78 g,
8.1 mmol)
was
added
to
a
solution
of
2-chloro-4-[(3S)-3-
pyrrolidinylamino]benzonitrile (1.88 g, 8.48 mmol) and Et3N (2 mL,
12.7 mmol) in CH2Cl2 (50 mL) at 0 °C. After stirring at 0 °C for 3 h, saturated
aqueous NaHCO3 was added and the layers were separated. The organic layer
was dried over Na2SO4 and concentrated. The residue was purified via flash
chromatography to give the product as a white foam (2.12 g, 89%). MS(ES) m/e
280.2 [M+H]+. Step c: NaH (60% in mineral oil, 0.043 g, 1.08 mmol) was added
to
a
solution of methyl (3S)-3-[(3-chloro-4-cyanophenyl)amino]-1-
pyrrolidinecarboxylate (0.1 g, 0.36 mmol) in DMF (3 mL) at 0 °C. After
stirring for 5 min, 1-(bromomethyl)-2-fluorobenzene (0.072 g, 0.38 mmol)
was added and the mixture was stirred for 2 h. Saturated aqueous NaHCO3
was added and the reaction mixture was extracted with EtOAc (2Â). The
organic layers were dried over Na2SO4 and concentrated. The residue was
purified via flash chromatography to give the product as a white foam (0.072 g,
52%).1H NMR (CDCl3): d 7.44 (d, J = 9.2 Hz, 1H), 7.30 (m, 1H), 7.12 (m, 2H), 7.01
(m, 1H), 6.80 (d, J = 2.4 Hz, 1H), 6.60 (dd, J = 8.8, 2.4 Hz, 1H), 4.60 (m, 3H), 3.83
(m, 1H), 3.72 (s, 3H), 3.60 (m, 1H), 3.45 (m, 1H), 3.31 (m, 1H), 2.23 (m, 1H), 2.05
(m, 1H). MS(ES) m/e 388.6 [M+H]+.