First chemoenzymatic stereodivergent synthesis of both enantiomers of promethazine and ethopropazine

Article abstract of DOI:10.3762/bjoc.10.322

Enantioenriched promethazine and ethopropazine were synthesized through a simple and straightforward four-step chemoenzymatic route. The central chiral building block, 1-(10H-phenothiazin-10-yl)propan-2-ol, was obtained via a lipase-mediated kinetic resolution protocol, which furnished both enantiomeric forms, with superb enantioselectivity (up to E = 844), from the racemate. Novozym 435 and Lipozyme TL IM have been found as ideal biocatalysts for preparation of highly enantioenriched phenothiazolic alcohols (up to >99% ee), which absolute configurations were assigned by Mosher's methodology and unambiguously confirmed by XRD analysis. Thus obtained key-intermediates were further transformed into bromide derivatives by means of PBr₃, and subsequently reacted with appropriate amine providing desired pharmacologically valuable (R)- and (S)-stereoisomers of title drugs in an ee range of 84-98%, respectively. The modular amination procedure is based on a solvent-dependent stereodivergent transformation of the bromo derivative, which conducted in toluene gives mainly the product of single inversion, whereas carried out in methanol it provides exclusively the product of net retention. Enantiomeric excess of optically active promethazine and ethopropazine were established by HPLC measurements with chiral columns.

Full text of DOI:10.3762/bjoc.10.322

First chemoenzymatic stereodivergent synthesis of both
enantiomers of promethazine and ethopropazine
Paweł Borowiecki*, Daniel Paprocki and Maciej Dranka
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2014, 10, 3038–3055.
Warsaw University of Technology, Faculty of Chemistry,
Noakowskiego St. 3, 00-664 Warsaw, Poland
doi:10.3762/bjoc.10.322
Received: 03 September 2014
Accepted: 01 December 2014
Published: 18 December 2014
Email:
Paweł Borowiecki* - pawel_borowiecki@onet.eu
*
Corresponding author
Associate Editor: J. Aubé
Keywords:
© 2014 Borowiecki et al; licensee Beilstein-Institut.
License and terms: see end of document.
ethopropazine; lipase-catalyzed kinetic resolution; Mosher
methodology; promethazine; stereodivergent synthesis
Abstract
Enantioenriched promethazine and ethopropazine were synthesized through a simple and straightforward four-step chemoenzy-
matic route. The central chiral building block, 1-(10H-phenothiazin-10-yl)propan-2-ol, was obtained via a lipase-mediated kinetic
resolution protocol, which furnished both enantiomeric forms, with superb enantioselectivity (up to E = 844), from the racemate.
Novozym 435 and Lipozyme TL IM have been found as ideal biocatalysts for preparation of highly enantioenriched phenothiazolic
alcohols (up to >99% ee), which absolute configurations were assigned by Mosher’s methodology and unambiguously confirmed
by XRD analysis. Thus obtained key-intermediates were further transformed into bromide derivatives by means of PBr3, and subse-
quently reacted with appropriate amine providing desired pharmacologically valuable (R)- and (S)-stereoisomers of title drugs in an
ee range of 84–98%, respectively. The modular amination procedure is based on a solvent-dependent stereodivergent transforma-
tion of the bromo derivative, which conducted in toluene gives mainly the product of single inversion, whereas carried out in
methanol it provides exclusively the product of net retention. Enantiomeric excess of optically active promethazine and etho-
propazine were established by HPLC measurements with chiral columns.
Introduction
Since enantiomorphs of biologically active compounds exhibit chemists. Nowadays, classical chemical manufacturing tech-
significant differences in their pharmacokinetic and pharmaco- nologies are increasingly displaced by pure biotechnological
dynamic behavior, optical purity of chiral drugs has emerged as processes or biotransformations coupled with chemical opera-
an important factor in efficacy and safety of their application tions [4-10]. The main advantages of these impactful tech-
[
1-3]. Thus, the development of an effective, inexpensive and niques are high substrate specificity as well as regio- and stereo-
environmentally friendly method for the formulation of single selectivity of the reaction [11-13]. In addition, the easy handling
enantiomers still remains a challenging task for organic (not involving the use of a complex chemical apparatus), possi-
3038

Beilstein J. Org. Chem. 2014, 10, 3038–3055.
bility of catalyst recovery, mild experimental (room tempera- tioenriched form, that is: ethopropazine (profenamine). In turn,
ture, atmospheric pressure, etc.) and non-hazardous environ- this antidyskinetic drug has been widely used in clinical prac-
mental conditions as well as organic solvent-, moisture- and air- tice for over the past 30 years in the treatment of Parkinson's
stability make tandem chemoenzymatic processes attractive for disease, although there is no data available regarding its phar-
both: laboratory and industrial applications. Since enzyme- macokinetic properties in humans. Only one study on the
catalyzed processes provide an access to enantiomerically mechanism of stereoselective interaction between butyryl-
enriched compounds, they became an extremely practical tool cholinesterase (BChE) and ethopropazine enantiomers was
for chemistry and biotechnology [14-21]. A vast number of performed leading to the final conclusion that BChE possesses a
publications have recently shown the enormous potential of significantly higher affinity for the (R)-configured etho-
enzymatic bioconversions in the synthesis of high value-added propazine [51]. In order to find whether ethopropazine displays
active ingredients and useful chiral building blocks for pharma- stereoselectivity in its pharmacokinetics and exhibits different
ceutical or fine-chemical industries [13,22-31]. Among the biological profiles of its stereoisomers, obtaining both enan-
enzymes applied, lipases occupy a prominent place in advanced tiomers in more than analytical quantities is particularly desir-
organic synthesis as biocatalysts of chiral recognition [32-39].
able. Moreover, the racemic mixture of this pharmaceutical is
commercially available, however it is sold at high price, as its
Nowadays, nervous and mental health problems mainly caused production by standard synthetic procedures is time-consuming
by the pace of life and the rapid development of civilization are and laborious. Therefore, a low-cost, practical, and scalable
very prevalent and dangerous diseases. An irregular and unsani- method for the straightforward preparation of ethopropazine is
tary lifestyle adversely affects the central nervous system still a great challenge.
resulting in neuroses and psychiatric disorders. Among the
antipsychotic medications phenothiazine derivatives [40,41], To the best of our knowledge, heretofore no reports concerning
and especially promethazine, widely known by its brand name the enzymatic preparative-scale synthesis of optically active
Phenergan®, hold well deserved place. As it posses excellent promethazine and ethopropazine have been reported. Only
antihistaminic activity (H1-antagonist) [42], pronounced enantioselective analytical methods towards enantioresolution
antimuscarinic [43], anticholinergic [44], sedative, antiemetic of promethazine employing various chiral selectors including
and adrenergic blocking action as well as some local anesthetic proteins, cyclodextrines, modified crown ethers or macrocyclic
properties [45], promethazine is approved for the effective antibiotics have been proposed [52-56]. Other two reports
therapy of various disorders clearly diverging from the original [57,58] containing information about ethopropazine enan-
application. It has gained widespread usage in the treatment of tiomers separation (mainly via fractional crystallization or
allergic reactions, itching dermatitis, cough (when applied column chromatography splitting of diastereomeric dibenzoyl-
together with codeine or dextromethorphan), sleeplessness or tartaric acid salts) have been described. Despite the above-
symptoms of motion, morning and Ménière's sicknesses mentioned results, the availability of a scalable stereoselective
including nausea and vomiting. For over four decades it has procedure has, as far as we know, been deemed insufficient. We
been considered that both enantiomers of promethazine posses believe that the lipase-catalyzed resolution procedure presented
exactly the same therapeutic activities [46,47], and thus it was herein, will be an interesting alternative to these published
administered in clinical settings as racemate. However, recent methods, and will open a novel route worth considering in
findings have shown that the different physiological and phar- multigram-scale synthesis of both aforementioned pharmaceut-
macological effects of the two enantiomers of promethazine are icals.
a fact, what makes this drug returned to the laboratories. For
example, it has been determined that (+)-promethazine reduced Therefore, the aim of this study was to take advantage of the
the cytokine IL-6 production in histamine-stimulated cells to extraordinary properties of enzyme catalysis and develop a new
9
0% while (−)-promethazine induced only 50% reduction in method based on lipase-mediated kinetic resolution (KR) of
cytokine IL-6 production [48]. Moreover, Boland and 1-(10H-phenothiazin-10-yl)propan-2-ol enantiomers, which
McDonought [49,50] found that preferably the (+)-enantiomer combined with convenient chemical reactions could provide a
of promethazine is particularly effective in inhibiting the forma- simple and straightforward approach for the preparation of opti-
tion of bone resorbing cells (osteoclasts) thus providing a new cally active promethazine and ethopropazine molecules, res-
class of agents useful for preventing or even treating bone loss, pectively.
mainly associated with periodontitis and osteoporosis. More-
over, kinetic resolution of the enantiomers of the key intermedi- Results and Discussion
ate 1-(10H-phenothiazin-10-yl)propan-2-ol may simultane- Herein, we wish to present an original chemoenzymatic proce-
ously provide access to another valuable compound in enan- dure for the enantioselective synthesis of promethazine 9 and
3039

