Integrated Biocatalysis in Multistep Drug Synthesis without Intermediate Isolation: A de Novo Approach toward a Rosuvastatin Key Building Block

Article abstract of DOI:10.1021/acs.oprd.5b00057

In this contribution, we report the chemoenzymatic preparation of a key building block for the active pharmaceutical ingredient rosuvastatin, one of the "top 5 blockbuster drugs" with a worldwide market value of 6.25 billion USD in 2012, via a seven-step synthesis without isolation of intermediates and with incorporation of two highly efficient biotransformations. This chemoenzymatic process reaches excellent space-time yields by using high substrate concentrations (several hundred grams per liter), emphasizing the potential of biocatalysis for industrial processes related to pharmaceutical drug synthesis and the compatibility of enzyme chemistry with classical organic synthesis.

Full text of DOI:10.1021/acs.oprd.5b00057

Article
pubs.acs.org/OPRD
Integrated Biocatalysis in Multistep Drug Synthesis without
Intermediate Isolation: A de Novo Approach toward a Rosuvastatin
Key Building Block
Richard Metzner,†,⊥ Werner Hummel,‡,# Frank Wetterich,*,§ Burghard Konig,∥ and Harald Groger*,†
̈
̈
†Faculty of Chemistry, Bielefeld University, Universitatsstr. 25, 33615 Bielefeld, Germany
̈
‡Institute of Molecular Enzyme Technology, Julich Research Center, Heinrich-Heine-University Dusseldorf, Stetternicher Forst,
̈
̈
52426 Julich, Germany
̈
§Sandoz International GmbH, Industriestr. 18, 83607 Holzkirchen, Germany
∥Sandoz Industrial Products GmbH, Bruningstraße 50, 65929 Frankfurt am Main, Germany
̈
S
* Supporting Information
ABSTRACT: In this contribution, we report the chemoenzymatic preparation of a key building block for the active
pharmaceutical ingredient rosuvastatin, one of the “top 5 blockbuster drugs” with a worldwide market value of 6.25 billion USD
in 2012, via a seven-step synthesis without isolation of intermediates and with incorporation of two highly efficient
biotransformations. This chemoenzymatic process reaches excellent space-time yields by using high substrate concentrations
(several hundred grams per liter), emphasizing the potential of biocatalysis for industrial processes related to pharmaceutical drug
synthesis and the compatibility of enzyme chemistry with classical organic synthesis.
ith a worldwide market value of 6.25 billion USD in
W2012 and a U.S. market value of 5.31 billion USD,
rosuvastatin (1) belongs to the top 5 of the so-called
“blockbuster drugs”, hence encouraging the development of
more efficient and sustainable processes for its production.1
The expiration of relevant patents in 2016 further increases the
ongoing interest in novel synthetic approaches, especially for
generic drug manufacturers. In the past, this issue of
establishing the most efficient process and the most favorable
synthetic route toward active pharmaceutical ingredients was
often focused on a multistep retrosynthesis depending entirely
on chemical reaction sequences.2 In addition, industrial
syntheses are also challenging in regard to minimizing waste
and maximizing the space-time yield of the overall process.
