Synthesis of bicyclic β-lactamase inhibitor relabactam derivatives from a relabactam intermediate

Article abstract of DOI:10.1016/j.tetlet.2017.06.018

We have developed highly efficient chemistry to prepare two key intermediates from a Relabactam intermediate available from the process chemistry. The new intermediates enabled us to quickly synthesize other Ralebactam derivatives such as compound 1.

Full text of DOI:10.1016/j.tetlet.2017.06.018

Accepted Manuscript
Synthesis of Bicyclic β-Lactamase Inhibitor Relabactam Derivatives From a
Relabactam Intermediate
Shu-Wei Yang, Xin Linghu, Elizabeth Smith, Jianping Pan, Victoria Sprague,
Jing Su
PII:
S0040-4039(17)30738-4
DOI:
http://dx.doi.org/10.1016/j.tetlet.2017.06.018
TETL 49010
Reference:
Tetrahedron Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
12 May 2017
2 June 2017
6 June 2017
Please cite this article as: Yang, S-W., Linghu, X., Smith, E., Pan, J., Sprague, V., Su, J., Synthesis of Bicyclic β-
Lactamase Inhibitor Relabactam Derivatives From a Relabactam Intermediate, Tetrahedron Letters (2017), doi:
http://dx.doi.org/10.1016/j.tetlet.2017.06.018
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Synthesis of Bicyclic β-Lactamase Inhibitor Relabactam Derivatives
From a Relabactam Intermediate
‡
§
*
Shu-Wei Yang,*, Xin Linghu, Elizabeth Smith,† Jianping Pan,† Victoria Sprague, Jing Su†,
†
†Discovery Chemistry, MRL, Merck & Co., Inc., Kenilworth, NJ, USA, 07033
‡Process Discovery Chemistry, MRL, Merck & Co., Inc., Kenilworth, NJ, USA, 07033
§Biocatalysis, MRL, Merck & Co., Inc., Rahway, NJ, USA, 07065
Supporting Information Placeholder
Graphical Abstract:
ABSTRACT: We have developed highly efficient chemistry to prepare two key intermediates from a Relabactam intermediate
available from the process chemistry. The new intermediates enabled us to quickly synthesize other Ralebactam derivatives such
as compound 1.
The rise of multidrug resistant Gram-negative bacteria has
posed a tremendous threat to human health.1 One of the main
mechanisms of the β-lactam antibiotic drug resistance is the over-
expression of β-lactamases by the resistant strains, resulting in the
degradation of β-lactam antibiotics that inhibit the bacterial cell
ate from the process synthesis of MK-7655.11,12 In this commu-
nication, we describe the highly efficient chemistry we developed
to convert the process intermediate 2 to intermediate 5 and the
subsequent synthesis of 1.
wall synthesis.2-4
β-Lactamase inhibitors (BLIs) such as
clavulanic acid inhibit β-lactamases and restore the susceptibility
of β-lactams against the resistant strains.5-6 The combination of
β-lactam antibiotics and BLIs has demonstrated restoration of the
original efficacy spectrum of β-lactams.7-8 One of the successful
examples in the clinic is Augmentin, a combination of amoxicillin
and clavulanic acid (Figure 1), which is widely prescribed for
various bacterial infections. However there are hundreds of
known β-lactamases, categorized into four classes (A, B, C, and
D).3 The clinically used BLIs only inhibit class A β-lactamases
which deactivate penicillins and early cephalosporins. There is an
unmet medical need for the treatment of serious bacterial infec-
tions caused by the resistant Gram-negative bacterial strains that
produce other classes of β-lactamases, for example, AmpC by
pseudomonas aeruginosa.9 Relabactam (MK-7655 is a potent
BLI currently in Phase III clinical trials in combination with
imipenem for the treatment of severe bacterial infection.
Relabactam displays inhibition against both classes A and C β-
lactamases (e.g. KPC-2 and AmpC).10 It effectively restored
imipenem’s activity against imipenem-resistant Klebsiella and
pseudomonasa strains.10 A similar compound avibactam com-
bined with ceftazidime was approved by FDA in 2015.
Figure 1. Example of BL/BLI used in clinic
Figure 3 illustrates both the chemical and enzymatic cleavage
of the amide bond in compound 2, which was available in kg
quantity. The chemical method required the protection of the
amide NH with the Boc group followed by the hydrolysis reac-
tion. Thus intermediate 2 was protected with Boc using di-t-butyl
dicarbonate and DMAP in THF. The yield for intermediate 3 was
only 26% after optimization, and the major product was di-Boc
protected 3a in > 50% yield (Figure 4). While the tris-Boc pro-
tected 3 could be cleanly converted to intermediate 4 using the
LiOH/H2O2 method, the conversion of the bis-Boc protected 3a to
its corresponding acid was much poorer. Our attention was turned
to the enzymatic cleavage method and a biocatalytic screen for
various enzymes was performed directly on compound 2. The
screening experiments quickly identified Amano protease P6 as
the enzyme of choice for this amide bond cleavage step.
In our BLI program, there was a need to prepare gram quanti-
ties of compound 1, a close analog to MK-7655, for various pur-
poses (Figure 2). A default solution was to use the original
lengthy synthetic route described by Blizzard et al.10 However, a
substantial quantity of compound 2 was available as an intermedi-