Beilstein J. Org. Chem. 2014, 10, 3038–3055.
ethopropazine 10. Most of our efforts during these studies have lectively opened by phenothiazine (1) in the presence of
been focused toward extensive screening of the conditions for n-butyllithium (n-BuLi) at ambient temperature providing
the lipase-catalyzed kinetic resolution of 1-(10H-phenothiazin- desired alcohol (±)-3 in 64–77% yield depending on the applied
1
0-yl)propan-2-ol racemate (±)-3. The other part of the work scale and the purification method (see Supporting Information
involved investigation of the enzymatic reactions stereoselec- File 1). Paradoxically, the product was obtained in higher yield
tivity, which was accomplished by means of the assignment of when the reaction was performed on a bigger scale, and when
the absolute configuration of the resolved enantiopure alcohol the product was isolated by vacuum distillation instead of
(
S)-(+)-5 determined by a modified Mosher’s methodology, and column chromatography (SiO2). It is worth to note, that the
was unambiguously confirmed by X-ray diffraction analysis. preparation of (±)-3 was very problematic. Before we decided
Optically active intermediates (S)-(+)-5 and (R)-(−)-7 achieved to repeat the reaction procedure proposed by Clement, we
in this manner were subsequently transformed into active attempted to obtain (±)-3 by adopting the original methodology
pharmaceuticals 9 and 10 as (R)- and (S)-enantiomers afforded reported by Dahlbom [60]. Unfortunately, quite surprisingly the
in high enantiopurity (84–98% ee) (Scheme 1) and in a stereodi- reaction conducted by this approach, which assumed usage of
vergent fashion.
sodium amide (NaNH2) as a base, produced enormous amounts
of byproducts, which significantly hampered both the isolation
and purification procedures. The same phenomenon was
observed if sodium hydride (NaH) was utilized instead of
NaNH2 under similar reaction conditions, what suggests that
Synthesis of the racemic 1-(10H-pheno-
thiazin-10-yl)propan-2-ol (±)-3 and its acyl
esters (±)-4a–c
In the first step, racemic 1-(10H-phenothiazin-10-yl)propan-2- this particular process is most probably inaccessible through the
ol (±)-3 was synthesized according to the method described by sodium salt of phenothiazine (1) due to multiple substitution at
Clement et al. [59], in which propylene oxide (2) was regiose- aromatic ring carbon atoms and some other byproducts of the
Scheme 1: Chemoenzymatic synthesis of enantioenriched enantiomers of promethazine 9 and ethopropazine 10. Reagents and conditions: (i) n-BuLi
(1.5 equiv), dry THF, 1 h at −78 °C, then propylene oxide 2 (2 equiv), 12 h at rt; (ii) CH3COCl (1.5 equiv) or C3H7COCl (1.5 equiv) or C9H19COCl
(1.5 equiv), NEt3 (1.5 equiv), DMAP (0.1 equiv), dry CH2Cl2, 12 h at rt; (iii) vinyl ester (3 equiv), lipase [20% (w/w)], MTBE, 25 °C, 500 rpm (magnetic
stirrer); (iv) NaOH (1.1 equiv), MeOH, 1 h at rt; (v) PBr3 (1 equiv), CH2Cl2, 2 h at rt; (vi) Me2NH (5 equiv) or Et2NH (20 equiv), MeOH, sealed tube, 4 d
at 90 °C (temperature of oil bath); (vii) Me2NH (25 equiv) or Et2NH (50 equiv), PhCH3, sealed tube, 7 d at 140 °C (temperature of oil bath).
3040

Beilstein J. Org. Chem. 2014, 10, 3038–3055.
non-regioselective propylene oxide ring opening. With the hope Lipase-catalyzed kinetic resolution of (±)-3
of eliminating the unwanted side-product formation, we thereby To address the challenges associated with the determination of
performed the reaction under phase transfer catalysis (PTC) the best reaction conditions for the asymmetric chemoenzy-
conditions. The results of this experiments have shown that matic total synthesis of promethazine (9) and ethopropazine
independently from the type of the used reaction media (10), necessary optimization studies of the lipase-catalyzed
(
toluene, CH2Cl2, diethyl ether), bases (50% NaOH or 60% kinetic resolution of (±)-3 were undertaken. For that purpose,
KOH), and the applied PTC-catalysts [tetrabutylammonium we considered the influence of major factors including: (i) type
bromide (TBAB) or tetrabutylammonium hydrogensulfate of the lipase, (ii) choice of the co-solvent, (iii) reaction time,
(
TBAHS)], the epoxide 2 ring opening proceeded unsuccess- and (iv) kind of acyl donor on the yield and enantioselectivity
fully. Again, the amount of the formed impurities was too large outcome of the overall process. For preliminary analysis, we
to isolate pure fraction. This provoked us to change the syn- followed the strategy of keeping all the other experimental para-
thetic strategy by excluding at first propylene oxide (2) as the meters constant, i.e., stirring speed (500 rpm), enzyme loading
reagent. Reported in the literature various PTC-mediated alkyla- in respect to substrate (20% w/w), substrate-to-acyl group donor
tions of phenothiazine (1) with various alkyl and alkenyl halides molar ratio (1:30), and the temperature (25 °C). Finally, after
[
61-64], inspired us to prepare the 10-(prop-2-en-1-yl)-10H- finding optimal biotransformation conditions in a mg-scale, we
phenothiazine derivative as potent substrate for the synthesis of have successfully performed the reaction in gram- and multi-
±)-3. This was smoothly achieved in high yield (76%) by using gram-scales providing sufficient amounts of optically active
(
allyl bromide in biphasic system composed of CH2Cl2/50% alcohol intermediates (S)-(+)-5 and (R)-(−)-7 enabling to
NaOH and the additive TBAHS as the catalyst at room tempera- continue the planned synthesis. Each of the reaction parameters
ture. The obtained 10-(prop-2-en-1-yl)-10H-phenothiazine was and up-scaling approaches are discussed in detail in the
subsequently conducted into an oxymercuration–demercuration following paragraphs.
reaction accordingly to the procedure proposed by Abreu et al.
Biocatalyst effect on the kinetic resolution of
[
65]. Although the recorded mass spectra (HRMS–ESI) for the
thus prepared product (±)-3 showed the quasi-molecular peak (±)-3
corresponding to the calculated value, and both the gas chroma- The enzymatic kinetic resolution of (±)-3 was initially
tography (GC) as well as thin layer chromatography (TLC) attempted using a representative set of 14 different lipases
showed a single peak and spot respectively, the signals of the isolated from various microorganisms (see Supporting Informa-
measured 1H NMR spectra were strongly diffused and could not tion File 1 for details). Among the broad panel of investigated
be easily interpreted. This can be explained by the presence of commercially available preparations, immobilized lipases of a
traces of diamagnetic impurities (Hg2+-complex) in the sample fungal origin such as Novozym 435, Chirazyme L-2, C2,
of (±)-3, which caused severe broadening of the signals due to Chirazyme L-2, C3, and Lipozyme TL IM were established as
reduction of relaxation times. This prevents (±)-3 to be accom- optimal biocatalysts for the enantioselective transesterification
plished in the required purity for pharmaceuticals production. when suspended in 30 equiv of vinyl acetate as the acyl donor,
All other attempts performed by us, which concerned N-alkyl- methyl tert-butyl ether (MTBE) as co-solvent, and carried out at
ation of 1 with various alkyl agents such as: 1-chloropropan-2- 25 °C, with agitation speed of the magnetic stirrer arranged at
ol, (1-chloropropan-2-yloxy)trimethylsilane or chloroacetone 500 rpm, and an enzyme loading of 20% w/w in respect to sub-
under PTC or NaH-base conditions, failed as well leading to strate (±)-3.
complex mixtures. The above mentioned drawbacks were
finally overcome by using the lithium salt of phenothiazine 1 Whereas preliminary experiments showed that four of the
for the oxirane 2 ring opening as it was described at the begin- above-mentioned lipase preparations exhibited good enzymatic
ning of this paragraph.
activities and excellent enantioselectivities towards racemic
alcohol (±)-3, it would have been insufficient to select the most
To obtain racemic esters (±)-4a–c which are required for robust promising catalyst only on the basis of single random results
analytical HPLC separation studies of the corresponding pairs obtained from the analysis of the reaction mixtures. By virtue of
of enantiomers and for the measurement of the enantiomeric the importance of enzyme selection, and to find the most suit-
excess values of the compounds prepared in the biocatalyzed able catalyst for the transesterification of racemic alcohol (±)-3
reactions, the afore-prepared alcohol (±)-3 was reacted with the as well as to provide better insight into the reaction progress of
appropriate acyl chloride in dry dichloromethane in the pres- each of the chosen lipases in general, comprehensive kinetic
ence of triethylamine and a catalytic amount of 4-N,N-dimethyl- studies were assessed (Figure 1). Effect of the reaction time on
aminopyridine (DMAP). This was leading to moderate yields the conversion degree of phenothiazinic alcohol (±)-3 and the
(
41–68%) of the acetates.
overall progress of the followed enzymatic process was esti-
3041

Beilstein J. Org. Chem. 2014, 10, 3038–3055.
Figure 1: Dependence of optical purities (% ee) of (R)-(−)-6a (red curve, ■) and (S)-(+)-5 (blue curve, ▲) on the conversion degree of (±)-3 during
lipase-catalyzed acetylation with vinyl acetate in a MTBE solution at 25 °C (magnetic stirrer, at 500 rpm).
mated by high-performance liquid chromatography (HPLC) nitely stopped at 49% conversion being reached after two
analysis using a Chiralcel OD-H column. The products could be weeks. The specificity of this lipase for the (R)-alcohol was
analyzed directly from the crude mixture since the conditions of evident since even after such a long reaction time, no trace of
the HPLC chiral stationary phases were adjusted to the needs of the (S)-configured ester was detected. As occurred at elevated
both resolution products (peaks of the alcohol and the acetate temperature (35 °C), the reaction did not proceed further with
enantiomers were well separated) (see the Supporting Informa- longer reaction times. Nevertheless, the results obtained with
tion File 1 for details). The plots presented in Figure 1 clearly this lipase were truly impressive, as all the investigated samples
indicate that in terms of enantioselectivity, the best results were revealed that generated ester (R)-(−)-6a was enantiomerically
obtained with Lipozyme TL IM with an enantiomeric ratio pure (>99% ee) regardless of the time and the conversion.
factor (E) reaching up to 790 for 49% conversion achieved after
1
4 days, whereas other lipase preparations generally displayed In turn, the reaction rate was considerably improved in
lower enantioselectivities (with E-values up to 180).
Novozym 435-catalyzed analytical scale acetylation. This
revealed that Candida antarctica lipase B (CAL-B) immobi-
While tracing the course of the Lipozyme TL IM-catalyzed lized on acrylic resin was superior to other enzymes in both the
reaction it became evident that in most cases substrate (±)-3 was reaction rate and the obtained enantiomeric excess values of the
slowly but ultra-selectively converted into the corresponding slower reacting enantiomer (S)-(+)-5. In this case, racemic
acetate (R)-(−)-6a obtained with excellent enantiomeric excess alcohol (±)-3 has been successfully resolved affording the unre-
(
>99% ee), while the remaining alcohol (S)-(+)-5 was recov- acted substrate (S)-(+)-5 in enantiopure form (>99% ee),
ered only with 96% ee due to the fact that the reaction defi- whereas the ester (R)-(−)-6a was yielded in high enantiomeric
3042

