High-yielding syntheses of 1-piperidin-4-yl butyro- and valerolactams through a tandem reductive animation-lactamization (reductive lactamization)

Article abstract of DOI:10.1021/op700016b

We report a procedure for the concise and high-yielding syntheses of 1-piperidin-4-yl-substituted butyro- and valero-lactams. Beginning with 1-benzyl-4-piperidone and γ- or δ-amino esters or acids, we have effected a tandem reductive animation lactamizati

Full text of DOI:10.1021/op700016b

Organic Process Research & Development 2007, 11, 482−486
High-Yielding Syntheses of 1-Piperidin-4-yl Butyro- and Valerolactams through
a Tandem Reductive Amination-Lactamization (Reductive Lactamization)
Christopher M. Mapes* and Neelakandha S. Mani
Johnson & Johnson Pharmaceutical Research and DeVelopment, L.L.C., 3210 Merryfield Row,
San Diego, California 92121, U.S.A.
Abstract:
We report a procedure for the concise and high-yielding
syntheses of 1-piperidin-4-yl-substituted butyro- and valero-
lactams. Beginning with 1-benzyl-4-piperidone and γ- or
δ-amino esters or acids, we have effected a tandem reductive
amination-lactamization using sodium triacetoxyborohydride.
This procedure represents an inexpensive and scaleable alterna-
tive to previous multistep syntheses of these important phar-
maceutical building blocks.
Figure 1. 1-Piperidin-4-yl lactam targets.
related compounds, which suffers from similar shortcomings,
8
1
-Piperidin-4-yl-substituted butyro- and valerolactam
has also been reported in the patent literature.
2
moieties have been used in a variety of biologically active
A second synthesis reported in the literature involved a
reductive amination between a lactam and 1-benzyl-4-
1
molecules, including tachykinin antagonists, dual NK1/NK2
2
3
inhibitors, and blood coagulation factor X (FXa) inhibitors.
3
piperidone using NaCNBH (Scheme 2). However, in our
Recently, these subunits have been used within our internal
medicinal chemistry programs. As the demand for these
piperidinyl lactams increased, we sought an improved and
scaleable synthesis for 1, 2, and 3 (Figure 1).
hands, this reaction failed to yield the desired product.
Our retrosynthetic analysis of 1 revealed a protected
4-piperidone and a linear amino acid as potential starting
materials (Scheme 3).
4
A previously reported synthesis of compound 1 (Scheme
An analogous method for the synthesis of tertiary lactams
has been published by Abdel-Magid et al. In this case, a
9
1
) employed a functionalized acid chloride in conjunction
5
with N-Cbz-4-aminopiperidine (5). N-Cbz-4-aminopiperi-
dine is derived from the 4-(N-Boc-amino)piperidine (4) in
lactam ring was installed onto aliphatic and carbocyclic
ketones using ethyl-4-aminobutyrate. The discovery of this
tandem reductive amination/lactamization, which the authors
coined “reductive lactamization”, was made while building
on their exploration into the uses of sodium triacetoxyboro-
6
modest yields by a protection/deprotection sequence. Fol-
lowing the two-step installation of the lactam, the Cbz
protecting group is removed by hydrogenation in the presence
1
0
of catalytic Pd(OH)
2
.
hydride. We were attracted to this method of synthesis by
both its operational simplicity and the potential for utilization
of inexpensive, commercially available starting materials.7
This methodology was used in the initial synthesis of
compounds 1 and 2 by our medicinal chemistry teams.
However, this sequence was lengthy, and it suffered from
use of expensive starting materials and reagents that were
Results and Discussion
7
not amenable to use on large scale. A sequence toward
Initially we chose 1-Boc-4-piperidone and ethyl 4-ami-
nobutyrate as the starting materials for our synthesis of
compound 1 (Scheme 4). The Boc protecting group was
*
To whom correspondence should be addressed. E-mail: cmapes@
prdus.jnj.com.
(
1) For examples, see: (a) Seto, S.; Tanioka, A.; Ikeda, M.; Izawa, S. PCT
Int. Publ. No. WO2003/080619, 2003. (b) Seto, S.; Tanioka, A.; Ikeda,
M.; Izawa, S. PCT Int. Publ. No. WO2003/062245, 2003. (c) Middleton,
D. S.; Stobie, A. PCT Int. Publ. No. WO2003/051868, 2003. (d) Seto, S.;
Tanioka, A.; Ikeda, M.; Izawa, S. PCT Int. Publ. No. WO2003/050123,
(7) A search using the Scifinder database was able to locate 4-(Boc-amino)-
piperidine priced at $6.6/g (250 g) and 1-benzyl-4-piperidone priced at
$0.31/g (5 kg).
