Allyl and propargyl substituted penam sulfones as versatile intermediates toward the syntheses of new β-lactamase inhbitors

Article abstract of DOI:10.1016/S0960-894X(01)00148-2

Several alkenyl derivatives were prepared using allyl penam sulfone as the key intermediate. Isomers of these derivatives having β configuration at C-6 showed potent acitivity against CcrA enzyme. A new method was developed to prepare propargyl penam sulfone. The majority of the triazoles prepared by this route exhibited good activity against all three representative enzymes used for the inhibition assay.

Full text of DOI:10.1016/S0960-894X(01)00148-2

Bioorganic & Medicinal Chemistry Letters 11 (2001) 997–1000
Allyl and Propargyl Substituted Penam Sulfones as
Versatile Intermediates Toward the Syntheses of New
ꢀ-Lactamase Inhibitors
Vincent P. Sandanayaka,* Gregg B. Feigelson, Amar S. Prashad, Youjun Yang
and Peter J. Petersen
Chemical Sciences and Infectious Diseases, Wyeth-Ayerst Research, 401 N. Middletown Road, Pearl River, NY 10965, USA
Received 8 December 2000; accepted 13 February 2001
Abstract—Several alkenyl derivatives were prepared using allyl penam sulfone as the key intermediate. Isomers of these derivatives
having b configuration at C-6 showed potent acitivity against CcrA enzyme. A new method was developed to prepare propargyl
penam sulfone. The majority of the triazoles prepared by this route exhibited good activity against all three representative enzymes
used for the inhibition assay. # 2001 Elsevier Science Ltd. All rights reserved.
Resistance to antibiotics is a major concern in the
infectious disease area.1 Today, bacterial resistance has
developed against vancomycin, which is considered an
antibiotic of last resort.2 b-Lactamases are the primary
cause for resistance against most widely used b-lactam
antibiotics.3 These enzymes efficiently hydrolyze the
amide bond of the b-lactam ring to give products that
are devoid of antibacterial activity. Therefore, the
administration of a b-lactamase inhibitor along with an
antibiotic represents an important strategy for combating
the resistance against these drugs.4
Therefore, we decided to introduce groups at C-6, that
might not only increase the binding but also act as
reactive moieties toward amino acid residues in the
active site. The presumption is that after initial acyl-
ation of the enzyme, the resulting acyl–enzyme inter-
mediate can further react with other catalytic site
residues leading to an enzyme-bound species.9 This
would result in complete inactivation of the enzyme. In
the past few years, several groups10 have used a similar
strategy in which substituents with such implications
were introduced at various positions of sulbactam to
enhance the broad-spectrum activity. Herein, we report
some of our efforts to find an inhibitor that is effective
for the most prevalent classes of enzymes.
Sulbactam (1)5 and tazobactam (2)6 are two currently
marketed b-lactamase inhibitors, which are used with an
antibiotic in this context (Fig. 1). These inhibitors have
potent activity against class A enzymes, but have little
or no activity against class B and C enzymes.7 There-
fore, an inhibitor with the broad-spectrum activity is
important to fight effectively against emerging resistant
pathogenic bacteria. It is known that the reactions of
the above inhibitors with b-lactamases involve branched
pathways with many putative intermediates, which have
been identified.8 Also, molecular modeling studies of the
above inhibitors with several b-lactamases revealed that
there is ample space in the active site to accommodate
sizable groups at the C-6 position of these molecules.
We envisioned that the penam sulfones containing allyl
and propargyl groups at the 6-position would be ideal
intermediates that can be easily converted into various
targets by functional group manipulations. Allyl sub-
stituted penam sulfones and their derivatives were
prepared according to Schemes 1–3.
*Corresponding author. Fax: +1-845-732-5561; e-mail: sandanv@
war.wyeth.com
Figure 1. Penam sulfones currently marketed as b-lactamase inhibitors.
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00148-2

998
V. P. Sandanayaka et al. / Bioorg. Med. Chem. Lett. 11 (2001) 997–1000
Preparation of compound 4ꢀ has been reported pre-
viously, starting with commercially available (+)-6-
aminopenicillanic acid (Scheme 1).11 The corresponding
a-isomer 4ꢁ was prepared according to Scheme 2.