Beilstein J. Org. Chem. 2014, 10, 3038–3055.
excess (82–92% ee) when the reaction slightly exceeded 52% hydrophilic media (such as methanol, ethanol, acetone, acetoni-
conversion. As can be seen in Figure 1, the time necessary to trile etc.) distort the active conformation of the enzyme mole-
achieve the appropriate 51% conversion for Novozym 435- cule by destabilizing electrostatic interactions, and even strip
catalyzed reaction was diminished by 2.5 orders-of-magnitude off the essential water layer (water of hydration) leading to loss
when compared to both Chirazyme-type preparations. Even of the biocatalytic activity. In turn, more lipophilic organic
more significant (by 5 orders-of-magnitude) difference in the solvents positively interact with the enzyme protein enhancing
speed of the reactions was observed when comparing the time its secondary structure mainly by stabilization of the internal
necessary for ca. 48% conversion achievement in the case of hydrogen-bonding network in the molecule, thus improving the
Novozym 435 and Lipozyme TL IM lipases. On the other hand, reactivity and selectivity of organic substrate transformations.
Chirazyme L-2, C2 and Chirazyme L-2, C3 displayed very high
enantioselectivity (E = 149–180) yielding stereoselectively ace- Therefore, an appropriate medium selection in which a rela-
tylated derivative (R)-(−)-6a in high enantiomeric excess tively high solubility of the phenothiazinic substrate (±)-3 as
(
94–95% ee) and unreacted alcohol (S)-(+)-5 with 98% ee. well as a good enzymatic activity and selectivity are retained, is
Examination of both reaction progresses showed very character- being of particular significance for the overall success of the
istic trends in kinetics for this type of lipase-catalyzed reaction, process. Five different apolar organic solvents (varying in
where an almost linear increase in the enantiomeric excess log P, dielectric constant ε, dipolar moment μ) including:
during the formation of optically active alcohol (S)-(+)-5 could n-hexane, pentane, toluene, MTBE, and diethyl ether (Et2O)
be observed. However, although good selectivities were were investigated as the reaction media. Additionally, both
attained in the kinetic resolution of (±)-3 catalyzed by both lipase catalysts were also screened in neat vinyl acetate
Chirazyme preparations, the reaction rates were low and 51% (Table 1). The courses of all the enzymatic reactions were
conversion was achieved after 166 h hampering the develop- monitored by gas chromatography (GC), arrested deliberately
ment of a straightforward route to optically active targeted after 7 days, and only then their final evolution were subse-
pharmaceuticals 9 and 10. The afore-outlined results forced us quently followed by chiral HPLC.
to cease further optimization studies with these particular
lipases. The enzyme selection stage leads to the conclusion that As a first attempt, Novozym 435-catalyzed KR of racemic
reactions catalyzed by Novozym 435 were beneficial for alcohol (±)-3 was carried out (Table 1, entries 1–6). After series
obtaining the alcohol (S)-(+)-5 in enantiopure form, while of experiments, the influence of various co-solvents could be
highly optically pure acetate (R)-(−)-6a was obtained with summarized as the following row orders in terms of enzyme
Lipozyme TL IM as the catalyst. These findings provoked us to activity: pentane > n-hexane ≈ MTBE ≈ Et2O > toluene ≈ vinyl
investigate the kinetic resolution of (±)-3 by using both acetate, and the reaction enantioselectivity: toluene > vinyl
Novozym 435 and Lipozyme TL IM preparations in the next acetate > n-hexane > MTBE ≈ Et2O > pentane. As it was
optimization stages.
observed, the acetylation reaction in toluene was mostly
enantioselective (E = 180), thus transforming the racemic
starting material (±)-3 into optically active ester (R)-(−)-6a of
Solvent effect on kinetic resolution of (±)-3
The next stage of enzymatic analytical scale studies was high enantiomeric excess (95% ee) and leaving thereby slower
designed to find the most suitable co-solvent system for the reacting enantiomer (S)-(+)-5 of very high enantiomeric purity
enantioselective O-acylation of racemic alcohol (±)-3 using (98% ee). On the other hand, from the view point of remaining
vinyl acetate. All experiments of the kinetic resolution of enan- the enatiopurity of alcohol (S)-(+)-5, ethereal solvents (Et2O
tiomers (±)-3 were conducted under the same reaction condi- and MTBE) which yielded (S)-(+)-5 with excellent enan-
tions as described above, using immobilized lipases: Novozym tiomeric excess (>99% ee) are even better. It is obvious that
4
35 and Lipozyme TL IM as the catalysts, respectively.
since methyl tert-butyl ether is free from dangerous peroxides,
hence it was chosen for further optimization studies concerning
It is well-known that the nature of solvent could significantly Novozym 435 lipase.
influence the activity and selectivity of the enzymatic reaction
as well as increase the thermal stability of enzymes [66]. The In turn, while investigating the acylation of (±)-3 catalyzed by
high sensitivity of lipases towards the medium environment Lipozyme TL IM suspended in chosen co-solvents systems
mainly stems from experimental observations that different (Table 1, entries 7–12) we noticed the following row order of
organic solvents had a different ability to interact with amino the reaction velocity: MTBE > pentane > Et2O > toluene >
acid residues which are responsible for the catalytic activity. vinyl acetate > n-hexane. Quite surprisingly, the same order was
For example, it is well documented [67-69] that polar solvents observed for the reaction enantioselectivity. As can be seen,
(
log P < 1) are less suitable for biocatalytic purposes since among the investigated apolar solvents for Lipozyme TL IM-
3043

Beilstein J. Org. Chem. 2014, 10, 3038–3055.
Table 1: Solvent influence on the selective O-acetylation of (±)-3 in the presence of lipase preparation (Novozym 435 or Lipozyme TL IM) with vinyl
acetate in MTBE at 25 °C after 7 days.
Entry
Enzymea
Solvent (log P)b
conv.c (%)
eesd (%)
eepd (%)
Ee
1
2
3
4
5
6
Novozym 435
hexane (3.00)
pentane (2.58)
toluene (2.52)
MTBE (0.96)
53
54
51
53
53
51
>99
>99
98
89
84
95
88
88
94
90
59
180
82
>99
>99
97
Et2O (0.76)
82
vinyl acetate (0.54)
136
7
8
9
Lipozyme TL IM
hexane (3.00)
pentane (2.58)
toluene (2.52)
MTBE (0.96)
18
41
29
44
37
28
22
70
40
78
58
39
98
123
419
295
474
359
292
>99
99
10
11
12
>99
>99
99
Et2O (0.76)
vinyl acetate (0.54)
aConditions: (±)-3 100 mg, lipase 20 mg, organic solvent 1 mL, vinyl acetate 948 mg (1 mL, 28 equiv), 25 °C, 500 rpm (magnetic stirrer). bLogarithm
of the partition coefficient of a given solvent between n-octanol and water according to ChemBioDraw Ultra 13.0 software indications. cBased on GC,
for confirmation the % conversion was calculated from the enantiomeric excess of the unreacted alcohol (ees) and the product (eep) according to the
formula conv. = ees/(ees + eep). dDetermined by chiral HPLC analysis by using a Chiralcel OD-H column. eCalculated according to Chen et al. [70],
using the equation: E = {ln[(1 – conv.)(1 – ees)]}/{ln[(1 – conv.)(1 + ees)]}.
catalyzed kinetic resolution of (±)-3, the best results were alcohol instantly and irreversibly tautomerizes to acetaldehyde,
achieved in MTBE both in terms of enzyme activity and reac- thus shifting the state of equilibrium in product direction.
tion enantioselectivity.
Therefore, in next experiments three different vinyl esters were
used: vinyl acetate, vinyl butanoate, and vinyl decanoate. The
influence of the acyl donor nature on the course of the alcohol
acylation reaction was examined using 3 equiv of the appro-
The effect of the acyl donor on the kinetic
resolution of (±)-3
A considerable number of fundamental researches have been priate vinyl ester under catalysis of the corresponding lipase
performed to explain the influence of the acyl group donor type (Novozym 435 or Lipozyme TL IM) suspended in MTBE as
on the lipase-catalyzed acylation of alcohols. Amongst the wide solvent, and stirred at 25 °C using a magnetic stirrer. All the
range of this reagent type, enol esters (i.e., vinyl acetate, assays were regularly traced by HPLC analyses after 1, 2, 4 and
isopropenyl acetate), activated esters (i.e., trifluoroethyl 7 days, respectively (Table 2). Consideration of the results
butyrate, S-ethyl thiooctanoate), non-activated esters (i.e., ethyl summarized in Table 2 leads to the conclusion that the highest
acetate, methyl acetate), and anhydrides (i.e., acetic acid anhy- enantioselectivity (E = 844) of the reaction has been observed in
dride, succinic acid anhydride) are most commonly used. the acylation of racemic alcohol (±)-3 in the presence of
Unfortunately, there are several drawbacks of incorporating the Lipozyme TL IM, and vinyl acetate as an acyl transfer reagent.
above-mentioned reagents into transesterification processes, In accordance with the previous experiments, this extremely
i.e., trifluoroethyl esters are very expensive, S-ethyl thioesters enantioselective lipase preparation afforded recovery of ester
generate undesirable flavor of the formed thiols, employment of (R)-(−)-6a with perfect optical purity (>99% ee) at a conver-
anhydrides yields free carboxylic acids causing enzyme deacti- sion in the rage of 39–49% (Table 2, entries 4–7). On the other
vation, and finally application of conventional esters, despite hand, since the Lipozyme TL IM-catalyzed acetylation tends to
their great availability and low price, could reverse acylation cease before achieving 50% conversion, the remaining alcohol
into alcoholysis due to alcohol formation during the course of (S)-(+)-5 could be obtained only with a moderate enantiomeric
reaction. Of course, removal of alcohol from the reaction excess (63–97% ee) reaching the highest optical purity (97%
system by, i.e., 4 Å molecular sieves could prevent lipase- ee) after 7 days (Table 2, entry 7). In turn, KR of (±)-3
catalyzed alcoholysis of the formed ester and thus shift the reac- conducted with vinyl acetate in the presence of Novozym 435
tion toward the product. However, a straightforward method to showed that after 24 h the enantiomeric purity of (S)-(+)-5
avoid the formation of alcohol molecules is available, and increased over 77% ee and reached the 98% ee after 48 h
assumes the use of vinyl esters as the acyl group donors. More- (Table 2, entry 1). Complete optical purification (>99% ee) of
over, from the application point of view, vinyl esters are signifi- the remaining alcohol (S)-(+)-5 was observed after a reaction
cantly practical reagents since in situ generated unstable vinyl time of 4 days (Table 2, entry 4). As can be also seen from data
3044