Table 1. Optimization of biocatalytic conversion of 2 to 5
Conversion Rates
Incubation timea
Conc. of 2
(g/L)
Figure 2. The structures of relabactam (MK-7655) and 1
20 h
29 h
3 days
ND
10
20
30
40
50
>98%
ND
90-94%
88-95%
90-91%
46-93%
88-97%
84-97%
90-91%
81-93%
>99%
>99%
>99%
94%-99%b
aThe incubation was carried out at three temperatures (30 °C,
35 °C, and 40 °C) for each concentration. The ranges of the con-
version rates were given based on the conversion at these three
b
temperatures. For 50 g/L concentration, the conversion rate was
reached to 99% at 40 °C. The conversion was monitored by
LC/MS.
To finish the synthesis of 1, a short route was proposed using
direct amide coupling without protecting the piperidine nitrogen
(Figure 5). The reaction was undertaken using the BOP
((Benzotriazol-1-yloxy)tris(dimethylamino)phophonium
Figure 3. Chemical and enzymatic cleavage of the amide bond
hexafluorophosphate) coupling conditions with amine 6 to afford
intermediate 7. Since compound 7 was highly water soluble, it
was not suitable for normal phase purification. Thus, the crude 7
was directly treated with triphosgene, and the key intermediate 8
was obtained in 38% yield after purification in two steps (Exp. 2
in Table 2). A self-coupling byproduct 1114was identified as the
major byproduct in the coupling reaction (Figure 6). Our strategy
to improve the yield of the desired product 7 was to allow the
hetero amide coupling to compete with the self-coupling early to
improve the reaction yield. By mixing the acid 5 and BOP fol-
lowed by the immediate addition of amine 6 and the Hunig’s base,
the self-coupling was minimized and the overall yield of com-
pound 8 in 2 steps was significantly improved to 76% (Table 2).
The benzyl protecting group of 8 was removed using Pd/C under
hydrogen atmosphere (1 atm, hydrogen balloon) in methanol to
obtain 9 in 93% yield. The hydroxyl group of 9 was converted to
sulfate using sulfur trioxide pyridine complex. The sulfate 10 was
converted to the tert-butyl ammonium salt to enhance its stability
and to reduce its polarity for easy handling. Finally the Boc group
was removed under TFA condition to obtain the desire product 1.
The pure compound 1 was obtained using the crystallization
method or purification on a prep HPLC column without any acid
or base modifier (see Supplemental material). In general this
route was very effective and could avoid complicated protecting
and de-protecting steps.
Figure 4. Major product of Boc-protection step
To further optimize the enzymatic cleavage condition, an addi-
tional screen with varying DMSO concentrations in the buffer
(10% to 30%) and the temperature (30 to 40 °C) was performed.
Only the condition with 10% DMSO in the buffer led to >97%
conversion at 30 °C and 40 °C. Another screen varying the con-
centrations of substrate 2 (10 g/L, 20 g/L, 30 g/L, 40 g/L, and 50
g/L) and the temperature (30 to 40 °C ) with fixed amount of
Amano protease P6 (conc. ~10 mg/mL) was also performed to
optimize the highest possible loading of 2. The results are dis-
played in Table 1. No reactions reached completion at 20 or 29 h,
except for the 10 g/L group, indicating longer reaction time was
required for full conversion in the high concentration groups (20
g/L-50 g/L). Although temperature did not play a significant
role, , for the high loading group (50 g/L), the conversion rate was
higher at 35 °C (98.5%) and 40 °C (99%) thanthat at 30 °C (94%).
All the tested conditions reached >98.5% conversion after 3 days
at 35 °C and 40 °C. Therefore a large scale reaction (183 g of
compound 2) was performed at 50 g/L substrate concentration
under 35-40 °C for three days. The desired product 5 was directly
obtained from crystallization from the reaction solution (42 g,
43% yield). The rest of the product in the mother solution was
not suitable for extraction due to the high polarity of 5. Instead,
the solution underwent direct Boc protection on the piperidine
nitrogen of 5 (details described below) and delivered 38 g of 12
(Figure 7). Through such 2-step synthesis, >95% conversion was
achieved.
2