Dibromide 5 was converted into monobromide 6 in
86% yield using 1 equivof MeMgBr followed by the
aqueous quench. Radical-mediated allyl addition with
allyltributyltin gave compound 7 as the sole diastereo-
isomer in 79% yield.12 Under these conditions, the tin
reagent approaches from the less-hindered a-face to
provide the observed stereochemical outcome. Ozono-
lysis of the allylic compound 7 at ꢀ78 ꢁC led to the
desired aldehyde 4ꢁ in 69% yield.13
the target compound 21. Overall yield of the final pro-
ducts by this method was relatively low mainly due to a
poor yield in the deoxygenation step.
To overcome the above deficiency, an alternative syn-
thetic route was devised (Scheme 4). First, dibromide 5
was converted to the corresponding monobromide 6
with EtMgBr at ꢀ78 ꢁC as shown in Scheme 2. Initial
attempts to obtain compound 23 directly by employing
propargyltributyltin under radical conditions provided
only the isomeric allenyl product. This observation
suggested that if the allenyltributyltin16 reagent is used
under similar conditions, the desired propargyl group
could be introduced at the 6-position of the b-lactam
ring. Indeed, treatment of monobromide 6 with allenyl-
tributyltin 22 in the presence of AIBN produced the
propargyl substituted compound 2317 in 50% yield. No
allenyl substituted product was detected in the crude
reaction mixture. The product crossover in the above
Aldehydes 4ꢁ and 4ꢀ were treated with various Wittig
reagents to provide the corresponding a,b-unsaturated
carbonyl compounds and nitriles 8–12 in 60–78% yields
(Scheme 3). With the exception of the nitrile, all cases
gave (E)-isomers as expected. Wittig reaction with
nitrile ylide gave a 2:1 mixture of (E)- and (Z)-isomers,
respectively. Conversion of the diphenylmethyl esters to
the corresponding acids was achieved by using TFA in
trifluoroethanol.
0
reactions is consistent with an SH process, which pre-
vails over the direct homolytic substitution (SH)
mechanism.18 Compound 23 also reacted efficiently with
various azides to give the corresponding cycloadducts in
45–70% yields. This method was one step shorter and
provided better overall yields compared to the previous
method.
Propargyl substituted penam sulfones and their deriva-
tives were prepared according to Scheme 4. Dibromide
5 was treated with EtMgBr at ꢀ78 ꢁC, followed by pro-
pargyl aldehyde to give 18 as a 1:1 mixture of diaster-
eoisomers in 61% yield. Debromination with Bu3SnH/
AIBN in toluene at 80 ꢁC provided the desired 6b-iso-
mer 19 in 71% yield. This propargyl derivative 19
underwent smooth dipolar cycloaddition reactions with
a series of substituted azides to give the corresponding
triazole 20, as the major regioisomer, in 40–50%
yields.14 Deoxygenation of 20 was effected by forming
the thiocarbonate with phenylchlorothiono formate fol-
lowed by reduction with Bu3SnH to give the corre-
sponding deoxygenated compound in 18% yield as a
single diastereoisomer.15 Deprotection of the benzhy-
dryl ester by catalytic hydrogenolysis (H2, Pd/C) gave
These compounds19 were evaluated for their inhibitory
activity against representative enzymes from three dif-
ferent enzyme classes.7 While TEM-1 and AmpC are
serine-based enzymes, CcrA is a metalloenzyme in
which zinc metal plays a key role in initiating catalytic
activity. Mechanistically, CcrA follows a unique inhibi-
tion pathway as compared to the serine-based
enzymes.20 In addition, metalloenzymes are currently
among the quickly spreading b-lactamases especially in
Japan and Europe. Our goal was to find an inhibitor
that is effective for all three enzymes. Inhibitory activ-
ities of these compounds were compared with reference
compounds, sulbactam and tazobactam.21 Overall, the
alkenyl series did not show any significant improvement
in activity against either TEM-1 or AmpC (Table 1).
However, the b-isomers in this series showed a dramatic
increase in activity against CcrA. Interestingly, the a-
isomers are significantly less active indicating the
importance of stereochemistry at the 6-position of these
compounds. This is evident by comparing the activities
of 14ꢁ and 14ꢀ. The fact that such stereochemical
Scheme 1. Five-step conversion of commercially available 6-(+)-ami-
nopenicillanic acid to the b-isomer of the aldehyde.