Beilstein J. Org. Chem. 2014, 10, 3038–3055.
Table 2: The influence of various acyl donors (3 equiv) upon the selectivity of Novozym 435 and Lipozyme TL IM for the O-acetylation of (±)-3 in
MTBE at 25 °C.
Entry
Enzymea
Acyl donor
t (d)
conv.b (%)
eesc (%)
eepc (%)
Ed
1
2
3
4
5
6
7
Novozym 435
vinyl acetate
1
2
4
1
2
4
7
44
51
52
39
45
48
49
77
98
>99
63
80
91
97
98
232
97
91
90
99
Lipozyme TL IM
>99
>99
>99
>99
382
492
637
844
8
9
Novozym 435
vinyl butanoate
1
2
4
1
2
4
51
53
57
49
50
51
>99
>99
>99
95
94
88
76
>99
98
94
170
82
10
11
12
13
37
Lipozyme TL IM
747
525
170
>99
>99
14
15
16
17
18
19
Novozym 435
vinyl decanoate
1
2
4
1
2
4
48
49
52
48
50
50
91
98
94
91
99
98
98
317
713
111
684
525
525
>99
>99
93
Lipozyme TL IM
>99
>99
aConditions: (±)-3 100 mg, lipase 20 mg, MTBE 2 mL, vinyl ester (3 equiv), 25 °C, 500 rpm (magnetic stirrer). b, c, dSee notes c, d and e above
Table 1).
(
presented in Table 2, both the rate and enantioselectivity of the lent enantiomer separation with 3 equiv of vinyl butanoate
enzymatic reactions could be significantly improved by using (Table 2, entries 11 and 12). However, enzymatic kinetic reso-
fatty acid vinyl esters. After analyzing the effect of those acyl lution with E = 747 has to be stopped at ca. 49% conversion in
group donors general findings can be presented: (i) (S)-(+)-5 order to obtain (R)-(−)-6b in fully enantiopure form (>99% ee),
alcohol with excellent enantiomeric excess (>99% ee) could be and somewhat over 50% conversion in order to acquire the
obtained after a relatively short time of 24 h when vinyl unreacted (S)-(+)-5 at >99% ee and superb enantioselectivity
butanoate was used, and the reaction was catalyzed by (E = 525). This observation hints that careful monitoring of the
Novozym 435 (Table 2, entry 8); (ii) in turn, the highest enan- reaction and its appropriate termination allows to receive both
tiopurity of the formed (R)-enantiomer could be attained in the resolution products (S)-(+)-5 and (R)-(−)-6b or (R)-(−)-6c with
enzymatic transesterification proceeded with Lipozyme TL IM very high enantiopurity (98–100% ee) what may facilitate
and vinyl butanoate as an acyl donor, appropriately arrested preparatory-scale synthesis. However, taking into account the
after 24 h, while 49% conversion was reached (Table 2, entry clear advantages of vinyl acetate such as (i) the relatively high
1
1); (iii) another characteristic tendency was observed in the volatility (hence possibility of easy recovery of the product and
case of the velocity of Novozym 435-catalyzed reactions, in remaining substrate), and (ii) lower cost when compared to the
which the influence of the acyl group donor was pronounced employed long-chain fatty acid vinyl esters, as well as the fact
more markedly than in the case of Lipozyme TL IM. For that (iii) vinyl acetate displayed acceptable impact on the
example, when the reaction was carried out with vinyl performance of the lipases, we reasoned to set it as the reagent
butanoate, Novozym 435 allowed 51% conversion to be of choice for further up-scaling investigations using both
achieved after remarkably 4-fold shorter reaction time heretofore examined enzymes.
compared with those conducted with vinyl acetate and vinyl
Preparative-scale lipase-catalyzed kinetic
decanoate (Table 2, Entry 8 vs 3 and 16), respectively.
resolution of (±)-3
As a conclusion, the enantioselectivity and rate of lipase- An optimized protocol for obtaining enantiopure key intermedi-
catalyzed KR of (±)-3 in MTBE were significantly affected by ates (S)-(+)-5 and (R)-(−)-6a–c on a milligram-scale (100 mg)
the nature of an acyl transfer reagent, thus allowing the excel- was successfully achieved. In order to provide further insight
3045

Beilstein J. Org. Chem. 2014, 10, 3038–3055.
into enzymatic kinetic resolution of (±)-3 enantiomers, and to excellent enantiomeric excess (>99%) reached when the reac-
show its potential as a method for asymmetric synthesis of tion was arrested at 46% conversion after 3 days (Table 3, entry
enantiomerically pure pharmaceuticals 9 and 10, this part of our 4). From the same reaction mixture unreacted (S)-(+)-5 was
study was aimed for investigating the feasibility of proceeding isolated with 84% ee. Independent experiment conducted with
the enantioselective transesterification on a significantly larger this lipase using 3 g of starting material (Table 3, entry 5)
scale. Bearing in mind that the fortune of this issue is critical for closely matched the above mentioned results showing that this
the development of a highly economic process at industrial approach effected the resolution of (±)-3 into (S)-(+)-5 alcohol
scale, we decided to examine respective 10-fold, 20-fold and (99% ee) and (R)-(−)-6a ester (86% ee) with a conversion value
3
0-fold linear enlargement of all the previously set parameters of 46% after 3 days. In the light of these findings, it became
including the reagents concentrations, enzyme quantities and clear that this process proved to be flexible enough for being
solvent volume under the optimized reaction conditions. up-scaled in a straightforward manner. The amount of both
Table 3 demonstrates the data of the lipase-catalyzed asym- enantiomers (S)-(+)-5 and (R)-(−)-7 obtained in the devised
metric O-acetylation of (±)-3 in one-, two-, and three-gram- lipase-catalyzed resolution of (±)-3 can be considered to be “in
scale regarding their final effect on the total stereochemical preparative scale” and ready-to-use for further transformations.
outcome for the employed lipase preparations. We were pleased
Determination of the stereochemistry of
to see that the results were almost the same, regardless of the
scale with, i.e., approx. 51% conversion after 4 days for the alcohol (+)-5
Novozym 435-mediated KR of racemic alcohol (±)-3 (Table 2, The determination of the absolute configuration is crucial for all
entry 1 vs Table 3, entry 1). The kinetic resolution in the pres- chiral molecules, but it is especially important for active phar-
ence of this enzyme provided slower reacting enantiomer maceutical ingredients (APIs) like those presented herein,
(
S)-(+)-5 with perfect optical purity (>99% ee), and its acetyla- whereas only one enantiomer has a high positive pharmaceu-
ted counterpart (R)-(−)-6a in enantioenriched form (94% ee). tical effect, while its counterpart is significantly less active. The
The separation of the resolution products on a silica gel column assignment of stereochemistry of one of the kinetically resolved
chromatography revealed that the unreacted alcohol (S)-(+)-5 products on this stage could not only serve the information
and the formed acetate (R)-(−)-6a could be prepared with about the catalytic behavior of the enzymes but also could help
isolated yield in the range of 81–95%, corresponding closely to us to plan further steps of the synthesis. Therefore, the stereo-
the theoretical 50% proportions of the enantiomers in the race- preference of the above-studied lipase preparations was evalu-
mate. Subsequent scaling of the reaction up to 2 and then up to ated by means of modified Mosher’s methodology described by
3
grams of the substrate (±)-3 gave almost the same isolated Riguera et al. [71] In this relatively simple and widely-used
yields of recovered alcohol (S)-(+)-5 and the acetate (R)-(−)-6a experimental approach, the absolute configuration of the chiral
with moderate to excellent enantiomeric excess for both com- substrate of unknown stereochemistry [in this case slower
pounds (69–100% ee) (Table 3, entries 2 and 3), respectively. reacting alcohol (+)-5 with an absolute enantiomeric purity
Again, a very similar chirality inducement occurred under (>99% ee)] was assigned by its independent reaction with the
Lipozyme TL IM- catalyzed conditions. In this manner, acetate two enantiomers of an appropriate chiral derivatizing agent
(
R)-(−)-6a was isolated chromatographically in 67% yield and (CDA) followed by comparison of the 1H NMR spectra of the
Table 3: Gram-scale enantioselective O-acetylation of (±)-3 under kinetically controlled conditions with vinyl acetate in MTBE at 25 °C.
Entry
Enzyme
t (d)
conv.a(%)
eesb (%)/Yieldc (%)/[α]Dd
eepb (%)/Yieldc (%)/[α]Dd
Ee
1
2
3
4
5
Novozym 435f
Novozym 435g
Novozym 435h
Lipozyme TL IMf
Lipozyme TL IMh
4
3
4
3
3
51
57
59
46
46
>99/95/+39.78 (c 1.02)
>99/83/+35.00 (c 1.10)
>99/86/N.D.i
94/81/-8.75 (c 1.03)
75/86/-7.43 (c 1.01)
69/N.D.i/N.D.i
170
50
27
84/72/+31.88 (c 1.49)
86/94/+29.56 (c 1.01)
>99/67/−8.30 (c 1.44)
99/92/−8.29 (c 0.91)
532
556
aThe % conversion was calculated from the enantiomeric excess of the unreacted alcohol (ees) and the product (eep) according to the formula
conv. = ees/(ees + eep). bDetermined by chiral HPLC analysis by using a Chiralcel OD-H column. cIsolated yield after column chromatography (calcu-
lated on the basis of the theoretical number of moles arising from conversion rate, relative to theoretical amount, i.e., when 50% conversion is
reached, up to the half of the acetate could be obtained). dSpecific rotation, c solution in methylene chloride, T = 22–23 °C, respectively (see
Supporting Information File 1 for details). eCalculated according to Chen et al. [70], using the equation: E = {ln[(1 – conv.)(1 – ees)]}/{ln[(1 – conv.)(1 +
ees)]}. fConditions: (±)-3 1 g, lipase 0.2 g, MTBE 10 mL, vinyl acetate 1 g (3 equiv), 25 °C, 500 rpm (magnetic stirrer). gConditions: (±)-3 2 g, lipase
0.4 g, MTBE 20 mL, vinyl acetate 2 g (3 equiv), 25 °C, 500 rpm (magnetic stirrer). hConditions: (±)-3 3 g, lipase 0.6 g, MTBE 30 mL, vinyl acetate 3 g
(3 equiv), 25 °C, 500 rpm (magnetic stirrer). iNot detected.
3046