(8) Reichard, G. A.; Aslanian, R. G.; Alaimo, C. A.; Kirkup, M. P.; Lupo, A.,
Jr.; Mangiaracina, P.; McCormick, K. D.; Piwinski, J. J.; Shankar, B. B.;
Shih, Neng-Yang; Spitler, J. M.; Ting, P. C.; Ganguly, A.; Carruthers, N.
I. (Cont.-in-part of U.S. Ser. No. 460,819, abandoned) U.S. Patent
5,696,267, 1997.
(9) Abdel-Magid, A. F.; Harris, B. D.; Maryanoff, C. A. Synlett 1994, 81.
(10) Abdel-Magid, A. F.; Maryanoff, C. A.; Carson, K. G. Tetrahedron Lett.
1990, 31, 5595. (b) Abdel-Magid, A. F.; Maryanoff, C. A. Synlett 1990,
537. (c) Abdel-Magid, A. F.; Carson, K. G.; Harris, B. D.; Maryanoff, C.
A.; Shah, R. D. J. Org. Chem. 1996, 61, 3849. (d) Abdel-Magid, A. F.;
Maryanoff, C. A. Use of Sodium Triacetoxyborohydride in Reductive
Amination of Ketones and Aldehydes. In Reductions In Organic Synthe-
sis: Recent AdVances and Practical Applications; Abdel-Magid, A. F., Ed.;
ACS Symposium Series 641; American Chemical Society: Washington,
DC, 1996; pp 201-216.
2
003.
(
2) Ting, P. C.; Lee, J. F.; Anthes, J. C.; Shih, N. Y.; Piwinski, J. J. Bioorg.
Med. Chem. Lett. 2001, 11, 491.
(
(
3) Kubo, K.; Imaeda, Y. PCT Int. Publ. No. WO2005/058823, 2005.
4) Axe, F. U.; Bembenek, S. D.; Butler, C. R.; Edwards, J. P.; Fourie, A. M.;
Grice, C. A.; Savall, B. M.; Tays, K. L.; Wei, J. PCT Int. Publ. No.
WO2005/012297, 2005. (b) Axe, F. U.; Bembenek, S. D.; Butler, C. R.;
Edwards, J. P.; Fourie, A. M.; Grice, C. A.; Savall, B. M.; Tays, K. L.;
Wei, J. PCT Int. Publ. No. WO2005/012296, 2005.
(
(
5) Miller, S. C. PCT Int. Publ. No. WO94/10146, 1994.
6) Recent literature details the direct and selective substitution of secondary
amines in the presence of primary amines: Laduron, F.; Tamborowski,
V.; Moens, L.; Horvath, A.; De Smaele, D.; Leurs, S. Org. Process Res.
DeV. 2005, 9, 102.
4
82
•
Vol. 11, No. 3, 2007 / Organic Process Research & Development
10.1021/op700016b CCC: $37.00 © 2007 American Chemical Society
Published on Web 04/19/2007

Scheme 1. Literature synthesis of lactam 1
Scheme 2. Literature synthesis of compounds 1 and 2
Scheme 5. First successful synthesis of lactam 1 through
reductive lactamization
Scheme 3. Retrosynthetic deconstruction of lactam 1
Scheme 4. Initial unsuccessful synthesis of lactam 10
Screening the commercially available N-acetyl, N-Cbz,
N-carbethoxy, and N-benzyl-4-piperidone we were able, in
each case, to produce the fully cyclized 1-piperdin-4-yl
lactam system in good yield. Ultimately, 1-benzyl-4-piperi-
done was chosen as our preferred starting material due to
its low cost, commercial availability, and ease of handling.
Using 1-benzyl-4-piperidone as the starting material, the
synthesis of 1-piperidin-4-yl-pyrrolidin-2-one (1) was com-
pleted in a two-step sequence (Scheme 5).
chosen for its expected relative ease of removal in the final
steps of the synthesis. The reductive amination is complete
after 2 h at room temperature. Unexpectedly, compound 8
did not cyclize to form the desired lactam system 9 due to
in situ Boc deprotection and subsequent degradation of the
resultant species.