ethyl-hexanoate and Pd(dppf)Cl2 gave rise to the final compound
16 in > 90% yield in two steps. No flash chromatography was
required in this 6-step synthesis (Figure 7).
Table 2. Correlation of wait time versus yields of 8
Wait time a
10 min
5 min
Yield of 8 b
14%
Exp 1
Exp 2
Exp 3
38%
0 min
76%
a Wait time for amine 6 and Hunig’s base addition after BOP
addition. b Two steps from compound 5.
All reactions were
a
conducted at rt with 1.2 eq. BOP, 1.1 eq. amine 6 and 3 eq.
Hunig’s base in DMF. The yields described here are isolated
yields after purification.
Figure 5. Synthetic scheme of 1 from intermediate 5
Figure 6. The structure of the byproduct, symmetric dimer 11
While intermediate 5 served as an excellent intermediate to
make new relabactam analogs for further structure-activity
relationship (SAR) studies, we were also interested in a more
advanced intermediate 16 for some practical reasons. After
failed attempts on direct cyclization from 5 to bicyclic 16, we
adopted a uneventful synthetic route shown in Figure 7. Tak-
ing the crude mother liquid of 5 directly from the enzymatic
amide cleavage step in Figure 3, the pH was adjusted to 10-11
using saturated sodium carbonate followed by the addition of
Boc2O.15 The overnight reaction afforded compound 12 after
the EtOAc extraction. Esterification with allyl alcohol in
Figure 7. Conversion of intermediate 5 to bicyclic intermedi-
ate 16
o
DMF at 60 C for one hour provided the clean compound 13
which was immediately treated with
4 M HCl in
dioxane/isopropanol. The desired intermediate 14 was ob-
tained by crystallization without any column purification (60-
70% yield in 3 steps from 5). The excellent yield of these
three steps further demonstrated the high yield of the enzymat-
ic amide bond cleavage step was achivable (estimated as
>95% yield, Figure 3).
With the cyclized product 16 in hand, Compound 1 could be
synthesized as shown in Figure 8. This route also allowed us to
completely avoid the formation of byproduct 11, enabling easy
access to new relabactam analogs with modification on the
piperdine ring.
To complete the synthesis of 16, the HCl salt of compound 14
was first converted to the free base followed by the intramolecular
cyclization with triphosgene. The optimal condition used 0.4
equivalent of triphosgene at -30 °C with pretreated Hunig’s Base
(2.5 eq.) in acetonitrile, followed by stirring at ~40-45 °C for one
hour. Additional triphosgene could cause the formation of other
unidentified byproducts. Removing the allyl group using sodium
3