Beilstein J. Org. Chem. 2014, 10, 3038–3055.
two diastereomeric derivatives obtained (11 and 12, Scheme 2). substrate system in the case of MPA esters of secondary alco-
As already exemplified by us in few other reports [72-75], the hols renders the sp rotamer, in which the methoxy group, the Cα
modified Mosher’s methodology is easy to carry out and to carbon, and the carbonyl group of the MPA unit as well as the
interpret, and can be successively applied toward secondary methine proton (CH) of the alcohol moiety [the substrate (+)-5
alcohols with various heterocyclic substituents with a minimum part] are all situated in the same plane, whereas the phenyl ring
of experimental work being invested as long as reliable CDA is is ca. perpendicular to the C=O bond and coplanar with the CαH
employed. In all of our previous studies α-methoxy-α-phenyl- bond. Both the methoxy and the carbonyl group are in a syn-
acetic acid (O-methylmandelic acid, MPA) used in both enan- periplanar disposition (see Newman projections of the appro-
tiomeric forms turned out to be an excellent chiral auxiliary priate (1R,2S)- and (1S,2S)-conformations in diastereomers 11
reagent for this type of derivatized compounds since it has and 12 depicted in Figure 3). (ii) The main induced anisotropic
given diastereomeric esters that were distinguishable by NMR magnetic field effect (shielding), which is generated by means
spectroscopy with reasonable levels of accuracy in terms of of high π electron density (present in the phenyl ring of the
determination of stereochemistry (Scheme 2).
chiral auxiliary regent MPA), showed appropriate strength
intensity) giving rise to perceptible shifts in the 1H NMR
(
Subsequently, the chemical shifts of the substituents directly signals of the L1/L2 groups located close to the chiral carbon in
bonded to the stereogenic centre of the investigated alcohol each of the investigated derivatives 11 and 12. (iii) The
(
(
+)-5 (L1/L2) were compared, and their spatial location 1H NMR signals of the chiral MPA shift reagent did not overlap
absolute configuration) was determined in accordance with the with those of the examined alcohol (+)-5 thus allowing easy
signs of their differences (ΔδRSL1 and ΔδRSL2) in both prepared interpretation of the recorded spectra.
MPA-derivatives 11 and 12 (Figure 2).
Comparing the recorded 1H NMR spectra of both CDAs deriva-
The chirality of asymmetric carbon in the alcohol (+)-5 could be tives 11 and 12 illustrated in Figure 3, the most remarkable
established due to the following facts: (i) It has been empiri- features should be outlined as follows. The 1H NMR data of
cally confirmed by Latypov et al. [76] (as a result of dynamic (R)-MPA ester 11 show a general shielding effect as a conse-
NMR studies) that a final prevalent conformation in the CDA- quence of the close proximity of the phenyl group, with impor-
Scheme 2: Assignment of the stereochemistry of enantiopure alcohol (+)-5 resulting from derivatization with (R)- and (S)-MPA 11 and 12, respective-
ly.
Figure 2: Description of substituents for determination of the absolute configuration of (+)-5 and ΔδRS values obtained for the MPA esters 11 and 12.
3047

Beilstein J. Org. Chem. 2014, 10, 3038–3055.
Figure 3: 1H NMR (CDCl3, 400 MHz) spectra of the (R)-MPA 11 (red colored line) and (S)-MPA and 12 (blue colored line) derivatives of enantiopure
alcohol (+)-5 (>99% ee). Additionally, with red color are marked protons shielded by the phenyl ring of chiral auxiliary, blue labels stand for unaffected
protons. Green arrows indicate the shielding effect caused by the aromatic system.
3048

Beilstein J. Org. Chem. 2014, 10, 3038–3055.
tant upfield shifts (lower frequency) for the signals assigned to step in optimizing the process to avoid racemization. To save
H(1’) hydrogen atoms.
one step less, we initiated our efforts by direct transformation of
the sec-alcohol functionality in afore-obtained compound (±)-3
In turn, this particular H(1’) signals in the spectrum for (S)- under modified Mitsunobu [78,79] reaction conditions using
MPA derivative 12 are observed as unaffected and thus a down- Ph3P (1.1 equiv), DEAD (1.1 equiv), and the respective amine
field shift of the resonance signal occurs. The opposite phenom- (1 equiv of Et2NH or Me2NH) conducted in dry THF. Since the
enon can be noticed when the protons of the methylene group Mitsunobu condensation protocol is considered as particularly
H(2’) in both MPA esters 11 and 12 are taken into account. In attractive due to the superior reactivity of the in situ-generated
this situation, the H(2’) protons of (S)-MPA derivative 12 are corresponding oxyphosphonium intermediate on nucleophilic
located under the shielding cone of the phenyl ring, which substitution, we envisioned that in this scenario the hydroxy
resulted in an upfield shift of the 1H NMR signals in compari- moiety would be inverted under mild reaction conditions and in
son to the signals from the (R)-MPA derivative 11.
a desired one-pot stereospecific manner. To validate this
hypothesis, the reaction was at first performed using model
The recognition of the absolute configuration of (+)-5 clearly racemic alcohol (±)-3, and in two variants of reagents addition
indicates that both Novozym 435 and Lipozyme TL IM lipases order. To our disappointment, none of these attempts led to
follow Kazlauskas’ rule [77], according to which the (R)-enan- success. When Mitsunobu chemistry fails, a hydroxy group
tiomer of the chiral substrate (±)-3 is preferentially acetylated.
inversion by tosylation and displacement using amine was
performed. Unfortunately, attempts to carry out the reaction at
In order to unambiguously confirm the adequacy of 1H NMR higher temperature led to complete decomposition of the tosy-
shift analysis, the single crystal of enantiomerically pure late substrate. In the forward direction to synthesize optically
alcohol (+)-5 (>99% ee) was subsequently analyzed by X-ray active pharmaceuticals 9 and 10, we decided to prepare both
diffraction (XRD) method (Figure 4), which confirmed (S)-con- enantiomers of corresponding bromide derivative 8. This was
figuration and allowed us to conclude that Mosher’s method- obtained by treatment of afore-prepared enantiopure alcohol
ology can serve as a reliable determination method of the (S)-(+)-5 with 1 equiv of phosphorus tribromide (PBr3) at room
absolute configuration toward this type of compounds. For temperature carried out for 2 h. This allows to access (R)-(+)-8
details concerning crystallization procedure and XRD measure- in >99% ee and in 27% yield. In turn, to afford its bromide
ments see Supporting Information File 1.
counterpart (S)-(−)-8 a two-step reaction sequence has to be
performed. At first, optically active acetate (R)-(−)-6a was
transformed into the corresponding alcohol (R)-(−)-7 through
simple cleavage of the acetyl group by means of NaOH-medi-
ated mathanolysis. This resulted in the formation of (R)-(−)-7 in
high 76% yield and almost without loss of the optical purity
(
(
98% ee). Secondly, thus-obtained (R)-(−)-7 was converted to
S)-(−)-8, but surprisingly with partial loss in enantiomeric
excess (96% ee) despite using the same procedure as for the
reaction between (S)-(+)-5 and PBr3. In addition, bromination
of (S)-(+)-5 under Appel reaction conditions employing tetra-
bromomethane (CBr4) as a halide ion source used along with
triphenylphosphine (Ph3P) was also investigated. Although this
approach proceeded to give (R)-(+)-8 in considerably higher
yield (77%), the obtained result was inferior from the stereo-
chemical point of view as the reaction was non-selective and led
to afford (R)-(+)-8 of poor 23% ee. With both enantiomers of
bromo derivative (R)-(+)-8 and (S)-(−)-8 in hand, we have
subsequently proceeded amination with Et2NH and Me2NH,
respectively. In all cases, a sealed glass vessel was used since
these reactions need to be operated mostly at elevated tempera-
Figure 4: An ORTEP plot of (S)-(+)-1-(10H-phenothiazin-10-yl)propan-
2-ol (S)-(+)-5. The following crystal structure has been deposited at the
Cambridge Crystallographic Data Centre and allocated the deposition
number CCDC-1018655.
Approaches to the preparation of both enan-
tiomeric forms of 9 and 10
Although, the remaining synthesis of the final products 9 and 10 tures, and low-boiling amines (bp. in the range of 54–60 °C)
appears deceptively straightforward, it actually turned out to be could be lost in the course of the process. The first set of reac-
a considerable challenge as the inversion of the chiral center in tions was performed by stirring racemic bromide (±)-8 in an
the intermediate molecules 3 and 8 has been identified as a key appropriate amine reagent for 24 h at room or elevated tempera-
3049