Scheme 2. Reagents and conditions: (a) EtMgBr, ꢀ78 ꢁC, quench
Scheme 3. Reagents and conditions: (a) RCH¼PPh3, benzene; (b)
(86%); (b) allyltributyltin/AIBN (79%); (c) O3, MeOH, Me2S (69%).
TFA/CF3CH2OH.

V. P. Sandanayaka et al. / Bioorg. Med. Chem. Lett. 11 (2001) 997–1000
999
differences have not led to any appreciable change in
activity for TEM-1 and AmpC further explains the
unique behavior of metalloenzymes.
Further in vitro evaluations of 21d against b-lactamase-
producing organisms indicated that this compound was
able to restore the activity of piperacillin to a certain
degree, but was less active compared to tazobactam
(Table 3). The MIC values of piperacillin in combina-
tion with 21d were improved to 64 mg/mL and 32 mg/mL
against AmpC producing E. coli (GC2894) and TEM-1
producing E. coli (GC2206), respectively.
Table 2 shows the data for the triazole series. The
majority of compounds in this series showed improved
activity, especially against CcrA and AmpC. Introduc-
tion of polar groups to the phenyl substituent in the
triazole ring improved TEM-1 activity about 2-fold
relative to sulbactam as exemplified by 21c and 21d.
Further extension of these groups away from the tri-
azole ring via an alkyl chain, as for compound 21f,
decreased the activity against all enzymes. The aniline
substituted triazole derivative 21d22 showed the best
activity profile of all and justified further efforts to find
a broad-spectrum b-lactamase inhibitor along these
lines.
In summary, a general method was devised to synthesize
various alkenyl penam sulfone derivatives. A new
method was developed to prepare the propargyl sub-
stituted penam sulfone, which was transformed into
various triazoles via a cycloaddition protocol. The
compounds prepared were evaluated as potential b-lac-
tamase inhibitors. Some of these compounds showed
good acitivity against several b-lactamases. Thus, we
have demonstrated that allyl and propargyl penam sul-
fones are valuble intermediates that can be converted
into various target molecules having b-lactamase
inhibitory activity.
Table 1. In vitro data for selected compounds of alkenyl series
IC50 (mM)b
Compda
R
TEM-1
(class A)
CcrA
(class B)
AmpC
(class C)
Table 2. In vitro data for selected compounds of triazole series
IC50 (mM)b
13ꢁ
14ꢁ
15ꢁ
16ꢁ
13ꢀ
14ꢀ
17ꢀ
1
COMe
CO2Me
CON(OMe)Me
CHO
8.4
8.6
12.0
12.0
6.5
2.8
24.0
1.4
71.0
>100
>100
>100
3.0
1.4
3.1
>100
100.0
29.0
17.0
17.0
40.0
64.0
>100
66.0
Compda
R0
TEM-1
(class A)
CcrA
(class B)
AmpC
(class C)
COMe
CO2Me
CN
21a
21bc
21cc
21dc
21e
1.0
1.6
>100
4.1
3.4
3.0
2.3
1.4
5.2
6.6
aAll compounds shown are (E)-isomers.
bDetermined graphically from six different concentrations of the inhi-
bitor.
0.7
2.1
0.6
1.8
7.9
22.0
24.0
21f
15.0
1
2
1.4
0.06
>100
>100
66.0
48.0
aAll compounds shown are single regioisomers.
bDetermined graphically from six different concentrations of the
inhibitor.
cSubstitution is at the para-position.
Table 3. In vitro data for 21d and tazobactam
MIC (mg/mL)
Compd
E. coli (GC2894)b
E. coli (GC2206)c
21da
Tazobactama
Piperacillin
64
16
>64
32
2
>64
Scheme 4. Reagents and conditions: (a) EtMgBr, propargyl aldehyde,
THF, ꢀ78 ꢁC to rt (61%); (b) Bu3SnH, AIBN, toluene, 80 ꢁC (71%);
(c) R0N3, toluene, reflux; (d) (i) phenylchlorothiono formate, pyr.; (ii)
Bu3SnH, AIBN, toluene (18% for 2 steps); (iii) H2, Pd–C, 3 h (65%);
(e) (i) EtMgBr, ꢀ78 ꢁC; (ii) 22, AIBN, toluene, 16 h (50%); (f) (i)
R0N3, toluene (45–70%); (ii) H2, Pd–C, 3 h (65%).
aPiperacillin to inhibitor ratio is 1:1.
bAmpC (class C).
cTEM-1 (class A).