The use of elevated temperatures and/or the addition of
NaOH or TsOH did not effect a cyclization of compound 8
to form lactam 9. The addition of TFA did result in complete
removal of the Boc protecting group; however, compound
In the first step of the sequence, ethyl 4-aminobutyrate
11
and sodium triacetoxyborohydride were employed to install
the lactam ring onto the 1-benzyl-4-piperidone. The butyrate
is commercially available as the hydrated HCl salt, and thus,
triethylamine was employed to buffer the reaction. We have
found that the preferred process for the installation of the
lactam ring involves aging the reaction at ambient temper-
ature until the reductive amination is complete. Heating is
subsequently applied to produce complete cyclization to the
(
11) The reductive lactamization, when using sodium borohydride in methanol,
required 60 h at 45 °C to yield 70% product. The mass balance was the
trans-esterified, uncyclized, reductive amination product.
1
was not observed. These findings led us to explore other
protected piperidones.
Vol. 11, No. 3, 2007 / Organic Process Research & Development
•
483

Scheme 6. Second-generation synthesis of compound 10
through reductive lactamization
Scheme 7. Synthesis of lactam 2 through reductive
lactamization
lactam.12 The obtained product is generally of sufficient
purity for further use; however, it may be purified by
recrystallization from hot EtOAc/heptane. We have found
that 1,2-dichloroethane, THF, 2-methyl-THF, MTBE, tolu-
ene, and (trifluoromethyl)benzene are all good solvent
choices for this reaction. Our initial experiments utilized 1,2-
dichloroethane in keeping with the previous literature
Scheme 8. Medicinal chemistry synthesis of hydroxy
lactam 3
9
precedent. Currently, our preferred solvent for this trans-
formation is toluene. In the second step of the sequence, the
benzyl group was cleaved by hydrogenation, using ethanol
and 10% Pd/C, to yield 1.
The initial method for the production of these lactams
required the portionwise addition of the reducing agent into
the reaction mixture (Method A). This method became
problematic when investigated on a 0.375 mol scale due to
an increase in solution viscosity, a potentially hazardous
exotherm, and the mechanical difficulties involved with
adding solid sodium triacetoxyborohydride. Due to these
potential process hazards, an alternate procedure, where the
reaction flask was charged with the butyrate, borohydride,
and triethylamine followed by controlled addition of the
piperidone, has also been developed (Method B). This
procedure alleviates the difficulties observed with Method
A and has emerged as our method of choice. This method
has been successfully demonstrated on a 0.75 mol scale.
Subsequently, it was found on a 5.0-g scale that we could
accomplish the same transformation in comparable yields
and ease by replacing the ethyl 4-aminobutyrate with
4-aminobutyric acid (GABA) (Method C). This variation of
the transformation obviates the use of triethylamine in the
reaction (Scheme 6). The direct application of free amino
acids should provide access to a wider range of analogues
while concomitantly lowering material costs.
We reasoned that our reductive lactamization process
could also be employed for the synthesis of 4-hydroxylactam
3. Our initial attempts utilized commercially available,
racemic 4-amino-3-hydroxylbutyric acid. However, use of
the acid directly for the reductive lactamization failed to yield
hydroxylactam 17 in any appreciable yield. When methyl
Direct application of this process using 1-benzyl-4-
piperidone and 5-aminovaleric acid as the feedstocks pro-
vided valerolactam 2 in good yield (Scheme 7). It is
interesting to note that an extended reaction time was
required to complete the lactamization of the acyclic
intermediate to form compound 11. The benzyl group was
removed by hydrogenation, as in the synthesis of 1.
In addition to syntheses of lactams 1 and 2, we also sought
a process for the synthesis of the 4-hydroxy-1-piperidin-4-
yl-pyrrolidin-2-one (3). The medicinal chemistry route
4
-amino-3-hydroxylbutyrate (16) was used, hydroxylactam
1
7 was produced in good yield and high purity (Scheme 9).
Deprotection of the benzyl group with Pd/C was found to
be ineffective. Employing Pd(OH) /C, the hydrogenolysis
proceeded cleanly and in excellent yield. The addition of
0% acetic acid to the reaction media was found to increase
2
1
3
(Scheme 8) was based on the work of Osamu Kanno et al.
1
(
12) This protocol was developed after an initial iteration presented a process
hazard on 10.0-g scale. When a vessel was charged with all of the reagents
and heated to 60 °C directly, a severe exotherm coupled with violent off
gassing was observed.
the rate of the deprotection; however, the isolated acetate
salt of compound 3 proved very hygroscopic and quickly
deliquesced into a dark-red oil.
(13) Kanno, O.; Miyauchi, M.; Kawamoto, I. Heterocycles 2000, 53, 173.