Beilstein J. Org. Chem. 2014, 10, 3038–3055.
ture (>60 °C). Unfortunately, all of these attempts failed to toward the product formation if (±)-10 precipitated over the
provide satisfactory results and led only to the recovery or the course of reaction in hydrobromide salt form. To our delight,
substantial decomposition of the starting material (±)-8. We when heating this reaction mixture at 110 °C for 96 h some
then turned our attention to the conventional protocol opti- progress has been observed. Therefore, to improve the effi-
mizing N-alkylation by screening several aprotic polar solvents ciency of this process, our next attempt involved the increase in
(
CH3CN, DMF, dioxane and CH2Cl2) at room temperature or at the operating temperature (up to 140 °C of an oil bath) as well
heating, respectively. Under most of the aforementioned condi- as elongation of the reaction time, what finally provided (±)-10
tions, (±)-8 proved to be reluctant to form respective tertiary in 38% yield after 7 days. An extension of this condition for the
amine products 9 or 10, and only in the case of substitution with reaction between (±)-8 and Me2NH gave even a better result in
Me2NH the reactions proceeded in both media (CH3CN and terms of the product (±)-9 yield (47%). However, when the
dioxane), however sluggishly and somewhat low-yielding typi- above-performed reactions were repeated with optically active
cally obtaining (±)-9 in less than 5%. In the course of investi- bromo derivatives (R)-(+)-8 (>99% ee) and (S)-(−)-8 (96% ee),
gating why (±)-8 failed to react with amines, we focused on it turned out that the respective products (S)-(−)-9 (84% ee) and
employing various basic conditions (EtN3, K2CO3, n-BuLi), (R)-(+)-9 (92% ee) or (S)-(−)-10 (90% ee) and (R)-(+)-10 (93%
under which activation of the N-nucleophile might increase the ee) were obtained with partial drop of the enantiomeric excess
reactivity and circumvent encountered drawbacks. Disappoint- (Scheme 3). These facts plus the modest yields (<50%) as well
ingly, treatment of (±)-8 with EtN3 showed no positive effect on as long reactions time (7 days) limited this method as being a
the reaction progress, and when K2CO3 or n-BuLi were applied preparative one, and forced us to search for an alternative
it led only to exhaustive decomposition of the starting material method.
(
±)-8 producing a complex mixture. To improve the efficiency
of the desired process, various catalytic conditions were subse-
quently evaluated. Initially, we decided to employ PTC-assisted
amination in the similar manner to those reported by O’Meara
[
80] and Durst et al. [81], respectively. A number of different
variants of these methods including use of N-benzyl-N,N,N-tri-
ethylammonium chloride [TEBA(Cl)] or dibenzo-18-crown-6
as the catalyst and a liquid–solid system comprising 1.6 equiv
of grounded NaOH suspended in DMSO or a liquid–liquid
systems composed of 50% NaOH or 60% KOH and the appro-
priate solvent (CH2Cl2, DMSO) stirred along with Et2NH at
room or slightly increased temperature (40 °C) have been
studied. However, none of the attempted changes led to an
improvement in the reaction progress. By the same token,
another catalytic approach using solid Ag2O was adopted in
accordance to a method proposed by D'Angeli et al. [82]. Since
the authors of this report suggested that solid Ag2O behaves as
both a Lewis acid and Brønsted base acting synergically on the
investigated 2-bromo amides, we have assumed that the exten-
sion of this concept on our reaction could activate substrate (±)-
Scheme 3: Amination of optically active bromo derivatives (R)-(+)-8 or
S)-(−)-8 in toluene.
(
8
and thus promote the desired N-alkylation. And this time,
application of Ag2O was not successful since this reaction has
been plagued by the formation of an intractable mixture of side Somewhat more surprisingly, the best results were obtained
products. The resistance of (±)-8 toward nucleophilic substitu- when the reaction was performed in MeOH similarly to the
tion in both studied catalytic systems (PTC- and Ag2O-medi- rivastigmine synthesis strategy published by Sethi and
ated approaches) may be explained by the fact that our sub- co-workers [83], but at higher temperature (90 °C of an oil bath)
strate (±)-8, in contrast to those described in the literature, lacks since at reported 25 °C the reaction with substrate (±)-8 did not
of the neighboring carbonyl group, which presumably favored occur. This modification led to obtain (±)-9 in yield of 38% and
the reported processes due to weakening of the carbon–bromine (±)-10 in yield of 71%, respectively. A simple change of the
bond. Next, we aimed to investigate the amination of (±)-8 solvent to methanol also drastically improved the rate as the
using Et2NH in toluene medium hoping that the employment of reactions needed at least 3 days less when compared to those
this aprotic and less polar solvent could shift the equilibrium conducted in toluene and what is crucial at lower temperature.
3050

Beilstein J. Org. Chem. 2014, 10, 3038–3055.
Moreover, under this conditions optically active substrate (R)-
+)-8 (>99% ee) underwent amination selectively almost
(
without loss of optical purity affording unexpectedly enantioen-
riched products (R)-(+)-9 (97% ee) and (R)-(+)-10 (98% ee)
with reverse stereochemistry at a chiral carbon. Compared to
the previously obtained products this seems to be more likely
than the net retention (double inversion) occurred during this
process. Their respective counterparts including (S)-(−)-9
(
94% ee) and (S)-(−)-10 (96% ee) were obviously obtained in
less enantiomerically pure forms as a consequence of preparing
them from less enantiopure substrate (S)-(−)-8 (96% ee)
(
Scheme 4).
The results of above described experiments can only be ratio-
nalized considering the formation of an intermediate aziri-
dinium ion (R)-13 upon heating in MeOH, which is then
attacked by amine at the more hindered carbon atom of the
aziridine ring. These observations allowed us to propose the
following plausible mechanism (Scheme 5), postulated mainly
on the basis of the signs of specific rotation and the enan-
Scheme 4: Amination of optically active bromo derivatives (R)-(+)-8 or
(S)-(−)-8 in methanol.
tiomeric elution order from chiral HPLC, which were opposite thiazine (S)-16 and 10-[(2S)-1-methoxypropan-2-yl]-10H-
to those corresponding to the appropriate products obtained phenothiazine (S)-17 as a result of a competitive reaction
when toluene was applied as a solvent (see Supporting Informa- between (R)-13 and methanol (Scheme 5) were isolated and
tion File 1). This phenomenon was also confirmed by isolation characterized as well.
of byproducts characteristic for this reaction mechanism
including (2S)-N,N-dimethyl-2-(10H-phenothiazin-10- Moreover, an unexpected regioselectivity upon ring opening by
yl)propan-1-amine (iso-promethazine) (S)-14 or (2S)-N,N- amines has been observed. Apparently, this aziridinium salt
diethyl-2-(10H-phenothiazin-10-yl)propan-1-amine (iso-etho- (R)-13 is opened exclusively at the more hindered aziridine
propazine) (S)-15 (depending on the amine used). What is even carbon atom since the respective products (S)-(−)-9 or (S)-(−)-
more compelling 10-[(2S)-2-methoxypropyl]-10H-pheno- 10 (23–46% yield) and (S)-16 (<25% yield) have been isolated
Scheme 5: The proposed reaction mechanism for amination of optically active (S)-(−)-8 in methanol.
3051

Beilstein J. Org. Chem. 2014, 10, 3038–3055.
in majority, while yield of (S)-17 and both unwanted isomers common NMR methodological approach, which assumes the
(
S)-14 or (S)-15 remained negligible (<5%). The outcome of the use of a lanthanide-like chiral shift reagent (CSR) such as
reaction is quite interesting since the conversion of, i.e., europium(III) tris[3-(heptafluoropropylhydroxymethylene)-(+)-
S)-bromide into the (S)-configurated product has proceeded camphorate [Eu(hfc)3] [84-86]. To our disappointment, the
(
with overall retention of configuration via regio- and stereospe- desirable enantiomeric shift difference magnitudes and
cific opening of the aziridinium ion (R)-13 at the activated enhanced spectral resolution for the studied (±)-10 were not
pseudo-benzylic position, where the partial positive charge (δ+) achieved. Under these particular circumstances we were forced
is obviously better stabilized than at the neighboring carbon to undertake another strategy, which would provide highly
atom. This clearly suggests that electrostatic factors (interac- accurate and rapid access to quantification of the enantiomeric
tions) are favored over steric hindrance effects giving the excess of optically active ethopropazine 10. Finally, the assess-
opportunity of a purely solvent-dependent stereodivergent syn- ment of enantiopurity of the synthesized (S)-(−)-10 and
thesis of therapeutically active ingredients 9 and 10 in a highly (R)-(+)-10 were determined by HPLC, but using two serially
stereoselective fashion.
connected Chiralcel OJ columns and mobile-phase system
composed of n-hexane/tert-BuOH/Et3N (98:1.5:0.5, v/v/v) at a
In view of this findings, it appeared that deterioration of the flow rate of 1.1 mL/min similarly to the method reported by
enantiomeric excess of the both APIs isomers [Δ% ee = 9–15 Šinko et al. [57].
for (R)-(+)-8 (>99% ee) and Δ% ee = 3–4 for (S)-(−)-8
(
96% ee)] synthesized in toluene may be related to the fact that Conclusion
this process proceeded at least in part through the formation of In summary, a straightforward chemoenzymatic preparation
activated aziridinium ion intermediate (S)-13 or (R)-13 (net strategy of both highly enantioenriched enantiomeric forms of
retention), and not only via direct attack of the amine on the promethazine 9 (up to 97% ee) and ethopropazine 10 (up to
appropriate bromo derivative (R)-(+)-8 or (S)-(−)-8 (single 98% ee), with the chirality being introduced by the action of
inversion). Toluene as an aprotic-lipophilic solvent is rather re- lipases has been devised. An extensive screening revealed both
sponsible for unfavorable conditions for the formation of aziri- Novozym 435 and Lipozyme TL IM lipase preparations to be
dinium salt due to ineffective stabilization of the cationic capable of asymmetric enantioselective transesterification of
species. Nevertheless some portions of the bromo derivative 8 racemic 1-(10H-phenothiazin-10-yl)propan-2-ol ((±)-3). Kinetic
most likely undergo intramolecular nucleophilic substitution by resolution of (±)-3 using the respective enzymes, and vinyl
means of the neighboring amine moiety present in the pheno- acetate as the acylating agent have provided an efficient access
thiazine ring, and by this affecting the optical purity of the prod- to highly enantioenriched alcohols (S)-(+)-5 (>99% ee) and
ucts.
(R)-(−)-7 (98% ee) accomplished in high isolated yields (up to
5%). Most of the critical enzymatic reaction factors such as
9
Determination of the enantiomeric purity of
medium, acyl group donor and operating time were investi-
gated in depth to establish the most optimal biotransformation
promethazine 9 and ethopropazine 10
In order to determine the enantiomeric excess of the final phar- conditions. The obtained results of enzymatic reactions
maceutical products 9 and 10, we initially examined the chiral conducted in small scale (100 mg) were scaled-up with high
HPLC conditions proposed by Ponder et al. [52]. Among correlation to a synthetically useful substrate amount of 3 g
various HPLC columns studied in that paper, we have owned (±)-3 (0.4 M concentration). Based on these findings, we can
the Chiralcel OJ with cellulose-derived chiral stationary phase speculate that the developed procedure can facilitate a potential
(
CSP) consisting of cellulose tris(4-methylbenzoate) as chiral commercial application in the production of title drugs 9 and 10
selector. Unfortunately, when using this column accordingly to with a good perspective in industrial scale since high conver-
Ponder’s indications we could only separate promethazine (±)-9 sions and excellent ee values can be achieved with low catalyst
base enantiomers (see Supporting Information File 1 for loadings. The 1H NMR spectroscopic assignment of the
details). Since ethopropazine (±)-10 enantiomers were not absolute configuration of the slower reacting enantiomer (+)-5
resolvable on a Chiralcel OJ column under various HPLC of enzymatic KR was performed by means of a modified
conditions performed in the normal phase mode, we have opted Mosher’s methodology, and was supported by single-crystal
to investigate several other polysaccharide type columns X-ray diffraction (XRD), which both led up to the final conclu-
including Chiralcel OD-H, Chiralcel OJ-H, Chiralcel OJ-RH sion that the employed lipase preparations exhibit (R)-stereo-
and Chiralpack IA. Unfortunately, all the studied chromato- preference towards the enantiomers of resolved alcohol (±)-3.
graphic analyses led to failure as no signs of enantiomeric sepa- The products of the enzymatic transformations were functional-
rations for (±)-10 were obtained. Therefore, we decided to ized en route to the title compounds via a two-step reaction
investigate enantiomeric separation conditions applying a sequence using the appropriate brominating agent (PBr3), and
3052