1000
V. P. Sandanayaka et al. / Bioorg. Med. Chem. Lett. 11 (2001) 997–1000
P.; Bush, K.; Steinberg, D. US Patent 5 488 106, 1996; Chem.
Acknowledgements
Abstr. 1996, 124, 316870. (e) Buynak, J. D.; Borate, H. B.;
Lamb, G. W.; Khasnis, D. D.; Husting, Ch.; Isom, H.; Sir-
iwardane, U. J. Org. Chem. 1993, 58, 1325. (f) Adam, S.;
Then, R.; Angehm, P. J. Antibiot. 1993, 46, 641. (g) Chen, Y.
L. US Patent 5 015 473, 1991; Chem. Abstr. 1990, 112,
157958.
The authors thank Dr. R. Nilakantan for molecular
modeling studies, and Drs. S. Lang, T. Mansour, and
D. Shlaes for support of this work.
11. Sandanayaka, V. P.; Yang, Y. Org. Lett. 2000, 2, 3087.
12. Hanessian, S.; Alpegiani, M. Tetrahedron 1989, 45, 941.
13. In the 1H NMR spectrum, 6ꢁ-hydrogen shows a coupling
constant of 2.0 Hz with the adjacent bridgehead hydrogen
whereas 6ꢀ-hydrogen shows a coupling constant of 4.8 Hz.
This is consistent with all the other derivatives as well. 4ꢁ: IR
References and Notes
1. (a) Tenover, F. C. JAMA 1996, 275, 300. (b) Nikaido, H.
Antimicrob. Agents Chemother. 1989, 33, 1831. (c) Davies, J.
Science 1994, 264, 360. (d) Neu, H. Science 1992, 257, 1064.
2. Nicolaou, K. C.; Hughes, R.; Cho, S. Y.; Winssinger, N.;
Smethurst, C.; Labischinski, H.; Endermann, R. Angew.
Chem., Int. Ed. Engl. 2000, 39, 3823.
3. (a) Marcaretti, O. A.; Danelon, G. O.; Setti, E. L.;
Laborde, M.; Mata, E. G. Curr. Pharm. Des. 1999, 5, 939. (b)
Yang, Y.; Rasmussen, B. A.; Shlaes, D. M. Pharmacol. Ther-
apeut. 1999, 83, 141. (c) Payne, D. J.; Du, W.; Bateson, J. H.
Exp. Opin. Invest. Drugs 2000, 9, 247. (d) Maiti, S. N.; Phillips,
O. A.; Micetich, R. G.; Livermore, D. M. Curr. Med. Chem.
1998, 5, 441.
2980, 1800, 1755, 1323, 1179, 1118, 700 cmꢀ1 1H NMR
;
(400 MHz, CDCl3) d 1.11 (s, 3H), 1.58 (s, 3H), 3.04 (dd, 1H,
J1=9.6 Hz, J2=19.2 Hz), 3.21 (dd, 1H, J1=4.8 Hz,
J2=19.2 Hz), 4.03 (m, 1H), 4.41 (d, 1H, J=2.0 Hz), 4.45 (s,
1H), 6.97 (s, 1H), 7.35 (m, 10H); MS (ES): 442.1 (M+H).
14. (a) Lwowski, W In 1,3-Dipolar Cycloaddition Chemistry;
Padwa, A., Ed., Wiley-Interscience: New York, 1984; Vol. 1,
pp 291–392. (b) Related triazole containing penem derivative
BRL 42715 is known, see: Farmer, T. H.; Page, J. W. J.;
Payne, D. J.; Knowles, D. J. C. Biochem. J. 1994, 303, 825.
15. Eby, P.; Cummings, M. D. Heterocycles 1996, 42, 653.
16. Ueno, Y.; Okawara, M. J. Am. Chem. Soc. 1979, 101,
1893.