484
•
Vol. 11, No. 3, 2007 / Organic Process Research & Development

Scheme 9. Synthesis of hydroxylactam 3 through reductive
lactamization
glass bottles and used directly. “Brine” refers to a saturated
aqueous solution of NaCl.
1-(1-Benzyl-piperidin-4-yl)-pyrrolidin-2-one (10). Method
B. To a 5-L jacketed reactor equipped with an overhead
mechanical stirrer, thermocouple probe, J-Kem dose control-
ler, and dynamic nitrogen inlet were added ethyl-4-aminobu-
tyrate hydrochloride (150.9 g, 0.90 mol), NaBH(OAc)
206.6 g, 0.97 mol), and anhydrous toluene (2.25 L). Stirring
3
(
was commenced, and triethylamine (379.5 g, 3.75 mol) was
added over a 5-min period. Once addition was complete, the
reactor was aged at 20 °C for 10 min. A solution of 1-benzyl-
Conclusion
By employing a reductive lactamization process, scaleable
syntheses of compounds 1 and 2 were developed by our
laboratories. As proof of the method’s generality, we have
also successfully applied this process to a concise, high-
yielding synthesis of racemic 3. The benefits of this process
are the use of widely available and inexpensive commercial
feedstocks as well as mild reaction conditions. A further
strength of our optimized process is that it does not require
any chromatography. The crude reaction products are
obtained in good purity and can be further purified by
recrystallization if necessary. The ability to use both amino
esters and, in some cases, the amino acids to construct the
lactam greatly widens the utility of this process. Furthermore,
this process is an example of a direct reductive amination
using an open-chain amino acid, a reaction that is rarely seen
4
-piperidone (141.9 g, 0.75 mol) in anhydrous toluene (250
mL) was then added over a 1-h period. A slight exotherm
of ∼6 °C was observed during addition. After completion
of addition, the solution was aged at 20 °C for 1 h and then
heated to 75 °C for an additional 2 h. The solution was then
cooled to 20 °C and quenched by the slow addition of water
(
1 L). A mild exotherm of ∼14 °C was observed during the
addition. The layers were separated after 20 min of stirring,
and the organic layer was extracted again with water (1 L).
The organic layer was then extracted with 1 N aqueous HCl
(
3 × 500 mL). The aqueous layers were combined, adjusted
to a pH of ∼12 with 6 N aqueous NaOH, and extracted with
i-PrOAc (2 × 1 L). All organic layers were combined,
4
washed with brine, dried over anhydrous MgSO , filtered,
and concentrated under vacuum until a thick, stirrable slurry
was achieved. This slurry was diluted with heptane (1.5 L)
with stirring. The resultant slurry was slowly cooled to 10
14
in organic synthesis. This modular approach to the lactam
formation should enable the assemblage of a wide range of
substitution patterns.
°
C, and the product was collected by filtration. The cake
was air-dried for 20 min and then placed in a 55 °C vacuum
oven for 24 h to yield 1-(1-benzyl-piperidin-4-yl)pyrrolidin-
Experimental Section
General Experimental Chemical Procedures. Nuclear
2
-one (11) (178.0 g, 92% yield) as an off-white solid.
1
13
magnetic resonance ( H NMR, C NMR) spectra were
recorded on a Bruker DPX 400 or Bruker DPX 500
spectrometer. Melting points were taken using a TA Instru-
ments DSCQ 100 apparatus. High-resolution mass spectra
were taken on a Bruker microTOF apparatus. Infrared
spectroscopy was performed on a Nicolet Avatar 360 FT-
IR as a neat pellet. Flash column chromatography was
performed using Merck silica gel 60. HPLC analysis was
performed on a Hewlett-Packard 1100 (Agilent ZORBAX
Eclipse XDB-C8, 5 µm, 4.6 mm × 150 mm, flow rate 1
mL/min, gradient (acetonitrile/water with 0.05% trifluoro-
acetic acid): 1% acetonitrile/99% water to 99% acetonitrile/
1
3
H NMR (400 MHz, CDCl ): δ 7.34-7.24 (m, 5H), 3.99
(tt, J ) 11.9 Hz, J ) 4.4 Hz, 1H), 3.50 (s, 2H), 3.35 (t, J )
7
2
7
1
1
3
1
.0 Hz, 2H), 2.95-2.91 (m, 2H), 2.38 (t, J ) 8.1 Hz, 2H),
.09 (dt, J ) 11.8 Hz, J ) 2.6 Hz, 2H), 1.99 (pentet, J )
.6 Hz, 2H), 1.72 (dq, J ) 12.2 Hz, J ) 8.3 Hz, 2H), 1.64-
13
.60 (m, 2H). C NMR (100 MHz, CDCl
3
): δ 174.52,
38.31, 129.15, 128.22, 127.05, 63.08, 52.78, 48.88, 42.93,
1.57, 29.30, 18.14. IR (neat): 1667, 1492, 1421, 1365, 1342,
310, 1265, 1220, 1145, 1124, 1023, 989, 792, 737, 698
-1
cm . Anal. Calcd (found): C, 74.38 (74.47); H, 8.58 (8.44);
N, 10.84 (10.90). MS (electrospray): exact mass calcd for
+
C
16
H
22
N
2
O 258.1732; m/z found 259.1803 [M + H] ; mp
1
1
% water ramp over 8 min, then hold at 99% acetonitrile/
% water). Combustion analyses were preformed by Numega
8
1.3 °C.