Beilstein J. Org. Chem. 2014, 10, 3038–3055.
5
6
7
.
.
.
Rasor, J. P.; Voss, E. Appl. Catal., A 2001, 221, 145–158.
subsequent amination of the thus-prepared halo derivative by
means of respective aliphatic amines. A systematic investi-
gation of the amine functionalization approach by examining
various reaction parameters including solvents, catalysts, bases,
additives, and operating temperatures revealed some unex-
pected results as the transformation of bromo derivative 8 is
highly dependent on the proper choice of the reaction medium
system. It turned out that in toluene solution a typical SN2 reac-
tion is observed (single inversion), whereas in methanol an
intramolecular cyclization occurred leading to aziridinium ion-
formation, which only then reacted with the amine giving the
product of double inversion. In the case of reactions carried out
in methanol only a negligible loss of the enantiomeric purity
doi:10.1016/S0926-860X(01)00804-3
Ran, N.; Zhao, L.; Chen, Z.; Tao, J. Green Chem. 2008, 10, 361–372.
doi:10.1039/b716045c
Rodrigues, R. C.; Fernandez-Lafuente, R. J. Mol. Catal. B: Enzym.
2010, 64, 1–22. doi:10.1016/j.molcatb.2010.02.003
8. Wenda, S.; Illner, S.; Mell, A.; Kragl, U. Green Chem. 2011, 13,
007–3047. doi:10.1039/c1gc15579b
3
9.
Muñoz Solano, D.; Hoyos, P.; Hernáiz, M. J.; Alcántara, A. R.;
Sánchez-Montero, J. M. Bioresour. Technol. 2012, 115, 196–207.
doi:10.1016/j.biortech.2011.11.131
1
0.Meyer, H.-P.; Eichhorn, E.; Hanlon, S.; Lütz, S.; Schürmann, M.;
Wohlgemuth, R.; Coppolecchia, R. Catal. Sci. Technol. 2013, 3, 29–40.
doi:10.1039/c2cy20350b
11.Schoffers, E.; Golebiowski, A.; Johnson, C. R. Tetrahedron 1996, 52,
3
769–3826. doi:10.1016/S0040-4020(95)01021-1
2.Faber, K., Ed. Biotransformations in Organic Chemistry: A Textbook,
th ed.; Springer Verlag: New York, NY, 2011.
(
94–97% ee for 9 and 96–98% ee for 10) was observed, while
1
the reactions conducted in toluene proceeded with higher level
of deterioration in terms of enantiomeric excess values
6
13.García-Urdiales, E.; Alfonso, I.; Gotor, V. Chem. Rev. 2011, 111,
(
84–92% ee for 9 and 90–93% ee for 10).
PR110–PR180. doi:10.1021/cr100330u
1
4.Wong, C.-H.; Whitesides, G. M., Eds. Enzymes in Synthetic Organic
Chemistry; Tetrahedron Organic Chemistry, Vol. 12; Pergamon:
Oxford, 1994.
Supporting Information
1
5.Drauz, K.; Waldmann, H., Eds. Enzyme Catalysis in Organic Synthesis;
VCH Verlagsgesellschaft mbH: Weinheim, 1995.
6.Sugai, T. Curr. Org. Chem. 1999, 3, 373–406.
7.Davis, B. G.; Boyer, V. Nat. Prod. Rep. 2001, 18, 618–640.
doi:10.1039/b003667f
Supporting Information File 1
1
1
Complete experimental procedures and characterization
data.
[http://www.beilstein-journals.org/bjoc/content/
1
8.Hudlicky, T.; Reed, J. W. Chem. Soc. Rev. 2009, 38, 3117–3132.
doi:10.1039/b901172m
supplementary/1860-5397-10-322-S1.pdf]
1
9.Borges, K. B.; de Souza Borges, W.; Durán-Patrón, R.;
Tallarico Pupo, M.; Sueli Bonato, P.; Gonzaléz Collado, I.
Tetrahedron: Asymmetry 2009, 20, 385–397.
Acknowledgements
doi:10.1016/j.tetasy.2009.02.009
The project was co-financed by The European Regional Devel-
opment Fund under The Innovative Economy Operational
Programme 2007–2013: ‘Biotransformations for Pharmaceu-
tical and Cosmetic Industry’ POIG.01.03.01-00-158/09. These
studies were partially supported by the Warsaw University of
Technology, Faculty of Chemistry. Paweł Borowiecki would
also like to gratefully acknowledge the financial support from
the project: ‘Development of science – the development of the
region – scholarship and accompanying support for Mazovia
2
0.Hollmann, F.; Arends, I. W. C. E.; Holtmann, D. Green Chem. 2011,
13, 2285–2314. doi:10.1039/c1gc15424a
21.Zhang, H.; Ren, J.; Wang, Y.; Sheng, C.; Wu, Q.; Diao, A.; Zhu, D.
Tetrahedron 2013, 69, 184–189. doi:10.1016/j.tet.2012.10.047
22.Schulze, B.; Wubbolts, M. G. Curr. Opin. Biotechnol. 1999, 10,
09–615. doi:10.1016/S0958-1669(99)00042-7
6
2
3.Liese, A.; Villela Filho, M. Curr. Opin. Biotechnol. 1999, 10, 595–603.
doi:10.1016/S0958-1669(99)00040-3
24.Panke, S.; Held, M.; Wubbolts, M. Curr. Opin. Biotechnol. 2004, 15,
272–279. doi:10.1016/j.copbio.2004.06.011
postgraduate students’ (Priority VIII Regional Economy; 25.Patel, R. N. Coord. Chem. Rev. 2008, 252, 659–701.
Human Resources Action 8.2 Knowledge Transfer; Inoppor-
tune 8.2.2 Regional Innovation Strategies, Operational Program
Human Capital 2007–2013, European Union fund).
doi:10.1016/j.ccr.2007.10.031
2
2
2
6.Tao, J.; Xu, J.-H. Curr. Opin. Chem. Biol. 2009, 13, 43–50.
doi:10.1016/j.cbpa.2009.01.018
7.Nestl, B. M.; Nebel, B. A.; Hauer, B. Curr. Opin. Chem. Biol. 2011, 15,
187–193. doi:10.1016/j.cbpa.2010.11.019
References
8.Clouthier, C. M.; Pelletier, J. N. Chem. Soc. Rev. 2012, 41, 1585–1605.
1.
2.
3.
4.
Carey, J. S.; Laffan, D.; Thomson, C.; Williams, M. T.
doi:10.1039/c2cs15286j
Org. Biomol. Chem. 2006, 4, 2337–2347. doi:10.1039/b602413k
Mentel, M.; Blankenfeldt, W.; Breinbauer, R. Angew. Chem., Int. Ed.
29.Simon, R. C.; Mutti, F. G.; Kroutil, W. Drug Discovery Today: Technol.
2013, 10, e37–e44. doi:10.1016/j.ddtec.2012.08.002
30.Huisman, G. W.; Collier, S. J. Curr. Opin. Chem. Biol. 2013, 17,
284–292. doi:10.1016/j.cbpa.2013.01.017
2009, 48, 9084–9087. doi:10.1002/anie.200902997
Kasprzyk-Hordern, B. Chem. Soc. Rev. 2010, 39, 4466–4503.
doi:10.1039/c000408c
31.Tokoshima, D.; Hanaya, K.; Shoji, M.; Sugai, T.
J. Mol. Catal. B: Enzym. 2013, 97, 95–99.
Liese, A.; Seelbach, K.; Wandrey, C., Eds. Industrial
Biotransformations, 2nd ed.; Wiley-VCH: Weinheim, 2006.
doi:10.1016/j.molcatb.2013.07.021
3053