Although currently biotransformations have been gaining
increasing attention as key steps for novel industrial drug
syntheses,3 as illustrated by the term biocatalytic retrosynthesis
coined for this field by Turner et al.,3c enzymatic reactions are
often still regarded as being incompatible with chemical
reactions, and usually the two are strictly separated from each
other with purification steps for most intermediates. Accord-
ingly, retrosynthetic approaches based on chemoenzymatic
one-pot processes for key intermediates remain rare. However,
even processes relying solely on chemical reactions often tend
to include purification steps of every intermediate. This
contradicts the aims of a sustainable process chemistry for
the production of pharmaceutical drugs as advised by the ACS
GCI Pharmaceutical Roundtable, established by the ACS Green
Chemistry Institute (GCI) and several of the world’s leading
pharmaceutical companies, as well as Pfizer’s “Green Chemistry
Metrics Programme”.4,5
herein we report a seven-step synthesis of an alkene key
intermediate of 1 that is prepared without purifying the
intermediates and with incorporation of two biotransforma-
tions. This chemoenzymatic process was developed to reach an
excellent space-time yield by using high substrate concen-
trations of several hundred grams per liter. Since this process
met and surpassed the required high productivity, it was
recently transferred to the industrial partner Sandoz for internal
work on process scale-up.6
Our initial retrosynthetic approach was based on the
development of an economically attractive synthesis of alkene
2 (Scheme 1). This compound is an industrial key intermediate,
since it is subsequently converted to the final rosuvastatin
Scheme 1. Retrosynthetic Wittig approach to rosuvastatin
calcium
Underlining the tremendous potential of biocatalysis for the
design of novel retrosynthetic approaches in drug synthesis,
Received: February 25, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.oprd.5b00057
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
calcium by well-established procedures (silyl ether cleavage,
diastereoselective Narasaka−Prasad reduction,7 ester hydrol-
ysis, salt formation).8 The formation of the CC double bond
in 2 can readily be achieved via Wittig reaction. Among the two
possibilities depicted in Scheme 1, the synthesis starting from
heteroaromatic aldehyde 5 and phosphorus ylide 6 appeared to
be more favorable for us, e.g., due to the presence of the oxo-
stabilized phosphorus ylide 6.8,9 However, while aldehyde 5 is
an easily accessible10 and commercial starting material, finding
an efficient synthetic route to the chiral phosphorus ylide 6
remained a challenge.
From a retrosynthetic point of view, such a type of stabilized
phosphorus ylide can be prepared by direct alkylation of
triphenylphosphine with a corresponding α-halo ketone11 or an
acylation of methyltriphenylphosphonium ylide with activated
carboxylic acid derivative 10 (as depicted in Scheme 2, route
As a result, none of the three initial routes (A, B, or C) could
be used as an overall process. It is, however, noteworthy that
the successful parts of the processes, i.e., the synthesis of chiral
monoester 9 starting from the prochiral bulk chemical 7 (route
A) and the synthesis of the chiral phosphorus ylide (R)-6b from
free 3-hydroxy monoester 14 (route B/C), are complementary
to each other. Thus, the missing “link” was the hydrolytic
cleavage of the O-acetyl group of monoester 9 to obtain the
required unprotected compound 14. At first glance this reaction
seemed trivial, but it became in fact quite challenging because
of the similar reactivities of the two carboxylic ester moieties in
this molecule. During a classical base-catalyzed saponification,
most of the product was lost through the formation of the
prochiral diacid (data not shown). Even enzymatic hydrolysis
turned out to be difficult. A range of examined hydrolases
showed no measurable activity for this compound. Fortunately,
a biocatalyst used for the synthesis of antibiotics, namely, a
cephalosporin C acetyl esterase,14 showed the desired
selectivity for the O-acetyl cleavage paired with excellent
efficiency. Hence, the missing “link” (Scheme 2, conversion of 9
to 14) was realized and successfully integrated into the process,
thus completing the overall reaction sequence starting from the
readily accessible and industrially attractive diester 7 to the
desired chiral ylide intermediate (R)-6b.
Scheme 2. Chemoenzymatic routes A, B, and C and their
“bottlenecks”
Next we focused on the development of a highly efficient
overall synthesis featuring excellent productivities, high
substrate concentrations, and facile catalyst separation while
avoiding isolation and purification steps. At first, we combined
the initial three reaction steps, especially the two consecutive
biotransformations for the preparation of chiral monoester 14
from acetylated diester 8. We were pleased to find that the
initial process development successfully reached an initial
substrate 8 concentration of 2.0 M, equivalent to 492 g/L, with
quantitative conversion and a very high enantioselectivity of
97% ee (Scheme 3). Since the resulting acid 9 has to be
A).12 An advantage of the latter route is the possibility of
obtaining this compound 10 by a straightforward enantiose-
lective ester hydrolysis of the readily available prochiral dialkyl
3-hydroxyglutarate of type 7. In this asymmetric step, the use of
a biocatalyst seems favorable. While the enzymatic desymmet-
rization using α-chymotrypsin was already proven to be R-
selective for different O-acyl derivatives of 7 and highly
enantioselective for the O-acetyl derivative 8 (in step 2 of route
A),13 the subsequent steps of route A in Scheme 2 became
troublesome. This was attributed to side reactions, e.g.,
elimination of acetic acid with formation of enoates due to
the lability of the O-acetyl group under strongly basic reaction
conditions. To suppress this undesired side reaction, different
derivatives of the chiral monoester 14 with an exchanged O-
protecting group were examined, revealing the tert-butyldime-
thylsilyl (TBS) ether as the most promising candidate (Scheme
2, route B/C). Starting from this TBS-protected monoester 12,
which is usually prepared over three steps via its L-mandelate
diastereomer,2b ylide 6b was prepared in two steps in high
yield. The drawback of this route B was the initial
desymmetrization reaction of silylated diester 11, since the
bulky O-substituent inhibited the enzymatic conversion of this
substrate. Furthermore, another alternative route C via
desymmetrization of the free hydroxy diester 7 itself could
not be established in an efficient fashion, since it was inferior in
terms of enantioselectivity. The enzymatic hydrolysis of 7
yielded the corresponding monoester 14 with only 60% ee,
even with the purified biocatalyst.