-Piperidin-4-yl-pyrrolidin-2-one (1). To a 2.25-L Parr
1
Resonance Labs, San Diego, California. All reagents were
purchased from Aldrich and used as received with the
exception of ethyl-4-aminobutyrate hydrochloride, which was
purchased from Alfa Aesar. All solvents utilized were
purchased in anhydrous form from E.M. Scientific and passed
through two columns of neutral alumina (DCE, THF, MTBE,
MeOH) or one column of neutral alumina and one column
of Q5 oxygen scavenger (toluene) with the exception of
flask were added 1-(1-benzyl-piperidin-4-yl)-pyrrolidin-2-
one (11) (91.6 g, 0.355 mol), ethanol (200 proof, 550 mL),
and 10 wt % Pd/C (9.0 g, wet catalyst). The Parr flask was
then shaken under a hydrogen atmosphere (45 psi) until
hydrogen consumption ceased, ∼36 h. The catalyst was
removed through filtration (Zapcap-CR, 0.45 µm Nylon) and
washed with EtOAc. The combined filtrates were concen-
trated under vacuum to yield 1-piperidin-4-yl-pyrrolidin-2-
one (1) (57.1 g, 95% yield) as a pale-yellow oil that solidified
to an off-white solid upon standing. On smaller scale (5.0
grams) the use of acidic media (3:1 EtOH/HOAc) reduced
the reaction time to ∼18 h.
2-methyl THF and (trifluoromethyl)benzene which were
purchased in anhydrous form from Aldrich in “Sure Seal”
(14) Levadala, M. K.; Banerjee, S. R.; Maresca, K. P.; Babich, J. W.; Zubieta,
J. Synthesis 2004, 1759.
Vol. 11, No. 3, 2007 / Organic Process Research & Development
•
485

1H NMR (500 MHz, CDCl
4.2 Hz, 1H), 3.37 (t, J ) 7.0 Hz, 2H), 3.13-3.11 (m,
H), 2.71 (dt, J ) 12.1 Hz, J ) 2.5 Hz, 2H), 2.39 (t, J )
chemical discussions. We also thank Dr. Pippel for his
assistance in the preparation of this manuscript.
3
): δ 4.06 (tt, J ) 11.9 Hz, J
)
2
8
2
.1 Hz, 2H), 1.99 (pentet, J ) 7.6 Hz, 2H), 1.67-1.64 (m,
Supporting Information Available
13
H) 1.58 (dq, J ) 12.1 Hz, J ) 4.1 Hz, 2H). C NMR (100
): δ 174.45, 49.05, 45.84, 42.95, 31.57, 30.42,
8.13. IR (neat): 2941, 1657, 1492, 1424, 1317, 1285, 1214,
MHz, CDCl
3
Experimental details concerning the syntheses of com-
pounds 10 (methods A and C), 11, 2, 16, 17, and 3. This
material is available free of charge via the Internet at http://
pubs.acs.org.
1
1
(
-1
141, 1091, 1006, 932, 870, 810, 682, 633 cm . Anal. Calcd
found): C, 64.25 (64.03); H, 9.59 (9.91); N, 16.65 (16.34).
MS (electrospray): exact mass calcd for C
m/z found 169.1338 [M + H] ; mp 106.2 °C.
9
H
16
2
N O 168.1263;
+
Received for review January 24, 2007.
OP700016B
Acknowledgment
We thank Heather McAllister for analytical support as
well as Danielle Neff and Dr. Dan Pippel for valuable
486
•
Vol. 11, No. 3, 2007 / Organic Process Research & Development