Beilstein J. Org. Chem. 2014, 10, 3038–3055.
3
3
3
2.Itoh, T.; Takagi, Y.; Tsukube, H. J. Mol. Catal. B: Enzym. 1997, 3,
56.Martínez-Gómez, M. A.; Sagrado, S.; Villanueva-Camañas, R. M.;
Medina-Hernández, M. J. Anal. Chim. Acta 2007, 582, 223–228.
doi:10.1016/j.aca.2006.09.036
259–270. doi:10.1016/S1381-1177(97)00007-6
3.Gandhi, N. N. J. Am. Oil Chem. Soc. 1997, 74, 621–634.
doi:10.1007/s11746-997-0194-x
57.Šinko, G.; Novak, P.; Žiher, D.; Vinković, V.; Šunjić, V.;
Simeon-Rudolf, V. Enantiomer 2002, 7, 149–156.
4.Kazlauskas, R. J.; Bornscheuer, U. T. Biotransformation 1. In
Biotransformations with lipases, Biotechnology, 2nd ed.; Kelly, D. R.,
Ed.; Wiley-VCH: Weinheim, 1998; Vol. 8, pp 37–191.
.
doi:10.1080/10242430212188
58.Puthca, L.; McDonough, J.; Boland, E. J.; Dixon, H.; Persyn, J. T.;
Vasishtha, N. Compositions a liberation progressive et son procede
d'utilisation. WO Patent WO2003/105811 A2 , Dec 24, 2003.
59.Clement, B.; Beckett, A. H. Arch. Pharm. 1981, 314, 716–722.
doi:10.1002/ardp.19813140812
3
3
3
5.Jaeger, K.-E.; Dijkstra, B. W.; Reetz, M. T. Annu. Rev. Microbiol. 1999,
53, 315–351. doi:10.1146/annurev.micro.53.1.315
6.Jaeger, K.-E.; Eggert, T. Curr. Opin. Biotechnol. 2002, 13, 390–397.
doi:10.1016/S0958-1669(02)00341-5
60.Dahlbom, R. Acta Chem. Scand. 1949, 3, 247–255.
doi:10.3891/acta.chem.scand.03-0247
7.Lambusta, D.; Nicolosi, G.; Patti, A.; Sanfilippo, C.
J. Mol. Catal. B: Enzym. 2003, 22, 271–277.
61.Gozlan, I.; Ladkani, D.; Halpern, M.; Rabinovitz, M.; Avnir, D.
J. Heterocycl. Chem. 1984, 21, 613–614. doi:10.1002/jhet.5570210271
62.Gomurashvili, Z.; Crivello, J. V. Macromolecules 2002, 35, 2962–2969.
doi:10.1021/ma0119272
doi:10.1016/S1381-1177(03)00042-0
3
8.Ghanem, A. Tetrahedron 2007, 63, 1721–1754.
doi:10.1016/j.tet.2006.09.110
3
9.Section C Lipases. In Industrial Enzymes: Structure, Function and
Applications; Polaina, J.; MacCabe, A. P., Eds.; Springer, 2007;
pp 263–341.
63.Hsu, Y.-C.; Gan, K.-H.; Yang, S.-C. Chem. Pharm. Bull. 2005, 53,
1266–1269. doi:10.1248/cpb.53.1266
64.Porumb, D.; Lovász, T.; Răchită, G.; Cristea, C.; Găină, L.;
Silaghi-Dumitrescu, L. Rev. Roum. Chim. 2010, 55, 879–884.
65.Abreu, A. R.; Costa, I.; Rosa, C.; Ferreira, L. M.; Lourenço, A.;
Santos, P. P. Tetrahedron 2005, 61, 11986–11990.
doi:10.1016/j.tet.2005.09.064
4
4
4
4
4
4
0.Lednicer, D.; Mitscher, L. A. Organic Chemistry of Drug Synthesis;
John Wiley & Sons, 1977; Vol. 1.
1.Taurand, G. Phenothiazine and Derivatives. Ullmann's Encyclopedia of
Industrial Chemistry; Wiley-VCH: Weinheim, 2005.
2.Halpern, B. N.; Ducrot, R. Compt. Rend. Soc. Biol. 1946, 140,
66.Carrea, G.; Riva, S. Angew. Chem., Int. Ed. 2000, 39, 2226–2254.
doi:10.1002/1521-3773(20000703)39:13<2226::AID-ANIE2226>3.0.CO
;2-L
361–363.
3.Reynolds, J. E. F., Ed. Martindale: the extra pharmacopoeia, 29th ed.;
Pharmaceutical Press: London, 1989.
67.Laane, C.; Boeren, S.; Vos, K.; Veeger, C. Biotechnol. Bioeng. 1987,
30, 81–87. doi:10.1002/bit.260300112
4.Yanai, K. Clin. Neurophysiol. 2012, 123, 633–634.
doi:10.1016/j.clinph.2011.09.005
68.Janssen, A. E. M.; van der Padt, A.; van't Riet, K. Biotechnol. Bioeng.
1993, 42, 953–962. doi:10.1002/bit.260420806
5.Biel, J. H.; Martin, Y. C. Organic synthesis as a source of new drugs. In
Drug Discovery, Advances in Chemistry Series no. 108; Gould, R. F.,
Ed.; American Chemical Society: Washington, DC, 1971; p 81.
6.Nauta, W. T.; Rekker, R. F. Structure–activity relationships of
H1receptor antagonists. In Handbook of Experimental Pharmacology;
Rocha e Silva, M., Ed.; Springer-Verlag: New York, 1978; pp 215 ff.
7.Toldy, L.; Vargha, L.; Toth, I. Acta Chim. Acad. Sci. Hung. 1959, 19,
69.Janssen, A. E. M.; van der Padt, A.; van Sonsbeek, H. M.; van't Riet, K.
Biotechnol. Bioeng. 1993, 41, 95–103. doi:10.1002/bit.260410113
70.Chen, C. S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J. J. Am. Chem. Soc.
1982, 104, 7294–7299. doi:10.1021/ja00389a064
4
71.Seco, J. M.; Quiñoá, E.; Riguera, R. Tetrahedron: Asymmetry 2001,
12, 2915–2925. doi:10.1016/S0957-4166(01)00508-0
72.Borowiecki, P.; Poterała, M.; Maurin, J.; Wielechowska, M.;
Plenkiewicz, J. ARKIVOC 2012, No. 8, 262–281.
4
273.
4
8.McDonough, J.; Dixon, H.; Niño, J. Intranasal Drug Delivery
Technology for space motion sickness, 01-9303. Southwest Research
Institute, inclusive dates 004/01/2002;
doi:10.3998/ark.5550190.0013.823
73.Borowiecki, P.; Milner-Krawczyk, M.; Brzezińska, D.;
Wielechowska, M.; Plenkiewicz, J. Eur. J. Org. Chem. 2013, 712–720.
doi:10.1002/ejoc.201201245
http://www.swri.org/3pubs/IRD2003/Synopses/019303.htm
9.Boland, E. J.; McDonough, J. Phenothiazine enantiomers as agents for
the prevention of bone loss. WO Patent WO2004/110458 A1, Dec 23,
4
74.Borowiecki, P.; Milner-Krawczyk, M.; Plenkiewicz, J.
Beilstein J. Org. Chem. 2013, 9, 516–525. doi:10.3762/bjoc.9.56
75.Borowiecki, P.; Fabisiak, M.; Ochal, Z. Tetrahedron 2013, 69,
4597–4602. doi:10.1016/j.tet.2013.04.014
2004.
50.Boland, E. J. Phenothiazine enantiomers as agents for the prevention
of bone loss. U.S. Patent US2006/0258650 A1, Nov 16, 2006.
51.Šinko, G.; Kovarik, Z.; Reiner, E.; Simeon-Rudolf, V.; Stojan, J.
Biochimie 2011, 93, 1797–1807. doi:10.1016/j.biochi.2011.06.023
52.Ponder, G. W.; Butram, S. L.; Adams, A. G.; Ramanathan, C. S.;
Stewart, J. T. J. Chromatogr., A 1995, 692, 173–182.
doi:10.1016/0021-9673(94)00997-N
76.Latypov, S. K.; Seco, J. M.; Quiñoá, E.; Riguera, R. J. Org. Chem.
1995, 60, 504–515. doi:10.1021/jo00108a008
77.Kazlauskas, R. J.; Weissfloch, A. N. E.; Rappaport, A. T.; Cuccia, L. A.
J. Org. Chem. 1991, 56, 2656–2665. doi:10.1021/jo00008a016
78.Edwards, M. L.; Stemerick, D. M.; McCarthy, J. R. Tetrahedron Lett.
1990, 31, 3417–3420. doi:10.1016/S0040-4039(00)97411-2
79.Busto, E.; Gotor-Fernández, V.; Gotor, V. J. Org. Chem. 2012, 77,
4842–4848. doi:10.1021/jo300552v
5
3.Chiari, M.; Cretich, M.; Desperati, V.; Marinzi, C.; Galbusera, C.;
De Lorenzi, E. Electrophoresis 2000, 21, 2343–2351.
doi:10.1002/1522-2683(20000701)21:12<2343::AID-ELPS2343>3.0.C
O;2-C
80.O'Meara, J. A.; Gardee, N.; Jung, M.; Ben, R. N.; Durst, T.
J. Org. Chem. 1998, 63, 3117–3119. doi:10.1021/jo9712714
81.Ben, R. N.; Durst, T. J. Org. Chem. 1999, 64, 7700–7706.
doi:10.1021/jo9811625
54.Bosáková, Z.; Kloučková, I.; Tesařová, E. J. Chromatogr., B 2002, 770,
63–69. doi:10.1016/S0378-4347(01)00559-X
55.Lin, C.-E.; Liao, W.-S.; Cheng, H.-T.; Kuo, C.-M.; Liu, Y.-C.
Electrophoresis 2005, 26, 3869–3877. doi:10.1002/elps.200500087
3054

Beilstein J. Org. Chem. 2014, 10, 3038–3055.
8
2.D'Angeli, F.; Marchetti, P.; Bertolasi, V. J. Org. Chem. 1995, 60,
013–4016. doi:10.1021/jo00118a017
4
8
3.Sethi, M. K.; Bhandya, S. R.; Maddur, N.; Shukla, R.; Kumar, A.;
Mittapalli, V. S. N. J. Tetrahedron: Asymmetry 2013, 24, 374–379.
doi:10.1016/j.tetasy.2013.01.023
8
8
8
4.McCreary, M. D.; Lewis, D. W.; Wernick, D. L.; Whitesides, G. M.
J. Am. Chem. Soc. 1974, 96, 1038–1054. doi:10.1021/ja00811a016
5.Hanna, G. M. J. Pharm. Biomed. Anal. 1997, 15, 1805–1811.
doi:10.1016/S0731-7085(96)02042-0
6.Kagawa, M.; Machida, Y.; Nishi, H.; Haginaka, J.
J. Pharm. Biomed. Anal. 2005, 38, 918–923.
doi:10.1016/j.jpba.2005.06.005
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
(http://creativecommons.org/licenses/by/2.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(http://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
doi:10.3762/bjoc.10.322
3055