Scheme 3. Combination of two biotransformations for the
production of the chiral unprotected monoester
neutralized during the biotransformation step, the added
volume of the aqueous base lowers the “overall” substrate
concentration accordingly.
The O-acetylated monoester 9 was obtained in 95% isolated
yield. In addition, the subsequent regioselective ester hydrolysis
with the cephalosporin C acetyl esterase proceeded in a highly
efficient manner, and the monoester 14 was obtained with
>95% conversion in 77% isolated yield. This second enzymatic
reaction was also done with a high substrate concentration of
0.8 M, corresponding to 174 g/L (Scheme 3).
Even with these yields of 95% and 77% (obtained without
the need of column chromatography), isolation steps on a
technical scale are disadvantageous from both economic and
ecological perspectives due to, e.g., the large amount of
extraction solvent, the resulting amount of waste, and the
B
DOI: 10.1021/acs.oprd.5b00057
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
increased time and reactor capacity. Thus, the next aim was the
combination of the three initial process steps (O-acetylation of
diester 7, α-chymotrypsin-catalyzed desymmetrization of 8, and
cephalosporin C acetyl esterase-catalyzed deacetylation of 9)
with practical unit operations steps for separation of the
enzymes from the reaction mixture without the necessity of
isolation or purification procedures of the formed products
(Table 1). After the initial solvent-free acetylation of diester 7
was added under the same pH-controlled reaction conditions.
This heterogeneous biocatalyst was removed via filtration over
a glass frit after quantitative formation of free hydroxy
monoester (R)-14. With this enzyme and its outstanding
chemoselectivity for the acetyl group, it was therefore possible
to produce chiral monoester (R)-14 with a substrate loading of
more than 430 g/L, an overall yield of 95% over three steps
starting from diester 7, and a very high enantioselectivity of
>97% ee. These three consecutive steps of chemical acetylation
followed by two complementary biotransformations proceeded
in a highly efficient way, with the (initial) 980 g/L
concentration of 8 being one of the highest substrate
concentrations reported to date in biocatalysis.
Table 1. Three-step synthesis with enzyme recycling
In addition to the avoidance of any isolation steps between
these three reaction steps, both enzymes could be reused after
their separation from the corresponding product mixture. The
catalyst recycling was examined over five reaction cycles, with
each cycle being stopped after reaching >95% conversion as
determined by the consumption of aqueous base to maintain
the reaction pH (Table 1). While the reaction time for α-
chymotrypsin increased slowly within each cycle (due to partial
inactivation), the cephalosporin C acetyl esterase demonstrated
impressive stability and high efficiency with a negligible
decrease in activity over five cycles.
a
conv. [%]
productivity [mg g−1 day−1
]
Furthermore, on the basis of this efficient three-step
synthesis of (R)-14 from 7 without intermediate isolation, a
seven-step overall process for alkene key intermediate (R)-2b
was developed that does not require isolation procedures for
purified intermediates (Scheme 4). Following the described
three-step chemoenzymatic synthesis of (R)-14, the crude
product was silylated by chemical means with tert-butyldime-
thylchlorosilane and 1H-imidazole. Although purification of
silylated monoester (R)-12 can be avoided as well, the isolation
of (R)-12 is beneficial and represents the only purification step
that improved the overall yield of the route by removing the
residual tert-butyldimethylsilanol byproduct, which forms a
hemihydrate16 that subsequently hydrolyzes parts of the
following reagents.