4. Knowles, J. R. Acc. Chem. Res. 1993, 31, 9.
5. (a) English, A. R.; Retsema, J. A.; Girard, A. E.; Lynch, J. E.;
Barth, W. E. Antimicrob. Agents Chemother. 1978, 14, 414. (b)
Brenner, D. J.; Knowles, J. R. Biochemistry 1981, 20, 3681.
6. (a) Livermore, D. M. J. Antimicrob. Chemother. 1993, 31, 9.
(b) Bush, K.; Macalintal, C.; Rasmussen, B. A.; Lee, V. J.;
Yang, Y. Antimicrob. Agents Chemother. 1993, 37, 851.
7. (a) Bush, K.; Jacoby, G. A.; Medeiros, A. A. Antimicrob.
Agents Chemother. 1995, 39, 1211. (b) Ambler, R. P. Philos.
Trans. R. Soc. London, Ser. B 1980, 289, 321.
8. (a) Brenner, D. J.; Knowles, J. R. Biochemistry 1981, 20,
3680. (b) Yang, Y.; Janota, K.; Tabei, K.; Huang, N.; Siegel,
M. M.; Lin, Y.-I.; Rasmussen, B. A.; Shlaes, D. M. J. Biol.
Chem. 2000, 275, 26674 and references cited therein.
9. For a comprehensive account on mechanism, see: Silver-
man, R. B. In Mechanism-Based Enzyme Inactivation:Chem-
istry and Enzymology; CRC: Boca Raton, 1988.
10. (a) Czajkowski, D. P.; Reddy, A. V. N.; Setti, E. L.; Phil-
lips, O. A.; Micetich, R. G.; Kunugita, C.; Maiti, S. N. Bioorg.
Med. Chem. Lett. 1997, 7, 11. (b) Tzouvalekis, L. S.; Gazouli,
M.; Prinarakis, E. E.; Tzelepi, E.; Legakis, N. I. Antimicrob.
Agents Chemother. 1997, 41, 475. (c) Bitha, P.; Li, Z.; Fran-
cisco, G. D.; Yang, Y.; Petersen, P. J.; Lenoy, E.; Lin, Y.-I.
Bioorg. Med. Chem. Lett. 1999, 9, 997. (d) Ziegler, C.; Fabio,
17. 23: IR 1760, 1780, 2850, 2950, 3300 cmꢀ1 1H NMR
;
(300 MHz, CDCl3) d 1.14 (s, 3H), 1.57 (s, 3H), 2.11 (t, 1H,
J=2.5 Hz), 2.78 (dd, 1H, J1=7.0 Hz, J2=2.6 Hz), 2.82 (dd,
1H, J1=7.0 Hz, J2=2.6 Hz), 3.86 (m, 1H), 4.49 (s, 1H), 4.54
(d, 1H, J=1.8 Hz), 6.95 (s, 1H), 7.35 (m, 10H); MS (ES): 437.1
(M+H).
18. Migita, T.; Kosugi, M.; Takayama, K.; Nakagawa, Y.
Tetrahedron 1973, 29, 51.
19. All new compounds were fully characterized (NMR, IR,
HRMS–MS).
20. Wang, Z.; Fast, W.; Valentine, A. M.; Benkovic, S. J.
Curr. Opin. Chem. Biol. 1999, 3, 614.
21. For details about the assay and experimental procedure,
see: Bush, K.; Macalintal, C.; Rasmussen, B. A.; Lee, V. J.;
Yang, Y. Antimicrob. Agents Chemother. 1993, 27, 851.
22. 21d: IR 3400, 2920, 1775, 1630, 1530, 1380, 1320,
1
1120 cmꢀ1; H NMR (300 MHz, D2O) d 1.41 (s, 3H), 1.56 (s,
3H), 3.36 (dd, 1H, J1=7.9 Hz, J2=15.5 Hz), 3.62 (dd, 1H,
J1=10.7 Hz, J2=15.5 Hz), 4.32 (s, 1H), 4.58 (m, 1H), 5.05 (d,
1H, J=4.7 Hz), 6.96 (d, 2H, J=8.7 Hz), 8.21 (s, 1H); MS (ES):
404.2 (MꢀH).