(10) Blizzard, T.A.; Chen, H.; Kim, S.; Wu, J.; Bodner, R.; Gude, C.;
Imbriglio, J.; Young, K.; Park, Y. W.; Ogawa, A.; Raghoobar, S.;
Hairston, N.; Painter, R. E.; Wisniewski, D.; Scapin, G.; Fitzger-
ald, P.; Sharma, N.; Lu, J.; Ha, S.; Hermes, J.; Hammond, M. L.
Bioorg. Med. Chem. Lett. 2014, 24, 780.
(11) Miller, S. P.; Zhong, Y.-L.; Liu, Z.; Simeone, M.; Yasuda, N.;
Limanto, J.; Chen, Z.; Lynch, J.; Capodanno, V. Org. Lett. 2014,
16, 174.
(12) Mangion, I. K.; Ruck, R. T.; Rivera, N.; Huffman, M. A.;
Shevlin, M.; Org. Lett. 2011, 13, 5480.
(13) The biological data of 1 will be published elsewhere.
(14) The structure of the dimer by-product 11 was confirmed by NMR
techniques, including COSY, HSQC, HMBC, and NOE.
(15) The Boc protection could not be achieved at pH ~8. The pH of the
solution was adjusted to ~10-11 to achieve the Boc addition.
Blizzard, T. A.; Chen, H.; Gude, C.; Hermes, J. D.; Imbriglio, J.
E.; Kim, S.; Wu, J. Y.; Ha, S.; Mortko, C. J.; Mangion, I.; Rivera,
N.; Ruck, R. T.; Shevlin, M. PTC WO 2009091856 A3, 2009.
Figure 8. Synthetic route of new relabactam derivatives from
intermediate 16
In conclusion we have developed highly efficient chemistry
for the synthesis of intermediates 5 and 16 from compound 2.
The novel chemistry empowered us to synthesize and scaled
up an important analog 1 of relabactam in an effective manner.
The key intermediates 5 and 16 also greatly advanced our
SAR studies on relabactam to assist future development of its
analogs for treatment of severe Gram negative bacterial infec-
tions.
ASSOCIATED CONTENT
Supporting Information
Experimental procedures, compound characterization, and copies
1
of H and 13C NMR spectra. This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Authors
*E-mail: jing.su2@merck.com, swyang101@yahoo.com
ACKNOWLEDGMENT
The authors wish to acknowledge the following individuals from
Merck & Co., Inc., Kenilworth, NJ, USA, for their assistance and
support: Drs. James Tata, Timothy Blizzard, Brian McKittrick and
Rebecca Ruck; Dr. Andy Evans for structure confirmation of 11;
Ms. Helen Chen for her initial effort on compound 1, and our
Process Chemistry Department for providing intermediates.
REFERENCES
(1) Hersh, A.; Newland, J.; Beekmann, S.; Polgreen, P.; Gilbert, D.
Clin. Infect. Dis. 2012, 54, 1677.
(2) Livermore, D. M. Clin. Microbiol. Rev. 1995, 8, 557.
(3) Bush, K.; Jacoby, G. A. Antimicrob. Agents Chemother. 2010, 54,
969.
(4) Davies, J. Science, 1994, 264, 375.
(5) Drawz, S. M.; Bonomo, R. A. Clin. Microbiol. Rev. 2010, 23,
160.
(6) Bebrone, C.; Lassaux, P.; Vercheval, L.; Sohier, J.-S.; Jehaes, A.;
Sauvage, E.; Galleni, M. Drugs 2010, 70, 651.
(7) Buynak, J. D. Biochem. Pharmacol. 2006, 71, 930.
(8) Perez-Llarena, F. J.; Bou, G. Curr. Med. Chem. 2009, 16, 3740.
(9) Jacoby, G. A. Clin. Microbiol. Rev. 2009, 22, 161.
4

• A very mild condition to enzymatically
cleave an amide bond was identified.
• Amide coupling condition optimized
without protecting other functional
groups.
• An alternative approach to the bicyclic
core was developed to enable convergent
synthesis.
• All reactions shown above were done in >
80% yield each step.
5