The conversion of (R)-12 into mixed anhydride (S)-13 and
the acylation of methyltriphenylphosphonium ylide to obtain
chiral ylide (R)-6b were again conducted without isolation
steps of the respective products (Scheme 4). The crude product
(R)-6b was reacted in situ with heteroaromatic aldehyde 5 in a
Wittig reaction, which gave the key intermediate (R)-2b in 59%
overall conversion relative to diester 8. It should be noted that
all of these reactions ran with high and in many cases even
cycle
B
C
CHY (B)
CAE (C)
1
2
3
4
5
>95
>95
>95
>95
>95
>95
>95
>95
>95
>95
31.51
32.12
38.27
68.34
124.43
44.35
44.64
49.36
49.65
49.91
b
b
b
a
Milligrams of biocatalyst per gram product and a reaction time of 24
b
h. The immobilized enzyme was stored for 14 days at 4−8 °C
between cycles 2 and 3.
in acetic anhydride, the subsequent desymmetrization reaction
was realized without isolation of 8 and with an excellent
(initial) substrate concentration of 4.0 M, corresponding to
more than 980 g/L. Because of the addition of aqueous base in
the biotransformation step for neutralization purposes, the
reaction volume is increased and the “overall” substrate
concentration is lowered accordingly. The homogeneous
biocatalyst α-chymotrypsin was subsequently separated via
ultrafiltration15 (and recycled), while the filtrate containing
chiral monoester (R)-9 was directly used for the next reaction.
For this step, the immobilized cephalosporin C acetyl esterase
Scheme 4. Final seven-step reaction sequence leading to the rosuvastatin key intermediate (R)-2b
C
DOI: 10.1021/acs.oprd.5b00057
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
(5) Assaf, G.; Checksfield, G.; Critcher, D.; Dunn, P. J.; Field, S.;
Harris, L. J.; Howard, R. M.; Scotney, G.; Scott, A.; Mathew, S.;
Walker, G. M. H.; Wilder, A. Green Chem. 2012, 14, 123−129.
quantitative conversions, which emphasizes the high compat-
ibility of the chemical and enzymatic reaction steps (Scheme 4).
In summary, a chemoenzymatic process for the production of
key intermediate (R)-2b leading to the “blockbuster” drug
rosuvastatin (1) was developed, comprising a seven-step
reaction sequence without mandatory isolation steps. The
two consecutive biotransformations run on excellent substrate
concentrations of up to 4.0 and 2.0 M respectively, to afford the
desired chiral monoester (R)-14 with a final product
concentration of 220 g/L. Recently this process was transferred
to the industrial partner Sandoz for internal work on process
scale-up, and a patent application on the biocatalytic
enantioselective synthesis of 3-hydroxy-glutaric acid mono-
esters and their use was filed.6 The successful implementation
of two recyclable biocatalysts for the enantioselective synthesis
of the chiral monoester (R)-14 in combination with organic-
chemical derivatization under retention of the stereogenic
information underlines the great potential of biocatalysis in the
development of novel industrially applicable retrosynthetic
chemoenzymatic routes to active pharmaceutical ingredients.
(6) Konig, B.; Wetterich, F.; Groger, H.; Metzner, R. Process for
̈
̈
enantioselective synthesis of 3-hydroxy-glutaric acid monoesters and
use thereof. PCT Int. Appl. WO 2014/140006, Sept 18, 2014.
(7) (a) Narasaka, K.; Pai, F.-C. Tetrahedron 1984, 40, 2233−2238.
(b) Chen, K.-M.; Hardtmann, G. E.; Prasad, K.; Repic,
J. Tetrahedron Lett. 1987, 28, 155−158.
̌
O.; Shapiro, M.
(8) For overviews of the industrial synthetic routes to rosuvastatin,
see: (a) Kleemann, A.; Engel, J.; Kutscher, B.; Reichert, D.
Pharmaceutical Substances: Syntheses, Patents, Applications; Thieme:
Stuttgart, Germany, 2001; pp 1234 ff. (b) Casar, Z. Curr. Org. Chem.
2010, 14, 816−845.
(9) For recent developments regarding an alternative route using a
formylated statin side chain, see: (a) Osl
Kopitar, G.; Mrak, P.; Casar, Z.; Marr, A. C. PLoS One 2013, 8,
̌
aj, M.; Cluzeau, J.; Orkic,
́
D.;
̌
̌
No. e62250. (b) Casar, Z.; Steinbucher, M.; Kosm
̌
rlj, J. J. Org. Chem.
̈
2010, 75, 6681−6684.
(10) Sterk, D.; Casar, Z.; Jukic,
2155−2160.
(11) Lin, W.; Zheng, H.; Liu, X. PCT Int. Appl. WO 2011/124050
A1, Oct 13, 2011.
̌
̌
̌
M.; Kosm
̌
rlj, J. Tetrahedron 2012, 68,
(12) Watanabe, M.; Koike, H.; Ishiba, T.; Okada, T.; Seo, S.; Hirai, K.
ASSOCIATED CONTENT
■
Bioorg. Med. Chem. 1997, 5, 437−444.
S
* Supporting Information
(13) (a) Santaniello, E.; Chiari, M.; Ferraboschi, P.; Trave, S. J. Org.
̈
Chem. 1988, 53, 1567−1569. (b) Ohrlein, R.; Baisch, G. Adv. Synth.
Experimental procedures and analytical methods. The Support-
ing Information is available free of charge on the ACS
Publications website at DOI: 10.1021/acs.oprd.5b00057.
Catal. 2003, 345, 713−715.
(14) Riethorst, W.; Reichert, A. Chimia 1999, 53, 600−607.
(15) For more examples of enzyme separation via ultrafiltration, see:
(a) Liese, A.; Kragl, U.; Kierkels, H.; Schulze, B. Enzyme Microb.
Technol. 2002, 30, 673−681. (b) Haberland, J.; Hummel, W.;
Daussmann, T.; Liese, A. Org. Process Res. Dev. 2002, 6, 458−462.
(16) Barry, S. M.; Mueller-Bunz, H.; Rutledge, P. J. Acta Crystallogr.
2008, E64, o1174.
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail: harald.groeger@uni-bielefeld.de.
*E-mail: frank.wetterich@sandoz.com.
Present Addresses
⊥R.M.: Asano Active Enzyme Molecule Project, ERATO, JST
Biotechnology Research Center, Toyama Prefectural Univer-
sity, 5180 Kurokawa, Imizu, Toyama 939-0398, Japan.
#W.H.: Faculty of Chemistry, Bielefeld University, Universi-
tatsstr. 25, 33615 Bielefeld, Germany.
̈
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Sandoz AG for financial and technical support and
student Susann Rath for experimental support.
REFERENCES
■
(1) Brooks, M. List of Top 100 Selling Drugs in 2013. Medscape
Medical News, Jan 14, 2014; http://www.medscape.com/viewarticle/
820011 (accessed Feb 25, 2015).
(2) (a) Kleemann, A.; Engel, J.; Kutscher, B.; Reichert, D.
Pharmaceutical Substances: Syntheses, Patents, Applications; Thieme:
Stuttgart, Germany, 2001. (b) Konoike, T.; Araki, Y. J. Org. Chem.
1994, 59, 7849−7854.
(3) (a) Bornscheuer, U. T.; Huisman, G. W.; Kazlauskas, R. J.; Lutz,
S.; Moore, J. C.; Robins, K. Nature 2012, 485, 185−194. (b) Renata,
H.; Wang, Z. J.; Arnold, F. H. Angew. Chem., Int. Ed. 2015, 54, 3351−
3367. (c) Turner, N. J.; O’Reilly, E. Nat. Chem. Biol. 2013, 9, 285−288.
(d) Wohlgemuth, R. Curr. Opin. Biotechnol. 2010, 21, 713−724.
(e) Liese, A.; Seelbach, K.; Wandrey, C. Industrial Biotransformations;
Wiley-VCH: Weinheim, Germany, 2006.
(4) Constable, D. J. C.; Dunn, P. J.; Hayler, J. D.; Humphrey, G. R.;
Leazer, J. L.; Linderman, R. J.; Lorenz, K.; Manley, J.; Pearlman, B. A.;
Wells, A.; Zaksh, A.; Zhang, T. Y. Green Chem. 2007, 9, 411−420.
D
DOI: 10.1021/acs.oprd.5b00057
Org. Process Res. Dev. XXXX, XXX, XXX−XXX